Annual Review of Immunology Volume 22 2004

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: FRK

February 27, 2004 22:25 Annual Reviews AR210-FM

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104727

Annu. Rev. Immunol. 2004. 22:1–31 doi: 10.1146/annurev.immunol.22.012703.104727 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on December 16, 2003

GENETICS, FACS, IMMUNOLOGY, AND REDOX: A Tale of Two Lives Intertwined

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Leonard A. Herzenberg and Leonore A. Herzenberg Genetics Department, Stanford University Medical School, Beckman Center, Stanford, California 94305–5318; email: [email protected], [email protected]

Key Words glutathione, B cells, flow cytometry, allotype, HIV

INTRODUCTION We (Len and Lee Herzenberg) have worked separately and together for more than 50 years. This blending of independence and mutual reliance is reflected here as we shift back and forth in telling the story of the laboratory we have led and the life we have lived. The space provided for this chapter is very generous. Yet, calculated out, it amounts to roughly 100 words per year for each of us. To make the most of this, we have written an autobiography rather than a history. In many instances, we have referred only briefly, or not at all, to work that had major influences on our thinking. In addition, we have adopted a policy of naming the many students, fellows, and collaborators with whom we have worked only by referring to our joint work with them. We hope the reader realizes there would be no biography worth writing were it not for the contributions made by these and all of our other colleagues.

THE CALTECH YEARS There were nine professors in genetics in the California Institute of Technology (Caltech) Biology Division when I arrived in 1952 as an entering graduate student. They worked with different organisms and taught different areas of biology, but they were united by a common theme—how genes are expressed and how they influence the appearance, physiology, function, and behavior of the organism. The Biology Division at that time was small—one three-story building housed the entire faculty, about a dozen postdocs, an equal number of graduate students, and a couple of undergraduate biology majors. Among the roughly 20 faculty members (visiting or permanent), there were seven Nobelists in the making: George Beadle, Max Delbruck, Ed Lewis, Renato Dulbecco, Roger Sperry, James Watson, and Barbara McClintock. Linus Pauling, Caltech Chemistry chair and winner of

LEN:

0732-0582/04/0423-0001$14.00

1

11 Feb 2004

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

2

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

two Nobel prizes, was in the next building, connected to ours by a much-used corridor. Creative thinking and challenging discussion were the rule; research productivity was the outcome. The development of tools and techniques that removed barriers to experimentation also played a central role in the Caltech Biology Division culture. We routinely used the pH meter and DU spectrophotometer recently invented by former Caltech chemistry professor Arnold Beckman. My thesis advisor, H. K. Mitchell, was an extraordinary glass blower and tinkerer who worked with me on an electrophoresis device. Most things couldn’t be bought, so we had to build them. As I think about it now, the automated fly counter that Ed Lewis built probably sowed the seeds for the Fluorescence-Activated Cell Sorter (FACS) that I developed some years later. In the same way, looking back at the cross-discipline culture in the Caltech Biology Division, I see the origins of the eclectic research goals that Lee and I have pursued. Over the years, we have ranged broadly and drawn our students and fellows into immunology studies as diverse as showing that H-2 antigens are surface proteins, using immunoglobulin (Ig) allotypes and classical genetics to define the Ig heavy chain (IgH) chromosome region, demonstrating IgH allelic and haplotype exclusion in B cells, defining functional subsets of T and B human and murine lymphocytes, cloning and sequencing lymphocyte surface markers, identifying fetal cells in maternal circulation, understanding redox influences on transcription factor activation, and doing clinical studies to characterize and treat the glutathione deficiency in HIV infection. However, the twin themes of genetics and somatic cell function that guided (and still guide) this work, and the love of reading and talking about these diverse areas with people of different scientific interests, are well rooted in Caltech tradition. Political activism was also important at Caltech. Joseph McCarthy, the Senator from Wisconsin who made a career of finding communists under every bed, was threatening to disrupt academic and personal freedom. In response, we joined Linus Pauling, Matt Messelson, George Streisinger, and Arthur Galston, as well as other faculty, students, and fellows in establishing a Federation of American Scientists chapter and in protesting this “witch hunt.” A portion of Lee’s and my life ever since has been devoted to helping the United States be the kind of country we want our children and children’s children to grow up in. LEE: When Len left Brooklyn for Caltech, he was 21 and I was 17. Logically, because he had three years of graduate work ahead of him and I had three years to finish my undergraduate degree, we decided that we would marry when we both finished school. However, logic couldn’t overrule the three-thousand-mile distance, the loneliness, or the $3/minute (1953 dollars) the telephone company charged young lovers just to say hello. By the end of the fall of 1952, Len urged me to apply to schools near Caltech, and when I was admitted to Pomona College in Claremont we set a wedding date for the coming summer. Needless to say, our parents thought we were too young, too innocent, too poor, and too crazy. They were probably right. But we got married anyway, with their

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TWO LIVES INTERTWINED

P1: IKH

3

blessings, and set off on an adventure that is as exciting today as it was the day we piled our stuff into the car Len’s parents gave us and started the long drive across the country to Pasadena. At Caltech, I was as enchanted as Len was by how immersed people were in their work. I had already selected biology as my major, but I had no idea how interesting it could be. For the rest of the summer, until the fall term started at Pomona, I went to the lab every day with Len. I attended seminars; I read in the library; I learned how to help Len with experiments; sometimes I even had the courage to ask questions of someone other than Len. It was just plain fun, and I couldn’t get enough of it. Pomona College, on the other hand, turned out to be a disappointment. It was an excellent school. There is no question about that. The liberal arts courses were wonderful. But the Biology Department was still teaching gram stains (to identify bacteria) and grilling us on the anatomy of flowers and reptiles. Meanwhile, Jim Watson had just brought the double helix back to Caltech and was teaching about it while I was sitting in a classroom learning things useful only to a stodgy highschool biology teacher. I would have enrolled as a Caltech undergraduate, but women weren’t even admitted to Caltech graduate programs (there were only a few women postdocs and research associates). Nevertheless, the biology faculty believed that women were educable and worth educating. By the beginning of the second semester of the academic year, they worked out an auditing program for me in which I would be treated, and graded, like a Caltech biology major. They allowed me to take whatever courses I wanted to, and even found a part-time job for me so that Len and I could afford to eat. For each course I completed, the professor gave me a letter certifying that I had met the course requirements and received a grade (always As, as it turned out). So, although I didn’t get formal credit, I managed to take classes such as virology from Max Delbruck, bacteriology from Renato Dulbecco, and immunology from Ray Owen. I learned how to think about science from these teachers. Perhaps the most formative event for me during our time at Caltech, though, was a dinner with Barbara McClintock, who had recently come to the Biology Department as a revered visiting professor. Len and I had decided to drive into Los Angeles for a Chinese meal and were about to leave when we noticed that the only light left on in the building was coming from Dr. McClintock’s door. We peeked in and saw that she was working alone. Then we drew back into the shadows, debating whether we should tiptoe away and not interrupt her, or whether she might actually like to take a break and join us. Finally, we screwed up our courage and asked her if she would like to go. “I’d be delighted,” she replied, and off we went. At dinner, I naively asked Dr. McClintock how she made such wonderful discoveries. Her answer, simple and straightforward, became my long-standing rudder. She said that in the course of her work, she occasionally got a surprising result that could not be reconciled with existing theory. First, she would decide whether she believed the “exception”—in other words, she could not see any technical or

11 Feb 2004

4

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

interpretive flaws that undermined it. Next, if she believed it, she would commit it to memory and compare it with any other exceptions she had come across. Ultimately, a constellation of exceptions would coalesce to yield a testable hypothesis that, if validated by additional experimentation, would provide the basis for extending or altering the current paradigm. Len and I have never forgotten that dinner with Dr. McClintock. Somehow, it must have been meaningful for her as well. When I next met her, some ten years later, I introduced myself by name and started to say, “Dr. McClintock, you may not remember me. . .” when she cut me off with, “Oh, you’re the people who took me to that Chinese restaurant in L.A.,” and she greeted me thus ever after.

THE PARIS YEARS Len defended his thesis in August 1955. We left immediately for Paris, where Len had organized a postdoctoral fellowship with Jacques Monod at the Pasteur Institute. My childhood friend who had recently returned to her native France met us at the boat, found us a room in a student hotel in the Parisian Latin Quarter, and introduced us to the life poor students lived in Paris. It was great! And so was the laboratory at Pasteur. Jacques Monod’s laboratory at Pasteur was physically separate but intellectually allied with Andre Lwoff’s laboratory two floors above. Francois Jacob, the third member of the trio later awarded the Nobel Prize for their seminal molecular biology studies, worked in the Lwoff laboratory. The two laboratories lunched together virtually every day at a single long table in a small atrium on an inner Institute courtyard, where huge glass vessels rumored to have been used by Louis Pasteur were stored. Lunch was marvelous. It wasn’t a formal seminar, but conversation revolved around science in the lab and the world at large. Findings were analyzed, theories debated, visitors questioned. Every day was an intellectual feast. There were also many first-person history lessons about the days before World War II and what the war was like in France. Monod was a major figure in the resistance against the Nazis. Francois Jacob had been to Algeria and North Africa and participated with the Free French in the liberation of France. Georges Cohen, one of the Monod senior scientists and still a close friend today, survived the war as a Jew in France and talked about things he did in the Resistance. We heard stories about how the laboratory hid Jewish scientists when the SS came knocking. It all sounded very romantic in 1955, ten years after the war ended. But as Georges and Jacques Monod often reminded us, it wasn’t much fun when it was happening. I should mention that, like the Caltech Biology Division, the group at Pasteur accepted Lee as an unofficial student. At the beginning, she was pregnant with our first child and spent most of her time working with me. Later, she brought the baby to the lab most afternoons and continued working.

LEN:

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

TWO LIVES INTERTWINED

P1: IKH

5

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Cradle of Molecular Biology This was a very exciting time at Pasteur. The characteristics of the β-galactosidase (LacZ) operon were unfolding before our eyes as each new piece of work was completed. Under Monod’s leadership, I contributed two pieces to the puzzle: I showed that the galactoside-concentrating mechanism encoded by the permease gene in the LacZ operon increases the internal concentration of inducers that upregulate expression of the LacZ operon genes and thus is responsible for the autocatalytic increase in LacZ induction. In addition, I showed that the inducers are acetylated, rather than phosphorylated (as people had thought), during LacZ induction, thus opening the way to adding (after I left Pasteur) the β-galactoside acetylase gene to the LacZ operon. These and other findings led Monod, Jacob, and Lwoff to the discovery of the LacZ operon, which laid the groundwork for much of modern molecular biology. For this, they received the Nobel Prize. While at Pasteur, I met Melvin Cohen, who later became a Salk Institute immunologist. Mel had worked with Monod and returned to visit several times. Conversations with him then, as always, were highly stimulating. His presence in the Stanford Biochemistry Department was a key motivation in my decision to move to Stanford when the opportunity arose several years after I left Pasteur. LEE: Aside from birthing a baby and learning to balance being a mom with being a scientist (albeit only a budding one), I don’t have too much to show for my time at Pasteur. I did do one independent piece of work, but it wasn’t well received. Monod had several times said that the thiogalactosides that we used to induce expression of LacZ operon genes were unnatural compounds that could not be digested by bacteria. This didn’t seem right to “wise guy” me. So I went out and scooped up some fresh Parisian soil, put it into a flask with minimal medium containing thiogalactosides as the only carbon source, and put the flask into the cabinet under the bench. About a week later, the medium in the flask was cloudy, and a clear sulfur smell wafted out when I opened the top. Something was clearly growing and “knew how” to break down thiogalactosides. Excitedly waving the flask, I went to Jacques’ office to show him my prize. He was quite surprised and, in a manner I hope I have learned, graciously said he was pleased to be wrong in this case. Some time later, however, everyone in the laboratory was ready to kill me. While we never had to sterilize thiogalactoside stock solutions before my little experiment, after I opened Pandora’s flask, all of the thiogalactoside stocks got contaminated. From then on, they all had to be sterilized immediately after they were made! This incident aside, I mainly spent my time at Pasteur helping Len. Because I was rather sedentary during the first year, and because Len loved hands-on experimentation, I took over much of the data recording, computation, and display (plotting) that was needed. The work was tedious (slide rules were the closest thing to computers at the time). However, it gave me the opportunity to do a preliminary analysis of the data and try novel approaches to analyzing LacZ induction kinetics. Len left this to me. He was more interested in developing methods and

11 Feb 2004

6

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

experiment designs that would enable clear conclusions without a lot of mathematical interference. This division of labor, which reflects Len’s innate preference for concreteness and my innate love for theory, remains with us even today.

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE NIH YEARS Just about the time that Len and I were considering what to do after Pasteur, an ominous postcard caught up with us. It had spent rather a long time traveling to France by surface mail, and it announced that Len should have reported for active duty in the U.S. Army several days before the postcard was loaded onto the slowest boat on the Atlantic. Late or not, the postcard made it quite clear that Len had been drafted! We immediately went to Jacques Monod for advice. He was as adamant as we were that it would be a pity to interrupt Len’s scientific career to serve in a peacetime army. “Why don’t you consider going to the National Institutes of Health, my boy? I have just had an inquiry from Harry Eagle looking for a fellow for his laboratory. He should be able to arrange for you to serve in the Public Health Service instead.” This was a shock, but I was not displeased with the idea of going to Harry Eagle’s laboratory. I had already been thinking about doing genetic studies with mammalian somatic cells. What better place to learn how to grow cells than the laboratory that had just developed Eagle’s medium? Leaving Pasteur and the Escherichia coli world would not be easy. But the challenges presented by mammalian studies would also be exciting. So, without further ado, I decided that I was lucky to have the opportunity to carry a pipette rather than a gun for my country and asked Jacques to write to Harry Eagle on my behalf. It took several months to untangle the draft board and Public Health Service mess and to wrap up my work in Paris. But by the summer of 1957, Lee, Berri (our toddler), and I were settled in the Bethesda area, and I began work at the NIH. Eagle’s laboratory operated with more of a top-down structure than Pasteur and lacked some of the intellectual and scientific excitement I was used to. However, my colleagues in the laboratory, notably Robert DeMars (now at the University of Wisconsin) and James Darnell (now at Rockefeller University), were great. “Captain Harry,” as we sometimes called Harry Eagle (to his face as well as behind his back), was focused on determining the nutritional conditions necessary to establish and maintain long-term cell lines. I was, too, because to do mutation and selection studies I needed to establish conditions that would allow individual clones to grow. My finding that adding pyruvate to Eagle’s medium was sufficient to support clonal growth let me begin exploring drug resistance markers for genetic studies. In addition, it led to the addition of pyruvate as a normal constituent of the medium. The Federation of American Scientists came back into our lives shortly after we arrived in Washington, because the national office was only a short distance

LEN:

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

TWO LIVES INTERTWINED

P1: IKH

7

from the NIH. Picking up where we left off at Caltech, we volunteered to work on the newsletter and do some administrative work for the organization. Ultimately, we helped to reorganize the office and put the administrative oversight into the hands of a liberal Washington, DC, law firm.

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Lee and the Salmonella Histidine Operon LEE: A few weeks after we arrived at the NIH, Len ran into Bruce Ames, a Caltech buddy who had just been appointed to a permanent staff position at the NIH. As luck would have it, Bruce was looking for his first technician and I was looking for my first job. Len made the match. Poor Bruce. No one should have had to put up with me as a technician. I was always asking “why” and looking for better ways to do things. As a graduate student, I would probably have been fine. But as a technician responsible for doing work that someone else gave me to do, and generating data that someone else was supposed to interpret, I was clearly a pain in the neck. Nevertheless, Bruce put up with me, and I learned to get the work done. I owe him a great deal. Bruce was working on the characterization of the enzymes in the histidine synthesis pathway in Salmonella. We had completed the work on three of the enzymes when Bruce left for a month to work in Arthur Kornberg’s department in St. Louis. As he went out the door, he handed me a tube containing the substrate for the last enzyme in the pathway and asked me to characterize that enzyme as we had the others. The only problem was that when I took the spectrum of the substrate, I found that its synthesis had gone wrong. I had no substrate to work with. Long-distance telephone calls, at that time, were very expensive. It was unthinkable to try to call Bruce and ask for instructions. So I took the question to one of the senior investigators in our department. “Find something useful to do. Bruce will be home soon,” he responded. I cogitated over this for a bit and then decided to apply some of the operon thinking I learned in Paris to the histidine pathway. I tried out some conditions I thought would reveal coordinated regulation of the expression of the enzymes that were already characterized and, to my surprise, readily found such conditions. Bruce was pleased with this finding when he returned, but he put it aside until the entire pathway was properly characterized. After a couple of months, I left Bruce’s lab because I was getting along in my second pregnancy. Bruce later completed the operon study with Barbara Garry. He included me as an author on the paper, which became my first peer-reviewed publication (1).

THE STANFORD YEARS The opportunity to move to Stanford came as a complete surprise. Joshua Lederberg visited Harry Eagle’s laboratory toward the end of my required Public Health Service “hitch.” We had a long discussion about the future of mammalian somatic cell genetics and the progress I had made thus far in developing useful

LEN:

11 Feb 2004

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

8

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

markers for the cell lines I had chosen. I was a bit preoccupied at the time because I was in the midst of negotiating a permanent position at the NIH. However, the talk with Josh was really a delight. Later in the week, Harry Eagle called me into his office and suggested that I delay a bit before making any commitments at the NIH. I didn’t make the connection with Josh’s visit and was somewhat mystified. However, a few days later, a letter arrived inviting me to consider a faculty position in the Genetics Department that Josh was in the midst of establishing at Stanford! Our parents considered the offer a disaster. California was still a long, long way away from Brooklyn, and they now had two grandchildren they wanted to help raise. Lee and I were also somewhat negative about returning to the West Coast. However, just after the offer came, we made a trip to New York that reset our direction. We came to New York so I could attend the annual Federation of American Scientists meeting. Martin Kamen also attended the meeting and wound up walking with me at its close from upper Manhattan to the Times Square subway station. I had met Martin when I was at Caltech, at a benefit party Linus Pauling gave to help him raise funds for his legal fees. He was fighting the INS decision to revoke his passport (another McCarthy victim). We talked a bit about this, and then I told him about the possibility of going to Stanford in Josh’s new department. By the end of the walk, I realized that going to Stanford was a chance of a lifetime and that there was no way I could turn down a position there, if I got it. Why was Stanford so exciting? Well, with urging from Henry Kaplan, head of Radiology at the Stanford Medical School and the pioneering developer of treatment protocols for Hodgkins disease, Stanford President Wallace Sterling had mustered the resources to upgrade the Medical School to a first-rate institution with the twin goals of forefront research and excellent clinical practice. Arthur Kornberg, who would win the Nobel Prize in 1959, was recruited as chairman of Biochemistry, and he in turn recruited the cream of the department he chaired in St. Louis. Joshua Lederberg, who couldn’t interview me until February (1959) because he had a date with the Nobel Prize in December 1958, was brought in as chairman of Genetics. I was the first faculty member Josh recruited. Josh’s (first) wife Esther was also in the department. One reason she and Josh had chosen Stanford was that, unlike Berkeley and many other schools at the time, there were no nepotism rules at Stanford that prevented her from working with Josh. I noted this, although Lee planned to look for a job in another department or possibly go back to school to get the degree(s) she wanted. In September 1959, the Medical School began moving into its new building, which housed both the basic sciences and the hospital. Lee and I (and our two children) arrived just as this was happening. The Biochemistry Department was already in its quarters. Space had been opened for Josh’s lab and the Genetics office, but little else was ready. The landscaping had not even been started, so the building stood in the middle of a hot, dusty field.

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TWO LIVES INTERTWINED

P1: IKH

9

I was given temporary laboratory space in the nearby Applied Physics building and set up my lab so that I could get the cultures that I had shipped from the NIH growing. But California weather was not kind. September that year turned out to be mercilessly hot, and the building I was in didn’t have air conditioning. As it turned out, it would have been better if my incubator had had a water cooled, rather than a water heated, jacket! Fortunately, I was able to recover from frozen stocks much of what I had lost. Lee began working with me around this time. I had already applied for and gotten a grant to support my somatic cell genetics work. The funding was available, but the delays in completion of the Genetics space put everything else into chaos. I had made a list of equipment I wanted purchased before I arrived, but none of it had been ordered. The Genetics office was overworked and understaffed, and my cultures were cooking in the incubator. Lee decided to pitch in for a while to help me get started. Best decision we ever made! Stanford was great for another reason. During McCarthy times, the University of California and many other schools required faculty to sign a loyalty oath swearing that they were not now, and had never been, a member of the Communist Party or any other organization that advocated the overthrow of the federal government. Because the list of proscribed organizations was created at the whim of people who rose to power ferreting out supposed communists, its sweep was extremely broad. Many faculty members found the requirement of a loyalty oath repugnant and refused to sign. Stanford supported this view by refusing to institute a loyalty oath and by hiring people who left other institutions rather than sign such an oath. A number of eminent Berkeley physicists moved en masse from the Berkeley to the Stanford Physics Department. We were pleased to have the opportunity to meet and work with these physicists in the years that followed.

We Become Immunologists Shortly before coming to Stanford, Josh Lederberg had spent some time in Australia with Sir MacFarlane Burnet, who was head of the Hall Institute in Melbourne. Josh and Sir Mac applied genetic thinking to the immune response and came up with the idea that antibody responses reflect the clonal selection of cells that are individually committed to producing antibodies that recognize, and are triggered by, the immunizing antigen. This so-called clonal selection theory predicted that individual cells would make antibodies specific for a single antigen and stood in opposition to instructive theories that predicted much more plasticity for individual cells. The clonal selection hypothesis ultimately won out. However, at the time Lee and I arrived at Stanford, the jury was still out. To do the studies that would test this hypothesis, Josh appointed two young visiting faculty members: Gustav Nossal, who later followed Burnet as head of the Hall Institute, and Olli Makela, who later returned to Finland to do immunology

11 Feb 2004

10

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

research and eventually became Dean of the Medical School in Helsinki. Gus and Olli, and the mouse facility they would need, were to be housed in the same corridor as the lab being completed for my use. So the move to the new building, which occurred about December 1959, put my somatic cell genetics group right next to one of the most exciting immunology projects of the time.

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The First Immunology Studies LEE: By the time we came to Stanford, Len had already developed an interest in the practical side of immunology. Just before leaving the NIH, he arranged to visit George Snell at Bar Harbor to discuss the idea of using mouse histocompatibility antigens, rather than drug sensitivity, as somatic cell markers in cultured cell lines. Len thought it would be neat to use cytotoxic antibodies to the H-2 antigen (thought to be a single entity at the time, now recognized as the MHC) to select cell surface antigen variants in lymphocyte and other cell lines. However, he wondered whether this would be practical. So he went to talk to George, who was very encouraging and offered some antibodies for this purpose in case Len needed them. This idea lay fallow until we sorted out all the problems involved in getting the lab set up. However, once this was accomplished, Len suggested to me that I take on the job of anti-H-2 antisera so that we would have our own reagents with which to select variants. Of course, I had never touched a mouse and knew nothing about how to proceed other than what I could read. Nevertheless, I took on the job. Len and I both liked it as a project for me because I could work independently at my own speed without creating for him a bottleneck on a critical path. To figure out how to start, I went knocking on Gus and Olli’s door. They did indeed know how to proceed, and they showed me how to take out spleens, use spleen cells to immunize the mice, and do tail bleeds to collect the sera. They weren’t much help in setting up the erythrocyte agglutination assays that were used at the time to titer the sera. However, with their and Len’s advice and several quite good papers on the subject, I managed to get a test going and learn to reliably read it. There was already a great deal of serologic evidence characterizing the genetically distinct H-2 antigens expressed by various mouse strains. Because C57BL mice were known to make strong antibodies to the DBA/2 H-2, and because both kinds of mice were available from a local commercial breeder, I chose this combination. In addition, because female C57BL mice that had been retired from the breeding colony were large and could be obtained quite cheaply, I chose these mice to immunize. I got very good responses and was able to collect lots of good antisera that Len could use for selection. Surprisingly, however, some of the control sera that I took from the breeders before immunization turned out to have low but clearly positive levels of antibodies that agglutinated DBA/2 rbc but were clearly negative against the serum producer strain (C57BL). A bit of detective work soon showed that many of the retired breeders that we had purchased had been out crossed to DBA/2 to make F1 animals

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

TWO LIVES INTERTWINED

P1: IKH

11

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

that were in high demand. Thus, by following my nose, I had discovered a potential model for human Rh immunization during pregnancy (2). Some time later, I showed these data to Ray Owen, the Caltech professor who taught the immunogenetics course I had taken. Ray shocked me by asking when I was going to publish it. I stammered and stuttered a bit until Ray finally said, “Well, if you are serious about being a scientist, then I guess you have to publish this.” So I did. For this and many other reasons, I often refer to Ray as the closest I ever had to a graduate professor.

Focus on H-2 Antigens With a plentiful supply of anti-H-2 antisera, I decided to phase out my drugresistance work and focus on using these sera for genetic studies with mouse cell lines. First, however, I needed to do some characterization of the H-2 antigen, at least to the point where I knew what it was. There was general confusion on this at the time. Immunologic evidence had located H-2 on the cell surface of many cell types. However, while some people thought the antigen was composed of carbohydrate or protein, no lesser a light than Peter Medawar, who would later be awarded the Nobel Prize for discovering adaptive immune tolerance, thought that H-2 was made of DNA. We soon laid this issue to rest by isolating plasma membranes and characterizing the H-2 antigen(s) associated with the membranes as a protein or glycoprotein (3). Working on H-2 drew me ever closer to the immunology community at Stanford. Gus and Olli became close friends as well as wonderful colleagues who loved discussing science as much as I did. Avrion Mitchison, who later headed a productive Immunology Department at University College, London, was also appointed as a visiting professor by Josh and began occupying the lab next door within the year. Together, we established an immunology journal club, which met one evening a week at my house as a no-holds-barred discussion in which we examined methodology, evaluated experiment design, questioned conclusions, and argued theory. The descendant of this journal club still functions in our laboratory today, with much the same rules. Gus and Olli were highly focused on testing the clonal selection theory (4) during this time. Their approach was to isolate individual antibody-producing cells and determine whether a single cell made antibodies to one or both of a pair of immunizing antigens. Their data, although limited by the number of cells they could isolate and test, clearly favored clonal selection. Mel Cohn, in the Biochemistry Department three floors up, with colleagues Lennox and Attardi at other institutions, were holding down the instructive corner of the argument. At the time all this was happening, I didn’t have a notion that I would one day develop an instrument (the FACS) that would make it possible to resolve this question. However, once we got the FACS running, we returned to these issues in studies that became a major focus of our laboratory for several years.

LEN:

11 Feb 2004

12

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Enter Immunoglobulin Allotypes In the spirit of the times, the clonal selection debate did not sour my relationship with Mel Cohn, with whom it has always been fun to argue about anything. In any event, when Mel decided to leave Stanford in 1962, he “willed” the medical student working in his laboratory to me. John Wunderlich thus joined our group, bringing with him a project focused on producing antisera that would distinguish between antibody molecules produced in different mouse strains and putatively encoded by different alleles in those strains (5). Ultimately, this project blossomed into a full-scale study of the genetics of the Ig heavy-chain (IgH) chromosome region (6). Long before the structure of the IgH region was defined by molecular methods, studies with anti-isotype and antiallotype antisera showed that IgH isotypes are encoded by a series of closely linked loci and that various mouse strains have distinctive alleles at these loci. The IgH isotypes were defined by other laboratories; we produced many of the antiallotype sera and used these sera in genetic studies (gel immunoprecipitation and radioimmune assay) to demonstrate the close linkage of several of the IgH (isotype) loci. In addition, we defined a series of IgH haplotypes based on the combinations of alleles represented at the IgH constant region loci on the IgH chromosome in each of the standard mouse strains and showed that these were codominantly inherited. Interestingly, the IgH haplotypes defined in this way provided the basis for the Jan Klein and Don Shreffler model for organization of the MHC chromosome region. We reported our IgH genetic studies at a Cold Spring Harbor Symposium (7) at which Henry Kunkel presented evidence for similar close linkage of the human IgH loci. Together, these studies solidified a paradigm that was extended by the recognition of additional loci and haplotypes and laid the groundwork for the modern understanding of Ig rearrangement, isotype switching, and haplotype (allelic) exclusion during B-cell development.

Regulation of Memory-B-Cell Expression LEE: Although I was working actively on allotype genetics, I maintained an independent interest in maternal immunization to fetal antigens (H-2 in particular) and in the effects of such immunization on the developing fetus. Therefore, when maternal antiallotype antibodies were shown to pass to the fetus and to delay the initial appearance of the paternal allotype in allotype heterozygotes, Len suggested that I get this to work with some of our mouse strains. I did, further extending my independent work in the lab. Ultimately, Len’s suggestion led us to the discovery of “chronic” allotype suppression. This occurs when SJL males are mated to immunized BALB/C females producing high-titer antiallotype antibodies reactive with the paternal Igh-1b (IgG2a) allotype. This finding then led to the discovery and characterization of CD8 suppressor T cells that control the expression of IgG2a (Igh-1b) memory B cells without impacting survival of the memory population (8).

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TWO LIVES INTERTWINED

P1: IKH

13

While exploring the mechanism(s) underlying allotype suppression, we recognized (as others had before us) that priming with typical protein antigens enabled a strong secondary response to the determinants present on the priming antigen. Such priming also resulted in suppressed responses to new epitopes such as haptens introduced on the priming antigen at the time of the secondary challenge. Because this suppression persists when carrier/hapten-carrier-immunized animals are challenged with the new epitope (hapten) on a different carrier protein, we refer to it as epitope-specific suppression (9). Although these studies were highly rigorous, they were not met with universal acclaim, perhaps because of the confusion they introduced and perhaps because they occurred just at the dawn of the molecular era in immunology. Nevertheless, the findings are “alive and well” in the vaccine world, where they have been confirmed with a variety of antigens and provide an important caveat when generating vaccine strategies. Similarly, the immunoregulatory-circuits model we had constructed just prior to beginning the epitope-specific work (10), which predicted much of what we found, was not roundly embraced by immunologists, but it too is alive and well, I am told, among today’s immune-system model builders.

THE FACS As I became more deeply involved in immunology, I became increasingly aware of the need to characterize and isolate the different kinds of lymphocytes that were beginning to be visualized with fluorescent-labeled antibodies under the microscope and studied functionally by sensitivity to complement-mediated depletion after treatment with antibodies (in conventional antisera). The need for better cell-isolation methods here dovetailed completely with the need for developing a method for positive selection of variants in the somatic-cell genetics projects that I was also engaged in. So I started asking around to see whether anyone had solved this problem. I soon found out that a group at Los Alamos (led by Mack Fulwyler and Marvin Van Dilla) had developed a machine that could examine and sort large numbers of cell-sized particles on the basis of particle volume. I immediately planned a trip to see whether I could convince them to add a fluorescence-detection system so I could use their machine to measure the amount of fluorescence associated with individual cells and to sort cells according to this measure in addition to volume. They demurred, saying that this “was not part of their mission.” [They were funded to build a machine to count and size particles, not cells, obtained from the lungs of mice and rats sent up in balloons to inhale debris generated by atomic-bomb testing (11).] I persisted, and they finally agreed to give me a set of engineering drawings and the permission to use them as the basis of a machine designed to distinguish cells labeled with fluorescent antibodies. Little did I know when I brought these plans back to Stanford that I was starting on a lifework that continues today as a major activity in our laboratory.

LEN:

11 Feb 2004

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

14

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

Back at Stanford, I took advantage of my close proximity to the instrumentation research laboratory set up by Joshua Lederberg to look for life in outer space (on Mars or on the Moon). I asked the engineer I knew best to look at the plans and estimate the cost to replicate the Los Alamos machine. He came to me a few days later and said, “Okay, I’ve got good news and I’ve got bad news. Which do you want to hear first?” I opted for the good news, and he said, “Well, I think the machine can be built here and I’ve completed a list of parts to be ordered.” I asked what it would cost, and he answered, “Something like $14,000.” That was a lot for those days, but it could probably be managed. So I asked him, “What’s the bad news?” He answered, “The bad news is that I’m leaving Stanford. I’ve got another position.” I next talked to Josh and the head of the Instrumentation Laboratory. They agreed that despite the loss of this key engineer, they could provide the engineering help I needed for the project. I went to Henry Kaplan, who was head of the Radiology Department and was working on thymic function and development (Irving Weissman worked with him). I told him how I thought a fluorescencebased cell-analysis and -sorting machine could be used to study the thymus and asked him to join me in funding the development of this machine. He agreed. I put up $7,000 from my somatic-cell genetics grant, he put up the remaining $7,000 needed to meet the estimate, and the project got under way. I didn’t do any of the engineering on this project. However, I was deeply involved in the daily development. I was essentially the head of the design team and took responsibility for assuring that the machine would be usable by scientists doing immunological or genetic studies. For example, at one meeting, the engineers told me that the best they could do was to take data from about one million cells in an hour. This was too few to be useful, so I insisted that they either increase the speed by an order of magnitude or close the project down. At first there was some discussion about “repealing the laws of physics,” but eventually an engineer came up with a solution and we were off and running again. I was also responsible for getting new capabilities designed and tested. I loved this role because it encouraged me to think broadly about potential applications for the nascent FACS and to develop collaborations within and outside our laboratory to generate and test these kinds of ideas. In fact, although FACS development has long since ceased to be an activity occurring solely within my purview, I still enjoy the development of new FACS applications and the scientific breakthroughs such development engenders. Our first cell-sorting paper was published in Science in 1969 and was entitled “Cell sorting: automated separation of mammalian [plasma] cells as a function of intracellular fluorescence” (12). The instrument we used for this study had a xenon light source, which we replaced with a laser shortly thereafter. By 1972, we had developed a much improved instrument and decided to call it the Fluorescence-Activated Cell Sorter (FACS). The engineering team was also much improved because I was able to recruit Richard Sweet, inventor of the ink-jet printer, to head the team. In essence, I pointed out to Dick that the sorting module

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

TWO LIVES INTERTWINED

P1: IKH

15

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

in the FACS was based on his invention and asked him to join our group. He responded, “There’s nothing I’d like to do more. I’d like to see biological applications of my inventions.” And see he did, as he applied himself to the development of several of the core features still with us in the modern FACS instrument. Dick’s initial work generated a paper, published in the Review of Scientific Instruments, that was really the first one describing the modern FACS (13). He also joined us as an author of a 1976 Scientific American article in which we introduced the FACS and the idea of using this novel instrument to track the expression of genes encoding surface molecules that distinguish various kinds of lymphocytes and other cells (14).

FACS Goes Commercial The next major milestone in the development of the FACS was a meeting I had with a vice president of Becton-Dickinson (BD), parent company of the current BD Biosystems (BDB), and with Bernie Shoor, then a local BD representative who ran an engineering group in the Stanford area. Bernie and the vice president came to me because they wanted help with making (conventional) antibodies. I changed the subject and said, “Well if you’re interested in making antibodies, then you’re interested in immunology. The most exciting thing in immunology right now is our fluorescence-activated cell analysis and sorting [FACS] instrument, which we have been developing for some time and is now working!” Bernie was interested in the machine but didn’t think it was commercially viable. “I think maybe we could sell 10 of these instruments worldwide,” he said. I thought 30, or possibly as high as 100 sales were more likely, but neither number seemed high enough to BD to support turning FACS into a commercial machine. BD, in the person of Bernie Shoor, would have walked away from the venture if I hadn’t gotten an NIH contract that would let me subcontract the building of two such instruments to Bernie’s group and let me collaborate in the effort. It’s also true, though, that I might not have gotten the contract if I did not have BD on board to build it. We eventually built two FACS instruments, one for Stanford and one for the National Cancer Institute (NCI), which put up the money as part of the “war on cancer” (15). Within a short time (as such projects go), we had a commercial instrument to replace “Whizzer,” the breadboard model we had been running until then. FACS-1 was later upgraded to FACS-2, which ran for many years both in our laboratory and at the NCI. Ultimately, the NCI instrument wound up in an NIH museum. Our instrument went to the Smithsonian Institute in Washington, DC, and is presently on display at the Walter Reed Army Institute of Pathology. The Walter Reed exhibit includes a tape recording of me describing some of the early work our group did and some of the work done by Bernie Shoor and his group. After the NIH contract work was completed, BD successfully marketed FACS-2 and we continued our independent development effort. Within the next few years,

11 Feb 2004

16

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

we had made some key improvements, including the addition of fluorescencecompensation circuitry to correct for spectral overlap between dyes and fourdecade logarithmic amplifiers to allow the full range of FACS data to be displayed on a single data plot. In addition, we introduced the use of computers for data collection and built the first software for FACS data computation and display.

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Monoclonal Antibodies as FACS Reagents LEE: In 1975, just about the time that Cesar Milstein and George Kohler succeeded in immortalizing spleen cells that produce antibodies by fusing them with a longestablished myeloma cell line, Len arranged a sabbatical in Cesar’s laboratory at the Medical Research Council (MRC) in Cambridge, England. He chose the MRC to learn the new molecular biology methods (which he did). However, by the time we reached Cambridge in the fall of 1976, the fusion work was in full swing, and Len was quick to realize that the ability to produce monoclonal antibodies to cell-surface determinants would remove what had come to be the most irritating restriction to FACS work at the time. The conventional antibody reagents that we were using for FACS studies were made primarily in mice or rats and were always in short supply. Furthermore, the specificity was always questionable because the animals were immunized with cell preparations that contained many different potential antigens. Finally, the ability to produce directly conjugated reagents was very limited, making background staining by the second-step reagents a major problem. No wonder then that Len was anxious to tap this new monoclonal reagent resource. Cesar, on the other hand, had not attended the many meetings we had at which discussions of potential problems with conventional antibody reagents had been narrowed down to the need for groups to exchange staining reagents before the findings could be evaluated. Therefore, Cesar was not highly motivated to have Len delve into the monoclonal technology and urged him instead to pursue the molecular biology studies he had come to do. The solution to this came when Vernon Oi, then a graduate student in our laboratory at Stanford, came to Cambridge for a prolonged stay. I was working, in principle, at the Babraham laboratories with Arnold Feinstein. However, I had not made much use of the space Arnold gave me because I had to write several chapters for the Weir Handbook of Experimental Immunology as well as several papers that had piled up before I left Stanford. Arnold was pleased to let Vernon take my place, and with agreement from Cesar, he outfitted a laboratory in which Vernon (with help from Len and me) could make a set of monoclonal antibody reagents that would detect allotypic determinants on IgG molecules (16). We brought this technology home at the end of the sabbatical year and, with a FACS available for screening for antibodies to cell surface determinants, began making a series of monoclonal reagents (17) to mouse MHC and other cell-surface molecules (16, 18). Shortly thereafter, we made a unilateral decision to make our monoclonal reagents, and the cell lines that produced them, freely available to the

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TWO LIVES INTERTWINED

P1: IKH

17

scientific community. I had the pleasure of announcing this at a large MHC workshop meeting and was pleased when Baruch Benacerraf approached me after the session to compliment our laboratory specifically on this decision. Breaking into the circle surrounding me, he said, “I would like to shake your hand.” And he did! Len also recognized at this time that distributing cell lines that produce important monoclonal reagents would not be sufficient to ensure the availability of these reagents to the overall immunology community. While still in Cambridge, he had phoned Bernie Shoor to suggest that he get BD to set up a commercial mechanism for producing and distributing monoclonal FACS reagents. It took some time for this to occur, but BD ultimately set up a business whose growth and importance to research and medical practice has well validated the original idea. Interestingly, neither BD nor we thought it necessary or appropriate to patent the monoclonal reagents that the BD monoclonal center was producing, or even to restrict the dissemination of the cell lines that produce these antibodies. Bernie felt, and was proven correct, that people would prefer to buy well-characterized fluorochrome-conjugated reagents rather than produce these reagents themselves. This view is probably more correct now than it was at the time. However, it became untenable as patents for biological material became commonplace and suits for patent infringement began invading the biomedical arena. Some time later (1982), Vernon Oi and I teamed up with Sherie Morrison (on sabbatical leave with Paul Berg at Stanford at the time and now at UCLA) to make human/mouse chimeric antibodies in which the antibody specificity was encoded by variable-region genes derived from mouse and the heavy-chain constant region was encoded by human IgH genes (19). Because we believed that chimeric antibodies of this type were likely to be useful as functional antibodies and therapeutic reagents, we applied for a patent for this molecular method (issued in 1998). We have been pleased to see the method applied by others, e.g., in the production of chimeric anti-TNF-α used in the treatment of human autoimmune diseases.

FACS: The First Biotech Instrument? If the biotechnology (biotech) industry can appropriately be characterized as an industry built around defining, measuring, and making use of gene expression in biology and medicine, then the FACS as we built, described, and used it in the early 1970s readily qualifies as a biotech instrument. In fact, to our knowledge, it is the first such instrument. In addition, Garry Nolan (Medical Microbiology and Immunology, Stanford) points out that FACS should be recognized as a key proteomics instrument, since it has been used in numerous studies to define the functions and demonstrate the interactions of surface and intracellular proteins. Although titles shouldn’t really matter in science, it is appropriate to grant the FACS these distinctions. Similarly, it is appropriate to congratulate Bernie Shoor and BD for having had the foresight to build the first biotech company and to lay the foundations for it to grow to its current status as Becton-Dickinson Biosciences (BDB).

11 Feb 2004

18

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

Len, of course, has been honored many times for his innovative role in developing the FACS and demonstrating its applications in biology and medicine. Notably, he was cited for this work when elected to the National Academy of Sciences in 1982.

Breaking the FACS Color Barrier At the beginning of the 1980s, we realized that the immunology and other studies that we wanted to do were limited by the number of individual fluorescence measurements (sometimes called parameters) that a single-laser instrument could make on individual cells. There were enough markers known on T cells, for example, to suggest that multiple subsets existed. However, we recognized that using these markers effectively requires their simultaneous measurement on individual cells. Measuring their expression two by two, or even three by three, is not adequate. Information is lost when the measurements are separated and cannot be regained by trying to merge them during analysis. David Parks, then a member and later the leader of our FACS Development Group, solved this problem by extending the FACS-2 to create a dual-laser FACS instrument that would, at a minimum, double the number of markers we could measure on a given cell. In addition, he independently developed single-cell cloning and added this capability to the dual laser instrument (20–22). This was not the first dual-laser FACS (one had already been created by an instrumentation research group in Germany). However, it was the first dual-laser instrument put into routine use for immunologic studies and hence was the first instrument to demonstrate the effectiveness of using multiparameter FACS methods for distinguishing lymphocyte subsets and for sorting these subsets for functional studies. Many of the key findings made over the years by our group and by other research groups at Stanford were enabled by the development of this dual-laser instrument and its installation in the Stanford Shared FACS Facility, which Len helped to organize some years ago and which David Parks now directs. We put the dual-laser instrument into routine service in 1983. Roughly 15 years later (1998), we put into operation a hybrid instrument (BD bench, Cytomation electronics) that provides three independent laser illuminations and can simultaneously measure up to 11 distinct fluorescence emissions from individual cells. The number of markers measured with this high-definition (Hi-D) FACS (23–25) instrument, and with our recently purchased BDB Hi-D instruments (FACS DIVA and ARIA), has grown from an initial 8 to the current 11, now the standard for most work in our laboratory. Mario Roederer and his group at the NIH have extended FACS DIVA and located additional fluorescent dyes that can be measured simultaneously to further increase the number of measurements that can be made per cell.

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

LEN:

The Soft Side of FACS LEE: The original FACS data were collected by photographing histograms traced on an oscilloscope screen. Although these were the early days for using computers

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TWO LIVES INTERTWINED

P1: IKH

19

to collect data from laboratory instruments, we once again were able to capitalize on our connection with Josh Lederberg’s exobiology engineering group and, with their help, began using the Digital PDP-8 computer to collect FACS data. Wayne Moore, who joined the FACS development group shortly thereafter and has since built or supervised the building of all of our FACS software, moved the FACS data collection and analysis to the PDP-11 platform. On this platform, he developed models for much of the data analysis and display methods that are still in use, including the equal-density (probability) contouring method that is today’s standard. FACS/Desk, which Moore introduced at about the same time that our duallaser FACS was put into operation (26), was built on a VAX-11/780 platform and offered a nonprocedural (keystroke rather than command line) user interface. This interface, which had windows that opened and asked for user input, foreshadowed what I was later to see in the Apple Macintosh windowing environment. Sometime around 1980, Len had to raise nearly half a million dollars to buy the VAX computer and build the specialized computer facility necessary to house it, but we have always considered this well worth the trouble. The capabilities that Moore’s full FACS/Desk system provided, and still provides, have enabled countless large multiparameter experiments and have provided a permanent, searchable record of all FACS experiments done in the Stanford Shared FACS Facility. For the past several years, we have been working on a replacement for FACS/ Desk. Some time ago, we did the initial designs for a new FACS analysis package. These provided the basis for Mario Roederer’s extensions and ultimately for the commercially developed FlowJo package (TreeStar.com), which is widely used today. At present, we are migrating data stored in FACS/Desk to our new FACS DataStore, whose capabilities are much improved over the older system, and have completed an initial version of a searchable Directory Server that can be closely integrated with the new DataStore. We, in collaboration with Mark Musen, Medical Information Sciences, Stanford, and Stephen Meehan, Meehan Metaspace, are also about to complete the first version of a FACS protocol editor (FacsXpert) that provides an advanced user interface coupled to knowledge-based technology to facilitate design of 12-color FACS staining protocols. FacsXpert is also designed to “painlessly” capture the information (metadata) necessary to annotate data for analysis output (e.g., for axes and table heads) and to facilitate searches with the Directory Server. Ultimately, we hope to make all these capabilities available to the scientific community and to extend the system to take and store data from multiple instruments. This goal, however, may have to wait until we can find a willing and appropriate commercial partner. Len has been both contributive and supportive in this software development effort. However, in some ways it has been very much “my baby.” Although I have written only a small part of the overall system (specifically, the FACS Facility instrument scheduler), I have frequently participated in the design process and have opened relevant collaborations with colleagues in the Stanford Computer Science

11 Feb 2004

20

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

and Statistics Departments. In addition, I pioneered the connection of FACS/Desk and FACS analysis output to the SAS Institute JMP statistics package to enable analysis of data from our HIV clinical trials. For this effort, and for the development of the overall FACS/Desk system, we were awarded the Computer World Smithsonian Award in recognition of our visionary use of information technology in the field of medicine.

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

A Note about Innovation Although all the key flow cytometry technologies that we developed were eventually adopted as standards by the commercial and academic flow-cytometry community, we have routinely encountered substantial resistance to the initial spread of these technologies. Biological innovations such as the use of monoclonal antibodies as FACS reagents were readily and rapidly accepted (27). However, technological innovations not part of the biological idiom fared less well, particularly when statistical or mathematical treatments were involved. Hopefully, this will change as these modes of data analysis become more common within the biomedical research community.

THE FACS IN THE SERVICE OF IMMUNOLOGY (AND VICE VERSA) This and the following sections summarize the work we have done over the past 50 years, organized longitudinally by subject area. Presenting the work this way provides a clear view of how our interests in various areas have played out, but it tends to obscure the ways in which the work in each area influenced the development of work in the others. This interplay, generated by the simultaneous pursuit of diverse studies within Lee’s and my jointly run group, is one of the key elements in what might be termed the Herzenberg laboratory experience. Interactions with our contemporaries at other institutions also played a key role in shaping our work. It’s hard to convey the fun we all had working individually, but nonetheless as a group, to unfold the immune system and pry its secrets loose. Immunology, particularly the study of cells and cell functions in the immune system, was a small discipline at the time. By and large, those of us working in this area communicated with each other frequently. We made a point of sharing reagents, knowledge, students, and fellows in a way that has become more difficult as we have gotten older and the field has gotten larger. But although our specific interests have diverged and the work of our close circle of collaborators has overlapped less, we remain friends and still love an evening of “talking science” over a good bottle of wine. Av Mitchison, Gus Nossal, and Olli Makela, as we have indicated, were part of this early immunology group, as were Tomio Tada, Klaus Rajewsky, Richard Gershon, Bill Paul, Max Cooper, Charlie Janeway, Eli Sercarz, Ray Owen, Hugh

LEN:

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

TWO LIVES INTERTWINED

P1: IKH

21

McDevitt, Ben Pernis, and Spedding Micklem. Elizabeth Simpson, Irving Weissman, Robert Mishell, Patricia Jones, Harvey Cantor, Ko Okumura, Lee Hood, and Fred Alt formed a younger contingent with whom we also interacted frequently. Of course there were many others, including the students and fellows in our laboratory (whom we have named by reference). But when we think of “the old days,” the people we have named here immediately come to mind as the friends and colleagues who contributed most to our development as immunologists and scientists.

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IgH Allelic and Haplotype Exclusion Our first serious application of FACS in immunology was done using the breadboard machine to characterize, sort, and transfer rabbit B cells according to Ig allotype expression. These early studies confirmed that the Ig-bearing cells in spleen and lymph nodes are precursors of antibody-producing cells, a conclusion that had been based, until this point, on sensible logic and correlations between enrichment of Ig-bearing cells and increases in functional activity in adoptive transfer studies (28). In addition, these studies provided the first indication that the allelic exclusion visible in IgG- and IgA-producing plasma cells in allotype heterozygotes has already occurred in the B cells that give rise to these plasma cells (29, 30). In subsequent studies conducted with FACS-1 and FACS-2, we used murine IgH allotype markers on IgM, IgD, IgG, and IgA to further characterize the Ig isotype and allotype commitment of naive B cells and their memory (IgG+)-B-cell progeny. These studies, which showed that a B cell and its progeny are committed to producing Ig heavy chains encoded by only one of the two parental IgH chromosomes (haplotypes), led us to propose that allotype exclusion should more appropriately be called haplotype exclusion (6, 7). Thus, this work laid the groundwork for current understanding of IgH rearrangement and isotype-switching mechanisms. In additional B cell studies, we showed that the IgG isotype expressed on memory cells indicates the commitment of its Ig-producing cell progeny (31) and that specificity of the surface Ig on memory cells indicates the specificity of the adoptive response they will produce (32, 33). This latter work, done by sorting cells that bound keyhole limpet hemocyanin (KLH), also provided direct evidence for the one-cell/one-antibody concept, which was still an issue at the time.

B Cells Subdivided LEE: Randy Hardy and Kyoko Hayakawa arrived in our laboratory just about the time that David Parks brought the dual-laser (multiparameter) FACS to a location where it could be used regularly for immunology studies. Several T-cell subsets were already known at this point, and others seemed imminent. B cells, in contrast, were thought to be largely homogenous except for the kinds and amounts of Ig molecules they expressed. Although we were principally interested in doing functional studies with T cells, we decided to exercise the new instrument on something

11 Feb 2004

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

22

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

relatively simple by focusing it on an investigation of B-cell heterogeneity in the spleen. We figured that three months should wrap up this B-cell work, and then we could get on to T cells. Wrong guess! We immediately uncovered several B-cell subsets (34–36) and found a third (B-1) shortly thereafter (37). In fact, two decades later, we and a host of other laboratories are still working on a definition of the B-cell subsets within the mature B-cell population and of the developmental subdivisions that occur as unrearranged progenitors differentiate to mature B cells in bone marrow. In our most recent work, we have used Hi-D FACS methods to examine these developmental subsets, as have Hardy and Hayakawa, whose pioneering work in their own laboratories some time ago defined the basic stages (so-called Fractions A-F) of B-cell development and identified many of the key genes expressed at these stages (38). Our studies of B-1 cells, which express low levels of CD5 and have quite different immune response properties, have spawned a great deal of controversy over whether B-1 and B-2 cells derive from the same or different progenitors (39, 40). However, the key evidence still supports placing these cells in separate developmental lineages. In essence, we and others have repeatedly confirmed the results of our original cotransfer studies, which show that progenitors from adult bone marrow give rise to few, if any, B-1 cells in adoptive recipients in which cotransferred progenitors from fetal liver fully repopulate the B-1 compartment (41–45). We agree that CD5 expression can be induced on bone marrow–derived (B-2) cells under some conditions and that selection impacts B-1 and B-2 development differently. However, these findings are consistent with either a one-lineage or a two-lineage model. In fact, all the data advanced so far in favor of the onelineage model are also explainable within a two-lineage model. Therefore, until a compelling reason emerges to ignore the cotransfer results, which clearly favor separate lineages for B-1 and B-2, we will stay in the two-lineage camp.

T Cells Subdivided and a Bit More on B Cells The idea that Ig-bearing (B) cells produce antibodies and T cells help them to do so emerged as a paradigm just about the time that FACS-1 was delivered to our laboratory. FACS sorting, transfer, and cell-culture studies with the new machine completed the proof needed for this basic concept and, by introducing adoptive cotransfer methods into the lab, set the stage for much of our memory-Bcell and allotype suppression studies over the next decade. Shortly before we left for sabbatical at Cesar Milstein’s laboratory, we completed a collaborative study with Harvey Cantor and Ted Boyse that provided the first evidence distinguishing the helper and suppressor/cytotoxic T-cell subsets from each other on the basis of reactivity with conventional antibodies to Lyt-1 and Lyt-2 (46). When we returned from sabbatical, the first monoclonal antibodies that we produced to mouse cell-surface antigens contained several that stained T cells (18,

LEN:

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TWO LIVES INTERTWINED

P1: IKH

23

47–50). Studies with these antibodies brought our focus back to the T-cell subsets (51, 52), but (amusingly, from a current perspective) in a way that also wound up creating a mini tempest in a pint-sized teapot. One of the antibodies we produced reacted with a T-cell surface molecule whose physical properties and distribution on T cells corresponded to Lyt-2 (now known as CD8) and a second antibody reacted with a molecule (now known as CD5) whose physical properties were identical to those reported for Lyt-1. However, the expression of this molecule on T cells, as detected by FACS, did not match the expected pattern. Rather than being restricted to a helper-T-cell subset, it was detectable on all T cells (18, 47). [In fact, as later studies showed, it is also expressed on developing thymocytes and is even present on a subset of B cells (37, 50).] The demonstration that the Lyt-1 (CD5) molecule is expressed at roughly the same level on all T cells caused some consternation. It clearly did not erode the functional distinction between the helper and suppressor/cytotoxic subsets; but, for the moment, it scuttled the idea that helper and suppressor/cytotoxic T cells each express a unique surface antigen. This was resolved (and the mini tempest dissipated) some time later by the isolation of a monoclonal antibody to human cell surface (CD4) and the isolation, by Frank Fitch, of a monoclonal antibody (L3T4) that reacts with the corresponding murine molecule (53). LEE: From T-cell subsets in mice, we extended our studies to human T cells. Bernie Shoor (still at BD) put us in contact with Robbie Evans at Rockefeller Institute, who was looking for someone to characterize the monoclonal antibodies he had made to human lymphocyte surface antigens. The work with these antibodies revealed the amazing homology, both at the molecular and the distributional level, of the markers defining human and mouse T-cell subsets (54). It also led Len to comment one day that, “really, the best study of man is man,” and it refocused much of our laboratory’s energy onto human lymphocyte studies.

T-Cell Subsets in Disease The loss of CD4 T cells in HIV disease is well known. However, the selectivity of this loss is overrated. Multiparameter FACS analyses of naive T cells in PBMC samples from HIV-infected people at various stages of disease shows that CD8 naive T cells are lost at the same rate as CD4 naive T cells (55). CD4 memory T cells are also lost, although not as quickly as CD4 and CD8 naive T cells. CD8 memory T cells, in contrast, increase in frequency as HIV disease progresses and are only lost at the end stages of the disease (55). Consistent with the idea that the coordinate loss of naive T cells in both subsets reflects the loss of thymic function during HIV progression, we have shown that both subsets of naive T cells are greatly decreased or missing entirely in recovered Hodgkin’s patients treated with radiation to the thymic region (56). In more recent studies on T-cell subsets, we have used 8 to 11 fluorescence colors in Hi-D FACS studies characterizing cytokine production, TCR antigen

11 Feb 2004

24

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

(“tetramer”) binding, glutathione levels, metabolic markers, and other properties of the naive- and memory-T-cell subsets as well as subsets within these subsets (57–59). The introduction of BD’s DIVA and ARIA instruments, both capable of 12-color FACS studies, has now enabled these kinds of studies in other laboratories. Mario Roederer, who pioneered much of the Hi-D FACS subset analysis while in our lab, is now doing forefront work in this area in his own laboratory at the NIH (60).

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Somatic Cell Genetics, Modern Style I have often pointed out that monoclonal antibodies and the FACS are complementary tools that synergize to enable studies for which neither alone is sufficient (27, 61). For somatic-cell genetics, the joint use of these tools allowed us (and others) to use the cell lines that produce monoclonal antibodies to investigate isotype expression (isotype switch variants) (62) and antigen-combining site variants (affinity variants) (20). In addition, of key importance for gene cloning and expression studies, these complementary tools provide the ability to select cells expressing particular molecules or variants thereof. Thus, among other things, they allowed us to clone the murine CD8 and CD5 genes (63–65) and to explore the gene amplification (double-minute/minichromosome) mechanism (66) that controlled the level of expression of the CD8 gene cloned into a nonlymphoid cell line.

LEN:

Fetal Cells in Maternal Circulation Many years ago, when the FACS was still a “baby,” we decided to use it to see if we could detect fetal cells in human maternal circulation as the first step toward developing a noninvasive method for prenatal diagnosis. We were indeed able to detect the fetal cells (67). However, we didn’t follow the project much farther, in part because there was reason to suspect that the cells we detected might not have derived from the current fetus. Indeed, current studies show that fetal cells persist for long times in the mother. In addition, these studies surprisingly now indicate that the presence of these cells is closely associated with the development of maternal autoimmune disease and that maternal cells are found in small numbers in multiple tissues of the newborn (68, 69).

HIV, NF–κb, and Redox LEE: Living in the San Francisco area in the early days of the AIDS epidemic was like living in a war zone, where the dead were counted weekly, and the casualty lists frequently included friends or friends of friends. In this climate, it was impossible to go to work in the laboratory without wondering whether something you were doing could be of help. Therefore, we took an interest in a letter sent by Wolf Droge (Heidelberg, Germany) reporting that glutathione (GSH), a multifunctional cysteine-containing tripeptide, is depleted in HIV disease and suggesting, on the basis of his in vitro work charting nutritional requirements for T-cell function, that

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TWO LIVES INTERTWINED

P1: IKH

25

the GSH depletion results in oxidative stress that may decrease the function of the dwindling number of T cells in HIV patients (70). He also showed that GSH levels were low at the later stages of HIV disease (71). Droge recommended treating HIV-infected people with N-acetylcysteine (NAC), a well-known nontoxic cysteine prodrug used clinically to prevent hepatotoxicity by providing the cysteine necessary to replenish GSH depleted by acetaminophen overdose. In addition, he reported that his neighbor’s HIV-infected son was greatly helped by oral NAC administration. This anecdotal finding was not highly convincing. However, we respected Droge’s logic and began seriously exploring the idea of NAC treatment for HIV disease. In the laboratory, we began what appeared to be an independent HIV project in which we made a construct that put the β-galactosidase (LacZ) gene under the control of the HIV-LTR and developed FACS methods for measuring the activity of this reporter gene. However, two projects merged when, after hearing Tony Fauci speak about TNF-α triggering of HIV replication, we arranged to test whether NAC would inhibit triggering of the HIV-LTR by TNF (72). The results of these studies provided the first evidence demonstrating that intracellular GSH levels regulate NF-κb induction and HIV-LTR activity in cell lines (73, 74). Importantly for T-cell function, we also showed that intracellular GSH regulates TCR-stimulated calcium flux in cell lines and in primary human T cells in vitro. The Heidelberg group extended this finding by showing, in their placebocontrolled studies with HIV patients, that NAC treatment improves the in vitro function of T cells taken from these patients (75). In the placebo-controlled trial that we conducted sometime before, we demonstrated that NAC treatment replenishes GSH in HIV-infected people (76). We did not detect any impact of the NAC treatment on HIV viral load. However, we found a marked correlation between NAC treatment and improved survival in the open-label portion of this trial, which was conducted prior to introduction of the more effective, current antiviral therapies (76, 77). Collectively, these findings are consistent with NAC being a useful adjunct therapy in HIV disease, and they support Droge’s initial suggestion that GSH replenishment may be most important for maintaining the overall health and defense against opportunistic infection in the HIV patient. As befits a primarily basic science laboratory, the redox studies we were doing were initially relevant to HIV disease but soon engendered a broader interest in the mechanisms through which intracellular GSH levels influence signal transduction and other physiologic properties of cells. To approach these questions more effectively, we established FACS assays for intracellular GSH levels and, more recently, for surface thiols (57), intracellular thioredoxin, and intracellular protein glutathionylation in primary lymphocytes (78). In addition, we began examining other consequences of GSH depletion (79, 80). This burgeoning project continues unabated in our laboratory and now involves studies with samples from metabolic disease, cystic fibrosis, CLL, HIV, and other types of patients. At the time we were working with HIV-infected subjects, we found that thioredoxin (Trx), a key intracellular redox molecule, is released into circulation and

11 Feb 2004

26

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

blocks neutrophil chemotaxis when present at high levels in blood. Consistent with this finding, we found that among HIV-infected subjects with T cell counts below 200/µl blood, those with elevated Trx levels died within 15 months, even though they had no signs of ill health at the time of assay. This led us to propose that Trx interference with innate responses may render immune-compromised HIV-infected patients more susceptible to progress to the kinds of opportunistic infections that led to death (81). Interestingly, in vitro studies indicate that NAC treatment decreases thioredoxin release (79).

Glutathione/Cysteine-Deficiency Disease Together with a broad series of collaborators with expertise in various aspects of GSH deficiency, we recently completed a literature survey of more than 50 placebocontrolled trials in which beneficial effects were demonstrated for NAC treatment in diseases ranging from chronic bronchitis to diabetes. Collectively, these studies indicate that GSH/cysteine deficiency commonly accompanies a wide variety of clinically important diseases and conditions and that treatment of this deficiency may significantly improve health. Thus, we hope this work will call physicians’ attention to this problem and the ways that it can be minimized.

FAMILY, FRIENDS, AND POLITICS LEN AND LEE: We have not spoken very much here about our family, although all who know us know that the dividing line between work and family is virtually indistinguishable in our lives. Our students, fellows, and laboratory personnel have in many different ways contributed to the upbringing of our children. Berri, our eldest, now runs the Bicycle Trip (http://www.bicycletrip.com), a bicycle shop in Santa Cruz, California. Janet, or Jana as she is now called, is a singer/songwriter who has just released a CD through Motema Music (http://www.motema.com), a small but up-and-coming record label for which she is the CEO. Rick has become a serious salesman and and a fine potter; and Michael, our Down’s Syndrome son, is living nearby and working one afternoon at the lab and the rest of the time at a local workshop. We are pleased that many of our friends and previous coworkers have established independent relationships with our children. We are also pleased that the entire family still puts in political time trying to make the world a better place to live in.

ACKNOWLEDGMENTS We owe much to Jeannette Colyvas, whose oral-history interviews of each of us independently and of both of us together helped us focus on and organize the material presented here. We also thank John Mantovani, our administrator, editorial assistant, friend, and friendly goad, without whom it is doubtful that we

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

TWO LIVES INTERTWINED

P1: IKH

27

could have completed this project. This work was supported in part by grants from the National Institutes of Health: grant numbers EB000231 and HL68522. The Annual Review of Immunology is online at http://immunol.annualreviews.org

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

LITERATURE CITED 1. Ames BN, Garry B, Herzenberg LA. 1960. The genetic control of the enzymes of histidine biosynthesis in Salmonella typhimurium. J. Gen. Microbiol. 22:369–77 2. Herzenberg LA, Gonzales B. 1962. Appearance of H-2 agglutinins in outcrossed female mice. Proc. Natl. Acad. Sci. USA 48:570–73 3. Herzenberg LA, Herzenberg LA. 1961. Association of H-2 antigens with the cell membrane fraction of mouse liver. Proc. Natl. Acad. Sci. USA 47:762–67 4. Burnet FM. 1959. The Clonal Selection Theory of Acquired Immunity. Cambridge, UK: Cambridge Univ. Press 5. Wunderlich J, Herzenberg LA. 1963. Genetics of a gamma globulin isoantigen (allotype) in the mouse. Proc. Natl. Acad. Sci. USA. 49:592–98 6. Herzenberg LA, Minna JD, Herzenberg LA. 1967. A chromosome region for immunoglobulin heavy chains in the mouse: allelic electrophoretic mobility differences and allotype suppression. Cold Spring Harb. Symp. Quant. Biol. 32:181–86 7. Herzenberg LA. 1964. A chromosome region for gamma2a and beta2A globulin H chain isoantigens in the mouse. Cold Spring Harb. Symp. Quant. Biol. 29:455462 8. Herzenberg LA, Okumura K, Metzler CM. 1975. Regulation of immunoglobulin and antibody production by allotype suppressor T cells in mice. Transplant. Rev. 27:57–83 9. Herzenberg LA, Tokuhisa T, Hayakawa K. 1983. Epitope-specific regulation. Annu. Rev. Immunol. 1:609–32 10. Herzenberg LA, Black SJ. 1980. Regulatory circuits and antibody responses. Eur. J. Immunol. 10:1–11

11. Van Dilla MA, Fulwyler MJ, Boone IU. 1967. Volume distribution and separation of normal human leucocytes. Proc. Soc. Exp. Biol. Med. 125:367–70 12. Hulett HR, Bonner WA, Barrett J, Herzenberg LA. 1969. Cell sorting: automated separation of mammalian cells as a function of intracellular fluorescence. Science 166:747–49 13. Bonner WA, Hulett HR, Sweet RG, Herzenberg LA. 1972. Fluorescence activated cell sorting. Rev. Sci. Instrum. 43: 404–9 14. Herzenberg LA, Sweet RG. 1976. Fluorescence-activated cell sorting. Sci. Am. 234: 108–17 15. Herzenberg LA. 1973. Curing cancer by federal fiat. Hosp. Pract. 8:16 16. Oi VT, Jones PP, Goding JW, Herzenberg LA. 1978. Properties of monoclonal antibodies to mouse Ig allotypes, H-2, and Ia antigens. Curr. Top. Microbiol. Immunol. 81:115–20 17. Milstein C, Herzenberg LA. 1977. T and B Cell Hybrids. Presented at Regul. Gen. Immune Syst.: ICN-UCLA Symp. Mol. Cell. Biol., Los Angeles 18. Ledbetter JA, Herzenberg LA. 1979. Xenogeneic monoclonal antibodies to mouse lymphoid differentiation antigens. Immunol. Rev. 47:63–90 19. Morrison SL, Johnson MJ, Herzenberg LA, Oi VT. 1984. Chimeric human antibody molecules: mouse antigen-binding domains with human constant region domains. Proc. Natl. Acad. Sci. USA 81: 6851–55 20. Parks DR, Bryan VM, Oi VT, Herzenberg LA. 1979. Antigen-specific identification and cloning of hybridomas with

11 Feb 2004

28

21.

22.

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

23.

24.

25.

26.

27.

28.

29.

30.

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

a fluorescence-activated cell sorter. Proc. Natl. Acad. Sci. USA 76:1962–66 Parks DR, Hardy Richard A, Herzenberg LA. 1983. Dual immunofluorescence— new frontiers in cell analysis and sorting. Immunol. Today 4:145–50 Parks DR, Hardy RR, Herzenberg LA. 1984. Three-color immunofluorescence analysis of mouse B-lymphocyte subpopulations. Cytometry 5:159–68 De Rosa SC, Herzenberg LA, Herzenberg LA, Roederer M. 2001. 11 color, 13 parameter flow cytometry: identification of naive T cells by phenotype, function and T cell receptor. Nat. Med. 7:245–48 Baumgarth N, Roederer M. 2000. A practical approach to multicolor flow cytometry for immunophenotyping. J. Immunol. Methods 243:77–97 Roederer M, Herzenberg LA. 1999. Flow cytometry. In Encyclopedia of Molecular Biology, ed. TE Creighton. pp. 1–4. New York: Wiley Moore W, Kautz R. 1986. Data analysis for flow cytometry. In The Handbook of Experimental Immunology, ed. LA Herzenberg, DM Weir, CC Blackwell, LA Herzenberg. Edinburgh: Blackwell Scientific, pp. 30–36 Herzenberg LA, Ledbetter JA. 1979. Monoclonal antibodies and the fluorescenceactivated cell sorter: complementary tools in lymphoid cell biology. In Molecular Basis of Immune Cell Function, pp. 315– 30. Amsterdam: Elsevier/North-Holland Biomedical Jones PP, Cebra JJ, Herzenberg LA. 1973. Immunoglobulin (Ig) allotype markers on rabbit lymphocytes: separation of cells bearing different allotypes and demonstration of the binding of Ig to lymphoid cell membranes. J. Immunol. 111:1334–48 Jones PP, Tacier-Eugster H, Herzenberg LA. 1974. Lymphocyte commitment to Ig allotype and class. Ann. Immunol. (Paris) 125C:271–76 Jones PP, Craig SW, Cebra JJ, Herzenberg LA. 1974. Restriction of gene expression in B lymphocytes and their progeny. II. Com-

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

mitment to immunoglobulin heavy chain isotype. J. Exp. Med. 140:452–69 Okumura K, Julius MH, Tsu T, Herzenberg LA. 1976. Demonstration that IgG memory is carried by IgG-bearing cells. Eur. J. Immunol. 6:467–72 Julius MH, Masuda T, Herzenberg LA. 1972. Demonstration that antigen-binding cells are precursors of antibody-producing cells after purification with a fluorescenceactivated cell sorter. Proc. Natl. Acad. Sci. USA 69:1934–38 Julius MH, Janeway CA Jr., Herzenberg LA. 1976. Isolation of antigen-binding cells from unprimed mice. II. Evidence for monospecificity of antigen-binding cells. Eur. J. Immunol. 6:288–92 Hardy RR, Hayakawa K, Haaijman J, Herzenberg LA. 1982. B-cell subpopulations identified by two-colour fluorescence analysis. Nature 297:589–91 Hardy RR, Hayakawa K, Parks DR, Herzenberg LA. 1983. Demonstration of B-cell maturation in X-linked immunodeficient mice by simultaneous three-colour immunofluorescence. Nature 306:270–72 Hardy RR, Hayakawa K, Parks DR, Herzenberg LA. 1984. Murine B cell differentiation lineages. J. Exp. Med. 159:1169– 88 Hayakawa K, Hardy RR, Parks DR, Herzenberg LA. 1983. The “Ly-1 B” cell subpopulation in normal immunodefective, and autoimmune mice. J. Exp. Med. 157:202–18 Li YS, Hayakawa K, Hardy RR. 1993. The regulated expression of B lineage associated genes during B cell differentiation in bone marrow and fetal liver. J. Exp. Med. 178:951–60 Herzenberg LA, Kantor AB. 1993. B-cell lineages exist in the mouse. Immunol. Today 14:79–83; discussion 8–90 Wortis HH, Berland R. 2001. Cutting edge commentary: origins of B-1 cells. J. Immunol. 166:2163–66 Stall AM, Adams S, Herzenberg LA, Kantor AB. 1992. Characteristics and

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

TWO LIVES INTERTWINED

42.

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

43.

44.

45.

46.

47.

48.

49.

50.

51.

development of the murine B-1b (Ly-1 B sister) cell population. Ann. NY Acad. Sci. 651:33–43 Kantor AB, Herzenberg LA. 1993. Origin of murine B cell lineages. Annu. Rev. Immunol. 11:501–38 de Waard R, Dammers PM, Tung JW, Kantor AB, Wilshire JA, et al. 1998. Presence of germline and full-length IgA RNA transcripts among peritoneal B-1 cells. Dev. Immunol. 6:81–87 Herzenberg LA. 2000. B-1 cells: the lineage question revisisted. Immunol. Reviews 175:9–21 Baumgarth N, Herman OC, Jager GC, Brown LE, Herzenberg LA, Chen J. 2000. B-1 and B-2 cell-derived immunoglobulin M antibodies are nonredundant components of the protective response to influenza virus infection. J. Exp. Med. 192:271–80 Herzenberg LA, Okumura K, Cantor H, Sato VL, Shen FW, Boyse EA. 1976. T-cell regulation of antibody responses: demonstration of allotype-specific helper T cells and their specific removal by suppressor T cells. J. Exp. Med. 144:330–44 Ledbetter J, Goding JW, Tokuhisa T, Herzenberg LA. 1980. Murine T cell differentiation and antigens detected by monoclonal antibodies. In Monoclonal Antibodies, pp. 235–49. New York: Plenum Goding JW, Herzenberg LA. 1980. Biosynthesis of lymphocyte surface IgD in the mouse. J. Immunol. 124:2540–47 Howard FD, Ledbetter JA, Wong J, Bieber CP, Stinson EB, Herzenberg LA. 1981. A human T lymphocyte differentiation marker defined by monoclonal antibodies that block E-rosette formation. J. Immunol. 126:2117–22 Lanier LL, Warner NL, Ledbetter JA, Herzenberg LA. 1981. Quantitative immunofluorescent analysis of surface phenotypes of murine B cell lymphomas and plasmacytomas with monoclonal antibodies. J. Immunol. 127:1691–97 Waldor MK, Sriram S, Hardy R, Herzenberg LA, Lanier L, et al. 1985. Reversal

52.

53.

54.

55.

56.

57.

58.

59.

P1: IKH

29

of experimental allergic encephalomyelitis with monoclonal antibody to a T-cell subset marker. Science 227:415–17 Kipps TJ, Herzenberg LA. 1986. Hybridoma immunoglobulin isotype switch variant selection using the fluorescence activated cell sorter. In The Handbook of Experimental Immunology, ed. LA Herzenberg, DM Weir, CC Blackwell, LA Herzenberg. Edinburgh: Blackwell Scientific, pp. 109.1–109.8 Dialynas DP, Quan ZS, Wall KA, Pierres A, Quintans J, et al. 1983. Characterization of the murine T cell surface molecule, designated L3T4, identified by monoclonal antibody GK1.5: similarity of L3T4 to the human Leu-3/T4 molecule. J. Immunol. 131:2445–51 Ledbetter JA, Evans RL, Lipinski M, Cunningham-Rundles C, Good RA, Herzenberg LA. 1981. Evolutionary conservation of surface molecules that distinguish T lymphocyte helper/inducer and cytotoxic/suppressor subpopulations in mouse and man. J. Exp. Med. 153:310–23 Roederer M, Dubs JG, Anderson MT, Raju PA, Herzenberg LA. 1995. CD8 naive T cell counts decrease progressively in HIVinfected adults. J. Clin. Invest. 95:2061– 66 Watanabe N, De Rosa SC, Cmelak A, Hoppe R, Herzenberg LA, Roederer M. 1997. Long-term depletion of naive T cells in patients treated for Hodgkin’s disease. Blood 90:3662–72 Sahaf B, Heydari K, Herzenberg LA. 2003. Lymphocyte surface thiol levels. Proc. Natl. Acad. Sci. USA 100:4001–5 Perez OD, Nolan GP, Magda D, Miller RA, Herzenberg LA, Herzenberg LA. 2002. Motexafin gadolinium (Gd-Tex) selectively induces apoptosis in HIV-1 infected CD4+ T helper cells. Proc. Natl. Acad. Sci. USA 99:2270–74 Lu L-S, Tung J, Baumgarth N, Herman O, Gleimer M, et al. 2002. Identification of a germ-line pro-B cell subset that distinguishes the fetal/neonatal from the adult

11 Feb 2004

30

60.

61.

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

62.

63.

64.

65.

66.

67.

68.

16:35

AR

AR210-IY22-01.tex

HERZENBERG

¥

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

P1: IKH

HERZENBERG

B cell development pathway. Proc. Natl. Acad. Sci. USA 99:3007–12 De Rosa SC, Brenchley JM, Roederer M. 2003. Beyond six colors: a new era in flow cytometry. Nat. Med. 9:112–17 Herzenberg LA, De Rosa SC, Herzenberg LA. 2000. Monoclonal antibodies and the FACS: complementary tools for immunobiology and medicine. Immunol. Today 21: 383–90 Kipps TJ, Herzenberg LA. 1986. Homologous chromosome recombination generating immunoglobulin allotype and isotype switch variants. EMBO J. 5:263–68 Huang HJ, Jones NH, Strominger JL, Herzenberg LA. 1987. Molecular cloning of Ly-1, a membrane glycoprotein of mouse T lymphocytes and a subset of B cells: molecular homology to its human counterpart Leu-1/T1 (CD5). Proc. Natl. Acad. Sci. USA 84:204–8 Kavathas P, Sukhatme VP, Herzenberg LA, Parnes JR. 1984. Isolation of the gene encoding the human T-lymphocyte differentiation antigen Leu-2 (T8) by gene transfer and cDNA subtraction. Proc. Natl. Acad. Sci. USA 81:7688–92 Nakauchi H, Nolan GP, Hsu C, Huang HS, Kavathas P, Herzenberg LA. 1985. Molecular cloning of Lyt-2, a membrane glycoprotein marking a subset of mouse T lymphocytes: molecular homology to its human counterpart, Leu-2/T8, and to immunoglobulin variable regions. Proc. Natl. Acad. Sci. USA 82:5126–30 Kavathas P, Herzenberg LA. 1983. Amplification of a gene coding for human T-cell differentiation antigen. Nature 306:385– 87 Herzenberg LA, Bianchi DW, Schroder J, Cann HM, Iverson GM. 1979. Fetal cells in the blood of pregnant women: detection and enrichment by fluorescence-activated cell sorting. Proc. Natl. Acad. Sci. USA 76: 1453–55 Johnson KL, Nelson JL, Furst DE, McSweeney PA, Roberts DJ, et al. 2001. Fetal cell microchimerism in tissue from multi-

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

ple sites in women with systemic sclerosis. Arthritis Rheum. 44:1848–54 Srivatsa B, Srivatsa S, Johnson KL, Bianchi DW. 2003. Maternal cell microchimerism in newborn tissues. J. Pediatr. 142:31–35 Droge W, Eck HP, Mihm S. 1992. HIVinduced cysteine deficiency and T-cell dysfunction—a rationale for treatment with N-acetylcysteine. Immunol. Today 13:211– 14 Kinscherf R, Fischbach T, Mihm S, Roth S, Hohenhaus-Sievert E, et al. 1994. Effect of glutathione depletion and oral N-acetylcysteine treatment on CD4+ and CD8+ cells. FASEB J. 8:448–51 Roederer M, Raju PA, Staal FJ, Herzenberg LA, Herzenberg LA. 1991. Nacetylcysteine inhibits latent HIV expression in chronically infected cells. AIDS Res. Hum. Retroviruses 7:563–67 Montano MA, Kripke K, Norina CD, Achacoso P, Herzenberg LA, et al. 1996. NFkappa B homodimer binding within the HIV-1 initiator region and interactions with TFII-I. Proc. Natl. Acad. Sci. USA 93:12376–81 Staal FJ, Roederer M, Raju PA, Anderson MT, Ela SW, et al. 1993. Antioxidants inhibit stimulation of HIV transcription. AIDS Res. Hum. Retroviruses 9:299–306 Breitkreutz R, Pittack N, Nebe CT, Schuster D, Brust J, et al. 2000. Improvement of immune functions in HIV infection by sulfur supplementation: two randomized trials. J. Mol. Med. 78:55–62 De Rosa SC, Zaretsky MD, Dubs JG, Roederer M, Anderson M, et al. 2000. N-acetylcysteine (NAC) replenishes glutathione in HIV infection. Eur. J. Clin. Invest. 30:841–56 Herzenberg LA, De Rosa SC, Dubs JG, Roederer M, Anderson MT, et al. 1997. Glutathione deficiency is associated with impaired survival in HIV disease. Proc. Natl. Acad. Sci. USA 94:1967–72 Ghezzi P, Romines B, Fratelli M, Eberini I, Gianazza E, et al. 2002. Protein glutathionylation: coupling and uncoupling

11 Feb 2004

16:35

AR

AR210-IY22-01.tex

AR210-IY22-01.sgm

LaTeX2e(2002/01/18)

TWO LIVES INTERTWINED

Annu. Rev. Immunol. 2004.22:1-31. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

of glutathione to protein thiol groups in lymphocytes under oxidative stress and HIV infection. Mol. Immunol. 38:773– 80 79. Nakamura H, De Rosa S, Roederer M, Anderson MT, Dubs JG, et al. 1996. Elevation of plasma thioredoxin levels in HIV-infected individuals. Int. Immunol. 8:603–11 80. Nakamura H, Herzenberg LA, Bai J, Araya

P1: IKH

31

S, Kondo N, et al. 2001. Circulating thioredoxin suppresses lipopolysaccharideinduced neutrophil chemotaxis. Proc. Natl. Acad. Sci. USA 98:15143–48 81. Nakamura H, De Rosa SC, Roederer M, Yodoi J, Holmgren A, et al. 2001. Chronic elevation of plasma thioredoxin: inhibition of chemotaxis and curtailment of life expectancy in AIDS. Proc. Natl. Acad. Sci. USA 98(5):2688–93

19 Feb 2004

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104558

Annu. Rev. Immunol. 2004. 22:33–54 doi: 10.1146/annurev.immunol.22.012703.104558 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on September 15, 2003

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS Teunis B.H. Geijtenbeek,1 Sandra J. van Vliet,1 Anneke Engering,1 Bert A. ’t Hart,2,3 and Yvette van Kooyk1 1

Department of Molecular Cell Biology and Immunology, Vrije Universiteit Medical Center Amsterdam, 1081 BT Amsterdam, Netherlands; email: Y.vankooyk @vumc.nl 2 Department of Immunobiology, Biomedical Primate Research Center, 2280 GH Rijswijk, Netherlands 3 Department of Immunology, Erasmus Medical Centre, 3015 GE Rotterdam, Netherlands

Key Words DC-SIGN, carbohydrates, antigen recognition, pathogen, dendritic cells ■ Abstract Dendritic cells (DCs) are highly efficient antigen-presenting cells (APCs) that collect antigen in body tissues and transport them to draining lymph nodes. Antigenic peptides are loaded onto major histocompatibility complex (MHC) molecules for presentation to na¨ıve T cells, resulting in the induction of cellular and humoral immune responses. DCs take up antigen through phagocytosis, pinocytosis, and endocytosis via different groups of receptor families, such as Fc receptors for antigen-antibody complexes, C-type lectin receptors (CLRs) for glycoproteins, and pattern recognition receptors, such as Toll-like receptors (TLRs), for microbial antigens. Uptake of antigen by CLRs leads to presentation of antigens on MHC class I and II molecules. DCs are well equipped to distinguish between self- and nonselfantigens by the variable expression of cell-surface receptors such as CLRs and TLRs. In the steady state, DCs are not immunologically quiescent but use their antigenhandling capacities to maintain peripheral tolerance. DCs are continuously sampling and presenting self- and harmless environmental proteins to silence immune activation. Uptake of self-components in the intestine and airways are good examples of sites where continuous presentation of self- and foreign antigens occurs without immune activation. In contrast, efficient antigen-specific immune activation occurs upon encounter of DCs with nonself-pathogens. Recognition of pathogens by DCs triggers specific receptors such as TLRs that result in DC maturation and subsequently immune activation. Here we discuss the concept that cross talk between TLRs and CLRs, differentially expressed by subsets of DCs, accounts for the different pathways to peripheral tolerance, such as deletion and suppression, and immune activation.

0732-0582/04/0423-0033$14.00

33

19 Feb 2004

34

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL.

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DENDRITIC CELLS Dendritic cells (DCs) are professional antigen-presenting cells (APCs) that are seeded throughout peripheral tissues to act as sentinels that process and present antigen to mount adequate immune responses. Depending on the type of antigen, and the tissue localization, the immune response is suppressed or activated (1, 2). DCs differentiate from bone marrow stem cells and migrate as precursor DCs into the blood. Immature DCs populate all body tissues where they sample either self- or nonself-antigens. Self-antigens can be derived from their innocuous environment, such as necrotic and apoptotic cells that need to be scavenged before they disintegrate. Alternatively, nonself-antigens are foreign products from invading pathogens that need to be eliminated. The main function of DCs is to capture antigen for processing and presentation as antigenic fragments on major histocompatibility complex (MHC) class I or II molecules to na¨ıve T cells (1). In a steady-state situation, prior to acute infection and inflammation, DCs are in an immature state and are not fully differentiated to carry out their known roles as inducers of immunity. They are not unresponsive as they actively circulate through tissues and into lymphoid organs, capturing self-antigens as well as innocuous environmental proteins. In a state of alarm such as a microbial invasion, or massive cell death, immature DCs receive simultaneous activation signals through the binding of conserved molecular motifs by pattern recognition receptors, such as Toll-like receptors (TLRs) or specific TNF family members (1, 3). This results in DC maturation and migration to secondary lymphoid organs, where DCs present the processed antigens to na¨ıve T cells and induce antigen-specific immune responses. The maturation and migration of DCs is carefully orchestrated by certain chemokines, adhesion molecules, and co-stimulatory molecules. These factors control the differentiation stages of the DC and direct the migration of the various DC subtypes (4). Chemokines generated within the lymph nodes attract na¨ıve T cells toward the DC, enabling maximal exposure of MHC-peptide complexes to na¨ıve T cells. Adhesion molecules are crucial for the cellular interactions that DCs undergo during their journey from bone marrow through blood into peripheral organs and subsequently lymphoid tissues, where they enable DC–T cell interactions necessary for T cell activation. Recently, many novel cell-surface molecules have been identified that are involved in antigen capture by DCs. In particular, a large diversity of C-type lectin receptors (CLRs) has been identified on DCs that are involved in the recognition of a wide range of carbohydrate structures on antigens (5, 6). Especially the resident immature tissue DCs express a wide variety of C-type lectins that seem to be involved in the specific recognition of both self-antigens and pathogens. Although little is known about their specific carbohydrate recognition profile and antigen specificity, it is becoming evident that some of the CLRs, such as Dectin1, DC-SIGN, and the mannose receptors (MRs), are more than just scavenger receptors and they may regulate signaling events or function as cell adhesion receptors.

19 Feb 2004

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC

35

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

One of the major enigmas in immunology concerns those regulatory mechanisms that maintain the unresponsiveness of self-reactive T cells in the normal repertoire to self-antigens while at the same time allowing effective immune reactions to foreign antigens to be mounted. We propose here that both of these functions are coordinated by DCs, using distinct characteristics of the various receptor classes, such as CLRs and TLRs, present on their surface.

ANTIGEN RECOGNITION RECEPTORS ON DCs: TLRs AND CLRs Next to MHC, CD1, and FcR, the main DC receptors engaged in the direct recognition of characteristic molecular patterns on antigens are the TLRs and CLRs (7–9). TLRs are pathogen recognition receptors (10) that recognize characteristic molecular patterns present in microbial lipids, lipoprotein, lipopolysaccharides (LPS), nucleic acids, or bacterial DNA, as well as factors secreted upon tissue damage such as heat shock proteins (Hsp 70) (11, 12). The recognition by TLRs triggers intracellular signaling cascades that result in DC maturation and the induction of inflammatory cytokines, ultimately leading to T cell activation (7, 10). In contrast, CLRs recognize specific carbohydrate structures on self-antigens or cell wall components of pathogens. Their main function is to internalize antigens for degradation in lysosomal compartments to enhance antigen processing and presentation by DCs (5, 13). To date, several CLRs have been found to function as pathogen recognition receptors, whereas only a few have been shown to recognize self-antigens. Specific pathogens target CLRs to circumvent processing and presentation and to prevent immune activation. By targeting certain CLRs, such as DEC-205, tolerance can be induced in vivo, indicating that the sampling of self-antigens by specific CLRs may set the stage for tolerance induction and immune suppression (2). These findings illustrate a possible function for CLRs in the recognition of a wide variety of carbohydrate structures on self-glycoproteins to allow specific homeostatic control to self-antigens and to mediate cellular processes such as cell signaling, cell adhesion, and migration (5). In contrast, specific pathogens benefit from their capacity to target host C-type lectins, using the function of the C-type lectins to induce nonresponsiveness against their processed antigens with the aim to promote their survival by escaping immune activation. There are new indications that TLRs and CLRs communicate with each other, and it is proposed that the cross talk between TLRs and CLRs may fine-tune the balance between immune activation and tolerance (14–17). Recognition of self-antigen by CLRs alone will favor immune suppression, whereas pathogen recognition or self-recognition in a situation of danger where both TLRs and CLRs are triggered induces immune activation. Thus, in a steady-state situation, antigens are captured by CLRs to maintain tolerance, whereas in a dangerous situation, CLR binding to the same antigen occurs in the presence of TLR triggering, and the immunostimulatory function of TLRs overrules the tolerizing function of CLRs, resulting in immune activation (18).

19 Feb 2004

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

36

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL.

Depending on their tissue localization and differentiation state, DCs express unique sets of TLRs and CLRs. Many different CLRs are expressed by immature monocyte-derived DCs (5), such as the MR (CD206) (19), DEC-205 (CD205) (20), DC-SIGN (CD209) (21), BCDA-2 (22), Dectin-1 (23), DCIR (24), DCAL-1 (25), C-LEC (26), and DC-ASGPR (27)/MGL-1 (28). In contrast, only a few Ctype lectins have been identified on blood DCs and Langerhans cells. Langerhans cells express specifically Langerin (CD207) (29) and DEC-205, whereas plasmacytoid DCs express BDCA-2, Dectin-1, and DEC-205. Many C-type lectins are not exclusively expressed by DCs but are also expressed by other APCs such as macrophages. However, other CLRs are DC-specific, such as DEC-205 on murine DCs (30), Langerin on human Langerhans cells (29), and DC-SIGN expression that is confined to human DCs both in vitro and in vivo (31). DC-SIGN is abundantly expressed on both monocyte- and CD34+-derived DCs, as well as dermal DCs, but not by Langerhans cells in the epidermis (21). In vivo, DC-SIGN is expressed by immature DCs in peripheral tissues like skin, gut mucosa, cervix, rectum, placenta, and lung, as well as on DCs present in lymphoid tissues, lymph nodes, tonsils, and spleen (21, 32). Two DC-SIGN-positive DC precursor populations that differ in CD14 expression were found to be present in peripheral blood (33). Importantly, the expression of DC-SIGN is strongly dependent on Th2 cytokines such as IL-4, linking high expression of DC-SIGN on DCs to Th2 polarization (34). CLRs are highly expressed on immature DCs that are efficient in antigen capture and processing, whereas upon maturation CLR expression is readily decreased. Reflecting the large variety of CLRs on DC subsets, a large diversity of TLRs are also expressed by these DC subsets (35). Myeloid DCs express TLR2, 4, and 6, whereas plasmacytoid DCs express TLR7 and 9 (36, 37). That distinct DC subsets carry different sets of TLRs and CLRs raises the possibility that subsets of DCs recognize distinct classes of self- and nonself-antigens to induce tolerance or activate immunity. The expression pattern of CLRs and TLRs in vivo is as yet an unexplored field, but it needs to be developed as knowledge about expression of certain sets of antigen recognition receptors is essential for our understanding of how these DC subsets handle antigens and induce or suppress immunity.

FUNCTIONAL CHARACTERISTICS OF C-TYPE LECTINS Most CLRs on DCs are type II transmembrane proteins, with the exception of the MR and DEC-205, which are both type I transmembrane proteins (5). All C-type lectins contain carbohydrate-binding activity based on the presence of at least one carbohydrate recognition domain (CRD) (9). The CRD of the C-type lectin DC-SIGN is a globular protein that contains two Ca2+-binding sites, of which one directly coordinates the binding specificity of the carbohydrate structures (38). Calcium binding is essential for the function of CLRs because mutation of either Ca2+-binding site in DC-SIGN leads to the loss of ligand binding

19 Feb 2004

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC

37

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

(39). Depending on the amino acid sequence, the CRD bears specificity for either mannose, galactose, or fucose structures. However, binding of these carbohydrate structures to the different CLRs is also dependent on carbohydrate branching, spacing, and multivalency. The involvement of specific C-type lectins in ligand binding is often demonstrated by the ability of specific carbohydrate components, such as mannan, or a calcium chelator to block the interaction. However, coexpression of several CLRs—with the same carbohydrate specificity and calcium-dependency—on one DC subset makes it essential to use CLRspecific blocking antibodies to demonstrate specific CLR functions.

RECOGNITION OF CARBOHYDRATES BY C-TYPE LECTINS Several C-type lectins expressed by DCs have specificity for mannose-containing carbohydrates. However, each C-type lectin may recognize a unique branching and positioning of mannose residues on a given pathogen or self-antigen structure (19, 21, 27, 40–44). Also, multimerization of the receptor influences carbohydrate specificity (42). For example, the neck domain of DC-SIGN consists of 7 repetitive sequences that are thought to effect oligomer formation. The oligomerization of lectin domains alters the affinity and specificity of carbohydrate recognition (45). Strikingly, MR and MGL-1 form trimers, whereas DC-SIGN has been shown to form tetramers (42), which may partly explain the differences in carbohydrate specificity (38). To date, little is known about the carbohydrate and antigen specificity of CLRs expressed by DCs. The MR, DC-SIGN, and Langerin have been demonstrated to recognize mannose-containing carbohydrates but with a different specificity that is dictated by the branching of the mannose structures. Whereas the MR recognizes end-standing single mannose structures or di-mannose clusters, DC-SIGN recognizes, besides end-standing di-mannoses, also internal mannosebranched structures (high mannose) (21, 38, 41). The MR has also been shown to recognize fucose-, glucose- and GlcNAc—but not galactose-containing structures (46). Screening panels of synthetic glycoconjugates, containing mannose, galactose, and fucose residues, including their multimeric derivatives, can be used to delineate the specificity of C-type lectins (47–49). These experiments revealed that DC-SIGN recognizes a profile of carbohydrates distinct from what was initially thought. Next to recognizing mannose-containing structures, DC-SIGN has a higher specificity for fucose-containing carbohydrates, which is present in Lewis blood group antigens (Lex, Ley, Lea, Leb) that contain fucose residues in different anomeric linkages (49, 50). In contrast, the MR does not recognize Lex structures even though it has specificities for fucose (41). Sialylation of Lex (yielding sialyl-Lex), a L-, E-, and P-selectin ligand, completely abrogates recognition by DC-SIGN, indicating that DC-SIGN has a distinct carbohydrate specificity from the selectins that mediate leukocyte rolling (49). The CLR ASGPR/MGL-1 is rather unique in its carbohydrate-recognition profile, as it primarily recognizes

15 Mar 2004

38

15:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL.

galactose but not mannose or fucose residues (27, 51). Some CLRs have overlapping carbohydrate specificity and recognize a wide variety of carbohydrate structures, yet others have a rather unique recognition profile. Many of the recognized carbohydrate structures, such as Lex and mannose-containing carbohydrates, can be present on both self-glycoproteins and pathogens, supporting a role for C-type lectins in both self- and pathogen-recognition.

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CLRs IN THE REGULATION OF ANTISELF IMMUNE REACTIONS Many C-type lectins serve as antigen receptors (5). The main function of C-type lectins on DCs is to interact with a large variety of conserved molecular patterns present on self- and nonself-antigens (52, 53). Most C-type lectins contain putative internalization motifs, such as the di-leucine motif and the tri-acidic clusters (54), and their function as endocytic receptors has been demonstrated by their capacity to internalize specific antibodies (5). For some C-type lectins, internalization of CLR-specific murine IgG antibodies was shown to lead to antigen presentation to murine IgG-specific CD4+ T cells. Most C-type lectins containing a tri-acidic cluster (DEC-205, DC-SIGN, BDCA-2, Dectin-1, and CLEC-1) target internalized antigens to lysosomes and MHC class II+ late endosomes (Figure 1) (20, 55). In contrast, other C-type lectins, such as the MR, quickly recycle via early endosomes to ensure uptake of large amounts of antigen (Figure 1) (19). This process is mediated by a tyrosine-based motif in the cytoplasmic tail of the MR (56). Similar to DEC-205, DC-SIGN contains a tri-acidic cluster in its cytoplasmic tail, and accordingly DC-SIGN-ligand complexes are targeted to lysosomal compartments where ligands are processed for MHC class II presentation to T cells, indicating an important function for DC-SIGN as an antigen receptor (13). However, DC-SIGN also contains a di-leucine motif that appears to be essential for rapid internalization of soluble ligands by DC-SIGN (Figure 1) (13). Although most C-type lectins are endocytic receptors, it is unknown whether they can direct antigens to different intracellular compartments. To date, the exact mechanism of antigen presentation by C-type lectins in vivo is poorly understood. Antigens targeted to the CLRs DEC-205 and the MR are processed by DCs for presentation on both MHC class I and II, leading to a more specific and efficient antigen presentation (57–59). Other studies show that MR is engaged in the presentation of glycolipids via CD1. Uptake of M. tuberculosis lipoarabinomannan (LAM) by DCs through MR resulted in presentation by CD1b to specific T cells (60). C-type lectin receptors are not only involved in the recognition of pathogens but they might also contribute to the capture and presentation of glycosylated selfantigens. For example, the MR recognizes lysosomal hydrolases, certain collagenlike peptides in serum (61), and thyroglobulin, a well known self-antigen (43), and may have a role in autoimmunity (62).

15 Mar 2004

15:1

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC

39

The natural ligands of DEC-205 and its carbohydrate specificity are currently unknown. However, using DEC-205-specific antibodies as surrogate ligands, it has been demonstrated that this CLR mediates routing to MHC class II compartments and facilitates a 30–100 times more efficient presentation of captured antigens as compared to the MR (20). Similarly, targeting of DEC-205 using anti-DEC205 antibodies facilitates efficient antigen presentation by MHC class I molecules (63). In vivo targeting of ovalbumin conjugated with DEC-205-specific antibodies to DCs induces tolerance, whereas coinjection of agonistic anti-CD40 antibody reverses the outcome from tolerance to prolonged T cell activation and immunity (Figure 2A) (55, 63). These experiments elegantly demonstrate that under steady-state conditions in vivo, DCs have the capacity to induce peripheral T cell unresponsiveness, but a robust immune response is induced to the same antigen in the presence of danger signals (Figure 2A). In summary, in vivo antigen targeting to DCs under steady-state conditions seems to induce deletional tolerance when immunity is induced in conditions where DC maturation is triggered. Tolerance can be achieved with small amounts of protein and is manifested as profound unresponsiveness, indicating that the mechanisms dictating tolerance and immune activation are strictly regulated. More than 50% of the proteins in the human body are glycosylated, and it can be envisaged that in the steady state, DCs define immunological self on the basis of sugar structures, thereby suppressing the self-reactivity of the T cell repertoire. When in a danger state, DC maturation is induced by a microbial infection: the repertoire and the immune attack are selectively focused on elimination of the pathogen. CLRs in particular play an important role in the capture and uptake of self-antigens by binding to specific carbohydrate structures present on cells or secreted proteins that need to be scavenged. Also, the in vivo localization of CLR expression hints to the important function of CLRs in self-antigen clearance and their potential role in tolerance induction. For example, DC-SIGN is highly expressed by DCs in placenta at the interface of mother/child antigen transmission, a site of immune tolerance (32, 64). DC-SIGN is also highly expressed on ellipsoids in spleen at those sites where a direct contact between blood and tissue exists to enable antigen clearance from blood, without induction of an immune response. DC-SIGN is highly expressed in lymph nodes on DCs located in the T cell area as well as on immature DCs located at the site close to the efferent sinus (21, 32). Moreover, the structural related CLR of DCSIGN, L-SIGN/DC-SIGNR, is highly expressed on liver sinusoidal endothelial cells (LSECs), a resident APC population of monocyte origin (65, 66). LSECs are known to mediate the clearance of many potentially antigenic proteins from the circulation in a similar manner as DCs in lymphoid organs/peripheral tissues (67). The tissue localization and ligand-binding properties of L-SIGN support a physiological role for this CLR in antigen clearance. LSECs can induce tolerance in the liver (68), and L-SIGN may be involved in the antigen capture. In particular, apoptotic cells alter their glycosylation pattern (69) and are therefore candidates to be cleared by L-SIGN in the liver. Another CLR, Endo-180, which is expressed on endothelium, appears to be involved in the clearance of collagen, which under

19 Feb 2004

40

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL.

steady-state conditions is continuously synthesized and degraded during the normal turnover of connective tissue (70). This process involves specific binding of collagen fibrils to the CLR, followed by the cellular uptake and degradation of the internalized collagen in the lysosomal compartment (71).

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CLR AND THE RECOGNITION OF SELF-GLYCOPROTEINS TO MEDIATE CELL ADHESION DC-SIGN functions as a cell-adhesion receptor that regulates DC migration (72) and DC-T cell interaction (21) through its interaction with the self-glycoproteins ICAM-2 and ICAM-3, respectively. The natural ligand of DC-SIGN on lymphocytes, ICAM-3, contains N-linked glycosylations consisting of high-mannose-type oligosaccharides (73), and enzymatic removal of the N-linked carbohydrates from ICAM-3Fc completely abrogates its binding to DC-SIGN (39). These data indicate that DC-SIGN recognizes high-mannose-type carbohydrates on ICAM-3, but the specific carbohydrate structure on ICAM-3 has not yet been identified. Results suggest that DC-SIGN does not preferentially bind to all ICAM-3-expressing cells, but only interacts with specific leukocyte subsets. These findings hint to cell-specific glycosylation of ICAM-3 enabling only binding of subsets of cells (K. van Gisbergen, unpublished results). The other self-glycoprotein, ICAM-2, plays a central role in DC migration and homing to secondary lymphoid tissues. Although DC-SIGN binds to both ICAM2 and ICAM-3 under static conditions, only the DC-SIGN-ICAM-2 interaction resists the shear stresses encountered under physiological flow conditions (33). Remarkably, DC-SIGN behaves as a DC-specific rolling receptor for ICAM-2 and is thus functionally similar to the selectins, which are well known for their regulation of leukocyte rolling upon carbohydrate-structure recognition (74). Despite the fact that both DC-SIGN and selectins mediate leukocyte rolling, they have a distinct carbohydrate-recognition repertoire, i.e., Lex versus sialyl-Lex, illustrating that they may be involved in transendothelial migration at distinct sites. Recently, it has been reported that the MR also recognizes sialyl-Lex structures expressed by endothelial cells; thus this CLR may also contribute to the rolling interactions of DCs (75). The MR is also expressed by lymphatic endothelial cells, where it can interact with L-selectin to mediate lymphocyte binding (76). This illustrates that several CLRs, including DC-SIGN, selectins, and the MR, can regulate leukocyte rolling and migration, processes regulated by the expression of the appropriate carbohydrate structures on target cells. Whereas DC-SIGN recognizes high-mannose and Lex structures, both the MR and the selectins mediate rolling upon recognition of endothelial ligands that present sialyl- Lex structures. We predict that tissue-restricted glycosylation of ICAM-2 and expression of sialyl Lex and Lex structures may direct the migration of DC to particular sites. Cell- and tissue-specific glycosylation of a protein (post-translational modification) is driven by the tissue-specific expression of certain glycosyltransferases and glycosidases,

19 Feb 2004

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC

41

which add or remove specific carbohydrate residues, respectively. The expression of a large variety of these enzymes is tightly regulated during the differentiation and activation of leukocytes and by specific cytokines. Because specific carbohydrate structures dictate the specificity of the interaction with certain CLRs, the impact of altered glycosylation of a given glycoprotein can change the recognition by CLRs and subsequently alter cell-cell interactions (77). Indeed, tissue- and cellspecific homing and migration by selectins are regulated by the expression of a selected set of fucosyltransferases resulting in expression of specific carbohydrate structures (78). Understanding the carbohydrate specificity of DC-SIGN and other DC-expressed CLRs is a recent topic of interest in the field of glycobiology and immunology (49) (http://web.mit.edu/glycomics/consortium).

CLRs: A TARGET FOR PATHOGENS Some CLRs have been implicated in the recognition of pathogens, yet specific pathogens target CLRs to escape immune activation. This paradox raises the question of whether the main function of CLRs is to capture pathogens or to recognize self-antigen and suppress immunity. Both the MR and DC-SIGN have been demonstrated to capture pathogens, and in particular DC-SIGN is targeted by pathogens that seek immune escape. The identification of DC-SIGN as a DC-specific adhesion receptor (21) revealed its 100% identity to the previously cloned (HIV)-1 envelope-binding C-type lectin, and as such the first pathogen that interacts with DC-SIGN was discovered (79, 80). DC-SIGN does not only interact with M- and T-tropic HIV-1, HIV-2, and simian immunodeficiency virus (SIV) (61, 80–82), but also with other viruses such as Ebola virus, Cytomegalovirus, Hepatitis C virus, and Dengue virus (83–90). DC-SIGN recognizes the viral envelope glycoproteins that contain a relatively large number of N-linked carbohydrates. In particular, for HIV-1, Ebola, and Dengue virus it has been demonstrated that differential glycosylation of the envelope glycoproteins affects DC-SIGN binding and subsequently infection and viral transmission (91, 92). Other viruses that do express heavily glycosylated glycoproteins on their surface (e.g., VSV) fail to interact with DCSIGN (93), suggesting a certain degree of specificity in DC-SIGN recognition. Even though it is likely that high-mannose structures on gp120 are recognized by DC-SIGN (91, 92, 94), it is not ruled out that protein-protein interactions may also be involved in binding (39). Additionally, nonviral pathogens can interact with DC-SIGN. Helicobacter pylori and certain Klebsiella pneumonia strains interact with DC-SIGN through LPS that contain Lex and mannose structures, respectively (49). Mycobacterium tuberculosis interacts with DC-SIGN on DCs via the mannose cap of the cellwall component ManLAM (49, 95). Parasites such as Leishmania amastigotes and Schistosoma mansoni are recognized by DC-SIGN through the mannosecapped surface lipophosphoglycan (49, 96) and the Lex-positive soluble egg antigen (50), respectively, whereas the specific structure for fungus Candida albicans

19 Feb 2004

42

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL.

is not yet identified (97). In conclusion, a large panel of pathogens are primarily captured by DCs through binding to the DC-specific C-type lectin DC-SIGN.

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DC-SIGN: A RECEPTOR TO ALLOW VIRAL DISSEMINATION In contrast to CD4, DC-SIGN does not function as a classical HIV-1 entry receptor (80) but acts as an HIV-1 transreceptor that binds HIV-1 and transmits the virus very efficiently to neighboring permissive target cells (80). The presence of DC-SIGN+ DCs in mucosal tissues and DC-SIGN+ DC precursors in blood that efficiently transmit HIV-1 to T cells (98) makes DC-SIGN a candidate to be a key molecule in HIV-1 dissemination both after sexual transmission and through blood contamination (99, 100). DC-SIGN not only captures and transmits HIV-1, but it also enhances T cell infection because at low virus titers, CD4/CCR5-expressing cells are not detectable as being infected by HIV-1 without the assistance of DCSIGN in trans (80). Conditions in which the number of HIV-1 particles is limiting likely occur shortly after infection in vivo. This suggests that DC-SIGN may not only be required for HIV-1 transmission from mucosa to lymphoid tissues, but also for efficient infection of T cells. Neutralization of the pH within the HIV-1-containing compartments or prevention of internalization by deletion of the DC-SIGN cytoplasmic region abrogates DC-SIGN-mediated enhanced transinfection of T cells (93). This indicates that the internalization of the infectious HIV-1 particle is essential for transinfection of T cells. The clathrin-dependent sorting pathway likely mediates DC-SIGN endocytosis and recycling through recognition of the di-leucin motif. Clathrinindependent pathways may additionally be used during virus-induced DC-SIGN internalization. It is presently unclear whether intact HIV-1 virions escape targeting to lysosomes, as was described for other internalized DC-SIGN-ligands (13), and how these virions are protected against full processing. Whereas DC-SIGN-bound ligands are internalized for processing in lysosomal compartments, HIV-1 bound to DC-SIGN is remarkably stable and remains infective for prolonged periods (Figure 1) (80). Enzymatic digestion of cell surface–bound HIV-1 demonstrated that HIV-1 is protected and probably hides within cellular compartments close to the cell membrane without being degraded. An immunofluorescence study by Kwon et al. has demonstrated that HIV-1 is indeed internalized upon binding to DC-SIGN into nonlysosomal acidic organelles (93). In mature DCs, DC-SIGN is targeted to early endosomal compartments, in which HIV-1 would be protected against degradation (13), suggesting that maturation of DCs by HIV-1 may lead to its altered internalization. Finding a way to override this mechanism and to target internalized DC-SIGN-HIV complexes to lysosomes would facilitate HIV-1 processing in DCs, and it would enhance specific anti-HIV-1 immune responses while reducing infection of T cells (Figure 1) (13, 93).

19 Feb 2004

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC

43

At high concentrations, HIV-1 can infect DCs in cultures that coexpress CD4 and chemokine receptors (57, 101, 102). HIV-1 can replicate in immature as well as in mature DCs that interact with T cells (82, 103). In particular, the initial quantity of virus that enters the mucosal tissues may be decisive in whether DCs become infected by HIV-1 or whether the virus is captured for efficient transinfection of T cells (80). The Nef protein of HIV-1 appears to be crucial for DC–T cell binding (104). In addition, Nef can interact with the cell-sorting machinery to downregulate expression levels of CD4 and MHC class I and thus facilitate immune evasion (101). Expression of Nef in immature DCs results in a redistribution of DC-SIGN to the cell surface, thus reducing DC-SIGN internalization in favor of cell-surface expression and facilitating increased cell adhesion and virus transmission to T cells (82). Redistribution of DC-SIGN requires the di-leucine motif in the cytoplasmic tail of DC-SIGN, as well as a di-leucine motif in Nef, indicating that Nef interferes with the recognition of DC-SIGN by the sorting machinery (82). Recently, other CLRs, such as Langerin and MR, were shown to bind HIV gp120 through recognition of mannose-containing carbohydrates present on HIV gp120 (105–108). However, only the MR has been shown to capture and transmit HIV-1 to permissive T cells similar to DC-SIGN (109). Interestingly, HIV bound to the MR on macrophages has a lower half-life than free HIV-1, and no transmission occurred beyond 24 h after initial capture of the virus (109). The finding that the longevity of HIV-1 when captured by DC-SIGN exceeds five days indicates that the HIV-1 internalization routes are different for the MR and DC-SIGN. On monocyte-derived immature DCs, DC-SIGN is the major HIV-1 receptor despite coexpression of the MR (80). It will be interesting to find out whether different subsets of DCs expressing different arrays of CLRs handle HIV-1 differently, leading either to immune escape and HIV dissemination or to immune activation through processing and presentation of the virus.

CLR AND TLR CROSS TALK Nonviral pathogens also target CLRs to infect DCs (89–90, 96). Most striking is that several pathogens subvert DC-SIGN function to escape immune surveillance by a mechanism other than HIV-1 (110). Mycobacterium tuberculosis is a potent inducer of T helper 1 (Th1)-polarized immune response, and mycobacterial components have often been shown to stimulate expression of costimulatory molecules and IL-12 production in DCs through TLR2 and TLR4 triggering. An immunocompetent host controls infection with M. tuberculosis, yet complete eradication of the pathogen does not occur. When the immune response is impaired, active disease can develop, normally through reactivation of quiescent organisms or in some cases through reinfection. Although alveolar macrophages are the primary targets for infection by mycobacteria, DCs are important for activation of the cellular immune response (111, 112). M. tuberculosis and M. bovis bacillus Calmette-Gu´erin (BCG) strongly bind to DC-SIGN

19 Feb 2004

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

44

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL.

through their mannose-capped cell-wall component lipoarabinomannan (ManLAM) (15, 95). ManLAM is abundant in slow-growing mycobacteria comprising virulent species such as M. tuberculosis and M. leprea. AraLAM, which is abundant in fast-growing atypical avirulent mycobacteria such as M. smegmatis, M. fortuitum, and M. chelonae, does not interact with DC-SIGN, indicating that particular virulent strains have adapted to target DC-SIGN (15, 113). DC-SIGN is a major receptor of DC for mycobacteria (15). Although immature DCs also express high levels of the receptors MR, CD11b, and CD11c, which can mediate binding of mycobacteria by macrophages (114, 115), DC-SIGNspecific antibodies, in contrast to MR-specific antibodies, inhibit the interaction of DCs with both M. bovis BCG and ManLAM by more than 80% (15). Captured mycobacteria are targeted to the lysosomal compartment (15, 113), although it is not yet clear whether some mycobacteria can escape degradation in DCs. However, DCs do not support mycobacterial growth due to IL-10-induced reversion of DC maturation (116, 117). The cell-wall component ManLAM, which is considered to be a virulence factor, is also secreted in vivo by macrophages infected with M. tuberculosis (118, 119). This suggests that mycobacteria may specifically secrete ManLAM to interfere with the immune function of bystander DCs. Strikingly, secreted ManLAM from mycobacteria-infected macrophages targets CLRs to alter the immune response through cross talk between CLRs and TLRs (Figure 2B) (13, 15, 120). In particular, binding of the mycobacterial component ManLAM to immature DCs inhibited LPS-mediated IL-12 induction (120). This study suggested that ManLAM binding to the MR on immature DCs interferes with TLR4 signaling. Nigou et al. implicated MR as the main ManLAM-binding C-type lectin on DCs (120). However, recent work shows that ManLAM inhibits LPS-induced DC maturation by interacting with DC-SIGN because LPS-induced DC maturation in the presence of ManLAM is fully restored by inhibiting DC-SIGN-ManLAM interaction with specific antibodies (15). Also, viable M. bovis BCG induced DC maturation (15), probably through TLR2 and TLR4 signaling (121), and as observed with LPS, ManLAM specifically blocked the M. bovis BCG-induced DC maturation (15). The inhibition of DC maturation caused by ManLAM binding to DC-SIGN could be fully restored by antibodies against DC-SIGN (15). This illustrates that DC-SIGN, upon binding ManLAM, delivers a negative signal for M. bovis BCGinduced DC maturation, presumably induced via TLR4 (Figure 2B). Furthermore, ManLAM binding to DC-SIGN induced the production of the anti-inflammatory cytokine IL-10 by LPS-activated DCs (15). The inhibition of DC maturation and the induction of IL-10 may contribute to the virulence of mycobacteria; immature DCs and IL-10-treated DCs are not only less efficient in the stimulation of T cell responses but also induce a state of antigen-specific tolerance (Figure 2B) (122). Thus, pathogen recognition by DC-SIGN may modulate DC-induced immune responses, shifting the balance from immune activation toward impairment of immune responses, which would be beneficial to pathogen survival. How DCSIGN signals are propagated within DCs is not yet clear, but the presence of

19 Feb 2004

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC

45

immunoreceptor tyrosine-based activation motifs (ITAMs) in its cytoplasmic tail suggests that DC-SIGN is capable of direct signaling (54). Both DC-SIGN and Dectin-1 contain ITAMs. DCIR contains an immunoreceptor tyrosine-based inhibitory motif (ITIM) (54, 123, 124). The function of these so-called activating or inhibitory signaling motifs is not yet clear, but recent data demonstrate that upon pathogen recognition by CLRs, such as DC-SIGN and Dectin-1, intracellular signaling processes initiated by TLR activation are inhibited (15–17). The C-type lectin Dectin-1, recognized as a receptor for the yeast component zymosan, acts in concert with TLR2/4 to enhance the production of IL-12 and TNFα in DCs, facilitating a Th1 response (16, 17). These studies demonstrated that the cytoplasmic tail of Dectin-1, and in particular the ITAM motif, is involved in enhancing stimulatory capacity. Remarkably, DC-SIGN also contains an ITAM motif, yet when targeted by M. tuberculosis, the immune response is driven toward immunosuppression by the induction of IL-10 and the inhibition of DC maturation. These studies demonstrate that the collaborative recognition of distinct microbial components by different classes of innate immune receptors (CLRs and TLRs) is crucial for orchestrating inflammatory or inhibitory responses. The balance between TLR stimulation and C-type lectin occupation may fine-tune regulatory mechanisms to allow appropriate immune responses. The inflammatory and pathogenic consequence of this recognition is dependent on both the receptor repertoire and the functional cooperation between the signals generated downstream of receptor-ligand interaction. Yet we should not forget that in some pathological situations, the pathogens have evolved to direct the balance toward immune suppression, for instance by the secretion and production of a large quantity of soluble factors that target CLRs such as DC-SIGN (15).

CONCLUDING REMARKS AND FUTURE DIRECTIONS Within the past few years, various CLRs have been identified on DCs. The pattern of CLR expression depends on the DC subset. Whereas CLRs are abundantly expressed by immature DCs in peripheral tissues, the expression of most CLRs is rapidly downregulated upon DC maturation. Although CLRs initially were regarded as scavenger receptors, it is now clear that they bind a variety of antigens via specific recognition of particular carbohydrate profiles with interesting and important consequences. CLR-bound antigens are processed and presented to T cells, thus enhancing antigen presentation and immune activation. Strikingly, in vivo targeting of CLRs induces immune suppression, hinting to an important function of CLRs in the tolerance toward self-antigen. It has been well established that T cells reactive to self-antigen are part of the normal immune repertoire of mice, nonhuman primates, and humans at frequencies comparable to patients with autoimmune diseases. The fact that most of us do not develop autoimmune diseases shows that the autoreactive cells are kept under tight control. As the majority of self-antigens are glycosylated, we postulate that tolerance is maintained by the

19 Feb 2004

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

46

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL.

interaction with CLRs on DCs. However, when DCs obtain stimuli that result in maturation, such as upon pathogen recognition by TLR or CD40 triggering, this signaling overrules the tolerizing effects of CLRs. Certain pathogens that can stimulate DCs via TLRs, likely in the presence of CD40 engagement, can break the tolerant state. In an experimental setting, this is modeled when self-antigen is injected in an emulsion with strong bacterial antigens. Under these conditions the mechanisms that keep autoreactive T cells quiescent are broken, and autoimmune diseases develop. The human body uses regulatory T cells (Tregs) to control unwanted immune reactivity (125). Natural Tregs are localized in the thymus and inhibit T cell activation primarily via cell-cell contact, whereas inducible Tregs are distributed over the lymphoid organs and exert their suppressive activity via secretion of cytokines. It has yet to be proven whether DCs that have taken up autoantigen via their CLRs present processed antigen to inducible Tregs and in this way maintain systemic tolerance. It will be very important to understand if all CLRs process antigen in a similar fashion and whether they all contribute to self-tolerance or whether some CLRs may also process antigen to induce antigen-specific immune activation, especially since some CLRs contain ITIM or ITAM motifs. Intriguingly, specific pathogens target CLRs to escape immunosurveillance and to promote their survival by sequestration in DCs and by reducing their powerful antigen-presenting capacities. After HIV-1 was shown to bind DC-SIGN to hide within DCs, the same sophisticated escape mechanism was shown to be exploited by a steadily expanding repertoire of viruses that cause serious diseases in the human population, including CMV, Ebola virus, and Dengue virus. Also, nonviral pathogens may target DC-SIGN to escape immune surveillance. M. tuberculosis targets DC-SIGN to inhibit DC maturation and induce IL-10 production by a mechanism that interferes with TLR signaling. Our present knowledge of the variety of CLRs expressed on DCs, the exact mechanisms via which they exert their functions, and the positive or negative interaction with other receptor families (TLRs) is still very limited. It will be challenging to understand the function of C-type lectins in signaling and communication to other receptors on DCs, such as TLRs. A common feature of pathogens that interact with DC-SIGN is that they cause chronic infections that may persist lifelong, and secondly, that a disturbed Th1/Th2 balance by the pathogens is central to their persistence. Therefore, pathogens that target DC-SIGN may not only infect DCs but also shift the Th1/Th2 balance toward a response in favor of their persistence. The intriguing question arises of whether oligosaccharides on cellular counterstructures allow DC migration to specific sites as well as DC interactions with specific T cell subsets. Future research addressing carbohydrate-recognition profiles by C-type lectins and the regulation of glycosylation on its cellular counterstructures by post-translational modification will provide insight as to how these cell-surface receptors mediate cellular interactions and regulate DC function. Also, understanding the carbohydrate-recognition element present on self-antigens or

19 Feb 2004

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC

47

pathogens that regulate the interaction with a specific CLR may have consequences for how the pathogen or antigen is processed by DC.

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ACKNOWLEDGMENTS We thank all former and present members of our group, including our collaborators whose work has helped shape the ideas expressed here. We are grateful to the Netherlands Organisation for Scientific Research (NWO) Pioneer grant 016.036.607 and VENI grant 916.36.009, the AIDS Foundation, the Dutch Digestive Diseases Foundation (MLDS), and the Dutch Cancer Foundation (NKB) for their financial support. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Banchereau J, Steinman RM. 1998. Dendritic cells and the control of immunity. Nature 392:245–52 2. Steinman RM, Hawiger D, Nussenzweig MC. 2003. Tolerogenic dendritic cells. Annu. Rev. Immunol. 21:685–711 3. Matzinger P. 2002. The danger model: a renewed sense of self. Science 296:301– 5 4. Sallusto F, Lanzavecchia A. 1999. Mobilizing dendritic cells for tolerance, priming, and chronic inflammation. J. Exp. Med. 189:611–14 5. Figdor CG, van Kooyk Y, Adema GJ. 2002. C-type lectin receptors on dendritic cells and Langerhans cells. Nat. Rev. Immunol. 2:77–84 6. Drickamer K. 1999. C-type lectin-like domains. Curr. Opin. Struct. Biol. 9:585– 90 7. Underhill DM, Ozinsky A. 2002. Tolllike receptors: key mediators of microbe detection. Curr. Opin. Immunol. 14:103– 10 8. Thoma-Uszynski S, Stenger S, Takeuchi O, Ochoa MT, Engele M, et al. 2001. Induction of direct antimicrobial activity through mammalian toll-like receptors. Science 291:1544–47 9. Weis WI, Taylor ME, Drickamer K. 1998. The C-type lectin superfamily in

10.

11.

12.

13.

14.

15.

16.

the immune system. Immunol. Rev. 163: 19–34 Janeway CA Jr, Medzhitov R. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197–216 Akira S, Hemmi H. 2003. Recognition of pathogen-associated molecular patterns by TLR family. Immunol. Lett. 85:85–95 Hemmi H, Takeuchi O, Kawai T, Kaisho T, Sato S, et al. 2000. A Toll-like receptor recognizes bacterial DNA. Nature 408:740–45 Engering A, Geijtenbeek TBH, van Vliet SJ, Wijers M, van Liempt E, et al. 2002. The dendritic cell-specific adhesion receptor DC-SIGN internalizes antigen for presentation to T cells. J. Immunol. 168:2118–26 Kaufmann SH, Schaible UE. 2003. A dangerous liaison between two major killers: Mycobacterium tuberculosis and HIV target dendritic cells through DCSIGN. J. Exp. Med. 197:1–5 Geijtenbeek TBH, van Vliet SJ, Koppel EA, Sanchez-Hernandez M, Vanden broucke-Grauls CMJE, et al. 2003. Mycobacteria target DC-SIGN to suppress dendritic cell function. J. Exp. Med. 197:7–17 Gantner BN, Simmons RM, Canavera SJ, Akira S, Underhill DM. 2003.

19 Feb 2004

12:0

48

17.

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

18.

19.

20.

21.

22.

23.

24.

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL. Collaborative induction of inflammatory responses by dectin-1 and Toll-like receptor 2. J. Exp. Med. 197:1107–17 Brown GD, Herre J, Williams DL, Willment JA, Marshall AS, Gordon S. 2003. Dectin-1 mediates the biological effects of beta-glucans. J. Exp. Med. 197:1119– 24 Ichikawa HT, Williams LP, Segal BM. 2002. Activation of APCs through CD40 or Toll-like receptor 9 overcomes tolerance and precipitates autoimmune disease. J. Immunol. 169:2781–87 Sallusto F, Cella M, Danieli C, Lanzavecchia A. 1995. Dendritic cells use macropinocytosis and the mannose receptor to concentrate macromolecules in the major histocompatibility complex class II compartment: downregulation by cytokines and bacterial products. J. Exp. Med. 182:389–400 Mahnke K, Guo M, Lee S, Sepulveda H, Swain SL, et al. 2000. The dendritic cell receptor for endocytosis, DEC-205, can recycle and enhance antigen presentation via major histocompatibility complex class II-positive lysosomal compartments. J. Cell Biol. 151:673–84 Geijtenbeek TBH, Torensma R, van Vliet SJ, van Duijnhoven GCF, Adema GJ, et al. 2000. Identification of DCSIGN, a novel dendritic cell-specific ICAM-3 receptor that supports primary immune responses. Cell 100:575–85 Dzionek A, Sohma Y, Nagafune J, Cella M, Colonna M, et al. 2001. BDCA-2, a novel plasmacytoid dendritic cellspecific type II C-type lectin, mediates antigen capture and is a potent inhibitor of interferon alpha/beta induction. J. Exp. Med. 194:1823–34 Willment JA, Gordon S, Brown GD. 2001. Characterization of the human beta-glucan receptor and its alternatively spliced isoforms. J. Biol. Chem. 276:43818–23 Bates EE, Fournier N, Garcia E, Valladeau J, Durand I, et al. 1999. APCs

25.

26.

27.

28.

29.

30.

31.

32.

express DCIR, a novel C-type lectin surface receptor containing an immunoreceptor tyrosine-based inhibitory motif. J. Immunol. 163:1973–83 Ryan EJ, Marshall AJ, Magaletti D, Floyd H, Draves KE, et al. 2002. Dendritic cell-associated lectin-1: a novel dendritic cell-associated, C-type lectinlike molecule enhances T cell secretion of IL-4. J. Immunol. 169:5638–48 Colonna M, Samaridis J, Angman L. 2000. Molecular characterization of two novel C-type lectin-like receptors, one of which is selectively expressed in human dendritic cells. Eur. J. Immunol. 30:697– 704 Valladeau J, Duvert-Frances V, Pin JJ, Kleijmeer MJ, Ait-Yahia S, et al. 2001. Immature human dendritic cells express asialoglycoprotein receptor isoforms for efficient receptor-mediated endocytosis. J. Immunol. 167:5767–74 Suzuki N, Yamamoto K, Toyoshima S, Osawa T, Irimura T. 1996. Molecular cloning and expression of cDNA encoding human macrophage C-type lectin. Its unique carbohydrate binding specificity for Tn antigen. J. Immunol. 156:128–35 Valladeau J, Ravel O, Dezutter-Dambuyant C, Moore K, Kleijmeer M, et al. 2000. Langerin, a novel C-type lectin specific to Langerhans cells, is an endocytic receptor that induces the formation of Birbeck granules. Immunity 12:71–81 Jiang W, Swiggard WJ, Heufler C, Peng M, Mirza A, et al. 1995. The receptor DEC-205 expressed by dendritic cells and thymic epithelial cells is involved in antigen processing. Nature 375:151– 55 Bleijs DA, Geijtenbeek TBH, Figdor CG, van Kooyk Y. 2001. DC-SIGN and LFA-1: a battle for ligand. Trends Immunol. 22:457–63 Soilleux EJ, Morris LS, Leslie G, Chehimi J, Luo Q, et al. 2002. Constitutive and induced expression of DCSIGN on dendritic cell and macrophage

19 Feb 2004

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC

33.

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

34.

35.

36.

37.

38.

39.

40.

41.

subpopulations in situ and in vitro. J. Leukoc. Biol. 71:445–57 Geijtenbeek TBH, Krooshoop DJEB, Bleijs DA, van Vliet SJ, van Duijnhoven GCF, et al. 2000. DC-SIGN-ICAM-2 interaction mediates dendritic cell trafficking. Nat. Immunol. 1:353–57 Relloso M, Puig-Kroger A, Pello OM, Rodriguez-Fernandez JL, de la Rosa G, et al. 2002. DC-SIGN (CD209) expression is IL-4 dependent and is negatively regulated by IFN, TGF-beta, and anti-inflammatory agents. J. Immunol. 168:2634–43 Jarrossay D, Napolitani G, Colonna M, Sallusto F, Lanzavecchia A. 2001. Specialization and complementarity in microbial molecule recognition by human myeloid and plasmacytoid dendritic cells. Eur. J. Immunol. 31:3388–93 Kadowaki N, Ho S, Antonenko S, Malefyt RW, Kastelein RA, et al. 2001. Subsets of human dendritic cell precursors express different toll-like receptors and respond to different microbial antigens. J. Exp. Med. 194:863–69 Liu YJ. 2001. Dendritic cell subsets and lineages, and their functions in innate and adaptive immunity. Cell 106:259–62 Feinberg H, Mitchell DA, Drickamer K, Weis WI. 2001. Structural basis for selective recognition of oligosaccharides by DC-SIGN and DC-SIGNR. Science 294:2163–66 Geijtenbeek TBH, van Duijnhoven GCF, van Vliet SJ, Krieger E, Vriend G, et al. 2002. Identification of different binding sites in the dendritic cell-specific receptor DC-SIGN for intercellular adhesion molecule 3 and HIV-1. J. Biol. Chem. 277:11314–20 Stambach NS, Taylor ME. 2003. Characterization of carbohydrate recognition by langerin, a C-type lectin of Langerhans cells. Glycobiology 13:401–10 Frison N, Taylor ME, Soilleux E, Bousser MT, Mayer R, et al. 2003. Oligolysine-based oligosaccharide clus-

42.

43.

44.

45.

46.

47.

48.

49.

49

ters: selective recognition and endocytosis by the mannose receptor and DCSIGN. J. Biol. Chem. 278:23922–29 Mitchell DA, Fadden AJ, Drickamer K. 2001. A novel mechanism of carbohydrate recognition by the C-type lectins DC-SIGN and DC-SIGNR. Subunit organization and binding to multivalent ligands. J. Biol. Chem. 276:28939– 45 Martinez-Pomares L, Linehan SA, Taylor PR, Gordon S. 2001. Binding properties of the mannose receptor. Immunobiology 204:527–35 Avrameas A, McIlroy D, Hosmalin A, Autran B, Debre P, et al. 1996. Expression of a mannose/fucose membrane lectin on human dendritic cells. Eur. J. Immunol. 26:394–400 Weis WI, Drickamer K. 1994. Trimeric structure of a C-type mannose-binding protein. Structure 2:1227–40 Hitchen PG, Mullin NP, Taylor ME. 1998. Orientation of sugars bound to the principal C-type carbohydrate-recognition domain of the macrophage mannose receptor. Biochem. J. 333(Pt. 3):601–8 Kogelberg H, Montero E, Bay S, Lawson AM, Feizi T. 1999. Re-evaluation of monosaccharide binding property of recombinant soluble carbohydrate recognition domain of the natural killer cell receptor NKR-P1A. J. Biol. Chem. 274:30335–36 Fukui S, Feizi T, Galustian C, Lawson AM, Chai W. 2002. Oligosaccharide microarrays for high-throughput detection and specificity assignments of carbohydrate-protein interactions. Nat. Biotechnol. 20:1011–17 Appelmelk BJ, van Die I, van Vliet SJ, Vandenbroucke-Grauls CMJE, Geijtenbeek TBH, van Kooyk Y. 2003. Cutting edge: carbohydrate profiling identifies new pathogens that interact with dendritic cell-specific ICAM-3-grabbing nonintegrin on dendritic cells. J. Immunol. 170:1635–39

19 Feb 2004

12:0

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

50

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL.

50. van Die I, van Vliet SJ, Nyame AK, Cummings RD, Bank CM, et al. 2003. The dendritic cell specific C-type lectin DC-SIGN is a receptor for Schistosoma mansoni egg antigens and recognizes the glycan antigen Lewis-x. Glycobiology 13:471–78 51. Higashi N, Fujioka K, Denda-Nagai K, Hashimoto S, Nagai S, et al. 2002. The macrophage C-type lectin specific for galactose/N-acetylgalactosamine is an endocytic receptor expressed on monocyte-derived immature dendritic cells. J. Biol. Chem. 277:20686–93 52. Drickamer K. 1995. Increasing diversity of animal lectin structures. Curr. Opin. Struct. Biol. 5:612–16 53. Kato M, Neil TK, Fearnley DB, McLellan AD, Vuckovic S, Hart DN. 2000. Expression of multilectin receptors and comparative FITC-dextran uptake by human dendritic cells. Int. Immunol. 12: 1511–19 54. Engering A, Geijtenbeek TBH, van Kooyk Y. 2002. Immune escape through C-type lectins on dendritic cells. Trends Immunol. 23:480–85 55. Hawiger D, Inaba K, Dorsett Y, Guo M, Mahnke K, et al. 2001. Dendritic cells induce peripheral T cell unresponsiveness under steady state conditions in vivo. J. Exp. Med. 194:769–79 56. Ezekowitz RA, Sastry K, Bailly P, Warner A. 1990. Molecular characterization of the human macrophage mannose receptor: demonstration of multiple carbohydrate recognition-like domains and phagocytosis of yeasts in Cos-1 cells. J. Exp. Med. 172:1785–94 57. Steinman RM. 2000. DC-SIGN: a guide to some mysteries of dendritic cells. Cell 100:491–94 58. Apostolopoulos V, Pietersz GA, Gordon S, Martinez-Pomares L, McKenzie IF. 2000. Aldehyde-mannan antigen complexes target the MHC class I antigenpresentation pathway. Eur. J. Immunol. 30:1714–23

59. Apostolopoulos V, Barnes N, Pietersz GA, McKenzie IF. 2000. Ex vivo targeting of the macrophage mannose receptor generates anti-tumor CTL responses. Vaccine 18:3174–84 60. Prigozy TI, Sieling PA, Clemens D, Stewart PL, Behar SM, et al. 1997. The mannose receptor delivers lipoglycan antigens to endosomes for presentation to T cells by CD1b molecules. Immunity 6:187–97 61. Lee B, Leslie G, Soilleux E, O’Doherty U, Baik S, et al. 2001. cis Expression of DC-SIGN allows for more efficient entry of human and simian immunodeficiency viruses via CD4 and a coreceptor. J. Virol. 75:12028–38 62. Gordon S. 2002. Pattern recognition receptors: doubling up for the innate immune response. Cell 111:927–30 63. Bonifaz L, Bonnyay D, Mahnke K, Rivera M, Nussenzweig MC, Steinman RM. 2002. Efficient targeting of protein antigen to the dendritic cell receptor DEC-205 in the steady state leads to antigen presentation on major histocompatibility complex class I products and peripheral CD8+ T cell tolerance. J. Exp. Med. 196:1627–38 64. Kammerer U, Eggert AO, Kapp M, McLellan AD, Geijtenbeek TBH, et al. 2003. Unique appearance of proliferating antigen-presenting cells expressing DC-SIGN (CD209) in the decidua of early human pregnancy. Am. J. Pathol. 162:887–96 65. Pohlmann S, Soilleux EJ, Baribaud F, Leslie GJ, Morris LS, et al. 2001. DCSIGNR, a DC-SIGN homologue expressed in endothelial cells, binds to human and simian immunodeficiency viruses and activates infection in trans. Proc. Natl. Acad. Sci. USA 98:2670– 75 66. Bashirova AA, Geijtenbeek TBH, van Duijnhoven GCF, van Vliet SJ, Eilering JB, et al. 2001. A dendritic cellspecific intercellular adhesion molecule

19 Feb 2004

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC

67.

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

68.

69.

70.

71.

72.

73.

74.

75.

3-grabbing nonintegrin (DC-SIGN)related protein is highly expressed on human liver sinusoidal endothelial cells and promotes HIV-1 infection. J. Exp. Med. 193:671–78 Knolle PA, Gerken G. 2000. Local control of the immune response in the liver. Immunol. Rev. 174:21–34 Limmer A, Ohl J, Kurts C, Ljunggren HG, Reiss Y, et al. 2000. Efficient presentation of exogenous antigen by liver endothelial cells to CD8+ T cells results in antigen-specific T-cell tolerance. Nat. Med. 6:1348–54 Kukuruzinska MA, Lennon K. 1998. Protein N-glycosylation: molecular genetics and functional significance. Crit. Rev. Oral. Biol. Med. 9:415–48 Howard MJ, Isacke CM. 2002. The Ctype lectin receptor Endo180 displays internalization and recycling properties distinct from other members of the mannose receptor family. J. Biol. Chem. 277:32320–31 Engelholm LH, List K, Netzel-Arnett S, Cukierman E, Mitola DJ, et al. 2003. uPARAP/Endo180 is essential for cellular uptake of collagen and promotes fibroblast collagen adhesion. J. Cell Biol. 160:1009–15 van Kooyk Y, Geijtenbeek TBH. 2002. A novel adhesion pathway that regulates dendritic cell trafficking and T cell interactions. Immunol. Rev. 186:47–56 Funatsu O, Sato T, Kotovuori P, Gahmberg CG, Ikekita M, Furukawa K. 2001. Structural study of N-linked oligosaccharides of human intercellular adhesion molecule-3 (CD50). Eur. J. Biochem. 268:1020–29 Vestweber D, Blanks JE. 1999. Mechanisms that regulate the function of the selectins and their ligands. Physiol. Rev. 79:181–213 Leteux C, Chai W, Loveless RW, Yuen CT, Uhlin-Hansen L, et al. 2000. The cysteine-rich domain of the macrophage mannose receptor is a multispecific lectin

76.

77.

78.

79.

80.

81.

82.

83.

51

that recognizes chondroitin sulfates A and B and sulfated oligosaccharides of blood group Lewis(a) and Lewis(x) types in addition to the sulfated N-glycans of lutropin. J. Exp. Med. 191:1117– 26 Irjala H, Johansson EL, Grenman R, Alanen K, Salmi M, Jalkanen S. 2001. Mannose receptor is a novel ligand for L-selectin and mediates lymphocyte binding to lymphatic endothelium. J. Exp. Med. 194:1033–42 Daniels MA, Hogquist KA, Jameson SC. 2002. Sweet ‘n’ sour: the impact of differential glycosylation on T cell responses. Nat. Immunol. 3:903–10 Lowe JB. 2002. Glycosylation in the control of selectin counter-receptor structure and function. Immunol. Rev. 186:19–36 Curtis BM, Scharnowske S, Watson AJ. 1992. Sequence and expression of a membrane-associated C-type lectin that exhibits CD4-independent binding of human immunodeficiency virus envelope glycoprotein gp120. Proc. Natl. Acad. Sci. USA 89:8356–60 Geijtenbeek TBH, Kwon DS, Torensma R, van Vliet SJ, van Duijnhoven GCF, et al. 2000. DC-SIGN, a dendritic cellspecific HIV-1-binding protein that enhances trans-infection of T cells. Cell 100:587–97 Pohlmann S, Baribaud F, Doms RW. 2001. DC-SIGN and DC-SIGNR: helping hands for HIV. Trends Immunol. 22:643–46 Sol-Foulon N, Moris A, Nobile C, Boccaccio C, Engering A, et al. 2002. HIV-1 Nef-induced upregulation of DC-SIGN in dendritic cells promotes lymphocyte clustering and viral spread. Immunity 16:145–55 Navarro-Sanchez E, Altmeyer R, Amara A, Schwartz O, Fieschi F, et al. 2003. Dendritic-cell-specific ICAM3grabbing non-integrin is essential for the productive infection of human dendritic

19 Feb 2004

12:0

52

84.

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

85.

86.

87.

88.

89.

90.

91.

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL. cells by mosquito-cell-derived Dengue viruses. EMBO Rep. 4:1–6 Tassaneetrithep B, Burgess TH, Granelli-Piperno A, Trumpfheller C, Finke J, et al. 2003. DC-SIGN (CD209) mediates Dengue virus infection of human dendritic cells. J. Exp. Med. 197: 823–29 Gardner JP, Durso RJ, Arrigale RR, Donovan GP, Maddon PJ, et al. 2003. LSIGN (CD 209L) is a liver-specific capture receptor for hepatitis C virus. Proc. Natl. Acad. Sci. USA 100:4498–503 Pohlmann S, Zhang J, Baribaud F, Chen Z, Leslie GJ, et al. 2003. Hepatitis C virus glycoproteins interact with DC-SIGN and DC-SIGNR. J. Virol. 77:4070–80 Lozach PY, Lortat-Jacob H, De Lacroix De Lavalette A, Staropoli I, Foung S, et al. 2003. DC-SIGN and L-SIGN are high-affinity binding receptors for hepatitis C virus glycoprotein E2. J. Biol. Chem. 278:20358–66 Simmons G, Reeves JD, Grogan CC, Vandenberghe LH, Baribaud F, et al. 2003. DC-SIGN and DC-SIGNR bind ebola glycoproteins and enhance infection of macrophages and endothelial cells. Virology 305:115–23 Alvarez CP, Lasala F, Carrillo J, Muniz O, Corbi AL, Delgado R. 2002. C-type lectins DC-SIGN and L-SIGN mediate cellular entry by Ebola virus in cis and in trans. J. Virol. 76:6841–44 Halary F, Amara A, Lortat-Jacob H, Messerle M, Delaunay T, et al. 2002. Human cytomegalovirus binding to DCSIGN is required for dendritic cell infection and target cell trans-infection. Immunity 17:653–64 Lue J, Hsu M, Yang D, Marx P, Chen ZW, Cheng-Mayer C. 2002. Addition of a single gp120 glycan confers increased binding to dendritic cell-specific ICAM-3-grabbing nonintegrin and neutralization escape to human immunodeficiency virus type 1. J. Virol. 76:10299– 306

92. Lin G, Simmons G, Pohlmann S, Baribaud F, Ni HP, et al. 2003. Differential N-linked glycosylation of human immunodeficiency virus and Ebola virus envelope glycoproteins modulates interactions with DC-SIGN and DC-SIGNR. J. Virol. 77:1337–46 93. Kwon DS, Gregorio G, Bitton N, Hendrickson WA, Littman DR. 2002. DCSIGN-mediated internalization of HIV is required for trans-enhancement of T cell infection. Immunity 16:135–44 94. Hong PW, Flummerfelt KB, de Parseval A, Gurney K, Elder JH, Lee B. 2002. Human immunodeficiency virus envelope (gp120) binding to DC-SIGN and primary dendritic cells is carbohydrate dependent but does not involve 2G12 or cyanovirin binding sites: implications for structural analyses of gp120DC-SIGN binding. J. Virol. 76:12855– 65 95. Maeda N, Nigou J, Herrmann JL, Jackson M, Amara A, et al. 2003. The cell surface receptor DC-SIGN discriminates between Mycobacterium species through selective recognition of the mannose caps on lipoarabinomannan. J. Biol. Chem. 278:5513–16 96. Colmenares M, Puig-Kroger A, Pello OM, Corbi AL, Rivas L. 2002. Dendritic cell (DC)-specific intercellular adhesion molecule 3 (ICAM-3)-grabbing nonintegrin (DC-SIGN, CD209), a C-type surface lectin in human DCs, is a receptor for Leishmania amastigotes. J. Biol. Chem. 277:36766–69 97. Cambi A, Gijzen K, de Vries JM, Torensma R, Joosten B, et al. 2003. The C-type lectin DC-SIGN (CD209) is an antigen-uptake receptor for Candida albicans on dendritic cells. Eur. J. Immunol. 33:532–38 98. Engering A, Van Vliet SJ, Geijtenbeek TBH, van Kooyk Y. 2002. Subset of DCSIGN(+) dendritic cells in human blood transmits HIV-1 to T lymphocytes. Blood 100:1780–86

19 Feb 2004

12:0

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC 99. Geijtenbeek TBH, Engering A, van Kooyk Y. 2002. DC-SIGN, a C-type lectin on dendritic cells that unveils many aspects of dendritic cell biology. J. Leukoc. Biol. 71:921–31 100. Baribaud F, Doms RW, Pohlmann S. 2002. The role of DC-SIGN and DCSIGNR in HIV and Ebola virus infection: Can potential therapeutics block virus transmission and dissemination? Exp. Opin. Ther. Targets 6:423–31 101. Piguet V, Gu F, Foti M, Demaurex N, Gruenberg J, et al. 1999. Nef-induced CD4 degradation: a diacidic-based motif in Nef functions as a lysosomal targeting signal through the binding of beta-COP in endosomes. Cell 97:63–73 102. Granelli-Piperno A, Delgado E, Finkel V, Paxton W, Steinman RM. 1998. Immature dendritic cells selectively replicate macrophagetropic (M-tropic) human immunodeficiency virus type 1, while mature cells efficiently transmit both Mand T-tropic virus to T cells. J. Virol. 72:2733–37 103. Granelli-Piperno A, Finkel V, Delgado E, Steinman RM. 1999. Virus replication begins in dendritic cells during the transmission of HIV-1 from mature dendritic cells to T cells. Curr. Biol. 9:21–29 104. Petit C, Buseyne F, Boccaccio C, Abastado JP, Heard JM, Schwartz O. 2001. Nef is required for efficient HIV-1 replication in cocultures of dendritic cells and lymphocytes. Virology 286:225–36 105. Turville SG, Arthos J, Mac Donald K, Lynch G, Naif H, et al. 2001. HIV gp120 receptors on human dendritic cells. Blood 98:2482–88 106. Turville SG, Cameron PU, Handley A, Lin G, Pohlmann S, et al. 2002. Diversity of receptors binding HIV on dendritic cell subsets. Nat. Immunol. 3:975– 83 107. Kavanagh DG, Bhardwaj N. 2002. A division of labor: DC subsets and HIV receptor diversity. Nat. Immunol. 3:891–93 108. Weissman D, Fauci AS. 1997. Role of

109.

110.

111.

112.

113.

114.

115.

116.

117.

53

dendritic cells in immunopathogenesis of human immunodeficiency virus infection. Clin. Microbiol. Rev. 10:358– 67 Nguyen DG, Hildreth JE. 2003. Involvement of macrophage mannose receptor in the binding and transmission of HIV by macrophages. Eur. J. Immunol. 33:483– 93 van Kooyk Y, Appelmelk B, Geijtenbeek TBH. 2003. A fatal attraction: Mycobacterium tuberculosis and HIV-1 target DC-SIGN to escape immune surveillance. Trends Mol. Med. 9:153–59 Manabe YC, Bishai WR. 2000. Latent Mycobacterium tuberculosis—persistence, patience, and winning by waiting. Nat. Med. 6:1327–29 Kaufmann SH. 2000. Is the development of a new tuberculosis vaccine possible? Nat. Med. 6:955–60 Tailleux L, Schwartz O, Herrmann J-L, Pivert E, Jackson M, et al. 2003. DCSIGN is the major Mycobacterium tuberculosis receptor on human dendritic cells. J. Exp. Med. 197:121–27 Ehlers MR, Daffe M. 1998. Interactions between Mycobacterium tuberculosis and host cells: Are mycobacterial sugars the key? Trends Microbiol. 6:328– 35 Schlesinger LS, Hull SR, Kaufman TM. 1994. Binding of the terminal mannosyl units of lipoarabinomannan from a virulent strain of Mycobacterium tuberculosis to human macrophages. J. Immunol. 152:4070–79 Jiao X, Lo-Man R, Guermonprez P, Fiette L, Deriaud E, et al. 2002. Dendritic cells are host cells for mycobacteria in vivo that trigger innate and acquired immunity. J. Immunol. 168:1294–301 Fortsch D, Rollinghoff M, Stenger S. 2000. IL-10 converts human dendritic cells into macrophage-like cells with increased antibacterial activity against virulent Mycobacterium tuberculosis. J. Immunol. 165:978–87

19 Feb 2004

12:0

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

54

AR

AR210-IY22-02.tex

AR210-IY22-02.sgm

LaTeX2e(2002/01/18)

P1: IKH

GEIJTENBEEK ET AL.

118. Chatterjee D, Khoo KH. 1998. Mycobacterial lipoarabinomannan: an extraordinary lipoheteroglycan with profound physiological effects. Glycobiology 8:113–20 119. Sada E, Brennan PJ, Herrera T, Torres M. 1990. Evaluation of lipoarabinomannan for the serological diagnosis of tuberculosis. J. Clin. Microbiol. 28:2587–90 120. Nigou J, Zelle-Rieser C, Gilleron M, Thurnher M, Puzo G. 2001. Mannosylated lipoarabinomannans inhibit IL-12 production by human dendritic cells: evidence for a negative signal delivered through the mannose receptor. J. Immunol. 166:7477–85 121. Tsuji S, Matsumoto M, Takeuchi O, Akira S, Azuma I, et al. 2000. Maturation of human dendritic cells by cell wall skeleton of Mycobacterium

122.

123.

124.

125.

bovis bacillus Calmette-Guerin: involvement of toll-like receptors. Infect. Immun. 68:6883–90 Jonuleit H, Schmitt E, Schuler G, Knop J, Enk AH. 2000. Induction of interleukin 10-producing, nonproliferating CD4(+) T cells with regulatory properties by repetitive stimulation with allogeneic immature human dendritic cells. J. Exp. Med. 192:1213–22 Grunebach F, Weck MM, Reichert J, Brossart P. 2002. Molecular and functional characterization of human Dectin1. Exp. Hematol. 30:1309–15 Brown GD, Gordon S. 2001. Immune recognition. A new receptor for betaglucans. Nature 413:36–37 Bluestone JA, Abbas AK. 2003. Natural versus adaptive regulatory T cells. Nat. Rev. Immunol. 3:253–57

Geijtenbeek.qxd

3/12/2004

1:47 PM

Page 1

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIGEN RECOGNITION BY C-TYPE LECTINS ON DC

C-1

Figure 1 Internalization of antigens by different C-type lectins leads to antigen routing to distinct intracellular compartments. The MR rapidly internalizes antigens and detaches from the antigen in early endosomes to cycle back to the cell surface. The antigen is subsequently targeted to the lysosomes. DC-SIGN internalizes antigens to lysosomes to allow loading on MHC class II molecules. HIV-1 interferes with the DC-SIGN internalization pathway to hide within early endosomes in an infectious form. Internalization pathways of CLRs are dependent on their cytoplasmic regions. Many CLRs contain different internalization signals.

Geijtenbeek.qxd

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

3/12/2004

1:47 PM

Page 2

GEIJTENBEEK ET AL.

Figure 2 Balanced activation of C-type lectins (CLRs) and Toll-like receptors (TLRs) determines DC differentiation and the resulting immune responses. (A) Recognition of self-antigens by CLRs can lead to a differential outcome of immune response. Targeting of CLRs with self-antigens, in the absence of TLR triggering, leads to processing of antigen, without any DC maturation leading to tolerance induction. During inflammatory responses, TLRs are triggered, and subsequently DCs maturate. When simultaneously self-antigens are captured by CLRs, T cell reactivity against self-antigens is initiated leading to autoimmunity. (B) Cross talk between CLRs and TLRs is essential in finetuning immune responses. Recognition of pathogens such as mycobacteria by the CLR DC-SIGN and TLRs can lead to immune activation when the TLR-signal overrules that of the CLR, and this includes DC differentiation. This can occur when low amounts of pathogens target the CLR. When high concentrations of pathogens target the CLR DCSIGN, the immune tolerizing signals can overrule the TLR-induced signals and inhibit DC differentiation, leading to immune suppression and pathogen survival.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22 Annu. Rev. Immunol. 2004.22:33-54. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104807

Annu. Rev. Immunol. 2004. 22:55–79 doi: 10.1146/annurev.immunol.22.012703.104807 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on September 15, 2003

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT1

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Meinrad Busslinger Research Institute of Molecular Pathology, Vienna Biocenter, A-1030 Vienna, Austria; email: [email protected]

Key Words Ikaros, PU.1, E2A, EBF, Pax5 ■ Abstract The generation of B-lymphocytes from hematopoietic stem cells is controlled by multiple transcription factors regulating distinct developmental aspects. Ikaros and PU.1 act in parallel pathways to control the development of lymphoid progenitors in part by regulating the expression of essential signaling receptors (Flt3, c-Kit, and IL-7Rα). The generation of the earliest B cell progenitors depends on E2A and EBF, which coordinately activate the B cell gene expression program and immunoglobulin heavy-chain gene rearrangements at the onset of B-lymphopoiesis. Pax5 restricts the developmental options of lymphoid progenitors to the B cell lineage by repressing the transcription of lineage-inappropriate genes and simultaneously activating the expression of B-lymphoid signaling molecules. LEF1 and Sox4 contribute to the survival and proliferation of pro-B cells in response to extracellular signals. Finally, IRF4 and IRF8 together control the termination of pre-B cell receptor signaling and thus promote differentiation to small pre-B cells undergoing light-chain gene rearrangements.

INTRODUCTION The pluripotent hematopoietic stem cell (HSC) with its extensive self-renewal potential regenerates all blood cell types throughout life by differentiating to progenitor cells with gradually restricted developmental potential. An early step in 1

Abbreviations used: bHLH, basic helix-loop-helix; BSAP, B-cell-specific activator protein; CLP, common lymphoid progenitor; CMP, common myeloid progenitor; DN, dominantnegative; EBF, early B cell factor; ELP, earliest lymphocyte progenitor; ETP, early T-lineage progenitor; FL, Flt3 ligand; HAT, histone acetyltransferase; HDAC, histone deacetylase; HMG, high mobility group; HMTase, histone methyltransferase; HSC, hematopoietic stem cell; IgH, immunoglobulin heavy-chain; IgL, immunoglobulin light-chain; MPP, multipotent progenitor; PDZ, postsynaptic density protein, disks large, zonula occludens (this abbreviation is based on the first proteins where this protein-protein interaction domain was identified); RSS, recombination signal sequences; SCF, stem cell factor; SET, Su(var)3-9, Enhancer of zeste, Trithorax (this abbreviation is based on the first proteins where this enzymatic domain was identified); TAD, transactivation domain; TCR, T cell receptor.

0732-0582/04/0423-0055$14.00

55

19 Feb 2004

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

56

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER

Figure 1 B cell development in the bone marrow. The expression of signaling receptors, the rearrangement status of immunoglobulin genes and the initiation of expression of characteristic cell surface proteins are indicated for successive progenitor cell stages of mouse B-lymphopoiesis. For further details, see references describing the characterization of the different cell types: HSC (hematopoietic stem cell) (1), MPP (multipotent progenitor) (3, 4), ELP (earliest lymphocyte progenitor) (5), ETP (early T-lineage progenitor) (6), CLP (common lymphoid progenitor) (1, 7), CLP-2 (8, 9), B-lymphocytes of fractions A–E (10–12), and CMP (common myeloid progenitor) (2). LIN−, negative for the expression of lineage-specific markers.

hematopoiesis is the commitment of multipotent progenitors (MPP) to either the lymphoid or erythro-myeloid lineages, resulting in the formation of the common lymphoid (CLP) or common myeloid (CMP) progenitors (1, 2) (Figure 1). The loss of long-term self-renewal capacity is accompanied by expression of the tyrosine kinase receptor Flt3 (also known as Flk2) in MPP cells (3, 4). The Flt3+ MPPs subsequently differentiate to the earliest lymphocyte progenitors (ELP), which initiate RAG1 and RAG2 expression and start to undergo DH-JH rearrangements at the immunoglobulin heavy-chain (IgH) locus (5) (Figure 1). The ELP cell is likely to give rise to the recently identified early T-lineage progenitor (ETP) in the thymus (6) and to the CLP in the bone marrow, which is able to develop into four distinct cell types: B, T, NK, and DC cells (1, 7) (Figure 1). Subsequent expression of the B cell marker B220 coincides with differentiation to CLP-2 cells (8, 9), constituting a subset of fraction A cells (10, 11) that enter the B cell pathway upon induction of CD19 expression and complete DH-JH rearrangements at the early pro-B cell stage (fraction B; Figure 1) (10, 12, 13). Productive VH-DJH recombination in late pro-B cells (fraction C) results in cell surface expression of the Igµ proteins as part of the pre-B cell receptor (pre-BCR), which acts as an important checkpoint to control the transition from the pro-B to the pre-B cell stage. Signaling from the pre-BCR promotes allelic exclusion at the IgH locus, stimulates proliferative cell expansion, and induces differentiation to small pre-B cells, which start to recombine immunoglobulin light-chain genes (14). Successful light-chain gene

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTROL OF EARLY B-LYMPHOPOIESIS

P1: IKH

57

rearrangement leads to the emergence of immature IgM+ B cells that emigrate from the bone marrow to peripheral lymphoid organs (14). Signaling via different transmembrane receptors is essential for guiding the differentiation of HSCs along the B cell pathway. Activation of the tyrosine kinase receptor c-Kit by its ligand SCF (stem cell factor) promotes the survival of long-term reconstituting HSCs in vitro (3, 15), although the absence of c-Kit does not affect the survival and engraftment of HSCs in vivo in a viable c-kitW/W (Vickid) mouse strain (16). The CLP and all later stages of B cell development are, however, severely depleted in the bone marrow of adult c-kitW/W mice (16). Similarly, signaling through the Flt3 receptor is required for efficient formation of the CLP and subsequent development of pro-B and pre-B cells, as these cell types are strongly reduced in the absence of the Flt3 ligand (FL) in the bone marrow of FL−/− mouse (17, 18). In contrast, normal numbers of the CLP are generated in γ c−/− mice lacking a functional IL-7 receptor, which is composed of the IL-7Rα protein and the signaling γ c chain (19). The CLP is, however, severely compromised in its ability to differentiate even to the earliest pro-B cell stage in the adult bone marrow of γ c−/−, IL-7Rα −/−, or IL-7−/− mice (19, 20). Consistent with this early developmental block, the lymphoid cytokine IL-7 not only signals pro-B cell survival, but also functions as a differentiation factor to induce B cell development of the CLP (19). Double-mutant flt3−/−IL-7Rα −/− mice entirely lack B-lymphocytes in the bone marrow (21). Likewise, early B cell development is more severely affected in flt3−/−c-kitW/Wv mice compared with single-mutant mice (22). Hence, the c-Kit, Flt3, and IL-7R signaling systems together account for the generation of all B-lymphocytes in the bone marrow of adult mice. In addition to cytokine signaling, a multitude of lineage-restricted transcription factors control B cell development from the HSC to immunoglobulin-secreting plasma cells (23). This review focuses on the transcription factors that regulate early B-lymphopoiesis in the bone marrow.

GENERATION OF LYMPHOID PROGENITORS FROM THE HSC Ikaros Controls the Development of Lymphoid Progenitors Ikaros together with Helios and Aiolos constitute a family of zinc finger transcription factors involved in hematopoietic cell fate decisions (24). These transcription factors form homo- or heterodimers via two C-terminal zinc fingers, while their DNA-binding specificity is determined by the N-terminal zinc fingers (Figure 2a). The Ikaros gene is expressed in all hematopoietic lineages including stem cells and multipotent progenitors (25–27) and gives rise, by alternative splicing, to eight different Ikaros isoforms, some of which lack the DNA-binding domain and thus function as dominant-negative proteins by sequestering members of the Ikaros

19 Feb 2004

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

58

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER

Figure 2 Ikaros and PU.1 control the development of multiple hematopoietic lineages. The functional domains of the longest Ikaros isoform 1 (a) and of PU.1 (b) are shown together with the corresponding consensus DNA-binding sequence, the null phenotype, and known target genes. Genes that have been identified as direct targets of these transcription factors are shown in bold face. The six zinc fingers (F1–F6) and three targeted mutations of the Ikaros gene are indicated. For further explanations, see text with its cited literature.

protein family into inactive heterodimers (28). Targeted deletion of the N-terminal zinc finger exons therefore results in a dominant-negative (DN) Ikaros (Ik) allele, which also interferes with Helios and Aiolos function (29), whereas inactivation of the C-terminal zinc fingers eliminates the dimerization function and thus creates an Ikaros null (−) allele (30) (Figure 2a). IkDN/DN mice fail to generate any B, T, NK, and DC cells, indicating that Ikaros together with its family members are required for the development of all lymphocytes (29, 31). The Ikaros null mutation causes a 30- to 40-fold reduction of the long-term reconstitution activity of HSCs (32), prevents B and NK cell development, and leads to a strong reduction of DC differentiation and to abnormal T-lymphopoiesis in postnatal mice (30, 31) (Figure 2a). These lymphoid-specific defects can be explained by the recent finding that the generation of the CLP is entirely lost, whereas development of the early T-lineage progenitor (ETP) is fairly normal in Ik−/− mice (6). Ikaros also plays an essential

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTROL OF EARLY B-LYMPHOPOIESIS

P1: IKH

59

role within the B-lymphoid lineage, as the development of B cells is arrested at the pro-B to pre-B cell transition in mice homozygous for a hypomorphic IklacZ allele (33) (Figure 2a). At later stages, Ikaros acts in concert with Aiolos (25) to control B cell maturation, proliferation, and germinal center formation (33, 34). The differentiation of erythroid and myeloid cell types is only mildly affected in Ikaros mutant mice (30, 31, 35, 36). The Ikaros null mutation even leads to a relative increase of erythroid and myeloid progenitors in the bone marrow and spleen, in contrast to its pronounced effects on the generation of lymphoid precursor cells (32). Ikaros therefore influences the outcome of early lineage decisions by promoting lymphocyte differentiation at the expense of myeloid development and thus contributes to the commitment of HSCs to the lymphoid lineages (24). At the molecular level, Ikaros acts as an unconventional transcription factor to repress or activate genes by recruiting potent corepressor complexes (containing Sin3 or CtBP) or chromatin remodeling machines (NuRD or SWI/SNF) to its target genes (37–39). Instead of regulating the transcription initiation process itself, Ikaros is thought to function as an epigenetic regulator by either modulating the local chromatin structure (38, 40) or targeting genes to the pericentromeric heterochromatin (41, 42). The GM-CSFRα gene is derepressed in Lin−c-Kitlow progenitors of the Ik−/− bone marrow, suggesting that Ikaros promotes lymphoid cell fates by repressing key myeloid genes (32). On the other hand, Ikaros is essential for activating the c-kit and flt3 genes in HSCs and multipotent progenitors (32). The loss of flt3 transcription and the reduction of c-kit expression are likely to account for the lack of the CLP and its lymphoid progeny in the bone marrow of Ik−/− mice, as the generation of the CLP depends on signaling of both tyrosine kinase receptors (16, 18).

PU.1 is Involved in the Myeloid Versus Lymphoid Lineage Decision PU.1 (Spi-1) is a member of the Ets transcription factor family, which is required for the generation of both the innate and adaptive immune systems (43, 44) (Figure 2b). PU.1 is expressed in hematopoietic stem cells, multipotent progenitors, and all differentiating cells except erythroblasts, megakaryocytes, and T cells (2, 45, 46). PU.1-deficient embryos die around birth and fail to generate myeloid as well as lymphoid cells, whereas erythrocytes and megakaryocytes develop normally in the fetal liver (43, 44, 47). PU.1 is stringently required for the development of macrophages, osteoclast, mast cells, and B-lymphocytes (43, 47–49), while the differentiation of granulocytes, NK, and T cells is delayed and severely affected in the absence of PU.1 (44, 50–52) (Figure 2b). PU.1 appears to act at the level of multipotent myeloid-lymphoid progenitors, which are reduced in the fetal liver of PU.1-deficient embryos (47). The PU.1−/− progenitors can be propagated in vitro in the presence of the multilineage cytokines IL-3, IL-6 and SCF, but fail to respond to the myeloid cytokines M-CSF, G-CSF, GM-CSF, and lymphoid cytokine IL-7 owing to the absence or reduced expression of the corresponding receptors (50, 53)

19 Feb 2004

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

60

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER

(Figure 2b). Hence, PU.1 controls the cytokine-dependent proliferation, survival, and differentiation of multipotent progenitor cells (50). Retroviral reconstitution experiments have implicated PU.1 in the myeloid versus lymphoid lineage decision, as low PU.1 expression induces B cell development in PU.1−/− progenitors, whereas high PU.1 levels suppress the B cell fate and promote instead macrophage differentiation under in vitro culture conditions (54). Hence, graded expression of PU.1 specifies either the myeloid or lymphoid fate of early hematopoietic progenitors. Interestingly, Ikaros is normally expressed in PU.1−/− progenitors (47, 53), suggesting that PU.1 and Ikaros act in parallel pathways to regulate lymphoid development in agreement with the distinct phenotypes of Ik−/− and PU.1−/− mice (Figure 2). The IL-7Rα gene was recently identified as a direct target of PU.1, as it contains a functional PU.1 binding site in its 50 region and fails to be transcribed in PU.1−/− progenitors (53). Moreover, retroviral expression of the IL-7Rα gene alone is sufficient to instruct PU.1−/− progenitors to differentiate into pro-B cells at low efficiency (53). Hence, PU.1 regulates early B cell development, at least in part, by activating the IL-7Rα gene and thus rendering lymphoid progenitors responsive to the pro-B cell differentiation and survival factor IL-7.

TRANSCRIPTIONAL CONTROL OF PRO-B CELL DEVELOPMENT Coordinate Regulation of Pro-B Cell Development and Survival by E2A and EBF Differentiation of the CLP to committed pro-B cells depends critically on the three transcription factors E2A, EBF, and Pax5 (23). The E2A gene codes for the alternative splice products E12 and E47 (55), which together with HEB and E2-2 constitute a family of related basic helix-loop-helix (bHLH) proteins (known as E proteins; Figure 3a). Although the widely expressed E2A proteins normally form heterodimers with tissue-restricted bHLH proteins, they specifically function as homodimers in the B-lymphoid lineage (56). The B-cell-specific homodimerization may be caused by hypophosphorylation (57) or, more likely, by increased expression (58) of the E2A protein at the onset of B-lymphopoiesis and is a possible reason why the B-lymphoid lineage is most severely affected by mutation of the E2A gene. B cell development in E2A−/− mice is arrested at its earliest stage in the absence of DH-JH rearrangements at the IgH locus (59, 60). The few B220+CD43+ cells (fraction A) in the E2A−/− bone marrow may even be precursors of other lineages, as they only express IL-7Rα, Igβ (B29), and immunoglobulin µo transcripts, whereas RAG1, Igα (mb-1), Iµ, λ5, CD19, and Pax5 expression could not be detected in these cells (59, 61). Importantly, B-lymphopoiesis is dependent on the combined dosage of all expressed E proteins, as pro-B cell development is severely affected in the absence of either E12 or E47 (58, 61), whereas it is reduced twofold in mice lacking HEB or E2-2 (62).

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTROL OF EARLY B-LYMPHOPOIESIS

P1: IKH

61

Figure 3 Cooperative activation of B-cell-specific genes by E2A and EBF. (a) The functional domains of E2A (55) and EBF (63) are shown together with the consensus recognition sequences. The E2A isoforms E12 and E47 differ only in the basic helixloop-helix (bHLH) region (55). TAD, transactivation domain. (b) A schematic diagram of the λ5 and mb-1 promoters indicates the arrangements of functional binding sites for the different transcription factors (70, 73).

The early B cell factor (EBF) also forms homodimers and recognizes DNA via an N-terminal domain containing an essential zinc coordination motif (63, 64) (Figure 3a). In contrast to E2A, EBF is specifically expressed in pro-B, pre-B, and B cells within the hematopoietic system (63). Interestingly, the loss of EBF results in a similar arrest of early B cell development, as observed in E2A mutant mice (65). The few B220+CD43+ cells (fraction A) present in the bone marrow of EBF−/− mice contain the IgH gene in germline configuration and express IL7Rα as well as immunoglobulin µo and Iµ transcripts, but fail to transcribe the RAG1, RAG2, Igα (mb-1), Igβ (B29), λ5, VpreB, CD19, and Pax5 genes (65). The similarity of the B cell developmental arrest in E2A and EBF mutant mice strongly suggests that the two transcription factors act in concert to control the earliest phase of B-lymphopoiesis. This concept is supported by the results of three different experiments. First, compound heterozygous E2A+/−EBF+/− mice provide genetic evidence for cooperation between E2A and EBF, as B cell development is arrested at the pro-B cell stage in these mice because of reduced expression of multiple B-lymphoid genes including RAG1 and RAG2 (66). Second, forced expression of E2A or EBF in non-B-lymphoid cell lines activates the expression of similar B-cell-specific genes (67–69). More importantly, E2A and EBF act in

19 Feb 2004

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

62

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER

strong synergy to induce the transcription of the endogenous λ5 and VpreB genes upon ectopic expression in a hematopoietic progenitor cell line (70). Third, detailed molecular analyses demonstrate that E2A and EBF bind to and cooperatively activate the promoters of the λ5 (70, 71), VpreB (72), and Igα (mb-1) (73) genes, whereas so far only EBF was shown to regulate the Igβ (B29) promoter (74) (Figure 3b). Moreover, chromatin immunoprecipitation experiments confirmed a direct interaction of E2A with all four target genes (75). This molecular and genetic evidence indicates that E2A and EBF cooperatively regulate the expression of the surrogate light chain (λ5, VpreB) and signaling (Igα, Igβ) components of the pre-BCR, as well as transcription of the RAG1 and RAG2 subunits of the V(D)J recombinase. None of the identified target genes can account for the early developmental block in E2A−/− and EBF−/− mice. Moreover, the inability to grow E2A- or EBFdeficient progenitors in vitro precludes a systematic search for E2A and EBF target genes. However, the HLH inhibitors of the Id family provide an alternative approach for studying E2A function, as these proteins lack the basic region required for DNA binding and thus act as negative regulators by sequestering the E2A protein into inactive heterodimers (76). Constitutive expression of a B-cell-specific Id1 transgene results in an early developmental block resembling that of E2Adeficient mice (77). Under physiological conditions, the endogenous Id3 gene is rapidly activated in B cell progenitors by TGF-β1, which is a potent inhibitor of lymphocyte growth and survival (78). Retroviral expression of Id3 in pro-B cells also results in the loss of cell viability, suggesting that the control of progenitor cell growth and survival is an essential function of E2A in early B cell development (78). Transgenic expression of the anti-apoptotic Bcl-2 protein is, however, unable to rescue the developmental block in E2A−/− mice, pointing to an additional role for E2A in controlling early differentiation (79). Pro-B cell development is characterized by successive rearrangements of the IgH locus first at the DH-JH segments, followed by VH-DJH recombination. E2A and EBF control the first rearrangement step in part by regulating the expression of RAG1 and RAG2 (59, 65, 66, 69). A more direct role of E2A and EBF in V(D)J recombination was indicated by reconstitution experiments in a human embryonic kidney cell line (80, 81). Upon ectopic expression, the E2A or EBF protein functions in concert with RAG1 and RAG2 to activate DH-JH rearrangements of the endogenous IgH locus in these non-B-lymphoid cells (80). The transactivation function of E2A is essential for inducing these rearrangements and the corresponding germline transcription (80), in agreement with the fact that the immunoglobulin Eµ enhancer with its E2A-binding sites is essential for proper regulation of V(D)J recombination (82, 83). However, not all DH and JH segments are equally accessible for V(D)J recombination, suggesting that the presence of E2A- or EBF-binding sites in the vicinity of each segment determines the local accessibility and thus the rearrangement frequency (81). E2A most likely opens up the local chromatin by using the N-terminal transactivation domain to recruit the SAGA histone acetyltransferase complex to its target sequences (84).

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

CONTROL OF EARLY B-LYMPHOPOIESIS

P1: IKH

63

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Circumstantial evidence indicates that E2A may act upstream of EBF in the genetic hierarchy of B cell development. First, fraction A cells of EBF−/− mice express almost normal levels of E2A mRNA (65), whereas EBF transcripts appear to be reduced in E2A−/− bone marrow cells (61). Second, ectopic expression of E2A activates the endogenous EBF gene in a 70Z/3 macrophage cell line in contrast to EBF, which is unable to induce E2A transcription in the same cells (69). Finally, the activity of the EBF promoter seems to depend on a functional E2A-binding site (85).

B-Lineage Commitment by Pax5 (BSAP) The mere activation of B-cell-specific genes and V(D)J recombination by E2A and EBF is not sufficient to commit B cell progenitors to the lymphoid lineage in the absence of Pax5 [also known as B-cell-specific activator protein (BSAP)] (86) (Figure 4a). B cell development is arrested at an early pro-B cell stage in the bone marrow of Pax5−/− mice (87). Pax5−/− pro-B cells, which can be cultured ex vivo in the presence of IL-7 and stromal cells (88), still retain a broad lymphomyeloid potential characteristic of uncommitted progenitors (89, 90). Upon IL-7 withdrawal and appropriate cytokine stimulation, the Pax5−/− pro-B cells are able to differentiate in vitro into functional NK cells, dendritic cells, macrophages,

Figure 4 Dual role of Pax5 in early B cell development. (a) The functional domains and consensus recognition sequence of Pax5 are shown. BSAP, B-cell-specific activator protein; HD, partial homeodomain; ID, inhibitory domain; OP, conserved octapeptide. (b) Pax5 represses lineage-inappropriate genes (such as MPO, Notch1, M-CSFR) and simultaneously activates B-cell-specific genes (such as BLNK, Igα, CD19) at B-lineage commitment (89). MPO, myeloperoxidase.

19 Feb 2004

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

64

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER

osteoclasts, and granulocytes (89). This multilineage potential of Pax5−/− pro-B cells is, however, suppressed by retroviral restoration of Pax5 expression, which rescues development to the mature B cell stage (89). Hence, Pax5 is the critical B-lineage commitment factor that restricts the developmental options of early progenitors to the B cell pathway. After injection into sublethally irradiated mice, the Pax5−/− pro-B cells home to the bone marrow, where they undergo self-renewal and develop into functional cells of all major hematopoietic lineages except for B-lymphocytes (89–91). Although the Pax5−/− pro-B cells share the characteristic features of long-term reconstitution, self-renewal, and multipotency with the HSC (91, 92), they differ from pluripotent stem cells in that they fail to radioprotect lethally irradiated mice. This inability is caused by the different efficiency of Pax5−/− pro-B cells in giving rise to the distinct blood cell types in vivo. Pax5−/− pro-B cells rapidly develop into T-lymphocytes within one week after cell transfer (90), whereas macrophages and granulocytes require 2 to 3 months and erythrocytes 4 to 6 months for their generation (91). However, ectopic expression of myeloid C/EBPα or GATA transcription factors strongly promotes myeloid differentiation of Pax5−/− pro-B cells in vitro as well as in vivo, which indicates that Pax5−/− pro-B cells are lymphoid progenitors with a latent erythro-myeloid developmental potential (93). Pax5 is exclusively expressed within the hematopoietic system from the pro-B to the mature B cell stage (86), raising the question of whether Pax5 is required to maintain B cell identity throughout B-lymphopoiesis. Indeed, conditional Pax5 inactivation leads to loss of the identity and function of mature B cells (94). Cre-mediated gene deletion in committed pro-B cells demonstrated that Pax5 is essential not only to initiate its B-lymphoid transcription program, but also to maintain it in early B cell development (95). As a consequence of Pax5 inactivation, previously committed pro-B cells regain the capacity to differentiate into macrophages in vitro and to reconstitute T cell development in vivo (95). The loss of Pax5 alone is therefore sufficient to reverse B-lineage commitment by converting pro-B cells with a restricted B-lymphoid potential into early progenitors with a broad developmental potential. Committed B-lymphocytes thus retain a surprising degree of epigenetic plasticity. As expected for uncommitted hematopoietic progenitors (96), the Pax5−/− proB cells not only transcribe genes characteristic of the pro-B cell stage (97), but also express genes of other lineage-specific programs (89). At lineage commitment, Pax5 fulfills a dual role by repressing lineage-inappropriate genes and simultaneously activating B-cell-specific genes, which leads to the consolidation of the B-lymphoid gene expression program (89) (Figure 4b). The Pax5-dependent repression of the M-CSFR and Notch1 genes illustrates at the molecular level how developmental options are suppressed in committed B-lymphocytes, as these cells are no longer responsive to the myeloid cytokine M-CSF or to T-cell-inducing Notch1 ligands (89, 98). On the other hand, activated Pax5 target genes code for essential components of the (pre)BCR signaling pathway, including the receptor signaling chain Igα (mb-1) (88, 99), the stimulatory coreceptor CD19 (97,

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTROL OF EARLY B-LYMPHOPOIESIS

P1: IKH

65

100), and the central adaptor protein BLNK (101) (Figure 4b). CD19 and BLNK are critical Pax5 target genes, as B cell development is stringently arrested in CD19−/−BLNK−/− mice at a stage similar to that of Pax5−/− mice (102). Pax5 also facilitates expression of the Igµ chain by controlling the second VH-DJH recombination step of the IgH gene, as rearrangement of the numerous distal VHJ558 genes is reduced 50-fold in Pax5−/− pro-B cells (88), although the proximal VH genes recombine at almost normal frequency (103). Pax5 was recently shown to cooperate with chromatin regulators to induce large-scale contraction of the IgH locus, which promotes long-range VH-DJH recombination by juxtaposition of distal VH genes next to the proximal DHJH-rearranged gene segment (M. Fuxa, J. Skok, & M. Busslinger, unpublished data). In conclusion, Pax5 seems to control B-lineage commitment, at least in part, by shutting down inappropriate signaling systems and by simultaneously facilitating signal transduction from the pre-BCR and BCR, which both execute important checkpoint functions in B cell development. The transcriptional activity of Pax5 is determined by its interaction with distinct partner proteins. For instance, Pax5 is converted from a transcriptional activator to a repressor through interaction with corepressors of the Groucho protein family (104). However, Pax5 requires the cooperation of another Groucho-interacting transcription factor (such as PU.1) to stably recruit Groucho proteins in a contextdependent manner to repressed target genes (105). A second well-studied example is the cooperative binding of Pax5 and Ets proteins to the mb-1 promoter, which results in transcriptional activation (99) (Figure 3b). In this case, the paired domain of Pax5 recruits a subset of Ets proteins to a low-affinity Ets-binding site by directly interacting with the Ets domain (99) and thereby altering its DNA sequence recognition (106). E2A, EBF, and their target genes are normally expressed in Pax5−/− pro-B cells, which demonstrates that Pax5 functions downstream of E2A and EBF in the control of B cell development (88, 97). In contrast, Pax5 expression is reduced in pro-B cells of compound heterozygous E2A+/−EBF+/− mice (66) and can be induced by ectopic expression of E2A or EBF in a 70Z/3 macrophage cell line (69). E2A and EBF are therefore thought to directly regulate the Pax5 gene, although so far there is no conclusive evidence for such a regulation in B-lymphocytes.

Cross-Repression of Lymphoid Pathways The development of NK cells depends on the inhibitory HLH protein Id2, which is specifically expressed in NK cell progenitors where it plays an essential role in controlling commitment to the NK cell lineage (107, 108). Id2−/− mice additionally lack two subsets of DCs, the splenic CD8α + DCs and Langerhans cells, in agreement with the observation that Id2 expression is strongly induced in differentiating DC progenitors (109, 110). Conversely, forced expression of Id2 or Id3 is compatible with in vitro differentiation of human hematopoietic progenitors into DC and NK cells, whereas it efficiently prevents the generation of B-lymphocytes (111, 112). Consequently, DC and NK cell development can take place only if

19 Feb 2004

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

66

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER

Id2 actively suppresses the B cell option of lymphoid progenitors by antagonizing E2A, the first essential regulator of the B-lymphoid lineage (Figure 5). B cell development therefore appears to be the primary or default fate of the CLP in the bone marrow, whereas Id2 functions as a molecular switch to activate the secondary DC and NK cell fates. Early T cell development requires Notch1 signaling, which is activated by Delta and Jagged ligands expressed on stromal cells of the thymus (113). Upon ligand binding, the intracellular domain is cleaved from the Notch1 receptor and then functions as a transcription factor to initiate T-cell-specific gene expression in lymphoid progenitors (113). Notch1 has been implicated in T-lineage specification by gain- and loss-of-function experiments. Retroviral expression of the active intracellular Notch1 domain induces T cell development at the expense of B-lymphopoiesis in hematopoietic progenitors of the bone marrow (114). Likewise, HSCs fail to develop into B cells, but differentiate into CD4+CD8+ T cells upon co-culture with Delta-1-expressing stromal cells (115, 116). In complementary experiments, conditional inactivation of Notch1 in HSCs arrests T cell development at the earliest precursor stage, while simultaneously promoting B cell development in the thymus (117, 118). Hence, lymphoid progenitors in the thymus adopt the default B cell fate in the absence of Notch1 signaling. Notch1 is thought to suppress B-lymphopoiesis by interfering with the transcriptional activity of E2A, although the exact molecular mechanism of the Notch1-E2A antagonism remains to be elucidated (114, 119) (Figure 5). In conclusion, the commitment of lymphoid progenitors to the DC, NK, and T cell lineages relies on a similar strategy for suppressing the primary B cell fate by inactivating the first regulator E2A of the B cell pathway. Interestingly, Pax5 and Notch1 influence the

Figure 5 Transcriptional control and cross-repression of lymphoid pathways. See text for explanations and references.

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

CONTROL OF EARLY B-LYMPHOPOIESIS

P1: IKH

67

B- versus T-lineage choice precisely in the opposite manner, as pan-hematopoietic expression of Pax5 strongly promotes B cell development at the expense of Tlymphopoiesis (98, 120). At the molecular level, Pax5 represses transcription of the Notch1 gene, thereby suppressing the T cell option at B-lineage commitment (98).

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

REGULATORS OF THE TRANSCRIPTIONAL RESPONSE TO EXTRACELLULAR SIGNALS LEF1 and Wnt Signaling The survival and proliferation of pro-B cells depends on IL-7 signaling (121, 122). A similar role has been described for Wnt signaling based on targeted inactivation of the HMG box-containing protein LEF1, which regulates the transcriptional response of the canonical Wnt/β-catenin pathway (123). The LEF1 gene is expressed under the control of Pax5 (97) in pro-B cells (123) and is then rapidly repressed in response to pre-BCR signaling (101). Although B cells still develop normally in the absence of LEF1, their number is considerably reduced because of impaired proliferation and increased apoptosis of LEF1−/− pro-B cells (123). Consistent with the notion that LEF1 mediates Wnt signaling, a complementary phenotype is seen in wild-type pro-B cells exposed to the Wnt3A ligand, which induces cell cycle entry and proliferation of these cells (123). Consequently, Wnt signaling has a mitogenic role in early B cell development similar to its self-renewal function in HSCs (124).

Signal-Dependent Activation of Sox4 at the Plasma Membrane Sox4 codes for another HMG box-containing transcription factor that is expressed in early B-lymphopoiesis similar to LEF1 (125). In contrast to the LEF1 mutation, B cell development is stringently arrested at the pro-B to pre-B cell transition in the absence of Sox4 (126). The pro-B cells in Sox4−/− mice are moderately reduced in number, but strongly impaired in their capacity to proliferate in response to IL-7 (126). Paradoxically, however, IL-5 signaling has been shown to stimulate the transactivation potential of Sox4 (127). The PDZ domain-containing adaptor protein syntenin thereby recruits the Sox4 protein directly to the α chain of the IL-5 receptor (127). Sox4 therefore resembles the SMAD and STAT transcription factors with regard to its receptor-proximal activation. IL-5 is unlikely to be the physiological signal for Sox4 activation, as conventional B-lymphocytes develop normally in IL-5Rα −/− mice, whereas the peritoneal B-1 cells are moderately reduced (128). In addition to its IL-5Rα interaction, the adaptor molecule syntenin associates with the cytoplasmic domains of other transmembrane receptors, including syndecans (129) and ephrins (130). Hence, the signaling system that is responsible for syntenin-mediated activation of Sox4 in early B cell development remains to be identified.

19 Feb 2004

68

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Termination of Pre-BCR Signaling by IRF4 and IRF8 The early regulators E2A, EBF, and Pax5 set up the pre-BCR signaling system during pro-B cell development by controlling the synthesis of all pre-BCR components (λ5, VpreB, Igα, Igβ, and Igµ) and the expression of the central adaptor protein BLNK (see above). In contrast, pro-B cell development proceeds normally in the absence of both IRF4 (Pip) and IRF8 (ICSBP), which belong to the family of interferon regulatory factors (131). B-lymphopoiesis is, however, arrested at the pre-B cell transition in IRF4−/−IRF8−/− mice, as B cell precursors continue to cycle owing to persistent expression and unimpeded signaling of the pre-BCR (131). The double-mutant cells fail to rearrange immunoglobulin light-chain (IgL) genes, do not express the corresponding germline transcripts, and are unable to repress the surrogate light-chain genes λ5 and VpreB in response to pre-BCR signaling (131). Hence, IRF4 and IRF8 together appear to control the termination of preBCR signaling by downregulating λ5 and VpreB expression (131), which results in differentiation to small pre-B cells and initiation of IgL gene rearrangements (132).

EPIGENETIC CONTROL OF B CELL DEVELOPMENT Gene expression is determined not only by the available combination of transcription factors, but also by the structure of the local chromatin, which is the physiological substrate for all nuclear processes including transcription and recombination. The chromatin structure and its accessibility to regulatory factors thus contribute to cell-fate decisions by facilitating the establishment of tissuespecific gene expression patterns that are then epigenetically maintained during multiple rounds of DNA replication and cell division. Below, I discuss two aspects of the rapidly evolving field of epigenetic research: the developmental changes of the chromatin structure at the IgH locus and the role of Ikaros in the epigenetic control of lymphopoiesis.

Developmental Regulation of the Chromatin Structure at the IgH Locus The histone tails on the nucleosome surface are subject to enzyme-catalyzed, posttranslational modifications that, singly or in combination, form a “code” specifying chromatin accessibility and gene expression patterns. These modifications, which include lysine acetylation, lysine and arginine methylation, as well as serine phosphorylation, are not only a means of reorganizing nucleosome structure, but also provide a rich source of epigenetic information (“histone code”), which is read by nonhistone proteins that bind in a modification-sensitive manner to the N-terminal histone tails (133, 134). Histone acetylation is a characteristic feature of open, transcriptionally competent chromatin and is determined by the opposing activities of histone acetyltransferases (HATs) and deacetylases (HDACs). In contrast, the more permanent methylation of specific lysine residues (in particular lysine 9

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTROL OF EARLY B-LYMPHOPOIESIS

P1: IKH

69

of histone H3) has been associated with repressive chromatin and is introduced by SET domain-containing histone methyltransferases (HMTases) (133). These histone-modifying enzymes are components of large coactivator (HATs) or corepressor (HDACs and HMTases) complexes that are recruited to target genes by specific transcription factors (133, 135, 136). V(D)J recombination of immunoglobulin and T cell receptor (TCR) genes is developmentally controlled by enhancers that regulate the differential accessibility and germline transcription of specific gene segments prior to rearrangement (137). As predicted by the accessibility hypothesis, V(D)J rearrangement must be preceded by the acquisition of an open chromatin structure, which allows RAG proteins to recognize and cleave DNA at the recombination signal sequences (RSS) (137). Analysis of the TCRα/δ locus has unequivocally demonstrated that enhancer activity imparts developmentally regulated, long-range changes in acetylation of histone H3 and that the H3 acetylation status is tightly linked to V(D)J recombination (138). During B cell development, the IgH locus is activated in discrete, independently regulated domains at the chromatin level. A 120-kb region encompassing the D, J, and Cµ gene segments is first hyperacetylated at histones H3 and H4 in early pro-B cells (139). Following DH-JH recombination, the chromatin of the JH-proximal VH genes also becomes acetylated, whereas IL-7 signaling is required for acetylation of the JH-distal VHJ558 genes (139). Interestingly, acetylation is narrowly confined to the VH gene segments, their promoters, and RSS elements, which indicates local recruitment of HAT complexes by DNA sequence-specific regulators controlling VH gene transcription (140). As IL-7Rα expression is downregulated in response to pre-BCR signaling (101), the unrearranged VHJ558 genes revert to a less accessible, hypoacetylated chromatin state in pre-B cells (141). Consequently, the loss of IL-7-mediated hyperacetylation seems to contribute to allelic exclusion of the IgH locus in small pre-B cells (141). Surprisingly, histone acetylation of VH genes is not sufficient to facilitate VHDJH recombination, as the acetylated distal VHJ558 genes are rearranged with only low efficiency in pro-B cells lacking the HMTase Ezh2 (142). The Ezh2 protein as part of a large Polycomb complex methylates lysine 27 of histone H3, although the specific regions characterized by an increase in this methylation mark have not yet been mapped within the VH gene cluster of pro-B cells (142). Hence, the rearrangement of distal VHJ558 genes critically depends on chromatin changes that are induced by both histone acetylation as well as methylation.

Epigenetic Control of Lymphopoiesis by Ikaros Proteins Although Ikaros is a key regulator of lymphoid cell fates (Figure 2a), it does not function as a conventional transcription factor, but instead is the prototype of a novel class of chromatin regulators (24, 143). The nuclear periphery and pericentromeric heterochromatin are the two major transcriptionally repressive compartments of the nucleus that are important for propagating the inactive state of genes in hematopoietic cells (41, 144). While the IgH loci are preferentially located at the nuclear periphery in nonexpressing cells (144), other transcriptionally

19 Feb 2004

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

70

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER

silent genes of B- and T-lymphocytes are associated with pericentromeric heterochromatin (41, 145), which is enriched in modifiers of repressive higher-order chromatin (133, 134). In cycling cells, the Ikaros protein colocalizes with silent genes at pericentromeric regions and is therefore thought to play an essential role in heritable gene silencing by directly recruiting repressed target genes to pericentromeric foci (41, 143, 145). Consistent with this hypothesis, the Ikaros protein is present in multimeric complexes (146) that may bind simultaneously to target genes and pericentromeric repeat sequences (42). In quiescent cells, the Ikaros protein is diffusely distributed throughout the nucleus (145, 147), and silent genes are not associated with pericentromeric heterochromatin (145). Mitogenic stimulation of resting lymphocytes leads to a rapid increase and accumulation of Ikaros at pericentromeric clusters (145, 147) and concomitant repositioning of inactive genes to these foci (145). Consistent with a role for Ikaros in gene silencing is the fact that Ikaros sets the threshold for B and T cell activation and thus regulates the proliferative expansion of mature lymphocytes in response to antigen receptor engagement (33, 147). A decrease or loss of Ikaros expression leads to T cell hyperproliferation and lymphomagenesis, indicating that Ikaros functions as a tumor suppressor by controlling cell cycle entry, DNA replication, and stable chromosome propagation (30, 147). Ikaros also has the potential to activate genes, although it is unable to stimulate transcription by itself (148). Instead, Ikaros is an integral component of two higher-order chromatin remodeling complexes, which function to promote a dynamic equilibrium between open and closed chromatin states (38, 149). The more abundant Ikaros-NuRD complex contains 10 to 12 Ikaros molecules together with HDAC1, HDAC2, and the ATPase Mi-2β; is active in chromatin remodeling as well as histone deacetylation; and is recruited to pericentromeric heterochromatin in activated lymphocytes (38). In contrast, the less abundant Ikaros-SWI/SNF complex is excluded from heterochromatin (38) and may contain the Sin3 corepressor together with its associated HDAC activities (37, 150). A role for Ikaros in genespecific targeting of chromatin remodeling machines has recently been suggested for the regulation of CD8α. In immature T cells, Ikaros functions as an activator of the CD8α gene by binding to upstream regulatory sequences, where the Ikarosassociated remodeling activity is thought to open up the local chromatin, thus enabling transcriptional activators to access their sites (40). In conclusion, Ikaros is an essential component of functionally diverse chromatin-remodeling networks.

EPILOGUE To date, we may not yet know all regulators of the transcriptional network controlling early B cell development. For instance, a novel zinc-finger transcription factor, Bcl11a (Evi9), was recently shown to play an essential role in early B-lymphopoiesis (151). Most of the known transcription factors are also involved in the development of other organs, leading to embryonic or postnatal lethality of the corresponding mutant mice. Conditional gene inactivation will be required for

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

CONTROL OF EARLY B-LYMPHOPOIESIS

P1: IKH

71

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

a more detailed analysis of the B-lymphoid function of these regulators. Moreover, the critical target genes causing the early B cell developmental arrest in mutant mice still need to be identified for most transcription factors. In addition, the characterization of B-cell-specific enhancers and associated DNA-binding factors of individual transcription factor genes will identify important upstream regulators, which in turn will provide further insight into the regulatory network of early B cell development. Finally, there is an urgent need to learn more about the epigenetic control of B-lymphopoiesis. ACKNOWLEDGMENTS I thank C. Bonifer, A. Rolink, L. Klein, J. Skok, and T. Jenuwein for critical reading of the manuscript and members of the Busslinger laboratory for helpful discussions. This work was supported by Boehringer Ingelheim and in part by the Austrian Industrial Research Promotion Fund. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Kondo M, Weissman IL, Akashi K. 1997. Identification of clonogenic common lymphoid progenitors in mouse bone marrow. Cell 91:661–72 2. Akashi K, Traver D, Miyamoto T, Weissman IL. 2000. A clonogenic common myeloid progenitor that gives rise to all myeloid lineages. Nature 404:193–97 3. Adolfsson J, Borge OJ, Bryder D, Theil˚ gaard-M¨onch K, Astrand-Grundstr¨ om I, et al. 2001. Upregulation of Flt3 expression within the bone marrow Lin−Sca1+c-kit+ stem cell compartment is accompanied by loss of self-renewal capacity. Immunity 15:659–69 4. Christensen JL, Weissman IL. 2001. Flk-2 is a marker in hematopoietic stem cell differentiation: A simple method to isolate long-term stem cells. Proc. Natl. Acad. Sci. USA 98:14541–46 5. Igarashi H, Gregory SC, Yokota T, Sakaguchi N, Kincade PW. 2002. Transcription from the RAG1 locus marks the earliest lymphocyte progenitors in bone marrow. Immunity 17:117–30 6. Allman D, Sambandam A, Kim S, Miller

7.

8.

9.

10.

11.

JP, Pagan A, et al. 2003. Thymopoiesis independent of common lymphoid progenitors. Nat. Immunol. 4:168–74 Traver D, Akashi K, Manz M, Merad M, Miyamoto T, et al. 2000. Development of CD8α-positive dendritic cells from a common myeloid progenitor. Science 290:2152–54 Gounari F, Aifantis I, Martin C, Fehling HJ, Hoeflinger S, et al. 2002. Tracing lymphopoiesis with the aid of a pTαcontrolled reporter gene. Nat. Immunol. 3:489–96 Martin CH, Aifantis I, Scimone ML, von Boehmer H, Gounari F. 2003. Efficient thymic immigration of B220+ lymphoidrestricted bone marrow cells with T precursor potential. Nat. Immunol. 4:866– 73 Li Y-S, Wasserman R, Hayakawa K, Hardy RR. 1996. Identification of the earliest B lineage stage in mouse bone marrow. Immunity 5:527–35 Tudor KS, Payne KJ, Yamashita Y, Kincade PW. 2000. Functional assessment of precursors from murine bone marrow

19 Feb 2004

11:59

72

12.

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

13.

14.

15.

16.

17.

18.

19.

20.

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER suggests a sequence of early B lineage differentiation events. Immunity 12:335– 45 Hardy RR, Carmack CE, Shinton SA, Kemp JD, Kayakawa K. 1991. Resolution and characterization of pro-B and pre-pro-B cell stages in normal mouse bone marrow. J. Exp. Med. 173:1213–25 Li Y-S, Hayakawa K, Hardy RR. 1993. The regulated expression of B lineageassociated genes during B cell differentiation in bone marrow and fetal liver. J. Exp. Med. 178:951–60 Meffre E, Casellas R, Nussenzweig MC. 2000. Antibody regulation of B cell development. Nat. Immunol. 1:379–85 Domen J, Weissman IL. 2000. Hematopoietic stem cells need two signals to prevent apoptosis; BCL-2 can provide one of these, Kitl/c-Kit signaling the other. J. Exp. Med. 192:1707–18 Waskow C, Paul S, Haller C, Gassmann M, Rodewald H. 2002. Viable c-KitW/W mutants reveal pivotal role for c-Kit in the maintenance of lymphopoiesis. Immunity 17:277–88 McKenna HJ, Stocking KL, Miller RE, Brasel K, De Smedt T, et al. 2000. Mice lacking flt3 ligand have deficient hematopoiesis affecting hematopoietic progenitor cells, dendritic cells, and natural killer cells. Blood 95:3489–97 Sitnicka E, Bryder D, Theilgaard-M¨onch K, Buza-Vidas N, Adolfsson J, Jacobsen SEW. 2002. Key role of flt3 ligand in regulation of the common lymphoid progenitor but not in maintenance of the hematopoietic stem cell pool. Immunity 17:463–72 Miller JP, Izon D, DeMuth W, Gerstein R, Bhandoola A, Allman D. 2002. The earliest step in B lineage differentiation from common lymphoid progenitors is critically dependent upon interleukin 7. J. Exp. Med. 196:705–11 Carvalho TL, Mota-Santos T, Cumano A, Demengeot J, Vieira P. 2001. Arrested B lymphopoiesis and persistence of ac-

21.

22.

23.

24.

25.

26.

27.

28.

29.

tivated B cells in adult interleukin 7−/− mice. J. Exp. Med. 194:1141–50 Vosshenrich CAJ, Cumano A, M¨uller W, Di Santo JP, Vieira P. 2003. Thymic stroma-derived lymphopoietin distinguishes fetal from adult B cell development. Nat. Immunol. 4:773–79 Mackarehtschian K, Hardin JD, Moore KA, Boast S, Goff SP, Lemischka IR. 1995. Targeted disruption of the flk2/flt3 gene leads to deficiencies in primitive hematopoietic progenitors. Immunity 3:147–61 Schebesta M, Heavey B, Busslinger M. 2002. Transcriptional control of B cell development. Curr. Opin. Immunol. 14:216–23 Georgopoulos K. 2002. Haematopoietic cell-fate decisions, chromatin regulation and Ikaros. Nat. Rev. Immunol. 2:162– 74 Morgan B, Sun L, Avitahl N, Andrikopoulos K, Ikeda T, et al. 1997. Aiolos, a lymphoid restricted transcription factor that interacts with Ikaros to regulate lymphocyte differentiation. EMBO J. 16:2004–13 Kelley CM, Ikeda T, Koipally J, Avitahl N, Wu L, et al. 1998. Helios, a novel dimerization partner of Ikaros expressed in the earliest hematopoietic progenitors. Curr. Biol. 8:508–15 Klug CA, Morrison SJ, Masek M, Hahm K, Smale ST, Weissman IL. 1998. Hematopoietic stem cells and lymphoid progenitors express different Ikaros isoforms, and Ikaros is localized to heterochromatin in immature lymphocytes. Proc. Natl. Acad. Sci. USA 95:657–62 Sun L, Liu A, Georgopoulos K. 1996. Zinc finger-mediated protein interactions modulate Ikaros activity, a molecular control of lymphocyte development. EMBO J. 15:5358–69 Georgopoulos K, Bigby M, Wang J-H, Molnar A, Wu P, et al. 1994. The Ikaros gene is required for the development of all lymphoid lineages. Cell 79:143–56

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTROL OF EARLY B-LYMPHOPOIESIS 30. Wang J-H, Nichogiannopoulou A, Wu L, Sun L, Sharpe AH, et al. 1996. Selective defects in the development of the fetal and adult lymphoid system in mice with an Ikaros null mutation. Immunity 5: 537–49 31. Wu L, Nichogiannopoulou A, Shortman K, Georgopoulos K. 1997. Cellautonomous defects in dendritic cell populations of Ikaros mutant mice point to a developmental relationship with the lymphoid lineage. Immunity 7:483–92 32. Nichogiannopoulou A, Trevisan M, Neben S, Friedrich C, Georgopoulos K. 1999. Defects in hematopoietic stem cell activity in Ikaros mutant mice. J. Exp. Med. 190:1201–13 33. Kirstetter P, Thomas M, Dierich A, Kastner P, Chan S. 2002. Ikaros is critical for B cell differentiation and function. Eur. J. Immunol. 32:720–30 34. Wang JH, Avitahl N, Cariappa A, Friedrich C, Ikeda T, et al. 1998. Aiolos regulates B cell activation and maturation to effector state. Immunity 9:543– 53 35. Lopez RA, Schoetz S, DeAngelis K, O’Neill D, Bank A. 2002. Multiple hematopoietic defects and delayed globin switching in Ikaros null mice. Proc. Natl. Acad. Sci. USA 99:602–7 36. Dumortier A, Kirstetter P, Kastner P, Chan S. 2003. Ikaros regulates neutrophil differentiation. Blood 101:2219– 26 37. Koipally J, Renold A, Kim J, Georgopoulos K. 1999. Repression by Ikaros and Aiolos is mediated through histone deacetylase complexes. EMBO J. 18: 3090–100 38. Kim J, Sif S, Jones B, Jackson A, Koipally J, et al. 1999. Ikaros DNAbinding proteins direct formation of chromatin remodeling complexes in lymphocytes. Immunity 10:345–55 39. Koipally J, Georgopoulos K. 2000. Ikaros interactions with CtBP reveal a repression mechanism that is independent

40.

41.

42.

43.

44.

45.

46.

47.

48.

P1: IKH

73

of histone deacetylase activity. J. Biol. Chem. 275:19594–602 Harker N, Naito T, Cortes M, Hostert A, Hirschberg S, et al. 2002. The CD8α gene locus is regulated by the Ikaros family of proteins. Mol. Cell 10:1403– 15 Brown KE, Guest SS, Smale ST, Hahm K, Merkenschlager M, Fisher AG. 1997. Association of transcriptionally silent genes with Ikaros complexes at centromeric heterochromatin. Cell 91:845– 54 Cobb BS, Morales-Alcelay S, Kleiger G, Brown KE, Fisher AG, Smale ST. 2000. Targeting of Ikaros to pericentromeric heterochromatin by direct DNA binding. Genes Dev. 14:2146–60 Scott EW, Simon MC, Anastasi J, Singh H. 1994. Requirement of transcription factor PU.1 in the development of multiple hematopoietic lineages. Science 265:1573–77 McKercher SR, Torbett BE, Anderson KL, Henkel GW, Vestal DJ, et al. 1996. Targeted disruption of the PU.1 gene results in multiple hematopoietic abnormalities. EMBO J. 15:5647–6658 Hromas R, Orazi A, Neiman R, Maki R, Van Beveran C, et al. 1993. Hematopoietic lineage- and stage-restricted expression of the ETS oncogen family member PU.1. Blood 82:2998–3004 Anderson MK, Hernandez-Hoyos G, Diamond RA, Rothenberg EV. 1999. Precise developmental regulation of Ets family transcription factors during specification and commitment to the T cell lineage. Development 126:3131– 48 Scott EW, Fisher RC, Olson MC, Kehrli EW, Simon MC, Singh H. 1997. PU.1 functions in a cell-autonomous manner to control the differentiation of multipotential lymphoid-myeloid progenitors. Immunity 6:437–47 Tondravi MM, McKercher SR, Anderson K, Erdmann JM, Quiroz M,

19 Feb 2004

11:59

74

49.

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

50.

51.

52.

53.

54.

55.

56.

57.

58.

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER et al. 1997. Osteopetrosis in mice lacking haematopoeitic transcription factor PU.1. Nature 386:81–84 Walsh JC, DeKoter RP, Lee HJ, Smith ED, Lancki DW, et al. 2002. Cooperative and antagonistic interplay between PU.1 and GATA-2 in the specification of myeloid cell fates. Immunity 17:665–76 DeKoter RP, Walsh JC, Singh H. 1998. PU.1 regulates both cytokine-dependent proliferation and differentiation of granulocyte/macrophage progenitors. EMBO J. 17:4456–68 Colucci F, Samson SI, DeKoter RP, Lantz O, Singh H, Di Santo JP. 2001. Differential requirement for the transcription factor PU.1 in the generation of natural killer cells versus B and T cells. Blood 97:2625–32 Spain LM, Guerriero A, Kunjibettu S, Scott EW. 1999. T cell development in PU.1-deficient mice. J. Immunol. 163: 2681–87 DeKoter RP, Lee H-J, Singh H. 2002. PU.1 regulates expression of the interleukin-7 receptor in lymphoid progenitors. Immunity 16:297–309 DeKoter RP, Singh H. 2000. Regulation of B lymphocyte and macrophage development by graded expression of PU.1. Science 288:1439–41 Murre C, McCaw PS, Baltimore D. 1989. A new DNA binding and dimerization motif in immunoglobulin enhancer binding, daughterless, MyoD, and myc proteins. Cell 56:777–83 Shen CP, Kadesch T. 1995. B-cellspecific DNA binding by an E47 homodimer. Mol. Cell. Biol. 15:4518–24 Sloan SR, Shen C-P, McCarrickWalmsley R, Kadesch T. 1996. Phosphorylation of E47 as a potential determinant of B-cell-specific activity. Mol. Cell. Biol. 16:6900–8 Zhuang Y, Barndt RJ, Pan L, Kelley R, Dai M. 1998. Functional replacement of the mouse E2A gene with a human HEB cDNA. Mol. Cell. Biol. 18:3340–49

59. Bain G, Maandag ECR, Izon DJ, Amsen D, Kruisbeek AM, et al. 1994. E2A proteins are required for proper B cell development and initiation of immunoglobulin gene rearrangements. Cell 79:885–92 60. Zhuang Y, Soriano P, Weintraub H. 1994. The helix-loop-helix gene E2A is required for B cell formation. Cell 79:875– 84 61. Bain G, Maandag ECR, te Riele HPJ, Feeney AJ, Sheehy A, et al. 1997. Both E12 and E47 allow commitment to the B cell lineage. Immunity 6:145–54 62. Zhuang Y, Cheng P, Weintraub H. 1996. B-lymphocyte development is regulated by the combined dosage of three basic helix-loop-helix genes, E2A, E2-2 and HEB. Mol. Cell. Biol. 16:2898–905 63. Hagman J, Belanger C, Travis A, Turck CW, Grosschedl R. 1993. Cloning and functional characterization of early Bcell factor, a regulator of lymphocytespecific gene expression. Genes Dev. 7: 760–73 64. Hagman J, Gutch MJ, Lin H, Grosschedl R. 1995. EBF contains a novel zinc coordination motif and multiple dimerization and transcriptional activation domains. EMBO J. 14:2907–16 65. Lin H, Grosschedl R. 1995. Failure of B-cell differentiation in mice lacking the transcription factor EBF. Nature 376:263–67 66. O’Riordan M, Grosschedl R. 1999. Coordinate regulation of B cell differentiation by the transcription factors EBF and E2A. Immunity 11:21–31 67. Schlissel M, Voronova A, Baltimore D. 1991. Helix-loop-helix transcription factor E47 activates germ-line immunoglobulin heavy-chain gene transcription and rearrangement in a pre-T cell line. Genes Dev. 5:1367–76 68. Choi JK, Shen C-P, Radomska HS, Eckhardt LA, Kadesch T. 1996. E47 activates the Ig-heavy chain and TdT loci in non-B cells. EMBO J. 15:5014–21

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTROL OF EARLY B-LYMPHOPOIESIS 69. Kee BL, Murre C. 1998. Induction of early B cell factor (EBF) and multiple B lineage genes by the basic helix-loophelix transcription factor E12. J. Exp. Med. 188:699–713 70. Sigvardsson M, O’Riordan M, Grosschedl R. 1997. EBF and E47 collaborate to induce expression of the endogenous immunoglobulin surrogate light chain genes. Immunity 7:25–36 71. Sigvardsson M. 2000. Overlapping expression of early B-cell factor and basic helix-loop-helix proteins as a mechanism to dictate B-lineage-specific activity of the λ5 promoter. Mol. Cell. Biol. 20:3640–54 72. Gisler R, Sigvardsson M. 2002. The human V-preB promoter is a target for coordinated activation by early B cell factor and E47. J. Immunol. 168:5130– 38 73. Sigvardsson M, Clark DR, Fitzsimmons D, Doyle M, Akerblad P, et al. 2002. Early B-cell factor, E2A, and Pax-5 cooperate to activate the early B-cellspecific mb-1 promoter. Mol. Cell. Biol. 22:8539–51 74. Akerblad P, Rosberg M, Leanderson T, Sigvardsson M. 1999. The B29 (immunoglobulin β-chain) gene is a genetic target for early B-cell factor. Mol. Cell. Biol. 19:392–401 75. Greenbaum S, Zhuang Y. 2002. Identification of E2A target genes in B lymphocyte development by using a gene tagging-based chromatin immunoprecipitation system. Proc. Natl. Acad. Sci. USA 99:15030–35 76. Benezra R, Davis RL, Lockshon D, Turner DL, Weintraub H. 1990. The protein Id: a negative regulator of helixloop-helix DNA-binding proteins. Cell 61:49–59 77. Sun X-H. 1994. Constitutive expression of the Id1 gene impairs mouse B cell development. Cell 79:893–900 78. Kee BL, Rivera RR, Murre C. 2001. Id3 inhibits B lymphocyte progenitor growth

79.

80.

81.

82.

83.

84.

85.

86.

P1: IKH

75

and survival in response to TGF-β. Nat. Immunol. 2:242–47 Kee BL, Bain G, Murre C. 2002. IL7Rα and E47: independent pathways required for development of multipotent lymphoid progenitors. EMBO J. 21:103– 13 Romanow WJ, Langerak AW, Goebel P, Wolvers-Tettero ILM, van Dongen JJM, et al. 2000. E2A and EBF act in synergy with the V(D)J recombinase to generate a diverse immunoglobulin repertoire in nonlymphoid cells. Mol. Cell 5:343– 53 Goebel P, Janney N, Valenzuela JR, Romanow WJ, Murre C, Feeney AJ. 2001. Localized gene-specific induction of accessibility to V(D)J recombination induced by E2A and early B cell factor in nonlymphoid cells. J. Exp. Med. 194:645–56 Serwe M, Sablitzky F. 1993. V(D)J recombination in B cells is impaired but not blocked by targeted deletion of the immunoglobulin heavy chain intron enhancer. EMBO J. 12:2321–27 Chen J, Young F, Bottaro A, Stewart V, Smith RK, Alt FW. 1993. Mutations of the intronic IgH enhancer and its flanking sequences differentially affect accessibility of the JH locus. EMBO J. 12:4635– 45 Massari ME, Grant PA, Pray-Grant MG, Berger SL, Workman JL, Murre C. 1999. A conserved motif present in a class of helix-loop-helix proteins activates transcription by direct recruitment of the SAGA complex. Mol. Cell. 4:63–73 Smith EM, Gisler R, Sigvardsson M. 2002. Cloning and characterization of a promoter flanking the early B cell factor (EBF) gene indicates roles for E-proteins and autoregulation in the control of EBF expression. J. Immunol. 169:261–70 Adams B, D¨orfler P, Aguzzi A, Kozmik Z, Urb´anek P, et al. 1992. Pax-5 encodes the transcription factor BSAP and is expressed in B lymphocytes, the

19 Feb 2004

11:59

76

87.

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

88.

89.

90.

91.

92.

93.

94.

95.

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER developing CNS, and adult testis. Genes Dev. 6:1589–607 Urb´anek P, Wang Z-Q, Fetka I, Wagner EF, Busslinger M. 1994. Complete block of early B cell differentiation and altered patterning of the posterior midbrain in mice lacking Pax5/BSAP. Cell 79:901– 12 Nutt SL, Urb´anek P, Rolink A, Busslinger M. 1997. Essential functions of Pax5 (BSAP) in pro-B cell development: difference between fetal and adult B lymphopoiesis and reduced V-to-DJ recombination at the IgH locus. Genes Dev. 11:476–91 Nutt SL, Heavey B, Rolink AG, Busslinger M. 1999. Commitment to the Blymphoid lineage depends on the transcription factor Pax5. Nature 401:556– 62 Rolink AG, Nutt SL, Melchers F, Busslinger M. 1999. Long-term in vivo reconstitution of T-cell development by Pax5-deficient B-cell progenitors. Nature 401:603–6 Schaniel C, Bruno L, Melchers F, Rolink AG. 2002. Multiple hematopoietic cell lineages develop in vivo from transplanted Pax5-deficient pre-B I-cell clones. Blood 99:472–78 Schaniel C, Gottar M, Roosnek E, Melchers F, Rolink AG. 2002. Extensive in vivo self-renewal, long-term reconstitution capacity, and hematopoietic multipotency of Pax5-deficient precursor B-cell clones. Blood 99:2760–66 Heavey B, Charalambous C, Cobaleda C, Busslinger M. 2003. Myeloid lineage switch of Pax5 mutant but not wild-type B cell progenitors by C/EBPα and GATA factors. EMBO J. 22:3887– 97 Horcher M, Souabni A, Busslinger M. 2001. Pax5/BSAP maintains the identity of B cells in late B lymphopoiesis. Immunity 14:779–90 Mikkola I, Heavey B, Horcher M, Busslinger M. 2002. Reversion of B cell

96.

97.

98.

99.

100.

101.

102.

103.

104.

commitment upon loss of Pax5 expression. Science 297:110–13 Hu M, Krause D, Greaves M, Sharkis S, Dexter M, et al. 1997. Multilineage gene expression precedes commitment in the hematopoietic system. Genes Dev. 11:774–85 Nutt SL, Morrison AM, D¨orfler P, Rolink A, Busslinger M. 1998. Identification of BSAP (Pax-5) target genes in early B-cell development by loss- and gain-of-function experiments. EMBO J. 17:2319–33 Souabni A, Cobaleda C, Schebesta M, Busslinger M. 2002. Pax5 promotes B lymphopoiesis and blocks T cell development by repressing Notch1. Immunity 17:781–93 Fitzsimmons D, Hodsdon W, Wheat W, Maira S-M, Wasylyk B, Hagman J. 1996. Pax-5 (BSAP) recruits Ets protooncogene family proteins to form functional ternary complexes on a B cell– specific promoter. Genes Dev. 10:2198– 211 Kozmik Z, Wang S, D¨orfler P, Adams B, Busslinger M. 1992. The promoter of the CD19 gene is a target for the B-cellspecific transcription factor BSAP. Mol. Cell. Biol. 12:2662–72 Schebesta M, Pfeffer PL, Busslinger M. 2002. Control of pre-BCR signaling by Pax5-dependent activation of the BLNK gene. Immunity 17:473–85 Hayashi K, Yamamoto M, Nojima T, Goitsuka R, Kitamura D. 2003. Distinct signaling requirements for Dµ selection, IgH allelic exclusion, pre-B cell transition, and tumor suppression of B cell progenitors. Immunity 18:825–36 Hesslein DGT, Pflugh DL, Chowdhury D, Bothwell ALM, Sen R, Schatz DG. 2003. Pax5 is required for recombination of transcribed, acetylated, 50 IgH V gene segments. Genes Dev. 17:37–42 Eberhard D, Jim´enez G, Heavey B, Busslinger M. 2000. Transcriptional repression by Pax5 (BSAP) through interaction

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

CONTROL OF EARLY B-LYMPHOPOIESIS

105.

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

106.

107.

108.

109.

110.

111.

112.

113.

with corepressors of the Groucho family. EMBO J. 19:2292–303 Linderson Y, Eberhard D, Malin S, Johansson A, Busslinger M, Pettersson S. 2003. Co-recruitment of the Grg4 repressor by PU.1 is critical for Pax5mediated repression of B-cell-specific genes. EMBO Rep. In press Garvie CW, Hagman J, Wolberger C. 2001. Structural studies of Ets-1/Pax5 complex formation on DNA. Mol. Cell 8:1267–76 Yokota Y, Mansouri A, Mori S, Sugawara S, Adachi S, et al. 1999. Development of peripheral lymphoid organs and natural killer cells depends on the helix-loophelix inhibitor Id2. Nature 397:702–6 Ikawa T, Fujimoto S, Kawamoto H, Katsura Y, Yokota Y. 2001. Commitment to natural killer cells requires the helix-loop-helix inhibitor Id2. Proc. Natl. Acad. Sci. USA 98:5164–69 Hacker C, Kirsch RD, Ju X-S, Hieronymus T, Gust TC, et al. 2003. Transcriptional profiling identifies Id2 function in dendritic cell development. Nat. Immunol. 4:380–86 Kusunoki T, Sugai M, Katakai T, Omatsu Y, Iyoda T, et al. 2003. TH2 dominance and defective development of a CD8+ dendritic cell subset in Id2-deficient mice. J. Allergy Clin. Immunol. 111:136– 42 Jaleco AC, Stegmann APA, Heemskerk MHM, Couwenberg F, Bakker AQ, et al. 1999. Genetic modification of human Bcell development: B-cell development is inhibited by the dominant negative helix loop helix factor Id3. Blood 94:2637–46 Spits H, Couwenberg F, Bakker AQ, Weijer K, Uttenbogaart CH. 2000. Id2 and Id3 inhibit development of CD34+ stem cells into predendritic cell (preDC)2 but not into pre-DC1: Evidence for a lymphoid origin of pre-DC2. J. Exp. Med. 192:1775–84 Izon DJ, Punt JA, Pear WS. 2002. Deciphering the role of Notch signaling

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

P1: IKH

77

in lymphopoiesis. Curr. Opin. Immunol. 14:192–99 Pui JC, Allman D, Xu L, DeRocco S, Karnell FG, et al. 1999. Notch1 expression in early lymphopoiesis influences B versus T lineage determination. Immunity 11:299–308 Jaleco AC, Neves H, Hooijberg E, Gameiro P, Clode N, et al. 2001. Differential effects of Notch ligands Delta-1 and Jagged-1 in human lymphoid differentiation. J. Exp. Med. 194:991–1001 Schmitt TM, Z´un˜ iga-Pfl¨ucker JC. 2002. Induction of T cell development from hematopoietic progenitor cells by Deltalike-1 in vitro. Immunity 17:749–56 Radtke F, Wilson A, Stark G, Bauer M, van Meerwijk J, et al. 1999. Deficient T cell fate specification in mice with an induced inactivation of Notch1. Immunity 10:547–58 Wilson A, MacDonald HR, Radtke F. 2001. Notch 1-deficient common lymphoid precursors adopt a B cell fate in the thymus. J. Exp. Med. 194:1003–12 Ordentlich P, Lin A, Shen C-P, Blaumueller C, Matsuno K, et al. 1998. Notch inhibition of E47 supports the existence of a novel signaling pathway. Mol. Cell. Biol. 18:2230–39 Cotta CV, Zhang Z, Kim H-G, Klug CA. 2003. Pax5 determines B- versus T-cell fate and does not block early myeloidlineage development. Blood 101:4342– 46 van Freeden-Jeffry U, Vieira P, Lucian LA, McNeil T, Burdach SEG, Murray R. 1995. Lymphopenia in interleukin (IL)7 gene-deleted mice identifies IL-7 as a nonredundant cytokine. J. Exp. Med. 181:1519–26 Peschon JJ, Morrissey PJ, Grabstein KH, Ramsdell FJ, Maraskovsky E, et al. 1994. Early lymphocyte expansion is severely impaired in interleukin 7 receptordeficient mice. J. Exp. Med. 180: 1955–60 Reya T, O’Riordan M, Okamura R,

19 Feb 2004

11:59

78

124.

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

125.

126.

127.

128.

129.

130.

131.

132.

133.

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUSSLINGER Devaney E, Willert K, et al. 2000. Wnt signaling regulates B lymphocyte proliferation through a LEF-1 dependent mechanism. Immunity 13:15–24 Reya T, Duncan AW, Ailles L, Domen J, Scherer DC, et al. 2003. A role for Wnt signalling in self-renewal of haematopoietic stem cells. Nature 423:409–14 van de Wetering M, Oosterwegel M, van Norren K, Clevers H. 1993. Sox-4, an Sry-like HMG box protein, is a transcriptional activator in lymphocytes. EMBO J. 12:3847–54 Schilham MW, Oosterwegel MA, Moerer P, Ya J, de Boer PAJ, et al. 1996. Defects in cardiac outflow tract formation and pro-B-lymphocyte expansion in mice lacking Sox-4. Nature 380:711–14 Geijsen N, Uings IJ, Pals C, Armstrong J, McKinnon M, et al. 2001. Cytokine-specific transcriptional regulation through an IL-5Rα interacting protein. Science 293:1136–38 Yoshida T, Ikuta K, Sugaya H, Maki K, Takagi M, et al. 1996. Defective B-1 cell development and impaired immunity against Angiostrongylus cantonensis in IL-5Rα-deficient mice. Immunity 4:483–94 Grootjans JJ, Zimmermann P, Reekmans G, Smets A, Degeest G, et al. 1997. Syntenin, a PDZ protein that binds syndecan cytoplasmic domains. Proc. Natl. Acad. Sci. USA 94:13683–88 Lin D, Gish GD, Songyang Z, Pawson T. 1999. The carboxyl terminus of B class ephrins constitutes a PDZ domain binding motif. J. Biol. Chem. 274:3726–33 Lu R, Medina KL, Lancki DW, Singh H. 2003. IRF-4/8 orchestrate the pre-B-toB transition in lymphocyte development. Genes Dev. 17:1703–8 Wang YH, Stephan RP, Scheffold A, Kunkel D, Karasuyama H, et al. 2002. Differential surrogate light chain expression governs B-cell differentiation. Blood 99:2459–67 Jenuwein T, Allis CD. 2001. Translat-

134. 135.

136.

137.

138.

139.

140.

141.

142.

143.

ing the histone code. Science 293:1074– 1080 Turner BM. 2002. Cellular memory and the histone code. Cell 111:285–91 Ogawa H, Ishiguro K, Gaubatz S, Livingston DM, Nakatani Y. 2002. A complex with chromatin modifiers that occupies E2F- and Myc-responsive genes in GO cells. Science 296:1132–36 Schultz DC, Ayyanathan K, Negorev D, Maul GG, Rauscher FJ III. 2002. SETDB1: a novel KAP-1-associated histone H3, lysine 9-specific methyltransferase that contributes to HP1mediated silencing of euchromatic genes by KRAB zinc-finger proteins. Genes Dev. 16:919–32 Krangel MS. 2003. Gene segment selection in V(D)J recombination: accessibility and beyond. Nat. Immunol. 4:624–30 McMurry MT, Krangel MS. 2000. A role for histone acetylation in the developmental regulation of V(D)J recombination. Science 287:495–98 Chowdhury D, Sen R. 2001. Stepwise activation of the immunoglobulin µ heavy chain gene locus. EMBO J. 20:6394–403 Johnson K, Angelin-Duclos C, Park S, Calame KL. 2003. Changes in histone acetylation are associated with differences in accessibility of VH gene segments to V-DJ recombination during Bcell ontogeny and development. Mol. Cell. Biol. 23:2438–50 Chowdhury D, Sen R. 2003. Transient IL-7/IL-7R signaling provides a mechanism for feedback inhibition of immunoglobulin heavy chain gene rearrangements. Immunity 18:229–41 Su IH, Basavaraj A, Krutchinsky AN, Hobert O, Ullrich A, et al. 2003. Ezh2 controls B cell development through histone H3 methylation and Igh rearrangement. Nat. Immunol. 4:124–31 Smale ST. 2003. The establishment and maintenance of lymphocyte identity through gene silencing. Nat. Immunol. 4:607–15

19 Feb 2004

11:59

AR

AR210-IY22-03.tex

AR210-IY22-03.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTROL OF EARLY B-LYMPHOPOIESIS 144. Kosak ST, Skok JA, Medina KL, Riblet R, Le Beau MM, et al. 2002. Subnuclear compartmentalization of immunoglobulin loci during lymphocyte development. Science 296:158–62 145. Brown KE, Baxter J, Graf D, Merkenschlager M, Fisher AG. 1999. Dynamic repositioning of genes in the nucleus of lymphocytes preparing for cell division. Mol. Cell 3:207–17 146. Trinh LA, Ferrini R, Cobb BS, Weinmann AS, Hahm K, et al. 2001. Downregulation of TDT transcription in CD4+CD8+ thymocytes by Ikaros proteins in direct competition with an Ets activator. Genes Dev. 15:1817– 32 147. Avitahl N, Winandy S, Friedrich C, Jones B, Ge Y, Georgopoulos K. 1999. Ikaros sets thresholds for T cell activation and

148.

149.

150.

151.

P1: IKH

79

regulates chromosome propagation. Immunity 10:333–43 Koipally J, Heller EJ, Seavitt JR, Georgopoulos K. 2002. Unconventional potentiation of gene expression by Ikaros. J. Biol. Chem. 277:13007–15 O’Neill DW, Schoetz SS, Lopez RA, Castle M, Rabinowitz L, et al. 2000. An Ikaros-containing chromatin-remodeling complex in adult-type erythroid cells. Mol. Cell. Biol. 20:7572–82 Sif S, Saurin AJ, Imbalzano AN, Kingston RE. 2001. Purification and characterization of mSin3A-containing Brg1 and hBrm chromatin remodeling complexes. Genes Dev. 15:603–18 Liu P, Keller JR, Ortiz M, Tessarollo L, Rachel RA, et al. 2003. Bcl11a is essential for normal lymphoid development. Nat. Immunol. 4:525–32

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22 Annu. Rev. Immunol. 2004.22:55-79. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104813

Annu. Rev. Immunol. 2004. 22:81–127 doi: 10.1146/annurev.immunol.22.012703.104813 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on September 15, 2003

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE Yun-Cai Liu Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Division of Cell Biology, La Jolla Institute for Allergy and Immunology, San Diego, California 92121; email: [email protected]

Key Words ubiquitination, HECT, RING, lymphocyte, immunity ■ Abstract Ubiquitin (Ub)-protein conjugation represents a novel means of posttranscriptional modification in a proteolysis-dependent or -independent manner. E3 Ub ligases play a key role in governing the cascade of Ub transfer reactions by recognizing and catalyzing Ub conjugation to specific protein substrates. The E3s, which can be generally classified into HECT-type and RING-type families, are involved in the regulation of many aspects of the immune system, including the development, activation, and differentiation of lymphocytes, T cell–tolerance induction, antigen presentation, immune evasion, and virus budding. E3-promoted ubiquitination affects a wide array of biological processes, such as receptor downmodulation, signal transduction, protein processing or translocation, protein-protein interaction, and gene transcription, in addition to proteasome-mediated degradation. Deficiency or mutation of some of the E3s like Cbl, Cbl-b, or Itch, causes abnormal immune responses such as autoimmunity, malignancy, and inflammation. This review discusses our current understanding of E3 Ub ligases in both innate and adaptive immunity. Such knowledge may facilitate the development of novel therapeutic approaches for immunological diseases.

INTRODUCTION Highly relevant to the immune system, ubiquitin (Ub) was first discovered as a lymphocyte differentiation-promoting factor close to three decades ago (1). The small peptide of 76 amino acids is highly conserved during evolution and forms a compact globular structure with C-terminal glycine residues protruding from the main body of the protein. Ub conjugation to the protein substrate, or ubiquitination, involves a cascade of enzymatic reactions (reviewed in Reference 2): Ub is first activated by a Ub-activating enzyme or E1 and forms a highly active thioester bond between the C-terminus glycine residue of Ub and an active cysteine group of E1; the activated Ub can then be transferred to one of a family of Ub conjugating enzymes (Ubc) or E2s via a similar thioester linkage; an E3 Ub protein ligase is required for the covalent isopeptide bond formation between Ub and the εamino group of lysine residues in the substrate protein; E3s recruit both Ub-loaded E2s and the target protein, and they facilitate the transfer of Ub from E2 to the substrate and also govern the formation of a poly-Ub chain on the substrate via 0732-0582/04/0423-0081$14.00

81

11 Feb 2004

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

82

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

an isopeptide bond between the ε-amino group of lysine residues on adjacent Ub molecules. Thus, the E3s are the critical components that provide specificity to the Ub conjugation system by direct and specific interaction with the substrate (3, 4). It is becoming increasingly clear that protein ubiquitination is involved in a wide array of cellular processes, including cell-cycle control, signal transduction, transcriptional regulation, DNA repair, receptor downregulation, antigen presentation, and apoptosis (2). Abnormalities in the Ub system have been shown to cause pathological responses, including malignant transformation, and several genetic diseases (5). Ub tagging to intracellular substrates serves as a marker that can be recognized by the 26S proteasome for degradation to remove excessive cellular responses (2), whereas Ub conjugation to cell surface receptors causes their downmodulation through the endosomal-lysosomal pathway (6). Recently, it has been appreciated that protein ubiquitination represents an important means of posttranscriptional modification in a proteolysis-independent manner, and like protein phosphorylation on tyrosine or serine/threonine residues of signaling molecules, Ub conjugation has been manifested in Ub-dependent protein processing, proteinprotein interaction, subcellular translocation, kinase activation, and transcriptional activation, as is discussed in this review. Most of the multicellular organisms are capable of defending themselves from infectious intruders by mounting the immune response through a mechanism of innate immunity (7). The innate immune response is induced by the binding of the pattern-recognition receptors in the host to the pathogen-derived substances. Triggering of these germline-encoded receptors allows the host to destroy the invading microbes as well as to initiate intracellular signaling events leading to the gene transcription of antimicrobial molecules or proinflammatory cytokines. In higher eukaryotes, microbial infection can also mount adaptive immune response through somatic DNA rearrangement to produce pathogen-specific receptors such as the B cell antigen receptor (BCR) on B cells and the T cell antigen receptor (TCR) on T cells or produce pathogenic antigen-specific antibodies. The innate immunity, as the initial defense against the pathogen, can modulate adaptive immune response by providing processed antigens or by inducing the expression of costimulatory molecules to the T cells (8). Evidence accumulated over the past decade clearly indicates that E3 Ub ligases and/or proteasomal-dependent degradation are involved in the regulation of both innate and adaptive immune responses. The intracellular signaling pathways that are activated by receptor triggering of both innate and adaptive immune systems share some similarities, such as the activation of nuclear factor-κB (NF-κB). For example, it is well known that the lipopolysaccharide of Gram-negative bacteria or the proinflammatory cytokine, interleukin-1 (IL-1), induces the activation of NFκB via a critical step of ubiquitination and subsequent degradation of the inhibitor of kB (IκB) (9). More recent data suggest that E3 ligases functioning in IL-1- or lipopolysaccharide-induced signaling events are also involved in the regulation of antigen receptor-mediated signaling pathways (10). Another well-known example of the Ub system in immune regulation is the proteasomal-dependent processing

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

83

of antigenic peptides in antigen-presenting cells (11). Given that the list of newly identified E3 Ub ligases emerges so rapidly, it is almost impossible to discuss each and every E3 ligase and its potential roles in the immune regulation. In this review, I first give a description of the Ub conjugation system and then introduce several examples of E3s that have been directly or indirectly implicated in the immune response, in the hope of casting a brick to attract jade.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE UB CONJUGATION SYSTEM The E1 and E2s Ub conjugation is initiated by the activation of Ub at the C-terminal glycine residue in an ATP-dependent manner by the E1 Ub activating enzyme. A single E1 enzyme in humans and the yeast is required for the catalysis of the Ub conjugation cascade (2). E1 is essential for cell viability because deletion of E1 is lethal in yeast (12). A temperature-sensitive inactivation of E1 in mammalian cells causes reduced antigen presentation (13), an early demonstration that the Ub pathway is involved in the immune system. In contrast to E1, there are 13 isoforms of E2s in yeast and more than 20 E2s in humans. All E2s contain a core structure of ∼150 amino acids, called the Ubc domain. In the core structure, there are interfaces for the interaction with Ub. More importantly, the active cysteine in the core structure forms a thioester bond with the activated Ub glycine residue transferred from the E1 (3). Unlike other Ubc E2s, Ubc9 does not form a thioester bond with Ub, but with a Ub-like molecule, small Ub-like modifier (SUMO) (14, 15). SUMO-conjugated protein substrates are not degraded by the 26S proteasome. Protein sumoylation has been implicated in protein stability enhancement, protein-protein interaction, subcellular translocation, and transcriptional control (16). In addition to the classical Ubc E2s, there are noncanonical Ub E2 variants (UEVs). One such UEV is the yeast MMS2, which resembles conventional E2s but does not contain the catalytically active site to form a thioester bond with the activated Ub (17). MMS2 forms a complex with a conventional E2, Ubc13, which is implicated in postreplicative DNA repair. Interestingly, the MMS2-Ubc13 complex is involved in a novel assembly of poly-Ub chains via the lysine-63 in the Ub molecule, instead of the usual lysine-48-linked chain formation. Uev1, a mammalian MMS2 homologue, also forms a complex with Ubc13 and is involved in lysine-63-linked polyubiquitination in the NF-κB signaling pathway (18). Another well-studied UEV is the tumor susceptibility gene 101 (Tsg101), which was originally identified as a tumor suppressor whose aberrant expression is associated with many human malignancies (19). It has a structure very similar to conventional E2s, but it lacks the catalytic cysteine residue for Ub conjugation (20). Tsg101 is involved in the traffic of endocytosis by delivering ubiquitinated cargo proteins into multivesicular bodies (20) and in the budding of viral particles from virus-infected cells (21).

11 Feb 2004

84

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The E3 Ub Ligases The E3 Ub ligases play a critical role in the Ub conjugation cascade by recruiting Ub-loaded E2, recognizing specific substrates, and facilitating the Ub transfer from the E2 to the lysine residues of the substrate. Compared with only dozens of E2s, a search of the gene database reveals that there are hundreds of E3s. The E3s are present in the form of either a single protein or multisubunit complexes. According to the homology of the amino acid sequences in the E2-binding domain, E3s can be generally divided into two large groups: the homology to the E6associated protein carboxyl terminus (HECT) domain-containing E3s, and the really interesting new gene (RING) domain-containing E3s. In the RING-type E3s, there are single subunit E3s and multisubunit E3s. More recently, RINGlike domain-containing E3 ligases have been identified, such as the PIAS family SUMO ligases, the plant homeodomain (PHD) domain-containing E3s, and the U-box E3s.

HECT-Type E3 Ligases Studies on human papillomaviruses led to the discovery of the viral E6-associated protein (E6-AP), which forms a complex with the oncogenic E6 to induce the degradation of the p53 tumor suppressor (22). Later on, the same group identified a family of proteins that have a highly conserved region of ∼350 amino acids similar to the C-terminus of E6-AP, named the HECT domain (23). In the Cterminus of the HECT domain, there exists a conserved active cysteine residue, which forms a high-energy thioester bond with Ub and constitutes a necessary step for the Ub transfer to the substrate. Interestingly, E6-AP seems not to be the physiological E3 ligase because it does not induce p53 degradation in HPVuninfected cells (24). Recent studies have demonstrated that Src family kinases are the potential targets for E6-AP (25, 26). Importantly, genetic analysis has linked the disruption of the maternal copy of E6-AP with Angelman syndrome, a genetic neurological disorder (27). E6-AP also acts as a coactivator for steroid hormones through ligase-dependent or -independent mechanisms (28). With the exception of E6-AP, other HECT domain–containing E3 ligases often contain an N-terminal Ca2+-binding, protein kinase C-related C2 domain, followed by multiple WW domains, in addition to the C-terminal HECT domain (29). WW domains derive their name from the presence of two highly conserved tryptophan (W) residues, which are spaced 20–22 amino acids apart. They normally contain 38–40 amino acids in a triple-stranded β sheet and are found in proteins that participate in cell signaling or regulation. These domains are implicated in mediating protein-protein interactions by binding to proline-rich motifs (30) or phosphoserine- and phosphothreonine-containing elements in their binding partners (31). One of the well-studied WW domain–containing HECT-type E3s in mammalian cells is Nedd4, which is implicated in the regulation of the epithelial Na+ channel in the kidney and other tissues, and whose deletion is

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

85

related to the human Liddle’s syndrome, a hereditary form of hypertension (32). The WW domains of Nedd4 associate with a PPXY motif in the channel protein, and mutations of the PPXY motif were found in the patients with Liddle’s syndrome. More relevant to the immune system is the HECT-containing Itch E3 Ub ligase, whose deficiency results in abnormal immunological and inflammatory responses and which is discussed later in this review.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

RING-Type E3 Ligases Compared with the early discovery of HECT-type E3s, the identification of RINGtype E3 ligases is a relatively recent event. RING finger is characterized by the presence of the consensus sequence of C3HC4 (C, cysteine; H, histidine) or C3H2C3, which coordinates two cross-braced Zinc-binding sites (33). Database search revealed that there are hundreds of RING-containing proteins, in contrast to the limited number of HECT-type E3s. Unlike the HECT-type E3s, which have an active cysteine in the HECT domain for the Ub binding and transfer, RING-type E3s do not form a thioester bond with Ub, but rather bring the Ub-loaded E2s and the substrate into proximity, and promote the Ub transfer directly from the E2 to the substrates (34). Depending on whether the RING finger is present in the form of a functional domain in a single protein or in the form of a subunit in a protein complex, the RING-type E3s are subdivided into single protein E3s and multisubunit E3s (3). SINGLE PROTEIN RING-TYPE E3S In this subgroup, the RING finger constitutes a functional portion of a single protein, in which a protein-interaction domain or domains are also present for the substrate recruitment. One of the earliest identified E3s of this subgroup is Cbl. Cbl is a 120 kDa protooncogene product that is comprised of an N-terminal tyrosine kinase–binding (TKB) domain, a RING finger, and C-terminal proline-rich sequences and tyrosine phosphorylation sites (35). Cbl acts as an E3 Ub ligase, whose RING finger recruits Ub-loaded E2, and its TKB domain binds to tyrosine phosphorylated receptor tyrosine kinases (36–38). The crystal structure of Cbl RING-E2 complex further supports a role of Cbl in Ub conjugation, in which the Cbl RING finger forms a shallow grove on to which the two loops of UbcH7 bind (39). More interestingly, the E2-binding grove in the Cbl RING domain is quite similar to that in the HECT domain of E6-AP (40), and UbcH7 uses the same structural elements for interaction with both domains, even though there is no similarity in amino acid sequences between a RING finger and a HECT domain. Another example of the single peptide E3s is MDM2, which binds to the tumor suppressor p53 and promotes its ubiquitination (41). The number of this subgroup of E3s has increased considerably, and I discuss several of them later in this review. MULTISUBUNIT RING-TYPE E3s This subgroup consists of a superfamily of E3s including the SCF (Skp1-Cullin 1-F box protein), the APC (Anaphase-promoting complex), and the VCB (VHL-elongin C/elongin B) (3, 42). All three E3s contains

11 Feb 2004

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

86

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

an ∼100 amino acid RING finger, Roc1/Rbx1 or Apc11, which together with its binding component, Cullin in the SCF or VCB E3s, or Apc2 in the APC, forms a core enzymatic structure for Ub-charged E2 recruitment. The substrate recognition is mediated via another subunit, which contains binding domains for another E3 component, Skp1, as well as binding domains for the substrate protein. In the SCF complex E3s, the F box proteins are responsible for the substrate recruitment and gene database revealed that there are hundreds of them in the yeast and humans, suggesting the diversity and specificity of these subfamily E3s in substrate targeting. A recent crystal structure study of the SCF complex revealed how these components are coordinated to facilitate Ub transfer from the E2 to a substrate (43), in which Cullin acts as a scaffold to position Rbx1/E2 complex and the Skp1/F box complex in a rigid distance. Such rigidity of the SCF complex may be important for the specific targeting of the lysine residues on a particular substrate. One of the well-characterized F box proteins implicated in the immune regulation is β-tranducin repeat-containing protein (β-TrCP), which binds to several critical molecules in lymphocyte signaling and is described below. Like the SCF complex, the VCB E3 ligase complex is also comprised of multiple components that include Roc1/Rbx1, Cullin 2, the Skp1-like protein elongin C, the Ub-like elongin B, and the substrate recruiting subunit pVHL (44). The pVHL was originally discovered as a protein product of the von Hippel-Lindau tumor suppressor gene, whose inactivation is associated with hereditary blood-vessel tumors in the nervous system and clear-cell kidney cancers (45). pVHL contains two domains, the α domain and the β domain. The α domain binds to elongin C, which forms a complex with Cullin 2 and Roc1 (46, 47). The β domain recognizes a substrate protein, hypoxia-inducible factor-α (HIF-α), and induces its ubiquitination and degradation (48, 49). The interaction between pVHL and HIF-α is regulated by oxygen, which induces hydroxylation of a particular proline residue on HIF-α (50–52). Crystal structure analysis showed that the hydroxyproline in HIF-α inserts into a gap in the pVHL hydrophobic core to form direct contact with the β domain of pVHL (53, 54). HIF-α forms a heterodimer with a constitutively expressed HIF-β subunit to function as a transcription factor central to cellular responses to hypoxia (55). Under normal conditions, HIF-α is hypoxylated and rapidly degraded by pVHLinduced ubiquitination. Under low-oxygen concentrations, HIF-α hydroxylation is inhibited and stabilized. Accumulation of HIF-α causes the gene transcription of several critical cellular factors involved in angiogenesis and glycolysis. Consistent with this notion, HIF is constitutively activated in the kidneys of patients with VHL disease (56), although it remains questionable whether stabilization of HIF-α alone is sufficient for tumorigenesis (57, 58).

PIAS Family SUMO Ligases Recent studies have identified a new family of RING-type E3 ligases, the PIAS (protein inhibitor of activated STAT) family, in catalyzing SUMO conjugation

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

87

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

(59, 60). The PIAS proteins contain a RING-like domain, in which (unlike the classical C3HC4 or C3H2C3 alignment) the two conserved cysteine residues for zinc binding are replaced with serine and aspartic acid, but it can associate with SUMO-conjugated Ubc9, and it promotes sumoylation to the PIAS-binding proteins such as septins in yeast (61, 62) and p53 in mammalian cells (63). PIASs have also been shown to act as SUMO E3 ligases for c-Jun and androgen receptors (64–66). Interestingly, sumoylation of these transcription factors leads to the repression of the transcriptional activity, although the molecular mechanism for the inhibition is not clear.

PHD Domain-Containing E3 Ligases The PHD or leukemia-associated protein domain consists of ∼60 amino acids, which are characterized by a consensus C4HC3 zinc-binding sequence, distinct from classical arrangement in a RING finger (67). The PHD domain has a structural feature that is very similar to the one in the RING domain, suggesting a role for this domain in the Ub conjugation pathway (68). As an example of mammalian PHDcontaining protein as an E3, MEKK1, a MAPK kinase kinase, exhibits E3 ligase activity and mediates ubiquitination and degradation of downstream targets, Erk1/2 (69). Interestingly, because MEKK1 is also an upstream activator of Erk, demonstration of MEKK1 as a negative regulator of Erk through its E3 ligase activity suggests a novel feedback mechanism for the regulation of this signaling network. In addition, a separate study showed that MEKK1 undergoes self-ubiquitination, which inhibits its kinase activity of its direct downstream substrates, MKK1 and MKK4 (70). The PHD domain is present in more than 400 eukaryotic proteins, many of which are involved in transcriptional regulation (67). The importance of this domain is further underscored by the findings that natural mutations of the PHD domain are implicated in the pathogenesis of human diseases including mental disorders, autoimmune diseases, and the formation of carcinomas (71). Further characterization of these PHD domain–containing proteins will reveal their physiological functions in normal cells.

U-Box E3 Ligases Originally identified as a novel ubiquitination factor, or E4 in yeast, UFD2 (Ub fusion degradation 2) works in conjunction with E1 and E2, and a HECT-type E3 to facilitate longer Ub chain assembly (72). A database search revealed homologues in other eukaryotes, including humans, that contain highly conserved C-terminal ∼100 residues similar to that in UFD2, designated as the U-box (UFD2-homology domain). A later study showed that the U-box of UFD2 and its homologues directly associates with E2s, including Ubc4 and UbcH5C, and the U-box-containing proteins themselves function as a new family of E3 ligases (73). In particular, the C-terminus of heat-shock protein (Hsc70)-interacting protein, CHIP, a mammalian U-box protein, is a cochaperone of heat-shock proteins, Hsc70 or Hsc90,

15 Mar 2004

88

15:3

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

and promotes Ub conjugation to the Hscs, or Hsc-capatured unfolded proteins (74, 75). Interestingly, structural modeling of the U-boxes predicts a feature similar to the RING finger (76). Although the U-box lacks the zinc-binding cysteine or histidine residues for the stabilization of the RING finger, it may use salt-bridges and hydrogen bonds to form a RING-like structural scaffold. The aromatic tryptophan resides conserved in the RING finger, which is required for E2 binding (36), is also present in the U-box and is exposed to contribute to E2 binding.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ITCH E3 LIGASE IN LYMPHOCYTE REGULATION Itch was identified through the study of the agouti locus on mouse chromosome 2, and mutation at this locus results in a wide variety of coat color alterations (77). The agouti protein induces the production of yellow pigment by melanocytes and thus determines the amount of yellow present in the hair. One of these mutations, 18H, which displays a darker coat, causes immunological disorders. The most obvious disorder is ear and skin scarring due to constant itching (thus called itchy mice) starting from 16-week-old or older mice. The mutant mice have enlarged spleens and lymph nodes, possibly due to lymphocyte hyperproliferation. Genetic studies have shown that 18H mutation results from a chromosomal inversion that deletes 18 and 20 bp from the proximal and distal inversion breaks, respectively. This inversion disrupts the expression of agouti and another novel protein, named Itch (78). Subsequent cDNA cloning showed that the Itch gene encodes 854 amino acids with a molecular weight of ∼113 kDa. It consists of an N-terminal protein kinase C-related C2 domain, four WW protein-protein interaction domains, and a C-terminal HECT Ub ligase domain. Itch therefore is a member of the HECT domain-containing E3 Ub protein ligases. The data based on itchy mice clearly suggest a critical role of E3 ligase or ubiquitination in the regulation of the immune system.

Itch in T Cell Differentiation A recent study showed that the thymocyte development in Itch-/- was relatively normal compared to wild-type controls (79). Itch-/- T cells showed slightly enhanced T cell proliferation and IL-2 production upon anti-CD3 engagement. In aging mice, Itch-/- T cells displayed increased cell surface expression of CD69 activation marker, suggesting a chronic activation of T cells in the absence of Itch. Interestingly, increased serum levels of IgG1 and IgE in itchy mice as well as a T helper cell type 2 (Th2)-biased differentiation were observed in itchy mice. The molecular mechanism underlying Itch-mediated T cell differentiation was further explored by using biochemical approaches. Itch was found to associate with Jun-B, through a PPXY motif in Jun-B and the WW domains of Itch, and to promote Jun-B ubiquitination (Figure 1A). Interestingly, Jun-B has been implicated in the gene regulation of Th2 cytokines such as IL-4 (80). In Itch-/T cells, the rate of Jun-B degradation was reduced, in parallel with increased

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

89

nuclear translocation and DNA binding activity of Jun-B. Na¨ıve CD4+ T cells differentiate into Th1 and Th2 effector cells based on their cytokine profiles and distinct functions (81, 82). Whereas Th1 cells produce IL-2 and IFN-γ , Th2 cells produce IL-4 and IL-5. In addition, the differential production of Th1 versus Th2 cytokines plays an important role in determining the immune response to intracellular and extracellular infections. Thus, the data point out a negative role for Itch in Th2 differentiation, which may be causal to the abnormal immunological responses in itchy mice. Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

c-Jun Ubiquitination and Activation Initiation of eukaryotic gene transcription requires the assembly of a multimolecular complex, including RNA polymerase II, transcription factors, and chromotinmodulating enzymes (83). Transcription factors are normally comprised of a transcription activation domain and a DNA-binding domain. The DNA-binding domain confers specificity by interacting with particular consensus nucleotide sequences, whereas the activation domain stimulates the activity of RNA polymerase II through direct protein-protein interaction (83). Given the critical role for transcription factors in gene expression, it can be assumed that they are tightly regulated by diverse mechanisms such as phosphorylation, translocation, or their own gene expression. One of the emerging mechanisms is Ub-dependent proteolysis, which has been implicated in the regulation of many critical transcription factors, including c-Jun, p53, c-Myc, NF-κB (84). However, our knowledge on the molecular details of how ubiquitination regulates transcription factors remains elusive. The Jun family of transcription factors consists of three members: c-Jun, Jun-B, and Jun-D (85, 86). They heterodimerize with Fos family transcription factors to form activation protein 1 (AP-1) transcription factor, which recognizes a consensus TPA responsive element and is implicated in the gene transcription of diverse cellular processes critical for cell proliferation, cell death, and tumorigenesis. It is particularly pertinent to immune regulation that many cytokine promoters contain AP-1 binding sites, and the AP-1 transcription factor is indeed involved in cytokine production and immune responses (87). c-Jun is one of the earliest identified transcription factors that is regulated by protein ubiquitination (88). Multiubiquitination of c-Jun was demonstrated both in vitro and in vivo, and a delta domain in the transcriptional activation domain was proposed to be responsible for Ub conjugation. Interestingly, this delta domain is missing in v-Jun, which is a transforming form of c-Jun. Thus, ubiquitination of c-Jun and its subsequent degradation by the 26S proteasome is implicated in the control of oncogenesis. Although c-Jun ubiquitination has previously been reported, the E3 Ub ligase that catalyzes Ub conjugation to c-Jun is still unclear. Because both Jun-B and c-Jun, but not Jun-D, contain a PPXY motif, it is possible that Itch is a promising E3 ligase for c-Jun as well. To support this, it was recently observed that Itch indeed induces Ub conjugation to c-Jun (D. Fang & Y.C. Liu, unpublished data). Recent studies have suggested a correlation between the transcription activation domain and proteolysis-dependent degradation (89). Many transcription

11 Feb 2004

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

90

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

factors such as c-Myc, p53, the viral protein VP16, or even c-Fos or c-Jun contain the signal for Ub-mediated proteolysis in their transcription activation domains (90). More intriguingly, using a fusion containing VP16 transcription activation domain and a bacterial DNA-binding domain of LexA, Salghetti et al. (91) further demonstrated that Ub conjugation to the VP16 transcription activation domain is required to activate transcription. Thus, ubiquitination serves a dual signal in regulating transcription factors, one for activation and another for destruction. The Jun family members are regulated by upstream kinases, and in the case of c-Jun, it is phosphorylated by Jun N-terminal kinase (JNK) at serines 63 and 73 of the transcriptional activation domain, which is required for its activation (86). A previous study demonstrated that phosphorylation of c-Jun by JNK in fact reduces its ubiquitination, thus suggesting an inverse relationship between c-Jun phosphorylation and its degradation (92). The data also call into question the “suicide” model for the regulation of transcriptional activators. Whether Itch-mediated ubiquitination of c-Jun is implicated in c-Jun degradation or its activation is not known at present. Obviously, further detailed function-structural analysis of Itch-mediated ubiquitination of c-Jun and its function is needed.

Itch Regulation of B Cell Development and Activation A potential role of Itch in the regulation of B cells emerges from the studies on Epstein-Barr virus latent membrane protein 2A (LMP2A) (93, 94). LMP2A is a membrane protein expressed on infected human B cells and implicated in lifelong infection with Epstein-Barr virus (95). It contains an N-terminal transmembrane domain, and intracellular consensus ITAM motif, and two highly conserved PPXY motifs. Using the PPXY motif as a probe, two groups isolated WW domaincontaining HECT E3 ligases including Nedd4 and the human Itch homologue, AIP4 (93, 94). It was shown that coexpression of LMP2A and AIP4 can induce Ub conjugation to Lyn and Syk, and the degradation of Lyn (93, 96). It has been known that LMP2A inhibits BCR signaling (95). The link of AIP4 (or other HECT E3s) to LMP2A suggests a novel Ub-dependent mechanism underlying LMP2Amediated regulation of B cell signaling. It should be mentioned that LMP2A can also block BCR signaling by prohibiting BCR from entering membrane rafts (97). In addition, whether AIP4 functions as an E3 ligase for Lyn or Syk only in the presence of LMP2A is not clear. To understand whether Itch deficiency affects B cells in itchy mice, we recently performed a systematic analysis of B cells in the mutant mice (D. Fang & Y.C. Liu, unpublished data). We found that B cell numbers are increased at different stages of B cell development, including those in the bone marrow and spleen of itchy mice. Itch-/- B cells from bone marrow display an enhanced response to interleukin-7 (IL-7) stimulation. Interestingly, the IL-7 receptor α and γ chains contain a consensus PPXY motif that binds directly to the WW domains of Itch, suggesting that Itch may target the IL-7 receptor for ubiquitination and induce its downmodulation. Both the IL-7 receptor α chain and the γ chain have been

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

91

implicated in the differentiation from pro-B cells into pre-B cells (98–100). Thus, Itch may regulate B cell development through the downregulation of the IL-7 receptor.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CBL PROTEINS AS RING-TYPE E3 LIGASES Numerous previous studies demonstrated that Cbl (Casitas B-cell lymphoma) protein functions as an adaptor protein by interacting with cell surface receptor and intracellular protein tyrosine kinases (PTKs), and other critical signaling molecules such as phosphatidylinositol (PI)-3 kinase, Crk-L, 14-3-3, and Grb2 (35, 101). In addition to Cbl, there are two close mammalian homologues, Cbl-b and Cbl3, which have similar N-terminal TKB domain, a RING finger, and C-terminal proline-rich sequences, except for Cbl-3, which contains a shorter C-terminal region (102, 103). The identification of Cbl as a RING-type E3 Ub ligase, as described earlier, suggests a distinct mechanism by which Cbl participates in the negative regulation of signal transduction. Recent studies have elegantly documented that Cbl is an E3 ligase for the activated protein tyrosine kinases of the Src and Syk/ZAP-70 families in lymphocytes, and Cbl-mediated ubiquitination of these kinases leads to their degradation, thus resulting in attenuation of antigenreceptor signaling (104–107), as recently reviewed (108, 109). Here I discuss some other Ub ligase-related aspects of Cbl and Cbl-b relevant to the lymphocyte regulation.

Cbl in T Cell Antigen Receptor Downmodulation Activation-induced internalization of cell surface receptor or receptor PTKs has long been known to be a very important means to terminate receptor signaling. One of the mechanisms to switch off such signaling is ubiquitination of the receptor that serves a signal for degradation (110). Cbl acts as an E3 Ub ligase to promote Ub conjugation to receptor PTKs such as EGFR and PDGFR (reviewed in Reference 109). It was proposed that Cbl is recruited to the activated receptor PTKs at the plasma membrane and induces their ubiquitination, which serves as a signal for the internalization and subsequent endocytotic sorting (111). Engagement of the TCR by the peptide major histocompatibility complex (MHC), or anti-TCR/CD3 antibodies also results in downregulation of the TCR/ CD3 complex (112–114). Downregulation of the TCR/CD3 complex is considered to be a self-limiting factor to terminate a sustained signaling in the MHC-TCR conjugated T cells, and it is also proposed to be responsible for unresponsiveness and/or tolerance induction of the self-reactive T cells (115, 116). Ub conjugation to the TCR/CD3 subunits, and particularly to the TCRζ chain, has been shown in activated T cells (117, 118). Multiple lysine residues in the intracellular domain of TCRζ chain are potential sites for Ub attachment (119). The enhanced cell surface expression of TCR/CD3 complex in Cbl-/- thymocytes clearly suggests a

11 Feb 2004

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

92

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

functional role of Cbl in their downmodulation. In support of this notion, it was demonstrated that Cbl acts as an E3 Ub ligase to promote Ub conjugation to TCRζ (120). More importantly, Zap-70 plays an adaptor’s role in facilitating Ub transfer to the TCRζ , instead of being a direct target for Cbl, as can be predicated from studies on receptor PTKs (Figure 1B). Therefore, the study provides a molecular link between the enhanced cell surface expression of TCR/CD3 in Cbl-/- thymocytes (121, 122) and its E3 Ub ligase activity. In addition, Cbl is implicated in the internalization of pre-TCR (123) as well as in the allelic exclusion of the T cell alpha chain in thymocyte development (124). A recent study on T cells deficient in both Cbl and Cbl-b showed a more profound effect on TCR/CD3 downregulation (125). Cbl may use alternative pathways to induce downregulation of activated receptor tyrosine kinases (126, 127). Cbl forms a complex with a Cbl-interacting protein of 85 kDa (CIN85) and endophilin. Endophilin has been implicated in receptor-mediated endocytosis through the invagination of the lipid bilayer (128). Although CIN85 constitutively binds to endophilin, the CIN85/endophilin complex is recruited to Cbl upon receptor activation. Therefore, Cbl can induce their downmodulation via the E3 ligase-independent endophilin-CIN85-Cbl complex formation. However, one of the two groups later showed that Cbl also induces monoubiquitination of CIN85 (129), and it seems that Cbl and CIN85 can form a complex with the activated receptor independent of endophilin. Thus, CIN85 plays a dual role in internalization and endocytic degradation. Whether similar mechanisms function in lymphocytes remains unclear. Another potential mechanism by which Cbl regulates TCR downmodulation is through the Src-like adaptor proteins (SLAPs). SLAP-deficient thymocytes display upregulation of TCR and CD5 at the CD4+CD8+ stage, a phenotype similar to the Cbl-/- thymocytes (130). In transiently tranfected cells, SLAP-2, a SLAP homologue, is shown to inhibit antigen receptor signaling (131) and reduces the surface expression of CD3 through a direct interaction with Cbl (132). Thus, Cbl may participate in the downmodulation of the TCR/CD3 complex via multiple pathways. It should be noted that not all Cbl-binding proteins are mediators for the regulation of receptor downregulation by Cbl. The Sprouty adaptor proteins associate with Cbl and are themselves the targets of Cbl-mediated polyubiquitination (133– 135). The interaction of Sprouty with Cbl can either reduce the E3 ligase activity of Cbl by blocking the E2-Ub loading to the RING finger (133) or sequester Cbl from its access to the activated receptor PTK, and its subsequent ubiquitination and degradation (134).

Proteolysis-Independent Regulation of PI3-kinase by Cbl-b The effective activation of naive T cells requires two signals: The first one is antigen-specific and is generated by the recognition of an antigenic peptide presented through MHC complex by the TCR; the second signal is antigen-nonspecific and is provided by interaction of the B7 ligand on the antigen-presenting cell with

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

93

its receptor, CD28, a costimulatory cell surface molecule on the T cell (136). In the presence of both signals, the T cell will proliferate and secrete cytokines. Genetic studies showed that Cbl-b-deficient peripheral T cells uncouple CD28 costimulation: Cbl-b-/- T cells proliferate and secrete IL-2 in the absence of CD28 costimulation. Cbl-b-deficient mice develop autoimmune responses (137–139). The findings that c-Cbl RING finger can recruit Ubc E2 and that it facilitates Ub conjugation to protein tyrosine kinases to which Cbl binds suggest that Cbl or Cbl family proteins function as Ub protein ligases or E3s to downregulate intracellular signaling (36). Consistent with this, it was recently identified that p85, the regulatory subunit of PI3-kinase, is a substrate for Cbl-b (140). More interestingly, ubiquitination of p85 seems not to induce its degradation because p85 is a very stable protein and its stability is not affected by either Cbl-b overexpression or Cbl-b deficiency (141). It was found that the interaction of p85 with the cell surface CD28 or TCR/CD3 complex is increased in Cbl-b-/- T cells. Thus, Cbl-b regulates p85 in a proteolysis-independent manner through affecting protein-protein interactions. It can be theorized that Ub conjugation to p85 may cause structural hindrance to prevent its interaction with other binding proteins such as CD28, although future study is needed to test this hypothesis.

Crk-L as a Common Target for Cbl and Cbl-b Cbl forms a complex with another adaptor protein, Crk-L, through two C-terminal tyrosine residues (tyrosine-700 and tyrosine-774) of Cbl and the SH2 domain of Crk-L (reviewed in Reference 109). Crk-L also constitutively interacts with a guanine exchange factor, C3G, which preferentially catalyzes the conversion from a GDP-bond form of Rap1 to a GTP bond form (142). Consistent with this, it was shown that in T cells, activation of Fyn kinase induces phosphorylation of Cbl and subsequently triggers the Cbl-Crk-L-C3G signaling pathway, which eventually leads to Rap1 activation (143). In insulin-stimulated adipocytes, a similar Cbl-mediated pathway causes the activation of the small GTP-binding protein TC10 (144). In fibroblast cells, triggering cAMP/PKA activates Src kinase, which in turn induces Cbl-Crk-L-C3G complex formation as well as Rap1 activation (145). All these data collectively point out a positive role of Cbl in the activation of Rap1. However, the genetic evidence to implicate Cbl in Rap1 signaling is lacking. Surprisingly, Cbl deficiency causes an activation of Rap1, which is supported by an increase in Crk-L-C3G association, C3G membrane translocation, and the exchange activity of C3G (146). The current model is that Cbl, as an E3 ligase, promotes ubiquitination of Crk-L, which inhibits the interaction with C3G. It seems that, genetically, Cbl is a negative regulator of Crk-L-C3G signaling, which is in sharp contrast to its previous documentation as a positive regulator. As described earlier, genetic studies showed that Cbl and Cbl-b target different lymphoid tissues. We (147) proposed that the differential functions of Cbl and Cblb are mediated by targeting different substrates for ubiquitination, with Cbl being

11 Feb 2004

94

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

an E3 ligase for a TCR/CD3 complex and Cbl-b for a PI3-kinase. However, given the high homology between these two molecules, it was not surprising to learn that they have overlapping functions. Cbl-b also associates with Crk-L (148). It seems that like Cbl, Cbl-b also negatively regulates Crk-L-C3G complex formation in a proteolysis-independent manner and that the activation of Rap1 in Cbl-b-/peripheral T cells is augmented (149). Thus, Crk-L may act as a common target for Cbl and Cbl-b E3 ligases.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

E3 Ub Ligases in T Cell Tolerance The immune system has evolved mechanisms by which it mounts defensive responses to invading pathogens and at the same time, the self-tissues are protected from destruction. The capacity to distinguish self and nonself results from at least two stages of tolerance induction (150, 151). The first stage is in the thymus, where maturing T cells of self-reactivity are eliminated by negative selection. However, thymic deletion is a leaky process and the self-reactive T cells can escape such negative selection and enter the peripheral lymphoid tissues including the spleen and lymph nodes. Then comes the second stage of immune tolerance induction that is acquired by several mechanisms, including ignorance, activation-induced cell death, anergy, and the generation of regulatory T cells. Genetic and/or environmental changes may cause the breakdown of immune tolerance, which can result in the development of autoimmune diseases. T cell anergy represents a state of hypo- or nonresponsiveness in response to restimulation with antigen plus antigen-presenting cells (152). In recent years, numerous studies have focused on the elucidation of the intracellular signaling pathways responsible for the anergy induction. Earlier studies showed that anergized T cells display a block in Ras or Erk and Jnk activation (153, 154). In anergic T cells, Rap1 is preferentially activated through a Fyn-dependent Cbl-Crk-L-C3G signaling pathway (143). Rap1 has been proposed to be an antagonist of Ras by sequesting Raf kinase from Ras (142). This model is further supported by findings that costimulation with anti-CD28 causes the inhibition of Rap1 activation (155, 156), thus rescuing the anergic state. However, as described earlier, Cbl and/or Cbl-b seem to be negative regulators in Rap1 activation (146, 149). A recent study showed that T cells from Rap1 transgenic mice do not have defects in Ras signaling; rather, Rap1 activates the TCR-induced integrin signaling (157). In fact, Cbl-/- thymocytes or Cbl-b-/- peripheral mature T cells also show increased integrin activation, including enhanced binding to ICAM-1 or augmented LFA-1 clustering in response to TCR stimulation (146, 149). A recent study shed new light on the mechanisms of T cell anergy induction, in which an unbalanced and sustained Ca2+-dependent activation of NFAT leads to the transcriptional expression of anergy-associated genes (158). The products of these genes are distinct from those induced by normal productive T cell triggering and can together render T cells into the anergic state. This new model is supported by the observation that NFAT1-/- T cells are resistant to anergy induction (158). Some of these genes

15 Mar 2004

15:4

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION

95

encode proteins involved in proteolytic pathways, such as caspases, or the RING finger-containing E3 ligases like SOCS-1 and TRAF6, suggesting that proteins functioning in the proteolytic pathways are involved in T cell anergy induction. In a subsequent study by the same group, it was shown that sustained Ca2+ signaling increases the expression of the E3 ligases Itch and Cbl-b, and the UEV Tsg101 (V. Heissmeyer & A. Rao, submitted manuscript), which results in selective downmodulation of critical signaling molecules PLC-γ 1 and PKCθ and the failure to form mature immunological synapse (Figure 2). The data thus clearly link the Ub conjugation pathway to the induction of lymphocyte tolerance. Another recent study identified a novel RING-containing protein, GRAIL (gene related to anergy in lymphocytes), in T cell unresponsiveness. GRAIL is a transmembrane protein residing in the endocytic compartment that exhibits in vitro E3 ligase activity (160). Expression of GRAIL in a T cell hybrid inhibits both IL-2 and IL-4 production, which depend on a functional endocytic pathway. It was suggested that GRAIL plays a role in T cell anergy induction by regulating endosome-dependent sorting of cell surface receptors and/or intracellular signaling molecules.

E3 LIGASES IN NF-κB SIGNALING TRAFs in the Formation of Variant Ub Chain and Kinase Activation Tumor necrosis factor (TNF) receptors are a superfamily of type II receptors involved in many aspects of the immune responses, including lymphoid organ formation, the development, activation, and death of lymphocytes, and inflammation (161). TNF receptors transduce signaling through adaptor molecules that bridge the receptor intracellular domains with downstream effectors. One group of the adaptor molecules is TNF receptor-associated factors (TRAFs), which have at least six members, TRAF1 to TRAF6. With the exception of TRAF1, all other TRAFs have an N-terminal RING domain, which suggests that this family of proteins plays a role in protein ubiquitination. The first evidence that TRAFs are involved in protein ubiquitination came, however, from studies on IL-1-induced NF-κB signaling (18). Like TNF receptors, the IL-1 receptor transduces signal leading to the activation of IkB kinase or IKK and NF-κB by the initial formation of multimolecular complex including Myd88, IRAK, and TRAF6 (162). Biochemical fractionation of this complex revealed the presence of two additional molecules, Ubc13 and Uea1A. The Ubc13/Uea1A complex is required for the activation of IKK by TRAF6. It turned out that TRAF6 facilitates the assembly of a lysine-63-mediated poly-Ub chain with the help of Ubc13/Uea1A. A subsequent study by the same group showed that lysine-63-linked polyubiquitination by TRAF6 and Ubc13/Uea1A is involved in TAK1 kinase complex activation and phosphorylation of IKK, and that TARF6 itself is the substrate for lysine-63-linked polyubiquitination (120). These studies also highlighted a proteolysis-independent role of protein ubiquitination.

11 Feb 2004

96

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

Another recent study on TNFα-induced activation of germinal center kinase also demonstrated that TRAF2 is involved in the activation of downstream signaling through a complex formation with Ubc13/Uea1A (163). Thus, TRAF proteins may have a general role in promoting self-ubiquitination, subsequent oligomerization, and activation of downstream kinases.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

E3 Ub Ligases for TRAFs If TRAFs promote lysine-63-linked polyubiquitination on their own and activate kinase complexes, one may wonder whether there is a mechanism by which TRAFs are degraded by the classical proteasomal dependent pathways. Indeed, stimulation of CD40, a member of TNF receptor family, on B cells caused 26S proteasomedependent degradation (164). Studies on inhibitors of apoptosis (IAPs) provide further clues to TRAF degradation. The IAPs are originally found in baculoviruses as inhibitors of cell death in infected cells and are a family of proteins containing one or more baculovirus IAP repeat (165). Some of the IAPs, like c-IAP1, c-IAP2, or X-linked IAP (XIAP) also contain a C-terminal RING finger. It is known that thymocytes become apoptotic upon stimulation with glucocorticoids, irradiation, and antigen receptor engagement. It was found that treatment of thymocytes with such stimuli selectively induces the loss of XIAP and c-IAP1, which can be prevented by proteasome inhibitors (166). In addition, both XIAP and c-IAP1 induces selfubiquitination that is dependent on a functional RING domain. Thus, the results provide an explanation for thymocyte apoptosis by inducing self-ubiquitination and their own degradation of IAP proteins. IAPs also directly interact with activated caspases through their baculovirus IAP repeat domains and inhibit the caspase activities (167). It was shown in an in vitro system that c-IAP1 catalyzes Ub conjugation to caspase-3 and caspase-7 (168), and XIAP promotes ubiquitination and proteasomal degradation of caspase-3, and thus exerts its antiapoptotic effect in Fas-induced cell death (169). Back to the degradation of TRAF proteins, it was found that stimulation of TNF receptor type II, but not type I, causes decreases in the TRAF2 protein level, which can be reversed by the addition of proteasomal inhibitors (170). In the study, c-IAP1 was identified as an E3 ligase for TRAF2 ubiquitination and degradation. It can be assumed that c-IAP-mediated TRAF2 ubiquitination is lysine-48-linked poly-Ub chain, which differs from TRAF and Ubc13/Uea1A-mediated lysine-63-linked Ub conjugation. A differential Ub chain formation thus results in totally opposite functional consequences. The molecular details when TRAF functions as an E3 ligase to activate downstream signaling, and when it functions as a substrate for IAPs to block apoptosis, have yet to be determined. It should be mentioned that in CD40-mediated signaling of B cells, TRAF2 is also shown to be self-ubiquitinated and induce its own degradation (164).

Ubiquitination of IκB Stimulation by proinflammatory cytokines such as TNFα, or IL-1, microbial products as well as stress exposure causes the activation of the IKK complex that

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

97

consists of IKKα, IKKβ, and a regulatory subunit, NEMO (9). The activated IKK then phosphorylates IκB, which is then recognized by β-TrCP, a component of SCF Ub ligase. Under resting conditions, IκB forms a complex with NF-κB and retains NF-κB in the cytoplasm. Upon stimulation, IκB is ubiquitinated and degraded by the 26S proteasome, liberating NF-κB to allow it to translocate into the nucleus and exert its transcriptional activity. The NF-κB signaling pathway is essential in mediating both innate and adaptive immunity in both mammals and insects (171). Likewise, the fly IκB homologue, Cactus, is also ubiquitinated and degraded by Slimb, a Drosophila homologue of β-TrCP (159). In mammalian cells, IKKβ, and to a less extent, IKKα, induces phosphorylation of Ser-32 and Ser-36 of the N-terminal IκB protein, which creates the docking site for the WD40 repeat in β-TrCP. β-TrCP also forms a complex with Skp1, Cullin 1, and Roc1 through its F-Box domain. The RING domain in Roc1 then recruits Ub-loaded Ubc E2 to transfer Ub to IκB protein (10). It appears that the proteasomal-dependent degradation of IκB is a very rapid process, occurring only minutes after stimulation, which may ensure a timely release of NF-κB from the IκB-NF-κB complex.

Ub-Dependent Processing of NF-κB2 p100 The NF-κB transcription factor consists of a family of structurally related proteins, including c-Rel, RelA, RelB, NF-κB1 p50/p105, and NF-κB2 p52/p100 (172). Both the Rel and NF-κB proteins share a highly conserved N-terminal Rel homology domain that mediates DNA binding, dimerization, and association with IκB. The NF-κB p105 and p100 are the precursors for p50 and p52, respectively. Like IκB, p105 and p100 contain a C-terminal ankyrin repeat domain. Therefore, these two precursor proteins not only generate p50 and p52 NF-κB molecules, but also act like IκB as inhibitors for the NF-κB dimeric complex. In addition to the canonical Ub-mediated degradation of IκB, both p105 and p100 are processed proteolytically to generate p50 and p52, as well as to release the sequestered NF-κB complex through a Ub-proteasomal pathway. An early study showed that p52 is generated from a p100 precursor along with B cell development (173). Transient transfection studies in 293 T cells suggested that the NF-κB-inducing kinase (NIK) acts as an upstream kinase to induce the processing of p100 (174). NIK binds to and induces phosphorylation of p100, which is required for p100 ubiquitination and subsequent processing. A natural mutation in the NIK gene identified in aly mice markedly reduces its ability to induce p100 processing. Studies on IKKα-deficient mice further support a role of NIK in regulating NF-κB2 p100 processing (175). It seems that NIK phosphorylates specifically IKKα, which preferentially phosphorylates NF-αB2 p100. NIK-induced phosphorylation of p100 at its C-terminal region creates a binding site for β-TrCP (176). It turns out that p100 contains a very conserved serine-rich motif compared to other β-TrCP-binding proteins such as IκB. It is intriguing that the same E3 ligase component β-TrCP induces Ub conjugation to both IκB

11 Feb 2004

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

98

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

and to NF-κB2, but with totally different outcomes: Ubiquitination of IκB results in complete degradation, whereas NF-κB2 ubiquitination causes regulated proteolytic processing. A recent study may shed light on the molecular mechanism underlying the proteasomal-dependent cleavage of p100 (177). By using a yeast two-hybrid screen, a C-terminal dead domain in p100 is shown to interact with S9, a 19S proteasome subunit. The p100/S9 interaction is dependent on NIK-induced phosphorylation and ubiquitination of p100. In addition, S9 is required for NIKinduced p100 processing. The data provide a link between p100 ubiquitination and its processing via the binding of S9. Two recent studies demonstrated that triggering of two TNF receptor family members, CD40 and B cell activating factor (BAFF) receptor, induces the processing of p100 to p52 (178, 179). In 293 cells transfected with CD40 plasmid, antiCD40 cross-linking induces p52 production, which occurs 3 to 6 h after stimulation and requires de novo protein synthesis (178). The second study demonstrated a critical role of BAFF-induced p100 processing in B cell maturation (179). In NF-κB1/2 double-deficient mice, the development of B cells is blocked at the early transitional 1 (T1) stage. Using B cells deficient in BAFF receptor, NIK, or NF-κB, the authors (179) further demonstrated that BAFF regulates B cell maturation through a signaling pathway that requires BAFF receptor, NIK, and NF- κB2, and induces p100 processing. The data clearly indicate that BAFF activates NF-κB through the processing of p100 to generate p52, which correlates with a T1 stage of B cell maturation. In addition to CD40 and BAFF receptors, stimulation of the BCR also activates NF-κB signaling. In PKCβ-deficient B cells, the activation of IKKα, and to a less extent, IKKβ, are decreased upon anti-IgM stimulation (180). Although the protein expression of other Rel family members seems relatively normal, PKCβdeficient B cells display an increased amount of unprocessed p100. It appears that PKCβ functions like NIK to phosphorylate IKKα and the subsequent processing of p100, which links the BCR signaling to the noncanonical NF-κB pathway.

Processing of NF-κB1 p105 Although generation of p52 requires stimulation-dependent processing of p100, the NF-κB1 precursor p105 undergoes constitutive cleavage to produce p50, which is abundant in cells (9). At least two mechanisms have been proposed for the generation of p50, and both models require the involvement of proteasome. In the first model, p50 is generated during translation and forms a homodimer with its precursor, p105 (181, 182). This homodimer formation may stabilize the Rel homology domain and further proteolytic degradation of p105 C-terminus results in the formation of p50 homodimer. The second model proposes that p105, like p100, is processed through a β-TrCP Ub ligase-dependent processing, in which C-terminal phosphorylation of p105 by IKK is required for the p105 cleavage (183). Indeed, two C-terminal serine residues, serine-927 and serine-932, in the C-terminus of p105 are phosphorylated by IKKα and IKKβ in response to TNFα stimulation (184). However, the phosphorylated p105 has lower affinity for β-TrCP than the phosphorylated IκB, probably due to the presence of an additional

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

99

amino acid residue in p105 compared with the consensus sequence for IKK phosphorylation in IκB. In addition to a potential role of ubiquitination of p105 in proteasomal-dependent processing, it was shown that IKK-induced phosphorylation of p105 and subsequent ubiquitination by β-TrCP causes rapid degradation without concomitant production of p50 (185). It remains to be resolved how the same E3 ligase component, β-TrCP, promotes Ub conjugation to three substrates IκB, p100, and p105, but causes different outcomes: degradation for IκB, regulated processing for p100, and processing or degradation for p105. It is possible that in addition to the kinase-mediated phosphorylation of these substrates and recognition by β-TrCP, other factors may also affect the outcomes. Recently, it was shown that β-TrCP-induced ubiquitination is regulated by NEDD8 modification (186), which may provide a clue to the differential role of the SCF-β-TrCP Ub complex in the regulation of NF-κB signaling. It should also be noted that the stimulation-induced degradation of IκB occurs rapidly, whereas Ub-dependent p100 processing is a slow process, usually hours after receptor engagement. IκB degradation can be triggered by various stimuli in many types of cells; p100 processing may be restricted to certain types of stimulation and/or cells, such as B cells by CD40- or BAFF receptor-mediated signaling. Therefore, β-TrCP-mediated ubiquitination of these substrates takes place in differentially temporal and spatial fashions, which may have different biological consequences.

TGF-β SIGNALING REGULATION BY SMURFS Transforming growth factor-β (TGF-β) is a multifunctional cytokine that controls a wide spectrum of biological processes including cell proliferation, differentiation, extracellular matrix formation, and apoptosis (187). It functions from the very early stage of embryogenesis to the tissue organization and to the cancer development. In addition, TGF-β has long been known to play an important role in immune regulation (188–190): First, TGF-β is produced by a wide range of lymphoid lineage cells, such as lymphocytes, macrophages, and dendritic cells; second, like in many other types of cells, TGF-β is a suppressive cytokine of T cell proliferation and inhibits the differentiation of naive T cells into effector cells; third, it is implicated in the induction of T cell anergy, and a membrane-bound form of TGF-β is important for the inhibitory function of CD4+CD25+ regulatory T cells; and fourth, TGF-β regulates the function of antigen-presenting cells by inhibiting MHC class II expression. The involvement of TGF-β in immune regulation is further underscored by the generation of TGF-β-deficient mice, which display profound autoimmune and inflammatory responses (191, 192). TGF-β signals through transmembrane serine/threonine kinase receptors (193): TGF-β binds to the type-II receptor of TGF-β (TGF-βRII), which recruits and phosphorylates TGF-βRI. The type-I and type-II receptors form heterotetrameric complexes and transduce signaling through a group of Smad proteins. The Smads

11 Feb 2004

16:49

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

100

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

are named from the original identification of Caenorhabditis elegans Sma and Drosophila Mad proteins and are grouped into receptor-restricted Smads (RSmads, Smad1, 2, 3, 5), common-partner Smad (co-Smad, Smad4), and inhibitory Smads (I-Smads, Smad6 and 7), based on their functions. Activation of a TGF-β receptor first induces the phosphorylation of Smad2 or 3, which then dissociates from the receptor and forms a complex with the co-Smad, Smad4. The Smads are then translocated into the nucleus, which then interacts with transcriptional coactivators, such as p300 and its homologue, CREB-binding protein (CBP), or corepressors like Ski or SnoN for the transcriptional regulation of TGF-β target genes. On the other hand, TGF-β signaling upregulates the expression of the inhibitory Smads like Smad7. Smad7 is then translocated into the cytoplasm to interfere with TGF-β signaling by directly associating with TGF-β receptor and blocking the phosphorylation of the R-Smads, thus providing a negative feedback mechanism for TGF-β signaling (194, 195). By using Xenopus Smad1 as bait, Zhu et al. (196) identified an interacting protein with high homology to the HECT subclass of E3 Ub ligases, which was named as Smad ubiquitination regulatory factor-1 or Smurf-1. Smurf-1 targets ubiquitination and degradation of Smad1, thus demonstrating the first E3 ligase in Smad signaling. A subsequent database search revealed a closely related homologue, Smurf-2 (197, 198). Smurf-2 was shown to reduce the steady-state levels of Smad1 and Smad2, but not Smad3 and Smad4, in transfected cells. Structurally, the Smurf-Smad interaction is mediated by the WW domains in the Smurfs and a conserved PPXY motif in the linker regions of Smads (197). A later study suggests that Smurf-2 associates with Smad2 but does not induce a direct degradation of Smad2; rather, TGF-β induces a complex formation of Smurf-2-Smad2 with another Smad2-binding protein, SnoN (199). Smad2 then acts as an adaptor protein to facilitate the ubiquitination and proteasome-dependent degradation of downstream molecules like SnoN. SnoN is a negative repressor of Smad function by directly associating with Smad, and upon TGF-β stimulation, SnoN becomes rapidly degraded, thus allowing the Smad to activate transcription of TGF-β-responsive genes (200). Thus, during the initiation of TGF-β signaling, Smurf-2-Smad2mediated degradation of SnoN helps the transduction of intracellular signaling. On the other hand, it was shown that Smurf-1 and Smurf-2 also play a role in opposing TGF-β signaling. TGF-β stimulation causes a complex formation between Smurf2 and Smad7, an I-Smad in the nucleus (201). The Smurf-2-Smad7 complex is then exported to the cytoplasm and binds to the activated TGF-β receptor through Smad7. When a multimolecular complex forms, Smurf-2 promotes Ub transfer to the activated TGF-β receptor and its subsequent downmodulation. It was later found that Smurf-1 has a similar role in the turnover of TGF-β receptors through Smad7 adaptor function (202). Thus, Smurf proteins have a dual function in regulating TGF-β-induced signaling transduction: first as a positive regulator in the initiation of signaling through the degradation of SnoN, and second as a negative regulator in terminating the signaling through the downmodulation of the activated TGF-β receptor.

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

101

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PROTEIN UBIQUITINATION IN NOTCH SIGNALING Notch signaling mediates evolutionarily conserved cellular processes important in cell-fate decisions in diverse organisms including worms, flies, and humans (203). Notch proteins are a family of cell surface receptors that consist of an extracellular ligand binding domain, transmembrane region, and intracellular signaling domains. The extracellular domain is comprised of epidermal growth factor (EGF)-repeats, and the intracellular domain contains multiple ankyrin repeats and a PEST (proline-glutamate-serine-threonine)-rich region. Upon binding to their ligands, Jagged or Delta, Notch proteins go through a series of proteolytic cleavage, which release the intracellular domain (Notch IC) from the membrane and allow it to translocate to the nucleus, where Notch IC binds to a complex of transcription corepressors of CSL/RBP-Jκ and converts them from repressors to activators. Given the important function of Notch in cell specification in many developmental systems, it is not surprising that Notch is also implicated in the regulation of the immune system. One of the earliest studies showed that transgenic overexpression of Notch IC in the developing T cells of the mouse leads to an increase in CD8 linage T cells, but a decrease in CD4 lineage T cells (204). A later study demonstrated that Notch, together with the newly formed antigen receptor, directs the αβ versus γ δ T cell lineage decision (205). Notch also confers resistance to glucocorticoid-induced cell death in transfected thymoma cells and in DP thymocytes harboring Notch IC transgene (206). Notch signaling in DP thymocytes promotes the maturation of both CD4 and CD8 SP thymocytes (207). A role of Notch in T lineage induction is further supported by the observation of deficiency in thymocyte development by induced inactivation of Notch gene (208). Retroviral transduction of Notch IC in the bone marrow promotes T lineage development, but inhibits B lineage commitment (209). A later study showed that Notch IC in fact abrogates differentiation of immature DP thymocytes into both CD4 and CD8 mature SP T cells (210). Such inhibitory function is accounted for by the interference with TCR signal strength. However, another study employing Notch antisense and anti-Notch antibody approaches showed that Notch signaling affects the duration of TCR signaling, and it is selectively required for CD8 T cell maturation (211). The discrepancies among the published data may be due to the different approaches used for the studies. Another explanation may be that the gain-of-function experiments mostly using constitutively active Notch IC do not represent the physiological function of the intact Notch protein. Indeed, a recent study using a conditional knockout technique showed that inactivation of Notch1 in immature thymocytes does not affect CD4+ or CD8+ T cell development (212), but αβ T cell lineage development (213). Because Notch has four isoforms (Notch 1 to 4) and at least three of them (Notch 1 to 3) are expressed in the thymus, it can be imagined that there might be overlapping redundancy among Notch proteins (214). To further complicate this issue, of the Notch ligands (Jagged 1 and 2, Delta-like-1 and 3), both Jagged 1 and 2 are known to be present in the thymus (215). It becomes obvious that an in vitro system is required to recapitulate the

11 Feb 2004

16:49

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

102

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

in vivo findings. An elegant study recently reported that ectopic expression of a Notch ligand, Delta-like-1, in a bone marrow stromal cell line, results in the loss of its ability to support B cell development, but it can help generate mature T cells (216). The data at least support a biological function of Notch-Delta interaction in T lineage cell commitment and differentiation. Given the conserved role of Notch in diverse organisms, it is no surprise that such important Notch signaling is regulated by protein ubiquitination. The first evidence that ubiquitination has a role in the downregulation of Notch signaling came from a genetic study in Drosophila (217). Mutations in Deltex, a Notch regulator, results in a phenotype resembling loss-of-function Notch mutation, which can be suppressed by mutation in another gene, suppressor of Deltex [Su(dx)], suggesting that Su(dx) is a genetically negative regulator of Notch signaling. Subsequent cloning of the Su(dx) revealed that it belongs to the HECT-type E3 ligases, with a typical Nterminal C2 domain, four WW domains, and a C-terminal HECT ligase domain. Inspired by this genetic observation, we examined whether Itch, a mammalian homologue of Su(dx), is involved in Notch signaling in cultured T cells (218). It was found that Itch induces Ub conjugation to Notch both in vitro and in vivo (218), suggesting that Itch is an E3 ligase for Notch in mammalian cells. Three later studies identified another E3 ligase component, Sel-10, as an inhibitor of Notch signaling that targets Notch for Ub-mediated degradation (219–221). Sel-10 protein contains an F-Box and seven WD-40 repeats, and belongs to the subgroup of SCF E3 ligase complexes. Sel-10 interacts with Notch through its WD-40 repeats and a region close to the PEST sequences in Notch, and the interaction is dependent on Notch phosphorylation. Interestingly, it seems that the Notch interaction with, and its ubiquitination by, Sel-10 takes place in the nucleus (219, 221). This suggests that Sel-10 is responsible for the proteasomal-dependent degradation after a nuclear phosphorylation event. More recently, Cbl was added to the list of E3 Ub ligases in Notch ubiquitination (222). Notch is tyrosine-phosphorylated after stimulation and becomes associated with Cbl. The data suggest another layer of Notch regulation by modulating its downmodulation at the cell surface. Because of its conserved role in regulating cell differentiation in diverse cell types and organisms, it can be postulated that Notch is tightly regulated by different mechanisms. In this context, it is not surprising that several different E3 ligases participate in the Ub conjugation to Notch, either at the cell surface, intracellular cytoplasm, or the nucleus, through lysosomal- or proteasomal-dependent degradation pathways. It is important to keep in mind, however, that almost all the published data are based on overexpression studies in cultured cells, even though the genetic observation in Drosophila may support such a conclusion. Obviously, genetic studies in mice are required to further confirm such biochemical observations. In addition to regulation of Notch by ubiquitination, studies from several groups have documented that the Notch ligand Delta is also targeted for protein ubiquitination (223–225). The protein product, Neur, contains a C-terminal RING finger that acts as an E3 ligase to promote Ub conjugation to the Notch ligand, Delta.

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

103

Ubiquitination of Delta enhances its endocytosis and decreases the cell surface levels of Delta in Drosophila. Interestingly, endocytosis of Delta also triggers the transendocytosis and the ectodomain shedding of Notch, which is critical for the activation of Notch signaling. Thus, Neur plays a positive role in Notch pathway by promoting ubiquitination of Delta. Similar to the function of Neur, a recent study showed that another RING-type E3 ligase, Mind bomb, is also able to promote Delta ubiquitination and internalization (226). Mind bomb contains multiple ankyrin repeats and three C-terminal RING domains. It directly associates with the intracellular regions of Delta and induces Delta ubiquitination. Together with Neur, Mind bomb participates in modulating Notch signaling by inducing a Ubdependent endocytosis of Delta.

UBIQUITINATION IN WNT SIGNALING PATHWAY Similar to the Notch pathway, Wnt-induced signaling represents another wellconserved cascade that plays important roles in cell fate specification, proliferation, and cell death (227). Wnt proteins are a family of secreted glycoproteins that bind to two groups of cell surface receptors: the Frizzled and the low-density lipoprotein-related protein. In the absence of Wnt, the intracellular protein βcatenin forms a complex with a number of proteins including the scaffold protein Axin, the tumor suppressor adenomatous plyposis coli, and the kinase glycogen synthase kinase-3β (GSK-3β). β-catenin is phosphorylated by GSK-3β at the N-terminal serine/threonine residues in this complex. Phosphorylation at these sites recruits β-TrCP (228–230). Thus, the SCF β-TrCP complex induces the ubiquitination of β-catenin and its degradation. When Wnt binds to the receptors, it induces the inhibition of the kinase activity of GSK-3β through another protein, Disheveled. β-catenin is then stabilized, translocated to the nucleus, and binds to transcription factors of T cell factor (TCF) or lymphoid enhancer factor (LEF). Much of our understanding on Wnt signaling in lymphocyte regulation is based on the studies of the TCF/LEF transcription factors. In unstimulated cells, TCF/LEF factors interact with transcriptional co-repressors to repress target gene transcription. In the presence of Wnt, β-catenin binds to TCF/LEF and switches from repression to transactivation. TCF/LEF factors have DNA-binding activity to the enhancers of the TCR complex components CD3ε and TCRα (231). Gene targeting of TCF-1 results in an age-dependent block of early T cell development, but normal proliferation and function of peripheral T cells (232, 233). TCF mutant LEF-1 knockout mice display more severe defects, with a complete block at the immature single-positive stage (234). A more recent study directly linked β-catenin to thymocyte development (235). In this study, a region of β-catenin containing the phosphorylation sites was conditionally deleted, which resulted in accumulation of β-catenin mutant protein in thymocytes. Surprisingly, stabilization of β-catenin in thymocytes led to the generation of single positive T cells lacking the αβ TCR

11 Feb 2004

16:49

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

104

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

expression in a Rag-2-/- background. Another recent study provided further information on the transgenic expression of TCF isoforms and found that only the intact form containing β-catenin-binding domain can rescue the thymocyte development in TCF-deficient mice (236). These studies together suggest that Wnt signaling is an important regulator in early thymocyte development, and β-catenin, whose stability is tightly controlled by protein ubiquitination, is a critical mediator to transduce Wnt signaling. In addition to its role in cell fate determination, β-catenin is also a protooncogene whose mutations have been associated with human cancers (237). Independently, two groups have demonstrated that another E3 ligase component, Siah, mediates β-catenin degradation (238, 239). Siah contains an N-terminal RING finger that binds to E2 and targets several proteins for proteasomal-dependent degradation (240–243). Siah is present at a very low amount in normal cells. However, activation of p53 by genotoxic stress or irradiation upregulates Siah expression (244). Thus, Siah may act as a mediator for p53-induced cell cycle arrest and tumor suppression through targeting Ub-dependent proteolysis of β-catenin. However, the results derived from gene-targeted mice seem at odds with the biochemical observations. Cells deficient in Siah proteins are relatively normal in p53 function (245). The genetic evidence may suggest that Siah targets other protein substrates than β-catenin for protein ubiquitination. Indeed, the same group showed that male mice deficient in Siah display a block in spermatogenesis, although the exact physiological targets for Siah E3 ligase activity are not known at present (246).

MDM2 AS AN E3 LIGASE FOR p53 p53 is a key tumor suppressor whose mutation is linked to many human cancers (247). The importance of p53 is further underscored by the development of multiple malignancies including the dominant occurrence of lymphoma in p53 gene-targeted mice (248). p53 acts as a transcription factor to induce gene transcription of multiple proteins involved in cell cycle arrest and apoptosis (247). In addition to its role as a tumor suppressor, p53 has been implicated in physiological immune regulations such as thymocyte development (249), and peripheral mature T cell apoptosis triggered by TCR engagement (250), whereas the role of p53 in regulating the expansion of antigen-specific CD8+ T cells seems to be relatively minor (251). In normal cells, p53 is kept at a very low level through an autoregulatory feedback loop by inducing the expression of Mdm2. Mdm2 is RING fingercontaining E3 ligase that promotes Ub conjugation to p53 and subsequent degradation via proteasome-mediated proteolysis (41, 252). The Ub conjugation region was mapped to several lysine residues at the extreme C-terminus of p53 (253, 254), and ubiquitination of these lysine residues by Mdm2 could unmask the buried nuclear export signal in p53 through its conformational changes and initiate the export

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

105

of p53 into the cytoplasm, where it is degraded by the 26S proteasome (255–257). To support a role of Mdm2 in regulating Mdm2 stability, conditional knockout of Mdm2 in mice leads to a greater accumulation of p53 and increased apoptosis in response to ionizing radiation (258). Interestingly, the mutant mice exhibit defects in multiple hematopoietic lineages, including the decreases in T and B lymphocyte numbers, and in the sizes of the thymus, lymph nodes, and the spleen. The genetic study clearly indicates an important role of Mdm2 in hematopoietic tissues, particularly during lymphopoiesis. Exposure of the cells to DNA damage such as ionizing irradiation or stress stimulation causes the activation of p53, primarily through the regulation of p53 stability (259). One current model is that DNA damage can activate a checkpoint kinase Chk2, which in turn induces the phosphorylation of p53 at serine-20. Phosphorylation of this serine residue disrupts the interaction of 53 with Mdm2 and thus increases the p53 stability (260, 261). Supporting this model, Chk2-deficient thymocytes were resistant to DNA damage–induced apoptosis and were defective for p53 stabilization following irradiation (262). In addition to serine-20, there are other potential serine phosphorylation sites in p53 that may also contribute to the regulation of p53 stability. Indeed, thymocytes harboring serine-20 mutation showed normal p53 stability and p53-dependent apoptosis (263). Another mechanism by which p53 is stabilized in response to DNA damage is acetylation of p53 by CBP/p300 (259). Interestingly, p53 is acetylated at the C-terminal lysine residues that overlap with Mdm2-mediated ubiquitination sites. Thus, acetylation of p53 potentially attenuates its Ub conjugation by Mdm2 and increases its stability (264). Once p53 has exerted its transcriptional activity, it needs to be deacetylated. Recent studies suggest that Mdm2 inhibits CBP/p300mediated acetylation (265). Furthermore, Mdm2 recruits the deacetylase HDAC1, which removes the acetyl groups from the p53 lysine residues and allows Mdm2 to promote p53 degradation (266). Thus, Mdm2 plays a dual role in safeguarding the p53 protein level by promoting direct Ub conjugation to p53 and by recruiting deacetylase to trigger p53 deacetylation. Contrary to a role of p300 in stabilizing p53 through acetylation, a recent study demonstrated that p300 acts as an E4 and together with Mdm2 to catalyze the polyubiquitination of p53, whereas Mdm2 only induces p53 monoubiquitination (267). In addition, a Parkin-like E3 ligase, Parc, and another RING-type E3, Pirh2, have been recently identified as the E3 ligases for p53 (268, 269). These new data suggest that the stability of p53 is regulated by multiple E3 ligases.

UBIQUITINATION IN THE PROCESSING OF ANTIGENIC PEPTIDE T cell activation is a central step in the development of the adaptive immune response, which is initiated by the binding of the TCR to the antigenic peptide presented by MHC on the antigen-presenting cells. There are two classes

11 Feb 2004

16:49

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

106

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

of MHCs: The MHC class I presents endogenous antigenic fragment to CD8+ T cells, and the MHC class II presents exogenous antigen peptide to the CD4+ T cells (270, 271). MHC class I is expressed in almost any cells in which antigens such as virus proteins or tumor antigens are processed in the cytosol and translocated into the endoplasmic reticulum to associate with the class I complex (270). MHC class II is only expressed in professional antigen-presenting cells, such as macrophages, dendritic cells, or B cells, where it picks up exogenous antigenic peptide derived from bacteria or dead cell debris processed in endolysomal compartments (271). The generation of antigenic peptide for MHC class I presentation is dependent on proteasome-dependent processing (11). By using cells with a temperaturesensitive defect in E1 Ub–activating enzyme, it was shown that the Ub-dependent proteolytic pathway is required for MHC class I–restricted presentation of ovalbumin introduced into the cytosol (13). In addition, during viral infection, IFNγ is produced and causes the production of two MHC-encoded enzymatic subunits— LMP2 and LMP7—and the non–MHC encoded MECL-1, which replace three proteasome subunits in the 20S catalytic proteasome to form the immunoproteasome (272, 273). The newly assembled immunoproteasome alters peptidase activity and produces antigenic peptides more suitable for binding to MHC class I (274). To further support a role of the proteasome in antigen presentation, a proteasomespecific inhibitor, lactacystin, blocks presentation of influenza-derived viral epitopes to cytotoxic T cells (275). More recently, it was shown that ubiquitinated proteins accumulate during dendritic cell maturation (276). Inflammation causes the transient formation of the Ub-containing aggresome-like structure, which requires continuous protein synthesis and stems from endogenous proteins because endocytosized exogenous protein is not incorporated into the structure. Such aggregate formation in dendritic cells may represent an antigen storage structure, which may have impact in determining MHC class I loading and peptide presentation. Although a relationship between the Ub-proteasome pathway and MHC class I antigen presentation is well established, it is not clear whether the generation of antigenic peptide is truly dependent on the classical Ub conjugation pathway. Indeed, it was also shown that presentation of both endogenous and exogenous antigens is not affected by the inactivation of the E1 Ub-activating enzyme (277). In addition, newly synthesized defective ribosomal products, or DRiPs, rather than stable intracellular proteins, are the major source of self- and viral peptides (278). Supporting this notion, nearly 30% of newly synthesized proteins are DRiPs, and some of these protein fragment products are tagged with Ub (279). During an acute influenza infection, the transporter associated with antigen processing (TAP) is rapidly loaded with newly synthesized proteins, which may ensure a timely presentation of viral antigen to the immune system (280). However, it remains unclear whether the E2 or E3 Ub conjugation components are involved in antigen processing and presentation by the MHC class I. In this regard, it has been shown that tumor cells of many types present antigenic peptides derived from both p53 and Mdm2 (281, 282). It might be possible that Mdm2 also helps process both p53

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

107

and itself as tumor antigens in the cytosol, which allows the presentation by MHC class I to cytotoxic T cells under pathological conditions.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

VIRAL E3 LIGASES IN IMMUNE EVASION Although the immune system has evolved machinery to eradicate virally infected cells, many viruses can persist inside cells to cause latent or chronic infection of the host. DNA viruses like human cytomegalovirus (hCMV) or herpesvirus are examples of viruses causing immune evasion. One mechanism used by the virus to avoid immune surveillance is the downregulation of MHC class I to reduce antigen presentation (283). For example, the US11 and US2 proteins of hCMV cause reverse translocation of MHC class I from the endoplasmic reticulum (ER) into the cytosol, where MHC class I is polyubiquitinated and degraded by the proteasome-dependent pathway (284, 285). Another mechanism of viral subversion is the identification of virus-encoded PHD-containing E3 Ub ligases. Several viral proteins contain a PHD domain, which is involved in the downmodulation of antigen-presenting MHC class I molecules (286, 287). Two herpesvirus proteins, K3 and K5, have E3 Ub ligase activity and downregulate cell surface molecules such as MHC class I (288, 289). The human herpesvirus K3 and K5 proteins have E3 ligase activity to induce Ub conjugation to MHC class I in transfected cells, and a critical cysteine residue in the PHD domain of K5 is required for its self-ubiquitination in vitro (288). Similarly, the K3 of murine γ -herpesvirus also downmodulates the class I molecule in a PHD domain–dependent manner (289). The C-terminal domain of K3 and the cytoplasmic tail of class I were essential for the association of K3 with the newly synthesized class I molecule in the ER membrane and its subsequent degradation. The K3-mediated Ub conjugation of class I results in subsequent binding by Tsg101, an E2 variant protein implicated in trafficking of membrane-associated proteins into the endolysosomal pathway (290). The K3 of murine γ -herpesvirus requires some of the MHC class I assembly components to exert its function (291), as it failed to downmodulate class I in cells deficient in TAP or tapasin and it directly associates with TAP and tapasin in the absense of class I. It seems that the K3 or K5 uses TAP and/or tapsin to form an indirect complex with class I that helps transfer Ub to the class I molecule. The PHD-containing viral E3 ligases also induce downmodulation of other cell surface molecules in addition to MHC class I. The cell surface molecules such as B7 and ICAM-1, which are important for the co-stimulation of T cells, are downregulated by K5 of human herpesvirus (288). Another PHD domain protein, M153R of myxomavirus, which causes myxomatosis in rabbit, induces internalization and lysosomal degradation of CD4 in T cells (292) as well as MHC class I (293). Viral proteins can also use the cellular E3 ligases to induce CD4 downregulation. The Vpu membrane protein of human immunodeficiency virus type-1 (HIV-1) causes downmodulation of CD4, which is dependent on an intact Ub conjugation system

11 Feb 2004

16:49

108

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

and C-terminal lysine residues in CD4 (294). Vpu interacts with the E3 ligase component of β-TrCP through a C-terminal intracellular domain of Vpu and the WD repeat of β-TrCP. The Ub is then transferred through a β-TrCP SCF complex to CD4 through the adaptor function of Vpu, which leads to CD4 ubiquitination (295). Taken together, it can be concluded that Ub-dependent downmodulation of cell surface proteins induced by either viral or cellular E3 Ub ligases represents an important means for the virus to evade immune surveillance.

UB AND RETROVIRUS BUDDING Not only do viruses contain E3 Ub ligases to downmodulate MHC class I and other cell surface molecules, some enveloped RNA viruses also exploit the cellular ubiquitination machinery for their own benefit. The critical steps in the retrovirus life cycle are virion assembly and fusion with the plasma membrane of infected cells, and the final release or budding from the host cell (296). Studies from three independent groups showed that proteasomal inhibitors interfere with the budding process of HIV, Rous sarcoma virus, or Ebola virus (297–299). The retroviral Gag protein is essential for the budding process, and it contains conserved motifs (or late budding domains) like PPXY or PTAP sequences (299). Using the PPXY motif as bait, several groups subsequently identified WW domain–containing HECT E3 ligases such as Nedd4 or Nedd4-like proteins (300–302). A functional E3 ligase is required for viral budding because expression of the WW domains alone inhibits the release of mature viral particles from the host cell (301, 302). In the case of Ebola virus, the late domain–containing VP40 protein is a target for Ub conjugation by the Nedd4 yeast homologue Rsp5 (300). Thus, HECT-type E3 ligases are involved in viral budding through their recruitment to the PPXY motifs in the late budding domains. Another exciting discovery on the Ub system involves HIV budding. Unlike other retroviruses, HIV does not contain the PPXY motif in the Gag protein, but has a PTAP motif, which has been implicated in viral budding (303). By using the yeast two-hybrid screen with the HIV Gag as bait, two groups identified Tsg101 as a binding partner (21, 304). The PATP motif in the HIV Gag protein mediates direct binding with the UEV domain in the N-terminus of Tsg101. More importantly, depletion of Tsg101 by using small interfering RNAs significantly reduces budding of HIV from infected cells (21). Structure studies of the Tsg101 UEV domain in the complex with the HIV PTAP motif demonstrated that the UEV domain forms a binding groove that makes close contacts with the PTAP residues (305). The importance of Tsg101 in HIV budding is further supported by the finding that overexpression of the N-terminal UEV domain of Tsg101 inhibits HIV budding, suggesting that the full-length Tsg101 is required for facilitating virus release (306). As described earlier, Tsg101 belongs to the UEV family that contains E2 homologous sequences but not the active cysteine, and it functions in late endosomal trafficking for the activated EGF receptor in mammalian

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

109

cells (20). In yeast, the Tsg101 homologue vacuolar-protein sorting 23 (Vps23) forms a multimolecular complex with other Vps proteins, called ESCRT-1, which recognizes monoubiquitinated cargo and helps sort it into multivesicular bodies (MVBs) (307). Interestingly, the MVB pathway is quite similar to the viral budding process, both of which involve invagination of lipid membrane. In addition, a recent study showed that Tsg101 is recruited together with other ESCRT-1 components to the late budding domain (308). One distinctive difference is that Tsg101 directly binds to the PTAP motif in HIV, instead of the ubiquitinated cargo protein, as in the MVB pathway in the yeast. Enveloped viruses therefore can utilize at least two cellular Ub machineries for release from the host cell: One is the recruiting of HECT-type E3 ligases by the PPXY motif, and another is the binding of Tsg101 and the ESCRT-1 complex. However, these two pathways may be not mutually exclusive because Tsg101 also contains a weak Ub binding site (309), and Ub conjugation to HIV Gag protein in fact enhances Tsg101 binding (21). Some viruses, like Ebola, contain both PPXY and PTAP motifs (299); the PTAP motif in Ebola indeed binds Tsg101, and this interaction is essential for virion release (310). Moreover, the late domain of Ebola virus recruits both Nedd4 and Tsg101 for effective budding (311, 312). However, the exact mechanism for the interplay between the late domains of budding viruses and cellular Ub factors such as Tsg101 or E3 ligases remains to be elucidated.

CONCLUDING REMARKS AND PERSPECTIVES Our understanding of Ub ligases in the regulation of cellular functions has tremendously improved with the identification of new E3 ligases and their specific substrates over the past few years. At the same time, it is clear that Ub conjugation to a substrate may not always leads to its degradation, but instead to proteolysisindependent functional modification. Ubiquitination of cell surface receptors may be a means of terminating signaling, whereas at other times receptor-induced ubiquitination of intracellular proteins is an essential step to initiate signaling transduction. Likewise, protein ubiquitination does not always play a negative role, but sometimes positively regulates kinase activation, protein-protein interaction, or gene transcription. Because the numbers of E3s are huge, it can be imagined that all these diverse biological functions primarily rely on the specificity of the E3s, in addition to the differential uses of Ub or Ub-like proteins, as well as the selection of Ubc E2s. However, the question of how a particular E3 mediates Ub transfer to a substrate and then induces totally different outcomes still awaits elucidation. In regard to the immune response, we now know that E3s are implicated in many aspects of both innate and adaptive immunity, such as the downregulation of antigen receptors, modulation of intracellular signaling molecules and transcription factors, and antigen processing or MHC expression (Table 1). Biochemical approaches will still be critical for the identification of the substrates for a particular E3 in lymphocytes. However, one has to be aware that a substrate identified in an

11 Feb 2004

16:49

110

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU TABLE 1 Selected E3 ligases, their substrates, and functional implications E3s HECT-type E6-AP Nedd4

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Smurf 1, 2 Itch

RING-type Cbl

Cbl-b MDM2 TRAF IAP SCF/β-TrCP

Siah VBC/VHL Viral K3, K5 ?? GRAIL

Substrates

Function

Src kinase Steroid hormone receptor Na channel Gag protein Smads, SnoN, TGF-β receptor Notch Jun-B IL-7 receptor PLCγ 1, PKCθ

B cell signaling Transcription Hypertension Virus budding TGF-β signaling ?? T helper cell differentiation B cell development T cell anergy

TCR/CD3 Syk/Zap70 and Lck Crk-L p85 of PI3-K Crk-L p53 IKK complex TRAF, Self IkB NF-kB (p100 and p105) β-catenin β-catenin HIFα MHC class I Antigen (native or unfolded) ??

Receptor downregulation Tyrosine kinase degradation Rap1 activation T cell activation Rap1 activation Cell cycle control NF-κB activation Apoptosis Degradation Processing Cell cycle/T cell development Cell cycle/T cell development Angiogenesis/inflammation Immune evasion Antigen presentation T cell anergy

in vitro system may not represent a physiologic target in vivo. Genetic approaches using mice deficient in an E3 ligase will definitely be important to complement the biochemical analysis. On the other hand, because lymphocytes go through a series of developmental and selection steps, the phenotypes revealed in the gene-targeted mice may not be the direct result of E3 deficiency, but rather a secondary effect. Thus, a combination of biochemical and genetic analysis is required to fully appreciate the biological significance of an E3 in the immune regulation. At present, we know extremely little about the regulation of E3 ligases, although recent studies have pointed out that conjugation by the Ub-like molecule Nedd8 to, or its removal from, the SCF E3 complex modulates the ligase activity of the latter (313, 314). Similarly, reversal of Ub conjugation or deubiquitination is emerging as another important regulatory mechanism for Ub-dependent protein modification. Further understanding of these molecular insights may lead to new therapeutic approaches for the treatment of immunological abnormalities. Obviously, we are

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

P1: IKH

111

facing many challenges ahead, and at the same time we are also provided with many opportunities to undertake steps to gradually fill in the gaps in our knowledge on this complex Ub conjugation system. As the Chinese proverb goes, the journey of a thousand miles starts with a single step.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ACKNOWLEDGMENTS I would like to thank former and present members of my laboratory for their contributions to the E3 Ub ligase projects, V. Heissmeyer, A. Rao, and D. Fang for the unpublished data, and my colleagues at the Division of Cell Biology for support. I apologize to those whose important literature is not in this review or is not properly cited in this review. The research in this laboratory is supported by the National Institutes of Health grants RO1DK56558, RO1AI50280, and R21AI48542, a research project grant from the American Cancer Society, and a biomedical science grant from the Arthritis Foundation. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Goldstein G, Scheid M, Hammerling U, Schlesinger DH, Niall HD, Boyse EA. 1975. Isolation of a polypeptide that has lymphocyte-differentiating properties is probably represented universally in living cells. Proc. Natl. Acad. Sci. USA 72:11–15 2. Hershko A, Ciechanover A. 1998. The ubiquitin system. Annu. Rev. Biochem. 67:425–79 3. Pickart CM. 2001. Mechanisms underlying ubiquitination. Annu. Rev. Biochem. 70:503–33 4. Weissman AM. 2001. Themes and variations on ubiquitylation. Nat. Rev. Mol. Cell Biol. 2:169–78 5. Schwartz AL, Ciechanover A. 1999. The ubiquitin-proteasome pathway and pathogenesis of human diseases. Annu. Rev. Med. 50:57–74 6. Hicke L. 2001. Protein regulation by monoubiquitin. Nat. Rev. Mol. Cell Biol. 2:195–201 7. Janeway CA Jr, Medzhitov R. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197–216

8. Medzhitov R, Janeway CA Jr. 1998. Innate immune recognition and control of adaptive immune responses. Semin. Immunol. 10:351–53 9. Karin M, Ben-Neriah Y. 2000. Phosphorylation meets ubiquitination: the control of NF-[kappa]B activity. Annu. Rev. Immunol. 18:621–63 10. Ben-Neriah Y. 2002. Regulatory functions of ubiquitination in the immune system. Nat. Immunol. 3:20–26 11. Rock KL, Goldberg AL. 1999. Degradation of cell proteins and the generation of MHC class I-presented peptides. Annu. Rev. Immunol. 17:739–79 12. McGrath JP, Jentsch S, Varshavsky A. 1991. UBA 1: an essential yeast gene encoding ubiquitin-activating enzyme. EMBO J. 10:227–36 13. Michalek MT, Grant EP, Gramm C, Goldberg AL, Rock KL. 1993. A role for the ubiquitin-dependent proteolytic pathway in MHC class I-restricted antigen presentation. Nature 363:552–54 14. Johnson ES, Blobel G. 1997. Ubc9p is the conjugating enzyme for the

11 Feb 2004

16:49

112

15.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

16.

17.

18.

19.

20.

21.

22.

23.

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU ubiquitin-like protein Smt3p. J. Biol. Chem. 272:26799–802 Schwarz SE, Matuschewski K, Liakopoulos D, Scheffner M, Jentsch S. 1998. The ubiquitin-like proteins SMT3 and SUMO-1 are conjugated by the UBC9 E2 enzyme. Proc. Natl. Acad. Sci. USA 95:560–64 Muller S, Hoege C, Pyrowolakis G, Jentsch S. 2001. SUMO, ubiquitin’s mysterious cousin. Nat. Rev. Mol. Cell Biol. 2:202–10 Hofmann RM, Pickart CM. 1999. Noncanonical MMS2-encoded ubiquitinconjugating enzyme functions in assembly of novel polyubiquitin chains for DNA repair. Cell 96:645–53 Deng L, Wang C, Spencer E, Yang L, Braun A, et al. 2000. Activation of the IkappaB kinase complex by TRAF6 requires a dimeric ubiquitin-conjugating enzyme complex and a unique polyubiquitin chain. Cell 103:351–61 Li L, Cohen SN. 1996. Tsg101: a novel tumor susceptibility gene isolated by controlled homozygous functional knockout of allelic loci in mammalian cells. Cell 85:319–29 Babst M, Odorizzi G, Estepa EJ, Emr SD. 2000. Mammalian tumor susceptibility gene 101 (TSG101) and the yeast homologue, Vps23p, both function in late endosomal trafficking. Traffic 1:248– 58 Garrus JE, von Schwedler UK, Pornillos OW, Morham SG, Zavitz KH, et al. 2001. Tsg101 and the vacuolar protein sorting pathway are essential for HIV-1 budding. Cell 107:55–65 Scheffner M, Huibregtse JM, Vierstra RD, Howley PM. 1993. The HPV-16 E6 and E6-AP complex functions as a ubiquitin-protein ligase in the ubiquitination of p53. Cell 75:495–505 Huibregtse JM, Scheffner M, Beaudenon S, Howley PM. 1995. A family of proteins structurally and functionally related to the E6-AP ubiquitin-protein lig-

24.

25.

26.

27.

28.

29.

30.

31.

32.

ase. Proc. Natl. Acad. Sci. USA 92:2563– 67 Talis AL, Huibregtse JM, Howley PM. 1998. The role of E6AP in the regulation of p53 protein levels in human papillomavirus (HPV)-positive and HPVnegative cells. J. Biol. Chem. 273:6439– 45 Harris KF, Shoji I, Cooper EM, Kumar S, Oda H, Howley PM. 1999. Ubiquitinmediated degradation of active Src tyrosine kinase. Proc. Natl. Acad. Sci. USA 96:13738–43 Oda H, Kumar S, Howley PM. 1999. Regulation of the Src family tyrosine kinase Blk through E6AP-mediated ubiquitination. Proc. Natl. Acad. Sci. USA 96:9557–62 Albrecht U, Sutcliffe JS, Cattanach BM, Beechey CV, Armstrong D, et al. 1997. Imprinted expression of the murine Angelman syndrome gene, Ube3a, in hippocampal Purkinje neurons. Nat. Genet. 17:75–78 Nawaz Z, Lonard DM, Smith CL, LevLehman E, Tsai SY, et al. 1999. The Angelman syndrome-associated protein, E6-AP, is a coactivator for the nuclear hormone receptor superfamily. Mol. Cell Biol. 19:1182–89 Pirozzi G, McConnell SJ, Uveges AJ, Carter JM, Sparks AB, et al. 1997. Identification of novel human WW domaincontaining proteins by cloning of ligand targets. J. Biol. Chem. 272:14611– 16 Sudol M. 1996. The WW module competes with the SH3 domain? Trends Biochem. Sci. 21:161–63 Lu PJ, Wulf G, Zhou XZ, Davies P, Lu KP. 1999. The prolyl isomerase Pin1 restores the function of Alzheimerassociated phosphorylated tau protein. Nature 399:784–88 Staub O, Dho S, Henry P, Correa J, Ishikawa T, et al. 1996. WW domains of Nedd4 bind to the proline-rich PY motifs in the epithelial Na+ channel deleted in

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

33. 34.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

35.

36.

37.

38.

39.

40.

41.

42.

Liddle’s syndrome. EMBO J. 15:2371– 80 Freemont PS. 2000. RING for destruction? Curr. Biol. 10:R84–87 Joazeiro CA, Weissman AM. 2000. RING finger proteins: mediators of ubiquitin ligase activity. Cell 102:549–52 Lupher ML Jr, Rao N, Eck MJ, Band H. 1999. The Cbl protooncoprotein: a negative regulator of immune receptor signal transduction. Immunol. Today 20:375– 82 Joazeiro CA, Wing SS, Huang H, Leverson JD, Hunter T, Liu YC. 1999. The tyrosine kinase negative regulator c-Cbl as a RING-type, E2-dependent ubiquitinprotein ligase. Science 286:309–12 Levkowitz G, Waterman H, Ettenberg SA, Katz M, Tsygankov AY, et al. 1999. Ubiquitin ligase activity and tyrosine phosphorylation underlie suppression of growth factor signaling by c-Cbl/Sli-1. Mol. Cell 4:1029–40 Yokouchi M, Kondo T, Houghton A, Bartkiewicz M, Horne WC, et al. 1999. Ligand-induced ubiquitination of the epidermal growth factor receptor involves the interaction of the c-Cbl RING finger UbcH7. J. Biol. Chem. 274: 31707–12 Zheng N, Wang P, Jeffrey PD, Pavletich NP. 2000. Structure of a c-Cbl-UbcH7 complex: RING domain function in ubiquitin-protein ligases. Cell 102:533– 39 Huang L, Kinnucan E, Wang G, Beaudenon S, Howley PM, et al. 1999. Structure of an E6AP-UbcH7 complex: insights into ubiquitination by the E2-E3 enzyme cascade. Science 286:1321–26 Fang S, Jensen JP, Ludwig RL, Vousden KH, Weissman AM. 2000. Mdm2 is a RING finger-dependent ubiquitin protein ligase for itself and p53. J. Biol. Chem. 275:8945–51 Tyers M, Willems AR. 1999. One ring to rule a superfamily of E3 ubiquitin ligases. Science 284:601–4

P1: IKH

113

43. Zheng N, Schulman BA, Song L, Miller JJ, Jeffrey PD, et al. 2002. Structure of the Cul1-Rbx1-Skp1-F boxSkp2 SCF ubiquitin ligase complex. Nature 416: 703–9 44. Jackson PK, Eldridge AG, Freed E, Furstenthal L, Hsu JY, et al. 2000. The lore of the RINGs: substrate recognition and catalysis by ubiquitin ligases. Trends Cell Biol. 10:429–39 45. Ivan M, Kaelin WG Jr. 2001. The von Hippel-Lindau tumor suppressor protein. Curr. Opin. Genet. Dev. 11:27– 34 46. Stebbins CE, Kaelin WG Jr, Pavletich NP. 1999. Structure of the VHL-ElonginC-ElonginB complex: implications for VHL tumor suppressor function. Science 284:455–61 47. Kamura T, Koepp DM, Conrad MN, Skowyra D, Moreland RJ, et al. 1999. Rbx1, a component of the VHL tumor suppressor complex SCF ubiquitin ligase. Science 284:657–61 48. Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, et al. 1999. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygendependent proteolysis. Nature 399: 271–75 49. Ohh M, Park CW, Ivan M, Hoffman MA, Kim TY, et al. 2000. Ubiquitination of hypoxia-inducible factor requires direct binding to the beta-domain of the von Hippel-Lindau protein. Nat. Cell Biol. 2:423–27 50. Ivan M, Kondo K, Yang H, Kim W, Valiando J, et al. 2001. HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science 292:464–68 51. Jaakkola P, Mole DR, Tian YM, Wilson MI, Gielbert J, et al. 2001. Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 292:468– 72 52. Yu F, White SB, Zhao Q, Lee FS. 2001.

11 Feb 2004

16:49

114

53.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

54.

55.

56.

57.

58.

59.

60.

61.

62.

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU HIF-1alpha binding to VHL is regulated by stimulus-sensitive proline hydroxylation. Proc. Natl. Acad. Sci. USA 98:9630–35 Min JH, Yang H, Ivan M, Gertler F, Kaelin WG Jr, Pavletich NP. 2002. Structure of an HIF-1alpha -pVHL complex: hydroxyproline recognition in signaling. Science 296:1886–89 Hon WC, Wilson MI, Harlos K, Claridge TD, Schofield CJ, et al. 2002. Structural basis for the recognition of hydroxyproline in HIF-1 alpha by pVHL. Nature 417:975–78 Semenza GL. 2001. HIF-1 mechanisms of hypoxia sensing. Curr. Opin. Cell Biol. 13:167–71 Mandriota SJ, Turner KJ, Davies DR, Murray PG, Morgan NV, et al. 2002. HIF activation identifies early lesions in VHL kidneys: evidence for site-specific tumor suppressor function in the nephron. Cancer Cell 1:459–68 Maranchie JK, Vasselli JR, Riss J, Bonifacino JS, Linehan WM, Klausner RD. 2002. The contribution of VHL substrate binding and HIF1-alpha to the phenotype of VHL loss in renal cell carcinoma. Cancer Cell 1:247–55 Mack FA, Rathmell WK, Arsham AM, Gnarra J, Keith B, Simon MC. 2003. Loss of pVHL is sufficient to cause HIF dysregulation in primary cells but does not promote tumor growth. Cancer Cell 3:75–88 Hochstrasser M. 2001. SP-RING for SUMO: new functions bloom for a ubiquitin-like protein. Cell 107:5–8 Jackson PK. 2001. A new RING for SUMO: wrestling transcriptional responses into nuclear bodies with PIAS family E3 SUMO ligases. Genes Dev. 15: 3053–58 Johnson ES, Gupta AA. 2001. An E3like factor that promotes SUMO conjugation to the yeast septins. Cell 106:735– 44 Takahashi Y, Kahyo T, Toh EA, Yasuda

63.

64.

65.

66.

67.

68.

69.

70.

71.

H, Kikuchi Y. 2001. Yeast Ull1/Siz1 is a novel SUMO1/Smt3 ligase for septin components and functions as an adaptor between conjugating enzyme and substrates. J. Biol. Chem. 276:48973–77 Kahyo T, Nishida T, Yasuda H. 2001. Involvement of PIAS1 in the sumoylation of tumor suppressor p53. Mol. Cell 8:713–18 Schmidt D, Muller S. 2002. Members of the PIAS family act as SUMO ligases for c-Jun p53 and repress p53 activity. Proc. Natl. Acad. Sci. USA 99:2872–77 Nishida T, Yasuda H. 2002. PIAS1 PIASxalpha function as SUMO-E3 ligases toward androgen receptor and repress androgen receptor-dependent transcription. J. Biol. Chem. 277:41311–17 Kotaja N, Karvonen U, Janne OA, Palvimo JJ. 2002. PIAS proteins modulate transcription factors by functioning as SUMO-1 ligases. Mol. Cell Biol. 22:5222–34 Aasland R, Gibson TJ, Stewart AF. 1995. The PHD finger: implications for chromatin-mediated transcriptional regulation. Trends Biochem. Sci. 20:56–59 Capili AD, Schultz DC, Rauscher IF, Borden KL. 2001. Solution structure of the PHD domain from the KAP-1 corepressor: structural determinants for PHD, RING and LIM zinc-binding domains. EMBO. J. 20:165–77 Lu Z, Xu S, Joazeiro C, Cobb MH, Hunter T. 2002. The PHD domain of MEKK1 acts as an E3 ubiquitin ligase and mediates ubiquitination and degradation of ERK1/2. Mol. Cell 9:945– 56 Witowsky JA, Johnson GL. 2003. Ubiquitylation of MEKK1 inhibits its phosphorylation of MKK1 MKK4 and activation of the ERK1/2 and JNK pathways. J. Biol. Chem. 278:1403–6 Coscoy L, Ganem D. 2003. PHD domains and E3 ubiquitin ligases: viruses make the connection. Trends Cell Biol. 13:7–12

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION 72. Koegl M, Hoppe T, Schlenker S, Ulrich HD, Mayer TU, Jentsch S. 1999. A novel ubiquitination factor, E4, is involved in multiubiquitin chain assembly. Cell 96:635–44 73. Hatakeyama S, Yada M, Matsumoto M, Ishida N, Nakayama KI. 2001. U box proteins as a new family of ubiquitinprotein ligases. J. Biol. Chem. 276: 33111–20 74. Jiang J, Ballinger CA, Wu Y, Dai Q, Cyr DM, et al. 2001. CHIP is a U-boxdependent E3 ubiquitin ligase: identification of Hsc70 as a target for ubiquitylation. J. Biol. Chem. 276:42938–44 75. Murata S, Minami Y, Minami M, Chiba T, Tanaka K. 2001. CHIP is a chaperonedependent E3 ligase that ubiquitylates unfolded protein. EMBO Rep. 2:1133– 38 76. Aravind L, Koonin EV. 2000. The U box is a modified RING finger—a common domain in ubiquitination. Curr. Biol. 10:R132–34 77. Hustad CM, Perry WL, Siracusa LD, Rasberry C, Cobb L, et al. 1995. Molecular genetic characterization of six recessive viable alleles of the mouse agouti locus. Genetics 140:255–65 78. Perry WL, Hustad CM, Swing DA, O’Sullivan TN, Jenkins NA, Copeland NG. 1998. The itchy locus encodes a novel ubiquitin protein ligase that is disrupted in a18H mice. Nat. Genet. 18: 143–46 79. Fang D, Elly C, Gao B, Fang N, Altman Y, et al. 2002. Dysregulation of T lymphocyte function in itchy mice: a role for Itch in TH2 differentiation. Nat. Immunol. 3:281–87 80. Li B, Tournier C, Davis RJ, Flavell RA. 1999. Regulation of IL-4 expression by the transcription factor JunB during T helper cell differentiation. EMBO J. 18:420–32 81. Mosmann TR, Coffman RL. 1989. TH1 and TH2 cells: Different patterns of lymphokine secretion lead to different func-

82.

83.

84.

85.

86. 87.

88.

89.

90.

91.

92.

93.

P1: IKH

115

tional properties. Annu. Rev. Immunol. 7:145–73 Paul WE, Seder RA. 1994. Lymphocyte responses and cytokines. Cell 76:241– 51 Conaway RC, Brower CS, Conaway JW. 2002. Emerging roles of ubiquitin in transcription regulation. Science 296: 1254–58 Fuchs SY, Fried VA, Ronai Z. 1998. Stress-activated kinases regulate protein stability. Oncogene 17:1483–90 Angel P, Karin M. 1991. The role of Jun, Fos and the AP-1 complex in cellproliferation and transformation. Biochim. Biophys. Acta 1072:129–57 Wagner EF. 2001. AP-1–Introductory remarks. Oncogene 20:2334–35 Macian F, Lopez-Rodriguez C, Rao A. 2001. Partners in transcription: NFAT and AP-1. Oncogene 20:2476–89 Treier M, Staszewski LM, Bohmann D. 1994. Ubiquitin-dependent c-Jun degradation in vivo is mediated by the delta domain. Cell 78:787–98 Thomas D, Tyers M. 2000. Transcriptional regulation: Kamikaze activators. Curr. Biol. 10:R341–43 Salghetti SE, Muratani M, Wijnen H, Futcher B, Tansey WP. 2000. Functional overlap of sequences that activate transcription signal and ubiquitin-mediated proteolysis. Proc. Natl. Acad. Sci. USA 97:3118–23 Salghetti SE, Caudy AA, Chenoweth JG, Tansey WP. 2001. Regulation of transcriptional activation domain function by ubiquitin. Science 293:1651–53 Musti AM, Treier M, Bohmann D. 1997. Reduced ubiquitin-dependent degradation of c-Jun after phosphorylation by MAP kinases. Science 275:400–2 Winberg G, Matskova L, Chen F, Plant P, Rotin D, et al. 2000. Latent membrane protein 2A of Epstein-Barr virus binds WW domain E3 protein-ubiquitin ligases that ubiquitinate B-cell tyrosine kinases. Mol. Cell Biol. 20:8526–35

11 Feb 2004

16:49

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

116

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

94. Ikeda M, Ikeda A, Longan LC, Longnecker R. 2000. The Epstein-Barr virus latent membrane protein 2A PY motif recruits WW domain-containing ubiquitinprotein ligases. Virology 268:178–91 95. Merchant M, Swart R, Katzman RB, Ikeda M, Ikeda A, et al. 2001. The effects of the Epstein-Barr virus latent membrane protein 2A on B cell function. Int. Rev. Immunol. 20:805–35 96. Ikeda M, Ikeda A, Longnecker R. 2001. PY motifs of Epstein-Barr virus LMP2A regulate protein stability and phosphorylation of LMP2A-associated proteins. J. Virol. 75:5711–18 97. Dykstra ML, Longnecker R, Pierce SK. 2001. Epstein-Barr virus coopts lipid rafts to block the signaling and antigen transport functions of the BCR. Immunity 14:57–67 98. Peschon JJ, Morrissey PJ, Grabstein KH, Ramsdell FJ, Maraskovsky E, et al. 1994. Early lymphocyte expansion is severely impaired in interleukin 7 receptordeficient mice. J. Exp. Med. 180:1955– 60 99. DiSanto JP, Muller W, Guy-Grand D, Fischer A, Rajewsky K. 1995. Lymphoid development in mice with a targeted deletion of the interleukin 2 receptor gamma chain. Proc. Natl. Acad. Sci. USA 92:377–81 100. Cao X, Shores EW, Hu-Li J, Anver MR, Kelsall BL, et al. 1995. Defective lymphoid development in mice lacking expression of the common cytokine receptor gamma chain. Immunity 2:223–38 101. Liu Y-C, Altman A. 1998. Cbl: complex formation and functional implications. Cell Signal. 10:377–85 102. Keane MM, Rivero-Lezcano OM, Mitchell JA, Robbins KC, Lipkowitz S. 1995. Cloning characterization of cbl-b: a SH3 binding protein with homology to the c-cbl proto-oncogene. Oncogene 10:2367–77 103. Keane MM, Ettenberg SA, Nau MM, Banerjee P, Cuello M, et al. 1999. cbl-

104.

105.

106.

107.

108.

109.

110.

111.

112.

3: a new mammalian cbl family protein. Oncogene 18:3365–75 Rao N, Lupher ML Jr, Ota S, Reedquist KA, Druker BJ, Band H. 2000. The linker phosphorylation site Tyr292 mediates the negative regulatory effect of Cbl on ZAP-70 in T cells. J. Immunol. 164:4616–26 Rao N, Ghosh AK, Ota S, Zhou P, Reddi AL, et al. 2001. The non-receptor tyrosine kinase Syk is a target of Cbl-mediated ubiquitylation upon B-cell receptor stimulation. EMBO J. 20:7085– 95 Rao N, Miyake S, Reddi AL, Douillard P, Ghosh AK, et al. 2002. Negative regulation of Lck by Cbl ubiquitin ligase. Proc. Natl. Acad. Sci. USA 99:3794– 99 Paolini R, Molfetta R, Beitz LO, Zhang J, Scharenberg AM, et al. 2002. Activation of Syk tyrosine kinase is required for c-Cbl-mediated ubiquitination of Fcepsilon RI Syk in RBL cells. J. Biol. Chem. 277:36940–47 Rao N, Dodge I, Band H. 2002. The Cbl family of ubiquitin ligases: critical negative regulators of tyrosine kinase signaling in the immune system. J. Leukoc. Biol. 71:753–63 Thien CB, Langdon WY. 2001. Cbl: many adaptations to regulate protein tyrosine kinases. Nat. Rev. Mol. Cell Biol. 2:294–307 Hicke L. 1999. Gettin’ down with ubiquitin: turning off cell-surface receptors, transporters and channels. Trends Cell Biol. 9:107–12 de Melker AA, van der Horst G, Calafat J, Jansen H, Borst J. 2001. c-Cbl ubiquitinates the EGF receptor at the plasma membrane remains receptor associated throughout the endocytic route. J. Cell Sci. 114:2167–78 Viola A, Lanzavecchia A. 1996. T cell activation determined by T cell receptor number and tunable thresholds. Science 273:104–6

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION 113. Valitutti S, Muller S, Salio M, Lanzavecchia A. 1997. Degradation of T cell receptor (TCR)-CD3-zeta complexes after antigenic stimulation. J. Exp. Med. 185:1859–64 114. San Jose E, Borroto A, Niedergang F, Alcover A, Alarcon B. 2000. Triggering the TCR complex causes the downregulation of nonengaged receptors by a signal transduction-dependent mechanism. Immunity 12:161–70 115. Pantaleo G, Olive D, Poggi A, Pozzan T, Moretta L, Moretta A. 1987. Antibodyinduced modulation of the CD3/T cell receptor complex causes T cell refractoriness by inhibiting the early metabolic steps involved in T cell activation. J. Exp. Med. 166:619–24 116. Schonrich G, Kalinke U, Momburg F, Malissen M, Schmitt-Verhulst AM, 1991. Down-regulation of T cell receptors on self-reactive T cells as a novel mechanism for extrathymic tolerance induction. Cell 65:293–304 117. Cenciarelli C, Hou D, Hsu KC, Rellahan BL, Wiest DL, et al. 1992. Activationinduced ubiquitination of the T cell antigen receptor. Science 257:795–97 118. Cenciarelli C, Wilhelm KG Jr, Guo A, Weissman AM. 1996. T cell antigen receptor ubiquitination is a consequence of receptor-mediated tyrosine kinase activation. J. Biol. Chem. 271:8709–13 119. Hou D, Cenciarelli C, Jensen JP, Nguygen HB, Weissman AM. 1994. Activation-dependent ubiquitination of a T cell antigen receptor subunit on multiple intracellular lysines. J. Biol. Chem. 269:14244–47 120. Wang HY, Altman Y, Fang D, Elly C, Dai Y, et al. 2001. Cbl promotes ubiquitination of the T cell receptor{zeta} through an adaptor function of Zap-70. J. Biol. Chem. 276:26004–11 121. Murphy MA, Schnall RG, Venter DJ, Barnett L, Bertoncello I, et al. 1998. Tissue hyperplasia enhanced Tcell signalling via ZAP-70 in c-Cbl-

122.

123.

124.

125.

126.

127.

128.

129.

130.

P1: IKH

117

deficient mice. Mol. Cell Biol. 18:4872– 82 Naramura M, Kole HK, Hu R-J, Gu H. 1998. Altered thymic positive selection intracellular signals in Cbldeficient mice. Proc. Natl. Acad. Sci. USA 95:15547–52 Panigada M, Porcellini S, Barbier E, Hoeflinger S, Cazenave PA, et al. 2002. Constitutive endocytosis and degradation of the pre-T cell receptor. J. Exp. Med. 195:1585–97 Niederberger N, Holmberg K, Alam SM, Sakati W, Naramura M, et al. 2003. Allelic exclusion of the TCR alphachain is an active process requiring TCRmediated signaling c-Cbl. J. Immunol. 170:4557–63 Naramura M, Jang IK, Kole H, Huang F, Haines D, Gu H. 2002. c-Cbl and Cbl-b regulate T cell responsiveness by promoting ligand-induced TCR downmodulation. Nat. Immunol. 3:1192– 99 Soubeyran P, Kowanetz K, Szymkiewicz I, Langdon WY, Dikic I. 2002. CblCIN85-endophilin complex mediates ligand-induced downregulation of EGF receptors. Nature 416:183–87 Petrelli A, Gilestro GF, Lanzardo S, Comoglio PM, Migone N, Giordano S. 2002. The endophilin-CIN85-Cbl complex mediates ligand-dependent downregulation of c-Met. Nature 416:187–90 Schmidt A, Wolde M, Thiele C, Fest W, Kratzin H, et al. 1999. Endophilin I mediates synaptic vesicle formation by transfer of arachidonate to lysophosphatidic acid. Nature 401:133–41 Haglund K, Shimokawa N, Szymkiewicz I, Dikic I. 2002. Cbl-directed monoubiquitination of CIN85 is involved in regulation of ligand-induced degradation of EGF receptors. Proc. Natl. Acad. Sci. USA 99:12191–96 Sosinowski T, Killeen N, Weiss A. 2001. The src-like adaptor protein downregulates the T cell receptor on cd4(+)cd8(+)

11 Feb 2004

16:49

118

131.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

132.

133.

134.

135.

136.

137.

138.

139.

140.

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU thymocytes and regulates positive selection. Immunity 15:457–66 Holland SJ, Liao XC, Mendenhall MK, Zhou X, Pardo J, et al. 2001. Functional cloning of Src-like adapter protein-2 (SLAP-2), a novel inhibitor of antigen receptor signaling. J. Exp. Med. 194:1263– 76 Loreto MP, Berry DM, McGlade CJ. 2002. Functional cooperation between cCbl and Src-like adaptor protein 2 in the negative regulation of T-cell receptor signaling. Mol. Cell Biol. 22:4241–55 Wong ES, Fong CW, Lim J, Yusoff P, Low BC, et al. 2002. Sprouty2 attenuates epidermal growth factor receptor ubiquitylation and endocytosis, and consequently enhances Ras/ERK signalling. EMBO J. 21:4796–808 Rubin C, Litvak V, Medvedovsky H, Zwang Y, Lev S, Yarden Y. 2003. Sprouty fine-tunes EGF signaling through interlinked positive negative feedback loops. Curr. Biol. 13:297–307 Hall AB, Jura N, DaSilva J, Jang YJ, Gong D, Bar-Sagi D. 2003. hSpry2 is targeted to the ubiquitin-dependent proteasome pathway by c-Cbl. Curr. Biol. 13:308–14 Chambers CA, Allison JP. 1999. Costimulatory regulation of T cell function. Curr. Opin. Cell Biol. 11:203–10 Bachmaier K, Krawczyk C, Kozieradzki I, Kong YY, Sasaki T, et al. 2000. Negative regulation of lymphocyte activation and autoimmunity by the molecular adaptor Cbl-b. Nature 403:211–16 Chiang YJ, Kole HK, Brown K, Naramura M, Fukuhara S, et al. 2000. Cbl-b regulates the CD28 dependence of T-cell activation. Nature 403:216–20 Krawczyk C, Bachmaier K, Sasaki T, Jones RG, Snapper SB, et al. 2000. Cbl-b is a negative regulator of receptor clustering and raft aggregation in T cells. Immunity 13:463–73 Fang D, Wang H-Y, Fang N, Altman Y, Elly C, Liu Y-C. 2001. Cbl-b, a RING-

141.

142.

143.

144.

145.

146.

147.

148.

149.

150.

151.

type E3 ubiquitin ligase, targets phosphatidylinositol 3-kinase for ubiquitination in T cells. J. Biol. Chem. 276:4872– 78 Fang D, Liu Y-C. 2001. Proteolysisindependent regulation of phosphatidylinositol 3-kinase by Cbl-b-mediated ubiquitination in T cells. Nat. Immunol. 2:870–75 Bos JL. 1998. All in the family? New insights and questions regarding interconnectivity of Ras, Rap1 and Ral. EMBO J. 17:6776–82 Boussiotis VA, Freeman GJ, Berezovskaya A, Barber DL, Nadler LM. 1997. Maintenance of human T cell anergy: blocking of IL-2 gene transcription by activated Rap1. Science 278:124–28 Chiang SH, Baumann CA, Kanzaki M, Thurmond DC, Watson RT, 2001. Insulin-stimulated GLUT4 translocation requires the CAP-dependent activation of TC10. Nature 410:944–48 Schmitt JM, Stork PJ. 2002. PKA phosphorylation of Src mediates cAMP’s inhibition of cell growth via Rap1. Mol. Cell 9:85–94 Shao Y, Elly C, Liu YC. 2003. Negative regulation of Rap1 activation by the Cbl E3 ubiquitin ligase. EMBO Rep. 4:425– 31 Liu Y-C, Gu H. 2002. Cbl and Cbl-b in T-cell regulation. Trends Immunol. 23: 140–43 Elly C, Witte S, Zhang Z, Rosnet O, Lipokowitz S, et al. 1999. Tyrosine phosphorylation and complex formation of Cbl-b upon T cell receptor stimulation. Oncogene 18:1153–62 Zhang W, Shao Y, Fang D, Huang J, Jeon MS, Liu YC. 2003. Negative regulation of TCR-mediated Crk-L-C3G signaling cell adhesion by Cbl-b. J. Biol. Chem. 278:23978–83 Kamradt T, Mitchison NA. 2001. Tolerance and autoimmunity. N. Engl. J. Med. 344:655–64 Walker LS, Abbas AK. 2002. The enemy

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

152. 153.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

154.

155.

156.

157.

158.

159.

160.

161.

within: keeping self-reactive T cells at bay in the periphery. Nat. Rev. Immunol. 2:11–19 Schwartz RH. 2003. T cell anergy. Annu. Rev. Immunol. 21:305–34 Fields PE, Gajewski TF, Fitch FW. 1996. Blocked Ras activation in anergic CD4+ T cells. Science 271:1276–78 Li W, Whaley CD, Mondino A, Mueller DL. 1996. Blocked signal transduction to the ERK and JNK protein kinases in anergic CD4+ T cells. Science 271:1272– 76 Reedquist KA, Bos JL. 1998. Costimulation through CD28 suppresses T cell receptor-dependent activation of the Ras-like small GTPase Rap1 in human T lymphocytes. J. Biol. Chem. 273:4944– 49 Carey KD, Dillon TJ, Schmitt JM, Baird AM, Holdorf AD, et al. 2000. CD28 and the tyrosine kinase Lck stimulate mitogen-activated protein kinase activity in T cells via inhibition of the small G protein Rap1. Mol. Cell Biol. 20:8409– 19 Sebzda E, Bracke M, Tugal T, Hogg N, Cantrell DA. 2002. Rap1A positively regulates T cells via integrin activation rather than inhibiting lymphocyte signaling. Nat. Immunol. 3:251–58 Macian F, Garcia-Cozar F, Im SH, Horton HF, Byrne MC, Rao A. 2002. Transcriptional mechanisms underlying lymphocyte tolerance. Cell 109:719– 31 Spenser E, Jiang J, Chen ZJ. 1999. Signal-induced ubiquitination of IkappaBalpha by the F-box protein Slimb/betaTrcp. Genes Dev. 13:284–94 Anandasabapathy N, Ford GS, Bloom D, Holness C, Paragas V, et al. 2003. GRAIL: an E3 ubiquitin ligase that inhibits cytokine gene transcription is expressed in anergic CD4+ T cells. Immunity 18:535–47 Locksley RM, Killeen N, Lenardo MJ. 2001. The TNF and TNF receptor su-

162.

163.

164.

165.

166.

167.

168.

169.

170.

P1: IKH

119

perfamilies: integrating mammalian biology. Cell 104:487–501 O’Neill LA, Dinarello CA. 2000. The IL1 receptor/toll-like receptor superfamily: crucial receptors for inflammation host defense. Immunol. Today 21:206–9 Shi CS, Kehrl JH. 2003. TNF-induced GCKR SAPK activation depends upon the E2/E3 complex Ubc13-Uev1A/ TRAF2. J. Biol. Chem. 278:15429– 34 Brown KD, Hostager BS, Bishop GA. 2002. Regulation of TRAF2 signaling by self-induced degradation. J. Biol. Chem. 277:19433–38 Deveraux QL, Reed JC. 1999. IAP family proteins—suppressors of apoptosis. Genes Dev. 13:239–52 Yang Y, Fang S, Jensen JP, Weissman AM, Ashwell JD. 2000. Ubiquitin protein ligase activity of IAPs and their degradation in proteasomes in response to apoptotic stimuli. Science 288:874– 77 Roy N, Deveraux QL, Takahashi R, Salvesen GS, Reed JC. 1997. The cIAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J. 16:6914–25 Huang H, Joazeiro CA, Bonfoco E, Kamada S, Leverson JD, Hunter T. 2000. The inhibitor of apoptosis, cIAP2, functions as a ubiquitin-protein ligase and promotes in vitro monoubiquitination of caspases 3 and 7. J. Biol. Chem. 275: 26661–64 Suzuki Y, Nakabayashi Y, Takahashi R. 2001. Ubiquitin-protein ligase activity of X-linked inhibitor of apoptosis protein promotes proteasomal degradation of caspase-3 and enhances its anti-apoptotic effect in Fas-induced cell death. Proc. Natl. Acad. Sci. USA 98: 8662–67 Li X, Yang Y, Ashwell JD. 2002. TNFRII and c-IAP1 mediate ubiquitination and degradation of TRAF2. Nature 416:345–47

11 Feb 2004

16:49

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

120

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU

171. Silverman N, Maniatis T. 2001. NFkappaB signaling pathways in mammalian insect innate immunity. Genes Dev. 15:2321–42 172. Ghosh S, Karin M. 2002. Missing pieces in the NF-kappaB puzzle. Cell 109(Suppl.):S81–96 173. Liou HC, Sha WC, Scott ML, Baltimore D. 1994. Sequential induction of NFkappa B/Rel family proteins during Bcell terminal differentiation. Mol. Cell Biol. 14:5349–59 174. Xiao G, Harhaj EW, Sun SC. 2001. NFkappaB-inducing kinase regulates the processing of NF-kappaB2 p100. Mol. Cell 7:401–9 175. Senftleben U, Cao Y, Xiao G, Greten FR, Krahn G, et al. 2001. Activation by IKKalpha of a second, evolutionary conserved, NF-kappa B signaling pathway. Science 293:1495–99 176. Fong A, Sun SC. 2002. Genetic evidence for the essential role of betatransducin repeat-containing protein in the inducible processing of NF-kappa B2/p100. J. Biol. Chem. 277:22111–14 177. Fong A, Zhang M, Neely J, Sun SC. 2002. S9, a 19 S proteasome subunit interacting with ubiquitinated NFkappaB2/p100. J. Biol. Chem. 277: 40697–702 178. Coope HJ, Atkinson PG, Huhse B, Belich M, Janzen J, et al. 2002. CD40 regulates the processing of NF-kappaB2 p100 to p52. EMBO J. 21:5375–85 179. Claudio E, Brown K, Park S, Wang H, Siebenlist U. 2002. BAFF-induced NEMO-independent processing of NFkappa B2 in maturing B cells. Nat. Immunol. 3:958–65 180. Saijo K, Mecklenbrauker I, Santana A, Leitger M, Schmedt C, Tarakhovsky A. 2002. Protein kinase C beta controls nuclear factor kappaB activation in B cells through selective regulation of the IkappaB kinase alpha. J. Exp. Med. 195:1647–52 181. Lin L, DeMartino GN, Greene WC.

182.

183.

184.

185.

186.

187.

188.

189.

190.

191.

1998. Cotranslational biogenesis of NFkappaB p50 by the 26S proteasome. Cell 92:819–28 Lin L, DeMartino GN, Greene WC. 2000. Cotranslational dimerization of the Rel homology domain of NF-kappaB1 generates p50-p105 heterodimers and is required for effective p50 production. EMBO J. 19:4712–22 Orian A, Gonen H, Bercovich B, Fajerman I, Eytan E, et al. 2000. SCF(beta) (-TrCP) ubiquitin ligase-mediated processing of NF-kappaB p105 requires phosphorylation of its C-terminus by IkappaB kinase. EMBO J. 19:2580– 91 Lang V, Janzen J, Fischer GZ, Soneji Y, Beinke S, et al. 2003. betaTrCP-mediated proteolysis of NF-kappaB1 p105 requires phosphorylation of p105 serines 927 and 932. Mol. Cell Biol. 23:402–13 Heissmeyer V, Krappmann D, Hatada EN, Scheidereit C. 2001. Shared pathways of IkappaB kinase-induced SCF (betaTrCP)-mediated ubiquitination and degradation for the NF-kappaB precursor p105 and IkappaBalpha. Mol. Cell Biol. 21:1024–35 Amir RE, Iwai K, Ciechanover A. 2002. The NEDD8 pathway is essential for SCF(beta -TrCP)-mediated ubiquitination and processing of the NFkappa B precursor p105. J. Biol. Chem. 277:23253–59 Massague J. 1998. TGF-beta signal transduction. Annu. Rev. Biochem. 67: 753–91 Letterio JJ, Roberts AB. 1998. Regulation of immune responses by TGF-beta. Annu. Rev. Immunol. 16:137–61 Miyazono K, ten Dijke P, Heldin CH. 2000. TGF-beta signaling by Smad proteins. Adv. Immunol. 75:115–57 Gorelik L, Flavell RA. 2002. Transforming growth factor-beta in T-cell biology. Nat. Rev. Immunol. 2:46–53 Shull MM, Ormsby I, Kier AB, Pawlowski S, Diebold RJ, et al. 1992.

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

192.

193.

194.

195.

196.

197.

198.

199.

200.

Targeted disruption of the mouse transforming growth factor-beta 1 gene results in multifocal inflammatory disease. Nature 359:693–99 Kulkarni AB, Huh CG, Becker D, Geiser A, Lyght M, et al. 1993. Transforming growth factor beta 1 null mutation in mice causes excessive inflammatory response and early death. Proc. Natl. Acad. Sci. USA 90:770–74 Moustakas A, Souchelnytskyi S, Heldin CH. 2001. Smad regulation in TGFbeta signal transduction. J. Cell Sci. 114:4359–69 Nakao A, Afrakhte M, Moren A, Nakayama T, Christian JL, et al. 1997. Identification of Smad7, a TGFbetainducible antagonist of TGF-beta signalling. Nature 389:631–35 Hayashi H, Abdollah S, Qiu Y, Cai J, Xu YY, et al. 1997. The MAD-related protein Smad7 associates with the TGFbeta receptor and functions as an antagonist of TGFbeta signaling. Cell 89:1165–73 Zhu H, Kavsak P, Abdollah S, Wrana JL, Thomsen GH. 1999. A SMAD ubiquitin ligase targets the BMP pathway and affects embryonic pattern formation. Nature 400:687–93 Lin X, Liang M, Feng XH. 2000. Smurf2 is a ubiquitin E3 ligase mediating proteasome-dependent degradation of Smad2 in transforming growth factor-beta signaling. J. Biol. Chem. 275:36818–22 Zhang Y, Chang C, Gehling DJ, Hemmati-Brivanlou A, Derynck R. 2001. Regulation of Smad degradation activity by Smurf2, an E3 ubiquitin ligase. Proc. Natl. Acad. Sci. USA 98:974– 79 Bonni S, Wang HR, Causing CG, Kavsak P, Stroschein SL, et al. 2001. TGF-beta induces assembly of a Smad2-Smurf2 ubiquitin ligase complex that targets SnoN for degradation. Nat. Cell Biol. 3:587–95 Stroschein SL, Bonni S, Wrana JL, Luo

201.

202.

203.

204.

205.

206.

207.

208.

209.

210.

P1: IKH

121

K. 2001. Smad3 recruits the anaphasepromoting complex for ubiquitination and degradation of SnoN. Genes Dev. 15:2822–36 Kavsak P, Rasmussen RK, Causing CG, Bonni S, Zhu H, et al. 2000. Smad7 binds to Smurf2 to form an E3 ubiquitin ligase that targets the TGF beta receptor for degradation. Mol. Cell 6:1365–75 Ebisawa T, Fukuchi M, Murakami G, Chiba T, Tanaka K, et al. 2001. Smurf1 interacts with transforming growth factor-beta type I receptor through Smad7 induces receptor degradation. J. Biol. Chem. 276:12477–80 Artavanis-Tsakonas S, Rand MD, Lake RJ. 1999. Notch signaling: cell fate control signal integration in development. Science 284:770–76 Robey E, Chang D, Itano A, Cado D, Alexander H, et al. 1996. An activated form of Notch influences the choice between CD4 and CD8 T cell lineages. Cell 87:483–92 Washburn T, Schweighoffer E, Gridley T, Chang D, Fowlkes BJ, et al. 1997. Notch activity influences the alphabeta versus gammadelta T cell lineage decision. Cell 88:833–43 Deftos ML, He YW, Ojala EW, Bevan MJ. 1998. Correlating notch signaling with thymocyte maturation. Immunity 9:777–86 Deftos ML, Bevan MJ. 2000. Notch signaling in T cell development. Curr. Opin. Immunol. 12:166–72 Radtke F, Wilson A, Stark G, Bauer M, van Meerwijk J, et al. 1999. Deficient T cell fate specification in mice with an induced inactivation of Notch1. Immunity 10:547–58 Pui JC, Allman D, Xu L, DeRocco S, Karnell FG, et al. 1999. Notch1 expression in early lymphopoiesis influences B versus T lineage determination. Immunity 11:299–308 Izon DJ, Punt JA, Xu L, Karnell FG, Allman D, et al. 2001. Notch1 regulates

11 Feb 2004

16:49

122

211.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

212.

213.

214.

215. 216.

217.

218.

219.

220.

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU maturation of CD4+ and CD8+ thymocytes by modulating TCR signal strength. Immunity 14:253–64 Yasutomo K, Doyle C, Miele L, Fuchs C, Germain RN. 2000. The duration of antigen receptor signalling determines CD4+ versus CD8+ T-cell lineage fate. Nature 404:506–10 Wolfer A, Bakker T, Wilson A, Nicolas M, Ioannidis V, et al. 2001. Inactivation of Notch 1 in immature thymocytes does not perturb CD4 or CD8T cell development. Nat. Immunol. 2:235–41 Wolfer A, Wilson A, Nemir M, MacDonald HR, Radtke F. 2002. Inactivation of Notch1 impairs VDJbeta rearrangement and allows pre-TCR-independent survival of early alpha beta lineage thymocytes. Immunity 16:869–79 Allman D, Punt JA, Izon DJ, Aster JC, Pear WS. 2002. An invitation to T and more: notch signaling in lymphopoiesis. Cell 109(Suppl.):S1–11 Osborne B, Miele L. 1999. Notch and the immune system. Immunity 11:653–63 Schmitt TM, Zuniga-Pflucker JC. 2002. Induction of T cell development from hematopoietic progenitor cells by deltalike-1 in vitro. Immunity 17:749–56 Cornell M, Evans DA, Mann R, Fostier M, Flasza M, et al. 1999. The Drosophila melanogaster Suppressor of deltex gene, a regulator of the Notch receptor signaling pathway, is an E3 class ubiquitin ligase. Genetics 152:567–76 Qiu L, Joazeiro C, Fang N, Wang HY, Elly C, et al. 2000. Recognition and ubiquitination of Notch by Itch, a hecttype E3 ubiquitin ligase. J. Biol. Chem. 275:35734–37 Gupta-Rossi N, Le Bail O, Gonen H, Brou C, Logeat F, et al. 2001. Functional interaction between SEL-10, an Fbox protein, and the nuclear form of activated Notch1 receptor. J. Biol. Chem. 276:34371–78 Wu G, Lyapina S, Das I, Li J, Gurney M, et al. 2001. SEL-10 is an inhibitor

221.

222.

223.

224.

225.

226.

227.

228.

229.

230.

of notch signaling that targets notch for ubiquitin-mediated protein degradation. Mol. Cell Biol. 21:7403–15 Oberg C, Li J, Pauley A, Wolf E, Gurney M, Lendahl U. 2001. The Notch intracellular domain is ubiquitinated and negatively regulated by the mammalian Sel10 homolog. J. Biol. Chem. 276:35847– 53 Jehn BM, Dittert I, Beyer S, von der Mark K, Bielke W. 2002. c-Cbl binding and ubiquitin-dependent lysosomal degradation of membrane-associated Notch1. J. Biol. Chem. 277:8033–40 Lai EC, Deblandre GA, Kintner C, Rubin GM. 2001. Drosophila neuralized is a ubiquitin ligase that promotes the internalization and degradation of delta. Dev. Cell 1:783–94 Deblandre GA, Lai EC, Kintner C. 2001. Xenopus neuralized is a ubiquitin ligase that interacts with XDelta1 and regulates Notch signaling. Dev. Cell 1:795–806 Yeh E, Dermer M, Commisso C, Zhou L, McGlade CJ, Boulianne GL. 2001. Neuralized functions as an E3 ubiquitin ligase during Drosophila development. Curr. Biol. 11:1675–9 Itoh M, Kim CH, Palardy G, Oda T, Jiang YJ, et al. 2003. Mind Bomb is a ubiquitin ligase that is essential for efficient activation of Notch signaling by Delta. Dev. Cell 4:67–82 Wodarz A, Nusse R. 1998. Mechanisms of Wnt signaling in development. Annu. Rev. Cell Dev. Biol. 14:59–88 Kitagawa M, Hatakeyama S, Shirane M, Matsumoto M, Ishida N, et al. 1999. An F-box protein, FWD1, mediates ubiquitin-dependent proteolysis of betacatenin. EMBO J. 18:2401–10 Latres E, Chiaur DS, Pagano M. 1999. The human F box protein beta-Trcp associates with the Cul1/Skp1 complex and regulates the stability of beta-catenin. Oncogene 18:849–54 Hart M, Concordet JP, Lassot I, Albert I, del los Santos R, et al. 1999. The F-box

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

231.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

232.

233.

234.

235.

236.

237. 238.

239.

240.

protein beta-TrCP associates with phosphorylated beta-catenin and regulates its activity in the cell. Curr. Biol. 9:207–10 van de Wetering M, de Lau W, Clevers H. 2002. WNT signaling and lymphocyte development. Cell 109(Suppl.):S13–19 Verbeek S, Izon D, Hofhuis F, RobanusMaandag E, te Riele H, et al. 1995. An HMG-box-containing T-cell factor required for thymocyte differentiation. Nature 374:70–74 Schilham MW, Wilson A, Moerer P, Benaissa-Trouw BJ, Cumano A, Clevers HC. 1998. Critical involvement of Tcf-1 in expansion of thymocytes. J. Immunol. 161:3984–91 Okamura RM, Sigvardsson M, Galceran J, Verbeek S, Clevers H, Grosschedl R. 1998. Redundant regulation of T cell differentiation and TCRalpha gene expression by the transcription factors LEF-1 and TCF-1. Immunity 8:11–20 Gounari F, Aifantis I, Khazaie K, Hoeflinger S, Harada N, et al. 2001. Somatic activation of beta-catenin bypasses pre-TCR signaling and TCR selection in thymocyte development. Nat. Immunol. 2:863–69 Ioannidis V, Beermann F, Clevers H, Held W. 2001. The beta-catenin–TCF1 pathway ensures CD4(+)CD8(+) thymocyte survival. Nat. Immunol. 2:691– 97 Polakis P. 2000. Wnt signaling and cancer. Genes Dev. 14:1837–51 Matsuzawa SI, Reed JC. 2001. Siah-1, SIP, and Ebi collaborate in a novel pathway for beta-catenin degradation linked to p53 responses. Mol. Cell 7:915–26 Liu J, Stevens J, Rote CA, Yost HJ, Hu Y, et al. 2001. Siah-1 mediates a novel betacatenin degradation pathway linking p53 to the adenomatous polyposis coli protein. Mol. Cell 7:927–36 Hu G, Fearon ER. 1999. Siah-1 Nterminal RING domain is required for proteolysis function, C-terminal sequences regulate oligomerization and

241.

242.

243.

244.

245.

246.

247.

248.

249.

P1: IKH

123

binding to target proteins. Mol. Cell Biol. 19:724–32 Lorick KL, Jensen JP, Fang S, Ong AM, Hatakeyama S, Weissman AM. 1999. RING fingers mediate ubiquitinconjugating enzyme (E2)-dependent ubiquitination. Proc. Natl. Acad. Sci. USA 96:11364–69 Germani A, Bruzzoni-Giovanelli H, Fellous A, Gisselbrecht S, Varin-Blank N, Calvo F. 2000. SIAH-1 interacts with alpha-tubulin and degrades the kinesin Kid by the proteasome pathway during mitosis. Oncogene 19:5997–6006 Tiedt R, Bartholdy BA, Matthias G, Newell JW, Matthias P. 2001. The RING finger protein Siah-1 regulates the level of the transcriptional coactivator OBF-1. EMBO J. 20:4143–52 Amson RB, Nemani M, Roperch JP, Israeli D, Bougueleret L, et al. 1996. Isolation of 10 differentially expressed cDNAs in p53-induced apoptosis: activation of the vertebrate homologue of the drosophila seven in absentia gene. Proc. Natl. Acad. Sci. USA 93:3953–57 Frew IJ, Dickins RA, Cuddihy AR, Del Rosario M, Reinhard C, et al. 2002. Normal p53 function in primary cells deficient for Siah genes. Mol. Cell Biol. 22: 8155–64 Dickins RA, Frew IJ, House CM, O’Bryan MK, Holloway AJ, et al. 2002. The ubiquitin ligase component Siah1a is required for completion of meiosis I in male mice. Mol. Cell Biol. 22:2294–303 Ashcroft M, Vousden KH. 1999. Regulation of p53 stability. Oncogene 18:7637– 43 Donehower LA, Harvey M, Slagle BL, McArthur MJ, Montgomery CA Jr, et al. 1992. Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours. Nature 356:215–21 Costello PS, Cleverley SC, Galandrini R, Henning SW, Cantrell DA. 2000. The GTPase rho controls a p53dependent survival checkpoint during

11 Feb 2004

16:49

124

250.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

251.

252.

253.

254.

255.

256.

257.

258.

259. 260.

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU thymopoiesis. J. Exp. Med. 192:77– 85 Boehme SA, Lenardo MJ. 1996. TCRmediated death of mature T lymphocytes occurs in the absence of p53. J. Immunol. 156:4075–78 Grayson JM, Lanier JG, Altman JD, Ahmed R. 2001. The role of p53 in regulating antiviral T cell responses. J. Immunol. 167:1333–37 Honda R, Tanaka H, Yasuda H. 1997. Oncoprotein MDM2 is a ubiquitin ligase E3 for tumor suppressor p53. FEBS Lett. 420:25–27 Nakamura S, Roth JA, Mukhopadhyay T. 2000. Multiple lysine mutations in the C-terminal domain of p53 interfere with MDM2-dependent protein degradation and ubiquitination. Mol. Cell Biol. 20:9391–98 Rodriguez MS, Desterro JM, Lain S, Lane DP, Hay RT. 2000. Multiple Cterminal lysine residues target p53 for ubiquitin-proteasome-mediated degradation. Mol. Cell Biol. 20:8458–67 Boyd SD, Tsai KY, Jacks T. 2000. An intact HDM2 RING-finger domain is required for nuclear exclusion of p53. Nat. Cell Biol. 2:563–68 Gu J, Nie L, Wiederschain D, Yuan ZM. 2001. Identification of p53 sequence elements that are required for MDM2mediated nuclear export. Mol. Cell Biol. 21:8533–46 Lohrum MA, Woods DB, Ludwig RL, Balint E, Vousden KH. 2001. C-terminal ubiquitination of p53 contributes to nuclear export. Mol. Cell Biol. 21:8521–32 Mendrysa SM, McElwee MK, Michalowski J, O’Leary KA, Young KM, Perry ME. 2003. mdm2 is critical for inhibition of p53 during lymphopoiesis and the response to ionizing irradiation. Mol. Cell Biol. 23:462–72 Prives C, Manley JL. 2001. Why is p53 acetylated? Cell 107:815–18 Chehab NH, Malikzay A, Stavridi ES, Halazonetis TD. 1999. Phosphorylation

261.

262.

263.

264.

265.

266.

267.

268.

269.

270.

of Ser-20 mediates stabilization of human p53 in response to DNA damage. Proc. Natl. Acad. Sci. USA 96:13777–82 Unger T, Juven-Gershon T, Moallem E, Berger M, Vogt Sionov R, et al. 1999. Critical role for Ser20 of human p53 in the negative regulation of p53 by Mdm2. EMBO J. 18:1805–14 Hirao A, Kong YY, Matsuoka S, Wakeham A, Ruland J, et al. 2000. DNA damage-induced activation of p53 by the checkpoint kinase Chk2. Science 287: 1824–27 Wu Z, Earle J, Saito S, Anderson CW, Appella E, Xu Y. 2002. Mutation of mouse p53 Ser23 and the response to DNA damage. Mol. Cell Biol. 22:2441– 49 Li M, Luo J, Brooks CL, Gu W. 2002. Acetylation of p53 inhibits its ubiquitination by Mdm2. J. Biol. Chem. 277: 50607–11 Jin Y, Zeng SX, Dai MS, Yang XJ, Lu H. 2002. MDM2 inhibits PCAF (p300/CREB-binding protein-associated factor)-mediated p53 acetylation. J. Biol. Chem. 277:30838–43 Ito A, Kawaguchi Y, Lai CH, Kovacs JJ, Higashimoto Y, et al. 2002. MDM2HDAC1-mediated deacetylation of p53 is required for its degradation. EMBO J. 21:6236–45 Grossman SR, Deato ME, Brignone C, Chan HM, Kung AL, et al. 2003. Polyubiquitination of p53 by a ubiquitin ligase activity of p300. Science 300:342– 44 Nikolaev AY, Li M, Puskas N, Qin J, Gu W. 2003. Parc: a cytoplasmic anchor for p53. Cell 112:29–40 Leng RP, Lin Y, Ma W, Wu H, Lemmers B, et al. 2003. Pirh2, a p53-induced ubiquitin-protein ligase, promotes p53 degradation. Cell 112:779–91 Pamer E, Cresswell P. 1998. Mechanisms of MHC class I–restricted antigen processing. Annu. Rev. Immunol. 16: 323–58

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION 271. Villadangos JA, Ploegh HL. 2000. Proteolysis in MHC class II antigen presentation: Who’s in charge? Immunity 12:233–39 272. Glynne R, Powis SH, Beck S, Kelly A, Kerr LA, Trowsdale J. 1991. A proteasome-related gene between the two ABC transporter loci in the class II region of the human MHC. Nature 353:357–60 273. Hisamatsu H, Shimbara N, Saito Y, Kristensen P, Hendil KB, et al. 1996. Newly identified pair of proteasomal subunits regulated reciprocally by interferon gamma. J. Exp. Med. 183:1807– 16 274. Gaczynska M, Rock KL, Goldberg AL. 1993. Gamma-interferon expression of MHC genes regulate peptide hydrolysis by proteasomes. Nature 365:264–67 275. Cerundolo V, Benham A, Braud V, Mukherjee S, Gould K, et al. 1997. The proteasome-specific inhibitor lactacystin blocks presentation of cytotoxic T lymphocyte epitopes in human murine cells. Eur. J. Immunol. 27:336–41 276. Lelouard H, Gatti E, Cappello F, Gresser O, Camosseto V, Pierre P. 2002. Transient aggregation of ubiquitinated proteins during dendritic cell maturation. Nature 417:177–82 277. Cox JH, Galardy P, Bennink JR, Yewdell JW. 1995. Presentation of endogenous and exogenous antigens is not affected by inactivation of E1 ubiquitin-activating enzyme in temperature-sensitive cell lines. J. Immunol. 154:511–19 278. Yewdell JW, Anton LC, Bennink JR. 1996. Defective ribosomal products (DRiPs): a major source of antigenic peptides for MHC class I molecules? J. Immunol. 157:1823–26 279. Schubert U, Anton LC, Gibbs J, Norbury CC, Yewdell JW, Bennink JR. 2000. Rapid degradation of a large fraction of newly synthesized proteins by proteasomes. Nature 404:770–74 280. Reits EA, Vos JC, Gromme M, Neefjes J. 2000. The major substrates for TAP

281.

282.

283.

284.

285.

286.

287.

288.

289.

P1: IKH

125

in vivo are derived from newly synthesized proteins. Nature 404:774–78 DeLeo AB. 1998. p53-based immunotherapy of cancer. Crit. Rev. Immunol. 18:29–35 Dahl AM, Beverley PC, Stauss HJ. 1996. A synthetic peptide derived from the tumor-associated protein mdm2 can stimulate autoreactive, high avidity cytotoxic T lymphocytes that recognize naturally processed protein. J. Immunol. 157:239–46 Furman MH, Ploegh HL. 2002. Lessons from viral manipulation of protein disposal pathways. J. Clin. Invest. 110:875– 79 Shamu CE, Flierman D, Ploegh HL, Rapoport TA, Chau V. 2001. Polyubiquitination is required for US11-dependent movement of MHC class I heavy chain from endoplasmic reticulum into cytosol. Mol. Biol. Cell 12:2546–55 Kikkert M, Hassink G, Barel M, Hirsch C, van der Wal FJ, Wiertz E. 2001. Ubiquitination is essential for human cytomegalovirus US11-mediated dislocation of MHC class I molecules from the endoplasmic reticulum to the cytosol. Biochem. J. 358:369–77 Ishido S, Choi JK, Lee BS, Wang C, DeMaria M, et al. 2000. Inhibition of natural killer cell-mediated cytotoxicity by Kaposi’s sarcoma-associated herpesvirus K5 protein. Immunity 13: 365–74 Coscoy L, Ganem D. 2000. Kaposi’s sarcoma-associated herpesvirus encodes two proteins that block cell surface display of MHC class I chains by enhancing their endocytosis. Proc. Natl. Acad. Sci. USA 97:8051–56 Coscoy L, Sanchez DJ, Ganem D. 2001. A novel class of herpesvirus-encoded membrane-bound E3 ubiquitin ligases regulates endocytosis of proteins involved in immune recognition. J. Cell Biol. 155:1265–73 Boname JM, Stevenson PG. 2001.

11 Feb 2004

16:49

126

290.

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

291.

292.

293.

294.

295.

296.

297.

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

P1: IKH

LIU MHC class I ubiquitination by a viral PHD/LAP finger protein. Immunity 15: 627–36 Hewitt EW, Duncan L, Mufti D, Baker J, Stevenson PG, Lehner PJ. 2002. Ubiquitylation of MHC class I by the K3 viral protein signals internalization and TSG101-dependent degradation. EMBO J. 21:2418–29 Lybarger L, Wang X, Harris MR, Virgin HWt, Hansen TH. 2003. Virus subversion of the MHC class I peptide-loading complex. Immunity 18:121–30 Mansouri M, Bartee E, Gouveia K, Hovey Nerenberg BT, Barrett J, et al. 2003. The PHD/LAP-domain protein M153R of myxomavirus is a ubiquitin ligase that induces the rapid internalization and lysosomal destruction of CD4. J. Virol. 77:1427–40 Guerin JL, Gelfi J, Boullier S, Delverdier M, Bellanger FA, et al. 2002. Myxoma virus leukemia-associated protein is responsible for major histocompatibility complex class I and Fas-CD95 downregulation and defines scrapins, a new group of surface cellular receptor abductor proteins. J. Virol. 76:2912–23 Schubert U, Anton LC, Bacik I, Cox JH, Bour S, et al. 1998. CD4 glycoprotein degradation induced by human immunodeficiency virus type 1 Vpu protein requires the function of proteasomes and the ubiquitin-conjugating pathway. J. Virol. 72:2280–88 Margottin F, Bour SP, Durand H, Selig L, Benichou S, et al. 1998. A novel human WD protein, h-beta TrCp, that interacts with HIV-1 Vpu connects CD4 to the ER degradation pathway through an F-box motif. Mol. Cell 1:565–74 Vogt VM. 2000. Ubiquitin in retrovirus assembly: actor or bystander? Proc. Natl. Acad. Sci. USA 97:12945–47 Schubert U, Ott DE, Chertova EN, Welker R, Tessmer U, et al. 2000. Proteasome inhibition interferes with gag polyprotein processing, release, and mat-

298.

299.

300.

301.

302.

303.

304.

305.

306.

uration of HIV-1 and HIV-2. Proc. Natl. Acad. Sci. USA 97:13057–62 Patnaik A, Chau V, Wills JW. 2000. Ubiquitin is part of the retrovirus budding machinery. Proc. Natl. Acad. Sci. USA 97:13069–74 Strack B, Calistri A, Accola MA, Palu G, Gottlinger HG. 2000. A role for ubiquitin ligase recruitment in retrovirus release. Proc. Natl. Acad. Sci. USA 97:13063– 68 Harty RN, Brown ME, Wang G, Huibregtse J, Hayes FP. 2000. A PPxY motif within the VP40 protein of Ebola virus interacts physically and functionally with a ubiquitin ligase: implications for filovirus budding. Proc. Natl. Acad. Sci. USA 97:13871–76 Kikonyogo A, Bouamr F, Vana ML, Xiang Y, Aiyar A, et al. 2001. Proteins related to the Nedd4 family of ubiquitin protein ligases interact with the L domain of Rous sarcoma virus and are required for gag budding from cells. Proc. Natl. Acad. Sci. USA 98:11199–204 Yasuda J, Hunter E, Nakao M, Shida H. 2002. Functional involvement of a novel Nedd4-like ubiquitin ligase on retrovirus budding. EMBO Rep. 3:636–40 Huang M, Orenstein JM, Martin MA, Freed EO. 1995. p6Gag is required for particle production from full-length human immunodeficiency virus type 1 molecular clones expressing protease. J. Virol. 69:6810–18 VerPlank L, Bouamr F, LaGrassa TJ, Agresta B, Kikonyogo A, et al. 2001. Tsg101, a homologue of ubiquitinconjugating (E2) enzymes, binds the L domain in HIV type 1 Pr55(Gag). Proc. Natl. Acad. Sci. USA 98:7724–29 Pornillos O, Alam SL, Davis DR, Sundquist WI. 2002. Structure of the Tsg101 UEV domain in complex with the PTAP motif of the HIV-1 p6 protein. Nat. Struct. Biol. 9:812–17 Demirov DG, Ono A, Orenstein JM, Freed EO. 2002. Overexpression of the

11 Feb 2004

16:49

AR

AR210-IY22-04.tex

AR210-IY22-04.sgm

LaTeX2e(2002/01/18)

UBIQUITINATION AND IMMUNE REGULATION

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

307.

308.

309.

310.

N-terminal domain of TSG101 inhibits HIV-1 budding by blocking late domain function. Proc. Natl. Acad. Sci. USA 99: 955–60 Katzmann DJ, Babst M, Emr SD. 2001. Ubiquitin-dependent sorting into the multivesicular body pathway requires the function of a conserved endosomal protein sorting complex, ESCRT-I. Cell 106:145–55 Martin-Serrano J, Zang T, Bieniasz PD. 2003. Role of ESCRT-I in retroviral budding. J. Virol. 77:4794–804 Pornillos O, Alam SL, Rich RL, Myszka DG, Davis DR, Sundquist WI. 2002. Structure and functional interactions of the Tsg101 UEV domain. EMBO J. 21: 2397–406 Martin-Serrano J, Zang T, Bieniasz PD. 2001. HIV-1 and Ebola virus encode small peptide motifs that recruit Tsg101 to sites of particle assembly to facilitate egress. Nat. Med. 7:1313–19

P1: IKH

127

311. Timmins J, Schoehn G, Ricard-Blum S, Scianimanico S, Vernet T, et al. 2003. Ebola virus matrix protein VP40 interaction with human cellular factors Tsg101 and Nedd4. J. Mol. Biol. 326:493–502 312. Licata JM, Simpson-Holley M, Wright NT, Han Z, Paragas J, Harty RN. 2003. Overlapping motifs (PTAP and PPEY) within the Ebola virus VP40 protein function independently as late budding domains: involvement of host proteins TSG101 and VPS-4. J. Virol. 77:1812– 19 313. Liu J, Furukawa M, Matsumoto T, Xiong Y. 2002. NEDD8 modification of CUL1 dissociates p120(CAND1), an inhibitor of CUL1-SKP1 binding SCF ligases. Mol. Cell 10:1511–18 314. Cope GA, Suh GS, Aravind L, Schwarz SE, Zipursky SL, et al. 2002. Role of predicted metalloprotease motif of Jab1/Csn5 in cleavage of Nedd8 from Cul1. Science 298:608–11

Liu.qxd

3/12/2004

1:50 PM

Page 1

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

UBIQUITINATION AND IMMUNE REGULATION

C-1

Figure 1 Representatives of HECT-type and RING-type E3 ligases. (A) The HECT-type E3 Itch promotes ubiquitination of JunB and affects IL-4 gene transcription. The WW domains of Itch recruit JunB, and Ub is transferred from E2 to the HECT domain and then to the substrate. (B) The RING-type E3 Cbl targets the TCR/CD3 complex for Ub conjugation through an adaptor’s role of Zap-70 to form Cbl, Zap-70, and the TCR/CD3 complex. Ub is directly transferred from E2 to the substrate.

Liu.qxd

3/12/2004

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

1:50 PM

Page 2

LIU

Figure 2 Protein ubiquitination in T cell anergy induction. TCR stimulation in the absence of co-stimulation or treatment of T cells with ionomycin results in an unbalanced activation of NFAT and the transcription of anergy-inducing genes. Some of these gene products turn out to be E3 Ub ligases like Cbl-b, Itch, or the Ub-binding protein, Tsg101 (V. Heissmeyer & A. Rao, submitted manuscript), which together induce the downmodulation of PLCγ–1 and PKCθ, and cause the T cell unresponsiveness. In anergized T cells, a novel RING-type E3 GRAIL is expressed and participates in T cell anergy induction, although the targets for this E3 remain unclear (160).

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:81-127. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.21.090501.080131

Annu. Rev. Immunol. 2004. 22:129–56 doi: 10.1146/annurev.immunol.21.090501.080131 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on September 15, 2003

LIGANDS FOR L-SELECTIN: Homing, Inflammation, and Beyond

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Steven D. Rosen Department of Anatomy and Program in Immunology, University of California, San Francisco, San Francisco, California 94143-0452; email: [email protected]

Key Words leukocyte trafficking, HEV, endothelium, glycosylation, sulfation ■ Abstract Understanding the molecular basis of lymphocyte homing to lymphoid organs was originally a problem of concern only to immunologists. With the discovery of L-selectin and its ligands, interested scientists have expanded to include glycobiologists, immunopathologists, cancer biologists, and developmental biologists. Going beyond its first discovered role in homing to lymph nodes, the L-selectin system is implicated in such diverse processes as inflammatory leukocyte trafficking in both acute and chronic settings, hematogenous metastasis of carcinoma cells, effector mechanisms for inflammatory demyelination of axons, and implantation of the early mammalian embryo. This review focuses on the ligands for L-selectin that are found on vascular endothelium, leukocytes, carcinoma cells, and at various extravascular sites. The discovery of selectins and their ligands has validated the long-predicted hypothesis that carbohydrate-directed cell adhesion is relevant in eukaryotic systems. Emphasis will be given to the carbohydrate and sulfation modifications of the ligands, which enable recognition by L-selectin. The rapid “homing” of labeled cells into the lymph nodes presumably had its basis in the special affinity of small lymphocytes for the endothelium of the postcapillary venules. Gowans & Knight (1)

DEFINING THE PROBLEM In the 1950s and 1960s, James Gowans at the Sir William Dunn School of Pathology at Oxford University discovered that lymphocytes recirculate from the blood into lymphoid organs (“homing”) and back to the blood (1). These landmark studies provided the first enunciation of the role of lymphocyte recirculation in immune surveillance; namely, the process enables small numbers of specific lymphocytes to be brought into contact with antigen in regional lymphoid organs, which underlies primary sensitization of na¨ıve lymphocytes and restimulation of memory cells (2– 4). The quote given above comes from the seminal study in which Knight & Gowans 0732-0582/04/0423-0129$14.00

129

11 Feb 2004

16:54

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

130

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN

(1) determined that the route of lymphocyte entry into lymph nodes is through postcapillary venules in the cortex “which are remarkable for the height of their endothelium.” These vessels are now commonly referred to as “high endothelial venules” (HEV), and the cells comprising them are called “high endothelial cells” (HEC). HEV are not unique to lymph nodes but are found in all secondary lymphoid organs with the exception of the spleen (2). Gowans & Knight (1) noted that the walls of HEV contain many small lymphocytes in various stages of extravasation. The restriction of infiltrating cells to this class of leukocyte led to their prediction of a “special affinity” between the endothelium and the small lymphocytes. In 1976, it became possible to approach the molecular basis of the special affinity through the development of an in vitro adhesion assay for lymphocyte-HEV binding. Overlaying viable lymphocytes onto cryostat-cut sections of lymph node, Stamper & Woodruff (5) observed highly specific adherence of exogenous lymphocytes to profiles of HEV (Figure 1). Taking a monoclonal antibody strategy to decipher the molecular basis of this interaction, Gallatin et al. (6) discovered a cell surface antigen on lymphocytes (gp90MEL-14) as the target of a functionblocking antibody (MEL-14). The subsequent cloning of this molecule led to the identification of L-selectin (CD62L), a type I membrane protein with a Ctype lectin at the amino terminus, an EGF domain, two short consensus repeats, a transmembrane domain, and a short cytoplasmic tail (7). The presence of the lectin domain validated earlier cell biological work that had shown that gp90MEL functions as a calcium-dependent lectin-like receptor that recognizes specific carbohydrate-based ligands on lymph node HEV (8–10). A definitive role for L-selectin in lymphocyte homing to lymph nodes was subsequently established by gene targeting (11). The other members of the selectin family (E- and P-selectin) have the same overall structural organization as L-selectin and also function in

Figure 1 The Stamper-Woodruff in vitro adhesion assay. A cryostat-cut section of a mouse lymph node is shown with exogenously added lymphocytes attached to a profile of an HEV, shown in a longitudinal view. The exogenous lymphocytes are above the plane of the section. The interaction measured by this assay accurately reflects the activity of L-selectin and its HEV-expressed ligands. The discovery of the L-selectin adhesion system emerged from studies with this assay. This experiment was performed by Mark Singer.

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

L-SELECTIN LIGANDS

P1: IKH

131

leukocyte-endothelial interactions as C-type lectins (12–14). P-selectin is on activated platelets and endothelial cells. E-selectin is restricted to activated endothelium. It is now recognized that L-selectin is essential for homing of na¨ıve lymphocytes to all HEV-bearing secondary lymphoid organs (2). Homing to secondary lymphoid organs occurs through a cascade of steps, which involves the step-wise action of adhesion and signaling receptors. For na¨ıve lymphocytes and some memory cells, L-selectin serves a primary adhesion receptor by mediating the tethering and rolling of lymphocytes along the apical aspects of HEV. The most fully elucidated adhesion cascade is for homing of na¨ıve lymphocytes to peripheral lymph nodes (PN) in mouse (4). First, L-selectin mediates the tethering and rolling of lymphocytes along the apical aspects of HEV. Second, HEV-associated chemokines such as CCL21 (SLC) engage specific G-protein couple receptors on lymphocytes (CCR7 in the case of SLC), which leads to the activation of LFA-1 on the lymphocyte. Firm arrest of the lymphocyte occurs when LFA-1 interacts with ICAMs on the HEV. Finally, the lymphocyte transmigrates into the lymph node parenchyma. The combinatorial usage of adhesion and signaling receptors provides a paradigm to explain the selective homing of effector/memory lymphocytes to anatomically distinct lymphoid compartments (2, 4). Generalization of this “cascade” or multistep model also accounts for the exquisite spatiotemporal specificity of inflammatory leukocyte trafficking by which leukocyte populations are selectively targeted with precise timing to particular vascular beds (2, 15). L-selectin is broadly distributed on most leukocytes in the blood. Beyond its general role in lymphocyte homing to secondary lymphoid organs, L-selectin is now implicated in many instances of inflammatory leukocyte trafficking (16, 17). Although its best understood function is as a primary adhesion receptor for ligands on the vascular endothelium, L-selectin also mediates leukocyte-leukocyte interactions during inflammatory reactions. Moreover, leukocyte-carcinoma cell interactions mediated through L-selectin promote tumor metastasis. Further expanding the functional repertoire of L-selectin, recent studies have revealed that it has important extravascular functions. The purpose of this review is to summarize the current knowledge about the ligands engaged by L-selectin during normal and pathological processes. Several classes of ligands have been identified with distinct posttranslational modifications and strategies for binding. The structures of the molecules and the enzymes responsible for the their posttranslational modifications will be reviewed.

HEV-EXPRESSED LIGANDS IN LYMPH NODES AND TONSILS Early Work Gesner & Ginsburg (18) were the first to suggest that cell surface carbohydrates serve as recognition determinants in lymphocyte homing. They observed that treatment of lymphocytes with mixed glycosidases, probably including a sialidase

11 Feb 2004

16:54

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

132

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN

activity, reduced their homing to lymph nodes and spleen but increased their accumulation in the liver. It is now appreciated that the accumulation of enzyme-treated lymphocytes in the liver was not due to inactivation of a normal homing mechanism but rather the activation of a liver clearance pathway. Ironically, sialic acid determinants were later shown to be critical elements of L-selectin ligands on HEV (9). The pivotal finding was that sialidase treatment of lymph node sections completely abrogates the binding of lymphocytes to HEV. A further intriguing finding was that HEV very avidly incorporate 35S-SO4 into macromolecular components, suggesting an important function of sulfated molecules in the process of lymphocyte homing (19). As detailed below, these early observations have now been rationalized by structural analysis of actual ligands.

Biochemical Identification of Ligands The first HEV-expressed ligands to be identified were GlyCAM-1, CD34, and podocalyxin (7, 20–22). Notably, all are sialomucins, a characteristic that generalizes to most other ligand candidates. Mucins are glycoproteins with multiple O-linked (via Ser or Thr) glycans. Whereas CD34 and podocalyxin are a type I membrane protein (Figure 2), GlyCAM-1 is a secreted protein that is found at high levels in plasma. GlyCAM-1 is also expressed in lactating mammary gland and milk in several species. However, milk GlyCAM-1 does not react with L-selectin (7). The human homologue of GlyCAM-1 is also expressed at the mRNA level in lactating mammary gland but not in lymphoid organs (23). CD34 and podocalyxin are broadly expressed on the vascular endothelium, as well as on hematopoietic

Figure 2 The CD34 family of sialomucins. All three of these glycoproteins exhibit L-selectin ligand activity when they possess the appropriate posttranslational modifications. Endoglycan differs from the other two in having an acidic amino region that contains two tyrosine sulfates. In addition, endoglycan is modified with chondroitin sulfate glycosaminoglycan (GAG) chains. The potential sites for GAG chain addition are the “SG” pairs. The two tyrosines (“Y”) in the acidic domain of endoglycan are modified by sulfation. The transmembrane regions, cysteine-containing regions, and C-terminal cytoplasmic tails are denoted by “M,” “C-C,” and “CT,” respectively. In addition to the overall structural similarity among the three glycoproteins, there is considerable sequence homology in the cytoplasmic tails.

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

L-SELECTIN LIGANDS

P1: IKH

133

precursor cells (24, 25). L-selectin-reactive forms of podocalyxin and CD34 are present in the HEV of tonsils in human, whereas nonreactive forms are found in noninflamed vessels (21, 22, 26). A second approach to the identification of HEV-expressed ligands has relied on a widely used mAb known as MECA-79 (27). This antibody strongly stains lymph node HEV in mouse, blocks lymphocyte attachment to HEV in the StamperWoodruff in vitro adherence assay, and inhibits short-term lymphocyte homing to lymph nodes. As revealed by intravital microscopy, MECA-79 effectively blocks the tethering and rolling of lymphocytes along HEV, thus preventing the initiation of the recruitment cascade (28). MECA-79 and an L-selectin-IgG chimera immunoprecipitate the same complex of proteins from mouse lymph nodes and human tonsils (29). In fact, the function-blocking activity of MECA-79 is explained by overlap of its sulfated carbohydrate epitope with the L-selectin recognition determinant (Figure 3 and see below). PNAd, or peripheral lymph node addressin, is

Figure 3 The structure of a disulfated biantennary O-glycan found in HEV-derived GlyCAM-1 from mouse (56). The upper arm is the core 2 branch and the lower arm is the core 1 extension. Both arms are capped by 6-sulfo sLex (indicated by boxes shown in panel A), which is likely to be the minimal recognition determinant for Lselectin. Panel B shows the MECA-79 epitope. This figure was modified from one kindly provided by Carolyn Bertozzi.

11 Feb 2004

16:54

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

134

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN

a widely used term to designate the complex of MECA-79-reactive glycoproteins within HEV (2). For mouse lymph nodes and human tonsils, there are four or more distinct members of this complex, including the aforementioned GlyCAM-1 (mouse), CD34 (mouse and human), and podocalyxin (human). Recently, endomucin, another type I membrane sialomucin, was identified as a member of the PNAd complex in mouse and human (30). MAdCAM-1, as isolated from mesenteric lymph nodes, is also a PNAd component, presumably attributable to its short mucin domain (31). It also belongs to the Ig superfamily and serves as a counterreceptor for α4β7 in the homing of lymphocytes to gut-associated lymphoid organs and tissues (2). MAdCAM-1 is not expressed in PN HEV. Other members of the PNAd complex, for example the >200 kd component, have not yet been molecularly identified, although they are susceptible to degradation by a mucindegrading enzyme (O-sialoendoglycoprotease, or OSGE) (22). Furthermore, it should be noted that human tonsillar HEV express L-selectin ligands that are insensitive to OSGE and lack MECA-79 reactivity (32). These nonmucin ligands also remain to be defined. Rolling studies performed in the parallel plate flow chamber have established that the entire PNAd complex as well as its individual members can support tethering and rolling of lymphocytes via L-selectin under physiological flow conditions (21, 22). Null mice have been generated for GlyCAM-1 (26) and CD34 (33) with no obvious abnormality in homing to lymph nodes. It is strongly suspected that the HEV-expressed sialomucins are functionally redundant such that the loss of one member is compensated by the activity of the others. In contrast, significant deficiencies in homing are seen when specific posttranslational modifications shared by these ligands are eliminated by gene targeting (see below). Thus, there is a not a strict requirement for a particular core protein but rather for a set of specific posttranslational modifications. The situation is strikingly different for P-selectin in that its principle leukocyte ligand is a single sialomucin called PSGL-1 (34, 35). Further description of PSGL-1 is provided below.

Biophysics of Selectin-Ligand Interactions For most eukaryotic cells, changes in cell adhesion to other cells or extracellular matrix takes minutes to hours. In contrast, leukocytes whose functions depend on rapid deployment over large distances develop adhesive interactions quickly, and the interactions are usually readily reversible. An extreme example of rapid kinetics is the rolling interaction of blood-borne leukocytes with the vascular endothelium. Here, within seconds of contact, the leukocyte is converted from a free-moving cell to a rolling cell propelled by the blood flow (2). The rolling interaction with the endothelium involves the rapid formation of bonds at the leading edge of the leukocyte and rapid bond dissociation at the trailing edge. From in vitro flow chamber measurements and intravital observations, L-selectin is specialized to mediate fast rolling, as compared to the other two selectins or the α4 integrins (36, 37). In vitro, rolling via L-selectin is about 7–12-fold faster than rolling through

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

L-SELECTIN LIGANDS

P1: IKH

135

E- or P-selectin at comparable substrate densities. This faster rolling is correlated with a faster unstressed koff, which is 7–9-fold faster for L-selectin than for the other two selectins. This faster rate of bond dissociation predicts that the rate of tether bond formation, kon, must also be faster for L-selectin in order to maintain stable rolling. L-selectin-mediated rolling exhibits a shear threshold phenomenon in which a minimum wall shear is required for rolling, after which the number of rolling increases and then decreases (38). This behavior is believed to limit the inappropriate interaction of leukocytes with one another in the center of vessels (low shear) or with the vessel wall in low-flow vessels. Rolling via L-selectin is also very stable in that rolling velocity plateaus with increasing wall shear stress (39); this characteristic minimizes variation in the velocity of rolling leukocytes with fluctuations in wall shear stress and provides sufficient time for the leukocytes to be exposed to activating stimuli present on the vessel wall.

Recognition Determinants for L-Selectin and MECA-79 All three selectins recognize sialyl Lewis x (sLex) (Table 1) but with relatively low affinity (40, 41). By NMR spectroscopy, the binding of sLex to L-selectin IgG fusion occurs with a Kd of 3.9 µM (42). Sialylation and fucosylation of the tetrasaccharide are required for binding of sLex to L-selectin (26, 41). An important line of evidence that sLex is actually a component of L-selectin ligands comes from the use of sLex-reactive mAbs. Several of these antibodies stain HEV in human lymphoid organs and in some cases block L-selectin dependent adherence of lymphocytes to HEV in in vitro adherence assays (26). sLex-reactive antibodies also stain capillaries and venules at sites of inflammation (24). In settings of sustained inflammation, the positive vessels frequently are high-walled and are referred to as HEV-like (26, 43). Vessels in nonlymphoid tissues under normal circumstances generally show little if any reactivity with these antibodies (24). The avid incorporation of 35S-SO4 incorporation by HEV mentioned above prompted the original investigation of the possible sulfation of the ligands associated with these vessels (20). In fact, HEV-expressed forms of L-selectin ligands

TABLE 1 Nomenclature and structure of complex carbohydrates Name

Structure

sLex

Siaα2 → 3Galβ1 → 4[Fucα1 → 3]GlcNAc

6-sulfo sLex

Siaα2 → 3Galβ1 → 4[Fucα1 → 3][SO3 → 6]GlcNAc

0

6 -sulfo sLex

Siaα2 → 3[SO3 → 6]Galβ1 → 4[Fucα1 → 3]GlcNAc

6-sulfo LacNAc

Galβ1 → 4[SO3 → 6]GlcNAc

Core 1

Galβ1 → 3GalNAc

Core 2

Galβ1 → 3[GlcNAcβ1 → 6]GalNAc

Extended core 1

GlcNAcβ1 → 3Galβ1 → 3GalNAc

11 Feb 2004

16:54

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

136

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN

are sulfated, as directly shown in mouse and human systems (20, 44, 45). Moreover, sulfation is, in fact, essential for L-selectin binding (46) and also for the MECA-79 epitope (29). Structural investigation of mouse GlyCAM-1 found sulfation at C-6 of GlcNAc (i.e., GlcNAc-6-SO4) (47). Importantly, this modification occurs in sLex to yield 6-sulfo sLex (Table 1) (47). In the simplest O-glycans of GlyCAM-1, 6-sulfo sLex occurs as a capping group on the core 2 branch (upper arm in Figure 3) (47). 6-sulfation of sLex enhances its interaction with L-selectin in cell-free assays (26, 48). Also, changing sLex to 6-sulfo sLex on the surface of transfected cells increases binding by L-selectin (49–51). Two mAbs, G72 and G152, recognize 6-sulfo sLex in a sulfation- and sialic acid–dependent manner (52). These antibodies strongly stain HEV in human lymphoid organs and competitively inhibit the binding of L-selectin to these vessels. Taken together, these findings qualify 6-sulfo sLex as a minimal recognition determinant for L-selectin. The original studies of mouse GlyCAM-1 found significant levels of Gal-6sulfate within its O-glycans and inferred the existence of a 60 -sulfo sLex capping structure (Table 1) (47). By contrast, the ligands on human lymph node or tonsillar HEV in humans lack detectable Gal-6-sulfation and 60 -sulfo sLex (44, 45, 52). Whether or not Gal-6-sulfation enhances the interaction of sLex with L-selectin is controversial (26, 53). Furthermore the existence of an enzymatic route for the synthesis of 60 -sulfo sLex is in question (54). Conceivably, Gal-6-sulfation in other structural contexts contributes to ligand activity in the murine system (55). The fact that many of the O-glycans of GlyCAM-1 possess multiple sulfation modifications (47) poses challenges for future structural and functional analyses. The structural elucidation of the MECA-79 epitope has provided further insights into the structure of L-selectin ligands (56). The critical structure is 6-sulfated N-acetyllactosamine (6-sulfo LacNAc) on an extended core 1 O-glycan (Table 1, Figure 3). This oligosaccharide is found in the O-glycans of both native and recombinant GlyCAM-1. In its fully substituted form, the structure is modified with α2 → 3 sialylation on galactose and α1 → 3 fucosylation on GlcNAc, thus providing the 6-sulfo sLex determinant as a capping group. Interestingly, sialylation and fucosylation are dispensable for the MECA-79 epitope, whereas they are required for optimal binding by L-selectin. Native GlyCAM-1 from murine lymph nodes exhibits biantennary O-glycans, both of which are capped by 6-sulfo sLex. The lower arm is the extended core 1 glycan containing the MECA-79 epitope, and the upper arm is the core 2 branch. Mass spectrometry analysis of human tonsillar CD34 is consistent with the existence of the sulfated extended core 1 structure (Figure 3) (44). The dual presentation of the capping group in the biantennary O-glycan potentiates the L-selectin ligand activity of recombinant GlyCAM-1 as compared to forms of GlyCAM-1 in which 6-sulfo sLex is presented only on a single branch (56). This result indicates that both branches are capable of engaging L-selectin. Biacore measurements have shown that the monovalent interaction of L-selectin with native GlyCAM-1 occurs with a Kd of 108 µM, whereas the interaction of oligomeric L-selectin has a much higher affinity (57). Clustering of L-selectin on the cell surface, as is known to occur for P-selectin (35), may allow it to more effectively engage arrays of 6-sulfo sLex on mucin scaffolds, thus promoting

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

L-SELECTIN LIGANDS

137

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 2 Enzymes implicated in the key posttranslational modifications of PNAd for L-selectin binding Structure

Enzyme

References

Core 2 branch

Core2GlcNAcT-I

60

β1 → 4-linked Gal

β4GalT-IV

71

Extended core 1

β3GlcNAcT-3

56

GlcNAc-6-sulfate

HEC-GlcNAc6ST and GlcNAc6T-1

67, 68

α2 → 3-linked Sia

ST3Gal-III, ST3Gal-IV or ST3Gal-VI

70

α1 → 3-linked Fuc

FucT-VII and FucT-IV

54

adhesive interactions through avidity effects. Indirect evidence for clustering of L-selectin has been presented (58).

Enzymes Involved in Posttranslational Modifications The essential role of glycans in the function of HEV-expressed ligands has stimulated considerable interest in the identification of the glycosyltransferases and sulfotransferases involved in their elaboration (59). Table 2 summarizes the known and suspected participation of cloned enzymes in ligand biosynthesis. The relevant enzymes are all type II transmembrane proteins, each with amino-terminal tail in the cytosol, a transmembrane domain, and a larger catalytic domain in the golgi/ER lumen. Glycosyltransferases catalyze the transfer of nucleotide donor sugars to acceptor structures, whereas sulfotransferases catalyze the formation of sulfate esters on acceptor glycans or tyrosines utilizing the high energy sulfate donor PAPS (30 -phosphoadenosine 50 -phosphosulfate) (59). The core 2 branch on the O-glycans of PNAd mucins, such as GlyCAM-1, requires the action of a core 2 β1 → 6-N-acetylglucosaminotransferase (Core2GlcNAcT), which adds GlcNAc in a β1 → 6 linkage to GalNAc in core 1 structures. For ligand formation in mouse lymph nodes, the relevant enzyme of this class is Core2GlcNAcT-I (60). The genetic deletion of this enzyme eliminates the formation of core 2-linked 6-sulfo sLex (56). However, substantial ligand activity and full MECA-79 reactivity persist in the HEV of these mice, apparently because of the extended core 1 glycan, which allows for the presentation of the 6-sulfo sLex determinant. The core 1 extension enzyme in mouse HEV is β3GlcNAcT-3, a member of the family that adds GlcNAc in a β1 → 3 linkage to Gal (61). This enzyme shows a restricted pattern of expression at the mRNA level that includes intestine, colon, placenta, and tonsillar HEV (56). A strong prediction is that null mice for this enzyme will not express MECA-79 reactivity on lymph node HEV. Mice that are doubly deficient in β3GlcNAcT-3 and Core2GlcNAcT-I are predicted to be devoid of 6-sulfo sLex on both arms (Figure 3) and to show a severe homing deficiency.

11 Feb 2004

16:54

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

138

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN

Considerable progress has been made in defining the enzymes responsible for the 6-sulfo sLex capping group, which is thought to be directly engaged by the L-selectin lectin domain. Genetic studies in mouse have clearly shown that α1 → 3 fucosylation of GlcNAc is necessary for lymph node HEV ligand function and that FucT-VII is the principal enzyme involved (41, 54, 62). Thus, FucT-VII null mice show a 80%–90% reduction of lymphocyte homing to peripheral lymph nodes, the loss of staining of HEV by an L-selectin/IgM chimera, and markedly reduced ligand activity of isolated GlyCAM-1. A small contribution to ligand function is attributable to another fucosyltransferase, FucT-IV. Whereas mice that are FucTIV null show no loss of homing or impairment of ligand function, mice that are null for both FucTs show profound deficits in both activities, exceeding the effects observed in FucT-VII null mice. Further evidence for the relevance of FucT-VII is that this enzyme can act on 6-sulfo α2→3 sialylated N-acetyllactosamine to form 6-sulfo sLex in vitro (62). Coinciding with a general interest in elucidating the entire sulfotransferase gene family has been the more focused objective of identifying the GlcNAc-6-Osulfotransferases (GlcNAc6STs) involved in the formation of L-selectin ligands. Six members of the subfamily of GlcNAc6STs have been cloned and characterized (63, 64). One of these, known principally as HEC-GlcNAc6ST or LSST (L-selectin ligand sulfotransferase) has received the most attention because of its highly restricted expression in lymph node and tonsillar HEV (49, 65). The enzyme is also referred to as GST-3, CHST4, or GlcNAc6ST-2. In combination with exogenously provided cDNAs (FucT-VII, Core2GlcNAcT-I) and endogenous enzymes (α2 → 3 sialyltransferases), HEC-GlcNAc6ST promotes the L-selectin ligand activity of GlyCAM-1 and CD34 (49, 65, 66). Importantly, HEC-GlcNAc6ST contributes to the formation of the 6-sulfo sLex epitope in transfected cells (49) and the MECA79 epitope when β3GlcNAcT-3 cDNA (core 1 extension enzyme) is included (56). A critical in vivo role for HEC-GlcNAc6ST was established through the analysis of gene-targeted mice (67). First, in vitro binding of lymphocytes to lymph node HEV and luminal staining of HEV by an L-selectin chimera are eliminated in the null mice. Additionally, MECA-79 reactivity of HEV is greatly diminished, with the residual staining being most apparent on the abluminal aspect of these vessels. The ability of lymphocytes to home to null lymph nodes is reduced to 50% of that in wild-type mice, but the residual homing remains entirely L-selectin dependent. An excellent candidate to account for the remaining MECA-79 reactivity in the null mice is another member of the GlcNAc-6-O-sulfotransferase family, namely GlcNAc6ST-1, also known as GST-2 or CHST2 (68). Like HECGlcNAc6ST, this enzyme promotes the formation of L-selectin ligand activity, the MECA-79 epitope, and 6-sulfo sLex determinant in transfected cells (50). Moreover, the enzyme is expressed in lymph node HEV, although its tissue expression is much broader than that of HEC-GlcNAc6ST. Galactose-6-O-sulfation (55) and other ligand-generating strategies (69) also have to be considered as potential contributing factors to the residual ligands. Although sialylation was the first recognized posttranslational modification required for L-selectin ligands (9), its enzymatic basis remains unresolved. Six

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

L-SELECTIN LIGANDS

P1: IKH

139

mammalian genes are known to add α2 → 3 sialylation to Gal (ST3Gal-I through VI) (70). Of these, three (ST3Gal-III, ST3Gal-IV, and ST3Gal-VI) can sialylate N-acetyllactosamine (Galβ1 → 4GlcNAc) in vitro and thus are candidates for involvement in the formation of sLex. The collaborative involvement of more than one sialyltransferase is indicated for the formation of ligands for E- and P-selectin (70). A similar scenario is likely to hold for L-selectin. Finally, the identification of the β1 → 4 galactosyltransferase (β4GalT) involved in the formation of the 6-sulfo sLex recognition determinant has received recent attention. It is clear that this action of this enzyme must follow the 6-sulfation of GlcNAc because the GlcNAc6STs act on GlcNAc only when it is terminal (64). Among the seven cloned β4GalTs, the relevant enzyme appears to be β4GalT-IV, as it is the only family member that efficiently adds Gal to the appropriate GlcNAc6-SO4-terminated glycans (71). Moreover, the broad distribution of this enzyme includes relatively strong expression in lymph nodes. Gene targeting has established that β4GalT-I makes a major contribution to the formation of P-selectin ligand activity on monocytes and neutrophils, whereas it has no detectable role in the elaboration of L-selectin ligands on HEV (72). Establishing a definitive role for β4GalT-IV in lymphocyte homing awaits a similar approach. In view of the foregoing discussion and the known acceptor preferences of the enzymes, a plausible scheme for the synthesis of 6-sulfo sLex would be the sequential action of the following enzymes on a GlcNAc-terminated acceptor: (A) HEC-GlcNAc6ST or GlcNAc6T-1; (B) β4GalT-IV; (C) α2 → 3 sialyltransferase (ST3Gal-III, ST3Gal-IV, or ST3Gal-VI); and (D) FucT-VII (54).

L-SELECTIN LIGANDS ON OTHER VASCULAR BEDS HEV-Expressed Ligands in Mucosal Lymphoid Organs Peyer’s patches (PP), one component of gut-associated lymphoid tissue, are found in the wall of the small intestine and participate in mucosal immune defense. Application of the Stamper-Woodruff assay to mouse PP failed to reveal the involvement of L-selectin in lymphocyte homing to this organ (6). Subsequently, in vivo homing studies and intravital microscopy established that naive lymphocytes tether and roll on PP HEV via L-selectin whereas activated/memory populations utilize α4β7 for these steps (73). Although L-selectin mediated rolling is faster in PP HEV than in PN HEV, efficient chemokine-mediated arrest is achieved because the rolling of naive lymphocytes is slowed through the braking action of the α4β7-MAdCAM-1 system. Very limited information is available about the nature of L-selectin ligands in mouse PP HEV. These molecules do not belong to the PNAd complex since administration of the MECA-79 mAb does not inhibit lymphocyte homing to PP, and the relatively weak staining of HEV by this antibody is predominantly in an abluminal pattern (for some strains of mice but not all) (27). Antibody blockade studies implicate MAdCAM-1 as a potential L-selectin ligand at this site, presumably through carbohydrate determinants presented by its mucin domain (73).

11 Feb 2004

16:54

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

140

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN

Gene targeting experiments have yet to be performed to substantiate this possibility. Core2GlcNAcT-I-deficient mice show only a slight impairment in ligand activity within PP HEV (74). However, FucT-VII-deficient mice show a ∼60% deficiency in homing to PP, consistent with the known participation of L-selectin in this process (62). From these results, it is plausible that the PP HEV luminal ligand is based on sLex presented on an N-linked chain or on an extended core 1 O-glycan. It should be noted that in others species (sheep, rabbit, and pig), MECA-79 staining of PP HEV is intense, comparable to that of lymph node HEV (75). Thus, in these cases, MECA-79-reactive molecules are likely to be the predominant ligands engaged by L-selectin. Bronchus-associated lymphoid tissue (BALT) is a mucosal lymphoid organ that is morphologically and functionally related to PP (76). It is somewhat surprising that the adhesion cascades are very different for these two anatomic sites. Thus, the L-selectin/PNAd system predominates in the homing of lymphocytes to mouse BALT in correspondence with strong staining of HEV by MECA-79 (76). The homing of memory lymphocytes to BALT relies on the α4β1-VCAM-1 adhesion pair rather than the α4β7-MAdCAM-1 system utilized in PP. In mouse nasal-associated lymphoid tissue, another example of a mucosal lymphoid organ, the L-selectin/PNAd system also makes a substantial contribution to lymphocyte adherence to HEV (77).

Induction of MECA-79+ Vessels at Sites of Chronic Inflammation Extralymphoid tissues are supplied with lymphocytes that perform surveillance and effector functions. Excessive recruitment of lymphocytes to these tissues results in the formation of chronic inflammatory lesions, which can cause disease. When inflammation persists for sustained periods, the inflammatory infiltrates frequently demonstrate features of lymphoid organs, including the compartmentalization of B and T cells, and the presence of dendritic cells and HEV-like vessels (78). Depending on the inflamed organ, these vessels express PNAd and/or MAdCAM-1. The overall process, referred to as lymphoid organ neogenesis, is observed in several common autoimmune diseases (e.g., rheumatoid arthritis, multiple sclerosis, thyroiditis, diabetes, and chronic inflammatory bowel disease) (78, 79). Lymphoid aggregates can also be induced experimentally in transgenic mice through the expression of various chemokines (e.g., CXCL13 and CCL21) or cytokines (e.g., LTαβ) at ectopic sites (78). Investigation of one of these models has shown a strong correlation between the expression of HEC-GlcNAc6ST protein in HEVlike vessels and MECA-79 staining (80). In other examples of chronic inflammation, infiltrates are more diffuse, lacking an organized lymphoid structure. Psoriatic skin lesions are generally of this character (81), and about half of the infiltrates in rheumatoid arthritis fall into this category (82a). MECA-79+ vessels, with either an HEV-like character or flatwalled, are frequently seen in a variety of inflamed sites (26, 43, 83). Renkonen

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

L-SELECTIN LIGANDS

P1: IKH

141

and coworkers (24) have carried out systematic surveys in humans, involving large numbers of independent samples. In asthma, thyroiditis, heart and kidney allograft rejection, ulcerative colitis, and psoriasis, MECA-79 reactivity is present on a substantial proportion of CD34+ vessels in the infiltrates. Similarly, Salmi et al. (82b) determined that this epitope is present on a subpopulation of synovial vessels in the majority of chronic arthritis patients. In several mouse models of inflammation, MECA-79 staining of HEV-like vessels has been confirmed to reflect L-selectin ligand activity. Thus, the staining has been correlated with positive reactivity with an L-selectin/IgG chimera (84) or L-selectin-dependent binding of lymphocytes in the Stamper-Woodruff assay (85). Most convincing has been the demonstration that an L-selectin mAb and MECA-79 both effectively inhibit short-term lymphocyte homing to the inflamed organs (86, 87). Substantiation of the importance of the PNAd complex in pathogenesis awaits the use of MECA-79 or a related antibody in a model of chronic inflammatory disease.

Ligands Associated with Acutely Inflamed Vessels Many studies have demonstrated an involvement of L-selectin in the recruitment of leukocytes to sites of acute inflammation (16). The standard approaches have been based on use of function-blocking antibodies or by exploiting L-selectin-deficient mice. For example, in the classic model of thioglycollate-induced inflammation in the peritoneum, the absence of L-selectin results in a significant reduction in the number of neutrophils, macrophages, and lymphocytes recruited between 24 and 48 h (88). L-selectin mAbs are protective in multiple models of ischemiareperfusion injury and hemorrhagic shock (17), thus establishing the relevance of L-selectin to actual disease processes. A humanized antibody against L-selectin is currently under development by Scil Biomedicals (http://www.scil.de) for the treatment of multi-organ failure following severe trauma. Intravital microscopy has allowed the visualization of L-selectin-dependent interactions of leukocytes with acutely inflamed vessels. The most thoroughly studied system has been the mouse cremaster muscle, in which surgical exteriorization induces rolling of leukocytes in venules (37). Initially, the rolling of granulocytes is exclusively due to P-selectin but by 1 h postsurgery shows a codependency between L-selectin and P-selectin. In TNF-α treated preparations, L-selectin cooperates with E- and P-selectin to mediate rolling. Other systems have been described in which L-selectin predominates in mediating rolling. For example, Lselectin transfectants lacking ligands for E- or P-selectin roll in mesenteric venules of rat (89). More physiologically, the rolling of na¨ıve T cells in the cremaster preparation is almost exclusively dependent on L-selectin (81). It is now clear that leukocyte rolling on the vascular endothelium is more complicated than just the primary interaction of the two cell types. This possibility was initially suggested by in vitro observations showing that neutrophils can roll on endothelial-adherent neutrophils (90). This process is mediated through

11 Feb 2004

16:54

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

142

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN

L-selectin on the rolling cells. Alon et al. (91) further observed L-selectin-dependent interleukocyte tethering events that enhance the accumulation of neutrophils (in “strings”) on several different adhesive substrata. The term secondary tethering (91), or secondary capture (92), have been coined for this phenomenon. High resolution imaging in mice has now confirmed the in vivo relevance of secondary capture and the molecular basis for it. Eriksson et al. (92) observed that free-flowing cells can tether on cells already rolling on vascular endothelium, converting the former cells into rolling cells. This phenomenon occurs on venules after surgical trauma, on arterial vessels after cytokine exposure, and on atherosclerotic lesions in the aorta. Secondary tethering can account for 20%–50% of the overall rolling flux. Importantly, the leukocyte-leukocyte interactions are completely dependent on Lselectin, whereas the primary interactions between leukocytes and endothelium are L-selectin independent (92). These new findings can rationalize the functional synergy between L-selectin and the vascular selectins discussed above. The nature of the L-selectin ligands operative in the acute settings of inflammation has been a subject of some interest for many years. MECA-79 staining of acutely inflamed vessels have not been reported. In the search for candidate molecules, various L-selectin ligand activities have been described on different types of cultured endothelium. These include a heparan sulfate-based activity on arterial endothelium (93), a sulfation-dependent ligand on TNF-α stimulated microvascular endothelial cells (94), and a sulfation- and sLex-dependent (but MECA-79-unreactive) ligand on TNF-α stimulated HUVEC (95). One molecularly defined ligand candidate is endoglycan (Figure 2) (96). It is found on vascular endothelium as well as leukocyte subpopulations. The appropriately modified recombinant protein can function as an L-selectin ligand (69). See the next section for a complete description of this molecule. Thus far, evidence is lacking on the in vivo relevance of these ligand candidates. The recent appreciation of the role of L-selectin in the secondary capture phenomenon has focused attention on leukocyte-expressed PSGL-1 as the relevant Lselectin ligand in acute inflammatory settings rather than an endothelial-expressed molecule. Antibody blockade of PSGL-1 or genetic deletion of this molecule eliminates the L-selectin-dependent component of rolling in the cremaster muscle model (97). Consistent with the findings of Eriksson et al. (92), L-selectin-dependent rolling events are attributable to secondary tethers to endothelial-adherent leukocytes or to fragments thereof (97).

LEUKOCYTE LIGANDS FOR L-SELECTIN PSGL-1 Antibody blockade and genetic deletion studies have established that PSGL-1 is the major leukocyte ligand for both P- and L-selectin (98–101). This sialomucin is a type I transmembrane protein that is broadly expressed on leukocytes (34). P- and L-selectin bind to the extreme amino-terminal region of PSGL-1,

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

L-SELECTIN LIGANDS

143

which is exposed after biosynthetic processing (35). Extensive studies have defined the posttranslational modifications of PSGL-1 that enable its binding to the selectins (35, 102–105). Essential are 3 tyrosine sulfate moieties (Tyr-46, Tyr-48, and Tyr-51) and a single O-glycan (Thr-58) bearing sLex on a core 2 branched O-glycan. All three tyrosine sulfates contribute to the ligand activity for L- and P-selectin, although the relative importance of the individual tyrosine sulfates differs for the two. Interestingly, the monovalent affinity of the aminoterminal fragment for L-selectin is about eightfold lower than for P-selectin (Kds of 5 µM versus 650 nM) (104). However, this affinity is 20-fold stronger than that reported for GlyCAM-1 (57). As is the case for L-selectin-mediated rolling on PNAd elements (see above), leukocytes require a minimum shear stress in order to roll on PSGL-1 (104). This shear-threshold phenomenon would help to minimize leukocyte-leukocyte interactions in the center of vessels where shear is absent and promote interactions (secondary capture) near the wall where shear is highest (38). Considerable information is available about the enzymes that provide the essential posttranslational modifications to L-selectin ligands (i.e., principally PSGL-1) on leukocytes. The available information is summarized in Table 3. Interestingly, GlcNAc-6-sulfation of PSGL-1 has a dramatic impact on its Lselectin ligand activity. Thus, PSGL-1 modified with 6-sulfo sLex, as imparted by transfection with FucT-VII and GlcNAc6ST-1, is a much superior L-selectin ligand than the sLex-modified form (106). Moreover, the amino-terminal tyrosine sulfates are rendered dispensable for its ligand activity. In contrast, P-selectin ligand activity is only marginally enhanced by the 6-sulfation of sLex while maintaining a critical dependency on tyrosine sulfation. These results indicate that L-selectin prefers ligands (PSGL-1 or other mucins) based on 6-sulfo sLex rather than sLex/Tyr sulfation, whereas the reverse appears to be true for P-selectin. The key question is whether 6-sulfo sLex actually decorates ligands on leukocytes. Staining of granulocytes by a 6-sulfo sLex-specific mAb has not been reported, although a small subpopulation of CD4 memory T cells is positive (107). Differentiated NK cells express a sulfated lactosamine epitope (PEN5) on PSGL-1 (108). This epitope appears to confer L-selectin binding activity, which is

TABLE 3 Enzymes implicated in the key posttranslational modifications of leukocyte ligands for L-selectin Structure

Enzyme

References

Core 2 branch

Core2GlcNAcT-I

60

β1 → 4-linked Gal

β4GalT-I

72

Tyrosine sulfate

TPST-1/TPST-2

105

α2 → 3-linked Sia

ST3Gal-IV

70

α1 → 3-linked Fuc

FucT-VII

60

11 Feb 2004

16:54

144

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN

independent of the amino-terminal tyrosine sulfates. Although PEN5 is suspected to involve GlcNAc-6-sulfate, its structure and relationship to 6-sulfo sLex remain to be defined.

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Other Ligands A variety of experiments point to the existence of PSGL-1 independent ligands on leukocytes. For example, neutrophil-neutrophil interactions that form under shear stress are completely dependent on L-selectin but are incompletely inhibited (60%) by a PSGL-1 mAb (109). Similarly, the rolling of myeloid cells on an L-selectin substratum is partially blocked by PSGL-1 antibodies, whereas rolling on a Pselectin substratum is totally inhibited (99). A detergent extract of HL60 retains Lselectin ligand activity after immunodepletion of PSGL-1. As this residual activity is sensitive to both sialidase and OSGE, a sialomucin character is indicated. Endoglycan (EG) is a member of the CD34 family of sialomucins that also includes CD34 and podocalyxin (Figure 2). Although its tissue distribution is incompletely defined at this point, it is known to be present on leukocytes subpopulations, hematopoietic precursors, and vascular endothelium (96). Like CD34 and podocalyxin, endoglycan possesses a mucin domain. However, in distinction, it possesses an amino terminal domain of 174 amino acids with an abundance of acidic residues and two tyrosine sulfates (Figure 2). Endoglycan exhibits striking parallels with PSGL-1 (69). Both form disulfide bond-dependent heterodimers in SDS gels. There is significant homology in a critical 9 amino acid span in the acidic region, including identity at the residues corresponding to Tyr 51 (Tyr-118 in EG) and Thr 58 (Thr-124 in EG) in PSGL-1. The acidic domain, when modified with FucT-VII, supports rolling of Jurkat T cells through L-selectin. The ligand activity depends on the two tyrosine sulfates and sLex on core 2-linked O-glycans (predominantly Tyr-124). Whether leukocyte-expressed endoglycan functions as ligand for E- and P-selectin awaits further studies. In view of its expression on vascular endothelium, the role of endoglycan as an endothelial ligand for L-selectin also merits attention. Finally, it is very noteworthy that endoglycan carries several chondroitin sulfate chains. This class of glycosaminoglycan chains, if appropriately sulfated, can impart binding to both L-selectin and chemokines (110), raising further functional possibilities for this molecule. Chrondroitin sulfate-modified ligands for L-selectin have been reported on cultured HEC from rat lymph node (111).

“BEYOND”: LIGANDS IN SURPRISING PLACES Tumor Metastasis Tumor cells frequently exhibit altered cell surface glycosylation (112). The cell surface expression of sLex/sLea epitopes (both of which are recognition determinants for selectins) is positively correlated with tumor progression and metastasis in humans (113). Furthermore, human carcinoma cell lines, such as LS180, express sialomucins that can serve as L- and P-selectin ligands (114). The vivo

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

L-SELECTIN LIGANDS

P1: IKH

145

relevance of these findings has been explored in mouse models of metastasis (115, 116). Utilizing sialomucin-based ligands on LS180 cells, L-selectin and P-selectin synergistically facilitate metastasis in immunodeficient mice. Similarly, the lung metastasis of a mouse adenocarcinoma cell in syngeneic mice is critically dependent on these selectins. In this case, the P-selectin ligands on the carcinoma cells are predominantly sulfated glycolipids (e.g., sulfatide), whereas the L-selectin ligand is carried by a combination of sulfated glycolipids, sialomucins, and glycosaminoglycans. P-selectin appears to act first in the metastatic process, probably by mediating the interaction of activated platelets with the carcinoma cells. Acting in subsequent steps, L-selectin may provide the platelet-carcinoma emboli with a leukocyte “coat.” The increased size of the aggregates could facilitate mechanical trapping of the emboli in the microvasculature; alternatively, the leukocyte layer could bridge the emboli to the vascular endothelium through L-selectin-mediated primary or secondary capture events. A related mechanism may explain why ectopic expression of L-selectin in carcinoma cells induces their metastasis to lymph nodes (117). The “commandeering” of selectin mechanisms by metastasizing cancer cells merits further study (118). Novel therapeutic approaches for the treatment of hematogenous metastasis may result.

Perivascular and Extravascular Ligands Streeter et al. (27) in their original description of the MECA-79 antibody noted that the epitope is found on abluminal aspects of HEV in addition to blood-exposed luminal regions. When sensitive immunohistochemical techniques are used, a reticular pattern of staining is observed, which emanates from the HEV and extends for a considerable distance into the parenchyma of the lymphoid organ (Figure 4). One

Figure 4 The reticular pattern of MECA-79 staining around HEV. Shown is a mouse lymph node section stained with MECA-79. The staining was performed using a fluorescently labeled second antibody. The epitope is present on both the luminal and abluminal aspects of HEV and extends into the lymph node parenchyma in a reticular pattern. The image was provided by Durwin Tsay.

11 Feb 2004

16:54

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

146

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN

provocative possibility is that these reticular ligands serve as tracks for lymphocytes, facilitating their dispersal from the HEV. Alternatively, these ligands could stimulate signaling in lymphocytes. Ligation of L-selectin on the surface of leukocytes activates signaling pathways that modulate integrin-mediated adhesivity and chemoattractant receptivity of the cells (119–122). Intriguingly, L-selectin ligands have been detected on efferent lymphatic vessels in lymph nodes, suggesting that L-selectin may be involved in exiting of lymphocytes from lymph nodes (123). Further study is needed to delineate the contributions of L-selectin to lymphocyte migration within the parenchyma of lymphoid organs. Studies by Hickey et al. (124) support a role for L-selectin in extravascular migration of leukocytes in an acute inflammatory setting. Superfusing the mouse cremaster muscle preparation with a chemoattractant (platelet activating factor or KC) induces emigration of leukocytes. In the absence of L-selectin or with this molecule functionally neutralized, the distance that extravasated leukocytes migrate is reduced by 50% or more. The nature of the postulated perivascular or interstitial ligands is open to speculation. Proteoglycans are candidates, since certain forms of chondroitin sulfate and heparan sulfate proteoglycans exhibit L-selectin ligand activity (125–127). In a rat kidney model of inflammation, interstitial sulfatide has been implicated as an L-selectin ligand for infiltrating mononuclear cells. (128). It is noteworthy that sulfatide, as distinct from PNAd elements, mediates L-selectin-dependent adhesive interactions under static conditions (129). This property is consistent with ligand function in the extravascular compartment where shear forces would be minimal (see below).

Ligands on Myelin In 1979, Woodruff adapted her in vitro adhesion assay to brain sections and showed that lymphocytes selectively adhere to white matter regions of the central nervous system (Figure 5) (130). After the discovery of L-selectin, it was shown that Lselectin is required for this interaction (131, 132). Ligands were directly demonstrated with an L-selectin fusion protein as a histochemical probe. The ligands are present on myelin sheaths of axons in the CNS but not in the peripheral nervous system. In contrast to HEV ligands, the myelin ligands are sialic acid–independent and are extractable with organic solvents. A prominent candidate for the activity is sulfatide, which is a major constituent of myelin sheaths and is known to bind to L-selectin (see above). From their results, Huang et al. (131) speculated that L-selectin may play a role in the targeting of leukocytes to myelinated axons during demyelinating diseases. Recent experiments of Grewal et al. (133) in a model of experimental allergic encephalomyelitis (EAE) have provided incentive to revisit this hypothesis. Transgenic mice with a T cell receptor specific for myelin basic protein (MBP) exhibit rapid onset of paralysis and demyelination after immunization with an MBP peptide. When the mice are put on an L-selectin null background, the animals fail to develop damage to myelin and do not show clinical signs of EAE. This protection is seen even though neither antigen activation of CD4 cells

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

L-SELECTIN LIGANDS

P1: IKH

147

Figure 5 In vitro adhesion of lymphocytes to white matter region of brain. The Stamper-Woodruff assay was performed with a section of mouse cerebellum. The darkly staining cells over the white matter tract (W) are adherent exogenous lymphocytes. Very few lymphocytes bind to the granular layer (G) or the molecular layer (M). The white matter region contains myelinated axons. Similar binding to other white matter regions of the CNS is seen. The binding to these regions requires L-selectin. The ligand is present on myelin sheaths of CNS axons.

nor leukocyte recruitment (T cells, B cells, and macrophages) into the CNS is compromised by the absence of L-selectin. Remarkably, EAE could be induced in L-selectin null animals by the adoptive transfer of wild-type macrophages. These experiments argue that the destruction of myelin during the effector phase of EAE requires L-selectin on extravasated macrophages. Whether L-selectin engagement of a myelin ligand is involved in this pathogenic process remains to be determined. Of great interest is whether L-selectin is a contributing factor to the effector phase of disease in other cases of inflammatory demyelination.

Implantation The first example of the presence of L-selectin on a nonhematopoietic cell is the remarkable finding that the human blastocyst-stage embryo, as well as trophoblasts, express this receptor in a functional form (134). Furthermore, uterine epithelium (human but not mouse) upregulates MECA-79-reactive determinants during the receptive period for implantation. That the ligand-receptor system might be functional at the fetal/maternal interface was shown through the application of the Stamper-Woodruff adhesion assay to endometrial biopsies. Trophoblasts interact

11 Feb 2004

16:54

148

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

with luteal-phase uterine epithelium (receptive) and much less so to follicular phase tissue (nonreceptive). The interaction is inhibited by L-selectin and MECA79 mAbs. As reviewed above, GlcNAc6ST-1 is one of the enzymes than can provide the sulfation modification recognized by MECA-79. Interestingly, gene profiling of endometrium shows that the expression of mRNA for this enzyme is upregulated during the window of implantation but is diminished in women with endometriosis, an infertility condition (135). Clearly, the L-selectin system will receive considerable attention from investigators who study implantation, placentation, and their disorders.

FUTURE PERSPECTIVES The study of L-selectin and its ligands will remain an active area for many years to come. Major structural and mechanistic questions remain as to the nature of the receptor ligand interactions and how these molecular interactions are translated into cellular interactions. Considerable efforts will be required to elucidate the full complement of ligand scaffolds and the posttranslational modifications involved in the vascular and extravascular ligand functions detailed in the present review. A second major theme for future work will be the continued investigation of the diverse and expanding functions for L-selectin. Further understanding of how Lselectin participates in the processes of inflammatory trafficking, metastasis, CNS demyelination, and implantation is likely to generate new opportunities for the treatment and diagnosis of human disorders. ACKNOWLEDGMENTS This work in the author’s lab is supported by grants from the NIH (R37GM23547 and GM57411), Sandler Foundation, and the Dana Foundation. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Gowans JL, Knight EJ. 1964. The route of recirculation of lymphocytes in the rat. Proc. R. Soc. London Ser. B 159:257–82 2. Butcher EC, Picker LJ. 1996. Lymphocyte homing and homeostasis. Science 272:60–66 3. Salmi M, Jalkanen S. 1997. How do lymphocytes know where to go: current concepts and enigmas of lymphocyte homing. Adv. Immunol. 64:139–218 4. von Andrian UH, Mackay CR. 2000. T-

cell function and migration. Two sides of the same coin. N. Engl. J. Med. 343:1020– 34 5. Stamper H, Woodruff J. 1976. Lymphocyte homing into lymph nodes: in vitro demonstration of the selective affinity of recirculating lymphocytes for highendothelial venules. J. Exp. Med. 144: 828–33 6. Gallatin W, Weissman I, Butcher E. 1983. A cell-surface molecule involved

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

L-SELECTIN LIGANDS

7.

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

in organ-specific homing of lymphocytes. Nature 304:30–34 Lasky LA. 1995. Selectin-carbohydrate interactions and the initiation of the inflammatory response. Annu. Rev. Biochem. 64:113–39 Stoolman LM, Rosen SD. 1983. Possible role for cell surface carbohydrate-binding molecules in lymphocyte recirculation. J. Cell Biol. 96:722–29 Rosen SD, Singer MS, Yednock TA, Stoolman LM. 1985. Involvement of sialic acid on endothelial cells in organ-specific lymphocyte recirculation. Science 228: 1005–7 Yednock TA, Butcher EC, Stoolman LM, Rosen SD. 1987. Receptors involved in lymphocyte homing: relationship between a carbohydrate-binding receptor and the MEL-14 antigen. J. Cell Biol. 104:725–31 Arbon´es ML, Ord DC, Ley K, Ratech H, Maynard-Curry C, et al. 1994. Lymphocyte homing and leukocyte rolling and migration are impaired in L-selectindeficient mice. Immunity 1:247–60 Drickamer K, Taylor ME. 1993. Biology of animal lectins. Annu. Rev. Cell Biol. 9: 237–64 Kansas GS. 1996. Selectins and their ligands: current concepts and controversies. Blood 88:3259–87 Vestweber D, Blanks JE. 1999. Mechanisms that regulate the function of the selectins and their ligands. Physiol. Rev. 79: 181–213 Springer TA. 1994. Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76:301–14 Tedder TF, Steeber DA, Chen A, Engel P. 1995. The selectins: vascular adhesion molecules. FASEB J. 9:866–73 Winn R, Vedder N, Ramamoorthy C, Sharar S, Harlan J. 1998. Endothelial and leukocyte adhesion molecules in inflammation and disease. Blood Coagul. Fibrinolysis 9(Suppl. 2):S17–23

P1: IKH

149

18. Gesner BM, Ginsburg V. 1964. Effect of glycosidases on the fate of transfused lymphocytes. Proc. Natl. Acad. Sci. USA 52:750–55 19. Andrews P, Ford W, Stoddart R. 1980. Metabolic studies of high-walled endothelium of postcapillary venules in rat lymph nodes. CIBA Symp. 71:211–30 20. Imai Y, Singer MS, Fennie C, Lasky LA, Rosen SD. 1991. Identification of a carbohydrate-based endothelial ligand for a lymphocyte homing receptor. J. Cell Biol. 113:1213–21 21. Puri KD, Finger EB, Gaudernack G, Springer TA. 1995. Sialomucin CD34 is the major L-selectin ligand in human tonsil high endothelial venules. J. Cell Biol. 131:261–70 22. Sassetti C, Tangemann K, Singer MS, Kershaw DB, Rosen SD. 1998. Identification of podocalyxin as an HEV ligand for L-selectin: parallels to CD34. J. Exp. Med. 187:1965–75 23. Rasmussen LK, Johnsen LB, Petersen TE, Sorensen ES. 2002. Human GlyCAM1 mRNA is expressed in the mammary gland as splicing variants and encodes various aberrant truncated proteins. Immunol. Lett. 83:73–75 24. Renkonen J, Tynninen O, Hayry P, Paavonen T, Renkonen R. 2002. Glycosylation might provide endothelial zip codes for organ-specific leukocyte traffic into inflammatory sites. Am. J. Pathol. 161:543– 50 25. McNagny KM, Pettersson I, Rossi F, Flamme I, Shevchenko A, et al. 1997. Thrombomucin, a novel cell surface protein that defines thrombocytes and multipotent hematopoietic progenitors. J. Cell Biol. 138:1395–407 26. Rosen SD. 1999. Endothelial ligands for L-selectin: from lymphocyte recirculation to allograft rejection. Am. J. Pathol. 155: 1013–20 27. Streeter PR, Rouse BTN, Butcher EC. 1988. Immunohistologic and functional characterization of a vascular addressin

11 Feb 2004

16:54

150

28.

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

29.

30.

31.

32.

33.

34.

35.

36.

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN involved in lymphocyte homing into peripheral lymph nodes. J. Cell Biol. 107: 1853–62 von Andrian UH. 1996. Intravital microscopy of the peripheral lymph node microcirculaton in mice. Microcirculation 3:287–300 Hemmerich S, Butcher EC, Rosen SD. 1994. Sulfation-dependent recognition of HEV-ligands by L-selectin and MECA 79, an adhesion-blocking mAb. J. Exp. Med. 180:2219–26 Samulowitz U, Kuhn A, Brachtendorf G, Nawroth R, Braun A, et al. 2002. Human endomucin: distribution pattern, expression on high endothelial venules, and decoration with the MECA-79 epitope. Am. J. Pathol. 160:1669–81 Berg EL, McEvoy LM, Berlin C, Bargatze RF, Butcher EC. 1993. Lselectin-mediated lymphocyte rolling on MAdCAM-1. Nature 366:695–98 Clark RA, Fuhlbrigge RC, Springer TA. 1998. L-selectin ligands that are Oglycoprotease resistant and distinct from MECA-79 antigen are sufficient for tethering and rolling of lymphocytes on human high endothelial venules. J. Cell Biol. 140:721–31 Suzuki A, Andrew DP, Gonzalo JA, Fukumoto M, Spellberg J, et al. 1996. CD34deficient mice have reduced eosinophil accumulation after allergen exposure and show a novel crossreactive 90-kD protein. Blood 87:3550–62 Sako D, Chang XJ, Barone KM, Vachino G, White HM, et al. 1993. Expression cloning of a functional glycoprotein ligand for P-selectin. Cell 75:1179– 86 McEver RP, Cummings RD. 1997. Role of PSGL-1 binding to selectins in leukocyte recruitment. J. Clin. Invest. 100:S97–103 Puri KD, Finger EB, Springer TA. 1997. The faster kinetics of L-selectin than of E-selectin and P-selectin rolling at comparable binding strength. J. Immunol. 158:405–13

37. Ley K, Tedder TF. 1995. Leukocyte interactions with vascular endothelium. New insights into selectin-mediated attachment and rolling. J. Immunol. 155:525–28 38. Finger EB, Puri KD, Alon R, Lawrence MB, Von Andrian UH, Springer TA. 1996. Adhesion through L-selectin requires a threshold hydrodynamic shear. Nature 379:266–69 39. Chen SQ, Springer TA. 1999. An automatic braking system that stabilizes leukocyte rolling by an increase in selectin bond number with shear. J. Cell Biol. 144:185–200 40. Varki A. 1997. Selectin ligands: Will the real ones please stand up? J. Clin. Invest. 99:158–62 41. Lowe JB. 2002. Glycosylation in the control of selectin counter-receptor structure and function. Immunol. Rev. 186:19–36 42. Poppe L, Brown GS, Philo JS, Nikrad PV, Shah BH. 1997. Conformation of sLex tetrasaccharide, free in solution and bound to E-, P-, and L-selectin. J. Am. Chem. Soc. 119:1727–36 43. Girard J-P, Springer TA. 1995. High endothelial venules: specialized endothelium for lymphocyte migration. Immunol. Today 16:449–57 44. Satomaa T, Renkonen O, Helin J, Kirveskari J, Makitie A, Renkonen R. 2002. O-glycans on human high endothelial CD34 putatively participating in L-selectin recognition. Blood 99:2609–11 45. Van Zante A, Rosen SD. 2003. Sulphated endothelial ligands for L-selectin in lymphocyte homing and inflammation. Biochem. Soc. Trans. 31:313–17 46. Imai Y, Lasky LA, Rosen SD. 1993. Sulphation requirement for GlyCAM-1, an endothelial ligand for L-selectin. Nature 361:555–57 47. Hemmerich S, Leffler H, Rosen SD. 1995. Structure of the O-glycans in GlyCAM1, an endothelial-derived ligand for Lselectin. J. Biol. Chem. 270:12035–47 48. Scudder PR, Shailubhai K, Duffin KL, Streeter PR, Jacob GS. 1994. Enzymatic

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

L-SELECTIN LIGANDS

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

49.

50.

51.

52.

53.

54.

55.

synthesis of a 6-sulfated sialyl-Lewisx which is an inhibitor of L-selectin binding to peripheral addressin. Glycobiology 4:929–33 Bistrup A, Bhakta S, Lee JK, Belov YC, Gunn MD, et al. 1999. Sulfotransferases of two specificities function in the reconstitution of high-endothelial-cell ligands for L-Selectin. J. Cell Biol. 145:899– 910 Kimura N, Mitsuoka C, Kanamori A, Hiraiwa N, Uchimura K, et al. 1999. Reconstitution of functional L-selectin ligands on a cultured human endothelial cell line by cotransfection of alpha1–3 fucosyltransferase VII and newly cloned GlcNAc beta: 6-sulfotransferase cDNA. Proc. Natl. Acad. Sci. USA 96:4530–35 Kanamori A, Kojima N, Uchimura K, Muramatsu T, Tamatani T, et al. 2002. Distinct sulfation requirements of selectins disclosed using cells that support rolling mediated by all three selectins under shear flow. L-selectin prefers carbohydrate 6-sulfation totyrosine sulfation, whereas P-selectin does not. J. Biol. Chem. 277:32578–86 Mitsuoka C, Sawada-Kasugai M, AndoFurui K, Izawa M, Nakanishi H, et al. 1998. Identification of a major carbohydrate capping group of the L-selectin ligand on high endothelial venules in human lymph nodes as 6-sulfo sialyl Lewis X. J. Biol. Chem. 273:11225–33 Bruehl RE, Bertozzi CR, Rosen SD. 2000. Minimal sulfated carbohydrates for recognition by L-selectin and the MECA79 antibody. J. Biol. Chem. 275:32642–48 Homeister JW, Thall AD, Petryniak B, Maly P, Rogers CE, et al. 2001. The alpha(1,3)fucosyltransferases FucT-IV and FucT-VII exert collaborative control over selectin-dependent leukocyte recruitment and lymphocyte homing. Immunity 15:115–26 Galustian C, Childs RA, Stoll M, Ishida H, Kiso M, Feizi T. 2002. Synergistic interactions of the two classes of ligand,

56.

57.

58.

59.

60.

61.

62.

63.

64.

P1: IKH

151

sialyl-Lewis(a/x) fuco-oligosaccharides and short sulpho-motifs, with the P- and L-selectins: implications for therapeutic inhibitor designs. Immunology 105:350– 59 Yeh JC, Hiraoka N, Petryniak B, Nakayama J, Ellies LG, et al. 2001. Novel sulfated lymphocyte homing receptors and their control by a Core1 extension beta 1, 3-N-acetylglucosaminyltransferase. Cell 105:957–69 Nicholson MW, Barclay AN, Singer MS, Rosen SD, van der Merwe PA. 1998. Affinity and kinetic analysis of L-selectin (CD62L) binding to GlyCAM-1. J. Biol. Chem. 273:763–70 Toppila S, Lauronen J, Mattila P, Turunen JP, Penttil¨a L, et al. 1997. L-selectin ligands in rat high endothelium: Multivalent sialyl Lewis x glycans are high-affinity inhibitors of lymphocyte adhesion. Eur. J. Immunol. 27:1360–65 Lowe JB, Marth JD. 2003. A genetic approach to mammalian glycan function. Annu. Rev. Biochem. 72:643–91 Ellies LG, Tsuboi S, Petryniak B, Lowe JB, Fukuda M, Marth JD. 1998. Core 2 oligosaccharide biosynthesis distinguishes between selectin ligands essential for leukocyte homing and inflammation. Immunity 9:881–90 Mitoma J, Petryniak B, Hiraoka N, Yeh JC, Lowe JB, Fukuda M. 2003. Extended core 1 and core 2 branched O-glycans differentially modulate sialyl Lewis x-type L-selectin ligand activity. J. Biol. Chem. 278:9953–61 Maly P, Thall AD, Petryniak B, Rogers CE, Mith PL, et al. 1996. The α(1,3)fucosyltransferase Fuc-TVII controls leukocyte trafficking through an essential role in L-, E- and P-selectin ligand biosynthesis. Cell 86:643–53 Hemmerich S, Rosen SD. 2000. Carbohydrate sulfotransferases in lymphocyte homing. Glycobiology 10:849–56 Grunwell JR, Bertozzi CR. 2002. Carbohydrate sulfotransferases of the GalNAc/

11 Feb 2004

16:54

152

65.

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

66.

67.

68.

69.

70.

71.

72.

73.

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN Gal/GlcNAc6ST family. Biochemistry 41:13117–26 Hiraoka N, Petryniak B, Nakayama J, Tsuboi S, Suzuki M, et al. 1999. A novel, high endothelial venule-specific sulfotransferase expresses 6-sulfo sialyl lewis x, an L-selectin ligand displayed by CD34. Immunity 11:79–89 Tangemann K, Bistrup A, Hemmerich S, Rosen SD. 1999. Sulfation of an HEVexpressed ligand for L-selectin: effects on tethering and rolling of lymphocytes. J. Exp. Med. 190:935–41 Hemmerich S, Bistrup A, Singer MS, van Zante A, Lee JK, et al. 2001. Sulfation of L-selectin ligands by an HEV-restricted sulfotransferase regulates lymphocyte homing to lymph nodes. Immunity 15:237–47 Uchimura K, Muramatsu H, Kadomatsu K, Fan QW, Kurosawa N, et al. 1998. Molecular cloning and characterization of an N-acetylglucosamine-6-O-sulfotransferase. J. Biol. Chem. 273:22577–83 Fieger CB, Sassetti CM, Rosen SD. 2003. Endoglycan, a member of the CD34 family, functions as a L-selectin ligand through modification with tyrosine sulfation and sialyl Lewis x. J. Biol. Chem. 278:27390–98 Ellies LG, Sperandio M, Underhill GH, Yousif J, Smith M, et al. 2002. Sialyltransferase specificity in selectin ligand formation. Blood 100:3618–25 Seko A, Dohmae N, Takio K, Yamashita K. 2003. Beta 1,4-galactosyltransferase (beta 4GalT)-IV is specific for GlcNAc 6-O-sulfate. J. Biol. Chem. 278:9150–58 Asano M, Nakae S, Kotani N, Shirafuji N, Nambu A, et al. 2003. Impaired selectin ligand biosynthesis and reduced inflammatory responses in {beta}-1,4galactosyltransferase-I-deficient mice. Blood. 102:1678–85 Bargatze RF, Jutila MA, Butcher EC. 1995. Distinct roles of L-selectin and integrins α4β7 and LFA-1 in lymphocyte homing to Peyer’s patch-HEV in situ: the

74.

75.

76.

77.

78.

79.

80.

81.

multistep model confirmed and refined. Immunity 3:99–108 Sperandio M, Forlow SB, Thatte J, Ellies LG, Marth JD, Ley K. 2001. Differential requirements for core2 glucosaminyltransferase for endothelial Lselectin ligand function in vivo. J. Immunol. 167:2268–74 Mackay CR, Marston WL, Dudler L, Spertini O, Tedder TF, Hein WR. 1992. Tissue-specific migration pathways by phenotypically distinct subpopulations of memory T cells. Eur. J. Immunol. 22:887– 95 Xu BH, Wagner N, Pham LN, Magno V, Shan ZY, et al. 2003. Lymphocyte homing to bronchus-associated lymphoid tissue (BALT) is mediated by L-selectin/PNAd, alpha(4)beta(1) integrin/VCAM-1, and LFA-1 adhesion pathways. J. Exp. Med. 197:1255–67 Csencsits KL, Jutila MA, Pascual DW. 1999. Nasal-associated lymphoid tissue: phenotypic and functional evidence for the primary role of peripheral node addressin in naive lymphocyte adhesion to high endothelial venules in a mucosal site. J. Immunol. 163:1382–89 Ruddle NH. 1999. Lymphoid neoorganogenesis: lymphotoxin’s role in inflammation and development. Immunol. Res. 19:119–25 Hjelmstrom P. 2001. Lymphoid neogenesis: de novo formation of lymphoid tissue in chronic inflammation through expression of homing chemokines. J. Leukoc. Biol. 69:331–39 Drayton DL, Ying XY, Lee J, Lesslauer W, Ruddle NH. 2003. Ectopic LT alpha beta directs lymphoid organ neogenesis with concomitant expression of peripheral node addressin and a HEV-restricted sulfotransferase. J. Exp. Med. 197:1153–63 Weninger W, Carlsen HS, Goodarzi M, Moazed F, Crowley MA, et al. 2003. Naive T cell recruitment to nonlymphoid tissues: a role for endothelium-expressed CC chemokine ligand 21 in autoimmune

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

L-SELECTIN LIGANDS

82a.

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

82b.

83.

84.

85.

86.

87.

88.

89.

disease and lymphoid neogenesis. J. Immunol. 170:4638–48 Takemura S, Braun A, Crowson C, Kurtin PJ, Cofield RH, et al. 2001. Lymphoid neogenesis in rheumatoid synovitis. J. Immunol. 167:1072–80 Salmi M, Rajala P, Jalkanen S. 1997. Homing of mucosal leukocytes to joints. Distinct endothelial ligands in synovium mediate leukocyte-subtype specific adhesion. J. Clin. Invest. 99:2165–72 Michie SA, Streeter PR, Bolt PA, Butcher EC, Picker LJ. 1993. The human peripheral lymph node vascular addressin. Am. J. Pathol. 143:1688–98 Onrust SV, Hartl PM, Rosen SD, Hanahan D. 1996. Modulation of L-selectin ligand expression during an immune response accompanying tumorigenesis in transgenic mice. J. Clin. Invest. 97:54–64 H¨anninen A, Taylor C, Streeter PR, Stark LS, Sarte JM, et al. 1993. Vascular addressins are induced on islet vessels during insulitis in nonobese diabetic mice and are involved in lymphoid binding to islet endothelium. J. Clin. Invest. 92:2509–15 Michie SA, Streeter PR, Butcher EC, Rouse RV. 1995. L-selectin and alpha 4 beta 7 integrin homing receptor pathways mediate peripheral lymphocyte traffic to AKR mouse hyperplastic thymus. Am. J. Pathol. 147:412–21 Mikulowska-Mennis A, Xu B, Berberian JM, Michie SA. 2001. Lymphocyte migration to inflamed lacrimal glands is mediated by vascular cell adhesion molecule-1/alpha(4)beta(1) integrin, peripheral node addressin/L-selectin, and lymphocyte function-associated antigen1 adhesion pathways. Am. J. Pathol. 159:671–81 Tedder TF, Steeber DA, Pizcueta P. 1995. L-selectin-deficient mice have impaired leukocyte recruitment into inflammatory sites. J. Exp. Med. 181:2259–64 Ley K, Tedder TF, Kansas GS. 1993. L-selectin can mediate leukocyte rolling in untreated mesenteric venules in vivo

90.

91.

92.

93.

94.

95.

96.

97.

98.

P1: IKH

153

independent of E- or P-selectin. Blood 82:1632–38 Bargatze RF, Kurk S, Butcher EC, Jutila MA. 1994. Neutrophils roll on adherent neutrophils bound to cytokine-induced endothelial cells via L-selectin on the rolling cells. J. Exp. Med. 180:1785–92 Alon R, Fuhlbrigge RC, Finger EB, Springer TA. 1996. Interactions through L-selectin and adherent leukocytes nucleate rolling adhesions on selectins and VCAM-1 in shear flow. J. Cell Biol. 135:849–65 Eriksson EE, Xie X, Werr J, Thoren P, Lindbom L. 2001. Importance of primary capture and L-selectin-dependent secondary capture in leukocyte accumulation in inflammation and atherosclerosis in vivo. J. Exp. Med. 194:205–18 Giuffr`e L, Cordey AS, Monai N, Tardy Y, Schapira M, Spertini O. 1997. Monocyte adhesion to activated aortic endothelium: role of L-selectin and heparan sulfate proteoglycans. J. Cell Biol. 136: 945–56 Zakrzewicz A, Grafe M, Terbeek D, Bongrazio M, Auch-Schwelk W, et al. 1997. L-selectin-dependent leukocyte adhesion to microvascular but not to macrovascular endothelial cells of the human coronary system. Blood 89:3228–35 Tu L, Delahunty MD, Ding H, Luscinskas FW, Tedder TF. 1999. The cutaneous lymphocyte antigen is an essential component of the L-selectin ligand induced on human vascular endothelial cells. J. Exp. Med. 189:241–52 Sassetti C, Van Zante M, Rosen SD. 2000. Identification of endoglycan, a member of the CD34/podocalyxin family of sialomucins. J. Biol. Chem. 275:9001–10 Sperandio M, Smith ML, Forlow SB, Olson TS, Xia L, et al. 2003. P-selectin glycoprotein ligand-1 mediates L-selectindependent leukocyte rolling in venules. J. Exp. Med. 197:1355–63 Moore KL, Patel KD, Bruehl RE, Fugang L, Johnson DA, et al. 1995. P-selectin

11 Feb 2004

16:54

154

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

99.

100.

101.

102.

103.

104.

105.

106.

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

P1: IKH

ROSEN glycoprotein ligand-1 mediates rolling of human neutrophils on P-selectin. J. Cell Biol. 128:661–71 Ramos CL, Smith MJ, Snapp KR, Kansas GS, Stickney GW, et al. 1998. Functional characterization of L-selectin ligands on human neutrophils and leukemia cell lines: evidence for mucinlike ligand activity distinct from P-selectin glycoprotein ligand-1. Blood 91:1067–75 Hirata T, Merrill-Skoloff G, Aab M, Yang J, Furie BC, Furie B. 2000. P-Selectin glycoprotein ligand 1 (PSGL-1) is a physiological ligand for E-selectin in mediating T helper 1 lymphocyte migration. J. Exp. Med. 192:1669–76 Xia L, Sperandio M, Yago T, McDaniel JM, Cummings RD, et al. 2002. P-selectin glycoprotein ligand-1-deficient mice have impaired leukocyte tethering to E-selectin under flow. J. Clin. Invest. 109:939– 50 Somers WS, Tang J, Shaw GD, Camphausen RT. 2000. Insights into the molecular basis of leukocyte tethering and rolling revealed by structures of P- and Eselectin bound Slex(X) and PSGL-1. Cell 103:467–79 Bernimoulin MP, Zeng XL, Abbal C, Giraud S, Martinez M, et al. 2003. Molecular basis of leukocyte rolling on PSGL-1. Predominant role of core-2 O-glycans and of tyrosine sulfate residue 51. J. Biol. Chem. 278:37–47 Leppanen A, Yago T, Otto VI, McEver RP, Cummings RD. 2003. Model glycosulfopeptides from PSGL-1 require tyrosine sulfation and a core-2 branched Oglycan to bind to L-selectin. J. Biol. Chem. 278:26391–400 Moore KL. 2003. The biology and enzymology of protein tyrosine O-sulfation. J. Biol. Chem. 278:24243–46 Kanamori A, Kojima N, Uchimura K, Muramatsu T, Tamatani T, et al. 2002. Distinct sulfation requirements of selectins disclosed using cells which support rolling mediated by all three selectins un-

107.

108.

109.

110.

111.

112.

113.

114.

115.

der shear flow. J. Biol. Chem. 277:32578– 86 Kannagi R. 2002. Regulatory roles of carbohydrate ligands for selectins in the homing of lymphocytes. Curr. Opin. Struct. Biol. 12:599–608 Andre P, Spertini O, Guia S, Rihet P, Dignat-George F, et al. 2000. Modification of P-selectin glycoprotein ligand-1 with a natural killer cell-restricted sulfated lactosamine creates an alternate ligand for L-selectin. Proc. Natl. Acad. Sci. USA 97:3400–5 Walcheck B, Moore KL, McEver RP, Kishimoto TK. 1996. Neutrophil-neutrophil interactions under hydrodynamic shear stress involve L-selectin and PSGL1. A mechanism that amplifies initial leukocyte accumulation of P-selectin in vitro. J. Clin. Invest. 98:1081–87 Kawashima H, Atarashi K, Hirose M, Hirose J, Yamada S, et al. 2002. Oversulfated chondroitin/dermatan sulfates containing GlcA beta 1/IdoA alpha 1-3GalNAc(4,6O-disulfate) interact with L- and Pselectin and chemokines. J. Biol. Chem. 277:12921–30 Derry CJ, Faveeuw C, Mordsley KR, Ager A. 1998. Novel chondroitin sulfate modified ligands for L-selectin on lymph node high endothelial venules. Eur. J. Immunol. 29:419–30 Fukuda M. 1996. Possible roles of tumorassociated carbohydrate antigens. Cancer Res. 56:2237–44 Nakamori S, Kameyama M, Imaoka S, Furukawa H, Ishikawa O, et al. 1993. Increased expression of sialyl Lewis x antigen correlates with poor survival in patients with colorectal carcinoma: clinicopathological and immunohistochemical study. Cancer Res. 53:3632–37 Mannori G, Crottet P, Cecconi O, Hanasaki K, Aruffo A, et al. 1995. Differential colon cancer cell adhesion to E-, P-, and Lselectin: role of mucintype glycoproteins. Cancer Res. 55:4425–31 Borsig L, Wong R, Feramisco J, Nadeau

11 Feb 2004

16:54

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

LaTeX2e(2002/01/18)

L-SELECTIN LIGANDS

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

116.

117.

118. 119.

120.

121.

122.

123.

DR, Varki NM, Varki A. 2001. Heparin and cancer revisited: mechanistic connections involving platelets, P-selectin, carcinoma mucins, and tumor metastasis. Proc. Natl. Acad. Sci. USA 98:3352–57 Borsig L, Wong R, Hynes RO, Varki NM, Varki A. 2002. Synergistic effects of L- and P-selectin in facilitating tumor metastasis can involve non-mucin ligands and implicate leukocytes as enhancers of metastasis. Proc. Natl. Acad. Sci. USA 99: 2193–98 Qian F, Hanahan D, Weissman IL. 2001. L-selectin can facilitate metastasis to lymph nodes in a transgenic mouse model of carcinogenesis. Proc. Natl. Acad. Sci. USA 98:3976–81 Coussens LM, Werb Z. 2002. Inflammation and cancer. Nature 420:860–67 Waddell TK, Fialkow L, Chan CK, Kishimoto TK, Downey GP. 1995. Signaling functions of L-selectin. Enhancement of tyrosine phosphorylation and activation of MAP kinase. J. Biol. Chem. 270:15403– 11 Hwang ST, Singer MS, Giblin PA, Yednock TA, Bacon KB, et al. 1996. GlyCAM-1, a physiological ligand for Lselectin, activates β2 integrins on naive peripheral lymphocytes. J. Exp. Med. 184:1343–48 Giblin PA, Hwang ST, Katsumoto TR, Rosen SD. 1997. Ligation of L-selectin on T lymphocytes activates beta 1 integrins and promotes adhesion to fibronectin. J. Immunol. 159:3498–507 Tsang YT, Neelamegham S, Hu Y, Berg EL, Burns AR, et al. 1997. Synergy between L-selectin signaling and chemotactic activation during neutrophil adhesion and transmigration. J. Immunol. 159:4566–77 Irjala H, Johansson EL, Grenman R, Alanen K, Salmi M, Jalkanen S. 2001. Mannose receptor is a novel ligand for Lselectin and mediates lymphocyte binding to lymphatic endothelium. J. Exp. Med. 194:1033–42

P1: IKH

155

124. Hickey MJ, Forster M, Mitchell D, Kaur J, De Caigny C, Kubes P. 2000. L-selectin facilitates emigration and extravascular locomotion of leukocytes during acute inflammatory responses in vivo. J. Immunol. 165:7164–70 125. Kawashima H, Li YF, Watanabe N, Hirose J, Hirose M, Miyasaka M. 1999. Identification and characterization of ligands for L-selectin in the kidney. I. Versican, a large chondroitin sulfate proteoglycan, is a ligand for L-selectin. Int. Immunol. 11:393–405 126. Wang L, Brown JR, Varki A, Esko JD. 2002. Heparin’s anti-inflammatory effects require glucosamine 6-O-sulfation and are mediated by blockade of L- and Pselectins. J. Clin. Invest. 110:127–36 127. Kawashima H, Watanabe N, Hirose M, Sun X, Atarashi K, et al. 2003. Collagen XVIII, a basement membrane heparan sulfate proteoglycan, interacts with L-selectin and monocyte chemoattractant protein-1. J. Biol. Chem. 278:13069–76 128. Shikata K, Suzuki Y, Wada J, Hirata K, Matsuda M, et al. 1999. L-selectin and its ligands mediate infiltration of mononuclear cells into kidney interstitium after ureteric obstruction. J. Pathol. 188:93– 99 129. Alon R, Feizi T, Yuen C-T, Fuhlbrigge RC, Springer TA. 1995. Glycolipid ligands for selectins support leukocyte tethering and rolling under physiologic flow conditions. J. Immunol. 154:5356–66 130. Kuttner B, Woodruff J. 1979. Selective adherence of lymphocytes to myelinated areas of rat brain. J. Immunol. 122:1666–68 131. Huang K, Geoffroy JS, Singer MS, Rosen SD. 1991. A lymphocyte homing receptor (L-selectin) mediates the in vitro attachment of lymphocytes to myelinated tracts of the central nervous system. J. Clin. Invest. 88:1778–83 132. Huang K, Kikuta A, Rosen SD. 1994. Myelin localization of a central nervous system ligand for L-selectin. J. Neuroimmunol. 53:133–41

11 Feb 2004

16:54

156

AR

AR210-IY22-05.tex

AR210-IY22-05.sgm

P1: IKH

ROSEN

133. Grewal IS, Foellmer HG, Grewal KD, Wang H, Lee WP, et al. 2001. CD62L is required on effector cells for local interactions in the CNS to cause myelin damage in experimental allergic encephalomyelitis. Immunity 14:291–302 134. Genbacev OD, Prakobphol A, Foulk RA, Krtolica AR, Ilic D, et al. 2003. Trophoblast L-selectin-mediated adhesion Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

LaTeX2e(2002/01/18)

at the maternal-fetal interface. Science 299:405–8 135. Kao LC, Germeyer A, Tulac S, Lobo S, Yang JP, et al. 2003. Expression profiling of endometrium from women with endometriosis reveals candidate genes for disease-based implantation failure and infertility. Endocrinology 144:2870– 81

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:129-156. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104649

Annu. Rev. Immunol. 2004. 22:157–80 doi: 10.1146/annurev.immunol.22.012703.104649 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on January 12, 2004

INTEGRINS AND T CELL–MEDIATED IMMUNITY

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu Department of Laboratory Medicine and Pathology, Center for Immunology, Cancer Center, University of Minnesota Medical School, Minneapolis, Minnesota 55455; email: [email protected], [email protected], [email protected], [email protected]

Key Words extracellular matrix, chemokine, signal transduction, dendritic cell, immunological synapse ■ Abstract Integrin receptors mediate adhesive events that are critical for a specific and effective immune response to foreign pathogens. Integrin-dependent interactions of lymphocytes and antigen-presenting cells (APCs) to endothelium regulate the efficiency and specificity of trafficking into secondary lymphoid organs and peripheral tissue. Within these sites, integrins facilitate cell movement via interactions with the extracellular matrix, and promote and stabilize antigen-specific interactions between T lymphocytes and APCs that are critical for initiating T cell–activation events. In this review, we discuss the role of integrins in T cell–mediated immunity, with a focus on how these receptors participate in lymphocyte recirculation and T cell activation, how antigen stimulation regulates integrin activity, and how integrins define functionally unique subsets of T cells and APCs.

INTRODUCTION Efficient recognition of foreign pathogens by T cells requires adhesive interactions between T cells and other cell types, such as endothelial cells and antigenpresenting cells (APCs), and with components of the extracellular matrix (ECM). These contacts are dynamic, in most cases transient, and subject to an unusually high degree of regulation. Integrins are αβ heterodimeric cell surface receptors that are major mediators of these adhesive interactions, which are essential for T cell recirculation, migration into inflammatory sites, and recognition of foreign antigens (1, 2). In mammals, the integrin family consists of 18 α subunits and 8 β subunits that combine to form 24 different integrin αβ pairs (Figure 1). The ligand specificity for each integrin differs, and therefore the spectrum of integrins expressed by a given cell type determines, to a large extent, how efficiently that cell can interact with a given microenvironment. The constellation of integrins expressed by circulating cells, such as lymphocytes, is particularly critical because they encounter a wide array of microenvironments as they 0732-0582/04/0423-0157$14.00

157

11 Feb 2004

17:3

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

158

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL.

Figure 1 Integrin receptors. A total of 18 α subunits and 8 β subunits combine to form 24 different integrins. Integrins in the circle are particularly relevant to T cell–mediated immunity.

recirculate through the body. In this review, we highlight how integrins participate in T cell–mediated immune responses by regulating the circulatory behavior of T cells and APCs, and the efficiency of antigen-specific interactions of T cells with APCs.

INTEGRINS AND LYMPHOCYTE RECIRCULATION The constant movement of na¨ıve T cells through lymph nodes (LNs) ensures that the immune system exposes as many T cells as possible to rare, foreign peptideMHC (pMHC) complexes in an environment conducive to optimal stimulation of na¨ıve T cells (1). The interaction of na¨ıve T cells with high endothelial venules (HEVs) is critical for efficient T cell recirculation through LNs. On na¨ıve T cells, L-selectin—the CCR7 chemokine receptor—and the LFA-1 integrin mediate distinct and sequential steps in T cell–HEV interactions: initial tethering of T cells to endothelium, activation, and the subsequent formation of stable adhesive contacts (Figure 2).

Tethering and Reversible Rolling Under conditions of shear flow, the initial interaction of T cells with endothelium is characterized by a rolling behavior of the T cell on the endothelial surface. Lselectin is uniquely suited to mediating transient adhesive interactions that result in rolling, as this receptor forms rapid but weak interactions with specific carbohydrate epitopes on the endothelium (3). Like L-selectin, α4 integrins mediate lymphocyte rolling (4–6), particularly in the absence of a selectin contribution (6). Lymphocyte rolling mediated by α4β1 does not appear to require high-affinity interactions with VCAM-1 because mutations that reduce α4β1 affinity support rolling but not firm adhesion (4). Recent evidence also suggests that α4β1 integrin-mediated rolling on VCAM-1 may be regulated by a very rapid (<0.1 second) induction of α4 integrin clustering by chemokines (5).

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTEGRINS AND T CELL IMMUNE RESPONSES

159

Figure 2 Integrin receptors mediate adhesive interactions essential to lymphocyte recirculation and T cell–APC interactions. T cell interactions with endothelium and T cell interactions with APCs in secondary lymphoid tissues both involve sequential, discrete steps that begin with brief, unstable interactions that are converted by an activation signal to stable, firm, integrin-dependent contacts. The figure depicts receptors expressed on na¨ıve T cells that have been implicated in the sequence of events that are critical for successful T cell extravasation into secondary lymphoid tissue and T cell interactions with antigen-laden APCs (see text for further details).

LFA-1 can also support rolling on both purified ICAM-1 and endothelial cells under shear flow (7–10). Although LFA-1/ICAM-1 interactions are not involved in initial tether formation, they appear to optimize L-selectin-mediated rolling by decreasing the velocity with which lymphocytes move over the endothelium (7, 10). The ability of LFA-1 to mediate rolling is dependent on weak interactions with ICAM-1 because mutations that either increase the lateral mobility of LFA-1 integrin in the membrane or increase the affinity of LFA-1 for ICAM-1 eliminate lymphocyte rolling in favor of firm adhesion to ICAM-1 (8, 9).

Activation and Stable Arrest The transient tethering and rolling of lymphocytes along the endothelium enhances the likelihood of sufficient lymphocyte stimulation to induce firm integrinmediated adhesion. Stimulation of rolling T cells by chemokines that are localized on the endothelial surface via interactions with apical proteoglycans results in rapid, but transient, increases in integrin adhesiveness (5, 11). As with other signals that modulate integrin activity, chemokines induce changes in integrin functional activity via two distinct, but complementary, mechanisms: conformational

11 Feb 2004

17:3

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

160

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL.

changes that alter integrin ligand binding affinity and changes in integrin lateral mobility that result in integrin clustering and increased avidity of cell adhesion. Although a vast majority of integrins on the cell surface exist in a low affinity state on unstimulated T cells, a small subset of integrins that are in a high-affinity state are particularly critical for spontaneous T cell arrest on endothelium (4). In addition, chemokines induce a rapid and transient increase in β1 and β2 integrin affinity that correlates with adhesion to endothelial ligands (12, 13). Structural studies of the inserted (I) domain of the αL subunit of LFA-1 have also demonstrated that the transition from LFA-1-mediated rolling to LFA-1-mediated stable adhesion can be mediated by conformational changes that result in increased LFA-1 affinity (9). Thus, increases in LFA-1 affinity are sufficient to promote firm adhesion under shear flow. Chemokines also regulate integrin function by altering the lateral mobility of integrins within the cell membrane, thereby facilitating integrin clustering (5, 13–15). Stimulation of primary na¨ıve T cells with the chemokines CCL19, CCL21, or CXCL12 induces changes in LFA-1 avidity that are sensitive to inhibitors of the lipid kinase phosphatidylinositol 3-kinase (PI 3-K) (13). In addition, CXCL12 induces α4β1 integrin clustering and results in either the transient tethering or the firm adhesion of cells to integrin ligands (5). The concentration of integrin ligand determines the relative contribution of affinity and avidity to changes in integrin-mediated adhesion under shear flow. At low ligand concentrations, avidity changes are critical, as blocking increased integrin avidity with PI 3-K inhibitors prevents lymphocyte arrest under shear flow (13). However, at high-ligand concentrations, affinity changes are sufficient to mediate chemokine-dependent firm adhesion (13). Although chemokines play a central role in inducing stable arrest of T cells on endothelium, other receptor signals may also participate. Integrin activation may be regulated, in part, through L-selectin-ligand interactions because L-selectin ligation increases the avidity of β2 integrin-mediated adhesion (16). The α4β1 integrin also modulates LFA-1 integrin–mediated migration (17) through a mechanism that involves binding of paxillin to the integrin α4 cytoplasmic tail (18). Thus, tethering mediated by α4 integrins may provide signals that enhance LFA-1dependent adhesion and migration. In addition, a recent study demonstrated that presentation of insulin by pancreatic endothelial cells is critical for the adhesion of insulin-specific CD8 T cells to endothelial cells and for homing of these cells to pancreatic islets (19). Thus, trafficking of effector cells to peripheral tissue sites may be enhanced by such cross-presentation of antigen by endothelial cells.

Diapedesis Following firm attachment to the endothelial surface, lymphocytes must crawl through interendothelial junctions and cross the underlying basement membrane in order to successfully enter tissue, a process referred to as diapedesis, transmigration, or extravasation (20). Studies utilizing inhibitory antibodies and β2

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTEGRINS AND T CELL IMMUNE RESPONSES

161

integrin–deficient T cell clones have demonstrated that LFA-1 mediates T cell transendothelial migration in response to several chemokines (21, 22). The transmigration process involves LFA-1 interactions with ICAM-1, as well as with the immunoglobulin superfamily member JAM-1, a new LFA-1 ligand that is expressed at the tight junctions of resting endothelium (23). LFA-1-dependent interactions with JAM-1 support both firm adhesion and transmigration, and this appears to be dependent in part on the activation state of the endothelial cell, as JAM-1 redistributes to the endothelial cell surface upon cytokine stimulation. Other nonintegrin receptors that are localized to interendothelial junctions, including CD31 (20) and CD99 (24), are likely the major receptors mediating diapedesis, and it now appears that, similar to tethering and stable arrest, diapedesis involves the sequential contribution of multiple receptor-ligand interactions. The α4β1 integrin also participates in T cell diapedesis (22), although its relative contribution to this process is dependent on cytokine activation of the endothelial cells, which enhances expression of the α4β1 ligand VCAM-1 and chemokine stimulation of the T cells (22). It should be noted that much of the work implicating β2 and β1 integrins in diapedesis has utilized experimental systems where the effects of blocking integrin function likely inhibited adhesion, which is required for successful diapedesis. Thus, it is difficult to precisely define the contribution of integrins specifically to the transmigration process in some of these studies.

Lymphocyte Movement in Lymph Nodes Following diapedesis, na¨ıve T cells enter the T cell–rich paracortical regions of LNs, where a network of reticular fibers extends from the capsule through the T cell–rich cortex and into the medulla (25, 26). Collagen types I, III, and IV form the core of these fibers and provide strength to the LN (25, 27, 28; JT Pribila, AA Itano, KL Mueller, Y Shimizu, submitted). These core fibers are surrounded by a meshwork of elastin, fibronectin, laminin-1, laminin-α3, tenascin, vitronectin, and several proteoglycans including heparan sulfate and keratan sulfate (25, 27, 30, 31). However, a majority of these fibrous cords are ensheathed by a layer of fibroblastic reticular cells (FRCs) that are interconnected by tight junctions (25). The FRCs likely limit exposure of T cells to ECM in LNs, as only approximately 10% of the ECM network is exposed to the cellular compartment of the LN (32). This structure likely enhances cell-cell contacts within the LN microenvironment. Recent studies using intravital two photon scanning laser microscopy have provided critical insight into lymphocyte behavior within the LN (33–36). In both explanted LNs and intact inguinal LNs of anesthetized mice, T cells exhibit a very high rate of random motility, averaging 11 µm per min (35, 36). In the absence of antigen, relatively short interactions occur between CD4 T cells and dendritic cells (DCs). However, a high percentage of T cells rapidly form stable interactions with antigen-laden DCs that last at least 15 h (34). At 36–48 h following antigen exposure, when T cells display evidence of activation, T cells dissociate from

11 Feb 2004

17:3

162

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL.

DCs and again move rapidly within the LN. Thus, like T cell interactions with endothelium, T cell interactions with APCs in LNs are characterized by short, relatively unstable initial interactions that can be converted into more stable, longterm interactions if an appropriate activation signal, in this case antigen, is delivered to the T cell (Figure 2).

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE LFA-1 INTEGRIN AND T CELL–APC INTERACTIONS A stable interaction between a T cell and APC is required to elicit the signaling cascades necessary for T cell activation and effector function (33, 34, 36). The TCR and MHC molecules alone cannot mediate this stable interaction owing to the low affinity of TCR interaction with pMHC, the small size of the TCR and MHC molecules, and the relatively low abundance of MHC expressing cognate antigen. The LFA-1 integrin, expressed on the T cell surface, and its ligand, ICAM-1, expressed on the surface of the APC, are key mediators of this interaction (37–39), and TCR stimulation promotes stable T cell–APC contact by rapidly enhancing LFA-1 functional activity (see below).

T Cell Scanning In the absence of antigen, T cells move rapidly within the LN, forming transient contacts with APCs that allow T cells to “scan” APCs for the presence of cognate antigen. Chemokine gradients present in the LN microenvironment may promote T cell scanning of APCs by activating LFA-1 because CXCL12 enhances antigenindependent T cell conjugation with DCs (40). Chemokine-stimulated increases in LFA-1 affinity are both extremely rapid and transient, characteristics that would support T cell scanning of APCs. However, imaging studies show that T cells within the LN move in a random rather than directed fashion (33, 36). The scanning process itself may promote integrin activation and facilitate transient interactions between T cells and APCs. The encounter of na¨ıve CD4 T cells with class II MHC-presenting self-peptide in the periphery promotes polarization of the TCR and signaling molecules, such as ZAP-70, to sites of T cell–APC contact (41). Although LFA-1 distribution was not assessed in these studies, this tonic level of activation may polarize LFA-1 or shift the conformational equilibrium toward high-affinity LFA-1 that facilitates T cell scanning. Antigen-independent interactions between na¨ıve T cells and DCs can lead to some T cell responses, including calcium signaling (42). Because calcium mobilizers enhance LFA-1 activity in T cells (43), low levels of antigen-independent calcium signaling may promote T cell scanning by modulating LFA-1 activity. Although these studies support a role for LFA-1 in T cell scanning of APCs, no direct evidence exists to suggest that T cell scanning requires integrins. Other nonintegrin adhesion molecules mediating low-affinity interactions, such as DCspecific ICAM-3 grabbing nonintegrin (DC-SIGN) and ICAM-3, are equally

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

INTEGRINS AND T CELL IMMUNE RESPONSES

163

attractive candidates because they can form transient adhesive interactions and do not require activation to mediate adhesion (44).

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

LFA-1 and the Immunological Synapse An external ring of LFA-1 (and the integrin-associated cytoskeletal protein talin) surrounding a central TCR-rich region is a structural hallmark of a mature immunological synapse (IS), which forms during antigen-dependent interactions with APCs (45–47). Furthermore, studies using planar lipid bilayers have established that the IS can form when a T cell interacts with only pMHC complexes and ICAM-1 (46). During the initial stages of T cell activation, LFA-1 is localized in the center of the T cell–APC contact site while the TCR localizes primarily to the periphery (46–48). Rapid clustering of LFA-1 coupled with its central localization within the contact site suggests that LFA-1-mediated adhesion is needed to stabilize T cell–APC contact, resulting in larger numbers of engaged TCRs and activation of signaling cascades required for T cell activation. This is supported by colocalization of the TCR and activated tyrosine kinases Lck and ZAP-70 in the periphery of the IS (47, 48), which suggests that signaling cascades necessary for T cell activation are initiated while LFA-1 and ICAM-1 are clustered in the center of the T cell–APC contact site. T cell commitment to proliferation and effector function requires sustained T cell–APC interaction (36, 47, 49) and extended interactions of a single T cell with an APC have been observed in ex vivo LN cultures (34, 36). In the mature IS, the TCR moves to the center of the T cell–APC contact site while LFA-1 translocates to the periphery (47, 48). Whereas the TCR is largely absent from the IS after 1 h of conjugation, clustered LFA-1 is present up to at least 4 h (47, 48, 50). The kinetics of LFA-1 localization to the IS correlate with the minimum amount of conjugation time necessary for T cell proliferation to occur (47, 51), suggesting that prolonged localization of LFA-1 may facilitate molecular engagements at the T cell–APC contact site that result in signaling events necessary for complete T cell activation. Recent insights into the structural basis of integrin-mediated adhesion also suggest an active role for LFA-1 in T cell–APC interactions. The negatively charged glycocalyx favors repulsion of the T cell from the APC. The major components of the glycocalyx, such as CD43 and CD45, extend at least 45 nm from the cell surface and span approximately 30% of the T cell surface area (52), whereas the TCR-pMHC complex extends only 15 nm from the cell surface (53, 54). Integrins are capable of adopting both open and closed conformations that are regulated by intracellular as well as extracellular stimuli (55). Integrins exhibit low ligandbinding affinity in the closed conformation, due in large part to a bend in the middle of the α and β subunits that brings the ligand binding site to within 5 nm of the cell surface (56, 57). This closed conformation likely would not efficiently promote T cell–APC interactions because the low-affinity integrin is unlikely to span the glycocalyx. In the open conformation, the two subunits straighten and the integrin

11 Feb 2004

17:3

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

164

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL.

displays high affinity for its ligand (55). The high-affinity integrin now extends approximately 20 nm from the cell surface and thus would likely be more effective in supporting T cell–APC interactions. Analysis of the I domain of LFA-1 bound to ICAM-1 also revealed that an intermediate conformational affinity state exists, where the αL and β2 subunits are partially bent (58). Thus, integrins are likely to exist in an allosteric equilibrium between these different conformational states. Antigen-dependent signaling through the TCR may shift the equilibrium toward the most open conformation, thereby promoting T cell–APC interaction. It is also important to note that factors such as the differentiation state of the T cell, the type of APC, and the microenvironment where the T cell is interacting with the APC affect the initiation, maintenance, and functional consequences of IS formation. Memory T cells express generally higher levels of LFA-1 and other integrins than do na¨ıve T cells, and thus memory T cells have enhanced basal and TCR-dependent integrin function (1). These differences in integrin function between na¨ıve and memory T cells likely affect both the T cell scanning process and IS formation. Furthermore, effector T cells have elevated integrin activity when compared to resting memory T cells, and they likely interact with collagen and other ECM proteins in peripheral tissue. We currently know little about the role of integrins in initiating interactions between effector T cells and target cells in peripheral microenvironments.

Integrins as Costimulatory Receptors In addition to mediating adhesion to cell surface and ECM ligands, integrins generate a diverse array of intracellular signals. In T cells, LFA-1 and several β1 integrins can provide costimulatory signals for TCR-induced T cell activation (59, 60). Differences in the regulation of IL-2 production following TCR stimulation by integrins and CD28 demonstrate the unique costimulatory properties of these molecules. In CD4 T cells, LFA-1-mediated costimulation modulates IL-2 transcription, whereas CD28 costimulation alters IL-2 levels by stabilizing IL-2 mRNA (61). In addition, although both LFA-1 and CD28 induce the activation of PI 3-K in CD8 T cells, PI 3-K activity is required only for LFA-1-meditated costimulation (62). A promising candidate protein that may couple LFA-1 to changes in IL-2 transcription is JAB1, which interacts with LFA-1 (63) and is a known coactivator of the c-Jun transcription factor. Integrin engagement on T cells increases the nuclear fraction of JAB1, resulting in increased AP-1 activity (63). This suggests that LFA-1 modulates the transcriptional activity of AP-1 by regulating the nuclear localization of JAB1. Upon T cell activation, both β1 and β2 integrins move into cholesterol-rich, detergent-insoluble areas in the cell membrane called lipid rafts (64). Although the importance of lipid rafts to integrin signaling is unclear, they likely facilitate the interaction of integrins with a subset of cytoplasmic and transmembrane proteins that are either constitutively or inducibly associated with rafts (65). For example, β1 integrin ligation on T cells induces activation of the src kinase fyn (66), which

15 Mar 2004

15:8

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTEGRINS AND T CELL IMMUNE RESPONSES

165

is constitutively localized to lipid rafts. Fyn associates with the adapter proteins SH2 domain-containing leukocyte phosphoprotein of 76 kDa (SLP-76), Crk-L, and adhesion- and degranulation-promoting adapter protein (ADAP) (66, 67), and both Crk-L and ADAP regulate integrin-dependent adhesion and migration (68–71). Because the Fyn/ADAP/SLP-76 complex is critical to the regulation of TCR-induced IL-2 production (72), these molecules may participate in integrin-mediated costimulation. CD3/TCR stimulation coupled with LFA-1 cross-linking also induces tyrosine phosphorylation of focal adhesion kinase (73), which constitutively associates with human enhancer of filamentation (74), a Crk-associated substrate that enhances β1 integrin-mediated costimulation when overexpressed in T cells (75).

REGULATION OF INTEGRIN FUNCTION BY THE TCR Stimulation of the CD3/TCR results within minutes in a transient increase in the functional activity of both LFA-1 and β1 integrins that is not dependent on increased expression of integrins on the T cell surface. Integrin activation is one of the earliest functional responses of T cells that encounter with pMHC and dramatically alters the migratory behavior of T cells (76). In addition, TCR-dependent activation of LFA-1 increases T cell sensitivity to antigen by stabilizing T cell interactions with APCs. TCR-dependent changes in integrin functional activity involve cytoskeletal-dependent integrin clustering that enhances the avidity of cell adhesion (14). Although TCR stimulation can also enhance integrin affinity (77), much less is currently known about how the TCR controls integrin affinity and how integrin affinity changes contribute to enhanced integrin-mediated adhesion following TCR stimulation.

Signaling Pathways Regulating TCR-Dependent Integrin Activation The proximal signaling events triggered by the TCR have been extensively characterized (78), and many of these signaling intermediates have been implicated in TCR-dependent regulation of integrin activity (Figure 3). Studies utilizing peripheral human T cells and genetic variants of the Jurkat T cell line, together with pharmacological inhibitors and dominant-negative signaling constructs, have identified important functions for Lck, ZAP-70, Itk, and PI 3-K in CD3/TCR-mediated integrin activation (39, 77, 79–82). Although ZAP-70 kinase activity is essential for CD3/TCR-mediated activation of β1 integrin function, a precise role for the ZAP-70 substrates LAT and SLP-76 has not been clearly established (81). The role of PI 3-K in regulating integrin activity appears to involve Itk, as binding of the pleckstrin homology domain of Itk to the PI 3-K lipid product PI(3,4,5)-P3 is important in membrane recruitment and subsequent activation of Itk. In addition, overexpression of just the pleckstrin homology domain of Itk inhibits TCR signaling to integrins (80).

11 Feb 2004

17:3

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

166

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL.

Because ZAP-70 phosphorylates LAT, and Itk associates with LAT following TCR stimulation and regulates PLC-γ 1 activity, a function for PLC-γ 1 in TCR-dependent integrin activation seems likely. A PLC inhibitor does, in fact, block integrin-mediated adhesion of mouse T cells and T cell–conjugate formation with APCs (39, 83). Furthermore, defects in PLC-γ 1 phosphorylation in anergic T cells are associated with defects in integrin activation in these cells (83). Signaling responses downstream of PLC-γ 1 are also important in regulating integrin activity in T cells, as intracellular calcium mobilizers and phorbol esters can enhance β1 and β2 integrin-mediated adhesion independently of the TCR (43, 84). Constitutively active forms of conventional PKC isotypes α, βI, and βII, and the novel isotype δ, but not the atypical isotype PKCζ , also enhance the basal adhesion of a LFA-1-expressing pro B cell line to ICAM-1 (85). In addition, TCR-mediated activation of integrins is blocked by PKC inhibitors (37, 39, 86). Studies with knockout mice have identified ADAP as a novel effector of integrin activation in T cells. Previously known as SLP-76-associated protein of 130 kDa (SLAP-130) and Fyn-T-binding protein (Fyb), ADAP is an intracellular adapter protein that becomes tyrosine phosphorylated following TCR stimulation and associates with SLP-76 (87), Fyn (88), and a number of proteins implicated in cytoskeletal rearrangement including the Ena/VASP family member protein Ev1, Wiskott-Aldrich syndrome protein, and the adapter protein Nck (89). Whereas ADAP is dispensable for T cell development, mature T cells from ADAP-deficient mice exhibit defective proliferative responses and are unable to efficiently activate β1 and β2 integrins following CD3/TCR stimulation (70, 71). This defect in integrin activation is associated with defective CD3-induced LFA-1 clustering in ADAP-deficient T cells. Although the mechanism by which ADAP regulates integrin activity is currently unknown, ADAP also associates with the adapter protein src kinase-associated phosphoprotein of 55 kDa (SKAP55), and overexpression of SKAP55 enhances CD3/TCR-mediated increases in adhesion to ICAM-1 and fibronectin, conjugate formation, and LFA-1 clustering (90). Regulation of integrin activity by SKAP55 appears to be dependent on its interaction with ADAP, as constructs lacking the SKAP55 SH3 domain, which binds to ADAP, do not enhance conjugate formation. Integrin activation and integrin-dependent conjugate formation induced by TCR stimulation is also impaired in T cells isolated from mice lacking expression of Vav1 (82, 91), a guanine nucleotide exchange factor for the GTPases Rac-1 and Rac-2 (92). GTPases regulated by Vav1 have also been implicated in integrin activation. Constitutively active Rac constructs induce cell spreading and enhance basal adhesion of nonlymphoid and lymphoid cells to fibronectin and ICAM-1 (85, 93, 94). Similarly, thymocytes from transgenic mice expressing the active form of Rac-1 show increased basal adhesion to fibronectin (95). These studies suggest that Rac-1 activation is sufficient to enhance integrin activity, but a role for Rac-1 in TCR-dependent activation of integrins is less clear. A role for other Rho family GTPases in controlling integrin activity in T cells is also not well

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTEGRINS AND T CELL IMMUNE RESPONSES

167

characterized. Although constitutively active forms of Rho and Cdc42 do not induce cell spreading or enhance adhesion to fibronectin or ICAM-1 (85, 93, 94), thymocytes from transgenic mice expressing active RhoA do show enhanced basal adhesion to fibronectin (96). In contrast, Rho inhibitors are reported to actually increase basal adhesion to ICAM-1 (97). More compelling evidence that GTPases regulate TCR signaling to integrins has come from the analysis of Rap1, a GTPase that was first identified by its ability to revert the morphological phenotype of Ras-transformed fibroblasts (98). Constitutively active Rap1 potently enhances basal LFA-1 integrin-mediated adhesion to ICAM-1 (85, 99). Rap1 activity is also enhanced in Cbl-b-deficient mice, owing to enhanced interactions between Crk-L and the Rap1 guanine nucleotide exchange factor C3G. Consequently, both basal LFA-1 function and CD3-induced increases in LFA-1 function are enhanced in Cbl-b-deficient T cells (100). A dominant-negative form of Rap1 or SPA-1, a Rap1 GTPase-activating protein, inhibits increased integrin activity induced by stimulation of the TCR (101), the CD31 antigen (102), and several cytokine and chemokine receptors (94, 103). Although these studies suggest a central function for Rap1 in the regulation of integrin function by multiple activating signals, it should be noted that TCR stimulation does not enhance integrin function in anergic T cells (83), even though anergy has been associated with enhanced Rap1 activity (104). There is no clear consensus on whether Rap1 mediates changes in integrin functional activity via effects on integrin affinity or integrin clustering. LFA-1 is constitutively clustered on thymocytes from transgenic mice expressing active Rap1, but there is no enhanced binding of soluble ICAM-1 (99). In contrast, active Rap1 expression in a pro B cell line expressing LFA-1 results in enhanced soluble ICAM-1 binding and increased expression of an antibody epitope indicative of conformational changes in LFA-1 (85). Rap1 activity is enhanced in Cbl-bdeficient T cells, and there is enhanced soluble ICAM-1 binding in these cells in the absence of stimulation (100). However, LFA-1 clustering is enhanced in these cells only following CD3 stimulation (100). Rap1 antagonists also block increases in β1 and β2 integrin function induced by PMA stimulation (105), which enhances integrin function via effects on integrin clustering, and by divalent cations and unique integrin-specific antibodies that induce conformational changes that enhance integrin affinity (106). However, inhibitory effects of Rap1 antagonists on PMA-induced activation of β2 integrin function have not been consistently observed (101). Ultimately, the identification of Rap1 effectors critical to Rap1-dependent regulation of integrin function will be essential. Katagiri et al. recently identified a novel protein, designated RapL, that associates with Rap1 following TCR or chemokine stimulation (107). RapL overexpression enhances LFA-1-dependent adhesion to immobilized as well as soluble ICAM-1 and induces cell polarization and LFA1 clustering. Expression of active Rap1 enhances the association of RapL with LFA-1, suggesting a potential mechanism by which RapL facilitates Rap1dependent effects on LFA-1 function (107).

11 Feb 2004

17:3

168

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL.

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integrin Structure and Activation of Integrin Function Recent studies of the cytoplasmic and transmembrane domains of the αIIb and β3 integrin subunits (2, 108) suggest an intimate connection between conformational changes that lead to increased integrin affinity and mechanisms of integrin clustering that lead to enhanced avidity of cell adhesion. Interaction between the αIIb and β3 subunit cytoplasmic domains (109, 110) restrains the αIIbβ3 integrin in an inactive state. Mutation of key residues that disrupt a putative salt bridge linking the αIIb and β3 cytoplasmic domains results in constitutively active αIIbβ3 integrin (111). This suggests that intracellular signals that enhance integrin activity ultimately result in separation of the integrin α and β subunit cytoplasmic tails. Activation-dependent separation of the integrin tails may involve competitive displacement of the α-β subunit association by proteins such as talin, which binds to integrin β subunits and colocalizes with LFA-1 in the IS. The phosphotyrosine binding-like (PTB) domain of talin binds to the β3 integrin cytoplasmic domain via a tyrosine residue (Y747) in a NPXY motif that is conserved among multiple integrin β subunit tails (112, 113). The interaction of talin with the β3 tail blocks the interaction of the αIIb tail with β3 (110), and binding of talin to the β3 tail results in integrin activation (112, 113). This suggests that proteins that interact with integrin cytoplasmic domains may mediate integrin activation via changes in the spacing between the integrin α and β tails. Structural studies of LFA-1 suggest a similar mechanism of regulation by the αL and β2 cytoplasmic domains (114). In addition, talin binding to integrin β tails requires exposure of the talin PTB domain, which can occur upon cleavage with calpain, an enzyme that has been implicated in calcium-dependent regulation of LFA-1 activity in T cells (43). Structural studies also show that isolated αIIb transmembrane helices can form homodimers and isolated β3 transmembrane helices can form homotrimers. These homomeric interactions would most likely occur when the integrin is active and the α and β tails are physically separated (55, 56, 108). Thus, activation-dependent conversion to a high-affinity integrin may be accompanied by integrin clustering mediated by transmembrane sequences that result in homomeric associations between integrin subunits (108). Two mutations in the β3 transmembrane helix have been identified that lead to a constitutively active integrin that is clustered on the cell surface (108).

INTEGRINS AS MARKERS OF T CELL AND DENDRITIC CELL SUBSETS The specificity by which integrins direct cell trafficking and retention within peripheral tissue makes them important markers of both T cells and DCs. Whereas na¨ıve T cells express uniform levels of β1 and β2 integrins, memory T cells can be divided into subsets based on differential integrin expression (1). A majority of memory T cells (>80%) express high levels of β1 integrins, but not α4β7

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

INTEGRINS AND T CELL IMMUNE RESPONSES

169

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

or αEβ7 integrins, and recirculate through nonmucosal tissue. Inflammatory cytokines induce VCAM-1 expression on the endothelium of nonmucosal tissue and preferentially recruit α4β1-bearing memory T cells to these sites. In contrast, two other subsets of memory T cells are defined by low levels of β1 integrin expression and by the differential expression of β7 integrins (αEβ7HIα4β7LO and αEβ7LOα4β7HI) (1, 115).

α4β7 and αEβ7 Integrins and Tissue-Specific Localization of T Cells Integrin α4β7 is expressed on a distinct subset of memory CD4 and CD8 T cells (115–117), and it directs the trafficking of effector and memory T cells to intestinal sites of inflammation via interactions with mucosal addressin cell adhesion molecule (MAdCAM-1) (1, 6, 115, 117–119). Although the importance of α4β7 to intestinal trafficking is clear, the factors that determine whether a na¨ıve T cell will become α4β7HI or α4β7− are not fully known. The site of initial antigen exposure can define the phenotype of the memory population, as CD4 T cells primed in intestinal lymphoid organs increase α4β7 expression, whereas those primed in cutaneous LNs lose α4β7 within two days of systemic immunization (120). Furthermore, intestinal DCs preferentially induce expression of α4β7 compared to peripheral DCs, regardless of the source of T cells (121). These studies suggest that the site of antigen exposure can differentially regulate integrin expression and, thereby, regulate the type of effector and memory T cell to which na¨ıve T cells give rise. Intraepithelial lymphocytes (IELs) are a distinct subset of memory T cells defined by expression of αEβ7 integrin (122, 123), which specifically binds to the epithelial-specific cadherin, E-cadherin (124, 125), and is critical to IEL retention between intestinal epithelial cells (1, 122) and to IEL effector function (126). Mice lacking αE integrin have dramatically reduced numbers of intestinal and vaginal IELs as well as decreased numbers of lamina propria T cells (127). However, αEβ7 is not important for generating or maintaining T cells in Peyer’s patches, as there are normal numbers of T cells in αE-deficient mice (127). In addition to IELs, the αEβ7 integrin is expressed on CD4+CD25+ regulatory T cells (TREG), as well as an uncharacterized subset of CD4+CD25− TREG (128). Integrin αEβ7 expression on TREG correlates with CTLA-4 expression, suppression of T cell proliferation in vitro, and protection against colitis in a severe combined immunodeficient model (128). However, other studies show that αEβ7-specific mAbs actually ameliorate existing immunization-induced colitis in IL-2 deficient mice, primarily by preventing the localization of CD4 T cells to the lamina propria (129). Because there are a reduced number of lamina propria T cells in αE-deficient mice, αEβ7 may regulate the localization of a specific subset of TREG to the lamina propria. Mice lacking αE are also predisposed to the development of inflammatory skin lesions (130), consistent with a role for αEβ7 integrin in localization or functional regulation of TREG.

11 Feb 2004

17:3

170

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL.

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Collagen-Binding Integrins and Effector T Cell Function The α1β1 and α2β1 integrins mediate cell adhesion to collagen and were originally identified as “very late activation” antigens expressed on activated T cells following several weeks of in vitro culture (131). Although resting na¨ıve and memory T cells lack expression of α1β1 and α2β1 and do not adhere to collagen in vitro, α1β1 is expressed on T cells isolated from the intestinal lamina propria (132). In addition, infiltrating T cells in arthritis and other inflammatory conditions express α1β1 and exhibit activation-dependent adhesion to collagen (133). Thus, α1β1 (and likely α2β1) are the major collagen receptors on T cells, and the highly regulated expression of these integrins suggests a unique function for collagen in T cell function. Studies in mice with blocking antibodies and α1 integrin-deficient mice demonstrate a role for α1β1 and α2β1 in both delayed-type and contact hypersensitivity models, as well as a model of collagen antibody–induced arthritis (134). The α1β1 integrin is expressed on infiltrating leukocytes, including T cells and subsets of monocytes and neutrophils, and antiα1 and antiα2 antibody treatment is associated with decreased leukocyte infiltration. Similar findings regarding α1β1 function have also been reported in two models of experimental colitis (135, 136) and are consistent with the reduced numbers of IELs in α1-deficient mice (137). Together, these studies directly implicate α1β1 and α2β1 in the function of effector T cells and suggest a critical function for T cell interactions with collagen in peripheral tissue (138).

Differential Expression of Multiple Integrins on Murine DC Subsets Integrins also define murine DCs and DC subsets (Figure 4). CD11c, the αX subunit of the αXβ2 integrin, was the first murine DC marker to be identified (139) and has been instrumental in the identification and characterization of DC subsets in the LN, spleen, and thymus (140–142). Although originally identified as a myeloidspecific antigen, the αM subunit of the αMβ2 (MAC-1; CR3) integrin, CD11b, distinguishes CD8α-negative myeloid DCs from CD8α-positive lymphoid DCs (140). In peripheral LNs from normal mice, lymphoid DCs, but not myeloid DCs, express the α1, α6, and αE integrin subunits (JT Pribila, AA Itano, KL Mueller, Y Shimizu, submitted; 143). Other integrin subunits, including αL, α4, and α5 are expressed at comparable levels on skin-derived lymphoid and myeloid DCs in the mouse (JT Pribila, AA Itano, KL Mueller, Y Shimizu, submitted). The α4 integrin subunit is also a maturation marker on monocyte-derived human DCs (144). The α1β1 and αEβ7 integrins are also expressed on a subset of CD40HICD8α INT skin-derived DCs in peripheral LNs (JT Pribila, AA Itano, KL Mueller, Y Shimizu, submitted). When compared to other skin-derived DCs, α1β1+αEβ7+ skinderived DCs exhibit an enhanced ability to interact with T cells in the absence of antigen and a reduced ability to acquire soluble antigen in vivo. Skin-derived DCs also exhibit α1β1 integrin-dependent adhesion to collagen and interact more

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTEGRINS AND T CELL IMMUNE RESPONSES

171

Figure 4 Integrin expression on murine DCs. Integrins are important phenotypic markers of functionally distinct subsets of DCs in peripheral LN of normal mice (JT Pribila, AA Itano, KL Mueller, Y Shimizu, submitted; 140, 142). These differences in integrin expression are also associated with differential adhesion of DC subsets to ECM components. The α1β1 and αEβ7 integrins are not expressed on Langerhans cells in the skin, which suggests that expression of these integrins is induced on a subset of skin-derived DCs after they have migrated to peripheral LNs (JT Pribila, AA Itano, KL Mueller, Y Shimizu, submitted).

efficiently with collagen and fibronectin than myeloid and lymphoid DCs, even though lymphoid DCs express comparable levels of α1β1, α4β1, and α5β1 integrins. These findings may be particularly relevant to DC function in LNs, as α1β1+ DCs colocalize in vivo with collagen, which is found in the reticular network that limits the diffusion of soluble antigens in LNs (26). Because the uptake of soluble antigen by skin-derived DCs resident in draining LNs is critical to the initial activation of naive T cells (28), DCs capable of adhering to these ECM fibers are uniquely situated to serve as the critical APCs that initiate T cell activation. The role of integrins in DC recruitment into LNs is incompletely understood. DCs and DC precursors interact with endothelium in a manner similar to lymphocytes, except that DC-SIGN expressed on DCs mediates tethering and rolling via interaction with ICAM-2 on endothelium (145). Blood-borne DCs, which enter the LN via HEVs, utilize both β1 and β2 integrins to adhere to human umbilical vein endothelial cells (146). The migration of Langerhans cells to LNs involves both LFA-1 (147, 148) and α6 laminin-binding integrins (149). Although the α1β1 and αEβ7 integrins are expressed on a subset of skin-derived DCs resident in LNs, these integrins are not expressed on Langerhans cells in the skin (JT Pribila, AA Itano, KL Mueller, Y Shimizu, submitted), and α1β1 is not required for the migration of skin-derived DCs to the LN (134).

11 Feb 2004

17:3

172

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL.

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONCLUSIONS In this review, we have focused on the role of integrins in T cell–mediated immunity. However, the impaired immune response in patients with leukocyte adhesion deficiency-1, where genetic mutations impair β2 integrin subunit expression (150), vividly illustrates the vital role that integrins play in virtually every aspect of immune system function. The identification of several cases of a novel form of LAD-1 characterized by loss not of integrin expression but of the ability of activation signals to enhance integrin function has provided additional evidence regarding the importance of integrin activation to immune system function (150). Further analysis of how integrins participate in the immune response will clearly facilitate ongoing efforts to utilize integrins as therapeutic targets for treatment of diseases such as inflammatory bowel disease, asthma, and multiple sclerosis. Because global loss of integrin function will have unwanted immunosuppressive effects, the challenge ahead lies in capitalizing on basic research findings discussed in this review to devise clinically effective therapies.

ACKNOWLEDGMENTS This work is supported by NIH grants AI31126 and AI38474 (Y.S.), the Harry Kay Chair in Biomedical Research (Y.S.), a University of Minnesota Doctoral Dissertation Fellowship (J.T.P.), NIH training grant T32 HL07741 (A.C.Q.), and NIH training grant T32 AI007313 (K.L.M.). The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Shimizu Y, Rose DM, Ginsberg MH. 1999. Integrins and the immune response. Adv. Immunol. 72:325–80 2. Hynes RO. 2002. Integrins: bidirectional, allosteric signaling machines. Cell 110:673–87 3. Lasky LA, Singer MS, Dowbenko D, Imai Y, Henzel WJ, et al. 1992. An endothelial ligand for L-selectin is a novel mucin-like molecule. Cell 69:927–38 4. Chen C, Mobley JL, Dwir O, Shimron F, Grabovsky V, et al. 1998. High affinity very late antigen-4 subsets expressed on T cells are mandatory for chemokineindependent adhesion strengthening but not for rolling on VCAM-1 in shear flow. J. Immunol. 162:1084–95

5. Grabovsky V, Feigelson S, Chen C, Bleijs DA, Peled A, et al. 2000. Subsecond induction of α4 integrin clustering by immobilized chemokines stimulates leukocyte tethering and rolling on endothelial vascular cell adhesion molecule 1 under flow conditions. J. Exp. Med. 192:495– 505 6. Berlin C, Bargatze RF, Campbell JJ, Von Andrian UH, Szabo MC, et al. 1995. α4 integrins mediate lymphocyte attachment and rolling under physiologic flow. Cell 80:413–22 7. Steeber DA, Campbell MA, Basit A, Ley K, Tedder TF. 1998. Optimal selectinmediated rolling of leukocytes during inflammation in vivo requires intercellular

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

INTEGRINS AND T CELL IMMUNE RESPONSES

8.

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

9.

10.

11.

12.

13.

14.

15.

adhesion molecule-1 expression. Proc. Natl. Acad. Sci. USA 95:7562–67 Sigal A, Bleijs DA, Grabovsky V, Van Vliet SJ, Dwir O, et al. 2000. The LFA1 integrin supports rolling adhesions on ICAM-1 under physiological sheer flow in a permissive cellular environment. J. Immunol. 165:442–52 Salas A, Shimaoka M, Chen SQ, Carman CV, Springer T. 2002. Transition from rolling to firm adhesion is regulated by the conformation of the I domain of the integrin lymphocyte function-associated antigen-1. J. Biol. Chem. 277:50255– 62 Kadono T, Venturi GM, Steeber DA, Tedder TF. 2002. Leukocyte rolling velocities and migration are optimized by cooperative L-selectin and intercellular adhesion molecule-1 functions. J. Immunol. 169:4542–50 Campbell JJ, Hedrick J, Zlotnik A, Siani MA, Thompson DA, Butcher EC. 1998. Chemokines and the arrest of lymphocytes rolling under flow conditions. Science 279:381–84 Chan JR, Hyduk SJ, Cybulsky MI. 2001. Chemoattractants induce a rapid and transient upregulation of monocyte α4 integrin affinity for vascular cell adhesion molecule 1 which mediates arrest: an early step in the process of emigration. J. Exp. Med. 193:1149–58 Constantin G, Majeed M, Giagulli C, Piccio L, Kim JY, et al. 2000. Chemokines trigger immediate β2 integrin affinity and mobility changes: differential regulation and roles in lymphocyte arrest under flow. Immunity 13:759–69 van Kooyk Y, Figdor CG. 2000. Avidity regulation of integrins: the driving force in leukocyte adhesion. Curr. Opin. Cell Biol. 12:542–47 Shamri R, Grabovsky V, Feigelson SW, Dwir O, van Kooyk Y, Alon R. 2002. Chemokine stimulation of lymphocyte α 4 integrin avidity but not of leukocyte function-associated antigen-1 avidity to

16.

17.

18.

19.

20.

21.

22.

23.

24.

173

endothelial ligands under shear flow requires cholesterol membrane rafts. J. Biol. Chem. 277:40027–35 Hwang ST, Singer MS, Giblin PA, Yednock TA, Bacon KB, et al. 1996. GlyCAM-1, a physiologic ligand for Lselectin, activates β2 integrins on naive peripheral lymphocytes. J. Exp. Med. 184:1343–48 Rose DM, Grabovsky V, Alon R, Ginsberg MH. 2001. The affinity of integrin α 4β 1 governs lymphocyte migration. J. Immunol. 167:2824–30 Rose DM, Liu S, Woodside DG, Han J, Schlaepfer DD, Ginsberg MH. 2003. Paxillin binding to the α 4 integrin subunit stimulates LFA-1 (integrin α Lβ 2)dependent T cell migration by augmenting the activation of focal adhesion kinase/proline-rich tyrosine kinase-2. J. Immunol. 170:5912–18 Savinov AY, Wong FS, Stonebraker AC, Chervonsky AV. 2003. Presentation of antigen by endothelial cells and chemoattraction are required for homing of insulin-specific CD8 + T cells. J. Exp. Med. 197:643–56 Muller WA. 2003. Leukocyte-endothelial-cell interactions in leukocyte transmigration and the inflammatory response. Trends Immunol. 24:326–33 Kavanaugh AF, Lightfoot E, Lipsky PE, Oppenheimer-Marks N. 1991. Role of CD11/CD18 in adhesion and transendothelial migration of T cells: analysis utilizing CD18-deficient T cell clones. J. Immunol. 146:4149–56 Ding ZQ, Xiong K, Issekutz TB. 2001. Chemokines stimulate human T lymphocyte transendothelial migration to utilize VLA-4 in addition to LFA-1. J. Leukoc. Biol. 69:458–66 Ostermann G, Weber KS, Zernecke A, Schroder A, Weber C. 2002. JAM-1 is a ligand of the β 2 integrin LFA-1 involved in transendothelial migration of leukocytes. Nat. Immunol. 3:151–58 Schenkel AR, Mamdouh Z, Chen X,

11 Feb 2004

17:3

174

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

25.

26.

27.

28.

29. 30.

31.

32.

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL. Liebman RM, Muller WA. 2002. CD99 plays a major role in the migration of monocytes through endothelial junctions. Nat. Immunol. 3:143–50 Kaldjian EP, Gretz JE, Anderson AO, Shi Y, Shaw S. 2001. Spatial and molecular organization of lymph node T cell cortex: a labyrinthine cavity bounded by an epithelium-like monolayer of fibroblastic reticular cells anchored to basement membrane-like extracellular matrix. Int. Immunol. 13:1243–53 Gretz JE, Norbury CC, Anderson AO, Proudfoot AE, Shaw S. 2000. Lymphborne chemokines and other low molecular weight molecules reach high endothelial venules via specialized conduits while a functional barrier limits access to the lymphocyte microenvironments in lymph node cortex. J. Exp. Med. 192:1425–39 Miyata K, Takaya K. 1981. Elastic fibers associated with collagenous fibrils surrounded by reticular cells in lymph nodes of the rat as revealed by electron microscopy after orcein staining. Cell Tissue Res. 220:445–48 Itano AA, McSorley SJ, Reinhardt RL, Ehst BD, Ingulli E, et al. 2003. Distinct dendritic cell populations sequentially present a subcutaneous antigen to CD4 T cells and stimulate different aspects of cell-mediated immunity. Immunity 19:47–57 Deleted in proof Karttunen T, Alavaikko M, ApajaSarkkinen M, Autio-Harmainen H. 1986. Distribution of basement membrane laminin and type IV collagen in human reactive lymph nodes. Histopathology 10:841–49 Ocklind G, Talts J, Fassler R, Mattsson A, Ekblom P. 1993. Expression of tenascin in developing and adult mouse lymphoid organs. J. Histochem. Cytochem. 41:1163– 69 Hayakawa M, Kobayashi M, Hoshino T. 1988. Direct contact between reticular fibers and migratory cells in the paracor-

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

tex of mouse lymph nodes: a morphological and quantitative study. Arch. Histol. Cytol. 51:233–40 Miller MJ, Wei SH, Parker I, Cahalan MD. 2002. Two-photon imaging of lymphocyte motility and antigen response in intact lymph node. Science 296:1869–73 Stoll S, Delon J, Brotz TM, Germain RN. 2002. Dynamic imaging of T celldendritic cell interactions in lymph nodes. Science 296:1873–76 Miller MJ, Wei SH, Cahalan MD, Parker I. 2003. Autonomous T cell trafficking examined in vivo with intravital two-photon microscopy. Proc. Natl. Acad. Sci. USA 100:2604–9 Bousso P, Robey E. 2003. Dynamics of CD8+ T cell priming by dendritic cells in intact lymph nodes. Nat. Immunol. 4:579– 85 Dustin ML, Springer TA. 1989. T-cell receptor cross-linking transiently stimulates adhesiveness through LFA-1. Nature 341:619–24 van Kooyk Y, van de Wiel-van Kemenade P, Weder P, Kuijpers TW, Figdor CG. 1989. Enhancement of LFA1-mediated cell adhesion by triggering through CD2 or CD3 on T lymphocytes. Nature 342:811–13 Morgan MM, Labno CM, van Seventer GA, Denny MF, Straus DB, Burkhardt JK. 2001. Superantigen-induced T cell:B cell conjugation is mediated by LFA-1 and requires signaling through Lck, but not ZAP-70. J. Immunol. 167:5708–18 Bromley SK, Dustin ML. 2002. Stimulation of naive T-cell adhesion and immunological synapse formation by chemokinedependent and -independent mechanisms. Immunology 106:289–98 Stefanova I, Dorfman JR, Germain RN. 2002. Self-recognition promotes the foreign antigen sensitivity of naive T lymphocytes. Nature 420:429–34 Revy P, Sospedra M, Barbour B, Trautmann A. 2001. Functional antigenindependent synapses formed between

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

INTEGRINS AND T CELL IMMUNE RESPONSES

43.

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

44.

45.

46.

47.

48.

49.

50.

51.

52.

T cells and dendritic cells. Nat. Immunol. 2:925–31 Stewart MP, McDowall A, Hogg N. 1998. LFA-1-mediated adhesion is regulated by cytoskeletal restraint and by a Ca2+dependent protease, calpain. J. Cell Biol. 140:699–707 van Kooyk Y, Geijtenbeek TB. 2002. A novel adhesion pathway that regulates dendritic cell trafficking and T cell interactions. Immunol. Rev. 186:47–56 Monks CRF, Freiberg BA, Kupfer H, Sciaky N, Kupfer A. 1998. Threedimensional segregation of supramolecular activation clusters in T cells. Nature 395:82–86 Grakoui A, Bromley SK, Sumen C, Davis MM, Shaw AS, et al. 1999. The immunological synapse: a molecular machine controlling T cell activation. Science 285:221–27 Lee KH, Holdorf AD, Dustin ML, Chan AC, Allen PM, Shaw AS. 2002. T cell receptor signaling precedes immunological synapse formation. Science 295:1539– 42 Freiberg BA, Kupfer H, Maslanik W, Delli J, Kappler J, et al. 2002. Staging and resetting T cell activation in SMACs. Nat. Immunol. 3:911–17 Iezzi G, Karjalainen K, Lanzavecchia A. 1998. The duration of antigenic stimulation determines the fate of naive and effector T cells. Immunity 8:89–95 Huang JY, Lo PF, Zal T, Gascoigne NRJ, Smith BA, et al. 2002. CD28 plays a critical role in the segregation of PKCθ within the immunologic synapse. Proc. Natl. Acad. Sci. USA 99:9369–73 van Stipdonk MJ, Lemmens EE, Schoenberger SP. 2001. Naive CTLs require a single brief period of antigenic stimulation for clonal expansion and differentiation. Nat. Immunol. 2:423–29 Cyster JG, Shotton DM, Williams AF. 1991. The dimensions of the T lymphocyte glycoprotein leukosialin and identification of linear protein epitopes that can

53.

54.

55.

56.

57.

58.

59.

60.

61.

175

be modified by glycosylation. EMBO J. 10:893–902 Garboczi DN, Ghosh P, Utz U, Fan QR, Biddison WE, Wiley DC. 1996. Structure of the complex between human T-cell receptor, viral peptide and HLA-A2. Nature 384:134–41 Garcia KC, Degano M, Stanfield RL, Brunmark A, Jackson MR, et al. 1996. An αβ T cell receptor structure at 2.5 A and its orientation in the TCR-MHC complex. Science 274:209–19 Takagi J, Petre BM, Walz T, Springer TA. 2002. Global conformational rearrangements in integrin extracellular domains in outside-in and inside-out signaling. Cell 110:599–611 Xiong JP, Stehle T, Diefenbach B, Zhang RG, Dunker R, et al. 2001. Crystal structure of the extracellular segment of integrin αVβ3. Science 294:339–45 Xiong JP, Stehle T, Zhang R, Joachimiak A, Frech M, et al. 2002. Crystal structure of the extracellular segment of integrin αVβ3 in complex with an Arg-Gly-Asp ligand. Science 296:151–55 Shimaoka M, Xiao T, Liu JH, Yang YT, Dong YC, et al. 2003. Structures of the αL I domain and its complex with ICAM-1 reveal a shape-shifting pathway for integrin regulation. Cell 112:99– 111 van Seventer GA, Shimizu Y, Horgan KJ, Shaw S. 1990. The LFA-1 ligand ICAM-1 provides an important costimulatory signal for T cell receptor-mediated activation of resting T cells. J. Immunol. 144:4579– 86 Shimizu Y, van Seventer GA, Horgan KJ, Shaw S. 1990. Costimulation of proliferative responses of resting CD4+ T cells by the interaction of VLA-4 and VLA-5 with fibronectin or VLA-6 with laminin. J. Immunol. 145:59–67 Abraham C, Miller J. 2001. Molecular mechanisms of IL-2 gene regulation following costimulation through LFA-1. J. Immunol. 167:5193–201

11 Feb 2004

17:3

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

176

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL.

62. Ni HT, Deeths MJ, Mescher MF. 2001. LFA-1-mediated costimulation of CD8+ T cell proliferation requires phosphatidylinositol 3-kinase activity. J. Immunol. 166:6523–29 63. Bianchi E, Denti S, Granata A, Bossi G, Geginat J, et al. 2000. Integrin LFA1 interacts with the transcriptional coactivator JAB1 to modulate AP-1 activity. Nature 404:617–18 64. Leitinger B, Hogg N. 2002. The involvement of lipid rafts in the regulation of integrin function. J. Cell Sci. 115:963– 72 65. Horejsi V. 2003. The roles of membrane microdomains (rafts) in T cell activation. Immunol. Rev. 191:148–64 66. Hunter AJ, Shimizu Y. 1997. α4β1 integrin-mediated tyrosine phosphorylation in human T cells: characterization of Crk- and Fyn-associated substrates (pp105, pp115, and human enhancer of filamentation-1) and integrindependent activation of p59fyn. J. Immunol. 159:4806–14 67. Hunter AJ, Ottoson NC, Boerth N, Koretzky GA, Shimizu Y. 2000. Cutting edge: a novel function for the SLAP-130/FYB adapter protein in β1 integrin signaling and T lymphocyte migration. J. Immunol. 164:1143–47 68. Uemura N, Salgia R, Ewaniuk DS, Little MT, Griffin JD. 1999. Involvement of the adapter protein CRKL in integrinmediated adhesion. Oncogene 18:3343– 53 69. Li LM, Guris DL, Okura M, Imamoto A. 2003. Translocation of CrkL to focal adhesions mediates integrin-induced migration downstream of src family kinases. Mol. Cell. Biol. 23:2883–92 70. Peterson EJ, Woods ML, Dmowski SA, Derimanov G, Jordan MS, et al. 2001. Coupling of the TCR to integrin activation by SLAP-130/Fyb. Science 293:2263–65 71. Griffiths EK, Krawczyk C, Kong YY, Raab M, Hyduk SJ, et al. 2001. Positive regulation of T cell activation and integrin

72.

73.

74.

75.

76.

77.

78.

79.

80.

adhesion by the adapter Fyb/Slap. Science 293:2260–63 Raab M, Kang H, Da Silva A, Zhu XC, Rudd CE. 1999. FYN-T-FYB-SLP76 interactions define a T-cell receptor ζ /CD3-mediated tyrosine phosphorylation pathway that up-regulates interleukin 2 transcription in T-cells. J. Biol. Chem. 274:21170–79 Tabassam FH, Umehara H, Huang JY, Gouda S, Kono T, et al. 1999. β 2integrin, LFA-1, and TCR/CD3 synergistically induce tyrosine phosphorylation of focal adhesion kinase (pp125FAK) in PHAactivated T cells. Cell. Immunol. 193:179– 84 van Seventer GA, Salmen HJ, Law SF, O’Neill GM, Mullen MM, et al. 2001. Focal adhesion kinase regulates β 1 integrindependent T cell migration through an HEF1 effector pathway. Eur. J. Immunol. 31:1417–27 Kamiguchi K, Tachibana K, Iwata S, Ohashi Y, Morimoto C. 1999. Cas-L is required for β 1 integrin-mediated costimulation in human T cells. J. Immunol. 163:563–68 Dustin ML, Bromley SK, Kan ZY, Peterson DA, Unanue ER. 1997. Antigen receptor engagement delivers a stop signal to migrating T lymphocytes. Proc. Natl. Acad. Sci. USA 94:3909–13 Woods ML, Caba˜nas C, Shimizu Y. 2000. Activation-dependent changes in soluble fibronectin binding and expression of β 1 integrin activation epitopes in T cells: relationship to T cell adhesion and migration. Eur. J. Immunol. 30:38–49 Kane LP, Lin J, Weiss A. 2000. Signal transduction by the TCR for antigen. Curr. Opin. Immunol. 12:242–49 Fagerholm S, Hilden TJ, Gahmberg CG. 2002. Lck tyrosine kinase is important for activation of the CD11a/CD18-integrins in human T lymphocytes. Eur. J. Immunol. 32:1670–78 Woods ML, Kivens WJ, Adelsman MA, Qiu Y, August A, Shimizu Y. 2001. A

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

INTEGRINS AND T CELL IMMUNE RESPONSES

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

81.

82.

83.

84.

85.

86.

87.

88.

novel function for the Tec family tyrosine kinase Itk in activation of β1 integrins by the T cell receptor. EMBO J. 20:1232–44 Epler JA, Liu R, Chung H, Ottoson NC, Shimizu Y. 2000. Regulation of β1 integrin-mediated adhesion by T cell receptor signaling involves ZAP-70 but differs from signaling events that regulate transcriptional activity. J. Immunol. 165:4941–49 Krawczyk C, Oliveira-dos-Santos A, Sasaki T, Griffiths E, Ohashi PS, et al. 2002. Vav1 controls integrin clustering and MHC/peptide-specific cell adhesion to antigen-presenting cells. Immunity 16:331–43 Wells AD, Liu QH, Hondowicz B, Zhang JD, Turka LA, Freedman BD. 2003. Regulation of T cell activation and tolerance by phospholipase Cγ -1-dependent integrin avidity modulation. J. Immunol. 170:4127–33 Kucik DF, Dustin ML, Miller JM, Brown EJ. 1996. Adhesion-activating phorbol ester increases the mobility of leukocyte integrin LFA-1 in cultured lymphocytes. J. Clin. Invest. 97:2139–44 Katagiri K, Hattori M, Minato N, Irie S, Takatsu K, Kinashi T. 2000. Rap1 is a potent activation signal for leukocyte function-associated antigen 1 distinct from protein kinase C and phosphatidylinositol-3-OH kinase. Mol. Cell. Biol. 20:1956–69 Mobley JL, Ennis E, Shimizu Y. 1994. Differential activation-dependent regulation of integrin function in cultured human T-leukemic cell lines. Blood 83:1039–50 Musci MA, Hendricks-Taylor LR, Motto DG, Paskind M, Kamens J, et al. 1997. Molecular cloning of SLAP-130, an SLP76-associated substrate of the T cell antigen receptor-stimulated protein tyrosine kinases. J. Biol. Chem. 272:11674–77 Da Silva AJ, Li ZW, De Vera C, Canto E, Findell P, Rudd CE. 1997. Cloning of a novel T-cell protein FYB that binds FYN and SH2-domain-containing leuko-

89.

90.

91.

92.

93.

94.

95.

96.

97.

177

cyte protein 76 and modulates interleukin 2 production. Proc. Natl. Acad. Sci. USA 94:7493–98 Krause M, Sechi AS, Konradt M, Monner D, Gertler FB, Wehland J. 2000. Fynbinding protein (Fyb)/SLP-76-associated protein (SLAP), Ena/vasodilator-stimulated phosphoprotein (VASP) proteins and the Arp2/3 complex link T cell receptor (TCR) signaling to the actin cytoskeleton. J. Cell Biol. 149:181–94 Wang H, Moon EY, Azouz A, Wu X, Smith A, et al. 2003. SKAP-55 regulates integrin adhesion and formation of T cellAPC conjugates. Nat. Immunol. 4:366– 74 Ardouin L, Bracke M, Mathiot A, Pagakis SN, Norton T, et al. 2003. Vav1 transduces TCR signals required for LFA-1 function and cell polarization at the immunological synapse. Eur. J. Immunol. 33:790–97 Bustelo XR. 2000. Regulatory and signaling properties of the Vav family. Mol. Cell. Biol. 20:1461–77 D’Souza-Schorey C, Boettner B, Van Aelst L. 1998. Rac regulates integrinmediated spreading and increased adhesion of T lymphocytes. Mol. Cell. Biol. 18:3936–46 Shimonaka M, Katagiri K, Nakayama T, Fujita N, Tsuruo T, et al. 2003. Rap1 translates chemokine signals to integrin activation, cell polarization, and motility across vascular endothelium under flow. J. Cell Biol. 161:417–27 Gomez M, Kioussis D, Cantrell DA. 2001. The GTPase Rac-1 controls cell fate in the thymus by diverting thymocytes from positive to negative selection. Immunity 15:703–13 Corre I, Gomez M, Vielkind S, Cantrell DA. 2001. Analysis of thymocyte development reveals that the GTPase RhoA is a positive regulator of T cell receptor responses in vivo. J. Exp. Med. 194:903– 13 Rodr´ıguez-Fern´andez JL, S´anchez-Mart´ın L, Rey M, Vicente-Manzanares M,

11 Feb 2004

17:3

178

98.

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

99.

100.

101.

102.

103.

104.

105.

106.

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL. Narumiya S, et al. 2001. Rho and Rhoassociated kinase modulate the tyrosine kinase PYK2 in T-cells through regulation of the activity of the integrin LFA-1. J. Biol. Chem. 276:40518–27 Bos JL, De Rooij J, Reedquist KA. 2001. Rap1 signalling: adhering to new models. Nat. Rev. Mol. Cell Biol. 2:369–77 Sebzda E, Bracke M, Tugal T, Hogg N, Cantrell DA. 2002. Rap1A positively regulates T cells via integrin activation rather than inhibiting lymphocyte signaling. Nat. Immunol. 3:251–58 Zhang W, Shao Y, Fang D, Huang J, Jeon MS, Liu YC. 2003. Negative regulation of T cell antigen receptor-mediated Crk-LC3G signaling and cell adhesion by Cbl-b. J. Biol. Chem. 278:23978–83 Katagiri K, Hattori M, Minato N, Kinashi T. 2002. Rap1 functions as a key regulator of T-cell antigen-presenting cell interactions and modulates T-cell responses. Mol. Cell. Biol. 22:1001–15 Reedquist KA, Ross E, Koop EA, Wolthuis RM, Zwartkruis FJ, et al. 2000. The small GTPase, Rap1, mediates CD31induced integrin adhesion. J. Cell Biol. 148:1151–58 Tsukamoto N, Hattori M, Yang HL, Bos JL, Minato N. 1999. Rap1 GTPaseactivating protein SPA-1 negatively regulates cell adhesion. J. Biol. Chem. 274:18463–69 Boussiotis VA, Freeman GJ, Berezovskaya A, Barber DL, Nadler LM. 1997. Maintenance of human T cell anergy: blocking of IL-2 gene transcription by activated Rap1. Science 278:124–28 Liu L, Schwartz BR, Tupper J, Lin N, Winn RK, Harlan JM. 2002. The GTPase Rap1 regulates phorbol 12-myristate 13acetate-stimulated but not ligand-induced β 1 integrin-dependent leukocyte adhesion. J. Biol. Chem. 277:40893–900 De Bruyn KMT, Rangarajan S, Reedquist KA, Figdor CG, Bos JL. 2002. The small GTPase Rap1 is required for Mn2+- and antibody-induced LFA-1- and VLA-4-

107.

108.

109.

110.

111.

112.

113.

114.

115.

mediated cell adhesion. J. Biol. Chem. 277:29468–76 Katagiri K, Maeda A, Shimonaka M, Kinashi T. 2003. RAPL, a Rap1-binding molecule that mediates Rap1-induced adhesion through spatial regulation of LFA1. Nat. Immunol. 4:741–48 Li RH, Mitra N, Gratkowski H, Vilaire G, Litvinov R, et al. 2003. Activation of integrin αIIbβ3 by modulation of transmembrane helix associations. Science 300:795–98 Weljie AM, Hwang PM, Vogel HJ. 2002. Solution structures of the cytoplasmic tail complex from platelet integrin αIIbβ3-subunits. Proc. Natl. Acad. Sci. USA 99:5878–83 Vinogradova O, Velyvis A, Velyviene A, Hu B, Haas TA, et al. 2002. A structural mechanism of integrin α IIb β 3 “inside-out” activation as regulated by its cytoplasmic face. Cell 110:587–97 Hughes PE, Diaz-Gonzalez F, Leong L, Wu CY, McDonald JA, et al. 1996. Breaking the integrin hinge—a defined structural constraint regulates integrin signaling. J. Biol. Chem. 271:6571–74 Calderwood DA, Zent R, Grant R, Rees DJ, Hynes RO, Ginsberg MH. 1999. The talin head domain binds to integrin β subunit cytoplasmic tails and regulates integrin activation. J. Biol. Chem. 274:28071– 74 Calderwood DA, Yan B, De Pereda JM, Alvarez BG, Fujioka Y, et al. 2002. The phosphotyrosine binding-like domain of talin activates integrins. J. Biol. Chem. 277:21749–58 Lu CF, Takagi J, Springer TA. 2001. Association of the membrane proximal regions of the α and β subunit cytoplasmic domains constrains an integrin in the inactive state. J. Biol. Chem. 276:14642–48 Rott LS, Briskin MJ, Andrew DP, Berg EL, Butcher EC. 1996. A fundamental subdivision of circulating lymphocytes defined by adhesion to mucosal addressin cell adhesion molecule-1—comparison

11 Feb 2004

17:3

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

INTEGRINS AND T CELL IMMUNE RESPONSES

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

116.

117.

118.

119.

120.

121.

122.

123.

124.

with vascular cell adhesion molecule-1 and correlation with β 7 integrins and memory differentiation. J. Immunol. 156:3727–36 Schweighoffer T, Tanaka Y, Tidswell M, Erle DJ, Horgan KJ, et al. 1993. Selective expression of integrin α4β7 on a subset of human CD4+ memory T cells with hallmarks of gut-trophism. J. Immunol. 151:717–29 Erle DJ, Briskin MJ, Butcher EC, GarciaPardo A, Lazarovits AI, Tidswell M. 1994. Expression and function of the MAdCAM-1 receptor, integrin α4β7, on human leukocytes. J. Immunol. 153:517– 28 Nakache M, Berg EL, Streeter PR, Butcher EC. 1989. The mucosal vascular addressin is a tissue-specific endothelial cell adhesion molecule for circulating lymphocytes. Nature 337:179–81 Bargatze RF, Jutila MA, Butcher EC. 1995. Distinct roles of L-selectin and integrin α4β7 in lymphocyte homing to Peyer’s patch-HEV in situ: the multistep model confirmed and refined. Immunity 3:99–108 Campbell DJ, Butcher EC. 2002. Rapid acquisition of tissue-specific homing phenotypes by CD4+ T cells activated in cutaneous or mucosal lymphoid tissues. J. Exp. Med. 195:135–41 Stagg AJ, Kamm MA, Knight SC. 2002. Intestinal dendritic cells increase T cell expression of α4β7 integrin. Eur. J. Immunol. 32:1445–54 Cepek KL, Parker CM, Madara JL, Brenner MB. 1993. Integrin α Eβ 7 mediates adhesion of T lymphocytes to epithelial cells. J. Immunol. 150:3459–70 Andrew DP, Rott LS, Kilshaw PJ, Butcher EC. 1996. Distribution of α 4β 7 and α Eβ 7 integrins on thymocytes, intestinal epithelial lymphocytes and peripheral lymphocytes. Eur. J. Immunol. 26:897– 905 Higgins JMG, Mandlebrot DA, Shaw SK, Russell GJ, Murphy EA, et al. 1998.

125.

126.

127.

128.

129.

130.

131.

132.

133.

179

Direct and regulated interaction of integrin α Eβ 7 with E-cadherin. J. Cell Biol. 140:197–210 Corps E, Carter C, Karecla P, Ahrens T, Evans P, Kilshaw P. 2001. Recognition of E-cadherin by integrin α Eβ 7-requirement for cadherin dimerization and implications for cadherin and integrin function. J. Biol. Chem. 276:30862–70 Sarnacki S, B`egue B, Buc H, Le Deist F, Cerf-Bensussan N. 1992. Enhancement of CD3-induced activation of human intestinal intraepithelial lymphocytes by stimulation of the β 7-containing integrin defined by HML-1 monoclonal antibody. Eur. J. Immunol. 22:2887–92 Sch¨on MP, Arya A, Murphy EA, Adams CM, Strauch UG, et al. 1999. Mucosal T lymphocyte numbers are selectively reduced in integrin α E (CD103)-deficient mice. J. Immunol. 162:6641–49 Lehmann J, Huehn J, de la Rosa M, Maszyna F, Kretschmer U, et al. 2002. Expression of the integrin αEβ7 identifies unique subsets of CD25+ as well as CD25-regulatory T cells. Proc. Natl. Acad. Sci. USA 99:13031–36 Ludviksson BR, Strober W, Nishikomori R, Hasan SK, Ehrhardt RO. 1999. Administration of mAb against αEβ7 prevents and ameliorates immunizationinduced colitis in IL-2−/− mice. J. Immunol. 162:4975–82 Sch¨on MP, Sch¨on M, Warren HB, Donohue JP, Parker CM. 2000. Cutaneous inflammatory disorder in integrin α E (CD103)-deficient mice. J. Immunol. 165:6583–89 Hemler ME. 1990. VLA proteins in the integrin family: structures, functions, and their role on leukocytes. Annu. Rev. Immunol. 8:365–400 Krivacic KA, Levine AD. 2003. Extracellular matrix conditions T cells for adhesion to tissue interstitium. J. Immunol. 170:5034–44 Bank I, Koltakov A, Goldstein I, Chess L. 2002. Lymphocytes expressing α1β1

11 Feb 2004

17:3

180

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

134.

135.

136.

137.

138.

139.

140.

141.

AR

AR210-IY22-06.tex

AR210-IY22-06.sgm

LaTeX2e(2002/01/18)

P1: IKH

PRIBILA ET AL. integrin (very late antigen-1) in peripheral blood of patients with arthritis are a subset of CD45RO(+) T-cells primed for rapid adhesion to collagen IV. Clin. Immunol. 105:247–58 De Fougerolles AR, Sprague AG, Nickerson-Nutter CL, Chi-Rosso G, Rennert PD, et al. 2000. Regulation of inflammation by collagen-binding integrins α1β1 and α2β1 in models of hypersensitivity and arthritis. J. Clin. Invest. 105:721–29 Fiorucci S, Mencarelli A, Palazzetti B, Sprague AG, Distrutti E, et al. 2002. Importance of innate immunity and collagen binding integrin α1β1 in TNBS-induced colitis. Immunity 17:769–80 Krieglstein CF, Cerwinka WH, Sprague AG, Laroux FS, Grisham MB, et al. 2002. Collagen-binding integrin α 1β 1 regulates intestinal inflammation in experimental colitis. J. Clin. Invest. 110:1773–82 Meharra EJ, Sch¨on M, Hassett D, Parker C, Havran W, Gardner H. 2000. Reduced gut intraepithelial lymphocytes in VLA1 null mice. Cell. Immunol. 201:1–5 Dustin ML, De Fougerolles AR. 2001. Reprograming T cells: the role of extracellular matrix in coordination of T cell activation and migration. Curr. Opin. Immunol. 13:286–90 Metlay JP, Witmer-Pack MD, Agger R, Crowley MT, Lawless D, Steinman RM. 1990. The distinct leukocyte integrins of mouse spleen dendritic cells as identified with new hamster monoclonal antibodies. J. Exp. Med. 171:1753–71 Henri S, Vremec D, Kamath A, Waithman J, Williams S, et al. 2001. The dendritic cell populations of mouse lymph nodes. J. Immunol. 167:741–48 Salomon B, Cohen JL, Masurier C, Klatzmann D. 1998. Three populations of mouse lymph node dendritic cells with different origins and dynamics. J. Immunol. 160:708–17

142. Ruedl C, Koebel P, Bachmann M, Hess M, Karjalainen K. 2000. Anatomical origin of dendritic cells determines their life span in peripheral lymph nodes. J. Immunol. 165:4910–16 143. Kilshaw PJ. 1993. Expression of the mucosal T cell integrin α M290β7 by a major subpopulation of dendritic cells in mice. Eur. J. Immunol. 23:3365–68 144. Puig-Kroger A, Sanz-Rodriguez F, Longo N, Sanchez-Mateos P, Botella L, et al. 2000. Maturation-dependent expression and function of the CD49d integrin on monocyte-derived human dendritic cells. J. Immunol. 165:4338–45 145. Geijtenbeek TB, Krooshoop DJ, Bleijs DA, Van Vliet SJ, Van Duijnhoven GC, et al. 2000. DC-SIGN-ICAM-2 interaction mediates dendritic cell trafficking. Nat. Immunol. 1:353–57 146. Brown KA, Bedford P, Macey M, McCarthy DA, LeRoy F, et al. 1997. Human blood dendritic cells: binding to vascular endothelium and expression of adhesion molecules. Clin. Exp. Immunol. 107:601– 7 147. Ma J, Wang J-H, Guo Y-J, Sy M-S, Bigby M. 1994. In vivo treatment with antiICAM-1 and anti-LFA-1 antibodies inhibits contact sensitization-induced migration of epidermal Langerhans cells to regional lymph nodes. Cell. Immunol. 158:389–99 148. Xu H, Guan H, Zu G, Bullard D, Hanson J, et al. 2001. The role of ICAM-1 molecule in the migration of Langerhans cells in the skin and regional lymph node. Eur. J. Immunol. 31:3085–93 149. Price AA, Cumberbatch M, Kimber I, Ager A. 1997. α6 integrins are required for Langerhans cell migration from the epidermis. J. Exp. Med. 186:1725– 35 150. Shimizu Y. 2003. Disabling multiple integrins from the inside out. J. Clin. Invest. 111:23–24

Pribila.qxd

3/12/2004

1:52 PM

Page 1

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTEGRINS AND T CELL IMMUNE RESPONSES

C-1

Figure 3 Signaling pathways implicated in TCR signaling to integrins. A large array of intracellular proteins, including tyrosine, serine/threonine and lipid kinases, adapter proteins, guanine nucleotide exchange factors, GTPases, and cytoskeletal proteins, are involved in increases in integrin function that occur following TCR stimulation. Potential relationships between these various signaling pathways are shown, but for many of these intracellular mediators, the biochemical basis by which they mediate integrin activation remains undefined.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:157-180. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104603

Annu. Rev. Immunol. 2004. 22:181–215 doi: 10.1146/annurev.immunol.22.012703.104603 First published online as a Review in Advance on January 12, 2004

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE∗ De Yang,1 Arya Biragyn,2 David M. Hoover,3 Jacek Lubkowski,3 and Joost J. Oppenheim4 1

Basic Research Program; 3Macromolecular Crystallography Laboratory, SAIC-Frederick; 4Laboratory of Molecular Immunoregulation, Center for Cancer Research; National Cancer Institute at Frederick, Frederick, Maryland 21702; 2 Laboratory of Immunology, Gerontology Research Center, National Institute of Aging, Baltimore, Maryland 20892; email: [email protected]

Key Words antimicrobial, chemotaxis, dendritic cells, GPCR, immune response ■ Abstract Mammals generate a diverse array of antimicrobial proteins, largely represented by defensins or cathelicidins. The direct in vitro microbicidal activity of antimicrobial proteins has long been considered an important innate immune defense, although the in vivo relevance has only very recently been established for certain defensins and cathelicidins. Mammalian defensins and cathelicidins have also been shown to have multiple receptor-mediated effects on immune cells. Beta-defensins interact with CCR6; murine β-defensin-2 in addition activates TLR4. Cathelicidins act on FPRL1-expressing cells. Furthermore, several defensins have considerable immunoenhancing activity. Thus, it appears that mammalian antimicrobial proteins contribute to both innate and adaptive antimicrobial immunity.

INTRODUCTION Mammals are constantly exposed to a myriad of microorganisms, yet individuals rarely become infected because of the barrier function of the skin and epithelia that prevent microbial entry based on mechanical obstruction and the presence of antimicrobial substances. If the barrier is breached, invading pathogens are contained and ultimately eliminated by the host immune system. Mammals have evolved two kinds of immunity: innate, nonclonal and nonspecific; and adaptive, inducible and antigen-specific (1, 2). Innate immunity represents the first line of constitutively pre-existing host defense that is rapidly mobilized following the ∗ The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

181

12 Feb 2004

20:51

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

182

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

detection of microbial invasion. The effector branch of innate immunity consists of two major components: the recruitment and/or activation of leukocytes (e.g., phagocytic granulocytes, monocytes/macrophages, etc.) capable of combating the invading pathogens and the release and/or activation of a variety of extracellular humoral mediators (e.g., complement, cytokines, antimicrobial substances, etc.). Adaptive immunity is induced when lymphocytes are activated in response to antigen (Ag) presented by antigen-presenting cells (APCs), in particular dendritic cells (DCs) (1, 3). The T cell antigen receptors recognize antigenic epitopes bound to the major histocompatibility complex (MHC) on the surface of APCs. CD8+ T cells, once activated after recognition of antigenic epitopes bound to the MHC class I, differentiate into cytotoxic T cells that directly kill cells infected by intracellular pathogens. The complex of MHC class II and antigenic epitope triggers the activation of CD4+, generating T-helper cells that, by producing various cytokines, promote B cell activation and enhance the efficiency of phagocytes to eliminate pathogens. B cell activation leads to the production of antigen-specific antibodies that neutralize pathogen-derived toxins, block the infectivity of invading pathogens, and promote their opsonization and elimination by phagocytes. Thus, innate and adaptive immune responses work in concert to eliminate microbial invaders. Antimicrobial substances comprise microbicidal chemicals (e.g., hydrogen peroxide, nitric oxide, etc.) and a wide variety of host gene-encoded antimicrobial proteins. Almost half a century ago, a crude protein fraction of polymorphonuclear leukocytes with antimicrobial properties was described and named “phagocytin” (4), the prologue for investigating antimicrobial components of inflammatory leukocytes. The subsequent exploration of the functional capabilities of defined proteins and peptides in inflammation paralleled the evolving methodologies of protein identification and purification, and molecular biology. In the early1960s, lysozyme was found to have the bacteriolytic activity toward Grampositive bacteria (5). The bactericidal effect of lactoferrin was discovered in 1977 (6). Study of antimicrobial activity in neutrophil lysates of rabbit and guinea pig identified the so-called arginine-rich basic antibacterial proteins that migrated faster than lysozyme toward the cathode (7). In the early 1980s, these arginine-rich basic antimicrobial proteins from rabbit lung macrophages and granulocytes and structurally similar peptides from human granulocytes were purified and named “defensins” (8–10). In the late 1980s, another type of antimicrobial proteins termed histatins were purified (11). The 1990s witnessed the discovery of constitutive βdefensins (12, 13), Paneth cell α-defensins (14–16), inducible β-defensins (17, 18), cathelicidins (19, 20), θ-defensin (21), and many others proteins with antimicrobial activities (22–24). Since then, the genomic organization, regulation of expression, structure, microbicidal activity, and mechanisms of many mammalian antimicrobial proteins have been investigated and reviewed extensively (20, 25–30). Although sporadic reports suggested that certain antimicrobial proteins might have direct effects on host monocytes (31, 32), only in 1996 were human

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIMICROBIAL PROTEINS IN IMMUNITY

P1: IKH

183

neutrophil-derived α-defensins demonstrated to be chemotactic for human T lymphocytes not only in vitro but also in vivo (33). This finding nurtured the investigation of potential chemotactic effects of other antimicrobial proteins and led to the detection of the chemotactic activities of human β-defensins and cathelicidin and identification of receptors that mediate their chemotactic activities on leukocytes (34, 35). Over the past 10 years, mammalian antimicrobial proteins have been reported to have numerous biological activities that promote host defense (28, 36–46). In this review, we discuss defensins in detail and also briefly outline the multiple effects of cathelicidins and eosinophil-derived neurotoxin (EDN) on host immune cells and their resultant roles in innate and adaptive antimicrobial immunity.

DEFENSIN Genes, Expression, and Regulation CLASSIFICATION AND STRUCTURE There are two major subfamilies of defensins in vertebrates (Figures 1A and 1B): α-defensins (47, 48), whose six cysteines are linked 1–6, 2–4, 3–5, and β-defensins (49, 50), linked 1–5, 2–4, 3–6. Because cysteines 5 and 6 are always adjacent to one another, the two subfamilies are structurally very similar. The canonical sequence of α-defensins in humans is x1–2CXCRx2–3Cx3Ex3GxCx3Gx5CCx1–4, where x represents any amino acid residue. Owing to their initial isolation from neutrophil primary (azurophilic) granules, human α-defensin-1 ∼ 4 are conventionally called human neutrophil peptides (HNP1 ∼ 4) (10, 51). Human α-defensin-5 ∼ 6 are products of Paneth cells, often referred to as human defensin-5 (HD5) and -6 (HD6) (15, 16). The canonical sequence for human β-defensins is x2–10Cx5–6(G/A)xCX3–4Cx9–13Cx4–7CCxn. Human β-defensin-1 (HBD1) was initially isolated from human plasma (13), βdefensin-2 (HBD2) was purified from psoriatic scales (18), β-defensin-3 (HBD3) was identified simultaneously from psoriatic scales and by bioinformatics (52–54), and β-defensin-4 (HBD4) was recently identified solely by genomics (55). The pace at which new defensin genes are being discovered has increased dramatically with the advent of genomics-based open-reading frame searches (53–56). The third subfamily, θ-defensins, is derived from two α-defensins prematurely truncated by stop codons present between the third and fourth cysteine residues. The translated products are shortened to nonapeptides, covalently dimerized by disulfide linkages, and cyclized via new peptide bonds between the first and ninth residues (21). The resulting structure (Figure 2C) is a β-hairpin stabilized by three disulfide bonds (57). As only one pseudogene may encode a human θ -defensin, there is no true consensus for sequence (58). A new and growing family of proteins with strong homology to β-defensins is expressed in the testis and epididymis. One of this family, termed HE2/EP2, has at least 6 ∼ 9 splice variants (59). Three of the variants contain β-defensin motifs (54, 59) of which one is bactericidal (60). Another epididymis-specific protein,

20:51

184

AR AR210-IY22-07.tex AR210-IY22-07.sgm

Figure 1 Alignment of human α-defensins (A) and selected β-defensins (B). Experimentally determined signal peptides are shown in white with black highlighting, predicted signal peptides are shown in white with gray highlighting, HNP proregions are shown in black with gray highlighting, and conserved residues are shown in bold. Sequences were selected only for those proteins that have been purified or whose mRNA has been detected. A 46 amino acid insert between the signal peptides and the defensin domains, indentical in EP2C and EP2D, is not shown. Signal sequences were predicted by neural networks using the software SignalP v2.0.

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

12 Feb 2004 LaTeX2e(2002/01/18) P1: IKH

YANG ET AL.

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

185

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIMICROBIAL PROTEINS IN IMMUNITY

P1: IKH

Figure 1 (Continued)

termed ESC42, also contains a β-defensin motif, although it is not clear whether this protein has defensin-like activity (56, 61). The distinct splicing mechanism of HE2/EP2 may indicate convergent evolution from an unrelated gene toward that of β-defensin-like products (60). The backbone of the tertiary structures of monomeric α- and β-defensins (Figure 2A,B) is termed the defensin-like fold, consisting of three antiparallel

12 Feb 2004

20:51

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

186

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

Figure 2 Structures of defensins. (A) HNP3 monomer. (B) HBD2 monomer. (C) RTD-1 from macaque. Disulfides are shown as ball-and-stick models, and the N and C termini are labeled for HNP3 and HBD2. One end of the cyclic molecule RTD-1 is disordered and is not displayed in the figure. The structures were made using the software Ribbons.

β-sheets constrained by three disulfide bridges (48, 50, 62–69). The archetype for the defensin-like fold is HNP3 (PDB: 1DFN) (48). Additionally, the sheet is sometimes flanked by one or two α-helices, as seen in β-defensins (64, 68, 69). In solution, most α- and β-defensins are monomeric, although HNP1, HNP3, and HBD3 form homodimers (48, 68), and HBD2 can even form higher-order oligomers (64). The molecular structures of representative α- and β-defensins are shown in Figures 1 and 2. Mammalian defensins show significant structural similarity to insect defensins (70). The evolution of defensins likely followed the standard model of gene evolution: duplication of the ancestral gene, followed by mutation and natural selection based on the needs of the organism. This has been proposed for both α-defensins (71) and mammalian β-defensins (71–74). Based on sequence homologies, there is a closer relationship between vertebrate β-defensins and insect defensins than between vertebrate α- and β-defensins (71). Furthermore, it is likely that α-defensins are a relatively new set of genes distinct from their progenitors because (a) αdefensins as a group are much more homologous than β-defensins, (b) α-defensins

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

ANTIMICROBIAL PROTEINS IN IMMUNITY

P1: IKH

187

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

exist only in mammals and are expressed in a small range of tissues, and (c) their genes are located in a very tight region of a single chromosome and present in variable nondeterminable multiple copies. GENOMIC STRUCTURE The gene for HNP1, HDEFA1, was first localized to chromosome 8 (75), followed by HNP3 gene, HDEFA1A (76). HDEFA1 and HDEFA1A (Figure 3) are almost identical (hence often referred to as HDEFA1/1A) except for one point mutation in the genome that leads to the conversion of the first amino acid residue from alanine in HNP1 to aspartic acid in HNP3 (Figure 1A). HDEFA1/1A has multiple copies that are inherited unequally, giving rise to individuals with either two or three copies of either HNP1 and/or HNP3 (77). The genes for HNP4 (78), HD5 and HD6 (79), and HBD1 ∼ 2 (74, 80, 81) are all mapped to the same chromosomal location, 8p23 (Figure 3). Analysis of this chromosomal region in detail has turned up genes for HBD3 (53, 54), HBD4 (55), and two new defensins of unknown activity, HBD5 and HBD6 (82). Because the gene for HNP2 has never been found, it is likely that HNP2 is generated from HNP1 and/or HNP3 by proteolytic processing (75–77). The genomic structure of HNP1, 3, and 4 (Figure 3) consists of three exons, with exon I, II, and III encoding the 50 -untranslated region (50 -UTR), the preproregion, and the mature peptide plus the 30 -UTR, respectively (76, 78). The HD5 and

Figure 3 The genomic location/organization, transcription, translation, and posttranslational processing of human α- and β-defensins. The short arm of human chromosome 8 is shown at the top. The intervening sequences between defensin genes are not drawn to scale. The circle on the right indicates the centromere. HDEFA/B, human α/β-defensin gene. UTR, untranslated region.

12 Feb 2004

20:51

188

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HD6 genes (Figure 3) have two exons, with exon I containing the 50 -UTR and preproregion, and exon II containing mature protein and 30 -UTR (79). The genomic structure of HBD1 ∼ 3 is similar to that of HD5 and HD6 (Figure 3), except that a portion of preproregion is present on the second exon (53, 74, 83). HBD4 and HBD6 show the same genomic structure as the β-defensin family, except that the internal intron is much larger (55, 82). The genomic structure of HBD5 is similar to the α-defensin family, except that a preproregion is encoded across the first two exons, rather than on only the second exon (82). EXPRESSING CELLS AND TISSUES The expression pattern of defensins varies greatly among different species; for example, human neutrophils have 4 α-defensins, bovine neutrophils contain 13 β-defensins (BNDB) (12), whereas mouse neutrophils lack defensins (84). In humans, HNP1 ∼ 4 are predominantly synthesized in promyelocytes and stored in the primary granules of neutrophils (10, 51). Recently, HNP1 has also been found to be expressed by NK and T cells (85). HD5 and HD6 are expressed by Paneth cells of intestine, placenta, aminion, chorion, and by nasal and bronchial epithelial cells (15, 16, 86, 87). HBD1 ∼ 6 are expressed by epithelial cells lining tissues facing the external environment, including keratinocytes (18, 52–55, 82, 83, 88, 89). REGULATORY SITES AND CORRESPONDING TRANSCRIPTION FACTORS The production of defensins can be either constitutive or inducible (Table 1). Neutrophilderived (e.g., HNP1 ∼ 4 and BNDB1 ∼ 13), Paneth cell–derived (e.g., HD5 ∼ 6 and mouse cryptidins), and β-defensin-1 are constitutively expressed (10, 12, 13, 15, 16, 51, 89–93). The promoter regions of HDEFA1/1A contain the binding sites for the myeloid factor PU.1 (94) and C/EBP-α (95) that control its constitutive transcription in promyelocytes. The transcription factors responsible for the constitutive expression of Paneth cell–derived defensins and β-defensin-1 are unknown. The expression of constitutive defensins can be modulated; for example, G-CSF can increase expression of HNP-1 ∼ 3 in neutrophils (96). HBD1 has recently been shown to be more highly expressed by peripheral blood leukocytes in response to LPS or dead bacteria (97). Expression of HBD1 is apparently downregulated during Shigella infections (potentially by bacterial DNA), which may contribute to the promotion of bacterial adherence and invasion (98). Most of the epithelial-derived β-defensins are induced by proinflammatory stimuli (17, 18, 83, 90, 99, 100). HBD2 is inducible by bacteria, LPS, TNFα, IL-1, or even L-isoleucine (18, 97, 100, 101) in an NF-κB-dependent manner (83, 101). In addition to multiple NF-κB binding sites, many regulatory elements including the binding sites for AP-1, AP-2, and NF-IL-6 exist in the promoter region of the HBD2 gene (102). The expression of HBD3 is induced in primary keratinocytes and tracheal epithelial cells by TNFα and by heat-inactivated Pseudomonas aeruginosa and Staphylococcus aureus (52). In fetal lung explants and gingival keratinocytes it is induced by IL-1β (54). However, IFNγ , but not TNFα, induces the expression of HBD3 in the cultured keratinocyte cell line HaCaT (53). The expression of HBD4

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANTIMICROBIAL PROTEINS IN IMMUNITY

189

TABLE 1 Cell sources and regulation of mammalian defensins, cathelicidins, and EDN∗

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Defensin Conventional name

Type

Cell source

Synthesis

Release

HNP1 ∼ 4

α

Neutrophils CD8 T cells

Constitutive Inducible

Degranulation Secretion?

HD5 ∼ 6

α

Paneth cells

Constitutive

Degranulation

Mouse cryptidin

α

Paneth cells

Constitutive

Degranulation

HBD1

β

Keratinocytes and epithelial cells

Constitutive and inducible

Secretion

HBD2 ∼ 4

β

Keratinocytes and epithelial cells

Inducible

Secretion

BNDBs

β

Neutrophils

Constitutive

Degranulation

Bovine TAP

β

Epithelial cells

Inducible

Secretion

RTD-1

θ

Neutrophils, monocytes

Constitutive

Degranulation

Cathelicidins

Neutrophils, keratinocytes, epithelial cells, mast cells, and monocytes/macrophages

Constitutive and inducible

Degranulation or secretion

EDN

Eosinophils, neutrophils, macrophages, and placental epithelial cells

Constitutive and inducible

Degranulation or secretion

∗ Abbreviations: EDN, eosinophil-derived neurotoxin; HNP, human neutrophil peptide; HD, human defensin; HBD, human β-defensin; BNDB, bovine neutrophil-derived β-defensin; TAP, tracheal antimicrobial peptide; RTD-1, rhesus theta defensin-1.

in cultured epithelial cells was induced to a limited extent by heat-inactivated P. aeruginosa, and markedly by heat-inactivated Streptococcus pneumoniae or PMA (55). POSTTRANSLATIONAL MODIFICATION Defensins are translated as larger precursor polypeptides consisting of the prepiece, the propiece (often anionic), and the mature peptide (Figures 1 and 3). Posttranslational modifications include the proteolytic removal of the preproregion. The prepiece is a typical signal peptide (highly hydrophobic), and is proteolytically cleaved in the Golgi body. Cleavage of the propiece differs for different defensins. For HNP1 ∼ 4, the propiece is also proteolytically removed in the Golgi body, and the mature peptide is sorted to the primary granules of neutrophils for storage (103, 104). HD5 is stored in granules of Paneth cells and released as the proform (105, 106). Removal of the propiece of HD5 occurs extracellularly and is mediated by trypsin (107). The cleavage of propieces of cryptidins (mouse Paneth cell–derived α-defensins) also occurs extracellularly, but by using another protease, matrilysin (37). The enzyme(s)

12 Feb 2004

20:51

190

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

responsible for the removal of the propiece of β-defensin is unknown, although the mature form is secreted onto the surface and immediate surroundings of epithelial cells. Further processing by removal of individual amino acids from the N terminus of mature peptide has been seen for both α- (107, 108) and β-defensins (88). The variable processing of defensins generates multiple forms differing in N-terminal truncation, a mechanism that may diversify the biological activity of antimicrobial peptides (33, 88).

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Defensins as Effectors and Regulators of Innate Immunity DIRECT ANTIMICROBIAL ACTIVITY Defensins have mutiple host defense functions (Table 2). All defensins characterized so far have direct antimicrobial effect (27, 30, 109). All mammalian defensins, including α-, β-, and θ-subtypes, have the capacity to kill a wide variety of Gram-positive and Gram-negative bacteria, fungi, and some parasites in vitro, especially when examined under low concentrations of salt and plasma proteins (9, 10, 13, 18, 52, 53, 55, 91). The direct antimicrobial activity of defensins is dose-dependent and can usually be observed in micromolar range. The mechanism(s) by which defensins kill microorganisms is not completely understood. The interaction of defensins with target microorganisms results in the destabilization and disruption of their cell membranes, leading to an increase in permeability and leakage of small molecules (110). The polar topology of defensins with spatially separated charged and hydrophobic regions allows them to insert themselves into the microbial cell membrane that contains more negatively charged phospholipids than mammalian cell membrane (111). This interaction is generally believed to cause the formation of multiple pores in the microbial cell membrane (112). Alpha-defensins are able to inactivate certain enveloped viruses (113). HNP1 ∼ 3 have recently been identified as the nonchemokine molecules responsible for the anti-HIV activity produced by CD8 T cells of HIV-nonprogressors (114), a finding in line with an earlier report showing that HNP1 ∼ 3 can inhibit HIV replication (115). Monkey θ-defensin and retrocyclin, an artificial human θ -defensin rescued from the pseudogene by correcting the mutated stop codon, also exibit considerable in vitro anti-HIV activity (58). The anti-HIV activity of α- or θ -defensins is effective for both M- and T-tropic HIV strains, but the mechanistic basis is not completely understood (58, 114). θ -defensin has recently been shown to bind with high affinity to glycoproteins and glycolipids such as gp120, CD4, and galactosylceramide, which may be partly responsible for inhibiting HIV infection of CD4+ T cells (116). The direct microbicidal activity of most defensins [except HBD3 (52) and θ defensin (21)] in vitro is often blunted by physiological concentration of salt (e.g., 150 mM NaCl), increasing concentrations of divalent cations, and serum proteins, although the magnitude of inhibition depends on the target bacteria (88, 100, 110, 117–119). Therefore, it has been speculated that the microbicidal activity of defensins may be restricted to sequestered environments where the salt and serum protein concentrations are low, such as the surface of epithelium or skin and the

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

ANTIMICROBIAL PROTEINS IN IMMUNITY

P1: IKH

191

TABLE 2 Diverse activities of mammalian defensins∗

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Activity relevant to immunity Defensin

Representative

In vitro

In vivo

α-defensin

HNP1 ∼ 3 HD5 Cryptidins

Microbicidal activity Antiviral activity Chemotactic activity Promoting phagocytosis

Recruiting immune cells Inducing cytokines Enhancing Ag-specific humoral and cellular immune responses Conferring resistance to bacterial invasion

Degranulating mast cells Stimulating the production of cytokines and chemokines Regulating complement activation Inhibiting ACTH receptor and glucocorticoid production β-defensin

HBD1 ∼ 4 mBD2 ∼ 3

Microbicidal activity Chemotactic activity using CCR6 Degranulating mast cells Stimulating the production of cytokines and prostaglandins Inducing dendritic cell maturation through TLR4 (only mBD2)

θ-defensin

RTD-1

Microbicidal Antiviral

Enhancing Ag-specific humoral and cellular immune responses Conferring resistance to bacterial invasion

Unknown

∗ Abbreviations: HNP1 ∼ 3, human neutrophil peptide 1 ∼ 3; HD5, human defensin 5; HBD1 ∼ 4, human β-defensin 1 ∼ 4; mBD2 ∼ 3, mouse β-defensin 2 ∼ 3; RTD-1, rhesus theta defensin 1.

phagosomes of neutrophils. However, evidence showing the physiologic relevance of the microbicidal activity of defensins has emerged in recent years (Table 2). In one study where HNP1 ∼ 3 were mixed with bacteria in the presence of lung tissue, defensins demonstrated more potent bactericidal activity (120), suggesting that the antimicrobial activity of defensins is effective under physiologic settings. Using a simple mouse model, HBD2 has been shown to have in vivo antimicrobial activity (121). More convincing evidence has been obtained for some defensins using transgenic or knockout models. For example, inactivation of the gene for matrilysin, an enzyme required for the generation of mature mouse Paneth cell α-defensins, leads to the deficiency of functional Paneth cell α-defensins and an increase in susceptibility of mice to oral challenge with Salmonella typhimurium (37). Knockout of a single defensin, mouse β-defensin-1, results in delayed clearance of Haemophilus influenzae from the lung (40) and increased colonization by Staphylococcus species in the bladder (41). More recently, HD5 has been shown to have antibacterial activity, using transgenic mice overexpressing HD5 gene (43). Compared with cryptidins, HD5 is more potent against murine S. typhimurium.

12 Feb 2004

20:51

192

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Transgenic HD5 mice were fully protected from orally administered murine S. typhimurium, which killed 100% of the wild-type mice (43). CHEMOTACTIC EFFECTS OF DEFENSINS ON PHAGOCYTIC AND MAST CELLS Phagocytes, including neutrophils and monocytes/macrophages, constitute the first line of effector cells of innate antimicrobial host defense after the entry of pathogen(s) into the host. For phagocytes to combat invading pathogens, they need to be recruited to sites of bacterial entry. Leukocyte recruitment occurs along gradients of chemotactic factors, including chemokines and other chemoattractants (122, 123). Many defensins examined to date are leukocyte chemoattractants at nanomolar concentrations (Table 3). In 1989, HNP1 ∼ 3 were reported to be chemotactic for monocytes (31). HBD3 and 4 are also chemotactic for monocytes/macrophages (53, 55, 124, 125), whereas HBD2 chemoattracts mast cells (126), which may promote phagocyte recruitment indirectly. Notably, several mammalian defensins also function as chemoattractants for DCs (discussed later in detail). DCs are also phagocytic and participate in innate antimicrobial immunity (127) in at least two ways. First, DC precursor cells and DCs can directly phagocytize and kill pathogens (127). Second, DCs, particularly upon activation, produce numerous mediators including cytokines, chemokines, and antimicrobial peptides (128) that participate in innate antimicrobial immunity. The capacity of various defensins to chemoattract phagocytes, mast cells, and DCs suggests that they can potentially

TABLE 3 The chemotactic target cells of defensins, cathelicidin, and EDN∗ T lymphocytes

DCs

CD4 Defensin

PMN

MC

Mo/Mφ

CD45RA

CD45RO

CD8

iDCs

mDCs

HNPm & 1

−

nt

−

+

−

+

+

−

HNP2

−

nt

−

+

+

−

HBD1

−

nt

−

−

+

+

+

−

HBD2

−

+

−

−

+

+

+

−

HBD3

−

nt

+

+

+

−

HBD4

nt

nt

+

nt

nt

mBD2 & 3

−

nt

(+)

nt

+

LL-37

+

+

+

+

−

−

PR-39

+

nt

−

−

nt

nt

Bac7

+

nt

+

−

nt

nt

EDN/mEAR2

−

nt

−

−

+

+

∗

−

Abbreviations: HNPm, a mixture of HNP1 ∼ 3; HNP1 or 2, human neutrophil peptide 1 or 2; HBD1, 2, and 3, HBD1, 2, 3, and 4, human β-defensin 1, 2, 3, and 4; mBD2 & 3, mouse β-defensin 2 and 3; EDN/mEAR2, eosinophil-derived neurotoxin/mouse eosinophil-associated ribonuclease 2; PMN, polymorphonuclear leukocytes (neutrophils); MC, mast cells; Mo/Mφ, monocytes/macrophages; DCs, dendritic cells; iDCs or mDCs, immature or mature DCs; nt, not tested.

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIMICROBIAL PROTEINS IN IMMUNITY

P1: IKH

193

contribute either directly or indirectly to the mobilization of host defense in response to pathogen entry. Defensins and chemokines not only overlap in chemotactic activity, but evidence is emerging that they share antimicrobial activity as well. Truncated forms of CXC chemokine CXCL7 with antimicrobial effects were initially isolated from platelet α-granules and named thrombocidins (129, 130). IFNγ -inducible ELR− CXC chemokines (including CXCL9 ∼ 11) were found to have defensin-like bactericidal activity (131). Based on the structural features shared by β-defensins and CCL20 (63, 64, 132), we showed that CCL20 has potent antibacterial activity (132, 133). CCL20 was also found to permeabilize the bacterial cell membrane and to be secreted by epithelial cells of the airway in response to proinflammatory stimuli in large quantity (approximately 170 ng/ml) (134). Screening of 30 human chemokines revealed that about two thirds of human chemokines have antibacterial activity (133). Together with the identification of the antimicriobial activity of CCL28 (135), it appears that many chemokines demonstrate defensin-like antimicrobial activity. INDUCTION OF INFLAMMATORY MEDIATORS Human, rabbit, and guinea pig αdefensins induce mast cell degranulation and histamine release (136, 137). Recently, HBD2 has also been shown to activate rat mast cells, resulting in the release of histamine and prostaglandin D2 (138). HNP1 ∼ 3 can augment the expression and/or production of CXCL8 and CXCL5 by bronchial epithelial cells (139–141). Because mast cell granule products increase neutrophil influx (142, 143) and CXCL8 is a potent neutrophil chemotactic factor (122, 123), defensins are likely to indirectly promote the recruitment of phagocytic neutrophils to inflammatory sites. Degranulation of the recruited neutrophils releases more defensins (33, 144) and consequently generates more CXCL8 and CXCL5, both of which result in a positive feedback loop. Furthermore, HNP1 ∼ 3 have been reported to increase the production of TNFα and IL-1 while decreasing the production of IL10 by monocytes (145). Increased levels of proinflammatory factors (IL-1, TNFα, and histamine) and suppressed levels of IL-10 at the site of microbial infection are likely to amplify local inflammatory responses. Intratracheal instillation of HNP1 ∼ 3 into mice led to the production of TNFα, MCP-1, and MIP-2 (146), providing in vivo support for the capacity of α-defensins to stimulate the production of inflammatory mediators. REGULATION OF THE FUNCTIONS OF PHAGOCYTES AND COMPLEMENT SYSTEM

Once recruited to the site of infection, phagocytes become activated and work with a variety of effector molecules to destroy microbial invaders. Guinea pig αdefensins are reported to induce neutrophil activation, as evidenced by enhanced expression of adhesion molecules including ICAM-1, CD11b, and CD11c by neutrophils (147). Upregulation of adhesion molecules on phagocytes not only facilitates their recruitment, but also promotes their activation, leading to enhanced microbicidal activity (122, 148, 149). In addition, human α-defensins can directly

15 Mar 2004

15:10

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

194

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

activate phagocytes to enhance phagocytosis (150) and to induce the production of reactive oxygen intermediates, which are potent bactericidal molecules (120). Moreover, human α-defensins can bind to complement C1q to either enhance (151) or suppress (152) the activation of the classical pathway of complement in vitro, depending on the experimental conditions. Presumably, α-defensins exhibit similar effects in vivo and therefore participate in regulating the function of complement system. Certain α-defensins have also been reported to inhibit the production of immunosuppressive adrenal steroid hormones, perhaps by blocking the ACTH receptor (153, 154). During systemic infections, α-defensin levels in plasma can reach up to 100 µg/ml, a concentration sufficient to interfere with the production of adrenal glucocorticoids (155, 156). Because glucocorticoids are potent immunosuppressive mediators, α-defensins may also enhance systemic antimicrobial immunity in vivo by inhibiting the production of glucocorticoids. Collectively, defensins are vital contributors to innate host antimicrobial defense. Although data from experiments using knockout and transgenic mice are often interpreted as being a result of the direct antimicrobial activity of defensins, it is more likely that the in vivo contribution of defensins to antimicrobial defense depends on their capabilities to directly kill or inactivate microorganisms; to enhance phagocytosis; to promote the recruitment of phagocytes, mast cells, and DCs; to enhance production of proinflammatory cytokines; to suppress antiinflammatory or immunosuppressive mediators; and to regulate complement activation (Figure 4).

Defensins as Enhancers of Adaptive Immunity INTERACTION OF DEFENSINS WITH G PROTEIN–COUPLED RECEPTOR(S) ON iDCs

The initiation of adaptive antimicrobial immune response begins at the sites of microbial entry where microbial Ags are taken up by immature DCs (iDCs) (1, 3, 157, 158). While processing Ags, iDCs undergo a maturational process in response to endogenous mediators and microbial products to become mature DCs (mDCs) with the capacity to migrate to secondary lymphoid organs and to stimulate Agspecific na¨ıve T cells (3, 158). Thus, recruitment to, and maturation of, DCs at sites of microbial entry are critical for the induction of adaptive antimicrobial immune responses. Indeed, one hallmark of the early phase of infection is the recruitment of iDCs to the inflammatory sites (158–160). Defensins have been found to be direct chemoattractants for iDCs (Table 2). For example, HNP1 ∼ 3 are selectively chemotactic for human iDCs generated from human CD34+ DC precursor cells (161). HBD1 and 2 are also selectively chemotactic for human iDCs, but not mDCs (34). In addition to being chemotactic for monocyte/macrophage (53), HBD3 also attracts iDCs (124). Similar to HBDs, mouse β-defensin 2 (mBD2) and mBD3, but not their proforms, are chemotactic for iDCs, but not mDCs, generated from mouse bone marrow progenitor cells (44). Since defensins are usually released in large quantities in response to inflammatory and microbial stimuli, they may contribute to the recruitment of iDCs to the sites of microbial entry.

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIMICROBIAL PROTEINS IN IMMUNITY

P1: IKH

195

The chemotactic activity of HBD1 ∼ 3 and mBD2 and 3 for iDCs is mediated by the chemokine receptor CCR6 (34, 44), which is also the receptor for the CC chemokine CCL20 (123, 162). The chemotaxis of monocytes by HBD3 must use a receptor other than CCR6 because monocytes do not express functional CCR6 (124, 162). Although the chemotactic receptor(s) for HNP1 ∼ 3 have not yet been identified, they must be a Giα-protein-coupled receptor(s) because their chemotactic activity can be inhibited by pretreatment of target cells with pertussis toxin (33, 161). Despite the lack of sequence homology between β-defensins and CCL20 (13, 18, 52, 53, 163, 164), they each interact with CCR6. The tertiary structures of both HBD2 and CCL20, determined by either X-ray crystallography or nuclear magnetic resonance (64, 69, 132, 165), possess a core folding of three antiparallel β-strands. Thus, BD2 can be structurally considered a simplified form of CCL20 lacking the C-terminal α-helix (69, 165). The Asp4-Leu9 motif in BD2 resembles the Asp5-Leu8 motif of CCL20, which is considered to be responsible for specific interaction with CCR6 (165), providing a structural basis for the capacity of βdefensins and CCL20 to interact with the same receptor (124, 165). STIMULATION OF DC MATURATION BY mBD2 In addition to its chemotactic effect on CCR6-expressing iDCs (44), mBD2 has the unique capacity to promote the phenotypic and functional maturation of murine DCs (42). Treatment of mouse DCs with mBD2 upregulates the expression of costimulatory molecules (CD40, CD80, and CD86), MHC class II, and chemokine receptor CCR7. In addition, mBD2-treated DCs develop the capacity to activate na¨ıve T cells since they induced a greater allogeneic mixed lymphocyte reaction than untreated DCs or DCs treated with promBD2. Furthermore, mBD2-matured DCs exhibit a Th1-polarizing cytokine profile with the production of proinflammatory cytokines IL-12, IL-1α, IL1β, and IL-6 (42). Because activated APCs, particularly mDCs, are responsible not only for the elicitation of Ag-specific immune responses, but also determination of Th1/Th2 polarization of T and B cells (3, 158), the capacity of mBD2, but not mBD3, to induce the production of a Th1-polarizing cytokine profile suggests that different defensins may regulate the type of immune responses (42, 44). The capacity of mBD2 to stimulate DC maturation is mediated by Toll-like receptor 4 (TLR4) based on several lines of evidence (42). Most critically, mBD2 cannot stimulate the maturation of iDCs derived from C3H/HeJ or C57BL10ScNcr mice (both lack functional TLR4). Furthermore, mBD2 can activate NF-κB in HEK293 cells transfected with TLR4 and MD-2, but not in untransfected HEK293 cells. Thus, mBD2, in addition to interacting with CCR6 (44), functions as an endogenous ligand of TLR4 signaling (42). DIRECT EFFECTS ON T LYMPHOCYTES Both α- and β-defensins function as chemoattractants for T lymphocytes (Table 3). HNP-1 and -2 are chemotactic for peripheral blood T cells both in vitro and in vivo, as tested using a human/SCID

12 Feb 2004

20:51

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

196

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

chimeric model (33). Subsequent study indicates that HNP1 ∼ 3 are chemotactic for CD4+/CD45RA+ and CD8+ T cells (161). On the other hand, HBD2 is chemotactic for CD4+/CD45RO+ T cells expressing CCR6 (34). The T cell– chemoattracting effect of defensins suggests that they may contribute to the recruitment of both CD4+ and CD8+ effector T cells to sites of microbial infection. An early report shows that HNP1 ∼ 3, in particular HNP2, can selectively inhibit the catalytic activity of PKC without affecting the binding of this enzyme to Ca2+, its substrate, or phosphatidylserine (a cofactor) (166). HNP1 ∼ 3 are present in the nucleus of human peripheral blood T cells (167). These findings suggest that defensins may profoundly regulate the functions of T lymphocytes. The T-cell chemotactic and microbicidal activities of HNP1 are differentially regulated in the airway (168). The airway epithelial cells secrete or express on their surface arginine-specific ADP-ribosyltransferase-1, which modify HNP1 in vivo by ADP-ribosylating Arg14 of HNP1. The modification greatly decreases its antimicrobial activity, but does not affect the T-cell chemotactic activity of HNP1 (168). ENHANCEMENT OF Ag-SPECIFIC IMMUNE RESPONSE IN VIVO The effects of defensins on DCs and T lymphocytes predicts that both α- and β-defensins may contribute to the induction of adaptive antimicrobial immunity (Table 2). The in vivo immunoenhancing activity of human α-defensins has been demonstrated in several studies. The earliest study showed that HNP1 ∼ 3 could enhance Ag-specific immune responses to simultaneous intranasal administration of ovalbumin (OVA) into C57BL/6 mice. Compared to immunization with OVA alone, coadministration of HNP1 ∼ 3 enhanced the systemic production of OVA-specific serum IgG antibody and the generation of IFNγ , IL-5, IL-6, and IL-10 by OVA-specific CD4+ T cells, providing the first evidence that α-defensins can promote both humoral and cellular adaptive immune responses (36). The immunoenhancing activity of α-defensins was further demonstrated in a study in which intraperitoneal injection of HNP1 ∼ 3 together with KLH or B-cell lymphoma idiotype Ag into mice was shown not only to augment the serum levels of Ag-specific IgG, but also to enhance the resistance of immunized mice to tumor challenge (169). A recent study of the immunoenhancing effect of individual HNP, indicated that the in vivo immunoenhancing role varies among defensins (170). The capacity of β-defensins to enhance Ag-specific immune responses has been demonstrated by the use of a DNA vaccine approach (42, 44). When mBD2 and mBD3 were fused with B-cell lymphoma epitope sFv38 (lymphoma-specific VH and VL fragments, an idiotypic Ag) and used as DNA vaccines to immunize mice, not only did the mice generate potent humoral immune responses against otherwise nonimmunogenic sFv38, but they also developed antitumor immunity (42, 44). Since the induction of antitumor protection in this model relies on the generation of tumor-specific cellular immunity (171), it seems that β-defensins can promote both humoral and cellular Ag-specific immune responses. In line with the fact that mBD2 and mBD3 share the same receptor with mCCL20, fusion

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIMICROBIAL PROTEINS IN IMMUNITY

P1: IKH

197

of mCCL20 with sFv38 also induced Ag-specific humoral and cellular immune responses (44). Using a similar approach with a different Ag, mBD2 was also demonstrated to induce both antibody and CTL responses to gp120, an envelope protein of HIV (172). In these studies, the immunoenhancing activity of the fusion construct of mDB2, mBD3, and mCCL20 depends on their interaction with CCR6 because the fusion constructs of promBD2 or mutated mCCL20 and target Ags, which were not able to interact with CCR6 to induce chemotaxis, failed to enhance immune responses to target Ags (42, 44, 172). In addition, fusion of mBD2, mBD3, or mCCL20 with target Ags was critical because immunization of the mice with a mixture of unfused target Ags and chemotactic ligands (either β-defensins or mCCL20) was not effective in inducing Ag-specific immune responses (44, 172). These characteristics are also true for the fusion constructs of other chemokine ligands such as MCP-3 (44, 171, 172). Therefore, besides promoting the recruitment of iDCs (or other Ag-presenting cells), another mechanism for chemotactic ligands such as β-defensins to promote Ag-specific immune responses may be the promotion of Ag uptake and processing following chemokine receptor-mediated internalization of the fused target Ags. When another even weaker B cell-lymphoma idiotypic Ag, sFv20, was used as the antigenic moiety in the fusion constructs, only mBD2, but not mBD3, could induce protective antitumor immunity (42, 44). The difference between mBD2 and mBD3 in this case may be due to the fact that only mBD2, but not mBD3, can induce phenotypic and functional maturation of DCs (42). The importance of DC maturation in the induction of adaptive immune responses has recently been demonstrated by a similar observation that linkage of Ag with anti-DEC205 mAb to facilitate efficient antigen uptake and processing by iDCs induces efficient Ag-specific immune responses only if a maturational signal is provided (173, 174). Moreover, IL-6 production by mBD2-treated DCs may also contribute to the induction of anti-sFv20 (42) as IL-6 contributes to the activation of adaptive immune response by overcoming the suppressive effect of CD4+ CD25+ regulatory T cells (175). Very recently, HBD1 and HBD2 were also investigated using the intranasal immunization model (36). In comparison with mice immunized with only OVA, mice immunized with a mixture of OVA and HBD1 or HBD2 generated significantly higher OVA-specific serum IgG antibody, confirming that human βdefensins have in vivo immunoenhancing activity (170). How can α- and β-defensins promote in vivo adaptive immunity to microbial Ags in the context of infection? Several scenarios can be envisaged (Figure 4). First, defensins potentially promote Ag uptake and processing by DCs. This can be achieved at least in two ways: the chemotactic recruitment of iDCs to sites of infection, and consequent facilitation of microbial Ag delivery to iDCs. Following the killing of microorganisms, defensins may still bind to a portion of antigenic microbial cell membrane to form “dead microbe-defensin” complexes. Such “defensin-Ag” complexes may be subject to defensin receptor-mediated internalization, and thus deliver microbial Ag(s) to iDCs more efficiently. In addition, defensins promote Ag presentation by inducing DC maturation directly or indirectly.

12 Feb 2004

20:51

198

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Although only mBD2 has been found to directly induce DC maturation so far, the possibility that additional defensin(s) with such activity will be identified is high given the recent realization that over 30 β-defensin genes are present in both humans and mice (56). Defensins induce production of TNFα and IL-1 by monocytes/macrophages (145), which are inducers of DC maturation (3, 158), and should contribute indirectly to DC maturation during microbial infection. Furthermore, defensins may also facilitate the chemotactic recruitment of memory CD4+ and/or CD8+ T cells to infected tissue, thus contributing to the effector phase of adaptive antimicrobial immune responses.

CATHELICIDIN Cathelicidin represents another major group of mammalian antimicrobial proteins found in many mammalian species (20, 25–27, 176). All cathelicidin proteins (Figure 5) contain an N-terminal putative signal peptide (preregion), a conserved cathelin-like domain (proregion), and a C-terminal microbicidal domain (hence the name cathelicidin) (20). About 35 cathelicidin members have been identified in various mammalian species; humans and mice, however, generate only one cathelicidin, called hCAP18 (19, 177–179) and CRAMP (180), respectively.

Gene, Expression, and Regulation Cathelicidin genes (Figure 5) consist of four exons (45, 181–183). The first three exons encode the N-terminal preproregion consisting of a putative signal peptide

Figure 5 The genomic organization, expression, and processing of cathelicidins and their activities. The last exon is not drawn to scale as it varies greatly among different cathelicidins.

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIMICROBIAL PROTEINS IN IMMUNITY

P1: IKH

199

(29 ∼ 30 a.a.) and the cathelin-like domain (94 ∼ 114 a.a.). The last exon encodes the cleavage site and the C-terminal antimicrobial domain, which varies remarkably in size (12 ∼ 97 a.a.). Cathelicidins are usually stored in the granules of neutrophils as an inactive proform (cathelin-like domain + antimicrobial domain) and undergo processing to mature peptide during or after secretion by appropriate proteases. For example, human cathelicidin/hCAP18 is cleaved by elastase (184) or proteinase 3 (185) to liberate its C-terminal antimicrobial domain, which is called “LL-37” because this peptide begins with two leucine residues and has 37 amino acid residues (19, 177–179, 184). In cattle and pigs, the processing of cathelicidins seems to be mediated exclusively by elastase (186, 187). The C-terminal antimicrobial peptides of cathelicidins are also markedly variable in structure. Most of this group of antimicrobial peptides have an α-helical conformation (e.g., hCAP18/LL37, rabbit CAP18, mouse CRAMP, etc.) (184, 188, 189). Some cathelicidins (e.g., porcine PR-39, bactenecin 5, etc.) are proline/arginine-rich, showing an extended α-helical structure like polyproline (190, 191). Pig protegrins, on the other hand, have β-sheet structures (192, 193). Cathelicidins are constitutively expressed in a variety of tissues such as bone marrow, thymus, liver, spleen, testis, stomach, and intestine in a developmentally regulated manner (180, 194, 195). At the cellular level, cathelicidins are most abundant in granules of neutrophils of various species (20, 176, 178); however, they are also expressed by other cell types either constitutively or in an inducible manner (176, 177, 196). For example, hCAP18/LL-37 is expressed by various epithelial cells and keratinocytes (177, 196, 197), monocytes, NK cells, B cells, and γ δT cells (85), and is induced in keratinocytes in response to inflammatory stimuli (188, 194) or injury (196). Very recently, cathelicidins have been found to be present in both mouse and human mast cells, and the expression can also be upregulated by bacterial components such as LPS or lipoteichoic acid (198). Several potential regulatory motifs have been found at the promoter regions of cathelicidin genes, such as the consensus binding sites for NF-κB, NF-IL-6, acute phase response factor, and IFNγ response element (182, 184, 188, 194, 199). The expression of LL-37 by colon epithelial cells is upregulated by short chain fatty acids in an Erk (extracellular signal regulated protein kinase)–dependent manner (200).

Roles of Cathelicidins in Host Antimicrobial Immunity DIRECT ANTIMICROBIAL ACTIVITY The N-terminal prepiece of cathelicidins is a classical signal peptide. The cathelin-like domain has long been considered to neutralize the cationic antimicrobial domain so that the whole molecule is maintained in an inactive form during intracellular transport and storage to avoid potential intracellular cytotoxicity (20, 176). However, a recent study found that the cathelin-like domain not only inhibits the protease activity of cathepsin L, a cysteine protease, but also demonstrates potent antibacterial activity (201). The Cterminal antimicrobial peptides of cathelicidins are microbicidal against a broad spectrum of microorganisms, including bacteria, fungi, and parasites, with a wide overlap in specificity, but they exhibit significant differences in potency from one

12 Feb 2004

20:51

200

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

another (20, 176). Similar to defensins, the mechanism of cathelicidin-mediated microbial killing depends on the formation of ion channels or pores in the microbial cell membrane (176). CHEMOTACTIC ACTIVITY Several cathelicidins are chemotactic for various leukocytes. ProBac7, a bovine cathelicidin, is chemotactic for monocytes/macrophages (125). PR-39, one member of the porcine cathelicidins, can induce chemotaxis of pig neutrophils (202). Recently, LL-37 has been found to be chemotactic for neutrophils, monocytes (35, 85), and mast cells (203). Cathelicidins generated at the sites of microbial entry form chemotactic gradients, which are likely to result in the recruitment of phagocytes. In addition to inducing chemotaxis, PR-39 can mobilize intracellular Ca2+ in pig neutrophils (202), whereas LL-37 can induce Ca2+ mobilization in phagocytes (35), suggesting that cathelicidins are endogenous activators of phagocytes. The capacity of LL-37 to induce Ca2+ mobilization in monocytes can be cross-desensitized by ligands specific for human formyl peptide receptor-like 1. This finding led to the identification of FPRL1 as a receptor for LL-37 (35). INDUCTION OF EXPRESSION OR RELEASE OF INFLAMMATORY MEDIATORS LL-37 enhances the expression of a variety of genes by macrophages (204). Of particular interest is the upregulation of the chemokines CXCL8 and CCL2 and their corresponding receptors, CXCR2 and CCR2, by LL-37 (204). CXCR2 is expressed by neutrophils, monocytes, and T cells, whereas CCR2 is expressed by monocytes and iDCs (122, 123, 158). Therefore, LL-37 may indirectly facilitate the recruitment of phagocytes, iDCs, and T cells to inflammatory sites through the induction of chemokines and their corresponding receptors. Furthermore, LL-37 can degranulate mast cells, leading to the release of proinflammatory mediators such as histamine and prostaglandins (138). The capacity of LL-37 to chemoattract human peripheral blood T cells (35, 85) indicates that it can participate in the recruitment of effector T cells to sites of microbial infection, thereby contributing to adaptive antimicrobial immunity. NEUTRALIZATION OF BACTERIAL ENDOTOXIN Killing of bacteria by antimicrobial peptides, phagocytes, and complement system result in the release of bacterial components, such as endotoxin from Gram-negative bacteria and lipotechoic acid from Gram-positive bacteria. These bacterial components, if allowed to enter the circulation, have a detrimental outcome such as septic shock by inducing the production of high levels of systemic proinflammatory cytokines, including TNFα, IL-1β, and IL-6. Many of the cathelicidins, including LL-37, SMAP-29, Bac 2A-NH2 (a linear form of Bac 2A), and indolicidin, bind LPS with high affinity and neutralize its biological activities (19, 28, 179, 189, 204). In addition, the activity of lipotechoic acid can also be inhibited by cathelicidins (204). This property has pathophysiological relevance because both rabbit CAP18106–142, intact or truncated LL-37 (a 27-residue LPS-binding fragment corresponding to hCAP18109–135), can protect

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

ANTIMICROBIAL PROTEINS IN IMMUNITY

P1: IKH

201

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

galactosamine-sensitized mice from LPS-mediated lethality (19, 204, 205). This protection has also been demonstrated by overexpression of LL-37 in mice through adenovirus-mediated gene transfer (206). Therefore, cathelicidins can serve as buffers of LPS and LTA, reducing the severity of host inflammatory responses to bacteria that can lead to lethal conditions such as septic shock, and this activity of cathelicidins can be considered one aspect of host innate antimicrobial defense. ANGIOGENESIS AND WOUND HEALING Certain cathelicidins have angiogenic effect. PR-39 can enhance angiogenesis in vitro and stimulate the formation of functional vasculature in vivo (207). This effect is thought to be mediated by inhibiting the ubiquitin-proteasome-dependent degradation of hypoxia-inducible factor-1α (207). LL-37 also stimulates angiogenesis (208). Similar to its chemotactic activity, the angiogenic activity of LL-37 is mediated by FPRL1 expressed on endothelial cells (208). PR-39 can induce the production of syndecan-1 and -4, major components of cell coat (209). LL-37 can stimulate re-epithelialization of skin wounds, and the wound repair can be inhibited by LL-37-specific neutralizing antibody (210). Thus, cathelicidins are potentially important for wound healing. IN VIVO ROLES OF CATHELICIDINS The participation of LL-37 in host antimicrobial defense has been demonstrated by in vivo studies showing that adenoviral vector–targeted systemic overexpression of LL-37 in mice results in decreased bacterial load and mortality of experimental mice following challenge with P. aeruginosa or E. coli (38, 206). Moreover, knockout of Cnlp, the mouse gene encoding CRAMP, results in increased susceptibility of the mice to infection by Group A Streptococcus (39). Furthermore, mice deficient for CRAMP show greatly decreased vascularization during skin wound repair (208).

EDN EDN, one of the four eosinophil-granule-derived cationic proteins, belongs to the ribonuclease superfamily (211). EDN was purified to homogeneity in 1981 as a protein of 18.4 kDa (212). Its cDNA, cloned in 1989, has an open reading frame encoding a 134-amino acid mature polypeptide with 4 intrachain disulfide bridges and a 27-residue amino-terminal hydrophobic leader sequence (213). Its threedimensional structure, solved by X-ray crystallography, shows a typical RNase folding, which is a V-shaped α + β-type polypeptide with the active-site cleft in the middle (214). The secondary structural elements consist of six β-strands and four α-helices. The tertiary structure is organized into two lobes, each consisting of three antiparallel β-strands and one α-helix, and with two α-helices located between the two lobes. The gene for EDN localizes to 14q24–q31 and contains two exons and one small intron (230 bases) in the 50 untranslated region, with the second exon encoding the whole mature protein (215). Although EDN is a single-copy gene, another form containing four additional amino acids (Ser-LeuHis-Val) on the N terminus originally considered as part of the signal sequence

12 Feb 2004

20:51

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

202

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

has also been identified, suggesting alternative splicing or different posttranscriptional modification (216, 217). The transcription of EDN gene is controlled by elements located at the intron and proximal promoter, including binding sites for AP-1, NF-ATp, PU.1, and C/EBP (218–220). EDN is stored in eosinophil granules or expressed by liver, spleen, neutrophils, and activated monocytes/macrophages (211, 221, 222). EDN is not antibacterial and its role(s) in host defense is unknown. In the late 1990s, EDN was found to have in vitro antiviral activity, in particular against respiratory syncytial virus (22), which thus marked it as one member of host gene-coded antimicrobial proteins. Recently, EDN has also been shown to be responsible, in part, for the HIV-1-inhibitory activities in the supernatant of allogeneic mixed lymphocyte reaction (223). EDN has recently been found to selectively chemoattract both human and mouse DCs (224). Mouse eosinophil-associated RNase 2 (mEAR2), one of a cluster of divergent orthologs of human EDN (211, 225), was also chemotactic for human and mouse DCs (224). In agreement with its DC chemotactic activity, EDN induced the activation of p42/44 Erk in DCs (224). Furthermore, injection of mEAR2 into the air pouches of mice resulted in the recruitment of DCs into the air pouches, indicating that EDN/mEAR2 acts as a DC chemoattractant in vivo (224). Most recently, EDN was found to be able to activate human dendritic in vitro, leading to the production of a variety of cytokines, chemokines, growth factor, and soluble receptors (D. Yang, Q. Fu, Q. Chen, D.L. Newton, H.F. Rosenberg, et al., submitted). Most of the mediators induced by EDN are proinflammatory, such as IL-6, TNFα, and many chemokines. Furthermore, EDN also induced the maturation of dendritic cells. Therefore, EDN can act as potent activators for human dendritic cells. Given the critical roles of DC recruitment, maturation, and IL-6 in the induction of Ag-specific immune response (3, 157, 158, 175), EDN is therefore another host gene-coded antimicrobial protein that potentially contributes to the induction and/or regulation of both innate and adaptive antimicrobial immunity.

CONCLUDING REMARKS Based on in vitro and animal model studies, it has become evident that many mammalian antimicrobial proteins (e.g., defensins, cathelicidins, EDN, etc.) have, in addition to their well-recognized direct antimicrobial activities, additional roles in initiating and amplifying host innate and adaptive immune responses against microbial invasion. However, their contribution to human antimicrobial immunity needs to be further explored. One study of the expression of human antimicrobial proteins by morbus Kostmann patients with normal level of neutrophils (as a result of G-CSF treatment) has shown that HNP1 ∼ 3 are greatly reduced and LL-37 is absent in patients’ plasma and saliva (226). The abnormal expression of α-defensins and LL-37 correlates with the occurrence of periodontal infectious disease in patients with morbus Kostmann, providing the strongest evidence to date

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIMICROBIAL PROTEINS IN IMMUNITY

P1: IKH

203

that defensins and cathelicidins are important in human antimicrobial immunity (226). New antimicrobial proteins keep emerging. Recent genomic surveys have revealed defensin-like genes in five separate syntenic clusters on four chromosomes (56, 82). The antimicrobial and immunomodulatory activities of these antimicrobial peptides have not yet been evaluated. Most of the known immunomodulatory effects of antimicrobial proteins overlap with other humoral components such as cytokines and chemotactic factors of the immune system. Additional studies are needed to distinguish their relative contribution to host antimicrobial immunity from that of other mediators using appropriate knockout or transgenic models. The receptor(s) utilized by antimicrobial proteins (e.g., EDN) to signal gene activation in target cells remains to be identified. Attempts to engineer antimicrobial peptides or proteins into clinically applicable antibiotics have been widespread (227). Such therapeutic effort may be complicated by their immunomodulatory activities, and any engineered products need to be evaluated to avoid potential undesirable immunomodulatory effects. Conversely, antimicrobial proteins or their modified forms have the potential to be developed into therapeutic agents with both antibiotic and immunoenhancing activities. ACKNOWLEDGMENTS This project has been funded in part with Federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. NO1-CO-12,400. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Hoffmann JA, Kafatos FC, Janeway CA Jr, Ezekowitz RAB. 1999. Phylogenetic perspectives in innate immunity. Science 284:1313–18 2. Medzhitov R, Janeway CA Jr. 1997. Innate immunity: the virtues of a nonclonal system of recognition. Cell 91:295–98 3. Banchereau J, Steinman RM. 1998. Dendritic cells and the control of immunity. Nature 392:245–51 4. Hirsch JG. 1956. Phagocytin: a bactericidal substance from polymorphonu-

clear leukocytes. J. Exp. Med. 103:589– 621 5. Glynn AA, Milne CM. 1965. Lysozyme and immune bacteriolysis. Nature 207: 1309–10 6. Arnold RR, Cole MF, McGhee A. 1977. Bactericidal effect for human lactoferrin. Science 197:263–65 7. Zeya HI, Spitznagel JK. 1963. Antibacterial and enzymic basic proteins from leukocyte lysosomes: separation and identification. Science 142:1085–87

12 Feb 2004

20:51

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

204

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

8. Lehrer RI, Selsted ME, Szklarek D, Fleischmann J. 1983. Antibacterial activity of microbicidal cationic proteins 1 and 2, natural peptide antibiotics of rabbit lung macrophages. Infect. Immun. 42:10– 14 9. Selsted ME, Szklarek D, Lehrer RI. 1984. Purification and antibacterial activities of antimicrobial peptides of rabbit granulocytes. Infect. Immun. 45:150–54 10. Ganz T, Selsted ME, Szklarek D, Harwig SSL, Daher K, et al. 1985. Defensins: Natural peptide antibiotics of human neutrophils. J. Clin. Invest. 76:1427–35 11. Oppenheim FG, Xu T, McMillian FM, Levitz SM, Diamond RD, et al. 1988. Histatins, a novel family of histidine-rich proteins in human parotid secretion. Isolation, characterization, primary structure, and fungistatic effects on Candida albicans. J. Biol. Chem. 263:7472–77 12. Selsted ME, Tang Y-Q, Morris WL, McGuire PA, Novotny MJ, et al. 1993. Purification, primary structures, and antimicrobial activities of β-defensins, a new family of antimicrobial peptides from bovine neutrophils. J. Biol. Chem. 268:6641–48 13. Bensch KW, Raida M, Magert H-J, Schulz-Knappe P, Forssmann W-G. 1995. hBD-1: a novel β-defensin from human plasma. FEBS Lett. 368:331–35 14. Ouellette AJ, Lualdi JC. 1990. A novel mouse gene family coding for cationic, cysteine-rich peptides. Regulation in small intestine and cells of myeloid origin. J. Biol. Chem. 265:9831–37 15. Jones DE, Bevins CL. 1992. Paneth cell of the human small intestine express an antimicrobial peptide gene. J. Biol. Chem. 267:23216–25 16. Jones DE, Bevins CL. 1993. Defensin-6 mRNA in human Paneth cells: implications for antimicrobial peptides in host defense of the human bowel. FEBS Lett. 315:187–92 17. Schonwetter BS, Stolzenberg ED, Zasloff MA. 1995. Epithelial antibiotics in-

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

duced at sites of inflammation. Science 276:1645–48 Harder J, Bartels J, Christophers E, Schr¨oder JM. 1997. A peptide antibiotic from human skin. Nature 387:861 Larrick JW, Hirata M, Zheng H, Zhong J, Bolin D, et al. 1994. A novel granulocyte-derived peptide with lipopolysaccharide-neutralizing activity. J. Immunol. 152:231–40 Zanetti M, Gennaro R, Romeo D. 1995. Cathelicidins: a novel protein family with a common proregion and a variable C-terminal antimicrobial domain. FEBS Lett. 374:1–5 Tang Y-Q, Yuan J, Osapay G, Osapay K, Tran D, et al. 1999. A cyclic antimicrobial peptide produced in primate leukocytes by the ligation of two truncated α-defensins. Science 286:498–502 Domachowske JB, Dyer KD, Bonville CA, Rosenberg HF. 1998. Recombinant human eosinophil-derived neurotoxin/RNase 2 functions as an effective antiviral agent against respiratory syncytial virus. J. Infect. Dis. 177:1458–64 Bellamy W, Takase M, Yamauchi K, Wakabayashi H, Kawase K, Tomita M. 1992. Identification of the bactericidal domain of lactoferrin. Biochim. Biophys. Acta 1121:130–36 Stenger S, Hanson DA, Teitelbaum R, Dewan P, Niazi KR, et al. 1998. An antimicrobial activity of cytolytic T cells mediated by granulysin. Science 282:121– 25 Boman HG. 1998. Gene-encoded peptide antibiotics and the concept of innate immunity: an update review. Scand. J. Immunol. 48:15–25 Lehrer RI, Ganz T. 1999. Antimicrobial peptides in mammalian and insect host defense. Curr. Opin. Immunol. 11:23–27 Ganz T, Lehrer RI. 1998. Antimicrobial peptides of vertebrates. Curr. Opin. Immunol. 10:41–44 Scott MG, Hancock REW. 2002. Cationic antimicrobial peptides and their

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

ANTIMICROBIAL PROTEINS IN IMMUNITY

29.

30.

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

31.

32.

33.

34.

35.

36.

37.

38.

multifunctional role in the immune system. Crit. Rev. Immunol. 20:407–31 Zasloff M. 2002. Antimicrobial peptides of multicellular organisms. Nature 415:389–95 Schr¨oder JM. 1999. Epithelial antimicrobial peptides: innate local host response elements. Cell. Mol. Life Sci. 56:32–46 Territo MC, Ganz T, Selsted ME, Lehrer R. 1989. Monocyte-chemotactic activity of defensins from human neutrophils. J. Clin. Invest. 84:2017–20 Pereira HA, Shafer WM, Pohl J, Martin LE, Spitznagel JK. 1990. CAP37, a human neutrophil-derived chemotactic factor with monocyte specific activity. J. Clin. Invest. 85:1468–76 Chertov O, Michiel DF, Xu L, Wang JM, Tani K, et al. 1996. Identification of defensin-1, defensin-2, and CAP37/azurocidin as T-cell chemoattractant proteins released from interleukin8-stimulated neutrophils. J. Biol. Chem. 271:2935–40 Yang D, Chertov O, Bykovskaia SN, Chen Q, Buffo MJ, et al. 1999. β-Defensins: Linking innate and adaptive immunity through dendritic and T cell CCR6. Science 286:525–28 Yang D, Chen Q, Schmidt AP, Anderson GM, Wang JM, et al. 2000. LL-37, the neutrophil granule- and epithelial cellderived cathelicidin, utilizes formyl peptide receptor-like 1 (FPRL1) as a receptor to chemoattract human peripheral blood neutrophils, monocytes, and T cells. J. Exp. Med. 192:1069–74 Lillard JW Jr, Boyaka PN, Chertov O, Oppenheim JJ, McGhee JR. 1999. Mechanisms for induction of acquired host immunity by neutrophil peptide defensins. Proc. Natl. Acad. Sci. USA 96:651–56 Wilson CL, Ouellette AJ, Satchell DP, Ayabe T, Lopez-Boado YS, et al. 1999. Regulation of intestinal α-defensin activation by the metalloproteinase matrilysin in innate host defense. Science 286:113–17 Bals R, Weiner DJ, Meegalla RL,

39.

40.

41.

42.

43.

44.

45.

46.

47.

P1: IKH

205

Wilson JM. 1999. Transfer of a cathelicidin peptide antibiotic gene restores bacterial killing in a cystic fibrosis xenograft model. J. Clin. Invest. 103:1113–17 Nizet V, Ohtake T, Lauth X, Trowbridge J, Rudisill J, et al. 2001. Innate antimicrobial peptide protects the skin from invasive bacterial infection. Nature 414:454– 57 Moser C, Weiner DJ, Lysenko E, Bals R, Weiser JN, Wilson JM. 2002. β-Defensin 1 contributes to pulmonary innate immunity in mice. Infect. Immun. 70:3068–72 Morrison G, Kilanowski F, Davidson D, Dorin J. 2002. Characterization of the mouse β defensin 1, Defb1, mutant mouse model. Infect. Immun. 70:3053–60 Biragyn A, Ruffini PA, Leifer CA, Klyushnenkova E, Shakhov A, et al. 2002. Toll-like receptor 4-dependent activation of dendritic cells by β-defensin 2. Science 298:1025–29 Salzman NH, Ghosh D, Huttner KM, Paterson Y, Bevins CL. 2003. Protection against enteric salmonellosis in transgenic mice expressing a human intestinal defensin. Nature 422:522–26 Biragyn A, Surenhu M, Yang D, Ruffini PA, Haines BA, et al. 2001. Mediators of innate immunity that target immature, but not mature, dendritic cells induce antitumor immunity when genetically fused with nonimmunogenic tumor antigens. J. Immunol. 167:6644–53 Gudmundsson GH, Agerberth B. 1999. Neutrophil antibacterial peptides, multifunctional effector molecules in the mammalian immune systen. J. Immunol. Methods 232:45–54 Yang D, Chertov O, Oppenheim JJ. 2001. The role of mammalian antimicrobial peptides and proteins in awakening of innate host defenses and adaptive immunity. Cell. Mol. Life Sci. 58:978–89 Selsted ME, Harwig SSL. 1989. Determination of the disulfide array in human defensin HNP-2. A covalently cyclized peptide. J. Biol. Chem. 264:4003–7

12 Feb 2004

20:51

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

206

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

48. Hill CP, Yee J, Selsted ME, Eisenberg D. 1991. Crystal structure of defensin HNP-3, an amphiphilic dimer: mechanisms of membrane permeabilization. Science 251:1481–85 49. Tang Y-Q, Selsted ME. 1993. Characterization of the disulfide motif in BNBD12, an antimicrobial β-defensin peptide from bovine neutrophils. J. Biol. Chem. 268:6649–53 50. Zimmermann GR, Legault P, Selsted ME, Pardi A. 1995. Solution structure of bovine neutrophil β-defensin-12: the peptide fold of the β-defensins is identical to that of the classical defensins. Biochemistry 34:13663–71 51. Wilde CG, Griffith JE, Marra MN, Snable JL, Scott RW. 1989. Purification and characterization of human neutrophil peptide 4, a novel member of the defensin family. J. Biol. Chem. 264:11200–3 52. Harder J, Bartels J, Christophers E, Schr¨oder JM. 2001. Isolation and characterization of human β-defensin-3, a novel human inducible peptide antibiotics. J. Biol. Chem. 276:5707–13 53. Garcia JRC, Jaumann F, Schulz S, Krause A, Rodriguez-Jimenez J, et al. 2001. Identification of a novel, multifunctional β-defensin (human β-defensin 3) with specific antimicrobial activity: its interaction with plasma membranes of Xenopus oocytes and the induction of macrophage chemoattraction. Cell Tissue Res. 306:257–64 54. Jia HP, Schutte BC, Schudy A, Linzmeier R, Guthmiller JM, et al. 2001. Discovery of new human β-defensins using a genomics-based approach. Gene 263:211–18 55. Garcia JRC, Krause A, Schulz S, Rodriguez-Jimenez FJ, Kluver E, et al. 2001. Human β-defensin 4: a novel inducible peptide with a specific saltsensitive spectrum of antimicrobial activity. FASEB J. 15:1819–21 56. Schutte BC, Mitros JP, Bartlett JA, Walters JD, Jia HP, et al. 2002. Discovery

57.

58.

59.

60.

61.

62.

63.

64.

65.

of five conserved β-defensin gene clusters using a computational search strategy. Proc. Natl. Acad. Sci. USA 99:2129–33 Trabi M, Schirra HJ, Craik DJ. 2001. Three-dimensional structure of RTD-1, a cyclic antimicrobial defensin from Rhesus macaque leukocytes. Biochemistry 40:4211–21 Cole AM, Hong T, Boo LM, Nguyen T, Zhao C, et al. 2002. Retrocyclin: a primate peptide that protects cells from infection by T- and M-tropic strains of HIV-1. Proc. Natl. Acad. Sci. USA 99:1813–18 Frohlich O, Po C, Young LG. 2001. Organization of the human gene encoding the epididymis-specific EP2 protein variants and its relationship to defensin genes. Biol. Reprod. 64:1072–79 von Horsten HH, Derr P, Kirchhoff C. 2002. Novel antimicrobial peptide of human epididymal duct origin. Biol. Reprod. 67:804–13 Liu Q, Hamil KG, Sivashanmugam P, Grossman G, Soundararajan R, et al. 2001. Primate epididymis-specific proteins: characterization of ESC42, a novel protein containing a trefoil-like motif in monkey and human. Endocrinology 142:4529–39 Pardi A, Zhang X-L, Selsted ME, Skalicky JJ, Yip PF. 1992. NMR studies of defensin antimicrobial peptides: 2. Threedimensional structures of rabbit NP-2 and human HNP-1. Biochemistry 31:11357– 64 Hoover DM, Chertov O, Lubkowski J. 2001. The structure of human β-defensin1: new insight into structural properties of β-defensins. J. Biol. Chem. 276:39021– 26 Hoover DM, Rajashankar KR, Blumenthal R, Puri A, Oppenheim JJ, et al. 2000. The structure of human β-defensin-2 shows evidence of higher order oligomerization. J. Biol. Chem. 275:32911–18 Zhang X-L, Selsted ME, Pardi A. 1992. NMR studies of defensin antimicrobial peptides: 1. Resonance assignment and

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

ANTIMICROBIAL PROTEINS IN IMMUNITY

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

66.

67.

68.

69.

70.

71.

72.

73.

74.

secondary structure determination of rabbit NP-2 and human HNP-1. Biochemistry 31:11348–56 Bauer F, Schweimer K, Kluver E, ConejoGarcia JR, Forssmann WG, et al. 2001. Structure determination of human and murine β-defensins reveals structural conservation in the absence of significant sequence similarity. Protein Sci. 10:2470– 79 McManus AM, Dawson NF, Wade JD, Carrington LE, Winzor DJ, Craik DJ. 2000. Three-dimentional structure of RK-1: a novel α-defensin peptide. Biochemistry 39:15757–64 Schibli DJ, Hunter HN, Aseyev V, Starner TD, Wiencek JM, et al. 2002. The solution structures of the human β-defensins lead to a better understanding of the potent bactericidal activity of HBD3 against Staphylococcus aureus. J. Biol. Chem. 277:8279–89 Sawai MV, Jia HP, Liu L, Aseyev V, Wiencek JM, et al. 2001. The NMR structure of human β-defensin-2 reveals a novel α-helical segment. Biochemistry 40:3810–16 Bulet P, Hetru C, Dimarcq J-L, Hoffmann D. 1999. Antimicrobial peptides in insects: structure and function. Dev. Comp. Immunol. 23:329–44 Hughes AL. 1999. Evolutionary diversification of the mammalian defensins. Cell. Mol. Life Sci. 56:94–103 Morrison GM, Semple CAM, Kilanowski FM, Hill RE, Dorin JR. 2003. Signal sequence conservation and mature peptide divergence within subgroups of the murine β-defensin gene family. Mol. Biol. Evol. 20:460–70 Semple CAM, Rolfe M, Dorin JR. 2003. Duplication and selection in the evolution of primate β-defensin genes. Genome Biol. 4:R31 Liu L, Zhao C, Heng HHQ, Ganz T. 1997. The human β-defensin-1 and α-defensins are encoded by adjacent genes: two peptide families with differing

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

P1: IKH

207

disulfide topology share a common ancestry. Genomics 43:316–26 Sparkes RS, Kronenberg M, Heinzmann C, Daher KA, Klisak I, et al. 1989. Assignment of defensin gene(s) to human chromosome 8p23. Genomics 5:240–44 Linzmeier R, Michaelson D, Liu L, Ganz T. 1993. The structure of neutrophil defensin genes. FEBS Lett. 321:267–73 Mars WM, Patmasiriwat P, Maity T, Huff V, Weil MM, Womack GF. 1995. Inheritance of unequal numbers of the genes that encode the human neutrophil defensins HP-1 and HP-3. J. Biol. Chem. 270:30371–76 Palfree RG, Sadro LC, Solomon S. 1993. The gene encoding the human corticostatin HP-4 precursor contains a recent 86base duplication and is located on chromosome 8. Mol. Endocrinol. 7:199–205 Bevins CL, Jones DE, Dutra A, Schaffzin J, Muenke M. 1996. Human enteric defensin genes: chromosomal map position and a model for possible evolutionary relationships. Genomics 31:95–106 Harder J, Siebert R, Zhang Y, Matthiesen P, Christophers E, et al. 1997. Mapping of the gene encoding human β-defensin2 (DEFB2) to chromosome region 8p22p23.1. Genomics 46:472–75 Linzmeier R, Ho CH, Hoang BV, Ganz T. 1999. A 450-kb contig of defensin genes on human chromasome 8p23. Gene 233:205–11 Yamaguchi Y, Nagase T, Mikita R, Fukuhara S, Tomita T, et al. 2002. Identification of multiple novel epididymisspecific β-defensin isoforms in humans and mice. J. Immunol. 169:2516–23 Liu L, Wang L, Jia HP, Zhao C, Heng HHQ, et al. 1998. Structure and mapping of the human β-defensin HBD-2 gene and its expression at sites of inflammation. Gene 222:237–44 Eisenhauer PB, Lehrer RI. 1992. Mouse neutrophils lack defensins. Infect. Immun. 60:3446–47 Agerberth B, Charo J, Werr J, Olsson B,

12 Feb 2004

20:51

208

86.

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

87.

88.

89.

90.

91.

92.

93.

94.

95.

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL. Idali F, et al. 2000. The human antimicrobial and chemotactic peptides LL-37 and α-defensins are expressed by specific lymphocyte and monocyte populations. Blood 96:3086–93 Svinarich DM, Wolf NA, Gomez R, Gonik B, Romero R. 1997. Detection of human defensin 5 in reproductive tissues. Am. J. Obstet. Gynecol. 176:470–75 Frye M, Bargon J, Dauletbaev N, Weber A, Wagner TOF, Gropp R. 2000. Expression of human α-defensin 5 (HD5) mRNA in nasal and bronchial epithelial cells. J. Clin. Pathol. 53:770–73 Valore EV, Park CH, Quayle AJ, Wiles KR, McCray PB Jr, Ganz T. 1998. Human β-defensin-1: an antimicrobial peptide of urogenital tissues. J. Clin. Invest. 101:1633–42 Zhao C, Wang I, Lehrer RI. 1996. Widespread expression of β-defensin hBD-1 in human secretory glands and epithelial cells. FEBS Lett. 396:319–22 Kaiser V, Diamond G. 2000. Expression of mammalian defensin genes. J. Leukoc. Biol. 68:779–84 Ouellette AJ, Selsted ME. 1996. Paneth cell defensins: endogenous peptide components of intestinal host defense. FASEB J. 10:1280–89 Huttner KM, Kozak CA, Bevins CL. 1997. The mouse genome encodes a single homolog of the antimicrobial peptide human β-defensin 1. FEBS Lett. 413:45– 49 Zhang G, Wu H, Shi J, Ganz T, Ross CR, Blecha F. 1998. Molecular cloning and tissue expression of porcine β-defensin-1. FEBS Lett. 424:37–40 Ma Y, Su Q, Tempst P. 1998. Differentiation-stimulated activity binds an ETS-like, essential regulatory element in the human promyelocytic defensin-1 promoter. J. Biol. Chem. 273:8727– 40 Tsutsumi-Ishii Y, Hasebe T, Nagaoka I. 2000. Role of CCAAT/enhancer-binding protein site in transcription of human neu-

96.

97.

98.

99.

100.

101.

102.

103.

104.

105.

trophil peptide-1 and -3 defensin genes. J. Immunol. 164:3264–73 Ashitani J, Nakazato M, Mukae H, Taniguchi H, Date Y, Matsukura S. 2000. Recombinant granulocyte colonystimulating factor induces production of human neutrophil peptides in lung cancer patients with neutropenia. Regul. Pept. 95:87–92 Fang X-M, Shu Q, Chen Q-X, Book M, Sahl H-G, et al. 2003. Differential expression of α- and β-defensins in human peripheral blood. Eur. J. Clin. Invest. 33:82– 87 Islam D, Bandholtz L, Nilsson J, Wigzell H, Christensson B, et al. 2001. Downregulation of bactericidal peptides in enteric infections: a novel immune escape mechanism with bacterial DNA as a potential regulator. Nat. Med. 7:180–85 Diamond G, Russell JP, Bevins CL. 1996. Inducible expression of an antibiotic peptide gene in lipopolysaccharidechallenged tracheal epithelial cells. Proc. Natl. Acad. Sci. USA 93:5156–60 Singh PK, Jia HP, Wiles K, Hesselberth J, Liu L, et al. 1998. Production of β-defensins by human airway epithelia. Proc. Natl. Acad. Sci. USA 95:14961–66 Fehlbaum P, Rao M, Zasloff M, Anderson GM. 2000. An essential amino acid induces epithelial β-defensin expression. Proc. Natl. Acad. Sci. USA 97:12723–28 Harder J, Meyer-Hoffert U, Teran LM, Schwichtenberg L, Bartels J, et al. 2000. Mucoid Pseudomonas aeruginosa, TNFα, IL-1β, but not IL-6, induce human β-defensin-2 in respiratory epithelia. Am. J. Respir. Cell Mol. Biol. 22:714–21 Valore EV, Ganz T. 1992. Posttranslational processing of defensins in immature human myeloid cells. Blood 79:1538–44 Harwig SS, Park AS, Lehrer RI. 1992. Characterization of defensin precursors in mature human neutrophils. Blood 79:1532–37 Porter EM, Liu L, Oren A, Anton PA, Ganz T. 1997. Localization of human

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

ANTIMICROBIAL PROTEINS IN IMMUNITY

106.

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

107.

108.

109.

110.

111.

112.

113.

114.

intestinal defensin 5 in Paneth cell granules. Infect. Immun. 65:2389–95 Cunliffe RN, Rose FR, Keyte J, Abberley L, Chan WC, Mahida YR. 2001. Human defensin 5 is stored in precursor form in normal Paneth cells is expressed by some villous epithelial cells and by metaplastic Paneth cells in the colon in inflammatory bowel disease. Gut 48:176–85 Ghosh D, Porter E, Shen B, Lee SK, Wilk D, et al. 2002. Paneth cell trypsin is the processing enzyme for human defensin5. Nat. Immunol. 3:583–90 Bateman A, Singh A, Shustik C, Mars WM, Solomon S. 1991. The isolation identification of multiple forms of the neutrophil granule peptides from human leukemic cells. J. Biol. Chem. 266:7524– 30 Lehrer RI, Ganz T. 2002. Defensins of vertebrate animals. Curr. Opin. Immunol. 14:96–102 Lehrer RI, Barton A, Daher KA, Harwig SS, Ganz T, Selsted ME. 1989. Interaction of human defensins with Escherichia coli: mechanism of bactericidal activity. J. Clin. Invest. 84:553–61 Lohner K, Latal A, Lehrer RI, Ganz T. 1997. Differential scanning microcalorimetry indicates that human defensin, HNP-2, interacts specifically with biomembrane mimetic systems. Biochemistry 36:1525–31 Wimley WC, Selsted ME, White SH. 1994. Interactions between human defensins and lipid bilayers: evidence for formation of multimeric pores. Protein Sci. 3:1362–73 Lehrer RI, Daher K, Ganz T, Selsted ME. 1985. Direct inactivation of viruses by MCP-1 and MCP-2, natural peptide antibiotics from rabbit leukocytes. J. Virol. 54:467–72 Zhang L, Yu W, He T, Yu J, Caffrey RE, et al. 2002. Contribution of human α-defensin 1, 2, and 3 to the anti-HIV-1 activity of CD8 antiviral factor. Science 298:995–1000

P1: IKH

209

115. Nakashima H, Yamamoto N, Masuda M, Fujii N. 1993. Defensins inhibit HIV replication in vitro. AIDS 7:1129 116. Wang W, Cole AM, Hong T, Waring AJ, Lehrer RI. 2003. Retrocyclin, an antiretroviral θ-defensin, is a lectin. J. Immunol. 170:4708–16 117. Bals R, Wang XR, Wu ZR, Freeman T, Bafna V, et al. 1998. Human β-defensin 2 is a salt-sensitive peptide antibiotic expressed in human lung. J. Clin. Invest. 102:874–80 118. Goldman MJ, Anderson GM, Stolzenberg ED, Kari UP, Zasloff M, Wilson JM. 1997. Human β-defensin-1 is a salt-sensitive antibiotic in lung that is inactivated in cystic fibrosis. Cell 88:553–60 119. Porter EM, van Dam E, Valore EV, Ganz T. 1997. Broad-spectrum antimicrobial activity of human intestinal defensin 5. Infect. Immun. 65:2396–401 120. Porro GA, Lee JH, de Azavedo J, Crandall I, Whitehead T, et al. 2001. Direct and indirect bacterial killing functions of neutrophil defensins in lung explants. Am. J. Physiol. Lung Cell Mol. Physiol. 281:L1240–47 121. Huang GT, Zhang HB, Kim D, Liu L, Ganz T. 2002. A model for antimicrobial gene therapy: demonstration of human β-defensin 2 antimicrobial activities in vivo. Hum. Gene Ther. 20:2017–25 122. Baggiolini M. 1998. Chemokines and leukocyte traffic. Nature 392:565–68 123. Zlotnik A, Yoshie O. 2000. Chemokines: a new classification system and their role in immunity. Immunity 12:121–27 124. Yang D, Biragyn A, Kwak LW, Oppenheim JJ. 2002. Mammalian defensins in immunity: more than just microbicidal. Trends Immunol. 23:291–96 125. Verbanac D, Zanetti M, Romeo D. 1993. Chemotactic and protease-inhibiting activities of antibiotic peptide precursors. FEBS Lett. 371:255–58 126. Niyonsaba F, Iwabuchi K, Matsuda H, Ogawa H, Nagaoka I. 2002. Epithelial cell-derived human β-defensin-2 acts as

12 Feb 2004

20:51

210

127.

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

128.

129.

130.

131.

132.

133.

134.

135.

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL. a chemotaxin for mast cells through a pertussis toxin-sensitive and phospholipase C-dependent pathway. Int. Immunol. 14:421–26 Liu YJ. 2001. Dendritic cell subsets and lineages, and their functions in innate and adaptive immunity. Cell 106:259–62 Duits LA, Ravensbergen B, Rademaker M, Hiemstra PS, Nibbering PH. 2002. Expression of β-defensin 1 and 2 mRNA by human monocytes, macrophages and dendritic cells. Immunology 106:517–25 Krijgsveld J, Zaat SA, Meeldijk J, van Veelen PA, Fang G, et al. 2000. Thrombocidins, microbicidal proteins from human blood platelets, are C-terminal deletion products of CXC chemokines. J. Biol. Chem. 275:20374–81 Tang Y-Q, Yeaman MR, Selsted ME. 2002. Antimicrobial peptides from human platelets. Infect. Immun. 70:6524–33 Cole AM, Ganz T, Liese AM, Burkick MD, Liu L, Strieter RM. 2001. Cutting Edge: IFN-inducible ELR− CXC chemokines display defensin-like antimicrobial activity. J. Immunol. 167:623–27 Hoover DM, Boulegue C, Yang D, Oppenheim JJ, Tucker KD, et al. 2002. The structure of human MIP-3α/CCL20: Linking antimicrobial and CCR6 receptor binding activities with human β-defensins. J. Biol. Chem. 277:37647–54 Yang D, Chen Q, Hoover DM, Staley P, Tucker KD, et al. 2003. Many chemokines including CCL20/MIP-3α display antimicrobial activity. J. Leukoc. Biol. 74:448– 55 Starner TD, Barker CK, Jia HP, Kang Y, McCray PB. 2003. CCL20 is an inducible product of human airway epithelia with innate immune properties. Am. J. Respir. Cell Mol. Biol. 10.1165/rcmb.2002-0272OC Hieshima K, Ohtani H, Shibano M, Izawa D, Nakayama T, et al. 2003. CCL28 has dual roles in mucosal immunity as a chemokine with broad-spectrum antimicrobial activity. J. Immunol. 170:1452–61

136. Yamashita T, Saito K. 1989. Purification, primary structure, and biological activity of guinea pig neutrophil cationic peptides. Infect. Immun. 57:2405–9 137. Befus AD, Mowat C, Gilchrist M, Hu J, Solomon S, Bateman A. 1999. Neutrophil defensins induce histamine secretion from mast cells: mechanisms of action. J. Immunol. 163:947–53 138. Niyonsaba F, Someya A, Hirata M, Ogawa H, Nagaoka I. 2001. Evaluation of the effects of peptide antibiotics human β-defensin-1/2 and LL-37 on histamine release and prostaglandin D2 production from mast cells. Eur. J. Immunol. 31:1066–75 139. van Wetering S, Mannesse-Lazeroms SPG, van Sterkenburg MAJA, Daha MR, Dijkman JH, Hiemstra PS. 1997. Effect of defensins on interleukin-8 synthesis in airway epithelial cells. Am. J. Physiol. Lung Cell Mol. Physiol. 272:L888–96 140. van Wetering S, Mannesse-Lazeroms SPG, Dijkman JH, Hiemstra PS. 1997. Effect of neutrophil serine proteinases and defensins on lung epithelial cells: modulation of cytotoxicity and IL-8 production. J. Leukoc. Biol. 62:217–26 141. van Wetering S, Mannesse-Lazeroms SPG, van Sterkenburg MAJA, Hiemstra PS. 2002. Neutrophil defensins stimulate the release of cytokines by airway epithelial cells: modulation by dexamethasone. Inflamm. Res. 51:8–15 142. Echtenacher B, Mannel DN, Hultner L. 1996. Critical protective role of mast cells in a model of acute septic peritonitis. Nature 381:75–77 143. Malaviya R, Ikeda T, Ross E, Abraham SN. 1996. Mast cell modulation of neutrophil influx and bacterial clearance at sites of infection through TNFα. Nature 381:77–80 144. Ganz T. 1987. Extracellular release of antimicrobial defensins by human polymorphonuclear leukocytes. Infect. Immun. 55:568–71 145. Chaly YV, Paleolog EM, Kolesnikova TS,

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

ANTIMICROBIAL PROTEINS IN IMMUNITY

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

146.

147.

148.

149.

150.

151.

152.

153.

154.

Tikhonov II, Petratchenko EV, Voitenok NN. 2000. Neutrophil α-defensin human neutrophil peptide modulates cytokine production in human monocytes and adhesion molecule expression in endothelial cells. Eur. Cytokine Netw. 11:257–66 Zhang H, Porro G, Orzech N, Mullen B, Liu MY, Slutsky AS. 2001. Neutrophil defensins mediate acute inflammatory response and lung dysfunction in doserelated fashion. Am. J. Physiol. Lung Cell Mol. Physiol. 280:L947–54 Yomogida S, Nagaoka I, Saito K, Yamashita E. 1996. Evaluation of the effects of defensins on neutrophil functions. Inflamm. Res. 45:62–67 Patarroyo M. 1994. Adhesion molecules mediating recruitment of monocytes to inflamed tissue. Immunobiology 191:474– 77 Hayflick JS, Kilgannon P, Gallatin WM. 1998. The intercellular adhesion molecule (ICAM) family of proteins: new members and novel functions. Immunol. Res. 17:313–27 Ichinose M, Asai M, Imai K, Sawada M. 1996. Enhancement of phagocytosis by corticostatin I (CSI) in cultured mouse peritoneal macrophages. Immunopharmacol 35:103–9 Prohaszka Z, Nemet K, Csermely P, Hudecs F, Mezo G, Fust G. 1997. Defensins purified from human granulocytes bind C1q and activate the classical complement pathway like the transmembrane glycoprotein gp41 of HIV-1. Mol. Immunol. 34:809–16 van den Berg RH, Faber-Krol MC, van Wetering S, Hiemstra PS, Daha MR. 1998. Inhibition of activation of classical pathway of complement by human neutrophil defensins. Blood 92:3898–903 Zhu Q, Hu J, Mulay S, Esch F, Shimasaki S, Solomon S. 1988. Isolation and structure of corticostatin peptides from rabbit fetal and adult lung. Proc. Natl. Acad. Sci. USA 85:592–96 Tominaga T, Fukata J, Naito Y, Funakoshi

155.

156.

157.

158.

159.

160.

161.

162.

163.

P1: IKH

211

S, Fujii N, Imura H. 1990. Effects of corticostatin-I on rat adrenal cells in vitro. J. Endocrinol. 125:287–92 Panyutich AV, Panyutich EA, Krapivin VA, Baturevich EA, Ganz T. 1993. Plasma defensin concentrations are elevated in patients with septicemia or bacterial meningitis. J. Lab. Clin. Med. 122:202– 7 Shiomi K, Nakazato M, Ihi T, Kangawa K, Matsuo H, Matsukura S. 1993. Establishment of radioimmunoassay for human neutrophil peptides and their increases in plasma and neutrophil in infection. Biochem. Biophys. Res. Commun. 195:1336–44 Medzhitov R, Janeway CA Jr. 1997. Innate immunity: impact on the adaptive immune response. Curr. Opin. Immunol. 9:4–9 Palucka K, Banchereau J. 2002. How dendritic cells and microbes interact to elicit or subvert protective immune responses. Curr. Opin. Immunol. 14:420–31 McWilliam AS, Nelson DJ, Thomas JA, Holt PG. 1994. Rapid dendritic cell recruitment is a hallmark of the acute inflammatory response at mucosal surfaces. J. Exp. Med. 179:1331–36 McWilliam AS, Napoli S, Marsh AM, Pemper FL, Nelson DJ, et al. 1996. Dendritic cells are recruited into the airway epithelium during the inflammatory response to a broad spectrum of stimuli. J. Exp. Med. 184:2429–32 Yang D, Chen Q, Chertov O, Oppenheim JJ. 2000. Human neutrophil defensins selectively chemoattract na¨ıve T and immature dendritic cells. J. Leukoc. Biol. 68:9– 14 Yang D, Howard OMZ, Chen Q, Oppenheim JJ. 1999. Cutting Edge: Immature dendritic cells generated from monocytes in the presence of TGF-β1 express functional C-C chemokine receptor 6. J. Immunol. 163:1737–41 Power CA, Church DJ, Meyer A, Alouani S, Proudfoot AEI, et al. 1997. Cloning

12 Feb 2004

20:51

212

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

164.

165.

166.

167.

168.

169.

170.

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL. and characterization of a specific receptor for the novel CC chemokine MIP-3α from lung dendritic cells. J. Exp. Med. 186:825–35 Hieshima K, Imai T, Opdenakker G, Van Damme J, Kusuda J, et al. 1997. Molecular cloning of a novel human CC chemokine liver and activation-regulated chemokine (LARC) expressed in liver: chemotactic activity for lymphocytes and gene localization on chromosome 2. J. Biol. Chem. 272:5846–53 P´erez-Ca˜nadillas JM, Zaballos A, Guti´errez J, Varona R, Roncal F, et al. 2001. NMR solution structure of murine CCL20/MIP-3α, a chemokine that specifically chemoattracts immature dendritic cells and lymphocytes through its highly specific interaction with the β-chemokine receptor CCR6. J. Biol. Chem. 276:28372–79 Charp PA, Rice WG, Raynor RL, Reimund E, Kinkade JM Jr, et al. 1988. Inhibition of protein kinase C by defensins, antibiotic peptides from human neutrophils. Biochem. Pharmacol. 37:951–56 Blomqvist M, Bergquist J, Westman A, Hakansson K, Hakansson P, et al. 1999. Identification of defensins in human lymphocyte nuclei. Eur. J. Biochem. 263:312– 18 Paone G, Wada A, Stevens LA, Matin A, Hirayama T, et al. 2002. ADP ribosylation of human neutrophil peptide-1 regulates its biological properties. Proc. Natl. Acad. Sci. USA 99:8231–35 Tani K, Murphy WJ, Chertov O, Salcedo R, Koh CY, et al. 2000. Defensins act as potent adjuvants that promote cellular and humoral immune responses in mice to a lymphoma idiotype and carrier antigens. Int. Immunol. 12:691–700 Brogden KA, Heidari M, Sacco RE, Palmquist D, Guthmiller JM, et al. 2003. Defensin-induced adaptive immunity in mice and its potential in preventing periodontal disease. Oral Microbiol. Immunol. 18:95–99

171. Biragyn A, Tani K, Grimm MC, Weeks S, Kwak LW. 1999. Genetic fusion of chemokines to a self tumor antigen induces protective, T-cell dependent antitumor immunity. Nat. Biotechnol. 17:253– 58 172. Biragyn A, Belyakov IM, Chow Y-H, Dimitrov DS, Berzofsky JA, Kwak LW. 2002. DNA vaccines encoding human immunodeficiency virus-1 glycoprotein 120 fusions with proinflammatory chemoattractants induce systemic and mucosal immune responses. Blood 100:1153– 59 173. Hawiger D, Inaba K, Dorsett Y, Guo M, Mahnke K, et al. 2001. Dendritic cells induce peripheral T cell unresponsiveness under steady state conditions in vivo. J. Exp. Med. 194:769–80 174. Bonifaz L, Bonnyay D, Mahnke K, Rivera M, Nussenzweig MC, Steinman RM. 2002. Efficient targeting of protein antigen to the dendritic cell receptor DEC-205 in the steady state leads to antigen presentation on major histocompatibility complex class I products and peripheral CD8+ T cell tolerance. J. Exp. Med. 196:1627– 38 175. Pasare C, Medzhitov R. 2003. Toll pathway-dependent blockade of CD4+ CD25+ T cell-mediated suppression by dendritic cells. Science 299:1033–36 176. Ramanathan B, Davis EG, Ross CR, Blecha F. 2002. Cathelicidins: microbicidal activity, mechanisms of action, and roles in innate immunity. Microbes Infect. 4:361–72 177. Agerberth B, Gunne H, Odeberg J, Kogner P, Boman HG, Gudmundsson GH. 1995. FALL-39, a putative human peptide antibiotic, is cysteine-free and expressed in bone marrow and testis. Proc. Natl. Acad. Sci. USA 92:195–99 178. Cowland JB, Johnsen AH, Borregaad N. 1995. hCAP-18, a cathelin/probactenecin-like protein of human neutrophil specific granules. FEBS Lett. 368:173–76

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIMICROBIAL PROTEINS IN IMMUNITY 179. Larrick JW, Hirata M, Balint RF, Lee J, Zhong J, Wright SC. 1995. Human CAP18: a novel antimicrobial lipopolysaccharide-binding protein. Infect. Immun. 63:1291–97 180. Gallo RL, Kim KJ, Bernfield M, Kozak CA, Zanetti M, et al. 1997. Identification of CRAMP, a cathelin-related antimicrobial peptide expressed in the embryonic and adult mouse. J. Biol. Chem. 272:13088–93 181. Gudmundsson GH, Magnusson KP, Chowdhary BP, Johansson M, Andersson L, Boman HG. 1995. Structure of the gene for porcine peptide antibiotic PR-39, a cathelin gene family member: comparative mapping of the locus for the human peptide antibiotic FALL-39. Proc. Natl. Acad. Sci. USA 92:7085–89 182. Zhao C, Ganz T, Lehrer RI. 1995. Structures of genes for two cathelin-associated antimicrobial peptides: prophenin-2 and PR-39. FEBS Lett. 376:130–34 183. Scocchi M, Wang S, Zanetti M. 1997. Structural organization of the bovine cathelicidin gene family and identification of a novel member. FEBS Lett. 417:311– 15 184. Gudmundsson GH, Agerberth B, Odeberg J, Bergman T, Olsson B, Salcedo R. 1996. The human gene FALL-39 and processing of the cathelin precursor to the antibacterial peptide LL-37 in granulocytes. Eur. J. Biochem. 238:325–32 185. Sorensen OE, Follin P, Johnsen AH, Calafat J, Tjabringa GS, et al. 2001. Human cathelicidin, hCAP-18, is processed to the antimicrobial peptide LL-37 by extracellular cleavage with proteinase 3. Blood 97:3951–59 186. Shi J, Ganz T. 1998. The role of protegrins and other elastase-activated polypeptides in the bactericidal properties of porcine inflammatory fluids. Infect. Immun. 66:3611–17 187. Zanetti M, Litteri L, Griffiths G, Gennaro R, Romeo D. 1991. Stimulus-induced maturation of probactenecins, precursors

188.

189.

190.

191.

192.

193.

194.

195.

P1: IKH

213

of neutrophil antimicrobial polypeptides. J. Immunol. 146:4295–300 Frohm M, Agerberth B, Ahangari G, Stahle-Backdahl M, Liden S, et al. 1997. The expression of the gene coding for the antimicrobial peptide LL-37 is induced in human keratinocytes during inflammatory disorders. J. Biol. Chem. 272:15258–63 Hirata M, Shimomura Y, Yoshida M, Morgan JG, Palings I, et al. 1994. Characterization of a rabbit cationic protein (CAP18) with lipopolysaccharide-inhibitory activity. Infect. Immun. 62: 1421–26 Cabiaux V, Agerberth B, Johansson J, Homble F, Goormaghtigh E, Ruysschaert JM. 1994. Secondary structure and membrane interaction of PR-39, a Pro+Argrich antibacterial peptide. Eur. J. Biochem. 224:1019–27 Shi J, Ross CR, Leto TL, Blecha F. 1996. PR-39, a proline-rich antibacterial peptide that inhibits phagocyte NADPH oxidase activity by binding to Src homology 3 domains of p47 phox. Proc. Natl. Acad. Sci. USA 93:6014–18 Aumelas A, Mangoni M, Roumestand C, Chiche L, Despaux E, et al. 1996. Synthesis and solution structure of the antimicrobial peptide protegrin-1. Eur. J. Biochem. 237:575–83 Fahrner RL, Dieckmann T, Harwig SS, Lehrer RI, Eisenberg D, Feigon J. 1996. Solution structure of protegrin1, a broad-spectrum antimicrobial peptide from porcine leukocytes. Chem. Biol. 3:543–50 Bals R, Wang X, Zasloff M, Wilson JM. 1998. The peptide antibiotic LL37/hCAP-18 is expressed in epithelia of the human lung where it has broad antimicrobial activity at the airway surface. Proc. Natl. Acad. Sci. USA 95:9541– 46 Wu H, Zhang G, Ross CR, Blecha F. 1999. Cathelicidin gene expression in porcine tissues: roles in ontogeny and tissue specificity. Infect. Immun. 67:439–42

12 Feb 2004

20:51

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

214

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

P1: IKH

YANG ET AL.

196. Dorschner RA, Pestonjamasp VK, Tamakuwala S, Ohtake T, Rudisill J, et al. 2001. Cutaneous injury induces the release of cathelicidin anti-microbial peptides active against group A Streptococcus. J. Invest. Dermatol. 117:91–97 197. Nilsson MF, Sandstedt B, Sorensen O, Weber G, Borregaad N, Stahle-Backdahl M. 1999. The human cationic antimicrobial protein (hCAP18), a peptide antibiotic, is widely expressed in human squamous epithelia and colocalizes with interleukin-6. Infect. Immun. 67:2561–66 198. Di Nardo A, Vitiello A, Gallo RL. 2003. Cutting Edge: mast cell antimicrobial activity is mediated by expression of cathelicidin antimicrobial peptide. J. Immunol. 170:2274–78 199. Wu H, Zhang G, Minton JE, Ross CR, Blecha F. 2000. Regulation of cathelicidin gene expression: induction by lipopolysaccharide, interleukin6, retinoic acid, and Salmonella enterica serovar typhimurium infection. Infect. Immun. 68:5552–58 200. Schauber J, Svanholm C, Termen S, Iffland K, Menzel T, et al. 2003. Expression of the cathelicidin LL-37 is modulated by short chain fatty acids in colonocytes: relevance of signalling pathways. Gut 52:735–41 201. Zaiou M, Nizet V, Gallo RL. 2003. Antimicrobial protease inhibitory functions of the human cathelicidin (hCAP18/LL37) prosequence. J. Invest. Dermatol. 120:810–16 202. Huang HJ, Ross CR, Blecha F. 1997. Chemoattractant properties of PR-39, a neutrophil antibacterial peptide. J. Leukoc. Biol. 61:624–29 203. Niyonsaba F, Iwabuchi K, Someya A, Hirata M, Matsuda H, et al. 2002. A cathelicidin family of human antibacterial peptide LL-37 induces mast cell chemotaxis. Immunology 106:20–26 204. Scott MG, Davidson DJ, Gold MR, Bowdish D, Hancock REW. 2002. The human antimicrobial peptide LL-37 is a mul-

205.

206.

207.

208.

209.

210.

211.

212.

213.

tifunctional modulator of innate immune response. J. Immunol. 169:3883–91 Kirikae T, Hirata M, Yamasu H, Kirikae F, Tamura H, et al. 1998. Protective effects of a human 18-kilodalton cationic antimicrobial protein (CAP18)-derived peptide against murine endotoxemia. Infect. Immun. 66:1861–68 Bals R, Weiner DJ, Moscioni AD, Meegalla RL, Wilson JM. 1999. Augmentation of innate host defense by expression of a cathelicidin antimicrobial peptide. Infect. Immun. 67:6084–89 Li J, Post M, Volk R, Gao Y, Li M, et al. 2000. PR39, a peptide regulator of angiogenesis. Nat. Med. 6:49–55 Koczulla R, von Degenfeld G, Kupatt C, Krotz F, Zahler S, et al. 2003. An angiogenic role for the human peptide antibiotic LL-37/hCAP-18. J. Clin. Invest. 111:1665–72 Gallo RL, Ono M, Povsic T, Page C, Eriksson E, et al. 1994. Syndecans, cell surface heparan sulfate proteoglycans, are induced by a proline-rich antimicrobial peptide from wounds. Proc. Natl. Acad. Sci. USA 91:11035–39 Heilborn JD, Nilsson MF, Kratz G, Weber G, Sorensen O, et al. 2003. The cathelicidin anti-microbial peptide LL37 is involved in re-epithelialization of human skin wounds and is lacking in chronic ulcer epithelium. J. Invest. Dermatol. 120:379–89 Rosenberg HF, Domachowke JB. 2001. Eosinophils, eosinophil ribonucleases, and their role in host defense against respiratory virus pathogens. J. Leukoc. Biol. 70:691–98 Durack DT, Ackerman SJ, Loegering DA, Gleich GJ. 1981. Purification of human eosinophil-derived neurotoxin. Proc. Natl. Acad. Sci. USA 78:5165–69 Rosenberg HF, Tenen DG, Ackerman SJ. 1989. Molecular cloning of human eosinophil-derived neurotoxin: a member of the ribonuclease gene family. Proc. Natl. Acad. Sci. USA 86:4460–64

12 Feb 2004

20:51

AR

AR210-IY22-07.tex

AR210-IY22-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTIMICROBIAL PROTEINS IN IMMUNITY 214. Mosimann SC, Newton DL, Youle RJ, James MN. 1996. X-ray crystallographic structure of recombinant eosinophilderived neurotoxin at 1.83 A resolution. J. Mol. Biol. 26:540–52 215. Hamann KJ, Ten RM, Loegering DA, Jenkins RB, Heise MT, et al. 1990. Structure and chromosome localization of the human eosinophil-derived neurotoxin and eosinophil cationic protein genes: evidence for intronless coding sequences in the ribonuclease gene superfamily. Genomics 7:535–46 216. Sakakibara R, Hashida K, Tominaga N, Sakai K, Ishiguro M, et al. 1991. A putative mouse oocyte maturation inhibitory protein from urine of pregnant women: Nterminal sequence homology with human nonsecretory ribonuclease. Chem. Pharm. Bull. 39:146–49 217. Newton DL, Rybak SM. 1998. Unique recombinant human ribonuclease and inhibition of Kaposi’s sarcoma cell growth. J. Natl. Cancer Inst. 90:1787–91 218. Tiffany HL, Handen JS, Rosenberg HF. 1996. Enhanced expression of the eosinophil-derived neurotoxin ribonuclease (RNS2) gene requires interaction between the promoter and intron. J. Biol. Chem. 271:12387–93 219. van Dijk TB, Caldenhoven E, Raaijmakers JAM, Lammers J-WJ, Koenderman L, de Groot RP. 1998. The role of transcription factor PU.I in the activity of the intronic enhancer of the eosinophil-derived neurotoxin (RNS2) gene. Blood 91:2126– 32 220. Baltus B, Buitenhuis M, van Dijk T, Vinson C, Raaijmakers J, et al. 1999. C/EBP regulates the promoter of the eosinophil-

221.

222.

223.

224.

225.

226.

227.

P1: IKH

215

derived neurotoxin/RNS2 gene in human eosinophilic cells. J. Leukoc. Biol. 66:683–88 Sur S, Glitz DG, Kita H, Kujawa SM, Peterson EA, et al. 1998. Localization of eosinophil-derived neurotoxin and eosinophil cationic protein in neutrophilic leukocytes. J. Leukoc. Biol. 63:715–22 Egesten A, Dyer KD, Batten D, Domachowske JB, Rosenberg HF. 1997. Ribonucleases and host defense: identification, localization and gene expression in adherent monocytes in vitro. Biochim. Biophys. Acta 1358:255–60 Rugeles MT, Trubey CM, Bedoya VI, Pinto LA, Oppenheim JJ, et al. 2003. Ribonuclease is partly responsible for the HIV-1 inhibitory effect activated by HLA alloantigen recognition. AIDS 17:481–86 Yang D, Rosenberg HF, Chen Q, Dyer KD, Kurosaka K, Oppenheim JJ. 2003. Eosinophil-derived neurotoxin (EDN), an antimicrobial protein with chemotactic activities for dendritic cells. Blood 10.1182/blood-2003-01-0151 Larson KA, Olson EV, Madden BJ, Gleich GJ, Lee NA, Lee JJ. 1996. Two highly homologous ribonuclease genes expressed in mouse eosinophils identify a larger subgroup of the mammalian ribonuclease superfamily. Proc. Natl. Acad. Sci. USA 93:12370–75 Putsep K, Carlsson G, Boman H, Andersson M. 2002. Deficiency of antibacterial peptides in patients with morbus Kostmann: an observation study. Lancet 360:1144–49 Hancock REW. 1999. Host defence (cationic) peptides: What is their future clinical potential? Drugs 57:469–73

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Yang.qxd 3/12/2004 1:53 PM

See legend on next page

Page 1

ANTIMICROBIAL PROTEINS IN IMMUNITY C-1

Yang.qxd

3/12/2004

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

1:53 PM

Page 2

YANG ET AL.

Figure 4 Schematic illustration of the potential mechanisms by which defensins enhance host innate and adaptive antimicrobial immunity. When the integrity of skin or epithelial barrier is breached, microorganisms invading the host stimulate the release of a-defensins from neutrophils (recruited by microbial fMLP) and b-defensins from epithelial cells (Ep, including keratinocytes). Defensins promote innate antimicrobial immunity through at least five ways. For the enhancement of adaptive antimicrobial immunity, defensins potentially enhance the recruitment of iDCs to sites of infection. After killing of microorganisms, defensins may form a “dead microbe-defensin” complex to facilitate Ag uptake by targeting Ag to iDCs through defensin receptor–mediated internalization. In addition, defensins can enhance the maturation of iDCs to mDCs directly or indirectly by inducing the production of DC-maturing cytokines such as TNF and IL-1. Furthermore, defensins may also contribute to the effector phase of adaptive antimicrobial immunity by facilitating the recruitment of effector T cells to infected tissues. Other antimicrobial proteins including chemokines (not illustrated here) may have similar roles.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:181-215. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104522

Annu. Rev. Immunol. 2004. 22:217–46 doi: 10.1146/annurev.immunol.22.012703.104522 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on October 15, 2003

STARTING AT THE BEGINNING: New Perspectives on the Biology of Mucosal T Cells

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Hilde Cheroutre Division of Developmental Immunology, La Jolla Institute for Allergy and Immunology, San Diego, California 92121; email: [email protected]

Key Words lymphopoiesis, intestine, thymus, epithelial cells, immune regulation ■ Abstract The gastrointestinal tract is the central organ for uptake of fluids and nutrients, and at the same time it forms the main protective barrier between the sterile environment of the body and the outside world. In mammals, the intestine has further evolved to harbor a vast load of commensal bacteria that have important functions for the host. Discrimination by the host defense system of nonself from self can prevent invasion of pathogens, but equivalent responses to dietary or colonizing bacteria can lead to devastating consequences for the organism. This dilemma imposed by the gut environment has probably contributed significantly to the evolutionary drive that has led to sophisticated mechanisms and diversification of the immune system to allow for protection while maintaining the integrity of the mucosal barrier. The immense expansion and specialization of the immune system is particularly mirrored in the phylogeny, ontogeny, organization, and regulation of the adaptive intraepithelial lymphocytes, or IEL, which are key players in the unique intestinal defense mechanisms that have evolved in mammals.

INTRODUCTION The intestine of higher vertebrates is a major T cell organ. Three defined anatomical compartments contain T lymphocytes. The gut-associated lymphoid tissues (GALT), including Peyer’s patches (PP) and mesenteric lymph nodes (MLN), harbor T cells in organized lymphoid structures. T cells residing in the GALT have the most in common with peripheral lymph node T cells. In the other two compartments, T lymphocytes are scattered among nonlymphoid cell types. These include T cells distributed within the lamina propria (lamina propria lymphocytes, or LPL). Additionally, numerous intraepithelial lymphocytes (IEL) reside on the other side of the basement membrane from the lamina propria. IEL are found as single cells between the intestinal epithelial cells, with about one lymphocyte for every four to nine epithelial cells in the small intestine. It has been estimated, based on immunohistology, that the IEL of the mouse small intestine alone amount to almost 50% of the total T cell number in all lymphoid organs (1), although the number 0732-0582/04/0423-0217$14.00

217

11 Feb 2004

17:10

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

218

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

that can be purified is somewhat less. Additionally, the T cells residing in these different intestinal compartments are significantly different from conventional T cells as well as from one another in terms of ontogeny, phenotype, and function, and it is therefore evident that the IEL provide qualitatively and quantitatively major contributions to the biology of the mucosal and systemic immune systems. This review discusses the biology of IEL, with an emphasis on the subsets that express an αβ TCR. Although they are heterogeneous, as a group IEL exhibit a number of properties that distinguish them from peripheral T cells. The distinct nature and complexity of IEL is undoubtedly dictated by their unique location at the critical interface between the intestinal epithelial barrier and the gut lumen.

THE IMMUNE FUNCTION ORIGINATED IN THE GUT Host defense by specialized immune mechanisms arose hundreds of millions of years ago from the nondefensive, food-absorbing mechanism of phagocytosis by single-cell organisms. With the evolution of complex, multicellular organisms, differentiation and specialization became apparent, and food-absorbing cells became distinct from other cell types. The initial “gastral layer” of phagocytic cells first appeared in Metazoans and showed much homology with the gastrointestinal epithelial cell layer of the higher vertebrates. When the function of phagocytosis assumed its defensive role, protection of the organism from self-destruction initiated the first immune recognition. Defensive phagocytosis most likely only occurred in the absence of self-recognition, established via inhibitory antiself receptors. This self-recognition-based defense system had much in common with inhibitory NK receptors, which are regulated by recognition of self-MHC class I molecules. In fact, NK-like receptors appeared before T and B cell antigen receptors, as evidenced by the finding of a homolog of the C94/NKR-P1 NK-type receptor in the urochordate Botryllus (2). The origin of the adaptive immune system probably goes back to the gastrointestinal regions of the first “jawed” fish in the Placoderms species some 500 mya. The intestine as a central primary immune organ is evident in primitive vertebrates including the sea lamprey and hagfish (3, 4), but this is less apparent in higher vertebrates. The dominant evolutionary drive that led to the expansion and specialization of immune defense mechanisms coincided with the increasing complexity of the organism itself and the constant race to keep up with the evolving and increasing number of pathogens. This could have been the evolutionary selection for the acquisition of the ancestral RAG gene for antigen-receptor gene recombination, which is likely to have arisen from an ancient bacterial transposon (5). A particularly strong selective pressure might have been imposed on the multiple tasks of the digestive tract with the dietary changes concurring with the emergence of the jaw function and the migration of these early vertebrates from water to land. This might have led to the first need for compartmentalization of the intestinal tract (6). The mechanisms controlling this differentiation and specialization of

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

THE BIOLOGY OF MUCOSAL T CELLS

P1: IKH

219

the gastrointestinal tract are poorly understood, but homeobox-containing genes, such as Hox-1.4 (7) and Hox-3 (8), have been implicated in this differentiation process.

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE EVOLUTION OF THE THYMUS Although the gastrointestinal tract has retained some of its ancestral ability for lymphopoiesis, particularly for B-lymphocytes, the thymus has evolved as the primary lymphoid organ for T cell development. The appearance of the thymus seems to coincide with the sudden appearance of a family of MHC antigen-presenting molecules and expanding repertoires of adaptive T and B cell antigen receptors. This dramatically increased the capacity of the immune system to cope with the staggering demand for recognition of nonself external molecules, and with the complexity of the organism itself, as well as the need to control for neoplasm. Paradoxically, this sudden expansion also increased the potential of self-destruction by the immune system, and strategies to educate immune cells became a necessity (9). The evolutionary origin of the thymus is unknown, but several phylogenetic as well as ontogenetic facts suggest a connection to the gastrointestinal tract. In humans, the primordial thymus develops from the anterior portion of the embryonic gut tube (the pharyngeal pouches). At one point during gestation, the third pair of pouches buds off the gut epithelium (endodermis) and forms a sac-like epithelial protrusion that bends toward the digestive tract. This becomes the thymus rudiment. As the thymus compartment grows, it pushes into the mesoderm that surrounds the primordial gut, and some of the mesenchymal tissue becomes incorporated into the thymus tissue. Subsequentially, it becomes colonized by bone marrow– derived lymphocyte precursors, macrophages, and DCs, all mesodermal in origin. The thymus epithelium further develops and differentiates from the gut endoderm, but only with the stimulus from mesenchymal cells. The lymphocyte precursors interact with the epithelium in a two-way fashion, whereby the epithelial cells induce lymphoid differentiation, and at the same time they themselves differentiate further by the stimulus of developing lymphoid cells (10, 11). As a result, the thymus tissue becomes organized and a clear cortex and medulla are formed. Unlike the typical layer of gut epithelium cells with tight connections and limited intercellular spaces, the thymic epithelial cells are loosely connected. Nevertheless, desmosomes or tight junctions remain and are witnesses of their epithelial origin (12, 13). It has been suggested recently that thymus epithelium is composed of a variety of cells with different antigen-presenting capacity and which display a mosaic of ectopically expressed self-antigens (14). Speculations have been put forward that this developmental plasticity of the thymus organ may be important for the generation of self-tolerance to some tissue-specific antigens, and we propose here that this ectopic expression of organ-specific self-antigens is equally important for the selection in the thymus of self-specific T cells that are plentiful among the IEL (discussed below).

11 Feb 2004

17:10

220

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Thymocyte precursors are believed to arrive in the subcapsullar region of the thymic cortex, where the initial events of conventional positive selection occur. The selected thymocytes subsequently move to the medulla for further maturation and differentiation, and eventually they exit the thymus. The thymus organ involutes later in life, when a peripheral repertoire of mature TCRαβ + T cells and memory T cells is established and the need for newly generated na¨ıve T cells is no longer desirable. As a consequence in older animals, the remnant lymphopoietic capacity of the gut occasionally becomes more apparent (15).

MODERN TIMES: MUCOSAL T CELL SUBSETS IN MAMMALS Mammalian IEL are almost exclusively T cells, but they have diverse phenotypes with regard to TCR and coreceptor expression, as well as the expression of other cell surface molecules. Unlike peripheral T cells, however, all the diverse populations of IEL display an antigen-experienced phenotype. Nevertheless, the locations and pathways the various IEL subsets follow to encounter antigens, and the nature of these antigens, distinguish them from one another. There are three broad categories of IEL: the subset of γ δTCR+ IEL and two distinct subsets of αβTCR+ IEL, those that express one of the conventional TCR coreceptors, CD4 or CD8αβ, and those that lack coreceptor expression and are double negative (DN). A common feature of all IEL subsets is the unique capacity to express CD8αα, a characteristic of the activated nature of the mucosal T cells and their adaptation to the gut environment. The ratio of γ δ to αβ TCRs in IEL varies among species. In mice, a significant portion of the IEL in the small intestine expresses a γ δTCR, and they predominate in young animals at three weeks of age, after which both TCRγ δ + and TCRαβ + IEL increase in numbers (16). They are less prevalent in the large intestine (17, 18), and in humans TCRγ δ + IEL are a relatively small (∼10%) population (19), but their numbers increase significantly during coeliac disease (20). γ δ TCRs generally do not recognize processed peptides presented by MHC class I or class II molecules (21, 22), and γ δ TCRs share properties with both adaptive receptors encoded by rearranging genes and invariant NK-like receptors. Similar to the self-based NK receptors, some γ δTCRs can interact directly with self-MHC molecules such as the mouse T10 and T22 molecules (23) and human MIC molecules (24), which are not loaded with processed antigens. The specificity of most mouse γ δTCR IEL, however, is unknown. γ δT cells are also capable of sensing stressed cells by a direct recognition of stress-induced self-proteins. The human MHC class I–like molecules MICA and MICB are examples of such inducible self-antigens. Stressed intestinal epithelial cells and some epithelial tumors induce their expression, and MIC molecules can be recognized by Vδ1+ γ δ TCRs that are most prevalent among IEL (25, 26). The NK receptor NKG2D, which also binds to MICA and MICB (27), likewise is expressed by those γ δTCR IEL (28). It is thus possible that the rearranged receptors and the invariant receptors recognize the same molecules,

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BIOLOGY OF MUCOSAL T CELLS

P1: IKH

221

providing an opportunity for cross-talk between these receptor systems. Mice do not have structural homologs of MICA and MICB, although they have NKG2D receptors and class I–like molecules, including REA-1 and H-60, that can bind to them (29, 30). The second major category of IEL are the αβTCR+ cells that express the typical TCR coreceptors, CD4 or CD8αβ, although the ratio of CD8αβ to CD4 is much higher on these cells compared with the spleen cells. Conventional αβTCR+ cells will migrate from the GALT and peripheral lymphoid tissues to the intestinal epithelium upon antigen stimulation in the periphery (31); these are essentially tissue-seeking effector or memory cells. Specialized M cells in the follicleassociated epithelium continuously sample the gut lumen and transport antigens without processing to the subepithelium and GALT (32). Local dendritic cells (DCs) then process and present these antigens and further distribute them via the thoracic duct (TD) into the bloodstream. Peripheral na¨ıve TCRαβ + T cells reactive to the specific antigens will then undergo activation and proliferation locally and migrate as antigen-experienced lymphocytes via the TD to the gut, from where they can seed the whole length of the small intestinal epithelium (33). Repeated identical antigenic stimulation progressively narrows the TCR repertoire of these conventional IEL, which consequently have an oligoclonal TCR repertoire (33). These cells, which have a memory function, have been shown to be protective against oral infections with several pathogens (34–38). Gut-seeking, antigen-experienced T cells have some differences from those in the spleen, including an initial dependence upon CD40-CD40L interactions (36), effector function that can readily be measured in vitro (38), and longer survival than the memory cells in the spleen (38). It has recently become apparent, however, that antigen-experienced T cells migrate to a variety of tertiary sites, including the lung, liver, and intestine (39). This has led to the concept that there are two types of memory T cells that can be distinguished by expression of CCR7 and other molecules (40). Central memory cells are localized to lymphoid organs, whereas effector memory cells are found in the effector organs. Although it is not fully understood what the specific signals are that determine the localization of these conventional T cells activated in the periphery to the intestinal epithelium, integrins containing a β7 subunit are important, and β7deficient mice have greatly reduced numbers of T cells throughout the mucosa (41–43). Chemokines are also likely to play an important role. The chemokine CCL25 (TECK) is produced by epithelial cells in the small intestine, but not by those in the large intestine (44, 45), and its receptor CCR9 likewise is found predominantly on small intestine IEL and LPL (46). Recent data suggest that this chemokine may be involved in the selective homing of conventional T cells to the small intestine (47), although the in vivo blocking of CCL25 with antibodies (48), or the analysis of CCR9−/− mice (49), indicated only modest decreases in IEL. It is therefore likely that there is redundancy in this chemokine/chemokinereceptor system. Nevertheless, the exposure to antigens that are present in mucosal lymphoid structures is important for the migration of conventional peripheral

11 Feb 2004

17:10

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

222

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

T cells to mucosal tissues. CD4+ T cells activated in peripheral lymph nodes have low expression of the α 4β 7 integrin important for homing to the intestine, whereas those activated in intestinal lymphoid structures selectively express high levels of α 4β 7 and migrate in response to CCL25 (50). Similarly, for CD8+ T cells, it has been shown that activation with DCs from PP induces α 4β 7 and CCL25 responsiveness, whereas activation of splenic or lymph node DC does not (51), suggesting an imprinting of a mucosal-seeking phenotype by PP DCs. The third category of IEL includes TCR αβ + DN cells, and the more numerous CD8αα single positive (SP) IEL. Like the γ δTCR IEL, the expression of a number of surface proteins characteristic of conventional T cells are reduced or lacking on these cells, including Thy-1 (52), CD28 (53), and CD2 (54). On the other hand, they exclusively express NK receptors and they have been dubbed “gut NKT cells” (55). Additionally, and typical of NKT cells, cells that express a canonical human Vα7.2-Jα33 or mouse Vα19-Jα33 TCR have been identified among the mucosal lymphocytes, particularly among the LPL. These mucosal-associated invariant T cells are selected and/or restricted by a nonclassical class I molecule, MR1 (56). The invariant TCR α chain and reactivity to a nonpolymorphic class I molecule are features similar to the CD1d-dependent NKT cells, suggesting that this MR1-reactive population might constitute an intestinal-specific NKT cell subset. CD8αα IEL preferentially use the invariant signaling component FcεRIγ chain (57, 58), which also is a subunit of the multimeric Fcγ RIII A (CD16) complex displayed by NK cells. The FcεRIγ chain is expressed as part of their CD3 complex, either as a homodimer or as a heterodimer together with CD3ζ . The presence of a different TCR signaling module, as well as the absence of coreceptor molecules and accessory/costimulatory molecules such as CD2 or CD28, indicate that the TCR-mediated selection and activation events for CD8αα TCRαβ + IEL are likely to be different from conventional T cells. Although CD8αα SP TCRαβ + IEL express an oligoclonal TCR repertoire, this repertoire does not overlap with the one expressed by the CD8αβ TCRαβ + IEL (59, 60). Furthermore, although potentially self-reactive TCRs are depleted from the peripheral repertoire during negative selection, these TCRs accumulate among the CD8ααSP TCRαβ + IEL (1). This implies that the CD8αα SP TCRαβ + IEL have not been submitted to conventional negative selection, and that the gut-specific CD8αα + IEL must have originated under distinct selection pressure with a distinct ontogeny.

T CELL LYMPHOPOIESIS AND THE INTESTINE Even before TCRs had been discovered, based on the presence of IEL in neonatally thymectomized mice, Fichtelius postulated that the small intestine is a “first-level,” or primary, lymphoid organ (61). Direct evidence for in situ development of T cells within the gut epithelium was provided by Ferguson & Parrott, who engrafted fetal small intestine tissue under the kidney capsule of athymic mice (62). They showed

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BIOLOGY OF MUCOSAL T CELLS

P1: IKH

223

that host lymphocytes developed to some extent in the intestinal grafts. With improved technology and new tools, other approaches were taken to provide further evidence to demonstrate and characterize T cell lymphopoiesis in the absence of a functional thymus. Although the results from a number of experiments provided evidence for an extrathymic origin for some IEL (63–65), in no case could it be demonstrated that functional TCRαβ + T cells had been subjected to a positive selection process. The results showed that the extrathymic pathway of development was inefficient and mostly skewed to the TCRγ δ + population. Consistent with this thymus dependence, thymic transplant gives rise to IEL in chickens (66, 67), neonatal thymectomy causes a drastic decrease in IEL in mice (68), and early thymectomy results in decreased IEL in Xenopus (69). Congenitally athymic (nu/nu) nude mice, which have a mutation in the winged-helix transcription factor, Whn, that is required for differentiation of thymic epithelial cells (70), have greatly reduced TCRαβ + IEL, although there is a residual γ δTCR+ IEL population (71). The level of TCRαβ and CD3 expression on the few TCRαβ + T cells in nude mice is low, and the percentage of splenocytes expressing CD4 or CD8αβ is greatly variable among individual nude mice (72). Furthermore, the repertoire of αβ TCRs in nude mice is pauciclonal (73, 74) and often shows extreme bias toward certain Vβs (73). In minor lymphocyte-stimulating (Mls)-1a or I-E+ strains, T cells expressing a Vβ6 or Vβ11 TCR are present in the periphery, whereas thymocytes expressing these TCRs in euthymic mice of the same strains are deleted during the process of negative selection in the thymus (73, 75). This provides further evidence that the TCRαβ + T cells that develop in the absence of a thymus do not undergo the selection events characteristic of thymic development. Evidence for lymphocyte development in the adult human intestine is scarce (76). mRNA for RAG-1 and RAG-2 could be isolated from IEL but not from LPL (77) while terminal deoxynucleotidyl transferase was undetectable (78). Nevertheless, adult IEL TCRβ mRNAs contain numerous N region additions (79, 80), indicating that any role in situ for RAG in building an IEL-specific TCRαβ repertoire must be very small. Human fetal intestine contains T cells in the LP and epithelium from 12 to 14 weeks into gestation (81), but the fetal thymus is already exporting T cells at this stage of gestation. Immunohistochemical staining of tissue sections of fetal intestine indicates that approximately 50% of the CD3+ IEL express CD8αα +, whereas the rest are predominantly DN cells. The LP also contains numerous CD3+ cells, but CD4+ T cells outnumber the CD8+ LPL regardless of gestation age, and there are large numbers of DN cells as well (76). The majority of fetal IEL already express an activated phenotype, and unlike adult IEL, they are actively proliferating. It is unlikely that IEL in the fetal intestine have migrated there as effectors in response to nonself antigens. Together with the fact that fetal IEL, unlike IEL of the adult intestine, display a very diverse TCR Vβ repertoire (76), it seems logical to suggest that fetal IEL represent a na¨ıve repertoire. The antigen-experienced phenotype of the fetal IEL together with the fact that most fetal CD8+ IEL express CD8αα (82) is consistent with the finding in mice that CD8αα TCRαβ + IEL have acquired this phenotype during self-agonist-based

11 Feb 2004

17:10

224

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

selection in the thymus (discussed below). In summary, the evidence from several species is consistent with a thymic origin of all or most IEL, and it is also consistent with the hypothesis that efficient positive selection of T cells occurs only in the thymus.

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

LYMPHOCYTE PRECURSORS AND THE INTESTINE Although abundant data support a thymic origin for IEL, the intestine retains some vestigial function in lymphopoiesis. In an immunohistochemical search for anatomical sites of lymphopoiesis within the gut, small clusters of lymphoid precursors called cryptopatches (CP) were discovered within the LP compartment adjacent to the epithelial crypts in the small intestine in mice (83). Similar structures in other species have been ambiguously described (84). In nude mice, none of these clusters contain CD3+ TCR+ cells, and the c-kit+ cells located there were negative for expression of CD3, B220, Mac-1, Gr-1, and TER 119 (lin−) (83). Transplantation of CP subsets isolated from athymic nude mice to irradiated severe combined immune deficient (SCID) recipient mice indicated a lymphopoietic capacity for the c-kit+ CP subset (83). Most c-kit+ CP cells also expressed the IL-7R (85), and in both IL7Rα −/− mice and common cytokine receptor γ -chaindeficient mice (CRγ −/−), formation of CP was defective (86). γ δTCR+ IEL, which are dependent on IL7-R signaling for their development, were also absent in these CP-deficient strains. Consistent with a CP-independent origin, however, αβ TCR+ IEL were still present in the CP-deficient strains but absent in athymic nude CRγ −/− mice (86). TCR− IEL isolated from these athymic CRγ −/− mice have characteristics resembling B220+ fetal liver progenitors (87). Furthermore, although the adoptive transfer of CP cells to SCID recipient mice suggested the presence of T cell precursors, when transferred to whole-lobe fetal thymic organ cultures, CP cells showed no such capacity (83). T cell precursor activity can be revealed among IEL as well. Co-culture of CD3−CD8−Ig−CD45+ IEL with newborn thymic stromal cells, which had been depleted of CD45+ thymocytes, led to the generation of DP (CD4+, CD8β +) and SP CD4+ or CD8αβ + cells (88). DP cells that were generated from these co-cultures of intestinal progenitors and thymic stromal cells expressed αβTCRs, whereas DN cells expressed either αβTCRs or γ δTCRs. Furthermore, a significant amount of the T cells expressed NK markers (88). A subset of the TCR− putative progenitors expresses low levels of CD45 (CD45low). Cells with this phenotype are abundant among the epithelial cells of nude mice. The CD45low subset also expressed markers typical of immature thymic precursors, such as CD44high, CD16, CD24 (heat-stable antigen), c-kit, CD127 (IL-7Rα), and CD122 (IL-2/15Rβ) but lacked CD25 (IL2Rα). The presence of the IL-7Rα and IL-2/15-Rβ is consistent with a role for IL-7 and IL-15 in the ontogeny of IEL (89, 90). Cocultures of these CD45low CD3−CD8− cells isolated from nude mice with thymic stromal cells from normal mice also led readily to the development of DP, SP, and DN T cells similar to the developmental stages of thymocytes (88).

15 Mar 2004

15:11

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BIOLOGY OF MUCOSAL T CELLS

P1: IKH

225

Studies on lymphopoiesis within the intestine are hampered by the dominant presence of mature T cells in the adult gut, as opposed to mostly immature precursors in the thymus. Using multiple lineage markers, progenitor cells isolated from CPs and IEL were further characterized and divided into six types (91, 92). Bone marrow from nude mice was transferred to thymectomized RAG-2−/− and CRγ −/− double-deficient recipient mice to establish the sequence of appearance of these precursors during lymphoid development (92). CP lin− cells showed very little evidence for T cell commitment, but they expressed GATA-2, a transcription factor found in pluripotent hematopoietic progenitor cells (93). It is therefore possible that CP cells represent predominantly precursors of other hematopoietic cells. Using several gene mutations that affect T cell differentiation, including preTα −/−, TCRα −/−, and CD3ε−/−, it was shown that none of these mutations affected the CP cells, further demonstrating that these cells are not committed to the T cell lineage (92). IEL progenitors were more committed to the T cell pathway, and the TCRα −/− mutation led to accumulation of DP IEL. On the other hand, TCRδ −/− did not block the cells at the DP stage, and it is thus more likely that TCRγ δ IEL developed before the DP stage. Such a scenario is typical of the developmental pathway for γ δ T cells in the thymus (See Figure 1). Pre-Tα mRNA, uniquely expressed by the precursors committed to the TCRαβ lineage in the thymus, was extremely low among the Lin− gut cells, further supporting the notion that T cell precursors in the gut are predominantly committed to the TCRγ δ T cell lineage (92). To determine if the extrathymic pathway of T cell development is active in euthymic mice, transgenic mice that carry a reporter gene (green fluorescent protein, or GFP) driven by the RAG-2 promoter were analyzed (15). Cells expressing high levels of GFP in these mice indicate recent transcription of RAG-2, reflecting TCR and Ig gene rearrangements (94). Using this reporter system, it was shown that athymic mice show receptor rearrangements predominantly in the MLN and PP. CP cells did not express RAG-2, indicating that rearrangements of TCR genes did not occur in these cells. Mature cells migrated from the MLN to the TD, and eventually to the gut mucosa. Interestingly, this pathway was totally suppressed in euthymic mice, except under conditions of severe lymphopenia. Nevertheless, in euthymic TCRβ −/− mice, which only have mature γ δ T cells, GFP+ cells were present in MLN and PP (15). This indicated that the lack of extrathymic T cell development in euthymic mice is not due merely to the presence of a thymus, but rather the thymusderived TCR αβ T cells themselves actively down-modulate lymphopoiesis in the gut. The suppression is less pronounced in older mice where newly generated TCRαβ T cells are much reduced due to the involution of the thymus (15).

CD8αα TCRαβ IEL ARE SELF-REACTIVE Conventional selection of TCRαβ T cells in the thymus is central in establishing self-MHC restriction (95–98) and self-tolerance (99, 100). The na¨ıve TCRαβ repertoire in the periphery is thus imprinted by the selective interactions of TCRs

15 Mar 2004

15:13

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

226

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

with self-antigens/MHC during this selection process in the thymus (101). The affinity/avidity-based model for selection proposes that TCR signal strength is central to the outcome of thymic selection, with intermediate TCR signal strength resulting in positive selection (Figure 2) and further maturation of the CD4+CD8αβ + double positive (DP) immature thymocytes to the SP mature stage (102). Strong TCR-mediated selection signals delete those TCRs from the repertoire of conventional T cells (see Figure 2). The CD8αα TCRαβ IEL subset, however, is governed by different avidity requirements than those outlined above for conventional T cells. In MHC class II, I-E+ mice bearing the Mls-1a allele of the Mls locus, T cells expressing TCRs (Vβ6, Vβ8.1, or Vβ11) reactive to the endogenous retroviral superantigen of the Mtv-7 provirus are deleted from the conventional repertoire while these TCR are present, and in some cases at frequencies higher than the nondeleting strain, in the CD8αα TCRαβ IEL subset (1). The notion that CD8αα TCRαβ IEL do not follow the rules of conventional selection was also apparent in TCRαβ transgenic mice that also express the cognate antigen recognized by the TCR transgene. H-Y TCR transgenic mice, with a TCR that recognizes a Y chromosome-encoded Smcy peptide presented by Db, provide a paradigm for these types of studies. When the TCR transgene was crossed on a RAG−/− background (103), it was shown that CD8αβ SP thymocytes matured in the thymus and populated the peripheral tissues of H-Y antigen-negative female mice. There were few IEL in these female mice, however, which is consistent with the belief that conventional T cells migrate to the intestine only upon antigen stimulation. This stimulation might be provided to some extent by endogenous peptides of the kind that are responsible for the homeostatic maintenance of na¨ıve T cells in the periphery. The H-Y TCR+ T cells are unusual in that they do not undergo this type of homeostatic maintenance in female mice (104, 105), perhaps reflecting the low affinity of this TCR for endogenous peptides in female mice. In sharp contrast, male H-Y TCR transgenic mice developed large numbers of TCR+ IEL, and they predominantly expressed CD8αα, despite deletion of conventional CD8αβ H-Y TCR+ T cells (103, 106, 107). The active generation of the CD8αα TCRαβ IEL under deleting conditions for the conventional TCRαβ T cells clearly demonstrated that a different outcome pertained to the CD8αα TCRαβ IEL with regard to self-recognition during selection. Similar observations were made in a number of other TCR transgenic systems expressing endogenous cognate antigen, including the 2C TCR system recognizing Ld as an alloantigen (108), and in double transgenic mice expressing a transgenic TCR and the cognate antigen as a transgene, including OT-I/RIPmOVA (107) and F5 TCR transgenic mice expressing influenza virus nucleoprotein (NP) peptides as transgenes (109). The selection of these CD8αα + IEL is most effective in the presence of highaffinity self-antigens. This was demonstrated using the clonotypic TCR F5. The NP68 peptide is recognized by the F5 TCR with high affinity and induces profound deletion of thymocytes in vitro, whereas the peptide NP34 is a peptide recognized with lower affinity. NP68-F5 TCR double transgenic mice generated large numbers

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BIOLOGY OF MUCOSAL T CELLS

P1: IKH

227

of CD8αα TCRαβ + IEL (109). Likewise, CD8αα IEL expressing the 2C TCR were readily generated in mice expressing the Ld high-affinity ligand, whereas CD8αβ + T cells developed when the low-affinity ligand presented by H-2Kb was present. Similarly, H-Y TCR+ CD8αα IEL were more efficiently generated in Db/b homozygous male mice than in their Db/d heterozygous counterparts (110). The expression of CD8αα on these IEL does not necessarily imply MHC class I restriction of the TCR. A similar, high-affinity self-antigen-dependent development of CD8αα expressing IEL also was observed using several MHC class II–restricted TCR transgenic systems (107, 111). These observations indicate that CD8αα does not function as a traditional TCR coreceptor and that selection of the CD8αα TCRαβ IEL might occur in a coreceptor-independent fashion. This was also supported by the observation that CD8αα H-Y TCR+ IEL are actively generated in male mice that are deficient for CD8β expression, whereas the female H-Y TCR transgenic mice on the CD8β −/− background failed to generate mature H-Y TCR transgenic T cells (107). Although CD8αα is most likely not directly involved as an MHC-interacting coreceptor in the selection of self-antigen-reactive TCRs, all CD8αα TCRαβ IEL depend on some kind of MHC class I expression for their development and/or homeostasis. CD8αα TCRαβ IEL, but not TCRγ δ IEL, are drastically reduced in β2-microglobulin-deficient mice (112–114). By contrast, CD8αα TCRαβ IEL are present in mice deficient for the transporter associated with Ag processing (TAP) gene, whereas conventional CD8αβ + MHC class I restricted T cells are nearly absent in these mice (112). The classical MHC class I molecules are highly TAP-dependent for their cell-surface expression, although some nonclassical class I molecules, including the thymic leukemia (TL) antigen and CD1d are not (115, 116). Similar to the TAP−/− mice, animals deficient for the classical class I molecules (Kb−/−Db−/−) efficiently generated CD8αα TCRαβ IEL, whereas the population of conventional CD8αβ-T cells was greatly reduced (117–119). The somewhat more efficient generation of CD8αα TCRαβ IEL in Kb−/−Db−/−mice, as compared with the TAP−/− mice, led to the suggestion that TAP-independent and TAP-dependent nonclassical class I molecules could promote the generation of the CD8αα TCRαβ IEL. It has been reported that mice deficient in the nonclassical MHC class I molecule, Qa-2, which is for the most part TAP dependent (120), were much less efficient in generating the CD8αα IEL, suggesting Qa-2 reactivity for some CD8αα TCRαβ IEL (121). By comparing BALB/c mice with the Qa2null Bailey substrain (BALB/cByJ), we were unable to confirm these findings (L. Gapin, M. Kronenberg & H. Cheroutre, unpublished data). The general rule that emerges from these studies is that CD8αα TCRαβ IEL are self-reactive, but they are not selected for self-reactivity to a single self-antigen or a single antigen-presenting molecule. It is difficult, however, to estimate the fraction of these cells that is reactive to classical class I molecules, nonclassical class I molecules, or MHC class II molecules, considering that in the absence of one specificity, cells with another specificity may be expanded. The diverse specificities of these cells, together with their requirement for β2m, indicates that

11 Feb 2004

17:10

228

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

a function mediated by an MHC class I molecule other than peptide presentation to a TCR is required for the selection of CD8αα TCRαβ IEL.

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD8αα TCRαβ IEL ARE THYMUS DERIVED The efficient generation of the CD8αα TCRαβ IEL, under conditions that caused negative selection in the thymus of CD8αβ T cells, was consistent with the view that the CD8αα-expressing cells were generated extrathymically. As discussed above, studies on the lymphopoietic capability of the small intestine indicated that IEL could develop locally in the absence of a functional thymus. Nonetheless, the local T cell development in the gut was biased toward the development of γ δ T cells. This, together with the drastic reduction of CD8αα TCRαβ IEL in nude mice and mice that were neonatally thymectomized, suggested that the thymus is required to efficiently generate all TCRαβ T cells, including the IEL. This is also supported by the observation that thymus grafts from normal mice efficiently generated CD8αα-expressing TCRαβ IEL in recipient nude mice (122). Interestingly, it was shown that the different subsets were generated with different kinetics, and that whereas fetal thymus grafts or grafts from mice up to weaning age predominantly repopulated the intestine of the recipient mice with CD8αα TCRαβ T cells, thymus grafts from older mice were more efficient in generating the conventional TCRαβ T cell subsets (122). This was not necessarily due to the reduced ability of the adult thymus to generate CD8αα TCRαβ IEL. Rather, it could result from the suppression of the selection and/or expansion of CD8αα TCRαβ IEL by the increasing population of conventional TCRαβ T cells. Support for this hypothesis is provided by the demonstration that adult thymus grafts isolated from male H-Y TCR donor mice on a RAG−/− background readily and efficiently repopulated the intestine of RAG−/− recipient mice with mature H-Y TCR transgenic CD8αα IEL (107). It is possible that in the absence of conventional positive selection, as in the case of the male H-Y TCR transgenic mice on a RAG−/− background, the generation of CD8αα TCRαβ IEL will continue as long as a functional thymus is present.

AGONIST SELECTION: A NOVEL PATHWAY OF T CELL DIFFERENTIATION Although it is thought that negative selection is a major tool for the immune system to eliminate self-reactivity from the T cell repertoire (123), the death of autoreactive cells is not all encompassing. New evidence suggests that an agonist-based positive selection pathway permits the immune system to preserve and specifically select for self-reactive TCRs (Figure 1) (107). As outlined above, strong correlative evidence indicates that self-agonists are required to promote the development of CD8αα + IEL. Consistent with active selection for high-affinity-based TCRs, it was found that among the selected TCR+ thymocytes in male H-Y TCR transgenic

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BIOLOGY OF MUCOSAL T CELLS

P1: IKH

229

mice on a RAG+ background, with the potential to edit their TCR to escape negative selection, there was no evidence for editing. On the other hand, on a nonselecting background, numerous thymocytes underwent editing of the TCRα chain (124). Similar high-affinity selection conditions have been described for CD4+ CD25+ regulatory T cells (125) and NKT cells, including those that express an invariant Vα14 TCR and are CD1d reactive (126), and those with other specificities (127, 128). Some of these agonist-selected T cell subsets have regulatory functions, suggesting that agonist selection enables the immune system to make use of T cells that recognize self to control the responses of nonself reactive T cells. The development and selection of self-reactive T cells is not an imperfection of the thymic selection mechanism, or a mere escape of these cells from conventional negative selection to create anergic or nonfunctional T cells. Evidence suggests that T lymphocytes selected by self-agonists require true ERK-dependent positive selection signals mediated by the TCR (107, 128). This was inferred by their requirement for the TCRα chain-connecting peptide domain or α-CPM. This unique motif appeared with the first ancestral TCRαβ receptor in jawed fish, and it has been conserved over 500 million years, underscoring its crucial function during TCRαβ T cell development (129). The α-CPM allows for efficient association of the CD3 subunits with the TCR, and imparts an ERK-mediated signaling function during thymocyte positive selection. TCR transgenic mice lacking the α-CPM motif have impaired positive selection of conventional TCRαβ T cells while negative selection still occurs (130). When TCR/Ag double transgenic mice that also had a mutated α-CPM α chain were analyzed for the presence of CD8αα TCRαβ TCR transgenic IEL, we showed a much reduced population of these cells in the α-CPM mutant mice, demonstrating that agonist selection is also a true ERK-dependent positive selection process (107). Together with the lack of dependence on CD8β, these data indicate that the agonist-driven selection of CD8αα TCRαβ IEL is not an escape from conventional negative selection. The concept of agonist-based selection is complementary to, and not mutually exclusive from, the deletion of thymocytes during conventional negative selection. The TCRs that are eliminated from the pool of conventional T cells might specifically accumulate in the repertoire of agonist-selected T cells. Such a prediction would indeed be consistent with the accumulation of the cognate transgenic TCRs among the CD8αα IEL of double TCR/Ag transgenic mice (107) as well as with the accumulation of self-reactive TCRs among the CD8αα IEL in normal mice (1). Furthermore, it remains to be determined if at very high affinities, agonist selection is no longer possible and negative selection is completely dominant (Figure 2). The data from multiple TCR transgenic systems suggests, however, that an affinity in which only negative selection is found may be difficult to reach. As a result of the agonist-selection process, T lymphocytes differentiate into highly specialized cells with an antigen-experienced phenotype and regulatory functions. Mature agonist-selected T cells migrate preferentially to specific organs and tissues including the intestinal epithelium or liver, as opposed to the migration of conventional na¨ıve T cells, which predominantly migrate to defined lymphoid

11 Feb 2004

17:10

230

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

organs in the periphery (Figure 2). The factors that are responsible for the selective localization of CD4+ Treg to lymph nodes and spleen, Vα14 invariant NKT cells to the liver, and CD8αα TCRαβ IEL remain to be determined.

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

AGONIST SELECTION AND SELF-ANTIGENS IN THE THYMUS If self-reactive T cells were primordial, then the thymus might have evolved initially to accommodate the agonist selection of self-Ag-specific T cells reactive to polymorphic self-molecules. Recently, it has been shown that tissue-specific self-antigens are expressed by a subset of medullary thymic epithelial cells (131). The ectopic expression of these self-antigens is controlled by the transcriptional regulator, AIRE (132–134). The human autoimmune polyendocrinopathy syndrome type 1 is a multiorgan autoimmune disease resulting from loss-of-function mutations in the AIRE gene (132, 134), and AIRE-deficient mice also showed autoimmune disorders similar to the human disease. These observations suggested strongly that the AIRE-controlled ectopic expression of self-antigens in the thymus was central to establish self-tolerance (135). It was proposed that the absence of expression of these self-antigens during selection in the thymus resulted in the absence of effective negative selection and the appearance of uncontrolled autoreactive conventional TCRαβ T cells (132, 133). This was illustrated in a model system, in which TCR transgenic T cells specific for a hen egg lysozyme (HEL) peptide were crossed to transgenic mice expressing HEL under the control of the rat insulin promoter. The normal induction of thymic tolerance in the TCR transgenic T cells to the insulin promoter–driven HEL was abolished when these mice were crossed onto an AIRE-deficient background (133). We speculate that AIRE, and a transcription factor that controls AIRE expression (see below), might be required for the positive selection of self-specific CD8αα TCRαβ IEL as well as the negative selection of conventional T cells. The observation that agonist-positive selection of specialized T cells, including the CD8αα TCRαβ IEL, is based on recognition of self-Ag, and the fact that efficient agonist-driven positive selection occurs simultaneously with negative selection of conventional TCRαβ T cells, suggests that the autoimmune pathology linked with defects in AIRE mutant mice and humans might reflect impaired agonist selection in addition to defective negative selection. The absence of self-Ag-based agonist selection might lead to reduced regulation of conventional T cells by these selfantigen-specific IEL and other regulatory T cells. The regulatory role of CD8αα TCRαβ IEL is discussed below. It has been proposed that mouse CD1d reactive NKT cells with an invariant Vα14 rearrangement, referred to here as Vα14i NKT cells, are selected by glycolipid self-agonists (126). Similar to the CD8αα TCRαβ + IEL, these NKT cells have a natural memory phenotype, and interestingly, their IL-4 and IFNγ gene loci are in the open configuration (136) and they become active as these cells mature in the

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BIOLOGY OF MUCOSAL T CELLS

P1: IKH

231

thymus. Although thymic epithelial cells are not required for the CD1d-mediated presentation of self-antigens during selection of Vα14i NKT cells, we and others have shown that a signaling pathway in a radio-resistant host cell, with the NFκB-inducing kinase (NIK) activating the NF-κB family transcription factor RelB, is required for Vα14i NKT cell differentiation (137, 138). Furthermore, evidence suggests that this signaling pathway is required in thymic medullary epithelial cells. Interestingly, RelB−/− mice also display a significant decrease in the number of IEL, including the CD8αα-expressing TCRαβ T cells (R. Shaikh, M. Kronenberg & H. Cheroutre, unpublished data) indicating that a RelB-dependent pathway is also crucial for the development, homing, or homeostasis of TCRαβ CD8αα IEL. The fact that these two agonist-selected T cell populations are defective in RelB−/− mice, while conventional T cells are numerous, has led us to speculate here that the agonist-dependent selection pathway is specifically impaired in the RelB−/− mice. Similar to the AIRE−/− mice, RelB−/− mice also failed to delete conventional T cells with a high affinity for self, and these mice also show a T cell–dependent autoimmune pathology (137, 139, 140). This suggests that the RelB-dependent contribution could be the same for both types of selection, the agonist-positive selection and the conventional negative selection. Interestingly, it has been shown that AIRE expression by thymic epithelial cells is RelB dependent, and RelB−/− mice show abnormal thymic medullary epithelial cells and complete absence of AIRE expression (134). It is thus possible that the autoimmunity displayed by RelB−/− mice could be attributed in part to the lack of RelB-dependent AIRE expression. Consequently, this would imply that defects in central tolerance, which lead to autoimmunity, are not solely the result of impaired conventional negative selection, but that concomitant defects in the process of agonist selection, and the resulting absence of self-specific CD8αα TCRαβ IEL and other agonist-dependent regulatory T cells, might contribute to the severity of autoimmune conditions that are particularly prevalent in the intestine.

AGONIST SELECTION AND DISTINCT FEATURES OF SIGNALING It is not understood what drives thymocyte precursors to the agonist-selection pathway or at what point in thymopoiesis precursors divert to the different selection pathways. We can only speculate that the quality of the TCR signal received during the initial selection events differentiates precursors along distinct ontogenetic pathways. Nearly all CD8αα TCRαβ IEL express the FcεRIγ chain as part of the CD3 complex, whereas conventional TCRαβ IELs express homodimers of CD3ζ or hetero dimers of CD3ζ and η (57), an alternatively spliced form of ζ . The genes for CD3ζ /η and FcεRIγ are related and genetically linked. Analyses of gene-deficient mice have indicated that CD3ζ /η is required for conventionally selected TCRαβ IEL (CD8β + or CD4), but less so for the agonist-selected CD8αα TCRαβ IEL, consistent with their usage of the FcεRIγ subunit (58, 141, 142).

11 Feb 2004

17:10

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

232

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

The differing composition of the TCR signaling components between agonistselected T cells and conventionally selected T cells suggests likely differences both in the downstream signaling cascades and in the developmental intermediates required to produce the different T cell types. Immature thymocytes at the DN stage initially express FcεRIγ as part of their CD3 complex (143), and upon transitioning to the immature DP stage they replace the FcεRIγ with CD3ζ /η. The fact that FcεRIγ is retained in the TCR complex of the CD8αα TCRαβ IEL, combined with the fact that these cells can be selected in a coreceptorindependent fashion, suggests the possibility that these agonist-selected T cells might be selected at the DN stage, deviating from the conventional pathway that requires a DP intermediate (see Figure 1). In many TCR/Ag double transgenic models that promote the development of the self-Ag-specific CD8αα TCRαβ IEL, the coreceptor-expressing thymocytes (DP) that proceed along the conventional pathway of selection are greatly reduced in number (107), indicating that the development of the precursors of these agonist-selected T cells could occur in a CD4/CD8αβ coreceptor–independent way. This was directly demonstrated using the class II restricted DO-TCR transgenic system. DN transgenic T cells are selected under agonist conditions on an I-Ab background. In H-2b CD4−/− mice expressing human CD2 under the control of the mouse CD4 promoter, selected mature DN DO-TCR transgenic T cells did not express the CD2 transgene, suggesting that the DO-TCR transgenic T cells had developed without passing the DP stage (144). The DN pathway was also suggested in a recent phenotypic analysis of thymocytes from H-Y TCR transgenic mice, where in the male transgenic mice the majority of the DN thymocytes have a mature phenotype, whereas the DN thymocytes from female transgenic mice are predominantly immature cells (111). The immediate precursors of DP thymocytes are the CD4− CD8αα + thymocytes (82). Although there are very few in normal adult mice, these precursors are numerous in the TCR/Ag double transgenic animals, and they constitute all the CD8+ thymocytes in the fetal thymus (82). Even though controversy exists as to what extent these CD8αα SP thymocytes are mature immune-competent cells, it was shown that TCR cross-linking of thymocytes at the CD8αα + SP stage interfered with their transition to the DP stage (145). Consistent with their development along an alternative selection pathway, as opposed to being a dead end, the CD4− CD8αα + immature thymocytes express high levels of antiapoptotic factors of the Bcl-2 family, and they lack DNA fragmentation (82). The difficulty in detecting these precursors in the thymus of normal adult mice could be due to very low numbers of these intermediates, or a different kinetics of their selection process with rapid export of newly matured cells.

AGONIST SELECTION AND CD8αα-TL INTERACTION Several factors, including the differential use of signaling molecules and the absence of coreceptor contributions, might influence the TCR signals received by the agonist-selected thymocytes. At the immature DN stage, however, where FcεRIγ

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BIOLOGY OF MUCOSAL T CELLS

P1: IKH

233

is expressed and CD4 and CD8αβ coreceptors are not, other mechanisms must operate to allow for deviation of the immature precursors and maturation along the agonist selection pathway. CD8αα may provide a key for understanding this. The coordinate binding of CD4 or CD8αβ with class II– or class I–restricted TCRs, respectively, and their association with the p56lck tyrosine kinase, are two requirements for efficient positive selection. The coexpression of CD8αα with CD4 or CD8αβ, and the expression of CD8αα on MHC class II–restricted T cells, are inconsistent with a coreceptor function for CD8αα and suggest strongly that CD8αα expression mediates another function. A number of studies have demonstrated that CD8αα is inefficient as a TCR coreceptor, even though it is the α chain of the CD8 molecule that associates with p56lck (146, 147), and it is the CD8α chain ectodomain that interacts with the α3 domain of the heavy chain of class I molecules (148, 149). For example, CD8β −/− mice, which express CD4 and CD8αα at their DP immature thymocytes, do not efficiently select class I– restricted conventional T cells (150, 151). It has been shown that CD4 and CD8αβ associate with p56lck and help sequester it into lipid rafts together with the TCR, thus localizing the kinase in proximity of the TCR/CD3 complex (152). Furthermore, it was demonstrated using CD8α transfection of CD4+ class II–restricted T cells that expression of the coreceptor not involved in the activation complex sequesters active p56lck away from the activation complex, thus down-modulating the TCR-mediated signaling cascade (153). These observations are very consistent with the effects observed when endogenous CD8αα is expressed by T cells. Unlike CD8αβ, CD8αα is not internalized with the TCR from the surface of activated cells (154), and associated p56lck is inefficiently provided to the TCR complex in part because of the inability of CD8αα to effectively merge into lipid rafts. It is CD8β that promotes raft association of CD8 (155), although this can be partially overcome if CD8αα can bind to the antigen-presenting MHC class I molecule. The TL antigen, a nonclassical MHC class I molecule that is abundantly expressed on the epithelial cells of the small intestine in mice (156), is a specific ligand for CD8αα (157). TL, however, is not a typical antigen-presenting molecule, and TL transgene expression in the absence of classical class I expression in TAP−/− mice does not promote the selection of CD8 T cells in the periphery (H. Cheroutre, unpublished data). Furthermore, the crystal structure of TL demonstrates a narrowed and completely occluded groove that cannot present peptides (158). These observations suggest a function for this MHC class I molecule that is different from antigen presentation. Direct evidence for this was provided by demonstrating that the TL interaction with its receptor CD8αα expressed by IEL mediates profound modulation of TCR activation signaling resulting in reduced proliferation and cytotoxicity of the activated T cells while promoting specific cytokine production (157). Furthermore, the TL-CD8αα interaction can mediate survival and further differentiation of the antigen-stimulated T cells (L. Madakamutil & H. Cheroutre, unpublished data). The pronounced expression of this receptor-ligand pair on immature thymocytes besides the IEL suggests a possible function for CD8αα-TL not only in the maintenance and regulation of mature IEL, but also in the development and

11 Feb 2004

17:10

234

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

selection of CD8αα-expressing thymocytes. We speculate that the interaction of CD8αα on immature thymocytes with its ligand TL could mediate the survival during agonist selection in the thymus of CD8αα TCRαβ mucosal T cells.

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

REGULATED IMMUNE RESPONSES IN THE INTESTINE All TCRαβ IEL subsets display functional and phenotypic traits of activated or memory T cells. Using adoptive transfer of na¨ıve TCRαβ transgenic lymph node T cells, it was shown that virus-encoded antigen, but not soluble antigens, could induce the generation of memory T cells in the LP and epithelium of the recipient mice (35–38). The mucosal memory T cells were retained locally and expressed specific adhesion molecules that differ from those expressed by memory T cells generated in the periphery (38). They showed functional characteristics of memory T cells, including rapid recall responses in in vivo and ex vivo lytic activity, which from a protective point of view are critical at interfaces such as the gut epithelium (38). These results therefore suggested that conventional T cells specific for nonself-antigens can differentiate into IEL following activation. These IEL represent a pool of antigen-specific memory T cells that can respond rapidly and effectively upon secondary exposure to their cognate antigen encountered in the intestine. IEL are surrounded by cells with antigen-presenting capability. Enterocytes have been shown to function as antigen-presenting cells (159–162), and DCs are widely spread throughout the LP and the intestinal epithelium. The epithelial DCs are specifically equipped to preserve the integrity of the epithelial barrier by expressing tight junction proteins, including occludin and claudin, which can establish rigid contact with the neighboring epithelial cells while protruding dendrites across the epithelium into the gut lumen (163). Furthermore, the antigendependent migration of conventionally selected T cells to the intestine also could be a protective measure to preserve an intact epithelial layer and prevent undesirable immune responses from nonself-specific na¨ıve T cells against nonpathogenic agents abundantly present in the gut. In this context, it is interesting to note that the intestinal epithelial cells lack significant expression of costimulatory molecules, including the B7-1 and B7-2 molecules, which are required for efficient antigenic stimulation of the na¨ıve T cells (164–166). Unlike the conventionally selected TCRαβ IEL, which have migrated from the periphery after encountering antigens in an organized lymphoid structure, there is no evidence that CD8αα TCRαβ IEL or their immediate precursors have resided in lymph nodes. We propose that agonist selection in the thymus imprints these cells with a “natural” memory phenotype during development, including expression of cell surface molecules typical of mature IEL. For example, these agonist-selected thymocytes can express the mucosal integrin, α Eβ 7, which can mediate binding to E cadherin expressed by the epithelial cells (167–168) and might function to direct these agonist-selected T cells effectively to the intestine and/or tether them to the epithelial cells.

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BIOLOGY OF MUCOSAL T CELLS

P1: IKH

235

The intestine displays a pronounced tone of immune regulation. Nowhere else in the body is there more evidence that the immune system has evolved numerous mechanisms to prevent useless or self-destructive immune responses. The selfspecific IEL subset has been conditioned during agonist selection in the thymus to remain immune quiescent in the antigen-rich environment of the gut. Using T cells from TCR transgenic mice that also expressed the cognate antigen during selection, it was shown that the CD8αα TCR transgenic IEL that express a TCR specific for a self-antigen were not destructive to the epithelium (107). By contrast, in a transfer model of colitis induction, which is induced by the injection of na¨ıve (CD45RBhigh) CD4+ T cells into immune-deficient mice, it was shown that these self-specific CD8αα TCRαβ IEL exert a regulatory role not carried out by other IEL populations (169). These cells controlled the migration to the intestine and immune responses of the pathogenic CD4+ CD45RBhigh T cells, and the regulation by the CD8αα TCRαβ + donor IEL was contingent upon expression and recognition of their specific cognate self-antigen in the intestine of the recipient mice (169). The absence or defect of these self-specific regulatory IEL might be the underlining factor in the predisposition of the animal to uncontrolled immune responses and tissue destruction mediated by conventional T cells, as is observed in several models of experimental colitis. The TCRγ δ IEL were not shown to display suppressive regulatory activity in these transfer experiments, but these cells have been shown to secrete keratinocyte growth factor, which can stimulate repair of damaged epithelium and restore the integrity of the barrier (170). The intestinal epithelial cells themselves play a key role in the immune regulation by expressing distinct surface molecules and restriction elements and secreting factors that allow them to uniquely modulate and differentiate regulatory subsets of mucosal T cells. Immune-modulatory cytokines, in particular TGFβ and IL-10, have been implicated as interrelating key regulatory factors (171–173), with TGFβ serving as the primary anti-inflammatory agent to down-modulate Th1 mucosal immune responses (174). IEL express receptors for growth factors that epithelial cells secrete, including stem cell factor, IL-7, and IL-15. The gut epithelial cells likely also express cell surface molecules that are specific ligands for regulatory receptors expressed by the IEL. This was demonstrated by the interaction of the TL antigen on epithelial cells with CD8αα expressed by the majority of the IEL (157). When CD8αα is engaged by TL, it is able to modulate TCR signals, thereby adapting the immune response of the IEL to an environment where maintaining the integrity of the epithelial barrier is a high priority. Unlike the clonal expansion and cytolytic activity of peripheral activated CD8+ T cells, IEL activated in the presence of the TL-CD8αα interaction show reduced proliferation and cytotoxicity but enhanced cytokine production allowing for protection without destruction of the epithelial cell layer (157). It is likely that other receptor and ligand pairs expressed by IEL and epithelial cells also play an important role in regulating mucosal immune responses. Antibodies to the human biliary glycoprotein (CD66a), a carcinoembryonic Ag-related molecule, could block the cytolytic activity of human IEL (175). This molecule

11 Feb 2004

17:10

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

236

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

is expressed on epithelial cells as well as on IEL, and it can engage in homotypic interactions (176). Cross-linking of the α Eβ 7 integrin, which may provide a mimic of the E cadherin interaction with this integrin, increased the TCR-mediated activation of human IEL (177). IEL, particularly the CD8αα TCRαβ subset, have been shown to express activating and inhibitory NK receptors, including members of the Ly49 family, CD94/NKG2 and NKR-PI (55). Like CD66a in human IEL, some of these receptors have ITIM motifs and therefore could suppress overly exuberant IEL responses. Similar to the TL-CD8αα interaction, and perhaps to some of the other examples mentioned here, it is possible that these NK receptor-ligand interactions also deliver signals that can control and modify TCR activation signals in order to adapt IEL to provide protective functions without epithelial destruction. It is therefore likely that the TL-CD8αα interaction provides a paradigm for these types of lympho-epithelial immune-regulation in the intestine, and we are just at the beginning stages of understanding them.

CONCLUSION Mucosal T cells of higher vertebrates have uniquely evolved to adapt to the gutspecific environment that imposes extremely challenging conditions for defensive mechanisms. Through evolution, the mucosal immune system has diverted immensely to provide rapid and effective protection against innumerable pathogens concurrently with maintaining the integrity of the epithelial barrier, tolerating commensal micro-organisms, and allowing for efficient nutrient absorption. The evolution of the thymus as the organ responsible for αβTCR+ T cell education has relieved the intestine of much of its primordial function in lymphopoiesis. CD8αβ TCRαβ IEL are a subtype of conventional memory T cells reactive to foreign antigens. By contrast, CD8αα TCRαβ IEL are self-reactive natural memory T cells with diverse specificities. Self-specificity is selected for in the thymus—a developmental pathway that may endow these cells with the ability to regulate immune responses. T cell localization and function in the intestine is influenced by selection in the thymus, the site of antigen encounter in the periphery for conventional cells, integrin and chemokine receptor expression, and a dense network of invariant and adaptive receptor-ligand interactions that maintain a continuous communication between the mucosal T cells and their neighboring intestinal epithelial cells. ACKNOWLEDGMENTS I would like to thank Dr. Mitchell Kronenberg for critical reading of this manuscript and for the very enlightening discussions and numerous suggestions that significantly enriched and enhanced this article. I would also like to thank Marieke Cheroutre for her very important contribution to this work, and all my colleagues and especially Drs. Loui Madakamutil, Denise Gangadharan, Raziya Shaikh, and

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

THE BIOLOGY OF MUCOSAL T CELLS

P1: IKH

237

Laurent Gapin for sharing unpublished data. My special thanks also goes to Drs. Madakamutil and Gapin for the design of the figures. The author’s research was supported by NIH grants DK54451 and AI50263. This is manuscript number 595 of the La Jolla Institute for Allergy and Immunology. The Annual Review of Immunology is online at http://immunol.annualreviews.org

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

LITERATURE CITED 1. Rocha B, Vassalli P, Guy-Grand D. 1991. The V beta repertoire of mouse gut homodimeric alpha CD8+ intraepithelial T cell receptor alpha/beta + lymphocytes reveals a major extrathymic pathway of T cell differentiation. J. Exp. Med. 173: 483–86 2. Khalturin K, Becker M, Rinkevich B, Bosch TC. 2003. Urochordates and the origin of natural killer cells: identification of a CD94/NKR-P1-related receptor in blood cells of Botryllus. Proc. Natl. Acad. Sci. USA 100:622–27 3. Flajnik MF. 1998. Churchill and the immune system of ectothermic vertebrates. Immunol. Rev. 166:5–14 4. Du Pasquier L, Wilson M, Greenberg AS, Flajnik MF. 1998. Somatic mutation in ectothermic vertebrates: musings on selection and origins. Curr. Top Microbiol. Immunol. 229:199–216 5. Agrawal A, Eastman QM, Schatz DG. 1998. Transposition mediated by RAG1 and RAG2 and its implications for the evolution of the immune system. Nature 394:744–51 6. Matsunaga T. 1998. Did the first adaptive immunity evolve in the gut of ancient jawed fish? Cytogenet. Cell Genet. 80:138–41 7. Wolgemuth DJ, Behringer RR, Mostoller MP, Brinster RL, Palmiter RD. 1989. Transgenic mice overexpressing the mouse homoeobox-containing gene Hox1.4 exhibit abnormal gut development. Nature 337:464–67 8. Manley NR, Capecchi MR. 1995. The role of Hoxa-3 in mouse thymus and thyroid

9.

10.

11.

12.

13.

14.

15.

16.

development. Development 121:1989– 2003 Ottaviani E, Valensin S, Franceschi C. 1998. The neuro-immunological interface in an evolutionary perspective: the dynamic relationship between effector and recognition systems. Front Biosci. 3:d431–35 Ritter MA, Boyd RL. 1993. Development in the thymus: It takes two to tango. Immunol. Today 14:462–69 Boyd RL, Tucek CL, Godfrey DI, Izon DJ, Wilson TJ, et al. 1993. The thymic microenvironment. Immunol. Today 14:445– 59 Hibi T, Fujisawa T, Kanai T, Akatsuka A, Habu S, et al. 1991. Establishment of epithelial cell lines from human and mouse thymus immortalized by the 12S adenoviral E1a gene product. Thymus 18:155– 67 Lepesant H, Reggio H, Pierres M, Naquet P. 1990. Mouse thymic epithelial cell lines interact with and select a CD3lowCD4 + CD8 + thymocyte subset through an LFA1-dependent adhesion-de-adhesion mechanism. Int. Immunol. 2:1021–32 Farr AG, Rudensky A. 1998. Medullary thymic epithelium: a mosaic of epithelial “self”? J. Exp. Med. 188:1–4 Guy-Grand D, Azogui O, Celli S, Darche S, Nussenzweig MC, et al. 2003. Extrathymic T cell lymphopoiesis: ontogeny and contribution to gut intraepithelial lymphocytes in athymic and euthymic mice. J. Exp. Med. 197:333–41 Maloy KJ, Mowat AM, Zamoyska R, Crispe IN. 1991. Phenotypic

11 Feb 2004

17:10

238

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

17.

18.

19.

20.

21.

22.

23.

24.

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE heterogeneity of intraepithelial T lymphocytes from mouse small intestine. Immunology 72:555–62 Camerini V, Panwala C, Kronenberg M. 1993. Regional specialization of the mucosal immune system. Intraepithelial lymphocytes of the large intestine have a different phenotype and function than those of the small intestine. J. Immunol. 151:1765–76 Beagley KW, Fujihashi K, Lagoo AS, Lagoo-Deenadaylan S, Black CA, et al. 1995. Differences in intraepithelial lymphocyte T cell subsets isolated from murine small versus large intestine. J. Immunol. 154:5611–19 Faure F, Jitsukawa S, Triebel F, Hercend T. 1988. Characterization of human peripheral lymphocytes expressing the CD3-gamma/delta complex with antireceptor monoclonal antibodies. J. Immunol. 141:3357–60 Spencer J, Isaacson PG, MacDonald TT, Thomas AJ, Walker-Smith JA. 1991. Gamma/delta T cells and the diagnosis of coeliac disease. Clin. Exp. Immunol. 85:109–13 Chien YH, Hampl J. 2000. Antigenrecognition properties of murine gamma delta T cells. Springer Semin. Immunopathol. 22:239–50 Morita CT, Mariuzza RA, Brenner MB. 2000. Antigen recognition by human gamma delta T cells: pattern recognition by the adaptive immune system. Springer Semin. Immunopathol. 22:191–217 Schild H, Mavaddat N, Litzenberger C, Ehrich EW, Davis MM, et al. 1994. The nature of major histocompatibility complex recognition by gamma delta T cells. Cell 76:29–37 Steinle A, Groh V, Spies T. 1998. Diversification, expression, and gamma delta T cell recognition of evolutionarily distant members of the MIC family of major histocompatibility complex class I-related molecules. Proc. Natl. Acad. Sci. USA 95:12510–15

25. Groh V, Steinle A, Bauer S, Spies T. 1998. Recognition of stress-induced MHC molecules by intestinal epithelial gamma delta T cells. Science 279:1737– 40 26. Wu J, Groh V, Spies T. 2002. T cell antigen receptor engagement and specificity in the recognition of stress-inducible MHC class I-related chains by human epithelial gamma delta T cells. J. Immunol. 169:1236–40 27. Bauer S, Groh V, Wu J, Steinle A, Phillips JH, et al. 1999. Activation of NK cells and T cells by NKG2D, a receptor for stressinducible MICA. Science 285:727–29 28. Poccia F, Cipriani B, Vendetti S, Colizzi V, Poquet Y, et al. 1997. CD94/NKG2 inhibitory receptor complex modulates both anti-viral and anti-tumoral responses of polyclonal phosphoantigen-reactive V gamma 9V delta 2 T lymphocytes. J. Immunol. 159:6009–17 29. Cerwenka A, Bakker AB, McClanahan T, Wagner J, Wu J, et al. 2000. Retinoic acid early inducible genes define a ligand family for the activating NKG2D receptor in mice. Immunity 12:721–27 30. Diefenbach A, Jamieson AM, Liu SD, Shastri N, Raulet DH. 2000. Ligands for the murine NKG2D receptor: expression by tumor cells and activation of NK cells and macrophages. Nat. Immunol. 1:119– 26 31. Kim SK, Reed DS, Olson S, Schnell MJ, Rose JK, et al. 1998. Generation of mucosal cytotoxic T cells against soluble protein by tissue-specific environmental and costimulatory signals. Proc. Natl. Acad. Sci. USA 95:10814–19 32. Neutra MR, Pringault E, Kraehenbuhl JP. 1996. Antigen sampling across epithelial barriers and induction of mucosal immune responses. Annu. Rev. Immunol. 14:275– 300 33. Arstila T, Arstila TP, Calbo S, Selz F, Malassis-Seris M, et al. 2000. Identical T cell clones are located within the mouse gut epithelium and lamina propia and

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

THE BIOLOGY OF MUCOSAL T CELLS

34.

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

35.

36.

37.

38.

39.

40.

41.

42.

circulate in the thoracic duct lymph. J. Exp. Med. 191:823–34 Lepage AC, Buzoni-Gatel D, Bout DT, Kasper LH. 1998. Gut-derived intraepithelial lymphocytes induce long term immunity against Toxoplasma gondii. J. Immunol. 161:4902–8 Corazza N, Muller S, Brunner T, Kagi D, Mueller C. 2000. Differential contribution of Fas- and perforin-mediated mechanisms to the cell-mediated cytotoxic activity of naive and in vivo-primed intestinal intraepithelial lymphocytes. J. Immunol. 164:398–403 Masopust D, Jiang J, Shen H, Lefrancois L. 2001. Direct analysis of the dynamics of the intestinal mucosa CD8 T cell response to systemic virus infection. J. Immunol. 166:2348–56 Pope C, Kim SK, Marzo A, Masopust D, Williams K, et al. 2001. Organ-specific regulation of the CD8 T cell response to Listeria monocytogenes infection. J. Immunol. 166:3402–9 Kim SK, Schluns KS, Lefrancois L. 1999. Induction and visualization of mucosal memory CD8 T cells following systemic virus infection. J. Immunol. 163:4125–32 Reinhardt RL, Khoruts A, Merica R, Zell T, Jenkins MK. 2001. Visualizing the generation of memory CD4 T cells in the whole body. Nature 410:101–5 Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. 1999. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401:708–12 Wagner N, Lohler J, Kunkel EJ, Ley K, Leung E, et al. 1996. Critical role for beta7 integrins in formation of the gut-associated lymphoid tissue. Nature 382:366–70 Rose JR, Williams MB, Rott LS, Butcher EC, Greenberg HB. 1998. Expression of the mucosal homing receptor alpha 4 beta7 correlates with the ability of CD8+ memory T cells to clear rotavirus infection. J. Virol. 72:726–30

P1: IKH

239

43. Steeber DA, Tang ML, Zhang XQ, Muller W, Wagner N, Tedder TF. 1998. Efficient lymphocyte migration across high endothelial venules of mouse Peyer’s patches requires overlapping expression of L-selectin and beta7 integrin. J. Immunol. 161:6638–47 44. Wurbel MA, Philippe JM, Nguyen C, Victorero G, Freeman T, et al. 2000. The chemokine TECK is expressed by thymic and intestinal epithelial cells and attracts double- and single-positive thymocytes expressing the TECK receptor CCR9. Eur. J. Immunol. 30:262–71 45. Campbell DJ, Butcher EC. 2002. Intestinal attraction: CCL25 functions in effector lymphocyte recruitment to the small intestine. J. Clin. Invest. 110:1079–81 46. Papadakis KA, Prehn J, Nelson V, Cheng L, Binder SW, et al. 2000. The role of thymus-expressed chemokine and its receptor CCR9 on lymphocytes in the regional specialization of the mucosal immune system. J. Immunol. 165:5069–76 47. Papadakis KA, Landers C, Prehn J, Kouroumalis EA, Moreno ST, et al. 2003. CC chemokine receptor 9 expression defines a subset of peripheral blood lymphocytes with mucosal T cell phenotype and Th1 or T-regulatory 1 cytokine profile. J. Immunol. 171:159–65 48. Svensson M, Marsal J, Ericsson A, Carramolino L, Broden T, et al. 2002. CCL25 mediates the localization of recently activated CD8 alpha beta(+) lymphocytes to the small-intestinal mucosa. J. Clin. Invest. 110:1113–21 49. Wurbel MA, Malissen M, Guy-Grand D, Meffre E, Nussenzweig MC, et al. 2001. Mice lacking the CCR9 CC-chemokine receptor show a mild impairment of early T- and B-cell development and a reduction in T-cell receptor gamma delta(+) gut intraepithelial lymphocytes. Blood 98:2626–32 50. Campbell DJ, Butcher EC. 2002. Rapid acquisition of tissue-specific homing phenotypes by CD4(+) T cells activated in

11 Feb 2004

17:10

240

51.

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

52.

53.

54.

55.

56.

57.

58.

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE cutaneous or mucosal lymphoid tissues. J. Exp. Med. 195:135–41 Mora JR, Bono MR, Manjunath N, Weninger W, Cavanagh LL, et al. 2003. Selective imprinting of gut-homing T cells by Peyer’s patch dendritic cells. Nature 424:88–93 Klein JR. 1986. Ontogeny of the Thy1-, Lyt-2+ murine intestinal intraepithelial lymphocyte. Characterization of a unique population of thymus-independent cytotoxic effector cells in the intestinal mucosa. J. Exp. Med. 164:309– 14 Ohteki T, MacDonald HR. 1993. Expression of the CD28 costimulatory molecule on subsets of murine intestinal intraepithelial lymphocytes correlates with lineage and responsiveness. Eur. J. Immunol. 23:1251–55 Van Houten N, Mixter PF, Wolfe J, Budd RC. 1993. CD2 expression on murine intestinal intraepithelial lymphocytes is bimodal and defines proliferative capacity. Int. Immunol. 5:665–72 Guy-Grand D, Cuenod-Jabri B, MalassisSeris M, Selz F, Vassalli P. 1996. Complexity of the mouse gut T cell immune system: identification of two distinct natural killer T cell intraepithelial lineages. Eur. J. Immunol. 26:2248–56 Treiner E, Duban L, Bahram S, Radosavljevic M, Wanner V, et al. 2003. Selection of evolutionarily conserved mucosalassociated invariant T cells by MR1. Nature 422:164–69 Guy-Grand D, Rocha B, Mintz P, Malassis-Seris M, Selz F, et al. 1994. Different use of T cell receptor transducing modules in two populations of gut intraepithelial lymphocytes are related to distinct pathways of T cell differentiation. J. Exp. Med. 180:673–79 Park SY, Arase H, Wakizaka K, Hirayama N, Masaki S, et al. 1995. Differential contribution of the FcR gamma chain to the surface expression of the T cell receptor among T cells localized in epithelia: anal-

59.

60.

61.

62.

63.

64.

65.

66.

67.

ysis of FcR gamma-deficient mice. Eur. J. Immunol. 25:2107–10 Regnault A, Cumano A, Vassalli P, GuyGrand D, Kourilsky P. 1994. Oligoclonal repertoire of the CD8 alpha-alpha and the CD8 alpha-beta TCR-alpha/beta murine intestinal intraepithelial T lymphocytes: evidence for the random emergence of T cells. J. Exp. Med. 180:1345–58 Muller S, Buhler-Jungo M, Mueller C. 2000. Intestinal intraepithelial lymphocytes exert potent protective cytotoxic activity during an acute virus infection. J. Immunol. 164:1986–94 Fichtelius KE. 1968. The gut epithelium—a first level lymphoid organ? Exp. Cell Res. 49:87–104 Ferguson A, Parrott DM. 1972. The effect of antigen deprivation on thymusdependent and thymus-independent lymphocytes in the small intestine of the mouse. Clin. Exp. Immunol. 12:477–88 Mosley RL, Styre D, Klein JR. 1990. Differentiation and functional maturation of bone marrow-derived intestinal epithelial T cells expressing membrane T cell receptor in athymic radiation chimeras. J. Immunol. 145:1369–75 Poussier P, Edouard P, Lee C, Binnie M, Julius M. 1992. Thymus-independent development and negative selection of T cells expressing T cell receptor alpha/beta in the intestinal epithelium: evidence for distinct circulation patterns of gut- and thymus-derived T lymphocytes. J. Exp. Med. 176:187–99 Rocha B, Vassalli P, Guy-Grand D. 1994. Thymic and extrathymic origins of gut intraepithelial lymphocyte populations in mice. J. Exp. Med. 180:681–86 Dunon D, Cooper MD, Imhof BA. 1993. Migration patterns of thymus-derived gamma delta T cells during chicken development. Eur. J. Immunol. 23:2545–50 Dunon D, Cooper MD, Imhof BA. 1993. Thymic origin of embryonic intestinal gamma/delta T cells. J. Exp. Med. 177:257–63

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BIOLOGY OF MUCOSAL T CELLS 68. Lin T, Takimoto H, Matsuzaki G, Nomoto K. 1995. Effect of neonatal thymectomy on murine small intestinal intraepithelial lymphocytes expressing T cell receptor alpha beta and “clonally forbidden V beta s”. Adv. Exp. Med. Biol. A 371:129–31 69. Horton JD, Horton TL, Dzialo R, Gravenor I, Minter R, et al. 1998. Tcell and natural killer cell development in thymectomized Xenopus. Immunol. Rev. 166:245–58 70. Nehls M, Pfeifer D, Schorpp M, Hedrich H, Boehm T. 1994. New member of the winged-helix protein family disrupted in mouse and rat nude mutations. Nature 372:103–7 71. Yoshikai Y, Reis MD, Mak TW. 1986. Athymic mice express a high level of functional gamma-chain but greatly reduced levels of alpha- and beta-chain T-cell receptor messages. Nature 324:482–85 72. Rocha B, Vassalli P, Guy-Grand D. 1992. The extrathymic T-cell development pathway. Immunol. Today 13:449–54 73. Rocha B. 1990. Characterization of V beta-bearing cells in athymic (nu/nu) mice suggests an extrathymic pathway for T cell differentiation. Eur. J. Immunol. 20:919–25 74. Maleckar JR, Sherman LA. 1987. The composition of the T cell receptor repertoire in nude mice. J. Immunol. 138:3873– 76 75. Fry AM, Jones LA, Kruisbeek AM, Matis LA. 1989. Thymic requirement for clonal deletion during T cell development. Science 246:1044–46 76. Howie D, Spencer J, DeLord D, Pitzalis C, Wathen NC, et al. 1998. Extrathymic T cell differentiation in the human intestine early in life. J. Immunol. 161:5862–72 77. Lynch S, Kelleher D, McManus R, O’Farrelly C. 1995. RAG1 and RAG2 expression in human intestinal epithelium: evidence of extrathymic T cell differentiation. Eur. J. Immunol. 25:1143–47 78. Taplin ME, Frantz ME, Canning C, Ritz J, Blumberg RS, Balk SP. 1996. Evidence

79.

80.

81.

82.

83.

84.

85.

86.

87.

P1: IKH

241

against T-cell development in the adult human intestinal mucosa based upon lack of terminal deoxynucleotidyltransferase expression. Immunology 87:402–7 Blumberg RS, Yockey CE, Gross GG, Ebert EC, Balk SP. 1993. Human intestinal intraepithelial lymphocytes are derived from a limited number of T cell clones that utilize multiple V beta T cell receptor genes. J. Immunol. 150:5144–53 Van Kerckhove C, Russell GJ, Deusch K, Reich K, Bhan AK, et al. 1992. Oligoclonality of human intestinal intraepithelial T cells. J. Exp. Med. 175:57–63 Spencer J, MacDonald TT, Finn T, Isaacson PG. 1986. The development of gut associated lymphoid tissue in the terminal ileum of fetal human intestine. Clin. Exp. Immunol. 64:536–43 Takahama Y, Shores EW, Singer A. 1992. Negative selection of precursor thymocytes before their differentiation into CD4+CD8+ cells. Science 258:653– 56 Saito H, Kanamori Y, Takemori T, Nariuchi H, Kubota E, et al. 1998. Generation of intestinal T cells from progenitors residing in gut cryptopatches. Science 280:275–78 Lefrancois L, Puddington L. 1998. Anatomy of T-cell development in the intestine. Gastroenterology 115:1588–91 Kanamori Y, Ishimaru K, Nanno M, Maki K, Ikuta K, et al. 1996. Identification of novel lymphoid tissues in murine intestinal mucosa where clusters of c-kit+ IL7R+ Thy1+ lympho-hemopoietic progenitors develop. J. Exp. Med. 184:1449– 59 Oida T, Suzuki K, Nanno M, Kanamori Y, Saito H, et al. 2000. Role of gut cryptopatches in early extrathymic maturation of intestinal intraepithelial T cells. J. Immunol. 164:3616–26 Sagara S, Sugaya K, Tokoro Y, Tanaka S, Takano H, et al. 1997. B220 expression by T lymphoid progenitor cells in mouse fetal liver. J. Immunol. 158:666–76

11 Feb 2004

17:10

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

242

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

88. Woodward J, Jenkinson E. 2001. Identification and characterization of lymphoid precursors in the murine intestinal epithelium. Eur. J. Immunol. 31:3329–38 89. Porter BO, Malek TR. 1999. IL-2R beta/IL-7R alpha doubly deficient mice recapitulate the thymic and intraepithelial lymphocyte (IEL) developmental defects of gamma c−/− mice: roles for both IL-2 and IL-15 in CD8 alpha alpha IEL development. J. Immunol. 163:5906–12 90. Suzuki H, Duncan GS, Takimoto H, Mak TW. 1997. Abnormal development of intestinal intraepithelial lymphocytes and peripheral natural killer cells in mice lacking the IL-2 receptor beta chain. J. Exp. Med. 185:499–505 91. Lambolez F, Rocha B. 2001. Molecular characterization of gut T cell precursors in euthymic and athymic mice. Adv. Exp. Med. Biol. 495:15–24 92. Lambolez F, Azogui O, Joret AM, Garcia C, von Boehmer H, et al. 2002. Characterization of T cell differentiation in the murine gut. J. Exp. Med. 195:437–49 93. Tsai FY, Keller G, Kuo FC, Weiss M, Chen J, et al. 1994. An early haematopoietic defect in mice lacking the transcription factor GATA-2. Nature 371:221–26 94. Yu W, Nagaoka H, Jankovic M, Misulovin Z, Suh H, et al. 1999. Continued RAG expression in late stages of B cell development and no apparent re-induction after immunization. Nature 400:682–87 95. Zinkernagel RM, Doherty PC. 1974. Immunological surveillance against altered self components by sensitised T lymphocytes in lymphocytic choriomeningitis. Nature 251:547–48 96. Bevan MJ. 1977. In a radiation chimaera, host H-2 antigens determine immune responsiveness of donor cytotoxic cells. Nature 269:417–18 97. Kisielow P, Teh HS, Bluthmann H, von Boehmer H. 1988. Positive selection of antigen-specific T cells in thymus by restricting MHC molecules. Nature 335:730–33

98. Teh HS, Kisielow P, Scott B, Kishi H, Uematsu Y, et al. 1988. Thymic major histocompatibility complex antigens and the alpha beta T-cell receptor determine the CD4/CD8 phenotype of T cells. Nature 335:229–33 99. Kappler JW, Roehm N, Marrack P. 1987. T cell tolerance by clonal elimination in the thymus. Cell 49:273–80 100. Kisielow P, Bluthmann H, Staerz UD, Steinmetz M, von Boehmer H. 1988. Tolerance in T-cell-receptor transgenic mice involves deletion of nonmature CD4+8+ thymocytes. Nature 333:742–46 101. Savage PA, Davis MM. 2001. A kinetic window constricts the T cell receptor repertoire in the thymus. Immunity 14:243–52 102. Starr TK, Jameson SC, Hogquist KA. 2003. Positive and negative selection of T cells. Annu. Rev. Immunol. 21:139–76 103. Cruz D, Sydora BC, Hetzel K, Yakoub G, Kronenberg M, Cheroutre H. 1998. An opposite pattern of selection of a single T cell antigen receptor in the thymus and among intraepithelial lymphocytes. J. Exp. Med. 188:255–65 104. Rocha B, von Boehmer H. 1991. Peripheral selection of the T cell repertoire. Science 251:1225–28 105. Ernst B, Lee DS, Chang JM, Sprent J, Surh CD. 1999. The peptide ligands mediating positive selection in the thymus control T cell survival and homeostatic proliferation in the periphery. Immunity 11:173–81 106. Rocha B, von Boehmer H, Guy-Grand D. 1992. Selection of intraepithelial lymphocytes with CD8 alpha/alpha co-receptors by self-antigen in the murine gut. Proc. Natl. Acad. Sci. USA 89:5336–40 107. Leishman AJ, Gapin L, Capone M, Palmer E, MacDonald HR, et al. 2002. Precursors of functional MHC class I- or class IIrestricted CD8alphaalpha(+) T cells are positively selected in the thymus by agonist self-peptides. Immunity 16:355–64 108. Guehler SR, Bluestone JA, Barrett TA. 1996. Immune deviation of 2C transgenic

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

THE BIOLOGY OF MUCOSAL T CELLS

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

109.

110.

111.

112.

113.

114.

115.

116.

intraepithelial lymphocytes in antigenbearing hosts. J. Exp. Med. 184:493– 503 Levelt CN, de Jong YP, Mizoguchi E, O’Farrelly C, Bhan AK, et al. 1999. Highand low-affinity single-peptide/MHC ligands have distinct effects on the development of mucosal CD8 alpha alpha and CD8 alpha beta T lymphocytes. Proc. Natl. Acad. Sci. USA 96:5628–33 Podd BS, Aberg C, Kudla KL, Keene L, Tobias E, Camerini V. 2001. MHC class I allele dosage alters CD8 expression by intestinal intraepithelial lymphocytes. J. Immunol. 167:2561–68 Guy-Grand D, Pardigon N, Darche S, Lantz O, Kourilsky P, Vassalli P. 2001. Contribution of double-negative thymic precursors to CD8 alpha alpha (+) intraepithelial lymphocytes of the gut in mice bearing TCR transgenes. Eur. J. Immunol. 31:2593–602 Sydora BC, Brossay L, Hagenbaugh A, Kronenberg M, Cheroutre H. 1996. TAPindependent selection of CD8+ intestinal intraepithelial lymphocytes. J. Immunol. 156:4209–16 Neuhaus O, Emoto M, Blum C, Yamamoto S, Kaufmann SH. 1995. Control of thymus-independent intestinal intraepithelial lymphocytes by beta 2microglobulin. Eur. J. Immunol. 25:2332– 39 Fujiura Y, Kawaguchi M, Kondo Y, Obana S, Yamamoto H, et al. 1996. Development of CD8 alpha alpha+ intestinal intraepithelial T cells in beta 2-microglobulinand/or TAP1-deficient mice. J. Immunol. 156:2710–15 Holcombe HR, Castano AR, Cheroutre H, Teitell M, Maher JK, et al. 1995. Nonclassical behavior of the thymus leukemia antigen: peptide transporter-independent expression of a nonclassical class I molecule. J. Exp. Med. 181:1433–43 Rodgers JR, Mehta V, Cook RG. 1995. Surface expression of beta 2-microglobulin-associated thymus-leukemia

117.

118.

119.

120.

121.

122.

123.

124.

125.

P1: IKH

243

antigen is independent of TAP2. Eur. J. Immunol. 25:1001–7 Gapin L, Cheroutre H, Kronenberg M. 1999. Cutting edge: TCR alpha beta+ CD8 alpha alpha+ T cells are found in intestinal intraepithelial lymphocytes of mice that lack classical MHC class I molecules. J. Immunol. 163:4100–4 Park SH, Guy-Grand D, Lemonnier FA, Wang CR, Bendelac A, Jabri B. 1999. Selection and expansion of CD8 alpha/alpha(1) T cell receptor alpha/beta(1) intestinal intraepithelial lymphocytes in the absence of both classical major histocompatibility complex class I and nonclassical CD1 molecules. J. Exp. Med. 190:885–90 Das G, Janeway CA Jr. 1999. Development of CD8 alpha/alpha and CD8 alpha/beta T cells in major histocompatibility complex class I-deficient mice. J. Exp. Med. 190:881–84 Tabaczewski P, Stroynowski I. 1994. Expression of secreted and glycosylphosphatidylinositol-bound Qa-2 molecules is dependent on functional TAP-2 peptide transporter. J. Immunol. 152: 5268–74 Das G, Gould DS, Augustine MM, Fragoso G, Sciutto E, et al. 2000. Qa-2dependent selection of CD8 alpha/alpha T cell receptor alpha/beta(+) cells in murine intestinal intraepithelial lymphocytes. J. Exp. Med. 192:1521–28 Lin T, Matsuzaki G, Yoshida H, Kenai H, Omoto K, et al. 1996. Thymus ontogeny and the development of TCR alpha beta intestinal intraepithelial lymphocytes. Cell Immunol. 171:132–39 von Boehmer H, Teh HS, Kisielow P. 1989. The thymus selects the useful, neglects the useless and destroys the harmful. Immunol. Today 10:57–61 Buch T, Rieux-Laucat F, Forster I, Rajewsky K. 2002. Failure of HY-specific thymocytes to escape negative selection by receptor editing. Immunity 16:707–18 Jordan MS, Boesteanu A, Reed AJ,

11 Feb 2004

17:10

244

126.

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

127.

128.

129.

130.

131.

132.

133.

134.

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE Petrone AL, Holenbeck AE, et al. 2001. Thymic selection of CD4+CD25+ regulatory T cells induced by an agonist selfpeptide. Nat. Immunol. 2:301–6 Bendelac A. 1995. Positive selection of mouse NK1+ T cells by CD1expressing cortical thymocytes. J. Exp. Med. 182:2091–96 Legendre V, Boyer C, Guerder S, Arnold B, Hammerling G, Schmitt-Verhulst AM. 1999. Selection of phenotypically distinct NK1.1+ T cells upon antigen expression in the thymus or in the liver. Eur. J. Immunol. 29:2330–43 Capone M, Troesch M, Eberl G, Hausmann B, Palmer E, MacDonald HR. 2001. A critical role for the T cell receptor alphachain connecting peptide domain in positive selection of CD1-independent NKT cells. Eur. J. Immunol. 31:1867–75 Backstrom BT, Muller U, Hausmann B, Palmer E. 1998. Positive selection through a motif in the alphabeta T cell receptor. Science 281:835–38 Werlen G, Hausmann B, Palmer E. 2000. A motif in the alphabeta T-cell receptor controls positive selection by modulating ERK activity. Nature 406:422–26 Derbinski J, Schulte A, Kyewski B, Klein L. 2001. Promiscuous gene expression in medullary thymic epithelial cells mirrors the peripheral self. Nat. Immunol. 2:1032– 39 Anderson MS, Venanzi ES, Klein L, Chen Z, Berzins SP, et al. 2002. Projection of an immunological self shadow within the thymus by the AIRE protein. Science 298:1395–401 Liston A, Lesage S, Wilson J, Peltonen L, Goodnow CC. 2003. AIRE regulates negative selection of organ-specific T cells. Nat. Immunol. 4:350–54 Zuklys S, Balciunaite G, Agarwal A, Fasler-Kan E, Palmer E, Hollander GA. 2000. Normal thymic architecture and negative selection are associated with AIRE expression, the gene defective in the autoimmune-polyendocrinopathy-

135.

136.

137.

138.

139.

140.

141.

142.

143.

candidiasis-ectodermal dystrophy (APECED). J. Immunol. 165:1976–83 Ramsey C, Winqvist O, Puhakka L, Halonen M, Moro A, et al. 2002. AIRE deficient mice develop multiple features of APECED phenotype and show altered immune response. Hum. Mol. Genet. 11:397–409 Matsuda JL, Gapin L, Baron JL, Sidobre S, Stetson DB, et al. 2003. Mouse V alpha 14i natural killer T cells are resistant to cytokine polarization in vivo. Proc. Natl. Acad. Sci. USA 100:8395–400 Elewaut D, Shaikh RB, Hammond KJ, De Winter H, Leishman AJ, et al. 2003. NIK-dependent RelB activation defines a unique signaling pathway for the development of V alpha 14i NKT cells. J. Exp. Med. 197:1623–33 Sivakumar V, Hammond KJ, Howells N, Pfeffer K, Weih F. 2003. Differential requirement for Rel/nuclear factor kappa B family members in natural killer T cell development. J. Exp. Med. 197:1613–21 DeKoning J, DiMolfetto L, Reilly C, Wei Q, Havran WL, Lo D. 1997. Thymic cortical epithelium is sufficient for the development of mature T cells in relB-deficient mice. J. Immunol. 158:2558–66 Valero R, Baron ML, Guerin S, Beliard S, Lelouard H, et al. 2002. A defective NFkappa B/RelB pathway in autoimmuneprone New Zealand black mice is associated with inefficient expansion of thymocyte and dendritic cells. J. Immunol. 169:185–92 Malissen M, Gillet A, Rocha B, Trucy J, Vivier E, et al. 1993. T cell development in mice lacking the CD3-zeta/eta gene. EMBO J. 12:4347–55 Simpson S, Hollander G, She J, Levelt C, Huang M, Terhorst C. 1995. Selection of peripheral and intestinal T lymphocytes lacking CD3 zeta. Int. Immunol. 7:287– 93 Rodewald HR, Moingeon P, Lucich JL, Dosiou C, Lopez P, Reinherz EL. 1992. A population of early fetal thymocytes

11 Feb 2004

17:10

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

THE BIOLOGY OF MUCOSAL T CELLS

144.

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

145.

146.

147.

148.

149.

150.

151.

152.

expressing Fc gamma RII/III contains precursors of T lymphocytes and natural killer cells. Cell 69:139–50 Liu CP, Kappler JW, Marrack P. 1996. Thymocytes can become mature T cells without passing through the CD4+ CD8+, double-positive stage. J. Exp. Med. 184:1619–30 Hunig T. 1988. Cross-linking of the T cell antigen receptor interferes with the generation of CD4+8+ thymocytes from their immediate CD4−8+ precursors. Eur. J. Immunol. 18:2089–92 Turner JM, Brodsky MH, Irving BA, Levin SD, Perlmutter RM, Littman DR. 1990. Interaction of the unique N-terminal region of tyrosine kinase p56lck with cytoplasmic domains of CD4 and CD8 is mediated by cysteine motifs. Cell 60:755– 65 Shaw AS, Chalupny J, Whitney JA, Hammond C, Amrein KE, et al. 1990. Short related sequences in the cytoplasmic domains of CD4 and CD8 mediate binding to the amino-terminal domain of the p56lck tyrosine protein kinase. Mol. Cell Biol. 10:1853–62 Gao GF, Jakobsen BK. 2000. Molecular interactions of coreceptor CD8 and MHC class I: the molecular basis for functional coordination with the T-cell receptor. Immunol. Today 21:630–36 Norment AM, Salter RD, Parham P, Engelhard VH, Littman DR. 1988. Cellcell adhesion mediated by CD8 and MHC class I molecules. Nature 336:79– 81 Crooks ME, Littman DR. 1994. Disruption of T lymphocyte positive and negative selection in mice lacking the CD8 beta chain. Immunity 1:277–85 Fung-Leung WP, Kundig TM, Ngo K, Panakos J, De Sousa-Hitzler J, et al. 1994. Reduced thymic maturation but normal effector function of CD8+ T cells in CD8 beta gene-targeted mice. J. Exp. Med. 180:959–67 Wallace VA, Penninger J, Mak TW. 1993.

153.

154.

155.

156.

157.

158.

159.

160.

161.

P1: IKH

245

CD4, CD8 and tyrosine kinases in thymic selection. Curr. Opin. Immunol. 5:235–40 Miceli MC, von Hoegen P, Parnes JR. 1991. Adhesion versus coreceptor function of CD4 and CD8: role of the cytoplasmic tail in coreceptor activity. Proc. Natl. Acad. Sci. USA 88:2623–27 Cawthon AG, Alexander-Miller MA. 2002. Optimal colocalization of TCR and CD8 as a novel mechanism for the control of functional avidity. J. Immunol. 169:3492–98 Arcaro A, Gregoire C, Boucheron N, Stotz S, Palmer E, et al. 2000. Essential role of CD8 palmitoylation in CD8 coreceptor function. J. Immunol. 165:2068–76 Hershberg R, Eghtesady P, Sydora B, Brorson K, Cheroutre H, et al. 1990. Expression of the thymus leukemia antigen in mouse intestinal epithelium. Proc. Natl. Acad. Sci. USA 87:9727–31 Leishman AJ, Naidenko OV, Attinger A, Koning F, Lena CJ, et al. 2001. T cell responses modulated through interaction between CD8alphaalpha and the nonclassical MHC class I molecule, TL. Science 294:1936–39 Liu Y, Xiong Y, Naidenko OV, Liu JH, Zhang R, et al. 2003. The crystal structure of a TL/CD8alpha alpha complex at ˚ resolution: implications for mod2.1 A ulation of T cell activation and memory. Immunity 18:205–15 Campbell N, Yio XY, So LP, Li Y, Mayer L. 1999. The intestinal epithelial cell: processing and presentation of antigen to the mucosal immune system. Immunol. Rev. 172:315–24 Hershberg RM, Cho DH, Youakim A, Bradley MB, Lee JS, et al. 1998. Highly polarized HLA class II antigen processing and presentation by human intestinal epithelial cells. J. Clin. Invest. 102:792–803 Hershberg RM, Framson PE, Cho DH, Lee LY, Kovats S, et al. 1997. Intestinal epithelial cells use two distinct pathways for HLA class II antigen processing. J. Clin. Invest. 100:204–15

11 Feb 2004

17:10

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

246

AR

AR210-IY22-08.tex

AR210-IY22-08.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEROUTRE

162. Hershberg RM, Mayer LF. 2000. Antigen processing and presentation by intestinal epithelial cells—polarity and complexity. Immunol. Today 21:123–28 163. Rescigno M, Urbano M, Valzasina B, Francolini M, Rotta G, et al. 2001. Dendritic cells express tight junction proteins and penetrate gut epithelial monolayers to sample bacteria. Nat. Immunol. 2:361–67 164. Sanderson IR, Ouellette AJ, Carter EA, Walker WA, Harmatz PR. 1993. Differential regulation of B7 mRNA in enterocytes and lymphoid cells. Immunology 79:434– 38 165. Framson PE, Cho DH, Lee LY, Hershberg RM. 1999. Polarized expression and function of the costimulatory molecule CD58 on human intestinal epithelial cells. Gastroenterology 116:1054–62 166. Byrne B, Madrigal-Estebas L, McEvoy A, Carton J, Doherty DG, et al. 2002. Human duodenal epithelial cells constitutively express molecular components of antigen presentation but not costimulatory molecules. Hum. Immunol. 63:977– 86 167. Cepek KL, Shaw SK, Parker CM, Russell GJ, Morrow JS, et al. 1994. Adhesion between epithelial cells and T lymphocytes mediated by E-cadherin and the alpha E beta 7 integrin. Nature 372:190–93 168. Schon MP, Arya A, Murphy EA, Adams CM, Strauch UG, et al. 1999. Mucosal T lymphocyte numbers are selectively reduced in integrin alpha E (CD103)deficient mice. J. Immunol. 162:6641–49 169. Poussier P, Ning T, Banerjee D, Julius M. 2002. A unique subset of self-specific intraintestinal T cells maintains gut integrity. J. Exp. Med. 195:1491–97

170. Boismenu R, Havran WL. 1994. Modulation of epithelial cell growth by intraepithelial gamma delta T cells. Science 266:1253–55 171. Asseman C, Mauze S, Leach MW, Coffman RL, Powrie F. 1999. An essential role for interleukin 10 in the function of regulatory T cells that inhibit intestinal inflammation. J. Exp. Med. 190:995– 1004 172. Asseman C, Powrie F. 1998. Interleukin 10 is a growth factor for a population of regulatory T cells. Gut 42:157–58 173. Annacker O, Asseman C, Read S, Powrie F. 2003. Interleukin-10 in the regulation of T cell-induced colitis. J. Autoimmun. 20:277–79 174. Fuss IJ, Boirivant M, Lacy B, Strober W. 2002. The interrelated roles of TGF-beta and IL-10 in the regulation of experimental colitis. J. Immunol. 168:900–8 175. Morales VM, Christ A, Watt SM, Kim HS, Johnson KW, et al. 1999. Regulation of human intestinal intraepithelial lymphocyte cytolytic function by biliary glycoprotein (CD66a). J. Immunol. 163:1363– 70 176. Watt SM, Teixeira AM, Zhou GQ, Doyonnas R, Zhang Y, et al. 2001. Homophilic adhesion of human CEACAM1 involves N-terminal domain interactions: structural analysis of the binding site. Blood 98:1469–79 177. Sarnacki S, Begue B, Buc H, Le Deist F, Cerf-Bensussan N. 1992. Enhancement of CD3-induced activation of human intestinal intraepithelial lymphocytes by stimulation of the beta 7-containing integrin defined by HML-1 monoclonal antibody. Eur. J. Immunol. 22:2887–92

Cheroutre.qxd

3/12/2004

1:55 PM

Page 1

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BIOLOGY OF MUCOSAL T CELLS

C-1

Figure 1 IEL are a diverse population of T cells. IEL consist of three major subsets: the TCRgd T cells that are thymus derived and develop along the DN pathway; the conventional CD4 or CD8ab TCRab IEL that have matured in the thymus along the conventional selection pathway and migrate as antigen-experienced T cells to the intestine; and the CD8aa TCRab IEL that have matured and differentiated in the thymus along the agonist-selection pathway and independent of coreceptor expression. The agonistselection process directs their migration as self-antigen-experienced T cells to the intestine.

Cheroutre.qxd

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

3/12/2004

1:55 PM

Page 2

CHEROUTRE

Figure 2 Thymic selection is determined by TCR signal strength. Conventional positive selection occurs under conditions of intermediate TCR signal strength received upon recognition of self-MHC and self-antigen. Agonist selection occurs under strong signal strength conditions that otherwise delete thymocytes along the conventional negative selection pathway. Conventional selected mature T cells migrate to the intestine upon agonistic nonself activation signals received in the periphery.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:217-246. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104753

Annu. Rev. Immunol. 2004. 22:247–306 doi: 10.1146/annurev.immunol.22.012703.104753 c 2004 by Annual Reviews. All rights reserved Copyright °

THE BCR-ABL STORY: Bench to Bedside and Back Stephane Wong1 and Owen N. Witte2 Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

1

Molecular Biology Interdepartmental PhD Program/UCLA, 2Howard Hughes Medical Institute/David Geffen School of Medicine at UCLA, Los Angeles, California 90095-1662; email: [email protected]; [email protected]

Key Words tyrosine kinase, leukemia, Imatinib mesylate, inhibitor mechanism, signal transduction ■ Abstract The twenty-first century is beginning with a sharp turn in the field of cancer therapy. Molecular targeted therapies against specific oncogenic events are now possible. The BCR-ABL story represents a notable example of how research from the fields of cytogenetics, retroviral oncology, protein phosphorylation, and small molecule chemical inhibitors can lead to the development of a successful molecular targeted therapy. Imatinib mesylate (Gleevec, STI571, or CP57148B) is a direct inhibitor of ABL (ABL1), ARG (ABL2), KIT, and PDGFR tyrosine kinases. This drug has had a major impact on the treatment of chronic myelogenous leukemia (CML) as well as other blood neoplasias and solid tumors with etiologies based on activation of these tyrosine kinases. Analysis of CML patients resistant to BCR-ABL suppression by Imatinib mesylate coupled with the crystallographic structure of ABL complexed to this inhibitor have shown how structural mutations in ABL can circumvent an otherwise potent anticancer drug. The successes and limitations of Imatinib mesylate hold general lessons for the development of alternative molecular targeted therapies in oncology.

IDENTIFICATION OF THE CHROMOSOMAL ABNORMALITY AND CELLULAR ORIGIN OF THE HUMAN DISEASE CHRONIC MYELOGENOUS LEUKEMIA (CML) The Philadelphia (Ph) Chromosome is Associated with CML In 1845, the pathologists Bennet, Craigie, and Virchow independently described the disease chronic myelogenous leukemia (CML) (1–3). In 1960, a major clue to the cause of CML was provided by Nowell & Hungerford’s landmark discovery of the Philadelphia (Ph) chromosome and its association with the disease CML (4). They examined leukemic cells from chronic phase CML and other leukemia patients. Surprisingly, the cells of all seven CML patients showed a consistent “minute chromosome” abnormality, which they named the Philadelphia (Ph) chromosome (4). Other leukemias did not show any chromosomal abnormalities (4). 0732-0582/04/0423-0247$14.00

247

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

248

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

Figure 1 Timeline of major ABL oncogenesis findings.

This discovery was the first demonstration of a chromosomal rearrangement being consistently linked to a specific cancer (Figure 1) and sparked searches for associations of additional chromosomal aberrations with specific forms of cancer. Numerous abnormalities have been subsequently identified and associated with cancer (reviewed in 5). In retrospect, it was fortunate that Nowell & Hungerford analyzed mainly chronic phase CML patient samples for chromosomal abnormalities, where the Ph chromosome is now known to be present in over 90% of CML cases (6). Had they analyzed samples from patients with adult B-ALL, where the Ph chromosome accounts for only 10% to 15% of such cases, the association of a specific chromosomal abnormality with a specific disease might not have been as apparent (7). Chronic phase CML has a consistent, relatively indolent presentation in patients, with an increase in immature and mature myeloid elements and retention of hematopoietic differentiation (8). This unique and consistent phenotype of chronic phase CML enabled Nowell & Hungerford to easily and accurately identify these patients. The phenotypes of accelerated and blast crisis CML is much more diverse and aggressive than chronic phase CML. Both accelerated and blast phases are characterized by a severe reduction in cellular differentiation, with a displacement of mature cells by immature blasts (8). In blast crisis, more than 50% of patients enter a myeloid blast stage resembling acute myeloblastic leukemia (AML) (8). A pre-B blast stage similar to acute lymphoid leukemia (B-ALL) accounts for 30% of patients (8), and erythroid blasts develop in 10% of patients (8). Rarely do T cell blasts (T-ALL) evolve (9, 10). The pleiotropic and aggressive phenotypes of blast crisis suggest that different oncogenic abnormalities could lead to the transition of chronic phase CML to blast phase CML. Numerous oncogenic events have been associated with blast crisis; these include trisomy 8, i(17q) (11, 12), loss of p53 function (13, 14), MYC amplification (12), RB (RB1) deletion/rearrangement (15), and p16INK4A (CDKN2) rearrangement/deletion (16). Loss of p53 function has been exclusively linked with

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

P1: IKH

249

myeloid and not lymphoid blast crisis, which suggests that p53 signaling is critical for the regulation of normal myelopoiesis (13). Over 80% of blast phase CML cases have definable genetic aberrations in addition to the Ph chromosome (11).

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Philadelphia Chromosome is Present in the Hematopoietic Stem Cell The transition from a predominantly myeloid disease in chronic phase CML to alternative lineages in blast crisis raised the question of which cell types contained the Ph chromosome. Fialkow and colleagues examined different cell types in chronic phase CML patients for the presence of the Ph chromosome (17, 18). Surprisingly, both granulocytes and erythroid lineage cells from chronic phase CML patients contained the Ph chromosome, even though only myeloid cells are expanded during chronic phase CML (17, 18). The presence of the Ph chromosome in granulocyte and erythroid lineages suggested that the Ph chromosome is either generated in multiple cell types or originates in a hematopoietic stem cell (HSC) from which it is passed down to other cell lineages. Fialkow and colleagues addressed the cellular origin of Ph+ cells by examining whether different lineage Ph+ cells were of polyclonal or monoclonal origin. Female CML patients heterozygous for the Glucose-6-Phosphate Dehydrogenase (G-6-PD) isoenzymes were examined for expression of a specific G-6-PD form in Ph+ cells. During the process of X-chromosome inactivation during embyrogenesis, one X chromosome is inactivated in each HSC, resulting in expression of a single G-6-PD isoenzyme type per HSC. Fialkow and colleagues demonstrated that individual CML patients expressed the same G-6-PD isoenzyme in Ph chromosome-positive granulocytic and erythroid lineage cells, whereas nonleukemic patient granulocytes expressed both enzymes (17, 18), substantiating the clonal and stem cell origin of CML (Figure 1). Subsequent purification of HSCs and different lineage–restricted cells from CML patients by cell surface markers has confirmed the presence of the Ph chromosome in the HSC (reviewed in 19, 20). The discovery of a clonal HSC origin of CML suggested that elimination of Ph+ HSCs and replacement of these cells with normal HSCs should be an effective therapy. Currently, this type of treatment regimen, using chemotherapy followed by bone marrow transplantation, is the only curative procedure for CML (reviewed in 21).

DISCOVERY OF A NOVEL TYROSINE KINASE ACTIVITY REQUIRED FOR ABELSON MURINE LEUKEMIA VIRUS (A-MULV)-INDUCED CELLULAR TRANSFORMATION AND LEUKEMOGENESIS A-MuLV Generates Lymphosarcomas Our understanding of the oncogenic activity generated from the Ph chromosome unexpectedly came in large part from studies of the Abelson murine leukemia virus

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

250

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

(A-MuLV). Inoculation of the Moloney Leukemia Virus (M-MuLV) into neonatal mice had been reported to generate thymomas (reviewed in 22). To determine whether this virus was capable of generating lymphomas of other cellular tropisms, Abelson & Rabstein destroyed the T cell compartment of mice with the steroid prednisolone prior to injecting animals with M-MuLV (23; reviewed in 24). Of 153 MLV-infected mice, one developed a distinct type of lymphosarcoma (23; reviewed in 24). This Abelson murine leukemia virus (A-MuLV) reproducibly generated thymic-independent lymphomas (23, 25, 26). Differences between A-MuLV and M-MuLV were noted by subsequent in vitro studies in which A-MuLV but not M-MuLV transformed NIH 3T3 cells (27) and lymphoid cells (28) (Figure 1). The direct in vitro B lymphoid transformation system developed by Rosenberg & Baltimore has proven exceptionally important for the studies of hematopoietic transformation and immune system development.

A-MuLV Expresses a Unique Chimeric Fusion Gene Product Gag-Abl To identify the unique transforming components in A-MuLV not present in MMuLV, the nucleic acid composition of both viruses were compared. Heteroduplex and endonuclease restriction digest analyses of A-MuLV RNA hybridized with MMuLV DNA indicated that A-MuLV encoded 50 and 30 sequences identical to the M-MuLV genome and a unique 3.6-kB fragment not present M-MuLV (29). Using this 3.6-kB insert sequence as a cDNA probe, it was demonstrated that the unique coding region of A-MuLV was cellular in origin (29). In parallel with viral genomic analysis, the unique protein components in AMuLV were analyzed using tryptic peptide analysis of [35S] methionine-labeled AMuLV and M-MuLV proteins and antibodies immunoreactive against major viral proteins. A 120-kD A-MuLV polyprotein contained p15, p12, and part of p30 Gag proteins in common with M-MuLV (30–32). The remaining 80–100-kD protein component in A-MuLV was unique (30, 31). To identify the unknown component in A-MuLV, sera specific for A-MuLV but not M-MuLV were obtained from mice that showed tumor regression upon A-MuLV transformed cell line challenge (reviewed in 24; 33). The tumor regressor sera reacted with the C terminus of A-MuLV, whereas the N terminus of A-MuLV reacted with Gag antibodies (33–35). The same tumor regressor sera cross-reacted with a normal cellular protein NCP 150, later named c-Abl (33, 36). These results demonstrated that A-MuLV encoded a chimeric fusion gene product Gag-Abl with the N-terminal Gag protein originating from M-MuLV and the C-terminal Abl protein being of cellular origin (Figures 1, 2A). Subsequently, A-MuLV was shown to encode a 160-kD chimeric protein. The initially identified 120-kD fusion protein contained an in-frame 789 base pairs (263 amino acids) deletion (Figure 2A). Both proteins were equally potent at transforming cells (reviewed in 24).

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

P1: IKH

251

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Tyrosine Kinase Activity is Essential for Gag-Abl Cellular Transformation The connection of the field of protein kinases and its significance to oncogenic transformation came from a critical observation by Collett & Erikson demonstrating that the protein kinase activity of the Rous Sarcoma Virus (RSV) SRC protein was connected to its transforming potency (37). This finding led to the question of whether other oncoproteins such as Gag-Abl possessed protein kinase activity. Immunoprecipitates of Gag-Abl proteins were shown to possess kinase activity and the P120 Gag-Abl protein itself was autophosphorylated (38, 39). Furthermore, of several variant strains of A-MuLV, the only strain (P92td) that was biologically inactive for cellular transformation had no detectable kinase activity (40). This result linked genetic variations in the Gag-Abl gene product to changes in enzyme activity and biological potency. In the late 1970s, cellular kinase activities were known to predominantly utilize serine and theonine residues as phosphate acceptors (41). However, in vitro autophosphorylated Gag-Abl had undetectable levels of phospho-serine and phosphothreonine (42). Alternative Gag-Abl protein phosphorylations on lysine, histidine, and acidic residues were ruled out based on chemical stabilities and reactivities (42). Further, no evidence of a protein–nucleic acid linkage was observed (42). Evaluation of Gag-Abl as a tyrosine-specific kinase was influenced from two types of prior studies. First, poliovirus RNA was shown to be linked to a small protein via a phosphoester linkage through a tyrosine residue (43), and second, the DNA topoisomerase (ω protein) involved in unwinding DNA superhelical turns was shown to utilize a phosphoester linkage to a tyrosine residue as a transient intermediate in the reaction between enzyme and DNA (44). Immunoprecipitated Gag-Abl autophosphorylated in vitro with 32γ -ATP and Gag-Abl harvested from cells labeled with 32P-orthophosphate were shown to contain phosphotyrosine when protein samples were hydroloyzed in acid at 110◦ C for only several hours instead of the customary 24-h period (42) (Figure 1). Phosphotyrosine has a limited half-life in acid at high temperatures, and extended hydrolysis converted the label to free phosphate. Hunter & Sefton independently discovered that the protein kinase activity associated with pp60Src of RSV specifically phosphorylated tyrosine residues (45). These combined observations led to the concept that tyrosine kinase activity might regulate both leukemias and solid tumors induced by oncogenic retroviruses. Aberrant tyrosine kinase regulation has been subsequently linked with cancer development in many systems, validating these early findings that deregulated tyrosine kinase activity is associated with malignant transformation (reviewed in 5). After the initial discovery of Gag-Abl tyrosine kinase activity, different experimental approaches provided further support for ABL tyrosine kinase activity being critical to Gag-Abl–induced cellular transformation. Within the next two years, it was shown that transformation-defective mutants of Gag-Abl, similar to its cellular homolog c-Abl (Abl NCP150), lacked the heightened tyrosine kinase activity

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

24 Feb 2004 19:26

252

AR

WONG

AR210-IY22-09.tex

¥

WITTE

AR210-IY22-09.sgm LaTeX2e(2002/01/18) P1: IKH

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

P1: IKH

253

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

present in transforming variants of Gag-Abl (34, 35). Subsequent improvements in kinase activity assays led to the discovery that c-Abl also possesses tyrosine kinase activity, albeit at a much lower specific activity than Gag-Abl (46). Random amino acid insertions into the tyrosine kinase domain, but not other regions of Gag-Abl, were shown to abrogate tyrosine kinase activity and cellular transformation (47). In addition, temperature-sensitive GAG-ABL mutants at nonpermissive temperatures were unable to transform NIH 3T3 and lymphoid cells due to inhibited tyrosine ←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 2 Generation of chimeric ABL oncogenes. (A) Generation of A-MuLV. The exact mechanism of A-MuLV formation is not clear. One possible mechanism involves recombination between M-MuLV viral RNA and Abl RNA via one or multiple steps to eventually generate the A-MuLV retrovirus. Another possible mechanism involves the integration of the M-MuLV genome into the abl genome at exon 3 to generate the replication defective Gag-Abl. Gag-Abl encodes a myristoylated fusion protein containing N-terminal p15, p12, and 21 amino acids of p30 from the M-MuLV Gag sequences combined with Abl exonic sequences starting from exon 3 onwards encoding the SH2 = src homology domain 2, SH1 = src homology tyrosine kinase domain, NLS = nuclear localization domains, DNA = DNA binding sites, and ACTIN = F and G actin binding sites. The A-MuLV genome encodes a 120-kD or 160-kD chimeric protein, both shown to contain elevated ABL tyrosine kinase activity and to induce cellular transformation, with the smaller initially identified form containing an inframe deletion of 789 base pairs (263 amino acids) in the Abl sequence of A-MuLV. U5 = unique to 50 sequences; U3 = unique to 30 sequences; R = repeated sequence. (B) Locations of breakpoints in BCR and ABL genes and structure of derived chimeric proteins. Three breakpoint regions within the BCR genome are responsible for generating the predominant BCR-ABL fusion proteins. The minor breakpoint cluster (m-bcr) spans 54-kb and results in an e1a2 7.0 mRNA that generates p185BCR-ABL. The major breakpoint cluster (M-bcr) spans 5.8 kb and results in either a b2a2 or b3a2 8.5 kb mRNA producing p210BCR-ABL. A third breakpoint located at the 30 end of the gene (µ-bcr) generates a c3a2 9.0 kb mRNA forming p230BCR-ABL. Regardless of the exact breakpoint in ABL, BCR sequences are most often fused to ABL exon a2 in the hybrid transcript. BCR protein domains include: OLIGO = oligomerization; A and B = A and B boxes; S/TKINASE = serine/threonine kinase; DBL/CDC24 = Dbl homology; PH = Pleckstrin homology; RACGAP = Rac GTPase. ABL protein domains include MYR = myristoylation signal; CAP = CAP hydrophobic residues; SH3 = src homology 3; and remaining ABL domains as described in Figure 2A. (C) Two forms of TEL-ABL are generated in human leukemias. TEL-ABL containing the first 4 exons of TEL and exon 2 of ABL onwards generates a p145TEL-ABLfusion protein encoding the first 154 amino acids of TEL and 1104 residues of ABL. TEL-ABL containing the first 5 exons of TEL and exon 2 of ABL onwards generates a p180TEL-ABL fusion protein encoding the first 336 amino acids of TEL and 1104 residues of ABL. Both chimeric proteins encode the SAM/PNT = SAM/pointed helix-loop-helix oligomerization domain of TEL and identical ABL domains to BCR-ABL (Figure 3B).

24 Feb 2004

19:26

254

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

kinase activity (48). However, it was still unclear whether the murine Gag-Abl oncoprotein and its biological transforming activity related to human cancer.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IDENTIFICATION OF THE GENETIC PRODUCT OF THE HUMAN PH CHROMOSOME AND ITS UNEXPECTED RELATIONSHIP TO THE MURINE GAG-ABL ONCOPROTEIN The Ph Chromosome Gene Product, BCR-ABL, Possesses Deregulated Tyrosine Kinase Activity Diverse experimental approaches were crucial to the identification of the Ph chromosome product. The advent of quinacrine fluorescence and Giemsa banding enabled Rowley and colleagues to show that the Ph chromosome resulted from a reciprocal translocation between the long arms of chromosomes 9 and 22 t(9;22)(q34;q11) (49). Another 9 years passed after the identification of the chromosome arms involved in the Ph chromosomal translocation before key studies from several investigators including Heisterkamp, Groffen, Stephenson, and Canaani, led to the definitive characterization of the Ph chromosome structure and its mRNA product (Figures 1 and 2B). Chromosome mapping studies demonstrated that the human ABL gene mapped to chromosome 9 (50). This gene was shown to be translocated to the Ph chromosome in CML cells (51–53). An ABL probe reacted with an ∼8-kb mRNA present in Ph-positive but not in control cells, suggesting that the ABL gene is part of the Ph chromosome product (54). Work from our laboratory independently identified ABL as part of the protein product of the Ph chromosome. ABL is highly conserved among species and ABLantisera generated from murine Gag-Abl studies effectively cross-reacted with human ABL. Murine Abl-antisera were shown to immuno-react with a large 210-kD protein present in the Ph+ K562 cell line and not control cell lines, providing evidence that part of the ABL protein is expressed in the Ph chromosome protein product (55). This was further confirmed by preparing antibodies against GST-ABL fusion determinants and demonstrating antibody reactivity towards the 210-kD protein in the CML cell line K562 and primary CML bone marrow cell lysates (56, 57). Could the human Ph chromosome product possess elevated ABL tyrosine kinase activity similar to murine Gag-Abl required for cellular transformation? Using ABL-antisera generated from Gag-Abl studies, the Ph chromosome P210 protein product was shown to have similar tyrosine kinase activity to Gag-Abl with respect to autophosphorylation, trans-phosphorylation, and kinase inhibition by antisera against the tyrosine kinase region of ABL (55, 58, 59) (Figure 1). Furthermore, similar to Gag-Abl, a mutation in the Ph chromosome protein that specifically inactivated the tyrosine kinase activity of ABL inhibited cellular transformation, demonstrating that, like Gag-Abl, the Ph chromosome protein product requires ABL tyrosine kinase activity for its oncogenic function (60).

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY

P1: IKH

255

Breakpoints along the Ph chromosome were found to occur within the breakpoint cluster region (BCR) gene, suggesting that BCR might also be affected by the translocation that creates the Ph chromosome (61). BCR was shown to be part of the unique Ph chromosome mRNA product not present in control cells (62). cDNA cloning of this Ph chromosome mRNA product demonstrated sequence homology to the BCR gene at the 50 end of the mRNA (62–65). Later, the unique Ph chromosome protein product was shown to cross-react with BCR-specific antiserum, demonstrating that this protein expressed at least part of BCR (56). Reactivity with both BCR and ABL probes to the unique mRNA product of the Ph chromosome demonstrated that this Ph chromosome mRNA was composed of BCR-ABL (56, 64, 66) (Figure 1). Later, through sequence analysis, the BCRABL transcript was shown to contain the first 13 to 14 BCR exons and exons 1a or a2 through a11 of ABL (reviewed in 67) (Figures 1, 2B). This generated a large mRNA product that after splicing, encoded an 8.5-kB BCR-ABL chimeric transcript (reviewed in 67). BCR-ABL was shown to be the 210-kD chimeric protein product of the Ph chromosome in K562 cells by coimmunoprecipitation of the protein with an ABL antiserum and detection of a 210-kD product with either ABL- or BCR-specific antibodies (56) (Figures 1, 2B).

Additional Forms of BCR-ABL are Preferentially Associated with Specific Human Leukemias A new association of the Ph chromosome with B-ALL was subsequently discovered. Advances in chromosome mapping and molecular biology enabled the specific B-ALL Ph chromosome gene product with its chromosome breakpoints and mRNA sequence to be analyzed and compared with that of CML samples. Initial Ph+ B-ALL studies showed that a smaller 7.0 kb mRNA as opposed to a CML Ph chromosome 8.5 kb mRNA product was formed (68, 69). Furthermore, the BCR-ABL protein product in B-ALL samples was 185/190-kD (henceforth referred to as p185BCR-ABL) as opposed to the 210-kD BCR-ABL protein product (henceforth referred to as p210BCR-ABL) in CML samples (68–71). The differences in the Ph chromosome gene product in B-ALL versus CML were due to differences within the BCR breakpoints. The translocation giving rise to the Ph chromosome in B-ALL was localized within the minor breakpoint cluster region (m-bcr) in BCR (71–75), whereas in CML, the translocation site was within the major breakpoint cluster region (M-bcr) in BCR (61, 62, 76) (Figure 2B). These studies showing a high but absolute correlation between the p210BCR-ABL form and CML, and p185BCR-ABL with B-ALL suggested that specific forms of BCR-ABL may play a role in the etiology of each leukemia. P210BCR-ABL is associated with over 90% of CML cases, approximately 33% of adult B-ALL cases with the Ph chromosome (reviewed in 77) and, on occasion, acute myeloid leukemias (AML) (reviewed in 78) and myelomas (79). P185BCR-ABL is associated with the remaining 66% of adult B-ALL cases, with the Ph chromosome not associated with p210BCR-ABL (67, 78), 3% of atypical CML cases with

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

256

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

monocytosis (80), and in some AML cases (81) and lymphomas (82). In addition, a third form of BCR-ABL, p230BCR-ABL was identified as being mainly associated with neutropenic CML (83) as well as some cases of CML (84) (Figure 2B). These findings demonstrate a preferential association between different BCR-ABL forms and distinct leukemias. This correlation is not absolute as some cases of Ph+ BALL are associated with p210BCR-ABL expression. Furthermore, although the Ph chromosome is detected in HSCs in CML patients, in B-ALL patients, it is not clear whether the Ph chromosome is present in HSCs, or whether it is restricted to the lymphoid lineage (85, 86). These findings suggest that differences in target cells and BCR-ABL isoforms can contribute to the type of leukemia formed.

BCR-ABL Transcripts are Detected in Some Healthy Individuals Using highly sensitive two-step RT-PCR, BCR-ABL mRNAs were detected in leukocytes of healthy individuals (87, 88). Depending on the specific sets of primers used to amplify the BCR-ABL transcript, between ∼30% to ∼70% of healthy individuals had detectable levels of BCR-ABL expression in their white blood cells (87, 88). The amount of cells bearing the BCR-ABL transcript in these people was extremely low, with an estimated incidence of 1–10 cells per 108 white blood cells (88). Many of the BCR-ABL transcripts formed in these healthy cells encoded for nonfunctional or truncated proteins (88). However, BCR-ABL transcripts that might encode for a functional BCR-ABL protein were detected in some leukocytes of healthy individuals (88). Why the BCR-ABL positive cells that encode for a functional BCR-ABL protein do not generate leukemia in these individuals is not known. One possibility is that the origin of the BCR-ABL transcripts is a cell type more mature than HSCs/progenitors and that these cells cannot be expanded sufficiently to generate CML. Alternatively, the BCR-ABL clones may have originated from a HSC/progenitor that has subsequently been inactivated, thereby limiting the expansive capabilities of these BCR-ABL–expressing cells. Healthy individuals, unlike untreated CML patients, have high-avidity cytotoxic T lymphocytes (CTL) capable of selectively killing CML cells owing to recognition of a proteinase 3 self-peptide or a BCR-ABL junction peptide highly expressed in leukemic cells (89; reviewed in 90). Most individuals with BCR-ABL–expressing cells may possess BCR-ABL-specific high-avidity CTLs that can destroy most Phpositive cells, preventing an expansion that could result in CML. Therapy aimed at generating specific CTL killing of BCR-ABL-expressing cells has had limited success in treating CML patients, however, due to the rapid loss or complete absence of high-avidity CTLs against leukemic cells in these patients (reviewed in 90). Discovering the mechanism by which healthy individuals maintain highavidity CTLs with specificity against Ph-positive cells may lead to the development of successful immunotherapy against CML. The alternative reciprocal translocation gene product ABL-BCR has been detected in Ph chromosome-positive leukemias but is thought not to play a role in

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

P1: IKH

257

BCR-ABL–induced leukemogenesis and is not discussed further in this review (91, 92).

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The TEL (ETV6)-ABL Oncoprotein is Associated with Rare Human Leukemias Another form of chimeric ABL fusion gene product, TEL (ETV6)-ABL, was identified in rare cases of human ALL, AML, and CML (93–96). Two TEL-ABL forms have been associated with these leukemias. One form encodes the first 4 exons of TEL and the other encodes the first 5 exons of TEL, both being fused to ABL from exon 2 onwards (Figure 2C). As in BCR-ABL and GAG-ABL, TEL-ABL possesses activated ABL tyrosine kinase activity, essential for its transformation properties (94).

MECHANISM OF ABL TYROSINE KINASE ACTIVATION Crystal Structure of the N-Terminal Myristoylation, cap, SH3, SH2, and SH1 Kinase Domains of ABL Reveal a Three-Step Mechanism of ABL Tyrosine Kinase Activation Myristoylation of Gag-Abl on the glycine residue of Gag is essential for Gag-Abl association with the plasma membrane and transformation of NIH 3T3 fibroblasts (97). Although BCR-ABL is not myristoylated, attachment of a Gag myristoylation sequence to the N terminus of BCR-ABL enhances its transformation properties (98). These results suggested that myristoylation of ABL oncoproteins could augment cellular transformation by bringing mitogenic and antiapoptotic proteins such as RAS and PI3K, which associates with the plasma membrane, into close proximity for activation by ABL. Alternatively, myristoylation of ABL oncoproteins could favor the folding of the ABL tyrosine kinase domain toward an active conformation, thereby enhancing ABL tyrosine kinase activation and oncogenesis. These findings raised the question of the effect of myristoylation in ABL activation. There are two isoforms of ABL; isoform 1b is myristoylated on its second glycine residue, a site absent in all chimeric ABL oncoproteins (numbering of ABL is based on isoform 1b throughout text) (99) (Figure 3). Surprisingly, loss of myristoylation in ABL resulting from mutation of the second glycine residue to an alanine, dramatically enhanced its tyrosine kinase activity compared to wildtype ABL (100). This result is in contrast to Gag-Abl findings where loss of myristoylation was shown to suppress Gag-Abl transforming abilities (97). Kuriyan and Superti-Furga examined whether ABL myristylation contributed to the generation of a structurally inactive form of ABL and whether demyristylation led to ABL tyrosine kinase activation in the context of both human and murine ABL structures comprised of the N-terminal myristoylation, cap, SH3, SH2, and SH1 kinase domains of ABL. Collaboratively, the groups demonstrated that a

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

258

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

Figure 3 Three-step mechanism of ABL tyrosine kinase activation (taken from 104). ABL is maintained in a closed inactive conformation by the “latching” of the N-terminal myristoyl group or N-terminal cap hydrophobic residues to a deep hydrophobic pocket in the ABL tyrosine kinase domain. This “latching” of the N terminus of ABL to its kinase domain places the SH2 and SH3 domains of ABL on the backside of the large and small lobes of the kinase respectively, “clamping” the structure together. This “clamp” sequesters the activation loop of the kinase, preventing its orientation to an active conformation and tyrosine 412 phosphorylation, required for full activation of the kinase. ABL activation requires an “unlatching” of the myristoyl group or hydrophobic residues in the ABL tyrosine kinase domain. Attachment of phosphotyrosine and proline-rich proteins to the SH2 and SH3 domains, respectively, could open up the kinase and expose the activation loop outwards leading to tyrosine 412 phosphorylation and full activation of the kinase.

nonmyristolyated peptide corresponding to the N terminus of ABL was unable to bind to an inactive conformation of the ABL tyrosine kinase domain, whereas the corresponding myristolyated peptide was able to complex with the inactive conformation of ABL (100, 101). These results suggested that ABL myristolyation contributed to the maintenance of an inactive conformation and that loss of ABL myristoylation may promote the conversion of ABL from an inactive to an active state (Figures 1, 3). The N-terminal myristoyl group of ABL binds to a deep hydrophobic pocket in the kinase domain, keeping ABL in a closed inactive conformation (100, 101) (Figure 3). Although ABL isoform 1a and BCR-ABL are not myristoylated, it was suggested that the hydrophobic residues in the cap domain of ABL, present in both forms of ABL, may substitute for the myristoyl latching function (100, 101). The N-terminal myristoyl group of ABL latching to its kinase domain was shown to place the SH2 and SH3 domains of c-ABL on the backside of the large and small lobes of the tyrosine kinase domain, respectively. Clamping of the SH2 and SH3 domains within the kinase domain lobes obstructed ATP binding (100, 101) (Figure 3). Addition of increasing concentrations of tyrosine phosphorylated peptides specific to ABL SH2 domain binding led to a dose-dependent opening up of the ABL structure and activation of ABL tyrosine kinase activity (100, 101).

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY

P1: IKH

259

Mutations in the SH2 and SH3 regions that disrupted the clamping of the ABL kinase lobes also generated active forms of ABL (100, 101). An SH3 domain mutation abolished clamping to the catalytic domain and led to ABL activation (102). This effect was reversed by a reciprocal mutation in the corresponding catalytic domain residue (102). These findings support the role of the SH3 domain in stabilizing the inactive conformation of ABL. The third step leading to full ABL tyrosine kinase activation involved an outward reorientation of the activation loop away from the protein and phosphorylation on tyrosine 412 (100, 101) (Figure 3). Phosphorylation of this site enhances the in vitro kinase activity of Abl by 2- to 3.3-fold (103). Abl inefficiently autophosphorylated this site, which required expression of the Src tyrosine kinase Hck for its phosphorylation (103). This general latching, clamping, and sequestration of the activation loop is very similar to the overall mechanism of maintaining SRC in an inactive state (100, 101; reviewed in 104). The up-regulated ABL tyrosine kinase activity of chimeric ABL oncoproteins is most likely the result of disruptions in these ABL inhibitory processes. The lack of a myristoyl group in BCR-ABL and TEL-ABL, for example, may favor a more open conformation of the ABL kinase domain, leading to the heightened tyrosine kinase activity detected in these chimeric ABL oncoproteins compared with ABL.

MECHANISMS OF TYROSINE KINASE UP-ACTIVATION IN CHIMERIC ABL ONCOPROTEINS Oligomerization Contributes to ABL Tyrosine Kinase Activation and Cellular Transformation BCR-ABL, TEL-ABL, and Gag-Abl all contain oligomerization motifs and form oligomers (94, 105; reviewed in 24). This commonality suggests that oligomerization may be a general mechanism for ABL tyrosine kinase activation. ABL, on the other hand, does not contain oligomerization domains and is not predicted to utilize oligomerization to regulate its tyrosine kinase activity. Deletion of the oligomerization motifs in BCR-ABL and TEL-ABL greatly suppresses tyrosine kinase activity and cellular transformation (94, 105–107) (Table 1). Substitution of the BCR oligomerization motif in BCR-ABL with an unrelated yeast transcription factor GCN4 leucine zipper oligomerization sequence was able to maintain a high level of BCR-ABL tyrosine kinase activity and transforming ability (108). However, attachment of alternative oligomerization motifs, such as GST (109), GAG (107), and BCR oligomerization sequences 1 through 63 (105) to ABL, generated chimeric fusion proteins with modest tyrosine kinase activity and cellular transformation abilities compared to BCR-ABL. These constructed ABL oligomers possessed higher levels of tyrosine kinase activity compared with ABL and transformed certain cell lines, demonstrating that ABL

− (105, 107, 114) ↓ (120, 121) − (107, 119) + (112) ↓ (120−123) ↓ (60, 120, 121) − (60)

↓/− (105, 107, 114, 172) + (114, 121, 128) ↓/− (107) + (112, 113, 319) + (121, 122, 128) + (60, 121, 128) − (60)

Oligomerization

Y177-GRB2 site

A + B boxes

SH3

SH2

412Y P site

SH1 kinase

− (60)

− (169, 170)

ND

LL (173)

+ (319)

ND

− (114, 127, 175)

− (114)

ND

ND

+ (174)

ND

ND

+ (175)

LL (172)

− (169, 170)

ND

ND

ND

ND

LL (114)

LL (114)

T-ALL

LaTeX2e(2002/01/18)

+ = positive; − = negative; ↓ = greatly reduced.

T-ALL = acute T-cell leukemia; LL = long latency; ND = not done.

− (60)

↓ (60, 121)

+ (121)

+ (112)

− (107, 119)

+ (121)

− (107, 114)

B-ALL

AR210-IY22-09.sgm

+ (60, 121, 128)

+ (121, 123, 128, 320)

+ (113)

− (128)

+ (121, 128)

↓/− (105, 114)

CML-like

AR210-IY22-09.tex

CML = chronic myelogenous leukemia; B-ALL = acute B-lymphoblastic leukemia;

Fibroblast

Tyrosine kinase

¥

Domains

In vivo leukemogenesis

WONG

Primary bone marrow

AR

Cytokine independence

260

In vitro cellular transformation

19:26

TABLE 1 BCR-ABL domain requirements for tyrosine kinase activation, in vitro cellular transformation, and in vivo leukemogenesis

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

24 Feb 2004 P1: IKH

WITTE

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

P1: IKH

261

oligomerization can be sufficient to activate ABL tyrosine kinase to levels required for cellular transformation.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Inactivation of the SH3 Domain Up-Regulates ABL Tyrosine Kinase Activity Leading to Cellular Transformation A direct correlation between the highest levels of ABL tyrosine phosphorylation and the greatest level of cellular transformation has been established such that the level of tyrosine kinase activity and transformation in Gag-Abl > p185BCR-ABL > p210 BCR-ABL (110). A major difference in the ABL structure of Gag-Abl compared to BCR-ABL is the absence of a functional SH3 domain in Gag-Abl (reviewed in 24) (Figure 2). This suggested that the SH3 domain may be a negative regulator of ABL tyrosine kinase activity. Further evidence for a negative function of the SH3 domain was demonstrated in the crystal structure of ABL, where contacts between the SH3 domain and the tyrosine kinase domain kept ABL in an inactive conformation (100–102) (Figure 3). Deletion of the SH3 domain in ABL greatly up-regulates its tyrosine kinase activity, leading to transformation in certain cell types (99, 111). Removal of the SH3 domain in BCR-ABL also enhances BCR-ABL cellular transformation (112, 113), and deletion of the SH3 domain in a transformation defective BCR-ABL oligomerization domain-deleted clone largely restored the tyrosine kinase activity and transforming abilities of the clone (113, 114). A random mutagenesis schema identified ABL point mutants within the SH3 domain with heightened catalytic and transforming activities. Surprisingly, these mutations were not at residues predicted to obstruct clamping of the SH3 domain with the tyrosine kinase domain leading to ABL activation. These ABL mutants were defective at binding PxxP ligands (115). These results suggest that the SH3 domain could negatively regulate ABL tyrosine kinase activity when ABL is in an open conformation by interacting with proline-rich proteins that suppress ABL catalytic activity (Figure 3). The ABL interactor proteins (ABI) contain PxxP motifs that bind to the SH3 domain of c-ABL (116). Expression of ABI suppressed Gag-Abl transformation (117), and BCR-ABL expression degraded ABI proteins (118). These results suggest that ABI proteins could act as negative regulators of ABL transformation via interaction with the SH3 domain of ABL during its open conformation. The SH3 domain may serve as a dual negative regulator of kinase activity in both the active and inactive conformations of ABL.

BCR A and B Boxes are Required for BCR-ABL Activation and Cellular Transformation Crystallographic analyses of ABL activation have demonstrated that addition of phospho-peptides that bind to the SH2 domain of ABL opens up the inactive conformation of ABL, leading to its activation in a dose-dependent manner (see section on Mechanism of ABL Tyrosine Kinase Activation; 100). Deletion of BCR A and B boxes in BCR-ABL that bind to the SH2 domain of ABL inhibit

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

262

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

BCR-ABL tyrosine kinase activity and cellular transformation (107, 119). These findings suggest that BCR A and B box regions may bind to the SH2 domain of ABL, preventing the SH2 domain from clamping to the large tyrosine kinase lobe, and maintaining BCR-ABL in an open active conformation. However, a reciprocal deletion of the SH2 domain predicted to reduce the chances of ABL from adopting an inactive conformation, based on ABL crystallographic data (see section on Mechanism of ABL Tyrosine Kinase Activation; 100, 101), does not lead to BCRABL activation, but rather its suppression in the context of fibroblast transformation (120–123). This is most likely due to the role of the SH2 domain in interacting with critical transforming signaling molecules when BCR-ABL is in an active conformation to induce fibroblast transformation. These findings suggest that the SH2 domain can have a positive or negative role in BCR-ABL transformation depending on the structural conformation of the tyrosine kinase.

DOMAIN AND SIGNALING REQUIREMENTS FOR BCR-ABL–INDUCED IN VITRO CELLULAR TRANSFORMATION ABL Tyrosine Kinase Activity is Essential for Oncogenic Signaling of Multiple BCR-ABL Domains The Y177 residue in the BCR portion of BCR-ABL complexes with the GRB2 adaptor protein linking BCR-ABL to the RAS-MAPK pathway (124–126) (Table 2). GAB2 is a scaffolding adaptor that binds to GRB2 in a BCR-ABL complex (127) (Table 2). BCR-ABL expression in GAB2 knockout cells shows greatly diminished in vitro cellular transformation coincident with a decrease in RAS-ERK activation (127). These results suggest that the GRB2-GAB2 complex could be a dominant RAS signaling component. Mutation of tyrosine 177 to phenylalanine suppressed the ability of BCR-ABL to transform Rat1 fibroblast cells (120, 121) (Table 1). However, expression of the BCR-ABL Y177F mutant maintained high levels of ABL tyrosine kinase activity and was sufficient to render cytokine-dependent hematopoietic cell lines growth factor independent (121, 128) (Table 1). Mutation in the FLVRES motif of the ABL SH2 domain or deletion of the SH2 domain in BCR-ABL prevented its association with tyrosine phosphorylated proteins such as p62DOK (121, 129, 130) and suppressed BCR-ABL transformation of fibroblasts (120–123) (Table 1). However, as in the case of the BCR-ABL Y177F mutant, expression of either BCR-ABL SH2 mutant led to high levels of ABL tyrosine kinase activity and transformation of hematopoietic cell lines (121, 123, 128) (Table 1). Without tyrosine kinase activity, the GRB2 binding site was unable to complex with its signaling molecule or regulate signaling cascades leading to heightened BCR-ABL activation (131). BCR-ABL regions such as the GRB2 binding site

VEGF (344) BACHS (348) C/EBPα (350) C/EBPε (350)

JNK (140, 151) p38MAPK (159) STAT5 (174, 179) NFκB (317, 318) c-MYC (335) BCL-XL (159, 352) BCL-2 (354, 355) BAD (147, 357) TRAIL (359) AATYK (359)

ERK1/2 (140, 174) c-MYC (335) RIN1 (340) RAC (343)

G2A (334)

GM-CSF (169, 170) G-CSF (324)

IL-3Rβc (330)

PTPIB (341)

SHP-1 (336)

SHIP2 (325)

SHIP1 (325, 326)

Phosphatases

c-CBL (337, 338)

PAXILLIN (331, 332)

CRKL (327, 328)

Cytoskeletal

DNA-PK (274) DNAPOLβ (267) RAD5I (345)

BRCA1 (273)

XPB (271)

DNA repair

SHC (121, 125) p62DOK (129, 130) c-CBL (337, 338)

GAB2 (127)

GRB4 (333)

GRB2 (124, 125)

Adaptors

AR210-IY22-09.sgm

CK2α (351) PKCι (353) JAK2 (356) VAV (358) p53 (156, 360) MDM2 (360) BAGE

LYN (323, 339) HCK (339, 342) FPS/FES (346, 347) GCKR (349)

ABI-2 (118)

ABI-1 (118)

Others

AR210-IY22-09.tex

bold italic text = primary cells; plain text = cell lines.

AKT (141, 174)

IL-3 (320, 324)

p27KIP (322, 329)

PI3K (146, 323)

RAS (140, 141)

Hematopoietic factors

CYCLIN D2 (321, 322)

Apoptosis

Mitogens

AR

Cell cycle

19:26

TABLE 2 BCR-ABL regulates a diverse range of signaling molecules and pathways

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

24 Feb 2004 LaTeX2e(2002/01/18) P1: IKH

THE BCR-ABL STORY

263

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

264

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

and SH2 domain need to work in concert with up-regulated ABL tyrosine kinase activity to heighten the cellular transforming ability of BCR-ABL. The functional dependence of these domains on ABL tyrosine kinase activity reinforces the critical role of activated tyrosine kinase in BCR-ABL–induced cellular transformation. BCR-ABL augments cellular adhesion to fibronection in a tyrosine kinase– independent manner (132). The cytoskeletal interacting F-actin and proline-rich CRKL binding domains of BCR-ABL are also capable of functioning in BCRABL tyrosine kinase inactive clones (133, 134). These results suggest that F-actin and proline-rich regions may mediate BCR-ABL tyrosine kinase–independent functions leading to enhanced cellular adherence. Deletion of these cytoskeletal interacting regions only marginally reduces BCR-ABL–induced foci formation and is dispensable for cytokine-independent growth of hematopoietic cell lines (123, 134, 135). These findings suggest that BCR-ABL cytoskeletal tyrosine kinase–independent functions are likely to be dispensable for BCR-ABL–induced oncogenesis.

BCR Serine/Threonine Kinase Activity is Dispensable for BCR-ABL Induced In Vitro Oncogenesis BCR is a serine/threonine kinase capable of transphosphorylating serine residues on histones and caesin (136). A cysteine pair of residues within amino acids 298– 333 was demonstrated to be essential for BCR serine/threonine kinase activity (136). Surprisingly, expression of a BCR-ABL mutant containing a deletion spanning this BCR region was competent in transforming cells (119). BCR-ABL inhibits the serine/threonine kinase activity of BCR via tyrosine phosphorylation of BCR residue 360 (137). Expression of a serine phosphorylated BCR peptide greatly reduced BCR-ABL tyrosine phosphorylation and interfered with the growth of BCR-ABL expressing cell lines, whereas expression of a BCR 360 tyrosine phosphorylated peptide had no such effects (138). Recently, BCR kinase activity has been shown to enhance BCR binding to the scaffolding protein AF-6, which led to a suppression of RAS-ERK activity (139). These results suggest that BCR-ABL may suppress BCR serine/threonine kinase activity via tyrosine 360 phosphorylation, thereby preventing BCR from negatively affecting BCR-ABL tyrosine kinase activity, RAS activation, and BCR-ABL–induced cellular transformation.

BCR-ABL Regulates a Diverse Range of Signaling Molecules of which RAS Activation Appears to be Critical for BCR-ABL–Induced Cellular Transformation BCR-ABL expression regulates a variety of signaling molecules ranging from mitogens, antiapoptotic proteins, hematopoietic factors, and cytoskeletal components, to negative regulators such as phosphatases (Table 2). Multiple studies of RAS activation have demonstrated that BCR-ABL expression leads to an increase in the amount of GTP loaded onto RAS (128, 140–143). Inactivation

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY

P1: IKH

265

of RAS signaling via expression of dominant negative forms of RAS or addition of farnesyl transferase inhibitors severely suppresses BCR-ABL–induced cell growth and focus formation (144, 145). These results indicate that activation of the RAS pathway is critical and potentially essential to BCR-ABL–induced oncogenesis. However, it is unclear whether the majority of signaling molecules regulated by BCR-ABL expression contribute to its transforming properties. The antiapoptotic molecule PI3K is activated upon BCR-ABL induction (141, 146) and suppression of the PI3K pathway by expression of dominant negative AKT or addition of the PI3K inhibitor LY 294002 suppressed BCR-ABL–induced cytokine-independent growth (147) and colony formation (141). However, in other reports, activation of the PI3K downstream substrate AKT was shown to be minimal (148), with expression of dominant negative AKT or addition of LY 294002 being unable to reverse BCR-ABL–induced cellular transformation (148, 149). The mitogen MAPK ERK1/2 is activated or phosphorylated at its activation site upon BCR-ABL expression in the cell lines 32D and BaF3 (140, 143). Addition of the ERK1/2 inhibitor PD 98059 suppressed BCR-ABL–induced cell expansion in the Ph-positive K562 cell line (150). However, BCR-ABL expression did not activate ERK1/2 in the 293T human kidney epithelial cell or cytokine-dependent DAGM cell lines (151) and addition of the ERK1/2 inhibitor PD 98059 did not suppress cell growth of Ph-positive cell lines EM2, BV173, and ALL-1 (152). These sets of apparently opposing data on the requirement of ERK1/2 activation in BCR-ABL cellular transformation utilized different cellular systems. BCR-ABL expression may regulate different sets of signaling molecules required to induce cellular transformation, depending on its cellular context.

IN VITRO SYSTEMS TO STUDY BCR-ABL EFFECTS IN MULTIPOTENT PROGENITOR CELLS Multipotent hematopoietic progenitors are a cell population receptive to BCRABL expression and generate an abnormal expansion of mature myeloid cells, a clinical feature of chronic phase CML (153, 154). Alternative in vitro cellular systems have been developed to study BCR-ABL in the cellular context of hematopoietic progenitors. Infection of 5-fluorouracil-treated bone marrow cells depleted of cycling cells and enriched for HSCs/progenitors with a retrovirus expressing BCR-ABL was shown to expand cells predominantly of the myeloid lineage (155) (Figure 1). The pluripotent FDCP-Mix cell line has been used to demonstrate that growth factor independence of myeloid cells required long-term BCR-ABL expression (156). Chronic BCR-ABL expression in human CD34+ multipotent progenitors via retroviral transduction delayed apoptosis induced by serum and cytokine withdrawal (153). Embryonic stem (ES) cell-derived multipotent progenitors expressing BCR-ABL via retroviral transduction or tetracycline regulation expanded blast cells (157, 158).

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

266

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

We have examined a set of signaling molecules shown to be regulated by BCR-ABL expression from cell line studies in the cellular context of ES-derived hematopoietic progenitors. The majority of these signaling molecules were not altered by acute BCR-ABL expression in this cellular context (159). The proapoptotic molecule p38MAPK was suppressed and the antiapoptotic molecule BCL-XL was up-regulated upon BCR-ABL expression coincident with the antiapoptotic effect of BCR-ABL in these progenitors (159) (Table 2). A gene microarray study has identified numerous signaling molecules regulated by BCR-ABL expression in Ph-positive multipotent progenitors (160) (Table 2). The contribution of these signaling molecules to Ph-positive leukemias awaits examination in BCR-ABL– induced in vivo leukemia animal models.

MURINE MODELS OF ABL ONCOGENESIS Deregulated Tyrosine Kinase Activity is Necessary for ABL-Induced Leukemias To confirm predictions based on initial GAG-ABL studies that expression of ABL forms with deregulated tyrosine kinase activity is sufficient to generate cancer in vivo, animal models expressing different ABL forms needed to be developed. An early transgenic study driving expression of a created trimeric bcr-gag-abl construct off an immunoglobulin heavy-chain enhancer (Eµ) or a myeloproliferative sarcoma virus (MPSV) promoter generated lymphomas in mice (161). Transgenic mice expressing p185BCR-ABL from a metallothionein-1 promoter (MT) developed acute leukemias (162) (Figure 1). Retroviral transduction of p210BCR-ABL expression into bone marrow cells transplanted into mice demonstrated that p210BCR-ABL expression induced multiple hematopoietic neoplasms including CML, ALL, erythroid leukemia, mast cell leukemia, T lymphoma, and macrophage tumors (reviewed in 163) (Figure 1). Different cell culture infection conditions and mouse strains affected the type of neoplasm induced by BCR-ABL expression. In fact, injection of BCR-ABL virus directly into the mouse thymus exclusively generated T lymphomas (164). These results validated earlier predictions that different forms of ABL oncoproteins with deregulated ABL tyrosine kinase activity could generate cancer in vivo and revealed that the target hematopoietic cell type expressing the ABL oncogene played a key role in determining the type of hematological disease generated. In 1990, several groups demonstrated that p210BCR-ABL expression could induce a CML-like disease in mice (165, 166) (Figure 1). The experimental strategy was to enrich for HSC/progenitor expression of p210BCR-ABL, the same target cell population from which human CML originates. Mice were treated with 5-fluorouracil (5-FU) to kill cycling cells and enrich for HSC/progenitor populations. Quiescent HSC/progenitors were induced to cycle with cytokine mixtures and these cells were subsequently infected with retroviral vectors encoding p210BCR-ABL. Lethally

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY

P1: IKH

267

irradiated mice reconstituted with this p210BCR-ABL–infected cell population generated a CML-like disease.The CML-like disease was transplantable into secondary and tertiary recipients, demonstrating that the target cell type from which the disease originates is a HSC (165, 167). The correlation between highest levels of ABL tyrosine kinase activity and greatest level of cellular transformation was extended to the rate of leukemia development with Gag-Abl > p185BCR-ABL > p210 BCR-ABL (166, 168). This infection condition coupled with improved retroviral technology is still the main CML-like mouse model used today to study BCR-ABL leukemogenesis (169, 170). Using this CML-like model, it was shown that a tyrosine kinase inactive p210BCR-ABL mutant could not generate a CML-like disease (169, 170). This contrasts to the situation in Drosophila where expression of kinase-active or kinase-defective ABL forms were shown to be effective at rescuing ABL-defective animals to wild-type, fertile flies (171).

Multiple BCR-ABL Domains Contribute to the Type and Latency of Leukemia Deletion of the oligomerization domain previously shown to severely suppress BCR-ABL tyrosine kinase activity and in vitro cellular transformation (see section on Domain and Signaling Requirements for BCR-ABL-Induced in Vitro Cellular Transformation) inhibited BCR-ABL from generating a CML-like disease in vivo (114) (Table 1). However, this BCR-ABL mutant was able to generate B cell leukemia/lymphoma (172) and T-ALL (114) with a significantly longer latency period than wild-type BCR-ABL (Table 1). Restoration of the tyrosine kinase activity in this BCR-ABL mutant by deleting the SH3 domain rescued the ability of the mutant to generate an aggressive CML-like disease (114). This direct correlation of enhanced ABL tyrosine kinase activation with leukemia severity again displays the dominant role of ABL tyrosine kinase activity in BCR-ABL oncogenesis. Inactivation of the SH2 domain previously shown to maintain high levels of ABL tyrosine kinase activity (see section on Domain and Signaling Requirements for BCR-ABL-Induced in Vitro Cellular Transformation) delayed the development of a BCR-ABL–induced CML-like disease compared with wild-type BCR-ABL (173, 174) (Table 1). However, the mutation did not affect the ability of BCR-ABL to induce B-ALL (174) (Table 1). These results demonstrate that a functional SH2 domain contributes to, but is not required for BCR-ABL to generate a CML-like disease and is dispensable for BCR-ABL–induced B-ALL. Expression of the Y177F BCR-ABL mutant shown to retain elevated levels of tyrosine kinase activity (see section on Domain and Signaling Requirements for BCR-ABL-Induced in Vitro Cellular Transformation) was unable to generate a lethal CML-like disease and instead developed a long-latency fatal lymphoid leukemia in mice (114, 127, 175) (Table 1). The Y177 site in BCR-ABL is complexed with GRB2 and GAB2, which can lead to RAS activation (see section

24 Feb 2004

19:26

268

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

on Domain and Signaling Requirements for BCR-ABL-Induced in Vitro Cellular Transformation; 127). RAS activation and expression of the GAB2 scaffolding adaptor is crucial for BCR-ABL–induced in vitro cellular transformation (see section on Domain and Signaling Requirements for BCR-ABL-Induced in Vitro Cellular Transformation). These findings suggest that BCR-ABL activation of RAS via its Y177 site interaction with GRB2 and GAB2 may be critical in the generation of CML (see 176 for a detailed review on domain requirements for BCR-ABL–induced leukemias).

The Significance of the RAS Pathway in BCR-ABL Cellular Transformation has been Confirmed in Murine Leukemogenesis Models In vivo administration of the RAS inhibitor SCH66336 suppressed p190BCR-ABL– induced B-ALL in 80% of mice for over 200 days, whereas 100% of control animals succumbed to leukemia from 18–103 days (177). GAB2-deficient BCRABL expressing progenitors with severely reduced RAS-MAPK activation were unable to generate a BCR-ABL–induced CML-like disease unlike control cells (B. Neel & R. Van Etten, personnal communication). These results suggest that GAB2 may be a critical signaling molecule connecting BCR-ABL to RAS activation and leukemogenesis. IL-3 up-regulation (170, 178), p62DOCK phosphorylation (129, 130), and STAT5 activation (179, 180) have all been demonstrated following BCR-ABL induction in numerous cell types (Table 2). However, knockouts of these signaling molecules did not prevent BCR-ABL–induced leukemias, and only in the case of p62 DOCK was there a minor delay in leukemogenesis (181–183). These findings suggest that it may be difficult to identify individual signaling molecules critical for BCR-ABL–induced leukemia.

BCR-ABL Expression is Required for the Development and Maintenance of a BCR-ABL–Induced Leukemia A transgenic BCR-ABL tetracycline-regulatable B-ALL mouse model demonstrated that continual BCR-ABL expression was required for the establishment and maintenance of a BCR-ABL–induced leukemia (184). This is in agreement with a BCR-ABL tetracycline-regulatable ES in vitro differentiation system where continual BCR-ABL expression was shown to be required for the expansion of progenitor and myeloid cells (148). Temperature-sensitive Gag-Abl and BCRABL mutants at nonpermissive temperatures were unable to transform NIH 3T3 and lymphoid cells owing to inhibited tyrosine kinase activity (48, 185) (see section on Discovery of Novel Tryosine Kinase Activity Required for Abelson Murine Leukemia Virus (A-MuLV)–Induced Cellular Transformation and Leukemogenesis).

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

P1: IKH

269

KNOCKOUT ANIMAL MODELS WITH CML PHENOTYPES

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Interferon Consensus Sequence Binding Protein (ICSBP)-Deficient Mice Develop a CML-Like Disease ICSBP is a transcription factor of the interferon regulatory factor (IRF) family, members of which regulate interferon (IFN)–stimulated gene expression (186). ICSBP knockout mice are 100% penetrant for a chronic phase CML-like disease by 10 to 16 weeks of age, of which 33% evolve into blast crisis by 50 weeks of age (187). The chronic stage disease in these mice is characterized by an elevation of neutrophils in hematopoietic tissues. Development of blast crisis is associated with a huge expansion of myeloblasts in the bone marrow and spleen, with B lymphoblasts being dominant in the blood and lymph nodes. ICSBP+/− mice have a lower penetrance of CML-like disease compared with ICSBP−/− mice, suggesting that development of CML by loss of ICSBP expression is dose dependent. The CML-like phenotype of ICSBP knockout mice prompted investigators to examine whether ICSBP expression was down-regulated in a p210BCR-ABL– induced leukemia. ICSBP down-regulation indeed correlated with the development of a p210BCR-ABL–induced CML-like disease in vivo (188). Moreover, enforced expression of ICSBP delayed the onset of a p210BCR-ABL–induced lethal CMLlike syndrome from 3 weeks to 5 weeks (188). However, ICSBP expression was unable to inhibit p210BCR-ABL–induced leukemogenesis. ICSBP is poorly active at regulating gene expression by itself (189) but is highly active when it is part of a multiprotein complex being bound to IRF-1 or IRF-2 and PU.1 (189). Overexpression of ISCBP as well as its partner(s) may be necessary to fully inhibit p210BCR-ABL–induced leukemogenesis. This has been shown in the myelomonocytic cell line U937 in which transfection of all three components (ICSBP, IRF-1, and PU.1) was required to produce a highly active transcriptional complex (189).

Lack of JunB Expression in the Myeloid Lineage Generates a CML-Like Disease in Mice JunB expression increases during myeloid differentiation (190, 191), suggesting it may play an important role in myelopoiesis. JunB knockout mice rescued from embryonic lethality via expression of JunB driven from a human ubiquitin promoter transgene (JunB−/− Ubi-junB) were shown to gradually develop elevated levels of mature myeloid cells in the blood, spleen, bone marrow, and lymph nodes beginning at 4 months of age concomitant with a reduction in JunB expression (192). These results suggested that down-regulation or loss of JunB expression contributed to the CML-like disease, which progressed to a syndrome resembling blast crisis in 16% of the mice. JunB expression inhibits cyclinD1 function by activating P16INK4a (193, 194). CyclinD1 overexpression complements BCR-ABL–induced transformation (195). Therefore, cyclinD1 up-regulation may be a common signaling event in myeloid

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

270

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

leukemogenesis induced by both BCR-ABL expression and loss of JunB activity. Although JunB−/− Ubi-junB transgenic mice lose JunB expression in the myeloid compartment during disease progression, JunB expression is maintained in T cells, B cells, and erythrocytes. This suggests that loss of JunB expression in HSCs is not required for the induction of the myeloproliferative disorder. Transgenic mice engineered with restricted expression of BCR-ABL to common myeloid progenitors and their progeny via the MRP8 promoter were shown to develop a CML-like disease (196). These findings demonstrate that a genetic lesion restricted to myeloid progenitors and their progeny can induce a CML-like disease and raise the question as to the involvement of the HSC in CML beyond that of the cellular origin/target.

Loss of SH2-Containing Inositol-5-Phosphatase (SHIP) Expression Leads to a CML-Like Disease in Mice Targeted disruption of SHIP was shown to generate a massive expansion of myeloid elements in the spleen, bone marrow, and lungs that proved fatal in 50% of mice by 10 weeks of age (197). SHIP mRNA and protein levels are down-regulated upon induction of ABL oncoproteins p185BCR-ABL, p210BCR-ABL, or TEL-ABL and exogenous expression of SHIP was shown to suppress BCR-ABL–induced spontaneous transwell migration (198). Both SHIP−/− and BCR-ABL expressing hematopoietic progenitors are hyperresponsive to cytokines such as IL-3, leading to an increase in the number of CFU colonies generated in methylcellulose compared with control cells (197, 199). SHIP down-regulation may contribute to BCRABL–induced leukemogenesis by enhancing the responsiveness of hematopoietic progenitors toward cellular expansion under growth factor limiting conditions.

CML TREATMENTS PRIOR TO BCR-ABL MOLECULAR TARGETED THERAPY Most CML treatments prior to BCR-ABL molecular targeted therapy were aimed at destroying actively proliferating cell populations. Busulfan and hydroxyurea treatments globally inhibited the growth of normal and leukemic cells and generated hematological responses in chronic phase CML patients. However, the treatments were unable to selectively destroy Ph+ cells or inhibit the oncogenic activity of BCR-ABL. The treatments did not lead to cytogenetic remission in most patients but could delay the onset of blast crisis CML (reviewed in 200). Ablative chemotherapy with or without irradiation therapy to destroy leukemic and normal proliferating cells followed by allogenic transplantation of normal bone marrow cells was and currently remains the only curative treatment for CML (reviewed in 21). However, only ∼20% of CML patients are candidates for bone marrow transplantation, and 5-year disease-free survival in transplanted patients can vary from 30% to 80% (reviewed in 201, 202).

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

P1: IKH

271

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Potential Signaling Pathways Regulated by IFN-α Treatment May Contribute to its Selectivity and Effectiveness as a CML Suppressor Interferon-α (IFN-α) treatment leads to both a hematological and cytogenetic response in many chronic phase CML patients (203). In a large study of 512 early chronic phase CML patients treated with IFN-α, 27% (140 out of 512) of patients achieved complete cytogenetic responses and 78% of those patients survived for greater than ten years with this treatment (204). The effectiveness of IFN-α at treating a significant portion of CML patients suggests that IFN-α therapy may have some selectively toward destroying Ph+ leukemic cells or inhibiting the oncogenic activity of BCR-ABL. However, up to 20% of CML patients do not tolerate IFN-α therapy because of toxicity problems (reviewed in 202). IFN-α activation leads to JAK1 phosphorylation followed by STAT1/2 translocation to the nucleus and activation of IFN-α responsive genes (reviewed in 205). This signal transduction cascade is perturbed in CML patient cells resistant to IFNα therapy with a lack of STAT1 expression (206). CML cell lines resistant to IFN-α treatment also suppress the JAK1-STAT1 activation pathway by expressing high levels of suppressor of cytokine signaling 3 (SOC3) (207), which in turn blocks JAK1-STAT1 (208, 209) and down-regulates IFN-α signaling (210). Activation of the JAK1-STAT1 pathway could be a plausible mechanism by which IFN-α treatment suppresses CML. IFN-α induced CML remission is correlated with the reappearance of ICSBP expression (211). High-avidity cytotoxic T lymphocytes (CTLs) against Ph+ cells in IFN-α−treated CML patients experiencing cytogenetic remission have been demonstrated (212). ICSBP expression in a BaF3 cell line expressing BCR-ABL suppressed leukemia development in mice due to the generation of a CD8 cytotoxic T cell response against the BCR-ABL–expressing cell line (213). Coincidentally, ICSBP knockout mice not only develop a CML-like disease, but they also have impaired T cell Th1 response with defective IFN-γ and IL-12 production and enhanced susceptibility to pathogens (187). These results suggest that ICSBP expression may be required to generate high-avidity CTLs and IFN-α treatment may suppress CML by generating high-avidity CTLs against Ph+ leukemic cells via ICSBP induction of IFN-γ and IL-12 production.

DEVELOPMENT OF MOLECULAR TARGETED THERAPY AIMED AT INHIBITING THE TYROSINE KINASE ACTIVITY OF ONCOPROTEINS Over 100 gain-of-function oncogenes have been defined that can contribute to carcinogenesis (reviewed in 5). Tyrosine kinases represent a large fraction of known dominant oncogenic proteins and were a prime target for the development of specific cancer inhibitors (reviewed in 5, 214). The human genome encodes 518

24 Feb 2004

19:26

272

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

serine/threonine and tyrosine kinases (215), all of which bind ATP in a highly conserved manner within their catalytic domains (216). This high degree of commonality among kinase ATP-binding regions suggested that it would be difficult to develop compounds that specifically inhibited a single or limited set of tyrosine kinases without having cross-reactivity toward others. The abundance of ATP in a cell raised another concern over the difficulty in developing inhibitors to be administered at concentrations that would effectively suppress oncogenic tyrosine kinases without cellular toxicity.

Identification of Naturally Occurring Compounds with Specific Inhibitory Activity Towards Tyrosine Kinases The first group of compounds tested for inhibitory activity against oncoproteins came from naturally occurring sources. Herbimycin A is a natural antibiotic originally isolated as an agent that reversed the transformation phenotype of RSVinfected rat kidney cells (217). Herbimycin A inhibits a wide range of tyrosine kinases including YES, FPS, ABL, ERBB, ROS (218), and RET (219) without affecting serine/threonine kinases such as protein kinase C (PKC) (220). Herbimycin A targeted protein tyrosine kinases for degradation via a 20S proteosome and ubiquitin-dependent mechanism (221). Genistein is an isoflavone compound isolated from soybeans that acts as an ATP competitive inhibitor toward the catalytic domain of different tyrosine kinases but not serine/threonine kinases (222). Erbstatin was isolated from Streptomyces (223) and shown to be a low-molecularweight substance that resembled tyrosine and blocked the kinase activity of different tyrosine kinases (reviewed in 224). These findings demonstrated that compounds with selectivity towards tyrosine kinases and not serine/threonine kinases could be generated. These naturally occurring compounds were also shown to inhibit Gag-Abl/BCR-ABL cellular transformation (218, 225, 226), suggesting that inhibition of deregulated ABL tyrosine kinase activity may be a viable therapeutic approach toward ABL-induced leukemias.

Development of Tyrphostins: Tyrosine Phosphorylation Inhibitors with Selectivity Toward Specific Tyrosine Kinases In 1988, Levitzki’s laboratory described the synthesis of a series of low-molecularweight protein tyrosine kinase inhibitors with progressive affinity for the epidermal growth factor receptor (EGFR) (227). The most potent of these tyrphostins inhibited the kinase activity of EGFR 1000-fold over that of insulin receptor kinase and suppressed EGF-dependent proliferation without affecting EGF-independent proliferation (227). This demonstrated that synthetic inhibitors could be engineered with selective activity towards specific tyrosine kinases (reviewed in 228). The first generation of tyrphostins was designed to resemble tyrosine and act as substrate inhibitors based on findings that the naturally occurring tyrosine kinase inhibitor erbstatin resembled tyrosine. The initial group of tyrphostins, however, covered a wide range of kinetic behavior: Some were substrate

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY

273

competitive inhibitors, some ATP competitive inhibitors, and some both substrate and ATP competitive inhibitors (reviewed in 228). Levitzki’s group further improved the selectivity and potency of tyrphostins against specific tyrosine kinases by converting one aromatic ring, largely substrate mimic inhibitors, into two-ring largely ATP mimic inhibitors (for details on the chemistry of tyrphostin development, see 228). Three tyrphostins were generated and shown to inhibit BCRABL tyrosine kinase activity, cellular transformation, and loss of stromal cell adhesion (229–231). These tyrphostins had a short serum half-life, and further modifications were required to generate an inhibitor with a longer serum halflife. Adaphostin was developed and proved to have a longer serum half-life, although selectivity toward BCR-ABL tyrosine kinase inhibition was greatly reduced (232). The tyrphostin inhibitor AG 490 was one of the first synthetic chemical compounds shown to inhibit the growth of human leukemic cells both in vitro and in vivo (233). The inhibitor blocked the growth of human B-ALL cells transplanted into mice by selectively suppressing JAK-2 activity and did not have deleterious effects on normal hematopoiesis in vivo (233). This finding demonstrated that it was possible to administer a tyrphostin at a dose sufficient to inhibit oncogenesis without lethal toxicities in an in vivo setting (for details of tyrphostins and their potential in clinical use, see 234, 235).

Development of the Small Molecule Inhibitor Imatinib Mesylate Investigators at Ciba-Geigy, now Novartis, originally initiated a random screening of archived inhibitors against protein kinase C-α (PKC-α). They identified the class of phenylamino-pyrimidines as being relatively selective ATP competitive inhibitors against PKC-α (reviewed in 234). Zimmermann, Buchdunger, Trexler, Lydon, and colleagues modified this class of compounds, which underwent highthroughput screenings to identify inhibitors against the platelet-derived growth factor receptor (PDGF-R) tyrosine kinase (234, 235) for the chemistry behind the development of phenylamino-pyrimidine inhibitors). CGP 53,716 was identified as a potent tyrosine kinase inhibitor against PDGF-R and suppressed tumor growth of PDGF-R-activated cell lines in vivo (236). Surprisingly, the CGP 53,716 compound not only suppressed PDGF-R tyrosine kinase activity but quantitatively inhibited Gag-Abl tyrosine kinase activity to a similar degree (236). PDGF-R belongs to the split tyrosine kinase domain type III receptor family, which includes members CSF-1R, c-KIT, and FLK2 (FLT3), whereby ∼100 residues separate two parts of the tyrosine kinase domain (reviewed in 237). ABL tyrosine kinase domain is contiguous and not separated by any intervening residues and is most similar to ARG at the primary amino acid analysis level (reviewed in 237). Primary sequence alignment of the tyrosine kinase domains of PDGF-R and ABL would not have predicted that the structure of these two tyrosine kinase domains be comparable enough to be selectively inhibited by CGP 53,716 without cross-inhibition to a broader range of tyrosine kinases (Figure 4). This result predicted the need to analyze three-dimensional structures of

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

274

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

Figure 4 Tyrosine kinase phylogenetic tree containing Imatinib mesylate-sensitive members. Phylogenetic tree includes SRC family members YES, SRC, FYN, FGR, BLK, LCK, LYN, and HCK; ABL family members ABL (ABL1) and ARG (ABL2); PDGFR family members PDGFRα, PDGFRβ, CSF1R, KIT, and FLK2 (FLT3); VEGFR family members VEGFR1 (FLT1), 2 (FLT4), and 3 (KDR/FLK1). Imatinib mesylate exhibits high selectivity toward inhibiting the tyrosine kinase activity of ABL, ARG, KIT, and PDGFR (r). Primary sequence alignment by ClusterW multiple sequence alignment (http://clustalw.genome.ad.jp/) demonstrates that CSF1R and VEGFR family tyrosine kinases are more closely related to KIT and PDGFR than ABL and ARG. However, CSF1R and VEGFR family members are not inhibited by Imatinib mesylate at doses that effectively block ABL, ARG, KIT, and PDGFR kinase activity.

kinases to obtain more reliable predictions towards the specificity of engineered inhibitors. CGP 53,716 was much more potent at inhibiting v-ABL tyrosine kinase activity than a previously engineered class of benzopyranone derivatives from Geissler and colleagues in the same company (238). CGP 53,716 was further optimized for vABL tyrosine kinase inhibition and Imatinib mesylate (Gleevec, STI571, CGP 57148B) was generated (reviewed in 234, 235). Imatinib mesylate demonstrated high selectivity at inhibiting a small group of tyrosine kinases including PDGF-R (239), KIT (240, 241), ABL (242), and ARG (243). Druker and colleagues demonstrated that Imatinib mesylate could selectively inhibit BCR-ABL–induced in vitro cell growth and in vivo tumor formation in mice from cell lines expressing BCR-ABL but not from control cell lines (244) (Figure 1). Imatinib mesylate inhibited Ph+ cell growth but not normal cell growth in a dose-dependent manner (245, 246). This high selectively of Imatinib mesylate at suppressing BCR-ABL–induced cell expansion in vitro and in vivo along with Imatinib mesylate’s pharmacological properties (reviewed in 247) led to the testing of Imatinib mesylate in CML patients refractory to IFN-α therapy.

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

P1: IKH

275

IMATINIB MESYLATE TREATMENT IN CML PATIENTS

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Imatinib Mesylate Treatment Leads to Remission in Most Chronic Phase CML Patients Resistant to IFN-α Therapy In 1998, a phase 1 study of Imatinib mesylate treatment on chronic phase CML patients refractory or resistant to interferon-based therapy began. When these patients became nonresponsive to IFNα therapy, no treatment options proven to be effective against CML were available. Druker, Talpaz, Sawyers, and colleagues reported that 53 of 54 of patients had complete hematologic responses by 4 weeks of Imatinib mesylate treatment (248). Of the Imatinib mesylate-treated patients, 96% (51 of 54) maintained normal blood counts and complete hematological responses for over a year (248) (Figure 1). The success of Imatinib mesylate treatment on chronic phase CML patients in phase I studies led to large-scale phase II and phase III studies. Results of Imatinib mesylate-treated chronic phase patients resistant to IFNα therapy were equally remarkable in phase II studies. Kantarjian and colleagues reported that of 532 patients, 95% of those treated with 400 mg of Imatinib mesylate daily, achieved complete hematologic responses, and 89% did not show disease progression (249) (Figure 1). Even more impressive, in a randomized phase III study of 553 patients not previously treated with IFNα, only 1.5% of Imatinib mesylate–treated patients had disease progression (201) (Figure 1).

A High Percentage of Accelerated and Blast Phase CML Patients are Resistant to or Relapse from Imatinib Mesylate Treatment The high rate of remission in Imatinib mesylate-treated chronic phase CML patients treated led the phase I research group to expand their study and determine the effect of Imatinib mesylate therapy on blast crisis CML patients and Ph+ BALL patients. In myeloid blast crisis patients, 55% (21 out of 38 patients) were responsive to Imatinib mesylate treatment, with 11% achieving complete hematological response. However, only 18% of patients had responses lasting over a year. The remaining 20 Imatinib mesylate–treated patients with lymphoid blast crisis CML or Ph+ B-ALL had a similar lower level of disease suppression compared with Imatinib mesylate–treated chronic phase CML patients, with only 20% achieving complete hematological response (250). During the aggressive phase of CML blast crisis, numerous secondary genetic aberrations including duplication of the Ph chromosome are common (11). These findings suggest that failure of Imatinib mesylate treatment in blast phase CML patients could be due to either the presence of added oncogenic events beyond the Ph chromosome or a heightened level of BCR-ABL tyrosine kinase activity ineffectively suppressed by the dose of Imatinib mesylate administered. The high relapse rate in Imatinib mesylate–treated blast crisis CML patients (250) and the frequent occurrence of an additional Ph chromosome in advanced stages of CML disease (11) prompted subsequent phase II accelerated and blast

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

276

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

crisis CML studies to examine if a higher dose of Imatinib mesylate would be more effective at treating CML patients at the acute phases of the disease. In both accelerated and blast crisis studies, 600 mg of daily Imatinib mesylate led to a higher percentage of patients with complete hematological response compared with 400 mg of Imatinib mesylate daily treatment (251, 252). In the accelerated phase study, Talpaz and colleagues reported that patients treated with 600 mg Imatinib mesylate daily had a statistically significant improvement in 12-month survival compared with 400 mg daily Imatinib mesylate–treated patients (78% versus 65%, respectively, and P = 0.014) (251). These results suggest that duplication of the Ph chromosome may be a mechanism by which accelerated and blast crisis leukemic cells express higher levels of BCR-ABL tyrosine kinase activity, thereby requiring higher doses of Imatinib mesylate for effective suppression of ABL tyrosine kinase. In a subsequent study, Imatinib mesylate treatment at a dose of 800 mg/daily led to severe pancytopenia in over half of CML patients who did not respond to Imatinib mesylate at 400 mg/day, and a dose reduction of Imatinib mesylate to 600 mg/day had to be implemented (253). This toxicity of Imatinib mesylate at a dose of 800 mg/day may limit the therapeutic applicability of Imatinib mesylate in treating CML patients expressing higher amounts of BCR-ABL.

MECHANISMS OF RESISTANCE TO IMATINIB MESYLATE TREATMENT BCR-ABL Tyrosine Kinase–Dependent Mechanisms of Imatinib Mesylate Resistance The majority of Imatinib mesylate–insensitive Ph+ patient cells continue to generate high levels of deregulated BCR-ABL tyrosine kinase activity even in the face of drug treatment (254–257). The presence of BCR-ABL mutants resistant to or with reduced sensitivity to Imatinib mesylate inhibition appears to be the most frequent mechanism by which CML patients relapse from Imatinib mesylate therapy (254, 255, 258) (Figures 1, 5). Some BCR-ABL–resistant mutants are highly resistant to Imatinib mesylate, whereas others have demonstrated a minor twofold or less enhancement of resistance to Imatinib mesylate (259). Genotyping BCR-ABL Imatinib mesylate–resistant clones from a patient and phenotyping the clone’s level of resistance to Imatinib mesylate may identify patients who could be effectively treated for CML with higher doses of Imatinib mesylate and define candidates who will need alternative treatments. BCR-ABL mRNA and protein levels have also been shown to be upregulated in a subpopulation of Imatinib mesylate–resistant leukemic cells, with amplification of the Ph chromosome being a likely mechanism (254, 256, 257). Expression of the multidrug resistance gene (MDR-1) encoding a P-glycoprotein (PGP) that suppressed Imatinib mesylate entry into the Ph+ K562 cell line has been demonstrated to be another mechanism by which BCR-ABL tyrosine kinase activity is maintained upon drug treatment (257, 260).

19:26 AR AR210-IY22-09.tex AR210-IY22-09.sgm

Figure 5 BCR-ABL mutations resistant to Imatinib mesylate suppression span across regions in addition to the ABL tyrosine kinase domain. Bold italic font: mutations identified in patients that relapsed from Imatinib mesylate treatment (254, 255, 258, 259). Normal font: mutations discovered in BCR-ABL expressing BaF3 cells resistant to Imatinib mesylate inhibition of cellular transformation (265). ABL numbering based on isoform 1b.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

24 Feb 2004 LaTeX2e(2002/01/18) P1: IKH

THE BCR-ABL STORY

277

24 Feb 2004

19:26

278

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

BCR-ABL Tyrosine Kinase–Independent Mechanisms of Imatinib Mesylate Resistance May Be Rare Events Over 80% of blast phase CML cases have definable additional genetic aberrations including trisomy 8, i(17q) (11, 12), loss of p53 function (13, 14, 261), MYC amplification (12), RB deletion/rearrangement (15), and p16INK4A rearrangement/deletion (16). These findings suggested that the high rate of relapse in blast crisis CML patients treated with Imatinib mesylate could be due to the acquisition of additional oncogenic events that render Ph+ leukemic cell growth independent of BCR-ABL tyrosine kinase activity. Surprisingly, the large majority of accelerated and blast crisis Ph+ leukemic cells from Imatinib mesylate–resistant CML patients maintained high levels of BCR-ABL tyrosine kinase activity, suggesting that the mechanism of Imatinib mesylate resistance in these cells depends on sustained deregulated BCR-ABL enzyme activity (254, 255, 258). These findings provide strong evidence that deregulated BCR-ABL tyrosine kinase activity is required for the maintenance of most cases of CML, irrespective of disease stage and presence of additional genetic aberrations (Figure 1). These results are in agreement with basic research findings using Gag-Abl and BCR-ABL inducible systems, which demonstrated that continual oncoprotein expression/tyrosine kinase activity was required for the establishment and maintenance of ABL-induced cellular transformation and leukemogenesis (48, 158, 184, 185). A Ph+ K562 cell line resistant to Imatinib mesylate suppression of cell growth was shown to express a form of BCR-ABL sensitive to Imatinib mesylate inhibition (262). Addition of Imatinib mesylate suppressed phosphorylation of BCR-ABL and several of its key substrates including STAT5, CRKL, and MAPK (262). The SRC family kinase LYN was highly expressed in this cell line (262). Suppression of LYN expression with antisense LYN oligonucleotides or inhibition of LYN kinase activity with the specific compound CGP 76030 abrogated cell growth of this cell line at concentrations that did not affect a control K562 cell line (262). This is a rare example whereby a Ph+ cell loses its dependence on deregulated BCR-ABL tyrosine kinase activity for oncogenic transformation.

MECHANISM OF ABL TYROSINE KINASE INHIBITION BY IMATINIB MESYLATE Imatinib Mesylate Inhibits ABL Tyrosine Kinase Activity by Displacing ATP and Trapping the Enzyme in an Inactive Conformation To determine how Imatinib mesylate achieves its high specificity at inhibiting BCR-ABL transformation, the crystal structure of the kinase domain of ABL complexed with Imatinib mesylate was resolved by Kuriyan’s and Zimmermann’s

15 Mar 2004

15:15

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY

P1: IKH

279

groups (235, 263, 264). Both groups showed that Imatinib mesylate bound to the inactive conformation of ABL in the region where the adenosine base of ATP would bind thereby obstructing ATP binding (numbering of c-ABL based on isoform 1b) (Figures 1, 6A). It was further demonstrated that Imatinib mesylate effectively inhibited the catalytic activity of ABL in its inactive conformation but not in its active conformation (263). A critical component in determining the inactive versus active conformation of tyrosine kinases is the orientation and phosphorylation status of residues in the activation loop of the kinase. During the transition from an inactive to an active conformation of the ABL tyrosine kinase, the activation loop is sharply reorientated from facing toward the protein to facing outward, accompanied by phosphorylation of tyrosine 412 leading to full kinase activity (see section on Mechanism of ABL Tyrosine Kinase Activation; Figure 3). A comparison of the three-dimensional structures of highly homologous tyrosine kinases (by primary sequence alignment) in their active form, revealed that ABL, Insulin receptor kinase (IRK), and hematopoietic cell specific kinase (HCK) have a similar overall structure and activation loop conformation (263, 264) (Figure 6C). However, although the overall crystal structures of the inactive kinase domains are comparable, the activation loop conformations were vastly different (263, 264) (Figure 6C). Kuriyan, Zimmermann, and colleagues propose that the high selectivity of Imatinib mesylate at inhibiting a small group of protein kinases is achieved by the specificity with which Imatinib mesylate recognizes unique structural motifs present in the inactive conformation of kinases such as the activation loop. By primary sequence analysis, KIT and PDGFR would be predicted to have a more divergent catalytic domain structure than SRC and IRK in relation to the ABL tyrosine kinase domain (Figure 4). Kuriyan and coworkers state that all the residues that contact Imatinib mesylate in the ABL kinase are identical or substituted in a conservative manner in the SRC-family tyrosine kinases (263, 264). However, Imatinib mesylate inhibits KIT, PDGFR, and ABL tyrosine kinase activity with comparable efficiencies and is largely ineffective in suppressing the tyrosine kinase activity of SRC and IRK (reviewed in 247). Zimmermann and colleagues are extrapolating a 3-D structure of KIT complexed with Imatinib mesylate based on 3-D interactions of ABL and Imatinib mesylate (235). They demonstrate that predicted key interactions between KIT and Imatinib mesylate directly parallel those of ABL and the drug (235).

Imatinib Mesylate and ABL Kinase Domain Interactions Can Only Account for a Portion of BCR-ABL Mutants Resistant to Imatinib Mesylate Treatment The initial search for BCR-ABL mutants resistant to Imatinib mesylate focused on the identification of mutants in the ABL kinase domain predicted to obstruct direct Imatinib mesylate binding. The Sawyers’ laboratory initially discovered the ABL

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

280

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

kinase T334I mutant as the most frequent BCR-ABL mutant in CML blast crisis patients resistant to Imatinib mesylate therapy (254) (Figures 1, 5). The crystal structure of Imatinib mesylate complexed to the ABL kinase domain demonstrated that T334 provided a key hydrogen bond (H-bond) interaction for Imatinib mesylate association with ABL (235, 263, 264) (Figure 6D). The BCR-ABL T334I amino acid substitution would be predicted to prevent formation of the H-bond with Imatinib mesylate as well as generate steric hindrance for drug binding. This Imatinib mesylate BCR-ABL mutation validated 3-D structural findings of key Imatinib mesylate and ABL kinase domain contact points. Numerous additional BCR-ABL mutants resistant to Imatinib mesylate have subsequently been identified in CML patients (255, 258, 259) (Figures 1, 5). Some of these mutants occurred in distal N-terminal and C-terminal regions of the ABL kinase domain, not predicted to directly interact with Imatinib mesylate (Figure 5). Using a random mutagenesis approach, a range of BCR-ABL mutants outside of the ABL kinase domain, in the cap, SH3, and SH2 domains of ABL were identified as resistant to Imatinib mesylate inhibition (265) (Figure 5). These domains play pivotal roles in the transition from an inactive to active conformation of ABL (see section on Mechanism of ABL Tyrosine Kinase Activation). Since Imatinib mesylate binds to the inactive conformation of ABL, mutations in these ABL domains could be predicted to abrogate Imatinib mesylate binding. The BCR portion of BCR-ABL has not been reported to be sequenced in BCRABL clones resistant to Imatinib mesylate treatment. As discussed above, different regions of BCR are critical for BCR-ABL activation. These findings suggest that mutations in BCR that up-regulate BCR-ABL tyrosine kinase activity may also be relatively resistant to Imatinib mesylate. Some Imatinib mesylate–resistant patient samples with an activated BCR-ABL tyrosine kinase domain contain no mutations in the ABL kinase domain (266). Sequencing BCR-ABL in its entirety in these samples may identify novel mutations in unexpected regions of BCR-ABL that render the protein insensitive to Imatinib mesylate treatment.

BCR-ABL EXPRESSION MAY INCREASE THE RATE OF CELLULAR MUTAGENESIS LEADING TO THE EVOLUTION OF BCR-ABL MUTANTS RESISTANT TO IMATINIB MESYLATE TREATMENT BCR-ABL expression has been shown to enhance the frequency of spontaneous mutations by three- to fivefold in the BaF3 cell line (267) and by two- to threefold in the preleukemic p185BCR-ABL transgenic mouse stain (268). This BCR-ABL effect was dependent on its tyrosine kinase activity as administration of Imatinib mesylate to preleukemic p185BCR-ABL transgenic mice reduced the mutation rate in these mice (269). Some Imatinib mesylate–resistant BCR-ABL mutants occur

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY

281

prior to Imatinib mesylate therapy in CML patients (270). These results suggest that Imatinib mesylate suppression of BCR-ABL tyrosine kinase activity during the early stages of chronic phase CML may prevent/reduce the generation of BCRABL Imatinib mesylate–resistant mutants in treated patients. BCR-ABL expression has been shown to down-regulate proteins involved in DNA repair. Mutation in the xeroderma pigmentosum group B protein results in a human DNA repair disorder, and BCR-ABL expression inactivates this enzyme by binding to and phosphorylating the enzyme (271). BRCA1 deficiency in mice leads to heightened sensitivity to DNA double-strand breaks (272), and this protein is down-regulated upon BCR-ABL induction in multiple cell types (273). The DNA-dependent protein kinase (DNA-PK) complex, a major DNA repair system in mammalian cells involved in DNA double-strand breaks repair, was demonstrated to be down-regulated via proteasome-induced degradation in a dosedependent manner concommitant with increasing levels of BCR-ABL expression (274) (Table 2). Determining which DNA repair mechanism is largely responsible for the mutator phenotype present in BCR-ABL expressing cells may identify therapeutic approaches that could suppress the generation of Imatinib mesylate– resistant BCR-ABL mutants in CML patients.

PH+ CELL POPULATIONS THAT MAY NOT BE ELIMINATED BY IMATINIB MESYLATE THERAPY Ph+ HSCs May Not Be Eliminated by Imatinib Mesylate Treatment The origin of CML begins in a HSC [see section on Identification of the Chromosomal Abnormality and Cellular Origin of the Human Disease Chronic Myelogenous Leukemia (CML); 17, 18; reviewed in 19, 20], a target population that is largely quiescent (275), which either does not express BCR-ABL (276) or is unaffected by BCR-ABL expression (277). Quiescent Ph+ HSCs are insensitive to Imatinib mesylate treatment in vitro (278). These findings suggest that the Ph+ HSC may not be eliminated by Imatinib mesylate treatment in CML patients owing to its unresponsiveness to both the Ph chromosome and Imatinib mesylate. Life-long Imatinib mesylate therapy may be required to continually suppress leukemia development in CML patients. Expansion of Ph+ progenitors is inhibited by Imatinib meyslate (244). However, this cell population persists in Imatinib mesylate–treated CML patients (279). Although Imatinib mesylate therapy is able to suppress the oncogenic activity of wild-type BCR-ABL expression, the treatment is unable to destroy all Ph+ cell populations. As in Ph+ HSCs, the inability of Imatinib mesylate to eliminate Ph+ progenitors suggests that CML patients will require long-term Imatinib mesylate monotherapy, which may lead to the generation and selection of BCR-ABL escape mutants that are resistant to the inhibitor.

24 Feb 2004

19:26

282

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

REGIONS IN CML PATIENTS THAT DO NOT RECEIVE HIGH LEVELS OF IMATINIB MESYLATE COULD LEAD TO LEUKEMIA RELAPSE

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Central Nervous System (CNS) May Not Receive Imatinib Mesylate at Concentrations High Enough to Suppress BCR-ABL–Induced Leukemogenesis Imatinib mesylate treatment was unable to prevent CNS leukemia in a BCR-ABL– induced CML-like mouse model (280) because Imatinib mesylate was unable to reach the CNS at concentrations high enough to effectively suppress BCRABL tyrosine kinase activity. A lymphoid CML blast crisis patient with excellent systemic response to Imatinib mesylate was shown to relapse due to a lymphoid leukemia in the CNS (281). The cerebrospinal fluid (CSF) had a 2-log lower level of Imatinib mesylate compared to the blood plasma. These results suggest that CML patients in remission with Imatinib mesylate could eventually relapse with CNS leukemia due to the inefficiency of Imatinib mesylate at crossing the blood-brain barrier. Imaging the location of CML cells in a patient along with the distribution of the drug Imatinib mesylate may help define locations in the body where drug delivery needs to be improved to effectively treat this disease. We have recently demonstrated the ability to noninvasively image the development of a BCR-ABL–induced B-ALL in a mouse model using positron emission tomography (PET) (282). PET imaging has been successfully used to image the progression of neurological diseases (283), and small molecule compounds have been created for imaging (284). Combining these technologies, it should be possible to identify key regions in the body that do not allow Imatinib mesylate to penetrate and to test CML treatments that may overcome this current limitation.

THERAPEUTIC OPTIONS FOR IMATINIB MESYLATE RESISTANT BCR-ABL LEUKEMIAS Alternative Small Molecule Inhibitors The small molecule inhibitors PD 166326, PD 173995, and PD 180970 belonging to the class of pyrido [2,3-d] pyrimidine compounds were shown to selectively suppress the tyrosine kinase activity of a small group of proteins including BCR-ABL (285–287). These compounds required about tenfold less drug compared with Imatinib mesylate to inhibit BCR-ABL tyrosine kinase activity at similar levels. Such inhibitors may be useful at treating Ph+ patients resistant to the highest doses of Imatinib mesylate because of high levels of BCR-ABL expression.

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY

P1: IKH

283

Structural analysis of this class of compounds complexed to the ABL tyrosine kinase domain revealed that these inhibitors bound to ABL with the activation loop being extended outwards, suggesting an active conformation of the protein (Figures 1, 6B). Addition of the inhibitor PD 173995 to both phosphorylated and unphosphorylated forms of ABL demonstrated that the drug suppressed ABL tyrosine kinase activity in both active and inactive conformations (101, 264, 285). The differing structural requirements of this group of inhibitors compared to Imatinib mesylate at suppressing BCR-ABL tyrosine kinase activity suggested that this class of compounds may be able to inhibit Imatinib mesylate–resistant forms of BCR-ABL. PD 166326 and PD 180970 effectively inhibit cellular transformation induced by Imatinib mesylate–resistant BCR-ABL forms Y272F and E274K (287, 288). These inhibitors could be an effective treatment in patients harboring such BCR-ABL mutants. However, the Imatinib mesylate T334I–resistant BCRABL clone was resistant to both pyrido [2,3-d] pyrimidine inhibitors (287, 288). Interestingly, a corresponding ABL T334I mutation made in EGFR by converting threonine 766 to methionine led to severely decreased sensitivity of the EGFR mutant to a different class of 4-anilinoquinazoline inhibitor, highly effective at suppressing wild-type EGFR tyrosine kinase activity (289). 4-anilino-3-quinolinecarbonitrile SKI-606 compound with dual SRC and ABL specificity was shown to completely regress xenografts of Ph+ K562 cells in mice and block BCR-ABL tyrosine phosphorylation (290). Whether this small molecule compound belonging to a class of inhibitors different from either Imatinib mesylate or PD 166326–180970 is able to suppress the tyrosine kinase activity of the multidrug-resistant BCR-ABL T334I mutant awaits determination.

Inhibiting BCR-ABL mRNA Production Inhibition of BCR-ABL production at the mRNA level via antisense oligonucleotide, small-interfering RNA, peptide nucleic acid, or ribozyme expression of BCR-ABL junction sequences effectively suppresses BCR-ABL in vitro transformation (291–295). However, these strategies aimed at preventing the generation of BCR-ABL transcripts are limited by the lack of proven technologies to introduce sufficient amounts of such agents into a patient and thereby suppress leukemia.

Suppressing BCR-ABL Protein Levels Arsenic trioxide decreases BCR-ABL protein levels via inhibition of BCR-ABL translation (296) and 17-allylaminogeldanamycin degrades both wild-type and Imatinib mesylate–resistant BCR-ABL protein forms via inactivation of the Heat Shock protein 90 (HSP 90), which is required for proper folding of the BCRABL molecule (297–300). Whether these compounds have sufficient selectively at blocking BCR-ABL protein production without severely affecting normal protein translation awaits further study.

24 Feb 2004

19:26

284

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Inhibiting BCR-ABL Tyrosine Kinase Activity Expression of a truncated BCR protein encoding amino acids 64–413 inhibits CML cell growth by inhibiting BCR-ABL tyrosine kinase activity (138, 301). Expression of a chimeric fusion protein encoding a BCR-ABL binding domain fused to an active phosphatase region was also demonstrated to suppress BCRABL tyrosine kinase activity as well as its in vitro and in vivo transformation properties (302). These systems further validated the concept that inhibition of BCR-ABL tyrosine kinase activity was a valuable approach to suppress BCRABL–induced oncogenesis. However, unlike small molecule inhibitors, there is no current delivery system of sufficient efficiency to administer these BCR-ABL inhibitory proteins uniformly into a patient’s leukemic cells.

Treatments that Obstruct Signaling Components Potentially Critical for BCR-ABL Oncogenesis Inhibition of the RAS pathway via expression of a GRB2-SOS peptide blocker or treatment with the farnesyl transferase inhibitor SCH 66336 effectively suppresses BCR-ABL–induced in vitro cellular growth and in vivo leukemogenesis (145, 177, 303). The farnesyl transferase inhibitor R115777 was used to treat 22 patients representing chronic, accelerated, and blast phases of CML (304). Seven patients responded favorably, with a reduction in their WBC count (304). However, the hematological response was transient, with a median duration of nine weeks (304). These apparently opposing results could be resolved if the farnesyl transferase inhibitor R115777 is inferior to SCH 66336 at inhibiting RAS activation in an in vivo setting. However, if it is shown that both farnesyl transferase inhibitors are equally potent at inactivating RAS and that SCH 66336 leads to the same poor therapeutic response in CML patients as R115777, the results will reaffirm gene knockout studies demonstrating the difficulty in identifying a single molecule essential for BCR-ABL–induced leukemogenesis (see section on Murine Models of ABL Oncogenesis). If an approach of inhibiting signaling pathways critical to BCRABL oncogenesis is taken, effective therapy against Imatinib mesylate–resistant Ph+ patients is likely to require inactivation of multiple and not individual signaling modules regulated by BCR-ABL expression.

CONCLUSION The BCR-ABL story demonstrated that identifying and selectively targeting the primary oncogenic event shown to be essential for the development of Ph+ leukemias was sufficient to suppress the cancer. The effectiveness of Imatinib mesylate at inhibiting the tyrosine kinase activity of BCR-ABL and its direct correlation to disease remission has broadened the testing of this inhibitor to treating cancers where the known dominant oncogenic activity has been demonstrated to be efficiently blocked by the drug. In a phase 1 study of gastrointestinal stromal

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY

P1: IKH

285

tumors (GISTs), where the dominant oncogenic activity is mostly due to either KIT or PDGFR activation (305), treatment with Imatinib mesylate, which suppresses the tyrosine kinase activity of both these oncoproteins led to major responses in over 80% of patients (306). Imatinib mesylate is also being tested for the treatment of systemic mastocytosis, AML, and germ-cell tumors where KIT activation has been detected (307, 308) and in chronic myeloproliferative diseases such as hypereosinophilic syndrome and CML, where rearrangements of the PDGFRα or PDGFRβ occur (309–311). Other small-molecule inhibitors and antibodies are being developed and tested for the inhibition of selective oncogenic activities and their associated diseases. Activated FMS-like tyrosine kinase 3 (FLT3) is the most common molecular defect associated with AML (312) and several FLT3 inhibitors are in clinical testing (reviewed in 313). Acute promyelocytic leukemia patients associated with the PML-RARα fusion can now be effectively treated with all-trans-retinoic acid, which selectively blocks the activity of the chimeric oncoprotein leading to a 5-year disease-free survival of 75%–80% (reviewed in 314). A significant portion of Ph+ patients treated with Imatinib mesylateat the acute phases of the disease are resistant to the therapy. Most patients relapse from treatment due to expression of BCR-ABL mutant proteins with reduced or complete insensitivity to Imatinib mesylate suppression of ABL tyrosine kinase activity. This is also turning out to be the case in the treatment of systemic mastocytosis, AML, and germ-cell tumors where expression of a KIT D816V mutant resistant to Imatinib mesylate has been detected (307, 308). These findings suggest that a multidrug treatment approach similar to HIV combination therapy (reviewed in 315) will be necessary to prevent the development of escape oncoprotein mutants resistant to the inhibitors. The BCR-ABL T334I mutant is resistant to tyrosine kinase inhibition by different forms of ATP competitive compounds (see section on Therapeutic Options for Imatinib Mesylate-Resistant BCR-ABL Leukemias). This mutant may not be suppressible by any type of ATP competitive inhibitor. Multidrug therapy aimed at inhibiting BCR-ABL tyrosine kinase activity via different mechanisms such as blockade of oligomerization, inactivation of the BCR A and B boxes, or suppression of an active conformation of ABL could be required to effectively treat Ph+ leukemias. Suppression of BCR-ABL protein production via 17-AAG inhibition of HSP 90 has been shown to block cellular transformation by the multidrugresistant BCR-ABL T334I mutant and could be another therapeutic approach (300). Therapeutic approaches aimed at suppressing the leukemic activity of a Ph+ cell by Imatinib mesylate-like mechanisms are unlikely to destroy all cancerous cells in a patient and would most probably require life-long treatment. Eliminating the cellular origin of leukemia is the only curative approach for patients; in CML, this would require destroying the Ph+ HSC that is largely quiescent. AML quiescent HSCs have been shown to constitutively activate the NF-κB pathway and treatment of these leukemic cells with the proteosome inhibitor carbobenzoxyl-L-leucylL-L-leucinal, which prevents IκB degradation and NF-κB activation, severely

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

286

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

reduces or completely abrogates the engraftment of these leukemia HSCs into NOD/SCIDs without affecting control HSCs (316). BCR-ABL expression has also been shown to activate the NF-κB pathway in hematopoietic cell lines (317, 318). Determining whether Ph+ quiescent HSCs also show constitutive NF-κB activation or contain other deregulated pathways may identify molecular targets whose inhibition could eliminate the leukemic stem cell. BCR-ABL expression has been shown to regulate a plethora of signaling molecules, and it is unclear whether inhibition of any single signal transduction pathway would be sufficient to block leukemogenesis. Suppression of multiple signaling molecules and pathways regulated upon BCR-ABL induction with a combination of low-toxicity selective drugs could be a therapeutic approach for Imatinib mesylate resistant patients. Alternatively, gene profiling of patient specimens against known oncoproteins where basic science findings suggest, may identify unexpected therapeutic targets. LYN upregulation, for example, was shown to replace the requirement of BCR-ABL activation in expanding a Ph+ cell line. This Imatinib mesylate–resistant cell line was sensitive to LYN inactivation and suppression of cellular transformation (262; see section on Mechanism of Resistance to Imatinib Treatment). The well-understood pathogenesis of Ph+ leukemias is turning out to be fertile ground for the testing of novel therapies. Defining the effectiveness of these treatment approaches against Imatinib mesylate–resistant patients may help guide the next wave of cancer therapeutics.

ACKNOWLEDGMENTS We are grateful to Drs. Benjamin Neel and Rick Van Etten (Harvard University, Cambridge) for providing unpublished results; Gregory Ferl (UCLA) for the generation of the tyrosine kinase phylogenetic tree containing Imatinib mesylate– sensitive members (Figure 4); Drs. John Colicelli and Emmanuel Beillard for helpful discussions and critical reading of the manuscript; and to Barbara Anderson for excellent preparation of the manuscript, tables, and figures. Work from this laboratory is supported by grants from the National Institutes of Health and the U.S. Department of Energy. ONW is an Investigator of the Howard Hughes Medical Institute. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Virchow R. 1845. Weisses Blut. Froriep’s Neue Notizen aus dem Gebiet der Natur.und Heilkunde. Weimer 36:151–55 2. Craigie D. 1845. Case of disease of the spleen, in which death took place in conse-

quence of the presence of purulent matter in the blood. Edinb. Med. Surg. J. 64:400– 13 3. Bennett J. 1845. Case of hypertrophy of the spleen liver, in which death took place

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

4.

5.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

6.

7.

8.

9.

10.

11.

12.

13.

from suppuration of the blood. Edinb. Med. Surg. J. 64:413–23 Nowell PC, Hungerford DA. 1960. A minute chromosome in human chronic granulocytic leukemia. Science 132:1497 Blume-Jensen P, Hunter T. 2001. Oncogenic kinase signalling. Nature 411:355– 65 Shepherd P, Suffolk R, Halsey J, Allan N. 1995. Analysis of molecular breakpoint and m-RNA transcripts in a prospective randomized trial of interferon in chronic myeloid leukaemia: no correlation with clinical features, cytogenetic response, duration of chronic phase, or survival. Br. J. Haematol. 89:546–54 Westbrook CA, Hooberman AL, Spino C, Dodge RK, Larson RA, et al. 1992. Clinical significance of the BCR-ABL fusion gene in adult acute lymphoblastic leukemia: a Cancer and Leukemia Group B Study (8762). Blood 80:2983– 90 Lichtman MA. 1995. Chronic myelogenous leukemia and related disorders. In Williams Hematology, ed. E Buetler, MA Lichtman, BS Coller, TJ Kipps, pp. 298– 324. New York: McGraw-Hill Colleoni GW, Yamamoto M, Kerbauy J, Serafim RC, Lindsey CJ, et al. 1996. BCRABL rearrangement in adult T-cell acute lymphoblastic leukemia. Am. J. Hematol. 53:277–78 Fabbiano F, Santoro A, Felice R, Catania P, Cannella S, Majolino I. 1998. bcrabl rearrangement in adult T-lineage acute lymphoblastic leukemia. Haematologica 83:856–57 Mitelman F. 1993. The cytogenetic scenario of chronic myeloid leukemia. Leuk. Lymphoma 11:11–15 Jennings BA, Mills KI. 1998. c-myc locus amplification and the acquisition of trisomy 8 in the evolution of chronic myeloid leukaemia. Leuk. Res. 22:899– 903 Prokocimer M, Rotter V. 1994. Structure function of p53 in normal cells and

14.

15.

16.

17.

18.

19.

20.

21.

22.

P1: IKH

287

their aberrations in cancer cells: projection on the hematologic cell lineages. Blood 84:2391–411 Feinstein E, Cimino G, Gale RP, Alimena G, Bertheir R, et al. 1991. p53 in chronic myelogenous leukemia in acute phase. Proc. Natl. Acad. Sci. USA 88:6293– 97 Ahuja HG, Jat PS, Foti A, Bar-Eli M, Cline MJ. 1991. Abnormalities of the retinoblastoma gene in the pathogenesis of acute leukemia. Blood 78:3259– 68 Sill H, Goldman JM, Cross NCP. 1995. Homozygous deletions of the p16 tumorsuppressor gene are associated with lymphoid transformation of chronic myeloid leukemia. Blood 85:2013–16 Fialkow PJ, Jacobson RJ, Papayannopoulou T. 1977. Chronic myelocytic leukemia: clonal origin in a stem cell common to the granulocyte, erythrocyte, and platelet monocyte/macrophage. Am. J. Med. 63:125–30 Fialkow PJ, Gartler SM, Yoshida A. 1967. Clonal origin of chronic myelocytic leukemia in man. Proc. Natl. Acad. Arts Sci. 58:1468–71 Kabarowski JHS, Witte ON. 2000. Consequences of BCR-ABL expression within the haematopoietic stem cell in chronic myeloid leukemia. Stem Cells 18:399– 408 Takahashi N, Miura I, Saitoh K, Miura AB. 1998. Lineage involvement of stem cells bearing the philadelphia chromosome in chronic myeloid leukemia in the chronic phase as shown by a combination of fluorescence-activated cell sorting and fluorescence in situ hybridization. Blood 92:4758–63 Druker BJ, Sawyers CL, Capdeville R, Ford JM, Baccarani M, Goldman JM. 2001. Chronic myelogenous leukemia. Hematology 2001:87–112 Moloney JB. 1962. The murine leukemias. Fed. Proc. Fed. Am. Soc. Exp. Biol. 21:19–31

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

288

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

23. Abelson HT, Rabstein LS. 1970. Lymphosarcoma: virus-induced thymic-independent disease in mice. Cancer Res. 30: 2213–22 24. Rosenberg N, Witte ON. 1988. The viral and cellular forms of the Abelson (abl) oncogene. Adv. Virus Res. 39:39–81 25. Rabstein LS, Gazdar AF, Chopra HC, Abelson HT. 1971. Early morphological changes associated with infection by a murine nonthymic lymphatic tumor virus. J. Natl. Cancer Inst. 46:481–91 26. Siegler R, Zajdel S. 1972. Pathogenesis of Abelson-virus-induced murine leukemia. J. Natl. Cancer Inst. 48:189–218 27. Scher CD, Siegler R. 1975. Direct transformation of 3T3 cells by Abelson murine leukemia virus. Nature 253:729–31 28. Rosenberg N, Baltimore D, Scher CD. 1975. In vitro transformation of lymphoid cells by Abelson murine leukemia virus. Proc. Natl. Acad. Sci. USA 72:1932–36 29. Goff SP, Gilboa E, Witte ON, Baltimore O. 1980. Structure of the Abelson murine leukemia virus and the homologous cellular gene: studies with cloned viral DNA. Cell 22:777–85 30. Reynolds FH, Sacks TL, Deobagkar DN, Stephenson JR. 1978. Cells nonproductively transformed by Abelson murine leukemia virus express a high molecular weight polyprotein containing structural and non-structural components. Proc. Natl. Acad. Sci. USA 75:3974–78 31. Van de Ven WJM, Reynolds FH, Nalewaik RP, Stephenson JR. 1979. The nonstructural component of the Abelson murine leukemia virus polyprotein P120 is encoded by a newly acquired genetic sequences. J. Virol. 32:1041–45 32. Witte ON, Rosenberg N, Paskind M, Shields A, Baltimore D. 1978. Identification of an Abelson murine leukemia virus encoded protein present in transformed fibroblast and lymphoid cells. Proc. Natl. Acad. Sci. USA 75:2488–92 33. Witte ON, Rosenberg N, Baltimore D. 1979. Preparation of syngeneic tumor re-

34.

35.

36.

37.

38.

39.

40.

41.

42.

gressor serum reactive with the unique determinants of the Abelson murine leukemia virus-encoded P120 protein at the cell surface. J. Virol. 31:776–84 Witte ON, Ponticelli A, Gifford A, Baltimore D, Rosenberg N, Elder J. 1981. Phosphorylation of the Ableson murine leukemia virus transforming protein. J. Virol. 39:870–78 Ponticelli AS, Whitlock CA, Rosenberg N, Witte ON. 1982. In vivo tyrosine phosphorylations of the Abelson virus transforming protein are absent in its normal cellular homolog. Cell 29:953–60 Witte ON, Rosenberg NE, Baltimore D. 1979. Identification of a normal cellular protein cross-reactive to the major Abelson murine leukaemia virus gene product. Nature 281:396–98 Collett MS, Erikson RL. 1978. Protein kinase activity associated with the avian sarcoma virus src gene product. Proc. Natl. Acad. Sci. USA 75:2021–24 Witte ON, Sun L, Rosenberg N, Baltimore D. 1979. A trans-acting kinase identified cells transformed by Abelson MuLV cells. Cold Spring Harbor Symp. Quant. Biol. 44:855–57 Van de Ven WJM, Reynolds FH Jr, Stephenson JR. 1980. The nonstructural components of polyproteins encoded by replication-defective mammalian transforming retroviruses are phosphorylated and have associated protein kinase activity. Virology 101:185–97 Witte ON, Goff S, Rosenberg N, Baltimore D. 1980. A transformation-defective mutant of Abelson murine leukemia virus lacks protein kinase activity. Proc. Natl. Acad. Sci. USA 8:4993–97 Krebs EG, Beavo JA. 1979. Phosphorylation-dephosphorylation of enzymes. Annu. Rev. Biochem. 48:923–59 Witte ON, Dasgupta A, Baltimore D. 1980. Abelson murine leukaemia virus protein is phophorylated in vitro to form phosphotyrosine. Nature 283:826– 31

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY 43. Ambros V, Baltimore D. 1978. Protein is linked to the 50 end of poliovirus RNA by a phosphodiester linkage to tyrosine. J. Biol. Chem. 253:5263–66 44. Kirkegaard K, Wang JC. 1978. Escherichia coli DNA topoisomerase I catalyzed linking of single-stranded rings of complementary base sequences. Nucleic Acids Res. 5:3811–20 45. Hunter T, Sefton BM. 1980. The transforming gene product of Rous sarcoma virus phosphorylates tyrosine. Proc. Natl. Acad. Sci. USA 77:1311–15 46. Konopka JB, Witte ON. 1985. Detection of c-abl tyrosine kinase activity in vitro permits direct comparison of normal and altered abl gene products. Mol. Cell. Biol. 5:3116–23 47. Rees-Jones RW, Goff SP. 1988. Insertional mutagenesis of the Abelson murine leukemia virus genome: Identification of mutants with altered kinase activity and defective transformation ability. J. Virol. 62:978–86 48. Engelman A, Rosenberg N. 1990. Temperature-sensitive mutants of Abelson murine leukemia virus deficient in protein tyrosine kinase activity. J. Virol. 64:4242–51 49. Rowley JD. 1973. A new consistent chromosomal abnormality in chronic myelogeneous leukemia identified by quinacrine fluorescence and giemsa staining. Nature 243:290–93 50. Heisterkamp N, Groffen J, Stephenson JR, Spurr NK, Goodfellow PN, et al. 1982. Chromosomal localization of human cellular homologues of two viral oncogenes. Nature 299:747–49 51. Groffen J, Heisterkamp N, Stephenson JR, van Kessel AG, de Klein A, et al. 1983. c-sis is translocated from chromosome 22 to chromosome 9 in chronic myelocytic leukemia. J. Exp. Med. 158:9–15 52. Bartram CR, de Klein A, Hagemeijer A, van Agthoven T, van Kessel AG, et al. 1983. Translocation of c-abl oncogene correlates with the presence of a Philadel-

53.

54.

55.

56.

57.

58.

59.

60.

P1: IKH

289

phia chromosome in chronic myelocytic leukaemia. Nature 306:277–80 de Klein A, van Kessel AG, Grosveld G, Bartram CR, Hagemeijer A, et al. 1982. A cellular oncogene is translocated to the Philadelphia chromosome in chronic myelocytic leukaemia. Nature 300:765– 67 Canaani E, Steiner-Saltz D, Aghai E, Gale RP, Berrebi A, Januszewicz E. 1984. Altered transcription of an oncogene in chronic myeloid leukaemia. Lancet 1: 593–95 Konopka JB, Watanabe SM, Witte ON. 1984. An alteration of the human c-abl protein in K562 leukemia cells unmasks associated tyrosine kinase activity. Cell 37:1035–42 Ben-Neriah Y, Daley GQ, Mes-Masson A-M, Witte ON, Baltimore D. 1986. The chronic myelogenous leukemia-specific P210 protein is the product of the bcr/abl hybrid gene. Science 233:212– 14 Konopka JB, Watanabe SM, Singer JW, Collins SJ, Witte ON. 1985. Cell lines and clinical isolates derived from Ph1-positive chronic myelogenous leukemia patients express c-abl proteins with a common structural alteration. Proc. Natl. Acad. Sci. USA 82:1810–14 Davis RL, Konopka JB, Witte ON. 1985. Activation of the c-abl oncogene by viral transduction or chromosomal translocation generates altered c-abl proteins with similar in vitro kinase properties. Mol. Cell. Biol. 5:204–13 Konopka JB, Davis RL, Watanabe SM, Ponticelli AS, Schiff-Maker L, et al. 1984. Only site-directed antibodies reactive with the highly conserved scrhomologous region of the v-abl protein neutralize kinase activity. J. Virol. 51:223–32 Pendergast AM, Gishizky ML, Havlik MH, Witte ON. 1993. SH1 domain autophosphorylation of P210 BCR/ABL is required for transformation but not

24 Feb 2004

19:26

290

61.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

62.

63.

64.

65.

66.

67.

68.

69.

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

growth factor-independence. Mol. Cell. Biol. 13:1728–36 Groffen J, Stephenson JR, Heisterkamp N, de Klein A, Bartram CR, Grosveld G. 1984. Philadelphia chromosomal breakpoints are clustered within a limited region, bcr, on chromosome 22. Cell 36:93– 99 Heisterkamp N, Stam K, Groffen J, de Klein A, Grosveld G. 1985. Structural organization of the bcr gene and its role in the Ph’ translocation. Nature 315:758–61 Mes-Masson A-M, McLaughin J, Daley G, Paskind M, Witte ON. 1986. Overlapping cDNA clones define the complete coding region for the P210 +cabl gene product associated with chronic myelogenous leukemia cells containing the Philadelphia chromosome. Proc. Natl. Acad. Sci. USA 83:9768–72 Shtivelman E, Lifshitz B, Gale RP, Roe BA, Canaani E. 1985. Fused transcript of abl and bcr genes in chronic myelogenous leukaemia. Nature 315:550–54 Grosveld G, Verwoerd T, van Agthoven T, de Klein A, Ramachandran KL, et al. 1986. The chronic myelocytic cell line K562 contains a breakpoint in bcr and produces a chimeric bcr/c-abl transcript. Mol. Cell. Biol. 6:607–16 Stam K, Heisterkamp N, Reynolds FH Jr, Groffen J. 1987. Evidence that the phl gene encodes a 160,000-dalton phosphoprotein with associated kinase activity. Mol. Cell. Biol. 7:1955–60 Melo JV. 1996. The diversity of BCRABL fusion proteins and their relationship to leukemia phenotype. Blood 88:2375– 84 Clark SC, McLaughlin J, Timmons M, Pendergast AM, Ben-Neriah Y, et al. 1988. Expression of a distinctive BCRABL oncogene in Ph1-positive acute lymphocytic leukemia (ALL). Science 239:775–77 Hermans A, Heisterkamp N, von Lindern M, van Baal S, Meijer D, et al. 1987. Unique fusion of bcr and c-abl genes in

70.

71.

72.

73.

74.

75.

76.

77.

78.

Philadelphia chromosome positive acute lymphoblastic leukemia. Cell 51:33– 40 Kurzrock R, Shtalrid M, Romero P, Kloetzer WS, Talpas M, et al. 1987. A novel c-abl protein product in Philadelphiapositive acute lymphoblastic leukaemia. Nature 325:631–35 Chan LC, Karhi KK, Rayter SI, Heisterkamp N, Eridani S, et al. 1987. A novel abl protein expressed in Philadelphia chromosome positive acute lymphoblastic leukaemia. Nature 325:635– 37 Denny CT, Shah N, Ogden S, Willman C, McConnell T, et al. 1989. Localization of preferential sites of rearrangement within the BCR gene in Philadelphia chromosome positive acute lymphoblastic leukemia. Proc. Natl. Acad. Sci. USA 86:4254–58 Clark SS, McLaughlin J, Crist WM, Champlin R, Witte ON. 1987. Unique forms of the abl tyrosine kinase distinguish Ph1-positive CML from Ph1positive ALL. Science 235:85–88 Fainstein E, Marcelle C, Rosner A, Canaani E, Gale RP, et al. 1987. A new fused transcript in Philadephia chromosome positive acute lymphocytic leukaemia. Nature 330:386–88 Walker LC, Ganesan TS, Dhut S, Gibbsons B, Lister TA, et al. 1987. Novel chimaeric protein expressed in Philadelphia positive acute lymphoblastic leukaemia. Nature 329:851–53 Heisterkamp N, Stephenson JR, Groffen J, Hansen PF, de Klein A, et al. 1983. Localization of the c-abl oncogene adjacent to a translocation break point in chronic myelocytic leukaemia. Nature 306:239– 42 Deininger MW, Goldman JM, Melo JV. 2000. The molecular biology of chronic myeloid leukemia. Blood 96:3343–56 Kantarjian HM, Talpaz M, Dhingra K, Estey E, Keating MJ, et al. 1991. Significance of the P210 versus P190

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

79.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

80.

81.

82.

83.

84.

85.

86.

molecular abnormalities in adults with Philadelphia chromosome-positive acute leukemia. Blood 78:2411–18 Martiat P, Mecucci C, Nizet Y, Stul M, Phillippe M, et al. 1990. P190 BCR/ABL transcript in a case of Philadelphiapositive multiple myeloma. Leukemia 4:751–54 Melo JV, Myint H, Galton DA, Goldman JM. 1994. P190BCR-ABL chronic myeloid leukaemia: the missing link with chronic myelomonocytic leukaemia? Leukemia 8:208–11 Kurzrock R, Shtalrid M, Talpaz M, Kloetzer WS, Gutterman JU. 1987. Expression of c-abl in Philadelphia-positive acute myelogenous leukemia. Blood 70:1584– 88 Mitani K, Sato Y, Tojo A, Ishikawa F, Kobayashi Y, et al. 1990. Philadelphia chromosome positive B-cell type malignant lymphoma expressing an aberrant 190 kDa bcr-abl protein. Br. J. Haematol. 76:221–25 Pane F, Frigeri F, Sindona M, Luciano L, Ferrara F, et al. 1996. NeutrophilicChronic Myeloid Leukemia: A distinct disease with a specific molecular marker (BCR/ABL with C3/A2 junction). Blood 88:2410–14 Wilson G, Frost L, Goodeve A, Vandenberghe E, Peake I, Reilly J. 1997. BCR-ABL transcript with an e19a2 (c3a2) junction in classical chronic myeloid leukemia. Blood 88:2410–14 Haferlach T, Winkemann M, RammPetersen L, Meeder M, Schoch R, et al. 1997. New insights into the biology of Philadelphia-chromosome-positive acute lymphoblastic leukaemia using a combination of May-Grunwald-Giemsa staining and fluorescence in situ hybridization techniques at the single cell level. Br. J. Haematol. 99:452–59 Kasprzyk A, Harrison CJ, Secker-Walker LM. 1999. Investigation of clonal involvement of myeloid cells in Philadelphiapositive and high hyperdiploid acute

87.

88.

89.

90.

91.

92.

93.

94.

95.

P1: IKH

291

lymphoblastic leukemia. Leukemia 13: 2000–6 Biernaux C, Sels A, Huez G, Stryckmans P. 1996. Very low level of major BCR-ABL expression in blood of some healthy individuals. Bone Marrow Transplant. (Suppl.) 3:S45–47 Bose S, Deininger M, Gora-Tybor J, Goldman JM, Melo JV. 1998. The presence of typical and atypical BCR-ABL fusion genes in leukocytes of normal individuals: biologic significance and implications for the assessment of minimal residual disease. Blood 92:3362–67 Molldrem JJ, Lee PP, Wang C, Champlin RE, Davis MM. 1999. A PR1human leukocyte antigen-A2 tetramer can be used to isolate low-frequency cytotoxic T lymphocytes from healthy donors that selectively lyse chronic myelogenous leukemia. Cancer Res. 59:2675–81 Pinilla-Ibarz J, Cathcart K, Scheinberg DA. 2000. CML vaccines as a paradigm of the specific immunotherapy of cancer. Blood Rev. 14:111–20 Melo JV, Gordon DE, Cross NCP, Goldman JM. 1993. The ABL-BCR fusion gene is expressed in chronic myeloid leukemia. Blood 81:158–65 Lazaridou A, Chase A, Melo J, Garicochea B, Diamond J, Goldman J. 1994. Lack of reciprocal translocation in BCR-ABL positive Ph-negative chronic myeloid leukaemia. Leukemia 8:454– 57 Papadopoulos P, Ridge SA, Boucher CA, Stocking C. 1995. The novel activation of A BL by fusion to an ets-related gene, TEL1. Cancer Res. 55:34–38 Golub TR, Goga A, Barker GF, Afar DE, McLaughlin J, et al. 1996. Oligomerization of the ABL tyrosine kinase by the ETS protein TEL in human leukemia. Mol. Cell. Biol. 16:4107–16 Van Limbergen H, Beverloo HB, van Drunen E, Janssens A, Hahlen K, et al. 2001. Molecular cytogenetic and clinical findings in ETV6/ABL1-positive

24 Feb 2004

19:26

292

96.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

97.

98.

99.

100.

101.

102.

103.

104. 105.

106.

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

leukemia. Genes Chromosomes Cancer 30:274–82 Andreasson P, Johansson B, Carlsson M, Jarlsfelt I, Fioretos T, et al. 1997. BCR/ABL-negative chronic myeloid leukemia with ETV6/ABL fusion. Genes Chromosomes Cancer 20:299–304 Daley GQ, Van Etten RA, Jackson PK, Bernards A, Baltimore D. 1992. Nonmyristoylated Abl proteins transform a factor-dependent hematopoietic cell line. Mol. Cell. Biochem. 12:1864–71 Daley GQ, McLaughlin J, Witte ON, Baltimore D. 1987. The CML-specific P210 bcr/abl protein, unlike v-abl, does not transform NIH-3T3 fibroblasts. Science 237:532–37 Jackson P, Baltimore D. 1989. N-terminal mutations activate the leukemogenic potential of the myristoylated form of c-abl. EMBO J. 8:449–56 Hantschel O, Nagar B, Guettler S, Kretzschmar J, Dorey K, et al. 2003. A myristoyl/phosphotyrosine switch regulates cAbl. Cell 112:845–57 Nagar B, Hantschel O, Young MA, Scheffzek K, Veach D, et al. 2003. Structural basis for the autoinhibition of c-Abl tyrosine kinase. Cell 112:859–71 Barila D, Superti-Furga G. 1998. An intramolecular SH3-domain interaction regulates c-Abl activity. Nat. Genet. 18:280– 82 Tanis KQ, Veach D, Duewel HS, Bornmann WG, Koleske AJ. 2003. Two distinct phosphorylation pathways have additive effects on Abl family kinase activation. Mol. Cell. Biol. 23:3884–96 Harrison SC. 2003. Variation on an Srclike theme. Cell 112:737–40 McWhirter JR, Galasso DL, Wang JYJ. 1993. A coiled-coil oligomerization domain of Bcr is essential for the transforming function of Bcr-Abl oncoproteins. Mol. Cell. Biol. 13:7587–95 McWhirter JR, Wang JYJ. 1991. Activation of tyrosine kinase and microfilamentbinding functions of c-abl by Ibcr se-

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

quences in bcr/abl fusion proteins. Mol. Cell. Biol. 11:1553–65 Muller AJ, Young JC, Pendergast AM, Pondel M, Landau NR, et al. 1991. BCR first exon sequences specifically activate Bcr/Abl tyrosine kinase oncogene of Philadelphia chromosome positive human leukemias. Mol. Cell. Biol. 11:1785– 92 McWhirter JR, Wang JY. 1997. Effect of Bcr sequences on the cellular function of the Bcr-Abl oncoprotein. Oncogene 15:1625–34 Maru Y, Afar DE, Witte ON, Shibuya M. 1996. The dimerization property of glutathione S-transferase partially reactivates Bcr-Abl lacking the oligermerization domain. J. Biol. Chem. 271:15353–57 Lugo TG, Pendergast A, Muller AJ, Witte ON. 1990. Tyrosine kinase activity transformation potency of bcr-abl oncogene products. Science 247:1079–82 Franz WM, Berger P, Wang JYJ. 1989. Deletion of an N-terminal regulatory domain of the c-abl tyrosine kinase activates its oncogenic potential. EMBO J. 8:137– 47 Goga A, McLaughlin J, Pendergast AM, Parmar K, Muller A, et al. 1993. Oncogenic activation of c-ABL by mutation within its last exon. Mol. Cell. Biol. 13:4967–75 Maru Y, Witte ON, Shibuya M. 1996. Deletion of the ABL SH3 domain reactivates de-oligomerized BCR-ABL for growth factor independence. FEBS Lett. 379:244–46 He Y, Wertheim JA, Xu L, Miller JP, Karnell FG, et al. 2002. The coiled-coil domain and Tyr177 of bcr are required to induce a murine chronic myelogenous leukemia-like disease by bcr/abl. Blood 99:2957–68 Brasher BB, Roumiantsev S, Van Etten RA. 2001. Mutational analysis of the regulatory function of the c-Abl Src homology 3 domain. Oncogene 20:7744–52 Dai Z, Pendergast AM. 1995. Abi-2, a

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

117.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

118.

119.

120.

121.

122.

123.

124.

125.

novel SH3-containing protein interacts with the c-Abl tyrosine kinase and modulates c-Abl transforming activity. Genes Dev. 9:2569–82 Shi Y, Alin K, Goff SP. 1995. Ablinteractor-1, a novel SH3 protein binding to the carboxy-terminal portion of the Abl protein, suppresses v-Abl transforming activity. Genes Dev. 9:2583–97 Dai Z, Quackenbush RC, Courtney KD, Grove M, Cortez D, et al. 1998. Oncogenic abl and src tyrosine kinases elicit the ubiquitin-dependent degradation of target proteins through a ras-independent pathway. Genes Dev. 12:1415–24 Pendergast AM, Muller AJ, Havlik MH, Maru Y, Witte ON. 1991. BCR sequences essential for transformation by the BCR/ABL oncogene bind to the ABL SH2 regulatory domain in a nonphosphotyrosine-dependent manner. Cell 66:161–71 Afar DEH, Goga A, McLaughlin J, Witte O, Sawyers CL. 1994. Differential complementation of BCR-ABL point mutants with c-MYC. Science 264:424–26 Goga A, McLaughlin J, Afar DEH, Saffran DC, Witte ON. 1995. Alternative signals to RAS for hematopoietic transformation by the BCR-ABL oncogene. Cell 82:981–88 Mayer BJ, Jackson PK, Van Etten RA, Baltimore D. 1992. Point mutations in the abl SH2 domain coordinately impair phosphotyrosine binding in vitro and transforming activity in vivo. Mol. Cell. Biol. 12:609–18 McWhirter JR, Wang JYJ. 1993. An actinbinding function contributes to transformation by the bcr-abl oncoprotein of philadelphia chromosome-positive human leukemias. EMBO J. 12:1533–46 Puil L, Liu J, Gish G, Mbamalu G, Bowtell D, et al. 1994. Bcr-Abl oncoproteins bind directly to activators of the Ras signalling pathway. EMBO J. 13:764–73 Tauchi T, Boswell HS, Leibowitz D, Broxmeyer HE. 1994. Coupling between

126.

127.

128.

129.

130.

131.

132.

133.

134.

P1: IKH

293

p210bcr-abl and Shc and Grb2 adaptor proteins in hematopoietic cells permits growth factor receptor-independent link to Ras activation pathway. J. Exp. Med. 179:167–75 Gishizky ML, Cortez D, Pendergast AM. 1995. Mutant forms of growth factorbinding protein-2 reverse BCR-ABLinduced transformation. Proc. Natl. Acad. Sci. USA 92:10889–93 Sattler M, Mohi MG, Pride YB, Quinnan LR, Malouf NA, et al. 2002. Critical role for Gab2 in transformation by BCR/ABL. Cancer Cell 1:479–92 Cortez D, Kadlec L, Pendergast AM. 1995. Structural and signaling requirements for BCR-ABL-mediated transformation and inhibition of Apoptosis. Mol. Cell. Biol. 15:5531–41 Carpino N, Wisniewski D, Strife A, Marshak D, Kobayashi R, 1997. p62dok: A constitutively tyrosine-phosphorylated, GAP-associated protein in chronic myelogenous leukemia progenitor cells. Cell 88:197–204 Yamanashi Y, Baltimore D. 1997. Identification of the Abl- and rasGAP-associated 62 kDa protein as a docking protein, Dok. Cell 88:205–11 Zhang X, Subrahmanyam R, Wong R, Gross AW, Ren R. 2001. The NH2terminal coiled-coil domain and tyrosine 177 play important roles in induction of a myeloproliferative disease in mice by Bcr-Abl. Mol. Cell. Biol. 21:840–53 Wertheim JA, Forsythe K, Druker BJ, Hammer D, Boettiger D, Pear WS. 2002. BCR-ABL-induced adhesion defects are tyrosine kinase-independent. Blood 99:4122–30 Skourides PA, Perera SA, Ren R. 1999. Polarized distribution of Bcr-Abl in migrating myeloid cells and co-localization of Bcr-Abl and its target proteins. Oncogene 18:1165–76 Heaney C, Kolibaba K, Bhat A, Oda T, Ohno S, et al. 1997. Direct binding of CRKL to BCR-ABL is not required

24 Feb 2004

19:26

294

135.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

136.

137.

138.

139.

140.

141.

142.

143.

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

for BCR-ABL transformation. Blood 89: 297–306 Senechal K, Halpern J, Sawyers CL. 1996. The CRKL adaptor protein transforms fibroblasts and functions in transformation by the BCR-ABL oncogene. J. Biol. Chem. 271:23255–61 Maru YM, Witte ON. 1991. The BCR gene encodes a novel serine/threonine kinase activity within a single exon. Cell 67:459–68 Liu JX, Wu Y, Ma GZ, Lu D, Haataja L, et al. 1996. Inhibition of Bcr serine kinase by tyrosine phosphorylation. Mol. Cell. Biol. 16:998–1005 Liu JX, Wu Y, Arlinghaus RB. 1996. Sequences within the first exon of BCR inhibit the activated tyrosine kinases of cAbl and the Bcr-Abl oncoprotein. Cancer Res. 56:5120–24 Radziwill G, Erdmann RA, Margelisch U, Moelling K. 2003. The Bcr kinase downregulates Ras signaling by phosphorylating AF-6 and binding to its PDZ domain. Mol. Cell. Biol. 23:4663–72 Cortez D, Reuther G, Pendergast AM. 1997. The Bcr-Abl tyrosine kinase activates mitogenic signaling pathways stimulates G1-to-S phase transition in hematopoietic cells. Oncogene 15:2333–42 Skorski T, Bellacosa A, NieborowskaSkorska M, Majewski M, Martinez R, et al. 1997. Transformation of hematopoietic cells by BCR/ABL requires activation of a PI-3k/Akt-dependent pathway. EMBO J. 16:6151–61 Mandanas RA, Leibowitz DS, Gharehbaghi K, Tauchi T, Burgess GS, et al. 1993. Role of p21 RAS in p210 bcr-abl transformation of murine myeloid cells. Blood 82:1838–47 Voss J, Posern G, Hannemann JR, Wiedemann LM, Turhan AG, et al. 2000. The leukaemic oncoproteins Bcr-Abl T and el-Abl (ETV6/Abl) have altered substrate preferences and activate similar intracellular signalling pathways. Oncogene 19:1684–90

144. Sawyers CL, McLaughlin J, Witte ON. 1995. Genetic requirement for ras in the transformation of fibroblasts hematopoietic cells by the Bcr-Abl oncogene. J. Exp. Med. 181:307–13 145. Peters DG, Hoover RR, Gerlach MJ, Koh EY, Zhang H, et al. 2001. Activity of the farnesyl protein transferase inhibitor SCH66336 against BCR/ABLinduced murine leukemia and primary cells from patients with chronic myeloid leukemia. Blood 97:1404–12 146. Varticovski L, Daley GQ, Jackson P, Baltimore D, Cantley LC. 1991. Activation of phosphatidylinositol 3-kinase in cells expressing abl oncogene variants. Mol. Cell. Biol. 11:1107–13 147. Neshat MS, Raitano AB, Wang HG, Reed JC, Sawyers CL. 2000. The survival function of the Bcr-Abl oncogene is mediated by Bad-dependent and -independent pathways: roles for phosphatidylinositol 3kinase and Raf. Mol. Cell. Biol. 20:1179– 86 148. Hoover RR, Gerlach MJ, Koh EY, Daley GQ. 2001. Cooperative and redundant effects of STAT5 and Ras signaling in BCR/ABL transformed hematopoietic cells. Oncogene 20:5826–35 149. Gaston I, Stenberg PE, Bhat A, Druker BJ. 2000. Abl kinase but not PI3-kinase links to the cytoskeletal defects in Bcr-Abl transformed cells. Exp. Hematol. 28:77– 86 150. Witt O, Sand K, Pekrun A. 2000. Butyrate-induced erythroid differentiation of human K562 leukemia cells involves inhibition of ERK and activation of p38 MAP kinase pathways. Blood 95:2391–96 151. Raitano AB, Halpern JR, Hambuch TM, Sawyers CL. 1995. The Bcr-Abl leukemia oncogene activates Jun kinase and requires Jun for transformation. Proc. Natl. Acad. Sci. USA 92:11746–50 152. Klejman A, Rushen L, Morrione A, Slupianek A, Skorski T. 2002. Phosphatidylinositol-3 kinase inhibitors enhance

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

153.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

154.

155.

156.

157.

158.

159.

160.

the anti-leukemia effect of STI571. Oncogene 21:5868–76 Zhao RC, Jiang Y, Verfaillie CM. 2001. A model of human p210(bcr/ABL)mediated chronic myelogenous leukemia by transduction of primary normal human CD34+ cells with a BCR/ABL-containing retroviral vector. Blood 97:2406–12 Moore S, Haylock DN, Levesque JP, McDiarmid LA, Samels LM, et al. 1998. Stem cell factor as a single agent induces selective proliferation of the Philadelphia chromosome positive fraction of chronic myeloid leukemia CD34+ cells. Blood 92:2461–70 Gishizky ML, Witte ON. 1992. Initiation of deregulated growth of multipotent progenitor cells by bcr-abl in vitro. Science 256:836–39 Pierce A, Spooncer E, Wooley S, Dive C, Francis JM, et al. 2000. Bcr-Abl protein tyrosine kinase activity induces a loss of p53 protein that mediates a delay in myeloid differentiation. Oncogene 19:5487–97 Perlingeiro RC, Kyba M, Daley GQ. 2001. Clonal analysis of differentiating embryonic stem cells reveals a hematopoietic progenitor with primitive erythroid and adult lymphoid-myeloid potential. Development 128:4597–604 Era T, Witte ON. 2000. Regulated expression of P210 Bcr-Abl during embryonic stem cell differentiation stimulates multipotential progenitor expansion and myeloid cell fate. Proc. Natl. Acad. Arts Sci. 97:1737–42 Wong S, McLaughlin J, Cheng D, Witte O. 2003. Cell context specific effects of the BCR-ABL oncogene monitored in hematopoietic progenitors. Blood 101:4088–97 Ohmine K, Ota J, Ueda M, Ueno S, Yoshida K, et al. 2001. Characterization of stage progression in chronic myeloid leukemia by DNA microarray with purified hematopoietic stem cells. Oncogene 20:8249–57

P1: IKH

295

161. Hariharan IK, Harris AW, Crawford M, Abud H, Webb E, et al. 1989. A bcr-v-abl oncogene induces lymphomas in transgenic mice. Mol. Cell. Biol. 9:2798–805 162. Heisterkamp N, Jenster G, ten Hoeve J, Zovich D, Pattengale PK, Groffen J. 1990. Acute leukaemia in bcr/abl transgenic mice. Nature 344:251–53 163. Wong S, Witte ON. 2001. Modeling Philadelphia chromosome positive leukemias. Oncogene 20:5644–59 164. Clark SS, Chen E, Fizzotti M, Witte ON, Malkovska V. 1993. BCR-ABL and vabl oncogenes induce distinct patterns of thymic lymphoma involving different lymphocyte subsets. J. Virol. 67:6033–46 165. Daley GQ, Van Etten RA, Baltimore D. 1990. Induction of chronic myelogenous leukemia in mice by the P210 bcr/abl gene of the Philadelphia chromosome. Science 247:824–30 166. Kelliher MA, McLaughlin J, Witte ON, Rosenberg N. 1990. Induction of a chronic myelogenous leukemia-like syndrome in mice with v-abl BCR/ABL. Proc. Natl. Acad. Sci. USA 87:6649–53 167. Gishizky ML, Johnson-White J, Witte ON. 1993. Efficient transplantation of BCR/ABL induced chronic myelogenous leukemia-like syndrome in mice. Proc. Natl. Acad. Sci. USA 90:3755–59 168. Kelliher M, Knott A, McLaughlin J, Witte ON, Rosenberg N. 1991. Differences in oncogenic potency but not target cell specificity distinguish the two forms of the BCR/ABL oncogene. Mol. Cell. Biol. 11:4710–16 169. Pear WS, Miller JP, Xu L, Pui JC, Soffer B, et al. 1998. Efficient and rapid induction of a chronic myelogenous leukemialike myeloproliferative disease in mice receiving P210 bcr/abl-transduced bone marrow. Blood 92:3780–92 170. Zhang X, Ren R. 1998. Bcr-Abl efficiently induces a myeloproliferative disease and production of excess interleukin-3 and granulocyte-macrophage colony-stimulating factor in mice: a novel

24 Feb 2004

19:26

296

171.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

172.

173.

174.

175.

176.

177.

178.

179.

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

model for chronic myelogenous leukemia. Blood 92:3829–40 Henkemeyer M, West SR, Gertier FB, Hoffmann FM. 1990. A novel tyrosine kinase-independent function of Drosophila abl correlates with proper subcellular localization. Cell 63:949–60 Heisterkamp N, Voncken JW, Senadheera D, Hemmeryckx B, Gonzalez-Gomez I, et al. 2001. The Bcr N-terminal oligomerization domain contributes to the full oncogenicity of P190 Bcr/Abl in transgenic mice. Int. J. Mol. Med. 7:351–57 Zhang X, Wong R, Hao SX, Pear WS, Ren R. 2001. The SH2 domain of Bcr-Abl is not required to induce a murine myeloproliferative disease; however, SH2 signaling influences disease latency phenotype. Blood 97:277–87 Roumiantsev S, de Aos IE, Varticovski L, Ilaria RL, Van Etten RA. 2001. The Src homology 2 domain of Bcr/Abl is required for efficient induction of chronic myeloid leukemia-like disease in mice but not for lymphoid leukemogenesis or activation of phosphatidylinositol 3-kinase. Blood 97:4–13 Million RP, Van Etten RA. 2000. The Grb2 binding site is required for the induction of chronic myeloid leukemia-like disease in mice by the Bcr/Abl tyrosine kinase. Blood 96:664–70 Ren R. 2002. The molecular mechanism of chronic myelogenous leukemia and its therapeutic implications: studies in a murine model. Oncogene 21:8629–42 Reichert A, Heisterkamp N, Daley GQ, Groffen J. 2001. Treatment of Bcr/Ablpositive acute lymphoblastic leukemia in P190 transgenic mice with the farnesyl transferase inhibitor. Blood 97:1399–403 Peters DG, Klucher KM, Perlingeiro RC, Dessain SK, Koh EY, Daley GQ. 2001. Autocrine and paracrine effects of an ES-cell derived, BCR/ABL-transformed hematopoietic cell line that induces leukemia in mice. Oncogene 20:2636–46 Shuai K, Halpern J, ten Hoeve J, Rao X,

180.

181.

182.

183.

184.

185.

186.

187.

Sawyers CL. 1996. Constitutive activation of STAT5 by the BCR-ABL oncogene in chronic myelogenous leukemia. Oncogene 13:247–54 Ilaria RL Jr, Van Etten RA. 1996. P210 P190(BCR/ABL) induce the tyrosine phosphorylation and DNA binding activity of multiple specific STAT family members. J. Biol. Chem. 271:31704– 10 Sexl V, Piekorz R, Moriggl R, Rohrer J, Brown MP, et al. 2000. Stat5a/b contribute to interleukin 7-induced B-cell precursor expansion, but abl- bcr/abl-induced transformation are independent of stat5. Blood 96:2277–83 Di Cristofano A, Niki M, Zhao M, Karnell FG, Clarkson B, et al. 2001. p62(dok), a negative regulator of Ras and mitogenactivated protein kinase (MAPK) activity, opposes leukemogenesis by p210(bcrabl). J. Exp. Med. 194:275–84 Li S, Gillessen S, Tomasson MH, Dranoff G, Gilliland DG, Van Etten RA. 2001. Interleukin 3 and granulocytemacrophage colony-stimulating factor are not required for induction of chronic myeloid leukemia-like myeloproliferative disease in mice by BCR/ABL. Blood 97:1442–50 Huettner CS, Zhang P, Van Etten RA, Tenen DG. 2000. Reversibility of acute B-cell leukaemia induced by BCR-ABL1. Nat. Genet. 24:57–60 Kabarowski JHS, Allen PB, Wiedemann L. 1994. A temperature sensitive p210 BCR-ABL mutant defines the primary consequences of BCR-ABL tyrosine kinase expression in growth factor dependent cells. EMBO J. 13:5887–95 Driggers PH, Ennist DL, Gleason SL, Mak WH, Marks MS, et al. 1990. An interferon gamma-regulated protein that binds the interferon-inducible enhancer element of major histocompatibility complex class I genes. Proc. Natl. Acad. Arts Sci. 87:3743–77 Holtschke T, Lohler J, Kanno Y, Fehr T,

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

188.

189.

190.

191.

192.

193.

194.

195.

196.

Giese N, et al. 1996. Immunodeficiency chronic and myelogenous leukemia-like syndrome in mice with a targeted mutation of the ICSBP gene. Cell 87:307–17 Hao SX, Ren R. 2000. Expression of interferon consensus sequence binding protein (ICSBP) is downregulated in BcrAbl-induced murine chronic myelogenous leukemia-like disease, and forced coexpression of ICSBP inhibits Bcr-Ablinduced myeloproliferative disorder. Mol. Cell. Biol. 20:1149–61 Eklund EA, Jalava A, Kakar R. 1998. PU.1, interferon regulatory factor 1, and interferon consensus sequence-binding protein cooperate to increase gp91(phox) expression. J. Biol. Chem. 273:13957–65 Lord KA, Abdollahi A, HoffmanLiebermann B, Liebermann DA. 1993. Proto-oncogenes of the fos/jun family of transcription factors are positive regulators of myeloid differentiation. Mol. Cell. Biol. 13:841–51 Mollinedo F, Vaquerizo MJ, Naranjo JR. 1991. Expression of c-jun, jun B and jun D proto-oncogenes in human peripheralblood granulocytes. Biochem. J. 273:477– 79 Passegue E, Jochum W, Schorpp-Kistner M, Mohle-Steinlein U, Wagner EF. 2001. Chronic myeloid leukemia with increased granulocyte progenitors in mice lacking JunB expression in the myeloid lineage. Cell 104:21–32 Bakiri L, Lallemand D, Bossy-Wetzel E, Yaniv M. 2000. Cell cycle-dependent variations in c-Jun and JunB phosphorylation: a role in the control of cyclin D1 expression. EMBO J. 19:2056–68 Passegue E, Wagner EF. 2000. JunB suppresses cell proliferation by transcriptional activation of p16(INK4a) expression. EMBO J. 19:2969–79 Afar DEH, McLaughlin J, Sherr CJ, Witte ON, Roussel MF. 1995. Signaling by ABL oncogenes through cyclin D1. Proc. Natl. Acad. Sci.USA 92:9540–44 Jaiswal S, Traver D, Miyamoto T, Akashi

197.

198.

199.

200. 201.

202.

203.

204.

205.

P1: IKH

297

K, Lagasse E, Weissman IL. 2003. Expression of BCR-ABL and BCL-2 in myeloid progenitors leads to myeloid leukemias. Proc. Natl. Acad. Sci. USA 100(17):10002–7 Helgason CD, Damen JE, Rosten P, Grewal R, Sorensen P, et al. 1999. Targeted disruption of SHIP leads to hemopoietic perturbations, lung pathology, and a shortened life span. Genes Dev. 12:1610–20 Sattler M, Verma S, Byrne CH, Shrikhande G, Winkler T, et al. 1999. BCR/ABL directly inhibits expression of SHIP, an SH2-containing polyinositol-5phosphatase involved in the regulation of hematopoiesis. Mol. Cell. Biol. 19: 7473–80 Gishizky ML, Witte ON. 1992. BCR/ABL enhances growth of multipotent progenitor cells but does not block their differentiation potential in vitro. Curr. Top. Microbiol. Immunol. 182:65–72 Sawyers CL. 1999. Chronic myeloid leukemia. N. Engl. J. Med. 340:1330–40 Druker BJ, O’Brien SG, Cortes J, Radich J. 2002. Chronic myelogenous leukemia. Hematology 2002:111–35 Salesse S, Verfaillie CM. 2002. BCR/ABL: from molecular mechanisms of leukemia induction to treatment of chronic myelogenous leukemia. Oncogene 21:8547–59 Talpaz M, Kantarjian HM, McCredie K, Trujillo JM, Keating MJ, Gutterman JU. 1986. Hematologic remission and cytogenetic improvement induced by recombinant human interferon alpha-A in chronic myelogenous leukemia. N. Engl. J. Med. 314:1065–69 Kantarjian HM, O’Brien S, Cortes JE, Shan J, Giles FJ, et al. 2003. Complete cytogenetic and molecular responses to interferon-alpha-based therapy for chronic myelogenous leukemia are associated with excellent long-term prognosis. Cancer 97:1033–41 Darnell JE Jr, Kerr IM, Stark GR. 1994. Jak-STAT pathways and transcriptional

24 Feb 2004

19:26

298

206.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

207.

208.

209.

210.

211.

212.

213.

214.

215.

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

activation in response to IFNs and other extracellular signaling proteins. Science 264:1415–21 Landolfo S, Guarini A, Riera L, Gariglio M, Gribaudo G, et al. 2000. Chronic myeloid leukemia cells resistant to interferon-alpha lack STAT1 expression. Hematol. J. 1:7–14 Sakai I, Takeuchi K, Yamauchi H, Narumi H, Fujita S. 2002. Constitutive expression of SOCS3 confers resistance to IFN-alpha in chronic myelogenous leukemia cells. Blood 100:2926–31 Starr R, Willson TA, Viney EM, Murray LJL, Rayner JR, et al. 1997. A family of cytokine-inducible inhibitors of signalling. Nature 387:917–21 Endo TA, Masuhara M, Yokouchi M, Suzuki R, Sakamoto H, et al. 1997. A new protein containing an SH2 domain that inhibits JAK kinases. Nature 387:921–24 Sakamoto H, Yasukawa H, Masuhara M, Tanimura S, Sasaki A, et al. 1998. A Janus kinase inhibitor, JAB, is an interferongamma-inducible gene and confers resistance to interferons. Blood 92:1668– 76 Schmidt M, Nagel S, Proba J, Thiede C, Ritter M, et al. 1998. Lack of interferon consensus sequence binding protein (ICSBP) transcripts in human myeloid leukemias. Blood 91:22–29 Molldrem JJ, Lee PP, Kant S, Wieder E, Jiang WD, et al. 2003. Chronic myelogenous leukemia shapes host immunity by selective deletion of high-avidity leukemia-specific T cells. J. Clin. Invest. 111:639–47 Deng M, Daley GQ. 2001. Expression of interferon consensus sequence binding protein induces potent immunity against BCR/ABL-induced leukemia. Blood 97: 3491–97 Levitzki A, Gazit A. 1995. Tyrosine kinase inhibition: an approach to drug development. Science 267:1782–88 Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. 2002. The pro-

216.

217.

218.

219.

220.

221.

222.

223.

224.

tein kinase complement of the human genome. Science 298:1912–34 Venter JC, Adams MD, Myers EW, Li PW, Mural RJ, et al. 2001. The sequence of the human genome. Science 291:1304–51 Uehara Y, Hori M, Takeuchi T, Umezawa H. 1985. Screening of agents which convert ‘transformed morphology’ of Rous sarcoma virus-infected rat kidney cells to ‘normal morphology’: identification of an active agent as herbimycin and its inhibition of intracellular src kinase. Jpn. J. Cancer Res. 76:672–75 Uehara Y, Murakami Y, Mizuno S, Kawai S. 1988. Inhibition of transforming activity of tyrosine kinase oncogenes by herbimycin A. Virology 164:294–98 Taniguchi M, Uehara Y, Matsuyama M, Takahashi M. 1993. Inhibition of ret tyrosine kinase activity by herbimycin A. Biochem. Biophys. Res. Commun. 195:208–14 Fukazawa H, Li PM, Yamamoto C, Murakami Y, Mizuno S, Uehara Y. 1991. Specific inhibition of cytoplasmic protein tyrosine kinases by herbimycin A in vitro. Biochem. Pharmacol. 42:1661–71 Sepp-Lorenzino L, Ma Z, Lebwohl DE, Vinitsky A, Rosen N. 1995. Herbimycin A induces the 20 S proteasome- ubiquitindependent degradation of receptor tyrosine kinases. J. Biol. Chem. 270:16580– 87 Akiyama T, Ishida J, Nakagawa S, Ogawara H, Watanabe S, et al. 1987. Genistein, a specific inhibitor of tyrosinespecific protein kinases. J. Biol. Chem. 262:5592–95 Umezawa H, Imoto M, Sawa T, Isshiki K, Matsuda N, et al. 1986. Studies on a new epidermal growth factor-receptor kinase inhibitor, erbstatin, produced by MH435hF3. J. Antibiot. 39:170–73 Umezawa K, Hori T, Tajima H, Imoto M, Isshiki K, Takeuchi T. 1990. Inhibition of epidermal growth factor-induced DNA synthesis by tyrosine kinase inhibitors. FEBS Lett. 260:198–200

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE BCR-ABL STORY 225. Carlo-Stella C, Dotti G, Mangoni L, Regazzi E, Garau D, et al. 1996. Selection of myeloid progenitors lacking BCR/ABL mRNA in chronic myelogenous leukemia patients after in vitro treatment with the tyrosine kinase inhibitor genistein. Blood 88:3091–100 226. Kawada M, Tawara J, Tsuji T, Honma Y, Hozumi M. 1993. Inhibition of Abelson oncogene function by erbstatin analogues. Drugs Exp. Clin. Res. 19:235–41 227. Yaish P, Gazit A, Gilon C, Levitzki A. 1988. Blocking of EGF-dependent cell proliferation by EGF receptor kinase inhibitors. Science 242:933–35 228. Levitzki A. 2002. Tyrosine kinases as targets for cancer therapy. Eur. J. Cancer 38(Suppl. 5):S11–18 229. Anafi M, Gazit A, Zehavi A, Ben-Neriah Y, Levitzki A. 1993. Tyrphostin-induced inhibition of p210bcr-abl tyrosine kinase activity induces K562 to differentiate. Blood 82:3524–29 230. Carlo-Stella C, Regazzi E, Sammarelli G, Colla S, Garau D, et al. 1999. Effects of the tyrosine kinase inhibitor AG957 and an Anti-Fas receptor antibody on CD34+ chronic myelogenous leukemia progenitor cells. Blood 93:3973–82 231. Bhatia R, Munthe HA, Verfaillie CM. 1998. Tyrphostin AG957, a tyrosine kinase inhibitor with anti-BCR/ABL tyrosine kinase activity restores beta1 integrin-mediated adhesion and inhibitory signaling in chronic myelogenous leukemia hematopoietic progenitors. Leukemia 12:1708–17 232. Mow BM, Chandra J, Svingen PA, Hallgren CG, Weisberg E, et al. 2002. Effects of the Bcr/abl kinase inhibitors STI571 and adaphostin (NSC 680410) on chronic myelogenous leukemia cells in vitro. Blood 99:664–71 233. Meydan N, Grunberger T, Dadi H, Shahar M, Arpaia E, et al. 1996. Inhibition of acute lymphoblastic leukaemia by a Jak-2 inhibitor. Nature 379:645–48 234. Traxler P, Bold G, Buchdunger E, Cara-

235.

236.

237.

238.

239.

240.

241.

242.

P1: IKH

299

vatti G, Furet P, et al. 2001. Tyrosine kinase inhibitors: from rational design to clinical trials. Med. Res. Rev. 21:499– 512 Manley PW, Cowan-Jacob SW, Buchdunger E, Fabbro D, Fendrich G, et al. 2002. Imatinib: a selective tyrosine kinase inhibitor. Eur. J. Cancer 38(Suppl. 5):S19–27 Buchdunger E, Zimmermann J, Mett H, Meyer T, Muller M, et al. 1995. Selective inhibition of the platelet-derived growth factor signal transduction pathway by a protein-tyrosine kinase inhibitor of the 2phenylaminopyrimidine class. Proc. Natl. Acad. Arts Sci. 92:2558–62 Hubbard SR, Till JH. 2000. Protein tyrosine kinase structure and function. Annu. Rev. Biochem. 69:373–98 Geissler JF, Roesel JL, Meyer T, Trinks UP, Traxler P, Lydon NB. 1992. Benzopyranones and benzothiopyranones: a class of tyrosine protein kinase inhibitors with selectivity for the v-abl kinase. Cancer Res. 52:4492–98 Carroll M, Ohno-Jones S, Tamura S, Buchdunger E, Zimmermann J, et al. 1997. CGP 57148, a tyrosine kinase inhibitor, inhibits the growth of cells expressing BCR-ABL, TEL-ABL, and TEL-PDGFR fusion proteins. Blood 90: 4947–52 Buchdunger E, Cioffi CL, Law N, Stover D, Ohno-Jones S, et al. 2000. Abl proteintyrosine kinase inhibitor STI571 inhibits in vitro signal transduction mediated by c-kit and platelet-derived growth factor receptors. J. Pharmacol. Exp. Ther. 295:139–45 Heinrich MC, Griffith DJ, Druker BJ, Wait CL, Ott KA, Zigler AJ. 2000. Inhibition of c-kit receptor tyrosine kinase activity by STI 571, a selective tyrosine kinase inhibitor. Blood 96:925–32 Buchdunger E, Zimmermann J, Mett H, Meyer T, M¨uller M, et al. 1996. Inhibition of the Abl protein-tyrosine kinase in vitro and in vivo by a

24 Feb 2004

19:26

300

243.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

244.

245.

246.

247.

248.

249.

250.

251.

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

2-phenylaminopyrimidine derivative. Cancer Res. 56:100–4 Okuda K, Weisberg E, Gilliland DG, Griffin JD. 2001. ARG tyrosine kinase activity is inhibited by STI571. Blood 97:2440–48 Druker BJ, Tamura S, Buchdunger E, Ohno S, Segal GM, et al. 1996. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat. Med. 2:561–66 Deininger MW, Goldman JM, Lydon N, Melo JV. 1997. The tyrosine kinase inhibitor CGP57148B selectively inhibits the growth of BCR-ABL-positive cells. Blood 90:3691–98 Gambacorti-Passerini C, le Coutre P, Mologni L, Fanelli M, Bertazzoli C, et al. 1997. Inhibition of the ABL kinase activity blocks the proliferation of BCR/ABL+ leukemic cells and induces apoptosis. Blood Cells Mol. Dis. 23:380–94 Buchdunger E, O’Reilly T, Wood J. 2002. Pharmacology of imatinib (STI571). Eur. J. Cancer 38(Suppl. 5):S28–36 Druker BJ, Talpaz M, Resta DJ, Peng B, Buchdunger E, et al. 2001. Efficacy and safety of a specific inhibitor of the BCRABL tyrosine kinase in chronic myeloid leukemia. N. Engl. J. Med. 344:1031–37 Kantarjian H, Sawyers C, Hochhaus A, Guilhot F, Schiffer C, et al. 2002. Hematologic and cytogenetic responses to imatinib mesylate in chronic myelogenous leukemia. N. Engl. J. Med. 346:645–52 Druker BJ, Sawyers CL, Kantarjian H, Resta DJ, Reese SF, et al. 2001. Activity of a specific inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N. Engl. J. Med. 344:1038–42 Talpaz M, Silver RT, Druker BJ, Goldman JM, Gambacorti-Passerini C, et al. 2002. Imatinib induces durable hematologic and cytogenetic responses in patients with accelerated phase chronic myeloid leukemia: results of a phase 2 study. Blood 99:1928–37

252. Sawyers CL, Hochhaus A, Feldman E, Goldman JM, Miller CB, et al. 2002. Imatinib induces hematologic and cytogenetic responses in patients with chronic myelogenous leukemia in myeloid blast crisis: results of a phase II study. Blood 99:3530– 39 253. Kantarjian HM, Talpaz M, O’Brien S, Giles F, Garcia-Manero G, et al. 2003. Dose escalation of imatinib mesylate can overcome resistance to standard-dose therapy in patients with chronic myelogenous leukemia. Blood 101:473–75 254. Gorre ME, Mohammed M, Ellwood K, Hsu N, Paquette R, et al. 2001. Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science 293:876–80 255. Shah NP, Nicoll JM, Nagar B, Gorre ME, Paquette RL, et al. 2002. Multiple BCR-ABL kinase domain mutations confer polyclonal resistance to the tyrosine kinase inhibitor imatinib (STI571) in chronic phase and blast crisis chronic myeloid leukemia. Cancer Cell 2:117–25 256. le Coutre P, Tassi E, Varella-Garcia M, Barni R, Mologni L, et al. 2000. Induction of resistance to the abelson inhibitor STI571 in human leukemic cells through gene amplification. Blood 95:1758–66 257. Mahon FX, Deininger MW, Schultheis B, Chabrol J, Reiffers J, et al. 2000. Selection and characterization of BCR-ABL positive cell lines with differential sensitivity to the tyrosine kinase inhibitor STI571: diverse mechanisms of resistance. Blood 96:1070–79 258. Branford S, Rudzki Z, Walsh S, Parkinson I, Grigg A, et al. 2003. Detection of BCRABL mutations in imatinib-treated CML patients is virtually always accompanied by clinical resistance and mutations in the ATP phosphate-binding loop (P-loop) are associated with a poor prognosis. Blood 102:276–83 259. Corbin AS, Rosee PL, Stoffregen EP, Druker BJ, Deininger MW. 2003. Several Bcr-Abl kinase domain mutants

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

260.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

261.

262.

263.

264.

265.

266.

267.

268.

associated with imatinib mesylate resistance remain sensitive to imatinib. Blood 101:4611–14 Mahon FX, Belloc F, Lagarde V, Chollet C, Moreau-Gaudry F, et al. 2003. MDR1 gene overexpression confers resistance to imatinib mesylate in leukemia cell line models. Blood 101:2368–73 Di Bacco A, Keeshan K, McKenna SL, Cotter TG. 2000. Molecular abnormalities in chronic myeloid leukemia: deregulation of cell growth apoptosis. Oncologist 5:405–15 Donato NJ, Wu JY, Stapley J, Gallick G, Lin H, et al. 2003. BCR-ABL independence and LYN kinase overexpression in chronic myelogenous leukemia cells selected for resistance to STI571. Blood 101:690–98 Schindler T, Bornmann W, Pellicena P, Miller WT, Clarkson B, Kurlyan J. 2000. Structural mechanism of STI-571 inhibition of abelson tyrosine kinase. Science 289:1938–42 Nagar B, Bornmann WG, Pellicena P, Schindler T, Veach DR, et al. 2002. Crystal structures of the kinase domain of cAbl in complex with the small molecule inhibitors PD173955 imatinib (STI-571). Cancer Res. 62:4236–43 Azam M, Latek RR, Daley GQ. 2003. Mechanisms of autoinhibition and STI571/imatinib resistance revealed by mutagenesis of BCR-ABL. Cell 112:831– 43 Hochhaus A, Kreil S, Corbin AS, La Rosee P, Muller MC, et al. 2002. Molecular and chromosomal mechanisms of resistance to imatinib (STI571) therapy. Leukemia 16:2190–96 Canitrot Y, Lautier D, Laurent G, Frechet M, Ahmed A, et al. 1999. Mutator phenotype of BCR–ABL transfected Ba/F3 cell lines and its association with enhanced expression of DNA polymerase beta. Oncogene 18:2676–80 Salloukh HF, Laneuville P. 2000. Increase in mutant frequencies in mice expressing

269.

270.

271.

272.

273.

274.

275.

276.

277.

278.

P1: IKH

301

the BCR-ABL activated tyrosine kinase. Leukemia 14:1401–4 Brain J, Saksena A, Laneuville P. 2002. The kinase inhibitor STI571 reverses the Bcr-Abl induced point mutation frequencies observed in pre-leukemic P190(Bcr-Abl) transgenic mice. Leuk. Res. 26:1011–16 Kreuzer KA, Le Coutre P, Landt O, Na IK, Schwarz M, et al. 2003. Preexistence and evolution of imatinib mesylate-resistant clones in chronic myelogenous leukemia detected by a PNA-based PCR clamping technique. Ann. Hematol. 82:284–89 Takeda N, Shibuya M, Maru Y. 1999. The BCR-ABL oncoprotein potentially interacts with the xeroderma pigmentosum group B protein. Proc. Natl. Acad. Sci. USA 96:203–7 Tomlinson GE, Chen TT, Stastny VA, Virmani AK, Spillman MA, et al. 1998. Characterization of a breast cancer cell line derived from a germ-line BRCA1 mutation carrier. Cancer Res. 58:3237–42 Deutsch E, Jarrousse S, Buet D, Dugray A, Bonnet ML, et al. 2003. Down-regulation of BRCA1 in BCR-ABL-expressing hematopoietic cells. Blood 101:4583–88 Deutsch E, Dugray A, AbdulKarim B, Marangoni E, Maggiorella L, et al. 2001. BCR-ABL down-regulates the DNA repair protein DNA-PKcs. Blood 97:2084– 90 Spangrude GJ, Heimfeld S, Weissman IL. 1988. Purification and characterization of mouse hematopoietic cells. Science 241:58–62 Negrin RS, Weissman I. 1992. Hematopoietic stem cells in normal and malignant states. Marrow Transpl. Rev. 2:23–26 Holyoake T, Jiang X, Eaves C, Eaves A. 1999. Isolation of a highly quiescent subpopulation of primitive leukemic cells in chronic myeloid leukemia. Blood 94: 2056–64 Graham SM, Jorgensen HG, Allan E, Pearson C, Alcorn MJ, et al. 2002. Primitive, quiescent, Philadelphia-positive

24 Feb 2004

19:26

302

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

279.

280.

281.

282.

283.

284.

285.

286.

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro. Blood 99:319–25 Bhatia R, Holtz M, Niu N, Gray R, Snyder DS, et al. 2003. Persistence of malignant hematopoietic progenitors in chronic myelogenous leukemia patients in complete cytogenetic remission following imatinib mesylate treatment. Blood 101:4701–7 Wolff NC, Richardson JA, Egorin M, Ilaria RL Jr. 2003. The CNS is a sanctuary for leukemic cells in mice receiving imatinib mesylate for Bcr/Abl-induced leukemia. Blood 101:5010–13 Petzer AL, Gunsilius E, Hayes M, Stockhammer G, Duba HC, et al. 2002. Low concentrations of STI571 in the cerebrospinal fluid: a case report. Br. J. Haematol. 117:623–25 Le LQ, Kabarowski JH, Wong S, Nguyen K, Gambhir SS, Witte ON. 2002. Positron emission tomography imaging analysis of G2A as a negative modifier of lymphoid leukemogenesis initiated by the BCRABL oncogene. Cancer Cell 1:381–91 Phelps ME. 2000. Inaugural article: positron emission tomography provides molecular imaging of biological processes. Proc. Natl. Acad. Sci. USA 97: 9226–33 Iyer M, Barrio JR, Namavari M, Bauer E, Satyamurthy N, et al. 2001. 8[18F]Fluoropenciclovir: an improved reporter probe for imaging HSV1-tk reporter gene expression in vivo using PET. J. Nucl. Med. 42:96–105 Wisniewski D, Lambek CL, Liu C, Strife A, Veach DR, et al. 2002. Characterization of potent inhibitors of the Bcr-Abl and the c-kit receptor tyrosine kinases. Cancer Res. 62:4244–55 Huang M, Dorsey JF, Epling-Burnette PK, Nimmanapalli R, Landowski TH, et al. 2002. Inhibition of Bcr-Abl kinase activity by PD180970 blocks constitutive activation of Stat5 and growth of CML cells. Oncogene 21:8804–16

287. Huron DR, Gorre ME, Kraker AJ, Sawyers CL, Rosen N, Moasser MM. 2003. A novel pyridopyrimidine inhibitor of Abl kinase is a picomolar inhibitor of Bcr-abl-driven K562 cells and is effective against STI571-resistant Bcr-abl mutants. Clin. Cancer Res. 9:1267–73 288. La Rosee P, Corbin AS, Stoffregen EP, Deininger MW, Druker BJ. 2002. Activity of the Bcr-Abl kinase inhibitor PD180970 against clinically relevant Bcr-Abl isoforms that cause resistance to imatinib mesylate (Gleevec, STI571). Cancer Res. 62:7149–53 289. Blencke S, Ullrich A, Daub H. 2003. Mutation of threonine 766 in the epidermal growth factor receptor reveals a hotspot for resistance formation against selective tyrosine kinase inhibitors. J. Biol. Chem. 278:15435–40 290. Golas JM, Arndt K, Etienne C, Lucas J, Nardin D, et al. 2003. SKI-606, a 4-anilino-3-quinolinecarbonitrile dual inhibitor of Src and Abl kinases, is a potent antiproliferative agent against chronic myelogenous leukemia cells in culture and causes regression of K562 xenografts in nude mice. Cancer Res. 63:375– 81 291. Wilda M, Fuchs U, Wossmann W, Borkhardt A. 2002. Killing of leukemic cells with a BCR/ABL fusion gene by RNA interference (RNAi). Oncogene 21:5716–24 292. Rapozzi V, Burm BE, Cogoi S, van der Marel GA, van Boom JH, et al. 2002. Antiproliferative effect in chronic myeloid leukaemia cells by antisense peptide nucleic acids. Nucleic Acids Res. 30:3712– 21 293. Szczylik C, Skorski T, Nicolaides NC, Manzella L, Malaguarnera L, et al. 1991. Selective inhibition of luekemia cell proliferation by BCR-ABL antisense oligodeoxynucleotides. Science 253:562– 65 294. Lange W, Cantin EM, Finke J, Dolken G. 1993. In vitro in vivo effects of synthetic

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

295.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

296.

297.

298.

299.

300.

301.

302.

ribozymes targeted against BCR/ABL mRNA. Leukemia 7:1786–94 Wright L, Wilson SB, Milliken S, Biggs J, Kearney P. 1993. Ribozyme-mediated cleavage of the bcr/abl transcript expressed in chronic myeloid leukemia. Exp. Hematol. 21:1714–18 Perkins C, Kim CN, Fang G, Bhalla KN. 2000. Arsenic induces apoptosis of multidrug-resistant human myeloid leukemia cells that express Bcr-Abl or overexpress MDR, MRP, Bcl-2, or Bclx(L). Blood 95:1014–22 Nimmanapalli R, O’Bryan E, Bhalla K. 2001. Geldanamycin and its analogue 17allylamino-17-demethoxygeldanamycin lowers Bcr-Abl levels and induces apoptosis and differentiation of Bcr-Ablpositive human leukemic blasts. Cancer Res. 61:1799–804 An WG, Schulte TW, Neckers LM. 2000. The heat shock protein 90 antagonist geldanamycin alters chaperone association with p210bcr-abl and v-src proteins before their degradation by the proteasome. Cell Growth Differ. 11:355–60 Blagosklonny MV, Fojo T, Bhalla KN, Kim JS, Trepel JB, et al. 2001. The Hsp90 inhibitor geldanamycin selectively sensitizes Bcr-Abl-expressing leukemia cells to cytotoxic chemotherapy. Leukemia 15:1537–43 Gorre ME, Ellwood-Yen K, Chiosis G, Rosen N, Sawyers CL. 2002. BCRABL point mutants isolated from patients with imatinib mesylate-resistant chronic myeloid leukemia remain sensitive to inhibitors of the BCR-ABL chaperone heat shock protein 90. Blood 100:3041–44 Wang Y, Liu J, Wu Y, Luo W, Lin S-H, et al. 2001. Expression of a truncated first exon BCR sequence in chronic myelogenous leukemia cells blocks cell growth induces cell death. Cancer Res. 61:138–44 Lim Y-M, Wong S, Lau G, Witte ON, Colicelli J. 2000. BCR/ABL inhibition by an escort/phosphatase fusion protein. Proc. Natl. Acad. Sci. USA 97:12233–38

P1: IKH

303

303. Kardinal C, Konkol B, Lin H, Eulitz M, Schmidt EK, et al. 2001. Chronic myelogenous leukemia blast cell proliferation is inhibited by peptides that disrupt Grb2SoS complexes. Blood 98:1773–81 304. Cortes J, Albitar M, Thomas D, Giles F, Kurzrock R, et al. 2003. Efficacy of the farnesyl transferase inhibitor R115777 in chronic myeloid leukemia other hematologic malignancies. Blood 101:1692–97 305. Heinrich MC, Corless CL, Duensing A, McGreevey L, Chen CJ, et al. 2003. PDGFRA activating mutations in gastrointestinal stromal tumors. Science 299:708–10 306. Demetri GD. 2002. Identification and treatment of chemoresistant inoperable or metastatic GIST: experience with the selective tyrosine kinase inhibitor imatinib mesylate (STI571). Eur. J. Cancer 38(Suppl. 5):S52–59 307. Zermati Y, De Sepulveda P, Feger F, Letard S, Kersual J, et al. 2003. Effect of tyrosine kinase inhibitor STI571 on the kinase activity of wild-type and various mutated c-kit receptors found in mast cell neoplasms. Oncogene 22:660–64 308. Frost MJ, Ferrao PT, Hughes TP, Ashman LK. 2002. Juxtamembrane mutant V560GKit is more sensitive to Imatinib (STI571) compared with wild-type ckit whereas the kinase domain mutant D816VKit is resistant. Mol. Cancer Ther. 1:1115–24 309. Baxter EJ, Kulkarni S, Vizmanos JL, Jaju R, Martinelli G, et al. 2003. Novel translocations that disrupt the plateletderived growth factor receptor beta (PDGFRB) gene in BCR-ABL-negative chronic myeloproliferative disorders. Br. J. Haematol. 120:251–56 310. Cools J, DeAngelo DJ, Gotlib J, Stover EH, Legare RD, et al. 2003. A tyrosine kinase created by fusion of the PDGFRA and FIP1L1 genes as a therapeutic target of imatinib in idiopathic hypereosinophilic syndrome. N. Engl. J. Med. 348:1201–14

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

304

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

311. Apperley JF, Gardembas M, Melo JV, Russell-Jones R, Bain BJ, et al. 2002. Response to imatinib mesylate in patients with chronic myeloproliferative diseases with rearrangements of the plateletderived growth factor receptor beta. N. Engl. J. Med. 347:481–87 312. Armstrong SA, Kung AL, Mabon ME, Silverman LB, Stam RW, et al. 2003. Inhibition of FLT3 in MLL. Validation of a therapeutic target identified by gene expression based classification. Cancer Cell 3:173–83 313. Sawyers CL. 2002. Finding the next Gleevec: FLT3 targeted kinase inhibitor therapy for acute myeloid leukemia. Cancer Cell 1:413–15 314. Gilliland DG, Tallman MS. 2002. Focus on acute leukemias. Cancer Cell 1:417– 20 315. Thaker HK, Snow MH. 2003. HIV viral suppression in the era of antiretroviral therapy. Postgrad. Med. J. 79:36–42 316. Guzman ML, Swiderski CF, Howard DS, Grimes BA, Rossi RM, et al. 2002. Preferential induction of apoptosis for primary human leukemic stem cells. Proc. Natl. Acad. Sci. USA 99:16220–25 317. Reuther JY, Reuther GW, Cortez D, Pendergast AM, Baldwin AS Jr. 1998. A requirement for NF-kappaB activation in Bcr-Abl-mediated transformation. Genes Dev. 12:968–81 318. Hamdane M, David-Cordonnier MH, D’Halluin JC. 1997. Activation of p65 NF-kappaB protein by p210BCR-ABL in a myeloid cell line (P210BCR-ABL activates p65 NF-kappaB). Oncogene 15: 2267–75 319. Gross AW, Zhang X, Ren R. 1999. BcrAbl with an SH3 deletion retains the ability to induce a myeloproliferative disease in mice, yet c-Abl activated by an SH3 deletion induces only lymphoid malignancy. Mol. Cell Biochem. 19:6918–28 320. Anderson SM, Mladenovic J. 1996. The BCR-ABL oncogene requires both kinase activity and src-homology 2 domain to

321.

322.

323.

324.

325.

326.

327.

induce cytokine secretion. Blood 87:238– 44 Jiang Y, Zhao RC, Verfaillie CM. 2000. Abnormal integrin-mediated regulation of chronic myelogenous leukemia CD34+ cell proliferation: BCR/ABL up-regulates the cyclin-dependent kinase inhibitor, p27Kip, which is relocated to the cell cytoplasm and incapable of regulating cdk2 activity. Proc. Natl. Acad. Arts Sci. 97:10538–43 Parada Y, Banerji L, Glassford J, Lea NC, Collado M, et al. 2001. BCR-ABL and interleukin 3 promote haematopoietic cell proliferation and survival through modulation of cyclin D2 and p27Kip1 expression. J. Biol. Chem. 276:23572–80 Ptasznik A, Urbanowska E, Chinta S, Costa MA, Katz BA, et al. 2002. Crosstalk between BCR/ABL oncoprotein and CXCR4 signaling through a Src family kinase in human leukemia cells. J. Exp. Med. 196:667–78 Jiang X, Lopez A, Holyoake T, Eaves A, Eaves C. 1999. Autocrine production and action of IL-3 and granulocyte colony-stimulating factor in chronic myeloid leukemia. Proc. Natl. Acad. Arts Sci. 96:12804–9 Wisniewski D, Strife A, Swendeman S, Erdjument-Bromage H, Geromanos S, et al. 1999. A novel SH2-containing phosphatidylinositol 3,4,5-trisphosphate 5-phosphatase (SHIP2) is constitutively tyrosine phosphorylated and associated with src homologous and collagen gene (SHC) in chronic myelogenous leukemia progenitor cells. Blood 93:2707–20 Odai H, Sasaki K, Iwamatsu A, Nakamoto T, Ueno H, et al. 1997. Purification molecular cloning of SH2- and SH3-containing inositol polyphosphate-5-phophatase, which is involved in the signaling pathway of granulocyte-macrophage colonystimulating factor, erythropoietin, and Bcr-Abl. Blood 89:2745–56 Oda T, Heaney C, Hagopian JR, Okuda K, Griffin JD, Druker BJ. 1994. Crkl is the

24 Feb 2004

19:26

AR

AR210-IY22-09.tex

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

THE BCR-ABL STORY

328.

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

329.

330.

331.

332.

333.

334.

335.

336.

major tyrosine-phosphorylated protein in neutrophils from patients with chronic myelogenous leukemia. J. Biol. Chem. 269:22925–28 ten Hoeve J, Arlinghaus RB, Guo JQ, Heisterkamp N, Groffen J. 1994. Tyrosine phosphorylation of CRKL in Philadelphia+ leukemia. Blood 84:1731– 36 Deininger MW, Vieira SA, Parada Y, Banerji L, Lam EW, et al. 2001. Direct relation between BCR-ABL tyrosine kinase activity and cyclin D2 expression in lymphoblasts. Cancer Res. 61:8005–13 Wilson-Rawls J, Xie S, Liu J, Laneuville P, Arlinghaus RB. 1996. P210 Bcr-Abl interacts with the interleukin 3 receptor beta(c) subunit and constitutively induces its tyrosine phosphorylation. Cancer Res. 56:3426–30 Salgia R, Uemura N, Okuda K, Li JL, Pisick E, et al. 1995. CRKL links p210BCR/ABL with paxillin in chronic myelogenous leukemia cells. J. Biol. Chem. 270:29145–50 Salgia R, Li J-L, Lo SH, Brunkhorst B, Kansas GS, et al. 1995. Molecular cloning of human paxillin, a focal adhesion protein phosphorylated by P210+BCR/ABL. J. Biol. Chem. 270:5039–47 Coutinho S, Jahn T, Lewitzky M, Feller S, Hutzler P, et al. 2000. Characterization of Grb4, an adapter protein interacting with Bcr-Abl. Blood 96:618–24 Weng Z, Fluckiger AC, Nisitani S, Wahl MI, Le LQ, et al. 1998. A DNA damage and stress inducible G protein-coupled receptor blocks cells in G2/M. Proc. Natl. Acad. Arts Sci. 95:12334–39 Stewart MJ, Litz-Jackson S, Burgess GS, Williamson EA, Leibowitz DS, Boswell HS. 1995. Role for E2F1 in p210 BCR-ABL downstream regulation of cmyc transcription initiation. Studies in murine myeloid cells. Leukemia 9:1499– 507 Bruecher-Encke B, Griffin JD, Neel BG, Lorenz U. 2001. Role of the tyrosine phos-

337.

338.

339.

340.

341.

342.

343.

344.

P1: IKH

305

phatase SHP-1 in K562 cell differentiation. Leukemia 15:1424–32 Andoniou CE, Thien CBF, Langdon WY. 1996. The two major sites of cbl tyrosine phosphorylation in abl-transformed cells select the crkL SH2 domain. Oncogene 12:1981–89 de Jong R, ten Hoeve J, Heisterkamp N, Groffen J. 1995. Crkl is complexed with tyrosine-phosphorylated Cbl in Ph-positive leukemia. J. Biol. Chem. 270:21468–71 Danhauser-Reidl S, Warmuth M, Druker BJ, Emmerich B, Hallek M. 1996. Activation of Src kinases p53/56lyn and p59hck by p210bcr/abl in myeloid cells. Cancer Res. 56:3589–96 Afar DEH, Han L, McLaughlin J, Wong S, Dhaka A, et al. 1997. Regulation of the oncogenic activity of BCR-ABL by a tightly bound substrate protein RIN1. Immunity 6:773–82 LaMontagne KR Jr, Flint AJ Jr, Franza BR, Pendergast AM, Tonks1 NK. 1998. Protein tyrosine phosphatase 1B antagonizes signalling by oncoprotein tyrosine kinase p210 bcr-abl in vivo. Mol. Cell. Biol. 18:2965–75 Klejman A, Schreiner SJ, NieborowskaSkorska M, Slupianek A, Wilson M, et al. 2002. The Src family kinase Hck couples BCR/ABL to STAT5 activation in myeloid leukemia cells. EMBO J. 21:5766–74 Skorski T, Wlodarski P, Daheron L, Salomoni P, Nieborowska-Skorska M, et al. 1998. BCR/ABL-mediated leukemogenesis requires the activity of the small GTPbinding protein Rac. Proc. Natl. Acad. Arts Sci. 95:11858–62 Mayerhofer M, Valent P, Sperr WR, Griffin JD, Sillaber C. 2002. BCR/ABL induces expression of vascular endothelial growth factor and its transcriptional activator, hypoxia inducible factor-1alpha, through a pathway involving phosphoinositide 3-kinase and the mammalian target of rapamycin. Blood 100:3767–75

24 Feb 2004

19:26

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

306

AR

WONG

AR210-IY22-09.tex

¥

AR210-IY22-09.sgm

LaTeX2e(2002/01/18)

P1: IKH

WITTE

345. Slupianek A, Schmutte C, Tombline G, Nieborowska-Skorska M, Hoser G, et al. 2001. BCR/ABL regulates mammalian recA homologs, resulting in drug resistance. Mol. Cell 8:795–806 346. Ernst TJ, Slattery KE, Griffin JD. 1994. p210Bcr/Abl p160v-Abl induce an increase in the tyrosine phosphorylation of p93c-Fes. J. Biol. Chem. 269:5764– 69 347. Maru Y, Peters KL, Afar DE, Shibuya M, Witte ON, Smithgall TE. 1995. Tyrosine phosphorylation of BCR by FPS/FES protein-tyrosine kinases induces association of BCR with GRB-2/SOS. Mol. Cell. Biol. 15:835–42 348. Vieira SA, Deininger MW, Sorour A, Sinclair P, Foroni L, et al. 2001. Transcription factor BACH2 is transcriptionally regulated by the BCR/ABL oncogene. Genes Chromosomes Cancer 32:353–63 349. Shi CS, Tuscano JM, Witte ON, Kehrl JH. 1999. GCKR links the Bcr-Abl oncogene and Ras to the stress-activated protein kinase pathway. Blood 93:1338–45 350. Schuster C, Forster K, Dierks H, Elsasser A, Behre G, et al. 2003. The effects of BcrAbl on C/EBP transcription-factor regulation and neutrophilic differentiation are reversed by the Abl kinase inhibitor imatinib mesylate. Blood 101:655–63 351. Heriche JK, Chambaz EM. 1998. Protein kinase CK2alpha is a target for the Abl and Bcr-Abl tyrosine kinases. Oncogene 17:13–18 352. Amarante-Mendes GP, McGahon AJ, Nishioka WK, Afar DE, Witte ON, Green DR. 1998. Bcl-2-independent Bcr-Ablmediated resistance to apoptosis: protection is correlated with up regulation of Bcl-xL. Oncogene 16:1383–90

353. Jamieson L, Carpenter L, Biden TJ, Fields AP. 1999. Protein kinase Ciota activity is necessary for Bcr-Abl-mediated resistance to drug-induced apoptosis. J. Biol. Chem. 274:3927–30 354. Sanchez-Garcia I, Grutz G. 1995. Tumorigenic activity of the BCR-ABL oncogenes is mediated by BCL2. Proc. Natl. Acad. Arts Sci. 92:5287–91 355. Cambier N, Chopra R, Strasser A, Metcalf D, Elefanty AG. 1998. BCR-ABL activates pathways mediating cytokine independence protection against apoptosis in murine hematopoietic cells in a dosedependent manner. Oncogene 16:335–48 356. Xie SH, Wang Y, Liu JX, Sun T, Wilson MB, et al. 2001. Involvement of Jak2 tyrosine phosphorylation in Bcr-Abl transformation. Oncogene 20:6188–95 357. Salomoni P, Condorelli F, Sweeney SM, Calabretta B. 2000. Versatility of BCR/ABL-expressing leukemic cells in circumventing proapoptotic BAD effects. Blood 96:676–84 358. Matsuguchi T, Inhorn RC, Carlesso N, Xu G, Druker B, Griffin JD. 1995. Tyrosine phosphorylation of p95Vav in myeloid cells is regulated by GM-CSF, IL-3 and steel factor and is constitutively increased by p210BCR/ABL. EMBO J. 14:257–65 359. Ghaffari S, Jagani Z, Kitidis C, Lodish HF, Khosravi-Far R. 2003. Cytokines and BCR-ABL mediate suppression of TRAIL-induced apoptosis through inhibition of forkhead FOXO3a transcription factor. Proc. Natl. Acad. Sci. USA 100:6523–28 360. Trotta R, Vignudelli T, Candini O, Intine RV, Pecorari L, et al. 2003. BCR/ABL activates mdm2 mRNA translation via the La antigen. Cancer Cell 3:145–60

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Wong.qxd 3/12/2004 1:56 PM

See legend on next page

Page 1

THE BCR-ABL STORY C-1

Wong.qxd

3/12/2004

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

WONG

1:56 PM

■

Page 2

WITTE

Figure 6 Structures of active and inactive tyrosine kinases complexed to inhibitors Imatinib mesylate and PD 166326 (reprinted from (101, 264)). (A) Structure of the murine Abl kinase domain complexed to a myristoylated peptide and Imatinib mesylate solved at a resolution of 1.75 Å. Imatinib mesylate binds to Abl with the activation loop pointed toward the protein indicative of an inactive Abl tyrosine kinase conformation (101). Abl numbering based on isoform 1b. (B) Structure of the murine Abl kinase domain bound to myristic acid and the inhibitor PD 166326 solved at a resolution of 1.8 Å. Unlike Imatinib mesylate, PD 166326 binds to the Abl kinase domain with the activation loop orientated outward of the protein indicative of an active tyrosine kinase conformation (101). Abl numbering based on isoform 1b. (C) Conformational changes in the activation loop upon activation of protein kinases. Left: three tyrosine kinases ABL, Insulin receptor kinase (IRK), and hematopoietic cell specific kinase (HCK) in inactive states showing distinct conformations of the activation loop (black). Right: the crystal structure of LCK, which illustrates the conformation of the activation loop that all kinases bear upon activation by phosphorylation (264). (D) Schematic diagram of interactions between Imatinib mesylate and the ABL kinase domain. Inhibitor carbon atoms are colored green and magenta and ABL carbon atoms are colored brown. Interacting nitrogen atoms are colored blue, oxygen atoms red, chlorine atoms green, and sulfur atoms yellow. Hydrogen bonds between inhibitor and amino acids are indicated by dotted lines along with their distances, and residues making Van Der Waal interactions are surrounded by green dots. A total of 6 H-bonds and 15 Van Der Waal interactions occur between Imatinib mesylate and the ABL tyrosine kinase domain (264). ABL numbering based on isoform 1b.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:247-306. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

17:30

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104533

Annu. Rev. Immunol. 2004. 22:307–28 doi: 10.1146/annurev.immunol.22.012703.104533 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on October 15, 2003

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY Sergio A. Quezada,∗ Lamis Z. Jarvinen,∗ Evan F. Lind, and Randolph J. Noelle Department of Microbiology and Immunology, Dartmouth Medical School, Lebanon, New Hampshire 03756; email: [email protected]

Key Words immature dendritic cell, transplantation tolerance, suppressor, donor transfusion, cell-mediated immunity ■ Abstract Development of the acquired immune response is dependent on the signaling of CD40 by its ligand, CD154. These molecules govern both the magnitude and quality of humoral- and cell-mediated immunity. A litany of studies have conclusively documented that blockade of this ligand-receptor pair can prevent, and also intervene in, the progression of antibody- and cell-mediated autoimmune diseases, and can instill long-lived allogeneic and xenogeneic graft tolerance. Many effector mechanisms of inflammation are abolished as a result of CD154 blockade, but we are now beginning to understand that CD154 blockade may, in some instances, engender long-lived, antigenspecific tolerance. In the context of transplantation tolerance, we present a hypothesis that αCD154 blockade is most effective at inducing long-lived allospecific tolerance if anergy and regulation can be elicited prior to the onslaught of inflammation that is induced by grafting (preemptive tolerance). This facet of αCD154-induced tolerance appears to co-opt the normal processes of peripheral tolerance induced by immature DCs and can be exploited to induce long-lived antigen-specific tolerance. The underlying science and the prospects for inducing long-lived antigen-specific tolerance in a model of allograft tolerance through CD154 blockade are presented and discussed.

INTRODUCTION The pivotal role played by CD40 in governing humoral and cell-mediated immunity is manifested by the capacity of this ligand-receptor pair to activate B cells and dendritic cells (DCs), respectively. With regard to the former, engagement of CD40 in the presence of cytokines can induce profound clonal expansion and differentiation to Ig secretion, and at the same time enhance the antigen-presenting cell (APC) capacity of the B cell. With regard to the latter, CD40 activation of DCs can, together with signals via toll-like receptors (TLRs), manifest the licensing ∗

These authors contributed equally to this paper.

0732-0582/04/0423-0307$14.00

307

11 Feb 2004

17:30

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

308

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL.

of DCs to engender heightened proficiency in antigen presentation, cytokine and chemokine production, and extended survival. The maturational impact on both the B cell and the DC induces the upregulation of a plethora of costimulatory molecules that can reciprocally trigger the cognate T cells to clonally expand and differentiate. Thus, indirectly, engagement of CD40 on B cells and DCs exerts a profound impact on the development of T cell immunity. Likewise, interruption of CD40/CD154 interactions can result in inadequate antigen presentation, a deficiency in the inflammatory environment, partial T cell responses, and in some cases, T cell tolerance. Understanding why CD154 blockade results in T cell tolerance is the focus of this review.

CD40 AND ITS LIGAND CD40 is a 48 kDa transmembrane glycoprotein cell surface receptor that shares sequence homology with the tumor necrosis factor α (TNF-α) receptor family and was initially identified as a B-cell surface molecule that induced B-cell growth upon ligation with monoclonal antibodies. DCs, macrophages, epithelial cells, hematopoietic progenitors (1), and activated T cells (2) have been shown to express CD40. The putative expression and function of CD40 on T cells and its possible function in regulating T cell tolerance and immunity is particularly pertinent and will be discussed below. Its ligand, CD154, is a 34–39 kDa type II integral membrane protein expressed on activated but not resting T cells (3), activated B cells (4), and activated platelets (5, 6). During inflammatory responses, other cell types such as peripheral blood monocytes (7, 8), human vascular endothelial cells, smooth muscle cells, and mononuclear phagocytes (9) have all been shown to express CD154. Engagement of CD40 leads to B cell clonal expansion, germinal center formation, isotype switching, affinity maturation, and generation of longlived plasma cells (10). As is discussed later in this review, within the cytoplasmic tail of CD40, multiple functional domains have been identified that contribute independently or jointly to the vast array of biological responses. An understanding of the contribution of each of these functional domains to the role CD40 plays in immunity is emerging, but not without controversy (11, 12). Blocking of the CD154 ligand, and thus CD40/CD154 ligation, is an effective means by which to abrogate autoimmune diseases (as reviewed in Reference 13) and induce transplantation tolerance (as reviewed in References 14, 15). Innumerable autoimmune models of disease in mice have been shown to be blocked by treatment with αCD154, including experimental autoimmune encephalomyelitis, diabetes, collagen-induced arthritis, uveitis, thyroiditis, lupus, and others, as reviewed in Reference 16. Blocking CD154 in autoimmune disease models such as EAE (13, 17, 18) and lupus (19–21) can be used as both a preventative means by which to block disease and a therapeutic means to treat the progression of disease. Aside from controlling autoimmune diseases, the prevention of transplant rejection by blocking CD40/CD154 interactions has also been repeatedly documented for the induction of long-term tolerance to skin (15, 22–25), islets (26), bone marrow

11 Feb 2004

17:30

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

CD40 AND PERIPHERAL TOLERANCE

P1: IKH

309

(27), and a myriad of other transplanted organs (28, 29). All of these findings on the blockade of CD40/CD154 interaction for the prevention of disease and transplant rejection demonstrate yet once again the importance of understanding the mechanisms of this receptor-ligand pair.

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD40 SIGNALING PATHWAYS THAT CONTROL APC FUNCTION Stimulation through CD40 results in a complex series of events within an APC. Engagement of CD40 by its ligand, CD154, leads to trimeric clustering of CD40 and the recruitment of adaptor proteins known as TNF receptor–associated factors (TRAFs) to the cytoplasmic tail (30). Binding of the TRAFs results in formation of a signaling complex that includes multiple kinases such as NF-κB inducing kinase (NIK), receptor interacting protein (RIP), members of the mitogen-activated protein kinase (MAPK) family and possibly others. Clustering of these kinases then initiates a downstream cascade of signaling events, resulting in activation of the MAPK and NF-κB pathways and finally transcription of target genes, leading to physiological effects such as production of inflammatory mediators, DC survival (31), and prolonged MHC/antigen complex presentation (32). The first level at which the signal through CD40 is modulated in the APC is through differential effects of binding of the TRAF molecules. Our lab (12) and others (33) have demonstrated that selective mutations to known TRAF binding sites on the cytoplasmic tail of CD40 result in differential signaling and phenotypic outcomes. In B cells, ablation of the TRAF 2,3 binding site results in blockade of p38 and JNK phosphorylation, and a decrease in the phosphorylation of IκBα (34). Despite these signaling defects, mice containing this mutation still have normal early immunoglobulin production. Mutation of the TRAF 6 binding site results in normal phosphorylation of IκBα, p38, and JNK, but deficient affinity maturation (12). Similar studies confirm the requirement of TRAF 2,3 binding in activation of signaling pathways but also implicate TRAF 2,3 binding as playing a role in class switching (11). In DCs, on the other hand, TRAF 6 binding to CD40 is required for activation of p38, JNK, and the production of the proinflammatory cytokine IL-12 p40 (33). The different functions of TRAFs in DCs versus B cells clearly show that CD40 signaling and the downstream effects are cell-type specific and that functional signaling diversity may be controlled at the level of TRAF recruitment in the APC. Activation of the NF-κB pathway via CD40 engagement on DCs has been shown to be critical for their maturation. NF-κB responsive genes include many inflammatory mediators as well as antiapoptotic factors central to DC maturation and longevity (35–38). The survival signal for the DC can be delivered either by CD40 or signaling through the TNF family member TRANCE (39). CD40 signaling has been shown to induce higher levels of the antiapoptotic protein Bcl-XL in mice (39) or Bcl-2 in human DCs (40). This effect has been shown to be NF-κB dependent (41). The NF-κB family consists of five known members

15 Mar 2004

15:18

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

310

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL.

in mammals, NF-κB1 (p50), NF-κB2 (p52), RelA (p65), c-Rel, and RelB. NF-κB1 and NF-κB2 are expressed as inactive precursor proteins (p105 and p100, respectively), which are held in the cytoplasm by their c-terminal ankyrin repeats (42). The broad NF-κB pathway can be further described as two separate pathways, referred to by Pomerantz and Baltimore as the “canonical” and “noncanonical” pathways (43). The canonical pathway refers to active NF-κB dimers consisting of mainly p50/RelA, and the noncanonical consisting of p52/RelB. The canonical pathway, and c-Rel in particular, has been shown to be responsible for immediate inflammatory responses from DCs such as production of IL-12 (44). Until recently, only one stimuli had been shown to induce processing and activation of NF-κB2—Lymphotoxin Beta signaling (45). Recently, however, several groups have shown that this processing occurs in B cells in response to BAFF binding its receptor (46, 47), as well as after CD40 ligation by CD154. The activation and function of the noncanonical, or NF-κB2, pathway has recently been described in human DCs as functioning to regulate several chemokines (48). Data from our lab indicates that the role of NF-κB2 activation via CD40 is critical for the prolonged survival of DCs after maturation with lipopolysaccharide (LPS) (49). We find that DCs from mice with defective NIK, a kinase critical in the activation of the NF-κB2 pathway, mature and migrate to the periarterial lymphoid sheath (PALS) in response to LPS stimulation, but rapidly undergo apoptotic death and cannot be rescued by stimulation through CD40 (see Figure 1, for a summary of these pathways in DCs). Specifically targeting the NF-κB2 pathway at the level of NIK would then allow one to limit antigen presentation during the duration of an inflammatory response, such as autoimmune disease or transplant engraftment, thus allowing the induction of tolerance after the secession of inflammation. The importance of the NF-κB2 signaling cascade in DC maturation is also underscored by the observation that RelB−/− DCs are incapable of eliciting productive immune responses and in fact are immunosuppressive by virtue of inducing regulatory cells (50). Therefore, understanding the complex biochemical signaling pathways that govern CD40-induced DC licensing will allow us to envision strategies to control immunity versus tolerance.

THE ROLE OF CD40 IN LICENSING DCs TO ELICIT IMMUNITY The two-signal hypothesis of T cell activation is an accepted paradigm that describes the early requirements for T cell activation versus anergy. This paradigm states that the proficient induction of an immune response requires TCR and MHC/peptide interaction (signal one) followed by the interaction between costimulatory molecules, namely CD80/86 and CD28 (signal two). If signal one is generated in absence of signal two, then the outcome is not immunity but tolerance. Providing signal one and signal two becomes the responsibility of the APC, which must efficiently engage and trigger multiple T cell surface molecules. To achieve

11 Feb 2004

17:30

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD40 AND PERIPHERAL TOLERANCE

P1: IKH

311

this proficiency, we have learned that APCs must “mature.” This maturation of APCs can be induced by a wide spectrum of factors including TNFα, CD154, cytokines, and a spectrum of TLR agonists. Therefore, decision of tolerance versus immunity is centered, at least in part, on the maturational status of the APC compartment. Appropriate maturation of APCs will license them to trigger productive cell-mediated immunity (CMI). If one plane of control of T cell tolerance versus immunity is at the level of APC maturation, a system of checks and balances must exist to regulate APC licensing. Whereas initially we thought that a single signal can adequately license APCs to trigger CMI responses, we are now beginning to realize that multiple signals are required for APC licensing. Although many studies initially showed that CD40 ligation of DCs alone was optimal to induce maturation (51, 52), far greater adjuvant activities will likely be observed when using coactivators like TLR agonists. A variety of single agents like LPS, TNFα, and CD154 have been shown to be DC activators (37, 53). However, recent in vivo studies have underscored the necessity of multiple signals to impinge on the DC compartment and to achieve true in vivo licensing for triggering CMI. Recent studies, as well as our own unpublished data, show that CD40 engagement alone is insufficient to induce IL-12 p70 production by DCs in vitro and in vivo. By evaluating mRNA for p40 and p35, the authors showed that coengagement via TLR (STAg, an extract from Toxoplasma gondi) and CD40 is critical for enhanced p35 mRNA expression and the production of IL-12 p70 (54). This study was followed by an investigation using human DCs, where it was shown that CpG DNA was a critical costimulus with CD40 signaling for IL-12 p70 production in vitro (55). Therefore, the study of CD40 signaling in DCs warrants considering the synergistic action of TLR triggering. These recent studies on TLR synergy bring into question work done by a large number of investigators, including work from our own lab that concludes that CD40 signaling alone can license DCs. These studies were likely compromised by a number of factors, including the use of supraphysiological levels of CD40 agonists, that the DCs were matured or grown in culture prior to activation with CD154 and/or αCD40, that the isolation procedure “activated” the DCs, or that TLR agonists were present (endotoxin) in the CD40 agonist preparations. Examples of this can be found in seminal papers by Bennet and coworkers (56) and Schoenberger and collaborators (57). With regard to the former, in order to induce CTL priming in absence of CD40 help, mice were immunized with antigen-loaded irradiated spleen cells and treated with high doses of agonistic αCD40 (four doses of 100 µg each). Besides the high levels of αCD40, the use of irradiated APCs probably allowed the release of proinflammatory mediators with the capacity of engaging TLR signaling, therefore licensing DC function and T cell priming (58–60). In fact, some studies have shown that, in isolation, CD40 agonists can impair CMI to tumors (61) and selfantigens (62). This once again argues for the necessity of supplementary signals in addition to CD40 stimulation in order to overcome tolerance and induce immunity. The new emerging paradigm is that combined TLR and CD40 triggering is critical for APC licensing. Pattern-recognition molecules (like TLRs) can recognize

11 Feb 2004

17:30

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

312

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL.

motifs found in bacterial glycolipids, DNA, and even viral RNA. The detailed mechanism of action and signaling of each one of these receptors has been recently summarized in reviews from the groups of Janeway and Medzhitov at Yale (63–66). Although they have been initially described as part of the innate component of immunity, they clearly are important cosignals in controlling the quality and magnitude of the ensuing acquired immune response. The importance of the synergy between the tumor necrosis factor receptor and TLR systems in licensing DCs to elicit CMI is underscored by a number of published studies (67–72), as well as collaborative studies from the laboratories of Ross Kedl (3M Pharmaceuticals in Minnesota) and our laboratory. These latter studies show that use of agonist αCD40 mAb together with a TLR agonist enhances the frequency of peptide-reactive T cells by more than tenfold. Thus, the administration of αCD40, TLR agonist, and ovalbumin (OVA) results in 10%– 25% of the CD8+ T cells to be antigen-specific in six days. Concordant with an increase in frequency, this synergistic effect was capable of profoundly activating antigen-specific CD8+ CTL responses as assessed by IFN-γ production and CTL activity. What is remarkable is that the magnitude of these responses is reminiscent of those observed to acute viral infections with lymphocytic choriomeningitis virus (LCMV) or vesicular stomatitis virus (VSV) (73, 74). The importance of combined TLR and CD40 signaling becomes self-evident when one histologically evaluates the distinctive impact of CD40 signaling alone versus CD40 and TLR on DC migration and survival. Previous studies by Moser and coworkers demonstrated that TLR agonists (like LPS) induce the rapid migration of the marginal zone DCs into the PALS (75). Once localized to the PALS, the DCs have a short half life unless provided a CD154 signal via antigen-specific helper T cells. Thus, the combined effects of TLR and agonistic αCD40 reposition DCs into the T cell zone for cognate interaction with T cells and provide a CD40dependent survival signal for long-lived presentation of antigen. The importance of this survival signal is underscored by the observation that loss of CD40-induced NIK function in the DC compartment (see CD40 Signaling Pathways That Control APC Function) impairs DC survival in the PALS and limits their capacity to support inflammatory responses. Finally, a number of studies have highlighted the importance of CD40 signaling in the context of DC longevity, as well as longevity of expression of class II MHC–peptide complex on the antigen-presenting DCs (32, 76). Thus, in addition to chemokine and cytokine production, longevity and anatomic location are critical players in the process of DC licensing. Although most studies that have employed CD40 agonists have assumed that they license the APC compartment, there may be alternative explanations to the immunogenic impact of these reagents. Bourgeois and colleagues have been able to demonstrate that the generation of memory CD8+ T cells can occur in the absence of CD40 expression by the APC but requires the help of CD4+ T cells (2). When they utilized APCs from a CD40 knockout mouse, they were able to show that CD4-CD8 T cell collaboration required the engagement of an APC but that activation of the CD8+ T cell was independent of CD40 expression by the APC.

11 Feb 2004

17:30

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD40 AND PERIPHERAL TOLERANCE

P1: IKH

313

The function of CD40 expressed on CD8+ T cells in regulating T cell expansion and differentiation in other systems awaits confirmation. Finally, it cannot go unnoticed that T cell function can be substantially influenced through the expression of CD40 on nonhematopoietic cells. For example, it is clear that CD40 plays a critical role in activating endothelial cells leading to the migration of leukocytes (77). CD40 is expressed on vascular endothelium, and its expression is modulated by proinflammatory cytokines (TNF-α), IL-1, IFN-γ , and LPS (78, 79). Ligation of CD40 on endothelial cells leads to the production of proinflammatory cytokines and enhanced expression of CD54 (intracellular adhesion molecule-1), E-selectin, and vascular cell adhesion molecule 1, which results in increased leukocyte binding (79). Therefore, studies that have employed the use of CD40 agonists to manipulate T cells responses have to be cautiously interpreted because the impact of the CD40 agonist could be exerted directly on DCs, T cells, or nonhematopoietic cells.

Denying DC Licensing: Peripheral Self-Tolerance Denying immune responses to self is also the responsibility of the peripheral APC compartment. We know that despite the central deletion of self-reactive T cells and B cells, autoreactive T and B cells manage to escape negative selection, find their way to the periphery, and present a constant threat to the generation of autoimmunity (80). Consequently, mechanisms to prevent the activation of mature, self-reactive lymphocytes are necessary. Among those mechanisms are the following: ignorance, deletion, anergy induction, and active suppression by endogenous CD4+CD25+ suppressor T cells. Ignorance presumes that antigen in the periphery is not available, and unless tissue damage or other type of stress allows its release to the periphery, autoimmunity is not elicited. In contrast, anergy is based on the premise that self-antigen is presented to the T cell compartment, but T cell recognition of self-antigen leads to unresponsiveness. In this case, APCs present self-antigen in absence of signal two (costimulation), leading to a state of hyporesponsiveness by the autoreactive T cell (81, 82). This process has been thought to be more efficient than simple deletion of autoreactive cells in the periphery because anergic autoreactive T cells will compete for the antigen and reduce the chances of na¨ıve autoreactive T cells triggering autoimmunity. In contrast, accumulating evidence suggests that upon presentation of antigen by DCs in absence of costimulation, T cells do indeed proliferate and then disappear (reviewed in Reference 83). In recent years, peripheral suppression has reemerged as a major regulatory phenomenon that controls mature, autoreactive effector T cells. Extensive insights into a specific population of regulatory cells (CD4+CD25+Foxp3+ subset) has been afforded by S. Sakaguchi’s group (84–87) and many others (88–91). These cells represent 5%–10% of the endogenous CD4+ T cells subset and are able to suppress CD4+ and CD8+ T cell responses in vitro and in vivo upon TCR ligation (92, 93). As is extensively reviewed (94), elimination of this subset or shutting down their

11 Feb 2004

17:30

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

314

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL.

regulatory function using agonistic αGITR mAbs leads to autoimmunity (87, 95, 96), proving that these cells play a critical function in controlling the immune response to self. Another group of regulatory cells recently described (97) are referred to as Type 1 T regulatory cells (Tr1). These cells have poor proliferative capacity and act mainly by secreting IL-10 and transforming growth factor β (TGFβ) (98). Although there is not yet a clear consensus, one difference between Tr1 and CD4+CD25+ suppressor cells is that the latter are a constitutive component of the immune system, whereas Tr1 seem to be induced in response to Ag presentation in presence of IL-10 and IFNα (99). The differences and similarities between these two subsets is clearly presented in a recent review by Bluestone & Abbas (100).

IMMATURE DCS AS MEDIATORS OF PERIPHERAL SELF-TOLERANCE The mechanisms for controlling immune responses to self in the periphery (anergy, regulatory T cell) have converged on the role of the immature DC in managing tolerance. A body of evidence has emerged to show that persistent presentation of self-antigens by immature DCs maintains peripheral self-tolerance (101–104). Supporting this is a recent article by Wilson et al., wherein they elegantly demonstrate that, in a steady state, DCs from most lymphoid organs are phenotypically and functionally immature (105). A body of evidence has also been generated regarding the capacity of immature DCs to sustain peripheral tolerance. In particular, studies show how immature DCs are capable of consuming cell-associated antigens and inducing T cell anergy or deletion as long as they are not induced to mature via CD40 ligation (101, 102, 104, 106). With regard to the capacity of immature DCs to elicit expansion of regulatory subsets, less evidence is available. Tr1 cells have been shown to be generated in vitro when antigen is presented in presence of IL-10 (97, 99). In these cases, the necessity for immature DCs in order to induce Tr1 cells is not clear. Recent experiments have shown that CD40−/− and RelB−/− DCs (which would resemble the immature phenotype) are capable of inducing Tr1 CD4+ T cells in vivo, therefore making a case for the capacity of immature DCs to sustain the generation of regulatory T cells (50). With regard to the endogenous CD4+CD25+ suppressor T cells, it has been recently published that in vivo targeting of antigen to immature DCs via DEC205 leads to the expansion of CD4+CD25+CTLA4+ suppressor T cells (107). Therefore, a link between CD40 signaling of immature DCs and the activities of regulatory T cells can be envisioned. Form and biochemical makeup of the self-antigens may be critical to the outcome of the events programmed by immature DCs. Several groups have shown that the uptake of apoptotic cells by DCs is an effective inducer of T cell tolerance (108–110), whereas the phagocytosis of necrotic cells has been proven to induce maturation of DCs and development of immunity (111). The difference resides in the proinflammatory components released by necrotic cells (like heat shock

11 Feb 2004

17:30

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD40 AND PERIPHERAL TOLERANCE

P1: IKH

315

proteins), which result in induction of the innate immune response (60); meanwhile, clearance of apoptotic cells by specific cell-surface receptors is accompanied by secretion of such anti-inflammatory cytokines as TGF-β (112, 113). In addition, new reports underscore the capacity of necrotic or stressed apoptotic cells in the induction of DC maturation as evidence for enhanced expression of CD40 and costimulatory molecules, together with heightened IL-12 secretion (114, 115). Taken together, it is apparent that the boundary line drawn for the induction of peripheral self-tolerance and immunity is a narrow one that is easily crossed by the accumulation of proinflammatory signals like CD40. Recently, studies have shown that CD40 agonist can cause the immune system to cross this line, and as a result, autoreactive responses can evolve (116). The vigilant presentation of self-antigens by immature DCs to maintain selftolerance is likely the reason why fully licensed DCs do not routinely “break” peripheral tolerance. Fully licensed DCs that are presenting peptides derived from pathogens no doubt simultaneously present self-peptides as well. We would contend that the preemptive and continuous induction of anergy to self-peptides in the absence of overt inflammation prohibits the subsequent induction of immune responses to self when licensed DCs are concomitantly presenting foreign and self-antigens.

CD154 AND CD40 IN REGULATING PERIPHERAL TOLERANCE Owing to the profound impact on DC maturation, CD40/CD154 interactions are decisive in the regulation of peripheral self-tolerance. It has been repeatedly demonstrated that induction of anergy by immature DCs can be overridden by the administration of an agonistic αCD40 antibody (117, 118). In most of these studies, it was shown that the delivery of a peptide (sometimes in adjuvant or via peptide-pulsed apoptotic cells) induces CD8+ T cell tolerance. The coadministration of αCD40 resulted in productive CD8+ T cell responses to the administered peptide. In one case, the use of a diabetogenic peptide (119) in conjunction with αCD40 resulted in diabetes. None of these reports have demonstrated the seminal changes in DC biology that are critical for making the transition from tolerance to immunity—no doubt they are many. However, it is of interest that recent studies have shown that DC persistence and antigen persistence is greatly enhanced as a result of CD40 agonists (37). Some insights into the biochemical and biological pathways that govern selftolerance are emerging. Particularly pertinent in this context is the data generated by Martin et al. (50). In their work they showed that RelB KO DCs or CD40 KO DCs loaded ex-vivo with Ag are highly tolerogenic and, most interesting, were able to impede ongoing immune responses. These studies suggest that the NF-κB2 pathway may be particularly important in triggering DC licensing. Similar studies are underway in our laboratory using DCs from mice in which specific CD40

11 Feb 2004

17:30

316

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL.

cytoplasmic functional domains have been mutated by site-directed mutagenesis. In the coming years, we will no doubt be able to deduce the biochemical pathways that are critical for inducing DC licensing.

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

WHEN DOES CD154 BLOCKADE INDUCE T CELL TOLERANCE: STUDIES IN ALLOGRAFT TOLERANCE If immature DCs are effective at silencing the self-reactive T cells, exploiting this function to control the response to allopeptides should allow enhanced graft survival. However, as we see from the many reports presented above, crossing the line between tolerance and immunity is, at times, unpredictable. Form of antigen, route of antigen delivery, the inflammatory environment, and the alloreactive pool size all circumvent the rather tenuous boundaries between tolerance and immunity. The discovery of costimulation provided a target to prophylactically intervene on behalf of the host and to instill a state of peripheral allospecific tolerance. Because CD154-CD40 and other costimulatory ligand-receptor pairs were believed critical for DC licensing, numerous studies evaluated the success of costimulation blockade in delaying graft rejection.

Altering the Alloreactive T Cell Response BLOCKING COSTIMULATORY PATHWAYS One of the first pathways to be targeted in an attempt to modify allograft rejection was the CD28/B7 pathway. It was demonstrated that by using a CTLA4-Ig (a soluble form of the high-affinity receptor to B7 molecules), long-term allograft survival of islet cells (120, 121) and cardiac allografts (122) could be achieved. By competitively binding to B7 molecules, CTLA4-Ig blocked CD28 signaling to the T cells, therefore limiting activation of the alloreactive T cell compartment. However, later studies utilizing CTLA4-Ig treatment in nonhuman primates were unable to induce long-lived tolerance with CTLA4-Ig alone and demonstrated that there was a requirement for coadministration of anti-CD154 mAb (123). In an attempt to analyze the immune events associated with tolerance, it was directly shown that a CD28-independent CTLA4 signal delivers a strong negative signal to CD4+ T cells and that by interrupting the CTLA4/B7 interaction, alloreactive T cells were not effectively shut down (124). This finding is supported by earlier results, which demonstrated that CTLA4 signaling was required for inactivating alloreactive T cells (125, 126). Other means by which to disrupt T cell activation have been through the blockade of the CD40/CD154 pathway. Originally, it was thought that αCD154 monotherapy would be effective because it limited APC maturation and downmodulated the B7-CD28 interaction, resulting in the lack of a signal 2 and T cell anergy (127, 128). However, αCD154 as a monotherapy has not been successful in inducing allograft tolerance (129, 130). Studies by Waldmann indicate that failure to induce tolerance using αCD154 therapy is insufficient because of its inability

11 Feb 2004

17:30

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD40 AND PERIPHERAL TOLERANCE

P1: IKH

317

to block rejection elicited by CD8+ T cells (131). Our own evidence suggests that αCD154 as a monotherapy is also ineffective at controlling the allogeneic response of CD4+ effector T cells in graft rejection (22). It is clear that blockade of any given pathway is less effective than blocking multiple pathways (i.e., CD40 and CD28 pathways), as has been demonstrated repeatedly (123, 131, 132). Highly immunogenic allografts, such as those of heart and skin, require stronger tolerogenic therapies. Among these, the combination of costimulatory blockade with immunosuppressive drugs (Rapamycin with αCD154 mAb and CTLA4-Ig) has been shown to result in long-term graft acceptance (133). Interestingly, the effect of cyclosporine A impedes the tolerogenic effects of αCD154 mAb and CTLA4-Ig. The explanation for this resides in the requirement of signal 1 to induce tolerance. Cyclosporine A (a calcineurin antagonist) inhibits signal 1 on the T cell, and together with signal 2 blockade results in the failure to induce apoptosis of the alloreactive T cells and thereby ablates tolerance induction (134, 135). The effectiveness of Rapamycin in combination with costimulation blockade to induce tolerance also appears to be dependent on the induction of apoptosis of the alloreactive T cells, as has been shown by Li and collaborators (136). The induction of apoptosis on the alloreactive T cell compartment seems to be an essential component of tolerance, as this contributes mostly to the reduction of the alloreactive clonal size (137). Altogether, enhanced allograft survival can be achieved by targeting independent costimulatory pathways, but greater success is obtained when several pathways are targeted simultaneously, and in some cases, when alloreactive T cells are sufficiently eliminated. Nonetheless, costimulation blockade, in the context of the inflammatory environment resulting from surgery and healing of a highly immunogenic graft is not an ideal context in which to ask immunotherapeutics to induce a state of immunologic tolerance. The immune system is most effective at inducing peripheral tolerance when the environment is quiescent, as we have seen in situations of peripheral self-tolerance. Based on this belief, tolerance may be achieved by exposing the recipient to donor antigens concomitantly with costimulation blockade prior to transplantation, in an effort to facilitate preemptive tolerance induction of the alloreactive T cell compartment. The next section discusses this type of approach, its relevance and effectiveness in experimental systems, and its possible application to clinics. DONOR-SPECIFIC TRANSFUSION We contend that the practice of donor-specific transfusion (DST) exploits the traditional practices of peripheral tolerance to modify the host immune response to alloantigens prior to the immunogenic introduction of an allograft. It is under these conditions that profound, enduring graft tolerance can be established. Historically, the infusion of whole blood from the donor into graft recipients modestly prolonged the longevity of grafts in humans (138) and mice (139). The enhanced graft survival following DST can be substantially lengthened or, in fact, rendered permanent if the DST is combined with blocking of CD154 (140). Recent studies by Xu and colleagues (141) and unpublished data

11 Feb 2004

17:30

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

318

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL.

from X. Zheng, Y. Li, X. Li, Y. Tian, K. Kawamoto & T. Strom have demonstrated long-term survival of allogeneic skin and islet allografts in monkeys that are treated with αCD154, DST, and Rapamycin. It is evident that we are on the path to bringing long-lived allospecific tolerance to the clinic for the management of allografts. Studies into this phenomenon have provided remarkable insights into the process of allospecific tolerance, peripheral tolerance, and the role of regulatory T cells in the survival of allografts. Long before DCs were recognized to be mediators of peripheral tolerance, Parker and colleagues championed the hypothesis that resting B cells (known for their inadequacies in antigen presentation) could be targeted with antigen to induce antigen-specific tolerance. In their studies (142), they demonstrated that a rabbit Fab αIgD could induce T cell tolerance to rabbit Ig epitopes. Although it was shown that targeting antigens to resting B cells in vivo was tolerogenic, the adoptive transfer of resting B cells carrying antigen or alloantigen was not reproducibly capable of inducing tolerance. At that time, it was reasoned that the transferred resting B cells carrying antigen or alloantigen were ineffective inducers of tolerance because upon entry into the host, they were activated by host CD154-bearing helper T cells. As a result, the transferred B cells were immunogenic, not tolerogenic. This hypothesis was supported by the observation that the adoptive transfer of allogeneic B cells and αCD154 induced profound allospecific unresponsiveness and allograft survival in grafted mice. Based on this observation, we and others hypothesized that the transfused allogeneic B cells were directly presenting alloantigen to the host in a tolerogenic fashion owing to the imposed blockade of CD40 signaling. More recent studies portray a very different picture of how the transfusion of allogeneic cells (DST) and αCD154 induce graft tolerance. Further testing of the proposed hypothesis demanded a critical evaluation of the role of direct or indirect presentation of DST to the host. To address this issue and also many of the remaining issues surrounding the fate of the alloreactive T cell compartment, a CD4+ TCR Tg T cell system was developed (TEa Tg T cells) (24). Briefly, CD4+ TCR Tg T cells that recognize an Eα peptide in the context of IAb were used as a population of alloreactive T cells that could be readily tracked. These cells are adoptively transferred into an H-2b Rag−/− mouse and upon grafting with an F1 graft (H-2bxd), graft rejection was followed over the next few weeks. Infusion of F1 B cells (DST) and αCD154 prolonged graft rejection, as shown in intact non-Tg systems. Among the most relevant findings were that indirect presentation of alloantigen was essential for tolerance induction. That is, the infusion of B cells that were virtually invisible to the TEa Tg T cells via direct recognition (i.e., Balb/c B cells bearing only Eα and not the restricting element, H-2b) induced profound tolerance. Thus, the infused DST was rapidly processed by host APCs and presented via indirect presentation to the TEa in a tolerogenic manner. The tolerogenic impact of DST and αCD154 in this system was to induce a rapid (day 3–4), systemic, abortive expansion of the alloreactive TEa Tg T cells that resulted in profound anergy. One additional unique feature of this system was that we could quantitate the magnitude of the T cell unresponsiveness

15 Mar 2004

15:19

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD40 AND PERIPHERAL TOLERANCE

P1: IKH

319

by purifying anergic cells. In these cases, the unresponsiveness was determined to be >95% on a per-cell basis. In striking contrast to the potent tolerogenic impact of alloantigen being delivered via DST, alloantigen present on an F1 skin graft exerted a profoundly immunogenic impact. Analysis of TEa Tg T cell responses to the F1 allograft (in the absence of DST and αCD154) revealed a delayed (6–8 days), regional [mostly draining lymph nodes (LNs)], productive T cell response with extensive expansion and cytokine production (IL-2 and IFN-γ ). Hence, infusion of DST and αCD154 resulted in a preemptive induction of tolerance within the alloreactive T cell compartment, thereby silencing the alloreactive response days prior to the time when the allograft had the opportunity to elicit an immune response. The finding that DST predominantly, if not exclusively, induces tolerance via indirect presentation provided the first clues that DST and αCD154 co-opted the fundamental mechanisms of peripheral tolerance to induce allotolerance. We now envision that the infused DST rapidly undergoes apoptosis and is presented by host APCs. It is of interest to note that αCD154 may facilitate the apoptosis of the DST by depriving it of a CD40 signal. At the same time, αCD154 impairs the maturation of host APCs, committing them to the tolerogenic presentation of DST-derived allopeptides. Delivery of peptides via apoptotic cells appears to be an extremely efficient means to induce peripheral tolerance (Figure 2). In analogous studies, Steinman and coworkers have shown that TAP−/− B cells that are hyperosmotically loaded with OVA can induce abortive expansion and anergy of OVA-specific CTLs in vivo via indirect presentation (143). Similarly, antigens expressed on dying pancreatic cells (144) induce tolerance via indirect presentation. Whether targeted via apoptotic cells, or by vectors that target directly to defined DC surface molecules (like DEC-205), antigens delivered to immature DCs induce profound antigen-specific tolerance, which is amplified in the presence of αCD154. INVOKING REGULATORY T CELLS Although the induction of clonal T cell anergy by peripheral tolerance mechanisms and by DST is near complete, these events are inadequate to achieve long-lived graft acceptance. Studies now suggest that clonal tolerance and graft permanence require the involvement of some regulatory activities (reviewed in Reference 145). Early studies by Waldmann and coworkers were the first to describe an infectious form of tolerance in a variety of allograft tolerance systems (146). Using DST and αCD154, these investigators showed that CD4+ T cells were critical for long-lived survival of the allograft (147). Subsequent studies in allogeneic bone marrow transplantation and other transplant models implicated an important role of CD4+CD25+ suppressor T cells in αCD154-induced graft tolerance (148, 149). Additional evidence substantiating the importance of regulatory T cells in longlived graft tolerance was provided by reconstitution studies from our laboratory. In these studies, we reconstituted RAG−/− mice with defined CD4+ T cell populations,

11 Feb 2004

17:30

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

320

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL.

treated them with DST and αCD154, and evaluated if graft tolerance could be induced following a skin allograft. Upon reconstitution with only CD4+CD25− T cells, DST and αCD154 significantly delayed the rejection of allogeneic skin, but the graft was ultimately rejected. In this case, the delay is due to extensive clonal abortion of the relevant alloreactive T cells; however, a small frequency remains that eventually rejected the graft. Upon the coadoptive transfer of CD4+CD25+ T cells with the CD25− T cells, permanent graft survival becomes evident. Hence, clonal abortion of the alloreactive CD4+ effectors is incomplete, as there appears to be a CD4+CD25− population that is resistant to tolerance induction. This residual population can be readily silenced by the cotransfer of regulatory T cells. One particularly intriguing observation in these studies was that the loss of CD154 expression on suppressor cells was sufficient to induce long-term skin graft survival with just the administration of DST (22), suggesting an important role of CD154 in regulatory T cell function.

FINAL COMMENTS Although licensed DCs elicit profound immunity to foreign peptides, the responses elicited to copresented self-peptides on the same DCs are silent. One must imagine that the silence to self-peptides afforded to a fully licensed DC is due to the constant presentation of self-peptides by the immature DC compartment and the preemptive silencing of the self-reactive T cell pool. Intervention into ongoing autoimmune responses solely with αCD154 can block many effector mechanisms and ablate the clinical and pathological features of autoimmunity but, in most cases, does not induce a state of tolerance. The delivery of allogeneic peptides via apoptotic cells (DST) to the quiescent immune system, and maintaining that quiescence with CD154 blockade, induces profound clonal abortion and elicits regulatory T cell function. It is under these circumstances that αCD154 instills long-lasting immune tolerance. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. van Kooten C, Banchereau J. 1997. Functions of CD40 on B cells, dendritic cells and other cells. Curr. Opin. Immunol. 9:330–37 2. Bourgeois C, Rocha B, Tanchot C. 2002. A role for CD40 expression on CD8+ T cells in the generation of CD8+ T cell memory. Science 297:2060–63 3. Klaus GG, Choi MS, Lam EW, JohnsonLeger C, Cliff J. 1997. CD40: a pivotal

receptor in the determination of life/death decisions in B lymphocytes. Int. Rev. Immunol. 15:5–31 4. Higuchi T, Aiba Y, Nomura T, Matsuda J, Mochida K, et al. 2002. Cutting edge: Ectopic expression of CD40 ligand on B cells induces lupus-like autoimmune disease. J. Immunol. 168:9–12 5. Henn V, Slupsky JR, Grafe M, Anagnostopoulos I, Forster R, et al. 1998. CD40

11 Feb 2004

17:30

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

CD40 AND PERIPHERAL TOLERANCE

6.

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

7.

8.

9.

10.

11.

12.

13.

ligand on activated platelets triggers an inflammatory reaction of endothelial cells. Nature 391:591–94 Danese S, de la Motte C, Sturm A, Vogel JD, West GA, et al. 2003. Platelets trigger a CD40-dependent inflammatory response in the microvasculature of inflammatory bowel disease patients. Gastroenterology 124:1249–64 Filion LG, Matusevicius D, GrazianiBowering GM, Kumar A, Freedman MS. 2003. Monocyte-derived IL12, CD86 (B7-2) and CD40L expression in relapsing and progressive multiple sclerosis. Clin. Immunol. 106:127–38 Katsiari CG, Liossis SN, Dimopoulos AM, Charalambopoulo DV, Mavrikakis M, Sfikakis PP. 2002. CD40L overexpression on T cells and monocytes from patients with systemic lupus erythematosus is resistant to calcineurin inhibition. Lupus 11:370–78 Schonbeck U, Gerdes N, Varo N, Reynolds RS, Horton DB, et al. 2002. Oxidized low-density lipoprotein augments and 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors limit CD40 and CD40L expression in human vascular cells. Circulation 106:2888–93 Garside P, Ingulli E, Merica RR, Johnson JG, Noelle RJ, Jenkins MK. 1998. Visualization of specific B and T lymphocyte interactions in the lymph node. Science 281:96–99 Jabara H, Laouini D, Tsitsikov E, Mizoguchi E, Bhan A, et al. 2002. The binding site for TRAF2 and TRAF3 but not for TRAF6 is essential for CD40-mediated immunoglobulin class switching. Immunity 17:265–76 Ahonen C, Manning E, Erickson LD, O’Connor B, Lind EF, et al. 2002. The CD40-TRAF6 axis controls affinity maturation and the generation of long-lived plasma cells. Nat. Immunol. 3:451–56 Burkly LC. 2001. CD40 pathway blockade as an approach to immunotherapy. Adv. Exp. Med. Biol. 489:135–52

P1: IKH

321

14. Kirk AD, Blair PJ, Tadaki DK, Xu H, Harlan DM. 2001. The role of CD154 in organ transplant rejection and acceptance. Philos. Trans. R. Soc. London Ser. B 356:691– 702 15. Markees TG, Phillips NE, Noelle RJ, Shultz LD, Mordes JP, et al. 1997. Prolonged survival of mouse skin allografts in recipients treated with donor splenocytes and antibody to CD40 ligand. Transplantation 64:329–35 16. Vogel LA, Noelle RJ. 1998. CD40 and its crucial role as a member of the TNFR family. Semin. Immunol. 10:435–42 17. Becher B, Durell BG, Miga AV, Hickey WF, Noelle RJ. 2001. The clinical course of experimental autoimmune encephalomyelitis and inflammation is controlled by the expression of CD40 within the central nervous system. J. Exp. Med. 193:967–74 18. Howard LM, Dal Canto MC, Miller SD. 2002. Transient anti-CD154-mediated immunotherapy of ongoing relapsing experimental autoimmune encephalomyelitis induces long-term inhibition of disease relapses. J. Neuroimmunol. 129:58–65 19. Davidson A, Wang X, Mihara M, Ramanujam M, Huang W, et al. 2003. Costimulatory blockade in the treatment of murine systemic lupus erythematosus (SLE). Ann. NY Acad. Sci. 987:188–98 20. Boumpas DT, Furie R, Manzi S, Illei GG, Wallace DJ, et al. 2003. A short course of BG9588 (anti-CD40 ligand antibody) improves serologic activity and decreases hematuria in patients with proliferative lupus glomerulonephritis. Arthritis. Rheum. 48:719–27 21. Kalunian KC, Davis JC Jr, Merrill JT, Totoritis MC, Wofsy D. 2002. Treatment of systemic lupus erythematosus by inhibition of T cell costimulation with anti-CD154: a randomized, double-blind, placebo-controlled trial. Arthritis. Rheum. 46:3251–58 22. Jarvinen LZ, Blazar BR, Adeyi OA, Strom TB, Noelle RJ. 2003. CD154 on the

11 Feb 2004

17:30

322

23.

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

24.

25.

26.

27.

28.

29.

30.

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL. surface of CD4+CD25+ regulatory T cells contributes to skin transplant tolerance. Transplantation 76:In press Gordon EJ, Markees TG, Phillips NE, Noelle RJ, Shultz LD, et al. 1998. Prolonged survival of rat islet and skin xenografts in mice treated with donor splenocytes and anti-CD154 monoclonal antibody. Diabetes 47:1199–206 Quezada SA, Fuller B, Jarvinen LZ, Gonzalez M, Blazar BR, et al. 2003. Mechanisms of donor specific transfusion tolerance: preemptive induction of clonal T cell exhaustion via indirect presentation. Blood 102:1920–26 Elster EA, Xu H, Tadaki DK, Montgomery S, Burkly LC, et al. 2001. Treatment with the humanized CD154-specific monoclonal antibody, hu5C8, prevents acute rejection of primary skin allografts in nonhuman primates. Transplantation 72:1473–78 Benda B, Ljunggren HG, Peach R, Sandberg JO, Korsgren O. 2002. Costimulatory molecules in islet xenotransplantation: CTLA4Ig treatment in CD40 ligand-deficient mice. Cell Transplant. 11:715–20 Kurtz J, Ito H, Wekerle T, Shaffer J, Sykes M. 2001. Mechanisms involved in the establishment of tolerance through costimulatory blockade and BMT: lack of requirement for CD40L-mediated signaling for tolerance or deletion of donor-reactive CD4+ cells. Am. J. Transplant. 1:339–49 Camirand G, Caron NJ, Turgeon NA, Rossini AA, Tremblay JP. 2002. Treatment with anti-CD154 antibody and donor-specific transfusion prevents acute rejection of myoblast transplantation. Transplantation 73:453–61 Tung TH, Mackinnon SE, Mohanakumar T. 2003. Long-term limb allograft survival using anti-CD40L antibody in a murine model. Transplantation 75:644–50 Pullen SS, Miller HG, Everdeen DS, Dang TT, Crute JJ, Kehry MR. 1998. CD40tumor necrosis factor receptor-associated

31.

32.

33.

34.

35.

36.

37.

38.

39.

factor (TRAF) interactions: regulation of CD40 signaling through multiple TRAF binding sites and TRAF hetero-oligomerization. Biochemistry 37:11836–45 Banchereau J, Steinman RM. 1998. Dendritic cells and the control of immunity. Nature 392:245–52 Frleta D, Lin JT, Quezada SA, Wade TK, Barth RJ, et al. 2003. Distinctive maturation of in vitro versus in vivo anti-CD40 mAb-matured dendritic cells in mice. J. Immunother. 26:72–84 Mackey MF, Wang Z, Eichelberg K, Germain RN. 2003. Distinct contributions of different CD40 TRAF binding sites to CD154-induced dendritic cell maturation and IL-12 secretion. Eur. J. Immunol. 33:779–89 Manning E, Pullen SS, Souza DJ, Kehry M, Noelle RJ. 2002. Cellular responses to murine CD40 in a mouse B cell line may be TRAF dependent or independent. Eur. J. Immunol. 32:39–49 Wong P, Pamer EG. 2003. Feedback regulation of pathogen-specific T cell priming. Immunity 18:499–511 De Smedt T, Pajak B, Klaus GG, Noelle RJ, Urbain J, et al. 1998. Antigen-specific T lymphocytes regulate lipopolysaccharide-induced apoptosis of dendritic cells in vivo. J. Immunol. 161:4476–79 Miga AJ, Masters SR, Durell BG, Gonzalez M, Jenkins MK, et al. 2001. Dendritic cell longevity and T cell persistence is controlled by CD154-CD40 interactions. Eur. J. Immunol. 31:959–65 Josien R, Li HL, Ingulli E, Sarma S, Wong BR, et al. 2000. TRANCE, a tumor necrosis factor family member, enhances the longevity and adjuvant properties of dendritic cells in vivo. J. Exp. Med. 191:495– 502 Wong BR, Josien R, Lee SY, Sauter B, Li HL, et al. 1997. TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a

11 Feb 2004

17:30

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

CD40 AND PERIPHERAL TOLERANCE

40.

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

dendritic cell-specific survival factor. J. Exp. Med. 186:2075–80 Cremer I, Dieu-Nosjean MC, Marechal S, Dezutter-Dambuyant C, Goddard S, et al. 2002. Long-lived immature dendritic cells mediated by TRANCE-RANK interaction. Blood 100:3646–55 Ouaaz F, Arron J, Zheng Y, Choi YW, Beg AA. 2002. Dendritic cell development and survival require distinct NF-kappaB subunits. Immunity 16:257–70 Ghosh S, Karin M. 2002. Missing pieces in the NF-kappaB puzzle. Cell 109(Suppl.):S81–96 Pomerantz JL, Baltimore D. 2002. Two pathways to NF-kappaB. Mol. Cell 10:693–95 Grumont R, Hochrein H, O’Keeffe M, Gugasyan R, White C, et al. 2001. c-Rel regulates interleukin 12 p70 expression in CD8(+) dendritic cells by specifically inducing p35 gene transcription. J. Exp. Med. 194:1021–32 Yin L, Wu L, Wesche H, Arthur CD, White JM, et al. 2001. Defective lymphotoxin-beta receptor-induced NFkappaB transcriptional activity in NIKdeficient mice. Science 291:2162–65 Kayagaki N, Yan MH, Seshasayee D, Wang H, Lee W, et al. 2002. BAFF/BLyS receptor 3 binds the B cell survival factor BAFF ligand through a discrete surface loop and promotes processing of NFkappaB2. Immunity 17:515–24 Claudio E, Brown K, Park S, Wang HS, Siebenlist U. 2002. BAFF-induced NEMO-independent processing of NFkappa B2 in maturing B cells. Nat. Immunol. 3:958–65 Saccani S, Pantano S, Natoli G. 2003. Modulation of NF-kappaB activity by exchange of dimers. Mol. Cell 11:1563–74 Lind EF, Kosaka Y, Noelle RJ. 2003. The NF-kappa B2 pathway controls dendritic cell longevity and immunity. Submitted Martin E, O’Sullivan B, Low P, Thomas R. 2003. Antigen-specific suppression of a primed immune response by dendritic

51.

52.

53.

54.

55.

56.

57.

58.

P1: IKH

323

cells mediated by regulatory T cells secreting interleukin-10. Immunity 18:155– 67 Mackey MF, Gunn JR, Maliszewsky C, Kikutani H, Noelle RJ, Barth RJ Jr. 1998. Dendritic cells require maturation via CD40 to generate protective antitumor immunity. J. Immunol. 161:2094–98 de Gruijl TD, Luykx-de Bakker SA, Tillman BW, van den Eertwegh AJ, Buter J, et al. 2002. Prolonged maturation and enhanced transduction of dendritic cells migrated from human skin explants after in situ delivery of CD40-targeted adenoviral vectors. J. Immunol. 169:5322–31 Morel Y, Truneh A, Sweet RW, Olive D, Costello RT. 2001. The TNF superfamily members LIGHT and CD154 (CD40 ligand) costimulate induction of dendritic cell maturation and elicit specific CTL activity. J. Immunol. 167:2479–86 Schulz O, Edwards AD, Schito M, Aliberti J, Manickasingham S, et al. 2000. CD40 triggering of heterodimeric IL-12 p70 production by dendritic cells in vivo requires a microbial priming signal. Immunity 13:453–62 Krug A, Towarowski A, Britsch S, Rothenfusser S, Hornung V, et al. 2001. Toll-like receptor expression reveals CpG DNA as a unique microbial stimulus for plasmacytoid dendritic cells which synergizes with CD40 ligand to induce high amounts of IL-12. Eur. J. Immunol. 31:3026–37 Bennett SR, Carbone FR, Karamalis F, Flavell RA, Miller JF, Heath WR. 1998. Help for cytotoxic-T-cell responses is mediated by CD40 signalling. Nature 393:478–80 Schoenberger SP, Toes RE, van der Voort EI, Offringa R, Melief CJ. 1998. T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L interactions. Nature 393:480–83 Asea A, Rehli M, Kabingu E, Boch JA, Bare O, et al. 2002. Novel signal transduction pathway utilized by extracellular

11 Feb 2004

17:30

324

59.

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL. HSP70: role of toll-like receptor (TLR) 2 and TLR4. J. Biol. Chem. 277:15028–34 Vabulas RM, Ahmad-Nejad P, Ghose S, Kirschning CJ, Issels RD, Wagner H. 2002. HSP70 as endogenous stimulus of the Toll/interleukin-1 receptor signal pathway. J. Biol. Chem. 277:15107–12 Vabulas RM, Wagner H, Schild H. 2002. Heat shock proteins as ligands of toll-like receptors. Curr. Top. Microbiol. Immunol. 270:169–84 Kedl RM, Jordan M, Potter T, Kappler J, Marrack P, Dow S. 2001. CD40 stimulation accelerates deletion of tumor-specific CD8(+) T cells in the absence of tumorantigen vaccination. Proc. Natl. Acad. Sci. USA 98:10811–16 Mauri C, Mars LT, Londei M. 2000. Therapeutic activity of agonistic monoclonal antibodies against CD40 in a chronic autoimmune inflammatory process. Nat. Med. 6:673–79 Janeway CA Jr, Medzhitov R. 1999. Lipoproteins take their toll on the host. Curr. Biol. 9:R879–82 Medzhitov R, Janeway C Jr. 2000. The Toll receptor family and microbial recognition. Trends Microbiol. 8:452–56 Barton GM, Medzhitov R. 2002. Toll-like receptors and their ligands. Curr. Top. Microbiol. Immunol. 270:81–92 Barton GM, Medzhitov R. 2003. Tolllike receptor signaling pathways. Science 300:1524–25 Sparwasser T, Koch ES, Vabulas RM, Heeg K, Lipford GB, et al. 1998. Bacterial DNA and immunostimulatory CpG oligonucleotides trigger maturation and activation of murine dendritic cells. Eur. J. Immunol. 28:2045–54 Brunner C, Seiderer J, Schlamp A, Bidlingmaier M, Eigler A, et al. 2000. Enhanced dendritic cell maturation by TNFalpha or cytidine-phosphate-guanosine DNA drives T cell activation in vitro and therapeutic anti-tumor immune responses in vivo. J. Immunol. 165:6278–86 Jakob T, Walker PS, Krieg AM, Udey

70.

71.

72.

73.

74.

75.

76.

MC, Vogel JC. 1998. Activation of cutaneous dendritic cells by CpG-containing oligodeoxynucleotides: a role for dendritic cells in the augmentation of Th1 responses by immunostimulatory DNA. J. Immunol. 161:3042–49 Melief CJ, Van der Burg SH, Toes RE, Ossendorp F, Offringa R. 2002. Effective therapeutic anticancer vaccines based on precision guiding of cytolytic T lymphocytes. Immunol. Rev. 188:177–82 Maxwell JR, Ruby C, Kerkvliet NI, Vella AT. 2002. Contrasting the roles of costimulation and the natural adjuvant lipopolysaccharide during the induction of T cell immunity. J. Immunol. 168: 4372–81 Cho HJ, Hayashi T, Datta SK, Takabayashi K, Van Uden JH, et al. 2002. IFN-alpha beta promote priming of antigen-specific CD8+ and CD4+ T lymphocytes by immunostimulatory DNAbased vaccines. J. Immunol. 168:4907–13 Gallimore A, Glithero A, Godkin A, Tissot AC, Pluckthun A, et al. 1998. Induction and exhaustion of lymphocytic choriomeningitis virus-specific cytotoxic T lymphocytes visualized using soluble tetrameric major histocompatibility complex class I-peptide complexes. J. Exp. Med. 187:1383–93 Murali-Krishna K, Altman JD, Suresh M, Sourdive DJ, Zajac AJ, et al. 1998. Counting antigen-specific CD8 T cells: a reevaluation of bystander activation during viral infection. Immunity 8:177–87 De Smedt T, Pajak B, Muraille E, Lespagnard L, Heinen E, et al. 1996. Regulation of dendritic cell numbers and maturation by lipopolysaccharide in vivo. J. Exp. Med. 184:1413–24 Inaba K, Turley S, Iyoda T, Yamaide F, Shimoyama S, et al. 2000. The formation of immunogenic major histocompatibility complex class II-peptide ligands in lysosomal compartments of dendritic cells is regulated by inflammatory stimuli. J. Exp. Med. 191:927–36

11 Feb 2004

17:30

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD40 AND PERIPHERAL TOLERANCE 77. Noelle RJ. 1996. CD40 and its ligand in host defense. Immunity 4:415–19 78. Karmann K, Hughes CC, Schechner J, Fanslow WC, Pober JS. 1995. CD40 on human endothelial cells: inducibility by cytokines and functional regulation of adhesion molecule expression. Proc. Natl. Acad. Sci. USA 92:4342–46 79. Hollenbaugh D, Mischel-Petty N, Edwards CP, Simon JC, Denfeld RW, et al. 1995. Expression of functional CD40 by vascular endothelial cells. J. Exp. Med. 182:33–40 80. Bouneaud C, Kourilsky P, Bousso P. 2000. Impact of negative selection on the T cell repertoire reactive to a self-peptide: A large fraction of T cell clones escapes clonal deletion. Immunity 13:829–40 81. Gimmi CD, Freeman GJ, Gribben JG, Gray G, Nadler LM. 1993. Human T-cell clonal anergy is induced by antigen presentation in the absence of B7 costimulation. Proc. Natl. Acad. Sci. USA 90:6586– 90 82. Johnson JG, Jenkins MK. 1993. Accessory cell-derived signals required for T cell activation. Immunol. Res. 12:48–64 83. Steinman RM, Hawiger D, Nussenzweig MC. 2003. Tolerogenic dendritic cells. Annu. Rev. Immunol. 21:685–711 84. Sakaguchi S, Fukuma K, Kuribayashi K, Masuda T. 1985. Organ-specific autoimmune diseases induced in mice by elimination of T cell subset. I. Evidence for the active participation of T cells in natural self-tolerance; deficit of a T cell subset as a possible cause of autoimmune disease. J. Exp. Med. 161:72–87 85. Sakaguchi S, Sakaguchi N, Asano M, Itoh M, Toda M. 1995. Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alphachains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J. Immunol. 155:1151–64 86. Takahashi T, Kuniyasu Y, Toda M, Sakaguchi N, Itoh M, et al. 1998. Im-

87.

88.

89.

90.

91.

92.

93.

94.

95.

P1: IKH

325

munologic self-tolerance maintained by CD25+CD4+ naturally anergic and suppressive T cells: induction of autoimmune disease by breaking their anergic/ suppressive state. Int. Immunol. 10:1969– 80 Shimizu J, Yamazaki S, Takahashi T, Ishida Y, Sakaguchi S. 2002. Stimulation of CD25(+)CD4(+) regulatory T cells through GITR breaks immunological selftolerance. Nat. Immunol. 3:135–42 Fontenot JD, Gavin MA, Rudensky AY. 2003. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat. Immunol. 4:330–36 Gavin MA, Clarke SR, Negrou E, Gallegos A, Rudensky A. 2002. Homeostasis and anergy of CD4(+)CD25(+) suppressor T cells in vivo. Nat. Immunol. 3:33–41 Belkaid Y, Piccirillo CA, Mendez S, Shevach EM, Sacks DL. 2002. CD4+CD25+ regulatory T cells control Leishmania major persistence and immunity. Nature 420:502–7 Piccirillo CA, Letterio JJ, Thornton AM, McHugh RS, Mamura M, et al. 2002. CD4(+)CD25(+) regulatory T cells can mediate suppressor function in the absence of transforming growth factor beta1 production and responsiveness. J. Exp. Med. 196:237–46 Thornton AM, Shevach EM. 1998. CD4+CD25+ immunoregulatory T cells suppress polyclonal T cell activation in vitro by inhibiting interleukin 2 production. J. Exp. Med. 188:287–96 Piccirillo CA, Shevach EM. 2001. Cutting edge: control of CD8+ T cell activation by CD4+CD25+ immunoregulatory cells. J. Immunol. 167:1137–40 Sakaguchi S, Sakaguchi N, Shimizu J, Yamazaki S, Sakihama T, et al. 2001. Immunologic tolerance maintained by CD25+ CD4+ regulatory T cells: their common role in controlling autoimmunity, tumor immunity, and transplantation tolerance. Immunol. Rev. 182:18–32 Sakaguchi S. 2000. Regulatory T cells:

11 Feb 2004

17:30

326

96.

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

97.

98.

99.

100.

101.

102.

103.

104.

105.

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL. key controllers of immunologic selftolerance. Cell 101:455–58 Shevach EM. 2000. Regulatory T cells in autoimmmunity. Annu. Rev. Immunol. 18:423–49 Groux H, O’Garra A, Bigler M, Rouleau M, Antonenko S, et al. 1997. A CD4+ T-cell subset inhibits antigen-specific Tcell responses and prevents colitis. Nature 389:737–42 Roncarolo MG, Bacchetta R, Bordignon C, Narula S, Levings MK. 2001. Type 1 T regulatory cells. Immunol. Rev. 182:68– 79 Levings MK, Sangregorio R, Galbiati F, Squadrone S, de Waal Malefyt R, Roncarolo MG. 2001. IFN-alpha and IL-10 induce the differentiation of human type 1 T regulatory cells. J. Immunol. 166:5530– 39 Bluestone JA, Abbas AK. 2003. Natural versus adaptive regulatory T cells. Nat. Rev. Immunol. 3:253–57 Hawiger D, Inaba K, Dorsett Y, Guo M, Mahnke K, et al. 2001. Dendritic cells induce peripheral T cell unresponsiveness under steady state conditions in vivo. J. Exp. Med. 194:769–79 Bonifaz L, Bonnyay D, Mahnke K, Rivera M, Nussenzweig MC, Steinman RM. 2002. Efficient targeting of protein antigen to the dendritic cell receptor DEC-205 in the steady state leads to antigen presentation on major histocompatibility complex class I products and peripheral CD8+ T cell tolerance. J. Exp. Med. 196:1627– 38 Munn DH, Sharma MD, Lee JR, Jhaver KG, Johnson TS, et al. 2002. Potential regulatory function of human dendritic cells expressing indoleamine 2,3-dioxygenase. Science 297:1867–70 Steinman RM, Turley S, Mellman I, Inaba K. 2000. The induction of tolerance by dendritic cells that have captured apoptotic cells. J. Exp. Med. 191:411–16 Wilson NS, El-Sukkari D, Belz GT, Smith CM, Steptoe RJ, et al. 2003. Most lym-

106.

107.

108.

109.

110.

111.

112.

113.

114.

phoid organ dendritic cell types are phenotypically and functionally immature. Blood 102:2187–94 Scheinecker C, McHugh R, Shevach EM, Germain RN. 2002. Constitutive presentation of a natural tissue autoantigen exclusively by dendritic cells in the draining lymph node. J. Exp. Med. 196:1079–90 Mahnke K, Qian Y, Knop J, Enk AH. 2003. Induction of CD4+/CD25+ regulatory T cells by targeting of antigens to immature dendritic cells. Blood 101:4862– 69 Albert ML, Sauter B, Bhardwaj N. 1998. Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs. Nature 392:86–89 Albert ML, Pearce SF, Francisco LM, Sauter B, Roy P, et al. 1998. Immature dendritic cells phagocytose apoptotic cells via alphavbeta5 and CD36, and crosspresent antigens to cytotoxic T lymphocytes. J. Exp. Med. 188:1359–68 Inaba K, Turley S, Yamaide F, Iyoda T, Mahnke K, et al. 1998. Efficient presentation of phagocytosed cellular fragments on the major histocompatibility complex class II products of dendritic cells. J. Exp. Med. 188:2163–73 Sauter B, Albert ML, Francisco L, Larsson M, Somersan S, Bhardwaj N. 2000. Consequences of cell death: Exposure to necrotic tumor cells, but not primary tissue cells or apoptotic cells, induces the maturation of immunostimulatory dendritic cells. J. Exp. Med. 191:423–34 Huynh ML, Fadok VA, Henson PM. 2002. Phosphatidylserine-dependent ingestion of apoptotic cells promotes TGFbeta1 secretion and the resolution of inflammation. J. Clin. Invest. 109:41–50 Fadok VA, Bratton DL, Rose DM, Pearson A, Ezekewitz RA, Henson PM. 2000. A receptor for phosphatidylserinespecific clearance of apoptotic cells. Nature 405:85–90 Kuppner MC, Gastpar R, Gelwer S, Nossner E, Ochmann O, et al. 2001. The role

11 Feb 2004

17:30

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

CD40 AND PERIPHERAL TOLERANCE

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

115.

116.

117.

118.

119.

120.

121.

122.

of heat shock protein (hsp70) in dendritic cell maturation: hsp70 induces the maturation of immature dendritic cells but reduces DC differentiation from monocyte precursors. Eur. J. Immunol. 31:1602–9 Feng H, Zeng Y, Graner MW, Likhacheva A, Katsanis E. 2003. Exogenous stress proteins enhance the immunogenicity of apoptotic tumor cells and stimulate antitumor immunity. Blood 101:245–52 Ichikawa HT, Williams LP, Segal BM. 2002. Activation of APCs through CD40 or Toll-like receptor 9 overcomes tolerance and precipitates autoimmune disease. J. Immunol. 169:2781–87 Roth E, Schwartzkopff J, Pircher H. 2002. CD40 ligation in the presence of selfreactive CD8 T cells leads to severe immunopathology. J. Immunol. 168:5124– 29 Diehl L, den Boer AT, Schoenberger SP, van der Voort EI, Schumacher TN, et al. 1999. CD40 activation in vivo overcomes peptide-induced peripheral cytotoxic Tlymphocyte tolerance and augments antitumor vaccine efficacy. Nat. Med. 5:774– 79 Garza KM, Chan SM, Suri R, Nguyen LT, Odermatt B, et al. 2000. Role of antigen-presenting cells in mediating tolerance and autoimmunity. J. Exp. Med. 191:2021–27 Linsley PS, Greene JL, Brady W, Bajorath J, Ledbetter JA, Peach R. 1994. Human B7-1 (CD80) and B7-2 (CD86) bind with similar avidities but distinct kinetics to CD28 and CTLA-4 receptors. Immunity 1:793–801 Lenschow DJ, Zeng Y, Thistlethwaite JR, Montag A, Brady W, et al. 1992. Longterm survival of xenogeneic pancreatic islet grafts induced by CTLA4lg. Science 257:789–92 Baliga P, Chavin KD, Qin L, Woodward J, Lin J, et al. 1994. CTLA4Ig prolongs allograft survival while suppressing cell-mediated immunity. Transplantation 58:1082–90

P1: IKH

327

123. Kirk AD, Harlan DM, Armstrong NN, Davis TA, Dong Y, et al. 1997. CTLA4Ig and anti-CD40 ligand prevent renal allograft rejection in primates. Proc. Natl. Acad. Sci. USA 94:8789–94 124. Onodera K, Chandraker A, Volk HD, Ritter T, Lehmann M, et al. 1999. Distinct tolerance pathways in sensitized allograft recipients after selective blockade of activation signal 1 or signal 2. Transplantation 68:288–93 125. Zheng XX, Markees TG, Hancock WW, Li Y, Greiner DL, et al. 1999. CTLA4 signals are required to optimally induce allograft tolerance with combined donorspecific transfusion and anti-CD154 monoclonal antibody treatment. J. Immunol. 162:4983–90 126. Markees TG, Phillips NE, Gordon EJ, Noelle RJ, Shultz LD, et al. 1998. Longterm survival of skin allografts induced by donor splenocytes and anti-CD154 antibody in thymectomized mice requires CD4(+) T cells, interferon-gamma, and CTLA4. J. Clin. Invest. 101:2446–55 127. Ranheim EA, Kipps TJ. 1993. Activated T cells induce expression of B7/BB1 on normal or leukemic B cells through a CD40dependent signal. J. Exp. Med. 177:925– 35 128. Ranheim EA, Kipps TJ. 1995. Tumor necrosis factor-alpha facilitates induction of CD80 (B7-1) and CD54 on human B cells by activated T cells: complex regulation by IL-4, IL-10, and CD40L. Cell. Immunol. 161:226–35 129. Jones ND, Van Maurik A, Hara M, Spriewald BM, Witzke O, et al. 2000. CD40-CD40 ligand-independent activation of CD8+ T cells can trigger allograft rejection. J. Immunol. 165:1111–18 130. Lehnert AM, Mottram PL, Han W, Walters SN, Patel AT, et al. 2001. Blockade of the CD28 and CD40 pathways result in the acceptance of pig and rat islet xenografts but not rat cardiac grafts in mice. Transpl. Immunol. 9:51–56 131. Honey K, Cobbold SP, Waldmann H.

11 Feb 2004

17:30

328

132.

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

133.

134.

135.

136.

137.

138.

139.

AR

AR210-IY22-10.tex

AR210-IY22-10.sgm

LaTeX2e(2002/01/18)

P1: IKH

QUEZADA ET AL. 1999. CD40 ligand blockade induces CD4+ T cell tolerance and linked suppression. J. Immunol. 163:4805–10 Larsen CP, Elwood ET, Alexander DZ, Ritchie SC, Hendrix R, et al. 1996. Longterm acceptance of skin and cardiac allografts after blocking CD40 and CD28 pathways. Nature 381:434–38 Li Y, Zheng XX, Li XC, Zand MS, Strom TB. 1998. Combined costimulation blockade plus rapamycin but not cyclosporine produces permanent engraftment. Transplantation 66:1387–88 Li Y, Li XC, Zheng XX, Wells AD, Turka LA, Strom TB. 1999. Blocking both signal 1 and signal 2 of T-cell activation prevents apoptosis of alloreactive T cells and induction of peripheral allograft tolerance. Nat. Med. 5:1298–302 Li W, Lu L, Wang Z, Wang L, Fung JJ, et al. 2001. Costimulation blockade promotes the apoptotic death of graftinfiltrating T cells and prolongs survival of hepatic allografts from FLT3L-treated donors. Transplantation 72:1423–32 Wells AD, Li XC, Li Y, Walsh MC, Zheng XX, et al. 1999. Requirement for T-cell apoptosis in the induction of peripheral transplantation tolerance. Nat. Med. 5:1303–7 Sho M, Yamada A, Najafian N, Salama AD, Harada H, et al. 2002. Physiological mechanisms of regulating alloimmunity: cytokines, CTLA-4, CD25+ cells, and the alloreactive T cell clone size. J. Immunol. 169:3744–51 Brennan DC, Mohanakumar T, Flye MW. 1995. Donor-specific transfusion and donor bone marrow infusion in renal transplantation tolerance: a review of efficacy and mechanisms. Am. J. Kidney Dis. 26:701–15 Wood ML, Gottschalk R, Monaco AP. 1984. Comparison of immune responsiveness in mice after single or multiple donor-specific transfusions. J. Immunol. 132:651–55

140. Parker DC, Greiner DL, Phillips NE, Appel MC, Steele AW, et al. 1995. Survival of mouse pancreatic islet allografts in recipients treated with allogeneic small lymphocytes and antibody to CD40 ligand. Proc. Natl. Acad. Sci.USA 92:9560–64 141. Xu H, Montgomery SP, Preston EH, Tadaki DK, Hale DA, et al. 2003. Studies investigating pretransplant donor-specific blood transfusion, rapamycin, and the CD154-specific antibody IDEC-131 in a nonhuman primate model of skin allotransplantation. J. Immunol. 170:2776–82 142. Eynon EE, Parker DC. 1992. Small B cells as antigen-presenting cells in the induction of tolerance to soluble protein antigens. J. Exp. Med. 175:131–38 143. Liu K, Iyoda T, Saternus M, Kimura Y, Inaba K, Steinman RM. 2002. Immune tolerance after delivery of dying cells to dendritic cells in situ. J. Exp. Med. 196:1091–7 144. Coulombe M, Yang H, Wolf LA, Gill RG. 1999. Tolerance to antigen-presenting cell-depleted islet allografts is CD4 T cell dependent. J. Immunol. 162:2503–10 145. Wood KJ, Sakaguchi S. 2003. Regulatory T cells in transplantation tolerance. Nat. Rev. Immunol. 3:199–210 146. Qin S, Cobbold SP, Pope H, Elliott J, Kioussis D, et al. 1993. “Infectious” transplantation tolerance. Science 259:974–77 147. Graca L, Honey K, Adams E, Cobbold SP, Waldmann H. 2000. Cutting edge: antiCD154 therapeutic antibodies induce infectious transplantation tolerance. J. Immunol. 165:4783–86 148. Hara M, Kingsley CI, Niimi M, Read S, Turvey SE, et al. 2001. IL-10 is required for regulatory T cells to mediate tolerance to alloantigens in vivo. J. Immunol. 166:3789–96 149. Taylor PA, Noelle RJ, Blazar BR. 2001. CD4(+)CD25(+) immune regulatory cells are required for induction of tolerance to alloantigen via costimulatory blockade. J. Exp. Med. 193:1311–18

Quezada.qxd

3/12/2004

1:58 PM

Page 1

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD40 AND PERIPHERAL TOLERANCE

C-1

Figure 1 DC licensing and CD40 activation of the NF-kB pathways in DCs. (A) DC licensing. DCs reside in the periphery in an immature state. Immature DCs sample their environment and express low levels of MHC molecules. Upon receipt of inflammatory signals such as TLR engagement, they rapidly migrate to the PALS and interact with T cells. Engagement of the TCR by MHC peptide on the T cell results in upregulation of CD154, which further matures the DC and allows costimulation and activation of the T cell. Interfering with the CD40 signal results in tolerance induction via multiple pathways discussed in this review. (B) CD40 activation of NF-kB in DCs. Stimulation of the DC through CD40 results in clustering of CD40 and recruitment of the TRAF molecules. TRAF-receptor complex then recruits multiple kinases such as NIK and MEKK1. Multiple different kinases may result in phosphorylation of IKKb, which in turn results in ubiquitination of the IkB protein and release of NF-kB transcription factor dimers. This pathway leads to rapid inflammatory response and production of IL-12 and IL-8. Phosphorylation of IKKa occurs via NIK and results in phosphorylation of the p100 NF-kB2 precursor protein. This causes p100 to become ubiquitinated and partially digested by the proteosome, thus releasing p52/ RelB dimers. The NF-kB2 pathway results in sustained DC longevity and chemokine production.

Quezada.qxd

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

3/12/2004

1:58 PM

Page 2

QUEZADA ET AL.

Figure 2 A model for preemptive tolerization of the alloreactive T cell compartment. Upon DST 1 aCD154 injection (1), DST will soon be cleared by host DCs, allowing indirect presentation of donor Ags (2). CD154 blockade also has a proapoptotic effect on the DST, therefore facilitating a noninflammatory environment. Donor antigen indirectly presented to the alloreactive T cells in absence of CD40 signals and DC licensing results in rapid, systemic, and abortive expansion of the T cells (3a). Part of the alloreactive compartment is deleted, and the remaining T cells are hyporesponsive to further stimulation (4). At a similar tempo, CD4+CD25+ suppressor T cells clonally expand to the alloantigen (3b). Some of them migrate to the graft where they suppress expansion of alloreactive T cells recruited to that site (5). When Ag bearing donor DCs finally reach the draining LN from the donor skin (6), the alloreactive T cell compartment is extremely reduced and is nonresponsive to restimulation. Induction of Tr1 might occur and be one of the components of infectious tolerance together with CD4+CD25+ suppressor T cells, but the physiological mechanisms behind this process are still to be resolved.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:307-328. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104803

Annu. Rev. Immunol. 2004. 22:329–60 doi: 10.1146/annurev.immunol.22.012703.104803 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on October 15, 2003

THE THREE ES OF CANCER IMMUNOEDITING Gavin P. Dunn,1 Lloyd J. Old,2 and Robert D. Schreiber1 Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

1

Department of Pathology and Immunology, Center for Immunology, Washington University School of Medicine, St. Louis, Missouri 63110; email: [email protected] 2 Ludwig Institute for Cancer Research, New York Branch at Memorial Sloan-Kettering Cancer Center, New York, NY 10021; email: [email protected]

Key Words immunosurveillance, tumor, lymphocytes, interferon, tumor sculpting ■ Abstract After a century of controversy, the notion that the immune system regulates cancer development is experiencing a new resurgence. An overwhelming amount of data from animal models—together with compelling data from human patients— indicate that a functional cancer immunosurveillance process indeed exists that acts as an extrinsic tumor suppressor. However, it has also become clear that the immune system can facilitate tumor progression, at least in part, by sculpting the immunogenic phenotype of tumors as they develop. The recognition that immunity plays a dual role in the complex interactions between tumors and the host prompted a refinement of the cancer immunosurveillance hypothesis into one termed “cancer immunoediting.” In this review, we summarize the history of the cancer immunosurveillance controversy and discuss its resolution and evolution into the three Es of cancer immunoediting— elimination, equilibrium, and escape.

INTRODUCTION The concept that the immune system can recognize and eliminate primary developing tumors in the absence of external therapeutic intervention has existed for nearly 100 years. However, the validity of this concept has, in the past, been difficult to establish. When first proposed in 1909 (1), the hypothesis could not be experimentally tested because so little was known at the time about the molecular and cellular basis of immunity. Later on, as the field of immunology developed and the concept acquired its name—cancer immunosurveillance (2, 3)—experimental testing became possible but failed to provide evidence for the process, using mice with spontaneous mutations that rendered them immunocompromised but not completely immunodeficient (4). Only recently, with the development of gene targeting and transgenic mouse technologies and the capacity to produce highly specific blocking monoclonal antibodies (mAb) to particular immune components, has the cancer immunosurveillance hypothesis become testable in unequivocal, molecularly defined murine models of immunodeficiency. Over the past ten years, the use of these 0732-0582/04/0423-0329$14.00

329

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

330

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

improved in vivo cancer models has provided strong and convincing data that have rekindled interest in the cancer immunosurveillance hypothesis. Most recently, this conundrum has been further clarified by the demonstration that the immune system not only can protect the host against tumor development but also, by selecting for tumors of lower immunogenicity, has the capacity to promote tumor growth. These dual effects of the immune system on developing tumors prompted us to refine the cancer immunosurveillance hypothesis into one we termed cancer immunoediting (5, 6). We envisage that this process is comprised of three phases that are collectively denoted the three Es of cancer immunoediting: elimination, equilibrium, and escape. In this review, we first present data supporting the existence of the elimination phase (i.e., cancer immunosurveillance) as it occurs in mice and humans and propose a model for the molecular and cellular events that underlie this process. Second, we provide evidence for a tumor-sculpting role of immunity and discuss the relationship between this function and the equilibrium and escape phases of cancer immunoediting. Third, we outline the implications of this concept for the understanding and treatment of human cancer.

CANCER IMMUNOSURVEILLANCE IN MICE Historical Perspective The validity of the cancer immunosurveillance hypothesis has emerged only recently from a long history of heated debate (reviewed in 6). The notion that the immune system could protect the host from neoplastic disease was initially proposed by Ehrlich (1) and formally introduced as the cancer immunosurveillance hypothesis nearly 50 years later by Burnet and Thomas (2, 3, 7–9). Based on an emerging understanding of the cellular basis of transplantation and tumor immunity (10–15), Burnet and Thomas predicted that lymphocytes were responsible for eliminating continuously arising, nascent transformed cells. However, when this prediction was put to the experimental test using nude mice, which were the most congenitally immunodeficient mice available at the time (16, 17), no convincing evidence for such a process was obtained. Specifically, CBA/H strain nude mice neither developed increased incidences of carcinogen [methylcholanthrene (MCA)]-induced or spontaneous tumors nor did they show shortened periods of tumor latency compared with wild-type controls (4, 18–22). However, in retrospect, there are several important caveats to these experiments that could not have been appreciated at the time. First, the nude mouse is now recognized to be an imperfect model of immunodeficiency. These mice produce low but detectable numbers of functional populations of αβ T cells (23–25) and therefore can manifest at least some degree of adaptive immunity. Second, the existence of natural killer (NK) cells (which are present and function normally in nude mice) was not well established at the time (26) and thus very little was known about their origins, actions, or roles in promoting innate immunity. In addition, the profound influence of innate immunity on adaptive immunity was

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CANCER IMMUNOEDITING

P1: IKH

331

not recognized (27). Thus, the residual adaptive immune system in the presence of a fully functional innate immune system may provide the nude mouse with at least some cancer immunosurveillance capacity. Third, the CBA/H strain mice used in Stutman’s MCA carcinogenesis experiments express the highly active isoform of the aryl hydroxylase enzyme that is required to metabolize MCA into its carcinogenic form (28, 29). Therefore, it is conceivable that MCA-induced cellular transformation in CBA/H strain mice occurred so efficiently that it masked any protective effect that immunity could provide. Nevertheless, since these caveats can only be appreciated in hindsight, the Stutman experiments were considered to be so convincing that by the end of the 1970s, the death knell had sounded for the cancer immunosurveillance hypothesis.

THE RENAISSANCE OF CANCER IMMUNOSURVEILLANCE IFN-γ , Perforin, and Lymphocytes in Tumor Immunity In the 1990s, two sets of studies incited renewed interest in cancer immunosurveillance. First, endogenously produced interferon-γ (IFN-γ ) was shown to protect the host against the growth of transplanted tumors and the formation of primary chemically induced and spontaneous tumors (30–33). The injection of neutralizing monoclonal antibodies specific for IFN-γ into mice bearing transplanted, established Meth A tumors blocked LPS-induced tumor rejection (30). In addition, transplanted fibrosarcomas grew faster and more efficiently in mice treated with IFN-γ -specific mAb. These observations were then extended to models of primary tumor formation. IFN-γ -insensitive 129/SvEv mice lacking either the IFNGR1 ligand-binding subunit of the IFN-γ receptor or STAT1, the transcription factor responsible for mediating much of IFN-γ ’s biologic effects on cells (34), were found to be 10–20 times more sensitive than wild-type mice to tumor induction by methylcholanthrene (31). Specifically, these mice developed more tumors, more rapidly, and at lower MCA doses than did wild-type controls. These results were subsequently confirmed by independent experiments using C57BL/6 strain mice lacking the gene encoding IFN-γ itself (32). Similarly, in models of genetically driven tumorigenesis, mice lacking the p53 tumor suppressor gene and either IFNGR1 or STAT1 formed a wider spectrum of tumors compared with IFN-γ -sensitive mice lacking only p53 (31). In addition, compared to their IFN-γ sufficient counterparts, IFN-γ −/− C57BL/6 mice showed an increased incidence of disseminated lymphomas, and IFN-γ −/− BALB/c mice displayed an increased incidence of spontaneous lung adenocarcinomas (33). Second, mice lacking perforin (pfp−/−) were found to be more susceptible to MCA-induced and spontaneous tumor formation compared with their wildtype counterparts (32, 33, 35–37). Perforin is a component of the cytolytic granules of cytotoxic T cells and NK cells that plays an important role in mediating lymphocyte-dependent killing (38). Following challenge with MCA, pfp−/− mice developed significantly more tumors compared with wild-type mice treated in the

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

332

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

same manner (32, 35, 36). Untreated pfp−/− mice also showed a high incidence of spontaneous disseminated lymphomas, which was accelerated on a p53−/− background (37). BALB/c mice lacking perforin also displayed a low incidence of spontaneous lung adenocarcinomas, which was not observed in wild-type mice (33). Taken together, these observations demonstrated that tumor development in mice was controlled by components of the immune system and stimulated a considerable amount of work aimed at better defining this process (Table 1). The definitive work demonstrating the existence of an IFN-γ - and lymphocytedependent cancer immunosurveillance process was based on experiments employing gene-targeted mice that lack the recombinase activating gene (RAG)-2 (5). Mice lacking RAG-2 (or its obligate partner RAG-1) cannot rearrange lymphocyte antigen receptors and thus lack T, B, and NKT cells (39). Since RAG-2 expression is limited to cells of the immune system, the use of RAG-2−/− mice provided an appropriate model to study the effects of host immunodeficiency on tumor development because, unlike other genetic models of immunodeficiency (such as SCID mice), the absence of RAG-2 would not result in impaired DNA repair in nonlymphoid cells undergoing transformation. Following challenge with MCA, 129/SvEv RAG-2−/− mice developed sarcomas more rapidly and with greater frequency than genetically matched wild-type controls (5) (Figure 1A). After 160 days, 30/52 RAG-2−/− mice formed tumors, compared with 11/57 wild-type mice. Similar findings were obtained in MCA tumorigenesis experiments that used RAG-1−/− C57BL/6 mice (40). Moreover, Helicobacter-free RAG-2−/− 129/SvEv mice aged in a specific pathogen-free mouse facility and maintained on broad-spectrum antibiotics formed far more spontaneous epithelial tumors than did wild-type mice housed in the same room (5; A.T. Bruce & R.D. Schreiber, unpublished observations) (Figure 1B). Specifically, 26/26 RAG-2−/− mice ranging in age from 13– 24 months developed spontaneous neoplasia, predominantly of the intestine; 8 of these mice had premalignant intestinal adenomas, 17 had intestinal adenocarcinomas, and 1 had both an intestinal adenoma and a lung adenocarcinoma. In contrast, only 5/20 wild-type mice aged 13–24 months developed spontaneous neoplasia, which was predominantly benign. Three wild-type mice developed adenomas of the Harderian gland, lung, and intestine, respectively; one developed a Harderian gland adenocarcinoma; and one developed an endometrial stromal carcinoma. Thus, lymphocytes protect mice against the formation of both chemically induced and spontaneous tumors. The overlap between the IFN-γ - and lymphocyte-dependent tumor suppressor pathways was explored by comparing tumor formation in 129/SvEv mice lacking either IFN-γ responsiveness (IFNGR1−/− or STAT1−/− mice), lymphocytes (RAG-2−/− mice), or both [RAG-2−/− X STAT1−/− (RkSk) mice] (5). Each of the four lines of gene-targeted mice formed three times more chemically induced tumors than syngeneic wild-type mice when injected with a single 100 µg dose of MCA (Figure 1A). Since no significant differences were detected between any of the gene-targeted mice, the conclusion was reached that the IFN-γ /STAT1 and lymphocyte-dependent extrinsic tumor suppressor mechanisms were heavily

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

CANCER IMMUNOEDITING

333

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 1 Enhanced susceptibility of immunodeficient mice to chemically induced and spontaneous tumors Tumor susceptibility relative to wild type

Technology

Immune status

References

RAG-2−/−

Lacks T, B, NKT cells

↑ MCA-induced sarcomas; ↑ spontaneous intestinal neoplasia

(5)

RAG-2−/− × STAT1−/− (RkSk)

Lacks T, B, NKT cells; IFNγ -, α/β-insensitive

↑ MCA-induced sarcomas; ↑ spontaneous intestinal and mammary neoplasia

(5)

RAG-1−/−

Lacks T, B, NKT cells

↑ MCA-induced sarcomas

(40)

BALB/c SCID

Lacks T, B, NKT cells

↑ MCA-induced sarcomas

(40)

TCRβ −/−

Lacks αβ T cells

↑ MCA-induced sarcomas

(58)

TCRδ −/−

Lacks γ δ T cells

↑ MCA-induced sarcomas; ↑ DMBA/TPA-induced skin tumors

(58)

Jα281−/−

Lacks NKT cell subset

↑ MCA-induced sarcomas

(32, 36, 40)

LMP2−/−

Lacks LMP2 subunit

↑ Spontaneous uterine neoplasms

(169)

Anti-asialo-GM1

Lacks NK cells, mono-cytes/macrophages

↑ MCA-induced sarcomas

(40)

Anti-NK1.1

Lacks NK, NKT cells

↑ MCA-induced sarcomas

(36, 40)

Anti-Thy1

Lacks T cells

↑ MCA-induced sarcomas

(36)

STAT1−/−

IFN-γ -, α/β-insensitive

↑ MCA-induced sarcomas; wider tumor spectrum in STAT1−/− × p53−/−

(5, 31)

IFNGR1−/−

IFN-γ -insensitive

↑ MCA-induced sarcomas; wider tumor spectrum in IFNGR1−/− × p53−/−

(5, 31)

IFN-γ −/−

Lacks IFN-γ

↑ MCA-induced sarcomas; B6: ↑ spontaneous disseminated lymphomas; BALB/c: ↑ spontaneous lung adenocarcinomas

(32, 33)

GM-CSF/IFN-γ −/−

Lacks GM-CSF, IFN-γ

↑ Spontaneous lymphomas; ↑ nonlymphoid solid cancers

(55)

Pfp−/− × IFN-γ −/−

Lacks Perforin, IFN-γ

↑ MCA-induced sarcomas; ↑ spontaneous disseminated lymphomas

(32, 33)

Pfp−/−

Lacks Perforin

↑ MCA-induced sarcomas; ↑ spontaneous disseminated lymphomas

(32, 33, 35–37)

TRAIL−/−

Lacks TRAIL

↑ MCA-induced sarcomas

(61)

Anti-TRAIL

Blockade of TRAIL function

↑ MCA-induced sarcomas; ↑ spontaneous sarcomas, disseminated lymphomas

(60)

IL-12p40−/−

Lacks IL-12

↑ MCA-induced sarcomas

(36)

Wt + IL-12

Exogenous IL-12

↓ MCA-induced sarcomas

(62)

Wt + α-GalCer

Exogenous NKT cell activation

↓ MCA-induced sarcomas

(63)

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

334

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

Figure 1 Increased incidence of chemically induced and spontaneous tumors in immunodeficient mice. (A) Age- and sex-matched mice were inoculated with 100 µg MCA and monitored for tumor development for 160 days. (B) Mice housed in a specific pathogen-free facility were monitored for spontaneous tumor development between 13–24 months. Adapted from Shankaran et al. (5).

overlapping. However, RkSk mice developed spontaneous breast tumors that were not observed in wild-type or RAG-2−/− mice, therefore demonstrating that the overlap between the two pathways was incomplete (Figure 1B). Similar findings were made in carcinogenesis experiments employing mice that lacked either perforin, IFN-γ , or both, where a small increase was observed in tumor induction in the doubly deficient mice compared with mice lacking only one of the two components (32).

Identification of the Components of the Immmunosurveillance Network TUMOR CELLS AS KEY TARGETS OF IFN-γ The finding that endogenously produced IFN-γ played a critical role in protecting mice against tumor development stimulated a search for the physiologically important cellular targets of this cytokine. Two approaches demonstrated that the tumor cell itself is an important IFN-γ target in tumor rejection. In the first, the effects of ablating IFN-γ sensitivity on the immunogenicity of IFN-γ -sensitive tumor cells was assessed using models of tumor cell transplantation (30). Meth A tumor cells, when

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CANCER IMMUNOEDITING

P1: IKH

335

engineered to be unresponsive to IFN-γ by overexpression of a dominant-negative IFNGR1 mutant (mgR.1IC) (41), grew more aggressively than mock-transfected controls when transplanted into na¨ıve syngeneic wild-type hosts and were resistant to LPS-induced tumor rejection (30, 41). Unlike their IFN-γ -sensitive counterparts, IFN-γ -insensitive Meth A.mgR.1IC cells failed to prime na¨ıve recipients for development of Meth A immunity and were poorly recognized when injected into mice with pre-established immunity to the parental wild-type tumor cell line. Similar results were obtained with a second fibrosarcoma derived from a C57BL/6 mouse (MCA-207). The second approach employed an opposite strategy where the effects on in vivo tumor growth were assessed following restoration of IFN-γ sensitivity to tumor cells generated in IFN-γ -insensitive IFNGR1−/− mice (31). When transplanted into wild-type mice, IFNGR1-deficient RAD.gR.28 tumor cells were highly tumorigenic and formed progressively growing tumors even when injected at very low cell number (10–100 cells/mouse). In contrast, when RAD.gR.28 cells were rendered responsive to IFN-γ by complementation with wild-type IFNGR1, the resulting tumor cell line (RAD.gR.28.mgR) was highly immunogenic and failed to form progressively growing tumors in wild-type recipients even when injected at high cell number (5 × 106 cells/mouse). Demonstration that RAD.gR.28.mgR rejection occurred via an IFN-γ -dependent immunologic mechanism was evidenced by the observations that (a) rejection of RAD.gR.28.mgR cells in wild-type mice was inhibited by administration of IFN-γ mAb (31), (b) rejection was inhibited if wild-type mice were depleted of either CD4+ or CD8+ T cells (A.T. Bruce & R.D. Schreiber, unpublished observations), and (c) RAD.gR.28.mgR cells formed progressively growing tumors when injected into RAG-2−/− mice (31). Thus, the effects of using IFN-γ -insensitive tumor cells are the same as blocking IFN-γ availability in the intact mouse: Immune rejection of the tumor is inhibited. Together, these results formed the basis for the conclusion that the tumor cell is a physiologically relevant target of IFN-γ in the tumor rejection process. Subsequent studies have pointed to several effects of IFN-γ on tumor cells that could promote tumor elimination. IFN-γ ’s capacity to enhance tumor cell immunogenicity by upregulating components of the MHC class I antigen processing and presentation pathway has been shown to be sufficient for tumor rejection. IFN-γ -insensitive RAD.gR.28 tumor cells engineered for enforced expression of either TAP-1 (5) or H-2Db (A.T. Bruce & R.D. Schreiber, unpublished observations) were rejected when transplanted into na¨ıve syngeneic recipients in an immunologic manner that was indistinguishable from that of IFN-γ -responsive RAD.gR.28.mgR cells. In contrast, RAD.gR.28 cells engineered for expression of H-2Kb were not rejected (5). The finding that enforced expression of H-2Db, but not H-2Kb, caused rejection of RAD.gR.28 corresponds to the H-2Db MHC restriction displayed by protective CD8+ T cells that arise naturally in mice immunized with RAD.gR.28.mgR (A.T. Bruce & R.D. Schreiber, unpublished observations). Thus, the capacity of IFN-γ to regulate tumor cell immunogenicity via enhancement of MHC class I pathway function is a physiologically relevant action that promotes tumor rejection.

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

336

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

Other known IFN-γ -dependent effects on developing tumors may also contribute to the rejection process; however, their physiologic relevance to the process has not yet been established. IFN-γ has profound antiproliferative and/or proapoptotic effects on certain tumor cells. In the former case, IFN-γ can induce expression of cell cycle inhibitors such as p21WAF1/CIP1 or p27KIP1 that bind to and inhibit the cyclin-dependent kinase CDK-2 (42) or CDK-4 (43), respectively. In the latter case, IFN-γ can induce expression of gene products such as caspase-1 (44, 45), Fas, and Fas ligand (46) that, under the proper conditions, can promote tumor cell apoptosis. IFN-γ can also stimulate tumor cells to produce the chemokines CXCL-9 (Mig) and CXCL-10 (IP-10), which, in addition to having potent chemoattractant activity for CXCR3-expressing leukocyte populations, also function as powerful inhibitors of angiogenesis (47–52). Although all the aforementioned processes likely contribute in some way to the antitumor response, the relative importance and interrelationships between the immunologic and nonimmunologic actions of IFN-γ on developing tumors in promoting tumor rejection requires further analysis. HOST CELLS AS POTENTIAL ADDITIONAL TARGETS OF IFN-γ Evidence has also been obtained supporting a role for IFN-γ and/or STAT1 at the level of the host immune system in the tumor rejection process. IFN-γ -unresponsive mice lacking STAT1 failed to reject highly immunogenic P198 tumor cells that were completely eliminated in wild-type mice (53). Similar findings have also been made using the highly immunogenic RAD.gR.28.mgR fibrosarcoma cell line that was rejected in wild-type mice but grew progressively in STAT1−/− mice (V. Shankaran & R.D. Schreiber, unpublished observations). In addition, T cells derived from STAT1−/− mice immunized with poorly immunogenic P1.HTR tumor cells in the presence of IL-12 failed to express cytolytic activity against the tumor. In contrast, T cells derived from similarly immunized wild-type mice developed potent cytocidal capacity. Mice lacking STAT6, which tend to polarize their CD4+ T cell compartment more easily into Th1 cells, spontaneously rejected poorly immunogenic P1.HTR tumor cells that grew progressively in wild-type mice (54). Thus, these studies suggest that IFN-γ ’s well-recognized STAT1-dependent promotion of CD4+ T cell polarization into Th1 cells facilitates development of the appropriate type of cellular immune response needed for tumor rejection. Another study revealed a more indirect immunological action of IFN-γ at the level of the host in preventing tumor development (55). Both GM-CSF/IFN-γ −/− doubly deficient and GM-CSF/IL-3/IFN-γ −/− triply deficient mice were found to be highly susceptible to bacterial infection, displayed acute and chronic inflammation in a variety of different organs, and developed high incidences of spontaneous lymphoma and nonlymphoid solid cancers. The incidences of infection, inflammation, and neoplasia were much reduced in mice lacking GM-CSF alone, IL-3 and GM-CSF only, or IFN-γ alone. Tumor development in the IL-3/GM-CSF/IFN-γ triply gene-targeted mice was prevented or delayed by maintaining the mice on broad-spectrum antibiotics from birth. These results suggest a role for IFN-γ , in combination with GM-CSF, in controlling chronic infections that can lead to

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CANCER IMMUNOEDITING

P1: IKH

337

a chronic inflammatory state that ultimately may result in cancer development. Clearly, the relationship between bacterial/microbial immunosurveillance and cancer immunosurveillance warrants further analysis but must await the development of in vivo models that can unequivocally differentiate between the two processes. Finally, other studies have suggested that host cells of nonimmunologic origin may also be important targets of IFN-γ in the antitumor response (56, 57). These studies report that IFN-γ can induce angiostatic effects in tumors by targeting nontransformed host cells that are in close proximity to the tumor. It is possible that the underlying mechanism of this effect is similar to the one that has already been discussed in the context of the tumor cells themselves—the IFN-γ -dependent induction in host stromal cells of the angiostatic chemokines IP-10 and Mig. THE CELLULAR EFFECTORS OF CANCER IMMUNOSURVEILLANCE Other studies have begun to shed light on the specific lymphocyte subsets that are involved in cancer immunosurveillance. Together, these studies have shown that components of both the adaptive and innate immune systems participate in the process. Girardi et al. (58) examined the relative contributions of different T-cell subsets in blocking primary tumor formation in mice lacking αβ T cells (TCRβ −/−) and/or γ δ T cells (TCRδ −/−). MCA treatment of either type of TCR−/− mouse led to an increased incidence of fibrosarcomas and spindle cell carcinomas compared with wild-type controls, thereby showing that both αβ and γ δ T-cell subsets play critical and nonredundant host-protective roles in this particular model of tumor development. However, in an initiation/promotion model of DMBA- and TPAinduced skin tumorigenesis, TCRδ −/− mice showed an increased susceptibility to tumor formation and a higher incidence of papilloma-to-carcinoma progression than wild-type mice, whereas TCRβ −/− mice did not. This result suggests that immunosurveillance may be a multivariable process requiring the actions of different immune effectors in a manner dependent on the tumor’s cell type of origin, mechanism of transformation, anatomic localization, and mechanism of immunologic recognition. NK and NKT cells represent cellular populations of the innate immune compartment that were shown to protect the host from tumor formation. C57BL/6 mice depleted of both NK and NKT cells using the NK1.1 mAb were two to three times more susceptible to MCA-induced tumor formation than wild-type controls (40). In the same study, C57BL/6 mice depleted of NK cells following anti-asialoGM1 treatment were two to three times more prone to developing MCA-induced tumors than control counterparts. Although anti-asialo-GM1 can also deplete activated macrophages, this study nevertheless supports the involvement of cells of innate immunity in blocking primary tumor development. A role for NKT cells in this process was implicated when Jα281−/− mice, which lack a large population of Vα14Jα281-expressing invariant NKT cells, were found to develop MCA-induced sarcomas at a higher incidence than their wild-type counterparts (36). Additional evidence pointing to cells of innate immunity as critical effectors of cancer immunosurveillance comes from studies of the TNF-related

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

338

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

apoptosis-inducing ligand (TRAIL). A member of the TNF superfamily that induces apoptosis through engagement of the TRAIL-R2 (DR5) receptor in mice, TRAIL is expressed constitutively on a subset of liver NK cells and is induced by either IFN-γ or IFN-α/β in monocytes, NK cells, and dendritic cells (59). When injected with low doses of MCA, C57BL/6 strain mice treated with neutralizing antibodies to TRAIL (60) or lacking the TRAIL gene (61) developed fibrosarcomas at a higher incidence than wild-type controls. Moreover, C57BL/6 strain p53+/− mice treated with the same neutralizing TRAIL antibody exhibited a higher incidence of spontaneous sarcoma and disseminated lymphoma formation over a two-year span than control IgG-treated mice (60). Further study will be required to identify the specific innate cell subsets that manifest the TRAIL-dependent antitumor effects. Finally, evidence also exists showing that enhancing immune system activity leads to reduced primary tumor formation in models of MCA tumorigenesis. Mice treated with either IL-12 (62) or the prototypic NKT cell activator αgalactosylceramide (α-GalCer) (63) throughout the MCA carcinogenesis process had a reduced incidence of tumors after longer latency periods than control mice. In summary, using a variety of well-characterized gene-targeted mice, specific immune system activators, and blocking monoclonal antibodies highly specific for distinct immunologic components, a large body of work has now accumulated to support the statement that the immune system indeed functions to protect the murine host against development of both chemically induced and spontaneous tumors (Table 1).

CANCER IMMUNOSURVEILLANCE IN HUMANS Given that there is significant evidence supporting the existence of a cancer immunosurveillance process in mice, does a similar process exist in humans? Analysis of individuals with congenital or acquired immunodeficiencies or patients undergoing immunosuppressive therapy has documented a highly elevated incidence of virally induced malignancies such as Kaposi’s sarcoma, non-Hodgkin’s lymphoma, and cancers of the anal and urogenital tracts compared with immunocompetent individuals (64–66). However, the study of the incidence of cancers of nonviral origins that may take many years to develop is confounded by the variety of viral and bacterial infections to which these immunodeficient/immunosuppressed patients are susceptible and by the more rapid appearance of virally induced tumors. Nevertheless, one can draw upon three lines of evidence to suggest that cancer immunosurveillance indeed occurs in humans: (a) immunosuppressed transplant recipients display higher incidences of nonviral cancers than age-matched immunocompetent control populations; (b) cancer patients can develop spontaneous adaptive and innate immune responses to the tumors that they bear, and (c) the presence of lymphocytes within the tumor can be a positive prognostic indicator of patient survival.

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

CANCER IMMUNOEDITING

P1: IKH

339

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Transplant Recipients Display Increased Incidences of Malignancies Increased relative risk ratios have indeed been observed in immunosuppressed transplant recipients for a broad subset of tumors that have no apparent viral origin. Assessment of 5692 renal transplant patients from 1964–1982 in Finland, Denmark, Norway, and Sweden showed increased standardized cancer incidence ratios for colon, lung, bladder, kidney, ureter, and endocrine tumors compared to the general population (67). For example, the relative risks for colon cancer were 3.2 for men and 3.9 for women. In addition, analysis of 925 patients who received cadaveric renal transplants from 1965 to 1998 in Australia and New Zealand exhibited increased risk ratios for the development of a variety of cancers, including those of the colon, pancreas, lung, and endocrine tumors as well as malignant melanomas (68). When tumor incidence was examined in 608 cardiac transplant patients at the University of Pittsburgh between 1980 and 1993, the prevalence of lung tumors was 25-fold higher than in the general population (69). Furthermore, Penn researchers documented several examples of the increased incidence of tumors of nonviral etiology in immunosuppressed transplant patients through analysis of the Cincinnati Transplant Tumor Registry (CTTR). A review of data accumulated by this database from 1968 to 1995 found a twofold increase in risk in transplant patients for developing melanoma over that of the general population (70). Moreover, whereas only 0.3% to 0.4% of melanomas occur in the general pediatric population, the occurrence in pediatric transplant patients followed in the CTTR was 4% (70). These data complemented other studies that showed approximately fourfold increases in the incidence of de novo malignant melanoma after organ transplantation (71, 72). Finally, analysis of the CTTR showed that transplant patients were three times more likely to develop non-Kaposi’s sarcomas (73). Thus, individuals with normal immune systems who undergo immunosuppression display an increased probability of developing a variety of cancers that have not been linked to a viral etiology. This observation may indicate that immunosuppressive intervention predisposed the transplant patients either to de novo tumor formation or allowed the outgrowth of occult tumors whose growth was contained by a functioning immune system. Either way, these results suggest a protective action of immunity in preventing human tumors.

Spontaneous Tumor Recognition by Adaptive and Innate Immunity ADAPTIVE IMMUNE RESPONSES Substantial amounts of data support the concept that cancer patients can spontaneously develop specific adaptive immune responses to tumor antigens. Because the transplantation techniques used to demonstrate the presence of tumor-specific antigens in the mouse could not be employed in humans, in vitro approaches to identify immune responses to human tumor antigens needed to be developed. A systematic survey of the humoral and cellular immune

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

340

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

responses of patients to their own tumors was initiated in the 1970s using an approach termed autologous typing (74). Tumor cell lines were established from a large series of patients with melanoma or other tumor types that could be propagated in vitro, and these cells were used as targets for analysis of the humoral or cellular antitumor immune responses of the autologous patient. Fibroblasts and other autologous normal cell types served as control targets to assess the specificity of the antitumor response. Using this system, a small subset of patients was identified who had specific antibody to cell-surface antigens (75, 76) or who had T cells that recognized the autologous tumor (77). The characterization of the molecular targets recognized by autologous typing was made possible by application of the gene cloning and expression systems developed by Boon and colleagues to identify tumor antigens recognized by CD8+ T cells (78, 79) and by Pfreundschuh and colleagues for antibody-defined tumor antigens (80). More recently, it has been possible to identify MHC class II restricted tumor antigens recognized by CD4+ T cells (81). A large array of immunogenic human tumor antigens has now been identified (82–84). These can be segregated into the following four classes: Differentiation Antigens, e.g., melanocyte differentiation antigens, Melan-A/MART-1, tyrosinase, gp-100; Mutational Antigens, e.g., abnormal forms of p53; Overexpressed/Amplified Antigens, e.g., HER-2/neu; Viral Antigens, e.g., EBV and HPV; and Cancer-Testis (CT) Antigens. Using the currently available methodologies, the search for immunogenic human tumor antigens continues. The ultimate objective of this work is to define the human cancer immunome—the complete repertoire of human tumor antigens eliciting an immune response in humans—and a human cancer immunome database containing over 1000 human tumor antigens has been established (https://www2.licr.org/CancerImmunomeDB/). Because of their unique characteristics, CT antigens are of particular interest (85). In adult normal tissues, their expression is limited to germ cells in the testis, whereas in cancer, a variable proportion of a wide range of different tumor types expresses CT antigens. The first members of the CT family of antigens (MAGE, BAGE, GAGE) were cloned by Boon and his colleagues using CD8+ T cells from a patient having strong CD8+ T cell reactivity to autologous melanoma cells (79). The serological expression cloning technique (SEREX) developed by Pfreundschuh and colleagues (80) to detect the humoral response to human cancer has greatly expanded the list of CT antigens as well as other categories of tumor antigens, and there are now more than 20 CT antigens or antigen families recognized in human cancer (85). The analysis of the immune response to NY-ESO-1, a SEREX-defined CT antigen, is one of the best-documented examples of an integrated, naturally occurring spontaneous immune response to a nonviral human cancer. NY-ESO-1 was identified using SEREX analysis of an esophageal squamous cell carcinoma (86). Analysis of NY-ESO-1 at the mRNA and protein levels showed that NYESO-1 expression is limited to testis, fetal ovary, and placenta, but is detected in a variety of tumors including melanoma, bladder cancer, lung cancer, and

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CANCER IMMUNOEDITING

P1: IKH

341

synovial sarcomas (87). Antibody to NY-ESO-1 has been found only in patients with NY-ESO-1-expressing tumors; no antibody has been detected in patients with NY-ESO-1-negative tumors or in normal individuals (88). NY-ESO-1 antibody is rare in patients with early stage cancer, but can be found in up to 50% of patients with advanced NY-ESO-1+ tumors. The presence of antibody appears to be antigen driven, as removal of the tumor by surgery or following chemotherapy is frequently followed by disappearance of antibody (89). CD8+ and CD4+ T cell responses to NY-ESO-1 have been detected in patients with NY-ESO-1+ tumors, and a large number of MHC class I– and II–restricted NY-ESO-1 epitopes have been defined (90–92). These cellular responses are almost invariably associated with a strong NY-ESO-1 humoral immune response, documenting the integrated spontaneous immune response to this tumor antigen. Although there are indications that patients with spontaneous NY-ESO-1 immunity have a more favorable prognosis, proof that such an association exists is difficult to establish because of the variable clinical course of cancer and the influence of chemotherapy and other therapeutic interventions. The development of immunogenic NY-ESO-1 vaccines and randomized clinical trials will undoubtedly be necessary before definitive evidence linking NY-ESO-1 immunity to patient benefit can be substantiated. A second, well-characterized example of spontaneous immune responses to developing tumors in humans comes from the analysis of paraneoplastic neurologic disorders/degenerations (PNDs). PNDs are rare autoimmune neurologic diseases that are thought to be caused by “remote effects of cancer on the nervous system” (93), i.e., they are not caused by either direct primary or metastatic tumor invasion into nervous tissue but rather may be caused by cross reactivity of host antitumor responses with cells of the nervous system. Clinically, PNDs may affect any part of the nervous system and are most commonly associated with tumors of the breast, lung, and ovary (93). In the 1980s, an immunologic link between neuronal degeneration and the presence of cancer was established by the discovery that the serum and cerebrospinal fluid of PND patients harbored high titers of antibodies that reacted with neuronal antigens present in both the affected neuronal population and the associated cancer [antigens are discussed in depth in (94, 95)]. Furthermore, CTLs have been identified in the peripheral blood (96) and cerebrospinal fluid (97) of PND patients that can react with peptides from one of these antigens. However, CTL reactivity and cytotoxicity against intact neuronal cells and antigenexpressing tumor cells has yet to be demonstrated. Data from several clinical studies suggest that the presence of neuronal-reactive autoantibodies is associated with improved prognosis in cancer patients (98–100). Specifically, these studies have noted a positive correlation between the presence of antibody and the extent of disease, response to anticancer therapy, and survival. INNATE IMMUNE RESPONSES Recent studies indicate that the innate arm of the human immune system may also discriminate between tumor cells and normal cells and thus has the potential of participating in cancer immunosurveillance. These studies have centered largely on the human MHC class I chain-related proteins

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

342

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

A and B (MICA/B) that are differentially expressed on tumor cells and function as ligands for two receptors expressed on cells of the innate immune system: NKG2D and the T cell receptor on Vδ1 γ δ T cells. MICA/B are highly polymorphic nonclassical MHC cell surface glycoproteins that do not associate with β2 microglobulin, nor do they require TAP for expression (101, 102). When a panel of normal tissues was screened by immunohistochemistry for expression of these proteins, MICA expression was found only on gastrointestinal epithelium of the stomach and large and small intestines. However, MICA/B gene expression could be induced in certain nontransformed cell lines by heat shock or viral infection (101, 103). In contrast, constitutive MICA/B expression has been documented in a high percentage of primary carcinomas of the lung, breast, kidney, ovary, prostate and colon (104), melanomas (105), and hepatocellular carcinomas (106). MICA/B are recognized by an activating receptor on NK cells that is also expressed on most human γ δ T cells and CD8+ αβ T cells (107). This receptor is comprised of two subunits: a ligand-binding NKG2D subunit and either a DAP10 or DAP12 signaling subunit (108). This receptor also reacts with other ligands such as those of the ULBP family that were independently identified as cell surface markers present on transformed cells and cells undergoing stress (109, 110). Tumor cells expressing MICA/B are killed by effector cells with functional NKG2D receptors, and lysis can be inhibited by pretreating the effector cell with blocking NKG2D mAb (107). However, recognition of MICA/B has also been ascribed to the direct binding of the γ δ TCR on Vδ1 γ δ T cells, as the lysis of MICAexpressing target cells by Vδ1 γ δ T cells can be inhibited by a Vδ1 γ δ TCR mAb (111), and soluble MICA tetramers can bind specifically to transfected cells expressing various Vδ1 γ δ TCRs but not NKG2D (112). Thus, γ δ T cells possess two mechanisms to recognize the MIC markers on tumors: one involving a direct interaction with the γ δ TCR and the other mediated by a more globally expressed NKG2D activating receptor. Two data sets link MICA/B recognition to immunosurveillance. First, Groh et al. (111) demonstrated that MIC-expressing cells were recognized and killed by the Vδ1 γ δ T-cell subset, and observed a strong in vivo correlation (p < 0.0001) between MICA/B expression on tumors and tumor infiltration by Vδ1 γ δ T cells (104). Second, recent data demonstrated a correlation between downregulation of NKG2D on tumor-infiltrating lymphocytes (TILs) and the expression of MICA/B in the tumor (113). Compared with NKG2D expression in lymphocytes from patients with MIC− tumors, NKG2D expression was reduced on tumor-infiltrating CD8+ αβ T cells, γ δ T cells, and NK cells and also on peripheral blood mononuclear cells (PBMCs) from individuals with MIC+ tumors. Further analysis revealed a correlation between the presence of soluble MIC proteins in the circulation of 7/14 cancer patients and a downregulated expression of NKG2D on lymphocytes. This downregulation could be recapitulated in vitro. Results of a separate study suggested that shedding of MIC proteins from tumor cell surfaces was the result of the actions of an unknown matrix metalloproteinase (114). These observations thus establish a common mechanism of tumor recognition—and potential

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CANCER IMMUNOEDITING

P1: IKH

343

elimination—by both the innate and adaptive immune systems. The finding that soluble MIC proteins may attenuate the expression/function of NKG2D on host immune cells provides one explanation for how a growing tumor could escape cancer immunosurveillance. Recent work has established the generalizable importance of NKG2D-dependent tumor recognition in murine tumor models as well. In these studies, the overexpression of NKG2D ligands H60 and Rae-1 family members in tumors capable of growing progressively led to their rejection in an NK cell-dependent manner (115–117). In summary, a large amount of data has begun to accumulate indicating that human cancer patients indeed develop immune responses to the tumors that they bear. Although these responses may not always be able to prevent cancer development, they may nevertheless function to restrain tumor growth.

The Presence of Tumor-Infiltrating Lymphocytes Correlates with Patient Survival The third line of evidence that a cancer immunosurveillance process exists in humans comes from a growing body of evidence showing that the presence of tumorinfiltrating lymphocytes (TILs) in a cancer patient’s tumor presages an improved clinical outcome for that individual. Some of the groundbreaking studies that established a strong correlation between patient survival and the presence of TILs involved collectively nearly 900 patients with primary or metastatic melanoma (118–120). The paradigm that was established in these studies has been upheld by several recent studies involving patients with other types of cancer. In a recent analysis, Zhang et al. (121) reported a relationship between the presence of CD3+ TILs and favorable clinical outcomes in patients with advanced ovarian adenocarcinoma. In this study, 186 frozen specimens of stage III or IV ovarian cancers from patients undergoing debulking surgery were assessed by immunostaining for the presence of TILs. Of the 174 tumors that could be evaluated, 102 contained TILs, whereas 72 did not. Patients with TIL-containing tumors had five-year overall survival rates of 38% compared with 4.5% for patients whose tumors lacked TILs. In a subset of 74 of these patients who experienced complete responses to surgical debulking and chemotherapy, the five-year overall survival rate was 73.9% for those with TIL-containing tumors versus 11.9% for patients with tumors that lacked TILs. Furthermore, in a multivariate analysis, it was shown that the presence or absence of TILs and the extent of residual tumor were the only independent prognostic factors of progression-free and overall survival in these patients; other variables such as the type of chemotherapy, histologic type of the tumor, tumor grade, or patient age were not predictive of both rates. Other studies examined the prognostic significance of individual T-cell subsets that infiltrate tumors. Naito et al. (122) found that the extent of CD8+ T cell infiltration specifically into cancer cell nests correlated with the survival of patients with colorectal cancer; 56 patients with no infiltration had five-year survival rates of 50%, whereas the 23 patients showing pronounced CD8+ T cell infiltration into

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

344

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

cancer cell nests had five-year survival rates of 100%. Moreover, multivariate analysis revealed that the presence CD8+ T cells in cancer cell nests was an independent prognostic factor with an impact on patient survival similar to conventional Dukes’ tumor-staging classifications. Analogous findings were made by Schumacher et al. (123) who tracked the clinical course of 70 patients with esophageal squamous cell carcinomas or adenocarcinomas. When histological analysis of the tumor was compared to clinical outcome, the presence of intratumoral lymphocytes correlated with both increased time to disease recurrence and also increased time to death over a five-year period after diagnosis. As in the aforementioned study, a multivariate analysis of the data showed that the presence of intratumoral CD8+ T cells was an independent prognostic factor for survival. Still other studies have shown similar positive correlations between NK cell infiltration and patient survival for gastric carcinoma (124), squamous cell lung carcinoma (125), and colorectal cancer (126). Thus, significant evidence links the presence of TILs to increased survival of cancer patients. Since tumors may attract distinct TIL subsets depending on their tissue of origin (127), it will be important in the future to clarify which particular immune cells are prognostic for each distinct type of cancer. Thus, after a century of controversy, substantial amounts of direct experimental data from mice coupled with correlative data from humans show that innate and adaptive immunity function together to protect the host against neoplastic disease and thereby converge on the original conviction of Burnet and Thomas: immunosurveillance exists.

IMMUNOLOGIC SCULPTING DURING TUMOR DEVELOPMENT Despite strong evidence supporting the existence of a functional cancer immunosurveillance process, immunocompetent individuals still develop cancer. This clinical reality may be explained by the existence of an immune process that facilitates the outgrowth of tumors with reduced immunogenicity that have a better chance of surviving in an immunocompetent host. Recent work from several laboratories now supports this hypothesis. Our laboratory used tumor transplantation approaches to assess the immunogenic characteristics of a large number of primary MCA-induced sarcomas generated in the presence or absence of a functional immune system (5). Tumor cells from either wild-type or RAG-2−/− mice grew progressively with similar kinetics when transplanted into RAG-2−/− recipients, indicating that there were no inherent growth differences between tumors generated in the presence or absence of an intact immune system (Figure 2A,B). Moreover, tumor cells derived from wildtype mice grew progressively when transplanted into na¨ıve immunocompetent 129/SvEv hosts (Figure 2C). In contrast, 8/20 of the tumors originally generated in RAG-2−/− mice were rejected when transplanted into immunocompetent hosts,

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CANCER IMMUNOEDITING

P1: IKH

345

Figure 2 Increased immunogenicity of tumours derived from MCA-treated RAG2−/− mice. Immunodeficient RAG-2−/− hosts were injected with a dose of 105 tumor cells derived from wild-type 129/SvEv mice (A) or RAG-2−/− mice (B). Tumor growth is plotted as mean tumor diameter of 3–5 mice inoculated with each tumor. Groups of 5–8 immunocompetent 129/SvEv × RAG-2−/− F1 mice were injected on day 0 with doses of 106 tumor cells derived from 17 individual 129/SvEv mice (C) or 20 individual RAG-2−/− mice (D) and tumor growth was monitored as above. In (D), the dashed lines denote tumors that grew progressively, whereas solid lines represent tumors that were rejected. Data from Shankaran et al. (5).

even when injected at high cell number (Figure 2D). Thus, tumors formed in the absence of an intact immune system are, as a group, more immunogenic than tumors that arise in immunocompetent hosts. Experiments performed in other laboratories have led to similar conclusions. MCA sarcomas derived from nude (128) or SCID mice (129) were rejected more frequently than similar tumors derived from wild-type mice when transplanted into wild-type hosts. In addition, two MCA-induced sarcomas derived from TCR Jα281−/− mice grew more slowly when transplanted into wild-type hosts than did sarcomas originally isolated from wild-type mice (36). In contrast, these tumors grew in a comparable manner when transplanted into Jα281−/− recipients. It was

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

346

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

also shown that lymphomas derived from pfp−/− mice grew avidly when transplanted into pfp−/− mice but most were rejected when transplanted into wild-type mice (33). Finally, results from one recent study suggest that immunocompetent mice select tumors that are less sensitive to TRAIL-mediated cytotoxicity (60). MCAinduced fibrosarcomas produced in C57BL/6 strain wild-type or p53+/− mice that were treated with either control IgG, a neutralizing monoclonal TRAIL-specific antibody, or an antibody specific for asialo-GM1 throughout tumor development were tested for susceptibility to TRAIL-mediated killing in vitro. Although only 1/6 tumors derived from control IgG-treated wild-type mice and 1/8 tumors derived from control IgG-treated p53+/− mice were lysed by TRAIL, 5/6 tumors generated in antiasialo-GM1-treated wild-type mice and 5/8 tumors derived from anti-TRAIL-treated p53+/− mice displayed susceptibility to TRAIL killing. Taken together, these results show that tumors are imprinted by the immunologic environment in which they form. By eliminating tumor cells of high intrinsic immunogenicity, this imprinting process may select for tumor cell variants of reduced immunogenicity and therefore favor the generation of tumors that are either poorly recognized by the immune system or that have acquired mechanisms that suppress immune effector functions. In this manner, the immunologic sculpting of developing tumor cells provides them with mechanisms to resist the extrinsic tumor-suppressor actions of the immune system. While the shaping of tumor immunogenicity most likely occurs continuously during tumor development, the major effects of this process probably occur early when the tumor is perhaps histologically—but not clinically—detectable. It follows, then, that the immunogenicity of most clinically apparent tumors has already been attenuated to some degree by the sculpting hand of immunity.

CANCER IMMUNOEDITING: REFINING CANCER IMMUNOSURVEILLANCE Based on the studies summarized in this review, the term “cancer immunosurveillance” no longer suffices to accurately describe the complex interactions that occur between a developing tumor and the immune system of the host. As originally conceived, cancer immunosurveillance was thought to be a host-protective function carried out by the adaptive immune system only at the earliest stages of cellular transformation. In contrast, we now recognize that both the innate and adaptive immune compartments participate in the process and serve not only to protect the host from tumor development but also to sculpt, or edit, the immunogenicity of tumors that may eventually form. Therefore, we have proposed the use of the broader term “cancer immunoediting” to more appropriately emphasize the dual roles of immunity in not only preventing but also shaping neoplastic disease (5, 6). Cancer immunoediting thus represents a refinement of the original cancer immunosurveillance hypothesis but is more comprehensive in its scope. As such,

15 Mar 2004

15:20

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

CANCER IMMUNOEDITING

P1: IKH

347

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

we envisage that the cancer immunoediting process is comprised of three phases that we have termed the “three Es of cancer immunoediting:” elimination, equilibrium, and escape. In the following sections, we discuss each of these three phases in more detail (Figure 3). Specifically, we attempt to integrate our enhanced and evolving understanding of immune system-tumor interactions with ongoing work in classical tumor biology. Our intention is not to be dogmatic but rather to present a testable model that will stimulate further work in defining the molecular and cellular basis of each of the three phases of cancer immunoediting.

Elimination The elimination phase represents the original concept of cancer immunosurveillance (Figure 3A; Figure 4). If this phase successfully eradicates the developing tumor, it represents the complete immunoediting process without progression to the subsequent phases. The immune components that participate in the elimination phase are now being identified but their precise roles need to be further clarified. As an extrinsic tumor suppressor, we envisage that the immune system manifests its effects only after transformed cells have circumvented their intrinsic tumorsuppressor mechanisms (130). Immunologic rejection of a developing tumor, as in host defense to microbial pathogens, likely requires an integrated response involving both the innate and adaptive arms of the immune system (27). Initiation of the antitumor immune response (Figure 4A) occurs when cells of the innate immune system become alerted to the presence of a growing tumor, at least in part owing to the local tissue disruption that occurs as a result of the stromal remodeling processes integral to the basic physiology of solid tumor development. This stromal remodeling could result from two of the six “hallmarks of cancer” (131): angiogenesis (132, 133) and tissue-invasive growth (134). The stromal remodeling induced during these processes could produce proinflammatory molecules that, together with chemokines that may be produced by the tumor cells themselves (135), summon cells of the innate immune system to this new source of local “danger” (136, 137). Once recruited to the developing tumor mass, NKT cells, γ δ T cells, NK cells, and/or macrophages may recognize molecules, such as the ligands for NKG2D previously discussed, that have been induced on tumor cells either by the incipient inflammation or the cellular transformation process itself. In addition, γ δ T cells and NKT cells may recognize developing tumors via TCR interaction with either NKG2D ligands or glycolipid-CD1 complexes expressed on tumor cells, respectively (138). Regardless of the precise mechanism of recognition, these events lead to a common outcome that is critical for progression of the antitumor response—the production of IFN-γ . In the second step (Figure 4B), the effects of innate immune recognition of the tumor are amplified. The initial amount of IFN-γ released at the tumor site induces the local production of chemokines that recruit more cells of the innate immune system to the tumor. Products generated during remodeling of the extracellular matrix may induce tumor-infiltrating macrophages to produce low amounts of

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

348

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

IL-12 (139) that stimulate tumor-infiltrating NK cells to produce low amounts of IFN-γ , which in turn activate macrophages in the tumor to produce more IL-12, leading to increased IFN-γ production by NK cells. In addition to this positive feedback system (140), the binding of NK cell–activating receptors to their cognate ligands on tumor cells stimulates even more NK cell IFN-γ production (115) that can now activate a number of IFN-γ -dependent processes—including antiproliferative (43), proapoptotic (141), and angiostatic (47, 52, 56) effects—that result in the killing of a proportion of the tumor. In addition, macrophages activated by IFN-γ that express tumoricidal products such as reactive oxygen and reactive nitrogen intermediates (142–144) and NK cells activated either by IFN-γ or via engagement of their activating receptors can kill tumor cells via TRAIL- (145, 146) or perforin-dependent (147) mechanisms, respectively. As a result of these processes, a source of tumor antigens from dead tumor cells becomes available and the adaptive immune system is recruited into the process. In the third step (Figure 4C), tumor antigens liberated by the effects of innate immunity on the tumor drive the development of tumor-specific adaptive immune responses. Immature dendritic cells (DCs) that have been recruited to the tumor site become activated either by exposure to the cytokine milieu created during the ongoing attack on the tumor by innate immunity or by interacting with tumor-infiltrating NK cells (148). The activated DCs can acquire tumor antigens directly by ingestion of tumor cell debris or potentially through indirect mechanisms involving transfer of tumor cell–derived heat shock protein/tumor antigen complexes to DCs (149, 150). Activated, antigen-bearing mature DCs then migrate to the draining lymph node (151), where they induce the activation of na¨ıve tumor-specific Th1 CD4+ T cells. Th1 cells facilitate the development of tumor-specific CD8+ CTL induced via cross-presentation of antigenic tumor peptides on DC MHC class I molecules (152–155). In the fourth step (Figure 4D), the development of tumor-specific adaptive immunity provides the host with a capacity to completely eliminate the developing tumor. Tumor-specific CD4+ and CD8+ T cells home to the tumor site, where they participate in the killing of antigen-positive tumor cells. CD4+ T cells produce IL-2 that, together with host cell production of IL-15, helps to maintain the function and viability of the tumor-specific CD8+ T cells. Tumor-specific CD8+ T cells will efficiently recognize their tumor targets [owing to the enhanced immunogenicity of tumor cells that have been exposed to the IFN-γ produced in steps 1 and 2 (5)] and will induce tumor cell death by both direct and indirect mechanisms. It is likely that these CD8+ T cells directly kill many of the tumor cells in vivo. However, these cells will also produce large amounts of IFN-γ following interaction with their tumor targets and thus should also induce tumor cell cytostasis and killing by the IFN-γ -dependent mechanisms of cell cycle inhibition, apoptosis, angiostasis, and induction of macrophage tumoricidal activity. These two scenarios are not mutually exclusive and most likely occur concomitantly; however, their relative contributions may vary among different tumors. Thus, the elimination phase of cancer immunoediting is a continuous process that must be repeated each time antigenically

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

CANCER IMMUNOEDITING

P1: IKH

349

distinct neoplastic cells arise. For this reason, it is particularly noteworthy that cancer is more prevalent in aged populations where immune system function, and therefore cancer immunosurveillance, begins to decline.

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Equilibrium In the equilibrium phase (Figure 3B), the host immune system and any tumor cell variant that has survived the elimination phase enter into a dynamic equilibrium, wherein lymphocytes and IFN-γ exert potent and relentless selection pressure on the tumor cells that is enough to contain, but not fully extinguish, a tumor bed containing many genetically unstable and mutating tumor cells. We envision this period to be a crucible of Darwinian selection: Although many of the original tumor cell escape variants are destroyed, new variants arise carrying different mutations that provide them with increased resistance to immune attack. The end result of the equilibrium process is a new population of tumor clones with reduced immunogenicity, hewn from a heterogeneous parental population by the sculpting forces of the immune system. Equilibrium is probably the longest of the three phases and may occur over a period of many years in humans. Indeed, it has been estimated that for many solid human tumors there can be a 20-year interval between initial carcinogen exposure and clinical detection of the tumor (156). During this period, the heterogeneity and genetic instability of cancer cells that survive the elimination phase are possibly the principal forces that enable tumor cells to eventually resist the host’s immunological siege. It has been proposed that the “mutator phenotype” of tumor cells (157) may result from the three types of genetic instability observed in cancer: nucleotide-excision repair instability (NIN), microsatellite instability (MIN), and chromosomal instability (CIN) (158). Of the three, CIN is thought to be the predominant mechanism responsible for destabilizing genomic integrity, and the observation that cancer cell genomes display gains or losses of whole chromosomes (i.e., aneuploidy) associated with an estimated loss of 25%–50% of their alleles reflects the degree of genomic upheaval associated with the CIN phenotype (158). Clearly, genomic instability has the potential to spawn tumor variants of reduced immunogenicity, and some of these will display an enhanced capacity to grow in an unfettered immune selecting environment. A complete mechanistic understanding of the equilibrium phase will require the development of new tumor models to better define the cell-intrinsic mechanisms that generate new tumor phenotypes and to identify the tumor-sculpting immune “editors.” One clinical scenario that may illustrate the equilibrium phase in humans is the transmission of cancer from transplant donors to recipients. In these cases, transplanted organs are grossly normal and cancer-free at the time of harvest. While some donors are subsequently found to harbor disease in other anatomic sites, other transplant donors either have no clinical history of cancer or have been in durable remission from cancer prior to transplantation. Recently, Mackie et al. (159) reported the occurrence of metastatic melanoma 1–2 years post-transplant

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

350

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

in two allograft recipients who had each received kidneys from the same donor. Upon subsequent analysis, it was found that the donor had been treated for primary melanoma 16 years before her kidneys were donated but was considered tumorfree at the time of her death. Two other case reports described the appearance of donor-derived melanoma in two renal transplant patients and one liver transplant recipient less than one year after these organs were transplanted from a donor with no known history of malignancy (160, 161). These observations, together with others that appear in the clinical literature (70, 162), suggest that the pharmacologic suppression of the immune systems of these transplant recipients facilitated the rapid and progressive outgrowth of occult tumors that had previously been maintained in the equilibrium phase by the donor’s competent immune system.

Escape In the escape phase (Figure 3C), tumor cell variants selected in the equilibrium phase now can grow in an immunologically intact environment. This breach of the host’s immune defenses most likely occurs when genetic and epigenetic changes in the tumor cell confer resistance to immune detection and/or elimination, allowing the tumors to expand and become clinically detectable. Because both the adaptive and innate compartments of the immune system function in the cancer immunosurveillance network, tumors most likely would have to circumvent either one or both arms of immunity in order to achieve progressive growth. Individual tumor cells may employ multiple immunoevasive strategies to elude the powerful integrated innate and adaptive antitumor immune responses to their immunogenic progenitors. Thus, it is likely that several distinct immunologically driven tumor sculpting events must occur before the final immunogenic phenotype of a malignant cell is ultimately established. Much work has recently focused on defining the molecular bases of tumor escape. It is now recognized that tumors can either directly or indirectly impede the development of antitumor immune responses either through the elaboration of immunosuppressive cytokines (such as TGF-β and IL-10) or via mechanisms involving T cells with immunosuppressive activities (i.e., regulatory T cells). Because the mechanisms that target the immune system to achieve tumor escape have been the subject of recent review articles (163, 164), they are not discussed further. Tumor escape can also result from changes that occur directly at the level of the tumor. These can include alterations that affect tumor recognition by immune effector cells [such as loss of antigen expression, loss of MHC components (165), shedding of NKG2D ligands (113), and development of IFN-γ insensitivity (31)] or provide tumors with mechanisms to escape immune destruction [such as defects in death-receptor signaling pathways (60) or expression of antiapoptotic signals such as those induced by constitutively active STAT3 (166)]. Two of these mechanisms, dysregulation of MHC class I processing and presentation and development of IFN-γ insensitivity in tumor cells, would allow tumors to escape from the events discussed in the elimination phase of the cancer immunoediting process and have,

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CANCER IMMUNOEDITING

P1: IKH

351

in fact, been identified in tumor cells. Analysis of large banks of human tumor specimens has shown that between 40%–90% of human tumors display total or selective allelic losses of HLA class I proteins (165, 167). Moreover, other components of this pathway, including TAP1 and the immunoproteasome subunits LMP2 and 7, are also frequently deficient in human tumors (168). The physiologic relevance of LMP2 deficiency in cancer is evidenced by the observation that LMP2−/− mice are more prone to the development of uterine neoplasms than their wild-type counterparts (169). IFN-γ receptor signaling dysfunction represents another potential mechanism of tumor immune escape. In one study, 4/17 (25%) human lung adenocarcinoma cell lines were found to be completely unresponsive to IFN-γ (31). The unresponsive state in these tumors was found to be caused by either the absence or abnormal function of distinct components of the IFN-γ receptor signaling pathway. In addition, unpublished work has shown that 15%–30% of primary MCA-induced fibrosarcomas derived from 129/SvEv mice display IFN-γ insensitivity (G.P. Dunn & R.D. Schreiber, unpublished observations). Clearly, identifying additional escape mechanisms will yield critical insights into how tumor cell immunogenicities are edited by the immune system.

CONCLUSIONS AND IMPLICATIONS In this review, we have summarized some of the salient data supporting the existence and physiologic relevance of a cancer immunoediting process. The recent development of sophisticated tumor models using genetically altered mice and function-blocking monoclonal antibodies has made possible the critical experiments that not only resolved the long-standing controversy surrounding the cancer immunosurveillance hypothesis of Burnet and Thomas but also led to its refinement into the cancer immunoediting hypothesis (5, 6). The continued clarification of the three Es of cancer immunoediting has important implications for cancer immunotherapy in humans. By gaining an improved understanding of the cellular and molecular processes that lead to immunologic tumor rejection in the elimination phase, it will be possible to identify which immune forces need to be augmented to facilitate natural protection against tumors of different tissue origins. By studying the equilibrium phase, it will be possible to understand the genetic processes that lead to development of tumors with reduced immunogenicities and identify the molecular targets of the cancer immunoediting process in order to gain insight into how tumor sculpting can be prevented by stabilizing tumor cell genomes. Finally, by elucidating how tumors escape immune detection and elimination, it will be possible to develop methods to determine the extent to which a tumor has been edited and devise molecular strategies to reverse these cloaking mechanisms and thus unmask tumor immunogenicity. In recent years, there has been a paradigm shift in how cancer is viewed. Rather than emphasizing the differences in the greater than 100 types of cancer, researchers have begun to consider the similarities between these seemingly disparate

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

352

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

malignancies (170). Hanahan &Weinberg have codified this view by proposing that cancer cells must acquire six enabling characteristics in order to form progressively growing tumors (131). Specifically, they must be able to grow autonomously, develop insensitivity to negative growth regulation, evade intrinsic apoptotic signals, display unlimited replicative potential, develop the capacity for angiogenesis, and develop competence for invasive growth and metastasis. In the current review, we have provided strong evidence that supports the existence of the seventh “hallmark of cancer:” the capacity of a malignant cell to evade the extrinsic tumor suppressor functions of the immune system. Moreover, we have discussed the possibility that this seventh hallmark is a result of a cancer immunoediting process, wherein the malignant cell’s immunogenic phenotype—forged by its interaction with the host immune system—may determine its fitness for continued survival and growth in an immunocompetent environment. We hope a generalized recognition of this new hallmark of cancer will stimulate new efforts to elucidate the pivotal events of cancer immunoediting so that the long history of thinking on the immune system and cancer will have as its d´enouement the enhanced understanding and treatment of neoplastic disease.

ACKNOWLEDGMENTS The authors are grateful to members of their laboratories, past and present, who contributed to the work from our groups that was quoted in this review. We are also particularly grateful to Vijay Shankaran, Hiroaki Ikeda, Allen Bruce, Kathleen Sheehan, Catherine Koebel, Jack Bui, Ravindra Uppaluri, Ruby Chan, and Mark Diamond for their contributions to the development of the cancer immunoediting hypothesis and for their helpful comments during the preparation of this review. The work from the authors was supported by grants from the National Cancer Institute, the Ludwig Institute for Cancer Research, and the Cancer Research Institute. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Ehrlich P. 1909. Ueber den jetzigen Stand der Karzinomforschung. Ned. Tijdschr. Geneeskd. 5(Part 1):273–90 2. Thomas L. 1959. Discussion. In Cellular and Humoral Aspects of the Hypersensitive States, ed. HS Lawrence, pp. 529–32. New York: Hoeber-Harper 3. Burnet FM. 1970. The concept of immunological surveillance. Prog. Exp. Tumor Res. 13:1–27

4. Stutman O. 1974. Tumor development after 3-methylcholanthrene in immunologically deficient athymic-nude mice. Science 183:534–36 5. Shankaran V, Ikeda H, Bruce AT, White JM, Swanson PE, et al. 2001. IFNγ lymphocytes prevent primary tumour development and shape tumour immunogenicity. Nature 410:1107–11 6. Dunn GP, Bruce AT, Ikeda H, Old LJ,

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

CANCER IMMUNOEDITING

7. 8.

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

9.

10.

11.

12.

13.

14.

15. 16.

17. 18.

19.

20.

Schreiber RD. 2002. Cancer immunoediting: from immunosurveillance to tumor escape. Nat. Immunol. 3:991–98 Burnet FM. 1971. Immunological surveillance in neoplasia. Transplant Rev. 7:3–25 Burnet FM. 1957. Cancer—a biological approach. Br. Med. J. 1:841–47 Burnet FM. 1964. Immunological factors in the process of carcinogenesis. Br. Med. Bull. 20:154–58 Medawar P. 1944. The behavior and fate of skin autografts and skin homografts in rabbits. J. Anat. 78:176–99 Medawar P. 1945. A second study of the behaviour and fate of skin homografts in rabbits. J. Anat. 79:157–76 Foley EJ. 1953. Antigenic properties of methylcholanthrene-induced tumors in mice of the strain of origin. Cancer Res. 13:835–37 Prehn RT, Main JM. 1957. Immunity to methylcholanthrene-induced sarcomas. J. Natl. Cancer Inst. 18:769–78 Old LJ, Boyse EA. 1966. Specific antigens of tumors and leukemias of experimental animals. Med. Clin. North Am. 50:901–12 Klein G. 1966. Tumor antigens. Annu. Rev. Microbiol. 20:223–52 Flanagan SP. 1966. ‘Nude’, a new hairless gene with pleiotropic effects in the mouse. Genet. Res. 8:295–309 Pantelouris EM. 1968. Absence of thymus in a mouse mutant. Nature 217:370–71 Stutman O. 1979. Chemical carcinogenesis in nude mice: comparison between nude mice from homozygous and heterzygous matings and effect of age and carcinogen dose. J. Natl. Cancer Inst. 2:353– 58 Stutman O. 1978. Spontaneous, viral, and chemically induced tumors in the nude mouse. In The Nude Mouse in Experimental and Clinical Research., ed. J Fogh, BC Giovanella, pp. 411–35. New York: Academic Stutman O. 1973. Tumor development in immunologically deficient nude mice after exposure to chemical carcinogens.

21.

22.

23.

24.

25.

26.

27.

28. 29.

30.

31.

P1: IKH

353

Proc. Int. Workshop Nude Mice. Scanticon, Aarhus, Denmark, Oct. 11–13, 1973. 1:257–64 Rygaard J, Povlsen CO. 974. Is immunological surveillance not a cell-mediated immune function? Transplantation 17: 135–36 Rygaard J, Povlsen CO. 1974. The mouse mutant nude does not develop spontaneous tumours. An argument against immunological surveillance. Acta Pathol. Microbiol. Scand. Sect. B 82:99– 106 Maleckar JR, Sherman LA. 1987. The composition of the T cell receptor repertoire in nude mice. J. Immunol. 138:3873– 76 Ikehara S, Pahwa RN, Fernandes G, Hansen CT, Good RA. 1984. Functional T cells in athymic nude mice. Proc. Natl. Acad. Sci. USA 81:886–88 Hunig T. 1983. T-cell function and specificity in athymic mice. Immunol. Today 4:84–87 Herberman RB, Holden HT. 1978. Natural cell-mediated immunity. Adv. Cancer Res. 27:305–77 Janeway CA, Jr. 1989. Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harbor Symp. Quant. Biol. 54(Part 1):1–13 Heidelberger C. 1975. Chemical carcinogenesis. Annu. Rev. Biochem. 44:79–121 Kouri RE, Nebert DW. 1977. Genetic regulation of susceptibility to polycyclichydrocarbon-induced tumors in the mouse. In Origins of Human Cancer, ed. HH Hiatt, JD Watson, JA Winsten, pp. 811–35. Cold Spring Harbor, NY: Cold Spring Harbor Lab. Press Dighe AS, Richards E, Old LJ, Schreiber RD. 1994. Enhanced in vivo growth resistance to rejection of tumor cells expressing dominant negative IFNγ receptors. Immunity 1:447–56 Kaplan DH, Shankaran V, Dighe AS, Stockert E, Aguet M, et al. 1998. Demonstration of an interferon γ -dependent

11 Feb 2004

18:54

354

32.

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

33.

34.

35.

36.

37.

38.

39.

40.

41.

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

tumor surveillance system in immunocompetent mice. Proc. Natl. Acad. Sci. USA 95:7556–61 Street SE, Cretney E, Smyth MJ. 2001. Perforin and interferon-γ activities independently control tumor initiation, growth, and metastasis. Blood 97:192–97 Street SE, Trapani JA, MacGregor D, Smyth MJ. 2002. Suppression of lymphoma and epithelial malignancies effected by interferon γ . J. Exp. Med. 196: 129–34 Bach EA, Aguet M, Schreiber RD. 1997. The IFNγ receptor: a paradigm for cytokine receptor signaling. Annu. Rev. Immunol. 15:563–91 van den Broek MF, Kagi D, Ossendorp F, Toes R, Vamvakas S, et al. 1996. Decreased tumor surveillance in perforindeficient mice. J. Exp. Med. 184:1781–90 Smyth MJ, Thia KY, Street SE, Cretney E, Trapani JA, et al. 2000. Differential tumor surveillance by natural killer (NK) and NKT cells. J. Exp. Med. 191:661–68 Smyth MJ, Thia KY, Street SE, MacGregor D, Godfrey DI, Trapani JA. 2000. Perforin-mediated cytotoxicity is critical for surveillance of spontaneous lymphoma. J. Exp. Med. 192:755–60 Russell JH, Ley TJ. 2002. Lymphocytemediated cytotoxicity. Annu. Rev. Immunol. 20:323–70 Shinkai Y, Rathbun G, Lam KP, Oltz EM, Stewart V, et al. 1992. RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell 68:855–67 Smyth MJ, Crowe NY, Godfrey DI. 2001. NK cells and NKT cells collaborate in host protection from methylcholanthreneinduced fibrosarcoma. Int. Immunol. 13: 459–63 Dighe AS, Farrar MA, Schreiber RD. 1993. Inhibition of cellular responsiveness to interferon-γ (IFNγ ) induced by overexpression of inactive forms of the IFNγ receptor. J. Biol. Chem. 268:10645– 53

42. Chin YE, Kitagawa M, Su WS, You Z, Iwamoto Y, Fu X. 1996. Cell growth arrest and induction of cyclin-dependent kinase inhibitor p21 mediated by Stat1. Science 272:719–22 43. Bromberg JF, Horvath CM, Wen Z, Schreiber RD, Darnell JE Jr. 1996. Transcriptionally active Stat1 is required for the antiproliferative effects of both interferon α and interferon γ . Proc. Natl. Acad. Sci. USA 93:7673–78 44. Chin YE, Kitagawa M, Kuida K, Flavell RA, Fu XY. 1997. Activation of the STAT signaling pathway can cause expression of caspase 1 and apoptosis. Mol. Cell. Biol. 17:5328–37 45. Detjen KM, Farwig K, Welzel M, Wiedenmann B, Rosewicz S. 2001. Interferon-γ inhibits growth of human pancreatic carcinoma cells via caspase-1 dependent induction of apoptosis. Gut 49:251–62 46. Xu X, Fu XY, Plate J, Chong AS. 1998. IFNγ induces cell growth inhibition by Fas-mediated apoptosis: requirement of STAT1 protein for up-regulation of Fas and FasL expression. Cancer Res. 58:2832–37 47. Luster AD, Leder P. 1993. IP-10, a –C– X–C– chemokine, elicits a potent thymusdependent antitumor response in vivo. J. Exp. Med. 178:1057–65 48. Voest EE, Kenyon BM, O’Reilly MS, Truitt G, D’Amato RJ, Folkman J. 1995. Inhibition of angiogenesis in vivo by interleukin 12. J. Natl. Cancer Inst. 87:581–86 49. Strieter RM, Kunkel SL, Arenberg DA, Burdick MD, Polverini PJ. 1995. Interferon γ -inducible protein 10 (IP-10), a member of the C–X–C chemokine family, is an inhibitor of angiogenesis. Biochem. Biophys. Res. Commun. 210:51–57 50. Sgadari C, Angiolillo AL, Cherney BW, Pike SE, Farber JM, et al. 1996. Interferon-inducible protein-10 identified as a mediator of tumor necrosis in vivo. Proc. Natl. Acad. Sci. USA 93:13791–96 51. Sgadari C, Farber JM, Angiolillo AL, Liao F, Teruya-Feldstein J, et al. 1997. Mig,

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

CANCER IMMUNOEDITING

52.

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

53.

54.

55.

56.

57.

58.

59.

60.

61.

the monokine induced by interferon-γ , promotes tumor necrosis in vivo. Blood 89:2635–43 Coughlin CM, Salhany KE, Gee MS, LaTemple DC, Kotenko S, et al. 1998. Tumor cell responses to IFNγ affect tumorigenicity and response to IL-12 therapy and antiangiogenesis. Immunity 9:25– 34 Fallarino F, Gajewski TF. 1999. Cutting edge: differentiation of antitumor CTL in vivo requires host expression of Stat1. J. Immunol. 163:4109–13 Kacha AK, Fallarino F, Markiewicz MA, Gajewski TF. 2000. Cutting edge: spontaneous rejection of poorly immunogenic P1.HTR tumors by Stat6-deficient mice. J. Immunol. 165:6024–28 Enzler T, Gillessen S, Manis JP, Ferguson D, Fleming J, et al. 2003. Deficiencies of GM-CSF and interferon γ link inflammation and cancer. J. Exp. Med. 197:1213–19 Qin Z, Blankenstein T. 2000. CD4+ T cell-mediated tumor rejection involves inhibition of angiogenesis that is dependent on IFNγ receptor expression by nonhematopoietic cells. Immunity 12:677–86 Ibe S, Qin Z, Schuler T, Preiss S, Blankenstein T. 2001. Tumor rejection by disturbing tumor stroma cell interactions. J. Exp. Med. 194:1549–59 Girardi M, Oppenheim DE, Steele CR, Lewis JM, Glusac E, et al. 2001. Regulation of cutaneous malignancy by γ δ T cells. Science 294:605–9 Smyth MJ, Takeda K, Hayakawa Y, Peschon JJ, van den Brink MR, Yagita H. 2003. Nature’s TRAIL—on a path to cancer immunotherapy. Immunity 18:1–6 Takeda K, Smyth MJ, Cretney E, Hayakawa Y, Kayagaki N, et al. 2002. Critical role for tumor necrosis factorrelated apoptosis-inducing ligand in immune surveillance against tumor development. J. Exp. Med. 195:161–69 Cretney E, Takeda K, Yagita H, Glaccum M, Peschon JJ, Smyth MJ. 2002. Increased susceptibility to tumor initiation

62.

63.

64.

65.

66. 67.

68.

69.

70.

71.

72. 73. 74.

P1: IKH

355

and metastasis in TNF-related apoptosisinducing ligand-deficient mice. J. Immunol. 168:1356–61 Noguchi Y, Jungbluth A, Richards E, Old LJ. 1996. Effect of interleukin 12 on tumor induction by 3-methylcholanthrene. Proc. Natl. Acad. Sci. USA 93:11798–801 Hayakawa Y, Rovero S, Forni G, Smyth MJ. 2003. α-Galactosylceramide (KRN7000) suppression of chemicaland oncogene-dependent carcinogenesis. Proc. Natl. Acad. Sci. USA http:// www.pnas.org/cgi/doi/10.1073/pnas.1630 663100 Penn I. 1970. Malignant Tumors in Organ Transplant Recipients. New York: Springer-Verlag. 51 pp. Gatti RA, Good RA. 1971. Occurrence of malignancy in immunodeficiency diseases. A literature review. Cancer 28:89– 98 Boshoff C, Weiss R. 2002. AIDS-related malignancies. Nat. Rev. Cancer 2:373–82 Birkeland SA, Storm HH, Lamm LU, Barlow L, Blohme I, et al. 1995. Cancer risk after renal transplantation in the Nordic countries, 1964–1986. Int. J. Cancer 60:183–89 Sheil AGR. 2001. Cancer in dialysis and transplant patients. In Kidney Transplantation, ed. PJ Morris, pp. 558–70. Philadelphia: W.B. Saunders Pham SM, Kormos RL, Landreneau RJ, Kawai A, Gonzalez-Cancel I, et al. 1995. Solid tumors after heart transplantation: lethality of lung cancer. Ann. Thorac. Surg. 60:1623–26 Penn I. 1996. Malignant melanoma in organ allograft recipients. Transplantation 61:274–78 Hoover RN. 1977. Effects of drugs— immunosuppression. See Ref. 29, pp. 369–79 Sheil AG. 1986. Cancer after transplantation. World J. Surg. 10:389–96 Penn I. 1995. Sarcomas in organ allograft recipients. Transplantation 60:1485–91 Old LJ. 1981. Cancer immunology: the

11 Feb 2004

18:54

356

75.

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

76.

77.

78.

79.

80.

81.

82.

83.

84.

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

search for specificity—G.H.A. Clowes memorial lecture. Cancer Res. 41:361–75 Carey TE, Takahashi T, Resnick LA, Oettgen HF, Old LJ. 1976. Cell surface antigens of human malignant melanoma: mixed hemadsorption assays for humoral immunity to cultured autologous melanoma cells. Proc. Natl. Acad. Sci. USA 73:3278–82 Ueda R, Shiku H, Pfreundschuh M, Takahashi T, Li LT, et al. 1979. Cell surface antigens of human renal cancer defined by autologous typing. J. Exp. Med. 150:564– 79 Knuth A, Danowski B, Oettgen HF, Old LJ. 1984. T-cell-mediated cytotoxicity against autologous malignant melanoma: analysis with interleukin 2-dependent Tcell cultures. Proc. Natl. Acad. Sci. USA 81:3511–15 Traversari C, van der Bruggen P, Van den Eynde B, Hainaut P, Lemoine C, et al. 1992. Transfection expression of a gene coding for a human melanoma antigen recognized by autologous cytolytic T lymphocytes. Immunogenetics 35:145– 52 van der Bruggen P, Traversari C, Chomez P, Lurquin C, De Plaen E, et al. 1991. A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma. Science 254:1643–47 Sahin U, Tureci O, Schmitt H, Cochlovius B, Johannes T, et al. 1995. Human neoplasms elicit multiple specific immune responses in the autologous host. Proc. Natl. Acad. Sci. USA 92:11810–13 Wang RF, Rosenberg SA. 1999. Human tumor antigens for cancer vaccine development. Immunol. Rev. 170:85–100 Boon T, van der Bruggen P. 1996. Human tumor antigens recognized by T lymphocytes. J. Exp. Med. 183:725–29 Old LJ, Chen YT. 1998. New paths in human cancer serology. J. Exp. Med. 187:1163–67 Rosenberg SA. 1999. A new era for cancer immunotherapy based on the genes

85.

86.

87.

88.

89.

90.

91.

92.

93.

that encode cancer antigens. Immunity 10:281–87 Scanlan MJ, Gure AO, Jungbluth AA, Old LJ, Chen YT. 2002. Cancer/testis antigens: an expanding family of targets for cancer immunotherapy. Immunol. Rev. 188:22–32 Chen YT, Scanlan MJ, Sahin U, Tureci O, Gure AO, et al. 1997. A testicular antigen aberrantly expressed in human cancers detected by autologous antibody screening. Proc. Natl. Acad. Sci. USA 94:1914–18 Jungbluth AA, Chen YT, Stockert E, Busam KJ, Kolb D, et al. 2001. Immunohistochemical analysis of NY-ESO1 antigen expression in normal malignant human tissues. Int. J. Cancer 92:856–60 Stockert E, Jager E, Chen YT, Scanlan MJ, Gout I, et al. 1998. A survey of the humoral immune response of cancer patients to a panel of human tumor antigens. J. Exp. Med. 187:1349–54 Jager E, Stockert E, Zidianakis Z, Chen YT, Karbach J, et al. 1999. Humoral immune responses of cancer patients against “Cancer-Testis” antigen NY-ESO-1: correlation with clinical events. Int. J. Cancer 84:506–10 Jager E, Nagata Y, Gnjatic S, Wada H, Stockert E, et al. 2000. Monitoring CD8 T cell responses to NY-ESO-1: correlation of humoral cellular immune responses. Proc. Natl. Acad. Sci. USA 97:4760–65 Zeng G, Touloukian CE, Wang X, Restifo NP, Rosenberg SA, Wang RF. 2000. Identification of CD4+ T cell epitopes from NY-ESO-1 presented by HLA-DR molecules. J. Immunol. 165:1153–59 Gnjatic S, Atanackovic D, Jager E, Matsuo M, Selvakumar A, et al. 2003. Survey of naturally occurring CD4+ T cell responses against NY-ESO-1 in cancer patients: correlation with antibody responses. Proc. Natl. Acad. Sci. USA 100:8862–67 Posner JB, Furneaux HM. 1990. Paraneoplastic syndromes. In Immunologic Mechanisms in Neurologic and Psychiatric

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

CANCER IMMUNOEDITING

94.

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

95.

96.

97.

98.

99.

100.

101.

102. 103.

Disease, ed. BH Waksman, pp. 187–219. New York: Raven Darnell RB. 1996. Onconeural antigens and the paraneoplastic neurologic disorders: at the intersection of cancer, immunity, and the brain. Proc. Natl. Acad. Sci. USA 93:4529–36 Musunuru K, Darnell RB. 2001. Paraneoplastic neurologic disease antigens: RNAbinding proteins and signaling proteins in neuronal degeneration. Annu. Rev. Neurosci. 24:239–62 Albert ML, Darnell JC, Bender A, Francisco LM, Bhardwaj N, Darnell RB. 1998. Tumor-specific killer cells in paraneoplastic cerebellar degeneration. Nat. Med. 4:1321–24 Albert ML, Austin LM, Darnell RB. 2000. Detection and treatment of activated T cells in the cerebrospinal fluid of patients with paraneoplastic cerebellar degeneration. Ann. Neurol. 47:9–17 Dalmau J, Furneaux HM, Gralla RJ, Kris MG, Posner JB. 1990. Detection of the anti-Hu antibody in the serum of patients with small cell lung cancer—a quantitative western blot analysis. Ann. Neurol. 27:544–52 Darnell RB, DeAngelis LM. 1993. Regression of small-cell lung carcinoma in patients with paraneoplastic neuronal antibodies. Lancet 341:21–22 Graus F, Dalmou J, Rene R, Tora M, Malats N, et al. 1997. Anti-Hu antibodies in patients with small-cell lung cancer: association with complete response to therapy and improved survival. J. Clin. Oncol. 15:2866–72 Groh V, Bahram S, Bauer S, Herman A, Beauchamp M, Spies T. 1996. Cell stressregulated human major histocompatibility complex class I gene expressed in gastrointestinal epithelium. Proc. Natl. Acad. Sci. USA 93:12445–50 Bahram S. 2000. MIC genes: from genetics to biology. Adv. Immunol. 76:1–60 Groh V, Rhinehart R, Randolph-Habecker J, Topp MS, Riddell SR, Spies T. 2001.

104.

105.

106.

107.

108.

109.

110.

111.

P1: IKH

357

Costimulation of CD8αβ T cells by NKG2D via engagement by MIC induced on virus-infected cells. Nat. Immunol. 2:255–60 Groh V, Rhinehart R, Secrist H, Bauer S, Grabstein KH, Spies T. 1999. Broad tumor-associated expression and recognition by tumor-derived γ δ T cells of MICA and MICB. Proc. Natl. Acad. Sci. USA 96:6879–84 Vetter CS, Groh V, Thor Straten P, Spies T, Brocker EB, Becker JC. 2002. Expression of stress-induced MHC class I related chain molecules on human melanoma. J. Invest. Dermatol. 118:600–5 Jinushi M, Takehara T, Tatsumi T, Kanto T, Groh V, et al. 2003. Expression and role of MICA and MICB in human hepatocellular carcinomas and their regulation by retinoic acid. Int. J. Cancer 104:354–61 Bauer S, Groh V, Wu J, Steinle A, Phillips JH, et al. 1999. Activation of NK cells and T cells by NKG2D, a receptor for stressinducible MICA. Science 285:727–29 Gilfillan S, Ho EL, Cella M, Yokoyama WM, Colonna M. 2002. NKG2D recruits two distinct adapters to trigger NK cell activation and costimulation. Nat. Immunol. 3:1150–55 Pende D, Rivera P, Marcenaro S, Chang CC, Biassoni R, et al. 2002. Major histocompatibility complex class I-related chain A and UL16-binding protein expression on tumor cell lines of different histotypes: analysis of tumor susceptibility to NKG2D-dependent natural killer cell cytotoxicity. Cancer Res. 62:6178– 86 Cosman D, Mullberg J, Sutherland CL, Chin W, Armitage R, et al. 2001. ULBPs, novel MHC class I-related molecules, bind to CMV glycoprotein UL16 and stimulate NK cytotoxicity through the NKG2D receptor. Immunity 14:123–33 Groh V, Steinle A, Bauer S, Spies T. 1998. Recognition of stress-induced MHC molecules by intestinal epithelial γ δ T cells. Science 279:1737–40

11 Feb 2004

18:54

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

358

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

112. Wu J, Groh V, Spies T. 2002. T cell antigen receptor engagement and specificity in the recognition of stress-inducible MHC class I-related chains by human epithelial γ δ T cells. J. Immunol. 169:1236–40 113. Groh V, Wu J, Yee C, Spies T. 2002. Tumour-derived soluble MIC ligands impair expression of NKG2D and T-cell activation. Nature 419:734–38 114. Salih HR, Rammensee HG, Steinle A. 2002. Cutting edge: down-regulation of MICA on human tumors by proteolytic shedding. J. Immunol. 169:4098–102 115. Diefenbach A, Jamieson AM, Liu SD, Shastri N, Raulet DH. 2000. Ligands for the murine NKG2D receptor: expression by tumor cells and activation of NK cells and macrophages. Nat. Immunol. 1:119– 26 116. Cerwenka A, Bakker AB, McClanahan T, Wagner J, Wu J, et al. 2000. Retinoic acid early inducible genes define a ligand family for the activating NKG2D receptor in mice. Immunity 12:721–27 117. Diefenbach A, Jensen ER, Jamieson AM, Raulet DH. 2001. Rae1 and H60 ligands of the NKG2D receptor stimulate tumour immunity. Nature 413:165–71 118. Clark WH Jr, Elder DE, Guerry D, Braitman LE, Trock BJ, et al. 1989. Model predicting survival in stage I melanoma based on tumor progression. J. Natl. Cancer Inst. 81:1893–904 119. Mihm MC Jr, Clemente CG, Cascinelli N. 1996. Tumor infiltrating lymphocytes in lymph node melanoma metastases: a histopathologic prognostic indicator and an expression of local immune response. Lab. Invest. 74:43–47 120. Clemente CG, Mihm MC Jr, Bufalino R, Zurrida S, Collini P, Cascinelli N. 1996. Prognostic value of tumor infiltrating lymphocytes in the vertical growth phase of primary cutaneous melanoma. Cancer 77:1303–10 121. Zhang L, Conejo-Garcia JR, Katsaros D, Gimotty PA, Massobrio M, et al. 2003. Intratumoral T cells, recurrence, and sur-

122.

123.

124.

125.

126.

127.

128.

129.

130.

vival in epithelial ovarian cancer. N. Engl. J. Med. 348:203–13 Naito Y, Saito K, Shiiba K, Ohuchi A, Saigenji K, et al. 1998. CD8+ T cells infiltrated within cancer cell nests as a prognostic factor in human colorectal cancer. Cancer Res. 58:3491–94 Schumacher K, Haensch W, Roefzaad C, Schlag PM. 2001. Prognostic significance of activated CD8+ T cell infiltrations within esophageal carcinomas. Cancer Res. 61:3932–36 Ishigami S, Natsugoe S, Tokuda K, Nakajo A, Che X, et al. 2000. Prognostic value of intratumoral natural killer cells in gastric carcinoma. Cancer 88:577–83 Villegas FR, Coca S, Villarrubia VG, Jimenez R, Chillon MJ, et al. 2002. Prognostic significance of tumor infiltrating natural killer cells subset CD57 in patients with squamous cell lung cancer. Lung Cancer 35:23–28 Coca S, Perez-Piqueras J, Martinez D, Colmenarejo A, Saez MA, et al. 1997. The prognostic significance of intratumoral natural killer cells in patients with colorectal carcinoma. Cancer 79:2320– 28 Balch CM, Riley LB, Bae YJ, Salmeron MA, Platsoucas CD, et al. 1990. Patterns of human tumor-infiltrating lymphocytes in 120 human cancers. Arch. Surg. 125:200–5 Svane IM, Engel AM, Nielsen MB, Ljunggren HG, Rygaard J, Werdelin O. 1996. Chemically induced sarcomas from nude mice are more immunogenic than similar sarcomas from congenic normal mice. Eur. J. Immunol. 26:1844– 50 Engel AM, Svane IM, Rygaard J, Werdelin O. 1997. MCA sarcomas induced in scid mice are more immunogenic than MCA sarcomas induced in congenic, immunocompetent mice. Scand. J. Immunol. 45:463–70 Macleod K. 2000. Tumor suppressor genes. Curr. Opin. Genet. Dev. 10:81–93

11 Feb 2004

18:54

AR

AR210-IY22-11.tex

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CANCER IMMUNOEDITING 131. Hanahan D, Weinberg RA. 2000. The hallmarks of cancer. Cell 100:57–70 132. Hanahan D, Folkman J. 1996. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell 86:353–64 133. Carmeliet P, Jain RK. 2000. Angiogenesis in cancer and other diseases. Nature 407:249–57 134. Sternlicht MD, Werb Z. 2001. How matrix metalloproteinases regulate cell behavior. Annu. Rev. Cell. Dev. Biol. 17:463–516 135. Vicari AP, Caux C. 2002. Chemokines in cancer. Cytokine Growth Factor Rev. 13:143–54 136. Matzinger P. 1994. Tolerance, danger, and the extended family. Annu. Rev. Immunol. 12:991–1045 137. Wrenshall LE, Stevens RB, Cerra FB, Platt JL. 1999. Modulation of macrophage B and cell function by glycosaminoglycans. J. Leukoc. Biol. 66:391–400 138. Benlagha K, Bendelac A. 2000. CD1drestricted mouse Vα14 and human Vα24 T cells: lymphocytes of innate immunity. Semin. Immunol. 12:537–42 139. Hodge-Dufour J, Noble PW, Horton MR, Bao C, Wysoka M, et al. 1997. Induction of IL-12 and chemokines by hyaluronan requires adhesion-dependent priming of resident but not elicited macrophages. J. Immunol. 159:2492–500 140. Bancroft GJ, Schreiber RD, Unanue ER. 1991. Natural immunity: a T-cellindependent pathway of macrophage activation, defined in the scid mouse. Immunol. Rev. 124:5–24 141. Kumar A, Commane M, Flickinger TW, Horvath CM, Stark GR. 1997. Defective TNF-α-induced apoptosis in STAT1-null cells due to low constitutive levels of caspases. Science 278:1630–32 142. Schreiber RD, Pace JL, Russell SW, Altman A, Katz D. 1983. Macrophage activating factor produced by a T cell hybridoma: physicochemical and biosynthetic resemblance to gamma-interferon. J. Immunol. 131:826–32

P1: IKH

359

143. Nathan CF, Murray HW, Wiebe ME, Rubin BY. 1983. Identification of interferon-γ as the lymphokine that activates human macrophage oxidative metabolism and antimicrobial activity. J. Exp. Med. 158:670–89 144. MacMicking J, Xie QW, Nathan C. 1997. Nitric oxide and macrophage function. Annu. Rev. Immunol. 15:323–50 145. Takeda K, Hayakawa Y, Smyth MJ, Kayagaki N, Yamaguchi N, et al. 2001. Involvement of tumor necrosis factor-related apoptosis-inducing ligand in surveillance of tumor metastasis by liver natural killer cells. Nat. Med. 7:94–100 146. Smyth MJ, Cretney E, Takeda K, Wiltrout RH, Sedger LM, et al. 2001. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) contributes to interferon γ -dependent natural killer cell protection from tumor metastasis. J. Exp. Med. 193:661–70 147. Hayakawa Y, Kelly JM, Westwood JA, Darcy PK, Diefenbach A, et al. 2002. Cutting edge: tumor rejection mediated by NKG2D receptor-ligand interaction is dependent upon perforin. J. Immunol. 169:5377–81 148. Gerosa F, Baldani-Guerra B, Nisii C, Marchesini V, Carra G, Trinchieri G. 2002. Reciprocal activating interaction between natural killer cells and dendritic cells. J. Exp. Med. 195:327–33 149. Srivastava P. 2002. Interaction of heat shock proteins with peptides and antigen presenting cells: chaperoning of the innate and adaptive immune responses. Annu. Rev. Immunol. 20:395–425 150. Li Z, Menoret A, Srivastava P. 2002. Roles of heat-shock proteins in antigen presentation and cross-presentation. Curr. Opin. Immunol. 14:45–51 151. Sallusto F, Mackay CR, Lanzavecchia A. 2000. The role of chemokine receptors in primary, effector, and memory immune responses. Annu. Rev. Immunol. 18:593– 620 152. Huang AY, Golumbek P, Ahmadzadeh M,

11 Feb 2004

18:54

360

153.

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

154.

155.

156.

157.

158.

159.

160.

161.

AR

DUNN

AR210-IY22-11.tex

¥

OLD

¥

AR210-IY22-11.sgm

LaTeX2e(2002/01/18)

P1: IKH

SCHREIBER

Jaffee E, Pardoll D, Levitsky H. 1994. Role of bone marrow-derived cells in presenting MHC class I-restricted tumor antigens. Science 264:961–65 Albert ML, Sauter B, Bhardwaj N. 1998. Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs. Nature 392:86–89 Schoenberger SP, Toes RE, van der Voort EI, Offringa R, Melief CJ. 1998. T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L interactions. Nature 393:480–83 Yu P, Spiotto MT, Lee Y, Schreiber H, Fu YX. 2003. Complementary role of CD4+ T cells and secondary lymphoid tissues for cross-presentation of tumor antigen to CD8+ T cells. J. Exp. Med. 197:985– 95 Loeb LA, Loeb KR, Anderson JP. 2003. Multiple mutations and cancer. Proc. Natl. Acad. Sci. USA 100:776–81 Loeb LA. 1991. Mutator phenotype may be required for multistage carcinogenesis. Cancer Res. 51:3075–79 Lengauer C, Kinzler KW, Vogelstein B. 1998. Genetic instabilities in human cancers. Nature 396:643–49 MacKie RM, Reid R, Junor B. 2003. Fatal melanoma transferred in a donated kidney 16 years after melanoma surgery. N. Engl. J. Med. 348:567–68 Elder GJ, Hersey P, Branley P. 1997. Remission of transplanted melanoma– clinical course and tumour cell characterisation. Clin. Transplant. 11:565–68 Suranyi MG, Hogan PG, Falk MC, Axelsen RA, Rigby R, et al. 1998. Advanced donor-origin melanoma in a renal transplant recipient: immunother-

162. 163.

164.

165.

166.

167.

168.

169.

170.

apy, cure, and retransplantation. Transplantation 66:655–61 Penn I. 1991. Donor transmitted disease: cancer. Transplant Proc. 23:2629–31 Sakaguchi S, Sakaguchi N, Shimizu J, Yamazaki S, Sakihama T, et al. 2001. Immunologic tolerance maintained by CD25+ CD4+ regulatory T cells: their common role in controlling autoimmunity, tumor immunity, and transplantation tolerance. Immunol. Rev. 182:18–32 Khong HT, Restifo NP. 2002. Natural selection of tumor variants in the generation of “tumor escape” phenotypes. Nat. Immunol. 3:999–1005 Marincola FM, Jaffee EM, Hicklin DJ, Ferrone S. 2000. Escape of human solid tumors from T-cell recognition: molecular mechanisms and functional significance. Adv. Immunol. 74:181–273 Catlett-Falcone R, Landowski TH, Oshiro MM, Turkson J, Levitzki A, et al. 1999. Constitutive activation of stat3 signaling confers resistance to apoptosis in human U266 myeloma cells. Immunity 10:105– 15 Algarra I, Cabrera T, Garrido F. 2000. The HLA crossroad in tumor immunology. Hum. Immunol. 61:65–73 Seliger B, Maeurer MJ, Ferrone S. 2000. Antigen-processing machinery breakdown and tumor growth. Immunol. Today 21:455–64 Hayashi T, Faustman DL. Development of spontaneous uterine tumors in low molecular mass polypeptide-2 knockout mice. Cancer Res. 62:24–27 Folkman J, Hahnfeldt P, Hlatky L. 2000. Cancer: looking outside the genome. Nat. Rev. Mol. Cell Biol. 1:76–79

3/12/2004 1:59 PM

CANCER IMMUNOEDITING

Figure 3 The three Es of cancer immunoediting: elimination, equilibrium, and escape. As indicated by the arrows, the immune system may eliminate the tumor in either the elimination or equilibrium phases, returning the tissue to normal.

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Dunn.qxd Page 1

C-1

Dunn.qxd

3/12/2004

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

DUNN

1:59 PM

■

OLD

■

Page 2

SCHREIBER

Figure 4 A proposed model for the elimination phase of the cancer immunoediting process. The events underlying this process are described in the text. Tumor cells are in blue; nontransformed cells in gray; lymphocytes, dendritic cells (DC), and macrophages (Mac) are marked and colored appropriately. Dead tumor cells are identified as white to gray gradient circles surrounded by a dashed black line, and tumor antigens are in blue squares. Panel (A) represents the initiation of the response, wherein cells of innate immunity recognize the nascent tumor. In panel (B), the initial amount of IFN-g produced starts a cascade of innate immune reactions that result in some tumor cell death by both immunologic and nonimmunologic mechanisms. In panel (C), events of innate immunity charge the adaptive response; tumor cells killed due to the increased cytocidal activities of NK cells and activated macrophages are ingested by DCs, which migrate to the draining lymph node and present antigen to naïve CD4+ and CD8+ T cells. In panel (D), tumor-specific CD4+ and CD8+ T cells home to the tumor along a chemokine gradient where they recognize and destroy tumor cells expressing distinctive tumor antigens.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:329-360. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

15 Mar 2004

15:23

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104644

Annu. Rev. Immunol. 2004. 22:361–403 doi: 10.1146/annurev.immunol.22.012703.104644 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on November 3, 2003

*View Erratum at http://arjournals.annualreviews.org/errata/immunol

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS Georg Wick,1 Michael Knoflach,1 and Qingbo Xu2 Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

1

Institute for Pathophysiology, University of Innsbruck, Medical School, Fritz-Pregl-Str. 3/IV, A-6020 Innsbruck, Austria; email: [email protected] 2 Department of Cardiovascular Medicine, St. George’s Hospital Medical School, Cranmer Terrace, London SW 17 0RE, United Kingdom; email: [email protected]

Key Words aging, inflammation, heat shock proteins, infections, signal transduction ■ Abstract The present review focuses on the concept that cellular and humoral immunity to the phylogenetically highly conserved antigen heat shock protein 60 (HSP60) is the initiating mechanism in the earliest stages of atherosclerosis. Subjecting arterial endothelial cells to classical atherosclerosis risk factors leads to the expression of HSP60 that then may serve as a target for pre-existent cross-reactive antimicrobial HSP60 immunity or bona fide autoimmune reactions induced by biochemically altered autologous HSP60. Endothelial cells can also bind microbial or autologous HSP60 via Toll-like receptors, providing another possibility for targetting adaptive or innate immunological effector mechanisms.

INTRODUCTION AND DEFINITIONS The arterial wall consists of four main layers: the innermost monolayer of endothelial cells, which is separated by a basement membrane from the intima consisting of connective tissue and some SMCs, the middle layer, the so-called media, consisting of SMCs, and finally an outer connective tissue layer, the adventititia, which embeds the vessel in its surroundings. Atherosclerotic lesions emerge in the intima, and for many years general dogma considered the so-called fatty streaks, cushion-like lipid-rich elevations protruding into the lumen, as the earliest stages of the disease that later transform into advanced lesions, so-called atherosclerotic plaques (1) (Figure 1). From the viewpoint of a pathologist, arteriosclerosis can be distinguished from atherosclerosis (2), where the former condition is characterized by the thickening and hardening (sclerosis) of the arterial wall associated with a loss of elasticity due to infiltration of the intima with mononuclear cells, SMCs, and increased ECM depositions. Atherosclerosis designates a special form of arteriosclerosis that is additionally characterized by the occurrence of so-called foam cells, i.e., lipid-laden macrophages and SMCs that may rupture and release 0732-0582/04/0423-0361$14.00

361

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

362

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

their contents into the lesional areas, so that even cholesterol crystals can form. Whereas all nucleated cells of the body express surface receptors for the cholesterol transport particles, i.e., native LDLs, macrophages and SMCs additionally carry scavenger receptors that bind biochemically altered, e.g. oxidized, LDL (oxLDL). In contrast to the classical LDL receptor, the scavenger receptor is not saturable, i.e., not downregulated by an excess of oxLDL, thus leading to an overload of cells with biochemically altered LDL (3). More recently, it has been shown that DC can also transform into foam cells, affording indirect evidence that they are capable of expressing the scavenger receptor (4). The atherosclerotic plaque is a complicated end-stage lesion that may exulcerate and even calcify. These severe lesions have a tendency to rupture at the edges, the so-called shoulder regions, provoke thrombus formation, and thus lead to the dreaded final complications such as myocardial infarction or stroke. Although arteriosclerosis is the umbrella term for these various forms of the disease, including atherosclerosis, the latter designation is now becoming the generally accepted term and is used throughout this review. It is only during the past 15–20 years that attention has also been given to the fact that inflammatory processes occur in the course of atherogenesis. This belated insight is astonishing in view of many earlier reports, some dating back more than a century, that had clearly described inflammatory processes in the walls of afflicted arteries (5). Historically, a controversy that arose in the middle of the nineteenth century relating to this issue is worth mentioning here. During this period, the two most prominent European pathologists, Karl von Rokitansky in Vienna, Austria, and Rudolf von Virchow in Berlin, Germany, had both recognized that inflammation takes place in the atherosclerotic vessel wall. Whereas von Rokitansky considered these changes to be of secondary nature, von Virchow emphasized that he assigned a primary role to them. Interestingly, our laboratory recently had the opportunity to study paraffin sections of early and late atherosclerotic lesions derived from autopsies that had been performed by von Rokitansky himself approximately 160 years ago. In the earliest of these lesions we found abundant CD3+ T cells, thus supporting the concept of von Virchow rather than that of von Rokitansky (6). Most studies during the last decade on inflammatory-immunologic aspects of atherosclerosis concentrated exclusively on advanced lesions, i.e., they did not relate to incipient stages of the disease. Furthermore, some of the classical papers as well as textbooks only relied on conventional histological procedures and did not harness immunofluorescence and immunohistological techniques.

CLASSICAL CONCEPTS OF ATHEROGENESIS Atherosclerosis is a multifactorial disease for which numerous pathogenetic concepts have been proposed. The main hypotheses are the following: Based upon a large body of earlier data, Ross formulated the response to injury hypothesis, stating that hyperlipidemia, hormone dysfunction, and the increased

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

363

shear stress in hypertension injure the endothelium and alter the nature of the endothelial barrier to the passage of blood constituents (7, 8). Until now, a long causative list of possible injuries consisting of classical risk factors for atherosclerosis, like high levels of LDL-cholesterol or oxLDL, free radicals caused by cigarette smoking, hyperglycemia as well as infectious agents (e.g., Chlamydia pneumoniae or herpes viruses) has been assembled (9). Also, genetic susceptibility to enthothelial damage (familial and male predisposition) has been proposed (10). Owing to this endothelial dysfunction, the permeability for lipoproteins and other plasma constituents—including growth factors— increases, leading to the deposition of lipids within the vessel wall and the activation and proliferation of macrophages and SMCs. In addition, differences in expression patterns of endothelial surface molecules alter the adhesion and migration properties of T lymphocytes and monocytes. The altered-lipoprotein hypothesis postulates an initiating role of chemically altered lipoproteins, most notably oxLDL, in forming foam cells and acting as chemoattractants for SMCs and mononuclear cells into the intima (11, 12). Advocates of the monoclonal hypothesis pretend that atherosclerosis is derived from a mono- or oligoclonal proliferation of SMCs, but data supporting this concept have become sparse in recent years (13). During the past 15 years, our group has developed a new immunological hypothesis of atherogenesis (14, 15), the roots of which were immunohistological data from very early arterial changes that do not yet lead to clinical symptoms. In essence, we have shown that the first cell types found in the arterial intima at sites known to be predisposed for the later development of atherosclerotic lesions are not foam cells, but rather lymphoid cells followed by macrophages and SMCs (16). Within the T cell population in these early lesions, CD4+ clearly prevail over CD8+ cells. This observation differs from findings by van der Wal et al. (17), who described a tendency of an increased CD8/CD4 ratio in both early and late lesions. Furthermore, in our specimens the CD4+ cells were mostly activated, i.e., positive for HLA-DR and CD25. The majority of the intima infiltrating T cells carry the T cell receptor αβ (TCRαβ), but an unexpectedly high proportion of these T cells (10%–15%) are TCRγ δ +, a frequency that far exceeds that of TCRγ δ + cells in the peripheral blood (1%–2%). Further characterization of TCRγ δ +-cells revealed that TCRγ 9δ2+, a configuration that is characteristic for human peripheral blood, made up only a small proportion (1%–2%, like in peripheral blood) of this subpopulation, whereas the majority expressed the TCR Vδ1 chain that is typical for the human MALT (18). A constitutive expression of HLA-DR by endothelial cells could not be detected in our sections of unafflicted arteries or in early lesions. As a matter of fact, HLADR+ endothelial cells were only found in those instances where activated IFN-γ producing T cells were present in the underlying intima (16). However, a recent report by Buono et al. (19) showed MHC class II antigen expression on endothelial cells overlying atherosclerotic lesions in LDL-R/ IFN-γ double knockout mice.

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

364

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

Granulocytes are usually not present in early atherosclerotic lesions (16, 20). B cells and NK cells are rare, but mast cells do occur in considerable numbers (21). This latter finding may not only be relevant with respect to the vascular permeability promoting effects of vasoactive peptides but also because of their profibrotic properties. In an extension of our first immunohistochemical studies, we then had the opportunity to analyze arterial specimens from a multicenter cooperative study that was begun in 1985 in the United States, the PDAY study (22). In the PDAY study, more than 3000 arterial fragments were collected from young (15–34-yearold) Americans who died from noncardiovascular causes, mostly accidents and suicides. Surprisingly, earlier histological studies of this valuable autopsy material did not address the question of an intimal inflammatory process in general, and the participation of T cells in particular. Our previous phenotypical observations summarized above could again be corroborated in the PDAY material with the following additional features: This time we also looked for CD1a DC and, in agreement with the previous results of Bobryshev et al. (23), found them to be present in the intima of all subjects. Because DC obviously form a link between the innate and the adaptive immune systems, it is of interest that Aicher et al. (24) have recently described a strong DC-activating potential of nicotine. This is reflected by the expression of costimulatory molecules (CD86, CD40) and MHC class II and adhesion molecules (LFA-1, CD54), and a sevenfold increase in secretion of the proinflammatory Th-1-activating cytokine IL-12. This observation is, of course, especially relevant in view of the known atherosclerosis-promoting effect of cigarette smoking. In contrast to our earlier investigations, PDAY specimens contained considerable numbers of CD19+ B cells, again depending on and increasing with severity of lesions. Although the present review concentrates on conventional atherosclerosis, i.e., atherosclerosis that can afflict everybody when applicable, albeit avoidable, risk factors are present, there are also other variants of this disease that emerge under very special conditions. On one hand, accelerated atherosclerosis occurs in allotransplant recipients; on the other hand, atherosclerosis develops on a clear-cut genetic basis, i.e., in experimental animals or humans lacking lipoprotein E (ApoE) or the LDL-R. While transplant atherosclerosis is mainly mediated by T cells, excessive hypercholesterolemia in the latter genetically compromised groups seems to entail a form of disease that is nearly exclusively based on foam cell formation and cholesterol deposition.

THE ROLE OF THE INNATE IMMUNE SYSTEM The repertoire of the highly conserved PRRs by cells of the innate defense system is based on a natural selection of germline-encoded genes and includes TLRs and collectins (mannan-binding lectin, surfactant). Probably fewer than 100 different

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

365

types of receptors exist, with each recognizing restricted patterns of entire classes of pathogen ligands, called PAMPs. PAMPs are present not only on the surface of pathogens, but also in solution, e.g., as bacterial endotoxins (LPS), denatured proteins (e.g., malondialdehyde derivatized or oxidized lipoproteins), bacterial and denatured eukaryontic DNAs, mannans, teichoic acids, and a variety of other compounds (25). The binding of ligands leads to internalization of the ligandreceptor complex, activation of the cell, and the mounting of an inflammatory defensive response via activation of various transcription factors, most notably NFκB. As will be detailed below in the section dealing with the vascular-associated lymphoid tissue, cellular components of the innate immune system are already assembled in the intima of arteries in children at certain branching points, known to be predilection sites for the possible later development of atherosclerotic lesions. It should be emphasized here that these sites could be considered to be part of the surface exposed areas of the body, albeit in this instance an inner surface exposed to PAMPs and exogenous as well as autoantigens contained in the blood stream. When the rolling, adhesion, and transendothelial migration of leucocytes is prevented, either by lack of expression of adhesion molecules or chemotactic molecules, respectively, atherosclerotic lesions are greatly reduced or do not develop. Thus, atherosclerosis develops only to a minor extent in hypercholesterolemic mice deficient in VCAM-1 (26). The same is true for P-selectin (27) and ICAM-1 (28), although these latter adhesion molecules seem to play a less prominent role. MCP-1 is essential for the recruitment of monocytes into the intima. MCP-1/Apo-E or LDL-R double knockout mice do not develop atherosclerosis (29). Also, crosses of Apo-E−/− or LDL-R−/− mice with mice deficient in the CCR-2 (the receptor for MCP-1) exhibit significantly impaired development of atherosclerotic lesions (30). It has to be remphasized that oxLDL deposited in the intima acts as a monocyte chemoattractant too, and oxLDL phospholipid constituents stimulate MCP-1 expression by endothelial cells manifested, by, the induction of VCAM-1 for example (31). VCAM-1 is an adhesion molecule for both monocytes and T cells that both express the respective ligand on their surface, i.e., the integrin VLA-4. It is not yet clear how cells of the innate system are recruited toward these sites. However, it is of interest that the arterial intima of wild-type mice contain only very few, if any, DC as compared to the elaborate network of such cells in human arteries (32). This may be one of the reasons why wild-type mice are relatively resistant to the induction of atherosclerosis, be it being fed a cholesterol-rich diet or by other means. The presence of monocytes seems to be absolutely necessary for the development of atherosclerosis because hypercholesterolemic MCP-1−/− or CCR-2−/− mice develop only minor, if any, atherosclerotic lesions. Arterial intimal macrophages derived from blood-borne monocytes exert both phagocytic functions via their scavenger receptors and innate defense functions via other PRRs. The main representatives of scavenger receptors are CD36, SR-P1, and SR-A. These receptors bind oxLDL and, due to the fact that they are not saturable, entail the formation of foam cells as mentioned above. Phagocytosis in general and

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

366

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

activation of macrophages via TLRs in particular lead to the generation of reactive oxygen species—the major microbial killer mechanisms—and thus transduction of inflammatory signals. An additional role for TLRs as target presenting structures that has so far not yet been clarified in detail may be due to the fact that CD14, the nontransmembrane receptor for LPS involved in the induction of nonspecific inflammation, interacts with different TLRs and that TLR-2 and TLR-4 have both been shown to bind HSPs, i.e., highly conserved molecules that act both as PAMPs and as potent antigens for adaptive immunity (33). Finally, humoral components of the innate immune system also have to be considered in this context. These include natural antibodies, complement components, acute-phase proteins, such as CRP, and MMPs and their antagonists, TIMPs. Deposits of complement components as a result of activation—via the classical, the alternative, or the mannose-dependent pathways—have been known for a long time within atherosclerotic lesions and seem to represent a transitional mechanism between innate and adaptive immunity in atherogenesis (34). It is of interest here that complement receptor-3 (CR-3)-deficient LDR−/− mice are more susceptible to atherosclerosis when compared to wild-type animals, showing a protective effect of complement activation, at least under certain circumstances (35). The association of inflammatory processes reflected by increased levels of CRP with an increased risk for atherosclerosis has now been supported by many groups, including ourselves (36, 37, 38). CRP is a useful parameter for the infectious load of a given individual. This issue is dealt with in more detail later in this review. Recently, the group of Witztum has described protection from atherosclerosis by natural antibodies with a certain idiotype (D15) directed against oxLDL (39). Natural antibodies are generated without known antigenic stimulation and belong to the IgM type produced by a special subset of innate B lymphocytes, the B1 cells. They are characterized by the expression of restricted germline-encoded B cell receptors conveying specificity for natural or self-antigens. Natural antibodies have been shown to bind oxidized phopholipids, such as POVPC, and serve to remove oxLDL, thus preventing its uptake by macrophages. Interestingly, these antibodies recognize similar oxidation-specific epitopes on apoptotic cells and are structurally and functionally identical to classical anti-PC antibodies with the D15 idiotype of B1 cell origin that is reported to provide optimal protection against pneumococcal infection (40). The authors suggest that a natural selection process for B1 cells secreting oxidation-specific D15 antibodies has been operative, both for their role in natural immune defense and for housekeeping purposes against oxidation-dependent epitopes under healthy and pathologic conditions. In a recent paper, Binder et al. (41) showed that mice immunized with pneumococci do develop D15 antibodies that protect from atherosclerosis. In conclusion, cellular and humoral components of the innate immune system certainly play a major role in atherogenesis. It is, however, not yet clear if this role is of a primary or secondary nature within the complicated network of factors contributing to the various stages of this disease.

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

INFLAMMATION IN ATHEROSCLEROSIS

367

THE ROLE OF ADAPTIVE IMMUNITY

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Humoral Antibodies Complement activation via the classical pathway could be the result of the deposition of immune complexes in the arterial intima, or it may reflect the binding of specific antibodies to autologous or exogenous antigens localized in the arterial wall. As detailed below, the former include oxLDL and HSP60, whereas the latter comprise various types of microorganisms such as EBV, CMV, or bacteria, such a Chlamydiae. A further possibility is the activation of complement via the alternative pathway. Here again, cholesterol crystals and other lipid components have been shown to activate this pathway in vitro (42). Interestingly, complement activation also occurs in the arterial intima of rabbits shortly after beginning a cholesterol-rich diet (34, 43, 44). That intralesional complement is activated by the classical pathway is supported by the codistribution with IgG, although the specificity of such locally deposited antibodies has not yet been demonstrated. Infection has an aggravating effect on atherogenesis, most probably due to the endothelial damaging effect of circulating immune complexes (45). The candidates for antigens against which such antibodies are directed are discussed below.

Cellular Immunity In human atherosclerosis, functional studies on intima-infiltrating T cells have so far been restricted to those harvested from advanced lesions. Obviously, these are easier to obtain in the form of surgically removed tissue fragments as compared to the more interesting (at least for us) early, clinically silent incipient stages. This is also the reason why most papers advocate an intralesional preponderance of macrophages over T cells in human atherosclerosis, as well as atherosclerosis in cholesterol-fed rabbits and ApoE knockout mice. Applying RT-PCR technology, Stemme et al. (46) showed more than a decade ago that T cells in human atherosclerotic plaques are of polyclonal origin, suggesting that they are either recruited to the plaque in an activated stage or activated locally by mechanisms that do not lead to clonal proliferation. However, the possibility that these populations contain pathogenic clones with locally relevant antigenic specificity cannot be ruled out. Nevertheless, oligoclonal T cell proliferation has been demonstrated in atherosclerotic lesions of ApoE−/− mice (47). In order to probe this idea, in vitro screenings of all populations of T cells reacting with predefined, potentially atherogenic antigens should be performed. Most importantly, this type of analysis has not yet been done on cells derived from early lesions. There is no question that knockout mouse models for cholesterol-induced atherosclerosis have contributed considerably to our understanding of atherogenesis. It has to be emphasized, however, that these models do not reflect those situations in common human atherosclerosis that are induced by a variety of

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

368

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

risk factors and hit nearly everybody in the Western world. The murine knockout models, such as LDL-R and ApoE knockout mice, in essence have their counterparts in genetically determined familial hypocholesterolemia in human patients. In common atherosclerotic lesions, Th1 cells expressing the cytokines IFN-γ and IL-2 prevail over IL-4-, IL-5-, and IL-10-producing Th2 cells (48, 49). This fact is also reflected by significantly elevated serum levels of neopterin, an IFNγ induced macrophage product (50, 51). IFN-γ has a proinflammatory and thus proatherogenic effect. This Th1 cell product locally activates macrophages and inhibits the proliferation of SMC and collagen synthesis. This then leads to a higher risk of plaque instability, i.e., plaque rupture. The proinflammatory effect of IFN-γ is augmented by the macrophage-derived cytokines IL-1 and TNF-α, and reflected by the induction of expression of acute-phase proteins such as IL-6 and CRP in the liver (52). In contrast, the anti-inflammatory cytokine TGF-β produced by macrophages, SMC, and Th3 cells, has a profibrotic effect, and thus promotes plaque stability. Blocking TGF-β signaling by appropriate monoclonal antibodies entails the development of larger and unstable lesions (53). Results obtained in ApoE−/− mice crossed with immunodeficient RAG1/2−/− mice originally seemed to support the idea that atherosclerosis can also develop without T cells (54, 55). However, careful recalculation of these data clearly shows a significant increase in the extent of lesion formation in these immunocompro- *Erratum mised combinations as compared to pure immunocompetent hypercholesterolemic strains (56). We have treated rabbits with high doses of monoclonal anti-CD43 antibodies, leading to nearly complete depletion of peripheral blood T cells and found that this regimen prevented the development of atherosclerosis (57). However, it must be emphasized that data published after the original production and commercial availability of this antibody have shown that CD43 is also expressed, albeit to a minor extent, by macrophages and endothelial cells, and its atherosclerosis-inhibiting effect can therefore perhaps not only be assigned to T cell depletion (57, 58). Finally, it has recently been shown that splenectomy, both of hypercholesterolemic ApoE−/− mice and human patients, leads to significantly more severe atherosclerosis (59, 60, 61). In mice, this effect could be compensated by the transfer of purified T or B cells from atherosclerotic ApoE−/− donors. Although an explanation of this interesting phenomenon is still lacking, it probably reflects the susceptibility to infections with an increased odds ratio to develop atherosclerosis.

Antigens Multifactorial diseases such as atherosclerosis are complicated by definition, but even complicated diseases have to begin by a single event, albeit not necessarily the same one in all instances. Later in this review, we show that we are of the opinion that conventional atherosclerosis is initiated by specific cellular and humoral immune reactions. Based on data from our laboratories and others, we hypothesize

*Erratum (24 Mar. 2004): See online log at http://arjournals.annualreviews.org/errata/immunol

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

369

that atherosclerosis is started by T cells that immigrate from the blood stream into the intima via the endothelial layer and that this process is accelerated by humoral antibodies (62). What are the antigens that are recognized by these antibodies and T cells, respectively? Over the years, a great number of antigens have been incriminated as playing a role in atherosclerosis-associated immune reactions. Originally, these included undefined “endothelial surface antigens,” sperm components, herpes viruses, biochemically altered lipoproteins, oxidized lipids, cardiolipin, structural vessel wall antigens, milk proteins, and others. In recent years, the list of candidate antigens has been narrowed down to a few major groups, both exogenous and autologous. The former comprise various viral and bacterial components, and the latter β-2 GP-1 (63), as well as biochemically modified arterial wall proteins and modified LDL, such as oxLDL. Our own group has introduced the concept that a family of phylogenetically highly conserved proteins, the stress or HSPs, play a major role in the induction of the earliest atherogenic immune responses. They also represent the missing link between exogenous and endogenous disease-promoting antigens, as is discussed in detail below. Another extensively studied candidate antigen is biochemically modified LDL, i.e., oxLDL or artificially derivatized LDL such as MDA-LDL or Cu SO4 oxLDL (Cu oxLDL) (64, 65). These modifications of LDL lead to the formation of several oxidation-specific epitopes, with the oxidized PCcontaining phospholipid forming an immunodominant epitope (66). As detailed in subsequent sections of this review, the notion now has emerged that immune reactions against HSP are proatherogenic, whereas—in contrast to assumptions in earlier reports—immune reactivity against oxLDL is protective.

Heat Shock Protein 60 HSPs are expressed in prokaryotic and eukaryontic cells under physiological conditions and in response to various forms of stress (Figure 2). These were first detected in Drosophila as a response to heat (hence the name), but it later turned out that many other forms of stress exert the same effect. They are classified into various families depending on their molecular weight, ranging from the 100 kDa over the 70 kDa, 60 kDa down to low-molecular-weight families (67). HSPs fulfill a wide variety of physiological functions during the process of intracellular protein transport, protein folding, cellular signaling, and protein degradation. Under stress conditions, some HSPs associate with denaturing cellular proteins and protect them from aggregation (chaperone function) (68). With regard to atherogenesis, the 60 kDa HSP family is of special importance because the members of this family were identified by us to represent the culprit antigens in the earliest stages of atherosclerosis (14, 69). The HSP60 family comprises, among others mammalian HSP60, the mycobacterial homologue mHSP65, cHSP60, and the Escherichia coli homologue GroEL (70). The members of the HSP60 family are phylogenetically highly conserved in mammalian and bacterial

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

370

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

Figure 2 Inducers of cellular stress. Various forms of stress induce transcription of HSPs via heat-shock factors (HSFs).

species. This fact forms the basis for extensive immunological cross-reactions between pathogens and autologous HSP60 (71). Other authors have provided evidence that other HSPs might also be involved in the development of atherosclerosis, for example HSP47 (72) or HSP70 (73). However, in contrast to the situation with HSP60, evidence has not been provided that immunization of experimental animals with these HSPs leads to the induction of atherosclerosis. More than 95% sequence homology exists on both the DNA and protein levels between HSP60 from various bacteria, e.g., mHSP65, cHSP60, and GroEL. Even between bacterial and hHSP60 an approximately 50%–55% sequence homology exists, and in highly conserved regions it reaches more than 70% (71). HSP60 can also be found in parasites, and even in the envelope layers of viruses such as HIV, the latter acquiring this molecule when budding off the surface of infected host cells (74). HSP60 is not only qualitatively a major microbial component, but also quantitatively constitutes an important and very immunogenic antigen. Practically all humans and animals develop humoral and cellular immunity against this antigen as a consequence of infection or vaccination, respectively. However, protective immunity may have to be “paid for” by the risk of cross-reactivity with autologous HSP60 that can, e.g., be expressed by stressed or otherwise maltreated vascular endothelial cells, as discussed below (75). Furthermore, copious amounts of HSP60 are released from the surface of stressed or damaged cells, and thus can be demonstrated in the supernatant of cell cultures or in the serum under in vivo conditions, respectively (76). We have previously shown that this latter sHSP60 can also be biochemically modified (77) and thus may lead to bona fide autoimmunity.

15 Mar 2004

15:23

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

P1: IKH

371

In eukaryontic cells, e.g., human endothelial cells, HSP60 expression can be induced by many different types of stressors, such as mechanical stress, temperature, oxygen radicals, infections, toxins, heavy metals, proinflammatory cytokines (e.g., TNF-α), etc. (69). Importantly, the same stressors lead to the simultaneous expression of HSP60 and adhesion molecules ICAM-1, ELAM-1, and VCAM-1 (78), thus providing the prerequisites not only for an interaction of potentially bacterial/human HSP60 cross-reactive antibodies, but also of T cells with endothelial targets. At later stages of atherogenesis, intralesional macrophages and SMC also express HSP60 (18), and the anti-HSP60 cellular immune reaction could, therefore, be perpetuated in situ. HSP60 is a mitochondrial protein that can be translocated into the cytoplasm under stress conditions (Figure 3), and it is even transported to the cell surface where it seems to provide a danger signal (79, 80). In vivo treatment of rats with LPS leads to the simultaneous expression of HSP60 and ICAM-1 by endothelial cells (81). It was further shown that leukocytes attached at exactly those sites, where EC had already been subject to major haemodynamic stress, e.g., the branching points of intracostal arteries from the aorta (81). More recently, an additional possible mechanism of cell surface presentation has been described for sHSP60, i.e., passive binding to surface receptors on endothelial cells (82). On the one hand, this provides another possibility for specific T cells and antibodies to interact with their target antigen; on the other hand, receptor-bound HSP60 may trigger signal transduction via the proinflammatory MAPK pathway. HSP60 might also be a ligand for TLR-4 and TLR-6, the former in conjunction with CD14 (83, 84). Interestingly, recent data obtained in a large number of volunteers from a well-controlled prospective atherosclerosis prevention study, the Bruneck Study (85, 86), provided evidence for a significant role of TLR-4 polymorphism in the susceptibility to atherosclerosis (87). However, functional data on possible alterations of HSP60 binding to and signal transduction via certain TLR isotypes are still missing.

EXPERIMENTAL MODELS FOR IMMUNOLOGICALLY INDUCED ATHEROSCLEROSIS Rabbits The first step of our own investigations into the possible role of (auto)immune mechanisms in atherogenesis consisted of the immunization of normocholesterolemic rabbits with appropriate antigens plus CFA (88). For this purpose, groups of normocholesterolemic New Zealand white rabbits were immunized either with delipidated proteins isolated from human atherosclerotic plaques or with similarly prepared plaque proteins from Watanabe rabbits (which lack the LDL-R and develop a hereditary form of atherosclerosis), emulsified with CFA. A control group of rabbits was immunized with OVA plus CFA. We reasoned that the first two groups would develop atherosclerotic lesions if appropriate autoantigens

19 Feb 2004

12:1

372

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

were present in the plaque extracts, and the controls should remain negative. Surprisingly, all three groups exhibited macroscopically visible lesions at the known predilection sites. Upon microscopic scrutiny it was found that these lesions consisted of infiltrations of the intima by mononuclear cells and SMC, however without foam cell formation. Because CFA was the only common denominator in these experiments, normocholesterolemic rabbits were then immunized with CFA alone, and atherosclerosis again emerged in these animals. Immunization with adjuvants that—in contrast to CFA—did not contain mycobacteria, and thus were devoid of mycobacterial HSP65, such as IFA, Ribi or lipopeptide, were without effect. HSP60 is known to be a major component of mycobacteria (mHSP65) and has been discussed as a culprit bacterial-mammalian cross-reactive antigen in a variety of diseases such as RA (89, 90). We therefore repeated our experiments by immunizing normocholesterolemic rabbits with recombinant mHSP65 alone, and the same results were again observed, i.e., mononuclear cell infiltrations of the intima at those arterial sites that are known to be subjected to major turbulent haemodynamic stress (88). In rabbits, a cholesterol rich diet has been the classical way to induce experimental atherosclerosis. Therefore, we subsequently combined immunization with mHSP65 by supplementation with a cholesterol-rich diet (2% cholesterol), and observed significantly more severe atherosclerotic lesions in these animals as compared to those that were either fed a cholesterol rich diet alone or were only immunized with mHSP65 (91). Using this model, we have shown that the earliest inflammatory stage of atherosclerosis induced by immunization with mHSP65 is still reversible, whereas the severe atherosclerotic changes induced by combining immunization and feeding a cholesterol-rich diet are not. As expected, the peripheral blood of rabbits immunized with mHSP65 contained antibodies and T cells reacting with this antigen (92). Analysis of mHSP65specific T cell lines derived from the peripheral blood, or atherosclerotic lesions of immunized rabbits showed an intralesional accumulation of such cells. It was, however, surprising that we also found abundant T cells in the atherosclerotic lesions (fatty streaks and plaques) of rabbits from the control group that was only fed a cholesterol-rich diet, without concomitant immunization. When T cell lines were produced from these cholesterol-rich lesions of a given rabbit, and compared to those derived from the peripheral blood of the same animal, a significant enrichment of HSP65-reactive T cells was observed in the lesions in comparison with cells derived from the peripheral blood (92). In conclusion, HSP60 has been found to be an optimal candidate for an antigen that plays a proatherogenic role in the very early inflammatory stages of the disease in the rabbit model. These changes may progress into more severe lesions when high serum concentrations of LDL are present, entailing the formation of foam cells and extracellular lipid deposition.

Mice Wild-type mice are notoriously resistant against the induction of atherosclerosis using the previously mentioned conventional dietary cholesterol supplementation.

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

373

Out of all the strains, C57BL6J is perhaps the most susceptible to developing some atherosclerosis upon being fed a high cholesterol diet, although still to a much lesser extent than that observed in rabbits. More recently, several knockout models have become available concerning genes and thus proteins that are crucial in lipid metabolism. These models have opened new possibilities for experimental atherosclerosis research. For example, ApoE−/− (93) and LDL-R−/− mice (94) accumulate excessive amounts of serum cholesterol and consequently develop severe atherosclerosis even when only given mildly cholesterol-enriched food, the so called Western diet. We have, therefore, also used these models in combination with various immunization procedures to study the role of immune reactions in atherogenesis. Our main interest was focused on the, at that time still-controversial, issue of whether biochemically altered LDL, i.e., oxLDL or MDA-LDL, is a possible antigenic candidate for a proatherogenic immune reaction. On the other hand, we were also interested to know if immunity to HSP60 may also lead to atherosclerosis in mice. The results of our experiments can be summarized as follows: The immunization of wild-type C57BL6J mice and ApoE−/− mice with mHSP65 or MDA-LDL, respectively, showed that the former led to an aggravation of the disease (95), whereas immunity against the latter was protective (96). Furthermore, the fact that—in contrast to the situation in rabbits—inbred mouse strains are available allowed for the demonstration of passive transfer of atherosclerosis from immunized donors to these hypercholesterolemic recipients (97). It is obvious that this new concept of atherogenesis provoked the idea that tolerization against atherogenic HSP60 epitopes may be a possible approach to prevent or even treat atherosclerosis. This principle has, in the meantime, been successfully applied in murine hypercholesterolemic ApoE−/− and LDL-R−/− models by treating these either intranasally or orally with whole mHSP65 preparations (98, 99). In contrast to the well-defined proarthritogenic and arthritis-preventing HSP60 epitopes that had been identified in the model of adjuvant arthritis in rats (100, 101), and human RA (102), atherogenic T cell epitopes have so far not yet been identified. This would, however, be of prime importance before embarking on such a preventive or therapeutic approach. The high frequency of HSP60 reactive T cells within the MALT and the occurrence of low titers of anti-HSP60 antibodies in most humans suggests that immunity against this phylogenetically highly conserved antigen has an important protective task, the complete circumvention of which may lead to higher susceptibility to a variety of different infectious diseases.

AUTOIMMUNE REACTIONS IN ATHEROSCLEROSIS HSP Expression in Atherosclerosis HSP60 expression in atherosclerotic lesions was shown by Kleindienst et al. (18). These authors reported that HSP60 was detected in endothelial cells, SMC, and

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

374

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

mononuclear cells of carotid and aortic specimens showing atherosclerotic lesions. In contrast, disease-free control vessels showed no expression of HSP60 (18). Increased expression of HSP70 has been shown in human (103) and rabbit arteries, and also in DCs in the arterial wall (104). Interestingly, expression of HSP60 and HSP70 was also found to correlate with the development of atherosclerotic lesions in the aortic tree of ApoE−/− mice (105). HSP47, acting as a chaperone for procollagen, has also been found to be involved in atherosclerosis (72). These findings identified HSPs as novel constituents of human coronary atheroma and suggested that selective upregulation of HSPs by stress may play a role in atherogenesis. However, only immunization with HSP60 seems to be capable of inducing atherosclerotic lesions in experimental animals. In vitro studies have demonstrated that the heat shock response is primarily regulated at the level of transcription. It is mediated by one or more members of a family of HSFs that interact with a specific regulatory element, the heat shock element (HSE), which is present in the promoters of the HSP genes (106, 107). In scrutinizing the molecular mechanisms of HSP expression involving HSF activation in atherosclerotic lesions, we recently demonstrated that HSF1—mainly present in the nuclei—was highly expressed in the cap and valve regions of atherosclerotic lesions in oneyear-old ApoE−/− mice (108). Immunohistochemical double labeling of vascular sections indicated that a proportion of SMC in the lesions were HSF1 positive. The level and activity of HSF1 in protein extracts from atherosclerotic lesions of hypercholesterolemic rabbits were significantly higher than those of normal vessels, as identified by Western blot and gel mobility shift assays. Furthermore, TRLP, oxidized-TRLP, LDL, and oxLDL did not activate HSF1 in cultured SMC, whereas HSF1 was highly activated in cells treated with TNF-α. Interestingly, mechanical stretching of SMC resulted in HSF1 hyperphosphorylation, translocation from cytoplasm to the nucleus, followed by increased HSP70 expression (108). Thus, our findings provide the first evidence that HSF1 is activated and highly expressed in atherosclerotic lesions, and that cytokine stimulation and excessive mechanical stress to the vessel wall may be responsible for such activation (Figure 4). So far, the kinases of other enzymes responsible for HSF hyperphosphorylation are unknown. However, there is evidence indicating that HSF1 activity can be inhibited through phosphorylation of HSF1 serine residues by ERKs (109). In fact, three families of MAPKs (ERKs, JNK/SAPK, and p38MAPK) are activated in vascular cells stimulated by heat shock (110), free radicals (111), LDL and oxLDL (112), arachidonic acid (113), hyperlipidemia (114, 115), and mechanical stress (116). This indicates a possible relationship between MAPK activation and HSP expression in cells of atherosclerotic vessels. Furthermore, different stimuli seem to activate different signal pathways leading to HSF activation. In SMC for instance, mechanical stress stimulation of the integrin-rac pathway results in HSP induction, whereas H2O2 activates the JAK2-STAT pathway leading to HSP expression (117). However, further studies will be needed to clarify the molecular mechanism of HSP expression in vivo and in vitro.

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

375

Figure 4 Signal pathways of HSP expression in vascular cells induced by various forms of stress. STRAIN indicates mechanical stress. ERK, extracellular regulatedprotein kinase; MEK, mitogen-activated protein kinase/ERK kinase; HSF, heat-shock transcription factor.

Surface Expression and Release of HSPs Although it is assumed that HSPs must be located in the cytoplasmic compartment in order to exert their function, Xu et al. (80) found that aortic endothelial cells express HSP60 on their surface after cytokine stimulation. Surface staining of endothelial cells was possible with the monoclonal antibody II-13, which recognizes amino acid residues 288–366 of HSP60, but not by ML-30, which binds to residues 315–318 or LK1, LK2-recognizing residues 383–447. Several groups (79, 118, 119, 120) have confirmed the surface expression of HSP60 in different cell types. Furthermore, Schett and coworkers (121) demonstrated that rat heart ischemia/reperfusion resulted in HSP60 release. These data suggest that HSPs are not only expressed within cells, but also on the cell surface, and can be released into the cultured medium or intercellular space in certain circumstances.

Autoimmune Reactivity Against HSP In Vivo in the Human System HUMORAL AUTOANTIBODIES Based on the experimental evidence reviewed above, the autoimmune hypothesis of atherogenesis featuring HSP60 as a major potential

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

376

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

autoantigen was formulated. In 1993, our group could demonstrate a correlation of the prevalence of sonographically demonstrable carotid atherosclerosis with antibody titers against mHSP65 in a large population-based study (122). In follow-up examinations, it became apparent that high antibody titers against mHSP65 are not only associated with atherosclerosis morbidity, but also with mortality due to cardiovascular disease (CVD) (123). In addition, anti-HSP60/65 antibodies were found to be significantly increased in the saliva of patients with periodontal disease that is also possibly associated with atherosclerosis (124). Since then, many groups have assessed the impact of antibodies against different HSPs in different clinical studies. Although the laboratory method measuring anti-HSP60 antibodies remained consistent between different publications, e.g., a classical sandwich ELISA, the clinical definition of atherosclerosis is a matter of constant change. In order to evaluate the role of immunoreactivity to HSP60, either large prospective studies focusing on clinical endpoints, like stroke and myocardial infarction, or good surrogate markers for atherosclerosis, like the sonographically determined IMT, are needed. Taking into account that acute cardiovascular events can alter anti-HSP60 antibody titers (125), interpretation of results from studies relying only on blood samples obtained directly after myocardial infarction or coronary angiography close to cardiovascular events has to be made very carefully. Anti-HSP60 antibody titers drop under these circumstances due to immune complex formation with sHSP60 released from disintegrated cells. Interestingly, probands who lacked a drop of anti-HSP65 antibody titers after coronary dilatation restenosed more frequently (126). Apart from differences in the clinical definition of atherosclerosis, different authors recruited probands in different age groups. Taking all these factors into account, the literature on antibodies to HSP is difficult to compare. Some of these data are summarized in Table 1. In conclusion, antibodies against mHSP65 (as a paradigmatic bacterial HSP60) have become a well established risk factor for prevalence, incidence, and mortality owing to cardiovascular events. According to the autoimmune hypothesis of atherogenesis, these antibodies are potentially cross-reactive to human HSP60. However, not all studies could demonstrate a positive correlation with antibodies against human HSP60. This might be due to the fact that only antibody subtypes reacting against certain epitopes shared by human and mycobacterial HSP 60/65 (127, 128), or complement activating antibodies, are the key players in atherogenesis. Positive correlation or lack of correlation in the studies summarized in Table 1 might thus reflect a higher or lower percentage of a certain subtype of anti-HSP60 antibodies. The effect of antibodies against HSP70 in atherosclerosis is still a matter of debate (129, 130). T-CELL REACTION AGAINST HSP60 IN ANIMAL EXPERIMENTS AND THE HUMAN SYSTEM The pathogenetic role and specificity of intralesional T-lymphocytes is

still the subject of intensive research. A central role of T cells specific against HSP60s has been established in animal experiments. Our results in mHSP65

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

INFLAMMATION IN ATHEROSCLEROSIS

377

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 1 Clinical studies with measurement of antibodies against different HSP60s, with method of assessment of atherosclerosis, the number of cases, and their age Antibodies against

Assessment of atherosclerosis

Cases/controls

Significant correlation

Agec

References

hHSP60

CA

276/129

Yes

58

(204)

hHSP60a

Familial predisposition

32/63

Yes

12

(205)

hHSP65

IMT

750

Yes

40

(122, 123)

mHSP65

Clinical + IMT + CA

203/76

Yes

60+

(125, 206)

mHSP65

CA pos./neg.

28/12

Yes

35+

(126)

mHSP65

CA pos./neg.

b

Yes

b

(207)

mHSP65

CA pos./neg.

357/67

No

58

hHSP60 mHSP65

Yes CVE

386/386

hHSP60 hHSP60

Yes

66

No Cerebral ischaemia

180/64

Yes

(209) (209)

69

mHSP65 hHSP 60

(208) (208)

(210) (210)

CVE in follow up

239/239

Chlamydial HSP60

Yes

48

No

Chlamydial HSP60

After CVE

179/100

HSP70

CA

99/99

HSP60, 65

(211) (211)

Yes

33+

(212)

No

52

(129)

Yes

(129)

HSP60, 65

CA pos. versus healthy

357/321

Yes

HSP60, 65, 70

IMT in hypertensive probands

b

No

50

(213)

HSP60, 70

Cerebral ischaemia (clinical)

292/485

No

72

(214)

HSP70

Peripheral atherosclerosis

61/21

Yes

60

(130)

HSP70

CA pos./neg.

421

No

57

(215)

a

Complement-activating antibodies.

b c

(129)

Data not available.

Mean age or age of youngest participant.

CA, coronary angiography; IMT, intima-media thickness; CVE, cardiovascular event.

immunized rabbits as well as in rabbits that—in addition to immunization— received a cholesterol-rich diet, have been summarized above. Also, immunization of LDL-R−/− mice with HSP65 induced specific T cell reactivity against mHSP65 as well as mammalian HSP60. Transfer of lymphocytes or

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

378

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

Figure 5 Bars reflect the impact (odds ratio per standard deviation unit) of the different vascular risk factors in the ARMY Study population (17- or 18-year-old clinically healthy males). Results from a stepwise logistic regression analysis of high IMT: Already in this young population, smoking is the most important risk factor. A high T cell reactivity against human HSP60 (hHSP60) equals the risk of an approximately 18 mm Hg rise in diastolic blood-pressure. Also, the humoral immune reaction against mycobacterial HSP65 (humoral reactivity) is an important risk factor. As shown in adults previously, HDL-cholesterol levels and alcohol consumption are protective factors, as is good pulmonary function measured by the maximum expiratory flow (MEF). ∗∗ P < 0.01; ∗ P < 0.05; (∗ ) P = 0.05 (130a).

isolated IgG of immunized mice into nonimmunized litter-mates enhanced lesion size (97). As mentioned, induction of tolerance against HSP65 by mucosal administration lowers the number of CD4+ T cells in plaques of cholesterol-fed LDL-R−/− mice, and can also lower the mitogenic in vitro response of lymph node T cells against HSP65 (98). In the course of the Bruneck Study (85, 86), we measured peripheral blood lymphocyte reaction against various HSPs60 in 50 males with, and 50 without, atherosclerosis, aged 50 years and above. Yet, no correlation could be found in this group with late stage atherosclerosis (B. Mayrl, unpublished results). On the other hand, specific T cell reactivity against HSP60 could be demonstrated as an independent major risk factor for early intima-media thickening in clinically healthy male 17- or 18-year-olds (ARMY Study). Antibodies against mHSP65 afforded less pronounced correlation with sonographically determinable surrogate of early atherosclerosis in this young population. The results of this study are illustrated in Figure 5 (130a). In summary, it has become increasingly clear that circulating HSP60-specific T cells are important in early atherogenesis, as their reactivity correlates with

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

INFLAMMATION IN ATHEROSCLEROSIS

379

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 2 The role of T cells and HSP60 in early and late atherosclerosis Early atherosclerosis

Late atherosclerosis

T cells in plaques are

Oligoclonal in mouse (47)

Polyclonal in human (216, 217)

T cells reactive against HSP60 in plaque

Demonstrated in rabbits (92)

Demonstrated in humans (160)

Peripheral blood T cells specific for HSP60

Positive correlation (130a)

No correlationa

Anti-HSP60 antibodies

Yes (130a)

Yes (122, 123)

a

B. Mayerl, S. Kiechl, A. Mayr, B. Grubeck-Loebenstein, G. Wick, unpublished results (Bruneck Study).

sonographically demonstrable arterial wall thickening in the young males of the ARMY Study, but not in the adult and old males of the Bruneck Study. Deducing from information obtained in other autoimmune disorders—like human autoimmune thyroiditis—lymphocytes specific against an autoantigen can mainly be detected in peripheral blood during the early phase of an autoimmune disease. As the inflammatory process continues, specific T cells can be found inside the target structure, i.e., the afflicted arteries, rather than in peripheral blood (131, 132). However, such similar information from early human lesions is still missing (Table 2).

IMMUNE REACTIONS TO oxLDL oxLDL Induces Innate Immune Response How does oxLDL trigger the events leading to the generation and/or enhancement of atherosclerotic lesions? Recently, there has been considerable progress in identifying the components of oxLDL that make it a ligand for macrophage receptors, including CD36, SR-B1, and CD14/ TLR4. For instance, the monoclonal antibody EO6 binds to intact oxLDL, blocks its binding and uptake by macrophages in a dose-dependent manner, and specifically prevents the binding of oxLDL to the scavenger receptors CD36 and SR-B1 (39, 133). Also, Miller et al. (134) demonstrated that minimally modified LDL binds to CD14, induces macrophage spreading via TLR4/MD2, and inhibits phagocytosis of apoptotic cells. Following TLR4/MD activation, NF-kB can be activated, and proinflammatory cytokines are released. These data strongly indicate a common innate immune response to apoptotic cells and oxLDL via competitive binding to macrophage scavenger receptor (135, 136). Therefore, oxLDL promotes foam cell formation participating in early lesion development in the intima mediated by innate immune reactions, which may progress to advanced lesions in the presence of other proatherogenic factors (137). One of the earliest responses induced by oxLDL is an increase in the expression of VCAM-1 on the arterial endothelial surface (26), a key adhesion molecule

19 Feb 2004

12:1

380

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

for monocytes and T cells. oxLDL is itself directly chemotactic for monocytes and T cells (3, 138). Among other biological effects, oxLDL is cytotoxic for endothelial cells (139), mitogenic for macrophages and smooth muscle cells, and it stimulates the release of MCP-1 and MCSF from endothelial cells (31). The oxidative modification hypothesis has been mentioned above and was extensively reviewed by Steinberg (137).

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Adaptive Immune Response to oxLDL As described above, circulating IgG and IgM antibodies against oxLDL, MDA– LDL and CuOx–LDL, are present in the plasma of animals and humans, and form immune complexes with oxLDL in atherosclerotic lesions (140, 141). The presence and titer of these antibodies closely correlate with atherosclerosis progression and regression, respectively, in murine models and also with the degree of lipid peroxidation (142). Several studies have shown that plasma titers of antibodies against oxLDL epitopes, particular of the IgG class, correlate with the presence of risk factors in patients with coronary heart disease (143). Because many variables affect such titers in humans, the clinical utility of such measurements remains to be determined. T cells isolated from human atherosclerotic plaques were shown to be specifically reactive to oxLDL (144). One fourth of all CD4+ T cells cloned from human plaques recognized oxLDL in an HLA-DR-restricted fashion. oxLDL-specific T cells are present in lymph nodes of ApoE−/− mice (145), which afford strong humoral, as well as cellular, immune responses to such modified lipoproteins (47). In humans, oxLDL induces activation of a subset of peripheral T cells. The immune response to oxLDL seems to play a pathogenetic role in atherosclerosis because lesion progression can be inhibited by immunization or induction of neonatal tolerance with oxLDL (59). It seems paradoxical that both tolerization and hyperimmunization can reduce the extent of disease; this may be due to the different effector pathways activated by these two kinds of treatment.

INFECTIONS, IMMUNITY, AND ATHEROSCLEROSIS It is well established that smoking, diabetes mellitus, hypertension, and high blood cholesterol levels are classical risk factors for atherosclerosis. On the other hand, growing evidence indicates that microorganisms play a role in the pathogenesis of the disease and may be a primary risk factor in people who do not suffer from other established risk factors (146). It was found that infectious organisms reside in the wall of atherosclerotic vessels, including CMV and C. pneumoniae. Seroepidemiological studies demonstrate an association between pathogen-specific IgG antibodies and atherosclerosis (147, 148). However, the data are inconsistent with other studies showing no increased risk for atherosclerosis (149). One possible explanation for this disparity is that infections contributing to atherosclerosis risk may depend, at least in part, on the host response to the pathogen, i.e., innate

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

381

and adaptive immune reactions. Trials by Zhu et al. (150), together with a clinical investigation by Espinola-Klein et al. (151), support the notion that a significant association exists between the number of infectious organisms a person has been exposed to and the extent of atherosclerosis. Epstein postulates the “pathogen burden” to be of significance, whereby multiple pathogens are involved in atherogenesis. The increased “pathogen load” augments the risk for CAD (152), implying that the coexistence of several organisms in the vessel wall has a synergistic effect in the induction of vessel injury. In vitro experiments have shown that C. pneumoniae infects macrophages, the vascular endothelium, and vascular SMC (153). This organism is capable of replicating within aortic endothelial cells. C. pneumoniae may also have a tropism for macrophages, which in turn accumulate in atherosclerotic plaques. This is supported by studies of postmortem specimens of vascular tissue, which found a high correlation between the distribution of atherosclerosis and C. pneumoniae as well as other microorganisms. In normal, nonatherosclerotic vessels these pathogens were absent (154). C. pneumoniae contributes to atherogenesis in a variety of ways. One mechanism is activating innate immune reactions by its ability to promote macrophages into becoming lipid-filled foam cells (155). This is due to an effect of LPS in the cell wall of the microorganism which, in one study, has been shown to be an independent factor in the accumulation of LDL by macrophages (156). It has also been reported that cHSP60 protein, a constituent of the bacteria, is involved in stimulating macrophages to secrete TNF-α and to synthesize MMPs (157). Thus, bacterial infections could also enhance lesion development via activation of innate immune responses. Bacterial infections can exacerbate a pre-existing plaque via activation of T lymphocytes and nonspecific inflammatory processes, which can lead to destabilization of the fibrous cap of the lesions (158). C. pneumoniae produces large amounts of HSP60 during infection, which is thought to play a role in atherogenesis (159). In fact, T cell lines established from human atherosclerotic plaques have been demonstrated to specifically react with bacterial products of C. pneumoniae, including membrane proteins and cHSP60 (160). These HSP reactive T cells and antibodies can in turn evoke autoimmune responses in humans owing to antigenic cross-reactivity.

THE VASCULAR-ASSOCIATED LYMPHOID TISSUE In the course of our immunohistological studies of early arteriosclerotic lesions, we originally also included, as a negative control, the carotid branching area from a single one-year-old baby who had died from an accident. Surprisingly, we observed accumulations of considerable numbers of mononuclear cells at those areas that are known to become predilection sites for atherosclerotic lesions later in life (16). We then embarked on a series of more detailed studies of this phenomenon, including carotid and other arterial specimens from many clinically healthy babies, children, and young adults aged 8 months to 16 years. These studies revealed that

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

382

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

mononuclear cell infiltration at these special sites started with T cells, followed by macrophages, SMC, and a few scattered mast cells. In very young children (8 months to 8 years), macrophages did not show the characteristics of foam cells, and extracellular lipid deposits were not yet detectable in the intima (161). This observation argues in favor of the pre-existence of mononuclear cell infiltrations that are not triggered by subendothelial lipid influx. Only older children showed the beginning of lipid accumulation paralleling that of healthy adults. In the CD3+ population, CD4+ cells generally predominate over CD8+ cells. Most T cells carry the TCR α/β, but TCR γ /δ + cells were again present in a proportion that significantly exceeds that in the peripheral blood of the same individual (161). In analogy to the MALT, we designated these mononuclear cell infiltrations as the VALT. Similar to the MALT, we tentatively assigned a similar function to the VALT, i.e., as a local immune-monitoring site for exogenous or autologous potentially harmful antigens passing by in the circulation. The mononuclear cells that make up the VALT are, of course, less abundant than those in accumulations of the MALT, e.g., the GALT, especially with respect to the lack of B cells (15, 162). Interestingly, however, the VALT has proven to be very rich in DC. We have recently made the unexpected discovery that the intima of arteries—but not of veins—contains a so-far-unreported network of DC that is especially dense in the VALT area, where it resembles that formed by Langerhans cells in the skin (32). Bobryshev and colleagues (23) earlier pointed out the existence of DC within advanced atherosclerotic lesions, whereas our studies were performed exclusively on arterial sections of children not yet affected by the disease. Comparing the DC phenotypes in atherosclerotic lesions and in healthy intima revealed a similar pattern of phenotypic characteristics, i.e., CD1a+ S100+ MHC II+ lag+, but a difference with respect to activation markers because DC in healthy arterial intima lack CD86 and CD83 (32). Because these latter two molecules, among others, are characteristic for mature DC that are able to stimulate naive T cells, it appears that immature DC found in the healthy intima transform into mature cells during atherogenesis. So far, we cannot present any functional data that would define the stimulators or timepoint of vascular-associated DC maturation because fresh arteries from children are fortunately only very rarely available. Finally, it has recently been shown that the vascular-associated DC may represent as a reservoir for HIV in AIDS patients (163) and pathogenic PrPsc in Creuzfeld-Jacob disease, respectively (164). In the latter case, this finding may be of relevance for the clarification of the question of how PrPsc from BSE-contaminated meat is transported from the gut to the central nervous system. Although we do not yet have any definitive data on the route of immigration of mononuclear cells to the VALT area, or the specificity of the involved T cells, indirect evidence based on the distribution of certain adhesion molecules on endothelial cells as well as the composition of ECM molecules may provide certain clues in this respect (165). A constitutive basic expression of ICAM-1 and P-selectin by endothelial cells in the VALT area of normal carotid arteries has been found, thus

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

383

confirming the work of other authors. ELAM-1 was not found on endothelial cells in any of the specimens investigated, explaining the lack of neutrophils that are known to preferentially adhere via this adhesion molecule. VCAM-1 expressed by both endothelial cells and by macrophages within the VALT may be instrumental in recruiting further mononuclear cells and the subsequent development of arteriosclerotic lesions. It should be re-emphasized here that different stress factors, notably the classical atherosclerosis risk factors, can simultaneously induce the expression of HSP60 as well as an upregulation of adhesion molecules ICAM-1, VCAM-1, and ELAM-1 (78). The ECM molecules, together with other factors, such as cytokines and growth factors, are known to crucially contribute to the microenvironment of any tissue, including arteries. The cells of the immune system are known to interact with ECM in many ways, resulting in the promotion or inhibition of their migration and function. They support cellular migration, e.g., collagen triple helices acting as conveyer belts for the movement of DC, whereas T cells show a similar pattern of migration depending on the three-dimensional structure of the ECM environment, e.g., via collagen-binding sites of VLA-1 and VLA-2. The ligand binding to these receptors induces T cell stimulation, resulting in the synthesis of cytokines and increased migration (166).

INFLAMMATION AND ATHEROGENESIS Inflammatory Mediators Although an association of infections with the development of atherosclerosis has been discussed for a long time, and was well supported by anecdotal evidence, strictly controlled epidemiological studies including the exact determination of mediators of inflammation have only been performed in recent years. In our own work within the framework of the Bruneck Study, chronic infection has been found to be significantly associated with the presence and extent of atherosclerotic lesions (38, 166a). Significant differences between study participants with and without chronic infections emerged with respect to the following serum markers of inflammation: white blood cell count, neutrophil count, CRP, α1-antitrypsin, ceruloplasmin, sICAM-1, sVCAM-1, sE-selectin, endotoxin, and sHSP60. When odds ratios (95% CIs) for new carotic plaques were calculated for chronic infection status according to subject subgroups and adjusted for age, sex and baseline atherosclerosis, only CRP, sVCAM-1, sE-selectin, endotoxin, and sHSP60 showed a significant association. In previous studies, CRP has been found to be elevated in patients with myocardial infarction and was also found to be an independent indicator for future coronary events (167). In our cohort, we did not determine IL-6 levels because it is considered to be a much less stabile marker of systemic inflammation than CRP. The role of inflammatory mediators in atherogenesis has been discussed both with respect to its inhibitory potential and its ability to promote (168, 169).

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

384

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

Proatherogenic mediators include IFN-γ , IL-8 (CXCL8), IL-12, IL-1β, IL-18, CD40–CD154, MCP-1, CCL-2, leukotriene P4 (LTP4), and 12/15 lipoxygenase products, whereas antiatherogenic properties have been assigned to IL-4, IL-10, PPR-γ , TGF- β, and PDGF-β. IFN-γ −/− × ApoE−/− double KO-mice develop less-pronounced atherosclerotic lesions than ApoE−/− only controls, and these are characterized by low abundance of macrophages and less-pronounced ECM deposition (170). IL-18, a Th-1 IFN-γ -inducing cytokine, promotes the development of atherosclerotic lesions in ApoE−/− mice, which are characterized by an excessively high number of activated (MHC II+) intralesional T cells (171). In contrast, the macrophagedeactivating cytokine IL-10 protects hypercholesterolemic LDL-R−/− mice from atherosclerosis (172). Considering what has been said so far about the cellular makeup of atherosclerotic lesions, it is not surprising that chemokines have been found to be of relevance in the development of the disease. Specifically, MCP-1 (CCL-2) and the expression of its receptor CCR-2 exert a proatherogenic effect (30). Activation and chemotaxis of monocytes by various chemokines, notably those of a CXC group, has been proven in vitro and in vivo, speaking for a role early in the disease process. Other members of the chemokine family, such as IFN-inducible protein-10 (CXCL-10), fractalkine (CX3 CR13), eotaxin (CCL-11) thymus- and activation-regulated chemokine (CCL-17), and macrophage-derived CCL-22 can be demonstrated in advanced lesions, although their potential role early in the disease process remains to be elucidated (168). These latter chemokines may also affect plaque stability and plaque remodeling, respectively. The possible contribution of fractalkine to atherogenesis has been deduced from the observation that certain polymorphisms in the CX3CR1 (human fractalkine receptor) gene are associated with an increased incidence of CAD (173). Involvement of LTP4, a potent leukocyte-attracting chemokine, has been suggested based on the antiatherogenic effect of an LTP4 receptor antagonist in LDL-R−/− mice. Furthermore, disruption of the 12/15-lipoxygenase gene diminished atherosclerosis in ApoE−/− mice (174). Finally, considerable attention has been paid in recent years to the CD40CD40 ligand (CD154) interaction as a mediator of inflammation in general, and during atherogenesis in particular. When CD40+ endothelial cells, macrophages, and SMC are treated with CD40-L, proatherogenic factors are induced in these cells, such as proinflammatory cytokines, chemokines, growth factors, and MMPs. When the CD40-CD154 interaction is blocked in LDL-R−/− mice with specific neutralizing monoclonal antibodies, the development of atherosclerosis is significantly reduced (175, 176). The role of CD40-CD40L interaction in atherosclerotic plaque rupture is discussed below.

Soluble HSP 60 In the Bruneck Study involving 826 patients, Xu et al. (77) demonstrated that sHSP60 levels were significantly elevated in subjects with prevalent/incident

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

385

Figure 6 Risk of carotid atherosclerosis according to baseline serum concentrations of sHSP60, chronic infection, and the logistic regression analysis of incident carotid atherosclerosis on sHSP60 categories and chronic infection (77).

carotid atherosclerosis and correlated with arterial IMT (Figure 6). Interestingly, sHSP60 was also correlated with anti-LPS, anti-Chlamydia, and anti-HSP60 antibodies, inflammation markers, and a history of chronic infections. The concentration of sHSP60 in some patients was >1 µg/ml, which is enough to produce extracellular-based signaling. Concomitantly, Pockley et al. (177) found that sHSP60 was increased in patients with borderline hypertension, and associated with an increased IMT and early atherosclerosis. Lewthwaite et al. (178) studied 229 civil servants and provided evidence that sHSP60 is associated with physiological and psychosocial stress. Previously, Sethna et al. (179) had reported that M. tuberculosis HSP60 could be detected in the sera of patients with tuberculosis. The source of the soluble HSP is currently not known, and the mechanisms of their release has not been identified. There are a number of theoretical possibilities. First, the presence of the infectious organism within the host cells could lead to increased synthesis of HSP as an immune defense mechanism by the host cell to protect itself from the effects of the bacteria. The effects these extracellular proteins exert on the body may include a cytokine-like activity (157). A second possibility could be that infectious agents such as C. pneumoniae, which eventually lyse the infected cell as part of their replication cycle, cause the release of these typically intracellular proteins (180). An argument in favor of this hypothesis is that sHSP60 levels are significantly correlated with antichlamydial antibodies (77), and both chlamydial and human HSP60s exist at high levels in human atherosclerotic lesions (159). Third, the elevation of soluble HSP could reflect the general inflammatory processes that are occurring within the vascular wall during CAD. Fourth, sHSP60 may be released from necrotic cells in the plaque, as studies have shown the presence of cell death within atherosclerotic lesions (181, 182). Lastly, cell-surface HSPs may be released into the circulation from apoptotic cells via the formation of microparticles (183–186), which have been detected in the circulation of patients suffering from acute coronary syndromes and in nonischemic patients.

19 Feb 2004

12:1

386

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

INFLAMMATION AND ATHEROSCLEROTIC PLAQUE RUPTURE

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-Reactive Protein CRP is an acute-phase protein that is involved in inflammatory processes. Levels of CRP were associated with the presence of carotid atherosclerosis in a large community-based cohort. Elevated CRP levels may increase the atherosclerotic burden, explaining part of the increased risk of cardiovascular events in individuals, particularly women (187). Liuzzo et al. (188) demonstrated early on that elevated CRP correlates with an adverse short-term prognosis in selected patients with coronary atherosclerosis. Half of all patients with coronary heart disease have persistently elevated CRP after discharge, a finding associated with recurrent episodes of plaque instability and infarction (37). CRP may not only be a marker of inflammation and atherosclerosis, it may also be an active component participating in atherogenesis (167, 189, 190, 191). It binds to lipoproteins and activates the complement system via the classical pathway. CRP deposits in the arterial wall early during lesion formation and is colocalized with the terminal complement complex. This suggests that CRP may promote atherosclerotic lesion formation by activating the complement system and is involved in foam cell formation, which may be caused in part by the uptake of CRP-opsonized LDL (191).

CD40/CD40L The CD40 receptor has been shown to be present on the surface of endothelial cells, smooth muscle cells, macrophages, T lymphocytes, and platelets within human atheroma (192, 193). The proatherogenic functions of CD40 ligation include augmented expression of matrix metalloproteinases, procoagulant tissue factor, chemokines, and cytokines (194–196). Indeed, interruption of CD40 signaling not only reduced the initiation and progression of atherosclerotic lesions in hypercholesterolemic mice in vivo (176, 197), but also modulated plaque architecture in ways that might lower the risk for causing thrombosis (175). Although in vitro and in vivo studies established that CD40 signaling participates in atherosclerosis, the initial trigger for CD40/CD40L expression within atheroma may be regulated by oxidized LDL. In addition to the 39 kDa cell membrane–associated form, CD40L also exists as a soluble protein, termed sCD40L (198). Although lacking the cytoplasmic, the transmembrane region, and parts of the extracellular domains, this, the soluble form of CD40L, is considered to possess biological activity. It was recently demonstrated that elevated plasma concentrations of sCD40L predict risk for future cardiovascular events (199). Of note, patients with unstable angina express higher sCD40L plasma levels than healthy individuals or patients with stable angina (200). Thus, CD40/CD40L may be a mediator in the inflammatory responses during the development of atherosclerosis.

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

INFLAMMATION IN ATHEROSCLEROSIS

387

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DIAGNOSTIC, PROGNOSTIC, AND THERAPEUTIC CONSEQUENCES It has been emphasized throughout this review that our own scientific interest in atherogenesis is highly focused on the earliest stages of the disease. Although it is common knowledge that numerous factors contribute to the pathohistological and clinical features of late and finally life-threatening complicated lesions, atherosclerosis—like any other disease—must start by a single pathogenic event, probably not necessarily the same in each patient. The elucidation of these earliest effector mechanisms stands at the center of our own scientific endeavors. We consider this issue not only to be of scientific interest, but also of utmost clinical importance. If this earliest stage can be appropriately diagnosed, most desirably, by a simple blood test, preventive and therapeutic measures could be taken to curtail further progression into more advanced, clinically relevant stages. Except for the genetically determined cases of hypercholesterolemia, we are convinced that the earliest events in atherogenesis are of an immunologic inflammatory nature with the reactivity of the adaptive and perhaps also innate immune system against HSP60 playing a primary role. Animal experiments as well as human studies performed by us and in other laboratories point to an initiation of the disease by HSP60-reactive T cells and an acceleration of the inflammatory intimal lesion development by humoral anti-HSP60 antibodies. A combination of HSP60-specific humoral and cellular immune reaction, therefore, had emerged as a new diagnostic parameter reflecting the risk to develop atherosclerosis independent of other classical risk factors. So far, the relative weight of microbial-human anti-HSP60 cross-reactivity and bona fide anti-HSP60 autoimmunity, respectively, has not been determined. Also, the identification of atherogenic B and T cell HSP60 epitopes is not yet concluded. Achieving these latter goals will allow for the design of more specific diagnostic tests. Furthermore, the induction of tolerance against such atherogenic peptides may minimize the risk for increased susceptibility to infections that may emerge when tolerance is induced against the whole HSP60 molecule. Finally, HSP60-specific anti-inflammatory and/or immunosuppressive measures may be indicated when early atherosclerosis is diagnosed. The significant association of serum levels of acute phase proteins as well as parameters reflecting an individual’s infectious load with sonographically demonstrable atherosclerotic lesions have prompted clinical trials aimed at assessing the effect of anti-inflammatory and antibiotic therapeutic strategies. In depth discussion of this is beyond the scope of the present review. It should be mentioned, however, that we have shown that aspirin leads to a downregulation of the expression of adhesion molecules but not HSP60 by endothelial cells in vitro (201). Furthermore, it is noteworthy that the immunosuppressive drug cyclosporin A has been identified as a potent inducer of HSP60 expression by endothelial cells (201), an observation that may explain the increased development of atherosclerosis in patients treated with this drug (202). The lipid-lowering drugs of the statin family have long been known to

15 Mar 2004

15:25

388

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

also have anti-inflammatory and immunosuppressive effects, the molecular basis of which may be a change in plasma membrane viscosity, and thus alteration of cell surface receptor function (203, 203a). Finally, it is conceivable that interference with HSP60 expression by endothelial cells may prevent atherosclerosis, and we are in the process of addressing this issue.

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONCLUSIONS In this review, we put our new autoimmune hypothesis for atherogenesis in the context of other data that speak for an important role of inflammatory-immunologic processes. Our particular focus of attention is on the very earliest stages of atherosclerosis, where HSP60 expression by arterial endothelial cells subjected to classical atherosclerosis risk factors, such as high blood pressure, smoking, increased blood cholesterol levels, diabetes, infections, etc., induce HSP60 expression at the known predilection sites for the development of atherosclerotic lesions. These HSP60 positive endothelial cells then become the target of pre-existent innate and adaptive cellular and humoral immunity against cross-reactive microbial HSP60 epitopes or bona fide autoimmunity to altered self-HSP60 (Figure 7). The first cells invading the intima have been identified as HSP60-reactive T cells followed by macrophages and SMC. Our concept is that HSP60-reactive T cells initiate the disease while anti-HSP60 antibodies play an accelerating role that still awaits confirmation in the human system. ACKNOWLEDGMENTS The work of the authors was supported by the Austrian Science Fund (FWF), the Merkur Insurance Company, and the Austrian Ministry of Defense. We would also like to acknowledge the contribution of those authors of our group over the years, the work of which is cited in the review. We thank Ms. Christina Mayerl and Dr. Gerald Pfister for providing Figures 1 and 3, respectively, Ms. Barbara Gschirr, Ms. Ilona Atzinger for the expert secretarial and artwork help, and Dr. Blair Henderson for the careful assessment of the manuscript for the proper use of the English language.

APPENDIX Abbreviatons ARMY Study—Atherosclerosis Risk-Factors in Male Youngsters Study BSE—bovine spongiform encephalopathy CAD—coronary artery disease CCR-2—chemokine receptor-2 CD—cluster of differentiation

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

INFLAMMATION IN ATHEROSCLEROSIS

389

CD40L—CD40 ligand CFA—complete Freund’s adjuvant cHSP60—chlamydial 60 kDa HSP CMV—cytomegalovirus

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CRP—C-reactive protein DC—dentiritic cells DNA—deoxyribonucleic acid EBV—Epstein Barr Virus ECM—extracellular matrix ELAM-1—endothelial leukocyte adhesion molecule 1 ERKs—extracellular signal-regulated kinases GALT—gut-associated lymphoid tissue GroEL—HSP60 of E. coli hHSP60—human 60 kDa HSP HLA—human leukocyte antigen HSF—heat-shock transcription factor HSP—heat-shock protein ICAM-1—intercellular adhesion molecule-1 IFA—incomplete Freund’s adjuvant IFN-γ —interferon gamma IgG—immunoglobulin G IL-2—interleukin-2 IMT—intima-media thickness JAK2-STAT-pathway—Janus kinase2 signal transducers and activators of transcription-pathway JNK/SAPK—c-Jun NH2-terminal kinase/stress-activated protein kinase LDL—low-density lipoproteins LFA-1—lymphocyte function–associated antigen-1 LPS—lipopolysaccharide MALT—mucosa-associated lymphoid tissue MAPK—mitogen-activated protein kinase MCP-1—macrophage chemoattractant protein-1 MCSF—monocyte colony-stimulating factor MDA-LDL—malondialdehyde LDL MD-LDL—malondialdehyde-modified LDL mHSP65—mycobacterial HSP65 MMP—matrix-metalloproteinase

19 Feb 2004

12:1

390

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

NF-κB—nuclear factor κB NK cells—natural killer cells OVA—ovalbumin oxLDL—oxidized LDL PAMPs—pathogen-associated molecular patterns PC—phosphatidylcholine

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PDAY study—Pathobiological Determinants of Atherosclerosis in Youth study PDGF-β—platelet-derived growth-factor beta POVPC—1-palmytoyl–2 (5-oxo-valeroyl)–sn glycerol-3 phosphorylcholine PPR—peroxisome proliferators–activated receptor γ PrPsc—prion protein PRR—pattern-recognition receptor RA—rheumatoid arthritis RT-PCR—reverse trascriptase polymerase chain reaction sHSP60—soluble form of HSP60 SMC—smooth muscle cell SR-B1—scavenger receptor B1 TCR—T cell receptor TGF-β—Transforming growth-factor beta Th1—T helper cell 1 TIMPs—tissue inhibitors of metalloproteinases TLR—Toll-like receptor TRLP—triglyceride-rich lipoprotein VALT—vascular associated lymphoid tissue VCAM-1—vascular cell adhesion molecule-1 VLA-1—very-late-activation antigen-1 β-2 GP-1—beta-2 glycoprotein-1 The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, et al. 1995. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis. A report from the Committee on Vascular Lesions of the

Council on Arteriosclerosis, American Heart Association. Arterioscler. Thromb. Vasc. Biol. 15:1512–31 2. Stary HC, Chandler AB, Glagov S, Guyton JR, Insull W, et al. 1994. A definition of initial, fatty streak, and

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

INFLAMMATION IN ATHEROSCLEROSIS

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

3.

4.

5.

6.

7.

8.

9.

10.

11.

intermediate lesions of atherosclerosis. A report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation 89:2462–78 Quinn MT, Parthasarathy S, Fong LG, Steinberg D. 1987. Oxidatively modified low density lipoproteins: a potential role in recruitment and retention of monocyte/macrophages during atherogenesis. Proc. Natl. Acad. Sci. USA 84:2995– 98 Harshyne LA, Zimmer MI, Watkins SC, Barratt-Boyes SM. 2003. A role for class A scavenger receptor in dendritic cell nibbling from live cells. J. Immunol. 170:2302–9 Nieto FJ. 1998. Infections and atherosclerosis: new clues from an old hypothesis? Am. J. Epidemiol. 148:937–48 Knoflach M, Mayrl B, Mayerl C, Sedivy R, Wick G. 2003. Atherosclerosis as a paradigmatic disease of the elderly: role of the immune system. Immunol. Allergy Clin. North Am. 23:117–32 Ross R, Glomset JA. 1976. The pathogenesis of atherosclerosis (first of two parts). N. Engl. J. Med. 295:369–77 Ross R, Glomset JA. 1976. The pathogenesis of atherosclerosis (second of two parts). N. Engl. J. Med. 295:420–25 Saikku P, Leinonen M, Mattila K, Ekman MR, Nieminen MS, et al. 1988. Serological evidence of an association of a novel Chlamydia, TWAR, with chronic coronary heart disease and acute myocardial infarction. Lancet 2:983–86 Shi W, Wang NJ, Shih DM, Sun VZ, Wang X, Lusis AJ. 2000. Determinants of atherosclerosis susceptibility in the C3H and C57BL/6 mouse model: evidence for involvement of endothelial cells but not blood cells or cholesterol metabolism. Circ. Res. 86:1078–84 Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. 1989. Beyond cholesterol. Modifications of lowdensity lipoprotein that increase its

12.

13.

14.

15.

16.

17.

18.

19.

391

atherogenicity. N. Engl. J. Med. 320: 915–24 Witztum JL, Steinberg D. 2001. The oxidative modification hypothesis of atherosclerosis: Does it hold for humans? Trends Cardiovasc. Med. 11:93–102 Benditt EP, Benditt JM. 1973. Evidence for a monoclonal origin of human atherosclerotic plaques. Proc. Natl. Acad. Sci. USA 70:1753–56 Wick G, Schett G, Amberger A, Kleindienst R, Xu Q. 1995. Is atherosclerosis an immunologically mediated disease? Immunol. Today 16:27–33 Wick G, Romen M, Amberger A, Metzler B, Mayr M, et al. 1997. Atherosclerosis, autoimmunity, and vascular-associated lymphoid tissue. FASEB J. 11:1199–207 Xu QB, Oberhuber G, Gruschwitz M, Wick G. 1990. Immunology of atherosclerosis: cellular composition and major histocompatibility complex class II antigen expression in aortic intima, fatty streaks, and atherosclerotic plaques in young and aged human specimens. Clin. Immunol. Immunopathol. 56:344–59 van der Wal AC, Das PK, Bentz vdB, van der Loos CM, Becker AE. 1989. Atherosclerotic lesions in humans. In situ immunophenotypic analysis suggesting an immune mediated response. Lab. Invest. 61:166–70 Kleindienst R, Xu Q, Willeit J, Waldenberger FR, Weimann S, Wick G. 1993. Immunology of atherosclerosis. Demonstration of heat shock protein 60 expression and T lymphocytes bearing alpha/beta or gamma/delta receptor in human atherosclerotic lesions. Am. J. Pathol. 142:1927–37 Buono C, Come CE, Stavrakis G, Maguire GF, Connelly PW, Lichtman AH. 2003. Influence of interferongamma on the extent and phenotype of diet-induced atherosclerosis in the LDLR-deficient mouse. Arterioscler. Thromb. Vasc. Biol. 23:454–60

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

392

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

20. Capron L. 1991. Leukocytes and arteriosclerosis. Arch. Mal Coeur Vaiss. 84:1845–50 21. Metzler B, Xu Q. 1997. The role of mast cells in atherosclerosis. Int. Arch. Allergy Immunol. 114:10–14 22. Wissler RW, Strong JP. 1998. Risk factors and progression of atherosclerosis in youth. PDAY Research Group. Pathological Determinants of Atherosclerosis in Youth. Am. J. Pathol. 153:1023–33 23. Bobryshev YV, Lord RS, Rainer S, Jamal OS, Munro VF. 1996. Vascular dendritic cells and atherosclerosis. Pathol. Res. Pract. 192:462–67 24. Aicher A, Heeschen C, Mohaupt M, Cooke JP, Zeiher AM, Dimmeler S. 2003. Nicotine strongly activates dendritic cell-mediated adaptive immunity: potential role for progression of atherosclerotic lesions. Circulation 107:604–11 25. Medzhitov R. 2001. Toll-like receptors and innate immunity. Nat. Rev. Immunol. 1:135–45 26. Cybulsky MI, Iiyama K, Li H, Zhu S, Chen M, et al. 2001. A major role for VCAM-1, but not ICAM-1, in early atherosclerosis. J. Clin. Invest. 107:1255–62 27. Johnson RC, Chapman SM, Dong ZM, Ordovas JM, Mayadas TN, et al. 1997. Absence of P-selectin delays fatty streak formation in mice. J. Clin. Invest. 99:1037–43 28. Collins RG, Velji R, Guevara NV, Hicks MJ, Chan L, Beaud AL. 2000. P-Selectin or intercellular adhesion molecule (ICAM)-1 deficiency substantially protects against atherosclerosis in apolipoprotein E-deficient mice. J. Exp. Med. 191:189–94 29. Gu L, Okada Y, Clinton SK, Gerard C, Sukhova GK, et al. 1998. Absence of monocyte chemoattractant protein-1 reduces atherosclerosis in low density lipoprotein receptor-deficient mice. Mol. Cell 2:275–81

30. Dawson TC, Kuziel WA, Osahar TA, Maeda N. 1999. Absence of CC chemokine receptor-2 reduces atherosclerosis in apolipoprotein E-deficient mice. Atherosclerosis 143:205–11 31. Cushing SD, Berliner JA, Valente AJ, Territo MC, Navab M, et al. 1990. Minimally modified low density lipoprotein induces monocyte chemotactic protein 1 in human endothelial cells and smooth muscle cells. Proc. Natl. Acad. Sci. USA 87:5134–38 32. Millonig G, Niederegger H, Rabl W, Hochleitner BW, Hoefer D, et al. 2001. Network of vascular-associated dendritic cells in intima of healthy young individuals. Arterioscler. Thromb. Vasc. Biol. 21:503–8 33. Chen W, Syldath U, Bellmann K, Burkart V, Kolb H. 1999. Human 60-kDa heatshock protein: a danger signal to the innate immune system. J. Immunol. 162:3212–19 34. Seifert PS, Kazatchkine MD. 1988. The complement system in atherosclerosis. Atherosclerosis 73:91–104 35. Buono C, Come CE, Witztum JL, Maguire GF, Connelly PW, et al. 2002. Influence of C3 deficiency on atherosclerosis. Circulation 105:3025–31 36. Ridker PM, Hennekens CH, Buring JE, Rifai N. 2000. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N. Engl. J. Med. 342:836– 43 37. Ridker PM, Stampfer MJ, Rifai N. 2001. Novel risk factors for systemic atherosclerosis: a comparison of C-reactive protein, fibrinogen, homocysteine, lipoprotein(a), and standard cholesterol screening as predictors of peripheral arterial disease. JAMA 285: 2481–85 38. Kiechl S, Egger G, Mayr M, Wiedermann CJ, Bonora E, et al. 2001. Chronic infections and the risk of carotid atherosclerosis: prospective results from

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

INFLAMMATION IN ATHEROSCLEROSIS

39.

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

40.

41.

42.

43.

44.

45.

46.

a large population study. Circulation 103:1064–70 Horkko S, Bird DA, Miller E, Itabe H, Leitinger N, et al. 1999. Monoclonal autoantibodies specific for oxidized phospholipids or oxidized phospholipidprotein adducts inhibit macrophage uptake of oxidized low-density lipoproteins. J. Clin. Invest. 103:117–28 Vos Q, Lees A, Wu ZQ, Snapper CM, Mond JJ. 2000. B-cell activation by Tcell-independent type 2 antigens as an integral part of the humoral immune response to pathogenic microorganisms. Immunol. Rev. 176:154–70 Binder CJ, Horkko S, Dewan A, Chang MK, Kieu EP, et al. 2003. Pneumococcal vaccination decreases atherosclerotic lesion formation: molecular mimicry between Streptococcus pneumoniae and oxidized LDL. Nat. Med. 9:736– 43 Ishida T, Yasukawa K, Kojima H, Harashima H, Kiwada H. 2001. Effect of cholesterol content in activation of the classical versus the alternative pathway of rat complement system induced by hydrogenated egg phosphatidylcholinebased liposomes. Int. J. Pharm. 224:69– 79 Seifert PS, Hansson GK. 1989. Complement receptors and regulatory proteins in human atherosclerotic lesions. Arteriosclerosis 9:802–11 Vlaicu R, Rus HG, Niculescu F, Cristea A. 1985. Immunoglobulins and complement components in human aortic atherosclerotic intima. Atherosclerosis 55:35–50 Minick CR, Murphy GE. 1973. Experimental induction of atheroarteriosclerosis by the synergy of allergic injury to arteries and lipid-rich diet. II. Effect of repeatedly injected foreign protein in rabbits fed a lipid-rich, cholesterol-poor diet. Am. J. Pathol. 73:265–300 Stemme S, Rymo L, Hansson GK. 1991. Polyclonal origin of T lymphocytes in

47.

48.

49.

50.

51.

52.

53.

54.

393

human atherosclerotic plaques. Lab. Invest. 65:654–60 Paulsson G, Zhou X, Tornquist E, Hansson GK. 2000. Oligoclonal T cell expansions in atherosclerotic lesions of apolipoprotein E-deficient mice. Arterioscler. Thromb. Vasc. Biol. 20:10–17 Pinderski LJ, Fischbein MP, Subbanagounder G, Fishbein MC, Kubo N, et al. 2002. Overexpression of interleukin-10 by activated T lymphocytes inhibits atherosclerosis in LDL receptor-deficient mice by altering lymphocyte and macrophage phenotypes. Circ. Res. 90:1064–71 Huber SA, Sakkinen P, David C, Newell MK, Tracy RP. 2001. T helper-cell phenotype regulates atherosclerosis in mice under conditions of mild hypercholesterolemia. Circulation 103:2610–16 Xu Q, Wick G, Wachter H, Reibnegger G. 1994. Relationship among serum hsp65 antibodies, neopterin, autoantibodies and atherosclerosis. Pteridines 5:139–41 Smith DA, Zouridakis EG, Mariani M, Fredericks S, Cole D, Kaski JC. 2003. Neopterin levels in patients with coronary artery disease are independent of Chlamydia pneumoniae seropositivity. Am. Heart J. 146:69–74 Teramoto S, Yamamoto H, Ouchi Y. 2003. Increased C-reactive protein and increased plasma interleukin-6 may synergistically affect the progression of coronary atherosclerosis in obstructive sleep apnea syndrome. Circulation 107:E40 Mallat Z, Gojova A, MarchiolFournigault C, Esposito B, Kamate C, et al. 2001. Inhibition of transforming growth factor-beta signaling accelerates atherosclerosis and induces an unstable plaque phenotype in mice. Circ. Res. 89:930–34 Dansky HM, Charlton SA, Harper MM, Smith JD. 1997. T and B lymphocytes play a minor role in atherosclerotic

19 Feb 2004

12:1

394

55.

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

56.

57.

58.

59.

60.

61.

62.

63.

64.

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

plaque formation in the apolipoprotein E-deficient mouse. Proc. Natl. Acad. Sci. USA 94:4642–46 Daugherty A, Pure E, Delfel-Butteiger D, Chen S, Leferovich J, et al. 1997. The effects of total lymphocyte deficiency on the extent of atherosclerosis in apolipoprotein E−/− mice. J. Clin. Invest. 100:1575–80 Wick G, Xu Q. 2002. Autoimmunity in Atherosclerosis. In The Molecular Pathology of Autoimmune Diseases, ed. CA Bona, AN Theofilopoulos, pp. 965– 77. New York: Taylor & Francis Metzler B, Mayr M, Dietrich H, Singh M, Wiebe E, et al. 1999. Inhibition of arteriosclerosis by T-cell depletion in normocholesterolemic rabbits immunized with heat shock protein 65. Arterioscler. Thromb. Vasc. Biol. 19:1905–11 Wick G. 2001. Correction for Metzler et al., Arterioscler Thromb Vasc Biol 19 (8) 1905–11. Arterioscler. Thromb. Vasc. Biol. 21:1706 Caligiuri G, Nicoletti A, Poirier B, Hansson GK. 2002. Protective immunity against atherosclerosis carried by B cells of hypercholesterolemic mice. J. Clin. Invest. 109:745–53 Nicoletti A, Caligiuri G, Paulsson G, Hansson GK. 1999. Functionality of specific immunity in atherosclerosis. Am. Heart J. 138:S438–43 Witztum JL. 2002. Splenic immunity and atherosclerosis: a glimpse into a novel paradigm? J. Clin. Invest. 109:721–24 Wick G, Xu Q. 1999. Atherosclerosis— an autoimmune disease. Exp. Gerontol. 34:559–66 George J, Harats D, Gilburd B, Afek A, Shaish A, et al. 2000. Adoptive transfer of beta(2)-glycoprotein I-reactive lymphocytes enhances early atherosclerosis in LDL receptor-deficient mice. Circulation 102:1822–27 Palinski W, Miller E, Witztum JL. 1995. Immunization of low density lipoprotein (LDL) receptor-deficient rabbits with

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

homologous malondialdehyde-modified LDL reduces atherogenesis. Proc. Natl. Acad. Sci. USA 92:821–25 Uusitupa MI, Niskanen L, Luoma J, Vilja P, Mercuri M, et al. 1996. Autoantibodies against oxidized LDL do not predict atherosclerotic vascular disease in noninsulin-dependent diabetes mellitus. Arterioscler. Thromb. Vasc. Biol. 16:1236– 42 Witztum JL. 1997. Immunological response to oxidized LDL. Atherosclerosis 131(Suppl.):S9–11 Craig EA, Gambill BD, Nelson RJ. 1993. Heat shock proteins: molecular chaperones of protein biogenesis. Microbiol. Rev. 57:402–14 Benjamin IJ, McMillan DR. 1998. Stress (heat shock) proteins: molecular chaperones in cardiovascular biology and disease. Circ. Res. 83:117–32 Xu Q, Wick G. 1996. The role of heat shock proteins in protection and pathophysiology of the arterial wall. Mol. Med. Today 2:372–79 Hartl FU, Hayer-Hartl M. 2002. Molecular chaperones in the cytosol: from nascent chain to folded protein. Science 295:1852–58 Young RA, Elliott TJ. 1989. Stress proteins, infection, and immune surveillance. Cell 59:5–8 Rocnik E, Chow LH, Pickering JG. 2000. Heat shock protein 47 is expressed in fibrous regions of human atheroma and is regulated by growth factors and oxidized low-density lipoprotein. Circulation 101:1229–33 Johnson AD, Berberian PA, Tytell M, Bond MG. 1993. Atherosclerosis alters the localization of HSP70 in human and macaque aortas. Exp. Mol. Pathol. 58:155–68 Bartz SR, Pauza CD, Ivanyi J, Jindal S, Welch WJ, Malkovsky M. 1994. An Hsp60 related protein is associated with purified HIV and SIV. J. Med. Primatol. 23:151–54

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS 75. Wick G, Kleindienst R, Schett G, Amberger A, Xu Q. 1995. Role of heat shock protein 65/60 in the pathogenesis of atherosclerosis. Int. Arch. Allergy Immunol. 107:130–31 76. Liao DF, Jin ZG, Baas AS, Daum G, Gygi SP, et al. 2000. Purification and identification of secreted oxidative stress-induced factors from vascular smooth muscle cells. J. Biol. Chem. 275:189–96 77. Xu Q, Schett G, Perschinka H, Mayr M, Egger G, et al. 2000. Serum soluble heat shock protein 60 is elevated in subjects with atherosclerosis in a general population. Circulation 102:14–20 78. Amberger A, Maczek C, Jurgens G, Michaelis D, Schett G, et al. 1997. Co-expression of ICAM-1, VCAM-1, ELAM-1 and Hsp60 in human arterial and venous endothelial cells in response to cytokines and oxidized lowdensity lipoproteins. Cell Stress Chaperones. 2:94–103 79. Soltys BJ, Gupta RS. 1997. Cell surface localization of the 60 kDa heat shock chaperonin protein (hsp60) in mammalian cells. Cell Biol. Int. 21:315– 20 80. Xu Q, Schett G, Seitz CS, Hu Y, Gupta RS, Wick G. 1994. Surface staining and cytotoxic activity of heat-shock protein 60 antibody in stressed aortic endothelial cells. Circ. Res. 75:1078–85 81. Seitz CS, Kleindienst R, Xu Q, Wick G. 1996. Coexpression of heat-shock protein 60 and intercellular-adhesion molecule-1 is related to increased adhesion of monocytes and T cells to aortic endothelium of rats in response to endotoxin. Lab. Invest. 74:241–52 82. Xu Q. 2002. Role of heat shock proteins in atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 22:1547–59 83. Bulut Y, Faure E, Thomas L, Karahashi H, Michelsen KS, et al. 2002. Chlamydial heat shock protein 60 activates macrophages and endothelial cells

84.

85.

86.

87.

88.

89.

90.

91.

92.

395

through Toll-like receptor 4 and MD2 in a MyD88-dependent pathway. J. Immunol. 168:1435–40 Habich C, Baumgart K, Kolb H, Burkart V. 2002. The receptor for heat shock protein 60 on macrophages is saturable, specific, and distinct from receptors for other heat shock proteins. J. Immunol. 168:569–76 Kiechl S, Willeit J. 1999. The natural course of atherosclerosis. Part I: incidence and progression. Arterioscler. Thromb. Vasc. Biol. 19:1484–90 Kiechl S, Willeit J. 1999. The natural course of atherosclerosis. Part II: vascular remodeling. Bruneck Study Group. Arterioscler. Thromb. Vasc. Biol. 19:1491–98 Kiechl S, Lorenz E, Reindl M, Wiedermann CJ, Oberhollenzer F, et al. 2002. Toll-like receptor 4 polymorphisms and atherogenesis. N. Engl. J. Med. 347:185– 92 Xu Q, Dietrich H, Steiner HJ, Gown AM, Schoel B, et al. 1992. Induction of arteriosclerosis in normocholesterolemic rabbits by immunization with heat shock protein 65. Arterioscler. Thromb. Vasc. Biol. 12:789–99 Goodall JC, Bledsoe P, Gaston JS. 1999. Tracking antigen-specific human T lymphocytes in rheumatoid arthritis by T cell receptor analysis. Hum. Immunol. 60:798–805 Van der Zee R, Anderton SM, Prakken AB, Liesbeth Paul AG, van Eden W. 1998. T cell responses to conserved bacterial heat-shock-protein epitopes induce resistance in experimental autoimmunity. Semin. Immunol. 10:35–41 Xu Q, Kleindienst R, Schett G, Waitz W, Jindal S, et al. 1996. Regression of arteriosclerotic lesions induced by immunization with heat shock protein 65-containing material in normocholesterolemic, but not hypercholesterolemic, rabbits. Atherosclerosis 123:145–55 Xu Q, Kleindienst R, Waitz W, Dietrich

19 Feb 2004

12:1

396

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

93.

94.

95.

96.

97.

98.

99.

100.

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

H, Wick G. 1993. Increased expression of heat shock protein 65 coincides with a population of infiltrating T lymphocytes in atherosclerotic lesions of rabbits specifically responding to heat shock protein 65. J. Clin. Invest. 91:2693–702 Zhang SH, Reddick RL, Piedrahita JA, Maeda N. 1992. Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein E. Science 258:468–71 Breslow JL. 1993. Transgenic mouse models of lipoprotein metabolism and atherosclerosis. Proc. Natl. Acad. Sci. USA 90:8314–18 George J, Shoenfeld Y, Afek A, Gilburd B, Keren P, et al. 1999. Enhanced fatty streak formation in C57BL/6J mice by immunization with heat shock protein65. Arterioscler. Thromb. Vasc. Biol. 19:505–10 George J, Afek A, Gilburd B, Levkovitz H, Shaish A, et al. 1998. Hyperimmunization of apo-E-deficient mice with homologous malondialdehyde low-density lipoprotein suppresses early atherogenesis. Atherosclerosis 138(1):147–52 George J, Afek A, Gilburd B, Shoenfeld Y, Harats D. 2001. Cellular and humoral immune responses to heat shock protein 65 are both involved in promoting fattystreak formation in LDL-receptor deficient mice. J. Am. Coll. Cardiol. 38:900– 5 Harats D, Yacov N, Gilburd B, Shoenfeld Y, George J. 2002. Oral tolerance with heat shock protein 65 attenuates Mycobacterium tuberculosis-induced and high-fat-diet-driven atherosclerotic lesions. J. Am. Coll. Cardiol. 40:1333–38 Maron R, Sukhova G, Faria AM, Hoffmann E, Mach F, et al. 2002. Mucosal administration of heat shock protein-65 decreases atherosclerosis and inflammation in aortic arch of low-density lipoprotein receptor-deficient mice. Circulation 106:1708–15 Barker RN, Easterfield AJ, Allen RF,

101.

102.

103.

104.

105.

106.

107.

108.

109.

Wells AD, Elson CJ, Thompson SJ. 1996. B- and T-cell autoantigens in pristane-induced arthritis. Immunology 89:189–94 Anderton SM, Van der Zee R, Prakken B, Noordzij A, van Eden W. 1995. Activation of T cells recognizing self 60-kD heat shock protein can protect against experimental arthritis. J. Exp. Med. 181:943–52 Gaston JS. 1998. Role of T-cells in the development of arthritis. Clin. Sci. (London) 95:19–31 Berberian PA, Myers W, Tytell M, Challa V, Bond MG. 1990. Immunohistochemical localization of heat shock protein70 in normal-appearing and atherosclerotic specimens of human arteries. Am. J. Pathol. 136:71–80 Bobryshev YV, Lord RS. 2002. Expression of heat shock protein-70 by dendritic cells in the arterial intima and its potential significance in atherogenesis. J. Vasc. Surg. 35:368–75 Kanwar RK, Kanwar JR, Wang D, Ormrod DJ, Krissansen GW. 2001. Temporal expression of heat shock proteins 60 and 70 at lesion-prone sites during atherogenesis in ApoE-deficient mice. Arterioscler. Thromb. Vasc. Biol. 21:1991–97 Baler R, Dahl G, Voellmy R. 1993. Activation of human heat shock genes is accompanied by oligomerization, modification, and rapid translocation of heat shock transcription factor HSF1. Mol. Cell Biol. 13:2486–96 Pirkkala L, Nykanen P, Sistonen L. 2001. Roles of the heat shock transcription factors in regulation of the heat shock response and beyond. FASEB J. 15:1118– 31 Metzler B, Abia R, Ahmad M, Wernig F, Pachinger O, Hu Y, Xu Q. 2003. Activation of heat shock transcription factor 1 in atherosclerosis. Am. J. Pathol. 162:1669–76 Chu B, Soncin F, Price BD, Stevenson MA, Calderwood SK. 1996.

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

INFLAMMATION IN ATHEROSCLEROSIS

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

110.

111.

112.

113.

114.

115.

116.

117.

Sequential phosphorylation by mitogenactivated protein kinase and glycogen synthase kinase 3 represses transcriptional activation by heat shock factor-1. J. Biol. Chem. 271:30847–57 Hu Y, Metzler B, Xu Q. 1997. Discordant activation of stress-activated protein kinases or c-Jun NH2-terminal protein kinases in tissues of heat-stressed mice. J. Biol. Chem. 272:9113–19 Guyton KZ, Liu Y, Gorospe M, Xu Q, Holbrook NJ. 1996. Activation of mitogen-activated protein kinase by H2O2. Role in cell survival following oxidant injury. J. Biol. Chem. 271:4138–42 Metzler B, Li C, Hu Y, Sturm G, Ghaffari-Tabrizi N, Xu Q. 1999. LDL stimulates mitogen-activated protein kinase phosphatase-1 expression, independent of LDL receptors, in vascular smooth muscle cells. Arterioscler. Thromb. Vasc. Biol. 19:1862–71 Metzler B, Hu Y, Sturm G, Wick G, Xu Q. 1998. Induction of mitogen-activated protein kinase phosphatase-1 by arachidonic acid in vascular smooth muscle cells. J. Biol. Chem. 273:33320–26 Hu Y, Dietrich H, Metzler B, Wick G, Xu Q. 2000. Hyperexpression and activation of extracellular signal-regulated kinases (ERK1/2) in atherosclerotic lesions of cholesterol-fed rabbits. Arterioscler. Thromb. Vasc. Biol. 20:18–26 Metzler B, Hu Y, Dietrich H, Xu Q. 2000. Increased expression and activation of stress-activated protein kinases/cJun NH(2)-terminal protein kinases in atherosclerotic lesions coincide with p53. Am. J. Pathol. 156:1875–86 Li C, Hu Y, Mayr M, Xu Q. 1999. Cyclic strain stress-induced mitogen-activated protein kinase (MAPK) phosphatase 1 expression in vascular smooth muscle cells is regulated by Ras/Rac-MAPK pathways. J. Biol. Chem. 274:25273–80 Madamanchi NR, Li S, Patterson C, Runge MS. 2001. Reactive oxygen species regulate heat-shock protein 70

118.

119.

120.

121.

122.

123.

124.

125.

397

via the JAK/STAT pathway. Arterioscler. Thromb. Vasc. Biol. 21:321–26 Di Cesare S, Poccia F, Mastino A, Colizzi V. 1992. Surface expressed heatshock proteins by stressed or human immunodeficiency virus (HIV)-infected lymphoid cells represent the target for antibody-dependent cellular cytotoxicity. Immunology 76:341–43 Soltys BJ, Gupta RS. 1996. Immunoelectron microscopic localization of the 60-kDa heat shock chaperonin protein (Hsp60) in mammalian cells. Exp. Cell Res. 222:16–27 Khan IU, Wallin R, Gupta RS, Kammer GM. 1998. Protein kinase A-catalyzed phosphorylation of heat shock protein 60 chaperone regulates its attachment to histone 2B in the T lymphocyte plasma membrane. Proc. Natl. Acad. Sci. USA 95:10425–30 Schett G, Metzler B, Kleindienst R, Amberger A, Recheis H, et al. 1999. Myocardial injury leads to a release of heat shock protein (hsp) 60 and a suppression of the anti-hsp65 immune response. Cardiovasc. Res. 42:685–95 Xu Q, Willeit J, Marosi M, Kleindienst R, Oberhollenzer F, et al. 1993. Association of serum antibodies to heat-shock protein 65 with carotid atherosclerosis. Lancet 341:255–59 Xu Q, Kiechl S, Mayr M, Metzler B, Egger G, et al. 1999. Association of serum antibodies to heat-shock protein 65 with carotid atherosclerosis: clinical significance determined in a follow-up study. Circulation 100:1169–74 Schett G, Metzler B, Kleindienst R, Moschen I, Hattmannsdorfer R, et al. 1997. Salivary anti-hsp65 antibodies as a diagnostic marker for gingivitis and a possible link to atherosclerosis. Int. Arch. Allergy Immunol. 114:246–50 Hoppichler F, Lechleitner M, Traweger C, Schett G, Dzien A, et al. 1996. Changes of serum antibodies to heatshock protein 65 in coronary heart

19 Feb 2004

12:1

398

126.

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

127.

128.

129.

130.

130a.

131.

132.

133.

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

disease and acute myocardial infarction. Atherosclerosis 126:333–38 Mukherjee M, De Benedictis C, Jewitt D, Kakkar VV. 1996. Association of antibodies to heat-shock protein-65 with percutaneous transluminal coronary angioplasty and subsequent restenosis. Thromb. Haemost. 75:258–60 Metzler B, Schett G, Kleindienst R, Van der Zee R, Ottenhoff T, Hajeer A, et al. 1997. Epitope specificity of anti-heat shock protein 65/60 serum antibodies in atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 17:536–41 Perschinka H, Mayr M, Millonig G, Mayerl C, Van der Zee R, Morrison SG, et al. 2003. Cross-reactive B-cell epitopes of microbial and human heat shock protein 60/65 in atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 23:1060–65 Kocsis J, Veres A, Vatay A, Duba J, Karadi I, et al. 2002. Antibodies against the human heat shock protein hsp70 in patients with severe coronary artery disease. Immunol. Invest. 31:219–31 Chan YC, Shukla N, Abdus-Samee M, Berwanger CS, Stanford J, et al. 1999. Anti-heat-shock protein 70 kDa antibodies in vascular patients. Eur. J. Vasc. Endovasc. Surg. 18:381–85 Knoflach M, Kiechl S, Kind M, Said M, Sief R, et al. 2003. Cardiovascular risk factors and atherosclerosis in young males: ARMY study (Atherosclerosis Risk-Factors in Male Youngsters). Circulation 108(9):1064–69 Wick G, Kromer G, Neu N, Fassler R, Ziemiecki A, et al. 1987. The multifactorial pathogenesis of autoimmune disease. Immunol. Lett. 16:249–57 Mackenzie WA, Davies TF. 1987. An intrathyroidal T-cell clone specifically cytotoxic for human thyroid cells. Immunology 61:101–3 Boullier A, Gillotte KL, Horkko S, Green SR, Friedman P, et al. 2000. The binding of oxidized low density lipoprotein to mouse CD36 is mediated in part by

134.

135.

136.

137.

138.

139.

140.

141.

142.

oxidized phospholipids that are associated with both the lipid and protein moieties of the lipoprotein. J. Biol. Chem. 275:9163–69 Miller YI, Viriyakosol S, Binder CJ, Feramisco JR, Kirkland TN, Witztum JL. 2003. Minimally modified LDL binds to CD14, induces macrophage spreading via TLR4/MD-2, and inhibits phagocytosis of apoptotic cells. J. Biol. Chem. 278:1561–68 Boullier A, Bird DA, Chang MK, Dennis EA, Friedman P, et al. 2001. Scavenger receptors, oxidized LDL, and atherosclerosis. Ann. NY Acad. Sci. 947:214–22 Podrez EA, Poliakov E, Shen Z, Zhang R, Deng Y, et al. 2002. A novel family of atherogenic oxidized phospholipids promotes macrophage foam cell formation via the scavenger receptor CD36 and is enriched in atherosclerotic lesions. J. Biol. Chem. 277:38517–23 Steinberg D. 2002. Atherogenesis in perspective: hypercholesterolemia and inflammation as partners in crime. Nat. Med. 8:1211–17 McMurray HF, Parthasarathy S, Steinberg D. 1993. Oxidatively modified low density lipoprotein is a chemoattractant for human T lymphocytes. J. Clin. Invest. 92:1004–8 Hessler JR, Morel DW, Lewis LJ, Chisolm GM. 1983. Lipoprotein oxidation and lipoprotein-induced cytotoxicity. Arteriosclerosis 3:215–22 Shaw PX, Horkko S, Chang MK, Curtiss LK, Palinski W, et al. 2000. Natural antibodies with the T15 idiotype may act in atherosclerosis, apoptotic clearance, and protective immunity. J. Clin. Invest. 105:1731–40 Yla-Herttuala S, Palinski W, Butler SW, Picard S, Steinberg D, Witztum JL. 1994. Rabbit and human atherosclerotic lesions contain IgG that recognizes epitopes of oxidized LDL. Arterioscler. Thromb. 14:32–40 Cyrus T, Pratico D, Zhao L, Witztum

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

INFLAMMATION IN ATHEROSCLEROSIS

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

143.

144.

145.

146.

147.

148.

149.

150.

151.

JL, Rader DJ, et al. 2001. Absence of 12/15-lipoxygenase expression decreases lipid peroxidation and atherogenesis in apolipoprotein e-deficient mice. Circulation 103:2277–82 Palinski W, Witztum JL. 2000. Immune responses to oxidative neoepitopes on LDL and phospholipids modulate the development of atherosclerosis. J. Intern. Med. 247:371–80 Stemme S, Faber B, Holm J, Wiklund O, Witztum JL, Hansson GK. 1995. T lymphocytes from human atherosclerotic plaques recognize oxidized low density lipoprotein. Proc. Natl. Acad. Sci. USA 92:3893–97 Zhou X, Stemme S, Hansson GK. 1996. Evidence for a local immune response in atherosclerosis. CD4+ T cells infiltrate lesions of apolipoprotein-Edeficient mice. Am. J. Pathol. 149:359– 66 Epstein SE, Zhou YF, Zhu J. 1999. Infection and atherosclerosis: emerging mechanistic paradigms. Circulation 100:E20–28 Folsom AR, Nieto FJ, Sorlie P, Chambless LE, Graham DY. 1998. Helicobacter pylori seropositivity and coronary heart disease incidence. Atherosclerosis Risk In Communities (ARIC) Study Investigators. Circulation 98:845–50 Danesh J, Collins R, Peto R. 1997. Chronic infections and coronary heart disease: is there a link? Lancet 350:430– 36 Danesh J, Peto R. 1998. Risk factors for coronary heart disease and infection with Helicobacter pylori: meta-analysis of 18 studies. BMJ 316:1130–32 Zhu J, Nieto FJ, Horne BD, Anderson JL, Muhlestein JB, Epstein SE. 2001. Prospective study of pathogen burden and risk of myocardial infarction or death. Circulation 103:45–51 Espinola-Klein C, Rupprecht HJ, Blankenberg S, Bickel C, Kopp H, et al. 2002. Impact of infectious burden

152.

153.

154.

155.

156.

157.

158.

159.

160.

399

on extent and long-term prognosis of atherosclerosis. Circulation 105:15– 21 Epstein SE, Zhu J, Burnett MS, Zhou YF, Vercellotti G, Hajjar D. 2000. Infection and atherosclerosis: potential roles of pathogen burden and molecular mimicry. Arterioscler. Thromb. Vasc. Biol. 20:1417–20 Gaydos CA, Summersgill JT, Sahney NN, Ramirez JA, Quinn TC. 1996. Replication of Chlamydia pneumoniae in vitro in human macrophages, endothelial cells, and aortic artery smooth muscle cells. Infect. Immun. 64:1614–20 Shor A. 2001. Mechanism of arterial infection by Chlamydia pneumoniae. Circulation 104:E75 Davidson M, Kuo CC, Middaugh JP, Campbell LA, Wang SP, et al. 1998. Confirmed previous infection with Chlamydia pneumoniae (TWAR) and its presence in early coronary atherosclerosis. Circulation 98:628–33 Hu H, Pierce GN, Zhong G. 1999. The atherogenic effects of chlamydia are dependent on serum cholesterol and specific to Chlamydia pneumoniae. J. Clin. Invest. 103:747–53 Kol A, Bourcier T, Lichtman AH, Libby P. 1999. Chlamydial and human heat shock protein 60s activate human vascular endothelium, smooth muscle cells, and macrophages. J. Clin. Invest. 103:571–77 Gura T. 1998. Infections: a cause of artery-clogging plaques? Science 281:35–37 Kol A, Sukhova GK, Lichtman AH, Libby P. 1998. Chlamydial heat shock protein 60 localizes in human atheroma and regulates macrophage tumor necrosis factor-alpha and matrix metalloproteinase expression. Circulation 98:300– 7 Curry AJ, Portig I, Goodall JC, Kirkpatrick PJ, Gaston JS. 2000. T lymphocyte lines isolated from atheromatous

19 Feb 2004

12:1

400

161.

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

162.

163.

164.

165.

166.

166a.

167.

168.

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

plaque contain cells capable of responding to Chlamydia antigens. Clin. Exp. Immunol. 121:261–69 Millonig G, Malcom GT, Wick G. 2002. Early inflammatory-immunological lesions in juvenile atherosclerosis from the Pathobiological Determinants of Atherosclerosis in Youth (PDAY)-study. Atherosclerosis 160:441–48 Waltner-Romen M, Falkensammer G, Rabl W, Wick G. 1998. A previously unrecognized site of local accumulation of mononuclear cells. The vascular-associated lymphoid tissue. J. Histochem. Cytochem. 46:1347–50 Bobryshev YV. 2000. Identification of HIV-1 in the aortic wall of AIDS patients. Atherosclerosis 152:529–30 Koperek O, Kovacs GG, Ritchie D, Ironside JW, Budka H, Wick G. 2002. Disease-associated prion protein in vessel walls. Am. J. Pathol. 161:1979–84 Millonig G, Schwentner C, Mueller P, Mayerl C, Wick G. 2001. The vascularassociated lymphoid tissue: a new site of local immunity. Curr. Opin. Lipidol. 12:547–53 Savino W, Villa-Verde DM, LannesVieira J. 1993. Extracellular matrix proteins in intrathymic T-cell migration and differentiation? Immunol. Today 14:158–61 Mayr M, Kiechl S, Willeit J, Wick G, Xu Q. 2000. Infections, immunity, and atherosclerosis: associations of antibodies to Chlamydia pneumoniae, Helicobacter pylori, and cytomegalovirus with immune reactions to heat-shock protein 60 and carotid or femoral atherosclerosis. Circulation 102(8):833–39 Torzewski J, Torzewski M, Bowyer DE, Frohlich M, Koenig W, et al. 1998. Creactive protein frequently colocalizes with the terminal complement complex in the intima of early atherosclerotic lesions of human coronary arteries. Arterioscler. Thromb. Vasc. Biol. 18:1386–92 Greaves DR, Channon KM. 2002.

169.

170.

171.

172.

173.

174.

175.

176.

177.

Inflammation and immune responses in atherosclerosis. Trends Immunol. 23:535–41 Libby P, Ridker PM, Maseri A. 2002. Inflammation and atherosclerosis. Circulation 105:1135–43 Gupta S, Pablo AM, Jiang X, Wang N, Tall AR, Schindler C. 1997. IFNgamma potentiates atherosclerosis in ApoE knock-out mice. J. Clin. Invest 99:2752–61 Whitman SC, Ravisankar P, Daugherty A. 2002. Interleukin-18 enhances atherosclerosis in apolipoprotein E(−/−) mice through release of interferon-gamma. Circ. Res. 90:E34– 38 Caligiuri G, Rudling M, Ollivier V, Jacob MP, Michel JB, et al. 2003. Interleukin10 deficiency increases atherosclerosis, thrombosis, and low-density lipoproteins in apolipoprotein E knockout mice. Mol. Med. 9:10–17 Moatti D, Faure S, Fumeron F, Amara M, Seknadji P, et al. 2001. Polymorphism in the fractalkine receptor CX3CR1 as a genetic risk factor for coronary artery disease. Blood 97:1925–28 Cyrus T, Witztum JL, Rader DJ, Tangirala R, Fazio S, et al. 1999. Disruption of the 12/15-lipoxygenase gene diminishes atherosclerosis in apo E-deficient mice. J. Clin. Invest. 103:1597–604 Lutgens E, Cleutjens KB, Heeneman S, Koteliansky VE, Burkly LC, Daemen MJ. 2000. Both early and delayed antiCD40L antibody treatment induces a stable plaque phenotype. Proc. Natl. Acad. Sci. USA 97:7464–69 Mach F, Schonbeck U, Sukhova GK, Atkinson E, Libby P. 1998. Reduction of atherosclerosis in mice by inhibition of CD40 signalling. Nature 394:200–3 Pockley AG, Wu R, Lemne C, Kiessling R, de Faire U, Frostegard J. 2000. Circulating heat shock protein 60 is associated with early cardiovascular disease. Hypertension 36:303–7

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS 178. Lewthwaite J, Owen N, Coates A, Henderson B, Steptoe A. 2002. Circulating human heat shock protein 60 in the plasma of British civil servants: relationship to physiological and psychosocial stress. Circulation 106:196–201 179. Sethna KB, Mistry NF, Dholakia Y, Antia NH, Harboe M. 1998. Longitudinal trends in serum levels of mycobacterial secretory (30 kD) and cytoplasmic (65 kD) antigens during chemotherapy of pulmonary tuberculosis patients. Scand. J. Infect. Dis. 30:363–69 180. Beatty WL, Morrison RP, Byrne GI. 1994. Persistent chlamydiae: from cell culture to a paradigm for chlamydial pathogenesis. Microbiol. Rev. 58:686–99 181. Bennett MR, Boyle JJ. 1998. Apoptosis of vascular smooth muscle cells in atherosclerosis. Atherosclerosis 138:3–9 182. Mayr M, Xu Q. 2001. Smooth muscle cell apoptosis in arteriosclerosis. Exp. Gerontol. 36:969–87 183. Boulanger CM, Scoazec A, Ebrahimian T, Henry P, Mathieu E, et al. 2001. Circulating microparticles from patients with myocardial infarction cause endothelial dysfunction. Circulation 104:2649– 52 184. Mallat Z, Hugel B, Ohan J, Leseche G, Freyssinet JM, Tedgui A. 1999. Shed membrane microparticles with procoagulant potential in human atherosclerotic plaques: a role for apoptosis in plaque thrombogenicity. Circulation 99:348–53 185. Mallat Z, Tedgui A. 2001. Current perspective on the role of apoptosis in atherothrombotic disease. Circ. Res. 88:998–1003 186. Mesri M, Altieri DC. 1999. Leukocyte microparticles stimulate endothelial cell cytokine release and tissue factor induction in a JNK1 signaling pathway. J. Biol. Chem. 274:23111–18 187. Wang TJ, Nam BH, Wilson PW, Wolf PA, Levy D, et al. 2002. Association of C-reactive protein with carotid atherosclerosis in men and women: the

188.

189.

190.

191.

192.

193.

194.

195.

401

Framingham Heart Study. Arterioscler. Thromb. Vasc. Biol. 22:1662–67 Liuzzo G, Biasucci LM, Rebuzzi AG, Gallimore JR, Caligiuri G, et al. 1996. Plasma protein acute-phase response in unstable angina is not induced by ischemic injury. Circulation 94:2373–80 Fong IW, Chiu B, Viira E, Tucker W, Wood H, Peeling RW. 2002. Chlamydial heat-shock protein-60 antibody and correlation with Chlamydia pneumoniae in atherosclerotic plaques. J. Infect. Dis. 186:1469–73 Klegeris A, Singh EA, McGeer PL. 2002. Effects of C-reactive protein and pentosan polysulphate on human complement activation. Immunology 106:381–88 Zwaka TP, Hombach V, Torzewski J. 2001. C-reactive protein-mediated low density lipoprotein uptake by macrophages: implications for atherosclerosis. Circulation 103:1194– 97 Mach F, Schonbeck U, Sukhova GK, Bourcier T, Bonnefoy JY, et al. 1997. Functional CD40 ligand is expressed on human vascular endothelial cells, smooth muscle cells, and macrophages: implications for CD40-CD40 ligand signaling in atherosclerosis. Proc. Natl. Acad. Sci. USA 94:1931–36 Reul RM, Fang JC, Denton MD, Geehan C, Long C, et al. 1997. CD40 and CD40 ligand (CD154) are coexpressed on microvessels in vivo in human cardiac allograft rejection. Transplantation 64:1765–74 Malik N, Greenfield BW, Wahl AF, Kiener PA. 1996. Activation of human monocytes through CD40 induces matrix metalloproteinases. J. Immunol. 156:3952–60 Miller DL, Yaron R, Yellin MJ. 1998. CD40L-CD40 interactions regulate endothelial cell surface tissue factor and thrombomodulin expression. J. Leukoc. Biol. 63:373–79

19 Feb 2004

12:1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

402

AR

WICK

AR210-IY22-12.tex

¥

KNOFLACH

¥

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

XU

196. Schonbeck U, Libby P. 2001. The CD40/CD154 receptor/ligand dyad. Cell Mol. Life Sci. 58:4–43 197. Lutgens E, Gorelik L, Daemen MJ, de Muinck ED, Grewal IS, et al. 1999. Requirement for CD154 in the progression of atherosclerosis. Nat. Med. 5:1313–16 198. Ludewig B, Henn V, Schroder JM, Graf D, Kroczek RA. 1996. Induction, regulation, and function of soluble TRAP (CD40 ligand) during interaction of primary CD4+ CD45RA+ T cells with dendritic cells. Eur. J. Immunol. 26:3137–43 199. Schonbeck U, Libby P. 2001. CD40 signaling and plaque instability. Circ. Res. 89:1092–103 200. Aukrust P, Muller F, Ueland T, Berget T, Aaser E, et al. 1999. Enhanced levels of soluble and membrane-bound CD40 ligand in patients with unstable angina. Possible reflection of T lymphocyte and platelet involvement in the pathogenesis of acute coronary syndromes. Circulation 100:614–20 201. Amberger A, Hala M, Saurwein-Teissl M, Metzler B, Grubeck-Loebenstein B, et al. 1999. Suppressive effects of antiinflammatory agents on human endothelial cell activation and induction of heat shock proteins. Mol. Med. 5:117–28 202. Miller LW. 2002. Cardiovascular toxicities of immunosuppressive agents. Am. J. Transplant. 2:807–18 203. Dichtl W, Dulak J, Frick M, Alber HF, Schwarzacher SP, et al. 2003. HMGCoA reductase inhibitors regulate inflammatory transcription factors in human endothelial and vascular smooth muscle cells. Arterioscler. Thromb. Vasc. Biol. 23:58–63 203a. Jurgens G, Xu QB, Huber LA, Bock G, Howanietz H, et al. 1989. Promotion of lymphocyte growth by high density lipoproteins (HDL). Physiological significance of the HDL binding site. J. Biol. Chem. 264(15):8549–56 204. Zhu J, Quyyumi AA, Rott D, Csako G,

205.

206.

207.

208.

209.

210.

211.

Wu H, et al. 2001. Antibodies to human heat-shock protein 60 are associated with the presence and severity of coronary artery disease: evidence for an autoimmune component of atherogenesis. Circulation 103:1071–75 Veres A, Szamosi T, Ablonczy M, Szamosi T Jr, Singh M, et al. 2002. Complement activating antibodies against the human 60 kDa heat shock protein as a new independent family risk factor of coronary heart disease. Eur. J. Clin. Invest. 32:405–10 Hoppichler F, Koch T, Dzien A, Gschwandtner G, Lechleitner M. 2000. Prognostic value of antibody titre to heat-shock protein 65 on cardiovascular events. Cardiology 94:220–23 Birnie DH, Holme ER, McKay IC, Hood S, McColl KE, Hillis WS. 1998. Association between antibodies to heat shock protein 65 and coronary atherosclerosis. Possible mechanism of action of Helicobacter pylori and other bacterial infections in increasing cardiovascular risk. Eur. Heart J. 19:387–94 Prohaszka Z, Duba J, Horvath L, Csaszar A, Karadi I, et al. 2001. Comparative study on antibodies to human and bacterial 60 kDa heat shock proteins in a large cohort of patients with coronary heart disease and healthy subjects. Eur. J. Clin. Invest. 31:285–92 Veres A, Fust G, Smieja M, McQueen M, Horvath A, et al. 2002. Relationship of anti-60 kDa heat shock protein and anticholesterol antibodies to cardiovascular events. Gromadzka G, Zielinska J, Ryglewicz D, Fiszer U, Czlonkowska A. 2001. Elevated levels of anti-heat shock protein antibodies in patients with cerebral ischemia. Cerebrovasc. Dis. 12:235– 39 Huittinen T, Leinonen M, Tenkanen L, Manttari M, Virkkunen H, et al. 2002. Autoimmunity to human heat shock protein 60, Chlamydia pneumoniae

19 Feb 2004

12:1

AR

AR210-IY22-12.tex

AR210-IY22-12.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS infection, and inflammation in predicting coronary risk. Arterioscler. Thromb. Vasc. Biol. 22:431–37 212. Ciervo A, Visca P, Petrucca A, Biasucci LM, Maseri A, Cassone A. 2002. Antibodies to 60-kilodalton heat shock protein and outer membrane protein 2 of Chlamydia pneumoniae in patients with coronary heart disease. Clin. Diagn. Lab. Immunol. 9:66–74 213. Pockley AG, de Faire U, Kiessling R, Lemne C, Thulin T, Frostegard J. 2002. Circulating heat shock protein and heat shock protein antibody levels in established hyp. J. Hypertens. 20(9): 1815–20 214. Kramer J, Harcos P, Prohaszka Z, Horvath L, Karadi I, et al. 2000. Frequencies of certain complement protein alleles and

403

serum levels of anti-heat-shock protein antibodies in cerebrovascular diseases. Stroke 31:2648–52 215. Zhu J, Quyyumi AA, Wu H, Csako G, Rott D, et al. 2003. Increased serum levels of heat shock protein 70 are associated with low risk of coronary artery disease. Arterioscler. Thromb. Vasc. Biol. 23(6):1055–59 216. Swanson SJ, Rosenzweig A, Seidman JG, Libby P. 1994. Diversity of T-cell antigen receptor V beta gene utilization in advanced human atheroma. Arterioscler. Thromb. Vasc. Biol. 14:1210–14 217. Oksenberg JR, Stavri GT, Jeong MC, Garovoy N, Salisbury JR, Erusalimsky JD. 1997. Analysis of the T-cell receptor repertoire in human atherosclerosis. Cardiovasc. Res. 36:256–67

Wick.qxd

3/12/2004

2:01 PM

Page 1

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

C-1

Figure 1 Macroscopic appearance of the inner surface of a human artery. I, normal intima; F, fatty streak; P, raised plaque; E, exulcerated plaque. The insert in the lower right corner shows immunohistochemical staining of CD3+ T-lymphocytes in the shoulder-region of an early atherosclerotic lesion.

Wick.qxd

3/12/2004

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

WICK

2:01 PM

■

Page 2

KNOFLACH

■

XU

Figure 3 Immunofluorescence staining for HSP60 (green) on cultured human umbilical vein endothelial cells (HUVEC) before (a, c) or after 30 min 42°C heat stress (b, d) on unfixed (a, b) or permeabilized, fixed (c, d) cells. HSP60 is expressed after stress on the cell surface (b) as well as in the cytosol (d). Nuclei are stained in red.

HI-RES-IY22-12-Wick

2/16/2004

12:08 PM

Page 3

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INFLAMMATION IN ATHEROSCLEROSIS

C-3

Figure 7 Atherosclerosis: the price we pay for immunity to heat-shock protein 60 (HSP60). Protection against microbial infection is mediated by humoral and cellular adaptive immunity directed against microbial HSP60 among other microbial antigens. In addition, the innate immunity is activated by HSP60 bound to Toll-like receptors. (a) Under physiological conditions, human vascular endothelial cells (EC) do not express HSP60, so no disease results. (b) The stress-induced expression of HSP60 on target ECs can result in cross-reactivity of the immune response between human and microbial HSP60 entailing immune destruction of ECs. This might also result from bona fide autoimmune reactions induced by altered self-HSP60. (c) In some cases, autoimmunity might not develop, despite the presence of human HSP60 on ECs, because the MHC class I and II grooves of these individuals do not accommodate the atherogenic HSP60 epitopes.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:361-403. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104711

Annu. Rev. Immunol. 2004. 22:405–29 doi: 10.1146/annurev.immunol.22.012703.104711 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on November 3, 2003

THE DYNAMIC LIFE OF NATURAL KILLER CELLS Wayne M. Yokoyama,1 Sungjin Kim,1 and Anthony R. French1,2 Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

1

Howard Hughes Medical Institute, Rheumatology Division, Department of Medicine, Department of Pathology and Immunology, and 2Department of Pediatrics, Washington University School of Medicine, St. Louis, Missouri 63110; email: [email protected], [email protected], French [email protected]

Key Words NK cells, development, proliferation, virus, innate immunity ■ Abstract Natural killer (NK) cells play important roles in immunological processes, including early defense against viral infections. This review provides an overview of the dynamic in vivo life of NK cells from their development in the bone marrow to their mature peripheral responses and their ultimate demise, with particular emphasis on mouse NK cells and viral infections.

INTRODUCTION Natural killer (NK) cells are bone marrow–derived lymphocytes, distinct from T and B cells, that are capable of lysing certain tumor cells without prior sensitization (1, 2). Because of this capacity, termed natural killing, NK cells were initially thought to serve primarily in defense against tumor cells. Over the years, however, NK cells have been shown to have other functions. They produce cytokines that regulate the development of acquired, specific immunity, and they can reject bone marrow (BM) transplants (3, 4). In view of their accumulation at the maternal-fetal interface, NK cells are also thought to play essential roles during pregnancy (5). They are also implicated in other pathological processes, including autoimmunity and tissue inflammation (6–8). A primary physiological role of NK cells is to provide early defense against pathogenic organisms during the initial response period while the adaptive immune system is being activated (reviewed in 9). Although NK cells respond to a variety of microorganisms, including bacteria and protozoa, they are particularly important in viral infections (10). This is well illustrated in the case report of an adolescent with a selective NK cell defect who had recurrent infections with particular difficulty in controlling herpes viruses (11). Other human patients with selective NK cell deficiencies have similar clinical courses (12). In mice, depletion of NK cells enhances susceptibility to several viral infections, including murine cytomegalovirus (MCMV) (13). Thus, NK cells are critical in host defense against infectious agents. 0732-0582/04/0423-0405$14.00

405

11 Feb 2004

19:16

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

406

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

NK cells have two major triggering mechanisms: (a) target recognition and (b) cytokine stimulation. Both are intertwined with their effector responses: target killing and cytokine production. These processes play important roles in NK cell effects especially during viral infections, as indicated by numerous viral strategies that limit NK cell responses that in turn emphasize the significance of NK cells in antiviral defense (reviewed in 14). Significant progress has been made in understanding NK cell target recognition that is guided by two general types of NK cell receptors, i.e., inhibitory and activation receptors (15–18). NK cells express inhibitory receptors specific for MHC class I molecules on target cells. As a result of direct ligand interaction, these inhibitory receptors prevent NK cell activation and killing, providing the molecular basis for the “missing-self” hypothesis (19). In the mouse, the inhibitory receptors are encoded in the NK gene complex (NKC) and include the Ly49 family of receptors, including Ly49A, and CD94/NKG2A (20, 21). NK cells also express activation receptors that recognize target cell ligands and can trigger perforindependent natural killing. Many of these receptors, such as Ly49D and Ly49H, are structurally related to the inhibitory receptors (22, 23) but their signaling activities are limited by the inhibitory receptors. Stimulation of NK cells through the activation receptors can lead to production of cytokines such as interferon-γ (IFN-γ ), tumor necrosis factor-α (TNF-α), and granulocyte-macrophage colony stimulating factor (GM-CSF) (24–30). Additionally, infected or activated dendritic cells (DCs) and macrophages produce cytokines and chemokines such as IFNα/β, IL-12, IL-15, and IL-18 that stimulate NK cells to rapidly produce other cytokines (including IFNγ , TNFα, and GM-CSF) and chemokines (such as ATAC/lymphotactin, Mig, and MIP-1α) (31–37; B. Dorner & W.M. Yokoyama, manuscript submitted). The cytokine effects on NK cells are intricately related. For example, during the course of viral infections, IFNα/β enhances NK cell cytotoxicity through STAT1 (signal transducer and activator of transcription 1) signaling, whereas IL-12 stimulates STAT4-dependent IFNγ production through STAT4 that is inhibited by STAT1 (38–41). Thus, sophisticated regulatory pathways control NK cell cytokine production. Considering the importance of NK cells in immune processes, this review provides an overview of the dynamic in vivo life of NK cells from their development in the BM to their mature cellular responses in the periphery and their ultimate demise (Figure 1), with particular emphasis on mouse NK cells and viral infections.

NK CELL DEVELOPMENT Natural killer (NK) cells comprise the third major lymphocyte population (2, 42) and can be distinguished from other lymphocytes by the absence of B- and T-cell antigen receptors, i.e., sIg and T-cell receptor (TCR), respectively. A unique feature of NK cells is that, unlike other lymphocytes, their development does not require events that are necessary for antigen receptor gene rearrangement. Indeed, they are

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMIC LIFE OF NK CELLS

P1: IKH

407

Figure 1 Dynamic life of NK cells. A simplified scheme shows the development of NK cells in the bone marrow from hematopoietic stem cells (HSC) to the common lymphoid progenitor (CLP) to bipotential T/NK progenitor (T/NKP) to the NK progenitor (NKP) and mature NK cells. Immature NK cells proliferate during development. In the periphery, NK cells are generally quiescent until stimulated to kill, produce cytokines, and proliferate.

present in scid and RAG-1- or -2-deficient mice (43–46). Despite close resemblance between T and NK cell effector functions, NK cells also develop normally in athymic nude mice that lack T cells (47, 48). The features distinguishing NK cells from other lymphocytes and apparent normal development in mice lacking B and T cells suggest that NK cells possess a unique developmental pathway.

In Vitro and In Vivo Models of NK Cell Development Available evidence indicates that the complete phenotypic and functional maturation of NK cells requires an intact BM microenvironment. Early studies of mice treated with estrogen (17-beta-estradiol) or bone-seeking isotopes (89Sr) demonstrated that BM ablation leads to the functional impairment of NK cells (49, 50). Moreover, NK cells derived from congenitally osteopetrotic (mi/mi) mice display impaired natural killing (51, 52). Interestingly, these mice possess substantial numbers of cells that express NK cell markers (50, 52, 53), suggesting a critical role of BM microenvironment in the full functional maturation of NK cells. Other studies support a role for direct interaction between stromal cells and the developing NK cell. Lymphotoxin-α (LTα)-deficient mice lack NK and NKT cells (54). Reciprocal BM transfer experiments suggest that close interactions between membrane LTα-expressing NK cell precursors and LTα-responsive stromal cells are necessary for normal NK and NKT cell development (54, 55). The LTαdeficient mouse also manifests a defect in lymph node development that resembles the phenotype of mice deficient in the helix-loop-helix inhibitor Id2 (56), suggesting that the NK cell defect in Id2-deficient mice may also involve similar NK cell interactions with stromal cells.

11 Feb 2004

19:16

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

408

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

Certain aspects of NK cell development are regulated by direct interactions between developing NK cells and stromal elements. In vitro models of NK cell differentiation have shown that cytolytic NK cells can be generated from early hematopoietic cells (57–61). Mouse NK cells can develop in vitro from BM precursor populations in a cytokine cocktail consisting of stem cell factor (c-kit ligand), IL-7, flt-3 ligand (FL), and IL-15 (62, 63). However, the presence of these cytokines alone is insufficient to generate phenotypically mature NK cells that express Ly49 receptors. Importantly, direct contact with stromal cells is required for acquisition of Ly49 receptors by developing NK cells. The in vitro studies led to identification of soluble factors capable of supporting the generation of NK cells and also demonstrated the importance of direct contact between precursors of NK cells and stromal cells. NK cell defects in mice with targeted mutations in cytokines and transcription factors have been instructive (Figure 2) (54–56, 64–77; reviewed in 78, 79). Analysis of gene-manipulated mice and reconstitution studies using alymphoid RAG-2−/− × IL-2R (common) γ (γ c)−/− mice (72, 80, 81) have also proven useful both in defining the steps of NK cell developmental process and in elucidating molecular mechanisms governing this process. Several other mouse models of NK cell deficiency have also been described. While the beige (bg) mouse has defective natural killing and has been used for many years as an “NK cell-deficient” model, the bg mutation does not dramatically affect NK cell number. It involves a molecule that affects the granules of many cells, not just CD3– NK cells (82), rather than affecting NK cell development per se. On the other hand, a mouse transgenic (tg) for the human CD3ε gene (tgε26) has an NK

Figure 2 In vivo models of NK cell development. Informative gene knockout and transgenic mice have helped establish major NK cell developmental steps from HSC to CLP to bipotential T/NK progenitor (T/NKP) to NK progenitor (NKP) to mature NK cells. Based on phenotypic analysis, the NK cell developmental block in these mice can be placed as indicated by knockout and transgenic (tg) mice. Partial defects (P) are also noted.

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

DYNAMIC LIFE OF NK CELLS

P1: IKH

409

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

cell deficiency and also a significant defect in T cell development, indicating a block in development at an early stage of T and NK cell development (83, 84), but this has not been characterized further. A similar phenotype is observed in FcεRIγ tg mice (85). Another mouse, transgenic for a granzyme A-Ly49a construct, also has an NK cell deficiency that appears to be more selective, suggesting that this mouse may be manifesting a block in a terminal differentiation step in NK cell development (86).

Generation of NK Cell Progenitors from Hematopoietic Stem Cells Based on the data currently available, a model has been proposed for the developmental process of NK cells in the BM that is divided into several major steps (Figures 1 and 2). The earliest step involves the commitment of hematopoietic stem cells (HSC) to the lymphoid cell lineage. Recent studies have identified common lymphoid progenitors (CLP) in the BM of adult mice and human (87, 88). In vivo reconstitution assays indicated that CLP give rise to NK, T, and B cells but not to myeloid cell lineages (87). Mice that are deficient in the zinc-finger transcription factor Ikaros or express a dominant-negative form display severe defects in the development of all lymphoid cell lineages including NK cells, whereas myeloid and erythroid lineages are less affected (Figure 2) (64, 89). A member of the Ets family, PU.1, has also been implicated as a critical transcription factor regulating early hematopoiesis and lymphopoiesis (90, 91). Compared to wild type, PU.1−/− fetal liver HSC differentiated less efficiently into NK cells (or B and T cells) when transferred into alymphoid RAG-2−/− × γ c−/− mice (72). Near-complete, partial, or minor defects in NK cell development were observed in FL-, c-Kit-, or IL-7-deficient mice, respectively, in addition to defects in the development of other cell lineages (73, 81, 92). The early impairment of hematopoiesis in Ikaros-deficient and PU.1-deficient mice may be explained by the observations that the Ikaros-deficient HSC express lower levels of the c-Kit and Flk2/flt3 (93), whereas the PU.1-deficiency leads to reduced expression of various cytokine receptors, including IL-7Rα (72, 91, 94). Ikaros and PU.1 may therefore influence NK cell development by promoting early hematopoiesis through up-regulation of receptors for early acting factors. Thus, the earliest step of NK cell development appears to share a common pathway with other lymphocytes involving differentiation of HSC to CLP. Another major step of NK cell development involves the differentiation of the CLP to a bipotential T/NK progenitor (T/NKP) (Figure 2). Consistent with the close resemblance between T and NK cell effector functions, T/NKP have been identified in various fetal organs including blood, thymus, spleen, and liver (95–99). Depending on culture conditions, T/NKP can give rise to T and/or NK cells but not to other lineages. Unexpectedly, T/NKP present in fetal thymus and blood express NK1.1 (96, 97), the most specific serological marker for NK cells (in C57BL/6 mice) at later stages of differentiation, whereas T/NKP present in

11 Feb 2004

19:16

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

410

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

fetal liver do not express this marker (99). T/NKP have not yet been identified in adult tissues, suggesting that adult NK cell development may differ from fetal development. However, unipotent NK cell progenitors (NKP) destined to become mature NK cells are likely to be generated via T/NKP even in adults because CD3ε and FcεRIγ tg mice exhibit selective defects in NK and T cell development (83, 85), and because IL-15 and IL-15Rα-deficiencies affect both NK and NKT cell development. The acquisition of IL-2/15Rβ subunit (CD122) is likely to mark the transition from T/NKP to NKP in both fetus and adult (98, 100) (Figure 2). This acquisition, in part, renders NKP responsive to IL-15, a critical cytokine for NK cell development, which mediates its effects through the IL-15R complex, consisting of IL-15Rα, IL-2/IL-15Rβ, and the IL-2R (common) γ (γ c) chains (101). IL-15or IL-15Rα−targeted mutations produce a relatively selective NK cell deficiency that also affects NK/T, intestinal epithelial lymphocytes (IEL), and memory CD8+ T cells (67, 68). This requirement for IL-15 in NK cell development explains why deficiencies in IL-2 or IL-2Rα do not affect NK cell development, whereas deficiency in IL-2Rβ, which forms part of the IL-15R complex, does affect NK cell development (66). On the other hand, γ c is required for response to many cytokines, including IL-2, -4, -7, -9, and -15, so it is not surprising that γ c deficiency produces a profound block in lymphocyte development, apparently affecting the CLP (65, 102). Interestingly, recent reports indicate that IL-15 can mediate its effects when presented in trans by IL-15Rα to responsive cells expressing IL-15Rβγ alone (103). Although the physiological relevance of this finding is still under investigation, the effect can be observed in vivo (104, 105). When IL-15Rα–specific antibodies become widely available, it may be possible to determine if IL-15Rα expression regulates either the CLP to T/NKP and/or T/NKP to NKP step. The transition from CLP and/or T/NKP to NKP is regulated by transcriptional activity of specific factors (Figure 2). Consistent with the essential role of IL-15 and its receptor in NK cell development, mice deficient in components of the IL-15R signaling pathway, such as Jak3 and STAT5a/b, exhibit defects in NK cell development (74, 77, 106). The ability of FL to enhance the expression of IL-2/15Rβ subunit on hematopoietic progenitor cells may account for the positive effect of FL on the generation of NKP (60–62, 98, 100). IRF-1 (interferon regulatory factor 1)-deficient mice lack NK cells, and this defect can be overcome by the addition of IL-15, suggesting that stromal cells produce IL-15 in an IRF-1-dependent manner (70). Thus, IL-15 and its receptor play significant roles in NK cell development, although detailed studies of the arrested NKPs in deficient mice have not been published. Other studies implicate Ets1 transcription factor and Id2 DNA-binding protein in NK cell development (Figure 2). In Ets1- or Id2-deficient mice, NK cells fail to develop, although T and B cells normally develop (56, 71). The Id proteins do not directly activate a target gene and instead appear to promote the generation of NKP by negatively regulating the activity of the basic helix-loop-helix family of transcription factors that affect B or T cell development (107–109). Thus, the

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

DYNAMIC LIFE OF NK CELLS

P1: IKH

411

commitment of CLP (and/or T/NKP) to NKP may be initiated by expression of specific transcription factors and/or by modulating interactions among already present transcription factors that affect the IL-15R signaling pathway and its downstream consequences.

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NK Cell Developmental Stages from NKP to Mature NK Cells The next major step of NK cell development involves the differentiation of NKPs into phenotypically and functionally mature NK cells (Figure 3). During this step, NKPs undergo a series of phenotypic changes, including acquisition of receptors (CD94/NKG2 and Ly49 in mice) specific for MHC class I molecules that guide target cell–specificity of mature NK cells. In addition, NK cell development during

Figure 3 Developmental stages of committed NK cells. During development in BM, informative surface markers characterize distinct stages from the NK cell progenitor to mature NK cells. After acquisition of NK cell receptors (NKG2 and Ly49), significant proliferation occurs while the cells are still immature.

11 Feb 2004

19:16

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

412

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

this step is characterized by significant proliferation before the cells become phenotypically and functionally mature. The changes during this step can be divided into several putative developmental intermediate stages, as described below, all of which occur in the BM (100, 110). Although NKP were previously identified from in vitro culture and from fetal thymus (60, 98), an equivalent cell was recently identified from adult BM (100). The BM-derived NKP expresses CD122 but neither NK1.1 nor DX5 (pan-NK cell markers), and gives rise exclusively to functional NK cells in vitro. This population is denoted as intermediate Stage I (Figure 3) (110). Thereafter, NK1.1 (also known as Nkrp1c) (111) is expressed (Stages II to V), comparable to studies of in vitro differentiation of human NK cells where NKR-P1A molecule is one of the earliest markers expressed (112). Among the mouse CD3– CD122+ NK1.1+ population, there is an integrin αv+ c-Kit– population (Stage II) that can express CD94/NKG2 receptors without concomitant expression of Ly49 receptors (110). They also express NKG2D (S.K. & W.M.Y., unpublished observations). NK cells at this stage appear to be equivalent to fetal or neonatal mouse NK cells that express CD94/NKG2 receptors without expressing Ly49 receptors (113, 114). Subsequently, Ly49 receptors are expressed on NK1.1+ cells that express both αv and c-Kit (Stage III). However, CD94/NKG2 expression is not a prerequisite for Ly49 expression because there are cells expressing Ly49 molecules without expressing CD94/NKG2. The initial expression of Ly49 receptors appears to be controlled by a stochastic process (reviewed in 115). This process may involve specific transcription factors. A deficiency in the transcription factor TCF-1 selectively affects Ly49a expression with minimal effects on other Ly49 genes (116, 117), although less is known about regulation of CD94 expression (118, 119). Additional studies are under way to explore the promoters and developmental regulation of Ly49 molecules (120– 124). Nevertheless, the acquisition of CD94/NKG2 and Ly49 receptor expression in the adult BM at Stage II and III recapitulates expression of these receptors on fetal and neonatal NK cells in which CD94/NKG2 receptors are first expressed, then the cells acquire Ly49 expression in an ordered manner (113, 114, 125). At Stage IV, the expression of integrin α2 (DX5) is increased and developing NK cells undergo a substantial expansion in the BM with vigorous proliferation. Thereafter, as NK cells finally acquire high-level expression of Mac-1 (αMβ1) and CD43 (Stage V), further NK cell proliferation is markedly reduced unless challenged by pathogens such as viruses that can stimulate mature NK cell proliferation (see below) (110, 126). Importantly, NK cells at Stage V become fully capable of functional activities associated with mature peripheral NK cells. Developing NK cells undergo significant proliferation at a defined, developmentally immature stage in the BM, as evidenced by acute in vivo BrdU (bromodeoxyuridine, a nucleotide analogue) incorporation by Mac-1lo DX5hi NK1.1+ cells (110). Regardless of the mechanism by which these receptors are initially expressed on the developing NK cell, the proliferation of immature NK cells in the BM occurs after they express CD94/NKG2 and Ly49 receptors because

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMIC LIFE OF NK CELLS

P1: IKH

413

most BrdU+ cells among CD122+ cells express NK1.1 and at least one of the CD94/NKG2 or Ly49 receptors. Expression of MHC class I–specific inhibitory receptors at this point may have consequences on the ultimate size and distribution of mature NK cell populations expressing the “missing-self” receptors as well as on functional maturation. The receptors themselves may be predicted to directly encounter MHC class I, although it is not currently known if the receptors themselves trigger proliferation and further development or whether the proliferation phase more simply affects the receptor repertoire of mature NK cells. Detailed examination of the receptor repertoire on mature NK cells in mice with different MHC class I haplotypes has indicated that there is a modest reduction of NK cells coexpressing receptors for self-MHC class I molecules (115). Consistent with this phenomenon, in vitro studies have shown that engagement of a Ly49 receptor with MHC class I on stromal cells affects further acquisition of other Ly49 receptors recognizing the same MHC class I molecule (63). There is also an MHC class I effect on the level of receptors expressed such that, in the presence of the self-MHC class I molecule, there is down-regulation of the corresponding Ly49 receptor (127). Studies of BM chimeric mice suggest that this is due to an extracellular interaction between Ly49 on the NK cell and host MHC class I (128). Indeed, recent studies show the presence of Ly49 receptors on the NK cell surface bound with its MHC class I ligand that was acquired from surrounding cells (129, 130). In addition, cis effects between Ly49 receptors and their ligands on the same NK cell have been noted (128, 131), but the significance of ligand capture and cis effects requires further investigation. Nevertheless, taken together, these studies indicate that there are direct in vivo interactions between Ly49 receptors and their cognate ligands. Thus, it will be important to investigate the possible influence of CD94 and Ly49 receptor engagement on proliferation and functional maturation of NK cells during development in the BM. Interestingly, phenotypically immature NK cells are present in the adult liver (110). Although these cells express CD122 and NK1.1, they do not express other markers such as DX5. Perhaps the most striking feature of the liver Mac-1lo DX5− NK cells is the rare expression of Ly49 receptors while they express CD94/NKG2 receptors at a high frequency. The relatively poor capacity of liver Mac-1lo DX5− NK cells to produce IFN-γ supports the idea that these cells are immature NK cells. The more frequent expression of CD94/NKG2 receptors on these cells is also consistent because these receptors are expressed earlier than Ly49 receptors during NK cell ontogeny. These Mac-1lo DX5− NK cells are phenotypically equivalent to NK cells at Stage II in the BM and are reminiscent of the previously termed “nonlytic NK1.1+ cells” found in osteopetrotic mice that display poor expression of Mac-1 and Ly49 receptors (53, 132). Thus, the liver contains NK cells with an immature phenotype, raising the interesting possibility that peripheral tissues serve as reservoirs for less differentiated NK progenitors, which may expand in the periphery under certain conditions. Our current model of NK cell development in the BM is based on the apparent sequential acquisition of certain informative molecules with concomitant

11 Feb 2004

19:16

414

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

down-regulation of other markers. Whether these molecules represent functionally important, rate-limiting steps needs to be determined. Nevertheless, the current model provides guidance for future investigation. Most importantly, the model highlights the points that developing NK cells expand at a phenotypically distinct stage and that NK cell development is characterized by a dynamic process involving up- and down-regulation of surface receptors for cytokines, extracellular matrix, and self- and target-recognition, rather than the cumulative, progressive acquisition of such receptors.

Human NK Cell Development NK cell development is likely to be similar in mice and humans. In the BM, acquisition of functional maturity of mouse NK cells, i.e., the capacity to kill targets and produce cytokines, is accompanied by changes in phenotypic markers (110). Similarly, immature human NK cells display regulated in vitro development of the capacity to secrete different cytokines and kill targets using different cytotoxic pathways (112, 133–135). On the other hand, there are some differences between mouse and human NK cell development. For example, IL-12 has been shown to be important in human NK cell development studied in vitro (136). However, studies on NK cell function in IL-12 or IL-12 receptor-deficient mice indicate that peripheral splenocytes have essentially normal natural killing and IL-2-activated NK cell activity against YAC-1 cells (137–139). Furthermore, IL-12 receptor-deficient mice backcrossed onto the C57BL/6 background have normal numbers of NK cells (140). Finally, human NK cell development appears to be dependent on IFN-γ or IL-18. However, there is no deficiency in NK cell number in mice with deficiencies of IFN-γ or IL-18 (140, 141). Therefore, there may be either in vitro versus in vivo or species effects to explain the similarities and differences in studies of mouse and human NK cell development.

MATURE NK CELLS IN THE PERIPHERY Mature NK cells in the periphery (outside the BM) are well suited to respond rapidly to infected or transformed cells and to play an important role in innate defense. NK cells, however, constitute only a small population of cells (about 2.5% of splenic leukocytes). How can this small population quickly respond with enough of a critical mass to effect significant innate defense? One mechanism involves the expression of multiple activation receptors by individual NK cells (23). By contrast to clonally distributed TCRs that endow the individual T cell with the ability to respond only to one antigen, an individual NK cell appears capable of responding to multiple activation receptor ligands. Furthermore, the na¨ıve T cell population contains only rare cells with a TCR for the relevant antigen, whereas large percentages of the NK cell population express any given activation receptor in an overlapping fashion. This multiple activation

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

DYNAMIC LIFE OF NK CELLS

P1: IKH

415

receptor expression on sizeable subpopulations would allow a substantial number of NK cells to quickly respond to a given specific insult. Another mechanism allowing a significant NK cell response is related to their constitutive expression of cytokine receptors that permit many NK cells to be stimulated by proinflammatory cytokines produced early in the course of an immune response. Thus, large numbers of NK cells can rapidly respond to a particular stimulus through their activation or cytokine receptors.

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Peripheral NK Cells Mature peripheral NK cells that have exited the BM are found primarily in the blood, spleen, and liver as well as in the uterus. The homing of NK cells to these locations is not well characterized but NK cells appear to have developmentally regulated expression of a number of integrins that may play a role in their localization as well as effector functions (110, 142). In addition, several chemokines, including CXCL12 and MIP1α, have been implicated in NK cell localization (143–148). Interestingly, immunohistochemical in situ staining has shown that NK cells are localized to the red pulp of the spleen and the sinusoidal regions of the liver (149, 150). Few NK cells are present in other solid organs, and surprisingly, there are relatively few NK cells in lymph nodes (149, 151). In experimental in vivo tumor models, NK cells are more frequently found when the targets are also sensitive to NK cell killing in vitro (152). In viral infections, NK cells infiltrate the liver parenchyma in the vicinity of infected foci (146, 149). However, NK cells may be less effective in killing when they migrate into parenchymal tissues (153). Thus, from the vantage point of the peripheral blood, spleen, and blood pool of the liver and lung, NK cells may be particularly suited for surveying the blood for pathogens and infected or transformed cells. Supporting a role for NK cells in controlling targets (tumors or infections) in the blood, early studies indicated that NK cells control experimental metastases (1). In more recent studies, NK cells have been shown to readily eradicate radiolabeled tumor cells injected intravenously. In as little as 4 h after injection, few NKsensitive targets can be found in the lungs, whereas when NK cells are absent or depleted, large numbers of injected cells migrate to the lungs via the blood circulation (86, 154). In viral infections, an absence of human NK cells is also associated with viremia due to viruses that are usually mucosal pathogens, such as herpes simplex (11). Thus, peripheral NK cells appear to be efficient in controlling dissemination of tumors and infections. How this peripheral pool is maintained is a topic of recent interest.

Peripheral NK Cell Homeostasis and Survival The half-life of mature NK cells in the periphery appears to be about seven to ten days based on the survival of adoptively transferred NK cells (104, 105, 155). Mature peripheral NK cells are relatively quiescent, undergoing low levels of

11 Feb 2004

19:16

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

416

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

proliferation (126). Acute short-term in vivo BrdU incorporation assays have demonstrated that only 1%–3% of splenic NK cells in na¨ıve mice are dividing (126), whereas longer-term BrdU incorporation studies indicate that peripheral NK cells have a relatively high turnover rate (156; S. Kim & W.M. Yokoyama, unpublished data). Mature NK cells can undergo homeostatic proliferation. When splenic CFSElabeled NK cells were adoptively transferred into RAG−/− × γ −/− mice that lack T, B, and NK cells, the transferred cells proliferated and accumulated in a manner analogous to that of transferred T cells in a lymphopenic host (105, 155). In contrast, mature NK cells did not proliferate or accumulate when transferred into wild-type or RAG−/− mice that have endogenous NK cells, suggesting that some factor, perhaps a cytokine or “space,” is limiting when there are endogenous NK cells present (155, 157). Comparative reconstitution studies using alymphoid or cytokine-deficient mice support the conclusion that the limiting factor is IL-15, with a smaller, partially redundant contribution from IL-7 (105, 155). This conclusion is bolstered by observations that the injection of exogenous IL-15 or transgenic overexpression of IL-15 results in higher in vivo NK cell numbers (41, 67, 158, 159), demonstrating that space was not the limiting factor in homeostatic proliferation. IL-15 is also involved in the homeostasis of CD8+ memory T cells and Vα14+ NKT cells, suggesting that all three cell types may compete for available IL-15 (160–166). Therefore, IL-15 appears to play a primary role in peripheral NK cell homeostasis in addition to its crucial role in NK cell development. IL-15 is also necessary for survival of peripheral NK cells (105, 155, 157). Splenic NK cells could be successively adoptively transferred into wild-type but not IL-15−/− mice. In addition, blockade of IL-15R signaling by treatment of mice with Fab fragments of anti-IL-2/IL-15Rβ resulted in the loss of 90% of splenic NK cells (157). The survival effect of IL-15 has been linked to its ability to maintain high levels of the antiapoptotic factor Bcl-2 in NK cells in vivo (155, 157). Thus, IL-15 promotes survival of peripheral NK cells apparently by maintaining antiapoptotic factors. In spite of their dependence on IL-15, NK cells themselves do not need to express IL-15Rα to survive (104). The adoptive transfer of mature NK cells into IL-15Rα −/− recipients resulted in rapid loss of the NK cells. However, IL-15Rαdeficient NK cells could develop when IL-15Rα-deficient BM is transplanted to IL-15Rα-sufficient hosts. Furthermore, the IL-15Rα-deficient NK cells from chimeric mice survive when adoptively transferred into wild-type hosts but not into IL-15Rα −/− recipients, demonstrating a role for IL-15Rα expression on nonNK cells in supporting the survival of peripheral NK cells. IL-15Rα on non-NK cells has been shown to bind IL-15 with high affinity, recycle through the endosomal compartment while complexed with IL-15, and present IL-15 in trans to NK cells, resulting in tight localization of IL-15 signaling as well as prolonged stimulation (103). Though this trans effect can act on na¨ıve NK cell development and homeostasis, it is not yet known if this effect will be observed in other NK cell in vivo responses, such as during viral infections when IL-15Rα on NK cells is upregulated on mature NK cells (A.R. French & W.M. Yokoyama, unpublished data).

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

DYNAMIC LIFE OF NK CELLS

P1: IKH

417

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NK Cell–Proliferative Responses to Viral Infections Mature peripheral NK cells rapidly proliferate during viral infections (126). Biron and colleagues have long noted indirect evidence of NK cell proliferation as measured by increased cell size (“blastogenesis”) during murine cytomegalovirus (MCMV) infection that correlated with increased NK cell numbers in the spleen and liver (34, 167, 168). Recent in vivo studies using intracellular BrdU staining as a direct measure of NK cell proliferation have demonstrated nonspecific NK cell proliferation early during the course of MCMV infection, regardless of NK cell expression of the MCMV-specific Ly49h receptor (126) (Figure 4, Phase I). This initial nonspecific proliferative phase resembles the cytokine-driven “bystander proliferation” observed in T cells in response to viral infections or stimulation with type I interferons (169), suggesting that this early phase of viral-induced NK cell proliferation represents a nonspecific response to proliferative cytokines such as IL-15. Although IL-15 levels are difficult to measure in vivo, accumulating evidence suggests that IFN α/β induction of IL-15 is involved in viral-induced NK cell proliferation (34, 41). Viral infections as well as polyinosinic:polycytidylic acid (pI:C) injections elicit strong IFN α/β responses that up-regulate IL-15 mRNA in dendritic cells and macrophages (41, 170, 171). The up-regulation of IL-15

Figure 4 Stages of NK cell responses in vivo. Following MCMV infection, NK cells undergo three phases of responses, as illustrated by their proliferation. During the first phase, proliferation is enhanced but nonspecific with respect to the Ly49h receptor specific for MCMV-infected cells. In this phase (I), there is no difference in proliferation of Ly49h+ (solid line) and Ly49h− (dashed line) NK cells. A second specific phase (II) then follows whereby significant proliferation of Ly49h+ NK cells occurs. In the third phase (III), specific proliferation wanes.

11 Feb 2004

19:16

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

418

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

mRNA during MCMV infection or pI:C injection was abrogated in the absence of IFNα/βR or STAT1 (41, 171). Correlated with this abrogation, the percentage of proliferating NK cells early during MCMV infection in IFNα/βR−/− or STAT1−/− mice was significantly lower than in wild-type littermates (41). However, direct confirmation of the role of IL-15 in mediating viral-induced NK cell proliferation has proven difficult due to a paucity of relevant reagents and because effective blockade of IL-15 signaling in the periphery results in NK cell death (157). In spite of this, strong circumstantial evidence suggests that IL-15 is involved in mediating viral-induced NK cell proliferation. After the early nonspecific phase of NK cell proliferation during MCMV infection, there is a later phase (Figure 4, Phase II) of selective proliferation of NK cells expressing the specific activation receptor (Ly49h) that is responsible for genetic NK cell-mediated resistance to MCMV by recognizing MCMV-infected cells, specifically m157 encoded by the virus (30, 126, 172–175). This period of preferential proliferation of Ly49h+ NK cells peaked at days 4 to 6 post-MCMV infection and resulted in an expansion of the pool of Ly49h+ NK cells. NK cells expressing other NK activation receptors such as Ly49d underwent proliferation during the initial nonspecific phase but did not undergo preferential expansion during the later specific phase. The specific proliferation of Ly49h+ NK cells was abrogated with F(ab0 )2 fragments of anti-Ly49h but not anti-Ly49d monoclonal antibodies supporting the hypothesis that Ly49h recognition of MCMV infected cells stimulates selective proliferation of Ly49h+ NK cells. In addition, the selective proliferation of Ly49h+ NK cells was shown to be MCMV-specific. Mice infected with vaccinia virus underwent a similar initial phase of nonspecific NK cell proliferation but did not undergo specific proliferation of Ly49h+ NK cells at later time points. However, vaccinia infection did result in the selective proliferation of a Ly49h-negative NK cell population, suggesting that the responding NK cells may express a different vaccinia-specific activation receptor. Interestingly, specific stimulation of Ly49h+ NK cells was not detectable at the earlier stage even though these cells should have been specifically triggered through Ly49h to control MCMV by day 2. It is therefore likely that the “nonspecific” stimulation of NK cells masks the detection of specific NK cell stimulation until later in the infection when it becomes manifest (Figure 4, Phase II). Following the selective expansion phase during MCMV infection, NK cells undergo a significant contraction of both Ly49h+ and Ly49h− NK cells (Figure 4, Phase III), which coincides with the onset of the adaptive immune response (126). This contraction that occurs around day 6 of MCMV infection is not well characterized but may be driven by antiproliferative cytokines derived from other cells such as activated T cells or simply the down-regulation of cytokines or cytokine receptors involved in stimulating proliferation. A recent report suggested that IL21 from activated CD4+ T cells might be involved in limiting NK cell responses and facilitating the transition to the adaptive immune response (176). However, the signals responsible for resolution of the NK cell response during MCMV infection remain to be identified.

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMIC LIFE OF NK CELLS

P1: IKH

419

Therefore, mature NK cells in the periphery remain primarily quiescent until they respond to a stimulus such as a viral infection. They then rapidly proliferate in an initial nonspecific mode resembling the “bystander proliferation” seen in T cells early during a viral infection or following stimulation with IFN α/β (169). This is followed by a specific phase of proliferation of NK cells expressing the activation receptor that recognizes virus-infected cells in a manner somewhat analogous to T-cell clonal expansion. This phase is followed by a resolution phase during which the number of NK cell is significantly contracted and rapidly returns back to its steady-state levels in a manner similar to the resolution of an adaptive immune response with a significant apoptosis of T cells.

SUMMARY NK cells undergo dynamic processes during development and immune responses. In the BM, developing NK cells up- and down-regulate surface receptors for cytokines and extracellular matrix that help define developmental stages. At an immature stage where they have already expressed receptors important in “missing-self,” proliferation that may modulate the receptor repertoire of mature NK cells occurs. In the periphery, mature NK cells are relatively quiescent. Upon viral challenge, they undergo both nonspecific and specific proliferative phases, the latter related to NK cell-activation receptors for infected cells, then return to a quiescent state. Numerous viral strategies to thwart NK cell responses have been discovered although not detailed here. Taken together, these data indicate that the life of NK cells is dynamic with respect to expression of receptors for cytokines, extracellular matrix, self- and target-recognition, and proliferation, and they signify the importance of NK cells in antiviral innate immunity.

ACKNOWLEDGMENTS The authors thank past and present members of the Yokoyama laboratory for their contributions to the understanding of NK cell biology, and R. Rodrigues for evaluation of the manuscript. Work in the Yokoyama laboratory is supported by the Howard Hughes Medical Institute, the Barnes-Jewish Hospital Foundation, and grants from the National Institute of Allergy and Infectious Diseases. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Herberman R. 1982. NK Cells and Other Natural Effector Cells. NY: Academic 2. Yokoyama WM. 1999. Natural killer cells. In Fundamental Immunology, ed.

WE Paul, pp. 575–603. New York: Lippincott-Raven 3. Tripp CS, Wolf SF, Unanue ER. 1993. Interleukin 12 and tumor necrosis factor

11 Feb 2004

19:16

420

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

alpha are costimulators of interferon gamma production by natural killer cells in severe combined immunodeficiency mice with listeriosis, and interleukin 10 is a physiologic antagonist. Proc. Natl. Acad. Sci. USA 90:3725–29 Yu YYL, Kumar V, Bennett M. 1992. Murine natural killer cells and marrow graft rejection. Annu. Rev. Immunol. 10:189–213 Moffett-King A. 2002. Natural killer cells and pregnancy. Nat. Rev. Immunol. 2:656– 63 Shi FD, Wang HB, Li H, Hong S, Taniguchi M, et al. 2000. Natural killer cells determine the outcome of B cellmediated autoimmunity. Nat. Immunol. 1:245–51 Flodstrom M, Maday A, Balakrishna D, Cleary MM, Yoshimura A, Sarvetnick N. 2002. Target cell defense prevents the development of diabetes after viral infection. Nat. Immunol. 3:373–82 Homann D, Jahreis A, Wolfe T, Hughes A, Coon B, et al. 2002. CD40L blockade prevents autoimmune diabetes by induction of bitypic NK/DC regulatory cells. Immunity 16:403–15 French AR, Yokoyama WM. 2003. Natural killer cells and viral infections. Curr. Opin. Immunol. 15:45–51 Yokoyama WM. 2002. The role of natural killer cells in innate immunity to infection. In Innate Immunity, ed. RAB Ezekowitz, JA Hoffman, pp. 321–39. Totowa, NJ: Humana Biron CA, Byron KS, Sullivan JL. 1989. Severe herpesvirus infections in an adolescent without natural killer cells. N. Engl. J. Med. 320:1731–35 Orange JS. 2002. Human natural killer cell deficiencies and susceptibility to infection. Microbes Infect. 4:1545–58 Bukowski JF, Woda BA, Habu S, Okumura K, Welsh RM. 1983. Natural killer cell depletion enhances virus synthesis and virus-induced hepatitis in vivo. J. Immunol. 131:1531–38

14. Orange JS, Fassett MS, Koopman LA, Boyson JE, Strominger JL. 2002. Viral evasion of natural killer cells. Nat. Immunol. 3:1006–12 15. Yokoyama WM. 1993. Recognition structures on natural killer cells. Curr. Opin. Immunol. 5:67–73 16. Lanier LL. 1998. NK cell receptors. Annu. Rev. Immunol. 16:359–93 17. Long EO. 1999. Regulation of immune responses through inhibitory receptors. Annu. Rev. Immunol. 17:875–904 18. Moretta A, Bottino C, Vitale M, Pende D, Cantoni C, et al. 2001. Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis. Annu. Rev. Immunol. 19:197– 223 19. K¨arre K. 1985. Role of target histocompatibility antigens in regulation of natural killer activity: a reevaluation and a hypothesis. In Mechanisms of Cytotoxicity by NK Cells, ed. RB Herberman, DM Callewaert, pp. 81–103. Orlando, FL: Academic 20. Karlhofer FM, Ribaudo RK, Yokoyama WM. 1992. MHC class I alloantigen specificity of Ly-49+ IL-2-activated natural killer cells. Nature 358:66–70 21. Yokoyama WM, Plougastel BF. 2003. Immune functions encoded by the natural killer gene complex. Nat. Rev. Immunol. 3:304–16 22. Mason LH, Anderson SK, Yokoyama WM, Smith HRC, Winklerpickett R, Ortaldo JR. 1996. The Ly-49D receptor activates murine natural killer cells. J. Exp. Med. 184:2119–28 23. Smith HR, Chuang HH, Wang LL, Salcedo M, Heusel JW, Yokoyama WM. 2000. Nonstochastic coexpression of activation receptors on murine natural killer cells. J. Exp. Med. 191:1341–54 24. Anegon I, Cuturi MC, Trinchieri G, Perussia B. 1988. Interaction of Fc receptor (CD16) ligands induces transcription of interleukin 2 receptor (CD25) and lymphokine genes and expression of their

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

DYNAMIC LIFE OF NK CELLS

25.

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

26.

27.

28.

29.

30.

31.

32.

33.

34.

products in human natural killer cells. J. Exp. Med. 167:452–72 Arase H, Arase N, Saito T. 1996. Interferon gamma production by natural killer (NK) cells and NK1.1+ T cells upon NKRP1 cross-linking. J. Exp. Med. 183:2391– 96 Kim S, Yokoyama WM. 1998. NK cell granule exocytosis and cytokine production inhibited by Ly-49A engagement. Cell Immunol. 183:106–12 Ho EL, Carayannopoulos LN, PoursineLaurent J, Kinder J, Plougastel B, et al. 2002. Co-stimulation of multiple NK cell activation receptors by NKG2D. J. Immunol. 169:3667–75 Gosselin P, Mason LH, Willette-Brown J, Ortaldo JR, McVicar DW, Anderson SK. 1999. Induction of DAP12 phosphorylation, calcium mobilization, and cytokine secretion by Ly49H. J. Leukoc. Biol. 66:165–71 Ortaldo JR, Young HA. 2003. Expression of IFN-gamma upon triggering of activating Ly49D NK receptors in vitro and in vivo: costimulation with IL-12 or IL18 overrides inhibitory receptors. J. Immunol. 170:1763–69 Smith HR, Heusel JW, Mehta IK, Kim S, Dorner BG, et al. 2002. Recognition of a virus-encoded ligand by a natural killer cell activation receptor. Proc. Natl. Acad. Sci. USA 99:8826–31 Biron CA, Nguyen KB, Pien GC, Cousens LP, Salazar-Mather TP. 1999. Natural killer cells in antiviral defense: function and regulation by innate cytokines. Annu. Rev. Immunol. 17:189–220 Zitvogel L. 2002. Dendritic and natural killer cells cooperate in the control/switch of innate immunity. J. Exp. Med. 195:F9– 14 Andrews DM, Scalzo AA, Yokoyama WM, Smyth MJ, Degli-Esposti MA. 2003. Functional interactions between dendritic cells and NK cells during viral infection. Nat. Immunol. 4:175–81 Orange JS, Biron CA. 1996. Characteri-

35.

36.

37.

38.

39.

40.

41.

42. 43.

P1: IKH

421

zation of early IL-12, IFN-alphabeta, and TNF effects on antiviral state and NK cell responses during murine cytomegalovirus infection. J. Immunol. 156:4746–56 Pien GC, Satoskar AR, Takeda K, Akira S, Biron CA. 2000. Cutting edge: selective IL-18 requirements for induction of compartmental IFN-gamma responses during viral infection. J. Immunol. 165:4787– 91 Akira S. 2000. The role of IL-18 in innate immunity. Curr. Opin. Immunol. 12:59– 63 Dorner BG, Scheffold A, Rolph MS, Huser MB, Kaufmann SH, et al. 2002. MIP-1alpha, MIP-1beta, RANTES, and ATAC/lymphotactin function together with IFN-gamma as type 1 cytokines. Proc. Natl. Acad. Sci. USA 99:6181–86 Cousens LP, Peterson R, Hsu S, Dorner A, Altman JD, et al. 1999. Two roads diverged: interferon alpha/beta- and interleukin 12-mediated pathways in promoting T cell interferon gamma responses during viral infection. J. Exp. Med. 189: 1315–28 Nguyen KB, Cousens LP, Doughty LA, Pien GC, Durbin JE, Biron CA. 2000. Interferon α/β-mediated inhibition and promotion of interferon-γ : STAT1 resolves a paradox. Nat. Immunol. 1:70–76 Nguyen KB, Watford WT, Salomon R, Hofmann SR, Pien GC, et al. 2002. Critical role for STAT4 activation by type 1 interferons in the interferongamma response to viral infection. Science 297:2063–66 Nguyen KB, Salazar-Mather TP, Dalod MY, Van Deusen JB, Wei XQ, et al. 2002. Coordinated and distinct roles for IFNalphabeta, IL-12, and IL-15 regulation of NK cell responses to viral infection. J. Immunol. 169:4279–87 Trinchieri G. 1989. Biology of natural killer cells. Adv. Immunol. 47:187–376 Dorshkind K, Pollack SB, Bosma MJ, Phillips RA. 1985. Natural killer (NK) cells are present in mice with severe

11 Feb 2004

19:16

422

44.

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

45.

46.

47.

48.

49.

50.

51.

52.

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

combined immunodeficiency (scid). J. Immunol. 134:3798–801 Hackett J, Jr, Bosma GC, Bosma MJ, Bennett M, Kumar V. 1986. Transplantable progenitors of natural killer cells are distinct from those of T and B lymphocytes. Proc. Natl. Acad. Sci. USA 83: 3427–31 Mombaerts P, Iacomini J, Johnson RS, Herrup K, Tonegawa S, Papaioannou VE. 1992. RAG-1-deficient mice have no mature B and T lymphocytes. Cell 68:869–77 Shinkai Y, Rathbun G, Lam KP, Oltz EM, Stewart V, et al. 1992. RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell 68:855–67 Kiessling R, Klein E, Pross H, Wigzell H. 1975. Natural killer cells in the mouse. II. Cytotoxic cells with specificity for mouse Moloney leukemia cells. Characteristics of the killer cell. Eur. J. Immunol. 5:117– 21 Herberman RB, Nunn ME, Lavrin DH. 1975. Natural cytotoxic reactivity of mouse lymphoid cells against syngeneic and allogeneic tumors. I. Distribution of reactivity and specificity. Int. J. Cancer 16: 216 Seaman WE, Blackman MA, Gindhart TD, Roubinian JR, Loeb JM, Talal N. 1978. beta-Estradiol reduces natural killer cells in mice. J. Immunol. 121:2193– 98 Kumar V, Ben-Ezra J, Bennett M, Sonnenfeld G. 1979. Natural killer cells in mice treated with 89strontium: normal targetbinding cell numbers but inability to kill even after interferon administration. J. Immunol. 123:1832–38 Seaman WE, Gindhart TD, Greenspan JS, Blackman MA, Talal N. 1979. Natural killer cells, bone, and the bone marrow: studies in estrogen-treated mice and in congenitally osteopetrotic (mi/mi) mice. J. Immunol. 122:2541–47 Ito A, Kataoka TR, Kim DK, Koma Y, Lee YM, Kitamura Y. 2001. Inhibitory

53.

54.

55.

56.

57.

58.

59.

60.

effect on natural killer activity of microphthalmia transcription factor encoded by the mutant mi allele of mice. Blood 97:2075–83 Puzanov IJ, Bennett M, Kumar V. 1996. IL-15 can substitute for the marrow microenvironment in the differentiation of natural killer cells. J. Immunol. 157:4282– 85 Iizuka K, Chaplin DD, Wang Y, Wu Q, Pegg LE, et al. 1999. Requirement for membrane lymphotoxin in natural killer cell development. Proc. Natl. Acad. Sci. USA 96:6336–40 Wu Q, Sun Y, Wang J, Lin X, Wang Y, et al. 2001. Signal via lymphotoxin-beta R on bone marrow stromal cells is required for an early checkpoint of NK cell development. J. Immunol. 166:1684–89 Yokota Y, Mansouri A, Mori S, Sugawara S, Adachi S, et al. 1999. Development of peripheral lymphoid organs and natural killer cells depends on the helix-loophelix inhibitor Id2. Nature 397:702–6 Miller JS, Alley KA, McGlave P. 1994. Differentiation of natural killer (NK) cells from human primitive marrow progenitors in a stroma-based long-term culture system: identification of a CD34+7+ NK progenitor. Blood 83:2594–601 Shibuya A, Nagayoshi K, Nakamura K, Nakauchi H. 1995. Lymphokine requirement for the generation of natural killer cells from CD34+ hematopoietic progenitor cells. Blood 85:3538–46 Mrozek E, Anderson P, Caligiuri MA. 1996. Role of interleukin-15 in the development of human CD56+ natural killer cells from CD34+ hematopoietic progenitor cells. Blood 87:2632–40 Williams NS, Moore TA, Schatzle JD, Puzanov IJ, Sivakumar PV, et al. 1997. Generation of lytic natural killer 1.1+, Ly49− cells from multipotential murine bone marrow progenitors in a stroma-free culture: definition of cytokine requirements and developmental intermediates. J. Exp. Med. 186:1609–14

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMIC LIFE OF NK CELLS 61. Yu H, Fehniger TA, Fuchshuber P, Thiel KS, Vivier E, et al. 1998. Flt3 ligand promotes the generation of a distinct CD34+ human natural killer cell progenitor that responds to interleukin-15. Blood 92:3647–57 62. Williams NS, Klem J, Puzanov IJ, Sivakumar PV, Bennett M, Kumar V. 1999. Differentiation of NK1.1+, Ly49+ NK cells from flt3+ multipotent marrow progenitor cells. J. Immunol. 163:2648–56 63. Roth C, Carlyle JR, Takizawa H, Raulet DH. 2000. Clonal acquisition of inhibitory Ly49 receptors on developing NK cells is successively restricted and regulated by stromal class I MHC. Immunity 13:143–53 64. Georgopoulos K, Bigby M, Wang JH, Molnar A, Wu P, et al. 1994. The Ikaros gene is required for the development of all lymphoid lineages. Cell 79:143–56 65. Cao X, Shores EW, Hu-Li J, Anver MR, Kelsall BL, et al. 1995. Defective lymphoid development in mice lacking expression of the common cytokine receptor gamma chain. Immunity 2:223–38 66. Suzuki H, Duncan GS, Takimoto H, Mak TW. 1997. Abnormal development of intestinal intraepithelial lymphocytes and peripheral natural killer cells in mice lacking the IL-2 receptor beta chain. J. Exp. Med. 185:499–505 67. Kennedy MK, Glaccum M, Brown SN, Butz EA, Viney JL, et al. 2000. Reversible defects in natural killer and memory CD8 T cell lineages in interleukin 15-deficient mice. J. Exp. Med. 191:771–80 68. Lodolce JP, Boone DL, Chai S, Swain RE, Dassopoulos T, et al. 1998. IL-15 receptor maintains lymphoid homeostasis by supporting lymphocyte homing and proliferation. Immunity 9:669–76 69. Ohteki T, Yoshida H, Matsuyama T, Duncan GS, Mak TW, Ohashi PS. 1998. The transcription factor interferon regulatory factor 1 (IRF-1) is important during the maturation of NK1.1+ T cell receptorαβ + (NK1+ T) cells, natural killer cells,

70.

71.

72.

73.

74.

75.

76.

77.

78.

P1: IKH

423

and intestinal intraepithelial T cells. J. Exp. Med. 187:967–72 Ogasawara K, Hida S, Azimi N, Tagaya Y, Sato T, et al. 1998. Requirement for IRF1 in the microenvironment supporting development of natural killer cells. Nature 391:700–3 Barton K, Muthusamy N, Fischer C, Ting CN, Walunas TL, et al. 1998. The Ets-1 transcription factor is required for the development of natural killer cells in mice. Immunity 9:555–63 Colucci F, Samson SI, DeKoter RP, Lantz O, Singh H, Di Santo JP. 2001. Differential requirement for the transcription factor PU.1 in the generation of natural killer cells versus B and T cells. Blood 97:2625– 32 McKenna HJ, Stocking KL, Miller RE, Brasel K, De Smedt T, et al. 2000. Mice lacking flt3 ligand have deficient hematopoiesis affecting hematopoietic progenitor cells, dendritic cells, and natural killer cells. Blood 95:3489–97 Park SY, Saijo K, Takahashi T, Osawa M, Arase H, et al. 1995. Developmental defects of lymphoid cells in Jak3 kinasedeficient mice. Immunity 3:771–82 Lohoff M, Duncan GS, Ferrick D, Mittrucker HW, Bischof S, et al. 2000. Deficiency in the transcription factor interferon regulatory factor (IRF)-2 leads to severely compromised development of natural killer and T helper type 1 cells. J. Exp. Med. 192:325–36 Lacorazza HD, Miyazaki Y, Di Cristofano A, Deblasio A, Hedvat C, et al. 2002. The ETS protein MEF plays a critical role in perforin gene expression and the development of natural killer and NK-T cells. Immunity 17:437–49 Moriggl R, Topham DJ, Teglund S, Sexl V, McKay C, et al. 1999. Stat5 is required for IL-2-induced cell cycle progression of peripheral T cells. Immunity 10:249– 59 Lian RH, Kumar V. 2002. Murine natural killer cell progenitors and their

11 Feb 2004

19:16

424

79.

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

80.

81.

82.

83.

84.

85.

86.

87.

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

requirements for development. Semin. Immunol. 14:453–60 Colucci F, Caligiuri MA, Di Santo JP. 2003. What does it take to make a natural killer? Nat. Rev. Immunol. 3:413– 25 Colucci F, Soudais C, Rosmaraki E, Vanes L, Tybulewicz VL, Di Santo JP. 1999. Dissecting NK cell development using a novel alymphoid mouse model: investigating the role of the c-abl proto-oncogene in murine NK cell differentiation. J. Immunol. 162:2761–65 Colucci F, Di Santo JP. 2000. The receptor tyrosine kinase c-kit provides a critical signal for survival, expansion, and maturation of mouse natural killer cells. Blood 95:984–91 Barbosa MD, Nguyen QA, Tchernev VT, Ashley JA, Detter JC, et al. 1996. Identification of the homologous beige and Chediak-Higashi syndrome genes. Nature 382:262–65 Wang B, Biron C, She J, Higgins K, Sunshine MJ, et al. 1994. A block in both early T lymphocyte and natural killer cell development in transgenic mice with high-copy numbers of the human CD3E gene. Proc. Natl. Acad. Sci. USA 91:9402–6 Wang BP, Hollander GA, Nichogiannopoulou A, Simpson SJ, Orange JS, et al. 1996. Natural killer cell development is blocked in the context of aberrant T lymphocyte ontogeny. Int. Immunol. 8:939– 49 Flamand V, Shores EW, Tran T, Huang K, Lee E, et al. 1996. Delayed maturation of CD4-CD8-Fc gamma RII/III+ T and natural killer cell precursors in Fc epsilon RI gamma transgenic mice. J. Exp. Med. 184:1725–35 Kim S, Iizuka K, Aguila HL, Weissman IL, Yokoyama WM. 2000. In vivo natural killer cell activities revealed by natural killer cell-deficient mice. Proc. Natl. Acad. Sci. USA 97:2731–36 Kondo M, Weissman IL, Akashi K. 1997. Identification of clonogenic com-

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

mon lymphoid progenitors in mouse bone marrow. Cell 91:661–72 Galy A, Travis M, Cen D, Chen B. 1995. Human T, B, natural killer, and dendritic cells arise from a common bone marrow progenitor cell subset. Immunity 3:459–73 Wang JH, Nichogiannopoulou A, Wu L, Sun L, Sharpe AH, et al. 1996. Selective defects in the development of the fetal and adult lymphoid system in mice with an Ikaros null mutation. Immunity 5:537–49 Scott EW, Fisher RC, Olson MC, Kehrli EW, Simon MC, Singh H. 1997. PU.1 functions in a cell-autonomous manner to control the differentiation of multipotential lymphoid-myeloid progenitors. Immunity 6:437–47 DeKoter RP, Singh H. 2000. Regulation of B lymphocyte and macrophage development by graded expression of PU.1. Science 288:1439–41 Moore TA, von Freeden-Jeffry U, Murray R, Zlotnik A. 1996. Inhibition of gamma delta T cell development and early thymocyte maturation in IL-7−/− mice. J. Immunol. 157:2366–73 Nichogiannopoulou A, Trevisan M, Neben S, Friedrich C, Georgopoulos K. 1999. Defects in hemopoietic stem cell activity in Ikaros mutant mice. J. Exp. Med. 190:1201–14 DeKoter RP, Lee HJ, Singh H. 2002. PU.1 regulates expression of the interleukin-7 receptor in lymphoid progenitors. Immunity 16:297–309 Rodewald HR, Moingeon P, Lucich JL, Dosiou C, Lopez P, Reinherz EL. 1992. A population of early fetal thymocytes expressing Fc gamma RII/III contains precursors of T lymphocytes and natural killer cells. Cell 69:139–50 Carlyle JR, Michie AM, Furlonger C, Nakano T, Lenardo MJ, et al. 1997. Identification of a novel developmental stage marking lineage commitment of progenitor thymocytes. J. Exp. Med. 186:173–82 Carlyle JR, Zuniga-Pflucker JC. 1998. Requirement for the thymus in alphabeta

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

DYNAMIC LIFE OF NK CELLS

98.

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

99.

100.

101.

102.

103.

104.

105.

106.

107.

T lymphocyte lineage commitment. Immunity 9:187–97 Ikawa T, Kawamoto H, Fujimoto S, Katsura Y. 1999. Commitment of common T/Natural killer (NK) progenitors to unipotent T and NK progenitors in the murine fetal thymus revealed by a single progenitor assay. J. Exp. Med. 190:1617– 26 Douagi I, Colucci F, Di Santo JP, Cumano A. 2002. Identification of the earliest prethymic bipotent T/NK progenitor in murine fetal liver. Blood 99:463–71 Rosmaraki EE, Douagi I, Roth C, Colucci F, Cumano A, Di Santo JP. 2001. Identification of committed NK cell progenitors in adult murine bone marrow. Eur. J. Immunol. 31:1900–9 Waldmann TA, Tagaya Y. 1999. The multifaceted regulation of interleukin-15 expression and the role of this cytokine in NK cell differentiation and host response to intracellular pathogens. Annu. Rev. Immunol. 17:19–49 DiSanto JP, Muller W, Guy-Grand D, Fischer A, Rajewsky K. 1995. Lymphoid development in mice with a targeted deletion of the interleukin 2 receptor gamma chain. Proc. Natl. Acad. Sci. USA 92:377–81 Dubois S, Mariner J, Waldmann TA, Tagaya Y. 2002. IL-15Ralpha recycles and presents IL-15 In trans to neighboring cells. Immunity 17:537–47 Koka R, Burkett PR, Chien M, Chai S, Chan F, et al. 2003. Interleukin (IL)15Rα-deficient natural killer cells survive in normal but not IL-15Rα-deficient mice. J. Exp. Med. 197:977–84 Prlic M, Blazar BR, Farrar MA, Jameson SC. 2003. In vivo survival and homeostatic proliferation of natural killer cells. J. Exp. Med. 197:967–76 Imada K, Bloom ET, Nakajima H, Horvath-Arcidiacono JA, Udy GB, et al. 1998. Stat5b is essential for natural killer cell-mediated proliferation and cytolytic activity. J. Exp. Med. 188:2067–74 Heemskerk MH, Blom B, Nolan G,

108.

109.

110.

111.

112.

113.

114.

115.

116.

P1: IKH

425

Stegmann AP, Bakker AQ, et al. 1997. Inhibition of T cell and promotion of natural killer cell development by the dominant negative helix loop helix factor Id3. J. Exp. Med. 186:1597–602 Spits H, Couwenberg F, Bakker AQ, Weijer K, Uittenbogaart CH. 2000. Id2 and Id3 inhibit development of CD34+ stem cells into predendritic cell (pre-DC)2 but not into pre-DC1. Evidence for a lymphoid origin of pre-DC2. J. Exp. Med. 192:1775–84 Ikawa T, Fujimoto S, Kawamoto H, Katsura Y, Yokota Y. 2001. Commitment to natural killer cells requires the helix-loophelix inhibitor Id2. Proc. Natl. Acad. Sci. USA 98:5164–69 Kim S, Iizuka K, Kang HS, Dokun A, French AR, et al. 2002. In vivo developmental stages in murine natural killer cell maturation. Nat. Immunol. 3:523–28 Ryan JC, Turck J, Niemi EC, Yokoyama WM, Seaman WE. 1992. Molecular cloning of the NK1.1 antigen, a member of the NKR-P1 family of natural killer cell activation molecules. J. Immunol. 149:1631–35 Loza MJ, Perussia B. 2001. Final steps of natural killer cell maturation: a model for type 1-type 2 differentiation? Nat. Immunol. 2:917–24 Sivakumar PV, Gunturi A, Salcedo M, Schatzle JD, Lai WC, et al. 1999. Cutting edge: expression of functional CD94/NKG2A inhibitory receptors on fetal NK1.1+Ly-49− cells: a possible mechanism of tolerance during NK cell development. J. Immunol. 162:6976–80 Salcedo M, Colucci F, Dyson PJ, Cotterill LA, Lemonnier FA, et al. 2000. Role of Qa-1β-binding receptors in the specificity of developing NK cells. Eur. J. Immunol. 30:1094–101 Raulet DH, Vance RE, McMahon CW. 2001. Regulation of the natural killer cell receptor repertoire. Annu. Rev. Immunol. 19:291–330 Held W, Kunz B, Lowin-Kropf B, van de

11 Feb 2004

19:16

426

117.

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

118.

119.

120.

121.

122.

123.

124.

125.

126.

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

Wetering M, Clevers H. 1999. Clonal acquisition of the Ly49A NK cell receptor is dependent on the trans-acting factor TCF1. Immunity 11:433–42 Kunz B, Held W. 2001. Positive and negative roles of the trans-acting T cell factor-1 for the acquisition of distinct Ly-49 MHC class I receptors by NK cells. J. Immunol. 166:6181–87 Lohwasser S, Wilhelm B, Mager DL, Takei F. 2000. The genomic organization of the mouse CD94 C-type lectin gene. Eur. J. Immunogenet. 27:149–51 Rodriguez A, Carretero M, Glienke J, Bellon T, Ramirez A, et al. 1998. Structure of the human CD94 C-type lectin gene. Immunogenetics 47:305–9 Kubo S, Itoh Y, Ishikawa N, Nagasawa R, Mitarai T, Maruyama N. 1993. The gene encoding mouse lymphocyte antigen Ly49: structural analysis and the 50 -flanking sequence. Gene 136:329–31 Gosselin P, Makrigiannis AP, Nalewaik R, Anderson SK. 2000. Characterization of the Ly49I promoter. Immunogenetics 51:326–31 McQueen KL, Wilhelm BT, Takei F, Mager DL. 2001. Functional analysis of 50 and 30 regions of the closely related Ly49c and j genes. Immunogenetics 52:212–23 Wilhelm BT, McQueen KL, Freeman JD, Takei F, Mager DL. 2001. Comparative analysis of the promoter regions and transcriptional start sites of mouse Ly49 genes. Immunogenetics 53:215–24 Saleh A, Makrigiannis AP, Hodge DL, Anderson SK. 2002. Identification of a novel Ly49 promoter that is active in bone marrow and fetal thymus. J. Immunol. 168:5163–69 Dorfman JR, Raulet DH. 1998. Acquisition of Ly49 receptor expression by developing natural killer cells. J. Exp. Med. 187:609–18 Dokun AO, Kim S, Smith HR, Kang HS, Chu DT, Yokoyama WM. 2001. Specific and nonspecific NK cell activation during virus infection. Nat. Immunol. 2:951–56

127. Karlhofer FM, Hunziker R, Reichlin A, Margulies DH, Yokoyama WM. 1994. Host MHC class I molecules modulate in vivo expression of a NK cell receptor. J. Immunol. 153:2407–16 128. Sykes M, Harty MW, Karlhofer FM, Pearson DA, Szot G, Yokoyama W. 1993. Hematopoietic cells and radioresistant host elements influence natural killer cell differentiation. J. Exp. Med. 178:223–29 129. Sjostrom A, Eriksson M, Cerboni C, Johansson MH, Sentman CL, et al. 2001. Acquisition of external major histocompatibility complex class I molecules by natural killer cells expressing inhibitory ly49 receptors. J. Exp. Med. 194:1519–30 130. Zimmer J, Ioannidis V, Held W. 2001. H2D ligand expression by Ly49A+ natural killer (NK) cells precludes ligand uptake from environmental cells: implications for NK cell function. J. Exp. Med. 194:1531– 39 131. Johansson MH, Bieberich C, Jay G, Karre K, Hoglund P. 1997. Natural killer cell tolerance in mice with mosaic expression of major histocompatibility complex class I transgene. J. Exp. Med. 186:353–64 132. Hackett J Jr, Tutt M, Lipscomb M, Bennett M, Koo G, Kumar V. 1986. Origin and differentiation of natural killer cells. II. Functional and morphologic studies of purified NK-1.1+ cells. J. Immunol. 136:3124–31 133. Loza MJ, Peters SP, Zangrilli JG, Perussia B. 2002. Distinction between IL-13+ and IFN-gamma+ natural killer cells and regulation of their pool size by IL-4. Eur. J. Immunol. 32:413–23 134. Loza MJ, Zamai L, Azzoni L, Rosati E, Perussia B. 2002. Expression of type 1 (interferon gamma) and type 2 (interleukin13, interleukin-5) cytokines at distinct stages of natural killer cell differentiation from progenitor cells. Blood 99:1273–81 135. Zamai L, Ahmad M, Bennett IM, Azzoni L, Alnemri ES, Perussia B. 1998. Natural killer (NK) cell-mediated cytotoxicity: differential use of TRAIL and Fas ligand

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

DYNAMIC LIFE OF NK CELLS

136.

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

137.

138.

139.

140.

141.

142.

143.

144.

by immature and mature primary human NK cells. J. Exp. Med. 188:2375–80 Bennett IM, Zatsepina O, Zamai L, Azzoni L, Mikheeva T, Perussia B. 1996. Definition of a natural killer NKRP1A+/CD56−/CD16− functionally immature human NK cell subset that differentiates in vitro in the presence of interleukin 12. J. Exp. Med. 184:1845–56 Magram J, Connaughton SE, Warrier RR, Carvajal DM, Wu CY, et al. 1996. IL-12deficient mice are defective in IFN gamma production and type 1 cytokine responses. Immunity 4:471–81 Wu C, Ferrante J, Gately MK, Magram J. 1997. Characterization of IL-12 receptor beta1 chain (IL-12Rbeta1)-deficient mice: IL-12Rbeta1 is an essential component of the functional mouse IL-12 receptor. J. Immunol. 159:1658–65 Wu C, Wang X, Gadina M, O’Shea JJ, Presky DH, Magram J. 2000. IL-12 receptor beta 2 (IL-12R beta 2)-deficient mice are defective in IL-12-mediated signaling despite the presence of high affinity IL-12 binding sites. J. Immunol. 165:6221–28 Wang LL, Chu DT, Dokun AO, Yokoyama WM. 2000. Inducible expression of the gp49B inhibitory receptor on NK cells. J. Immunol. 164:5215–20 Takeda K, Tsutsui H, Yoshimoto T, Adachi O, Yoshida N, et al. 1998. Defective NK cell activity and Th1 response in IL-18-deficient mice. Immunity 8:383–90 Frey M, Packianathan NB, Fehniger TA, Ross ME, Wang WC, et al. 1998. Differential expression and function of Lselectin on CD56bright and CD56dim natural killer cell subsets. J. Immunol. 161:400–8 Salazar-Mather TP, Ishikawa R, Biron CA. 1996. NK cell trafficking and cytokine expression in splenic compartments after IFN induction and viral infection. J. Immunol. 157:3054–64 Salazar-Mather TP, Orange JS, Biron CA. 1998. Early murine cytomegalovirus (MCMV) Infection induces liver natural

145.

146.

147.

148.

149.

150.

151.

152.

P1: IKH

427

killer (NK) cell inflammation and protection through macrophage inflammatory protein 1-alpha (MIP-1-alpha)-dependent pathways. J. Exp. Med. 187:1–14 Salazar-Mather TP, Hamilton TA, Biron CA. 2000. A chemokine-to-cytokine-tochemokine cascade critical in antiviral defense. J. Clin. Invest. 105:985–93 Salazar-Mather TP, Lewis CA, Biron CA. 2002. Type I interferons regulate inflammatory cell trafficking and macrophage inflammatory protein 1alpha delivery to the liver. J. Clin. Invest. 110:321–30 Hanna J, Wald O, Goldman-Wohl D, Prus D, Markel G, et al. 2003. CXCL12 expression by invasive trophoblasts induces the specific migration of CD16 negative human natural killer cells. Blood. 102:1569– 77 Beider K, Nagler A, Wald O, Franitza S, Dagan-Berger M, et al. 2003. Involvement of CXCR4 and IL-2 in the homing and retention of human NK and NK T cells to the bone marrow and spleen of NOD/SCID mice. Blood. 102:1951–58 Dokun AO, Chu DT, Yang L, Bendelac AS, Yokoyama WM. 2001. Analysis of in situ NK cell responses during viral infection. J. Immunol. 167:5286–93 Andrews DM, Farrell HE, Densley EH, Scalzo AA, Shellam GR, Degli-Esposti MA. 2001. NK1.1+ cells and murine cytomegalovirus infection: What happens in situ? J. Immunol. 166:1796–802 Fehniger TA, Cooper MA, Nuovo GJ, Cella M, Facchetti F, et al. 2003. CD56bright natural killer cells are present in human lymph nodes and are activated by T cell-derived IL-2: a potential new link between adaptive and innate immunity. Blood 101:3052–57 Glas R, Franksson L, Une C, Eloranta ML, Ohlen C, et al. 2000. Recruitment and activation of natural killer (NK) cells in vivo determined by the target cell phenotype. An adaptive component of NK cell-mediated responses. J. Exp. Med. 191:129–38

11 Feb 2004

19:16

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

428

AR

AR210-IY22-13.tex

YOKOYAMA

¥

KIM

¥

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRENCH

153. Tay CH, Welsh RM. 1997. Distinct organdependent mechanisms for the control of murine cytomegalovirus infection by natural killer cells. J. Virol. 71:267–75 154. Hackett J, Jr., Bennett M, Kumar V. 1985. Origin and differentiation of natural killer cells. I. Characteristics of a transplantable NK cell precursor. J. Immunol. 134:3731– 38 155. Ranson T, Vosshenrich CA, Corcuff E, Richard O, Muller W, Di Santo JP. 2003. IL-15 is an essential mediator of peripheral NK cell homeostasis. Blood 101:4887–93 156. Wang JW, Howson JM, Ghansah T, Desponts C, Ninos JM, et al. 2002. Influence of SHIP on the NK repertoire and allogeneic bone marrow transplantation. Science 295:2094–97 157. Cooper MA, Bush JE, Fehniger TA, VanDeusen JB, Waite RE, et al. 2002. In vivo evidence for a dependence on interleukin 15 for survival of natural killer cells. Blood 100:3633–38 158. Marks-Konczalik J, Dubois S, Losi JM, Sabzevari H, Yamada N, et al. 2000. IL2-induced activation-induced cell death is inhibited in IL-15 transgenic mice. Proc. Natl. Acad. Sci. USA 97:11445–50 159. Fehniger TA, Suzuki K, Ponnappan A, VanDeusen JB, Cooper MA, et al. 2001. Fatal leukemia in interleukin 15 transgenic mice follows early expansions in natural killer and memory phenotype CD8+ T cells. J. Exp. Med. 193:219–31 160. Goldrath AW, Sivakumar PV, Glaccum M, Kennedy MK, Bevan MJ, et al. 2002. Cytokine requirements for acute and basal homeostatic proliferation of naive and memory CD8+ T cells. J. Exp. Med. 195:1515–22 161. Tan JT, Ernst B, Kieper WC, LeRoy E, Sprent J, Surh CD. 2002. Interleukin (IL)-15 and IL-7 jointly regulate homeostatic proliferation of memory phenotype CD8+ cells but are not required for memory phenotype CD4+ cells. J. Exp. Med. 195:1523–32

162. Prlic M, Lefrancois L, Jameson SC. 2002. Multiple choices: regulation of memory CD8 T cell generation and homeostasis by interleukin (IL)-7 and IL-15. J. Exp. Med. 195:F49–52 163. Becker TC, Wherry EJ, Boone D, MuraliKrishna K, Antia R, et al. 2002. Interleukin 15 is required for proliferative renewal of virus-specific memory CD8 T cells. J. Exp. Med. 195:1541–48 164. Judge AD, Zhang X, Fujii H, Surh CD, Sprent J. 2002. Interleukin 15 controls both proliferation and survival of a subset of memory-phenotype CD8+ T cells. J. Exp. Med. 196:935–46 165. Matsuda JL, Gapin L, Sidobre S, Kieper WC, Tan JT, et al. 2002. Homeostasis of V alpha 14i NKT cells. Nat. Immunol. 3:966–74 166. Ranson T, Vosshenrich CA, Corcuff E, Richard O, Laloux V, et al. 2003. IL-15 availability conditions homeostasis of peripheral natural killer T cells. Proc. Natl. Acad. Sci. USA 100:2663–68 167. Biron CA, Turgiss LR, Welsh RM. 1983. Increase in NK cell number and turnover rate during acute viral infection. J. Immunol. 131:1539–45 168. Biron CA, Sonnenfeld G, Welsh RM. 1984. Interferon induces natural killer cell blastogenesis in vivo. J. Leukoc. Biol. 35: 31–37 169. Tough DF, Borrow P, Sprent J. 1996. Induction of bystander T cell proliferation by viruses and type I interferon in vivo. Science 272:1947–50 170. Zhang X, Sun S, Hwang I, Tough DF, Sprent J. 1998. Potent and selective stimulation of memory-phenotype CD8+ T cells in vivo by IL-15. Immunity 8:591– 99 171. Mattei F, Schiavoni G, Belardelli F, Tough DF. 2001. IL-15 is expressed by dendritic cells in response to type I IFN, doublestranded RNA, or lipopolysaccharide and promotes dendritic cell activation. J. Immunol. 167:1179–87 172. Brown MG, Dokun AO, Heusel JW, Smith

11 Feb 2004

19:16

AR

AR210-IY22-13.tex

AR210-IY22-13.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMIC LIFE OF NK CELLS HR, Beckman DL, et al. 2001. Vital involvement of a natural killer cell activation receptor in resistance to viral infection. Science 292:934–37 173. Lee SH, Girard S, Macina D, Busa M, Zafer A, et al. 2001. Susceptibility to mouse cytomegalovirus is associated with deletion of an activating natural killer cell receptor of the C-type lectin superfamily. Nat. Genet. 28:42–45 174. Lee SH, Zafer A, de Repentigny Y, Kothary R, Tremblay ML, et al. 2003. Transgenic expression of the activating natural killer receptor Ly49H confers re-

P1: IKH

429

sistance to cytomegalovirus in genetically susceptible mice. J. Exp. Med. 197:515– 26 175. Arase H, Mocarski ES, Campbell AE, Hill AB, Lanier LL. 2002. Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors. Science 296:1323–26 176. Kasaian MT, Whitters MJ, Carter LL, Lowe LD, Jussif JM, et al. 2002. IL-21 limits NK cell responses and promotes antigen-specific T cell activation: a mediator of the transition from innate to adaptive immunity. Immunity 16:559–69

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:405-429. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104549

Annu. Rev. Immunol. 2004. 22:431–56 doi: 10.1146/annurev.immunol.22.012703.104549 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on November 10, 2003

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS Anthony P. Manderson,1 Marina Botto,1 and Mark J. Walport1,2 1

Rheumatology Section, Division of Medicine, Faculty of Medicine, Imperial College, Hammersmith Campus, London W12 0NN, United Kingdom; email: [email protected], [email protected] 2 The Wellcome Trust, London NW1 2BE, United Kingdom; email: [email protected]

Key Words autoimmunity, tolerance, apoptosis ■ Abstract Complement has both beneficial and deleterious roles in the pathogenesis of systemic lupus erythematosus (SLE). On the one hand, patients with SLE present with decreased complement levels and with complement deposition in inflammed tissues, suggestive of a harmful role of complement in the effector phase of disease. On the other hand, homozygous deficiency of any of the classical pathway proteins is strongly associated with the development of SLE. There are two main hypotheses to explain these observations. The first invokes an important role for complement in the physiological waste-disposal mechanisms of dying cells and immune complexes. The second hypothesis is based around the role of complement in determining the activation thresholds of B and T lymphocytes, with the proposal that complement deficiency causes incomplete maintenance of peripheral tolerance. These two hypotheses are not mutually exclusive. In addition, there is evidence for a contribution from other genetic factors in determining the phenotype of disease in the absence of complement.

INTRODUCTION Links between the complement system and systemic lupus erythematosus (SLE) were first identified when it was discovered that complement levels were decreased in patients with SLE. In addition, it was noted that autoantibodies and complement were deposited in inflammatory lesions in the glomeruli of patients with renal involvement. The original model proposed to explain these observations was that immune complexes containing autoantigens and autoantibodies activate complement, and deposition of these complexes within tissues causes inflammatory injury. This model implies that deficiency in complement would protect patients from the development of tissue injury in SLE. 0732-0582/04/0423-0431$14.00

431

11 Feb 2004

19:19

432

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

However, the reverse appears to be the case. It was discovered that hereditary homozygous deficiency of the early proteins of the classical pathway of complement is strongly associated with the development of SLE. This suggests a protective role for complement against the development of SLE. In this review, we first discuss these clinical associations between the complement system and SLE. We review findings from animals with complement deficiency that have confirmed the link between deficiency of classical pathway proteins and predisposition to the development of SLE. Second, we review the hypotheses that have been advanced to explain these associations.

CLINICAL ASSOCIATIONS OF COMPLEMENT DEFICIENCIES AND SLE Inherited Complement Deficiencies in Humans HOMOZYGOUS COMPLEMENT DEFICIENCY In recent years, compelling evidence has emerged that homozygous deficiency of any of the early components of the classical pathway of complement activation (C1q, C1r, C1s, C4, and C2) predisposes to the development of SLE. In fact, these deficiencies are the strongest susceptibility factors for the development of SLE identified up to now in humans. It has also become apparent that the disease association is rather precise, with subtle but important variations according to the position of the missing protein in the triggered cascade. Overall, there appears to be a hierarchy of association of both disease prevalence and severity within the classical pathway, with patients deficient in one of C1-complex proteins (1–3) or C4 (1, 4) exhibiting the strongest prevalence (>80%) and the most severe disease, whereas the strength of the association decreases significantly in C2-deficient patients (1, 5). In this context, as most individuals with homozygous deficiency of C2 appear to remain asymptomatic, it is difficult to ascertain the correct strength of the association of C2 deficiency with SLE. One report estimated that SLE occurs in approximately one third of C2deficient patients (5); however, population studies have shown that the prevalence of homozygous C2-deficient individuals is of the order of 1/20,000 in western Europe (1). In light of this, the true strength of the association between SLE and C2 deficiency is likely to be that no more than 10% of C2-deficient subjects develop SLE. Furthermore, C2 deficiency is associated with a female-to-male ratio of 7:1, respectively, which is similar to the ratio observed among patients with SLE in the absence of complement deficiencies. This is in direct contrast to the ratio of females to males seen in patients with SLE who are deficient in C1 or C4, which is approximately 1:1, illustrating the very powerful effect of these deficiencies that overcomes the normal female preponderance seen in SLE. Finally, C3 deficiency, described only in 23 subjects, has only been associated with lupus-like disease in three patients in two families (1, 6–8). A small number of SLE patients have also been shown to carry deficiencies in components of the membrane attack complex (C5, C6, C7, C8, and C9). However, these represent only very rare or even unique

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

COMPLEMENT AND SLE

P1: IKH

433

associations, and therefore it is likely that there is no causal association between deficiency in these proteins and increased susceptibility to SLE. In these cases, the association between membrane attack complex protein deficiencies and SLE may be an example of an ascertainment artifact. A detailed description of the clinical cases falls outside the scope of this review, and the interested reader is referred to a recent review by Pickering et al. (1). Although homozygous complement deficiencies are extremely rare, they tell us a great deal about the normal physiological activities of the complement system in humans and may provide important insights for understanding the underlying mechanisms leading to the development of an autoimmune response. Collectively, the clinical observations suggest that a physiological activity of the early part of the classical pathway of complement, independent of C3 activation, provides a protective role against the development of SLE. In support of this hypothesis, there are three additional sets of observations in humans. First, two large population surveys in Switzerland and in Japan failed to identify healthy individuals with homozygous deficiency of any classical pathway protein or of C3 (9, 10). Second, studies of the inbred Turkish populations among which C1q-deficient patients had been identified revealed no asymptomatic C1q-deficient individuals (11). Third, there is a very high concordance of SLE among siblings with C1q, C1r/C1s, and C4 deficiency (90%, 67%, and 80%, respectively) (1). These concordance data are strikingly higher when compared to those observed in twin studies, which showed concordance of SLE of 2% among dizygotic twin and 24% among monozygotic twins (12). PARTIAL INHERITED COMPLEMENT DEFICIENCIES, NULL ALLELES, AND POLYMORPHISMS Although there is strong circumstantial evidence that homozygous de-

ficiency of one of the early components of the classical complement pathway increases susceptibility to SLE, the results are less conclusive with respect to partial complement deficiencies. This is in part due to the difficulty of identifying heterozygous individuals using phenotypic analyses. The serum levels of complement proteins are highly variable in normal individuals, and there is overlap between the protein concentrations observed in normal homozygous individuals and heterozygous subjects. In addition, active SLE is commonly associated with activation and consumption of classical pathway complement proteins. These difficulties can be overcome only if the molecular basis of common null alleles of complement proteins and polymorphisms affecting complement levels is known. Subsequent genotypic analysis of patient cohorts can then be performed to determine possible disease associations. So far this has been possible for only a few complement components. C1q-deficient patients have a very high incidence of SLE; however, no clinical manifestations have been observed among any heterozygous C1q-deficient siblings of the homozygous-deficient patients. Interestingly, a recent report demonstrated a single nucleotide polymorphism (SNP) in the C1qA gene, resulting in decreased serum levels of C1q and an increased incidence of this SNP in patients with

11 Feb 2004

19:19

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

434

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

subacute cutaneous lupus erythematosus (13). The incidence of this SNP in patients with systemic disease has not been reported. In human serum C4 exists in two isoforms encoded by separate genes that have arisen by gene duplication, C4A and C4B, with the genes located in tandem within the MHC class III region. Both genes are highly polymorphic, and the relatively high frequency of null alleles in the C4A and C4B genes in the normal population highlighted the potential for association between partial deficiency of C4 and SLE. Several independent studies investigating different ethnic groups have shown a strong association between null alleles in C4A and SLE (1, 14). However, not all studies have confirmed this association. In contrast to the associations between C4A deficiency and SLE, most studies have found no increase of C4B null alleles in patients with SLE. The reason for this discrepancy between the associations of C4A and C4B deficiencies and SLE is most likely related to the effector functions of the individual proteins. Although C4A and C4B probably arose by gene duplication, the proteins encoded by these genes have slightly different activities, especially in respect to their acceptor sites (15). C4A binds preferentially to amine groups and C4B to hydroxyl groups. It is estimated that approximately 1 in 170 western European Caucasians are carriers for the C2 null allele. However, with the exception of one study, no significant association has been found linking C2 heterozygosity in any population to increased susceptibility to the development of SLE (1, 16). The mannose-binding lectin (MBL) pathway is homologous to the classical complement pathway, with MBL replacing C1q as the recognition component. MBL is a pattern-recognition molecule that binds to arrays of mannose found on many bacteria. A number of polymorphisms have been identified within both structural and promoter regions of the gene, and these polymorphisms result in wide variation in the level of circulating MBL (17). In most studies, polymorphisms associated with low levels of MBL activity were found at higher frequencies in patients with SLE, and conversely, polymorphisms that result in high-level MBL expression were underrepresented in SLE patients (18–21). Two studies have also demonstrated that individuals carrying null alleles for C4B, in combination with a dysfunctional MBL allele, had a higher likelihood of developing SLE (20, 21). Taken together, the current data suggest that MBL may play a similar role to C1q in conferring protection against the development of SLE, although with a much weaker effect, possibly reflecting the lower levels of MBL in circulation compared to those of C1q.

Acquired Complement Deficiencies in Humans In addition to hereditary defects of the classical complement pathway components, there is supporting evidence from other disease conditions that acquired low levels of complement components predispose individuals to the development of SLE. There is an increased prevalence of lupus disease among patients with hereditary angioedema who, because of an autosomal dominant functional deficiency of

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

COMPLEMENT AND SLE

P1: IKH

435

C1-inhibitor, fail to regulate classical pathway complement activation, resulting in chronically low levels of C4 and C2. Among these patients, there is a female predominance and a high incidence of antinuclear antibodies, skin lesions, and photosensitivity, all common characteristics in SLE (1, 22). Patients with C3 nephritic factors, autoantibodies that bind and stabilize the alternative pathway C3 convertase, leading to uncontrolled complement activation, have chronic hypocomplementemia. These patients can initially present with a range of clinical symptoms; however, a number of them have been reported to subsequently develop features of SLE, in each case many years after the initial development of nephritic factor (23, 24).

Complement Activation in SLE Complement is strongly activated in patients with SLE. Deposits of C3, C4, and associated complement proteins can be easily detected in biopsies from inflamed tissues from patients with SLE. Complement activity and classical pathway protein levels are generally reduced in relation to disease activity and increase following treatment. The initial cause of complement activation in SLE is thought to be the formation of high levels of immune complexes (IC) that, in turn, activate complement via the classical pathway. However, following the initial disease onset, there are a number of factors that may influence the degree of reduction of serum levels of complement components. These include disease activity per se, the rate of production versus catabolism, and importantly, the presence of autoantibodies directed against complement proteins, such as anti-C1q antibodies. Consistent with this, studies on cohorts of SLE patients have demonstrated that complement levels in the circulation provide only a rough guide to disease activity (25, 26). The associations between complement and SLE seem paradoxical. Active SLE is accompanied by activation of the complement pathway, and there is circumstantial evidence that complement participates in causing inflammatory damage to tissues in SLE. This proinflammatory role of complement in SLE is difficult to reconcile with the clinical observations, described above, that inherited deficiency of classical pathway complement proteins plays a causal role in the induction of SLE. Thus, it is now generally thought that complement is a double-edged sword in SLE. On the one hand, it provides important protective roles against the development of SLE. On the other hand, a direct pathogenic role for complement may still influence disease expression. The initial observations showing a link between complement deficiency and autoimmunity have all come from studies in humans. Thus, in recent years it became necessary to generate animal models of complement deficiency in order to perform the experiments that may provide an explanation for this association. A number of mouse strains have now been generated with classical pathway complement deficiencies, and their analysis has already provided important insights into the pathogenesis of SLE. We first review the relevant data that have emerged

11 Feb 2004

19:19

436

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

from the studies of these mice and then discuss some of the current hypotheses on the role of complement in the pathogenesis of SLE.

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Animal Models Linking Complement Deficiency and SLE The genes for C1q, C4, C3, and CD35/CD21 (CR1/2) have been successfully targeted in mice, and these mouse strains showed a hierarchy of disease susceptibility similar to that observed in humans. C1q- and C4-deficient mice on the hybrid (129 × C57BL/6) genetic background were shown to develop higher levels of autoantibodies compared to strain-matched controls and to have histological evidence of glomerulonephritis by 8–10 months of age (27–30). In contrast, no disease manifestations were detected in C3- or CR1/2-deficient mice on this genetic background (27), indicating that, like in humans, the lack of C3 was not critical for the development of SLE. Interestingly, in mice the effects of complement deficiency on disease expression is dependent on the background genes (Table 1). In initial studies of C1qdeficient mice, it was observed that C1q deficiency is associated with antinuclear antibody production and glomerulonephritis in hybrid 129 × C57BL/6 mice but not in either parental strain (28, 30), suggesting that additional susceptibility genes were required for disease expression. Similarly, C4-deficient mice backcrossed onto C57BL/6 had no histological evidence of glomerulonephritis, although increased levels of anti-double-strand DNA antibodies were still detected (29). Further evidence that background genes provide essential contributions to the development of SLE in the absence of complement was given by studies introducing complement deficiency onto lupus-prone genetic backgrounds. Backcrossing C1q-deficient mice onto the autoimmune MRL/Mp background accelerated both the onset and severity of the autoimmune disease (28), and the introduction of the lpr mutation to C4-deficient mice (C57BL/6 × 129) (32) and CR1/2-deficient mice (C57BL/6 × 129 or C57BL/6) (32, 34) had the same effect. However, it is of note that the introduction of the lpr mutation to C1q-deficient mice

TABLE 1 Effect of genetic background on the level of autoimmune disease in complement-deficient mice 129 × C57BL/6

C57BL/6

MRL/Mp

C57BL/6 × lpr/lpr

MRL × lpr/lpr

Control

+

−

++

++

+++

C1qa−/−

++

−

+++

++

+++

(28, 30)

C4−/−

++

+

ND

+++b

ND

(29, 31, 32)

C3−/−

+

ND

ND

++

Cr2−/−

+

ND

ND

+++

a

a

ND = not determined.

b

On hybrid 129 × C57BL/6 background.

b

References

++

(27, 31–33)

ND

(32b, 34)

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

COMPLEMENT AND SLE

P1: IKH

437

backcrossed onto C57BL/6 or MRL did not have any effect on disease progression compared to strain-matched lpr/lpr mice (28). Several environmental and genetic factors may contribute to these differences, including the chromosome location of each complement gene. For example, the C4 gene resides within the MHC region, and it is possible that 129-specific polymorphisms tightly linked to the C4 gene may explain the preserved autoreactivity observed in the C4-deficient mice after backcrossing onto C57BL/6. In this context, it has been suggested that the influence of background genes on the development of SLE appears to be greater in complement-deficient mice than in humans. However, although more than 90% of humans with C1q deficiency develop at least one manifestation of SLE, only a proportion of them (approximately one third) develop glomerulonephritis (1). This variation in prevalence of lupus glomerulonephritis in humans with C1q deficiency may be analogous to the genetic effects that have been observed in different strains of C1q-deficient mice. In summary, both the clinical observations and animal models have clearly established that the classical complement pathway proteins play a protective role against the development of SLE. So what are the complement-dependent mechanisms that could mediate this protection from disease, resulting in predisposition to the development of autoimmunity in their absence?

CURRENT WORKING HYPOTHESES FOR LINKING COMPLEMENT DEFICIENCY AND SLE There are two main hypotheses to explain the causal link between complement deficiency and the development of SLE, neither of which are mutually exclusive (Figure 1). The first involves the role of complement in physiological wastedisposal mechanisms, in particular the clearance of dying cells and immune complexes. It has been proposed that deficiency of complement impairs a normal mechanism of waste disposal and that dying cells and tissue injury provide a source of autoantigens and inflammatory cues that drive the production of autoantibodies and further tissue inflammation. The second hypothesis proposes that complement plays a role in determining the thresholds of activation of B and T lymphocytes and that complement deficiency causes autoantibody production and SLE by impairing the normal mechanisms of tolerance induction and maintenance. We discuss each of these hypotheses in turn.

Clearance of Dying Cells AUTOANTIGENS DERIVED FROM APOPTOTIC CELLS DRIVE SLE Despite the clinical heterogeneity of SLE, different murine lupus models and SLE patients are unified by the presence of autoantibodies specific for ubiquitous self-antigens. Indeed, the repertoire of these target autoantigens is surprisingly limited, and among these, antibodies directed against nuclear components feature prominently. This implies

11 Feb 2004

19:19

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

438

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

Figure 1 Model describing the association of SLE and complement deficiency.

that a general immune dysregulation cannot account for the restricted autoantibody repertoire observed in lupus, and understanding how these specific autoantigens come to drive the pathogenic autoimmune response is paramount to deciphering the etiology of disease. A breakthrough in determining the likely source of the autoantigenic drive in SLE was provided by Rosen and colleagues, who established that a number of autoantigens targeted in SLE are localized in high concentrations within and on the surface of apoptotic cell blebs (35, 36). This led to the hypothesis that dying cells provide the source of autoantigens responsible for driving autoantibody production in SLE. Defects in the clearance mechanisms for dying cells could enhance the chance of developing SLE. There is a now a large amount of direct and indirect evidence describing defects in the uptake of apoptotic cells being associated with autoimmunity. In humans, monocyte-derived macrophage isolated from SLE patients have significantly reduced ability to phagocytose apoptotic cells when incubated in the presence of autologous serum in vitro (37). In some patients with SLE, nucleosomal deposits can be observed within the skin and in renal lesions (38, 39). In another group of

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

COMPLEMENT AND SLE

P1: IKH

439

SLE patients, apoptotic cells were observed within lymph node germinal centers, and the numbers of tingible body macrophage, which usually remove apoptotic nuclei, were significantly reduced in these patients (40). In addition, free nucleosomes were found in the blood of SLE patients, with the highest levels observed in patients with the most active disease (41, 42). Similar observations have been made in murine models of lupus. High levels of circulating nucleosomes, including histones and DNA, were detected in the serum from NZB/W F1 mice and MRL/lpr mice (43, 44). In addition, injection of apoptotic cells into nonautoimmune mouse strains induced the formation of autoantibodies characteristic of SLE (45). These results directly imply that the material from apoptotic cells can induce specific autoimmunity characteristic of SLE. Additional insight linking the ineffective clearance of apoptotic cell material and the development of SLE is provided by mice carrying genetic deficiencies that also develop a progressive SLE-like disease. These include genes that are thought to protect the immune system from potentially pathogenic chromatin by masking its presence (serum amyloid P component) (46) or directly digesting it (DNase I) (47, 48). Finally, mice deficient in membrane tyrosine kinase c-mer carry a defect that is thought to cause a loss of phosphatidylserine recognition by phagocytic cells, resulting in decreased ability to bind to and ingest apoptotic cells. Mice lacking membrane tyrosine kinase c-mer develop a progressive autoimmunity characterized by the presence of high levels of antinuclear antibodies (49, 50). So what evidence is there that complement is involved in the clearance of apoptotic cells, which could promote the development of SLE in its absence? Activation of complement by necrotic cells is well established; however, the importance of complement in the process of scavenging apoptotic cells is a recent discovery. Complement was first implicated in this process by the observation that C1q could bind directly and specifically to surface blebs on apoptotic keratinocytes, with deposition increasing as the blebs mature (51). Subsequently, the binding of C1q to apoptotic blebs was shown to occur via the globular heads of C1q, and it was also shown to induce the activation of the classical pathway (52, 53). These observations led to the hypothesis that deficiency of complement, especially C1q, may predispose to autoimmunity as a consequence of impaired clearance of apoptotic cells. In vitro studies in humans have implicated both the classical and alternative complement pathways in enhancing the uptake of apoptotic cells by peripheral blood-derived macrophage (54). The presence of activated C3 fragments on the surface of the apoptotic cells, and direct blocking experiments, have suggested that binding and ingestion is mediated through CR3 and/or CR4 on the surface of phagocytes (54). In addition, in a small number of patients carrying hereditary deficiencies in C1q, monocyte-derived macrophages showed defective uptake of apoptotic cells in the presence of autologous serum, a defect completely reversible by the addition of exogenous C1q (55). Interestingly, a similar defect is observed in the uptake of apoptotic cells by monocyte-derived macrophage from SLE patients (37).

11 Feb 2004

19:19

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

440

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

The strongest evidence that complement may play an essential role in the clearance of apoptotic cells came from studies in complement-deficient mice. In addition to in vitro data demonstrating enhanced uptake of apoptotic cells by serum complement (54), in vivo studies showed the existence of a hierarchy for classical complement proteins in mediating the internalization of apoptotic cells by peritoneal macrophage, with deficiency in C1q displaying the most severe phenotype (55). In addition, C1q-deficient mice developed a proliferative glomerulonephritis characterized by the presence of multiple apoptotic cell bodies, suggesting impaired ability to clear cellular debris (30). Importantly, these findings parallel the hierarchy of disease susceptibility and severity observed in patients and in murine models of complement deficiency. Collectively these data strongly support the proposed hypothesis that deficiency in complement predisposes to the development of SLE through inefficient removal of potentially pathogenic apoptotic cell debris. The process that mediates the ingestion of apoptotic cells is remarkably complex, pointing to possible roles for many phagocytic receptors, various bridging molecules, and several specific molecules on the dying cells that mark them for uptake (56). With respect to complement, in vitro studies have demonstrated that collectins such as C1q, as well as MBL, can drive the ingestion of apoptotic cells through interaction with calreticulin and CD91 on the phagocyte (57, 58). In addition, in humans, CR3 and CR4 have also been implicated in the recognition of iC3b bound to the surface of apoptotic cells (54). To date, the identification of additional receptors mediating the complement-dependent uptake of opsonized apoptotic material by phagocytes has remained elusive. In addition to the direct binding of C1q to apoptotic cells, a number of other molecules that can bind C1q and activate the classical complement pathway have been shown to recognize the modified cell membranes that form during the process of apoptosis (59). Among them, human c-reactive protein (CRP) (60) and natural IgM, but not IgG (61), have been shown to bind to apoptotic cells and activate the classical pathway of complement. In this context it is of note that mice lacking soluble IgM develop a lupus-like disease (62, 63). The molecules on the surface of apoptotic cells recognized by CRP and IgM have been determined, and independent studies have suggested almost identical recognition patterns, binding to lysophospholipids containing phosphorylcholine (60, 61). In addition, the binding of CRP to apoptotic cells was shown to protect the cells from assembling the membrane attack complex, and to mediate the noninflammatory uptake of apoptotic cells, a mechanism dependent on the presence of C1q (60). These accessory proteins may, in part, account for some of the complement deposition on dying cells, leading to their safe clearance by complement-dependent mechanisms. Further investigations are required to clarify and further define their importance for the complement-dependent clearance of apoptotic cells. It is interesting to note that the binding of complement components, as well as IgM and pentraxins, to apoptotic cells appears to occur only during a late phase of the process of apoptosis, and to cells that have undergone secondary necrosis (64). This could suggest that a hierarchy exists in the clearance mechanisms of

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

COMPLEMENT AND SLE

P1: IKH

441

apoptotic cells, with uptake by local macrophage representing a very early event, and complement-mediated processes a rather late event. Recently, it was reported that cells undergoing apoptosis only release low levels of nucleosomes, one of the major autoantigens in SLE, during the first 24 hours postinduction of apoptosis (65). Under normal situations, this should provide ample opportunity for efficient clearance by macrophage phagocytosis; however, in situations with high local rates of apoptosis and low or impaired phagocytic capacity, the additional opsonic activity of complement could become essential to ensure the safe clearance of the nuclear material. Therefore, a selective defect in the complement-dependent mechanisms could result in an increased release of apoptotic material into the circulation, driving an autoimmune response. This hypothesis may also explain why approximately one third of SLE patients, who are not C1q-deficient, develop autoantibodies to C1q. If C1q bound to apoptotic cells normally promotes their physiological clearance, in SLE patients poor clearance may lead to C1q-containing complexes on the surface of dying cells becoming antigenic and triggering an autoantibody response against C1q. So how might an increased peripheral load of autoantigens from dying cells lead to the development of SLE? IMMUNITY VERSUS TOLERANCE The process of apoptosis and the clearance of apoptotic cells by phagocytic cells usually occurs in the absence of tissue injury or inflammation. Breakdown in the physiological process of clearance can lead to the release of material with powerful proinflammatory properties as well as releasing potentially immunogenic self-antigens into the circulation. The release of selfantigens in a proinflammatory environment is proposed to result in their uptake by professional antigen presenting cells and the inappropriate presentation of selfantigens to T cells, leading to autoimmunity. What evidence exists to support this model for the development of autoimmune disease? The processes involved in the uptake and processing of apoptotic and necrotic material and how abnormalities in these pathways could lead to the development of SLE are currently an extremely active area of research. Immature dendritic cells (iDCs) avidly take up both apoptotic and necrotic cells, and the level of maturation regulates the consequential immune response, with maturation being the main difference between the induction of tolerance and immune activation (56). Therefore, a great potential for autoimmunity exists, and deciphering the mechanisms responsible for dendritic cell (DC) maturation is crucial to understanding the etiology of SLE. A number of specific signals that lead to the maturation of DCs have been identified and include viral and bacterial proteins, proinflammatory cytokines, high antigen load (66), and potentially uptake of necrotic cells (67, 68). The process of phagocytosis of apoptotic cells by macrophage and iDCs actively suppresses inflammation and promotes peripheral tolerance through the increased production of anti-inflammatory mediators such as transforming growth factor-β (TGF-β) and IL-10, and inhibition of proinflammatory cytokines such as tumor necrosis factor-α (TNF-α) (69–72). Even strong proinflammatory signals, such as LPS-driven maturation of DCs, can be suppressed by the engulfment of apoptotic

11 Feb 2004

19:19

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

442

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

cells by iDCs, also reducing the capacity of the DCs to directly stimulate T cells. In addition, cells that are actively undergoing apoptosis can also directly secrete cytokines with immunosuppressive properties (73, 74). So how can autoimmune proinflammatory responses to apoptotic tissue be generated when abundant data demonstrate that the clearance of apoptotic cells in vivo is normally a tolerogenic event? In contrast to the uptake of apoptotic cells, a number of studies have demonstrated that uptake of necrotic cells by iDCs can induce their maturation. This is an important consideration to the development of SLE, as it has been proposed that as a result of an impaired clearance the apoptotic cells may undergo secondary necrosis, releasing proinflammatory signals that could affect the maturation state of the ingesting DC and inducing a break in peripheral tolerance. A number of studies using tumor cell lines have shown the ability of necrotic cells to induce DC maturation in vitro (75–77). Another study also confirmed these observations in tumor cell lines; however, it did not observe DC maturation following uptake of necrotic cells induced from primary cell cultures (67). Furthermore, a recent study casts doubt on the interpretations of these findings. The investigators examined the ability of a number of apoptotic and necrotic cells to induce DC maturation and found that necrotic cells induced DC maturation only from cell lines that were contaminated by mycoplasma (68). Clearly, further investigations will be required to resolve whether cells that have undergone secondary necrosis play a role in inducing autoimmunity. There is also now a large body of evidence suggesting that in SLE the maturation of DCs could be mediated by the cytokine interferon-α (IFN-α) (78, 79). This cytokine is usually produced in response to infections and, in addition to mediating the maturation of DCs, may also promote the development of autoimmunity through its direct ability to regulate the activation thresholds of T and B cells (80). An increased proportion of functionally active monocyte-derived DCs has been observed in the blood of SLE patients. In addition, monocytes, normally quiescent cells, when isolated from the blood of SLE patients, are able to induce a strong mixed lymphocyte reaction, a property usually used to define mature DCs (81). Both of these observations suggest that DC maturation is abnormal in SLE. It was also shown that incubating normal monocytes with serum of some SLE patients induced the differentiation of normal monocytes to DCs, with the factor in SLE serum responsible for this spontaneous maturation of DCs shown to be IFN-α (81). Although high levels of IFN-α are not present in serum from all SLE patients, it has been recently demonstrated, using microarray technology, that peripheral blood mononuclear cells from all SLE patients with active disease have a dysregulated expression of genes induced through the IFN pathway (82–84). This discrepancy may reflect the production of IFN-α at local sites by DCs not measurable in the circulation. In addition, it has become apparent that prolonged IFN treatment in patients with virus infections or malignant diseases can induce symptoms much like those found among patients with active lupus (78). In agreement with the observations in humans, IFN-α has also been implicated in disease exacerbation

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

COMPLEMENT AND SLE

P1: IKH

443

in murine models of lupus. Direct administration of IFN-α accelerated disease in NZB/W mice (85), and introduction of a deficiency in the type I IFN receptor reduced lupus-like disease in NZB mice, characterized by decreases in autoantibodies, kidney disease, and mortality (86). So, what cells are responsible for producing the IFN-α, and what causes its production? Plasmacytoid DCs (pDCs) are capable of producing unusually large amounts of IFN-α in response to inducers such as viral or bacterial infections, and surprisingly also when exposed to sera of patients with lupus (78, 87). This is consistent with the observation that very high numbers of pDCs are found in lymphoid tissues in patients with active SLE (88). An inducer of IFN-α in the sera was shown to comprise small immune complexes containing DNA and IgG anti-DNA autoantibodies as essential components (89, 90). Apoptotic cells were shown to provide the antigen for these complexes, and expression of FcgRII and toll-like receptor 9 (TLR9) on DCs were required for recognition of these complexes and IFN-α production (91). Importantly, DNA-IgG immune complexes (ICs) have recently also been implicated in the aberrant activation of autoreactive B cells, mediated through synergistic B cell receptor/TLR9 signaling (92). How mammalian DNA, usually poorly immunogenic, can form activatory ICs is still poorly understood; however, it is becoming evident that even methylated CpG motifs, present within mammalian DNA, do have the potential to induce B cell activation (93, 94). It is also interesting to note that in addition to anti-DNA ICs, other events that can trigger secretion of excessive amounts of IFN-α, such as infection and UV skin injury, are both known activators of lupus. In conclusion, there is strong circumstantial evidence suggesting that chronic stimulation of DCs can lead to a break in peripheral tolerance and the development of SLE. In the absence of complement, elevated levels of apoptotic cell debris lead to uptake and inappropriate presentation of these autoantigens to T cells, leading to the development of autoimmunity. What still remains to be confirmed is the role of complement in regulating the maturation of DCs. Is the presence of increased circulating autoantigens sufficient, or does complement regulate additional mechanisms that lead to the break of tolerance toward self-antigens? What stimulates the production of IFN-α?

Clearance of Immune Complexes The formation of ICs is an important process of the adaptive immune response, promoting the removal of foreign antigens. Complement plays two main roles in the process: maintaining solubility of the complexes and, via complement receptors, promoting their recognition and clearance by phagocytic cells in conjunction with Fcγ receptors. IC precipitation is prevented primarily by the binding of the C1 complex and the subsequent covalent deposition of C4 and C3 binding to both antigen and the heavy chain of IgG (95, 96). The deposition of the complement components within the IC is thought to prevent the formation of large insoluble ICs by reducing the effective

11 Feb 2004

19:19

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

444

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

valency of antigen and antibody, as well as interfering with Fc-Fc interactions (97, 98). Complement is also able to induce the solubilization of insoluble ICs, a process mediated by the alternative complement pathway. However, this process is relatively inefficient, requiring consumption of large amounts of complement (99). In addition to maintaining the solubility and minimizing the size of IC, complement fragments bound within the ICs provide ligands for the clearance of these complexes by phagocytic cells. In primates, CR1 on erythrocytes is predominantly responsible for the transporting of IC to tissue macrophage present in the liver and spleen (100). Although CR1 is not expressed by rodent erythrocytes, soluble IC are also predominantly removed through the liver and spleen (101). There are numerous studies demonstrating that IC processing is abnormal in SLE patients (102, 103). The in vivo studies have been performed using both particulate and soluble ICs with similar observations made in both models (104, 105). In patients with SLE or complement deficiency, the initial clearance of immune complexes was impaired in the spleen; however, the uptake by the liver was more rapid. Most strikingly, the ICs were poorly retained within the liver with a proportion released back to the circulation. These ICs have the potential to deposit in tissues throughout the body, inducing inflammation. Owing to the similarities in the observations in complement-deficient and SLE patients with acquired reduction in complement levels, the cause of this phenomenon is presumably due to reduced complement binding to ICs, as well as the decreased levels of erythrocyte CR1 that have been characterized in SLE patients (106). In addition, an accelerated hepatic and decreased splenic uptake of IC was demonstrated in C1q-deficient mice, analogous to the observations in humans (101). Collectively, these observations led to the hypothesis that in the absence of complement, or in the presence of acquired hypocomplementaemia, the impaired clearance and decreased solubility of IC could lead to their deposition within tissues, inducing inflammatory injury that may cause the release of autoantigens, augmenting an autoimmune response. This theory is complementary to the hypothesis proposed in the previous section, suggesting that defective clearance of dying cells in the absence of complement leads to the development of SLE.

Complement and Lymphocyte Responses PHYSIOLOGICAL REGULATION OF B CELL RESPONSES BY COMPLEMENT B cells are responsible for the production of pathogenic autoantibodies characteristic of SLE. Therefore, dissection of the mechanisms by which complement regulates these cells is crucial for understanding disease etiology. B cells of the B-1 subtype are known to produce “natural” IgM with specificities for autoantigens, even in healthy individuals, with higher titers observed in the sera of SLE patients. In patients with SLE, B cells of the B-2 subtype produce a large array of pathogenic IgG autoantibodies. The autoreactive B-2 B cells have been shown to undergo antigendriven hypermutation, isotype switching, and selection to produce antibodies with

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

COMPLEMENT AND SLE

P1: IKH

445

a very high affinity for specific autoantigens. These observations suggest that autoantibody generation is a T cell–dependent process. There is evidence linking complement to regulation of B cell development, activation, and tolerance. In the absence of C1q, C4, or C3, humoral immune responses to T-dependent antigens are severely impaired with poor IgM responses and an inability of the B cells to switch antibody isotypes (107). Two mechanisms were proposed to explain enhanced humoral responses in the presence of complement. Initially, it was proposed that retention of ICs on follicular dendritic cells (FDCs) was essential for the development of normal T-dependent antibody responses. In mice depleted of C3 or inherently deficient in complement components (C3, C4, and C1q) or CR1/2, loss of antigen retention (as IC) on FDCs was noted (108–112). More recently, however, extensive biochemical and in vivo studies (113, 114), using anti-CR1/2 blocking antibodies (115–118) and CR1/2 knockout mice (119, 120), have revealed that an important complement effect is mediated directly on B cells via the complement receptors CD21/CD35. Coligation of the B cell receptor with CD21 (CR2) on the surface of B cells provides the necessary costimulation for enhanced B cell activation to T-dependent antigens (113). Direct evidence for this was provided by Dempsey and colleagues, who found that coupling multiple oligomers of C3d to an antigen, hen egg lysozyme (HEL), lowered the amount of the antigen required for the induction of HEL-specific IgG responses (121). How this coupling of C3d to antigen occurs in vivo is poorly understood, although it is believed to involve IC formation, possibly involving natural IgM (122). Evidence for the involvement of natural IgM comes from findings that deficiency in secreted IgM leads to impaired B cell responses to low doses of T-dependent antigens that are similar to those seen in the absence of complement (123). Importantly, although expression of CR1/2 on B cells is sufficient for the development of short-term antibody responses, the retention of ICs bound to CR1/2 on FDCs is important for the long-term maintenance of the memory B cell response (124). Do these findings help to decipher the mechanisms responsible for linking complement deficiency to the development of SLE? On the contrary, they show that one of the normal activities of complement is to lower the threshold for B cell activation by T-dependent antigens. Therefore, as the autoantibodies commonly found in SLE are thought to form via T-dependent B cell responses, individuals devoid of complement should be protected from developing disease, rather than deficiency promoting the development of SLE. This apparent paradox can be explained by the fact that the necessity for complement and complement receptors in humoral immune responses can be overcome by the presence of large doses of antigen (110, 125, 126). In the context of SLE it has been demonstrated, both in patients (41, 42) and animal models (44, 127), that high levels of circulating autoantigen are present, as previously discussed. Therefore, the effect of complement deficiency on B cell activation, through providing costimulation via the complement receptors on the B cell, is probably minimal or even negated by other effects of complement, such as the increase of antigen available to the B cells.

11 Feb 2004

19:19

446

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

However, there is some evidence demonstrating that an absence of complement can cause a breakdown in B cell tolerance, resulting in the escape of autoreactive B cells and predisposition to autoimmunity in the absence of complement. The reports published to date testing this hypothesis are somewhat contradictory and are reviewed below (31, 32, 128). COMPLEMENT DEFICIENCY, B CELLS, AND TOLERANCE An alternative hypothesis to the waste disposal hypothesis to explain the association between complement deficiencies and SLE is that a breakdown in B cell tolerance results in the escape from deletion/anergy of self-reactive B cells that are subsequently activated on encountering autoantigen within the periphery (32, 125, 129–131). A transgenic model comprising soluble HEL (sHEL) as autoantigen and an antiHEL as the autoantibody has been extensively studied in the presence and absence of complement proteins. Prodeus and colleagues found that self-tolerance to sHEL was dependent on the presence of C4 and CR1/2, but not C3. They hypothesized that the attachment of C4b to self-antigens and localization of these complexes to CD35 on stromal cells within the bone marrow may regulate the selection of potentially autoreactive B cells (32). However, in complete contrast to these results, the interbreeding of C1q-deficient mice with the same transgenic model (sHEL/antiHEL) demonstrated no alterations in B cell tolerance to sHEL expressed as a self-antigen in the absence of C1q (128). The different genetic background of the mice used could account for the opposing observations made in these two studies. As disease in SLE is known to be influenced by multiple genetic factors (28, 132), when employing mice of mixed genetic background, as in the study of C4- and CR1/2-deficient mice, a number of different genetic loci could affect the break in tolerance observed. In addition, the transgenic expression of sHEL is not the best model for studying the maintenance of tolerance in SLE. Soluble secreted HEL is not a natural autoantigen, and the majority of the autoantigens targeted by autoantibodies in SLE are cell-associated (1). Experimental studies are currently underway exploring the role of complement deficiency in tolerance induction in experimental animals expressing transgenic lupus autoantibodies and transgenic cell-associated autoantigens. These mice will provide a better tool for investigating the extent to which complement contributes to the regulation of autoreactive B cells. COMPLEMENT AND T LYMPHOCYTE ACTIVATION Extensive experimental evidence clearly underlines the critical role of CD4+ T cells in the pathogenesis of SLE (133–137). It is thought that the effector T cells induce their pathogenic effects in two ways, indirectly, through the provision of T cell help to autoreactive B cells (138–140), resulting in the production of high-affinity autoantibodies, and directly, inducing tissue damage (141). Until recently, any effect of complement in mediating activation of autoreactive T cells was thought to be largely indirect, with altered APC function resulting in inappropriate signaling to helper T cells. However, there is now growing evidence that complement may directly influence T cell activation/regulation mechanisms.

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

COMPLEMENT AND SLE

P1: IKH

447

Subpopulations of human T cells have been shown to express receptors for C1q (142–144), and proliferation of the T cells was significantly inhibited when soluble C1q was present in mitogen-induced T cell cultures (144) and in a number of T cell lines (142). The physiological significance of these observations is not known. Some insight to these earlier observations may be provided by a recent report showing that C1q-containing ICs, but not nonopsonised ICs, can activate human T cells, inducing the secretion of both TNF-α and IFN-γ (145). Crosslinking of C1q on the surface of the T cells had the same effect. To date no direct T cell abnormalities have been observed in the complementdeficient animals. However, in C1q-deficient mice, cytokine production by antigenspecific T cells was altered in response to immunization with a T-dependent antigen, with diminished levels of IFN-γ (112). The mechanism responsible is not known. The primary function of complement regulatory proteins is to prevent complement deposition on the surface on which they reside. However, new data suggest an additional costimulatory role for these proteins in T cell activation. In humans, ligation of CD3 with membrane-bound complement regulatory proteins, either CD46 [also known as membrane cofactor protein (MCP)] or CD55 [decay accelerating factor (DAF)], induced a synergistic activation of human T cells, promoting T cell proliferation and altered cytokine production (146–150). Similarly, in mice the coligation of CD3 and Crry, which possesses both MCP and DAF activity, led to increased CD4 T cell proliferation and decreased IFN-γ production (151). In addition, the coligation of CD3 and CD46 on human CD4+ T cells induced development of a T-regulatory phenotype, when cultured in the presence of IL-2 (148). What could be the physiological significance of these observations, and do they have any relevance to the pathogenesis of SLE associated with complement deficiencies? Complement fragments bound within immune complexes could ligate a number of receptors expressed on the membrane of T cells and alter their activation threshold. In addition, as regulatory T cells have been shown to prevent and/or regulate systemic autoimmunity and have been directly implicated in a number of autoimmune diseases (152, 153), a link to SLE is an attractive hypothesis.

SUMMARY The evidence is overwhelming from both the clinical observations and animal models that the classical complement pathway proteins play a protective role against the development of SLE. A flow diagram linking the current hypotheses that may account for the development of autoimmunity in the absence of complement is presented in Figure 1. First, the absence of complement-dependent clearance of dying cells and/or immune complexes could lead to elevated levels of potentially immunogenic self-antigens. The uptake and inappropriate presentation of these antigens could drive the production of autoantibodies. In addition, the formation of DNA/anti-DNA immune complexes could directly induce tissue inflammation, as well as mediating the inappropriate maturation of dendritic cells with

11 Feb 2004

19:19

448

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

potentially pathogenic consequences. Second, the absence of regulatory signals to B and T lymphocytes, usually provided by activated complement fragments, could lead to a breakdown in the maintenance of peripheral tolerance to self-antigens. Thus, the presence of elevated levels of autoantigens and inappropriate presentation of these antigens by dendritic cells may explain the loss of self-tolerance associated with complement deficiency.

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Pickering MC, Botto M, Taylor PR, Lachmann PJ, Walport MJ. 2000. Systemic lupus erythematosus, complement deficiency, and apoptosis. Adv. Immunol. 76: 227–324 2. Stone NM, Williams A, Wilkinson JD, Bird G. 2000. Systemic lupus erythematosus with C1q deficiency. Br. J. Dermatol. 142:521–24 3. Dragon-Durey MA, Quartier P, Fremeaux-Bacchi V, Blouin J, de Barace C, et al. 2001. Molecular basis of a selective C1s deficiency associated with early onset multiple autoimmune diseases. J. Immunol. 166:7612–16 4. Rupert KL, Moulds JM, Yang Y, Arnett FC, Warren RW, et al. 2002. The molecular basis of complete complement C4A and C4B deficiencies in a systemic lupus erythematosus patient with homozygous C4A and C4B mutant genes. J. Immunol. 169:1570–78 5. Agnello V. 1978. Association of systemic lupus erythematosus and SLE-like syndromes with hereditary and acquired complement deficiency states. Arthritis Rheum. 21:S146–52 6. Imai K, Nakajima K, Eguchi K, Miyazaki M, Endoh M, et al. 1991. Homozygous C3 deficiency associated with IgA nephropathy. Nephron 59:148–52 7. Nilsson UR, Nilsson B, Storm KE, SjolinForsberg G, Hallgren R. 1992. Hereditary dysfunction of the third component of complement associated with a systemic lupus erythematosus-like syndrome

8.

9.

10.

11.

12.

13.

14.

and meningococcal meningitis. Arthritis Rheum. 35:580–86 Sano Y, Nishimukai H, Kitamura H, Nagaki K, Inai S, et al. 1981. Hereditary deficiency of the third component of complement in two sisters with systemic lupus erythematosus-like symptoms. Arthritis Rheum. 24:1255–60 Inai S, Akagaki Y, Moriyama T, Fukumori Y, Yoshimura K, et al. 1989. Inherited deficiencies of the late-acting complement components other than C9 found among healthy blood donors. Int. Arch. Allergy Appl. Immunol. 90:274–79 Hassig A, Borel JF, Amman P, Thon M, Butler R. 1964. Essentielle hypokomplementamie. Pathol. Microbiol. 27:542– 49 Berkel AI, Birben E, Oner C, Oner R, Loos M, Petry F. 2000. Molecular, genetic and epidemiologic studies on selective complete C1q deficiency in Turkey. Immunobiology 201:347–55 Deapen D, Escalante A, Weinrib L, Horwitz D, Bachman B, et al. 1992. A revised estimate of twin concordance in systemic lupus erythematosus. Arthritis Rheum. 35:311–18 Racila DM, Sontheimer CJ, Sheffield A, Wisnieski JJ, Racila E, Sontheimer RD. 2003. Homozygous single nucleotide polymorphism of the complement C1QA gene is associated with decreased levels of C1q in patients with subacute cutaneous lupus erythematosus. Lupus 12:124–32 Traustadottir KH, Sigfusson A, Steinsson

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

COMPLEMENT AND SLE

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

15.

16.

17.

18.

19.

20.

21.

22.

K, Erlendsson K. 2002. C4A deficiency and elevated level of immune complexes: the mechanism behind increased susceptibility to systemic lupus erythematosus. J. Rheumatol. 29:2359–66 Schifferli JA, Steiger G, Paccaud JP, Sjoholm AG, Hauptmann G. 1986. Difference in the biological properties of the two forms of the fourth component of human complement (C4). Clin. Exp. Immunol. 63:473–77 Glass D, Raum D, Gibson D, Stillman JS, Schur PH. 1976. Inherited deficiency of the second component of complement. Rheumatic disease associations. J. Clin. Invest. 58:853–61 Lipscombe RJ, Sumiya M, Summerfield JA, Turner MW. 1995. Distinct physicochemical characteristics of human mannose binding protein expressed by individuals of differing genotype. Immunology 85:660–67 Sullivan KE, Wooten C, Goldman D, Petri M. 1996. Mannose-binding protein genetic polymorphisms in black patients with systemic lupus erythematosus. Arthritis Rheum. 39:2046–51 Ip WK, Chan SY, Lau CS, Lau YL. 1998. Association of systemic lupus erythematosus with promoter polymorphisms of the mannose-binding lectin gene. Arthritis Rheum. 41:1663–68 Davies EJ, Teh LS, Ordi-Ros J, Snowden N, Hillarby MC, et al. 1997. A dysfunctional allele of the mannose binding protein gene associates with systemic lupus erythematosus in a Spanish population. J. Rheumatol. 24:485–88 Davies EJ, Snowden N, Hillarby MC, Carthy D, Grennan DM, et al. 1995. Mannose-binding protein gene polymorphism in systemic lupus erythematosus. Arthritis Rheum. 38:110–14 Koide M, Shirahama S, Tokura Y, Takigawa M, Hayakawa M, Furukawa F. 2002. Lupus erythematosus associated with C1 inhibitor deficiency. J. Dermatol. 29:503– 7

P1: IKH

449

23. Ohi H, Yasugi T. 1994. Occurrence of C3 nephritic factor and C4 nephritic factor in membranoproliferative glomerulonephritis (MPGN). Clin. Exp. Immunol. 95:316– 21 24. Walport MJ, Davies KA, Botto M, Naughton MA, Isenberg DA, et al. 1994. C3 nephritic factor and SLE: report of four cases and review of the literature. Q. J. Med. 87:609–15 25. Cameron JS, Lessof MH, Ogg CS, Williams BD, Williams DG. 1976. Disease activity in the nephritis of systemic lupus erythematosus in relation to serum complement concentrations. DNA-binding capacity and precipitating anti-DNA antibody. Clin. Exp. Immunol. 25:418–27 26. Valentijn RM, van Overhagen H, Hazevoet HM, Hermans J, Cats A, et al. 1985. The value of complement and immune complex determinations in monitoring disease activity in patients with systemic lupus erythematosus. Arthritis Rheum. 28:904–13 27. Chen Z, Koralov SB, Kelsoe G. 2000. Complement C4 inhibits systemic autoimmunity through a mechanism independent of complement receptors CR1 and CR2. J. Exp. Med. 192:1339–52 28. Mitchell DA, Pickering MC, Warren J, Fossati-Jimack L, Cortes-Hernandez J, et al. 2002. C1q deficiency and autoimmunity: the effects of genetic background on disease expression. J. Immunol. 168:2538–43 29. Paul E, Pozdnyakova OO, Mitchell E, Carroll MC. 2002. Anti-DNA autoreactivity in C4-deficient mice. Eur. J. Immunol. 32:2672–79 30. Botto M, Dell’Agnola C, Bygrave AE, Thompson EM, Cook HT, et al. 1998. Homozygous C1q deficiency causes glomerulonephritis associated with multiple apoptotic bodies. Nat. Genet. 19:56– 59 31. Einav S, Pozdnyakova OO, Ma M, Carroll MC. 2002. Complement C4 is protective

11 Feb 2004

19:19

450

32.

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

33.

34.

35.

36.

37.

38.

39.

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

for lupus disease independent of C3. J. Immunol. 168:1036–41 Prodeus AP, Goerg S, Shen LM, Pozdnyakova OO, Chu L, et al. 1998. A critical role for complement in maintenance of self-tolerance. Immunity 9:721– 31 Sekine H, Reilly CM, Molano ID, Garnier G, Circolo A, et al. 2001. Complement component C3 is not required for full expression of immune complex glomerulonephritis in MRL/lpr mice. J. Immunol. 166:6444–51 Wu X, Jiang N, Deppong C, Singh J, Dolecki G, et al. 2002. A role for the Cr2 gene in modifying autoantibody production in systemic lupus erythematosus. J. Immunol. 169:1587–92 Rosen A, Casciola-Rosen L, Ahearn J. 1995. Novel packages of viral and selfantigens are generated during apoptosis. J. Exp. Med. 181:1557–61 Casciola-Rosen LA, Anhalt G, Rosen A. 1994. Autoantigens targeted in systemic lupus erythematosus are clustered in two populations of surface structures on apoptotic keratinocytes. J. Exp. Med. 179:1317–30 Herrmann M, Voll RE, Zoller OM, Hagenhofer M, Ponner BB, Kalden JR. 1998. Impaired phagocytosis of apoptotic cell material by monocyte-derived macrophages from patients with systemic lupus erythematosus. Arthritis Rheum. 41:1241–50 Grootscholten C, van Bruggen MC, Van Der Pijl JW, De Jong EM, Ligtenberg G, et al. 2003. Deposition of nucleosomal antigens (histones and DNA) in the epidermal basement membrane in human lupus nephritis. Arthritis Rheum. 48:1355– 62 van Bruggen MC, Kramers C, Walgreen B, Elema JD, Kallenberg CG, et al. 1997. Nucleosomes and histones are present in glomerular deposits in human lupus nephritis. Nephrol. Dial Transplant. 12:57–66

40. Baumann I, Kolowos W, Voll RE, Manger B, Gaipl U, et al. 2002. Impaired uptake of apoptotic cells into tingible body macrophages in germinal centers of patients with systemic lupus erythematosus. Arthritis Rheum. 46:191–201 41. Amoura Z, Piette JC, Chabre H, Cacoub P, Papo T, et al. 1997. Circulating plasma levels of nucleosomes in patients with systemic lupus erythematosus: correlation with serum antinucleosome antibody titers and absence of clear association with disease activity. Arthritis Rheum. 40:2217–25 42. Raptis L, Menard HA. 1980. Quantitation and characterization of plasma DNA in normals and patients with systemic lupus erythematosus. J. Clin. Invest. 66:1391– 99 43. Lambert PH, Dixon FJ. 1968. Pathogenesis of the glomerulonephritis of NZB/W mice. J. Exp. Med. 127:507–22 44. Licht R, van Bruggen MC, OppersWalgreen B, Rijke TP, Berden JH. 2001. Plasma levels of nucleosomes and nucleosome-autoantibody complexes in murine lupus: effects of disease progression and lipopolyssacharide administration. Arthritis Rheum. 44:1320–30 45. Mevorach D, Zhou JL, Song X, Elkon KB. 1998. Systemic exposure to irradiated apoptotic cells induces autoantibody production. J. Exp. Med. 188:387–92 46. Bickerstaff MC, Botto M, Hutchinson WL, Herbert J, Tennent GA, et al. 1999. Serum amyloid P component controls chromatin degradation and prevents antinuclear autoimmunity. Nat. Med. 5:694– 97 47. Napirei M, Karsunky H, Zevnik B, Stephan H, Mannherz HG, Moroy T. 2000. Features of systemic lupus erythematosus in Dnase1-deficient mice. Nat. Genet. 25:177–81 48. Walport MJ. 2000. Lupus, DNase and defective disposal of cellular debris. Nat. Genet. 25:135–36 49. Cohen PL, Caricchio R, Abraham V,

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

COMPLEMENT AND SLE

50.

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

51.

52.

53.

54.

55.

56.

57.

58.

Camenisch TD, Jennette JC, et al. 2002. Delayed apoptotic cell clearance and lupus-like autoimmunity in mice lacking the c-mer membrane tyrosine kinase. J. Exp. Med. 196:135–40 Scott RS, McMahon EJ, Pop SM, Reap EA, Caricchio R, et al. 2001. Phagocytosis and clearance of apoptotic cells is mediated by MER. Nature 411:207–11 Korb LC, Ahearn JM. 1997. C1q binds directly and specifically to surface blebs of apoptotic human keratinocytes: complement deficiency and systemic lupus erythematosus revisited. J. Immunol. 158:4525–28 Nauta AJ, Trouw LA, Daha MR, Tijsma O, Nieuwland R, et al. 2002. Direct binding of C1q to apoptotic cells and cell blebs induces complement activation. Eur. J. Immunol. 32:1726–36 Navratil JS, Watkins SC, Wisnieski JJ, Ahearn JM. 2001. The globular heads of C1q specifically recognize surface blebs of apoptotic vascular endothelial cells. J. Immunol. 166:3231–39 Mevorach D, Mascarenhas JO, Gershov D, Elkon KB. 1998. Complementdependent clearance of apoptotic cells by human macrophages. J. Exp. Med. 188:2313–20 Taylor PR, Carugati A, Fadok VA, Cook HT, Andrews M, et al. 2000. A hierarchical role for classical pathway complement proteins in the clearance of apoptotic cells in vivo. J. Exp. Med. 192:359–66 Savill J, Dransfield I, Gregory C, Haslett C. 2002. A blast from the past: Clearance of apoptotic cells regulates immune responses. Nat. Rev. Immunol. 2:965–75 Ogden CA, deCathelineau A, Hoffmann PR, Bratton D, Ghebrehiwet B, et al. 2001. C1q and mannose binding lectin engagement of cell surface calreticulin and CD91 initiates macropinocytosis and uptake of apoptotic cells. J. Exp. Med. 194:781– 95 Vandivier RW, Ogden CA, Fadok VA, Hoffmann PR, Brown KK, et al. 2002.

59.

60.

61.

62.

63.

64.

65.

66.

P1: IKH

451

Role of surfactant proteins A, D, and C1q in the clearance of apoptotic cells in vivo and in vitro: calreticulin and CD91 as a common collectin receptor complex. J. Immunol. 169:3978–86 Nauta AJ, Daha MR, Kooten C, Roos A. 2003. Recognition and clearance of apoptotic cells: a role for complement and pentraxins. Trends Immunol. 24:148–54 Gershov D, Kim S, Brot N, Elkon KB. 2000. C-reactive protein binds to apoptotic cells, protects the cells from assembly of the terminal complement components, and sustains an antiinflammatory innate immune response: implications for systemic autoimunity. J. Exp. Med. 192:1353–64 Kim SJ, Gershov D, Ma X, Brot N, Elkon KB. 2002. I-PLA(2) activation during apoptosis promotes the exposure of membrane lysophosphatidylcholine leading to binding by natural immunoglobulin M antibodies and complement activation. J. Exp. Med. 196:655–65 Ehrenstein MR, Cook HT, Neuberger MS. 2000. Deficiency in serum immunoglobulin (Ig)M predisposes to development of IgG autoantibodies. J. Exp. Med. 191: 1253–58 Boes M, Schmidt T, Linkemann K, Beaudette BC, Marshak-Rothstein A, Chen J. 2000. Accelerated development of IgG autoantibodies and autoimmune disease in the absence of secreted IgM. Proc. Natl. Acad. Sci. USA 97:1184–89 Gaipl US, Kuenkele S, Voll RE, Beyer TD, Kolowos W, et al. 2001. Complement binding is an early feature of necrotic and a rather late event during apoptotic cell death. Cell Death Differ. 8:327–34 van Nieuwenhuijze AE, van Lopik T, Smeenk RJ, Aarden LA. 2003. Time between onset of apoptosis and release of nucleosomes from apoptotic cells: putative implications for systemic lupus erythematosus. Ann. Rheum. Dis. 62:10– 14 Rovere P, Vallinoto C, Bondanza A, Crosti

11 Feb 2004

19:19

452

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

67.

68.

69.

70.

71.

72.

73.

74.

75.

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

MC, Rescigno M, et al. 1998. Bystander apoptosis triggers dendritic cell maturation and antigen-presenting function. J. Immunol. 161:4467–71 Sauter B, Albert ML, Francisco L, Larsson M, Somersan S, Bhardwaj N. 2000. Consequences of cell death: Exposure to necrotic tumor cells, but not primary tissue cells or apoptotic cells, induces the maturation of immunostimulatory dendritic cells. J. Exp. Med. 191:423–34 Salio M, Cerundolo V, Lanzavecchia A. 2000. Dendritic cell maturation is induced by mycoplasma infection but not by necrotic cells. Eur. J. Immunol. 30:705–8 Huynh ML, Fadok VA, Henson PM. 2002. Phosphatidylserine-dependent ingestion of apoptotic cells promotes TGFbeta1 secretion and the resolution of inflammation. J. Clin. Invest. 109:41–50 Fadok VA, Bratton DL, Konowal A, Freed PW, Westcott JY, Henson PM. 1998. Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-beta, PGE2, and PAF. J. Clin. Invest. 101:890–98 McDonald PP, Fadok VA, Bratton D, Henson PM. 1999. Transcriptional and translational regulation of inflammatory mediator production by endogenous TGF-beta in macrophages that have ingested apoptotic cells. J. Immunol. 163:6164–72 Voll RE, Herrmann M, Roth EA, Stach C, Kalden JR, Girkontaite I. 1997. Immunosuppressive effects of apoptotic cells. Nature 390:350–51 Gao Y, Herndon JM, Zhang H, Griffith TS, Ferguson TA. 1998. Antiinflammatory effects of CD95 ligand (FasL)-induced apoptosis. J. Exp. Med. 188:887–96 Chen W, Frank ME, Jin W, Wahl SM. 2001. TGF-beta released by apoptotic T cells contributes to an immunosuppressive milieu. Immunity 14:715–25 Sato K, Yamashita N, Matsuyama T.

76.

77.

78.

79.

80.

81.

82.

83.

2002. Human peripheral blood monocytederived interleukin-10-induced semimature dendritic cells induce anergic CD4(+) and CD8(+) T cells via presentation of the internalized soluble antigen and cross-presentation of the phagocytosed necrotic cellular fragments. Cell Immunol. 215:186–94 Chen Z, Moyana T, Saxena A, Warrington R, Jia Z, Xiang J. 2001. Efficient antitumor immunity derived from maturation of dendritic cells that had phagocytosed apoptotic/necrotic tumor cells. Int. J. Cancer 93:539–48 Basu S, Binder RJ, Suto R, Anderson KM, Srivastava PK. 2000. Necrotic but not apoptotic cell death releases heat shock proteins, which deliver a partial maturation signal to dendritic cells and activate the NF-kappa B pathway. Int. Immunol. 12:1539–46 Ronnblom L, Alm GV. 2002. The natural interferon-alpha producing cells in systemic lupus erythematosus. Hum. Immunol. 63:1181–93 Hardin JA. 2003. Directing autoimmunity to nucleoprotein particles: the impact of dendritic cells and interferon alpha in lupus. J. Exp. Med. 197:681–85 Biron CA. 2001. Interferons alpha and beta as immune regulators–a new look. Immunity 14:661–64 Blanco P, Palucka AK, Gill M, Pascual V, Banchereau J. 2001. Induction of dendritic cell differentiation by IFN-alpha in systemic lupus erythematosus. Science 294:1540–43 Baechler EC, Batliwalla FM, Karypis G, Gaffney PM, Ortmann WA, et al. 2003. Interferon-inducible gene expression signature in peripheral blood cells of patients with severe lupus. Proc. Natl. Acad. Sci. USA. 100:2610–15 Bennett L, Palucka AK, Arce E, Cantrell V, Borvak J, et al. 2003. Interferon and granulopoiesis signatures in systemic lupus erythematosus blood. J. Exp. Med. 197:711–23

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

COMPLEMENT AND SLE 84. Hooks JJ, Moutsopoulos HM, Geis SA, Stahl NI, Decker JL, Notkins AL. 1979. Immune interferon in the circulation of patients with autoimmune disease. N. Engl. J. Med. 301:5–8 85. Steinberg AD, Baron S, Talal N. 1969. The pathogenesis of autoimmunity in New Zealand mice, I. Induction of antinucleic acid antibodies by polyinosinicpolycytidylic acid. Proc. Natl. Acad. Sci. USA. 63:1102–7 86. Santiago-Raber ML, Baccala R, Haraldsson KM, Choubey D, Stewart TA, et al. 2003. Type-I interferon receptor deficiency reduces lupus-like disease in NZB mice. J. Exp. Med. 197:777–88 87. Cederblad B, Blomberg S, Vallin H, Perers A, Alm GV, Ronnblom L. 1998. Patients with systemic lupus erythematosus have reduced numbers of circulating natural interferon-alpha-producing cells. J. Autoimmun. 11:465–70 88. Cella M, Jarrossay D, Facchetti F, Alebardi O, Nakajima H, et al. 1999. Plasmacytoid monocytes migrate to inflamed lymph nodes and produce large amounts of type I interferon. Nat. Med. 5:919– 23 89. Vallin H, Perers A, Alm GV, Ronnblom L. 1999. Anti-double-stranded DNA antibodies and immunostimulatory plasmid DNA in combination mimic the endogenous IFN-alpha inducer in systemic lupus erythematosus. J. Immunol. 163:6306– 13 90. Vallin H, Blomberg S, Alm GV, Cederblad B, Ronnblom L. 1999. Patients with systemic lupus erythematosus (SLE) have a circulating inducer of interferon-alpha (IFN-alpha) production acting on leucocytes resembling immature dendritic cells. Clin. Exp. Immunol. 115:196–202 91. Bave U, Alm GV, Ronnblom L. 2000. The combination of apoptotic U937 cells and lupus IgG is a potent IFN-alpha inducer. J. Immunol. 165:3519–26 92. Leadbetter EA, Rifkin IR, Hohlbaum AM, Beaudette BC, Shlomchik MJ, Marshak-

93.

94.

95.

96.

97.

98.

99.

100.

101.

P1: IKH

453

Rothstein A. 2002. Chromatin-IgG complexes activate B cells by dual engagement of IgM and Toll-like receptors. Nature 416:603–7 Goeckeritz BE, Flora M, Witherspoon K, Vos Q, Lees A, et al. 1999. Multivalent cross-linking of membrane Ig sensitizes murine B cells to a broader spectrum of CpG-containing oligodeoxynucleotide motifs, including their methylated counterparts, for stimulation of proliferation and Ig secretion. Int. Immunol. 11:1693– 700 Rui LX, Vinuesa CG, Blasioli J, Goodnow CC. 2003. Resistance to CpG DNA-induced autoimmunity through tolerogenic B cell antigen receptor ERK signaling. Nat. Immunol. 4:594–600 Schifferli JA, Steiger G, Schapira M. 1985. The role of C1, C1-inactivator and C4 in modulating immune precipitation. Clin. Exp. Immunol. 60:605–12 Schifferli JA, Steiger G, Hauptmann G, Spaeth PJ, Sjoholm AG. 1985. Formation of soluble immune complexes by complement in sera of patients with various hypocomplementemic states. Difference between inhibition of immune precipitation and solubilization. J. Clin. Invest. 76:2127–33 Hong K, Takata Y, Sayama K, Kozono H, Takeda J, et al. 1984. Inhibition of immune precipitation by complement. J. Immunol. 133:1464–70 Lachmann PJ, Walport MJ. 1987. Deficiency of the effector mechanisms of the immune response and autoimmunity. Ciba Found Symp. 129:149–71 Miller GW, Nussenzweig V. 1975. A new complement function: solubilization of antigen-antibody aggregates. Proc. Natl. Acad. Sci. USA 72:418–22 Walport MJ, Lachmann PJ. 1988. Erythrocyte complement receptor type 1, immune complexes, and the rheumatic diseases. Arthritis Rheum. 31:153–58 Nash JT, Taylor PR, Botto M, Norsworthy PJ, Davies KA, Walport MJ.

11 Feb 2004

19:19

454

102.

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

103.

104.

105.

106.

107.

108.

109.

110.

111.

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

2001. Immune complex processing in C1q-deficient mice. Clin. Exp. Immunol. 123:196–202 Hebert LA, Cosio G. 1987. The erythrocyte-immune complex-glomerulonephritis connection in man. Kidney Int. 31:877–85 Walport MJ, Davies KA. 1996. Complement and immune complexes. Res. Immunol. 147:103–9 Schifferli JA, Ng YC, Estreicher J, Walport MJ. 1988. The clearance of tetanus toxoid/anti-tetanus toxoid immune complexes from the circulation of humans. Complement- and erythrocyte complement receptor 1-dependent mechanisms. J. Immunol. 140:899–904 Davies KA, Erlendsson K, Beynon HL, Peters AM, Steinsson K, et al. 1993. Splenic uptake of immune complexes in man is complement-dependent. J. Immunol. 151:3866–73 Davies KA, Peters AM, Beynon HL, Walport MJ. 1992. Immune complex processing in patients with systemic lupus erythematosus. In vivo imaging and clearance studies. J. Clin. Invest. 90:2075–83 Heyman B. 2000. Regulation of antibody responses via antibodies, complement, and Fc receptors. Annu. Rev. Immunol. 18:709–37 Pepys MB. 1972. Role of complement in induction of the allergic response. Nat. New Biol. 237:157–59 Feldmann M, Pepys MB. 1974. Role of C3 in in vitro lymphocyte cooperation. Nature 249:159–61 Fischer MB, Ma M, Goerg S, Zhou X, Xia J, et al. 1996. Regulation of the B cell response to T-dependent antigens by classical pathway complement. J. Immunol. 157:549–56 Fang YF, Xu CG, Fu YX, Holers VM, Molina H. 1998. Expression of complement receptors 1 and 2 on follicular dendritic cells is necessary for the generation of a strong antigen-specific IgG response. J. Immunol. 160:5273–79

112. Cutler AJ, Botto M, van Essen D, Rivi R, Davies KA, et al. 1998. T cell-dependent immune response in C1q-deficient mice: defective interferon gamma production by antigen-specific T cells. J. Exp. Med. 187:1789–97 113. Fearon DT, Carroll MC. 2000. Regulation of B lymphocyte responses to foreign and self-antigens by the CD19/CD21 complex. Annu. Rev. Immunol. 18:393– 422 114. Fearon DT, Carter RH. 1995. The CD19/CR2/TAPA-1 complex of B lymphocytes: linking natural to acquired immunity. Annu. Rev. Immunol. 13:127–49 115. Thyphronitis G, Kinoshita T, Inoue K, Schweinle JE, Tsokos GC, et al. 1991. Modulation of mouse complement receptors 1 and 2 suppresses antibody responses in vivo. J. Immunol. 147:224–30 116. Gustavsson S, Kinoshita T, Heyman B. 1995. Antibodies to murine complement receptor 1 and 2 can inhibit the antibody response in vivo without inhibiting T helper cell induction. J. Immunol. 154:6524–28 117. Heyman B, Wiersma EJ, Kinoshita T. 1990. In vivo inhibition of the antibody response by a complement receptorspecific monoclonal antibody. J. Exp. Med. 172:665–68 118. Hebell T, Ahearn JM, Fearon DT. 1991. Suppression of the immune response by a soluble complement receptor of B lymphocytes. Science 254:102–5 119. Ahearn JM, Fischer MB, Croix D, Goerg S, Ma M, et al. 1996. Disruption of the Cr2 locus results in a reduction in B-1a cells and in an impaired B cell response to T-dependent antigen. Immunity 4:251– 62 120. Molina H, Holers VM, Li B, Fung Y, Mariathasan S, et al. 1996. Markedly impaired humoral immune response in mice deficient in complement receptors 1 and 2. Proc. Natl. Acad. Sci. USA 93:3357–61 121. Dempsey PW, Allison ME, Akkaraju S, Goodnow CC, Fearon DT. 1996. C3d

11 Feb 2004

19:19

AR

AR210-IY22-14.tex

AR210-IY22-14.sgm

LaTeX2e(2002/01/18)

COMPLEMENT AND SLE

122.

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

123.

124.

125.

126.

127.

128.

129.

130.

131. 132.

133.

of complement as a molecular adjuvant: bridging innate and acquired immunity. Science 271:348–50 Carroll M. 1999. Role of complement receptors CD21/CD35 in B lymphocyte activation and survival. Curr. Top Microbiol. Immunol. 246:63–68 Boes M, Esau C, Fischer MB, Schmidt T, Carroll M, Chen J. 1998. Enhanced B-1 cell development, but impaired IgG antibody responses in mice deficient in secreted IgM. J. Immunol. 160:4776–87 Croix DA, Ahearn JM, Rosengard AM, Han S, Kelsoe G, et al. 1996. Antibody response to a T-dependent antigen requires B cell expression of complement receptors. J. Exp. Med. 183:1857–64 Tedder TF, Inaoki M, Sato S. 1997. The CD19-CD21 complex regulates signal transduction thresholds governing humoral immunity and autoimmunity. Immunity 6:107–18 Bitter-Suermann D, Burger R. 1989. C3 deficiencies. Curr. Top Microbiol. Immunol. 153:223–33 Jiang N, Reich CF 3rd, Monestier M, Pisetsky DS. 2003. The expression of plasma nucleosomes in mice undergoing in vivo apoptosis. Clin. Immunol. 106:139–47 Cutler AJ, Cornall RJ, Ferry H, Manderson AP, Botto M, Walport MJ. 2001. Intact B cell tolerance in the absence of the first component of the classical complement pathway. Eur. J. Immunol. 31:2087–93 Carroll MC. 2000. The role of complement in B cell activation and tolerance. Adv. Immunol. 74:61–88 Gommerman JL, Carroll MC. 2000. Negative selection of B lymphocytes: a novel role for innate immunity. Immunol. Rev. 173:120–30 Carroll MC. 1998. The lupus paradox. Nat. Genet. 19:3–4 Mohan C. 2001. Murine lupus genetics: lessons learned. Curr. Opin. Rheumatol. 13:352–60 Herron LR, Eisenberg RA, Roper E,

134.

135.

136.

137.

138.

139.

140.

141.

P1: IKH

455

Kakkanaiah VN, Cohen PL, Kotzin BL. 1993. Selection of the T cell receptor repertoire in Lpr mice. J. Immunol. 151:3450–59 Kotzin BL, Kappler JW, Marrack PC, Herron LR. 1989. T cell tolerance to self antigens in New Zealand hybrid mice with lupus-like disease. J. Immunol. 143:89– 94 Mihara M, Ohsugi Y, Saito K, Miyai T, Togashi M, et al. 1988. Immunologic abnormality in NZB/NZW F1 mice. Thymus-independent occurrence of B cell abnormality and requirement for T cells in the development of autoimmune disease, as evidenced by an analysis of the athymic nude individuals. J. Immunol. 141:85– 90 Steinberg AD, Roths JB, Murphy ED, Steinberg RT, Raveche ES. 1980. Effects of thymectomy or androgen administration upon the autoimmune disease of MRL/Mp-lpr/lpr mice. J. Immunol. 125:871–73 Wofsy D, Seaman WE. 1985. Successful treatment of autoimmunity in NZB/NZW F1 mice with monoclonal antibody to L3T4. J. Exp. Med. 161:378–91 Ando DG, Sercarz EE, Hahn BH. 1987. Mechanisms of T and B cell collaboration in the in vitro production of anti-DNA antibodies in the NZB/NZW F1 murine SLE model. J. Immunol. 138:3185–90 Datta SK, Patel H, Berry D. 1987. Induction of a cationic shift in IgG anti-DNA autoantibodies. Role of T helper cells with classical and novel phenotypes in three murine models of lupus nephritis. J. Exp. Med. 165:1252–68 Shivakumar S, Tsokos GC, Datta SK. 1989. T cell receptor alpha/beta expressing double-negative (CD4−/CD8−) and CD4+ T helper cells in humans augment the production of pathogenic antiDNA autoantibodies associated with lupus nephritis. J. Immunol. 143:103–12 Shlomchik MJ, Craft JE, Mamula MJ. 2001. From T to B and back again:

11 Feb 2004

19:19

456

142.

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

143.

144.

145.

146.

AR

AR210-IY22-14.tex

MANDERSON

¥

BOTTO

AR210-IY22-14.sgm

¥

LaTeX2e(2002/01/18)

P1: IKH

WALPORT

positive feedback in systemic autoimmune disease. Nat. Rev. Immunol. 1:147– 53 Ghebrehiwet B, Habicht GS, Beck G. 1990. Interaction of C1q with its receptor on cultured cell lines induces an antiproliferative response. Clin. Immunol. Immunopathol. 54:148–60 Ghebrehiwet B, Lu PD, Zhang W, Keilbaugh SA, Leigh LE, et al. 1997. Evidence that the two C1q binding membrane proteins, gC1q-R and cC1q-R, associate to form a complex. J. Immunol. 159:1429– 36 Chen A, Gaddipati S, Hong Y, Volkman DJ, Peerschke EI, Ghebrehiwet B. 1994. Human T cells express specific binding sites for C1q. Role in T cell activation and proliferation. J. Immunol. 153:1430–40 Jiang K, Chen Y, Xu CS, Jarvis JN. 2003. T cell activation by soluble C1q-bearing immune complexes: implications for the pathogenesis of rheumatoid arthritis. Clin. Exp. Immunol. 131:61–67 Zaffran Y, Destaing O, Roux A, Ory S, Nheu T, et al. 2001. CD46/CD3 costimulation induces morphological changes of human T cells and activation of Vav, Rac, and extracellular signal-regulated kinase mitogen-activated protein kinase. J. Immunol. 167:6780–85

147. Marie JC, Astier AL, Rivailler P, Rabourdin-Combe C, Wild TF, Horvat B. 2002. Linking innate and acquired immunity: divergent role of CD46 cytoplasmic domains in T cell induced inflammation. Nat. Immunol. 3:659–66 148. Kemper C, Chan AC, Green JM, Brett KA, Murphy KM, Atkinson JP. 2003. Activation of human CD4+ cells with CD3 and CD46 induces a T-regulatory cell 1 phenotype. Nature 421:388–92 149. Davis LS, Patel SS, Atkinson JP, Lipsky PE. 1988. Decay-accelerating factor functions as a signal transducing molecule for human T cells. J. Immunol. 141:2246– 52 150. Cerny J, Stockinger H, Horejsi V. 1996. Noncovalent associations of T lymphocyte surface proteins. Eur. J. Immunol. 26:2335–43 151. Fernandez-Centeno E, de Ojeda G, Rojo JM, Portoles P. 2000. Crry/p65, a membrane complement regulatory protein, has costimulatory properties on mouse T cells. J. Immunol. 164:4533–42 152. Shevach EM. 2000. Regulatory T cells in autoimmmunity. Annu. Rev. Immunol. 18:423–49 153. von Herrath MG, Harrison LC. 2003. Antigen-induced regulatory T cells in autoimmunity. Nat. Rev. Immunol. 3:223–32

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:431-456. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104626

Annu. Rev. Immunol. 2004. 22:457–83 doi: 10.1146/annurev.immunol.22.012703.104626 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on November 10, 2003

DROSOPHILA: The Genetics of Innate Immune Recognition and Response

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Catherine A. Brennan and Kathryn V. Anderson Sloan-Kettering Institute, Memorial Sloan-Kettering Cancer, New York, New York 10021; email: [email protected], [email protected]

Key Words Toll, Imd, PGRP, autoimmunity, pathogenesis ■ Abstract Because of the evolutionary conservation of innate mechanisms of host defense, Drosophila has emerged as an ideal animal in which to study the genetic control of immune recognition and responses. The discovery that the Toll pathway is required for defense against fungal infection in Drosophila was pivotal in studies of both mammalian and Drosophila immunity. Subsequent genetic screens in Drosophila to isolate additional mutants unable to induce humoral responses to infection have identified and ordered the function of components of two signaling cascades, the Toll and Imd pathways, that activate responses to infection. Drosophila blood cells also contribute to host defense through phagocytosis and signaling, and may carry out a form of self-nonself recognition that is independent of microbial pattern recognition. Recent work suggests that Drosophila will be a useful model for dissecting virulence mechanisms of several medically important pathogens.

INTRODUCTION In 1989, Janeway proposed that innate immune mechanisms, those that rely on detection of microbes by germline encoded receptors, are ancient and essential for earliest detection of and defense against infection in mammals (1). In 1996, Lemaitre and coworkers demonstrated that the Toll receptor, previously known for its essential role during Drosophila embryonic development, is required for antifungal defense in Drosophila (2). This finding stimulated the identification of the mammalian Toll-like receptors (TLRs) and the demonstration of their importance in mammalian innate immunity. We now know that mice that lack TLRs are susceptible to infection and are impaired in the ability to activate adaptive immune mechanisms, supporting Janeway’s predictions (3–5). Well before the power of Drosophila genetics was harnessed to study regulation of immune responses, insects were already known to have sophisticated immune systems, involving phagocytic blood cells, serine proteolytic cascades, and inducible humoral responses, thanks to decades of biochemical work with larger insects. In particular, insects induce a number of antimicrobial peptides 0732-0582/04/0423-0457$14.00

457

11 Feb 2004

19:23

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

458

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

upon immune challenge that are effective against Gram-negative or Gram-positive bacteria or fungi. Following the discovery of the immune function of Drosophila Toll, genetic screens were designed to isolate Drosophila mutants that could not induce particular antimicrobial peptides in response to infection. Genetic methods have identified about two dozen genes required for induction of the antimicrobial peptides. Many of these genes encode proteins in two signaling pathways that control the activation of NF-κB-like factors in response to infection, the Toll and Imd pathways, both named for the first gene to be discovered in the pathway. Immune signaling through Toll leads to the activation of two NF-κB factors, Dif and Dorsal. Activation of the Imd signaling pathway culminates in the activation and nuclear translocation of the third Drosophila NF-κB-like factor, Relish. The early finding that Toll mutants are impaired in survival to fungal infection and imd mutants impaired in antibacterial responses suggested that distinct pathways are used to detect and induce responses against bacteria and fungi. However, we now know that survival to Gram-positive bacterial infection also requires the Toll pathway. In addition to its importance in activation of antifungal responses, Toll is a central regulator of multiple aspects of Drosophila immunity, including resistance to bacterial infection, blood cell activation, and regulation of a melanization cascade. Genetic screens have also identified two peptidoglycan recognition proteins (PGRPs) that bind bacterial components directly. One recognizes Gram-negative bacteria and activates the Imd pathway, and the other detects Gram-positive bacteria and triggers Toll signaling. These findings fulfil Janeway’s prediction that in innate immunity, pattern recognition receptors (PRRs) would recognize conserved molecular features of microbes, or pathogen-associated molecular patterns (PAMPs). What was not foreseen in 1989, and what Drosophila has revealed, is that innate immune systems can discriminate among PAMPs that are characteristic of different microbial classes and activate the most appropriate defenses. Still under intense scrutiny in Drosophila are the mechanisms linking detection of microbes to signaling through the Imd and Toll pathways. Other aspects of the Drosophila immune response are not as well understood as the signaling pathways that lead to the humoral responses, but are ripe for genetic dissection. Blood cells are activated in response to infection, but our understanding of the mechanisms and consequences of blood cell activation are fragmentary. Serine protease cascades, which also activate several aspects of the mammalian immune response (6, 7), are required for activation of Toll signaling and the melanization response in Drosophila. Drosophila can also activate immune responses in the absence of microbial PAMPs—both in response to infestation with parasites, and under autoimmune conditions generated by a variety of mutations. Although the mechanisms of immune activation under these circumstances are not known, insect blood cells are able to discriminate between self and nonself, and both aberrant basement membrane patterns and endogenous DNA may be immunostimulatory.

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DROSOPHILA IMMUNITY

P1: IKH

459

Drosophila has recently emerged as a suitable model organism to investigate virulence mechanisms of a wide range of medically important pathogens, including Pseudomonas, Serratia, Mycobacteria, and malaria parasites. Many of the bacterial virulence mechanisms that are essential to establish infection in mammals also contribute to pathogenesis in flies, including type three secretion systems (TTSS) and ability to proliferate inside macrophages. Major questions in innate immunity concern how germline encoded receptors distinguish self from nonself and discriminate among various types of foreign invaders in order to activate the most effective response. The striking commonalities between Drosophila and mammalian innate immunity, including Toll-NF-κB signaling, phagocytosis, serine protease cascades, PGRPs, and autoimmune defects suggest that in the years to come studies in Drosophila will continue to shed light on immune mechanisms that are also important in humans.

OVERVIEW OF THE DROSOPHILA IMMUNE RESPONSE A Diversity of Infectious Threats Because Drosophila is not an agricultural pest, there is not a long history of entomological study of their pathogens. Much of our knowledge of Drosophila immunity thus concerns the responses to microbes that are not normally pathogenic and do not infect wild-type Drosophila unless directly injected. Studying immune responses to these opportunistic infections may be particularly relevant to mammalian immunity, as generalized immune defenses tend to be more evolutionarily conserved than ones specific to individual virulent pathogens. Nevertheless, some microorganisms and parasites are known that can naturally infect Drosophila, permitting finer analysis of immune mechanisms without the complications that a wound can introduce (8–10). This review primarily addresses Drosophila immunity to extracellular bacteria and fungi, as well as recognition of parasites, but a brief survey of a wider range of pathogens is presented. Drosophila are very adept at eliminating invading bacteria: Larvae injected with 5000 E. coli CFU clear the infection within 6 h (11). Two Gram-negative bacteria are known that can naturally infect larvae through the gut (9, 12). Otherwise, Drosophila antibacterial responses are assessed in the lab by directly injecting Gram-negative and Gram-positive bacteria. Neither of two intracellular bacterial types studied in Drosophila, Mycobacteria, or the obligate intracellular rickettsial symbiont Wolbachia, elicits a humoral antibacterial response (13, 14). Specific immune defenses against intracellular pathogens are not known in Drosophila. BACTERIA

Several fungal species, including Beauveria bassiana, can penetrate the cuticle of Drosophila and establish a lethal infection. Other fungi can be injected, and are lethal only in immunocompromised mutants (2, 8).

FUNGI

11 Feb 2004

19:23

460

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PARASITES Parasitoid wasps that lay their eggs in fly larvae represent a significant threat to wild Drosophila. In a successful infestation, the wasp egg hatches, the larva eats the fly pupa from inside, and an adult wasp eventually emerges (10, 15). The host protective response includes the encapsulation of the wasp egg by specialized lamellocyte blood cells (see below). Specific genetic loci in Drosophila correlate with ability to resist wasp infestation, although the genes have not yet been identified (16). A flagellate protozoan can induce systemic antimicrobial expression from the fly gut, and kills Drosophila when injected (17).

VIRUSES Several viruses that can infect Drosophila have been characterized, including rhabdoviruses, picornaviruses, baculoviruses, retroviruses, and birnaviruses (18–23). General insect antiviral strategies are not understood, although one Drosophila gene required for resistance to a rhabdovirus encodes a protein that may interfere with virus replication (24).

Overview of Innate Immune Responses There is no evidence in Drosophila or other insects for an adaptive immune system like that of mammals: specific antisera are not produced, and no sign of somatic gene rearrangement or a system resembling MHC antigen presentation has been found (25). Drosophila and most other invertebrates are thought to rely exclusively on innate immune mechanisms. Drosophila has an open circulatory system that disseminates the mediators and effectors of immune responses, most notably the blood cells and the antimicrobial peptides. The blood, also called the hemolymph, circulates in the extracellular space, or hemocoel, which is lined with a basement membrane (26). The fat body, an analog of the mammalian liver, is an extensive monolayer sheet of cells and is the source of most of the antimicrobial peptides produced in response to systemic infection (27, 28). ANTIMICROBIAL PEPTIDES Within hours of infection, transcription of a battery of antimicrobial peptides is induced in the fat body and the peptides are secreted into the blood. Insect antimicrobial peptides were originally isolated from larger insects based on their activities against different types of microbes and are active against fungi (Drosomycin, Metchnikowin, Cecropin), Gram-negative bacteria (Attacin, Cecropin, Diptericin, Drosocin), or Gram-positive bacteria (Defensin, Metchnikowin). Many of the peptides work by disrupting bacterial membranes (29, 30). Mutants impaired in the Imd and Toll signaling pathways that induce the antimicrobial peptide genes are severely immunocompromised (2). The ability of these mutants to resist infection can be rescued by transgenic expression of the appropriate peptides, attesting to the importance of the antimicrobial peptides in fighting infection (31).

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

DROSOPHILA IMMUNITY

P1: IKH

461

Drosophila also relies on blood cells to protect against infection. Plasmatocytes, phagocytic macrophage-like cells, comprise about 90% of the blood cell population (26, 32). Crystal cells are a source of enzymes for the melanization reaction (see below) (33). Lamellocytes are extremely flattened cells that differentiate in response to certain immune challenges, and encapsulate large invaders such as parasite eggs (26, 32). The plasmatocytes may have functionally distinct subgroups. Some are sessile while others circulate, and some are more highly phagocytic than others (32). Several genes are differentially expressed among plasmatocytes, and efforts are under way to generate monoclonal antibodies that can discriminate among Drosophila blood cells (34, 35). Phagocytosis is a vital contribution of blood cells to immunity; most bacteria injected into a fly are taken up by the blood cells within minutes (26, 36). Blood cells are required for signaling to the fat body under some infection conditions (9, 37). They also accumulate at wound sites and help form clots (33, 38). The importance of blood cells in fighting infection is shown by sensitization to infection seen when phagocytosis is blocked or in mutants that lack blood cells (36, 39). Although it is beyond the scope of this review, there is significant homology between Drosophila and mammalian hematopoieisis. Both require the function of genes in the GATA, NF-κB, Notch, Runt/AML1, JAK-STAT, Ras, and VEGF families and pathways (40–48).

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CELLULAR RESPONSES

MELANIZATION The deposition of melanin is a rapid, highly localized defense triggered by wounding and the presence of foreign invaders. Melanization contributes to wound clotting and encapsulation of wasp eggs, and produces toxic intermediates including reactive oxygen species. Phenoloxidase, which catalyzes melanin production, is maintained as an inactive zymogen and is activated by a serine protease cascade (49). A mutant lacking hemolymph phenoloxidase is sensitized to infection and is vulnerable to death from wounds (38, 39). BARRIER EPITHELIA All the surface epithelia of Drosophila that contact the environment, including the exterior, the gut, and the tracheae, induce antimicrobial gene expression upon contact with microbes (50–52). The Imd pathway regulates the induction of all peptides in the epithelia, including antifungal peptides (50– 52). In mammals, antimicrobial peptides also play key roles in epithelial defenses against infection (53, 54).

The Question of Specificity Can innate immune systems tailor responses to the type of immune challenge? Some aspects of the Drosophila humoral response are highly specific. For example, fungal infection specifically induces Metchnikowin and Drosomycin, the two antifungal peptides (8). Infection with Gram-negative bacteria, on the other hand, induces many antimicrobial peptides, even the antifungal Drosomycin (9, 37). However, some specificity is apparent because Drosomycin is induced only transiently by Gram-negative bacterial infection, whereas the expression of antibacterial

11 Feb 2004

19:23

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

462

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

peptides is sustained (8, 55). Selective activation by different microbes of either the Imd or the Toll pathway has been proposed to account for the specificity of immune response in Drosophila. However, there is evidence that the system is more complex than this, which is discussed below. Although there is no evidence in insects for a system of adaptive immunity like that in mammals, the possibility of specific immunological memory has not been excluded. Indeed, the recent finding that some arthropods are able to transfer specific immunity to their offspring suggests that a system of inducible immunological memory could also exist in insects (55a). Recognition and memory of evolved pathogens likely involves molecules that remain to be characterized.

THE IMD AND TOLL PATHWAYS CONTROL THE HUMORAL RESPONSES Signaling through the Imd and Toll pathways results in the translocation of distinct NF-κB factors to the fat body nucleus and accounts for most of the transcriptional induction of genes in response to fungal and bacterial infection (2, 56–59) (Figure 1).

Toll Pathway Adult Drosophila mutants lacking the function of Toll pathway elements are unable to induce Drosomycin in response to fungi and are susceptible to fungal infection, −−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→ Figure 1 Imd and Toll signaling pathways activate humoral antimicrobial defenses in the Drosophila fat body. Toll signaling is activated by fungal and Gram-positive invaders by different mechanisms. By an unknown mechanism, fungi trigger a cascade involving the serine protease Persephone, which results in the proteolytic activation of Sp¨atzle, a ligand for Toll. Gram-positive invaders are recognized by an independent process that requires a circulating peptidoglycan recognition protein. Toll signaling culminates in the translocation of the NF-κB factor Dif to the nucleus where it activates transcription of the antifungal Drosomycin and other genes. Gram-negative bacteria are recognized by a transmembrane peptidoglycan receptor, PGRP-LC. In at least some cases (see text), blood cells are required for induction of defenses against Gram-negative bacteria, suggesting that the blood cells may signal to the fat body. Nitric oxide (NO) is implicated in blood cell activation of antibacterial defenses in the fat body, although its specific role is unknown. PGRP is upstream of the Imd pathway that culminates in the phosphorylation and cleavage of the Relish NF-κB factor, which enters the nucleus and activates many genes including the antibacterial Diptericin. Overexpression of Imd and several other constituents of the Imd pathway causes constitutive Diptericin expression. Genetic epistasis between overexpressing transgenes and mutations in other genes has tentatively ordered the pathway as shown. The dotted line indicates relationships suggested by overexpression experiments or physical associations.

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DROSOPHILA IMMUNITY

P1: IKH

463

yet they are able to induce Diptericin normally and resist most Gram-negative bacterial infections (2, 59–62). Toll signaling was thus originally considered an antifungal pathway. Interestingly, however, mutants lacking Toll pathway elements are also susceptible to Gram-positive infection (59, 63) (see Figure 1). Toll, first discovered in Drosophila, is a transmembrane protein with extracellular leucine-rich repeats and an intracellular signaling domain similar to that of the Interleukin-1 receptor (64). The ligand for Drosophila Toll is a circulating endogenous protein, Sp¨atzle, which is proteolytically activated by a serine protease cascade in response to infection (65, 66). Ligand binding to Toll activates a cytoplasmic cascade involving dMyD88, the IRAK-like kinase Pelle, and the adaptor

11 Feb 2004

19:23

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

464

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

protein Tube. This results in the degradation of the IκB-like Cactus, permitting the nuclear translocation of the NF-κB-like proteins Dif and Dorsal (2, 56, 57, 62, 67– 69). Most of the Toll pathway mutants were originally isolated for their maternal effect embryonic patterning defects; the mutants are viable, but females produce abnormally patterned embryos (70). Dif and dMyD88 mutants were isolated by reverse genetic approaches (56, 60–62). Although Drosophila has 9 Toll receptors, they are not believed to provide specificity by recognizing different PAMPs as they are in mammals. Toll does not directly bind microbial components. None of the other Drosophila Tolls have been identified in genetic screens for immunodeficient mutants or are upregulated by immune challenge (71). An early report that 18Wheeler (the second Drosophila Toll) had specific antibacterial defects has not been confirmed (72, 73). Several of the other Drosophila Tolls have morphogenetic and neural functions (74, 75). Sequence comparisons suggest that the last common ancestor of mammals and invertebrates may have had only one or a few Toll receptors that subsequently duplicated and diverged under different selection pressures in the two lineages (76). Some of the other Drosophila Tolls may have immune functions, but assessment of their contributions awaits loss-of-function analysis (Table 1).

Imd Pathway The Imd signaling pathway mediates the induction of Diptericin and other antibacterial peptide genes in the fat body in response to Gram-negative bacterial infection, and bears some resemblance to the mammalian TNF-α pathway (Figure 1) (77–79). Although Imd itself was discovered serendipitously, most Imd pathway components were identified in genetic screens for mutants unable to induce reporters of Diptericin expression in response to bacterial challenge, or to survive infection (36, 80–83). The Imd signaling pathway culminates in the activation of the NF-κB-family member Relish. Relish, like mammalian p100 and p105, is a compound protein with an N-terminal Rel domain and C-terminal ankyrin repeat domain and is proteolytically processed in response to upstream signals to generate an N-terminal Rel protein that enters the nucleus and activates transcription of target genes including Diptericin (58, 84). Proteolytic activation of Relish differs from that of p100 and p105 in that it is proteasome-independent and mediated directly by the caspase Dredd (58, 85). Relish activation also requires phosphorylation by an IκB kinase (IKK) complex (11, 83, 86). The genes encoding dTAK (a MAPKKK) and the adaptor protein dFADD are also required for induction of antibacterial peptide genes (87–90). All the Imd pathway mutants are unable to induce Diptericin and are susceptible to bacterial infection (2, 36, 82–84, 88, 89, 91, 92), whereas in vivo overexpression of Imd and other pathway elements causes constitutive Diptericin expression (87, 88, 91). Imd pathway mutants are often able to resist Gram-positive infection almost as well as wild-type (59, 63, 92). Negative regulators in the Imd pathway cannot be recovered in screens for immunodeficient mutants, but a screen for mutants that overexpress Diptericin

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

DROSOPHILA IMMUNITY

465

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 1 Proteins that may activate immune responses in response to microbial invasion. The roles of these proteins have not been confirmed by mutant analysis Gene(s)

Type of protein encoded

Evidence

References

PGRP-LE

PRR

In vivo overexpression induces constitutive Diptericin and melanization

(117)

dSR-C1

Scavenger receptor

RNAi in S2 blood cell line impairs ability to bind and phagocytose bacteria

(118)

GNBPs

Secreted; similar to bacterial β-1,3 glucanases and CD14 (3 in Drosophila)

Binds LPS, β-1,3 glucan; overexpression in S2 cells increases ability to induce antimicrobial peptides

(116, 119)

TEPs

Thiolester-containing; complement factor C3-like; opsonin?

Upregulated upon infection; RNAi of mosquito homolog impairs Gram-negative phagocytosis

(99, 120)

Masquerade

“Inactive” serine protease (null mutations lethal)

Upregulated upon Gram-positive, fungal infection; crayfish homolog is an opsonin

(55, 121)

dToll5

Toll-like receptor

Chimeric constitutively active protein with dToll5 cytoplasmic domain activates Drosomycin in S2 cells; coreceptor with Toll?

(122, 123)

dToll9

Toll-like receptor

Wild-type protein constitutively activates Drosomycin in S2 cells in absence of infection; through canonical Toll pathway

(124)

identified a ubiquitin ligase complex which targets Relish for destruction in the absence of infection (93).

Other Signaling Pathways The JAK-STAT and JNK pathways are important in the mammalian immune response (94, 95) and also signal during the Drosophila immune response. LPS stimulation of Drosophila blood cells activates JNK within minutes (96). Microarray analysis of RNAi-treated blood cells exposed to LPS indicates that the JNK pathway controls the rapid upregulation of cytoskeletal genes in response to infection (90). Although null JNK pathway mutants die as embryos due to a requirement for this pathway for early developmental events including epithelial fusion (97), a hypomorphic allele of DFos, a transcriptional effector of JNK signaling, impairs wound healing (38).

11 Feb 2004

19:23

466

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

JAK-STAT signaling in Drosophila is required for the induction in the fat body of a number of genes in response to infection, including the stress-induced gene totA, as well as Tep1, which encodes a thiolester-containing protein that may be an opsonin (see Table 1) (90, 98, 99). Upd3 is a cytokine-like protein produced by the blood cells upon infection, and it activates transcription of totA in the fat body by signalling through the receptor Domeless and the JAK-STAT pathway (98). In general, dependence on JAK-STAT signaling correlates with delayed, transient induction following immune challenge (90). In addition, constitutive JAK signaling hyperactivates the blood cells (see below).

PATTERN RECOGNITION AND ACTIVATION OF IMD AND TOLL SIGNALING The Imd and Toll pathways can be activated through binding by specific PGRPs of Gram-negative or Gram-positive bacterial molecules, respectively. It is not yet clear how fungal recognition activates Toll signaling, nor have the functions of other candidate PRRs been defined.

Recognition of Microbes PGRPs were first identified in moths as infection-induced proteins that bind peptidoglycan, triggering the proteolytic melanization cascade (100, 101). The Drosophila genome encodes at least 13 PGRPs, some with multiple splice-forms, and the human genome encodes 4 that also have splice variants (102–104). PGRPs share a 160 amino acid peptidoglycan recognition domain, and both mammals and Drosophila have genes that encode secreted (S) and transmembrane (L) forms (101). In Drosophila, a secreted PGRP is required for survival to Gram-positive bacteria, whereas mutants lacking the function of a transmembrane PGRP are susceptible to Gram-negative bacterial infection (92, 105–107). In addition, at least one of the Drosophila secreted PGRPs scavenges and degrades peptidoglycan (108). A hydrophobic groove shared by both secreted and transmembrane PGRPs may function to bring downstream effectors into proximity, promoting signaling (109), while the cytoplasmic tails of transmembrane PGRPs may also contribute to downstream signaling. PGRP-LC AND GRAM-NEGATIVE DETECTION Mutants in PGRP-LC fail to induce the antibacterial peptides and are susceptible to Gram-negative but not Grampositive infection (92, 106, 107). The transmembrane PGRP-LC has two major splice variants that share common transmembrane and cytoplasmic domains but have very different extracellular peptidoglycan recognition domains (103, 106). PGRP-LC acts upstream of the Imd pathway: Mutants are unable to proteolytically activate Relish during infection, and constitutive expression of Diptericin caused by PGRP-LC overexpression requires wild-type Imd function (92, 106).

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DROSOPHILA IMMUNITY

P1: IKH

467

It was surprising to find a PGRP implicated in the recognition of Gram-negative bacteria because Gram-negative peptidoglycan (PG) is in the inner cell wall layer, which is covered by the outer membrane. In contrast, the PG of Gram-positive bacteria is much more accessible on the cell wall surface. Lipopolysaccharide (LPS), on the other hand, is an abundant constituent of the Gram-negative outer membrane, and is an extremely immunogenic molecule in mammals (110). Because LPS is not found in Gram-positive bacteria, it was expected to be the key to the discrimination between Gram-negative and Gram-positive bacteria by Drosophila. However, Drosophila is able to discriminate between Gram-negative and Grampositive PG, which differ in a single amino acid, whereas LPS is not a potent inducer of the humoral response in vivo (111, 112). The PGRPs play an essential role in the recognition of and disrimination between Gram-negative and Gram-positive bacteria. Gram-negative PG, but not Gram-positive PG, is a potent inducer of Diptericin in flies and cultured blood cells, and requires PGRP-LC function (103, 112, 113). PGRP-LC was also identified in a blood cell RNAi screen as a protein involved in Gram-negative, but not Gram-positive, binding and phagocytosis (107). PGRP-SA AND GRAM-POSITIVE DETECTION A mutant lacking the function of PGRP-SA was identified in a screen for mutants unable to induce Drosomycin following a mixed bacterial infection (105). Although Drosomycin is an antifungal peptide, it is induced through the Toll pathway by both fungi and Gram-positive bacteria. PGRP-SA mutants are specifically susceptible to Gram-positive infection; resistance to fungi and Gram-negative bacteria is normal (105). Consistent with this, PGRP-SA mutants cannot induce Drosomycin in response to Gram-positive infection or Gram-positive PG, although they induce Drosomycin normally in response to fungal infection. Diptericin is induced normally in response to Gramnegative PG or whole bacteria (105, 112). PGRP-SA is a secreted PGRP consisting mainly in a single peptidoglycan recognition domain, and binds Gram-positive PG with high affinity (102, 105). The inability of the mutants to induce Drosomycin suggests that PGRP-SA may activate the Toll pathway, although this has not been tested genetically. Some aspects of secreted PGRP function may be evolutionarily conserved. A mouse mutant lacking the function of a secreted PGRP is also vulnerable to Gram-positive infection (114). OTHER POSSIBLE PATTERN RECOGNITION PROTEINS Other Drosophila proteins may have roles in recognizing microbial invaders, but because no mutants have yet been isolated, their requirements in immune function have not been directly tested. Some of these are listed in Table 1. Currently, there are no genetic data concerning host molecules involved in detection of fungi, although fungal β-1,3D-glucans induce antimicrobial gene expression in a Drosophila blood cell line, and a β-1,3-glucan-binding protein contributes to this response (115, 116).

11 Feb 2004

19:23

468

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

From Recognition to Fat Body Signaling

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

A major focus of current research is how pattern recognition events involving PGRPs or other host receptors leads to activation of the fat body NF-κB signaling pathways that induce the antimicrobial peptides. The importance of a circulating PGRP for Gram-positive detection and a transmembrane PGRP for Gram-negative detection suggests that the fly immune system detects these two classes of bacteria rather differently. Gram-positive detection has more in common with fungal pathways in the dependence on the Toll pathway, and possibly also on serine proteolytic cascades. ACTIVATION OF THE IMD PATHWAY The fat body is the source of the majority of the antimicrobial peptides produced in a systemic immune response. Relish translocation can be visualized here, and Imd signaling presumably takes place in fat body cells. However, the site of microbial detection by PGRP-LC may be elsewhere. Whereas constitutive fat body Diptericin expression can be induced by overexpressing PGRP-LC in the adult fat body, endogenous larval PGRP-LC expression is higher in the blood cells than the fat body (92, 102, 106). Consistent with this, in larvae, blood cells are required for the induction of Diptericin in the fat body in response to a Gram-negative gut infection (9, 37). These data suggest that blood cells might detect Gram-negative microbes and signal this information to the fat body. Blood cells are not required for induction of fat body Diptericin when Gram-negative bacteria are introduced through a wound (39), which may be the result of a signal generated at the site of injury. NO is implicated in the blood cell–dependent induction of fat body Diptericin. Exogenous NO induces Diptericin, and a pharmacological inhibitor of nitric oxide synthase (NOS) prevents Diptericin induction in response to a Gram-negative infection (37, 125). The response to NO requires Imd, suggesting that NO acts upstream of the Imd pathway. However, NO is unlikely to be the signal from blood cells to the fat body, as exogenous NO does not stimulate Diptericin induction in a mutant that lacks blood cells (37). NO appears to be important in some step in blood cells downstream of microbial recognition and upstream of a signal that is relayed to the fat body, possibly in blood cell activation. ACTIVATION OF THE TOLL PATHWAY In the Drosophila embryo, the Toll pathway is triggered by a cascade of four serine proteases that proteolytically activate the endogenous ligand Sp¨atzle (126). Although Sp¨atzle is also required for immune responses, the embryonic proteases are not (2). Necrotic is a serine protease inhibitor (serpin) that prevents constitutive activation of Toll-dependent immune responses in the absence of immune challenge, suggesting that a distinct serine proteolytic cascade activates Sp¨atzle in immune responses (65). Persephone, an immune response serine protease, was identified in a screen for mutations that suppressed the ability of necrotic mutants to constitutively activate Toll signaling. persephone mutants are susceptible to fungal infection and unable

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DROSOPHILA IMMUNITY

P1: IKH

469

to activate the Toll pathway in response to fungi (127). It is unclear how fungal infection leads to Persephone activation, but the possession of a prodomain suggests that Persephone may be activated by another serine protease (127). Necrotic may directly inhibit Persephone; fungal infection is likely to shift the balance between protease and serpin activities (127, 128). It is also not known whether Persephone directly activates Sp¨atzle, or whether there are additional intervening proteases. Activation of the Toll pathway by Gram-positive bacteria does not require Persephone, suggesting that fungi and Gram-positive bacteria activate Toll signaling by distinct mechanisms (127). Because Sp¨atzle is required for responses to some Gram-positive bacteria, PGRP-SA can likely activate a distinct serine protease cascade that leads to Sp¨atzle cleavage (59, 63, 105).

Do Imd and Toll Really Provide Specificity of Response? The model that selective activation of the Imd or Toll pathways confers specificity of response against Gram-negative bacteria or against Gram-positive bacteria and fungi, respectively, is an oversimplification. For example, Imd Toll pathway double mutants are more susceptible than single Imd pathway mutants to E. coli infection, and Toll pathway mutants are susceptible to Pseudomonas infection, arguing that Toll is important for resistance to Gram-negative infection (59, 129). On the other hand, Imd signaling is important for resistance to some Gram-positive bacteria such as Micrococcus luteus (59, 82). Another difficulty with the model that the Imd and Toll pathways confer specificity of response is the sharing of the Toll pathway by fungi and Gram-positive bacteria. Despite the use of distinct mechanisms for detecting fungi and Grampositive bacteria, the signals apparently converge at Sp¨atzle and Toll. In fact, some responses to Gram-positive bacteria and fungi are remarkably similar: Both types of infection trigger the massive induction of Masquerade, a serine protease-like protein, whereas Gram-negative infection does not (55). Another similarity between fungal and Gram-positive immune induction is the apparent reliance on circulating, rather than cell-associated detection mechanisms. Why would an immune system have one pathway dedicated to Gram-negative defense and another for responses to microorganisms as disparate as fungi and Gram-positive bacteria? It could be that the Toll pathway is the ancestral immune induction pathway in insects (Figure 2), and that a blood cell-mediated system to recognize Gramnegative bacteria evolved later, triggering a distinct pathway. The Imd pathway may also have an apoptotic role, suggesting that its immune function may have been co-opted secondarily (91). There may be features of Gram-negative bacteria that selected for an additional immune detection and induction mechanism, such as a higher growth rate or concealment of PG beneath the outer membrane. Indeed, a genome-wide microarray analysis of the kinetics of induction of immuneresponsive genes in wild-type and mutant conditions found a correlation between Imd regulation of early activated genes and Toll regulation of genes with a more delayed response (90). Specialization of Imd signaling for rapid responses would be

11 Feb 2004

19:23

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

470

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

Figure 2 Toll is a central regulator of the Drosophila immune response. In agreement with the idea that Toll may represent the ancestral immune signaling pathway in insects, Toll regulates many aspects of the immune response. In addition to activating antifungal defenses, Toll is also required for survival to many Gram-positive and some Gram-negative bacterial infections, for regulation of the melanization cascade and for regulation of blood cell proliferation, and is implicated in blood cell differentiation and activation. Toll may fulfill these roles in several tissues, such as the fat body and the blood cells. Sp¨atzle is the only known ligand for Toll in Drosophila; however, Toll mutants are more impaired in the induction of antimicrobial peptides than are null sp¨atzle mutants (2), suggesting that there may be additional Toll ligands.

consistent with the independence of Imd of both Drosomycin induction, which has slow kinetics, and survival to fungi, which are slower-growing microorganisms.

ACTIVATION OF MELANIZATION AND BLOOD CELL RESPONSES In addition to the well-characterized antimicrobial peptide induction, infection also triggers blood cell activation and melanization, events that are less well understood in Drosophila.

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

DROSOPHILA IMMUNITY

P1: IKH

471

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Activation of the Prophenoloxidase Pathway Biochemical experiments in the silkworm and crayfish revealed that a serine protease cascade regulates the activation of prophenoloxidase (PPO), which catalyzes melanin production (49, 130–132). An important regulator of the melanization response is Serpin27A (Spn27A), which specifically inhibits the PPO-activating enzyme (PPAE) (133, 134). Spn27A mutants exhibit sporadic melanization in the absence of immune challenge, and the entire animal becomes melanized after septic injury (133, 134). The inhibitory effect of Spn27A on melanization appears to be overcome in two ways under conditions of immune challenge. The melanization cascade can be rapidly triggered by microbial products, by mechanisms that involve PGRPs (100, 117, 131) and that likely activate an upstream protease that leads to hyperproduction of PPAE, depleting Spn27A. An additional uncharacterized inhibitor of Spn27A may be transcriptionally induced by Sp¨atzle-Toll signaling (134). It appears that rapid activation of a protease cascade, triggered by pattern recognition, initiates an immediate melanization response, and that sustained activation is ensured by local Toll-mediated depletion of Spn27A.

Activation of Blood Cells PHAGOCYTOSIS Phagocytic uptake of invaders by blood cells is a vital defense strategy in both flies and mammals (36, 135). Drosophila and mammalian phagocytosis are homologous actin-mediated processes (118). Critical unanswered questions in both mammalian and Drosophila phagocytosis include how ingested particles are trafficked and the role of phagocytosis in stimulating other immune responses (135). The ability of a blood cell to phagocytose a particular particle is influenced by the affinity of host receptors for surface molecules on the particle. For example, a Drosophila scavenger receptor-like protein, dSR-C1, confers the ability to phagocytose bacteria but not yeast on a heterologous cell line (118). However, Drosophila blood cells are able to phagocytose abiotic particles such as polystyrene beads, so pattern recognition of microbial PAMPs is not an absolute requirement (36). Few Drosophila mutants are detectably impaired in phagocytosis, possibly reflecting redundancy of mechanisms promoting particle uptake and requirements for cytoskeletal proteins for viability. An RNAi screen in cultured blood cells tested the requirements for 1000 randomly selected genes in phagocytosis of different microbes, and defined roles for proteins involved in cytoskeletal function and vesicle formation and transport (107). In addition, the removal of PGRP-LC function impaired the binding and phagocytosis of Gram-negative bacteria by approximately 30% (107). Null PGRP-LC mutants are not detectably impaired in phagocytosis, suggesting either that a 30% reduction in phagocytic ability is not detectable in vivo or that additional mechanisms compensate in vivo for the loss of this receptor (106). Several Drosophila proteins have been proposed to bind microbes and promote their uptake by blood cells (Table 1).

15 Mar 2004

15:26

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

472

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

OTHER ASPECTS OF BLOOD CELL ACTIVATION Wasp parasitization triggers blood cell proliferation, the differentiation of the lamellocyte blood cell type, and an encapsulation response (see below) (32, 35, 136). Although the regulation of these dramatic events is not understood, they are mimicked in mutants with hyperactivated Toll and JAK-STAT signaling (42, 137, 138) (Figure 3). Toll also regulates steady-state hemocyte numbers: There are fewer blood cells in loss-of-function mutants (42). In addition, blood cells accumulate at wound sites (38); because blood cells with overactivated Toll signaling also aggregate, Toll could also be involved in this type of activation (42). In short, Toll signaling is implicated in several aspects of blood cell activation in Drosophila, but the regulation of blood cell activation in response to actual immune challenge is not understood.

IMMUNE RESPONSES IN THE ABSENCE OF MICROBIAL PATHOGENS PAMPs are most clearly defined for bacteria and fungi, microorganisms from entirely different kingdoms than animals. However, like mammals, Drosophila is able to mount immune responses in the absence of PAMPs, for example during parasite infestation and under autoimmune conditions. Deviations from the normal basement membrane pattern as well as presence of endogenous DNA in the blood are both associated with immune activation.

Basement Membrane PARASITES A parasitic wasp egg in a Drosophila larva triggers an encapsulation immune response designed to seal off and kill the wasp embryo before it can hatch and kill the larva. Circulating plasmatocytes appear to recognize the wasp as foreign, and attach to it. Lamellocytes then adhere in layers, and the entire capsule is melanized (32, 139). This protective response has some similarities to granuloma formation by mammalian macrophages and T cells (140). Because wasps are also insects, they are not expected to have obligate molecular signatures, or PAMPs, that allow them to be recognized as invaders. How, then, are wasp eggs recognized as foreign? Transplantation experiments suggest that Drosophila blood cells may recognize the absence of endogenous or presence of foreign basement membrane on invaders as nonself. Drosophila does not encapsulate fat bodies of within-species transplants, but fat bodies transplanted from other species are encapsulated (141).

In some mutants, Drosophila blood cells aberrantly encapsulate the fly’s own tissue, representing a kind of autoimmune defect (Figure 3). The resulting melanotic capsules resemble encapsulated parasitoid eggs, with layers of melanized lamellocytes. Although melanotic capsules can appear in mutants with constitutively activated blood cells (Figure 3), they also occur in mutants in which damaged or aberrant tissues trigger an immune response (142).

MELANOTIC CAPSULES

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

DROSOPHILA IMMUNITY

P1: IKH

473

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Several melanotic capsule mutants show disruption of the basement membrane surrounding various tissues and the subsequent encapsulation of the exposed tissues. For example, in Tu-Sz[ts] mutants, disruption of the basement membrane over part of the fat body precedes the adhesion of lamellocytes and formation of melanotic capsules there (26). Transplanted Drosophila fat bodies whose basement membranes have been damaged are encapsulated, whereas intact fat bodies are not, supporting the notion that disruptions to the endogenous basement membrane can trigger immune responses (141). BLOOD CELLS AND BASEMENT MEMBRANE The blood cells are implicated in the recognition of basement membrane abnormalities in parasite infestation and autoimmune activation. The notion of blood cells constantly surveying the basement membrane lining the hemocoel is consistent with their roles in basement membrane secretion, repair, and degradation (26, 33, 143). Blood cells seem to be able to distinguish the basement membrane of healthy self from that of damaged self, as well as absence of basement membrane on abiotic material, and possibly presence of foreign basement membrane on parasites. The basement membrane is a proteoglycan-rich matrix, and experiments in other insects, in which beads coated with different materials were transplanted, suggest that the self characteristics may be related to carbohydrate composition (33, 144).

Endogenous DNA In addition to those with basement membrane defects, some other mutants with melanotic capsules have defects in apoptosis. Mutants lacking the function of the fly proapoptotic homologs of Ced-3 caspase and Ced-4/Apaf-1 develop melanotic capsules (145, 146). Drosophila blood cells do normally recognize and phagocytose apoptotic cells through a CD36-like receptor called Croquemort (147). It is possible that cells failing to undergo proper apoptosis are recognized as abnormal, triggering encapsulation. Alternatively, cells undergoing aberrant apoptosis might release immunostimulatory molecules. Ced-3 protein is able to fragment DNA; perhaps DNA in the blood of Ced-3 caspase mutants triggers an immune response resulting in encapsulation of self tissue (145). Fly mutants lacking the function of two other DNAses required for the degradation of apoptotic cell DNA show constitutive expression of Diptericin (148). This further suggests that endogenous DNA may stimulate the innate immune system of Drosophila as it does in mammals (149).

STUDYING MEDICALLY IMPORTANT PATHOGENS IN DROSOPHILA Drosophila has recently emerged as a very promising system in which to study the virulence of a variety of medically important pathogens. Several microbes have been shown to infect flies with similar mechanisms to those known from mammals.

11 Feb 2004

19:23

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

474

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

Because Drosophila is a genetically tractable animal with a sophisticated, bloodcell-dependent innate immune system, the possibilities for increasing our understanding of pathogenesis are tremendous. Pseudomonas aeroginosa is a ubiquitous Gram-negative bacterium that is responsible for opportunistic infections in wounds and in immunocompromised patients. Remarkably, some of the mechanisms used by this bacterium to infect and kill the host are common to mammals, insects, nematodes, and plants (150). All 11 virulence factors required for maximal pathogenicity in mammals are also required for maximum virulence in flies (129). The TTSS, which contributes to virulence in both mammals and insects, is activated upon entry into Drosophila (151). This is one of the first reports of activation of the TTSS of any bacterial species in a whole animal, suggesting that Drosophila may be a suitable in vivo alternative to simulating combinations of signals in vitro. A screen for P. aeroginosa mutants impaired in Drosophila killing led to the identification of a gene cluster that regulates motility factors important for virulence in flies and mammals (152). Serratia marcescens is a Gram-negative insect pathogen responsible for opportunistic infections in humans. Several S. marcescens mutants impaired in Drosophila killing are also attenuated in mammalian infection models (153). Like Mycobacterium tuberculosis, M. marinum can proliferate inside vertebrate macrophages in which phagosome acidification has been blocked. M. marinum can also infect Drosophila blood cells by a similar mechanism, killing the fly. M. marinum upregulates some of the same genes in vertebrate and insect phagosomes, and at least one bacterial virulence factor contributes to pathogenesis in both systems (13). A Gram-negative plant pathogen, Erwinia carotovora, that may be spread by Drosophila, has also provided tools for the study of mammalian disease. A novel virulence factor was identified from a genetic screen for E. carotovora mutants unable to infect Drosophila larvae. This gene is sufficient to confer ability to infect Drosophila not only on noninfectious E. carotovora strains, but also on other Enterobacteria such as E. coli and Salmonella typhimurium (154). These new strains of medically important bacteria that are infectious in Drosophila may prove to be powerful tools to investigate virulence mechanisms. In addition to serving as lab models for medically important bacterial diseases, insects are also the natural vectors for many pathogens that infect mammals. Although malaria-causing Plasmodia are transmitted by mosquitoes, Plasmodia are able to infect Drosophila in the lab and progress through several steps of their complex life cycle (155). In addition, knowledge of insect immunity that Drosophila studies have yielded is being applied to studies of mosquito-Plasmodium infections with the goal of developing antimalaria strategies (156).

FUTURE DIRECTIONS In the eight years since Toll and imd mutants were found to be differentially susceptible to fungal and bacterial infection (2), and the six years since the first genetic screen for immune defects was reported (57), genetic analysis has revealed many of

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

DROSOPHILA IMMUNITY

P1: IKH

475

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

the essential aspects of these two signaling pathways. Continued forward genetic screens, as well as reverse genetic analysis of genes regulated by infection, will identify more components required for diverse aspects of the immune response. We can anticipate that future research in Drosophila immunity is likely to identify more novel pattern recognition mechanisms, define mechanisms of blood cell activation, illuminate interactions between NF-κB and other signaling pathways, define mechanisms that allow recognition of intracellular bacterial and viral infections, provide perspectives on autoimmunity, and define specific responses to pathogenic organisms. ACKNOWLEDGMENTS The authors wish to thank Derek Sant’Angelo for comments on the manuscript. CAB is the Keyser Family Fellow supported by the Damon Runyon Cancer Research Foundation (DRG-3582). KVA is supported by National Institutes of Health grant # AI45149. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Janeway CA. 1989. Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harbor Symp. Quant. Biol. 54:1–13 2. Lemaitre B, Nicolas E, Michaut L, Reichhart J-M, Hoffmann JA. 1996. The dorsoventral regulatory gene cassette sp¨atzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell 86:973–83 3. Poltorak A, He X, Smirnova I, Liu MY, Huffel CV, et al. 1998. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282:2085–88 4. Takeuchi O, Hoshino K, Akira S. 2000. Cutting edge: TLR2-deficient and MyD88-deficient mice are highly susceptible to Staphylococcus aureus infection. J. Immunol. 165:5392–96 5. Schnare M, Barton GM, Holt AC, Takeda K, Akira S, Medzhitov R. 2001. Toll-like receptors control activation of adaptive immune responses. Nat. Immunol. 2:947– 50 6. Volanakis J. 1998. Overview of the com-

7. 8.

9.

10.

11.

12.

plement system. In The Human Complement System in Health and Disease, ed. J Volanakis, M Frank, pp. 9–32. New York: Marcel Dekker Dahlback B. 2000. Blood coagulation. Lancet 355:1627–32 Lemaitre B, Reichhart J-M, Hoffmann JA. 1997. Drosophila host defense: differential induction of antimicrobial peptide genes after infection by various classes of microorganisms. Proc. Natl. Acad. Sci.USA 94:14614–19 Basset A, Khush RS, Braun A, Gardan L, Boccard F, et al. 2000. The phytopathogenic bacteria Erwinia carotovora infects Drosophila and activates an immune response. Proc. Natl. Acad. Sci. USA 97:3376–81 Carton Y, Nappi AJ. 1997. Drosophila cellular immunity against parasitoids. Parasitol. Today 13:218–27 Lu Y, Wu LP, Anderson KV. 2001. The antibacterial arm of the Drosophila innate immune response requires an IκB kinase. Genes Dev. 15:104–10 Flyg C, Kenne K, Boman HG. 1980.

11 Feb 2004

19:23

476

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

Insect pathogenic properties of Serratia marcescens: phage-resistant mutants with a decreased resistance to Cecropia immunity and a decreased virulence to Drosophila. J. Gen. Microbiol. 120:173– 81 Dionne MS, Ghori N, Schneider DS. 2003. Drosophila melanogaster is a genetically tractable model host for Mycobacterium marinum. Infect. Immun. 71: 3540–50 Bourtzis K, Pettigrew MM, O’Neill SL. 2000. Wolbachia neither induces nor suppresses transcripts encoding antimicrobial peptides. Insect Mol. Biol. 9:635– 39 Melk JP, Govind S. 1999. Developmental analysis of Ganaspis xanthopoda, a larval parasitoid of Drosophila melanogaster. J. Exp. Biol. 202:1885–96 Carton Y, Nappi AJ. 2001. Immunogenetic aspects of the cellular immune response of Drosophila against parasitoids. Immunogenetics 52:157–64 Boulanger N, Ehret-Sabatier L, Brun R, Zachary D, Bulet P, Imler J. 2001. Immune response of Drosophila melanogaster to infection with the flagellate parasite Crithidia spp. Insect. Biochem. Mol. Biol. 31:129–37 Seecof R. 1968. The sigma virus infection of Drosophila melanogaster. Curr. Top. Microbiol. Immunol. 42:59–93 Teninges D, Plus N. 1972. P virus of Drosophila melanogaster, as a new picornavirus. J. Gen. Virol. 16:103–9 Teninges D, Ohanessian A, RichardMolard C, Contamine D. 1979. Isolation and biological properties of Drosophila X virus. J. Gen. Virol. 42:241–54 Brun G, Plus N. 1980. The viruses of Drosophila. In The Genetics and Biology of Drosophila, Part C ed. M Ashburner, pp. 625–702. New York: Academic Morris TD, Miller L. 1993. Characterization of productive and non-productive AcMNPV infection in selected insect cell lines. Virology 197:339–48

23. Kim A, Terzian C, Santamaria P, Pelisson A, Prud’homme N, Bucheton A. 1994. Retroviruses in invertebrates: the gypsy retrotransposon is apparently an infectious retrovirus of Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 91:1285–89 24. Wyers F, Petitjean AM, Dru P, Gay P, Contamine D. 1995. Localization of domains within the Drosophila Ref(2)P protein involved in the intracellular control of sigma rhabdovirus multiplication. J. Virol. 69:4463–70 25. Bernheimer AW, Caspari E, Kaiser AD. 1952. Studies on antibody formation in caterpillars. J. Exp. Zool. 119:23–35 26. Rizki TM, Rizki RM. 1984. The cellular defense system of Drosophila melanogaster. In Insect Ultrastructure, ed. RC King, HH Akai, pp. 579–604. New York: Plenum 27. Rizki TM. 1978. Fat body. In The Genetics and Biology of Drosophila, ed. M Ashburner, TRF Wright, pp. 561–99. New York: Academic 28. Sondergaard L. 1993. Homology between the mammalian liver and the Drosophila fat body. Trends Genet. 9:193 29. Bulet P, Hetru C, Dimarcq JL, Hoffmann D. 1999. Antimicrobial peptides in insects; structure and function. Dev. Comp. Immunol. 23:329–44 30. Meister M, Hetru C, Hoffmann J. 2000. The antimicrobial host defense of Drosophila. Curr. Top. Microbiol. Immunol. 248:17–36 31. Tzou P, Reichhart J-M, Lemaitre B. 2002. Constitutive expression of a single antimicrobial peptide can restore wild-type resistance to infection in immunodeficient Drosophila mutants. Proc. Natl. Acad. Sci. USA 99:2152–57 32. Lanot R, Zachary D, Holder F, Meister M. 2001. Postembryonic hematopoiesis in Drosophila. Dev. Biol. 230:243–57 33. Lackie AM. 1988. Haemocyte behaviour. Adv. Insect Physiol. 21:85–178 34. Braun A, Lemaitre B, Lanot R, Zachary

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

DROSOPHILA IMMUNITY

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

35.

36.

37.

38.

39.

40.

41. 42.

43.

44.

45.

D, Meister M. 1997. Drosophila immunity: analysis of larval hemocytes by Pelement-mediated enhancer trap. Genetics 147:623–34 Kurucz E, Zettervall CJ, Sinka R, Vilmos P, Pivarcsi A, et al. 2003. Hemese, a hemocyte-specific transmembrane protein, affects the cellular immune response in Drosophila. Proc. Natl. Acad. Sci. USA 100:2622–27 Elrod-Erickson M, Mishra S, Schneider D. 2000. Interactions between the cellular and humoral immune responses in Drosophila. Curr. Biol. 10:781–84 Foley E, O’Farrell PH. 2003. Nitric oxide contributes to induction of innate immune responses to gram-negative bacteria in Drosophila. Genes Dev. 17:115–25 R¨amet M, Lanot R, Zachary D, Manfruelli P. 2002. JNK signaling pathway is required for efficient wound healing in Drosophila. Dev. Biol. 241:145–56 Braun A, Hoffmann JA, Meister M. 1998. Analysis of the Drosophila host defense in domino mutant larvae, which are devoid of hemocytes. Proc. Natl. Acad. Sci. USA 95:14337–42 Mathey-Prevot B, Perrimon N. 1998. Mammalian and Drosophila blood: JAK of all trades? Cell 92:697–700 Dearolf CR. 1998. Fruit fly “leukemia.” Biochim. Biophys. Acta 1377:M13–23 Qiu P, Pan PC, Govind S. 1998. A role for the the Drosophila Toll/Cactus pathway in larval hematopoiesis. Development 125:1909–20 Lebestky T, Chang T, Hartenstein V, Banerjee U. 2000. Specification of Drosophila hematopoietic lineage by conserved transcription factors. Science 288: 146–49 Fossett N, Schulz RA. 2001. Functional conservation of hematopoietic factors in Drosophila and vertebrates. Differentiation 69:83–90 Duvic B, Hoffmann J, Meister M, Royet J. 2002. Notch signaling controls lineage specification during Drosophila lar-

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

55a.

P1: IKH

477

val hematopoiesis. Curr. Biol. 12:1923– 27 Cho NK, Keyes L, Johnson E, Heller J, Ryner L, et al. 2002. Developmental control of blood cell migration by the Drosophila VEGF pathway. Cell 108:865–76 Lebestky T, Jung SH, Banerjee U. 2003. A Serrate-expressing signaling center controls Drosophila hematopoiesis. Genes Dev. 17:348–53 Asha H, Nagy I, Kovacs G, Stetson D, Ando I, Dearolf CR. 2003. Analysis of ras-induced overproliferation in Drosophila hemocytes. Genetics 163: 203–15 S¨oderh¨all K, Cerenius L. 1998. Role of the prophenoloxidase-activating system in invertebrate immunity. Curr. Opin. Immunol. 10:23–28 Ferrandon D, Jung AC, Criqui M, Lemaitre B, Uttenweiler-Joseph S, et al. 1998. A drosomycin-GFP reporter transgene reveals a local immune response in Drosophila that is not dependent on the Toll pathway. EMBO J. 17:1217–27 ¨ Onfelt Tingvall T, Roos E, Engstr¨om Y. 2001. The imd gene is required for local Cecropin expression in Drosophila barrier epithelia. EMBO Rep. 2:239–43 Tzou P, Ohresser S, Ferrandon D, Capovilla M, Reichhart J-M, et al. 2000. Tissue-specific inducible expression of antimicrobial peptide genes in Drosophila surface epithelia. Immunity 13:737–48 Ganz T, Lehrer RI. 1998. Antimicrobial peptides of vertebrates. Curr. Opin. Immunol. 10:41–44 Huttner KM, Bevins C. 1999. Antimicrobial peptides as mediators of epithelial host defense. Pediatr. Res. 45:785–94 Irving P, Troxler L, Heuer TS, Belvin M, Kopczynski C, et al. 2001. A genomewide analysis of immune responses in Drosophila. Proc. Natl. Acad. Sci. USA 98:15119–24 Little TJ, O’Connor B, Colegrave N, Watt K, Read AF. 2003. Maternal transfer of

11 Feb 2004

19:23

478

56.

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

57.

58.

59.

60.

61.

62.

63.

64.

65.

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

strain-specific immunity in an invertebrate. Curr. Biol. 13:489–92 Ip YT, Reach M, Engstr¨om Y, Kadalayil L, Cai H, et al. 1993. Dif, a dorsal-related gene that mediates an immune response in Drosophila. Cell 75:753–63 Wu LP, Anderson KV. 1998. Regulated nuclear import of Rel proteins in the Drosophila immune response. Nature 392:93–97 St¨oven S, Ando I, Kadalayil L, Engstr¨om Y, Hultmark D. 2000. Activation of the Drosophila NF-κB factor Relish by rapid endoproteolytic cleavage. EMBO Rep. 1:347–52 De Gregorio E, Spellman PT, Tzou P, Rubin GM, Lemaitre B. 2002. The Toll and Imd pathways are the major regulators of the immune response in Drosophila. EMBO J. 21:2568–79 Meng X, Khanuja BS, Ip YT. 1999. Toll receptor-mediated Drosophila immune response requires Dif, an NF-κB factor. Genes Dev. 13:792–97 Rutschmann S, Jung AC, Hetru C, Reichhart J-M, Hoffmann JA, Ferrandon D. 2000. The Rel protein DIF mediates the antifungal but not the antibacterial host defense in Drosophila. Immunity 12:569– 80 Tauszig-Delamasure S, Bilak H, Capovilla M, Hoffmann JA, Imler J. 2002. Drosophila MyD88 is required for the response to fungal and Gram-positive bacterial infections. Nat. Immunol. 3:91–97 Rutschmann S, Kilinc A, Ferrandon D. 2002. Cutting edge: the Toll pathway is required for resistance to gram-positive bacterial infections in Drosophila. J. Immunol. 168:1542–46 Schneider DS, Hudson KL, Lin TY, Anderson KV. 1991. Dominant and recessive mutations define functional domains of Toll, a transmembrane protein required for dorsal-ventral polarity in the Drosophila embryo. Genes Dev. 5:797–807 Levashina EA, Langley E, Green C, Gubb D, Ashburner M, et al. 1999. Constitutive

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

activation of Toll-mediated antifungal defense in serpin-deficient Drosophila. Science 285:1917–19 Weber AN, Tauszig-Delamasure S, Hoffmann JA, Leli`evre E, Gascan H, et al. 2003. Binding of the Drosophila cytokine Sp¨atzle to Toll is direct and establishes signaling. Nat. Immunol. 4:794–800 Belvin MP, Anderson KV. 1996. A conserved signaling pathway: the Drosophila Toll-Dorsal pathway. Annu. Rev. Cell Dev. Biol. 12:393–416 Nicolas E, Reichhart J-M, Hoffmann JA, Lemaitre B. 1998. In vivo regulation of the IκB homologue cactus during the immune response of Drosophila. J. Biol. Chem. 273:10463–69 Imler JL, Hoffmann J. 2002. Toll receptors in Drosophila: a family of molecules regulating development and immunity. Curr. Top Microbiol. Immunol. 270:63–79 Morisato D, Anderson KV. 1995. Signaling pathways that establish the dorsalventral pattern of the Drosophila embryo. Annu. Rev. Genet. 29:371–99 De Gregorio E, Spellman PT, Rubin GM, Lemaitre B. 2001. Genome-wide analysis of the Drosophila immune response by using oligonucleotide microarrays. Proc. Natl. Acad. Sci. USA 98:12590–95 Williams MJ, Rodriguez A, Kimbrell DA, Eldon ED. 1997. The 18-wheeler mutation reveals complex antibacterial gene regulation in Drosophila host defense. EMBO J. 16:6120–30 Ligoxygakis P, Bulet P, Reichhart J-M. 2002. Critical evaluation of the role of the Toll-like receptor 18-Wheeler in the host defense of Drosophila. EMBO Rep. 3:666–73 Seppo A, Matani P, Sharrow M, Tiemeyer M. 2003. Induction of neuronspecific glycosylation by Tollo/Toll-8, a Drosophila Toll-like receptor expressed in non-neural cells. Development 130: 1439–48 Eldon E, Kooyer S, D’Evelyn D, Duman M, Lawinger P, et al. 1994. The

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

DROSOPHILA IMMUNITY

76.

77.

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

78.

79.

80.

81.

82.

83.

84.

85.

86.

Drosophila 18 wheeler is required for morphogenesis and has striking similarities to Toll. Development 120:885–99 Kimbrell DA, Beutler B. 2001. The evolution and genetics of innate immunity. Nat. Rev. Genet. 2:256–67 Hultmark D. 2003. Drosophila immunity: paths and patterns. Curr. Opin. Immunol. 15:12–19 Hoffmann JA, Reichhart J-M. 2002. Drosophila innate immunity: an evolutionary perspective. Nat. Immunol. 3:121– 26 Khush RS, Leulier F, Lemaitre B. 2001. Drosophila immunity: two paths to NFκB. Trends Immunol. 22:260–64 Lemaitre B, Kromer-Metzger E, Michaut L, Nicolas E, Meister M, et al. 1995. A recessive mutation, immune deficiency (imd), defines two distinct control pathways in the Drosophila host defense. Proc. Natl. Acad. Sci. USA 92:9465–69 Wu LP, Choe KM, Lu Y, Anderson KV. 2001. Drosophila immunity: Genes on the third chromosome required for the response to bacterial infection. Genetics 159:189–99 Leulier F, Rodriguez A, Khush RS, Abrams JM, Lemaitre B. 2000. The Drosophila caspase Dredd is required to resist gram-negative bacterial infection. EMBO Rep. 1:353–58 Rutschmann S, Jung AC, Rui Z, Silverman N, Hoffmann JA, Ferrandon D. 2000. Role of Drosophila IKKγ in a Toll-independent antibacterial immune response. Nat. Immunol. 1:342–47 ˚ Hedengren M, Asling B, Dushay MS, Ando I, Ekengren S, et al. 1999. Relish, a central factor in the control of humoral, but not cellular immunity in Drosophila. Mol. Cell 4:827–37 St¨oven S, Silverman N, Junell A, Hedengren-Olcott M, Erturk D, et al. 2003. Caspase-mediated processing of the Drosophila NF-κB factor Relish. Proc. Natl. Acad. Sci. USA 100:5991–96 Silverman N, Zhou R, St¨oven S, Pandey

87.

88.

89.

90.

91.

92.

93.

94.

95.

P1: IKH

479

N, Hultmark D, Maniatis T. 2000. A Drosophila IkappaB kinase complex required for Relish cleavage and antibacterial immunity. Genes Dev. 14:2461–71 Vidal S, Khush RS, Leulier F, Tzou P, Nakamura M, Lemaitre B. 2001. Mutations in the Drosophila dTAK1 gene reveal a conserved function for MAPKKKs in the control of rel/NF-kappaB-dependent innate immune responses. Genes Dev. 15:1900–12 Naitza S, Rosse C, Kappler C, Georgel P, Belvin M, et al. 2002. The Drosophila immune defense against gram-negative infection requires the death protein dFADD. Immunity 17:575–81 Leulier F, Vidal S, Saigo K, Ueda R, Lemaitre B. 2002. Inducible expression of double-stranded RNA reveals a role for dFADD in the regulation of the antibacterial response in Drosophila adults. Curr. Biol. 12:996–1000 Boutros M, Agaisse H, Perrimon N. 2002. Sequential activation of signaling pathways during innate immune responses in Drosophila. Dev. Cell 3:711–22 Georgel P, Naitza S, Kappler C, Ferrandon D, Zachary D, et al. 2001. Drosophila Immune Deficiency (IMD) is a death domain protein that activates antibacterial defense and can promote apoptosis. Dev. Cell 1:503–14 Gottar M, Gobert V, Michel T, Belvin M, Duyk G, et al. 2002. The Drosophila immune response against Gram-negative bacteria is mediated by a peptidoglycan recognition protein. Nature 416:640–44 Khush RS, Cornwell WD, Uram JN, Lemaitre B. 2002. A ubiquitin-proteasome pathway represses the Drosophila immune deficiency signaling cascade. Curr. Biol. 12:1728–37 Ihle J. 2001. The Stat family in cytokine signaling. Curr. Opin. Cell Biol. 13:211– 17 Dong C, Davis RJ, Flavell R. 2002. MAP kinases in the immune response. Annu. Rev. Immunol. 20:55–72

11 Feb 2004

19:23

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

480

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

96. Sluss HK, Han Z, Barrett T, Davis RJ, Ip YT. 1996. A JNK signal transduction pathway that mediates morphogenesis and an immune response in Drosophila. Genes Dev. 10:2745–58 97. Noselli S, Agnes F. 1999. Roles of the JNK signaling pathway in Drosophila morphogenesis. Curr. Opin. Genet. Dev. 9:466–72 98. Agaisse H, Petersen U-M, Boutros M, Mathey-Prevot B, Perrimon N. 2003. Signaling role of hemocytes in Drosophila JAK/STAT-dependent response to septic injury. Dev. Cell 5:441–50 99. Lagueux M, Perrodou E, Levashina EA, Capovilla M, Hoffmann JA. 2000. Constitutive expression of a complement-like protein in Toll and JAK gain-of-function mutants of Drosophila. Proc. Natl. Acad. Sci. USA 97:11427–32 100. Yoshida H, Kinoshita K, Ashida M. 1996. Purification of a peptidoglycan recognition protein from hemolymph of the silkworm, Bombyx mori. J. Biol. Chem. 271:13854–60 101. Kang D, Liu G, Lundstr¨om A, Gelius E, Steiner H. 1998. A peptidoglycan recognition protein in innate immunity conserved from insects to humans. Proc. Natl. Acad. Sci. USA 95:10078–82 102. Werner T, Liu G, Kang D, Ekengren S, Steiner H, Hultmark D. 2000. A family of peptidoglycan recognition proteins in the fruit fly Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 97:13772–77 103. Werner T, Borge-Renberg K, Mellroth P, Steiner H, Hultmark D. 2003. Functional diversity of the Drosophila PGRPLC gene cluster in the response to LPS and peptidoglycan. J. Biol. Chem. 278:26319– 22 104. Liu C, Xu ZJ, Gupta D, Dziarski R. 2001. Peptidoglycan recognition proteins: a novel family of four human innate immunity pattern recognition molecules. J. Biol. Chem. 276:34686–94 105. Michel T, Reichhart J-M, Hoffmann JA, Royet J. 2001. Drosophila Toll is activated

106.

107.

108.

109.

110.

111.

112.

113.

114.

by Gram-positive bacteria through a circulating peptidoglycan recognition protein. Nature 414:756–59 Choe KM, Werner T, St¨oven S, Hultmark D, Anderson K. 2002. Requirement for a peptidoglycan recognition protein (PGRP) in relish activation antibacterial immune responses in Drosophila. Science 296:359–62 R¨amet M, Manfruelli P, Pearson A, Mathey-Prevot B, Ezekowitz RAB. 2002. Functional genomic analysis of phagocytosis and identification of a Drosophila receptor for E. coli. Nature 416:644– 48 Mellroth P, Karlsson J, Steiner H. 2003. A scavenger function for a Drosophila peptidoglycan recognition protein. J. Biol. Chem. 278:7059–64 Kim MS, Byun M, Oh B. 2003. Crystal structure of peptidoglycan recognition protein LB from Drosophila melanogaster. Nat. Immunol. 4:787–93 Horn DL, Morrison DC, Opal SM, Silverstein R, Visvanathan K, Zabriskie JB. 2000. What are the microbial components implicated in the pathogenesis of sepsis? Report on a symposium. Clin. Infect. Dis. 31:851–58 Schleifer KH, Kandler O. 1972. Peptidoglycan types of bacterial cell walls and their taxonomic implications. Bacteriol. Rev. 36:407–77 Leulier F, Parquet C, Pili-Floury S, Ryu JH, Caroff M, et al. 2003. The Drosophila immune system detects bacteria through specific peptidoglycan recognition. Nat. Immunol. 4:478–84 Kaneko T, Goldman W, Mellroth P, Steiner H, Kosma P, et al. 2003. Gramnegative bacterial recognition: Do flies recognize LPS? Presented at Annu. Drosophila Res. Conf., 44th, Chicago Dziarski R, Platt KA, Gelius E, Steiner H, Gupta D. 2003. Defect in neutrophil killing and increased susceptibility to infection with nonpathogenic gram-positive bacteria in peptidoglycan

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

DROSOPHILA IMMUNITY

115.

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

116.

117.

118.

119.

120.

121.

122.

recognition protein-S (PGRP-S)-deficient mice. Blood 102:689–97 ˚ Samakovlis C, Asling B, Boman HG, Gateff E, Hultmark D. 1992. In vitro induction of cecropin genes—an immune response in a Drosophila blood cell line. Biochem. Biophys. Res. Commun. 188:1169–75 Kim YS, Ryu JH, Han SJ, Choi KH, Nam KB, et al. 2000. Gram-negative bacteria binding protein, a pattern recognition receptor for lipopolysaccharide and β-1,3glucan, which mediates the signaling for the induction of innate immune genes in Drosophila melanogaster cells. J. Biol. Chem. 275:32721–27 Takehana A, Katsuyama T, Yano T, Oshima Y, Takada H, et al. 2002. Overexpression of a pattern-recognition receptor, peptidoglycan-recognition protein-LE, activates imd/relish-mediated antibacterial defense and the prophenoloxidase cascade in Drosophila larvae. Proc. Natl. Acad. Sci. USA 99:13705–10 R¨amet M, Pearson A, Manfruelli P, Li XH, Koziel H, et al. 2001. Drosophila scavenger receptor CI is a pattern recognition receptor for bacteria. Immunity 15:1027– 38 Lee WJ, Lee JD, Kravchenko VV, Ulevitch RJ, Brey PT. 1996. Purification and molecular cloning of an inducible gramnegative bacteria-binding protein from the silkworm, Bombyx mori. Proc. Natl. Acad. Sci. USA 93:7888–93 Levashina EA, Moita LF, Blandin S, Vriend G, Lagueux M, Kafatos FC. 2001. Conserved role of a complement-like protein in phagocytosis revealed by dsRNA knockout in cultured cells of the mosquito, Anopheles gambiae. Cell 104:709–18 Lee SY, S¨oderh¨all K. 2001. Characterization of a pattern recognition protein, a masquerade-like protein, in the freshwater crayfish Pacifastacus leniusculus. J. Immunol. 166:7319–26 Luo C, Shen B, Manley JL, Zheng L. 2001. Tehao functions in the Toll path-

123.

124.

125.

126.

127.

128.

129.

130.

131.

P1: IKH

481

way in Drosophila melanogaster: possible roles in development and innate immunity. Insect Mol. Biol. 10:457–64 Tauszig S, Jouanguy E, Hoffmann JA, Imler JL. 2000. Toll-related receptors and the control of antimicrobial peptide expression in Drosophila. Proc. Natl. Acad. Sci. USA 97:10520–25 Ooi JY, Yagi Y, Hu XD, Ip YT. 2002. The Drosophila Toll-9 activates a constitutive antimicrobial defense. EMBO Rep. 3:82– 87 Nappi AJ, Vass E, Frey F, Carton Y. 2000. Nitric oxide involvement in Drosophila immunity. Nitric Oxide 4:423–30 LeMosy EK, Hong CC, Hashimoto C. 1999. Signal transduction by a protease cascade. Trends Cell Biol. 9:102–7 Ligoxygakis P, Pelte N, Hoffmann JA, Reichhart J-M. 2002. Activation of Drosophila Toll during fungal infection by a blood serine protease. Science 297:114–16 Robertson AS, Belorgey D, Lilley KS, Lomas DA, Gubb D, Dafforn T. 2003. Characterization of the Necrotic protein that regulates the Toll-mediated immune response in Drosophila. J. Biol. Chem. 278:6175–80 Lau GW, Goumnerov BC, Walendziewicz CL, Hewitson J, Xiao W, et al. 2003. The Drosophila melanogaster Toll pathway participates in resistance to infection by the gram-negative human pathogen Pseudomonas aeruginosa. Infect. Immun. 71:4059–66 Ashida M, Brey PT. 1998. Recent advances in research in the insect prophenoloxidase cascade. In Molecular Mechanisms of Immune Response in Insects, ed. PT Brey, D Hultmark, pp. 135–72. London: Chapman & Hall Yoshida H, Ochiai M, Ashida M. 1986. β-1,3-glucan receptor and peptidoglycan receptor are present as separate entities within insect prophenoloxidase activating system. Biochem. Biophys. Res. Commun. 141:1177

11 Feb 2004

19:23

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

482

AR

AR210-IY22-15.tex

BRENNAN

¥

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

P1: IKH

ANDERSON

132. Asp´an A, Huang TS, Cerenius L, S¨oderh¨all K. 1995. cDNA cloning of prophenoloxidase from the freshwater crayfish Pacifastacus leniusculus and its activation. Proc. Natl. Acad. Sci. USA 92: 939–43 133. De Gregorio E, Han SJ, Lee WJ, Baek MJ, Osaki T, et al. 2002. An immuneresponsive Serpin regulates the melanization cascade in Drosophila. Dev. Cell 3: 581–92 134. Ligoxygakis P, Pelte N, Ji C, Leclerc V, Duvic B, et al. 2002. A serpin mutant links Toll activation to melanization in the host defence of Drosophila. EMBO J. 21:6330–37 135. Underhill DM, Ozinsky A. 2002. Phagocytosis of microbes: complexity in action. Annu. Rev. Immunol. 20:825–52 136. Vass E, Nappi AJ. 2000. Developmental and immunological aspects of Drosophila-parasitoid relationships. J. Parasitol. 86:1259–70 137. Luo H, Hanratty WP, Dearolf CR. 1995. An amino acid substitution in the Drosophila hopTum-l Jak kinase causes leukemia-like hematopoietic defects. EMBO J. 14:1412–20 138. Harrison DA, Binari R, Nahreini TS, Gilman M, Perrimon N. 1995. Activation of a Drosophila Janus kinase (JAK) causes hematopoietic neoplasia and developmental defects. EMBO J. 14:2857– 65 139. Russo J, Dupas S, Frey F, Carton Y, Breh´elin M. 1996. Insect immunity: early events in the encapsulation process of parasitoid (Leptopilina boulardi) eggs in resistant and susceptible strains of Drosophila. Parasitology 112:135–42 140. Hogan LH, Weinstock JV, Sandor M. 1999. TCR specificity in infection– induced granulomas. Immunol. Lett. 68: 115–20 141. Rizki RM, Rizki TM. 1980. Hemocyte responses to implanted tissues in Drosophila melanogaster larvae. Wilhelm Roux Arch. Dev. Biol. 189:207–13

142. Watson KL, Johnson TK, Denell RE. 1991. Lethal(1) aberrant immune response mutations leading to melanotic tumor formation in Drosophila melanogaster. Dev. Genet. 12:173–87 143. Murray MA, Fessler LI, Palka J. 1995. Changing distributions of extracellular matrix components during early wing morphogenesis in Drosophila. Dev. Biol. 168:150–65 144. Lavine MD, Strand MR. 2001. Surface characteristics of foreign targets that elicit an encapsulation response by the moth Pseudoplusia includens. J. Insect Physiol. 47:965–74 145. Song ZW, McCall K, Steller H. 1997. DCP-1, a Drosophila cell death protease essential for development. Science 275:536–40 146. Rodriguez A, Oliver H, Zou H, Chen P, Wang XD, Abrams JM. 1999. Dark is a Drosophila homologue of Apaf-1/CED4 and functions in an evolutionarily conserved death pathway. Nat. Cell Biol. 1:272–79 147. Franc NC, Heitzler P, Ezekowitz RA, White K. 1999. Requirement for Croquemort in phagocytosis of apoptotic cells in Drosophila. Science 284:1991–94 148. Mukae N, Yokoyama H, Yokokura T, Sakoyama Y, Nagata S. 2002. Activation of the innate immunity in Drosophila by endogenous chromosomal DNA that escaped apoptotic degradation. Genes Dev. 16:2662–71 149. Napirei M, Karsunky H, Zevnik B, Stephan H, Mannherz HG, Moroy T. 2000. Features of systemic lupus erythematosus in Dnase1-deficient mice. Nat. Genet. 25:177–81 150. Rahme LG, Ausubel FM, Cao H, Drenkard E, Goumnerov BC, et al. 2000. Plants and animals share functionally common bacterial virulence factors. Proc. Natl. Acad. Sci. USA 97:8815–21 151. Fauvarque MO, Bergeret E, Chabert J, Dacheux D, Satre M, Attree I. 2002. Role and activation of type III secretion

11 Feb 2004

19:23

AR

AR210-IY22-15.tex

AR210-IY22-15.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DROSOPHILA IMMUNITY system genes in Pseudomonas aeruginosa-induced Drosophila killing. Microb. Pathog. 32:287–95 152. D’Argenio DA, Gallagher LA, Berg CA, Manoil C. 2001. Drosophila as a model host for Pseudomonas aeruginosa infection. J. Bacteriol. 183:1466–71 153. Kurz CL, Chauvet S, Andres E, Aurouze M, Vallet I, et al. 2003. Virulence factors of the human opportunistic pathogen Serratia marcescens identified by in vivo screening. EMBO J. 22:1451–60 154. Basset A, Tzou P, Lemaitre B, Boccard

P1: IKH

483

F. 2003. A single gene that promotes interaction of a phytopathogenic bacterium with its insect vector, Drosophila melanogaster. EMBO Rep. 4:205–9 155. Schneider D, Shahabuddin M. 2000. Malaria parasite development in a Drosophila model. Science 288:2376– 79 156. Dimopoulos G, Muller HM, Levashina EA, Kafatos F. 2001. Innate immune defense against malaria infection in the mosquito. Curr. Opin. Immunol. 13:79– 88

Brennan.qxd

3/12/2004

2:02 PM

Page 1

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DROSOPHILA IMMUNITY

C-1

Figure 3 Drosophila blood cells in normal and autoimmune conditions. (A) Two plasmatocytes stained with phalloidin, showing filamentous actin. (B) Phase contrast images of two blood cells from a larva with a gain-of-function allele of Toll, Toll10b. Constitutive Toll signaling causes the differentiation of lamellocytes, the large flat cells. The arrow indicates a plasmatocyte. (C) The blood cells of a Toll10b larva carrying an enhancer trap that expresses lacZ in lamellocytes. Here the lamellocytes are beginning to encapsulate self-tissue, but melanization has not yet begun. Scale bars in (A), (B), and (C) are all 10 µm. Melanotic capsules (arrows) are visible through the cuticle of Toll10b larvae (D), and adults (E) and (F).

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:457-483. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

19:20

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104707

Annu. Rev. Immunol. 2004. 22:485–501 doi: 10.1146/annurev.immunol.22.012703.104707 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on November 10, 2003

RAGs AND REGULATION OF AUTOANTIBODIES Mila Jankovic,1 Rafael Casellas,2 Nikos Yannoutsos,1 Hedda Wardemann,1 and Michel C. Nussenzweig1,3 Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

1

Laboratory of Molecular Immunology, 3Howard Hughes Medical Institute, The Rockefeller University, New York, New York 10021; email: [email protected]; [email protected]; [email protected]; [email protected] 2 Division of Biology, California Institute of Technology, Pasadena, California 91125; email: [email protected]

Key Words B cell development, tolerance, receptor editing, RAG1 and RAG2 expression, antiself antibodies ■ Abstract Autoreactive antibodies are etiologic agents in a number of autoimmune diseases. Like all other antibodies these antibodies are produced in developing B cells by V(D)J recombination in the bone marrow. Three mechanisms regulate autoreactive B cells: deletion, receptor editing, and anergy. Here we review the prevalence of autoantibodies in the initial antibody repertoire, their regulation by receptor editing, and the role of the recombinase proteins (RAG1 and RAG2) in this process.

INTRODUCTION Antigen receptor diversity is generated by random rearrangements between variable (V) diversity (D) and joining (J) gene segments in developing B cells and T cells (1). This receptor diversity ensures efficient immune responses to a universe of rapidly evolving potential pathogens. However, random gene reassortment can also produce self-reactive antibodies that could cause “horror autotoxicus,” as pointed out by Ehrlich (2). Burnett, Talmage, and Lederberg provided a framework for understanding how self-reactive lymphocytes might be removed from the nascent repertoire as part of the clonal selection theory (3–5): They proposed that any self-reactive antibody producing lymphocytes would be deleted during development. In the past 20 years transgenic experiments in mice have shown that at least three mechanisms account for silencing of self-reactive antibodies during B cell development in the bone marrow: receptor editing, deletion, and anergy (6–9). One of these mechanisms, receptor editing, relies on V(D)J recombination to obviate autoantibody production in vivo. Here we review recent work on aspects of V(D)J recombination relevant to production and regulation of autoreactive antibodies in vivo. 0732-0582/04/0423-0485$14.00

485

15 Mar 2004

15:28

486

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

P1: IKH

JANKOVIC ET AL.

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MOST NASCENT B CELLS PRODUCE AUTOANTIBODIES How B cells produce a diverse antibody repertoire while maintaining self-tolerance has been an important question since the time of Ehrlich (2). Landsteiner’s finding that the antibody repertoire is almost infinitely diverse further stressed the potential importance of avoiding self-reactivity and raised the problem of prevalence of autoantibodies in the initial repertoire (10). A number of investigators found that the majority of nascent immature B cells are deleted before they reach the mature B cell compartment, suggesting that autoreactive B cells might be a large part of the initial repertoire (11–15). Only 10%–20% of the bone marrow immature cells reach the periphery, and as few as 3% contribute to the mature B cell pool (14, 16–18). Furthermore, deletion was not random because there was a specific shift in the representation of antibodies in the repertoire between the immature and mature B cell stages (19–21). The number of self-reactive antibodies was determined for human B cells by single cell analysis of the bone marrow and blood (22) (Figure 1). Two types of autoreactive antibodies were measured by ELISA, antinuclear antibodies (ANAs) and antibodies to defined antigens such as DNA, insulin and LPS. Antibodies that bound to more than one of these defined antigens were considered polyreactive. The ANAs were assayed by ELISA on HEp-2 cell lysates and verified by indirect immunofluorescence. Fifty-five percent to 75% of all antibodies in the early immature B cell compartment (B cells with the surface phenotype of pre-B cells that contained in-frame Igκ or Igλ chains) produced self-reactive antibodies (22). The self-reactive antibodies were lost from the B cell repertoire at two discrete checkpoints in development. The first checkpoint was between the early immature B cell stage and the immature B cell stage. Nearly all antibodies (poly)reactive with defined antigens were lost in this early transition in the bone marrow. ANA antibodies were lost from the repertoire in the immature B cell compartment and also in the periphery at the transition between the immature and mature B cell pool. Therefore, the majority of nascent human B cells is self-reactive. Future experimentation should focus on understanding changes in the efficiency of autoantibody removal that might lead to increased susceptibility to autoimmunity.

RAG EXPRESSION AND ANTIBODY ASSEMBLY IN DEVELOPING B CELLS RAG1 and RAG2 proteins catalyze V(D)J recombination and antibody gene assembly (23, 24). RAGs bind to and cleave DNA at conserved recombination signal sequences (RS) that flank immunoglobulin (Ig) and T cell receptor (TCR) gene segments (25). These proteins are essential for the recombination reaction, and loss of either one in mice or humans results in complete block of lymphocyte development (26–28). RAG1 and RAG2 are only coexpressed in developing lymphocytes, and the level of expression is regulated in a developmental stage–specific manner (29, 30)

11 Feb 2004

19:20

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

REGULATION OF AUTOANTIBODIES

P1: IKH

487

(see Figure 1). The first B cells to express RAG1 and RAG2 are early lymphoid committed progenitors found in fraction A1 (31, 32). These cells are Lin−CD27+ckitHiScaHi and express low levels of RAGs and exhibit low levels of recombinase activity as measured by recombination of antibody D and J gene segments (33). RAGs are further induced in pro-B cells where Ig heavy chain gene assembly begins (31, 32, 34). Successful heavy chain gene recombination leads to expression of membrane Ig heavy chains (µ), which in turn induce clonal expansion, RAG downregulation, heavy chain allelic exclusion, and progression to the pre-B cell stage (35–40). It is essential for heavy chain allelic exclusion to be established at this stage of B cell development because persistent recombination might inactivate the productively rearranged heavy chain or produce a cell with two heavy chains and multiple antibody specificities. RAG1 and RAG2 downregulation is likely an important initial aspect of establishing allelic exclusion at the heavy chain locus. However, exclusion is maintained in later B cell stages by RAG-independent changes in heavy chain accessibility that appear to be induced through IL-7 (41, 42). RAG1 and RAG2 expression is downregulated in cycling pre-B cells but it is not known whether this downregulation is mediated at the transcriptional level or at the level of mRNA stability (43). Decreased transcription is suggested by decreased GFP expression in mice that carry YFP and GFP inserted into the RAG1 or RAG2 genes, respectively (32, 44). RAG protein levels are also regulated by phophorylation of the RAG protein, which decreases the half-life of the protein (45). A second wave of RAG expression is induced when pre-B cells stop dividing, and it is associated with onset of high levels of Ig light chain gene recombination (35, 46). Upon successful expression of an Ig light chain gene and assembly of a functional BCR, pre-B cells become immature B cells. Immature B cells are the first B cells to express surface IgM, and it is in this compartment that antibodies are initially tested for self-reactivity and that central B cell tolerance is established by receptor editing or deletion (6, 7). Immature B cells display two features that facilitate induction of central tolerance: persistent RAG expression that allows for receptor editing, and apoptosis in response to high levels of BCR cross-linking that promotes deletion (32, 44, 47, 48).

PERSISTENT RAG EXPRESSION AND RECEPTOR EDITING Receptor editing involves replacement of self-reactive antibody V region (7, 49) genes by V(D)J recombination (6). Gene replacement by persistent V(D)J recombination was first described in B cell lines in vitro (50–52). In mice, it was noted that Igκ transgenes did not induce high levels of allelic exclusion and that there was frequent persistent V(D)J recombination with some antibody transgenes (53, 54). However, the potential physiologic significance of secondary recombination of Ig genes in maintaining tolerance was only appreciated from studies of transgenic mice carrying anti–double stranded DNA or anti-MHC antibodies (6, 7). In these mice, immature B cells are exposed to multivalent self-antigens at early stages

11 Feb 2004

19:20

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

488

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

P1: IKH

JANKOVIC ET AL.

of development (8, 55). Most of these B cells are deleted, but a small cohort of B cells escapes deletion and migrates to the spleen and lymph nodes where they accumulate with age (6). It was shown that while the transgenic heavy chain was present in the B cells that escape deletion, they expressed an endogenous, nontransgenic κ or λ gene, and the resulting chimera was no longer self-reactive (6, 7, 49). Thus, the nontransgenic κ or λ gene was said to “edit” the original transgenic self-reactivity. Further insight into light chain editing came from the analysis of the transgenic mice that carry only the 3H9 heavy chain of an anti-DNA antibody (49, 56). 3H9 produces anti-DNA when combined with most light chains but a small group of light chains with low isoelectric points such as Vκ12A block anti-DNA reactivity. Despite the high level of reactivity of 3H9 with DNA, splenic B cells from these mice did not produce anti-DNA antibodies. 3H9 peripheral B cells showed highly skewed Vκ and Jκ usage with a large number of cells expressing Vκ12A, and the majority of the Vκ genes were joined to Jκ5, indicating that multiple recombination events had occurred on the same chromosome (49). Efficient light chain “editors” differed from other κ light chains in that their CDRs are very acidic and have the capacity to prevent DNA binding by neutralizing the arginines in the CDR3 of 3H9 (57). Enhanced editing was associated with high levels of RAG expression, suggesting that editing induced or prolonged RAG expression (7). The efficiency of editing was assessed using mice produced by gene targeting to express anti-MHC (H-2Kk-b) or anti-DNA heavy and light chains (58, 59). In contrast to the severely reduced numbers of mature B cells reported in conventional anti-MHC transgenic mice, the bone marrow and peripheral B cell compartments of these knock-in mice were nearly normal, suggesting that editing was highly efficient (58, 59). IgVH genes can also be replaced by secondary V(D)J recombination through the use of cryptic recombination signal sequences embedded in the VH gene (52, 60–62). Studies in mice and humans have shown evidence for VH replacement in autoantibody-producing B cells (63–65). However, edited heavy chains frequently show N nucleotide addition, suggesting that this process can occur in progenitor B cells before light chain assembly (63, 66, 67). Such replacements would not be a response to self-antigen binding. Nevertheless, it has been estimated that 5%–12% of the human antibody repertoire is produced by VH replacement (68).

SITE OF EDITING RAG expression was believed to be upregulated for a third time in immature B cells that carry self-reactive receptors. In vitro experiments with both transgenic and wild-type B cells showed increased RAG mRNA levels in cultures of developing B cells following BCR cross-linking (69, 70). In addition bone marrow cells from mice that carry anti-MHC (H-2Kk-b) antibody transgenes show higher levels of RAG expression in the self-reactive background (71). Whether this increase in RAG

11 Feb 2004

19:20

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

REGULATION OF AUTOANTIBODIES

P1: IKH

489

expression is due to increased transcription or selection for cells that express high levels of RAG remains to be determined. Immature B cells are not a homogeneous population with respect to the level of RAG or BCR expression (Figure 1). Analysis of the RAG2-GFP indicator mice showed that RAG levels were inversely correlated with the density of the surface B cell receptor and B cell maturation (44, 72). More mature B cells with higher levels of cell surface BCR expression showed lower levels of RAG expression, suggesting that RAG expression is gradually turned off as B cells mature (44, 72). Immature B cells that express high levels of surface IgM are also more susceptible to negative selection (47, 48). Thus, the apparent upregulation of RAG mRNA by BCR cross-linking in vitro could also be a consequence of the selective survival of the cells that expressed higher levels of RAGs and lower levels of BCR to begin with. Similarly, increased RAG expression in transgenic B cells altered B cell maturation and enrichment for less mature B cells that have lower levels of surface BCR and higher RAG expression. To visualize the editing compartment in vivo, an Igκ allelic polymorphism was created by targeted replacement of the mouse constant region gene (mCκ) by its human homologue (hCκ) (73). In mice heterozygous for both a prerecombined light chain (at the mCκ locus) and the hCκ allele, edited B cells were distinguished as those cells that lost expression of the prerecombined mCκ allele (73). Kinetic studies showed that B lymphocytes undergoing editing were arrested for at least two hours at the pre-BII cell stage when compared to unedited counterparts. The delayed and edited B cell population showed higher levels of RAG expression than unedited cells (73). Similar kinetic data have been obtained using mice that carry a prerecombined Vλ gene instead of hCκ (74). These observations are consistent with a model where self-recognition activates secondary gene recombination by inducing a developmental arrest at a stage where RAG1 and RAG2 are expressed. In this model, cross-linking of the cell surface receptor by self-antigen does not reinduce but sustains RAG1 and RAG2 expression and allows for secondary V(D)J recombination. The model further clarifies how light chain allelic exclusion is maintained in the presence of high levels of editing: B cells expressing innocuous receptors transit rapidly from this stage to a compartment where RAG expression and V(D)J recombination are turned off. Thus, B cells with a nonself-reactive receptor are not exposed to prolonged V(D)J recombination, and the self-reactive cells that are trapped in the editing compartment remain so only until they produce a nonself-reactive receptor. In this regulated model for allelic exclusion, both Igκ alleles could be accessible to the V(D)J recombinase simultaneously without interfering with allelic exclusion.

EXTENT OF EDITING Several lines of evidence suggest that secondary gene rearrangements are a common occurrence. For example, analysis of hybridomas generated from Vκ4Jκ4 and Vκ8Jκ5 knock-in mice shows that both prerearranged light chains are replaced by

11 Feb 2004

19:20

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

490

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

P1: IKH

JANKOVIC ET AL.

editing in a large number of B lymphocytes (75). Likewise, the Vκ3-83Jκ2 light chain is edited by secondary recombination occurring at the targeted allele and this is accompanied by V(D)J recombination at the wild-type allele (58). Editing has also been examined in Igλ expressing nontransgenic B cells in mice and man (76, 77). Forty-seven percent of all mouse Igλ expressing B cells that have had their Igκ genes inactivated by RS rearrangements show in-frame and potentially functional Igκ genes (77). In man, almost all λ expressing B cells also carry a rearranged κ locus, whereas only 2% of κ expressing B cells carry rearrangement Ig λ genes (76). Furthermore, inactivated in-frame κ joints are found in 30% of human λ expressing cells (76). However, it was not determined why the in-frame Igκ alleles were replaced in mouse or man. A direct measurement of the extent of receptor editing was made using the hCκ and Vλ knock-in alleles (73, 74). These studies revealed that on average 25% of the repertoire is the result of secondary gene replacements, indicating that editing makes an important contribution to the normal antibody repertoire.

TRANSCRIPTION AND SECONDARY V(D)J RECOMBINATION Ig gene expression has been proposed to be essential for primary V(D)J recombination based on the direct correlation between the onset of germline transcription and antibody gene recombination (78–80). In support of this model, targeted deletion of transcription regulatory elements and some transcription factors impairs both expression and recombination of TCR and Ig genes in vivo (81–85). In addition, V(D)J recombination can only be induced in nonlymphoid cells when Ig genes are transcribed (86). However, experiments with cell lines, transgenic recombination substrates, and Pax5−/− B cells show that transcription does not always correlate with recombinational accessibility (87–91). Furthermore, the question of whether there is a role of transcription per se as opposed to indirect enhancement of locus accessibility has not been resolved. The role of transcription in secondary recombination was investigated in mice mutant for the transcriptional coactivator OcaB. OcaB (also known as Bob-1 and OBF-1) (92–94) forms a ternary complex with the transcription factor Oct-1 (or Oct-2) (95) on octamer motifs present in some Ig gene promoters. This octamer/ Oct-1(2)/OcaB ternary complex is thought to activate Ig gene transcription by specific interactions with the TATA-box-associated basal transcription machinery (96, 97). A direct correlation was found between light chain promoter transcription and receptor editing (98). In the absence of OcaB, the Igκ chromatin domain is accessible as measured by DNA demethylation, histone acetylation, promoter loading of Oct-1, and low level transcription (98). However, open chromatin and low level transcription were not sufficient for gene replacement by V(D)J recombination. Only strongly transcribed promoters were replaced (98). Thus, a specific threshold of Vκ gene transcription is required for secondary gene recombination.

11 Feb 2004

19:20

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

REGULATION OF AUTOANTIBODIES

P1: IKH

491

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TERMINATION OF RAG EXPRESSION RAG expression was initially thought to be confined to pro-B, pre-B, and immature B cells in the bone marrow. However, several reports showed RAG mRNA and protein expression in germinal centers and in vitro activated B cells (99, 100). RAG1 and RAG2 mRNAs were detected in IgD+ B cells cultured in the presence of CD40 (or lipopolysaccharide) and IL4, whereas splenic, Peyer’s patches, or lymph node germinal center B cells showed abundant RAG1 protein expression as determined by immunostaining (99, 100). Unlike most peripheral B cells, RAG+ lymphocytes (in both mice and humans) expressed several cell surface markers characteristic of pro-B or pre-B cells such as λ5, VpreB, TdT, GL7, low levels of B220, and IL7R (99, 101–105). Ongoing recombination in these peripheral cells was implied by the presence of DNA double-stranded break V(D)J intermediates (106–108) and de novo formation of Vλ1-Jλ1 signal joints. However, the similarities between bone marrow and peripheral RAG+ B cells raised the possibility that these cells might in fact represent B cell precursors. Experiments with transgenic and GFP knock-in mice supported this hypothesis (44, 72). In both cases, GFP expression in developing B and T cells mimics endogenous RAGs with the caveat that the GFP expression level in the transgenic mice was one order of magnitude higher than in the knock-in mice. Despite this, the results obtained with both models were similar in that they showed that RAG-GFP expression in spleens was restricted to immature new immigrant B cells (44, 72). Furthermore, adoptive transfer experiments showed that mature B cells could not be induced to re-express RAGs during an immune response in vivo or by culturing them in the presence of LPS and IL4 (44, 72). Interestingly, the RAG+ immature B cell compartment in the spleen was increased in animals immunized with antigen, infectious agents, or adjuvant only at late stages of the immune response (44, 72, 109, 110). Increase in production and export of immature B cells from the bone marrow (109, 110) accounted for accumulation of these RAG+ immigrants with kinetics similar to germinal center formation (99, 100). The levels of RAG expression in developing B cells decreased as they acquired higher levels of surface IgM expression (see above). New immigrant B cells in the periphery express substantially lower levels of RAG than immature B cells in the bone marrow (44, 72). Nevertheless, alterations in the kinetics of immature B cell migration to the periphery under conditions of immunization might explain reports on RAG expression in peripheral B cells even in the absence of reinduction.

FROM EDITING TO DELETION Termination of RAG expression prevents further V(D)J recombination and receptor editing. To maintain tolerance, B cells that cannot edit self-reactive receptors must be deleted or rendered anergic. The idea that self-reactive clones are deleted during development was put forward by Joshua Lederberg as a refinement of the clonal selection theory (5). This

11 Feb 2004

19:20

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

492

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

P1: IKH

JANKOVIC ET AL.

concept was confirmed by experiments in which mice were treated with anti-Igs from birth and were found to be B cell depleted (111). The assumption linking this result to tolerance was that anti-IgM antibodies must mimic autoantigens by cross-linking the cell surface receptor. Likewise, bone marrow B cells challenged in vitro with anti-IgM antibodies downregulate surface Ig and undergo apoptosis (11, 13, 15, 19). In transgenic mice, expression of certain self-specificities also induced clonal B cell deletion (8, 55, 112–114). In anti-MHC expressing B cells exposure to antigen induced a developmental block that was concurrent with secondary VL gene recombination and replacement of cell surface receptors (48, 69, 70). Cells that failed to undergo successful editing were destined for death by apoptosis. This idea was confirmed in experiments with RAG-deficient mice that carried anti-DNA or self-MHC antibodies (115, 116). In the absence of secondary recombination all autoreactive B cells were deleted and there were no peripheral B cells (115, 116). Thus, antigen-induced apoptosis is a relatively late event in bone marrow B cell development that is preceded by an editing permissive stage that appears to be dependent on the bone marrow microenvironment (21, 48, 117).

CIS REGULATION OF RAG EXPRESSION Highly regulated RAG expression is required for successful assembly of Ig and TCR genes and normal lymphocyte development. On the other hand, silencing RAGs in mature lymphoid cells is necessary in order to ensure genomic stability. The RAG locus is peculiar in that RAG1 and RAG2 are adjacent but convergently transcribed genes that entered the vertebrate genome at the time of emergence of jawed fish (23–25, 118, 119). This unusual evolutionarily conserved genomic organization and the finding that RAG proteins can mediate transposition in vitro lead to the hypothesis that these genes originate from a mobile element that entered the vertebrate genome at the time of emergence of jawed fish (120, 121). Regulation of RAG expression has been studied by transfection in cell lines in vitro and in vivo using bacterial artificial chromosome (BAC) transgenic mice, using green or yellow fluorescent protein (GFP or YFP) as reporters in reconstitution experiments with transfected ES cells, and by gene targeting. The disadvantage of the transfection systems is that the transfected cells represent static stages in B or T cell development and therefore the dynamic aspects of RAG regulation in developing lymphocytes cannot be explored. Nevertheless, much has been learned about the chromatin structure and transcription factors that bind the murine and human RAG1 and RAG2 TATA-less promoters in vitro (122–126). The RAG1 promoter is not active or tissue specific in transient transfections in the absence of a heterologous enhancer such as the Ig Eµ (125). In contrast, the murine RAG2 promoter is active in transient transfection experiments in B and T cell lines, and the activity is dependent on PAX5 and GATA3 (127, 128). However, stable transfectants showed little activity for either RAG1 or RAG2 promoters in the absence of exogenous enhancers (129). A second method used to evaluate RAG regulation was RAG2−/− blastocyst reconstitution by transfected RAG2 transgenes (130). These experiments suggested

11 Feb 2004

19:20

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

REGULATION OF AUTOANTIBODIES

P1: IKH

493

that all of the information required for RAG2 regulation was found in an 18 kb genomic fragment extending from 9 kb upstream of the RAG2 promoter to 2.4 kb 30 of the untranslated region because it rescued thymocyte development (130). Similar rescue of RAG2−/− mice was also obtained with a yeast artificial chromosome (RYAC) that contained both functional RAG genes and 12 kb of DNA 50 of RAG2 (131). The double positive (DP) compartment was fully reconstituted, and mature CD4+CD8− and CD4−CD8+ thymocytes were present at one third of normal levels in the RYAC/RAG2−/− mice (131). However, this reconstitution of the DP and single positive (SP) stage differed significantly from the normal wild-type T cell development in that the mature cells were reduced by up to 70% and that the TCRα rearrangements were restricted to only the initial recombination at the 50 end of the Jα locus (131). Reconstitution was nonphysiological because RYAC failed to induce detectable RAG1 or RAG2 transcription in DP T cells. Thus, neither the 18 kb genomic fragment (130) nor the RYAC contained sequences required for RAG expression in DP T cells (131), and the RAG2−/− blastocyst reconstitution is not a sensitive assay to study RAG regulation in vivo. Transgenic mouse and gene targeting experiments have confirmed the requirement for distal enhancers in regulating RAG expression in vivo. Two elements with RAG promoter-enhancing activity have been identified: One element is active in B cells and the second in DP T cells (32, 129, 131). Both of these elements are on the RAG2 side of the locus, and both elements have effects on RAG1 and on RAG2 promoters, suggesting a mechanism for achieving coordinate regulation of the two genes by one set of genetic elements (32). The B cell specific element is found in a 2.3 kb area located approximately 22 kb 50 of RAG2. In cell lines this element enhances transcription over background levels in a highly variegated manner (36% of the transfected cells show no expression at all and only 10% of the clones show high levels of expression) (129). Nevertheless, enhancement is found only in pro-B cells and is absent from T, adult B, or nonlymphoid cell lines. Targeted deletion of this element from the mouse genome leads to a B cell specific tenfold and a twofold decrease in RAG1 and RAG2 mRNA levels, respectively, with no effect on RAG expression in DN or DP cells (129). Although the targeted deletion shows that this element is required for complete regulation, no ectopic or developmentally aberrant expression is reported in its absence. It remains to be determined whether this element is sufficient to drive tissue-specific and developmentally regulated expression of RAG, as is the case for other enhancers such as the Igµ enhancer. The T cell specific element was identified in transgenic experiments with bacterial artificial chromosomes (BACs) that carry fluorescent protein indicator genes in place of the RAG genes (32). These experiments showed that the transcription of both RAG1 and RAG2 in DP thymocytes is coordinated by one or more cis element(s) located between 32 and 87 kb 50 of RAG2. In addition to regulating expression in DP thymocytes, this region was also required to prevent position effect variegation in developing T cells and B cells. Therefore, an element (or elements) in this region is a required part of a traditional locus control region (132, 133).

11 Feb 2004

19:20

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

494

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

P1: IKH

JANKOVIC ET AL.

Like developing B cells, thymocytes express RAGs in two waves. V(D)J recombination is initiated with the first wave of RAG expression at the TCRβ locus in CD4−CD8− (DN) T cells. The TCRβ locus is analogous to the Ig heavy chain locus, and DN T cells are analogous to pro-B cells (134). Once a TCRβ chain is expressed, it combines with pre-Tα and CD3 components to produce the pre-TCR (135, 136). Similar to the pre-BCR, the pre-TCR downregulates RAG expression and induces T cells to mature to the CD4+CD8+ double positive (DP) stage. Upon entering the DP stage there is a second wave of RAG expression and progressive V(D)J recombination along the Jα locus (29, 131, 137–139). V(D)J recombination is only turned off in DP cells when they express a TCR that recognizes self-MHC with an affinity that allows for positive selection (140, 141). Thus, TCR cross-linking in DP T cells is the signal that extinguishes RAG expression. This is different from pre-B cells where receptor cross-linking delays extinction of RAG expression and prolongs recombination leading to receptor editing (73). We speculate that the separate B cell and DP T cell specific elements for RAG regulation may provide a molecular explanation for why T cells turn off RAG expression upon TCR cross-linking, whereas immature B cells prolong RAG expression and induce receptor editing upon BCR cross-linking. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Tonegawa S. 1983. Somatic generation of antibody diversity. Nature 302:575–81 2. Silverstein AM. 2001. Autoimmunity versus horror autotoxicus: the struggle for recognition. Nat. Immunol. 2:279–81 3. Burnet MF. 1959. The Clonal Selection Theory of Acquired Immunity. Cambridge, UK: Cambridge Univ. Press 4. Talmage DW. 1959. Tumor development in immunologically deficient nude mice after exposure to chemical carcinogens. Science 129:1643–48 5. Lederberg J. 1959. Genes and antibodies. Science 129:1649–53 6. Gay D, Saunders T, Camper S, Weigert M. 1993. Receptor editing: an approach by autoreactive B cells to escape tolerance. J. Exp. Med. 177:999–1008 7. Tiegs SL, Russell DM, Nemazee D. 1993. Receptor editing in self-reactive bone marrow B cells. J. Exp. Med. 177:1009– 20 8. Nemazee DA, Burki K. 1989. Clonal dele-

9.

10. 11.

12. 13.

14.

tion of B lymphocytes in a transgenic mouse bearing anti-MHC class I antibody genes. Nature 337:562–66 Goodnow CC, Crosbie J, Adelstein S, Lavoie TB, Smith-Gill SJ, et al. 1988. Altered immunoglobulin expression and functional silencing of self-reactive B lymphocytes in transgenic mice. Nature 334:676–82 Landsteiner K. 1936. The Specificity of Serological Reactions. New York: Dover Sidman CL, Unanue ER. 1975. Receptormediated inactivation of early B lymphocytes. Nature 257:149–51 Teale JM, Klinman NR. 1980. Tolerance as an active process. Nature 288:385–87 Nossal GJ. 1983. Cellular mechanisms of immunologic tolerance. Annu. Rev. Immunol. 1:33–62 Allman DM, Ferguson SE, Lentz VM, Cancro MP. 1993. Peripheral B cell maturation. II. Heat-stable antigen (hi) splenic B cells are an immature developmental

11 Feb 2004

19:20

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

REGULATION OF AUTOANTIBODIES

15.

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

16.

17.

18.

19.

20.

21.

22.

23.

24.

intermediate in the production of longlived marrow-derived B cells. J. Immunol. 151:4431–44 Norvell A, Mandik L, Monroe JG. 1995. Engagement of the antigen-receptor on immature murine B lymphocytes results in death by apoptosis. J. Immunol. 154:4404–13 Rolink AG, Andersson J, Melchers F. 1998. Characterization of immature B cells by a novel monoclonal antibody, by turnover and by mitogen reactivity. Eur. J. Immunol. 28:3738–48 Forster I, Rajewsky K. 1990. The bulk of the peripheral B-cell pool in mice is stable and not rapidly renewed from the bone marrow. Proc. Natl. Acad. Sci. USA 87:4781–84 Hao Z, Rajewsky K. 2001. Homeostasis of peripheral B cells in the absence of B cell influx from the bone marrow. J. Exp. Med. 194:1151–64 Raff MC, Owen JJ, Cooper MD, Lawton ARd, Megson M, et al. 1975. Differences in susceptibility of mature and immature mouse B lymphocytes to antiimmunoglobulin-induced immunoglobulin suppression in vitro. Possible implications for B-cell tolerance to self. J. Exp. Med. 142:1052–64 Gu H, Tarlinton D, Muller W, Rajewsky K, Forster I. 1991. Most peripheral B cells in mice are ligand selected. J. Exp. Med. 173:1357–71 Sandel PC, Monroe JG. 1999. Negative selection of immature B cells by receptor editing or deletion is determined by site of antigen encounter. Immunity 10:289–99 Wardemann H, Yurasov S, Schaefer A, Young JW, Meffre E, Nussenzweig MC. 2003. Predominant autoantibody production by early human B cell precursors. Science 301(5638):1374-77 Schatz DG, Oettinger MA, Baltimore D. 1989. The V(D)J recombination activating gene, RAG-1. Cell 59:1035–48 Oettinger MA, Schatz DG, Gorka C, Baltimore D. 1990. RAG-1 and RAG-2, ad-

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

P1: IKH

495

jacent genes that synergistically activate V(D)J recombination. Science 248:1517– 23 McBlane JF, van Gent DC, Ramsden DA, Romeo C, Cuomo CA, et al. 1995. Cleavage at a V(D)J recombination signal requires only RAG1 and RAG2 proteins and occurs in two steps. Cell 83:387–95 Mombaerts P, Iacomini J, Johnson RS, Herrup K, Tonegawa S, et al. 1992. RAG1-deficient mice have no mature B and T lymphocytes. Cell 68:869–77 Shinkai Y, Rathbun G, Lam KP, Oltz EM, Stewart V, et al. 1992. RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell 68:855–67 Schwarz K, Gauss GH, Ludwig L, Pannicke U, Li Z, et al. 1996. RAG mutations in human B cell-negative SCID. Science 274:97–99 Wilson A, Held W, MacDonald HR. 1994. Two waves of recombinase gene expression in developing thymocytes. J. Exp. Med. 179:1355–60 Grawunder U, Leu TM, Schatz DG, Werner A, Rolink AG, et al. 1995. Downregulation of RAG1 and RAG2 gene expression in preB cells after functional immunoglobulin heavy chain rearrangement. Immunity 3:601–8 Allman D, Li J, Hardy RR. 1999. Commitment to the B lymphoid lineage occurs before DH-JH recombination. J. Exp. Med. 189:735–40 Yu W, Misulovin Z, Suh H, Hardy RR, Jankovic M, et al. 1999. Coordinate regulation of RAG1 and RAG2 by cell typespecific DNA elements 50 of RAG2. Science 285:1080–84 Igarashi H, Gregory SC, Yokota T, Sakaguchi N, Kincade PW. 2002. Transcription from the RAG1 locus marks the earliest lymphocyte progenitors in bone marrow. Immunity 17:117–30 Alt FW, Yancopoulos GD, Blackwell TK, Wood C, Thomas E, et al. 1984. Ordered rearrangement of immunoglobulin heavy

11 Feb 2004

19:20

496

35.

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

36.

37.

38.

39.

40.

41.

42.

43.

44.

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

P1: IKH

JANKOVIC ET AL. chain variable region segments. EMBO J. 3:1209–19 Li YS, Hayakawa K, Hardy RR. 1993. The regulated expression of B lineage associated genes during B cell differentiation in bone marrow and fetal liver. J. Exp. Med. 178:951–60 Ehlich A, Martin V, Muller W, Rajewsky K. 1994. Analysis of the B-cell progenitor compartment at the level of single cells. Curr. Biol. 4:573–83 Nussenzweig MC, Shaw AC, Sinn E, Danner DB, Holmes KL, et al. 1987. Allelic exclusion in transgenic mice that express the membrane form of immunoglobulin mu. Science 236:816–19 Manz J, Denis K, Witte O, Brinster R, Storb U. 1988. Feedback inhibition of immunoglobulin gene rearrangement by membrane mu, but not by secreted mu heavy chains [published erratum appears in J. Exp. Med. 1989 June 1. 169(6):2269]. J. Exp. Med. 168:1363–81 Kitamura D, Rajewsky K. 1992. Targeted disruption of mu chain membrane exon causes loss of heavy-chain allelic exclusion. Nature 356:154–56 Melchers F, Rolink A, Grawunder U, Winkler TH, Karasuyama H, et al. 1995. Positive and negative selection events during B lymphopoiesis. Curr. Opin. Immunol. 7:214–27 Bassing CH, Swat W, Alt FW. 2002. The mechanism and regulation of chromosomal V(D)J recombination. Cell 109(Suppl.):S45–55 Chowdhury D, Sen R. 2001. Stepwise activation of the immunoglobulin mu heavy chain gene locus. EMBO J. 20:6394–403 Melchers F. 1995. The role of B cell and pre-B cell receptors in development and growth control of the B-lymphocyte lineage. In Immunoglobulin Genes, ed. T Honjo, FW Alt. pp. 33–56. London: London Academic Ltd. 3rd ed. Monroe RJ, Seidl KJ, Gaertner F, Han S, Chen F, et al. 1999. RAG2:GFP knockin mice reveal novel aspects of RAG2 ex-

45.

46.

47.

48.

49.

50.

51.

52.

53.

pression in primary and peripheral lymphoid tissues. Immunity 11:201–12 Li Z, Dordai DI, Lee J, Desiderio S. 1996. A conserved degradation signal regulates RAG-2 accumulation during cell division and links V(D)J recombination to the cell cycle. Immunity 5:575–89 Ehlich A, Schaal S, Gu H, Kitamura D, Muller W, et al. 1993. Immunoglobulin heavy and light chain genes rearrange independently at early stages of B cell development. Cell 72:695–704 Carsetti R, Kohler G, Lamers MC. 1995. Transitional B cells are the target of negative selection in the B cell compartment. J. Exp. Med. 181:2129–40 Melamed D, Benschop RJ, Cambier JC, Nemazee D. 1998. Developmental regulation of B lymphocyte immune tolerance compartmentalizes clonal selection from receptor selection. Cell 92:173–82 Radic MZ, Erikson J, Litwin S, Weigert M. 1993. B lymphocytes may escape tolerance by revising their antigen receptors. J. Exp. Med. 177:1165–73 Feddersen RM, Van Ness BG. 1985. Double recombination of a single immunoglobulin kappa-chain allele: implications for the mechanism of rearrangement. Proc. Natl. Acad. Sci. USA 82: 4793–97 Levy S, Campbell MJ, Levy R. 1989. Functional immunoglobulin light chain genes are replaced by ongoing rearrangements of germline V kappa genes to downstream J kappa segment in a murine B cell line. J. Exp. Med. 170:1–13 Kleinfield R, Hardy RR, Tarlinton D, Dangl J, Herzenberg LA, et al. 1986. Recombination between an expressed immunoglobulin heavy-chain gene and a germline variable gene segment in a Ly 1+ B-cell lymphoma. Nature 322:843–46 Ritchie KA, Brinster RL, Storb U. 1984. Allelic exclusion and control of endogenous immunoglobulin gene rearrangement in kappa transgenic mice. Nature 312:517–20

11 Feb 2004

19:20

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

REGULATION OF AUTOANTIBODIES 54. Rusconi S, Kohler G. 1985. Transmission and expression of a specific pair of rearranged immunoglobulin mu and kappa genes in a transgenic mouse line. Nature 314:330–34 55. Erikson J, Radic MZ, Camper SA, Hardy RR, Carmack C, et al. 1991. Expression of anti-DNA immunoglobulin transgenes in non-autoimmune mice. Nature 349:331– 34 56. Prak EL, Trounstine M, Huszar D, Weigert M. 1994. Light chain editing in kappa-deficient animals: a potential mechanism of B cell tolerance. J. Exp. Med. 180:1805–15 57. Li H, Jiang Y, Prak EL, Radic M, Weigert M. 2001. Editors and editing of anti-DNA receptors. Immunity 15:947–57 58. Pelanda R, Schwers S, Sonoda E, Torres RM, Nemazee D, et al. 1997. Receptor editing in a transgenic mouse model: site, efficiency, and role in B cell tolerance and antibody diversification. Immunity 7:765– 75 59. Chen C, Prak EL, Weigert M. 1997. Editing disease-associated autoantibodies. Immunity 6:97–105 60. Kleinfield RW, Weigert MG. 1989. Analysis of VH gene replacement events in a B cell lymphoma. J. Immunol. 142:4475–82 61. Reth M, Gehrmann P, Petrac E, Wiese P. 1986. A novel VH to VHDJH joining mechanism in heavy-chain-negative (null) pre-B cells results in heavy-chain production. Nature 322:840–42 62. Wilson PC, Wilson K, Liu YJ, Banchereau J, Pascual V, et al. 2000. Receptor revision of immunoglobulin heavy chain variable region genes in normal human B lymphocytes. J. Exp. Med. 191:1881–94 63. Chen C, Nagy Z, Prak EL, Weigert M. 1995. Immunoglobulin heavy chain gene replacement: a mechanism of receptor editing. Immunity 3:747–55 64. Pewzner-Jung Y, Friedmann D, Sonoda E, Jung S, Rajewsky K, et al. 1998. B cell deletion, anergy, and receptor editing in “knock in” mice targeted with

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

P1: IKH

497

a germline-encoded or somatically mutated anti-DNA heavy chain. J. Immunol. 161:4634–45 Itoh K, Meffre E, Albesiano E, Farber A, Dines D, et al. 2000. Immunoglobulin heavy chain variable region gene replacement as a mechanism for receptor revision in rheumatoid arthritis synovial tissue B lymphocytes. J. Exp. Med. 192:1151– 64 Sonoda E, Pewzner-Jung Y, Schwers S, Taki S, Jung S, et al. 1997. B cell development under the condition of allelic inclusion. Immunity 6:225–33 Cascalho M, Ma A, Lee S, Masat L, Wabl M. 1996. A quasi-monoclonal mouse. Science 272:1649–52 Zhang Z ZM, Wang YH, Munfus D, Huye LE, Findley HW, et al. 2003. Contribution of Vh gene replacement to the primary B cell repertoire. Immunity 19:21–31 Hertz M, Nemazee D. 1997. BCR ligation induces receptor editing in IgM+IgDbone marrow B cells in vitro. Immunity 6:429–36 Melamed D, Nemazee D. 1997. Selfantigen does not accelerate immature B cell apoptosis, but stimulates receptor editing as a consequence of developmental arrest. Proc. Natl. Acad. Sci. USA 94:9267–72 Nemazee D, Buerki K. 1989. Clonal deletion of autoreactive B lymphocytes in bone marrow chimeras. Proc. Natl. Acad. Sci. USA 86:8039–43 Yu W, Nagaoka H, Jankovic M, Misulovin Z, Suh H, et al. 1999. Continued RAG expression in late stages of B cell development and no apparent re-induction after immunization. Nature 400:682–87 Casellas R, Shih TA, Kleinewietfeld M, Rakonjac J, Nemazee D, et al. 2001. Contribution of receptor editing to the antibody repertoire. Science 291:1541–44 Oberdoerffer P, Novobrantseva TI, Rajewsky K. 2003. Expression of a targeted lambda 1 light chain gene is developmentally regulated and independent of

11 Feb 2004

19:20

498

75.

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

76.

77.

78.

79.

80.

81.

82.

83.

84.

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

P1: IKH

JANKOVIC ET AL. Ig kappa rearrangements. J. Exp. Med. 197:1165–72 Prak EL, Weigert M. 1995. Light chain replacement: a new model for antibody gene rearrangement. J. Exp. Med. 182:541– 48 Brauninger A, Goossens T, Rajewsky K, Kuppers R. 2001. Regulation of immunoglobulin light chain gene rearrangements during early B cell development in the human. Eur. J. Immunol. 31:3631– 37 Retter MW, Nemazee D. 1998. Receptor editing occurs frequently during normal B cell development. J. Exp. Med. 188:1231– 38 Blackwell TK, Moore MW, Yancopoulos GD, Suh H, Lutzker S, et al. 1986. Recombination between immunoglobulin variable region gene segments is enhanced by transcription. Nature 324:585–89 Schlissel MS, Baltimore D. 1989. Activation of immunoglobulin kappa gene rearrangement correlates with induction of germline kappa gene transcription. Cell 58:1001–7 Yancopoulos GD, Alt FW. 1985. Developmentally controlled and tissue-specific expression of unrearranged VH gene segments. Cell 40:271–81 Chen J, Young F, Bottaro A, Stewart V, Smith RK, et al. 1993. Mutations of the intronic IgH enhancer and its flanking sequences differentially affect accessibility of the JH locus. EMBO J. 12:4635–45 Serwe M, Sablitzky F. 1993. V(D)J recombination in B cells is impaired but not blocked by targeted deletion of the immunoglobulin heavy chain intron enhancer. EMBO J. 12:2321–27 Xu Y, Davidson L, Alt FW, Baltimore D. 1996. Deletion of the Ig kappa light chain intronic enhancer/matrix attachment region impairs but does not abolish V kappa J kappa rearrangement. Immunity 4:377– 85 Whitehurst CE, Schlissel MS, Chen J. 2000. Deletion of germline promoter PD

85.

86.

87.

88.

89.

90.

91.

92.

93.

beta 1 from the TCR beta locus causes hypermethylation that impairs D beta 1 recombination by multiple mechanisms. Immunity 13:703–14 Corcoran AE, Riddell A, Krooshoop D, Venkitaraman AR. 1998. Impaired immunoglobulin gene rearrangement in mice lacking the IL-7 receptor. Nature 391:904–7 Romanow WJ, Langerak AW, Goebel P, Wolvers-Tettero IL, van Dongen JJ, et al. 2000. E2A and EBF act in synergy with the V(D)J recombinase to generate a diverse immunoglobulin repertoire in nonlymphoid cells. Mol. Cell 5:343–53 Alvarez JD, Anderson SJ, Loh DY. 1995. V(D)J recombination and allelic exclusion of a TCR beta-chain minilocus occurs in the absence of a functional promoter. J. Immunol. 155:1191–202 Okada A, Mendelsohn M, Alt F. 1994. Differential activation of transcription versus recombination of transgenic T cell receptor beta variable region gene segments in B and T lineage cells. J. Exp. Med. 180:261–72 Cherry SR, Baltimore D. 1999. Chromatin remodeling directly activates V(D)J recombination. Proc. Natl. Acad. Sci. USA 96:10788–93 Angelin-Duclos C, Calame K. 1998. Evidence that immunoglobulin VH-DJ recombination does not require germ line transcription of the recombining variable gene segment. Mol. Cell Biol. 18:6253– 64 Hesslein DG, Pflugh DL, Chowdhury D, Bothwell AL, Sen R, et al. 2003. Pax5 is required for recombination of transcribed, acetylated, 50 IgH V gene segments. Genes Dev. 17:37–42 Gstaiger M, Knoepfel L, Georgiev O, Schaffner W, Hovens CM. 1995. A B-cell coactivator of octamer-binding transcription factors. Nature 373:360–62 Strubin M, Newell JW, Matthias P. 1995. OBF-1, a novel B cell-specific coactivator that stimulates immunoglobulin promoter

11 Feb 2004

19:20

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

REGULATION OF AUTOANTIBODIES

94.

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

95.

96.

97.

98.

99.

100.

101.

102.

103.

activity through association with octamerbinding proteins. Cell 80:497–506 Luo Y, Fujii H, Gerster T, Roeder RG. 1992. A novel B cell-derived coactivator potentiates the activation of immunoglobulin promoters by octamer-binding transcription factors. Cell 71:231–41 Staudt LM, Lenardo MJ. 1991. Immunoglobulin gene transcription. Annu. Rev. Immunol. 9:373–98 Luo Y, Roeder RG. 1995. Cloning, functional characterization, and mechanism of action of the B-cell-specific transcriptional coactivator OCA-B. Mol. Cell Biol. 15:4115–24 Schubart DB, Rolink A, Kosco-Vilbois MH, Botteri F, Matthias P. 1996. B-cellspecific coactivator OBF-1/OCA-B/Bob1 required for immune response and germinal centre formation. Nature 383:538–42 Casellas R, Jankovic M, Meyer G, Gazumyan A, Luo Y, et al. 2002. OcaB is required for normal transcription and V(D)J recombination of a subset of immunoglobulin κ genes. Cell 110:575–85 Han S, Zheng B, Schatz DG, Spanopoulou E, Kelsoe G. 1996. Neoteny in lymphocytes: Rag1 and Rag2 expression in germinal center B cells. Science 274:2094– 97 Hikida M, Mori M, Takai T, Tomochika K, Hamatani K, et al. 1996. Reexpression of RAG-1 and RAG-2 genes in activated mature mouse B cells. Science 274:2092– 94 Giachino C, Padovan E, Lanzavecchia A. 1998. Re-expression of RAG-1 RAG-2 genes and evidence for secondary rearrangements in human germinal center B lymphocytes. Eur. J. Immunol. 28:3506– 13 Girschick HJ, Grammer AC, Nanki T, Mayo M, Lipsky PE. 2001. RAG1 and RAG2 expression by B cell subsets from human tonsil and peripheral blood. J. Immunol. 166:377–86 Hikida M, Ohmori H. 1998. Rearrangement of lambda light chain genes in

104.

105.

106.

107.

108.

109.

110.

111.

112.

P1: IKH

499

mature B cells in vitro and in vivo. Function of reexpressed recombinationactivating gene (RAG) products. J. Exp. Med. 187:795–99 Meffre E, Papavasiliou F, Cohen P, de Bouteiller O, Bell D, et al. 1998. Antigen receptor engagement turns off the V(D)J recombination machinery in human tonsil B cells. J. Exp. Med. 188:765–72 Meffre E, Davis E, Schiff C, CunninghamRundles C, Ivashkiv LB, et al. 2000. Circulating human B cells that express surrogate light chains edited receptors. Nat. Immunol. 1:207–13 Hikida M, Nakayama Y, Yamashita Y, Kumazawa Y, Nishikawa SI, et al. 1998. Expression of recombination activating genes in germinal center B cells: involvement of interleukin 7 (IL-7) and the IL-7 receptor. J. Exp. Med. 188:365–72 Papavasiliou F, Casellas R, Suh H, Qin XF, Besmer E, et al. 1997. V(D)J recombination in mature B cells: a mechanism for altering antibody responses. Science 278:298–301 Han S, Dillon SR, Zheng B, Shimoda M, Schlissel MS, et al. 1997. V(D)J recombinase activity in a subset of germinal center B lymphocytes. Science 278:301–5 Gartner F, Alt FW, Monroe RJ, Seidl KJ. 2000. Antigen-independent appearance of recombination activating gene (RAG)-positive bone marrow B cells in the spleens of immunized mice. J. Exp. Med. 192:1745–54 Nagaoka H, Gonzalez-Aseguinolaza G, Tsuji M, Nussenzweig MC. 2000. Immunization and infection change the number of recombination activating gene (RAG)expressing B cells in the periphery by altering immature lymphocyte production. J. Exp. Med. 191:2113–20 Lawton AR 3rd, Asofsky R, Hylton MB, Cooper MD. 1972. Suppression of immunoglobulin class synthesis in mice. I. Effects of treatment with antibody to µchain. J. Exp. Med. 135:277–97 Chen C, Nagy Z, Radic MZ, Hardy RR,

11 Feb 2004

19:20

500

113.

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

114.

115.

116.

117.

118.

119.

120.

121.

122.

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

P1: IKH

JANKOVIC ET AL. Huszar D, et al. 1995. The site and stage of anti-DNA B-cell deletion. Nature 373:252–55 Hartley SB, Cooke MP, Fulcher DA, Harris AW, Cory S, et al. 1993. Elimination of self-reactive B lymphocytes proceeds in two stages: arrested development and cell death. Cell 72:325–35 Okamoto M, Murakami M, Shimizu A, Ozaki S, Tsubata T, et al. 1992. A transgenic model of autoimmune hemolytic anemia. J. Exp. Med. 175:71–79 Spanopoulou E, Roman CA, Corcoran LM, Schlissel MS, Silver DP, et al. 1994. Functional immunoglobulin transgenes guide ordered B-cell differentiation in Rag-1-deficient mice. Genes Dev. 8:1030–42 Xu H, Li H, Suri-Payer E, Hardy RR, Weigert M. 1998. Regulation of anti-DNA B cells in recombination-activating genedeficient mice. J. Exp. Med. 188:1247–54 Sandel PC, Gendelman M, Kelsoe G, Monroe JG. 2001. Definition of a novel cellular constituent of the bone marrow that regulates the response of immature B cells to B cell antigen receptor engagement. J. Immunol. 166:5935–44 Fugmann SD, Lee AI, Shockett PE, Villey IJ, Schatz DG. 2000. The RAG proteins and V(D)J recombination: complexes, ends, and transposition. Annu. Rev. Immunol. 18:495–527 Flajnik MF. 1998. Churchill and the immune system of ectothermic vertebrates. Immunol. Rev. 166:5–14 Hiom K, Melek M, Gellert M. 1998. DNA transposition by the RAG1 and RAG2 proteins: a possible source of oncogenic translocations. Cell 94:463–70 Agrawal A, Eastman QM, Schatz DG. 1998. Transposition mediated by RAG1 and RAG2 and its implications for the evolution of the immune system. Nature 394:744–51 Zarrin AA, Fong I, Malkin L, Marsden PA, Berinstein NL. 1997. Cloning and characterization of the human recombina-

123.

124.

125.

126.

127.

128.

129.

130.

131.

132.

tion activating gene 1 (RAG1) and RAG2 promoter regions. J. Immunol. 159:4382– 94 Kurioka H, Kishi H, Isshiki H, Tagoh H, Mori K, et al. 1996. Isolation and characterization of a TATA-less promoter for the human RAG-1 gene. Mol. Immunol. 33:1059–66 Kitagawa T, Mori K, Kishi H, Tagoh H, Nagata T, et al. 1996. Chromatin structure and transcriptional regulation of human RAG-1 gene. Blood 88:3785–91 Fuller K, Storb U. 1997. Identification and characterization of the murine Rag1 promoter. Mol. Immunol. 34:939–54 Brown ST, Miranda GA, Galic Z, Hartman IZ, Lyon CJ, et al. 1997. Regulation of the RAG-1 promoter by the NF-Y transcription factor. J. Immunol. 158:5071–74 Lauring J, Schlissel MS. 1999. Distinct factors regulate the murine RAG-2 promoter in B- and T-cell lines. Mol. Cell. Biol. 19:2601–12 Kishi H, Wei XC, Jin ZX, Fujishiro Y, Nagata T, et al. 2000. Lineage-specific regulation of the murine RAG-2 promoter: GATA-3 in T cells and Pax-5 in B cells. Blood 95:3845–52 Hsu LY, Lauring J, Liang HE, Greenbaum S, Cado D, et al. 2003. A conserved transcriptional enhancer regulates RAG gene expression in developing B cells. Immunity 19:105–17 Monroe RJ, Chen F, Ferrini R, Davidson L, Alt FW. 1999. RAG2 is regulated differentially in B and T cells by elements 50 of the promoter. Proc. Natl. Acad. Sci. USA 96:12713–18 Yannoutsos N, Wilson P, Yu W, Chen HT, Nussenzweig A, et al. 2001. The role of recombination activating gene (RAG) reinduction in thymocyte development in vivo. J. Exp. Med. 194:471–80 Grosveld F, van Assendelft GB, Greaves DR, Kollias G. 1987. Positionindependent, high-level expression of the human beta-globin gene in transgenic mice. Cell 51:975–85

11 Feb 2004

19:20

AR

AR210-IY22-16.tex

AR210-IY22-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

REGULATION OF AUTOANTIBODIES 133. Li Q, Peterson KR, Fang X, Stamatoyannopoulos G. 2002. Locus control regions. Blood 100:3077–86 134. Godfrey D, Kennedy J, Mombaerts P, Tonegawa S, Zlotnik A. 1994. Onset of TCR-beta gene rearrangement and role of TCR-beta expression during CD3CD4-CD8-thymocyte differentiation. J. Immunol. 152:4783–92 135. von Boehmer H, Fehling HJ. 1997. Structure and function of the pre-T cell receptor. Annu. Rev. Immunol. 15:433– 52 136. Kruisbeek AM, Haks MC, Carleton M, Wiest DL, Michie AM, et al. 2000. Branching out to gain control: how the pre-TCR is linked to multiple functions. Immunol. Today 21:637–44 137. Petrie H, Livak F, Schatz D, Strasser A, Crispe I, et al. 1993. Multiple rearrangements in T cell receptor alpha chain genes

138.

139.

140.

141.

P1: IKH

501

maximize the production of useful thymocytes. J. Exp. Med. 178:615–22 Shinkai Y, Koyasu S, Nakayama K, Murphy KM, Loh DY, et al. 1993. Restoration of T cell development in RAG-2-deficient mice by functional TCR transgenes. Science 259:822–25 Guo J, Hawwari A, Li H, Sun Z, Mahanta SK, et al. 2002. Regulation of the TCRalpha repertoire by the survival window of CD4(+)CD8(+) thymocytes. Nat. Immunol. 3:469–76 Turka LA, Schatz DG, Oettinger MA, Chun JJ, Gorka C, et al. 1991. Thymocyte expression of RAG-1 and RAG-2: termination by T cell receptor cross-linking. Science 16:778–81 Borgulya P, Kishi H, Uematsu Y, von Boehmer H. 1992. Exclusion and inclusion of α and β T cell receptor alleles. Cell 69:529–37

3/12/2004 2:05 PM

Figure 1 B cell development. Stages, status of the immunoglobulin heavy (IgH) and light chain (IgL) loci, expression of the RAG proteins (the shades of green represent the intensity of expression of the RAG2 protein in mouse RAG2-GFP indicator strains). Expression of the pre-B and B cell receptors, induction of receptor editing, and apoptosis by self-antigens.

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Jankovic.qxd Page 1

REGULATION OF AUTOANTIBODIES C-1

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:485-501. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.091003.090312

Annu. Rev. Immunol. 2004. 22:503–29 doi: 10.1146/annurev.immunol.22.091003.090312 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on November 17, 2003

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE Warren S. Alexander and Douglas J. Hilton The Walter and Eliza Hall Institute of Medical Research and The Cooperative Research Center for Cellular Growth Factors, Parkville, 3052 Victoria, Australia; email: [email protected], [email protected]

Key Words SOCS proteins, negative regulation, cytokines, signal transduction, attenuation ■ Abstract Cytokines are an integral component of the adaptive and innate immune responses. The signaling pathways triggered by the engagement of cytokines with their specific cell surface receptors have been extensively studied and have provided a profound understanding of the intracellular machinery that translates exposure of cells to cytokine to a coordinated biological response. It has also become clear that cells have evolved sophisticated mechanisms to prevent excessive responses to cytokines. In this review we focus on the suppressors of cytokine signaling (SOCS) family of cytoplasmic proteins that completes a negative feedback loop to attenuate signal transduction from cytokines that act through the janus kinase/signal transducer and activator of transcription (JAK/STAT) pathway. SOCS proteins inhibit components of the cytokine signaling cascade via direct binding or by preventing access to the signaling complex. The SOCS proteins also appear to target signal transducers for proteasomal destruction. Analyses of genetically modified mice in which SOCS proteins are overexpressed or deleted have established that this family of negative regulators has indispensable roles in regulating cytokine responses in cells of the immune system as well as other tissues. Emerging evidence also suggests that disruption of SOCS expression or activity is associated with several immune and inflammatory diseases, raising the prospect that manipulation of SOCS activity may provide a novel future therapeutic strategy in the management of immunological disorders.

INTRODUCTION Cytokines, Receptors and the JAK/STAT Pathway Communication between cells can occur at a variety of levels, from the intimate “pillow-talk” of cell-cell contact, to the “satellite TV” of widely acting, circulating hormones and growth factors. Among the most numerous and functionally diverse 0732-0582/04/0423-0503$14.00

503

15 Mar 2004

15:32

504

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

group of cellular messengers are the cytokines (1) that bind to members of the hemopoietin receptor superfamily (2–4) and signal via the four members of the Janus kinase family (JAK1-3 and TYK2) (reviewed in 5) and the seven members of signal transduction and activators of transcription family (STAT1-4, 5a, 5b, and 6) (reviewed in 6, 7) (Figure 1).

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Importance of Maintaining a Balanced Response Regulation of the initiation, duration, and magnitude of cytokine signaling occurs at multiple levels, including limiting the availability of cytokine to initiate a response, regulating the expression and half-life of cell surface receptor components, and controlling the duration of activation and half-life of intracellular signal transduction machinery (Figure 1). Conceptually, one of the simplest means of attenuating a response is via a negative feedback loop. This review focuses on the physiological role of an important class of negative feedback inhibitors of signal transduction, SOCS proteins, with particular emphasis on the immune system.

SUPPRESSORS OF CYTOKINE SIGNALING The SOCS Protein Family There are eight members of the SOCS protein family: the cytokine-inducible SH2 domain-containing protein (CIS) and SOCS1 through SOCS7 (8–13). In addition to a central SH2 domain, the eight SOCS proteins contain a conserved and previously undescribed C-terminal motif that is termed the SOCS box (Figure 2) (9) that was also found in three other novel protein families and known as the ankyrin repeat-containing proteins with a SOCS box (ASBs), SPRY domain-containing proteins with a SOCS box (SSBs), and WD40 repeat-containing proteins with a SOCS box (WSBs), as well as other miscellaneous proteins (13, 14). Since their identification, the different SOCS proteins have enjoyed a host of aliases (Figure 2); however, the SOCS nomenclature has now gained widespread, if not universal, acceptance.

Basic Principles Elucidated from In Vitro Studies A plethora of studies have shown that SOCS1, SOCS2, SOCS3, and CIS mRNA and protein are generally present at low levels in unstimulated cells, perhaps because of active repression (15–17), and that mRNA and protein levels are induced rapidly in response to cytokines, with the STATs playing an important part in regulating SOCS gene transcription (17–25). Agents other than cytokines, for example pathogens and their products such as LPS, also induce SOCS expression (26–31). Because overexpression of SOCS proteins can inhibit signaling by a variety of cytokines that act via the JAK/STAT pathway, it has been proposed that SOCS proteins may act as part of a negative feedback loop and may be induced

15 Mar 2004

15:32

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SOCS PROTEINS AND THE IMMUNE RESPONSE

P1: IKH

505

by one stimulus to inhibit a cell’s response to subsequent stimuli (Figures 3 and 4). In the case of SOCS1 there is good evidence that the SH2 domain interacts with a key regulatory tyrosine in the activation loop of JAKs (10, 32–34), whereas for SOCS2, SOCS3, and CIS the evidence points to phosphotyrosines in the cytoplasmic domains of cytokine receptors being the primary site of interaction (8, 21, 35–56). For SOCS1 and SOCS3, an N-terminal domain, known as the kinase inhibitory region (KIR), has also been hypothesized to contribute to negative regulation by acting as a pseudosubstrate for JAKs and by increasing the strength of the interaction between SOCS and the signal transduction complex (32, 34, 57). Finally, there is now increasingly good evidence that the SOCS box, present at the C terminus of all SOCS proteins, acts to couple the substrate-specific interactions of the SH2 domains to generic components of the ubiquitin ligation machinery. This leads to the polyubiquitylation of key signaling proteins, their degradation in the proteasome, and the termination of signaling (14, 58–63).

THE PHYSIOLOGY OF SOCS FUNCTION Because the SOCS proteins inhibit cytokine signaling via a negative feedback loop, it was envisaged that the specific biological actions of the SOCS proteins would emerge from an understanding of the specific cell types and cytokines that induce each SOCS. However, as discussed above, SOCS proteins appear to be induced in many cell types by a multitude of different cytokines. Moreover, enforced expression of these SOCS proteins in cell culture models has resulted in a remarkably promiscuous range of activity (reviewed in 64–66). Thus, although overexpression studies have provided valuable insights into the mechanisms of SOCS action, the breadth of SOCS activity in these systems is likely to exaggerate the physiological roles of these regulators in vivo. Over recent years, production of transgenic mice expressing various SOCS proteins has focused attention on the in vivo actions of these proteins, and the indispensable physiological roles of SOCS have been revealed in studies of mice genetically engineered to lack functional Socs genes (Table 1). We focus below on the data gleaned from analyses of genetically modified mice in which SOCS activity has been enhanced or ablated, particularly where phenotypes affecting the immune system have ensued. These studies have established that SOCS proteins are important biological regulators in adaptive and innate immune responses. In several instances, key physiological roles of individual SOCS proteins have clearly and definitively emerged from these studies, such as the indispensable role for SOCS1 in regulating IFNγ signaling and T cell homeostasis. In other areas, it is clear that no consensus has yet become apparent. For example, in T helper cell development, SOCS1, SOCS3, CIS, and SOCS5 all appear to be differentially expressed in Th1 versus Th2 cells, and conflicting evidence exists regarding the specific roles for these SOCS proteins in the regulation of Th

11 Feb 2004

22:3

506

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 1 Major phenotypic consequences of SOCS gene manipulation in vivo Gene

Knockout mice

Transgenic mice

CIS

Reported to have no abnormalities (96).

Low body weight; lactation failure; fewer splenic γ δT, NK, and NKT cells; preferential Th2 differentiation–reduced IL-2 signaling (109). Anomalous T cell receptor responses (110).

SOCS1

Neonatal lethality with fatty degeneration of the liver, hematopoietic infiltration of multiple organs, lymphopenia, apoptosis in lymphoid organs, and aberrant T-cell activation (67, 68). Lethality due to deregulated responses to IFNγ (72, 74), but responses to γ c receptor-dependent cytokines as well as IL-12, TNFα, LPS, insulin, and prolactin also altered (30, 31, 67, 69, 81, 83).

Suppressed signaling by several cytokines, T cell developmental defects including increased in CD4+ cells, fewer γ δT cells and spontaneous T cell activation (87).

SOCS2

Gigantism with evidence of deregulated GH signaling (111).

Gigantism with evidence of deregulated GH signaling (56).

SOCS3

Midgestational embryonic lethality due to placental insufficiency (96, 97, 99). Deregulated responses to LIF (placenta) and IL-6 (macrophages and liver) (99–102).

Embryonic lethality with anemia (96).

SOCS5 SOCS6

Disrupted Th2 cell responses. Attenuated IL-4 signaling (107). Mild growth retardation (138).

IFNγ , interferon gamma; GH, growth hormone; IL, interleukin; LIF, leukemia inhibitory factor; Th2, T helper cell type 2; TNF, tumor necrosis factor; LPS, lipopolysaccharide; γ c, common gamma chain.

polarization in vivo. In cases such as these, it seems likely that analyses of genetically modified mice in which multiple SOCS proteins have been ablated, as well as complementary studies of SOCS in human health and disease, will ultimately be required to delineate the precise roles of specific SOCS proteins in complex immune cell physiology.

SOCS1 is a Key Regulator of Interferon-γ Signaling Mice lacking the Socs1 gene die within the first three weeks of life, displaying low body weight and complex pathology (67, 68). Sick Socs1−/− mice have major liver damage, with parenchymal cells showing marked accumulation of lipid and consequent fatty degeneration and necrosis that can affect large areas of the organ.

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SOCS PROTEINS AND THE IMMUNE RESPONSE

507

Diseased livers also exhibited significant hematopoietic infiltration, characterized by the presence of aggregates of granulocytes, eosinophils, and macrophages. In most Socs1−/− mice, monocytic invasion of the pancreas, heart, and lungs was also evident (68). Analyses of Socs1−/− mice have suggested they have defects in regulation of signaling in response to prolactin and insulin (69, 70); however, the most striking defects in these mice are found in the acquired and innate immune systems, as discussed below. Analysis of peripheral blood in Socs1−/− mice revealed modest reductions in hematocrit and circulating eosinophil and platelet numbers, and a variable increase in neutrophils, but most strikingly, a consistent and marked reduction in blood lymphocytes (68). A relative deficit in lymphocytes was also evident in the spleen, in which lymphoid follicles were often absent or rudimentary and composed of immature cells. The thymus was markedly reduced in size in sick Socs1−/− mice, reflecting a progressive depletion of cortical thymocytes with declining health, and lymph nodes and Peyer’s patches were similarly hypocellular (67, 68). Increased numbers of apoptotic cells were observed in the thymus and spleen of mice lacking SOCS1, accompanied by increased expression of the Bax protein, suggesting that the lymphopenia in these mice may be associated with accelerated apoptosis (67). Flow cytometric analysis of B-lymphoid populations revealed relatively normal numbers of pro-B cells in Socs1−/− mice but severe deficiencies in pre-B and B lymphocytes (67, 68). The deficiency in T lymphocyte numbers in the thymus was accompanied by a reduction in the ratio of mature (CD3+) CD4:CD8 cells, due predominantly to elevated numbers of CD8+ T lymphocytes, and this altered ratio was also evident in the peripheral lymphoid organs (71). T cells in Socs1−/− mice also displayed features of activation: increased cell size and expression of elevated levels of activation markers including CD44, CD25, and CD69 (72). The similarities in the pathology of sick Socs1−/− mice to that observed in wild-type mice administered with interferon (IFN) (73) led to the hypothesis that the disease developing in Socs1−/− mice might be due to excessive responses to IFN. Consistent with this notion, Socs1−/− mice displayed evidence of an ongoing response to IFNγ including constitutive activation of STAT1, the primary STAT that mediates IFNγ signaling, in the liver, and markedly elevated expression of IFNγ -inducible genes in several SOCS1-deficient tissues (74). Direct evidence that IFNγ is required for the development of lethal disease in Socs1−/− mice emerged when these mice were treated from birth with neutralizing antibodies to IFNγ . At three weeks of age, when all untreated Socs1−/− mice had succumbed to lethal disease, anti-IFNγ -treated Socs1−/− mice remained in good health and displayed only minor histopathological or hematological signs of disease (74). The necessity for IFNγ in Socs1−/− disease was unequivocally demonstrated when double knockout Socs1−/− Ifng−/− mice were generated. These mice remained healthy and showed none of the histological features of typical Socs1−/− disease (72, 74). Interestingly, Ifng gene dosage is a critical determinant of disease in SOCS1-deficient mice. Whereas the absence of SOCS1 leads to neonatal lethality in the presence of two functional Ifng alleles but is tolerated without apparent

11 Feb 2004

22:3

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

508

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

disease in the absence of IFNγ , Socs1−/− mice with a single functional Ifng allele develop disease as young adult mice. Intriguingly, Socs1−/− Ifng+/− mice, which typically die between 30 and 70 days of age, do not simply develop an attenuated form of the neonatal disease that characterizes Socs1−/− Ifng +/+ mice, but display a distinct pathology dominated by myocarditis and polymyositis (75). Extensive infiltrates involving T lymphocytes, macrophages, and eosinophils were evident in all muscles, the heart and the cornea, with minimal involvement of other organs or tissues. Although slightly excessive numbers of hematopoietic cells were often observed in the livers of Socs1−/− Ifng+/− mice, the fatty degeneration typical of Socs1−/− neonatal liver disease was absent. It is noteworthy that the reduced CD4:CD8 T cell ratio and excessive T cell activation associated with SOCS1deficient mice were also clearly evident in lymphoid organs of Socs1−/− Ifng+/− and Socs1−/− Ifng−/− animals (75), and these mice have been exploited to investigate the basis of these T cell anomalies (see below). Although elevated circulating concentrations of IFNγ have been observed in some mice lacking SOCS1 (72, 74), increased sensitivity of SOCS1-deficient tissues to IFNγ clearly plays a significant role in Socs1−/− disease. Hematopoietic progenitor cells committed to production of granulocytes and macrophages from the bone marrow of Socs1−/− mice exhibited hypersusceptibility to IFNγ mediated inhibition of proliferation in vitro (76). Similarly, macrophages derived from Socs1−/− bone marrow proved capable of eliminating intracellular Leishmania major parasites following stimulation with IFNγ at a concentration two orders of magnitude lower than required by wild-type cells (74). SOCS1-deficient mice also displayed marked hypersensitivity to IFNγ in vivo. Administration of IFNγ to neonatal Socs1−/− Ifng−/− mice was acutely toxic and recapitulated the pathology evident in Socs1−/− mice, at doses tolerated without symptoms in wildtype neonates (77). Consistent with hyper-responsiveness to IFNγ , Socs1−/− mice also showed increased resistance to infection with Semliki Forest virus, surviving infectious doses lethal to wild-type mice (74). The role of SOCS1 in attenuating IFNγ signaling has also been scrutinized at the biochemical level. The actions of IFNγ in cells are largely mediated via the phosphorylation and activation of STAT1. Following injection of IFNγ , STAT1 phosphorylation in the livers of Socs1+/+ mice is evident within 15 min and remains prominent for approximately 2 h before declining to undetectable levels. In contrast, while STAT1 phosphorylation is induced to a similar absolute level in SOCS1-deficient livers following injection of IFNγ , the phosphorylated protein persists and remains detectable 8 h after cytokine administration (77). Prolonged activation of the IFNγ signaling pathway is also evident in isolated hepatocytes in vitro and can be demonstrated not only by extended phosphorylation of STAT1 following IFNγ stimulation but also by the prolonged presence of activated, DNAbinding STAT1 complexes in the nuclei of these cells (77). Together, these biological and biochemical data clearly establish that SOCS1 is a key physiological negative regulator of IFNγ signaling. The actions of SOCS1 are necessary to attenuate the duration of IFNγ signaling in cells, allowing the

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

SOCS PROTEINS AND THE IMMUNE RESPONSE

509

beneficial immunological effects of IFNγ , but preventing the pathological consequences of uncontrolled responses to this inflammatory cytokine.

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Immune Cells in Socs1−/− Disease A significant reduction in the numbers of maturing B cells is a feature of Socs1−/− mice. However, production of SOCS1-deficient B cells is not intrinsically defective because in vitro development of mature B cells from purified Socs1−/− precursors was as efficient as that observed from wild-type cells (68). There is little evidence for a direct role for B cells in Socs1−/− disease, and the loss of mature B cells in Socs1−/− mice is likely to reflect the sensitivity of these cells to the cytotoxic actions of IFNγ (78). In contrast, T lymphocytes appear to play a crucial role in the development of disease in mice lacking SOCS1. Transplantation of bone marrow cells from Socs1−/− mice into JAK3- or RAG2-deficient mice, which allows donor lymphoid reconstitution amid a background of host myelopoiesis, resulted in lethality of the reconstituted animals and was accompanied by expression of an activated phenotype by donor SOCS1-deficient T cells (67, 72). When Socs1−/− mice were crossed with RAG2-deficient animals, perinatal disease was eliminated in Socs1−/− Rag2−/− double knockout mice. Thus, the capacity to generate mature T cells appears necessary for the neonatal disease in Socs1−/− mice (72) and may reflect the fact that T cells are the primary cellular source of IFNγ . SOCS1 has been shown to coprecipitate with the CD8/ζ and Syk components of the T cell receptor (TCR) signaling cascade, and overexpression of SOCS1 blocks this pathway in a model of reconstituted TCR signaling in heterologous cells (79). The role of the T cell antigen receptor in Socs1−/− disease has been further examined using transgenic mice in which T cells exclusively express specificity for an exogenous, foreign antigen. To determine whether the T cell activation and/or the disease that develops in the absence of SOCS1 is driven by antigen stimulation, Socs1−/− mice were bred with OT-I MHC class I-restricted TCR transgenic mice. These transgenic mice express chicken ovalbumin (OVA)-specific TCRs on CD8+ T cells; in the absence of administered OVA all these cells should be na¨ıve. Untreated OT-I Socs1−/− Rag1−/− mice, in which all CD8+ T cells are antigen-na¨ıve, survive the neonatal period but succumb to a disease in early adulthood resembling that in Socs1−/− Ifng+/− mice (71). Significantly, CD8+ T cells in these mice exhibited a blast-like activated morphology and expressed high levels of the CD44 activation marker. Similarly, Socs1−/− mice on an OTII transgenic background, in which MHC class II–restricted OVA-specific CD4+ T cells are generated, displayed a similar lifespan and phenotype to OT-I Socs1−/− mice. The abrogation of neonatal disease in OT-I Socs1−/− and OT-II Socs1−/− mice argues that high-affinity TCR interaction contributes to this inflammatory syndrome. However, the fact that the TCR transgenic Socs1−/− mice do ultimately succumb to disease in young adulthood is in clear contrast to Socs1−/− Rag1−/− mice, which remain healthy for this period (71). The difference between these models

11 Feb 2004

22:3

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

510

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

is the presence of T and NKT cells in the TCR transgenic Socs1−/− mice, and it implies that SOCS1-deficient T and/or NKT cells can contribute to inflammatory disease in a manner distinct from classical TCR-mediated autoimmunity. Hepatic lymphocytes from Socs1−/− mice are cytotoxic for syngeneic wild-type hepatocytes, and cell-depletion experiments suggest that this activity resides within the NKT cell population (80). The number of hepatic NKT cells was shown to be significantly elevated in Socs1−/− mice compared with wild-type controls and selective stimulation of NKT cells in vivo accelerated liver disease in these animals (80). Whereas IFNγ appears to be the pivotal player in development of liver degeneration in Socs1−/− mice, primarily via hypersensitivity of cytotoxic hepatic NKT cells to IFNγ stimulation, several other cytokines appear likely to contribute to disease. Administration of IL-4 prior to disease in Socs1−/− mice, in combination with IFNγ , accelerated liver disease in an NKT-cell dependent manner and prolonged activation of STAT6, through which IL-4 signals, has been observed in Socs1−/− cells treated with IL-4 (67). Moreover, double knockout mice lacking SOCS1 in addition to STAT6 were rescued from neonatal lethality (80). Similarly, mice lacking both SOCS1 and STAT4, through which IL-12 and IL-23 signal, show prolonged survival, and IL-12-stimulated T cell proliferation and NK cell cytotoxicity are enhanced in Socs1−/− cells (81). Thus, hypersensitivity to IL-12 and/or IL-23 may also contribute to neonatal Socs1−/− disease, possibly via IL12-stimulated increases in IFNγ production. Finally, hypersensitivity to tumor necrosis factor α (TNFα) has also been reported in Socs1−/− cells (82, 83), and it has been proposed that the apoptosis observed in tissues of neonatal Socs1−/− mice might, at least in part, result from IFNγ -induced TNFα production (83). Although the evidence supporting a key role for T and NKT cells in Socs1−/− pathology is compelling, the absence of SOCS1 in these cells is not in itself sufficient to cause disease. Mice in which the Socs1 gene can be deleted in specific tissues have been created using loxP-cre recombinase technology. While the inactivation of the Socs1 gene specifically throughout the T and NKT compartments was sufficient to invoke the activated T cell phenotype typical of Socs1−/− models, disease did not develop in these mice, at least within the first 6 months of life (84). In contrast, upon transplantation of bone marrow cells from Socs1−/− mice into recipients in which endogenous hematopoiesis has been ablated by irradiation, disease develops but exhibits hallmarks of classical graft-versus-host disease and lacks the fatty degeneration of the liver or extensive infiltration of muscles that typify neonatal or adult Socs1−/− disease (85). The implication of these observations is that the absence of SOCS1 in target tissues as well as in T, NK, or other hematopoietic cells that contribute to disease, is required to achieve the extensive inflammatory lesions typical of Socs1−/− mice. It is already well established that the neonatal liver is acutely sensitive to IFNγ (73) and it seems feasible that the absence of SOCS1 in other tissues such as the pancreas, heart, and skeletal muscles might confer hypersensitivity to cytokines that promote a local inflammatory response.

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

SOCS PROTEINS AND THE IMMUNE RESPONSE

511

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SOCS1 and the Regulation of T Cell Homeostasis It has emerged that SOCS1 also has important IFNγ -independent roles in T lymphoid development and function. As previously discussed, SOCS1 is expressed at high levels in the thymus and, in contrast to the usual pattern of SOCS1 expression, this does not appear to depend on stimulation of thymocytes with T cell cytokines or the TCR (72). The Socs1 gene appears to be transcribed in cells at all major stages of T cell development in the thymus (86), although there is evidence that expression is downregulated during immature T cell development and becomes particularly high in double positive CD4+ CD8+ thymocytes (84, 86). Studies in transgenic mice in which SOCS1 expression was targeted to T lymphocytes provided in vivo confirmation that enforced SOCS1 expression results in inhibition of signaling by multiple T cell cytokines, including IFNγ , IL-6, and cytokines that act via the common gamma (γ c) receptor chain, such as IL-4 and IL-7 (87). Moreover, these mice exhibited reduced numbers of thymocytes resulting from a partial blockade in early thymocyte development (87). These observations were confirmed upon constitutive retroviral-mediated expression of SOCS1 in primitive hematopoietic cells. Enforced expression of SOCS1 suppressed expansion of lymphoid progenitors beyond the earliest pre-TCR stages of development, but did not appear to interfere with the T cell differentiation program (86). These studies imply that high levels of SOCS1 expression at various stages of T lymphocyte development may maintain cells in a cytokine-refractory state until they receive the appropriate developmental cues to proliferate and differentiate in an orderly and controlled manner. Several of the T cell anomalies that characterize Socs1−/− disease were also evident in healthy mice lacking both SOCS1 and IFNγ . The reduced CD4:CD8 T cell ratio in the developing thymus was prominent from an early age in Socs1−/− Ifng−/− mice. Adoptive transfer studies demonstrated that this anomaly was cellintrinsic, and fetal thymic organ cultures confirmed that the altered CD4:CD8 ratio originated in the thymus and resulted from increased production of CD8+ cells (71). Consistent with an intrinsic origin of this abnormality, a similarly reduced CD4:CD8 ratio was observed in mice in which SOCS1 was specifically deleted in T cells (84). T cells in healthy Socs1−/− Ifng−/− mice also displayed increased expression of activation markers such as CD44, which was most profound in peripheral CD8+ cells (71). Analysis of other cell surface markers revealed that these cells expressed a phenotype most similar to memory T cells (71, 84). Again, this phenotype was prominent in mice in which only T cells lack SOCS1 (84), but it was not evident in fetal thymic organ cultures, suggesting that factors external to the thymus are involved (71). Specific high-affinity stimulation of the TCR appears not to be required for expression of this activated T cell phenotype because Socs1−/− mice in which CD8+ T cells express an ovalbumin-specific TCR also exhibit this phenotype in the absence of antigen (71). Functional analysis of T cells from Socs1−/− Ifng−/− mice revealed no apparent difference in cell survival or cytotoxic activity compared with cells expressing SOCS1 (71). However, transplantation

11 Feb 2004

22:3

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

512

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

studies showed that SOCS1-deficient CD8+ T cells have an increased proliferative capacity (71). Because the altered balance of CD4 versus CD8 T cell development and the unusual prominence of cells with a memory T cell–like phenotype appear independent of exogenous antigen-stimulation and are evident in healthy Socs1−/− Ifng−/− mice, they cannot simply arise in response to autoimmunity or other aspects of disease in the absence of SOCS1, but must reflect in vivo actions of SOCS1 that are independent of its role in IFNγ signaling. Since the phenotype of Socs1−/− T cells resembles in many aspects T cells undergoing homeostatic proliferation (88), it is feasible that this process is deregulated in Socs1−/− mice and that an important role of SOCS1 is to regulate cytokines involved in T cell homeostasis. In this context, the cytokines that signal through the common gamma chain receptor have been closely examined. Expression of SOCS1 is potently induced in T cells stimulated with IL-2, IL-4, IL-7, or IL-15 (33, 89), and overexpression of SOCS1 can inhibit responses to several of these cytokines (33, 86, 87, 90). T cells lacking SOCS1 exhibited hypersensitivity to signals from cytokines that act through γ c, with activation of STAT5 evident following stimulation with significantly lower concentrations of IL-7 in thymocytes and each of IL-2, IL-7, and IL-15 in peripheral T cells than was observed in wild-type cells (84, 89). Moreover, in vitro, purified SOCS1-deficient CD4+ CD8lo cells exhibited significantly increased IL-7-induced differentiation to CD8+ single positive cells relative to wild-type controls, and this was amplified when IL-7 was combined with IL-2 and IL-15 (84). Proliferative responses of peripheral CD8+ T cells from Socs1−/− Ifng−/− mice also revealed hypersensitivity to IL-2 and IL-4 (89). Similarly, upregulation of CD44 expression on sorted populations of CD44lo cells from mice lacking SOCS1 revealed that whereas only minor changes were induced by cytokine signaling through the γ c receptor in wild-type cells, robust induction of CD44 was evident in Socs1−/− cells, particularly in IL-2-stimulated peripheral T cells and IL-7-stimulated thymocytes (71, 84). Together, these data reveal that increased sensitivity to signals from γ cdependent cytokines correlates with changes in Socs1−/− T cells in vitro that are similar to those observed in SOCS1-deficient mice in vivo and provide compelling evidence that SOCS1 is an important regulator of this family of cytokines. Although Socs1−/− Ifng−/− mice survive the lethal neonatal and young adult inflammatory diseases typical of Socs1−/− Ifng +/+ and Socs1−/− Ifng+/− mice, later in life they succumb to a range of diseases including chronic inflammatory lesions, polycystic kidneys, and pneumonia (91). Thus, the IFNγ -independent actions of SOCS1 are essential for continued health of the animal, and it is tempting to hypothesize that control of the immune system via regulation of γ c-dependent cytokines is also an important function of SOCS1 throughout life. SOCS1 has also been implicated in T helper cell development. Differentiation of na¨ıve CD4+ T cells into Th1 cells has been correlated with expression of high levels of SOCS1 (92). However, na¨ıve Socs1+/− CD4+ cells underwent enhanced differentiation in vitro under either Th1- or Th2-polarizing conditions (93). Enhanced in vivo polarization was also evident in Socs1+/− mice upon infection

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SOCS PROTEINS AND THE IMMUNE RESPONSE

513

with Listeria monocytogenes, which induces Th1 responses as well as following inoculation with Neobythitis braziliensis, an inducer of Th2 differentiation (93). In contrast, unlike the enhanced Th1 response observed following L. monocytogenes infection (93), in response to intradermal infection with Leishmania major, Socs1+/− mice developed a normal Th1 response and controlled parasitemia in a manner indistinguishable to that of wild-type mice. In this model, skin lesions and lymphadenopathy persisted in Socs1+/− mice, suggesting that other anomalies in the SOCS1-deficient immune system might promote increased tissue damage in response to infection (94). An additional study has suggested a role for IL-6 in preventing Th1 differentiation via stimulation of SOCS1 and the subsequent blockade of IFNγ signal transduction (95). Thus, although evidence is mounting that SOCS1 may be involved in T helper cell differentiation, a consistent model has not yet emerged for a specific role in Th polarization, and further studies will be needed to clarify whether SOCS1 contributes to this process in vivo.

The Role of SOCS3 in the Immune System The deletion of Socs3 by gene targeting resulted in embryonic lethality at midgestation (96, 97) because of defects in the structure of the placenta that may be due to dysregulated LIF signaling (98, 99), rather than defects in erythropoiesis, as first suggested (96). The lethal phenotype in Socs3−/− mice has frustrated efforts to define the key physiological roles of SOCS3 in adult tissues, including the immune system. However, recent exploitation of conditional gene targeting technology as well as studies of adult recipients of Socs3−/− fetal liver transplants has begun to overcome this obstacle. Mice in which the coding sequence of the Socs3 gene has been specifically deleted in the liver or in macrophages have been produced, and analysis has revealed a key role for SOCS3 in the regulation of IL-6 signaling. In mice bearing SOCS3-deficient livers, injection of IL-6 led to prolonged activation of both STAT3 and STAT1 relative to that observed in wild-type livers (100). Similarly, in macrophages lacking SOCS3, enhanced IL-6-induced STAT3 and STAT1 (100– 102) as well as SHP2 activation (102) were also observed. Prolonged biochemical responses to IL-6 correlated with increased sensitivity of Socs3−/− macrophages to the inhibitory effects of IL-6 on proliferation (100). Although enforced expression of SOCS3 can inhibit responses to IFNγ (103), the regulation of IFNγ signaling was unperturbed in livers lacking SOCS3: The kinetics and magnitude of IFNγ -induced STAT1 activation in SOCS3-deficient livers were indistinguishable from those in wild-type livers (100). Thus, responses to IL-6 and IFNγ in SOCS3-deficient livers are the reciprocal of that observed in Socs1−/− tissue, where prolonged responses to IFNγ , but not IL-6, prevail (77, 100). Although SOCS1 and SOCS3 are both induced by IL-6 and IFNγ , and are capable of inhibiting responses to these cytokines when overexpressed, in vivo the actions of these SOCS proteins are not interchangeable but specific and reciprocal.

11 Feb 2004

22:3

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

514

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

Analysis of cytokine-responsive genes in IL-6-stimulated cells revealed that whereas expression of a number of genes was induced as anticipated in wild-type livers or macrophages, the expression of a significantly larger group was stimulated in IL-6-treated Socs3−/− cells. Intriguingly, many of these additional genes are not normally associated with IL-6 responses, but are typically induced upon exposure of cells to IFNγ (100, 102) and may reflect prolonged activation of STAT1 by IL-6 in the absence of SOCS3. Previous studies have shown a similar IFNγ -like response in STAT3-null cells exposed to IL-6 (104). Because SOCS3 expression appears to be STAT3-dependent (25), it follows that the altered response to IL-6 observed in STAT3-deficient cells occurs because these cells are unable to induce SOCS3. Together, these data indicate that SOCS proteins can influence the quality as well as quantity of cellular responses to cytokines. They raise the possibility that significant overlap exists between the signaling pathways triggered by IL-6 and IFNγ and that a key role for SOCS3 is to sculpt the specific response observed in cells exposed to IL-6, perhaps particularly by restricting activation of STAT1. In addition to IL-6, IL-10 and LPS also potently induce expression of SOCS3. However, in Socs3−/− macrophages, LPS- and IL-10-induced responses were normal, suggesting that SOCS3 is not a physiological regulator of the signaling cascades initiated by these agents (101, 102). However, the altered qualitative response to IL-6 in Socs3−/− cells was also evident when responses of SOCS3-deficient macrophages to IL-6 and IL-10 were compared. Whereas LPS-induced induction of inflammatory cytokine production by macrophages is normally inhibited by IL10 and unaffected by exposure to IL-6, in Socs3−/− macrophages, IL-6 mimicked IL-10 with potent inhibition of LPS-induced IL-12 and TNFα production (101). Like SOCS1, SOCS3 is expressed in na¨ıve T cells. However, in contrast to SOCS1, which is associated with differentiation to Th1 cells, SOCS3 becomes prominently expressed in the Th2 class of helper T lymphocytes (92, 105–107). Transgenic mice in which SOCS3 is constitutively expressed in T cells are healthy but show a reduced proliferative response to T cell mitogens (106). TCR-mediated signaling appeared to be intact in Socs3 transgenic T cells, but a defect in the augmentation of IL-2 production and NF-κB activation by CD28 costimulation of anti-TCR-activated T cells was evident (106). Conversely, in Socs3+/− mice, CD28-mediated IL-2 production was enhanced relative to wild-type controls (106). This pattern of observations was reproduced using a differentiated Th2 cell line in which SOCS3 was overexpressed or inhibited by antisense RNA expression. In this model, SOCS3 levels also inversely correlated with cellular proliferation and CD28-mediated cytokine production (105). Biochemical analysis revealed that SOCS3, via its SH2 domain, specifically binds to the phosphorylated form of CD28. Interestingly, unlike the requirement for the N-terminal KIR region of SOCS3 for effective interaction with many cytokine receptors (see above), this domain appeared dispensable for the interaction of SOCS3 with CD28 (106). Although no specific tyrosine on the CD28 protein appeared to mediate interaction with SOCS3, some evidence suggests that SOCS3 inhibits the association of CD28 with PI3 kinase, which is known to require Tyr-189 on CD28 for recruitment to the

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

SOCS PROTEINS AND THE IMMUNE RESPONSE

515

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

complex (106). Together these data suggest a model in which SOCS3 contributes to maintaining quiescence in T lymphocytes and that antigen-stimulated downregulation of SOCS3 expression allows T-helper cell activation. SOCS3 appears to have an important supplementary role in controlling CD28-mediated responses in differentiated Th2 cells. However, it should be noted that T cell development was reported to be unaffected in adult mice reconstituted with blood cells from Socs3−/− fetal liver stem cells (96). The imminent availability of mice in which the Socs3 gene has specifically been deleted in T cells will help clarify the nonredundant roles of SOCS3 in T lymphopoiesis.

The Physiological Actions of Other SOCS Family Members Widespread expression of CIS in transgenic mice resulted in a panoply of phenotypes including growth retardation and a failure to lactate, which closely resembled abnormalities evident in mice lacking STAT5a and/or STAT5b (108, 109). Thus, these observations suggested a particular role for CIS in regulating responses to cytokines that signal via STAT5. Indeed, T cells from CIS transgenic mice were refractory to the effects of IL-2: A failure to activate STAT5 was observed and proliferative responses were attenuated (109). Constitutive expression of CIS in the spleen and the thymus resulted in a significant reduction in the number of γ δ T cells, NK and NKT cells, and T cells from these mice exhibited a tendency for Th2 polarized differentiation in vitro (109). Again, similar immune phenotypes are evident in mice lacking STAT5, and although the specific cytokine signaling pathways have not been defined, it is likely that blockade of STAT5-mediated cytokine responses in CIS transgenic mice accounts for these immune cell anomalies. Curiously, in an independent study of cells from transgenic mice expressing CIS in CD4+ T cells, enhanced TCR-mediated proliferation and survival, cytokine production, and superantigen-mediated T cell activation were observed in vitro (110). Phosphorylation of ZAP-70, an immediate post-TCR signaling event, appeared to occur normally in CIS transgenic CD4+ cells, but enhanced activation of MAP kinases was evident and interaction of CIS with PKCθ was observed (110). The authors propose that via regulation of PKCθ , CIS modulates signals transmitted from TCR activation to the MAP kinases (110). Nevertheless, T cell development appeared to occur normally in these CD4+ CIS transgenic mice and so the biological consequences of these observations are unclear. T cell development is also reportedly normal in mice in which the gene encoding CIS has been inactivated (96). Thus, whereas CIS clearly has profound effects on enforced expression in immune cells, its precise physiological role remains obscure. The complexity of the contribution of SOCS proteins to T cell regulation is further illustrated by an apparent role for SOCS5 in regulation of IL-4 signaling and Th2 cell differentiation. SOCS5 appears to be preferentially expressed in Th1 cells and could coprecipitate with the Box1 region of the IL-4 receptor in extracts from these cells (107). SOCS5 coprecipitates with the IL-4 receptor in a phosphorylation-independent manner, and consistent with this observation, the Nterminal 50 amino acids and not the SH2 domain of SOCS5 appear to mediate the

11 Feb 2004

22:3

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

516

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

interaction. These observations have led to the proposal that SOCS5 inhibits IL-4 signaling by precluding association of JAK1 with the IL-4 receptor. In support of this model, high levels of SOCS5 attenuated IL-4-induced STAT6 phosphorylation in cell lines, and comparison of T cells from transgenic mice expressing SOCS5 with wild-type controls revealed reduced Th2 cell development and reduced production of the Th2 cytokines IL-4, IL-5, and IL-10 (107). Finally, mice lacking SOCS2 exhibit increased growth in a pattern characteristic of deregulated growth hormone (GH) signaling (111). Socs2−/− cells displayed moderately prolonged activation of STAT5, and the enhancement of growth in the absence of SOCS2 was substantially attenuated in mice also lacking STAT5b (112). Counterintuitively, transgenic mice with widespread overexpression of SOCS2 do not show reduced growth, but exhibit a gigantism similar to mice lacking SOCS2 (56). Thus, it appears that SOCS2 can both positively and negatively regulate GH signaling, an observation that has also been made in vitro (113). SOCS2 has been shown to interact with the GH receptor and may do so at multiple sites within the intracellular domain (56). As discussed above, one model for the dual activity of SOCS2 in GH regulation invokes that at physiological concentrations SOCS2 prevents STAT5 activation, thus inhibiting signaling, but at very high levels it can preclude access of other important negative regulators to the GH receptor complex. STAT5 is known to play an essential role in immune cell development (109). If SOCS2 is a key regulator of STAT5 activation by GH, an obvious corollary is that SOCS2 may also regulate activation of STAT5 by other receptor systems, including those operating in lymphocytes. However, to date, no anomalies in lymphocyte development or function have been reported in either Socs2−/− or Socs2 transgenic mice.

SOCS and Innate Immunity It is clear from the foregoing discussion that the study of SOCS proteins in immunity has primarily focused on the adaptive immune response, and clearly several SOCS proteins have a central and indispensable role in the regulation of this arm of host defense, primarily via attenuation of the actions of cytokines that influence T cell development and function. The phagocytes of the innate immune system are also regulated by several cytokines that are controlled by SOCS proteins, including, for example, IL-12 and the IFNs. It therefore follows that SOCS proteins make an important contribution to the regulation of the innate immune response. Indeed, macrophages lacking SOCS1 show enhanced killing of intracellular Leishmania parasites in cultures stimulated with LPS and IFNγ (74). The possibility that SOCS proteins are directly induced by microbial infection has been raised in recent studies. Exposure to CpG-DNA triggers expression of SOCS1 and SOCS3 in macrophages and dendritic cells via a pathway that does not require protein synthesis and appears independent of JAK-STAT signaling (114). Similarly, exposure of macrophages to Leishmania donovani or Listeria monocytogenes appears to directly induce expression of SOCS3 (26, 115). These observations have provoked speculation that SOCS expression may be induced directly via signals from the Toll receptor family.

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SOCS PROTEINS AND THE IMMUNE RESPONSE

517

LPS is a component of bacterial cell walls that binds the Toll receptor TLR4 and induces expression of SOCS1 and SOCS3, although at least some component of this activity may be indirect and due to LPS-induced autocrine factors such as IFNs (28–31). Nevertheless, Socs1+/− or Socs1−/− mice showed dramatically increased sensitivity to the lethal effects of LPS. Induction of nitric oxide synthesis and TNFα production were enhanced in SOCS1-deficient mice exposed to LPS, and LPS tolerance was significantly reduced (30, 31). Phosphorylation of NF-κB and MAP kinase was increased in LPS-stimulated Socs1−/− macrophages relative to wild-type control cells, and enforced SOCS1 expression blocked LPSmediated activation of NF-κB. These data were interpreted to indicate that SOCS1 is induced by LPS and feeds back to inhibit signals from TLR4 in a direct inhibitory loop. However, until the specific proteins in the TLR signaling cascade that are the targets of SOCS1 inhibition are identified, this model remains somewhat conjectural. Unlike Socs1−/− mice, mice in which the Socs3 gene has been ablated in macrophages show resistance to challenge with LPS. Enhanced inhibition of macrophage activation by IL-6 in macrophages lacking SOCS3 has been proposed to account for the reduced sensitivity in these mice (101), but the precise role of SOCS proteins in LPS responses clearly remains enigmatic. The prospect that SOCS proteins control signaling from chemokine receptors has also emerged in recent studies. Upon ligand binding, chemokine receptors associate with guanine nucleotide-binding proteins (G proteins) resulting in activation of signaling pathways commonly associated with regulation of the cytoskeleton. However, it is clear that at least some chemokine receptors also activate JAK kinases and recruit and activate members of the STAT family (116, 117). In a human B cell line, stimulation with CXCL12, which acts through the CXCR4 receptor, induced expression of SOCS3 via a JAK kinase-dependent and G protein–independent mechanism (118). Overexpression of SOCS1 or SOCS3, but not SOCS2, inhibited migration of fibroblasts in response to a gradient of CXCL12, but had no effect on migration stimulated by CCL20 binding to CCR6. Biochemical studies indicated that SOCS1 and SOCS3 could associate with the CXCR4 receptor and the complex appeared to be stabilized by CXCL12 ligand binding (118). Thus, in addition to the well-established role of SOCS proteins as classical negative feedback inhibitors of signaling from the hemopoietin class cytokine receptors, the SOCS may also be part of feedback regulation of distinct receptor classes including the Toll and chemokine receptors. Given the central role that these receptor systems play in modulating innate immune responses, dissection of the specific roles of SOCS proteins in innate immunity and diseases where these responses are disrupted is an important endeavor.

SOCS PROTEINS IN DISEASE Cytokines play a pivotal role in the development and pathology of human disease, including diseases of the immune system. As it is now clear that SOCS proteins have profound physiological actions, it seems inevitable that disruption of normal

11 Feb 2004

22:3

518

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

SOCS function will contribute to disease onset and progression. Moreover, the potent inhibitory actions of SOCS proteins on cytokine signaling raise the exciting possibility that the SOCS proteins may prove to be excellent targets for the discovery of drugs that can manipulate cytokine outcomes to resolve disease.

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SOCS Proteins in Malignancy It is well documented that a number of hematological malignancies are characterized by constitutive activation of the JAK-STAT pathway (119). A well-studied example is the chromosomal translocation t (9, 12) (p24;p13), which fuses the TEL gene to the sequence encoding the kinase domain of JAK2, resulting in expression of a fusion protein with deregulated kinase activity. This translocation has been described in a portion of patients with T cell acute lymphoblastic leukemia (ALL), pre-B ALL, and atypical chronic myelogenous leukemia (120, 121). Recent data suggest that SOCS1 can inhibit transformation of cell lines expressing the TEL-JAK fusion protein. SOCS1 inhibited kinase activity and autophosphorylation of TEL-JAK2, but the inhibitory effect of SOCS1 was also dependent on the SOCS box domain, and it was demonstrated that expression of SOCS1 induced proteasomal degradation of the oncogenic fusion protein (62, 122). Significantly, coexpression of SOCS1 with TEL-JAK2 in primary murine hematopoietic progenitor cells prolonged the latency of leukemia in a transplantation model of leukemia (122). Ectopic expression of SOCS1 has also been shown to suppress proliferation induced by activated forms of the c-Kit receptor and v-Abl protein as well as suppressing metastasis of cells transformed by the Bcr-abl fusion oncoprotein (123). Socs1−/− fibroblasts were also found to be more sensitive than wild-type controls to transformation mediated by these tyrosine kinase oncogenes, as well as exhibiting enhanced spontaneous transformation (123). In a similar context, levels of SOCS2 expression appear significantly higher in cells from patients in CML blast crisis relative to that observed in the chronic phase of the disease, and the level of SOCS2 was downmodulated upon exposure of cells to drugs that inhibit the Bcr-abl oncoprotein (124). Overexpression of SOCS2 in cells expressing Bcr-abl led to diminution of the transformed phenotype, suggesting that disruption of a negative regulatory loop involving SOCS2 may contribute to CML. These studies raise the exciting possibility that SOCS proteins may prove useful in treatment of malignancies in which activation of specific tyrosine kinase oncoproteins plays a central role. The role of SOCS1 as a potential tumor suppressor has also been raised in analyses of hepatocellular carcinoma. Aberrant DNA methylation at the SOCS1 locus resulting in transcriptional silencing has been observed in a large portion of human primary hepatocellular carcinomas and hepatoblastomas (125–128). Restoration of SOCS1 expression in cells in which the SOCS1 gene was silenced led to a reduction in the transformed phenotype (125). Methylation and silencing at the SOCS1 locus has now also been associated with significant numbers of patients with multiple myeloma or acute myeloid leukemia (129, 130).

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SOCS PROTEINS AND THE IMMUNE RESPONSE

519

Conversely, in cell lines generated from cutaneous T cell lymphoma and in chronic myeloid leukemia, constitutive SOCS3 expression has been documented (20, 131). Although the basis for this apparent deregulation of SOCS3 expression and any role it may play in disease etiology remain unresolved, in both cases it was observed that excessive SOCS3 expression correlated with resistance to IFNα. Because insensitivity to IFN has been observed in subsets of patients with several leukemias, as well as in the treatment of hepatitis, these early studies raise the possibility that specific patterns of SOCS gene expression in individual patients or tumors may influence the impact of IFN therapy and that therapeutic strategies designed to inhibit SOCS action may prove beneficial in this context.

SOCS Proteins in Inflammatory Disease Activation of STAT3 is a hallmark of animal models of colitis as well as human Crohn’s disease and ulcerative colitis, and concomitant constitutive expression of SOCS3 has been documented in these inflammatory diseases (132, 133). In a murine model of induced colitis in mice, transgenic mice expressing a dominantnegative mutant form of SOCS1 exhibited increased expression of STAT3 and suffered a more severe colitis than wild-type control mice (132). Because the mutant form of SOCS1 expressed in these mice interferes both with endogenous SOCS1 and SOCS3 activity, these data are consistent with the potential role for SOCS3 in regulating cytokines that contribute to inflammation in the bowel. Similarly, activation of STAT3 and deregulated SOCS3 expression have also been observed in the joint tissue of patients with rheumatoid arthritis (134). In murine models of antigen- or collagen-induced arthritis, periarticular administration of an adenovirus producing SOCS3 led to reduced inflammation and joint swelling and significantly reduced cartilage and bone destruction (134). As in the colitis models, it is proposed that the inflammatory cytokines that drive pathology act through STAT3 activation, and SOCS3 is induced in an attempt to control these signaling cascades. Because increasing local SOCS3 concentrations via gene therapy appeared to tip the balance between disease progression and suppression, therapeutic strategies that target SOCS3 expression or activity in patients with inflammatory diseases may prove effective. Although most attention has focused on SOCS3 in inflammatory disease, recent studies have also demonstrated that mice lacking SOCS1 display increased synovial inflammation and joint destruction in an IL-1-driven model of experimental arthritis that is dependent on a number of cytokines including GM-CSF, TNF-α, and IL-6 (135). In these animals, SOCS1 appeared to act in synovial macrophages and fibroblasts to limit inflammation and joint destruction, as well as regulating T cell proliferation (135). The observation that SOCS3 appears to be selectively expressed in Th2 cells has led to an examination of the role of SOCS3 in allergies such as atopic asthma, which is characterized by extensive infiltration of the airways by T cells expressing Th2 cytokines. A positive correlation was evident between SOCS3 expression

11 Feb 2004

22:3

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

520

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

and asthma pathology, as well as serum IgE levels in patients with allergy (136). Socs3+/− mice or transgenic mice expressing a dominant-negative form of SOCS3 exhibited decreased Th2 development. Conversely, transgenic mice constitutively expressing the wild-type SOCS3 protein in splenic T cells showed increased Th2 responses and in a mouse airway hypersensitivity model of asthma exhibited enhanced pathological features (136). These observations support a role for SOCS3 as a regulator of Th2 development and suggest that modulation of SOCS3 may represent a worthwhile therapeutic strategy in immunological diseases characterized by a Th1/Th2 imbalance. SOCS proteins have also been implicated in inflammatory diseases of the skin. For example, biopsies taken from patients with psoriasis or allergic contact dermatitis exhibited high levels of SOCS1, SOCS2, and SOCS3 expression, whereas these proteins were not detected in healthy skin (137). Together, these observations provide a promising basis for pursuing the contribution of SOCS proteins to inflammatory and other immune disorders. However, in complex inflammatory diseases the contribution to pathology and host responses by a range of cytokines is likely to be reflected by the induction and action of multiple SOCS proteins. Dissecting the crucial regulators of disease outcome from those SOCS proteins simply expressed as a secondary consequence of disease is an important experimental challenge.

CONCLUSION Cytokines are crucial to maintaining health and play an important role in the onset and progression of disease. Since the discovery of the SOCS proteins, researchers have recognized that negative feedback regulation of signal transduction also plays a central role in balancing the positive and deleterious consequences of cytokine action. Despite the progress that has been made in understanding how cytokine signalling is controlled, many challenges remain. For example, how does exposure to one cytokine modify a cell’s response to a subsequent stimuli? Do SOCS proteins act in concert to regulate signalling? Can modulation of SOCS protein production or activity lead to beneficial clinical outcomes? Addressing these questions is likely to require the coordinated efforts of chemists, biochemists, cell biologists, physiologists, and clinicians for many years to come. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Nicola NA. 1994. An introduction to the cytokines. In Guidebook to Cytokines and Their Receptors, ed. NA Nicola. New York: Oxford Univ. Press

2. Hilton DJ. 1994. An introduction to cytokine receptors. In Guidebook to Cytokines and Their Receptors, ed. NA Nicola. New York: Oxford Univ. Press

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SOCS PROTEINS AND THE IMMUNE RESPONSE 3. Gadina M, Hilton D, Johnston JA, Morinobu A, Lighvani A, et al. 2001. Signaling by type I and II cytokine receptors: ten years after. Curr. Opin. Immunol. 13:363–73 4. Bazan JF. 1990. Haemopoietic receptors and helical cytokines. Immunol. Today. 11:350–54 5. Ihle JN, Kerr IM. 1995. Jaks and Stats in signalling by the cytokine receptor superfamily. Trends Genet. 11:69–74 6. Darnell JE Jr, Kerr IM, Stark GR. 1994. Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264:1415–21 7. Darnell JE Jr. 1997. STATs and gene regulation. Science 277:1630–35 8. Yoshimura A, Ohkubo T, Kiguchi T, Jenkins NA, Gilbert DJ, et al. 1995. A novel cytokine-inducible gene CIS encodes an SH2-containing protein that binds to tyrosine-phosphorylated interleukin 3 and erythropoietin receptors. EMBO J. 14:2816–26 9. Starr R, Willson TA, Viney EM, Murray LJ, Rayner JR, et al. 1997. A family of cytokine-inducible inhibitors of signalling. Nature 387:917–21 10. Endo TA, Masuhara M, Yokouchi M, Suzuki R, Sakamoto H, et al. 1997. A new protein containing an SH2 domain that inhibits JAK kinases. Nature 387:921–24 11. Naka T, Narazaki M, Hirata M, Matsumoto T, Minamoto S, et al. 1997. Structure and function of a new STAT-induced STAT inhibitor. Nature 387:924–29 12. Minamoto S, Ikegame K, Ueno K, Narazaki M, Naka T, et al. 1997. Cloning and functional analysis of new members of STAT induced STAT inhibitor (SSI) family: SSI-2 and SSI-3. Biochem. Biophys. Res. Commun. 237:79–83 13. Hilton DJ, Richardson RT, Alexander WS, Viney EM, Willson TA, et al. 1998. Twenty proteins containing a C-terminal SOCS box form five structural classes. Proc. Natl. Acad. Sci. USA 95:114–19

521

14. Kile BT, Schulman BA, Alexander WS, Nicola NA, Martin HM, Hilton DJ. 2002. The SOCS box: a tale of destruction and degradation. Trends Biochem. Sci. 27:235–41 15. Jegalian AG, Wu H. 2002. Regulation of Socs gene expression by the proto-oncoprotein GFI-1B: two routes for STAT5 target gene induction by erythropoietin. J. Biol. Chem. 277:2345–52 16. Gregorieff A, Pyronnet S, Sonenberg N, Veillette A. 2000. Regulation of SOCS-1 expression by translational repression. J. Biol. Chem. 275:21596–604 17. Schluter G, Boinska D, Nieman-Seyde SC. 2000. Evidence for translational repression of the SOCS-1 major open reading frame by an upstream open reading frame. Biochem. Biophys. Res. Commun. 268:255–61 18. He B, You L, Uematsu K, Matsangou M, Xu Z, et al. 2003. Cloning and characterization of a functional promoter of the human SOCS-3 gene. Biochem. Biophys. Res. Commun. 301:386–91 19. Donnelly RP, Dickensheets H, Finbloom DS. 1999. The interleukin-10 signal transduction pathway and regulation of gene expression in mononuclear phagocytes. J. Interferon Cytokine Res. 19:563–73 20. Brender C, Nielsen M, Kaltoft K, Mikkelsen G, Zhang Q, et al. 2001. STAT3-mediated constitutive expression of SOCS-3 in cutaneous T-cell lymphoma. Blood 97:1056–62 21. Matsumoto A, Masuhara M, Mitsui K, Yokouchi M, Ohtsubo M, et al. 1997. CIS, a cytokine inducible SH2 protein, is a target of the JAK-STAT5 pathway and modulates STAT5 activation. Blood 89:3148– 54 22. Verdier F, Rabionet R, Gouilleux F, Beisenherz-Huss C, Varlet P, et al. 1998. A sequence of the CIS gene promoter interacts preferentially with two associated STAT5A dimers: a distinct biochemical difference between STAT5A and STAT5B. Mol. Cell Biol. 18:5852–60

11 Feb 2004

22:3

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

522

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

23. Davey HW, McLachlan MJ, Wilkins RJ, Hilton DJ, Adams TE. 1999. STAT5b mediates the GH-induced expression of SOCS-2 and SOCS-3 mRNA in the liver. Mol. Cell Endocrinol. 158:111–16 24. Feldman GM, Rosenthal LA, Liu X, Hayes MP, Wynshaw-Boris A, et al. 1997. STAT5A-deficient mice demonstrate a defect in granulocyte-macrophage colonystimulating factor-induced proliferation and gene expression. Blood 90:1768–76 25. Lee CK, Raz R, Gimeno R, Gertner R, Wistinghausen B, et al. 2002. STAT3 is a negative regulator of granulopoiesis but is not required for G-CSF-dependent differentiation. Immunity 17:63–72 26. Stoiber D, Stockinger S, Steinlein P, Kovarik J, Decker T. 2001. Listeria monocytogenes modulates macrophage cytokine responses through STAT serine phosphorylation and the induction of suppressor of cytokine signaling 3. J. Immunol. 166:466–72 27. Gautron L, Lafon P, Tramu G, Laye S. 2003. In vivo activation of the interleukin-6 receptor/gp130 signaling pathway in pituitary corticotropes of lipopolysaccharide-treated rats. Neuroendocrinology 77:32–43 28. Crespo A, Filla MB, Russell SW, Murphy WJ. 2000. Indirect induction of suppressor of cytokine signalling-1 in macrophages stimulated with bacterial lipopolysaccharide: partial role of autocrine/paracrine interferon-alpha/beta. Biochem. J. 349:99–104 29. Stoiber D, Kovarik P, Cohney S, Johnston JA, Steinlein P, Decker T. 1999. Lipopolysaccharide induces in macrophages the synthesis of the suppressor of cytokine signaling 3 and suppresses signal transduction in response to the activating factor IFN-gamma. J. Immunol. 163:2640–47 30. Nakagawa R, Naka T, Tsutsui H, Fujimoto M, Kimura A, et al. 2002. SOCS-1 participates in negative regulation of LPS responses. Immunity 17:677–87

31. Kinjyo I, Hanada T, Inagaki-Ohara K, Mori H, Aki D, et al. 2002. SOCS1/ JAB is a negative regulator of LPSinduced macrophage activation. Immunity 17:583–91 32. Yasukawa H, Misawa H, Sakamoto H, Masuhara M, Sasaki A, et al. 1999. The JAK-binding protein JAB inhibits Janus tyrosine kinase activity through binding in the activation loop. EMBO J. 18:1309– 20 33. Sporri B, Kovanen PE, Sasaki A, Yoshimura A, Leonard WJ. 2001. JAB/SOCS1/SSI-1 is an interleukin-2induced inhibitor of IL-2 signaling. Blood 97:221–26 34. Nicholson SE, Willson TA, Farley A, Starr R, Zhang JG, et al. 1999. Mutational analyses of the SOCS proteins suggest a dual domain requirement but distinct mechanisms for inhibition of LIF and IL-6 signal transduction. EMBO J. 18:375–85 35. Fairlie WD, De Souza D, Nicola NA, Baca M. 2003. Negative regulation of gp130 signalling mediated through tyrosine-757 is not dependent on the recruitment of SHP2. Biochem. J. 372:495–502 36. De Souza D, Fabri LJ, Nash A, Hilton DJ, Nicola NA, Baca M. 2002. SH2 domains from suppressor of cytokine signaling-3 and protein tyrosine phosphatase SHP-2 have similar binding specificities. Biochemistry 41:9229–36 37. Nicholson SE, De Souza D, Fabri LJ, Corbin J, Willson TA, et al. 2000. Suppressor of cytokine signaling-3 preferentially binds to the SHP-2-binding site on the shared cytokine receptor subunit gp130. Proc. Natl. Acad. Sci. USA 97:6493–98 38. Schmitz J, Weissenbach M, Haan S, Heinrich PC, Schaper F. 2000. SOCS3 exerts its inhibitory function on interleukin6 signal transduction through the SHP2 recruitment site of gp130. J. Biol. Chem. 275:12848–56 39. Lehmann U, Schmitz J, Weissenbach M, Sobota RM, Hortner M, et al. 2003.

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

SOCS PROTEINS AND THE IMMUNE RESPONSE

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

40.

41.

42.

43.

44.

45.

46.

47.

SHP2 and SOCS3 contribute to Tyr-759dependent attenuation of interleukin-6 signaling through gp130. J. Biol. Chem. 278:661–71 Friederichs K, Schmitz J, Weissenbach M, Heinrich PC, Schaper F. 2001. Interleukin-6-induced proliferation of pre-B cells mediated by receptor complexes lacking the SHP2/SOCS3 recruitment sites revisited. Eur. J. Biochem. 268:6401–7 Bjorbak C, Lavery HJ, Bates SH, Olson RK, Davis SM, et al. 2000. SOCS3 mediates feedback inhibition of the leptin receptor via Tyr985. J. Biol. Chem. 275:40649–57 Eyckerman S, Broekaert D, Verhee A, Vandekerckhove J, Tavernier J. 2000. Identification of the Y985 and Y1077 motifs as SOCS3 recruitment sites in the murine leptin receptor. FEBS Lett. 486:33–37 Hermans MH, van de Geijn GJ, Antonissen C, Gits J, van Leeuwen D, et al. 2003. Signaling mechanisms coupled to tyrosines in the granulocyte colonystimulating factor receptor orchestrate GCSF-induced expansion of myeloid progenitor cells. Blood 101:2584–90 Hortner M, Nielsch U, Mayr LM, Johnston JA, Heinrich PC, Haan S. 2002. Suppressor of cytokine signaling-3 is recruited to the activated granulocytecolony stimulating factor receptor and modulates its signal transduction. J. Immunol. 169:1219–27 Hansen JA, Lindberg K, Hilton DJ, Nielsen JH, Billestrup N. 1999. Mechanism of inhibition of growth hormone receptor signaling by suppressor of cytokine signaling proteins. Mol. Endocrinol. 13:1832–43 Ram PA, Waxman DJ. 1999. SOCS/CIS protein inhibition of growth hormonestimulated STAT5 signaling by multiple mechanisms. J. Biol. Chem. 274:35553– 61 Hortner M, Nielsch U, Mayr LM, Hein-

48.

49.

50.

51.

52.

53.

54.

55.

523

rich PC, Haan S. 2002. A new high affinity binding site for suppressor of cytokine signaling-3 on the erythropoietin receptor. Eur. J. Biochem. 269:2516–26 Sasaki A, Yasukawa H, Shouda T, Kitamura T, Dikic I, Yoshimura A. 2000. CIS3/SOCS-3 suppresses erythropoietin (EPO) signaling by binding the EPO receptor and JAK2. J. Biol. Chem. 275: 29338–47 Cohney SJ, Sanden D, Cacalano NA, Yoshimura A, Mui A, et al. 1999. SOCS-3 is tyrosine phosphorylated in response to interleukin-2 and suppresses STAT5 phosphorylation and lymphocyte proliferation. Mol. Cell Biol. 19:4980–88 Lejeune D, Demoulin JB, Renauld JC. 2001. Interleukin 9 induces expression of three cytokine signal inhibitors: cytokineinducible SH2-containing protein, suppressor of cytokine signalling (SOCS)-2 and SOCS-3, but only SOCS-3 overexpression suppresses interleukin 9 signalling. Biochem. J. 353:109–16 Losman J, Chen XP, Jiang H, Pan PY, Kashiwada M, et al. 1999. IL-4 signaling is regulated through the recruitment of phosphatases, kinases, and SOCS proteins to the receptor complex. Cold Spring Harbor Symp. Quant. Biol. 64:405–16 Eyckerman S, Verhee A, der Heyden JV, Lemmens I, Ostade XV, et al. 2001. Design and application of a cytokinereceptor-based interaction trap. Nat. Cell Biol. 3:1114–19 Dif F, Saunier E, Demeneix B, Kelly PA, Edery M. 2001. Cytokine-inducible SH2-containing protein suppresses PRL signaling by binding the PRL receptor. Endocrinology 142:5286–93 Aman MJ, Migone T-S, Sasaki A, Ascherman DP, Zhu M, et al. 1999. CIS associates with the interleukin-2 receptor β chain and inhibits interleukin2-dependent signaling. J. Biol. Chem. 274:30266–72 Pezet A, Favre H, Kelly PA, Edery M. 1999. Inhibition and restoration of

11 Feb 2004

22:3

524

56.

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

57.

58.

59.

60.

61.

62.

63.

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

prolactin signal transduction by suppressors of cytokine signaling. J. Biol. Chem. 274:24497–502 Greenhalgh CJ, Metcalf D, Thaus AL, Corbin JE, Uren R, et al. 2002. Biological evidence that SOCS-2 can act either as an enhancer or suppressor of growth hormone signaling. J. Biol. Chem. 277:40181–14 Narazaki M, Fujimoto M, Matsumoto T, Morita Y, Saito H, et al. 1998. Three distinct domains of SSI-1/SOCS-1/JAB protein are required for its suppression of interleukin 6 signaling. Proc. Natl. Acad. Sci. USA 95:13130–34 Zhang JG, Farley A, Nicholson SE, Willson TA, Zugaro LM, et al. 1999. The conserved SOCS box motif in suppressors of cytokine signaling binds to elongins B and C and may couple bound proteins to proteasomal degradation. Proc. Natl. Acad. Sci. USA 96:2071–76 Kamura T, Sato S, Haque D, Liu L, Kaelin WG, et al. 1998. The elongin BC complex interacts with the conserved SOCS-box motif present in members of the SOCS, ras, WD-40 repeat, and ankyrin repeat families. Genes Dev. 12:3872–81 Ungureanu D, Saharinen P, Junttila I, Hilton DJ, Silvennoinen O. 2002. Regulation of Jak2 through the ubiquitinproteasome pathway involves phosphorylation of Jak2 on Y1007 and interaction with SOCS-1. Mol. Cell Biol. 22:3316– 26 Monni R, Santos SC, Mauchauffe M, Berger R, Ghysdael J, et al. 2001. The TEL-Jak2 oncoprotein induces Socs1 expression and altered cytokine response in Ba/F3 cells. Oncogene 20:849–58 Kamizono S, Hanada T, Yasukawa H, Minoguchi S, Kato R, et al. 2001. The SOCS box of SOCS-1 accelerates ubiquitin-dependent proteolysis of TELJAK2. J. Biol. Chem. 276:12530–38 Zhang JG, Metcalf D, Rakar S, Asimakis M, Greenhalgh CJ, et al. 2001. The SOCS box of suppressor of cytokine signaling-1

64.

65.

66.

67.

68.

69.

70.

71.

72.

is important for inhibition of cytokine action in vivo. Proc. Natl. Acad. Sci. USA 98:13261–65 Alexander WS. 2002. Suppressors of cytokine signalling (SOCS) in the immune system. Nat. Rev. Immunol. 2:410–16 Krebs DL, Hilton DJ. 2001. SOCS proteins: negative regulators of cytokine signaling. Stem Cells 19:378–87 Yoshimura A. 1998. The CIS/JAB family: novel negative regulators of JAK signaling pathways. Leukemia 12:1851–57 Naka T, Matsumoto T, Narazaki M, Fujimoto M, Morita Y, et al. 1998. Accelerated apoptosis of lymphocytes by augmented induction of Bax in SSI-1 (STATinduced STAT inhibitor-1) deficient mice. Proc. Natl. Acad. Sci. USA 95:15577–82 Starr R, Metcalf D, Elefanty AG, Brysha M, Willson TA, et al. 1998. Liver degeneration and lymphoid deficiencies in mice lacking suppressor of cytokine signaling1. Proc. Natl. Acad. Sci. USA 95:14395– 99 Kawazoe Y, Naka T, Fujimoto M, Kohzaki H, Morita Y, et al. 2001. Signal transducer and activator of transcription (STAT)-induced STAT inhibitor 1 (SSI1)/suppressor of cytokine signaling 1 (SOCS1) inhibits insulin signal transduction pathway through modulating insulin receptor substrate 1 (IRS-1) phosphorylation. J. Exp. Med. 193:263–69 Lindeman GJ, Wittlin S, Lada H, Naylor MJ, Santamaria M, et al. 2001. SOCS1 deficiency results in accelerated mammary gland development and rescues lactation in prolactin receptor-deficient mice. Genes Dev. 15:1631–36 Cornish AL, Davey GM, Metcalf D, Purton JF, Corbin JE, et al. 2003. Suppressor of cytokine signaling-1 has IFN-gammaindependent actions in T cell homeostasis. J. Immunol. 170:878–86 Marine JC, Topham DJ, McKay C, Wang D, Parganas E, et al. 1999. SOCS1 deficiency causes a lymphocyte-dependent perinatal lethality. Cell 98:609–16

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SOCS PROTEINS AND THE IMMUNE RESPONSE 73. Gresser I, Tovey MG, Maury C, Chouroulinkov I. 1975. Lethality of interferon preparations for newborn mice. Nature 258:76–78 74. Alexander WS, Starr R, Fenner JE, Scott CL, Handman E, et al. 1999. SOCS1 is a critical inhibitor of interferon gamma signaling and prevents the potentially fatal neonatal actions of this cytokine. Cell 98:597–608 75. Metcalf D, Di Rago L, Mifsud S, Hartley L, Alexander WS. 2000. The development of fatal myocarditis and polymyositis in mice heterozygous for IFN-gamma and lacking the SOCS-1 gene. Proc. Natl. Acad. Sci. USA 97:9174–79 76. Metcalf D, Alexander WS, Elefanty AG, Nicola NA, Hilton DJ, et al. 1999. Aberrant hematopoiesis in mice with inactivation of the gene encoding SOCS-1. Leukemia 13:926–34 77. Brysha M, Zhang JG, Bertolino P, Corbin JE, Alexander WS, et al. 2001. Suppressor of cytokine signaling-1 attenuates the duration of interferon gamma signal transduction in vitro and in vivo. J. Biol. Chem. 276:22086–89 78. Grawunder U, Melchers F, Rolink A. 1993. Interferon-gamma arrests proliferation and causes apoptosis in stromal cell/interleukin-7-dependent normal murine pre-B cell lines and clones in vitro, but does not induce differentiation to surface immunoglobulinpositive B cells. Eur. J. Immunol. 23:544– 51 79. Matsuda T, Yamamoto T, Kishi H, Yoshimura A, Muraguchi A. 2000. SOCS1 can suppress CD3zeta- and Sykmediated NF-AT activation in a nonlymphoid cell line. FEBS Lett. 472:235– 40 80. Naka T, Tsutsui H, Fujimoto M, Kawazoe Y, Kohzaki H, et al. 2001. SOCS-1/SSI-1deficient NKT cells participate in severe hepatitis through dysregulated cross-talk inhibition of IFN-gamma and IL-4 signaling in vivo. Immunity 14:535–45

525

81. Eyles JL, Metcalf D, Grusby MJ, Hilton DJ, Starr R. 2002. Negative regulation of interleukin-12 signaling by suppressor of cytokine signaling-1. J. Biol. Chem. 277:43735–40 82. Chong MM, Thomas HE, Kay TW. 2002. Suppressor of cytokine signaling-1 regulates the sensitivity of pancreatic beta cells to tumor necrosis factor. J. Biol. Chem. 277:27945–52 83. Morita Y, Naka T, Kawazoe Y, Fujimoto M, Narazaki M, et al. 2000. Signals transducers and activators of transcription (STAT)-induced STAT inhibitor-1 (SSI-1)/suppressor of cytokine signaling1 (SOCS-1) suppresses tumor necrosis factor alpha-induced cell death in fibroblasts. Proc. Natl. Acad. Sci. USA 97:5405– 10 84. Chong MM, Cornish AL, Darwiche R, Stanley EG, Purton JF, et al. 2003. Suppressor of cytokine signaling-1 is a critical regulator of interleukin-7-dependent CD8+ T cell differentiation. Immunity 18:475–87 85. Metcalf D, Mifsud S, Di Rago L, Alexander WS. 2003. The lethal effects of transplantation of Socs1−/− bone marrow cells into irradiated adult syngeneic recipients. Proc. Natl. Acad. Sci. USA 100:8436–41 86. Trop S, De Sepulveda P, Zuniga-Pflucker JC, Rottapel R. 2001. Overexpression of suppressor of cytokine signaling-1 impairs pre-T-cell receptor-induced proliferation but not differentiation of immature thymocytes. Blood 97:2269–77 87. Fujimoto M, Naka T, Nakagawa R, Kawazoe Y, Morita Y, et al. 2000. Defective thymocyte development and perturbed homeostasis of T cells in STAT-induced STAT inhibitor-1/suppressors of cytokine signaling-1 transgenic mice. J. Immunol. 165:1799–806 88. Jameson SC. 2002. Maintaining the norm: T-cell homeostasis. Nat. Rev. Immunol. 2:547–56 89. Cornish AL, Chong MM, Davey GM, Darwiche R, Nicola NA, et al. 2003.

11 Feb 2004

22:3

526

90.

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

91.

92.

93.

94.

95.

96.

97.

98.

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

SOCS-1 regulates signaling in response to IL-2 and other gamma c-dependent cytokines in peripheral T cells. J. Biol. Chem. 278:22755–61 Losman JA, Chen XP, Hilton D, Rothman P. 1999. Cutting edge: SOCS-1 is a potent inhibitor of IL-4 signal transduction. J. Immunol. 162:3770–74 Metcalf D, Mifsud S, Di Rago L, Nicola NA, Hilton DJ, Alexander WS. 2002. Polycystic kidneys and chronic inflammatory lesions are the delayed consequences of loss of the suppressor of cytokine signaling-1 (SOCS-1). Proc. Natl. Acad. Sci. USA 99:943–48 Egwuagu CE, Yu CR, Zhang M, Mahdi RM, Kim SJ, Gery I. 2002. Suppressors of cytokine signaling proteins are differentially expressed in Th1 and Th2 cells: implications for Th cell lineage commitment and maintenance. J. Immunol. 168:3181– 87 Fujimoto M, Tsutsui H, YumikuraFutatsugi S, Ueda H, Xingshou O, et al. 2002. A regulatory role for suppressor of cytokine signaling-1 in T(h) polarization in vivo. Int. Immunol. 14:1343–50 Bullen DV, Baldwin TM, Curtis JM, Alexander WS, Handman E. 2003. Persistence of lesions in suppressor of cytokine signaling-1-deficient mice infected with Leishmania major. J. Immunol. 170:4267–72 Diehl S, Anguita J, Hoffmeyer A, Zapton T, Ihle JN, et al. 2000. Inhibition of Th1 differentiation by IL-6 is mediated by SOCS1. Immunity 13:805–15 Marine JC, McKay C, Wang D, Topham DJ, Parganas E, et al. 1999. SOCS3 is essential in the regulation of fetal liver erythropoiesis. Cell 98:617–27 Roberts AW, Robb L, Rakar S, Hartley L, Cluse L, et al. 2001. Placental defects and embryonic lethality in mice lacking suppressor of cytokine signaling 3. Proc. Natl. Acad. Sci. USA 98:9324–29 Bischof P, Haenggeli L, Campana A. 1995. Effect of leukemia inhibitory fac-

99.

100.

101.

102.

103.

104.

105.

106.

107.

tor on human cytotrophoblast differentiation along the invasive pathway. Am. J. Reprod. Immunol. 34:225–30 Takahashi Y, Carpino N, Cross JC, Torres M, Parganas E, Ihle JN. 2003. SOCS3: an essential regulator of LIF receptor signaling in trophoblast giant cell differentiation. EMBO J. 22:372–84 Croker BA, Krebs DL, Zhang JG, Wormald S, Willson TA, et al. 2003. SOCS3 negatively regulates IL-6 signaling in vivo. Nat. Immunol. 4:540–45 Yasukawa H, Ohishi M, Mori H, Murakami M, Chinen T, et al. 2003. IL-6 induces an anti-inflammatory response in the absence of SOCS3 in macrophages. Nat. Immunol 4:551–56 Lang R, Pauleau AL, Parganas E, Takahashi Y, Mages J, et al. 2003. SOCS3 regulates the plasticity of gp130 signaling. Nat. Immunol. 4:546–50 Song MM, Shuai K. 1998. The suppressor of cytokine signaling (SOCS) 1 and SOCS3 but not SOCS2 proteins inhibit interferon-mediated antiviral and antiproliferative activities. J. Biol. Chem. 273:35056–62 Costa-Pereira AP, Tininini S, Strobl B, Alonzi T, Schlaak JF, et al. 2002. Mutational switch of an IL-6 response to an interferon-gamma-like response. Proc. Natl. Acad. Sci. USA 99:8043–47 Yu CR, Mahdi RM, Ebong S, Vistica BP, Gery I, Egwuagu CE. 2003. Suppressor of cytokine signaling 3 (SOCS3) regulates proliferation and activation of T-helper cells. J. Biol. Chem. 278:29752–59 Matsumoto A, Seki Y, Watanabe R, Hayashi K, Johnston JA, et al. 2003. A role of suppressor of cytokine signaling 3 (SOCS3/CIS3/SSI3) in CD28-mediated interleukin 2 production. J. Exp. Med. 197:425–36 Seki Y, Hayashi K, Matsumoto A, Seki N, Tsukada J, et al. 2002. Expression of the suppressor of cytokine signaling5 (SOCS5) negatively regulates IL-4dependent STAT6 activation and Th2

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

SOCS PROTEINS AND THE IMMUNE RESPONSE

108.

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

109.

110.

111.

112.

113.

114.

115.

differentiation. Proc. Natl. Acad. Sci. USA 99:13003–8 Teglund S, McKay C, Schuetz E, van Deursen JM, Stravopodis D, et al. 1998. Stat5a and Stat5b proteins have essential and nonessential, or redundant, roles in cytokine responses. Cell 93:841–50 Matsumoto A, Seki Y, Kubo M, Ohtsuka S, Suzuki A, et al. 1999. Suppression of STAT5 functions in liver, mammary glands, and T cells in cytokineinducible SH2-containing protein 1 transgenic mice. Mol. Cell Biol. 19:6396–407 Li S, Chen S, Xu X, Sundstedt A, Paulsson KM, et al. 2000. Cytokine-induced Src homology 2 protein (CIS) promotes T cell receptor-mediated proliferation and prolongs survival of activated T cells. J. Exp. Med. 191:985–94 Metcalf D, Greenhalgh CJ, Viney E, Willson TA, Starr R, et al. 2000. Gigantism in mice lacking suppressor of cytokine signalling-2. Nature 405:1069–73 Greenhalgh CJ, Bertolino P, Asa SL, Metcalf D, Corbin JE, et al. 2002. Growth enhancement in suppressor of cytokine signaling 2 (SOCS-2)-deficient mice is dependent on signal transducer and activator of transcription 5b (STAT5b). Mol. Endocrinol. 16:1394–406 Favre H, Benhamou A, Finidori J, Kelly PA, Edery M. 1999. Dual effects of suppressor of cytokine signaling (SOCS-2) on growth hormone signal transduction. FEBS Lett. 453:63–66 Dalpke AH, Opper S, Zimmermann S, Heeg K. 2001. Suppressors of cytokine signaling (SOCS)-1 and SOCS-3 are induced by CpG-DNA and modulate cytokine responses in APCs. J. Immunol. 166:7082–89 Bertholet S, Dickensheets HL, Sheikh F, Gam AA, Donnelly RP, Kenney RT. 2003. Leishmania donovani-induced expression of suppressor of cytokine signaling 3 in human macrophages: a novel mechanism for intracellular parasite suppression of activation. Infect. Immun. 71:2095–101

527

116. Vila-Coro AJ, Rodriguez-Frade JM, Martin De Ana A, Moreno-Ortiz MC, Martinez AC, Mellado M. 1999. The chemokine SDF-1alpha triggers CXCR4 receptor dimerization and activates the JAK/STAT pathway. FASEB J. 13:1699– 710 117. Mellado M, Rodriguez-Frade JM, Aragay A, del Real G, Martin AM, et al. 1998. The chemokine monocyte chemotactic protein 1 triggers Janus kinase 2 activation and tyrosine phosphorylation of the CCR2B receptor. J. Immunol. 161:805–13 118. Soriano SF, Hernanz-Falcon P, Rodriguez-Frade JM, De Ana AM, Garzon R, et al. 2002. Functional inactivation of CXC chemokine receptor 4-mediated responses through SOCS3 up-regulation. J. Exp. Med. 196:311–21 119. Ward AC, Touw I, Yoshimura A. 2000. The Jak-Stat pathway in normal and perturbed hematopoiesis. Blood 95:19– 29 120. Lacronique V, Boureux A, Valle VD, Poirel H, Quang CT, et al. 1997. A TELJAK2 fusion protein with constitutive kinase activity in human leukemia. Science 278:1309–12 121. Peeters P, Raynaud SD, Cools J, Wlodarska I, Grosgeorge J, et al. 1997. Fusion of TEL, the ETS-variant gene 6 (ETV6), to the receptor-associated kinase JAK2 as a result of t(9;12) in a lymphoid and t(9;15;12) in a myeloid leukemia. Blood 90:2535–40 122. Frantsve J, Schwaller J, Sternberg DW, Kutok J, Gilliland DG. 2001. Socs-1 inhibits TEL-JAK2-mediated transformation of hematopoietic cells through inhibition of JAK2 kinase activity and induction of proteasome-mediated degradation. Mol. Cell Biol. 21:3547–57 123. Rottapel R, Ilangumaran S, Neale C, La Rose J, Ho JM, et al. 2002. The tumor suppressor activity of SOCS-1. Oncogene 21:4351–62 124. Schultheis B, Carapeti-Marootian M, Hochhaus A, Weisser A, Goldman JM,

11 Feb 2004

22:3

528

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

125.

126.

127.

128.

129.

130.

131.

132.

AR

AR210-IY22-17.tex

ALEXANDER

¥

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

HILTON

Melo JV. 2002. Overexpression of SOCS2 in advanced stages of chronic myeloid leukemia: possible inadequacy of a negative feedback mechanism. Blood 99:1766–75 Yoshikawa H, Matsubara K, Qian GS, Jackson P, Groopman JD, et al. 2001. SOCS-1, a negative regulator of the JAK/STAT pathway, is silenced by methylation in human hepatocellular carcinoma and shows growth-suppression activity. Nat. Genet. 28:29–35 Nagai H, Kim YS, Konishi N, Baba M, Kubota T, et al. 2002. Combined hypermethylation and chromosome loss associated with inactivation of SSI1/SOCS-1/JAB gene in human hepatocellular carcinomas. Cancer Lett. 186:59– 65 Nagai H, Naka T, Terada Y, Komazaki T, Yabe A, et al. 2003. Hypermethylation associated with inactivation of the SOCS1 gene, a JAK/STAT inhibitor, in human hepatoblastomas. J. Hum. Genet. 48:65– 69 Nagai H, Kim YS, Lee KT, Chu MY, Konishi N, et al. 2001. Inactivation of SSI1, a JAK/STAT inhibitor, in human hepatocellular carcinomas, as revealed by twodimensional electrophoresis. J. Hepatol. 34:416–21 Chen CY, Tsay W, Tang JL, Shen HL, Lin SW, et al. 2003. SOCS1 methylation in patients with newly diagnosed acute myeloid leukemia. Genes Chromosomes Cancer 37:300–5 Galm O, Yoshikawa H, Esteller M, Osieka R, Herman JG. 2003. SOCS-1, a negative regulator of cytokine signaling, is frequently silenced by methylation in multiple myeloma. Blood 101:2784–88 Sakai I, Takeuchi K, Yamauchi H, Narumi H, Fujita S. 2002. Constitutive expression of SOCS3 confers resistance to IFN-alpha in chronic myelogenous leukemia cells. Blood 100:2926–31 Suzuki A, Hanada T, Mitsuyama K, Yoshida T, Kamizono S, et al. 2001.

133.

134.

135.

136.

137.

138.

139.

140.

CIS3/SOCS3/SSI3 plays a negative regulatory role in STAT3 activation and intestinal inflammation. J. Exp. Med. 193:471– 81 Lovato P, Brender C, Agnholt J, Kelsen J, Kaltoft K, et al. 2003. Constitutive STAT3 activation in intestinal T cells from patients with Crohn’s disease. J. Biol. Chem. 278:16777–81 Shouda T, Yoshida T, Hanada T, Wakioka T, Oishi M, et al. 2001. Induction of the cytokine signal regulator SOCS3/CIS3 as a therapeutic strategy for treating inflammatory arthritis. J. Clin. Invest. 108:1781– 88 Egan PJ, Lawlor KE, Alexander WS, Wicks IP. 2003. Suppressor of cytokine signaling-1 regulates acute inflammatory arthritis and T cell activation. J. Clin. Invest. 111:915–24 Seki YI, Inoue H, Nagata N, Hayashi K, Fukuyama S, et al. 2003. SOCS-3 regulates onset and maintenance of T(H)2mediated allergic responses. Nat. Med. 9:1047–54 Federici M, Giustizieri ML, Scarponi C, Girolomoni G, Albanesi C. 2002. Impaired IFN-gamma-dependent inflammatory responses in human keratinocytes overexpressing the suppressor of cytokine signaling 1. J. Immunol. 169:434–42 Krebs DL, Uren RT, Metcalf D, Rakar S, Zhang JG, et al. 2002. SOCS-6 binds to insulin receptor substrate 4, and mice lacking the SOCS-6 gene exhibit mild growth retardation. Mol. Cell Biol. 22:4567– 78 Jones SA, Rose-John S. 2002. The role of soluble receptors in cytokine biology: the agonistic properties of the sIL6R/IL-6 complex. Biochim. Biophys. Acta 1592:251–63 Jiao H, Berrada K, Yang W, Tabrizi M, Platanias LC, Yi T. 1996. Direct association with and dephosphorylation of Jak2 kinase by the SH2-domain-containing protein tyrosine phosphatase SHP-1. Mol. Cell. Biol. 16:6985–92

11 Feb 2004

22:3

AR

AR210-IY22-17.tex

AR210-IY22-17.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SOCS PROTEINS AND THE IMMUNE RESPONSE 141. Ihle JN, Witthuhn BA, Quelle FW, Silvennoinen O, Tang B, Yi T. 1994. Protein tyrosine phosphorylation in the regulation of hematopoiesis by receptors of the cytokine-receptor superfamily. Blood Cells 20:65–80, discussion 2 142. Qu CK. 2002. Role of the SHP-2 tyrosine phosphatase in cytokine-induced signaling and cellular response. Biochim. Biophys. Acta 1592:297–301 143. Irie-Sasaki J, Sasaki T, Matsumoto W, Opavsky A, Cheng M, et al. 2001. CD45 is a JAK phosphatase and negatively regulates cytokine receptor signalling. Nature 409:349–54 144. Yamada T, Zhu D, Saxon A, Zhang K. 2002. CD45 controls interleukin-4mediated IgE class switch recombination

529

in human B cells through its function as a Janus kinase phosphatase. J. Biol. Chem. 277:28830–35 145. Yamamoto T, Sekine Y, Kashima K, Kubota A, Sato N, et al. 2002. The nuclear isoform of protein-tyrosine phosphatase TC-PTP regulates interleukin-6-mediated signaling pathway through STAT3 dephosphorylation. Biochem. Biophys. Res. Commun. 297:811–17 146. Aoki N, Matsuda T. 2002. A nuclear protein tyrosine phosphatase TC-PTP is a potential negative regulator of the PRLmediated signaling pathway: dephosphorylation and deactivation of signal transducer and activator of transcription 5a and 5b by TC-PTP in nucleus. Mol. Endocrinol. 16:58–69

3/12/2004 2:09 PM

Figure 1 Cytokine signal transduction is regulated at many levels. Cytokines act by binding to members of the hemopoietin receptor family and inducing their dimerization, resulting in cross-phosphorylation of noncovalently associated janus kinases (JAKs). JAKs phosphorylate tyrosine residues in the receptor cytoplasmic domain, creating docking sites for proteins such as the signal transducers and activators of transcription (STATs). Dimeric STATs migrate to the nucleus where they increase transcription of genes important for eliciting the biological effect of the cytokine. Signaling is controlled at many levels outside the cell and within the cell. Secreted and membrane associated receptors can act as antagonists by binding free cytokine (139). Activated receptors with bound cytokine are internalized and degraded, whereas activated signaling components can be dephosphorylated by a plethora of phosphatases including the SH2 domain containing phosphatases SHP1 (140, 141) and SHP2 (142), the transmembrane phosphatase CD45 (143, 144) and STAT phosphatases (145, 146).

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HI-RES-IY22-17-Alexander Page 1

SOCS PROTEINS AND THE IMMUNE RESPONSE C-1

HI-RES-IY22-17-Alexander

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

3/12/2004

ALEXANDER

■

2:09 PM

Page 2

HILTON

Figure 2 Structure of SOCS proteins. All eight members of the SOCS protein family contain an N-terminal region of varying length and sequence, a central SH2 domain, and a Cterminal SOCS box. SOCS proteins are known by a range of aliases: SOCS1 = SSI1 = JAB, SOCS2 = SSI2 = CIS2, SOCS3 = SSI3 = CIS3, SOCS4 = CIS7, SOCS5 = CIS6, SOCS6 = CIS4, SOCS7 = CIS5 = NAP4.

3/12/2004 2:09 PM

Figure 3 SOCS proteins act in a negative feedback loop to attenuate signaling. Upon binding to their receptor, cytokines activate the JAK/STAT pathway, resulting in an increase in the transcription of not only the genes mediating the biological effect of the cytokine, but also the SOCS genes. Once produced, SOCS proteins can inhibit signaling by binding to phosphorylated JAKs and receptors and can interact with components of E3 ubiquitin ligases to polyubiquitylate JAKs and target them for proteasomal degradation.

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HI-RES-IY22-17-Alexander Page 3

SOCS PROTEINS AND THE IMMUNE RESPONSE C-3

C-4

3/12/2004

ALEXANDER ■

2:09 PM

Figure 4 SOCS proteins induced by one stimulus can inhibit signaling by subsequent stimuli. Agents such as LPS can stimulate production of SOCS proteins. Once produced, these can inhibit signaling induced by subsequent stimuli, such as cytokines, by acting on phosphorylated receptors and JAKs.

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HI-RES-IY22-17-Alexander Page 4

HILTON

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:503-529. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.21.120601.141122

Annu. Rev. Immunol. 2004. 22:531–62 doi: 10.1146/annurev.immunol.21.120601.141122 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on November 17, 2003

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NATURALLY ARISING CD4+ REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES Shimon Sakaguchi Department of Experimental Pathology, Institute for Frontier Medical Sciences, Kyoto University, Kyoto 606–8507, Japan; Laboratory for Immunopathology, Research Center for Allergy and Immunology, The Institute for Physical and Chemical Research (RIKEN), Yokohama 230–0045, Japan; email: [email protected]

Key Words regulatory T cells, immunologic self-tolerance, imunoregulation, CD25, Foxp3 ■ Abstract Naturally occurring CD4+ regulatory T cells, the majority of which express CD25, are engaged in dominant control of self-reactive T cells, contributing to the maintenance of immunologic self-tolerance. Their depletion or functional alteration leads to the development of autoimmune disease in otherwise normal animals. The majority, if not all, of such CD25+CD4+ regulatory T cells are produced by the normal thymus as a functionally distinct and mature subpopulation of T cells. Their repertoire of antigen specificities is as broad as that of naive T cells, and they are capable of recognizing both self and nonself antigens, thus enabling them to control various immune responses. In addition to antigen recognition, signals through various accessory molecules and via cytokines control their activation, expansion, and survival, and tune their suppressive activity. Furthermore, the generation of CD25+CD4+ regulatory T cells in the immune system is at least in part developmentally and genetically controlled. Genetic defects that primarily affect their development or function can indeed be a primary cause of autoimmune and other inflammatory disorders in humans. Based on recent advances in our understanding of the cellular and molecular basis of this T cell–mediated immune regulation, this review discusses how naturally arising CD25+CD4+ regulatory T cells contribute to the maintenance of immunologic self-tolerance and negative control of various immune responses, and how they can be exploited to prevent and treat autoimmune disease, allergy, cancer, and chronic infection, or establish donor-specific transplantation tolerance.

INTRODUCTION The immune system discriminates between self and nonself, establishing and maintaining unresponsiveness to self (i.e., self-tolerance). How this is achieved has been a key issue in immunology for nearly 50 years since the proposition of the clonal 0732-0582/04/0423-0531$14.00

531

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

532

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

selection theory (1). There is clear evidence that clonal deletion of self-reactive T and B cells exposed to self-antigens at immature stages of their development is a major mechanism of self/nonself discrimination and self-tolerance (2–4). The deletion mechanism is not complete, however, and there is ample evidence that potentially hazardous self-reactive lymphocytes are present in the periphery of normal individuals. A typical illustration of this is that immunization of normal animals with a normal self-constituent along with potent adjuvant can elicit T cell–mediated autoimmune tissue damage. As a more recent illustration of this, tissue-specific transgenic expression of a costimulatory molecule and a cytokine can trigger self-reactive T cells to attack the cells expressing the transgenes (5, 6). Several plausible mechanisms have been proposed that deal with those pathogenic self-reactive lymphocytes that have escaped the deletional mechanism in the central generative organs (7–9). For example, they may be rendered anergic or further deleted upon encounter with self-antigens. Self-reactive T cells may fail to be activated (i.e., they ignore self-antigens) because of low avidities of their T cell antigen receptors (TCRs) for self-antigens, lack of costimulation from antigen-presenting cells (APCs), or their seclusion from the target self-antigens. In addition to these passive or recessive mechanisms of self-tolerance, there appears to be a dominant one: Certain T cells actively downregulate the activation and expansion of self-reactive lymphocytes (reviewed in References 10–12). Assuming that multiple mechanisms contribute to the maintenance of selftolerance to various degrees, at different levels, and in complementary manners, a crucial question would be which of these, when it goes awry, leads to the development of autoimmune disease, and, more practically, which of these can prevent autoimmune disease when strengthened. It is also of keen medical interest to know whether natural self-tolerance could hamper immune surveillance of autologous tumor cells—in other words, how one can break tolerance to tumor-associated quasi-self antigens and provoke effective tumor immunity by breaching any mechanism of self-tolerance. Furthermore, given that the ideal of organ transplantation is to establish graft tolerance as naturally and stably as self-tolerance, one should ask which mechanism of self-tolerance can be exploited to achieve this. Among the mechanisms of immunologic self-tolerance listed above, the contribution of regulatory (or suppressor) T cells (TR cells) or even their existence as a cellular entity has been controversial until recently, mainly because of lack of reliable markers to identify them and ambiguity of their functions at the molecular level (10, 11). In the past several years, however, we have witnessed explosive interest in T cell–mediated suppression as a key mechanism of self-tolerance and immune regulation. There are several reasons for this move. First, there has been a gradual reappraisal of the finding that autoimmune diseases immunopathologically similar to the human counterparts can be produced in normal rodents by simply depleting a TR-cell population, and reconstitution of the population inhibits the autoimmunity (13–18). Second, a cell surface marker was discovered that could differentiate operationally, if not specifically, such a TR-cell population from other T cells (13, 16–18). Third, the use of the marker enabled identification of human

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NATURAL CD25+CD4+ REGULATORY T CELLS

533

TR cells with a common phenotype and function to those found in rodents (19). There is also recent unequivocal evidence that genetic abnormality in the generation and function of naturally arising CD4+ TR cells and resulting immune dysregulation are indeed the primary cause and the mechanism of a human genetic disease accompanying autoimmune and other immunological disorders (20). Furthermore, in addition to self-tolerance and autoimmunity, evidence is now accumulating that natural CD4+ TR cells actively engage in negative control of a broad spectrum of immune responses to quasi-self or nonself antigens as in tumor immunity, organ transplantation, allergy, and microbial immunity (12, 21). TR cells are heterogeneous in phenotype, function, and the way of generation. Some are naturally occurring; others are induced by specific ways of antigenic stimulation (22). In this review, I focus on naturally arising CD25+CD4+ TR cells because there is now substantial evidence that they play indispensable roles in the maintenance of natural self-tolerance and negative control of pathological as well as physiological immune responses.

NATURAL TR CELLS FOR THE MAINTENANCE OF SELF-TOLERANCE Induction of Autoimmune Disease in Normal Rodents by Eliminating a Specific Subpopulation of CD4+ T Cells There are two findings, made nearly 30 years ago, that have since contributed to the formation of the current notion that T cell–mediated control of self-reactive T cells is a key mechanism of self-tolerance. Nishizuka & Sakakura showed in 1969 that neonatal thymectomy (NTx) of normal mice, especially between day 2 and 4 after birth, led to the destruction of ovaries, which later turned out to be of autoimmune nature (23). In 1973, Penhale et al. reported that thymectomy of adult rats (ATx) followed by several exposures to sublethal X-irradiation resulted in the development of autoimmune thyroiditis (24). Based on the fact that inoculation of normal T cells, especially CD4+ T cells, prevented the diseases, both groups suspected depletion of suppressor T cells as a plausible mechanism of the autoimmunities (25, 26). Other interpretations of these autoimmune phenomena have also been proposed, including the possible role of profound T-lymphocytopenia that might incur microbial infections triggering autoimmunity, inefficient thymic clonal deletion of self-reactive T cells in the neonatal period, and altered T cell homeostasis that might elicit homeostatic proliferation of certain self-reactive T cells (27). The existence of CD4+ T cells with autoimmune-inhibitory activity has also been shown since the 1980s with spontaneous models of autoimmune disease. For example, in NOD (non-obese diabetic) mice and BB (Bio-Breeding) rats, which spontaneously develop type 1 diabetes mellitus (T1D), inoculation of CD4+ T cells from histocompatible normal animals prevented T1D. Cotransfer of normal CD4+ T cells with diabetogenic T cells effectively prevented the disease in T cell– deficient recipients (28, 29).

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

534

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

Figure 1 (A) Transfer of T cell suspensions depleted of CD25+CD4+ TR cells induces autoimmune disease and IBD, and heightens immune responses to nonself-antigens in athymic nude or SCID mice. (B) Male children are afflicted with IPEX. Their mothers with hemizygous defects of the FOXP3 gene bear defective and normal TR cells as a mosaic because of random inactivation of the X chromosome in each TR cell. They are completely normal because normal TR cells dominantly control the activation and expansion of effector T (TE) cells that mediate autoimmune disease, IBD, and allergy. Open circles mean intact TR or TE cells; closed circles mean defective TR cells.

A direct experiment to assess the possible regulatory role of normal CD4+ T cells in self-tolerance would be to examine whether direct removal of putative CD4+ TR cells from normal animals can break self-tolerance, leading to the development of autoimmune disease, and whether reconstitution of the removed population can reestablish self-tolerance and prevent autoimmune disease. Attempts have been made to differentiate a TR-cell population from other T cells by the expression levels of a particular cell surface molecule (13–18) (Figure 1A). Sakaguchi et al. showed in 1985 that, when CD4+ splenic cell suspensions from normal BALB/c mice were depleted of CD5highCD4+ T cells ex vivo (by in vitro treatment of the cell suspensions with a mixture of anti-CD8 and anti-CD5 antibodies and complement) and the remaining CD5lowCD4+ T cells were transferred to congenitally T cell–deficient BALB/c athymic nude mice, the nude mice spontaneously developed autoimmune disease in multiple organs (stomach, thyroid, ovaries, or testes) in a few months after cell transfer (13). Cotransfer of normal untreated CD4+ T cells with CD5lowCD4+ T cells inhibited the autoimmunity. Once diseases developed, T cells were able to transfer the diseases adoptively to naive nude mice in a disease-specific manner, indicating that they were bona fide T cell–mediated autoimmune diseases (13). Likewise, transfer of CD5lowCD4+ T cells from normal C3H mice to T cell–depleted C3H mice produced autoimmune thyroiditis (14). Powrie & Mason subsequently reconstituted PVG athymic

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NATURAL CD25+CD4+ REGULATORY T CELLS

535

nude rats with splenic T cells depleted of CD45RClowCD4+ T cells, showing that the transferred CD45RChighCD4+ T cells elicited graft-versus-host disease-like systemic disease and autoimmune tissue damage in multiple organs including thyroid and Langerhans’ islets (16). McKeever et al. similarly showed that transfer of spleen cell suspensions depleted of RT6.1+ T cells was able to produce T1D and thyroiditis in histocompatible athymic nude rats (17). Powrie et al. and Morrisey et al. then independently showed that transfer of BALB/c CD45RBhigh CD4+ T cells to T/B cell–deficient BALB/c SCID mice induced inflammatory bowel disease (IBD) (30, 31). These findings prompted Sakaguchi et al. to search for a cell surface molecule that would be more specific than CD5 or CD45RB in defining such autoimmunity/inflammation-preventive CD4+ T cells (“More specific” in this setting means that removal of a smaller fraction of peripheral T cells can produce autoimmune disease in a more severe form and in a wider spectrum of organs.). They revealed the CD25 molecule (the IL-2 receptor α-chain) as a candidate because CD25+ T cells, which constitute 5%–10% of peripheral CD4+ T cells and less than 1% of peripheral CD8+ T cells in normal naive mice and humans, are contained in the CD5high and CD45RBlow fraction of CD4+ T cells (18) (see also Figure 2). Transfer of BALB/c splenic cell suspensions depleted of CD25+CD4+ T cells indeed produced in BALB/c athymic nude mice histologically and serologically evident autoimmune diseases at higher incidences and in a wider spectrum of organs (including stomach, thyroid, ovaries, adrenal glands, and Langerhans’ islets) than the transfer of CD5low or CD45RBhigh T cells prepared from the same number of splenic cell suspensions. Large doses of CD25− T cells even produced lethal systemic autoimmunity. Cotransfer of a small number of

Figure 2 Cell surface markers for naturally arising CD4+ TR cells. The most specific and reliable marker for naturally occurring CD4+ TR cells is the transcription factor Foxp3, which is expressed by the majority of CD25+CD4+ T cells and a fraction of CD25−CD4 T cells. CD45RBlow T cells, GITRhigh T cells, and CTLA-4+ T cells in the CD4+ T cell population include Foxp3-expressing T cells in normal naive mice.

19 Feb 2004

12:6

536

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD25+CD4+ T cells with CD25− T cells, on the other hand, completely prevented these autoimmunities. Furthermore, transfer of CD25−CD4+ T cells alone sufficed to mediate the autoimmune disease by giving rise to CD4+ helper T cells for humoral and cell-mediated autoimmunity, although the presence of CD25−CD8+ T cells enhanced the autoimmune induction presumably as a source of self-reactive cytotoxic T-lymphocytes. Importantly, the experiments also showed that removal of CD25+CD4+ T cells not only elicited autoimmune disease but also enhanced immune responses to nonself antigens including xenogeneic proteins and allografts (see also Figure 3B) (18).

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NATURAL CD25+CD4+ REGULATORY T CELLS

537

This spontaneous development of autoimmune disease following depletion of CD25+CD4+ T cells can hardly be attributed to T-lymphocytopenia and consequent vulnerability of the hosts to microbial infection because the depletion reduced the number of T cells only by 5%–10% (18, 27). It cannot be due to homeostatic proliferation of T cells (including self-reactive T cells), which might occur following transfer of a small number of T cells to a T cell–deficient environment, because the incidence, severity, and the range of autoimmune diseases in nude mice were proportional to the number of transferred CD25− cells. Taken together, these results indicate the following: First, despite thymic negative selection, the normal immune system still harbors self-reactive T cells (CD4+ T cells in particular) that are sufficiently pathogenic in TCR specificity and affinity to mediate various autoimmune diseases similar in immunopathology to their human counterparts, such as autoimmune gastritis/pernicious anemia, premature ovarian failure with autoimmune oophoritis, Hashimoto’s thyroiditis, adrenalitis/ Adison’s disease, and insulitis/T1D. Second, the activation/expansion of such self-reactive T cells is normally kept in check by regulatory CD4+ T cells, many if not all of which physiologically express CD25. Third, elimination of this CD25+CD4+ regulatory population suffices to break natural self-tolerance and incite chronic and destructive autoimmune diseases. In addition, the appearance of various disease-specific autoantibodies in the TR cell–depleted animals implies that the breakdown of this mode of T cell self-tolerance and the development of autoimmune CD4+ helper T cells result in breakdown of B cell self-tolerance as well. The activated CD4+ helper T cells presumably provide stimulatory signals to relevant self-reactive B cells, rescue them from apoptosis, and stimulate them to form autoantibodies (32). Thus, one aspect of natural self-tolerance in T cells, and ←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 3 Contribution of altered immunoregulation and host genetic factors to the development of autoimmune disease. (A) Genetic alterations or environmental insults that affect the balance between TR cells and self-reactive T cells toward the dominance of the latter can be a cause of autoimmune disease. (B) Immune responsiveness to self and nonself forms a continuum, and autoimmune disease may develop as a result of altered immunoregulation. Ordinate denotes immune responsiveness (or TCR avidity). Each peak denotes individual T cell clones. The horizontal line indicates the level of T cell–mediated immunoregulation. The peaks above the line represent overt immune responses to self or nonself antigens. When the level goes down (i.e., the immunoregulation becomes weaker, for example, by the reduction of CD25+ TR cells), immune responses to certain self-antigens become apparent, and immune responses to nonself-antigens are enhanced [as the development of autoimmune disease, IBD, and allergy in IPEX (Figure 1)]. The immune responsiveness to each self-antigen is mainly determined by the genetic makeup of the host. In this particular individual, the mild degree of reduction in TR cell–mediated immunoregulation, depicted as “a,” may elicit autoimmune gastritis (G) and thyroiditis (T), and with more severe reduction depicted as “b,” diabetes (D) as well. (C) See text.

19 Feb 2004

12:6

538

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

for that matter in B cells, is maintained by a regulatory subpopulation of CD4+ T cells. Furthermore, such naturally present TR cells also engage in the control of immune responses to nonself antigens as well.

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Immune Dysregulation, Polyendocrinopathy, Enteropathy, and X-Linked Syndrome (IPEX) as Evidence for the Key Role of TR Cells in Natural Self-Tolerance in Humans In 1982, IPEX was described as an X-linked immunodeficiency syndrome associated with autoimmune disease in multiple endocrine organs (such as T1D and thyroiditis), IBD, atopic dermatitis, and fatal infections (33). The Scurfy strain of mice is an X-linked recessive mutant with lethality in hemizygous males within a month after birth, exhibiting hyperactivation of CD4+ T cells and overproduction of proinflammatory cytokines (34). The gene defective in Scurfy mice was identified and designated Foxp3, which encodes Scurfin, a new member of the forkhead/winged-helix family of transcription factors (35). Subsequently, mutations of the human gene FOXP3, the ortholog of the murine Foxp3, were found to be the cause of IPEX (36–38). Recent studies have revealed the specific role of Foxp3 in the development and function of natural CD25+CD4+ TR cells (39–41). CD25+CD4+ peripheral T cells and CD25+CD4+CD8− thymocytes predominantly expressed Foxp3 mRNA, whereas other thymocytes/T cells and B cells did not. Importantly, activation of CD25−CD4+ T cells or differentiated Th1/Th2 cells failed to induce Foxp3 expression (39, 40). Furthermore, retroviral transduction of Foxp3 to naive CD25−CD4+ T cells converted them into CD25+CD4+ TR-like cells phenotypically and functionally (39, 40). For example, Foxp3 transduction induced expression of CD25, CTLA-4, CD103, and GITR (glucocorticoid-induced TNF receptor-related gene), which are closely associated with the functions of natural TR cells (see below) (39). Foxp3-transduced CD25−CD4+ T cells were able to suppress proliferation of other T cells in vitro and the development of autoimmune disease and IBD in vivo (39). Analyses of Foxp3-transgenic or -deficient mice also revealed an indispensable role of Foxp3 for the development of CD25+CD4+ TR cells (40, 41). For example, the number of CD25+CD4+ T cells increased in Foxp3-transgenic mice. In addition, CD25−CD4+ T cells and CD8+ T cells in these transgenic mice also expressed high levels of Foxp3 and exhibited in vitro suppressive activity (41). Foxp3-deficient mice, on the other hand, showed hyperactivation of T cells as observed in Scurfy mice. In bone marrow (BM) chimeras with a mixture of BM cells from wild-type and Foxp3-deficient mice, Foxp3-deficient BM cells failed to give rise to CD25+CD4+ T cells, whereas Foxp3-intact BM cells generated them (40). Scurfy mice, whose Scurfin protein lacks the fork-head domain, contained few CD25+CD4+ TR cells, although they developed many CD25+ chronically activated T cells (41). Furthermore, transduction of the Foxp3 gene with a similar mutation failed to confer a regulatory phenotype to normal naive T cells (39).

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NATURAL CD25+CD4+ REGULATORY T CELLS

539

Thus, Foxp3/FOXP3 appears to be a master control gene for the development and function of natural CD25+CD4+ TR cells. Given that humans bear natural CD25+CD4+ TR cells with a common phenotype and function to those found in rodents (19), it is most likely in IPEX patients that mutations of the FOXP3 gene abrogate the development of the TR cells or impair their suppressive function, leading to hyperactivation of T cells reactive with self-antigens, commensal bacteria in the intestine, or innocuous environmental substances, thus causing autoimmune polyendocrinopathy, IBD, and allergy, respectively (Figure 1B). IPEX has several implications for the mechanism of self-tolerance and autoimmune/inflammatory disease in humans. First, this is so far the most clear example that abnormality in naturally arising TR cells is a primary cause of human autoimmune disease and for that matter IBD and allergy. Second, the development of natural TR cells is, at least in part, genetically and developmentally programmed, indicating that autoimmunity is in part a primary T cell immunodeficiency. Third, females with hemizygous defects of the FOXP3 gene illustrate that the mechanism of dominant self-tolerance is physiologically operating in humans. Because of random inactivation of the X-chromosome (Lyonization) in individual TR cells or their precursors, the hemizygous females develop FOXP3-normal TR cells together with or without FOXP3-defective TR cells, and they are completely normal and do not show intermediate disease phenotypes (42) (Figure 1B). This means that the normal TR cells dominantly control self-reactive T cells in the hemizygous females. It also indicates that a partial reconstitution of IPEX patients with normal TR cells (for example by BM transplantation) or FOXP3-transduced autologous T cells may suffice to control the disease dominantly.

PHENOTYPIC DEFINITION OF NATURAL TR CELLS: HOW SPECIFIC IS CD25 FOR NATURALLY ARISING CD4+ TR CELLS? As depicted in the short historical note above, CD25 was found to be a useful marker for distinguishing the autoimmune-preventive CD4+ cells present in normal rodents from other CD4+ T cells that include dormant self-reactive CD4+ T cells (18, 27) (Figure 2). All T cells, however, express CD25 upon activation. Inevitable questions are then whether CD25 is merely a good marker for naturally arising CD4+ TR cells or a key molecule specific and essential for their generation and/or function; how specifically and stably they express it—in other words, whether CD25−CD4+ T cells with a similar regulatory function exist in the physiological state; and, if this is the case, how they can be defined by other cell surface markers. The following findings indicate that CD25 is an indispensable molecule for the generation and maintenance of natural CD4+ TR cells. IL-2-deficient mice bear few CD25+CD4+ T cells and spontaneously develop severe autoimmunity, although they develop a normal number of other T cells with normal composition of CD4/CD8 subsets (43, 44). CD25-deficient or CD122 (the IL-2Rβ-chain)-deficient

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

540

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

mice are also afflicted with similar autoimmunity, which can be prevented by the inoculation of CD25+CD4+ T cells from normal syngeneic mice (44–47). In addition, in vivo neutralization of IL-2 by the administration of anti-IL-2 monoclonal antibody (mAb) substantially reduced CD25+CD4+ T cells, but not other T cells, and consequently produced autoimmune disease in otherwise normal mice (48; R. Setoguchi and S. Sakaguchi, manuscript in preparation). Furthermore, Foxp3 controls the expression of CD25 in natural TR cells but not in activated T cells in general (39). Taken together, assuming that IL-2 is a key growth/survival factor for natural CD25+CD4+ TR cells, the CD25 molecule as a component of the high affinity IL-2R is a key functional molecule for them. Its expression, a high level expression in particular, is an excellent marker for natural CD4+ TR cells (49). Although attempts have been unsuccessful so far to specify the suppressive activity of the CD25+CD4+ T cell population to a smaller population included in it, individual CD25+CD4+ TR cells may be different in the degree of suppressive activity (49). Supporting this, CD103+ T cells were reportedly more suppressive in vitro than CD103− T cells among CD25+CD4+ T cells, and CD62highCD25+ CD4+ TR cells were more potent in preventing T1D in NOD mice partly owing to different tissue homing specificities (50, 51). Regarding the stability of CD25 expression on natural TR cells, it was noted that the majority of CD25+CD4+ T cells derived from normal naive mice lost the surface expression of CD25 when transferred to SCID mice (52, 53). These CD25+CD4+ cell-derived CD25−CD4+ T cells nevertheless exhibited an equivalent suppressive activity both in vivo and in vitro to that of CD25+CD4+ T cells in normal naive mice and stably expressed Foxp3. In addition, when CD25+CD4+ T cells were cotransferred with normal T cells to T cell–deficient mice, the CD25 expression level of the former was sustained as nearly normal (E. Nishimura, T. Sakihama, R. Setoguchi, K.J. Wood, K. Tanaka, and S. Sakaguchi, manuscript submitted). Thus, natural CD4+ TR cells may change their CD25 expression levels while retaining their suppressive activity, especially in a T cell–deficient condition. The above findings, however, do not necessarily mean that naturally arising CD4+ TR cells are confined to the CD25+CD4+ T cell population. There is indeed a substantial amount of data that T cells in the CD25−CD4+ T cell population in normal naive rodents bear similar suppressive activity in vivo and in vitro in various experimental systems. For example, in rat T1D induced by ATx and fractionated X-irradiations, the CD25−CD45RClow fraction of splenic CD4+ T cells from normal syngeneic rats showed diabetes-preventive activity, although the activity was much less potent than CD25+CD4+ T cells or CD25+CD4+CD8− thymocytes (54). CD25−CD45RBlowCD4+ T cells enriched from normal mice inhibited the development of IBD in SCID mice transferred with CD45RBhighCD4+ T cells (55). In addition, the CD25−CD45lowCD4+ T cell fraction of normal naive mice also expresses Foxp3 at a low level (39). Based on these findings, our recent efforts to further characterize the TR cells in the CD25−CD45RBlowCD4+ population in terms of Foxp3 expression and in vitro suppressive activity have revealed that they are CTLA-4+and GITRhigh, similar to natural CD25+CD4+ TR cells (see below).

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NATURAL CD25+CD4+ REGULATORY T CELLS

541

Indeed, depletion of CTLA-4+GITRhigh T cells from normal spleen cells and transfer of the remaining cells to nude mice produced more severe forms of autoimmune disease in a wider spectrum of organs than the transfer of the same number of CD25− splenic T cells (M. Ono and S. Sakaguchi, manuscript in preparation). Taken together, these findings indicate that, in normal naive rodents, the regulatory activity detectable in the CD25−CD4+ T cell population can be attributed to CTLA4+CD45RBlowGITRhigh cells, which constitute at most 5% of the CD25−CD4+ T cell population. It remains to be determined whether these TR cells in the CD25+ or CD25− fraction of CD4+ T cells belong to functionally and developmentally the same population of TR cells, which are only phenotypically convertible in CD25 expression. With these findings on the phenotype of natural CD4+ TR cells, it may still be hoped to identify a cell surface molecule, if any, that is more specific for natural CD4+ TR cells than CD25, CTLA-4, or GITR, and more closely associated with Foxp3 expression (Figure 2). Nevertheless, CD25 is a useful marker for obtaining a highly pure population of TR cells to analyze their function, as discussed below, because the CD25+CD4+ T cell population in normal naive animals is highly enriched for TR cells (21).

FUNCTIONS OF NATURAL TR CELLS Cellular Characteristics of CD25+CD4+ T Cell–Mediated Suppression To analyze the suppressive function of CD25+CD4+ TR cells, a simple in vitro assay was established in which CD25+CD4+ T cells were cocultured with CD25−CD4+ T cells (or CD8+ T cells), and the degree of suppression was assessed by measuring the proliferation of cocultured cells upon antigenic stimulation in the presence of APCs (56–58). The analysis of this in vitro suppression as well as in vivo suppression of autoimmune/inflammatory disease in rodents has revealed the following characteristics of CD25+CD4+ TR cell–mediated suppression. First, stimulation via TCR is required for CD25+CD4+ TR cells to exert suppression. Antigen-specific as well as polyclonal TCR stimulation can activate CD25+CD4+ T cells to exert in vitro suppression, whereas irrelevant antigens incapable of activating CD25+CD4+ T cells cannot (56, 57). This implies that natural CD25+CD4+ TR cells should recognize self-antigens and must be stimulated by them to control self-reactive T cells in the normal internal environment. Second, the CD25+CD4+ TR cell–mediated suppression is highly sensitive to antigenic stimulation: much lower concentration of antigen can stimulate CD25+ CD4+ TR cells to exert suppressive activity than the antigen concentration required for the activation/proliferation of naive CD25−CD4+ T cells with the same antigen specificity. For example, when CD25+ or CD25−CD4+ T cells are prepared from a TCR-transgenic mouse and stimulated with a specific peptide, the antigen concentration required for stimulating the former to exert suppression is 10- to

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

542

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

100-fold lower than that required for triggering the latter to proliferate (56). This high antigen sensitivity of CD25+CD4+ TR cells may be partly attributed to their high expression of various accessory molecules (including CTLA-4 and adhesion molecules) and possibly to their specific mode of signal transduction via TCR and/or accessory molecules such as CTLA-4 (see below). Third, once CD25+CD4+ TR cells are stimulated by a specific antigen, they can suppress both CD4+ and CD8+ T cells, and the suppression they mediate is antigennonspecific: They suppress the proliferation of not only T cells with the same antigen specificity but also those specific for other antigens (56, 59). Likewise, in the allogeneic mixed lymphocyte reaction (MLR), CD25+CD4+ T cells activated by allogeneic stimulator cells suppressed CD25−CD4+ T cells of the same strain as well as those from a third party strain reactive with the same stimulator cells (21). This may form the cellular basis of linked suppression in transplantation (60). It also implies that histocompatibility between CD25+CD4+ TR cells and T cells to be suppressed is not required for effective suppression. Finally, CD25+CD4+ TR cells are highly differentiated in their function and ready to conduct their specific function upon encountering stimulating antigens. This is illustrated by the finding that CD25+CD4+ T cells freshly prepared from normal naive mice or TCR-transgenic mice can mediate suppression from the first day of in vitro TCR stimulation (56).

CD25+CD4+ TR Cells are Hypoproliferative In Vitro, but Proliferative In Vivo, Upon Antigenic Stimulation Although CD25+CD4+ TR cells require antigenic stimulation for their functional activation, they themselves are anergic to in vitro antigenic stimulation, if one defines anergy as an antiproliferative state (56–58). This anergic state is closely linked with suppression: Abrogation of the anergic state by in vitro TCR stimulation in the presence of a high dose of IL-2 or CD28 ligation results in simultaneous loss of suppressive activity (56). Importantly, this anergic/suppressive state of CD25+CD4+ TR cells appears to be their basal and default condition: TCR-stimulated and IL-2-treated (or anti-CD28 antibody-treated), hence anergy/suppression-broken CD25+CD4+ TR cells spontaneously revert to the original anergic state and reacquire the suppressive activity once IL-2 (or anti-CD28 antibody) is removed (56, 59). The expanded TR cells showed more potent suppressive activity on a per cell basis than before (59). This unique property of CD25+CD4+ TR cells distinguishes them from other anergic or regulatory T cells because T cells may show a suppressive activity when rendered anergic in vitro by antigenic stimulation without costimulation, but they will never spontaneously revert to an anergic/suppressive state once anergy is abrogated (61, 62). This reversible anergy of CD25+CD4+ TR cells can be exploited to expand them by TCR stimulation and a high dose of IL-2. For example, in vitro stimulation of CD25+CD4+ T cells with allogeneic cells along with a high dose of IL-2 enables alloantigen-specific TR cells to expand (63). The CD25+CD4+ T cells thus

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NATURAL CD25+CD4+ REGULATORY T CELLS

543

expanded suppress MLR more potently than nonstimulated CD25+CD4+ T cells, suppressing even secondary MLR (E. Nishimura, T. Sakihama, R. Setoguchi, K.J. Wood, K. Tanaka, and S. Sakaguchi, manuscript submitted). Furthermore, intermittent stimulation of CD25+CD4+ T cells with a relevant antigen and a high dose of IL-2 enables the preparation of anergic and suppressive T cell clones, illustrating that CD25+CD4+ TR cells are anergic and suppressive at a single cell level (21; J. Shimizu, T. Takahashi, S. Hori, and S. Sakaguchi, manuscript submitted). In contrast to their in vitro resistance to proliferation upon TCR stimulation, CD25+CD4+ TR cells show active proliferation in vivo by antigenic stimulation (64–67; E. Nishimura, T. Sakihama, R. Setoguchi, K.J. Wood, K. Tanaka, and S. Sakaguchi, manuscript submitted). For example, when CD25+CD4+ T cells from naive mice were transferred to syngeneic nude mice with allogeneic skin grafts, alloantigen-specific CD25+CD4+ TR cells actively proliferated and as a consequence enhanced alloantigen-specific suppressive activity. Furthermore, a fraction of CD25+CD4+ T cells in normal naive mice are continuously proliferating without exogenous antigenic stimulation, presumably by recognizing self-antigens in the periphery (68). Thus, natural CD25+CD4+ TR cells can be adaptive in that they show antigen-specific expansion and consequently augment antigen-specific suppression with each successive exposure to a particular antigen.

The Molecular Basis of CD25+CD4+ TR Cell–Mediated Suppression The precise molecular mechanism by which CD25+CD4+ TR cells suppress the activation and proliferation of other T cells is currently controversial and under active investigation. Some findings support essential roles of cytokines in the regulation; others support the contribution of cell-to-cell cognate interactions on APCs; and still others support the modification of APCs by TR cells. CD25+CD4+ TR cells, which express various adhesion molecules at high levels, may also outcompete naive T cells with the same antigen specificity in adhesion to APCs and thereby contribute to the suppression. POSSIBLE ROLES OF CYTOKINES FOR SUPPRESSION A critical role of IL-10 in CD25+CD4+ T cell–mediated suppression has been shown mainly in murine IBD induced by the transfer of CD25−CD45RBhighCD4+ T cells to SCID mice and prevented by cotransfer of CD25+ or CD45RBlowCD4+ T cells (52). Administration of anti-IL-10 receptor-blocking mAb to the mice with the preventive cell transfer neutralized the suppression, resulting in the development of IBD (55). In addition, CD25+CD4+ or CD45RBlowCD4+ T cells from IL-10-deficient mice failed to prevent IBD in this model (52, 55). Contribution of IL-10 to CD25+CD4+ TR cell–mediated suppression was also shown in murine models of transplantation tolerance, graft-versus-host disease, chronic parasite infection, autoimmunity, and a rat model of T1D (69). In contrast to these results, IL-10-deficient T cells effectively prevented autoimmune disease produced by depletion of CD25+CD4+ TR cells (such as autoimmune gastritis in BALB/c mice) (70). IL-10-deficient BALB/c

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

544

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

mice spontaneously developed IBD, but not gastritis (55, 71). Furthermore, addition of neutralizing anti-IL-10 mAb to an in vitro proliferation assay had no effect on the ability of CD25+CD4+ TR cells to mediate suppression (56, 57). This discrepancy in the effect of IL-10 on the control of colitis versus gastritis and in vivo versus in vitro results remains to be resolved. The contribution of TGF-β1 to CD25+CD4+ T cell–mediated suppression is also controversial, although TGF-β mRNA can be predominantly detected in the CD25+CD4+ population in normal naive mice (27). For example, expression of a T cell–specific dominant-negative form of the TGF-β receptor II produced IBD in mice (72). Similar to the effect of IL-10 to murine IBD (55), anti-TGF-β antibody treatment abrogated the suppression, resulting in IBD (73). Administration of anti-TGF-β and anti-IL-4 antibody neutralized suppressive activity of CD45RClowCD4+ T cells in rat T1D and thyroiditis produced by ATx and fractionated X-irradiations (74). In addition, it was reported that activated CD25+CD4+ TR cells, but not CD25−CD4+ T cells, expressed TGF-β on the cell surface and anti-TGF-β antibody neutralized suppression in vitro (75), although these findings are somehow at variance with the results by others (76). REQUIREMENT OF DIRECT CELL-TO-CELL INTERACTIONS FOR SUPPRESSION The in vitro CD25+CD4+ TR cell–mediated suppression appears to depend on cell-tocell cognate interactions between the two T cell populations on APCs (56–58). This notion is supported by the following findings: Supernatants recovered from activated CD25+CD4+ T cells or the mixture of CD25+CD4+ T cells and CD25−CD4+ T cells failed to suppress the response of CD25−CD4+ T cells (56); when CD25+ CD4+ T cells and CD25−CD4+ T cells were physically separated by a semipermeable membrane, the former failed to suppress the response of the latter (56, 57). Furthermore, activated CD25+CD4+ TR cells suppressed responder T cells in the absence of APCs, at least in vitro (77, 78). This in vitro suppressive cell contact was not due to killing of the responder population via Fas/FasL- or TNF/TNF receptordependent pathway (56). Together with the failure of cytokine-neutralizing antibodies to abrogate in vitro suppression as discussed above, these findings indicate that the in vitro suppressive activity of CD25+CD4+ TR cells does not depend on paracrine or long-lasting soluble factors. MODULATION OF APCs TR cells may also down-modulate APC functions and thereby make APCs unable to activate effector T cells. For example, it was reported that CD25+CD4+ T cells downregulated the expression levels of CD80 and CD86 on dendritic cells (DCs) (79). MORE THAN ONE MECHANISM OF CD25+CD4+ TR CELL–MEDIATED SUPPRESSION?

These results, when taken together, indicate that more than one mechanism of CD25+CD4+ T cell–mediated suppression is operative in vivo, and that one TR cell may exert suppression by more than one mechanism depending on the situation— for example, the intensity of immune responses, especially that of TR cell stimulation, and the site of regulation, especially the cytokine milieu.

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

NATURAL CD25+CD4+ REGULATORY T CELLS

545

The Molecular Basis of the Control of TR Cell Activation and Survival

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

In addition to signaling through TCR, signals via accessory molecules, such as CTLA-4, CD28, and GITR, contribute to the activation and proliferation of TR cells, and hence the tuning of the intensity of suppression. CTLA-4 AND CD28 IN ACTIVATION OF TR CELLS CD28 provides naive T cells with a costimulatory signal for promoting IL-2 formation and cell expansion as well as preventing anergy induction and cell death, whereas CTLA-4, which is expressed upon T cell activation, transduces a negative signal to activated T cells, thereby attenuating T cell responses (80). These two molecules appear to play different roles in CD25+CD4+ TR cells. A strong signal through CD28 to CD25+CD4+ TR cells by strong ligation of the molecules abrogates their anergic and suppressive state (56, 57). In contrast, they apparently do not need CD28 for their activation because CD25+CD4+ T cells from CD28-deficient or -intact mice exhibited an equally potent in vitro suppressive activity (81). On the other hand, CD28 seems to play a key role in the generation of CD25+CD4+ T cells in the thymus and presumably in their survival in the periphery because CD28-deficient mice develop a substantially reduced number of CD25+CD4+ T cells in the thymus and periphery (82). In contrast to naive T cells, CD25+CD4+ T cells in normal naive mice constitutively express CTLA-4 (81–83). The following findings indicate that these CTLA4 molecules should play a critical role in TR cell–mediated suppression. First, inoculation of anti-CTLA-4 mAb to normal mice over a limited period elicited autoimmune diseases similar to those produced by depletion of CD25+CD4+ TR cells, without reducing the number of CD25+CD4+ T cells (81). Similarly, administration of anti-CTLA-4 mAb abolished the protective activity of CD25+CD4+ TR cells in the murine IBD model (83). Second, in vitro blockade of CTLA-4 by Fab fragments of anti-CTLA-4 mAb neutralized the CD25+CD4+ TR cellmediated suppression (81). CTLA-4-expressing CD25+CD4+ T cells from normal mice suppressed the in vitro proliferation of CD25−CD4+ T cells from CTLA4-deficient mice upon polyclonal TCR stimulation; anti-CTLA-4 Fab fragments neutralized the suppression (81). Third, a lethal lymphoproliferative and autoimmune syndrome that spontaneously develops in CTLA-4-deficient mice is not T cell autonomous but can be inhibited by wild-type T cells, indicating that CTLA4-deficiency may lead to impaired dominant regulation (84). In addition, Foxp3 appears to instruct CTLA-4 gene expression directly or indirectly (39). These results collectively indicate that signals through CTLA-4 may activate CD25+CD4+ TR cells to exert suppression and that blockade of the signal lead to the failure in their activation and thereby to attenuation of the TR cell–mediated suppression. The results, however, do not necessarily mean that CTLA-4 is the only accessory molecule that is required for the activation of TR cells because CD25+CD4+ T cells from CTLA-4-deficient mice also exhibited a significant suppressive activity in vitro (81).

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

546

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

Thus, signal through CD28 to naive T cells and TR cells synergistically enhances immune responses by activating naive T cells and attenuating TR cell–mediated suppression. Similarly, CTLA-4 plays two roles that are synergistic in attenuating T cell immune responses, i.e., transduction of a negative signal to activated T cells and an activating signal to CD25+CD4+ TR cells for suppression. The effects of in vivo CTLA-4 blockade to exacerbate autoimmunity, enhance rejection of transplanted organs, or provoke tumor immunity have been attributed to the former role of CTLA-4 (80). The effects could also be attributed, at least in part, to the blockade of CTLA-4 on CD25+CD4+ TR cells and consequent abrogation of TR cell–mediated suppression. It remains to be examined at molecular levels how the balance between signals through CTLA-4 and CD28, both of which interact with CD80 and CD86 on APCs, contribute to the tuning of the regulatory activity of CD25+CD4+ TR cells. CD25+CD4+ T cells in normal naive mice predominantly express GITR (also known as TNFRSF18), as revealed by raising a specific mAb or conducting DNA microarray analyses (53, 85–87). Other T cells (CD25−CD4+ and CD8+ T cells), B cells, DCs, and macrophages express GITR at low levels, increasing their expression levels upon activation (85, 86). Cross-linking of GITR by a specific mAb, together with TCR stimulation, abrogated in vitro CD25+CD4+ TR cell–mediated suppression without breaking their anergic state (85). Unlike CTLA-4, mere blockade of GITR by Fab fragments of anti-GITR mAb failed to neutralize suppression (85). Furthermore, administration of anti-GITR mAb elicited autoimmune disease in normal mice (85). Taken together, these results indicate that a signal through GITR attenuates the ability of CD25+CD4+ TR cells to exert suppression. As discussed above, not only CD25+CD4+ TR cells but also TR cells in the CD25−CD4+ T cell population are GITRhigh and express CTLA-4. It remains to be determined how the signals through GITR and CTLA-4, which are apparently in opposition, are integrated in these CD25+ or CD25−CD4+ TR cells. GITR

TLRs are germline-encoded receptors that recognize pathogen-associated molecular patterns shared by large groups of microbes or certain endogenous molecules released during inflammation (88, 89). A recent report showed that natural CD25+CD4+ TR cells selectively expressed several members of the TLR family, such as TLR4 (90). In vitro stimulation of CD25+CD4+ T cells with a high concentration of lipopolysaccharide (LPS) through TLR4 elicited their proliferation, prolonged their survival, and augmented their in vitro suppressive activity even in the absence of APCs, indicating that LPS directly acts on TLR4 molecules expressed by TR cells (90). It is likely that, if a large amount of LPS (several orders of magnitude higher than the concentration required for in vitro activation of DCs) is produced upon Gram-negative bacterial infection, TR cells activated by LPS through TLRs may augment their suppressive activity and thereby prevent local or systemic immunopathology (such as septic shock) owing to production of large amounts of pro-inflammatory cytokines (such as TNFα and IL-1) by macrophages TLRs

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

NATURAL CD25+CD4+ REGULATORY T CELLS

547

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

(91). Thus, expression of the two types of receptors, TCRs specific for MHCbound self or nonself peptide antigens and TLRs for conserved molecular patterns shared by microbial components or certain endogenous molecules (such as heat shock proteins), may make CD25+CD4+ TR cells unique in controlling immune responses against various self and nonself antigens, especially against microbes. ADHESION MOLECULES, CHEMOKINE RECEPTORS, AND HOMING RECEPTORS The expression pattern of other accessory molecules on CD25+CD4+ TR cells [e.g., CD45RBlow, CD44high, CD5high, CD54 (ICAM-1)high, CD11a/CD18 (LFA-1)high, and partly CD62Llow] is in part similar to that of primed, activated effector or memory T cells (18, 49, 57, 92). The profile suggests that the TR cells may be continuously stimulated by self-antigens in the normal internal environment. Furthermore, expression of particular chemokine or homing receptors may enable CD25+CD4+ TR cells to be preferentially guided to the sites of antigen presentation in the secondary lymphoid tissues and also recruited to the site of inflammation and tissue damage to control physiological and pathological immune responses (69, 92a, 92b).

IL-2 is an essential growth/survival factor for CD25+CD4+ TR cells, as discussed above, and may also be required for their function (93). TGF-β can stimulate and expand them in human peripheral blood (94). A recent report showed that IL-6 secreted by activated DCs acts on CD25−CD4+ T cells and makes them resistant to the suppression by TR cells (95). It remains to be determined whether other cytokines are physiologically involved in the activation and survival of TR cells. CYTOKINES

THE ORIGIN OF NATURAL CD25+CD4+ TR CELLS AND THEIR ANTIGEN SPECIFICITIES Thymic Production of CD25+CD4+ TR Cells as a Functionally Distinct and Mature T Cell Subpopulation: Another Role of the Thymus in Immunologic Self-Tolerance The following findings provide evidence that the normal thymus produces the majority, if not all, of CD25+CD4+ TR cells as a functionally mature and distinct T cell subpopulation, which appears to constitute a distinct cellular lineage to the periphery. First, transfer of CD4+CD8+ mature thymocyte suspensions depleted of CD25+ thymocytes produced various autoimmune diseases in syngeneic nude mice, as shown with the transfer of CD25−CD4+ spleen cells (92). This indicates that the normal thymus is continuously producing pathogenic selfreactive CD4+ T cells as well as functionally mature CD25+CD4+ TR cells. Second, both CD25+CD4+CD8− thymocytes and CD25+CD4+ T cells express Foxp3, and Foxp3 deficiency abrogates both populations (39, 40). Third, both populations are

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

548

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

functionally and phenotypically similar. For example, both are naturally anergic to in vitro TCR stimulation, exhibiting an equivalent in vitro suppressive activity, constitutively expressing CTLA-4 and being GITRhigh, and being resistant to a superantigen-induced clonal deletion (81, 85, 96). Thus, the thymus contributes to the maintenance of self-tolerance not only by deleting or inactivating self-reactive T cells but also by producing TR cells. Accumulated findings indicate that the thymic development of CD25+CD4+ TR cells requires unique interactions of their TCRs with self-peptide/MHC complexes expressed on the thymic stromal cells. For example, in TCR-transgenic mice, a large number of CD25+CD4+ T cells express endogenous TCR α-chains paired with transgenic β-chains, whereas the majority of CD25−CD4+ T cells express TCRs composed of transgenic α- and β-chains. Furthermore, RAG-2 deficiency, which blocks the gene rearrangement of the endogenous TCR α-chain locus, abrogates the development of CD25+CD4+ TR cells in TCR transgenic mice (92). This indicates that developing T cells that happen to acquire a high affinity for self-peptide/MHC ligands as the result of receptor editing may be positively selected to become TR cells. Supporting this, in a double-transgenic strain that expressed a transgene-encoded specific peptide in the thymic stromal cells at a high level, the majority of T cells expressing transgenic TCR α- and β-chains specific for the peptide differentiated to CD25+CD4+ TR cells (97, 98). The TR cells failed to develop, however, when double-transgenic mice expressed either low-affinity transgenic TCR or high concentrations of the peptide, presumably because of insufficient positive selection or strong negative selection, respectively (97, 98). Furthermore, H2-DMα-deficient mice, in which class II MHC molecules display a limited array of self-peptides, developed CD25+CD4+ TR cells, whereas MHC class II–deficient mice, which generate a small number of CD4+ T cells restricted to classical or nonclassical MHC class I antigens, did not (99). These findings altogether indicate that, compared with thymic selection of other T cells, the development of CD25+CD4+ TR cells requires higher avidity agonistic interactions of their TCRs with self-peptide/MHC or class II MHC itself expressed on the thymic stromal cells [especially cortical epithelial cells (97, 99)], and that the required avidity must not be so high as to lead to their deletion. This possible role of thymic epithelial cells to generate CD25+CD4+ TR cells may form a common basis to the observation that grafting of allogeneic thymic epithelium to MHCdisparate, T cell–deficient hosts generated TR cells that specifically suppressed immune responses to the alloantigens (100). Antigen-independent interactions via accessory molecules expressed on developing thymocytes and the thymic stromal cells also contribute to the thymic generation of CD25+CD4+ TR cells, presumably by increasing the avidity of the interaction between thymocytes and the stromal cells. For example, the number of CD25+CD4+CD8− thymocytes and T cells substantially reduced in CD28-, B7-, or CD40-deficient mice, or mice treated with CTLA-4-Ig, which blocks interaction between B7 and CD28/CTLA-4 (82, 101). The deficiencies and blockade enhanced the development of T1D in NOD mice (82). Mice deficient of CD40L, CD11a/CD18 (LFA-1,) and CD54 (ICAM-1) also developed significantly smaller

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

NATURAL CD25+CD4+ REGULATORY T CELLS

549

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

numbers of CD25+CD4+ TR cells in the thymus and periphery (T. Takahashi, unpublished data). Transfer of spleen cells from CD40-deficient mice to syngeneic athymic nude mice indeed produced various autoimmune diseases similar to those produced by the removal of CD25+CD4+ TR cells; reconstitution with normal CD25+CD4+ T cells inhibited the autoimmune development (101). Further study is required to determine how the antigen-dependent and antigenindependent interaction between developing thymocytes and thymic stromal cells leads to the development of CD25+CD4+ TR cells and what the relationship between this unique selection event and Foxp3 expression in developing TR cells is.

Antigen Specificity of CD25+CD4+ TR Cells CD25+CD4+ T cells in normal naive mice are as diverse as CD25−CD4+ T cells in the usage of TCR α- and β-chain gene subfamilies and equally susceptible to negative selection by endogenous superantigens (56, 102, 103). They are alloreactive: Their stimulation with MHC or Mls-disparate stimulator cells along with IL-2 can specifically expand alloreactive TR cells (21, 63). Use of transgenic mouse strains in which every class II MHC molecule binds the same single peptide revealed that the single peptide/MHC can positively select functional CD25+CD4+ TR cells that have a TCR repertoire as broad as that of CD25−CD4+ T cells (68, 103). Interestingly, these TR cells are more reactive with the selecting single peptide/MHC complexes than CD25−CD4+ T cells, indicating that they have higher avidity for the ligand compared with other T cells that are also selected by the same ligand (68; T. Takahashi, S. Hori, Y. Fukui, T. Sasazuki & S. Sakaguchi, unpublished data). In normal animals, summation of each broad repertoire selected by each selfpeptide/MHC ligand may well form in total a broad repertoire of CD25+CD4+ TR cells, which is almost duplicated in the CD25+ and CD25−CD4+ population, but with higher reactivity of the former to the thymic self-peptide/MHC ligands. This implies that, if T cell clones reactive to a particular self-antigen are detected in the CD25−CD4+ T cell population, for example, by the transfer of CD25−CD4+ T cells to athymic nude mice (18), regulatory T cell clones capable of recognizing the same self antigen are also present in the CD25+CD4+ T cell population. The high self-reactivity and broad repertoire of CD25+CD4+ TR cells would ensure their roles to dominantly control various immune responses against self and nonself antigens.

Developmentally Determined Generation of CD25+CD4+ TR Cells CD25+CD4+ T cells become detectable in the periphery of normal mice from around day 3 after birth, rapidly increasing to the adult level (i.e., 5%–10% of CD4+ T cells) in 3 weeks (27). As discussed above, NTx on day 3 produces autoimmune diseases similar to those induced by depletion of CD25+CD4+ TR cells. NTx on day 3 substantially reduces peripheral CD25+CD4+ T cells, and the inoculation of CD25+CD4+ T cells from normal mice prevents the autoimmune

19 Feb 2004

12:6

550

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

development in NTx mice (27, 104). These results collectively indicate that the ontogenic time course of the thymic production and the peripheral migration of CD25+CD4+ TR cells is developmentally predetermined (i.e., around day 3 after birth in mice), and the abrogation of the thymic production from the very beginning of their ontogeny results in their selective paucity in the periphery, leading to the activation of self-reactive T cells that have migrated to the periphery before NTx, producing autoimmune disease.

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Peripheral Generation of TR Cells It is known that TR cells can be produced from peripheral T cells by ex vivo stimulation with antigens in the presence of particular cytokines (such as IL-10) or by in vitro propagation after in vivo induction of oral tolerance (105–107). These TR cells, called Tr1 cells or Th3 cells, respectively, are apparently different from natural CD25+CD4+ TR cells in their dependency on cytokines for the maintenance and exertion of suppression (108). On the other hand, TR cells phenotypically and functionally similar to natural CD25+CD4+ TR cells can be induced in vitro from CD25−CD4+ T cells by particular ways of antigen stimulation, for example, on immature or cytokine (IL-10/TGF-β)-treated DCs or those expressing a Notchligand (109–111). It remains to be determined whether these TR cells are de novo induced from naive T cells or derived from naturally present Foxp3-expressing CTLA-4+GITRhigh TR cells in the CD25−CD4+ T cell population (see above). CD25+CD4+ T cells could also be induced from CD25−CD4+ T cells in RAGdeficient TCR transgenic mice by intravenous administration of low-dose peptide antigen, oral administration of the antigen, or adoptive transfer of naive transgenic T cells to antigen-expressing transgenic mice (112, 113). Further study is required to determine how such apparently de novo induced TR cells are phenotypically and functionally similar or disimilar to natural CD25+CD4+ TR cells, for example, in Foxp3 expression.

THE ROLES OF NATURAL TR CELLS IN IMMUNOLOGICAL DISEASES, TUMOR IMMUNITY, AND TRANSPLANTATION TOLERANCE Imbalance Between Natural TR Cells and Self-Reactive T Cells as a Cause of Autoimmune Disease Any genetic abnormality or environmental insult can be a causal or predisposing factor to autoimmune disease if it would tip the balance between natural TR cells and self-reactive T cells toward the dominance of the latter (Figure 3A). It has been shown with rodents that physical, chemical, and biological agents or genetic alteration can indeed cause autoimmune disease by reducing natural TR cells in the periphery or affecting their thymic production. For example, administration of cyclosporin A (CsA), infection with mouse T-lymphotropic virus (MTLV), low-dose

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NATURAL CD25+CD4+ REGULATORY T CELLS

551

fractionated X-irradiations, or alteration in TCR gene expression elicited autoimmune diseases in mice similar to those produced by depletion of CD25+ CD4+ TR cells (114–117). These autoimmune inductions can be attributed to specific immunological properties of natural CD25+CD4+ TR cells. For example, natural CD25+CD4+ TR cells recognizing self-antigens are continuously proliferating in the periphery and hence are more radiosensitive than other T cells (68) (see above). Inhibition of IL-2 production, for example by CsA, may reduce natural TR cells in the periphery by affecting their survival (R. Setoguchi and S. Sakaguchi, unpublished data). In addition, an early period in life seems to be more susceptible to such environmental insults because it is relatively easy to deplete a small number of natural TR cells in the periphery, or self-reactive T cells produced by the thymus can more easily expand owing to the paucity of TR cells in the periphery, as illustrated by neonatal CsA treatment or MTLV infection in mice. In humans, an autoimmune disease with an epidemiologically evident environmental cause is T1D and other autoimmune endocrinopathies in congenital rubella syndrome (118–120). Infection with rubella virus at a particular stage of intrauterine life might affect developing natural TR cells and thereby trigger autoimmunity as rubella virus inhibits proliferation of various tissue cells including T cells. As autoimmune-causing genetic alterations, deficiency of the Foxp3, CTLA-4, CD40, IL-2, CD25, and CD122 genes elicits severe autoimmune diseases that are not T cell autonomous and can be controlled by normal T cells, as discussed above. The polymorphisms of these genes may also contribute to determining the genetic susceptibility to autoimmune disease as recently shown with the CTLA-4 and IL-2 genes, which are the main susceptibility genes for T1D and other organ-specific autoimmune diseases in humans (121, 122). Of note, however, is that deficiency or dysfunction of natural TR cells per se cannot determine which organs/tissues are targeted by the triggered autoimmune responses (18). For example, NTx of mice predominantly produces autoimmune oophoritis in the A strain, autoimmune gastritis in the BALB/c but not DBA/2 strain (which shares H-2d MHC haplotype), while ATx and fractionated X-irradiations predominantly induces thyroiditis in PVG/c rats with the RT1c MHC haplotype, and T1D in MHC congenic PVG-RT1u rats (119, 120) (Figure 3B). These findings, when taken together, lead to the hypothesis that the development of autoimmune diseases, especially organ-specific ones, may be in part determined by two elements: One is the degree of deficiency or dysfunction of natural TR cells (or the balance between natural TR cells and self-reactive T cells) and the other the host genes, including MHC and non-MHC genes, which determine the phenotype of the autoimmune disease (i.e., the specificity and intensity of the autoimmune responses) (118–120) (Figure 3C). Possible evolutionary conservation of the genetic defects or polymorphisms [including transspecies polymorphism of MHC genes (123)] that contribute to determining autoimmune phenotypes may account for the development of a similar spectrum of autoimmune diseases in TR cell–depleted mice and in other species including humans. This possible scheme of autoimmune pathogenesis connotes the following: First, unless the abnormality of TR cells is present, host genes per se are unable

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

552

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

to induce autoimmune disease, e.g., mothers of IPEX patients are healthy even if they share phenotype-determining genes with their affected children (42). Second, the degree of the abnormality of TR cells is able to influence the manifestation of the genetically predetermined phenotype. In general, the longer and/or the more severe is the reduction or dysfunction of CD25+CD4+ TR cells, the higher is the incidence of the genetically predisposed particular autoimmune diseases, and the wider is the spectrum of autoimmune diseases in a genetically determined hierarchical pattern (Figure 3B). In the BALB/c strain, for example, the order of incidence in the hierarchy is autoimmune gastritis, oophoritis, thyroiditis, adrenalitis, insulitis/T1D, and others (18). In IPEX, a high incidence of T1D [more than 80% of patients (20, 33)] could be attributed to so severe a deficiency or dysfunction of natural TR cells. Third, this theory implies that a single causative agent or a single genetic abnormality affecting the TR cell–mediated control may lead to the occurrence of different autoimmune diseases, frequently more than one, in a single individual, as in IPEX. On the other hand, different causative agents affecting the control may produce the same autoimmune disease in genetically susceptible individuals through a common mechanism (e.g., BALB/c mice predominantly develop autoimmune gastritis following CsA treatment, MTLV infection, fractionated Xirradiation, or genetical alteration of TR cell ontogeny). It is thus likely that many, if not all, autoimmune diseases have a common mechanism, and not necessarily a specific etiology for each autoimmune disease, especially in organ-specific ones. Furthermore, assuming that the goal of the treatment of autoimmune disease is to reestablish natural tolerance to the self-antigens targeted in autoimmune disease, one way of achieving this is to reestablish or newly establish dominant tolerance by helping naturally present TR cells to expand or strengthen their suppressive activity to the degree capable of stably controlling autoimmune T cells.

Tumor Immunity Given that many tumor-associated antigens recognized by autologous T cells are antigenically normal self-constituents, natural CD25+CD4+ TR cells engaged in the maintenance of self-tolerance may impede the generation and activation of tumor-effector T cells recognizing autologous tumor cells (10, 124). This is indeed the case because reduction of CD25+CD4+ T cells by anti-CD25 mAb treatment provoked effective immune responses to syngeneic tumors in otherwise nonresponding mice (125–127). Tumor effector cells can also be generated in vitro by simply eliminating CD25+CD4+ T cells from splenic cell suspensions prepared from tumor-nonsensitized mice: In the absence of CD25+CD4+ TR cells, selfreactive CD25−CD4+ T cells responded to self-peptides/class II MHC molecules on autologous APCs, spontaneously proliferated, and secreted a large amount of IL-2, which generated NK-like tumor effector cells as lymphokine-activated killer cells capable of promiscuously killing various tumor cells (125). Thus, removal of CD25+CD4+ TR cells can breach immunological unresponsiveness to syngeneic tumors in vivo and in vitro, leading to spontaneous development of tumor-specific effector cells as well as tumor-nonspecific ones. This novel

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

NATURAL CD25+CD4+ REGULATORY T CELLS

553

way of provoking tumor immunity would help to devise effective immunotherapy for cancer in humans. A caveat is possible development of autoimmunity depending on the degree and the period of in vivo TR cell depletion, and the genetic makeup of the host (128).

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Transplantation Tolerance The Holy Grail of organ transplantation is to establish graft tolerance as stably and naturally as in natural tolerance to self-constituents, without continuous general immunosuppression. One way of achieving this is to induce graft-specific TR cells sufficiently potent to maintain long-term graft acceptance (60, 129, 130). The finding that removal of CD25+CD4+ T cells from normal mice enhanced graft rejection (i.e., reduced the survival of the grafts) suggests that increase in the number or the suppressive activity of natural CD25+CD4+ TR cells may be able to induce and maintain dominant transplantation tolerance (18). Indeed, when CD25+CD4+ T cells enriched from normal syngeneic mice are inoculated, together with naive T cells, to syngeneic T cell–deficient mice with allografts, the graft survival was significantly prolonged, and even permanent graft tolerance can be established at large doses of natural CD25+CD4+ TR cells (21; E. Nishimmua, T. Sakihama, R. Setoguchi, K.J. Wood, K. Tanaka, and S. Sakaguchi, manuscript submitted). It has been shown that administration of mAbs specific for various cell surface molecules (including accessory and costimulatory molecules) involved in T cell activation can induce long-term allograft acceptance (60, 129, 130). Such molecules include CD4, CD8, CD28, and CD40L on T cells; CD40, CD80, and CD86 on APCs; and CD11a and CD54 on both. TR cells appear to play a critical role in transplantation tolerance induced by these treatments. It remains to be determined how these treatments induce TR cells and whether these TR cells are derived from na¨ıve alloreactive T cells or naturally arising CD25+CD4+ TR cells.

Immune Responses to Microbes or Allergens Naturally occurring CD25+CD4+ TR cells are engaged in suppressing excessive immune responses to microbes invading or cohabiting with the host (Figure 1). For example, SCID mice transferred with CD25−CD45RBhighCD4+ T cells spontaneously develop IBD as TR cell–defective IPEX accompanies IBD, whereas germ-free SCID mice similarly treated fail to develop the disease (131). Likewise, the depletion precipitates severe pneumonitis in mice that have opportunistic infection with Pneumocystis carinii (132). Depletion of natural CD25+CD4+ TR cells can also enhance protective immune responses against invading microbes including bacteria, viruses, fungi, and intracellular parasites, leading to their eradication from the host (91). The depletion of TR cells may thus lead to complete eradication of the microbes but may advertently prevent induction of long-term immunity because of insufficient maintenance of memory T cells owing to lack of

19 Feb 2004

12:6

554

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

microbe persistence (133). Natural TR cells may be required for the maintenance of balanced synbiosis between the host and microbes. Deficiency or functional defects of TR cells may also make the host prone to develop allergic reactions to innocuous environmental substances, as observed in IPEX (Figure 1) (21, 134). Thus, naturally arising CD25+CD4+ TR cells can be exploited to tune the intensity of antimicrobial immune responses in acute and chronic infections, to develop effective ways of vaccination against microbes, and to treat or prevent allergic diseases.

CONCLUSION It is now firmly established that T cell–mediated dominant immunoregulation is essential for maintaining immunologic self-tolerance and control of immune responses to nonself antigens. Among various kinds of TR cells, CD25+CD4+ TR cells play a key role in natural self-tolerance. They are distinct from other TR cells in that the majority of them are produced by the normal thymus as a functionally mature population, their ontogenic generation is developmentally programmed, and their deficiency or dysfunction may directly lead to the development of autoimmune and other inflammatory diseases in animals and humans. Further analysis of the function and the generation of this natural TR cell population at cellular and molecular levels will facilitate our ability to develop better ways for controlling physiological as well as pathological immune responses.

ACKNOWLEDGMENTS The author apologizes to authors whose citations were omitted because of space constraints. He thanks Drs. Zoltan Fehervari and Takeshi Takahashi for critical reading of the manuscript and colleagues in his laboratory for allowing him to mention their unpublished results. This work was supported by grants-in-aid from the Ministry of Education, Science, Sports, and Culture and the Ministry of Human Welfare of Japan. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Burnet FM. 1959. The Clonal Selection Theory of Acquired Immunity. London: Cambridge Univ. Press 2. Kappler JW, Roehm N, Marrack P. 1987. T cell tolerance by clonal elimination in the thymus. Cell 49:273–80

3. Kisielow P, Bluthmann H, Staerz UD, Steinmetz M, von Boehmer H. 1988. Tolerance in T-cell-receptor transgenic mice involves deletion of nonmature CD4+8+ thymocytes. Nature 333:742–46 4. Goodnow CC, Cyster JG, Hartley SB, Bell

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

NATURAL CD25+CD4+ REGULATORY T CELLS

5.

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

6.

7.

8.

9.

10.

11.

12.

13.

14.

SE, Cooke MP, et al. 1995. Self-tolerance checkpoints in B lymphocyte development. Adv. Immunol. 59:279–368 Weigle WO. 1980. Analysis of autoimmunity through experimental models of thyroiditis and allergic encephalomyelitis. Adv. Immunol. 30:159–273 Guerder S, Picarella DE, Linsley PS, Flavell RA. 1994. Costimulator B7-1 confers antigen-presenting-cell function to parenchymal tissue and in conjunction with tumor necrosis factor alpha leads to autoimmunity in transgenic mice. Proc. Natl. Acad. Sci. USA 91:5138–42 Arnold B, Schoenrich G, Haemmerling GJ. 1993. Multiple levels of peripheral tolerance. Immunol. Today 14:12–14 Van Parijs L, Abbas AK. 1998. Homeostasis and self-tolerance in the immune system: turning lymphocytes off. Science 280:243–48 Miller JF, Heath WR. 1993. Selfignorance in the peripheral T-cell pool. Immunol. Rev. 133:131–50 Sakaguchi S. 2000. Regulatory T cells: key controllers of immunologic selftolerance. Cell 101:455–58 Shevach EM. 2000. Regulatory T cells in autoimmunity. Annu. Rev. Immunol. 18:423–49 Maloy KJ, Powrie F. 2001. Regulatory T cells in the control of immune pathology. Nat. Immunol. 2:816–22 Sakaguchi S, Fukuma K, Kuribayashi K, Masuda T. 1985. Organ-specific autoimmune diseases induced in mice by elimination of T-cell subset. I. Evidence for the active participation of T cells in natural self-tolerance: deficit of a T-cell subset as a possible cause of autoimmune disease. J. Exp. Med. 161:72–87 Sugihara S, Izumi Y, Yoshioka T, Yagi H, Tsujimura T, et al. 1988. Autoimmune thyroiditis induced in mice depleted of particular T-cell subset. I. Requirement of Lyt-1dull L3T4bright normal T cells for the induction of thyroiditis. J. Immunol. 141:105–13

555

15. Smith H, Lou YH, Lacy P, Tung KSK. 1992. Tolerance mechanism in experimental ovarian and gastric autoimmune disease. J. Immunol. 149:2212–18 16. Powrie F, Mason D. 1990. OX-22high CD4+ T cells induce wasting disease with multiple organ pathology: prevention by OX-22low subset. J. Exp. Med. 172:1701– 8 17. McKeever U, Mordes JP, Greiner DL, Appel MC, Rozing J, et al. 1990. Adoptive transfer of autoimmune diabetes and thyroiditis to athymic rats. Proc. Natl. Acad. Sci. USA 87:7618–22 18. Sakaguchi S, Sakaguchi N, Asano M, Itoh M, Toda M. 1995. Immunologic tolerance maintained by activated T cells expressing IL-2 receptor α-chains (CD25): breakdown of a single mechanism of selftolerance causes various autoimmune diseases. J. Immunol. 155:1151–64 19. Shevach EM. 2001. Certified professionals: CD4+CD25+ suppressor T cells. J. Exp. Med. 193:F41–46 20. Gambineri E, Torgerson TR, Ochs HD. 2003. Immune dysregulation, polyendocrinopathy, enteropathy, and X-linked inheritance (IPEX), a syndrome of systemic autoimmunity caused by mutations of FOXP3, a critical regulator of Tcell homeostasis. Curr. Opin. Rheumatol. 15:430–35 21. Sakaguchi S, Sakaguchi N, Shimizu J, Yamazaki S, Sakihama T, et al. 2001. Immunologic tolerance maintained by CD25+CD4+ regulatory T cells: their common role in controlling autoimmunity, tumor immunity, and transplantation tolerance. Immunol. Rev. 182:18–32 22. Bluestone JA, Abbas AK. 2003. Natural versus adaptive regulatory T cells. Nat. Rev. Immunol. 3:253–57 23. Nishizuka Y, Sakakura T. 1969. Thymus and reproduction: sex-linked dysgenesia of the gonad after neonatal thymectomy in mice. Science 166:753–55 24. Penhale WJ, Farmer A, McKenna RP, Irvine WJ. 1973. Spontaneous thyroiditis

19 Feb 2004

12:6

556

25.

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

26.

27.

28.

29.

30.

31.

32.

33.

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI in thymectomized and irradiated Wistar rats. Clin. Exp. Immunol. 15:225–36 Sakaguchi S, Takahashi T, Nishizuka Y. 1982. Study on cellular events in postthymectomy autoimmune oophoritis in mice. II. Requirement of Lyt-1 cells in normal female mice for the prevention of oophoritis. J. Exp. Med. 156:1577–86 Penhale WJ, Irvine WJ, Inglis JR, Farmer A. 1976. Thyroiditis in T cell-depleted rats: suppression of the autoallergic response by reconstitution with normal lymphoid cells. Clin. Exp. Immunol. 25:6–16 Asano M, Toda M, Sakaguchi N, Sakaguchi S. 1996. Autoimmune disease as a consequence of developmental abnormality of a T cell subpopulation. J. Exp. Med. 184:387–96 Greiner DL, Mordes JP, Handler ES, Angelillo M, Nakamura N, Rossini AA. 1987. Depletion of RT6.1+ T lymphocytes induces diabetes in resistant biobreeding/Worcester (BB/W) rats. J. Exp. Med. 166:461–75 Boitard C, Yasunami R, Dardenne M, Bach JF. 1989. T cell-mediated inhibition of the transfer of autoimmune diabetes in NOD mice. J. Exp. Med. 169:1669–80 Powrie F, Leach MW, Mauze S, Caddle LB, Coffman RL. 1993. Phenotypically distinct subsets of CD4+ T cells induce or protect from chronic intestinal inflammation in C. B-17 scid mice. Int. Immunol. 5:1461–71 Morrissey PJ, Charrier K, Braddy S, Liggitt D, Watson JD. 1993. CD4+ T cells that express high levels of CD45RB induce wasting disease when transferred into congenic severe combined immunodeficient mice. Disease development is prevented by cotransfer of purified CD4+ T cells. J. Exp. Med. 178:237–44 Tsubata T, Wu J, Honjo T. 1993. Bcell apoptosis induced by antigen receptor crosslinking is blocked by a T-cell signal through CD40. Nature 364:645–48 Powell BR, Buist NR, Stenzel P. 1982. An X-linked syndrome of diarrhea, polyen-

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

docrinopathy, and fatal infection in infancy. J. Pediatr. 100:731–37 Godfrey VL, Wilkinson JE, Russell LB. 1991. X-linked lymphoreticular disease in the scurfy (sf) mutant mouse. Am. J. Pathol. 138:1379–87 Brunkow ME, Jeffery EW, Hjerrild KA, Paeper B, Clark LB, et al. 2001. Disruption of a new forkhead/winged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfy mouse. Nat. Genet. 27:68–73 Chatila TA, Blaeser F, Ho N, Lederman HM, Voulgaropoulos C, et al. 2000. JM2, encoding a fork head-related protein, is mutated in X-linked autoimmunityallergic disregulation syndrome. J. Clin. Invest. 106:R75–81 Wildin RS, Ramsdell F, Peake J, Faravelli F, Casanova JL, et al. 2001. X-linked neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome is the human equivalent of mouse scurfy. Nat. Genet. 27:18–20 Bennett CL, Christie J, Ramsdell F, Brunkow ME, Ferguson PJ, et al. 2001. The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nat. Genet. 7:20–21 Hori S, Nomura T, Sakaguchi S. 2003. Control of regulatory T cell development by the transcription factor Foxp3. Science 299:1057–61 Fontenot JD, Gavin MA, Rudensky AY. 2003. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat. Immunol. 4:330–36 Khattri R, Cox T, Yasayko SA, Ramsdell F. 2003. An essential role for Scurfin in CD4+CD25+ T regulatory cells. Nat. Immunol. 4:337–42 Tommasini A, Ferrari S, Moratto D, Badolato R, Boniotto M, et al. 2002. Xchromosome inactivation analysis in a female carrier of FOXP3 mutation. Clin. Exp. Immunol. 130:127–30 Schorle H, Holtschke T, Hunig T, Schimpl

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

NATURAL CD25+CD4+ REGULATORY T CELLS

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

44.

45.

46.

47.

48.

49.

50.

51.

A, Horak I. 1991. Development and function of T cells in mice rendered interleukin-2 deficient by gene targeting. Nature 352:621–24 Almeida AR, Legrand N, Papiernik M, Freitas AA. 2002. Homeostasis of peripheral CD4+ T cells: IL-2R alpha and IL2 shape a population of regulatory cells that controls CD4+ T cell numbers. J. Immunol. 169:4850–60 Willerford DM, Chen J, Ferry JA, Davidson L, Ma A, Alt FW. 1995. Interleukin-2 receptor alpha chain regulates the size and content of the peripheral lymphoid compartment. Immunity 3:521–30 Suzuki H, Zhou YW, Kato M, Mak TW, Nakashima I. 1999. Normal regulatory alpha/beta T cells effectively eliminate abnormally activated T cells lacking the interleukin 2 receptor beta in vivo. J. Exp. Med. 190:1561–72 Malek TR, Yu A, Vincek V, Scibelli P, Kong L. 2002. CD4 regulatory T cells prevent lethal autoimmunity in IL2Rbeta-deficient mice. Implications for the nonredundant function of IL-2. Immunity 17:167–78 Murakami M, Sakamoto A, Bender J, Kappler J, Marrack P. 2002. CD25+CD4+ T cells contribute to the control of memory CD8+ T cells. Proc. Natl. Acad. Sci. USA 99:8832–37 Kuniyasu Y, Takahashi T, Itoh M, Shimizu J, Toda G, Sakaguchi S. 2000. Naturally anergic and suppressive CD25+ CD4+ T cells as a functionally and phenotypically distinct immunoregulatory T-cell subpopulation. Int. Immunol. 12:1145–55 Lehmann J, Huehn J, de la Rosa M, Maszyna F, Kretschmer U, et al. 2002. Expression of the integrin alpha Ebeta 7 identifies unique subsets of CD25+ as well as CD25− regulatory T cells. Proc. Natl. Acad. Sci. USA 99:13031–36 Szanya V, Ermann J, Taylor C, Holness C, Fathman CG. 2002. The subpopulation of CD4+CD25+ splenocytes that delays adoptive transfer of diabetes expresses L-

52.

53.

54.

55.

56.

57.

58.

59.

60.

557

selectin and high levels of CCR7. J. Immunol. 169:2461–65 Annacker O, Burlen-Defranoux O, Pimenta-Araujo R, Cumano A, Bandeira A. 2000. Regulatory CD4 T cells control the size of the peripheral activated/memory CD4 T cell compartment. J. Immunol. 164:3573–80 Gavin MA, Clarke SR, Negrou E, Gallegos A, Rudensky A. 2002. Homeostasis and anergy of CD4+CD25+ suppressor T cells in vivo. Nat. Immunol. 3:33–41 Stephens LA, Mason D. 2000. CD25 is a marker for CD4+ thymocytes that prevent autoimmune diabetes in rats, but peripheral T cells with this function are found in both CD25+ and CD25− subpopulations. J. Immunol. 165:3105–10 Asseman C, Mauze S, Leach MW, Coffman RL, Powrie F. 1999. An essential role for interleukin 10 in the function of regulatory T cells that inhibit intestinal inflammation. J. Exp. Med. 190:995–1004 Takahashi T, Kuniyasu Y, Toda M, Sakaguchi N, Itoh M, et al. 1998. Immunologic self-tolerance maintained by CD25+ CD4+ naturally anergic and suppressive T cells: induction of autoimmune disease by breaking their anergic/suppressive state. Int. Immunol. 10:1969–80 Thornton AM, Shevach EM. 1998. CD4+ CD25+ immunoregulatory T cells suppress polyclonal T cell activation in vitro by inhibiting interleukin 2 production. J. Exp. Med. 188:287–96 Read S, Mauze S, Asseman C, Bean A, Coffman R, Powrie F. 1998. CD38+ CD45RBlow CD4+ T cells: a population of T cells with immune regulatory activities in vitro. Eur. J. Immunol. 28:3435– 47 Thornton AM, Shevach EM. 2000. Suppressor effector function of CD4+ CD25+ immunoregulatory T cells is antigen nonspecific. J. Immunol. 164:183–90 Wood KJ, Sakaguchi S. 2003. Regulatory T cells in transplantation tolerance. Nat. Rev. Immunol. 3:199–210

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

558

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

61. Lombardi G, Sidhu S, Batchelor R, Lechler R. 1994. Anergic T cells as suppressor cells in vitro. Science 264:1587–89 62. Taams LS, van Rensen AJ, Poelen MC, van Els CA, Besseling AC, et al. 1998. Anergic T cells actively suppress T cell responses via the antigen-presenting cell. Eur. J. Immunol. 28:2902–12 63. Cohen JL, Trenado A, Vasey D, Klatzmann D, Salomon BL. 2002. CD4+CD25+ immunoregulatory T cells: new therapeutics for graft-versus-host disease. J. Exp. Med. 196:401–6 64. Yamazaki S, Iyoda T, Tarbell K, Olson K, Velinzon K, et al. 2003. Direct expansion of functional CD25+ CD4+ regulatory T cells by antigen-processing dendritic cells. J. Exp. Med. 198:235–47 65. Oldenhove G, de Heusch M, UrbainVansanten G, Urbain J, Maliszewski C, et al. 2003. CD4+ CD25+ regulatory T cells control T helper cell type 1 responses to foreign antigens induced by mature dendritic cells in vivo. J. Exp. Med. 198:259–66 66. Walker LS, Chodos A, Eggena M, Dooms H, Abbas AK. 2003. Antigen-dependent proliferation of CD4+ CD25+ regulatory T cells in vivo. J. Exp. Med. 198:249–58 67. Klein L, Khazaie K, von Boehmer H. 2003. In vivo dynamics of antigenspecific regulatory T cells not predicted from behavior in vitro. Proc. Natl. Acad. Sci. USA 100:8886–91 68. Sakaguchi S, Hori S, Fukui Y, Sasazuki T, Sakaguchi N, Takahashi T. 2003. Thymic generation and selection of CD25+CD4+ regulatory T cells: implications of their broad repertoire and high self-reactivity for the maintenance of immunologic self-tolerance. In Generation and Effector Functions of Regulatory Lymphocytes. Novartis Found. Symp. 252:6– 23 69. Hori S, Takahashi T, Sakaguchi S. 2003. Control of autoimunity by natural regulatory T cells. Adv. Immunol. 81:329–69 70. Suri-Payer E, Cantor H. 2001. Differ-

71.

72.

73.

74.

75.

76.

77.

78.

ential cytokine requirements for regulation of autoimmune gastritis and colitis by CD4+CD25+ T cells. J. Autoimmun. 16:115–23 Kuhn R, Lohler J, Rennick D, Rajewsky K, Muller W. 1993. Interleukin-10deficient mice develop chronic enterocolitis. Cell 75:263–74 Gorelik L, Flavell RA. 2000. Abrogation of TGFβ signaling in T cells leads to spontaneous T cell differentiation and autoimmune disease. Immunity 12:171–81 Powrie F, Carlino J, Leach MW, Mauze S, Coffman RL. 1996. A critical role for transforming growth factor-beta but not interleukin 4 in the suppression of T helper type 1-mediated colitis by CD45RBlow CD4+ T cells. J. Exp. Med. 183:2669–74 Seddon B, Mason D. 1999. Regulatory T cells in the control of autoimmunity: the essential role of transforming growth factor beta and interleukin 4 in the prevention of autoimmune thyroiditis in rats by peripheral CD4+CD45RC− cells and CD4+CD8− thymocytes. J. Exp. Med. 189:279–88 Nakamura K, Kitani A, Strober W. 2001. Cell contact-dependent immunosuppression by CD4+CD25+ regulatory T cells is mediated by cell surface-bound transforming growth factor beta. J. Exp. Med. 194:629–44 Piccirillo CA, Letterio JJ, Thornton AM, McHugh RS, Mamura M, et al. 2002. CD4+CD25+ regulatory T cells can mediate suppressor function in the absence of transforming growth factor beta1 production and responsiveness. J. Exp. Med. 196:237–46 Piccirillo CA, Shevach EM. 2001. Cutting edge: control of CD8+ T cell activation by CD4+CD25+ immunoregulatory cells. J. Immunol. 167:1137–40 Ermann J, Szanya V, Ford GS, Paragas V, Fathman CG, Lejon K. 2001. CD4+CD25+ T cells facilitate the induction of T cell anergy. J. Immunol. 167:4271–75

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NATURAL CD25+CD4+ REGULATORY T CELLS 79. Cederbom L, Hall H, Ivars F. 2000. CD4+CD25+ regulatory T cells downregulate co-stimulatory molecules on antigen-presenting cells. Eur. J. Immunol. 30:1538–43 80. Salomon B, Bluestone JA. 2001. Complexities of CD28/B7: CTLA-4 costimulatory pathways in autoimmunity and transplantation. Annu. Rev. Immunol. 19:225–52 81. Takahashi T, Tagami T, Yamazaki S, Uede T, Shimizu J, et al. 2000. Immunologic self-tolerance maintained by CD25+ CD4+ regulatory T cells constitutively expressing cytotoxic lymphocyte-associated antigen 4. J. Exp. Med. 192:303–10 82. Salomon B, Lenschow DJ, Rhee L, Ashourian N, Singh B, et al. 2000. B7/CD28 costimulation is essential for the homeostasis of the CD4+ CD25+ immunoregulatory T cells that control autoimmune diabetes. Immunity 12:431–40 83. Read S, Malmstroem V, Powrie F. 2000. CTLA-4 plays an essential role in the function of CD25+ CD4+ regulatory cells which control intestinal inflammation. J. Exp. Med. 192:295–302 84. Bachmann MF, Kohler G, Ecabert B, Mak TW, Kopf M. 1999. Cutting edge: lymphoproliferative disease in the absence of CTLA-4 is not T cell autonomous. J. Immunol. 163:1128–31 85. Shimizu J, Yamazaki S, Takahashi T, Ishida Y, Sakaguchi S. 2002. Stimulation of CD25+CD4+ regulatory T cells through GITR breaks immunological selftolerance. Nat. Immunol. 3:135–42 86. McHugh RS, Whitters MJ, Piccirillo CA, Young DA, Shevach EM, et al. 2002. CD4+CD25+ immunoregulatory T cells: gene expression analysis reveals a functional role for the glucocorticoid-induced TNF receptor. Immunity 16:311–23 87. Zelenika D, Adams E, Humm S, Graca L, Thompson S, et al. 2002. Regulatory T cells overexpress a subset of Th2 gene transcripts. J. Immunol. 168:1069–79 88. Takeda K, Kaisho T, Akira S. 2003.

89.

90.

91.

92.

92a.

92b.

93.

94.

95.

559

Toll-like receptors. Annu. Rev. Immunol. 21:335–76 Janeway CA Jr, Medzhitov R. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197–216 Caramalho I, Lopes-Carvalho T, Ostler D, Zelenay S, Haury M, Demengeot J. 2003. Regulatory T cells selectively express toll-like receptors and are activated by lipopolysaccharide. J. Exp. Med. 197:403–11 Sakaguchi S. 2003. Control of immune responses by naturally arising CD4+ regulatory T cells that express tolllike receptors. J. Exp. Med. 197:397– 401 Itoh M, Takahashi T, Sakaguchi N, Kuniyasu Y, Shimizu J, et al. 1999. Thymus and autoimmunity: production of CD25+CD4+ naturally anergic and suppressive T cells as a key function of the thymus in maintaining immunologic selftolerance. J. Immunol. 162:5317–26 Bystry RS, Aluvihare V, Welch KA, Kallikourdis M, Betz AG. 2001. B cells and professional APCs recruit regulatory T cells via CCL4. Nat. Immunol. 2:1126– 32 Iellem A, Mariani M, Lang R, Recalde H, Panina-Bordignon P, et al. 2001. Unique chemotactic response profile and specific expression of chemokine receptors CCR4 and CCR8 by CD4+CD25+ regulatory T cells. J. Exp. Med. 194:847–53 Furtado GC, Curotto de Lafaille MA, Kutchukhidze N, Lafaille JJ. 2002. Interleukin 2 signaling is required for CD4+ regulatory T cell function. J. Exp. Med. 196:851–57 Yamagiwa S, Gray JD, Hashimoto S, Horwitz DA. 2001. A role for TGF-beta in the generation and expansion of CD4+CD25+ regulatory T cells from human peripheral blood. J. Immunol. 166:7282–89 Pasare C, Medzhitov R. 2003. Toll pathway-dependent blockade of CD4+ CD25+ T cell-mediated suppression by dendritic cells. Science 299:1033–36

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

560

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

96. Papiernik M. 2001. Natural CD4+CD25+ regulatory T cells. Their role in the control of superantigen responses. Immunol. Rev. 182:180–89 97. Jordan MS, Boesteanu A, Reed AJ, Petrone AL, Holenbeck AE, et al. 2001. Thymic selection of CD4+CD25+ regulatory T cells induced by an agonist selfpeptide. Nat. Immunol. 2:301–6 98. Kawahata K, Misaki Y, Yamauchi M, Tsunekawa S, Setoguchi K, et al. 2002. Generation of CD4+CD25+ regulatory T cells from autoreactive T cells simultaneously with their negative selection in the thymus and from nonautoreactive T cells by endogenous TCR expression. J. Immunol. 168:4399–405 99. Bensinger SJ, Bandeira A, Jordan MS, Caton AJ, Laufer TM. 2001. Major histocompatibility complex class II-positive cortical epithelium mediates the selection of CD4+25+ immunoregulatory T cells. J. Exp. Med. 194:427–38 100. Modigliani Y, Bandeira A, Coutinho A. 1996. A model of developmentally acquired thymus-dependent tolerance to central and peripheral antigens. Immunol. Rev. 149:155–75 101. Kumanogoh A, Wang X, Lee I, Watanabe C, Kamanaka M, et al. 2001. Increased T cell autoreactivity in the absence of CD40CD40 ligand interactions: a role of CD40 in regulatory T cell development. J. Immunol. 166:353–60 102. Romagnoli P, Hudrisier D, van Meerwijk JP. 2002. Preferential recognition of self antigens despite normal thymic deletion of CD4+CD25+ regulatory T cells. J. Immunol. 168:1644–48 103. Pacholczyk R, Kraj P, Ignatowicz L. 2002. Peptide specificity of thymic selection of CD4+CD25+ T cells. J. Immunol. 168:613–20 104. Suri-Payer E, Amar AZ, Thornton AM, Shevach EM. 1998. CD4+CD25+ T cells inhibit both the induction and effector function of autoreactive T cells and represent a unique lineage of immunoreg-

105.

106.

107.

108.

109.

110.

111.

112.

ulatory cells. J. Immunol. 160:1212– 18 Groux H, O’Garra A, Bigler M, Rouleau M, Antonenko S, et al. 1997. A CD4+ T-cell subset inhibits antigen-specific Tcell responses and prevents colitis. Nature 389:737–42 Barrat FJ, Cua DJ, Boonstra A, Richards DF, Crain C, et al. 2002. In vitro generation of interleukin 10-producing regulatory CD4+ T cells is induced by immunosuppressive drugs and inhibited by T helper type 1 (Th1)- and Th2-inducing cytokines. J. Exp. Med. 195:603–16 Chen Y, Kuchroo VK, Inobe J, Hafler DA, Weiner HL. 1994. Regulatory T cell clones induced by oral tolerance: suppression of autoimmune encephalitis. Science 265:1237–40 Levings MK, Sangregorio R, Sartirana C, Moschin AL, Battaglia M, et al. 2002. Human CD25+CD4+ T suppressor cell clones produce transforming growth factor beta, but not interleukin 10, and are distinct from type 1 T regulatory cells. J. Exp. Med. 196:1335–46 Jonuleit H, Schmitt E, Schuler G, Knop J, Enk AH. 2000. Induction of interleukin 10-producing, nonproliferating CD4+ T cells with regulatory properties by repetitive stimulation with allogeneic immature human dendritic cells. J. Exp. Med. 192:1213–22 Sato K, Yamashita N, Baba M, Matsuyama T. 2003. Modified myeloid dendritic cells act as regulatory dendritic cells to induce anergic and regulatory T cells. Blood 101:3581–89 Hoyne GF, Dallman MJ, Champion BR, Lamb JR. 2001. Notch signalling in the regulation of peripheral immunity. Immunol. Rev. 182:215–27 Thorstenson KM, Khoruts A. 2001. Generation of anergic and potentially immunoregulatory CD25+ CD4 T cells in vivo after induction of peripheral tolerance with intravenous or oral antigen. J. Immunol. 167:188–95

19 Feb 2004

12:6

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

NATURAL CD25+CD4+ REGULATORY T CELLS 113. Apostolou I, Sarukhan A, Klein L, von Boehmer H. 2002. Origin of regulatory T cells with known specificity for antigen. Nat. Immunol. 3:756–63 114. Sakaguchi S, Sakaguchi N. 1989. Organspecific autoimmune disease induced in mice by elimination of T-cell subsets. V. Neonatal administration of cyclosporin A causes autoimmune disease. J. Immunol. 142:471–80 115. Morse SS, Sakaguchi N, Sakaguchi S. 1999. Virus and autoimmunity: induction of autoimmune disease in mice by mouse T lymphotropic virus (MTLV) destroying CD4+ T cells. J. Immunol. 162:5309– 16 116. Sakaguchi N, Miyai K, Sakaguchi S. 1994. Ionizing radiation and autoimmunity. Induction of autoimmune disease in mice by high dose fractionated total lymphoid irradiation and its prevention by inoculating normal T cells. J. Immunol. 152:2586–95 117. Sakaguchi S, Ermak TH, Toda M, Berg LJ, Ho W, et al. 1994. Induction of autoimmune disease in mice by germline alteration of the T cell receptor gene expression. J. Immunol. 152:1471–84 118. Sakaguchi S, Sakaguchi N. 1994. Thymus, T cells and autoimmunity: various causes but a common mechanism of autoimmune disease. In Autoimmunity: Physiology and Disease, ed. A Coutinho, M Kazatchkine, pp. 203–27. New York: Wiley-Liss 119. Sakaguchi S. 2000. Animal models of autoimmunity and their relevance to human diseases. Curr. Opin. Immunol. 12:684– 90 120. Sakaguchi S, Sakaguchi N. 2000. Role of genetic factors in organ-specific autoimmune diseases induced by manipulating the thymus or T cells, and not self-antigens. Rev. Immunogenetics 2: 147–53 121. Ueda H, Howson JM, Esposito L, Heward J, Snook H, et al. 2003 Association of the T-cell regulatory gene CTLA4 with sus-

122.

123.

124.

125.

126.

127.

128.

129.

130.

561

ceptibility to autoimmune disease. Nature 423(6939):506–11 Podolin PL, Wilusz MB, Cubbon RM, Pajvani U, Lord CJ, et al. 2000. Differential glycosylation of interleukin 2, the molecular basis for the NOD Idd3 type 1 diabetes gene? Cytokine 12:477–82 Klein J. 1987. Origin of major histocompatibility complex polymorphism: the trans-species hypothesis. Hum. Immunol. 19:155–62 Boon T, Cerottini J-C, Van den Eynde B, van der Bruggen P, Van Pel A. 1994. Tumor antigens recognized by T lymphocytes. Annu. Rev. Immunol. 12:337– 65 Shimizu J, Yamazaki S, Sakaguchi S. 1999. Induction of tumor immunity by removing CD25+CD4+ T cells: a common basis between tumor immunity and autoimmunity. J. Immunol. 163:5211–18 Onizuka S, Tawara I, Shimizu J, Sakaguchi S, Fujita T, Nakayama E. 1999. Tumor rejection by in vitro administration of anti-CD25 (Interleukin-2 receptor α) monoclonal antibody. Cancer Res. 59:3128–33 Sutmuller RP, van Duivenvoorde LM, van Elsas A, Schumacher TN, Wildenberg ME, et al. 2001. Synergism of cytotoxic T lymphocyte-associated antigen 4 blockade and depletion of CD25+ regulatory T cells in antitumor therapy reveals alternative pathways for suppression of autoreactive cytotoxic T lymphocyte responses. J. Exp. Med. 194:823–32 Phan GQ, Yang JC, Sherry RM, Hwu P, Topalian SL, et al. 2003. Cancer regression and autoimmunity induced by cytotoxic T lymphocyte-associated antigen 4 blockade in patients with metastatic melanoma. Proc. Natl. Acad. Sci. USA 100:8372–77 Zhai Y, Kupiec-Weglinski JW. 1999. What is the role of regulatory T cells in transplantation tolerance? Curr. Opin. Immunol. 11:497–503 Waldmann H, Cobbold S. 2001.

19 Feb 2004

12:6

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

562

AR

AR210-IY22-18.tex

AR210-IY22-18.sgm

LaTeX2e(2002/01/18)

P1: IKH

SAKAGUCHI

Regulating the immune response to transplants: a role for CD4+ regulatory cells? Immunity 14:399–406 131. Singh B, Read S, Asseman C, Malmstrom V, Mottet C, et al. 2001. Control of intestinal inflammation by regulatory T cells. Immunol. Rev. 182:190–200 132. Hori S, Carvalho TL, Demengeot J. 2002. CD25+CD4+ regulatory T cells suppress CD4+ T cell-mediated pulmonary hyperinflammation driven by Pneumocystis

carinii in immunodeficient mice. Eur. J. Immunol. 32:1282–91 133. Belkaid Y, Piccirillo CA, Mendez S, Shevach EM, Sacks DL. 2002. CD4+CD25+ regulatory T cells control Leishmania major persistence and immunity. Nature 420:502–7 134. Curotto de Lafaille MA, Lafaille JJ. 2002. CD4+ regulatory T cells in autoimmunity and allergy. Curr. Opin. Immunol. 14:771–78

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:531-562. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104721

Annu. Rev. Immunol. 2004. 22:563–98 doi: 10.1146/annurev.immunol.22.012703.104721 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on November 17, 2003

PHOSPHOINOSITIDE 3-KINASE: Diverse Roles in Immune Cell Activation

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Jonathan A. Deane and David A. Fruman Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California 92697; email: [email protected], [email protected]

Key Words 1-phosphatidylinositol 3-kinase, PI3K, second messengers, signal transduction, lymphocyte activation ■ Abstract Cells of the immune system carry out diverse functions that are controlled by surface receptors for antigen, costimulatory molecules, cytokines, chemokines, and other ligands. A shared feature of signal transduction downstream of most receptors on immune cells, as in nonhematopoietic cell types, is the activation of phosphoinositide 3-kinase (PI3K). The mechanism by which this common signaling event is elicited by distinct receptors and contributes to unique functional outcomes is an intriguing puzzle. Understanding how specificity is achieved in PI3K signaling is of particular significance because altered regulation of this pathway is observed in many disease states, including leukemia and lymphoma. Here we review recent advances in the understanding of PI3K signaling mechanisms in different immune cells and receptor systems. We emphasize the concept that PI3K and its products are components of complex networks of interacting proteins and second messengers, rather than simple links in linear signaling cascades.

INTRODUCTION The term phosphoinositide 3-kinase (PI3K; several other names appear in the literature, including PI 3-kinase and phosphatidylinositol 3-kinase) refers to a family of enzymes that phosphorylate D-myo-phosphatidylinositol (PtdIns) or its derivatives on the 3-hydroxyl of the inositol head group (Figure 1). The primary function of 3-phosphorylated inositol lipids (3-phosphoinositides) is to serve as membrane targeting signals to mediate membrane recruitment of selected proteins. PI3Ks and their function in regulated protein translocation are conserved through eukaryotes. In nearly all cases studied, genetic ablation or enhancement of PI3K function has a profound effect on cellular and organismal function (1–4). Most early studies of PI3K essentially catalogued the receptors that activate the enzyme and the cellular responses that ensue. Experimental approaches employed in this work included the use of pharmacological inhibitors wortmannin and LY294002, expression of mutated forms of PI3K genes and/or upstream receptors, 0732-0582/04/0423-0563$14.00

563

11 Feb 2004

22:12

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

564

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

Figure 1 Structure and synthesis of 3-phosphoinositides. (A) Chemical structure of D-myo-phosphatidylinositol (PtdIns). With permission, from the Annual Review c 2001 by Annual Reviews, www.AnnualReviews.org. of Biochemistry, Volume 70 ° (B) Diagram of pathways for synthesis and degradation of D-3-phosphoinositides. Enzymes involved in this pathway that are not discussed in the text are not included.

and measurement of cellular 3-phosphoinositide levels by chromatographic methods. Together these approaches made clear that elevation in 3-phosphoinositides is characteristic of a host of responses to a great variety of stimuli, including those important for immune responses. A subsequent phase of PI3K research involved the identification of cellular proteins that bind selectively to 3-phosphoinositides. The list of modular lipid binding domains and the proteins in which they are found, herein termed PI3K effectors, is ever growing.

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

565

Although the cataloguing of PI3K signaling responses and effectors continues, progress in recent years has been driven by methodological innovations that have allowed a greater mechanistic understanding of this important signaling pathway in different cell types. Knockout and knock-in mouse models have helped define the roles of individual PI3K genes in normal physiology, overcoming the limitations of global inhibitors and dominant negatives. The development of fluorescent probes to localize and quantitate 3-phosphoinositides has enabled the comparison of lipid distribution within and between cells that could not be achieved by earlier biochemical methods. Expanded studies of PI3K signaling in primary cells have helped to resolve controversies arising from the study of cultured cell lines. This review presents a brief overview of PI3K structure, activation mechanism, and function (for more extensive reviews, see 1, 5). We then describe notable recent advances in the understanding of PI3K signaling in cells of the immune system, with emphasis on results gained from primary cells studied with the novel approaches outlined above. Headings are organized according to types of receptor signaling systems rather than by cell types. We focus on physiological responses of mature cells; the reader is referred to detailed reviews on the role of PI3K in lymphocyte development, leukemogenesis, and autoimmunity (4, 6, 7). Space constraints limit discussion of PI3K function downstream of receptors for mitogenic cytokines and inflammatory mediators (for recent reviews, see 2, 8).

PI3K BACKGROUND Phosphoinositides and PI3K Enzyme Families Phosphoinositides are found in the cytoplasmic leaflet of cellular membranes where they regulate activities that include vesicle trafficking, cytoskeletal reorganization, and signal transduction (1, 9). Among the 3-phosphoinositides, PtdIns(3)P is the most abundant and its levels are relatively constant; it is primarily found in endosomes where its major function appears to be in protein sorting, though acute increases in PtdIns(3)P synthesis play a role in specific processes such as phagocytosis (10). Little is known about the regulation or function of PtdIns(3,5)P2 in mammalian cells. PtdIns(3,4)P2 and PtdIns(3,4,5)P3 (also known as PIP3) are essentially absent in quiescent cells but are transiently increased at the plasma membrane in response to a broad array of extracellular cues. Although these lipids contribute to a host of cellular responses, their production is strongly correlated with proliferation and survival in many cell types (1, 2, 7). PI3Ks are categorized as class I, II, or III, depending on their subunit structure, regulation, and substrate selectivity (Figure 2) (1, 5). Class I PI3Ks are the only enzymes capable of converting PtdIns(4,5)P2 to the critical second messenger PtdIns(3,4,5)P3. Class I PI3Ks are heterodimers composed of a catalytic subunit of approximately 110 kDa, and a tightly associated regulatory subunit that modulates its activity and cellular location. The class IA subgroup exists as multiple isoforms, with three catalytic subunits (p110α, p110β, and p110δ) encoded by three distinct

11 Feb 2004

22:12

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

566

AR

Figure 2 Schematic diagram of the PI3K catalytic and regulatory subunits. Class IA regulatory subunits have several modular domains that can regulate function of the heterodimer. Each isoform has two SH2 domains selective for binding pTyr-X-X-Met sequences, an interaction that appears critical for enzyme activation. The Rac-binding domain in p85α and p85β, also known as the breakpoint cluster region homology or BH domain, is homologous to RacGAPs for Rho family small G proteins but lacks GAP activity. The SH3 and proline-rich motifs can also participate in intramolecular and intermolecular interactions.

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

567

genes and five regulatory subunits (p85α, p85β, p55γ , p55α, and p50α) encoded by three genes (Figure 2). Alternative transcripts of the Pik3r1 gene encode the p85α, p55α, and p50α proteins. A single class IB isoform is composed of the p110γ catalytic subunit and the p101 regulatory subunit (Figure 2). Class IB PI3K is expressed most highly in cells of the immune system (1). Most of the immune receptors discussed in this review activate class IA and/or IB PI3Ks, leading to the production of PtdIns(3,4,5)P3. The class II PI3K subgroup in mammals comprises three catalytic isoforms (α, β, and γ ) that are larger in size than those of class I PI3Ks and do not constitutively associate with a regulatory subunit (Figure 2) (1, 5). These enzymes selectively use PtdIns as a substrate to produce PtdIns(3)P, although they may also contribute to the production of PtdIns(3,4)P2. Although some data suggest that class II PI3K enzyme activity and location are regulated by extracellular signals, little is known about their function downstream of immune cell receptors. A single isoform of class III PI3K is found in organisms ranging from yeast to humans. Class III PI3K is only able to produce PtdIns(3)P, and is localized to endocytic vesicles where its major function appears to be in protein and vesicle trafficking. There is growing evidence that class III PI3K contributes to trafficking processes unique to immune cells, especially phagocytosis (10).

Activation and Regulation of Class I PI3Ks Class I PI3Ks in resting cells are cytoplasmic proteins whose substrates reside in cellular membranes (1, 5). A number of protein-protein interactions contribute to the proper localization and activation of these enzymes following cell stimulation. For class IA PI3Ks, the best-understood mechanism of activation involves the binding of two Src-Homology 2 (SH2) domains found in all forms of the regulatory subunit (Figure 2) to phosphorylated tyrosines (pTyr) within the sequence context pTyr-X-X-Met (X is any amino acid). This SH2-pTyr interaction is triggered by receptors with either intrinsic or associated tyrosine kinase activity, bringing PI3K to membrane-associated signaling complexes and, in many cases, allowing further activation by additional interactions. Of particular importance for PI3K function are small G proteins of the Ras and Rho families. GTP-bound Ras binds to class IA catalytic subunits, an interaction shown recently to activate PI3K only in the context of a SH2-pTyr interaction (11). Rac proteins represent a subgroup of Rho family G proteins that when bound to GTP can stimulate PI3K activity by two potential mechanisms. Rac-GTP can bind directly to a domain in p85α and/or p85β that shows homology to GTPase-activating proteins (GAPs) for Rho family members (Figure 2). In addition, Rac-GTP associates with PtdIns(4)P-5kinase (PIP5K), an enzyme that can generate local increases in PtdIns(4,5)P2, the substrate for PI3K (12). Additional modular domains within the amino-terminal portions of p85α and p85β can mediate further associations that regulate PI3K function (see legend to Figure 2). Whereas the class IA PI3Ks are activated primarily by signaling pathways that involve tyrosine kinase activation, the class IB PI3K is activated by βγ subunits

11 Feb 2004

22:12

568

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

of heterotrimeric G proteins (1, 5). A growing list of G protein–coupled receptors (GPCRs) have been demonstrated to trigger PI3K pathway activation, and genetic evidence exists for a required function of class IB PI3K in PtdIns(3,4,5)P3 production triggered by certain GPCRs on leukocytes (13–15). The overall structure of the p110γ catalytic isoform is similar to the class IA enzymes, including a Ras-binding domain (Figure 2). However, the amino terminus of p110γ is distinct and mediates interaction with the p101 subunit, whose structure does not contain recognizable sequence motifs yet appears to be required for class IB enzyme activation by βγ subunits.

Multiple PI3K Effectors A number of modular protein domains have evolved to recognize specific phosphoinositides (1, 16–18). The pleckstrin homology (PH) domain is a small (∼60 aa) module found in more than 100 proteins. Although most PH domains have demonstrable affinity for phosphoinositides, only a subset is selective for PI3K products. Among this group, further specificity has been defined, with some members selective for PtdIns(3,4,5)P3 or PtdIns(3,4)P2 and others capable of binding either lipid with equivalent affinity (17). The phox homology (PX) domain has a distinct structure from PH domains but the family also contains members that bind selectively to one or more 3-phosphoinositides (18). The FYVE domain (originally named based on the first letter of four yeast proteins with the module) was identified as a PtdIns(3)P-specific binding module and is found in a number of proteins involved in protein and vesicle trafficking (16). Several proteins have been identified as likely “PI3K effectors” based on the presence of a domain selective for one or more 3-phosphoinositides (1, 2). In general, proteins with PH and PX domains are involved in signal transduction, whereas proteins with FYVE domains regulate trafficking. Phosphoinositide binding by a given PH or PX domain in some cases can directly increase activity of linked enzymatic domains, by allosteric changes and/or relieving intramolecular inhibitory interactions. However, in most cases, the membrane recruitment of a PI3K effector is one step in an activation process that can also involve posttranslational modification (i.e., phosphorylation) and additional intermolecular associations. Indeed, a paradigm has emerged in antigen receptor signaling in which PI3K lipid production is one of several crucial steps leading to the formation of large signaling complexes, or “signalosomes,” composed of numerous enzymes and scaffolding/adapter proteins (19, 20). Two groups of PI3K effectors that have received particular attention in studies of immune cell signaling are tyrosine kinases of the Tec family and serine/threonine kinases of the AGC family (21, 22). Members of the Tec family include Tec itself, Btk, Itk, Etk, and Rlk. Each member of the Tec family, except Rlk, possesses a PH domain with apparent selectivity for PtdIns(3,4,5)P3. Traditional models have posited that PtdIns(3,4,5)P3 binding by Tec family PH domains is critical for activating tyrosine kinase activity, in part by facilitating activation loop

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

569

phosphorylation by membrane-associated Src family kinases (20, 21). However, recent data have challenged that view, and modifications to this model are described below. AGC family serine/threonine kinases regulated by PI3K include phosphoinositide-dependent kinase (PDK-1), Akt (also termed PKB), certain isoforms of protein kinase C (PKC), and S6 kinase. Akt exists as three isoforms and its activation is strongly linked to PI3K-dependent proliferation and survival signals. Akt activation depends upon phosphorylation by PDK-1, which also has a 3-phosphoinositide-selective PH domain that brings it to the membrane to interact with Akt. Other important PDK-1 substrates include PKC and S6 kinase. Of note, PH and FYVE domains linked to green fluorescent protein (GFP) or its derivatives have proven highly useful as genetically encoded fusion protein probes for localization of specific 3-phosphoinositides by fluorescence microscopy. Many of the papers cited herein utilize fusions of fluorescent proteins with the PH domain of Akt to detect PtdIns(3,4,5)P3, and/or fusions with tandem FYVE domains from the protein EEA1 to detect PtdIns(3)P.

Phosphatases Two enzymes are primarily responsible for dephosphorylating PtdIns(3,4,5)P3: PTEN (Phosphatase and TENsin homolog) and SHIP (SH2-containing inositol phosphatase) (1). PTEN hydrolyzes the 3-phosphate and plays a central role in limiting cellular levels of PtdIns(3,4,5)P3, thereby opposing proliferation and survival responses (23). Loss of PTEN function is observed in a large fraction of human cancers (7). Much has been learned in recent years about PTEN function in immune cells (24). SHIP1 and SHIP2 are SH2 domain–containing phosphoinositide phosphatases that selectively remove the 5-phosphate from PtdIns(3,4,5)P3 to generate PtdIns(3,4)P2. Thus, SHIP1/2 activity may alter the spectrum of PI3Kdependent signals rather than simply opposing all PI3K signaling. SHIP1 is selectively expressed in cells of the immune system and is important for setting activation thresholds and maintaining homeostasis of a variety of hematopoietic lineages (25).

PI3K IN ANTIGEN RECEPTOR AND Fc RECEPTOR SIGNALING Signal transduction pathways initiated by the B cell receptor (BCR), T cell receptor (TCR), high-affinity IgE receptor (FCεRI), and Fcγ receptors share many common features (19, 26–28). Clustering of the receptors triggers the activation of Src family tyrosine kinases that phosphorylate ITAMs (immunoreceptor tyrosine-based activation motifs) in the cytoplasmic tails of receptor signaling chains. Subsequent activation of Syk/ZAP-70 family and Tec family tyrosine kinases controls downstream events including generation of sustained calcium (Ca2+) mobilization, GTP loading of small G proteins, and initiation of mitogen-activated protein

11 Feb 2004

22:12

570

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

(MAP) kinase cascades. Each receptor also triggers the accumulation of PI3K products and activation of Akt. Many antigen-dependent functional responses of primary cells are blocked by PI3K inhibitors (2). However, important differences exist among antigen and Fc receptor systems in the mechanisms by which PI3K and its effectors are activated, and the specific PI3K isoforms involved.

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

B Cell Receptor The mechanisms by which BCR engagement leads to the recruitment and activation of PI3K are complex. Many clues have been provided by studies in the DT40 B lymphoma system, a chicken cell line in which genes can be readily disrupted by homologous recombination. DT40 cells lacking Syk exhibit a severe reduction in BCR-dependent PI3K activation, as determined by PtdIns(3,4,5)P3 measurements (29). Several proteins in B cells are likely targets for Syk-mediated phosphorylation of Tyr-X-X-Met motifs, including CD19 and the adapter proteins BCAP and Gab1. CD19 is a major contributor to PI3K activation in mouse B cells (3, 4). This transmembrane protein is loosely associated with the BCR and the cytoplasmic tail of CD19 contains tandem pTyr-X-X-Met motifs that become phosphorylated following BCR crosslinking. Co-crosslinking CD19 with the BCR augments PI3K activation and increases the sensitivity of B cell responses by 3 to 4 orders of magnitude. Transgenic and knockout studies have confirmed that CD19 plays a role both in B cell development and in setting the activation threshold for mature B cells (30). In splenic B cells from CD19 knockout mice, anti-Ig-stimulated Akt activation is reduced by greater than tenfold (31). Two recent studies showed elegantly that activation of PI3K is the primary signaling function of CD19. In the first report, a CD19 transgene with mutations in the tyrosine residues that interact with PI3K was unable to restore function when introduced into CD19-deficient mice (32). The second study showed that the developmental and functional phenotypes of CD19 knockout mice could be complemented by B cell–specific deletion of the PTEN phosphatase (33). This paper also used flow cytometry (FACS) to measure PI3K activation with anti-PtdIns(3,4,5)P3 antibodies, a novel assay that should facilitate quantitative measurements of PI3K signaling in lymphocyte subsets and when cell yield is limiting (see also 34, 35). The adapter protein BCAP has four Tyr-X-X-Met motifs and BCAP-deficient DT40 cells have greatly impaired activation of PI3K and Akt following engagement of either the BCR or CD19 (36). However, targeting of the mouse BCAP gene had no effect on activation of PI3K or Akt in splenic B cells stimulated with anti-IgM (37). BCAP-deficient B cells did show impaired Ca2+ mobilization and proliferation. Gab1 is a member of a subgroup of adapter proteins (also including Gab2 and Gab3) that can serve as response amplifiers in PI3K signaling (38). These proteins possess PH domains selective for PtdIns(3,4,5)P3 and are recruited to membranes following PI3K activation. Subsequent phosphorylation on specific pTyr residues recruits more PI3K. Overexpression of Gab1 in an immature B cell line suggested that Gab1 can augment BCR-mediated Akt activation (39).

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

571

However, analysis of Gab1-deficient mouse B cells revealed an inhibitory role for Gab1 in antibody responses to T-independent type-2 antigens (40). Other studies are consistent with the notion that Gab family adapter proteins may have both positive and negative influences on cell activation (41). Evidence is emerging that Vav proteins also contribute to PI3K activation downstream of the BCR and related receptors. Vav1, Vav2, and Vav3 have guanine nucleotide exchange factor (GEF) activity for Rac family small G proteins, and GTP-bound Rac proteins can increase 3-phosphoinositide production by two mechanisms, as described above. Thus, Vav proteins might contribute to PI3K activation by increasing the fraction of Rac-GTP. Consistent with this, PI3K activation is reduced in DT40 cells lacking Vav3 or expressing dominant negative Rac1 (42). Mouse B cells express both Vav1 and Vav2 and combined deletion impairs B cell development and function to a degree similar to loss of specific PI3K gene products (43, 44). Vav proteins have traditionally been considered to act downstream of PI3K, as they possess PH domains and GEF activity of Vav1 was reported to be enhanced by PtdIns(3,4,5)P3 (45). However, it has not been conclusively demonstrated that PI3K activation is required for Vav activation in cells (1). To summarize the available data from BCR signaling studies, PI3K activation is driven by interactions with phosphorylated CD19 and probably Rac-GTP, whereas the functions of BCAP and Gab1 adapter proteins with respect to PI3K are not yet clear. Of the multiple catalytic and regulatory isoforms of class IA PI3K expressed in B cells, unique roles have been identified for p85α and p110δ. Two strains of p85α-deficient mice have been generated. One lacks expression of all gene products including p85α, p55α, and p50α, whereas the other lacks only p85α. Both strains exhibit defects in development, proliferation, and survival (46, 47). Mice with a null mutation in the p110δ gene, or with a kinase-dead knock-in mutation, exhibit defects in B cell development, activation, and antibody response that are generally similar to those in p85α-deficient mice (48–50). B cell function is apparently normal in mice lacking p85β (J. Deane, M. Trifilo, C. Yballe, S. Choi, T. Lane, D. Fruman, submitted manuscript). Proliferation in anti-Ig-stimulated B cells lacking p85α or p110δ is reduced to a degree similar to that in wild-type cells treated with global PI3K inhibitors, suggesting that p85α/p110δ complexes are responsible for a major fraction of PI3K signaling output. This idea was supported by one study of p110δ-deficient B cells, in which PtdIns(3,4,5)P3 generation was found to be completely blocked (49). Consistent with failure to generate PtdIns(3,4,5)P3, Akt phosphorylation was nearly abolished (48–50), as in wild-type cells treated with PI3K inhibitors. Although PtdIns(3,4,5)P3 measurements have not been reported in p85α-deficient B cells, one group demonstrated markedly reduced Akt activation (52). Ca2+ mobilization in response to anti-Ig, shown previously to be attenuated in primary B cells treated with wortmannin (53), is moderately to severely impaired in cells deficient in p85α (our unpublished data) or p110δ (48–50). Of note, Ca2+ mobilization and Akt activation are also impaired by coengagement of the BCR with the inhibitory

11 Feb 2004

22:12

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

572

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

receptor Fcγ RIIB1, which recruits SHIP, leading to metabolism of PI3K lipid products (2). Similar defects in B cell development, function, and Ca2+ mobilization are observed in mice lacking either p85α, p110δ, Btk, Vav1/Vav2, CD19, BCAP, the adapter BLNK (B cell linker), or phospholipase C-γ -2 (PLCγ 2) (2, 20, 37, 43, 44, 46–50). Ca2+ mobilization is also impaired in Rac2-deficient B cells (54). Although detailed comparison of these different genetic models reveals important differences, especially with respect to development and B cell subset differentiation (4), the overall similarity in phenotypes has suggested that these gene products function in a common pathway leading to Ca2+ mobilization (20). The intricate signaling connections among these components has suggested a “signalosome” model in which BCR engagement leads to assembly of a large complex of signaling proteins, including both upstream activators (CD19, Vav/Rac, BCAP) and downstream effectors (Btk, PLCγ 2) of PI3K (Figure 3A). 3-Phosphoinositides are thought to help with assembly and proper localization of the signalosome at the B cell membrane. In particular, various observations suggested that binding of the Btk PH domain to PtdIns(3,4,5)P3 was essential for Btk activation and subsequent phosphorylation of PLCγ 2 (20). Especially striking is the finding that Xid mice, which harbor a naturally occurring point mutation in the Btk PH domain that impairs binding to PtdIns(3,4,5)P3, have B cell defects nearly as severe as those in Btk-null mice (55). Consistent with these findings, one group reported that p110δ-deficient B cells exhibit impaired phosphorylation of tyrosines in the Btk activation loop and an autophosphorylation site (49). In contrast, others have reported that Btk kinase activation and PLCγ 2 phosphorylation are intact in B cells lacking either p85α or p110δ, as well as in wild-type cells treated with PI3K inhibitors (50, 52). The basis for the discrepant findings is not fully clear, but may be in part due to the use of different antibodies for detection of Btk phosphorylation (4). However, some of the conflicting data might be reconciled by the recent finding of a new kinase-independent function for Btk: association of the PH domain with PtdIns(4)P-5-kinase (PIP5K) (56). This enzyme can generate PtdIns(4,5)P2, the substrate for PLCγ 2. Thus, a modified model can be proposed in which PtdIns(3,4,5)P3 is not essential for Btk activation or phosphorylation of PLCγ 2, but that proper membrane targeting by PtdIns(3,4,5)P3 allows Btk to bring along an enzyme necessary to generate locally high concentrations of PtdIns(4,5)P2, allowing PLCγ 2 to generate sufficient IP3 levels for maximal Ca2+ mobilization (Figure 3A). This mechanism could help solve the problem of potential competition between PLCγ 2 and PI3K for the same substrate. However, Btk-mediated increases in local PtdIns(4,5)P2 production do not appear to be necessary for resupply of substrate for PI3K, as activation of Akt is intact in Btk-deficient B cells (52). Activation of PLCγ 2 in the signalosome leads to the production of two second messengers, diacylglycerol (DAG) and IP3, the latter important for triggering Ca2+ release. Both DAG and Ca2+ contribute to the activation of conventional PKC isoforms. The defect in proliferation to anti-Ig in B cells lacking Btk or p85α

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

573

Figure 3 Role of PI3K in early BCR signaling and transcription factor activation. Cross shapes in this figure and Figures 4–7 represent PtdIns(3,4,5)P3. (A) Current model for activation of PI3K and its effectors, including components of the BCR signalosome that promotes Ca2+ mobilization. PtdIns(3,4,5)P3 is important for proper assembly of the complex at the membrane but its role in activation of Btk has been debated (see text). Blockade by the inhibitory receptor Fcγ RIIB1 is also shown. SFK = Src-family kinase. (B) Diagram showing how PI3K can regulate multiple critical transcription factors in B cells. The activity of NFκB, AP-1, and NFAT can be enhanced, whereas FOXO function is suppressed. Another downstream target shown, S6 kinase (S6K), participates in regulation of mRNA translation. Hatched lines in this figure and others depict pathways with one or more steps omitted for simplicity.

11 Feb 2004

22:12

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

574

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

can be restored by low concentrations of phorbol ester, a DAG analog (57; our unpublished data). This suggests that failure to generate DAG and activate PKC is the central signaling defect in these cells. The PKCβ isoform may be critical in this pathway as B cells lacking PKCβ fail to proliferate to anti-Ig (58). This link has been strengthened by the finding that B cells lacking Btk, p85α, BCAP, or PKCβ are all defective in activation of the NFκB pathway and upregulation of the NFκB target gene Bcl-xL (52, 59–63). NFκB transcription factors are sequestered in the cytoplasm by IκB subunits, but are released for nuclear entry following IκB phosphorylation by the IκB kinase (IKK) complex. PKCβ is required for BCR-mediated activation of IKKα (61, 62). The importance of IKKα and other components of the NFκB pathway for B cell survival and function is well-established (64 and references therein). Notably, overexpression of a Bcl-xL transgene in B cells is sufficient to restore development and proliferation in B cells lacking Btk or p85α (52, 65). Failure to upregulate cyclin D2 expression is a common defect in cells lacking signalosome components or treated with PKC inhibitors (52, 66–68); however, this may be a direct consequence of impaired Bcl-xL induction leading to cell death (52). Several observations suggest that PI3K activation in B cells triggers important signals that are independent of Btk. Mice lacking both p85α and Btk have more severe B cell defects than single knockouts have (52). In addition, PI3K and Btk have both shared and distinct target genes, as determined by microarray analysis (69). PDK-1 and Akt are likely mediators of critical Btk-independent signals downstream of PI3K. Akt phosphorylates and inactivates FOXO proteins, a family of transcription factors that promote quiescence (70). Several potential FOXO target genes are not appropriately downregulated in PI3K-deficient cells (69, 70). Akt and novel PKC isoforms (also PDK-1 substrates) may also feed into the NFκB pathway independently of the signalosome and PKCβ (52, 71, 72). In addition, PDK-1 and Akt contribute to the activation of the S6 kinase pathway and subsequent increases in translation and cell size that are critical for cell cycle progression (73, 74). PI3K also promotes B cell size increases via the NFκBdependent upregulation of c-Myc (75). In this regard, we recently reported that sustained PI3K activation and concomitant increases in cell size are required for cell cycle progression and for continued mitotic activity of daughter cells (76). Although activation of the NFκB pathway and inactivation of FOXO proteins have emerged as critical downstream events in PI3K signaling in B cells, PI3K also can influence other key transcription factors (Figure 3B). PI3K activation could promote NFAT nuclear accumulation in two ways: by enhancing Ca2+ mobilization and calcineurin-dependent NFAT dephosphorylation, leading to nuclear import, and by Akt-mediated inactivation of GSK-3 (glycogen synthase kinase-3), a kinase that can phosphorylate NFAT and drive nuclear export (72). PI3K inhibitors have been reported to diminish BCR-dependent activation of Erk, likely leading to impaired AP-1 transcriptional activation (77). Studies of B cells lacking PTEN or SHIP1 have demonstrated that precise regulation of PI3K signaling is essential for normal BCR responses. In PTEN-deficient

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

575

B cells, treatment with anti-Ig results in elevated production of PtdIns(3,4,5)P3, enhanced and sustained phosphorylation of Akt, and markedly increased proliferation (33, 78). However, class switch recombination is impaired, owing to an inability to induce expression of activation-induced cytidine deaminase. As might be expected, one group reported enhanced survival of PTEN-deficient B cells (78); in contrast, another group reported increased susceptibility to apoptosis, which they attributed to inappropriate cell cycle entry (33). Analysis of SHIP1-deficient B cells confirmed its central role in inhibitory signaling downstream of Fcγ RIIB1, and showed that loss of this control mechanism correlated with elevated basal levels of serum Ig (79, 80).

T Cell Receptor Physiological engagement of the TCR triggers the rapid and sustained production of PtdIns(3,4,5)P3. Elegant studies of T cells expressing GFP-PH domain fusion probes showed that the lipid is concentrated at the site of antigen contact, although it distributes throughout the plasma membrane (81–83). When PI3K inhibitors or antibodies that block TCR-MHC contact are added at various time points after activation, PtdIns(3,4,5)P3 disappears rapidly from the membrane, indicating that active phosphatases turn off the signal in the absence of prolonged TCR engagement and PI3K enzyme activation (81, 83). 3-Phosphoinositide production appears functionally important in primary T cells as PI3K inhibitors block proliferation driven by antigen, and in some experiments, anti-CD3ε antibodies. Moreover, studies of T cells lacking PTEN or expressing activated forms of PI3K or Akt have uniformly supported a role for the PI3K/Akt pathway in promoting T cell proliferation and survival (6). On the other hand, some experiments have suggested that PI3K activation plays a role in opposing T cell-activation signals (72, 84). Some of these data were gathered from studies of the Jurkat cell line, which has aberrant PI3K signaling due to the absence of PTEN and SHIP1 (84). Nevertheless, it is becoming clear that the inputs and outputs of PI3K signaling in T cells are distinct in many ways from B cells. The process of antigen recognition by T cells is more complex than that of B cells. TCR engagement by MHC-peptide on antigen-presenting cells (APC) is accompanied by multiple receptor-ligand interactions that can influence signaling. Indeed, productive T cell activation requires coreceptor engagement (CD4 or CD8) and a second costimulatory signal in addition to the primary TCR-mediated set of signals. Even purified T cells stimulated with crosslinking antibodies may receive signals from neighboring cells. Thus, it has been difficult to study TCR activation of PI3K in isolation. To briefly review data acquired primarily from antibody crosslinking experiments in T cell lines, TCR engagement has been reported to lead to association of p85 regulatory subunits with Rac1 and various adapter proteins, as well as cytoplasmic domains of the TCR signaling chains (72). There are few definitive data establishing a role for these interactions in PI3K activation in primary mature T cells; however, the link between Rac and PI3K was strengthened

11 Feb 2004

22:12

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

576

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

Figure 4 Role of PI3K in TCR signaling and regulation of gene expression. The costimulatory receptor CD28 is shown as the major contributor to PI3K/Akt activation but signaling from the TCR itself can also activate PI3K. 3-Phosphoinositides may not be required for Ca2+ flux and Ca2+-dependent events in T cells due to the expression of Rlk.

by the finding that levels of Rac-GTP and Akt phosphorylation in TCR-activated thymocytes are markedly reduced in the absence of Vav1 (85) (Figure 4). One clear mechanism to activate PI3K in T cells is via engagement of costimulatory receptors CD28 or ICOS (inducible costimulator) (Figure 4). CD28 is the primary costimulatory molecule on resting T cells, and ligation of CD28 with antibodies or physiological ligands of the B7 family results in rapid activation of PI3K and Akt. The cytoplasmic tail of CD28 contains a Tyr-Met-Gln-Met sequence that can bind p85 regulatory subunits when phosphorylated. Notably, the Gln residue also makes the phosphotyrosine residue competent to bind the adapter protein Grb2, perhaps in competition with PI3K. The ICOS molecule is related to CD28 but binds distinct ligands and appears to function more in responses of activated T cells than in initial activation. The ICOS cytoplasmic tail contains a sequence Tyr-Met-Phe-Met that can bind PI3K but not Grb2. ICOS engagement leads to remarkably greater recruitment of PI3K than CD28 engagement (86). Studies of T cell function in mice lacking class IA regulatory isoforms have yielded unexpected results. The predominant regulatory isoform p85α (or p85α/ p55α/p50α), while critical for B cell function, is not essential for proliferation driven by the TCR or TCR+CD28 (46, 47). T cells lacking the p85β isoform

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

577

proliferate normally in response to anti-CD3 plus anti-CD28 but show a surprising enhancement in cell division and survival when treated with anti-CD3 alone (+/− exogenous IL-2) (J. Deane, M. Trifilo, C. Yballe, S. Choi, T. Lane, D. Fruman, submitted manuscript). It seems reasonable to propose that p85α and p85β (and perhaps p50α, which is also expressed in T cells) have redundant functions in mitogenic signaling, but this has not yet been tested in compound mutant mice. The roles of catalytic isoforms p110δ and p110γ in T cell function have been somewhat controversial. In mice with a kinase-dead knock-in mutation in p110δ, several aspects of T cell function are impaired, including Akt activation, Ca2+ mobilization, and anti-CD3-mediated proliferation (48). Although proliferation in response to co-crosslinking of CD3 and CD28 was not impaired in the absence of p110δ, proliferation of TCR transgenic T cells stimulated by cognate antigen was reduced. Another group analyzing p110δ-null mice reported normal T cell proliferation and Ca2+ mobilization, although the response to ConA was partially defective (50). It is possible that compensatory upregulation of p110α and p110β isoforms occurred in p110δ-null T cells but not the p110δ kinase-deficient T cells. Three groups disrupted the p110γ gene and two reported defects in T cell development (13–15, 87). In one study, p110γ -deficient mature T cells stimulated with anti-CD3 +/− anti-CD28 showed impaired proliferation, yet early signaling appeared to be intact (15). A plausible model is that p110γ becomes activated via an as yet undefined GPCR and contributes to sustained PtdIns(3,4,5)P3 production in T cells activated via the TCR. Similar mechanisms have been described in mast cells stimulated via FcεRI, monocytes stimulated with Fcγ RI, and fibroblasts stimulated with platelet-derived growth factor (88–90). T cell proliferation is driven by the autocrine cytokine IL-2 that is produced in response to TCR engagement with appropriate costimulation. Like many cytokines, IL-2 is a potent activator of PI3K and Akt (2). IL-2-dependent events linked to the PI3K/Akt pathway include upregulation of Bcl-XL, activation of E2F transcription factors, and increased S6 kinase activity (2). The mechanisms of PI3K activation by IL-2 and other mitogenic cytokines are reviewed in Reference (2). There is conflicting evidence concerning whether PI3K is essential for IL-2 production. Wortmannin does not inhibit IL-2 production by murine T cells stimulated with anti-CD3 in the absence or presence of B7+ B cell blasts (91, 92), whereas another study found the opposite (93). When T cells are stimulated with APC+ peptide, PI3K has generally been found to be required for maximal IL-2 production (48, 92). Other early activation events are likely to be PI3K-independent. Recently, it was reported that LY294002 does not disrupt the immunological synapse nor prevent the upregulation of activation markers in T cells stimulated with APC+ peptide (81). Interestingly, the drug was completely effective at blocking cell division when added 9 h after mixing T cells with APC. Thus, T cells share with B cells a requirement for sustained PI3K activation to proceed through the cell cycle (76). The Ca2+ mobilization response following antigen recognition by T cells is less dependent on PI3K than in B cells. PI3K inhibitors have little effect on Ca2+ flux

11 Feb 2004

22:12

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

578

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

in response to APC-peptide (82, 94). A partial impairment in Ca2+ mobilization was reported in the p110δ kinase-deficient T cells, but not in p110δ-null cells (48, 50). The lack of a marked effect on the Ca2+ response is consistent with the observations that PI3K inhibitors fail to completely block IL-2 production or activation marker induction. A role for 3-phosphoinositides in aspects of Ca2+ mobilization cannot be fully discounted, especially given the novel observation that PtdIns(3,4,5)P3 directly stimulates Ca2+ entry into T cells, but not B cells, via a membrane channel (95). TCR-dependent Ca2+ mobilization does appear to require assembly of a signalosome analogous to the one in B cells, but with distinct components (2, 4) (Figure 4). T cells express Tec and Itk, Tec family kinases with PH domains homologous to the Btk PH domain. However, T cells also express Rlk, which is targeted to the membrane by palmitoylation and is presumably PI3K-independent. Rlk is essential for Ca2+ mobilization and subsequent functional responses in T cells that also lack Itk (96). The binding of Itk to PtdIns(3,4,5)P3 is likely to be functionally significant as Itk is constitutively membrane-associated in Jurkat T cells (97), which have elevated basal PtdIns(3,4,5)P3 levels due to the absence of PTEN and SHIP phosphatases (84). Interestingly, a null mutation in the Itk gene partially impairs Ca2+ mobilization, along with T cell proliferation and differentiation (98–100). The fact that PI3K inhibitors have little effect on Ca2+ mobilization whereas Itk is required, even with Rlk present, suggests that Itk has critical functions that do not require PtdIns(3,4,5)P3 binding. It is worth considering whether Itk contributes to Ca2+ mobilization by associating with PIP5K, as demonstrated for Btk in B cells. The importance of CD28 for physiological T cell activation and its defined role in PI3K activation has stimulated inquiry to understand the role of PI3K downstream of CD28. Several groups have approached this problem by expressing CD28 transgenes in the CD28-deficient background (101–103). These studies tend to support the conclusion that mutation of the Tyr residue in the Tyr-Met-Gln-Met motif impairs T cell survival and Bcl-xL upregulation, but not IL-2 production. These defects cannot be attributed solely to impaired PI3K recruitment as Grb2 also interacts with this sequence. Nevertheless, other work has established that PI3K and Akt are important for CD28-dependent enhancement of NFκB transcriptional function, and likely Bcl-xLexpression (72). A recent mechanistic advance in this regard was the identification of the MAP3K Cot1/Tpl2 as an Akt substrate upstream of the IKK complex (104). Interestingly, costimulation via ICOS does not lead to upregulation of Bcl-xL despite the greater activation of PI3K (86). In apparent conflict with the finding that CD28 can costimulate production of IL-2 in the absence of PI3K recruitment (101–103), it was reported that expression of a constitutively membrane-targeted form of Akt could restore IL-2 production in CD28-deficient T cells (105). One possibility is that a high level of sustained Akt activation is required for increased IL-2 transcription, and this is provided by the membrane-targeted allele but not by regulated synthesis of 3-phosphoinositides downstream of CD28. An intriguing observation made more recently by the same

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

579

group (72) is that NFκB activation by membrane-targeted Akt is dependent upon an intact PH domain. Thus, as in Btk, the PH domain of Akt appears to have critical functions in addition to mediating membrane recruitment. The CD28/PI3K/Akt pathway may promote survival not only by upregulating Bcl-xL, thus promoting mitochondrial integrity, but also via inhibition of death receptor signaling. A recent study found that CD28/Akt signaling blocked Fasmediated apoptosis by preventing assembly of the death-inducing signaling complex (106). This mechanism is consistent with the finding that T cells lacking PTEN, or expressing constitutively active forms of PI3K or Akt, are resistant to Fas-mediated apoptosis (6). A recent advance in the understanding of CD28 function was the finding that signaling via PI3K and Akt is essential for increased glucose transport, metabolism, and glycogen synthesis (107). These responses are essential for the ability of CD28 to costimulate proliferation and survival. The molecular link between Akt and altered metabolism is not yet established. The PI3K/Akt pathway has long been known to have an essential and evolutionarily conserved function in metabolic changes promoted by insulin and related growth factors (1). Akt-dependent phosphorylation and inactivation of FOXO transcription factors and GSK-3 are essential for metabolic responses in other systems, but it is not yet certain that these events contribute to CD28 costimulation. Of possible relevance, transgenic expression of constitutively active GSK-3β impairs T cell proliferation (108). It has also been established that FOXO proteins are phosphorylated in a PI3K/Aktdependent manner in activated T cells, and expression of an Akt-independent form of FOXO3a promotes cell cycle arrest and apoptosis in primary T cell blasts and IL-2-dependent clones (109). These effects have been linked to the FOXO target genes p27kip (a cell cycle inhibitor) and Bim (a proapoptotic protein) (Figure 4) rather than well-defined metabolic targets of FOXO such as phosphoenolpyruvate carboxylase kinase and glucose-6-phosphatase. Cbl and Cbl-b are multifunctional proteins that are phosphorylated following BCR and TCR engagements and have complex roles in signal transduction (110). Cbl proteins have multiple protein-interaction motifs, including proline-rich and phosphotyrosine sequences that together recruit PI3K p85 subunits via their SrcHomology 3 (SH3) and SH2 domains. Cbl proteins also function as E3 ubiquitin ligases that promote degradation of other associated signaling proteins, including p85 (111). A negative role for Cbl proteins in antigen receptor signaling is suggested by the hyperproliferation observed in T cells lacking Cbl or Cbl-b, and B cells lacking Cbl-b (112, 113). Strikingly, loss of Cbl-b bypasses the requirement for CD28-mediated costimulation of T cell proliferation (114). However, these phenotypes do not appear to be the result of enhanced signaling protein stability (115). Cbl-b-deficient T cells have enhanced Vav activity (114), which could increase the efficiency of PI3K activation by TCR engagement alone. The understanding of Cbl protein function is further complicated by studies showing positive regulatory functions for Cbl in BCR-driven Ca2+ mobilization in DT40 cells and in CD40-dependent PI3K activation (116, 117). Elucidating the complex functions

11 Feb 2004

22:12

580

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

of Cbl proteins will likely require the generation of knock-in mice with targeted mutations affecting individual interaction motifs.

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IgE Receptor FcεRI is the high-affinity IgE receptor that is central to allergic and inflammatory responses mediated by mast cells and basophils (27). FcεRI binds IgE and crosslinking by multivalent antigen triggers Ca2+ mobilization and other signaling pathways essential for degranulation and other responses. PI3K signaling contributes to Ca2+ mobilization and degranulation as well as to cytokine gene induction. Recent work has indicated that FcεRI crosslinking leads to association of PI3K with two signaling complexes that are spatially and functionally distinct in mast cells (118, 119). One complex is nucleated by tyrosine-phosphorylated Gab2. The other complex is analogous to the Ca2+ signalosomes of T cells and B cells and contains some of the same components, including Syk, LAT, Btk and Itk, SLP-76, Vav1, and PLCγ 1 (Figure 5). Assembly of this “LAT” complex requires Syk and the Src family kinase Lyn, whereas the Gab2 complex requires the Src family kinase Fyn (118). Class IA PI3K regulatory subunits associate directly with tyrosine-phosphorylated Gab2. This interaction appears to be essential for FcεRI-mediated Akt

Figure 5 Role of PI3K in FcεRI signaling and amplification by GPCRs. The complexes containing either LAT or Gab2 are shown as physically separated, in keeping with recent findings. Also shown is an inhibitory receptor that recruits SHP-1 tyrosine phosphatase leading to Gab2 dephosphorylation; other inhibitory receptors on mast cells can recruit SHIP-1 as well. Note that Lyn can both promote mast cell activation (by phosphorylating signaling chains of FcεRI and components of the LAT complex) and oppose activation (by phosphorylating inhibitory receptor tails).

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

581

activation and allergic responses, based on studies of Gab2-deficient mast cells (120). The mechanism of PI3K association and activation in the LAT complex is not yet clear. PtdIns(3,4,5)P3 production and Akt activation are attenuated in mast cells lacking Vav1 (121), suggesting that interaction with GTP-bound Rac proteins may contribute to PI3K activation, as reported in DT40 B cells (42). This model is consistent with a report that FcεRI-mediated Akt activation and Bcl-xL upregulation are diminished in Rac2-deficient mast cells (122). One issue that remains to be resolved is why deletion of either Gab2 or Rac2, which presumably activate distinct pools of PI3K in spatially separated signaling complexes, each results in a near-blockade of Akt activation. One possibility is that PtdIns(3,4,5)P3 acts as a diffusible second messenger to mediate crosstalk between the separate signaling complexes. In this view, small amounts of PtdIns(3,4,5)P3 are produced initially by PI3K in the LAT complex, and the lipid then acts to recruit Gab2 to another membrane site via its PH domain (Figure 5). Subsequent Gab2 phosphorylation recruits more PI3K and amplifies PtdIns(3,4,5)P3 production to a level necessary for Akt activation. Ca2+ mobilization and degranulation can be attenuated by coligation of FcεRI with various inhibitory receptors expressed on mast cells. As in B cells, binding of IgE to antigen coated with IgG leads to coengagement of Fcγ RIIB1 and resultant recruitment of SHIP to the membrane and dephosphorylation of PtdIns(3,4,5)P3 (123). Mast cells lacking SHIP have elevated basal and stimulated Ca2+-mobilization responses (25). Other inhibitory receptors may inhibit PI3K signaling by recruiting the tyrosine phosphatase SHP-1 (123), leading to dephosphorylation of critical signaling components such as Gab2. Recruitment of SHIP-1 or SHP-1 to inhibitory receptor cytoplasmic tails involves tyrosine phosphorylation of a sequence known as the ITIM (immunoreceptor tyrosine-based inhibitory motif). Phosphorylation of ITIMs by Lyn may play a central role in this feedback inhibition pathway (Figure 5), as Lyn-deficient mast cells have elevated basal Gab2 phosphorylation, PtdIns(3,4,5)P3 levels, and Akt activity (124). The roles of individual class IA PI3K isoforms in FcεRI signaling are not fully understood. Deletion of p85α/p55α/p50α does not impair FcεRI-dependent responses, including Akt phosphorylation (125). The same mast cells showed partial impairments in proliferation driven by Kit ligand, a transmembrane tyrosine kinase receptor with tandem Tyr-X-X-Met motifs. Mast cell function in vivo has not been studied in p85α/p55α/p50α knockouts because of the perinatal lethal phenotype of these mice. Mice lacking only p85α are viable and exhibit profound defects in mast cell function in vivo, but this has been attributed to defects in mast cell and T helper cell differentiation rather than a signaling defect in mature mast cells (3, 126). The phenotypes of mast cells lacking p85β or class IA catalytic subunits have not been reported. However, studies in which p110 isoform-specific antibodies were injected in the rat basophilic leukemia (RBL) cell line have suggested that p110α, p110β, and p110δ each have required functions in FcεRI-dependent degranulation, the latter two essential for Ca2+ mobilization (127, 128). Class IB and class III PI3Ks can also regulate degranulation. The p110γ isoform is required for amplification of mast cell degranulation by GPCR ligands such as

11 Feb 2004

22:12

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

582

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

adenosine (Figure 5) (90). This appears to be functionally important as p110γ deficient mice showed impairments in passive systemic anaphylaxis. Another study reported a novel function for class III PI3K in degranulation triggered by carbachol, a muscarinic receptor agonist (128). The mechanisms by which 3-phosphoinositides promote Ca2+ mobilization and Akt activation downstream of FcεRI are probably similar to those described in the BCR and TCR sections. Btk and Itk are expressed in mast cells, and Btk-deficient mast cells show impaired responses to FcεRI crosslinking (129). One group studying mast cells has also reported that PtdIns(3,4,5)P3 can directly stimulate PLCγ 1 activity (127), a finding supported by investigators working on nonhematopoietic cell types (130). PLCγ proteins possess two domains reported to interact with PtdIns(3,4,5)P3: a PH domain and a SH2 domain, and lipid binding may increase enzyme activity by altering the tertiary structure of the protein and/or stabilizing membrane association. Whether these mechanisms are also involved in PI3K-dependent Ca2+ mobilization downstream of the BCR is not known. The role of Akt in FcεRI signaling has not been studied in as much detail as in BCR or TCR/CD28 systems. However, Akt has been linked to Bcl-xL upregulation in mast cells, as in other cell types (122).

Fc Receptors for IgG Fcγ receptors (Fcγ Rs) are expressed in diverse hematopoietic cell types and mediate numerous cellular functions including phagocytosis and antibody-dependent cytotoxicity (28). Whereas the cytoplasmic tail of the inhibitory Fc receptor Fcγ RIIB1 contains an ITIM, the cytoplasmic tails of “activating” Fcγ R signaling chains contain ITAMs and initiate signaling mechanisms similar to those discussed above for the BCR, TCR, and FcεRI. In several cell types, crosslinking of Fcγ Rs has been linked to PI3K activation (2). Here we focus on Fcγ R-dependent phagocytosis, as notable recent advances have clarified the role of PI3K in this process. PI3K activity is required for phagocytosis of large (>0.5 µm) IgG-coated particles, with specific roles in pseudopod extension and contractile processes during phagosome closure (131–133). In the presence of PI3K inhibitors, smaller particles can be engulfed but phagosome maturation is impaired (134). Engulfment of large particles requires local expansion of plasma membrane, and PI3K activity is specifically required for exocytosis of membranes for this purpose (133). Fluorescence microscopy of cells expressing GFP fusion probes has been particularly informative in studies of 3-phosphoinositide dynamics during phagocytosis (134, 135). Consistent with a role in pseudopod extension and phagosome closure, PtdIns(3,4,5)P3 accumulates in the nascent phagocytic cup and disappears shortly after the phagosome is sealed (Figure 6) (135). Interestingly, the 3phosphoinositide species required for phagosome maturation is PtdIns(3)P, which appears in the phagosome membrane only after sealing (Figure 6). Immunofluorescent staining with antibodies to EEA1, an endosomal protein with a FYVE domain, and LAMP-1, a lysosome marker, was used to demonstrate impaired phagosome maturation in macrophages treated with wortmannin (Figure 6) (134).

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

583

Figure 6 Role of PI3K in Fcγ R-mediated phagocytosis and phagosome maturation. Numbers represent temporal sequence of events. The initial activation of PI3K in the LAT complex downstream of Vav and Rac is speculative (see text). PtdIns(3)P in the membrane is represented by the inverted L-shapes.

Src family kinases and Syk are critical for tyrosine phosphorylation events that initiate Fcγ R signaling (2, 28). As in the FcεRI system, Gab2 contributes to activation of PI3K and Akt, and is required for Fcγ R-dependent phagocytosis (137). Gab2 is recruited to the phagocytic cup, a process that requires its PH domain and PI3K activity. Mutation of the p85 binding site on Gab2 reduces PtdIns(3,4,5)P3 levels in the phagocytic cup. Thus, Gab2 is both downstream and upstream of PI3K, serving as an amplifier of 3-phosphoinositide production at a focused membrane site critical for phagocytosis. How does Fcγ R engagement trigger initial activation of PI3K required for Gab2 recruitment? As in mast cells, a complex involving LAT may be important. Class IA PI3K associates with tyrosinephosphorylated LAT following Fcγ R crosslinking (138), and macrophages lacking LAT exhibit impaired phagocytosis (139). Vav is involved in Fcγ R-dependent Rac activation (140), but whether this is involved in PI3K activation is not known (Figure 6). Importantly, phagocytosis is enhanced in cells lacking Fcγ RIIB1 or SHIP, indicating that attenuating the PI3K pathway is a shared mechanism of inhibitory signaling in phagocytes as in mast cells and B cells (141). Consistent with detection of PtdIns(3,4,5)P3 in the phagocytic cup and PtdIns(3)P in sealed phagosomes, class I PI3K is essential for particle engulfment,

11 Feb 2004

22:12

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

584

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

whereas class III PI3K is essential for phagosome maturation (Figure 6) (134). In this report, class IA isoforms were shown to contribute to phagocytosis in experiments in which fibroblasts lacking most regulatory isoforms (p85α/p55α/p50α/ p85β) were transfected with Fcγ RIIA. Compared with wild-type fibroblast transfectants, PtdIns(3,4,5)P3 production and engulfment of large particles was significantly impaired. Specific roles for the class IA catalytic isoform p110β in phagocytosis, and the mammalian class III PI3K (VPS34) in phagosome maturation, were demonstrated recently by microinjection of specific antibodies (134, 142, 143). As in mast cells stimulated via FcεRI, PtdIns(3,4,5)P3 production in monocytes stimulated via Fcγ RI is amplified by activation of the class IB isoform p110γ (88). The PtdIns(3,4,5)P3-binding proteins that mediate exocytosis of intracellular membrane reserves have not been established, but may include regulators of the small G protein Arf6 (10, 144). 3-Phosphoinositides have defined roles in actin cytoskeletal remodeling events involved in membrane ruffling and directional movement (see below), yet PI3K appears to be dispensable for actin polymerization during phagocytosis (133, 134). However, an atypical form of myosin that contains a PH domain was recently implicated as a link between PtdIns(3,4,5)P3 and pseudopod extension (145).

PI3K IN OTHER LEUKOCYTE RECEPTOR SIGNALING SYSTEMS ITAM/ITIM-Containing Receptors on NK Cells Natural killer (NK) cells are important for the innate immune response to viruses and tumor cells, yet share many attributes with effector cells of the adaptive immune response (146). NK cell activation is based on the integration of signals from activating receptors containing ITAMs and inhibitory receptors containing ITIMs (147). As for inhibitory receptors in other cell types, ITIM sequences in NK inhibitory receptors can recruit SHIP1 as well as SH2-containing tyrosine phosphatases (148, 149). An important role for 3-phosphoinositides in NK cell development and function is supported further by the finding that mice lacking SHIP1 have a marked increase in NK cells, combined with a decrease in NKmediated bone marrow graft rejection (148). Several NK-activating receptors share the ability to activate PI3K, including NKG2D, CD28, and 2B4 (reviewed in 150). We focus here on NKG2D, whose signaling mechanisms have been clarified recently. NKG2D associates with two adapter proteins, DAP10 and DAP12, which mediate assembly of distinct signaling complexes (151). DAP10 possesses a Tyr-Ile-Gln-Met motif that can interact with class IA PI3K, whereas DAP12 possesses ITAMs that are linked to activation of Syk and ZAP-70 (151, 152). Recent studies of both human and mouse NK cells have indicated that DAP12 is primarily responsible for triggering cytokine production, whereas DAP10 is important for cytotoxicity downstream of NKG2D (153, 154).

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

585

Cytotoxicity is blocked by LY294002 or mutation of the Tyr-Ile-Gln-Met motif in DAP10, emphasizing the importance of PI3K activation in the DAP10 complex.

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Receptors for Chemoattractants Regulated synthesis and localization of 3-phosphoinositides play an evolutionarily conserved role in directional movement of cells exposed to gradients of chemoattractants (155, 156). PI3K inhibitors inhibit leukocyte chemotaxis induced by microbial products (e.g., fMLP), inflammatory products (e.g., C5a), and numerous chemokines. Fluorescence imaging of cells expressing GFP-PH fusion probes has revealed a striking concentration of PtdIns(3,4,5)P3 at the leading edge of migrating neutrophils (157), and there is considerable evidence that 3-phosphoinositides help determine cell polarity during directional movement (10). Recent work has provided new mechanistic insights into how PtdIns(3,4,5)P3 is concentrated at the leading edge of migrating cells, even in shallow gradients of chemoattractant (158, 159). These studies took advantage of two methodological advances: detection of PtdIns(3,4,5)P3 with a GFP-PH fusion probe, and the ability to deliver exogenous PtdIns(3,4,5)P3 to the cytoplasmic leaflet of the membrane using cationic vesicles. Using these techniques, it was shown that PtdIns(3,4,5)P3 delivery to neutrophils induces a positive feedback loop leading to production of more PtdIns(3,4,5)P3 by endogenous PI3K. The mechanism involves PtdIns(3,4,5)P3dependent activation of Rho family G proteins, which in turn amplify PI3K activity (Figure 7). Interestingly, actin polymerization at the leading edge, which requires PtdIns(3,4,5)P3 and Rho family G proteins, also plays a role in the positive feedback loop by helping to maintain proper polarity in PtdIns(3,4,5)P3 accumulation (159). The Rac subfamily of Rho G proteins is probably involved in the feedback amplification loop. Recently, the RacGEF pRex-1 was purified from neutrophil extracts based on its ability to mediate PtdIns(3,4,5)P3-dependent GTP loading of Rac (160). pRex-1 has a PH domain but this has not yet been demonstrated to mediate the effects of PtdIns(3,4,5)P3. Importantly, pRex-1 activation is also promoted by G protein βγ subunits, and antisense inhibition of pRex-1 expression inhibits Rac-dependent superoxide production induced by C5a. These findings support the model that pRex-1 is a critical mediator of PtdIns(3,4,5)P3-dependent Rac activation downstream of GPCRs (Figure 7) (161). Most chemoattractant receptors are members of the GPCR superfamily. Thus, it is not surprising that the sole class IB PI3K catalytic isoform p110γ is required for PtdIns(3,4,5)P3 production and Akt activation downstream of the receptors for fMLP, C5a, and IL-8 (13–15). Mice lacking p110γ show reduced neutrophil and macrophage chemotaxis in vitro and impaired leukocyte recruitment to inflammatory sites in vivo. p110γ -deficient neutrophils show impaired Rac activation and F-actin accumulation at the leading edge (14, 162). Class IA PI3K can also be activated by GPCRs and contribute to chemotaxis (162a). A specific role for p110δ in neutrophil chemotaxis to GPCR ligands was recently demonstrated using a novel

11 Feb 2004

22:12

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

586

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

Figure 7 Role of PI3K in chemotaxis to GPCR ligands (fMLP shown in model). In this two-step model, initial activation of class IB PI3K synergizes with Gβγ subunits to activate the Rac exchange factor pRex-1. Subsequent Rac activation and actin polymerization (hatched area) contribute to feedback amplification of PtdIns(3,4,5)P3 production, in part via class IA PI3K isoforms. This could be mediated directly via a Rac-class IA PI3K interaction, via recruitment of PIP5K to generate more substrate, or through other mechanisms such as polarized accumulation of tyrosine kinases that are transactivated by GPCR signaling.

inhibitor specific for this class IA isoform (163). Notably, F-actin synthesis was not blocked. Thus, a plausible model is that p110δ is involved in the Rac- and Factin–dependent feedback amplification loop, whereas PtdIns(3,4,5)P3 produced by p110γ cooperates with Gβγ subunits to initiate actin polymerization via pRex1 and Rac activation (Figure 7). Class IA catalytic isoforms are also involved in chemotaxis of macrophages to colony stimulating factor, whose receptor possesses intrinsic tyrosine kinase activity (164). Certain chemotactic ligands (e.g., fMLP, C5a) also trigger neutrophils to release reactive oxygen species via the activation of a complex known as the NADPH

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

587

oxidase. This response, also known as oxidative burst, requires activation of class IB PI3K, pRex-1, and Rac (13–15, 160, 165). Interestingly, priming of neutrophils with TNFα or GM-CSF, which greatly enhances fMLP-mediated oxidative burst, correlates with enhanced production of PtdIns(3,4,5)P3 via p110γ (166). In addition to the indirect role of 3-phosphoinositides in NADPH oxidase activation via pRex-1 and Rac, PtdIns(3)P and PtdIns(3,4)P2 bind directly to cytosolic components of the complex to facilitate assembly of an active complex at the phagosomal membrane (reviewed in 2, 18). Lymphocyte migration within lymphoid organs and to sites of infection and inflammation is controlled by a variety of chemokines that act via GPCRs. The mechanisms by which PI3K contributes to lymphocyte chemotaxis are less well studied than those in neutrophils and macrophages. The p110γ isoform seems likely to play a role, but this has not been directly tested. However, defects in T and B cell development observed in p110γ -deficient mice (14, 15) may be the result of impaired migration. A role for Rac proteins in lymphocyte chemotaxis is supported by the finding of impaired chemokine-induced migration of T and B cells lacking Rac2 (54, 167). However, it is not yet clear whether PI3K acts upstream and/or downstream of Rac in lymphocyte chemotaxis. It has been reported that B cells lacking one or both copies of the PTEN gene show enhanced chemotaxis in response to a number of chemokines (78, 168). On the other hand, another group found impaired chemotaxis in PTEN-deficient B cells despite increased basal Rac activation (33). Although the reason for the conflicting results is not clear, it is important to note that PTEN in Dictyostelium has an essential role in directional migration by metabolizing PtdIns(3,4,5)P3 at the trailing edge of the cell (155). These findings emphasize that augmented PtdIns(3,4,5)P3 production can promote chemotaxis only when produced at a focal area of the cell.

PERSPECTIVES PI3K activation is a common signal transduction event in a remarkable variety of functional responses in leukocytes. How does production of 3-phosphoinositides promote different responses downstream of distinct receptors? As emphasized throughout this review, an important mechanism is that PI3K effector proteins must integrate signals from lipid binding and other intermolecular interactions in order to be recruited to signaling complexes and activated. Thus, the spectrum of PI3K effectors engaged by distinct receptors differs. Examples include synergistic activation of pRex-1 by PtdIns(3,4,5)P3 and Gβγ downstream of GPCRs, and activation of Tec family tyrosine kinases by membrane-associated Src family kinases that are activated by ITAM-containing receptors. Feedback amplification loops further enhance localized synthesis of 3-phosphoinositides, augmenting and sustaining the activation of PI3K effectors at focal areas of the cell where dynamic changes in membrane dynamics occur. Examples include Gab adapter-based amplification (Figure 5 and 6) and the Rac-mediated feedback loop in chemotaxis (Figure 7).

11 Feb 2004

22:12

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

588

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

A paradigm has also emerged in antigen and Fc receptor systems in which receptor occupancy leads to the assembly of large macromolecular complexes containing PI3K, its lipid products, along with PI3K activators and effectors (Figures 3–6). In these large complexes or signalosomes, information is exchanged among components and the distinctions between upstream and downstream events can be obscured. Removal of any component can disrupt function of the entire complex. An additional level of organization is probably provided by the localization of antigen/Fc receptor signaling complexes in glycolipid-enriched membrane microdomains, also known as lipid rafts. Although not emphasized in this review, lipid rafts clearly play an important role in assembly of signalosomes; conversely, receptor signaling can promote aggregration of lipid rafts (169, 170). There is growing evidence for a relationship between PI3K signaling and lipid raft aggregation. In B cells, recruitment of SHIP to lipid rafts by the inhibitory receptor Fcγ RIIB1 inhibits raft aggregation (171). In T cells lacking p110δ, raft aggregation is impaired following cocrosslinking of CD3 and CD28 (48). Clearly, there is much still to learn about the interplay between lipid rafts, PI3K and signalosome assembly. Other questions worthy of further investigation include: How does the composition of membrane-associated signalosomes differ downstream of different receptors, especially those less well studied than the BCR and TCR? What are the unique functions of PI3K isoforms in various signaling complexes, and do these functions change at different times during development? Which modular domains of PI3K catalytic and regulatory isoforms are critical for enzyme activation in different contexts? How does the role of PI3K signaling differ at distinct points during cell cycle progression of activated lymphocytes? New methodological tools are helping to address these questions more easily under physiological conditions. Given the heterogeneity of primary cells, it will be especially important to study signal transduction at the single-cell level, using fluorescence-based technologies such as confocal microscopy and flow cytometry (34). When reviewing the literature on PI3K, an easier task would perhaps be to list the few responses that are not dependent on this ubiquitous signaling pathway. Here we have noted a few instances in which a particular biological response can be either PI3K-dependent or -independent, depending on cell type or receptor system. For example, actin polymerization is regulated by PI3K signaling during chemotaxis but appears to be PI3K-independent during pseudopod extension and phagocytosis. Ca2+ mobilization is clearly influenced by PI3K following BCR crosslinking, but apparently less so following TCR engagement. Further detailed analysis of which biological responses require PI3K activation, and which PI3K isoforms and effector proteins are involved, will eventually allow predictions to be made about components within the pathway that could be viable pharmacological targets for different disease states. Already a compound has been developed that inhibits p110δ, a leukocyte-specific PI3K catalytic isoform, and shown to impair neutrophil migration (163). It is hoped that continued efforts to define specificity in PI3K signaling will uncover opportunities to manipulate the pathway for therapeutic benefit (172).

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

PI-3 KINASE IN THE IMMUNE SYSTEM

P1: IKH

589

ACKNOWLEDGMENTS

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

We thank Bart Vanhaesebroeck, Klaus Okkenhaug, Henry Bourne, Juan Rivera, Steven Greenberg, Amber Donahue, Michael Kharas, and Jean Oak for helpful comments on the manuscript, and Chris Carpenter for sharing unpublished data. Unpublished work from our lab that is cited in this review was supported by NIH grant AI50831 to D.A.F. No review of PI3K can adequately describe every interesting paper, and we apologize to those investigators whose work could not be cited because of space limitations.

NOTE ADDED IN PROOF Strong evidence that Rac proteins promote BCR-mediated PI3K activation has been provided by a study of mice with B cell-specific deletion of Rac1 and Rac2 (173). In addition, a study of Bam32-deficient mice has implicated this PH domaincontaining protein in linking BCR-mediated PI3K activation to the Erk pathway (174). The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Vanhaesebroeck B, Leevers SJ, Ahmadi K, Timms J, Katso R, et al. 2001. Synthesis and function of 3-phosphorylated inositol lipids. Annu. Rev. Biochem. 70: 535–602 2. Fruman DA, Cantley LC. 2002. Phosphoinositide 3-kinase in immunological systems. Semin. Immunol. 14:7–18 3. Koyasu S. 2003. The role of PI3K in immune cells. Nat. Immunol. 4:313– 19 4. Okkenhaug K, Vanhaesebroeck B. 2003. PI3K in lymphocyte development, differentiation and activation. Nat. Rev. Immunol. 3:317–30 5. Fruman DA, Meyers RE, Cantley LC. 1998. Phosphoinositide kinases. Annu. Rev. Biochem. 67:481–507 6. Ohashi PS. 2002. T-cell signalling and autoimmunity: molecular mechanisms of disease. Nat. Rev. Immunol. 2:427– 38 7. Vivanco I, Sawyers CL. 2002. The phosphatidylinositol 3-kinase AKT pathway

8.

9.

10.

11.

12.

13.

in human cancer. Nat. Rev. Cancer 2: 489–501 Fukao T, Koyasu S. 2003. PI3K and negative regulation of TLR signaling. Trends Immunol. 24:358–63 Rameh LE, Cantley LC. 1999. The role of phosphoinositide 3-kinase lipid products in cell function. J. Biol. Chem. 274:8347–50 Stephens L, Ellson C, Hawkins P. 2002. Roles of PI3Ks in leukocyte chemotaxis and phagocytosis. Curr. Opin. Cell Biol. 14:203–13 Jimenez C, Hernandez C, Pimentel B, Carrera AC. 2002. The p85 regulatory subunit controls sequential activation of phosphoinositide 3-kinase by Tyr kinases and Ras. J. Biol. Chem. 277: 41556–62 Welch HC, Coadwell WJ, Stephens LR, Hawkins PT. 2003. Phosphoinositide 3-kinase-dependent activation of Rac. FEBS Lett. 546:93–97 Hirsch E, Katanaev VL, Garlanda C,

11 Feb 2004

22:12

590

14.

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

Azzolino O, Pirola L, et al. 2000. Central role for G protein-coupled phosphoinositide 3-kinase γ in inflammation. Science 287:1049–53 Li Z, Jiang H, Xie W, Zhang Z, Smrcka AV, Wu D. 2000. Roles of PLC-β2 and -β3 and PI3Kγ in chemoattractantmediated signal transduction. Science 287:1046–49 Sasaki T, Irie-Sasaki J, Jones RG, Oliveira-dos-Santos AJ, Stanford WL, et al. 2000. Function of PI3Kγ in thymocyte development, T cell activation, and neutrophil migration. Science 287:1040– 46 Fruman DA, Rameh LE, Cantley LC. 1999. Phosphoinositide binding domains: embracing 3-phosphate. Cell 97: 817–20 Lemmon MA, Ferguson KM. 2000. Signal-dependent membrane targeting by pleckstrin homology (PH) domains. Biochem. J. 350(Part 1):1–18 Sato TK, Overduin M, Emr SD. 2001. Location, location, location: membrane targeting directed by PX domains. Science 294:1881–85 Kurosaki T. 2000. Functional dissection of BCR signaling pathways. Curr. Opin. Immunol. 12:276–81 Fruman DA, Satterthwaite AB, Witte ON. 2000. Xid-like phenotypes: a B cell signalosome takes shape. Immunity 13:1–3 Schaeffer EM, Schwartzberg PL. 2000. Tec family kinases in lymphocyte signaling and function. Curr. Opin. Immunol. 12:282–88 Cantrell D. 2002. Protein kinase B (Akt) regulation and function in T lymphocytes. Semin. Immunol. 14:19–26 Cantley LC, Neel BG. 1999. New insights into tumor suppression: PTEN suppresses tumor formation by restraining the phosphoinositide 3-kinase/AKT pathway. Proc. Natl. Acad. Sci. USA 96:4240–45 Kishimoto H, Hamada K, Saunders M,

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

Backman S, Sasaki T, et al. 2003. Physiological functions of pten in mouse tissues. Cell Struct. Funct. 28:11–21 Rohrschneider LR, Fuller JF, Wolf I, Liu Y, Lucas DM. 2000. Structure, function, and biology of SHIP proteins. Genes Dev. 14:505–20 Kane LP, Lin J, Weiss A. 2000. Signal transduction by the TCR for antigen. Curr. Opin. Immunol. 12:242–49 Turner H, Kinet JP. 1999. Signalling through the high-affinity IgE receptor Fc εRI. Nature 402:B24–30 Ravetch JV, Bolland S. 2001. IgG Fc receptors. Annu. Rev. Immunol. 19:275– 90 Beitz LO, Fruman DA, Kurosaki T, Cantley LC, Scharenberg AM. 1999. SYK is upstream of phosphoinositide 3-kinase in B cell receptor signaling. J. Biol. Chem. 274:32662–66 Fujimoto M, Poe JC, Inaoki M, Tedder TF. 1998. CD19 regulates B lymphocyte responses to transmembrane signals. Semin. Immunol. 10:267–77 Otero DC, Omori SA, Rickert RC. 2001. CD19-dependent activation of Akt kinase in B-lymphocytes. J. Biol. Chem. 276:1474–78 Wang Y, Brooks SR, Li X, Anzelon AN, Rickert RC, Carter RH. 2002. The physiologic role of CD19 cytoplasmic tyrosines. Immunity 17:501–14 Anzelon AN, Wu H, Rickert RC. 2003. Pten inactivation alters peripheral B lymphocyte fate and reconstitutes CD19 function. Nat. Immunol. 4:287–94 Perez OD, Nolan GP. 2002. Simultaneous measurement of multiple active kinase states using polychromatic flow cytometry. Nat. Biotechnol. 20:155–62 Perez OD, Kinoshita S, Hitoshi Y, Payan DG, Kitamura T, et al. 2002. Activation of the PKB/AKT pathway by ICAM-2. Immunity 16:51–65 Okada T, Maeda A, Iwamatsu A, Gotoh K, Kurosaki T. 2000. BCAP: the tyrosine kinase substrate that connects B cell

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

PI-3 KINASE IN THE IMMUNE SYSTEM

37.

38.

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

39.

40.

41.

42.

43.

44.

45.

receptor to phosphoinositide 3-kinase activation. Immunity 13:817–27 Yamazaki T, Takeda K, Gotoh K, Takeshima H, Akira S, Kurosaki T. 2002. Essential immunoregulatory role for BCAP in B cell development and function. J. Exp. Med. 195:535–45 Liu Y, Rohrschneider LR. 2002. The gift of Gab. FEBS Lett. 515:1–7 Ingham RJ, Santos L, Dang-Lawson M, Holgado-Madruga M, Dudek P, et al. 2001. The Gab1 docking protein links the B cell antigen receptor to the phosphatidylinositol 3-kinase/Akt signaling pathway and to the SHP2 tyrosine phosphatase. J. Biol. Chem. 276:12257–65 Itoh S, Itoh M, Nishida K, Yamasaki S, Yoshida Y, et al. 2002. Adapter molecule Grb2-associated binder 1 is specifically expressed in marginal zone B cells and negatively regulates thymusindependent antigen-2 responses. J. Immunol. 168:5110–16 Lynch DK, Daly RJ. 2002. PKBmediated negative feedback tightly regulates mitogenic signalling via Gab2. EMBO J. 21:72–82 Inabe K, Ishiai M, Scharenberg AM, Freshney N, Downward J, Kurosaki T. 2002. Vav3 modulates B cell receptor responses by regulating phosphoinositide 3-kinase activation. J. Exp. Med. 195:189–200 Doody GM, Bell SE, Vigorito E, Clayton E, McAdam S, et al. 2001. Signal transduction through Vav-2 participates in humoral immune responses and B cell maturation. Nat. Immunol. 2:542–47 Tedford K, Nitschke L, Girkontaite I, Charlesworth A, Chan G, et al. 2001. Compensation between Vav-1 and Vav-2 in B cell development and antigen receptor signaling. Nat. Immunol. 2:548–55 Han J, Luby-Phelps K, Das B, Shu X, Xia Y, et al. 1998. Role of substrates and products of PI 3-kinase in regulating activation of Rac-related guanosine triphosphatases by Vav. Science 279:558–60

P1: IKH

591

46. Fruman DA, Snapper SB, Yballe CM, Davidson L, Yu JY, et al. 1999. Impaired B cell development and proliferation in absence of phosphoinositide 3kinase p85α. Science 283:393–97 47. Suzuki H, Terauchi Y, Fujiwara M, Aizawa S, Yazaki Y, et al. 1999. Xid-like immunodeficiency in mice with disruption of the p85α subunit of phosphoinositide 3-kinase. Science 283:390–92 48. Okkenhaug K, Bilancio A, Farjot G, Priddle H, Sancho S, et al. 2002. Impaired B and T cell antigen receptor signaling in p110δ PI 3-kinase mutant mice. Science 297:1031–34 49. Clayton E, Bardi G, Bell SE, Chantry D, Downes CP, et al. 2002. A crucial role for the p110δ subunit of phosphatidylinositol 3-kinase in B cell development and activation. J. Exp. Med. 196:753– 63 50. Jou ST, Carpino N, Takahashi Y, Piekorz R, Chao JR, et al. 2002. Essential, nonredundant role for the phosphoinositide 3-kinase p110δ in signaling by the B-cell receptor complex. Mol. Cell. Biol. 22:8580–91 51. Deleted in proof 52. Suzuki H, Matsuda S, Terauchi Y, Fujiwara M, Ohteki T, et al. 2003. PI3K and Btk differentially regulate B cell antigen receptor-mediated signal transduction. Nat. Immunol. 4:280–86 53. Buhl AM, Pleiman CM, Rickert RC, Cambier JC. 1997. Qualitative regulation of B cell antigen receptor signaling by CD19: selective requirement for PI3-kinase activation, inositol-1,4,5trisphosphate production Ca2+ mobilization. J. Exp. Med. 186:1897–910 54. Croker BA, Tarlinton DM, Cluse LA, Tuxen AJ, Light A, et al. 2002. The Rac2 guanosine triphosphatase regulates B lymphocyte antigen receptor responses and chemotaxis and is required for establishment of B-1a and marginal zone B lymphocytes. J. Immunol. 168:3376– 86

11 Feb 2004

22:12

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

592

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

55. Khan WN, Alt FW, Gerstein RM, Malynn BA, Larsson I, et al. 1995. Defective B cell development and function in Btk-deficient mice. Immunity 3:283–99 56. Saito K, Tolias KF, Saci A, Koon HB, Humphries LA, et al. 2003. BTK regulates PtdIns-4,5-P2 synthesis. Importance for calcium signaling and PI3K activity. Immunity 19:669–77 57. Forssell J, Nilsson A, Sideras P. 1999. Bruton’s tyrosine-kinase-deficient murine B lymphocytes fail to enter S phase when stimulated with anti-immunoglobulin plus interleukin-4. Scand. J. Immunol. 49:155–61 58. Leitges M, Schmedt C, Guinamard R, Davoust J, Schaal S, et al. 1996. Immunodeficiency in protein kinase cβdeficient mice. Science 273:788–91 59. Bajpai UD, Zhang K, Teutsch M, Sen R, Wortis HH. 2000. Bruton’s tyrosine kinase links the B cell receptor to nuclear factor κB activation. J. Exp. Med. 191:1735–44 60. Petro JB, Rahman SM, Ballard DW, Khan WN. 2000. Bruton’s tyrosine kinase is required for activation of IκB kinase and nuclear factor κB in response to B cell receptor engagement. J. Exp. Med. 191:1745–54 61. Saijo K, Mecklenbrauker I, Santana A, Leitger M, Schmedt C, Tarakhovsky A. 2002. Protein kinase C β controls nuclear factor κB activation in B cells through selective regulation of the IκB kinase α. J. Exp. Med. 195:1647–52 62. Su TT, Guo B, Kawakami Y, Sommer K, Chae K, et al. 2002. PKC-B controls IκB kinase lipid raft recruitment and activation in response to BCR signaling. Nat. Immunol. 3:780–86 63. Yamazaki T, Kurosaki T. 2003. Contribution of BCAP to maintenance of mature B cells through c-Rel. Nat. Immunol. 4:780–786 64. Kaisho T, Takeda K, Tsujimura T, Kawai T, Nomura F, et al. 2001. IκB kinase α is essential for mature B cell development

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

and function. J. Exp. Med. 193:417– 26 Solvason N, Wu WW, Kabra N, LundJohansen F, Roncarolo MG, et al. 1998. Transgene expression of bcl-xL permits anti-immunoglobulin (Ig)-induced proliferation in xid B cells. J. Exp. Med. 187:1081–91 Brorson K, Brunswick M, Ezhevsky S, Wei DG, Berg R, et al. 1997. xid affects events leading to B cell cycle entry. J. Immunol. 159:135–43 Glassford J, Holman M, Banerji L, Clayton E, Klaus GG, et al. 2001. Vav is required for cyclin D2 induction and proliferation of mouse B lymphocytes activated via the antigen receptor. J. Biol. Chem. 276:41040–48 Glassford J, Soeiro I, Skarell SM, Banerji L, Holman M, et al. 2003. BCR targets cyclin D2 via Btk and the p85α subunit of PI3-K to induce cell cycle progression in primary mouse B cells. Oncogene 22:2248–59 Fruman DA, Ferl GZ, An SS, Donahue AC, Satterthwaite AB, Witte ON. 2002. Phosphoinositide 3-kinase and Bruton’s tyrosine kinase regulate overlapping sets of genes in B lymphocytes. Proc. Natl. Acad. Sci. USA 99:359–64 Yusuf I, Fruman DA. 2003. Regulation of quiescence in lymphocytes. Trends Immunol. 24:380–86 Martin P, Duran A, Minguet S, Gaspar ML, Diaz-Meco MT, et al. 2002. Role of ζ PKC in B-cell signaling and function. EMBO J. 21:4049–57 Kane LP, Weiss A. 2003. The PI-3 kinase/Akt pathway and T cell activation: pleiotropic pathways downstream of PIP3. Immunol. Rev. 192:7–20 Manning BD, Tee AR, Logsdon MN, Blenis J, Cantley LC. 2002. Identification of the tuberous sclerosis complex-2 tumor suppressor gene product tuberin as a target of the phosphoinositide 3kinase/akt pathway. Mol. Cell 10:151–62 Pullen N, Dennis PB, Andjelkovic M,

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

PI-3 KINASE IN THE IMMUNE SYSTEM

75.

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

76.

77.

78.

79.

80.

81.

82.

83.

Dufner A, Kozma SC, et al. 1998. Phosphorylation and activation of p70s6k by PDK1. Science 279:707–10 Grumont RJ, Strasser A, Gerondakis S. 2002. B cell growth is controlled by phosphatidylinosotol 3-kinase-dependent induction of Rel/NF-κB regulated c-myc transcription. Mol. Cell 10: 1283–94 Donahue AC, Fruman DA. 2003. Proliferation and survival of activated B cells requires sustained antigen receptor engagement and phosphoinositide 3kinase activation. J. Immunol. 170:5851– 60 Jacob A, Cooney D, Pradhan M, Coggeshall KM. 2002. Convergence of signaling pathways on the activation of ERK in B cells. J. Biol. Chem. 277:23420–26 Suzuki A, Kaisho T, Ohishi M, TsukioYamaguchi M, Tsubata T, et al. 2003. Critical roles of Pten in B cell homeostasis and immunoglobulin class switch recombination. J. Exp. Med. 197:657– 67 Liu Q, Oliveira-Dos-Santos AJ, Mariathasan S, Bouchard D, Jones J, et al. 1998. The inositol polyphosphate 5phosphatase ship is a crucial negative regulator of B cell antigen receptor signaling. J. Exp. Med. 188:1333–42 Brauweiler A, Tamir I, Dal Porto J, Benschop RJ, Helgason CD, et al. 2000. Differential regulation of B cell development, activation, and death by the src homology 2 domain-containing 50 inositol phosphatase (SHIP). J. Exp. Med. 191:1545–54 Costello PS, Gallagher M, Cantrell DA. 2002. Sustained and dynamic inositol lipid metabolism inside and outside the immunological synapse. Nat. Immunol. 3:1082–89 Harriague J, Bismuth G. 2002. Imaging antigen-induced PI3K activation in T cells. Nat. Immunol. 3:1090–96 Huppa JB, Gleimer M, Sumen C, Davis MM. 2003. Continuous T cell receptor

84.

85.

86.

87.

88.

89.

90.

91.

P1: IKH

593

signaling required for synapse maintenance and full effector potential. Nat. Immunol. 4:749–55 Astoul E, Edmunds C, Cantrell DA, Ward SG. 2001. PI 3-K and T-cell activation: limitations of T-leukemic cell lines as signaling models. Trends Immunol. 22:490–96 Reynolds LF, Smyth LA, Norton T, Freshney N, Downward J, et al. 2002. Vav1 transduces T cell receptor signals to the activation of phospholipase C-γ 1 via phosphoinositide 3-kinasedependent and -independent pathways. J. Exp. Med. 195:1103–14 Parry RV, Rumbley CA, Vandenberghe LH, June CH, Riley JL. 2003. CD28 and inducible costimulatory protein Src homology 2 binding domains show distinct regulation of phosphatidylinositol 3-kinase, Bcl-x(L), and IL-2 expression in primary human CD4 T lymphocytes. J. Immunol. 171:166–74 Rodriguez-Borlado L, Barber DF, Hernandez C, Rodriguez-Marcos MA, Sanchez A, et al. 2003. Phosphatidylinositol 3-kinase regulates the CD4/CD8 T cell differentiation ratio. J. Immunol. 170:4475–82 Melendez AJ, Gillooly DJ, Harnett MM, Allen JM. 1998. Aggregation of the human high affinity immunoglobulin G receptor (Fcγ RI) activates both tyrosine kinase and G protein-coupled phosphoinositide 3-kinase isoforms. Proc. Natl. Acad. Sci. USA 95:2169–74 Hobson JP, Rosenfeldt HM, Barak LS, Olivera A, Poulton S, et al. 2001. Role of the sphingosine-1-phosphate receptor EDG-1 in PDGF-induced cell motility. Science 291:1800–3 Laffargue M, Calvez R, Finan P, Trifilieff A, Barbier M, et al. 2002. Phosphoinositide 3-kinase γ is an essential amplifier of mast cell function. Immunity 16:441– 51 Truitt KE, Shi J, Gibson S, Segal LG, Mills GB, Imboden JB. 1995. CD28

11 Feb 2004

22:12

594

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

92.

93.

94.

95.

96.

97.

98.

99.

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

delivers costimulatory signals independently of its association with phosphatidylinositol 3-kinase. J. Immunol. 155:4702–10 Shi J, Cinek T, Truitt KE, Imboden JB. 1997. Wortmannin, a phosphatidylinositol 3-kinase inhibitor, blocks antigenmediated, but not CD3 monoclonal antibody-induced, activation of murine CD4+ T cells. J. Immunol. 158:4688–95 Eder AM, Dominguez L, Franke TF, Ashwell JD. 1998. Phosphoinositide 3-kinase regulation of T cell receptormediated interleukin-2 gene expression in normal T cells. J. Biol. Chem. 273: 28025–31 Delon J, Bercovici N, Liblau R, Trautmann A. 1998. Imaging antigen recognition by naive CD4+ T cells: compulsory cytoskeletal alterations for the triggering of an intracellular calcium response. Eur. J. Immunol. 28:716–29 Hsu AL, Ching TT, Sen G, Wang DS, Bondada S, et al. 2000. Novel function of phosphoinositide 3-kinase in T cell Ca2+ signaling. A phosphatidylinositol 3,4,5-trisphosphate-mediated Ca2+ entry mechanism. J. Biol. Chem. 275:16242– 50 Schaeffer EM, Debnath J, Yap G, McVicar D, Liao XC, et al. 1999. Requirement for Tec kinases Rlk and Itk in T cell receptor signaling and immunity. Science 284:638–41 Bunnell SC, Diehn M, Yaffe MB, Findell PR, Cantley LC, Berg LJ. 2000. Biochemical interactions integrating Itk with the T cell receptor-initiated signaling cascade. J. Biol. Chem. 275:2219–30 Liao XC, Littman DR. 1995. Altered T cell receptor signaling and disrupted T cell development in mice lacking Itk. Immunity 3:757–69 Liu KQ, Bunnell SC, Gurniak CB, Berg LJ. 1998. T cell receptor-initiated calcium release is uncoupled from capacitative calcium entry in Itk-deficient T cells. J. Exp. Med. 187:1721–27

100. Fowell DJ, Shinkai K, Liao XC, Beebe AM, Coffman RL, et al. 1999. Impaired NFATc translocation and failure of Th2 development in Itk-deficient CD4+ T cells. Immunity 11:399–409 101. Okkenhaug K, Wu L, Garza KM, La Rose J, Khoo W, et al. 2001. A point mutation in CD28 distinguishes proliferative signals from survival signals. Nat. Immunol. 2:325–32 102. Burr JS, Savage ND, Messah GE, Kimzey SL, Shaw AS, et al. 2001. Cutting edge: Distinct motifs within CD28 regulate T cell proliferation and induction of Bcl-XL. J. Immunol. 166:5331– 35 103. Harada Y, Tokushima M, Matsumoto Y, Ogawa S, Otsuka M, et al. 2001. Critical requirement for the membraneproximal cytosolic tyrosine residue for CD28-mediated costimulation in vivo. J. Immunol. 166:3797–803 104. Kane LP, Mollenauer MN, Xu Z, Turck CW, Weiss A. 2002. Akt-dependent phosphorylation specifically regulates Cot induction of NFκB-dependent transcription. Mol. Cell. Biol. 22:5962–74 105. Kane LP, Andres PG, Howland KC, Abbas AK, Weiss A. 2001. Akt provides the CD28 costimulatory signal for upregulation of IL-2 and IFN-γ but not TH2 cytokines. Nat. Immunol. 2:37– 44 106. Jones RG, Elford AR, Parsons MJ, Wu L, Krawczyk CM, et al. 2002. CD28-dependent activation of protein kinase B/Akt blocks Fas-mediated apoptosis by preventing death-inducing signaling complex assembly. J. Exp. Med. 196:335–48 107. Frauwirth KA, Riley JL, Harris MH, Parry RV, Rathmell JC, et al. 2002. The CD28 signaling pathway regulates glucose metabolism. Immunity 16:769–77 108. Ohteki T, Parsons M, Zakarian A, Jones RG, Nguyen LT, et al. 2000. Negative regulation of T cell proliferation and interleukin 2 production by the serine

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

PI-3 KINASE IN THE IMMUNE SYSTEM

109.

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

110.

111.

112.

113.

114.

115.

116.

117.

118.

threonine kinase GSK-3. J. Exp. Med. 192:99–104 Stahl M, Dijkers PF, Kops GJ, Lens SM, Coffer PJ, et al. 2002. The forkhead transcription factor FoxO regulates transcription of p27Kip1and Bim in response to IL-2. J. Immunol. 168:5024–31 Liu YC, Gu H. 2002. Cbl and Cbl-b in T-cell regulation. Trends Immunol. 23: 140–43 Fang D, Wang HY, Fang N, Altman Y, Elly C, Liu YC. 2001. Cbl-b, a RINGtype E3 ubiquitin ligase, targets phosphatidylinositol 3- kinase for ubiquitination in T cells. J. Biol. Chem. 276:4872– 78 Thien CB, Bowtell DD, Langdon WY. 1999. Perturbed regulation of ZAP-70 and sustained tyrosine phosphorylation of LAT SLP-76 in c-Cbl-deficient thymocytes. J. Immunol. 162:7133–39 Bachmaier K, Krawczyk C, Kozieradzki I, Kong YY, Sasaki T, et al. 2000. Negative regulation of lymphocyte activation and autoimmunity by the molecular adaptor Cbl-b. Nature 403:211–16 Chiang YJ, Kole HK, Brown K, Naramura M, Fukuhara S, et al. 2000. Cbl-b regulates the CD28 dependence of T-cell activation. Nature 403:216–20 Fang D, Liu YC. 2001. Proteolysisindependent regulation of PI3K by Cblb-mediated ubiquitination in T cells. Nat. Immunol. 2:870–75 Yasuda T, Tezuka T, Maeda A, Inazu T, Yamanashi Y, et al. 2002. Cbl-b positively regulates Btk-mediated activation of phospholipase C-γ 2 in B cells. J. Exp. Med. 196:51–63 Arron JR, Vologodskaia M, Wong BR, Naramura M, Kim N, et al. 2001. A positive regulatory role for Cbl family proteins in tumor necrosis factor-related activation-induced cytokine (trance) and CD40L-mediated Akt activation. J. Biol. Chem. 276:30011–17 Rivera J. 2002. Molecular adapters in FcεRI signaling and the allergic re-

119.

120.

121.

122.

123.

124.

125.

126.

127.

P1: IKH

595

sponse. Curr. Opin. Immunol. 14:688– 93 Wilson BS, Pfeiffer JR, Surviladze Z, Gaudet EA, Oliver JM. 2001. High resolution mapping of mast cell membranes reveals primary and secondary domains of FcεRI LAT. J. Cell Biol. 154:645– 58 Gu H, Saito K, Klaman LD, Shen J, Fleming T, et al. 2001. Essential role for Gab2 in the allergic response. Nature 412:186–90 Manetz TS, Gonzalez-Espinosa C, Arudchandran R, Xirasagar S, Tybulewicz V, Rivera J. 2001. Vav1 regulates phospholipase cγ activation and calcium responses in mast cells. Mol. Cell. Biol. 21:3763–74 Yang FC, Kapur R, King AJ, Tao W, Kim C, et al. 2000. Rac2 stimulates Akt activation affecting BAD/Bcl-XL expression while mediating survival and actin function in primary mast cells. Immunity 12:557–68 Ott VL, Cambier JC. 2000. Activating and inhibitory signaling in mast cells: new opportunities for therapeutic intervention? J. Allergy Clin. Immunol. 106:429–40 Parravicini V, Gadina M, Kovarova M, Odom S, Gonzalez-Espinosa C, et al. 2002. Fyn kinase initiates complementary signals required for IgE-dependent mast cell degranulation. Nat. Immunol. 3:741–48 Lu-Kuo JM, Fruman DA, Joyal DM, Cantley LC, Katz HR. 2000. Impaired kit- but not FcεRI-initiated mast cell activation in the absence of phosphoinositide 3-kinase p85α gene products. J. Biol. Chem. 275:6022–29 Fukao T, Yamada T, Tanabe M, Terauchi Y, Ota T, et al. 2002. Selective loss of gastrointestinal mast cells and impaired immunity in PI3K-deficient mice. Nat. Immunol. 3:295–304 Smith AJ, Surviladze Z, Gaudet EA, Backer JM, Mitchell CA, Wilson BS.

11 Feb 2004

22:12

596

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

128.

129.

130.

131.

132.

133.

134.

135.

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

2001. p110β and p110δ phosphatidylinositol 3-kinases up-regulate FcεRIactivated Ca2+ influx by enhancing inositol 1,4,5-trisphosphate production. J. Biol. Chem. 276:17213–20 Windmiller DA, Backer JM. 2003. Distinct phosphoinositide 3-kinases mediate mast cell degranulation in response to Gprotein-coupled versus Fcε RI receptors. J. Biol. Chem. 278:11874–78 Hata D, Kawakami Y, Inagaki N, Lantz CS, Kitamura T, et al. 1998. Involvement of Bruton’s tyrosine kinase in FcεRI-dependent mast cell degranulation and cytokine production. J. Exp. Med. 187:1235–47 Rameh LE, Rhee SG, Spokes K, Kazlauskas A, Cantley LC, Cantley LG. 1998. Phosphoinositide 3-kinase regulates phospholipase Cγ -mediated calcium signaling. J. Biol. Chem. 273: 23750–57 Araki N, Johnson MT, Swanson JA. 1996. A role for phosphoinositide 3kinase in the completion of macropinocytosis and phagocytosis by macrophages. J. Cell Biol. 135:1249–60 Swanson JA, Johnson MT, Beningo K, Post P, Mooseker M, Araki N. 1999. A contractile activity that closes phagosomes in macrophages. J. Cell. Sci. 112(Part 3):307–16 Cox D, Tseng CC, Bjekic G, Greenberg S. 1999. A requirement for phosphatidylinositol 3-kinase in pseudopod extension. J. Biol. Chem. 274:1240–47 Vieira OV, Botelho RJ, Rameh L, Brachmann SM, Matsuo T, et al. 2001. Distinct roles of class I and class III phosphatidylinositol 3-kinases in phagosome formation and maturation. J. Cell Biol. 155:19–25 Marshall JG, Booth JW, Stambolic V, Mak T, Balla T, et al. 2001. Restricted accumulation of phosphatidylinositol 3kinase products in a plasmalemmal subdomain during Fcγ receptor-mediated phagocytosis. J. Cell Biol. 153:1369–80

136. Deleted in proof 137. Gu H, Botelho RJ, Yu M, Grinstein S, Neel BG. 2003. Critical role for scaffolding adapter Gab2 in Fcγ R-mediated phagocytosis. J. Cell Biol. 161:1151– 61 138. Gibbins JM, Briddon S, Shutes A, van Vugt MJ, van de Winkel JG, et al. 1998. The p85 subunit of phosphatidylinositol 3-kinase associates with the Fc receptor γ -chain and linker for activitor of T cells (LAT) in platelets stimulated by collagen and convulxin. J. Biol. Chem. 273:34437–43 139. Tridandapani S, Lyden TW, Smith JL, Carter JE, Coggeshall KM, Anderson CL. 2000. The adapter protein LAT enhances fcγ receptor-mediated signal transduction in myeloid cells. J. Biol. Chem. 275:20480–87 140. Patel JC, Hall A, Caron E. 2002. Vav regulates activation of Rac but not Cdc42 during Fcγ R-mediated phagocytosis. Mol. Biol. Cell 13:1215–26 141. Cox D, Dale BM, Kashiwada M, Helgason CD, Greenberg S. 2001. A regulatory role for Src homology 2 domaincontaining inositol 50 -phosphatase (SHIP) in phagocytosis mediated by Fcγ receptors and complement receptor 3 (α(M)B(2); CD11b/CD18). J. Exp. Med. 193:61–71 142. Fratti RA, Backer JM, Gruenberg J, Corvera S, Deretic V. 2001. Role of phosphatidylinositol 3-kinase and Rab5 effectors in phagosomal biogenesis and mycobacterial phagosome maturation arrest. J. Cell Biol. 154:631–44 143. Leverrier Y, Okkenhaug K, Sawyer C, Bilancio A, Vanhaesebroeck B, Ridley AJ. 2003. Class I PI3K p110β is required for apoptotic cell and Fcγ R-mediated phagocytosis by macrophages. J. Biol. Chem. 278:38437–42 144. Cox D, Greenberg S. 2001. Phagocytic signaling strategies: Fc(γ ) receptormediated phagocytosis as a model system. Semin. Immunol. 13:339–45

11 Feb 2004

22:12

AR

AR210-IY22-19.tex

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PI-3 KINASE IN THE IMMUNE SYSTEM 145. Cox D, Berg JS, Cammer M, Chinegwundoh JO, Dale BM, et al. 2002. Myosin X is a downstream effector of PI(3)K during phagocytosis. Nat. Cell Biol. 4:469–77 146. Yokoyama WM, Plougastel BF. 2003. Immune functions encoded by the natural killer gene complex. Nat. Rev. Immunol. 3:304–16 147. Moretta A, Bottino C, Vitale M, Pende D, Cantoni C, et al. 2001. Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis. Annu. Rev. Immunol. 19:197– 223 148. Wang JW, Howson JM, Ghansah T, Desponts C, Ninos JM, et al. 2002. Influence of SHIP on the NK repertoire and allogeneic bone marrow transplantation. Science 295:2094–97 149. Long EO. 1999. Regulation of immune responses through inhibitory receptors. Annu. Rev. Immunol. 17:875–904 150. Trinchieri G. 2003. The choices of a natural killer. Nat. Immunol. 4:509–10 151. Wu J, Cherwinski H, Spies T, Phillips JH, Lanier LL. 2000. DAP10 and DAP12 form distinct, but functionally cooperative, receptor complexes in natural killer cells. J. Exp. Med. 192:1059–68 152. Wu J, Song Y, Bakker AB, Bauer S, Spies T, et al. 1999. An activating immunoreceptor complex formed by NKG2D DAP10. Science 285:730–32 153. Billadeau DD, Upshaw JL, Schoon RA, Dick CJ, Leibson PJ. 2003. NKG2DDAP10 triggers human NK cellmediated killing via a Syk-independent regulatory pathway. Nat. Immunol. 4: 557–64 154. Zompi S, Hamerman JA, Ogasawara K, Schweighoffer E, Tybulewicz VL, et al. 2003. NKG2D triggers cytotoxicity in mouse NK cells lacking DAP12 or Syk family kinases. Nat. Immunol. 4:565–72 155. Comer FI, Parent CA. 2002. PI 3kinases PTEN: How opposites chemoattract. Cell 109:541–44

P1: IKH

597

156. Iijima M, Huang YE, Devreotes P. 2002. Temporal and spatial regulation of chemotaxis. Dev. Cell 3:469–78 157. Servant G, Weiner OD, Herzmark P, Balla T, Sedat JW, Bourne HR. 2000. Polarization of chemoattractant receptor signaling during neutrophil chemotaxis. Science 287:1037–40 158. Weiner OD, Neilsen PO, Prestwich GD, Kirschner MW, Cantley LC, Bourne HR. 2002. A PtdInsP(3)- and Rho GTPasemediated positive feedback loop regulates neutrophil polarity. Nat. Cell Biol. 4:509–13 159. Wang F, Herzmark P, Weiner OD, Srinivasan S, Servant G, Bourne HR. 2002. Lipid products of PI(3)Ks maintain persistent cell polarity and directed motility in neutrophils. Nat. Cell Biol. 4:513– 18 160. Welch HC, Coadwell WJ, Ellson CD, Ferguson GJ, Andrews SR, et al. 2002. P-Rex1, a PtdIns(3,4,5)P3- and GBγ regulated guanine-nucleotide exchange factor for Rac. Cell 108:809–21 161. Weiner OD. 2002. Rac activation: PRex1—a convergence point for PIP(3) and GBγ ? Curr. Biol. 12:R429–31 162. Hannigan M, Zhan L, Li Z, Ai Y, Wu D, Huang CK. 2002. Neutrophils lacking phosphoinositide 3kinase γ show loss of directionality during N-formyl-Met-Leu-Phe-induced chemotaxis. Proc. Natl. Acad. Sci. USA 99:3603–8 162a. Vicente-Manzanares M, Rey M, Jones DR, Sancho D, Mellado M, et al. 1999. Involvement of phosphatidylinositol 3kinase in stromal cell derived factor1α-induced lymphocyte polarization and chemotaxis. J. Immunol. 163:4001–12 163. Sadhu C, Masinovsky B, Dick K, Sowell CG, Staunton DE. 2003. Essential role of phosphoinositide 3-kinase δ in neutrophil directional movement. J. Immunol. 170:2647–54 164. Vanhaesebroeck B, Jones GE, Allen WE, Zicha D, Hooshmand-Rad R, et al.

11 Feb 2004

22:12

598

165.

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

166.

167.

168.

169.

AR

DEANE

AR210-IY22-19.tex

¥

AR210-IY22-19.sgm

LaTeX2e(2002/01/18)

P1: IKH

FRUMAN

1999. Distinct PI(3)Ks mediate mitogenic signalling and cell migration in macrophages. Nat. Cell Biol. 1:69–71 Dinauer MC. 2003. Regulation of neutrophil function by Rac GTPases. Curr. Opin. Hematol. 10:8–15 Cadwallader KA, Condliffe AM, McGregor A, Walker TR, White JF, et al. 2002. Regulation of phosphatidylinositol 3-kinase activity and phosphatidylinositol 3,4,5-trisphosphate accumulation by neutrophil priming agents. J. Immunol. 169:3336–44 Croker BA, Handman E, Hayball JD, Baldwin TM, Voigt V, et al. 2002. Rac2deficient mice display perturbed T-cell distribution and chemotaxis, but only minor abnormalities in T(H)1 responses. Immunol. Cell Biol. 80:231–40 Fox JA, Ung K, Tanlimco SG, Jirik FR. 2002. Disruption of a single Pten allele augments the chemotactic response of B lymphocytes to stromal cell-derived factor-1. J. Immunol. 169:49–54 Cherukuri A, Dykstra M, Pierce SK. 2001. Floating the raft hypothesis: Lipid

170.

171.

172.

173.

174.

rafts play a role in immune cell activation. Immunity 14:657–60 Sedwick CE, Altman A. 2002. Ordered just so: lipid rafts and lymphocyte function. Sci. STKE 2002: www.stke.org/cgi/ content/full/OC sigtrans;2002/122/re2 Aman MJ, Tosello-Trampont AC, Ravichandran K. 2001. Fc γ RIIB1/SHIPmediated inhibitory signaling in B cells involves lipid rafts. J. Biol. Chem. 276:46371–78 Wymann MP, Zvelebil M, Laffargue M. 2003. Phosphoinositide 3-kinase signaling—which way to target? Trends Pharmacol. Sci. 24:366–76 Walmsley MJ, Ooi SK, Reynolds LF, Smith SH, Ruf S, et al. 2003. Critical roles for Rac1 and Rac2 GTPases in B cell development and signaling. Science 302:459–62 Han A, Saijo K, Mecklenbrauker I, Tarakhovsky A, Nussenzweig MC. 2003. Bam32 links the B cell receptor to ERK and JNK and mediates B cell proliferation but not survival. Immunity 19:621– 32

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:563-598. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104635

Annu. Rev. Immunol. 2004. 22:599–623 doi: 10.1146/annurev.immunol.22.012703.104635 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on December 12, 2003

IMMUNITY TO TUBERCULOSIS Robert J. North and Yu-Jin Jung Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Trudeau Institute, Saranac Lake, New York 12983; email: [email protected]

Key Words Th1 immunity, CD4 T cells, interferon-γ , nitric oxide synthase, lung pathology ■ Abstract Only 5 to 10% of immunocompetent humans are susceptible to tuberculosis, and over 85% of them develop the disease exclusively in the lungs. Human immunodeficiency virus (HIV)-infected humans, in contrast, can develop systemic disease that is more quickly lethal. This is in keeping with other evidence showing that susceptible humans generate some level of Th1 immunity to Mycobacterium tuberculosis (Mtb) infection. Tuberculosis in mice is also exclusively a lung disease that is progressive and lethal, in spite of the generation of Th1-mediated immunity. Thus mouse tuberculosis is a model of tuberculosis in susceptible humans, as is tuberculosis in guinea pigs and rabbits. Inability to resolve infection and prevent disease may not be a consequence of the generation of an inadequate number of Th1 cells but of an intrinsic deficiency in macrophage function that prevents these cells from expressing immunity. If this proves to be true, vaccinating susceptible humans against tuberculosis will be a difficult task.

INTRODUCTION Much of what is known about the immunology of tuberculosis has come from studies of immunity to tuberculosis in mice. The emphasis of this article, therefore, is on a discussion of the mouse model of tuberculosis and its appropriateness, compared with other models, for studying immunity to the disease in humans. Animal models are essential for investigating the immune response in vivo where the ability to control infection can be assessed in a physiological setting after the selective removal of one or more components of the host response suspected of being involved. We argue that the immunology of mouse tuberculosis is similar in important ways to that in humans and can be used to investigate key questions about the human disease. We also argue that mouse tuberculosis and tuberculosis in the guinea pig and rabbit are models of tuberculosis in susceptible humans.

0732-0582/04/0423-0599$14.00

599

11 Feb 2004

22:13

600

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TUBERCULOSIS Tuberculosis is a major world disease estimated to kill over 2 million people annually. The majority of cases occur in developing countries, with sub-Sahara Africa having the highest incidence rate per capita, and the Southeast Asian region having the largest number of cases. It is estimated (1) that there were 8.3 million new cases worldwide in 2000 and that tuberculosis was the cause of 11% of all adult acquired immune deficiency (AIDS) deaths in 2000. In South Africa alone there were 2 million people coinfected with Mycobacterium tuberculosis (Mtb) and HIV. Attempts over the years to vaccinate adults against pulmonary tuberculosis with the attenuated BCG strain of Mycobacterium bovis, a closely related species that shares many antigens with Mtb, have proved disappointing. Moreover, an immunologic explanation for the lack of efficacy of BCG has not been forthcoming. Therefore, the scientific rationale behind current attempts to design more efficacious antituberculosis vaccines is not obvious. It will require a thorough understanding of why the anti-Mtb immune response is successful at defending most immunocompetent humans against Mtb infection, but unsuccessful at defending others. Or, put another way, it will require an understanding of why Mtb is virulent for only a small percentage of immunocompetent humans.

The Pathogen Mtb is a slow-growing, facultative intracellular pathogen that can survive and multiply inside macrophages and other mammalian cells. It is gram-positive, nonsporeforming, and aerobic. It shares with other members of the Mycobacterium genus a cell wall of unique composition due to the dominant presence of mycolic acids that make up more than 50% of its dry weight. It is the cell wall that gives Mtb its acid fastness, enabling it to retain basic dyes in the presence of acid alcohol. A large number of different strains of Mtb exist that can be distinguished from one another on the basis of restriction fragment length polymorphism (RFLP). The RFLP method most commonly employed is one based on the polymorphism and copy number of the IS6110 chromosomal insertion sequence (2) that inserts in different numbers and at different sites along the chromosome. It is its insertion at different sites that is responsible for the polymorphism of the IS6110-containing fragments generated by restriction enzymes. The usefulness of the IS6110 profile is based on its relative stability. Southern blot hybridization analysis using a standard restriction enzyme and a standard DNA probe can allow the IS6110 profile to be used to determine whether tuberculosis outbreaks in the same or in different geographical areas are caused by the same or by multiple Mtb strains. It also can be used to determine whether recurrent disease is caused by endogenous or exogenous reinfection. Other molecular methods are available to aid in strain identification (2). The genome of Mtb has been sequenced and shown to be 4.41 Mb in size and to contain about 4000 protein-coding genes of which 52% can be assigned

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

ANTITUBERCULOSIS IMMUNITY

P1: IKH

601

a function. Only 376 putative proteins share no homology with known proteins and presumably are unique to Mtb (3). The availability of this type of information is important for identifying genes that code for virulence factors and for antigens against which host immunity is directed. The genome sequence is also important for identifying targets for chemotherapy. Except for a few possible exceptions, there is little reason to believe at this time that recent clinical isolates of Mtb are more virulent for humans than standard laboratory strains isolated over 100 years ago.

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Pathology Mtb has evolved to avoid destruction by innate and adaptive mechanisms of immunity in certain immunocompetent humans and to induce lung pathology of a type that will ensure its transmission by infectious aerosol for relatively long periods of time. Therefore, in its natural host, Mtb has evolved to induce chronic immunopathology of a type that has been described in numerous books and articles over the years. It begins with the accumulation of macrophages at sites of bacterial implantation and multiplication to form compact granulomas that contain the pathogen at these sites (4). The granuloma can later show central caseous necrosis and give rise to cavities, although this does not occur in all cases of the disease. A key aspect of granuloma formation is the development of fibrosis within the granuloma and in surrounding parenchyma, which produces macroscopic nodules (tubercles). In adults, the disease advances as a necrotizing pneumonic process that can involve bronchioles and result in the spread of infection to other areas of the lungs. Undoubtedly, these events depend on the presence of metabolically active Mtb bacilli.

Resistant and Susceptible Humans The natural history of tuberculosis shows that most humans are resistant, presumably because of an ability to generate a successful immune response against Mtb infection. Of those who are exposed to Mtb, only 10–30% actually become infected, indicating that some humans may be capable of preventing infection from becoming established after Mtb implants in the lungs. Of those who become infected (acquire delayed sensitivity to Mtb proteins), 90% or more do not develop tuberculosis, meaning that they are capable of resolving infection completely, or to a level incapable of causing disease. In the latter case, infection is said to be latent and capable of reactivating and of causing outbreaks of disease at a future time. Humans with latent tuberculosis represent a large reservoir of Mtb. Of the 10% of humans who are susceptible, half develop active disease within 1 year and the other half do so thereafter owing to reactivation. Thus, predisposed humans show a spectrum of susceptibilities. There is little doubt, in this regard, that resistance/susceptibility to tuberculosis is genetically determined and that many genes are involved. Linkage-based, genome-wide screens of populations to determine the chromosomal locations of genes involved in susceptibility to tuberculosis (5), as well as case-control association studies of candidate genes (5–7), have been

11 Feb 2004

22:13

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

602

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

performed. Polymorphisms in genes for natural resistance-associated macrophage protein (NRAMP1), vitamin D receptor, and mannose-binding protein appear to be associated with tuberculosis susceptibility. A key aspect of tuberculosis in susceptible humans that has immunologic implications is it exclusively involves the lungs in 85% of cases. In contrast, in immunocompromised humans, such as those infected with HIV, tuberculosis can be a systemic disease and involve multiple organs (8). This suggests that the disease is confined to the lungs in immunocompetent susceptible humans because they possess a systemic mechanism of defense capable of controlling the disease in all other organs.

Susceptible Humans Generate Th1 Immunity Most humans who are susceptible to tuberculosis are not immunodeficient in that they are not susceptible to infectious agents in general. On the contrary, convincing evidence indicates that susceptible humans with active disease acquire Mtb-specific immunity. For example, it has been accepted for many years (9, 10) that humans with active tuberculosis generate and maintain high levels of cellular immunity to Mtb antigens in the form of delayed-type hypersensitivity (DTH), which enables them to mount a delayed-type inflammatory reaction at a site of intracutaneous injection of Mtb proteins. It is generally accepted that DTH represents the expression of Th1 immunity. Cause-and-effect evidence that anti-Mtb immunity is Th1 mediated has mostly come from in vivo experiments with mice (see below). However, there is no doubt that humans with tuberculosis can generate a Th1 response to Mtb, as evidenced by the presence in their blood and lungs of CD4 and CD8 T cells capable of responding specifically to Mtb antigens by replicating and by synthesizing IFN-γ and other Th1 cytokines in vitro (11–13). It is apparent that the favorite source of Mtb-specific T cells for in vitro analysis is the blood of persons with active tuberculosis. An additional reason for believing that people with active disease develop a level of protective immunity against Mtb infection is that they form macrophage-dominated granulomas at sites of lung infection, a process believed to represent the expression of DTH. Hence the ongoing debate about whether DTH is helpful or harmful to the host. The reason for believing it is helpful is that the histopathological response to Mtb infection in the lungs of AIDS patients is different (14), being diffuse and necrotic and incapable of restricting Mtb growth or of containing Mtb at original sites of infection. Consequently, tuberculosis in HIV-infected persons can involve multiple organs. Again, tuberculosis is a severe systemic disease in humans with recessive mutations in genes for the IFN-γ -receptor ligand-binding chain, the IL12p40 subunit, and the IL-12 receptor β1 chain (15). All of this points to the importance of Th1 immunity, not only in protecting resistant humans, but also in enabling most immunocompetent susceptible humans with active disease to exert a growth-restricting influence on Mtb and consequently on the pathology it induces.

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

ANTITUBERCULOSIS IMMUNITY

P1: IKH

603

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

EXPERIMENTAL TUBERCULOSIS In vivo analysis of the host response to Mtb infection requires the use of animal models of the disease. Most studies of immunity to tuberculosis have been performed in mice. This host species has also been used to investigate other aspects of the host-pathogen relationship, such as the molecular basis of Mtb virulence, a property that can be investigated only in terms of changes induced in the host. The reasons for using mice are numerous: (a) Mice are much less expensive to purchase and to maintain than other host species, (b) they are available as numerous inbred strains that differ from one another in terms of their ability to resist infection and survive, and (c) they are available as targeted mutant strains that have been functionally deleted of one or more genes involved in the generation and expression of immunity. Many more analytical reagents are available with which to investigate immunity in mice compared with those for other host species. However, these reasons for using mice would be of limited value if mouse tuberculosis were not an appropriate model with which to investigate problems relevant to the human disease. Key questions about human tuberculosis for which answers are needed relate to (a) why only 5–10% of humans are susceptible, (b) why more than 85% of susceptible humans develop disease exclusively in the lungs, and (c) why it is difficult to vaccinate susceptible people against the disease. It will be argued that mouse tuberculosis can be used to investigate these key questions. However, to make full use of the mouse model it is necessary to be aware of its characteristics.

Resistant Versus Susceptible Mice A comparative study (16) of the survival times of mice of selected inbred strains inoculated intravenously (iv) with 105 Mtb showed that the strains were either resistant (C57BL/6 and BALB/c), with approximate median survival times (MSTs) of about 250 days, or susceptible (DBA/2, C3H, CBA, 129Sv), with MSTs of about 100 days. A similar result was obtained with mice infected with 102 Mtb via the airborne route (16). Resistance/susceptibility is not determined by Nramp1 because Mtb-resistant strains carry the susceptibility allele of this gene, whereas susceptible strains carry the resistance allele. Evidence against a role for Nramp1 in resistance to infection with virulent Mtb has been discussed elsewhere (17). Resistance/susceptibility is also not determined by MHC haplotype (16). However, it should be noted when discussing susceptibility and resistance that resistance is a relative term and that mice of all strains eventually succumbed to infection-induced disease, the difference being that susceptible strains succumb much earlier. It is also worth noting that demonstrating a difference in survival between resistant and susceptible strains is dependent on using a small enough inoculum of Mtb to initiate infection. Whereas an iv dose of 105 Mtb revealed a considerable difference in survival times between a resistant DBA/2 and susceptible C57BL/6 mice, a dose of 107 reveals no difference at all (16). In the latter case, survival times were greatly shortened, but were shortened to a much greater extent in mice of the resistant

11 Feb 2004

22:13

604

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

strain. It is apparent that overdosed resistant mice die of a lung disease that is different from that induced by a smaller dose of Mtb. They do not survive long enough for chronic lung pathology to develop.

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Pattern of Infection in Resistant Mice It should not be surprising that mice of resistant strains have a superior ability to control lung infection with Mtb. In resistant C57BL/6 and BALB/c strains (17– 19), initiation of infection via the respiratory route with a small number (102 or fewer) of Mtb is followed by about a 20-day period of log linear bacterial growth in the lungs that results in about a 106.5 level of infection. The average doubling time of the pathogen during this progressive phase of infection is about 28 h. The progressive phase ends with the inhibition of further Mtb growth, which is followed by an approximately stationary level of infection that persists from day 20 until the mice succumb in about 250 days. This pattern of lung infection seems to have been first described for mice by Smith and colleagues (20) more than 30 years ago. Needless to say, the onset of stationary level infection is taken as evidence of the onset of expression of adaptive immunity. Infection in mice initiated via the respiratory route does not stay confined to the lungs but disseminates via lymph and blood (21) to infect other organs. Infection in the liver and spleen, for example, is evident by about day 15 (21), and progresses until about day 20 when further Mtb growth ceases and infection is held at a stationary level, almost certainly because of the expression of systemic immunity (22). Because of the short period of Mtb growth in the liver and spleen, the level of stationary infection reached in these organs is much lower than that reached in the lungs. Most of the foregoing discussion also applies to infection initiated via the iv route with larger numbers of bacilli (18, 23). The bulk of an iv inoculum implants in the liver (95%) and spleen (about 4%); however, it is the 0.1% that implants in the lungs that is responsible for the disease that follows. To ensure that the lung is infected with 102 cfu of Mtb it is necessary to give about 105 cfu via the iv route. It is in the lungs that the pathogen grows progressively for the longest period of time (20 days) before further Mtb growth ceases and stationary infection ensues. This compares with only about 10 days of progressive growth in the liver and spleen before infection is controlled and held at a stationary level (23). The knowledge that Mtb infection is eventually controlled and held at a stationary level in the organs of mice infected via the iv route was revealed by McCune & Tompsett nearly 50 years ago (24).

Pattern of Infection in Susceptible Mice In mice of susceptible strains, Mtb infection is different. When initiated via the respiratory route of susceptible DBA/2 mice (18, 25), lung infection progresses for 20 days at the same rate as in the lungs of resistant strains. However, after day 20, Mtb growth in the lungs of susceptible mice is not inhibited, but progresses at

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

ANTITUBERCULOSIS IMMUNITY

P1: IKH

605

a slower rate until death. Growth is fast enough after day 20 of infection to result in a 2 log higher level of lung infection than in resistant strains by day 100. The foregoing description does not apply to the liver and spleen where growth of Mtb in susceptible mice is controlled and held at the same stationary level as in resistant mice (18, 23).

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Pattern of Infection in the Mouse Versus the Rabbit and Guinea Pig The literature shows that the characteristics of Mtb infection in the lungs of resistant mice are similar to those for lung infection in guinea pigs and rabbits. In all three host species, lung infection progresses for about 20 days before further Mtb growth is inhibited and infection is held at a stationary level. This can be seen in the growth curves (Figure 1) of Mtb published by Smith and colleagues for outbred mice in 1974 (20), for guinea pigs in 1973 (26), and by Lurie et al. for rabbits in 1955 (27). The mouse results have been repeated many times over the years. Thus in all three host species it is apparent that immunity is not expressed in the lungs until 3 weeks of infection and that it is incapable of causing infection to resolve. Therefore, according to the ability to control lung infection, guinea pigs are not more susceptible to tuberculosis than mice of resistant strains, as is frequently stated, and are even less susceptible than mice of susceptible strains that are incapable of stabilizing lung infection. Therefore, greater susceptibility cannot be used as a reason for proposing that the guinea pig is more valuable than the mouse for vaccine testing. On the other hand, guinea pigs show progressive disease in the spleen, a characteristic that has limited relevance to the disease in most susceptible humans. This is not to say, however, that the guinea pig model is not a useful model with which to investigate key aspects of the human disease.

Infection Induces Progressive Lung Disease As pointed out above, Mtb infection in mice of resistant strains, as well as in rabbits and guinea pigs, is not resolved in the lungs, but is stabilized at a stationary level from about day 20 of infection on. Stationary lung infection, however, induces progressive lung disease that is eventually lethal. In the mouse, lung pathology develops at discrete sites that presumably represent original sites of bacterial implantation. Lesions that develop near the surface of the lung are easily visible to the naked eye by 30 days or so of infection. The histological response at each site at day 20 of infection (28) is seen as alveolitis dominated by Mtb-infected macrophages that occupy the space of a large number of contiguous air sacs. By day 50 of infection, macrophages are larger and epithelioid-like in appearance (29), and at a still later stages, large aggregates of lymphoid cells appear in close proximity to but separated from the sites of macrophage accumulation (22, 29). The histopathology begins to look chronic sometime after day 100 with the onset of extensive and progressive thickening of the interalveolar septa, and an associated

11 Feb 2004

22:13

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

606

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

Figure 1 The pattern of lung infection is the same in the mouse (20), guinea pig (26), and rabbit (27). In all cases Mtb grew progressively for 3 weeks, after which further growth was inhibited and infection held at an approximately stationary level. Thus immunity in these three hosts is not expressed in the lungs until 3 weeks, and it is not capable of resolving infection.

swelling of the air sacs themselves results in an unfolding of the alveolar outpocketings. This process is referred to as fibrosing alveolitis in the case of certain human lung diseases. It is misleading to refer to sites of histopathology in the lung of mice as granulomas because the macrophages and other mononuclear cells that make up the alveolitis are distributed among intact contiguous air sacs that are separated from one another by septa. The lung eventually takes on a honey comb

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTITUBERCULOSIS IMMUNITY

607

Figure 2 Macroscopic examination of the lungs, liver, spleen, and kidney of B6 mice. At day 200 of an airborne infection initiated with 100 Mtb bacilli, only the lungs (top left) show signs of progressive disease (white nodules).

appearance (29, 30) resembling honey comb lung seen with pulmonary fibrotic diseases of humans. The expansion and coalescence of lesions eventually leads to consolidation of large areas of lung (Figure 2) and loss of respiratory function. It is apparent that mice die of respiratory insufficiency. In contrast to events in the lung, stationary infection in the liver and spleen does not cause progressive pathology. Instead, sites of infection are seen as small, compact, macrophage-dominated granulomas that appear stable in number and size throughout the course of infection (23, 29) Macroscopically and microscopically, the lungs are the only organs that show signs of progressive disease during the course of infection. The foregoing description of lung tuberculosis applies to mice of genetically resistant strains. The situation is quite different in mice that are genetically susceptible in that lung infection does not plateau after day 20 but remains progressive, although at a reduced rate. This eventually induces a different type of pathology. For example, the histological response at sites of Mtb infection in the lungs of susceptible DBA/2 (18, 31) and C3H mice (32) is similar to that in the lungs of resistant mice up to days 40–50 of infection, i.e., mononuclear cell-dominated. However, neutrophils eventually replace macrophages as the dominant cell type in many parts of the lesions, and this process continues until neutrophils occupy most of the lesions. Eventually the lung becomes occupied by a fulminating neutrophildominated necrotizing pathology that causes early death. Mice of susceptible strains, however, are not more susceptible in the liver and spleen. In these organs, infection is stabilized at an approximately stationary level as efficiently as in

11 Feb 2004

22:13

608

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

resistant strains. In both types of mice, infection is confined to compact granulomas that stay stable in size and number.

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Lung Pathology in the Mouse Versus Guinea Pig and Rabbit As in mice, stationary lung infection in the rabbit (33) and guinea pig (34) causes progressive pathology. However, the pathology that develops is different from that which develops in the mouse. The macrophages that accumulate at sites of infection are not distributed among intact air sacs but form true granulomas composed of compact aggregates of Mtb-infected macrophages. The granulomas appear to be anatomically distinct structures, resembling those that form early in the lungs of Mtb-infected humans. The centers of granulomas can undergo necrosis and liquification, and this can lead to the formation of cavities, particularly in the lungs of rabbits. There is no doubt, therefore, that the histopathology of tuberculosis in the rabbit and guinea pig is more human-like than in the mouse during early stages of infection. The reason for this is not known, but it is worth considering two key differences between the lungs of mice and of guinea pig, rabbits, and other larger mammals. First, although the lungs of mice are smaller, they are not structurally equivalent to a small piece of guinea pig or rabbit lung. Instead, the mouse lung has a miniaturized architecture (35), with alveoli that are a fraction of the volume of those in the lungs of rabbits and guinea pigs, which are smaller, in turn, than those in the human lung. This is in keeping with the rule (36) that the smaller the mammal, the higher its rate of its metabolism, and the larger the respiratory surface area needed per unit volume of lung to facilitate gas exchange. This is achieved by having smaller and more numerous air sacs per unit volume of lung and, consequently, denser blood supply. The second difference is that the lungs of rabbits and guinea pigs, but not of mice, contain intrapulmonary lymphoid tissue in the form of bronchus-associated lymphoid tissue (BALT). In the mouse lung, in contrast, BALT is inducible, meaning that it forms in response to infection. It seems possible that some Mtb-induced granulomas in guinea pigs, rabbits, and humans form within BALT. This should not be used as a case for arguing against the use of mice to investigate immunity to tuberculosis. Infection-induced pathology in all these host species is of a chronic type that leads to lung consolidation. The purpose of studying the immunology of the disease is to generate knowledge that will aid in the development of a vaccine capable of providing the host with the means to destroy Mtb before it can induce lung pathology. There is no reason to suspect that the mouse is inferior to the guinea pig and rabbit for investigating anti-Mtb immunity.

Genetics of Susceptibility/Resistance Because some inbred strains of mice are susceptible and others are resistant, it is possible to investigate the genetic basis of anti-Mtb immunity by performing whole

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTITUBERCULOSIS IMMUNITY

P1: IKH

609

genome scans of informative F2 generation progeny of susceptible and resistant strains to determine whether susceptibility/resistance maps to particular chromosomal locations. The chromosomal regions containing genes controlling a complex phenotypic trait can be identified by quantitative trait locus analysis (QTL), which allows determination of whether differences in the trait are linked with the location of polymorphic, microsatellite framework markers distributed throughout the genome (37). Either of two measurements of susceptibility/resistance has been used as a quantitative trait for QTL analysis: survival time after iv inoculation of Mtb or control of Mtb multiplication after infecting via the airborne route. Using survival time as a quantitative trait, QTL analysis of (C3H × C57BL/6) F2 progeny with a wide range of survival times revealed strong association between survival time and a 9 cM interval on chromosome 1, at a distance from the location of Nramp1 (32). This finding was subsequently confirmed by a linkage study of (DBA/2 × C57BL/6) F2 mice (31) that showed, in addition, linkage with an interval on the proximal portion of chromosome 7 and the proximal portion of chromosome 3. A subsequent study (36) of progeny of a (A/Sn × I/St) F1 × I/St backcross-infected iv showed linkage with an interval at a different region on chromosome 3 and with an interval on chromosome 9. Thus different laboratories have identified different loci using different strains of mice. This should not be surprising given the likelihood that the genes that control anti-Mtb resistance are numerous and have multiple alleles. Even so, there may be technical reasons for some of the differences observed. For example, mice of the A/Sn and I/St strains infected iv with 5 × 105 Mtb showed survival times of about 45 and 21 days, respectively (38). Therefore, both strains have survival times short enough to warrant classifying them as susceptible. The chronic pathology that takes 250 or more days to kill C57BL/6 and BALB/c mice would not have time to develop in mice that die before 50 days of infection. This type of problem can be partly overcome by using the ability to control Mtb growth as a quantitative trait. This was recently done with (DBA/2 × C57BL/6) F2 mice infected with 100 cfu of Mtb via the respiratory route (25). Whereas C57BL/6 mice were shown to hold lung infection at a stationary level of about 6.5 logs after day 21 of progressive Mtb growth, DBA/2 mice permitted progressive Mtb growth after day 21, resulting in about 2 logs more Mtb by day 100. Informative (C57BL/6 × DBA/2) F2 mice showed a wide range of abilities to control Mtb multiplication in their lungs, as assessed by the number of Mtb at day 90 of infection. QTL analysis revealed a strong linkage with an interval on chromosome 19 that accounted for 25% of the phenotypic variance. The results of this scan, therefore, were different from those obtained from the scan that used survival time as the quantitative trait (31) and point to the possibility that the genes involved in the control of Mtb growth are different from those involved in infection-induced lung pathology and survival. Mapping of resistance/susceptibility to chromosomal regions is the first in a number of steps that need to be taken for high-resolution mapping of the regions and identification of candidate genes within them. Demonstrating linkage of

11 Feb 2004

22:13

610

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

resistance/susceptibility with regions of mouse chromosomes provides reason for looking for the same linkage in syntenic regions of human chromosomes.

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Mouse Tuberculosis as a Model of the Human Disease On the basis of the foregoing discussion it is proposed that mouse tuberculosis can be used to investigate key aspects of the disease in humans. In the mouse, as in more than 85% of susceptible humans, tuberculosis is exclusively a disease of the lungs. Also in the mouse, as in susceptible humans, lung pathology develops progressively in spite of the ability of anti-Mtb defenses to exert a growth inhibitory influence on Mtb. In immunocompromised mice, the disease is more progressive and quickly lethal (see below), as is the case in humans immunocompromised by HIV infection. Therefore, mouse tuberculosis is a model of the disease in immunocompetent susceptible humans, and, as such, can be used to investigate why the lung is the most susceptible organ, and why immunity fails to resolve infection in a susceptible host. The guinea pig and rabbit can be used to investigate the same problems, but at a much higher cost and with fewer experimental tools.

IMMUNITY TO EXPERIMENTAL TUBERCULOSIS That immunity to Mtb and to certain other facultative intracellular bacterial pathogens is cell mediated was proposed by Mackaness in the 1960s (39). He saw lymphocytes as the cells that specifically mediate immunity and macrophages as the cells that nonspecifically express it. According to this investigator, before macrophages can express immunity, they need to be activated to a bactericidal state, and it is the function of antigen-specific lymphocytes to direct them to do this. In the 1970s it was established that antigen-specific T cells cause macrophages to become bacteriostatic or bactericidal by way of the secretion of humoral factors, later to be called lymphokines.

Innate Immunity It is generally believed that the first cells in the lungs that become infected with Mtb and that support its growth are alveolar macrophages. It is now well established (40, 41) that ingestion of Mtb by macrophages triggers, via NF-κB activation, the transcriptional activation of numerous macrophage genes including those that code for proinflammatory cytokines and chemokines. Thus macrophages respond to infection by initiating the inflammatory cascade needed for the extravasation of leukocytes at sites of infection, an essential event not only for the eventual expression of immunity, but almost certainly for its induction. Ingestion of Mtb by mouse bone marrow–derived macrophages in vitro induces transcriptional activation of a surprisingly large number of genes (41). Therefore, it might be expected that infected lung macrophages would exert an inhibitory action on growth of the pathogen in vivo. However, there seems to be no convincing evidence that

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTITUBERCULOSIS IMMUNITY

P1: IKH

611

the growth of Mtb in the lungs is under macrophage-mediated growth restriction during the initial 20-day period of log linear growth, before adaptive immunity is expressed. This is not to say that growth restriction does not occur. To show that it does, however, would require demonstrating that selective suppressing of macrophage function results in a faster rate of Mtb growth. One reason for believing that macrophages exert growth inhibitory action on Mtb is that the pathogen grows faster in glucocorticoid-treated macrophages in vitro (42), presumably because of inhibition of Mtb-activated gene transcription (43). Ingestion of Mtb by macrophages is believed to depend on the engagement of surface receptors of macrophages, including Toll-like receptors (TLRs). TLRs are pattern recognition receptors that enable macrophages and dendritic cells to recognize bacteria, thus ensuring an appropriate immune response is generated to defend against the particular pathogen causing infection. As evidenced by in vitro studies (44, 45), TLRs of macrophages are essential in resistance of mice to Mtb infection, with TLR2 and TLR4 being important. The importance of TLR4 in vivo is shown by a study of the ability of TLR4-deficient versus TLR4-sufficient mice to defend against airborne Mtb infection (46). According to this study, the absence of TLR4 causes a significant decrease in resistance, as evidenced by increased Mtb growth in the lungs and other organs, which results in more extensive pathology and consequently in a shorter survival time. Interestingly, TLR4-deficient mice showed no increase in bacterial growth in the lungs during the initial progressive phase of infection, but did so at a later stage when adaptive immunity is known to be expressed. These results are contradicted, however, by others showing (47) that there is no difference between the ability of TLR4-deficient and TLR4-sufficient mice to defend against airborne infection with Mtb. Additional negative evidence came from a study of TLR4- and TLR2-deficient mutant mice showing (48) that mice deficient in either of these TLRs were no less capable than WT (wildtype) mice in defending against airborne infection with 100 Mtb, although TLR2deficiency resulted in less resistance against a much higher dose of Mtb. However, the subject is controversial and in need of further investigation. Regardless, there is no reason to believe that the engagement of TLRs induces mechanisms of innate resistance capable of exerting a growth inhibitory influence on Mtb prior to the expression of adaptive immunity.

Adaptive Th1-Mediated Immunity Studies over the past few years of anti-Mtb immunity in mice concluded that immunity is mediated predominantly by CD4 Th1 cells with the aid of CD8 T cells. These studies have been the subject of previous reviews (49–51). For the most part, they were based on the use of targeted gene-deleted mice incapable of making certain types of T cells or cytokines involved in the generation and expression of Th1 immunity. The mice were infected either iv with a relatively large number of Mtb or with much smaller numbers via the airborne route. A recent comparative study of targeted gene-deleted mice incapable of making αβ

11 Feb 2004

22:13

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

612

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

T cells, CD4 αβ T cells (Class II−/−), CD8 αβ T cells (Class I−/−), or γ δ T cells showed (22) that both CD4 and CD8 αβ T cells contribute to the ability of mice to inhibit Mtb growth from about day 20 of infection initiated with 100 cfu of Mtb via the respiratory route. In contrast, mice incapable of making γ δ T cells were identical to WT mice in their ability to control infection and survive. Thus in spite of a substantial literature dealing with the in vitro response of γ δ T cells to Mtb antigens (52), these cells appear to have no protective function against Mtb infection in mice. Anti-Mtb immunity is considered to be Th1 mediated, as opposed to Th2 mediated, because of the fact (53, 54) that gene-deleted mice incapable of making IFN-γ fail to acquire the ability to inhibit Mtb growth in their lungs and other organs. This has been taken to indicate that the protective role of CD4 T cells, and most likely CD8 T cells, is based on their ability to synthesize and secrete this key Th1 cytokine and thereby activate the mycobacteriostatic function of macrophages at sites of infection. Additional evidence that Th1 immunity is responsible for control of Mtb growth is that gene-deleted mice incapable of making IL-12 are incapable of expressing anti-Mtb immunity (55). It is well established that secretion of IL-12 by antigen-presenting cells is essential for the induction and generation of Th1 immune responses. A recent study of changes in the level of transcription of genes for Th1 and Th2 cytokines during the course of airborne Mtb infection clearly showed (56) that immunity is based on a dominant Th1 response.

The Relative Importance of CD8 and CD4 T Cells The need for CD8 T cells in immunity can be determined by measuring immunity in Class I gene-deleted mice, which are incapable of the positive and negative selection of CD8 T cells and of presenting antigen to CD8 T cells. The need for CD4 T cells, on the other hand, can be determined with Class II gene-deleted mice, which are incapable of the positive and negative selection of CD4 T cells and of presenting antigen to these cells. However, the relative importance of CD8 and CD4 T cells in anti-Mtb immunity can be assessed only in the same experiment with mice of the same age and sex that are infected with the same number of Mtb at the same time in an aerosol infection chamber. A comparative study of the ability of Class I−/− and Class II−/− mice to defend against Mtb infection showed that Class II-dependent immunity is much more important (22). Whereas in the absence of Class I-dependent immunity, lung infection progressed to a 1 log higher level than in WT mice and was controlled at a stationary level for a long period of time; in the absence of Class II-dependent immunity, infection remained progressive and was quickly lethal. A diagrammatic representation of these findings is shown in Figure 3. The situation with Class I-dependent immunity is more complex, however, in that there are two types of Class I molecules: classical polymorphic Class Ia (H2K,D) and nonclassical nonpolymorphic Class Ib (CD1). Class 1b molecules present

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTITUBERCULOSIS IMMUNITY

P1: IKH

613

Figure 3 Course of Mtb infection in the lungs of B6 WT mice and in B6 gene-deleted mice incapable of making IFN-γ , NOS2, αβ T cells, Class II MHC, Class I MHC, or γ δ T cells (see 22, 67) In WT and γ δ TCR−/− mice, Mtb infection was controlled and stabilized at a stationary level after 3 weeks of progressive Mtb growth. The least capable at controlling Mtb growth were IFN-γ −/− mice, followed by NOS2−/−, αβ TCR−/−, Class II−/−, and Class I−/− mice. The growth rate of the pathogen was the same in all mice until the third week of infection when protective Th1 immunity began to be expressed. The faster the pathogen grew in the lungs after 3 weeks the shorter was host survival time. Whereas WT mice almost certainly die from pulmonary insufficiency, the exact cause of death in gene-deleted mice is not known.

bacterial lipids and glycolipids to CD8 T cells. Most studies of the importance of CD8 T cells in anti-Mtb immunity have used β2-microglobulin (β2m) gene-deleted mice. However, β2m is a component of both Class Ia and Class Ib, meaning that β2m−/− mice are devoid of both types of Class I molecules. Studies aimed at looking for a possible role for MHC Class Ib (CD1) in anti-Mtb immunity using CD1 gene-deleted mice found no difference between the ability of these animals and WT mice to defend against Mtb infection initiated via the iv route (57). Studies that looked for a difference between the ability of β2m−/− mice and mice deleted of classical Class Ia to resist lung infection and to control lung pathology showed that β2m−/− mice succumbed earlier than Class Ia−/− mice (58–60). The reason for this difference has yet to be determined. A study showing (61) that increased Mtb growth in β2m−/− mice is due to defective iron metabolism and iron overload in these mice dealt with a stage of infection that was too early to apply to adaptive Th1-mediated immunity. The way that CD8 T cells contribute to anti-Mtb immunity is not fully understood. In vitro evidence suggests that these cells function via the secretion of IFN-γ to activate macrophages to a mycobacteriostatic state by lysing

11 Feb 2004

22:13

614

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Mtb-infected macrophages or by secreting products that can directly kill Mtb (reviewed in 50). Given that CD8 T cells generated in response to Mtb infection are both cytolytic and capable of synthesizing IFN-γ , it is possible that they contribute to immunity by performing both functions. However, on the basis of results of T cell-depletion studies, there is no reason to believe that CD8 T cells can compensate for an absence of CD4 T cells and vice versa. This could mean that each T cell subset performs a different protective function.

Macrophages as the Effectors of Th1 Immunity According to conventional wisdom, protection afforded by CD4 Th1 cells is based almost entirely on their ability to secrete IFN-γ and other Th1 cytokines and thus activate macrophages to a bacteriostatic state at sites of infection. To some extent this belief is based on faith because cause-and-effect in vivo evidence for an essential role for macrophages has not been presented. There has been no follow-up on immunity to Mtb infection in mice in which macrophage function has been selectively abolished. Belief in an essential role for macrophages rests on the knowledge that macrophages are the cells in which Mtb almost exclusively resides at sites of infection in the lungs and elsewhere during the expression of immunity, and on numerous demonstrations that macrophages activated by IFNγ can express mycobacteriostatic or mycobacteriocidal function in vitro (62). Presumably, macrophage anti-Mtb function is not expressed in the lungs until Mtb-specific CD4 and CD8 T cells extravasate from blood into alveoli in which Mtb-infected macrophages reside. According to the Mtb growth curves, bacteriostatic function is not acquired by the lungs until about day 20 of infection, which is about the time that CD4 T cells begin entering the lungs in large numbers (63). Nitric oxide and its metabolites represent one of two major antimicrobial defense mechanisms of macrophages (discussed in 64). It is generated from Larginine by action of the inducible isoform of nitric oxide synthase 2 (NOS2). The other antimicrobial defense mechanism is based on reactive oxygen, which is generated by the transfer of an electron from NADPH to molecular oxygen by NADPH-oxidase. Mice deleted of the gene for NOS2 are incapable of defending against Mtb infection (65–67), whereas mice incapable of making NADPHoxidase are only slightly susceptible (68) or not more susceptible (67) to Mtb infection than WT mice. In this connection, it was recently shown in vitro (41) that mouse macrophages become partially activated by ingesting Mtb but do not become fully activated and produce NOS2 unless they are treated with IFN-γ . This is in keeping with the view that macrophages do not become activated to a mycobacteriostatic state in the lungs until Th1 immunity is expressed at about day 20 of infection. According to immunocytochemistry, most of the NOS2 synthesized in the lungs on day 20 is present in Mtb-infected macrophages at sites of infection (28).

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

ANTITUBERCULOSIS IMMUNITY

P1: IKH

615

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

It is worth mentioning that although macrophages are a dominant cell type in the histological response to Mtb infection, the origin of these cells is unknown and has received little attention. They might be derived from alveolar macrophages by division, from blood monocytes, or from both sources. Monocytes would seem the most likely source because they are equipped to extravasate at sites of infection in response to proinflammatory stimuli. This subject is important because of the ability of monocytes to differentiate into macrophages or dendritic cells.

Induction, Generation, and Continuous Expression of Th1 Immunity It is now generally believed that immune responses are induced and generated in secondary lymphoid tissue that drains sites of infection. In the case of mice challenged aerogenically, this would be the mediastinal lymph nodes. Presumably, in order for the immune response to Mtb to be initiated, Mtb antigens or Mtb itself needs to reach the draining nodes via afferent lymph. This is likely to be achieved by dendritic cells that are specialized to enter sites of infection from blood, leave via afferent lymph, and locate to the paracortical region of lymph nodes through which naive T cells traffic. Dendritic cells are able to ingest and restrict the growth of Mtb (69), to subsequently mature (70), and to process and present Mtb antigens (71). In mice infected via the airborne route, Mtb reaches the draining node by day 15 (21), which is 5 days before immunity is expressed in the lungs. It is apparent that the generation of Mtb-specific T cells in the draining node has not been studied in any detail. Knowing the time that these T cells are generated in draining nodes is important for understanding why it takes 20 days for immunity to be expressed in the lungs of mice, guinea pigs, and rabbits. If Mtb-specific T cells are made in the nodes well before day 20, it would indicate that they are unable to perform a protective function in the lungs until sites of infection are ready to attract them. The finding (23) that immunity is expressed in the liver and spleen 10 days before it is expressed in the lungs in mice infected via the iv route supports this notion. As discussed above, lung infection in the mouse, guinea pig, and rabbit does not resolve but persists at an approximately stationary level from about day 20 of infection onward. That the stationary level of infection is maintained by the continuous mediation and expression of Th1 immunity is evidenced by the demonstration (72) that depleting mice with stationary lung infection of CD4 T cells with an anti-CD4 mAb results in a resumption of Mtb growth, as does treatment with an inhibitor of NOS2 (73). Additional evidence that active immunity continues to be expressed is seen in the large increases in the levels of IFN-γ , IL-12, and NOS2 gene transcription, attained between days 15 and 20 of infection, that are sustained for at least 100 days thereafter (56, 67). Again, stationary level lung infection has been shown (74) to be associated with the presence in the lungs of replicating CD4 T cells capable of making IFN-γ in response to Mtb antigens.

11 Feb 2004

22:13

616

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

REASONS FOR FAILURE OF IMMUNITY TO RESOLVE INFECTION A central problem in tuberculosis research is to explain why immunity to infection does not enable mice, guinea pigs, rabbits, or susceptible humans to resolve lung infection and thereby stop the development of disease.

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Deficiency in Th1 Immunity Versus Macrophage Function Because macrophages are almost certainly the cells that express immunity, failure of mice and other hosts to resolve Mtb infection must ultimately be determined by the inability of these cells to acquire mycobactericidal function. This might be due to an intrinsic deficiency that prevents macrophages of susceptible hosts from acquiring this function or, alternatively, to an inability of the host to generate a sufficient number of Th1 cells to activate macrophages to a mycobactericidal state. Neither possibility appears to have been seriously investigated. It is simply not known whether macrophages would acquire the ability to kill Mtb and cause infection to resolve if the number of Mtb-specific Th1 cells were to be substantially increased. Needless to say, if it were shown that failure to resolve infection is due to an intrinsic deficiency in macrophage responsiveness, the number of Mtb-specific T cells produced might not matter. Considering that in mice given 100 Mtb via the respiratory route the level of stationary lung infection is over 6 logs and Mtb is distributed among 100 or so lesions (the number of bacilli that implant), the number of Mtb per lesion and per macrophage is likely to be high. It has been our recent experience (R.J. North & Y.-J. Jung, unpublished ongoing research), however, that using chemotherapy to reduce the Mtb load in the lungs by 2 logs does not enable immunity to cause the much lower level of infection to resolve. Given that it is not known whether failure to resolve infection is the result of the generation of too few Mtb-specific Th1 cells, the suggestion that Th1 immunity is incapable of resolving infection because it is downregulated by a Th2 response (75) is premature. Regardless, it has been shown (56) that mice incapable of generating a Th2 response are no more capable than WT mice at dealing with Mtb infection. Likewise, the ability of IL-10 to inhibit macrophage and T cell function is not responsible because the growth of Mtb in gene-deleted mice incapable of making IL10 is essentially the same as its growth in WT mice (76, 77). The demonstration of increased Mtb growth in transgenic mice in which IL-10 is made by cells that do not normally make it and at sites at which it is not necessarily made is of limited value in explaining why an established state of immunity fails to cause infection to resolve. Lastly, the possibility that Mtb-mediated inhibition of macrophage phagosomelysosome fusion (reviewed in 78) is responsible for the inability of macrophages to kill Mtb in vivo is worth considering. However, because phagosome maturation proceeds in Mtb-infected macrophages that are activated by IFN-γ (79), it seems unlikely that this phenomenon is responsible for inability to resolve infection after the onset of IFN-γ -dependent expression of Th1 immunity. It is noteworthy that

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

ANTITUBERCULOSIS IMMUNITY

P1: IKH

617

infection with an avirulent strain of Mtb is resolved after day 20 of infection (67), even though this strain is as capable at inhibiting macrophage phagosomelysosome fusion in vitro as virulent strains are (80). Indeed, avirulent mycobacteria are commonly used to study inhibition of macrophage phagosome maturation.

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Counter Response of the Pathogen Mtb is virulent because it is equipped to avoid destruction by host immunity and to induce pathology. However, the pathogen is virulent in only a susceptible host, meaning that it is virulent in only about 10% of humans. In mice, guinea pigs, and rabbits, immunity succeeds in inhibiting Mtb growth and stabilizing infection at a stationary level. However, there is no evidence that inhibition of Mtb growth by macrophages is associated with structural damage to the pathogen or with inhibition of any of its metabolic pathways. It is possible, therefore, that inhibition of Mtb growth is initiated by the pathogen itself in response to changes in the macrophage phagosome caused by macrophage activation, and that this is the way the pathogen ensures its resistance to antimycobacterial mechanisms of macrophages. It is known (81, 82), in this connection, that ingestion of Mtb by macrophages results in transcriptional changes in the pathogen, and it has been demonstrated in mice (28) that the expression of Th1-mediated inhibition of Mtb growth in the lungs is associated with a change in the transcription pattern of Mtb, as evidenced by a sharp increase in the expression of some Mtb genes, including the gene for the stress protein α-crystallin (acr) and a decrease in the expression of others. Although this type of study does not reveal genes responsible for Mtb survival, it indicates that the pathogen is capable of changing gene transcription in response to changes that occur in the macrophages in which it resides.

PROSPECTS FOR A SUCCESSFUL VACCINE Protection afforded experimental animals by vaccination against Mtb infection can be measured as a reduction in the level of infection, increased survival time, or decreased pathology. It was argued above that it is not known why Th1 immunity fails to resolve primary Mtb infection in mice and other susceptible hosts. It is generally assumed that it is because of the generation of an inadequate level of Th1 immunity in response to infection. According to this line of reasoning, the purpose of vaccination would be to provide the host with a level of immunologic memory that would enable it to generate Mtb-specific Th1 cells faster and in larger numbers in response to subsequent Mtb infection. Current attempts to design a vaccine that is more protective than BCG are based on the assumption that BCG is of insufficient immunogenicity to invoke the generation of a protective level of Mtb-specific immunological memory. However, there is no hard evidence to support this assumption, and, until there is, it remains possible that low immunogenicity is not responsible for the lack of protective efficacy of BCG. Granted, BCG is missing some major Mtb antigens, but there has been no immunologic

11 Feb 2004

22:13

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

618

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

explanation for why the many antigens that it shares with Mtb should not be as protective. It is surely telling that attempts to vaccinate mice with a variety of recently designed vaccines, including subunit vaccines, DNA vaccines, or recombinant BCG, have proved no more, or only marginally more, protective than BCG. All these vaccines enable vaccinated mice to maintain an Mtb challenge infection at about a 1 log lower level than in unvaccinated mice (83). However, there is no reason to believe that the lower level of lung infection does not cause progressive pathology and eventual death, as is the case with BCG-vaccinated guinea pigs (84). Moreover, there is also no reason to believe that the macrophages responsible for inhibiting Mtb growth and maintaining infection at a lower level in vaccinated mice are any more capable at dealing with Mtb than those that maintain infection at a higher level in unvaccinated mice. It is likely, instead, that vaccination simply enables them to become activated earlier. If this were shown to be the case, it would indicate that activated macrophages of susceptible hosts are intrinsically defective in their ability to kill Mtb. This would mean that providing a host with a capacity to generate larger numbers of Mtb-specific T cells would be to no avail. The knowledge (85, 86) that humans cured of tuberculosis by chemotherapy can become reinfected with a different strain of Mtb within weeks suggests that at least some susceptible humans are unlikely to be protected by vaccination. ACKNOWLEDGMENTS Supported by National Institutes of Health grants AI-37,844 and HL-64,565. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Corbett EL, Watt JC, Walker N, Maher D, Williams BG, et al. 2003. The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Arch. Intern. Med. 163:1009–21 2. Mostrom P, Gordon M, Sola C, Ridell M, Rastogi N. 2002. Methods used in molecular epidemiology of tuberculosis. Clin. Microbiol. Infect. 8:694–704 3. Camus JC, Pryor MJ, Medigue C, Cole ST. 2002. Re-annotation of the genome sequence of Mycobacterium tuberculosis H37Rv. Microbiology 148:2967– 73 4. Dannenberg AM. 1989. Immune mechanisms in the pathogenesis of pulmonary

5.

6.

7.

8.

tuberculosis. Rev. Infect. Dis. 11(Suppl.) 2:S369–78 Hill AV. 2001. The genomics and genetics of human infectious disease susceptibility. Annu Rev. Genomics Hum. Genet. 2:373–400 Cervino AC, Lackiss S, Sow O, Bellamy R, Beyers N, et al. 2002. Fine mapping of a putative tuberculosis-susceptibility locus on chromosome 15q11–13 in African families. Hum. Mol. Genet. 11:1599–603 Marquet S, Schurr E. 2001. Genetics of susceptibility to infectious diseases: tuberculosis and leprosy as examples. Drug Metab. Dispos. 29:479–83 Shafer RW, Kim DS, Weiss JP, Quale JM.

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

ANTITUBERCULOSIS IMMUNITY

9.

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

10.

11.

12.

13.

14.

15.

16.

1991. Extrapulmonary tuberculosis in patients with human immunodeficiency virus infection. Medicine 70:384–97 Furcolow ML, Hewell B, Nelson WE. 1942. Quantitative studies of the tuberculin reaction: tuberculin sensitivity in relation to active tuberculosis. Am. Rev. Tuberc. 45:504–20 Al Zahrani K, Al Jahdali H, Menzies D. 2000. Does size matter? Utility of tuberculin reactions for the diagnosis of mycobacterial disease. Am. J. Respir. Crit. Care Med. 162:1419–22 Arend SM, Anderson P, van Meijgaarden KE, Skjot RL, Subronto YW, et al. 2000. Detection of active tuberculosis infection by T cell responses to early-secreted antigenic target 6-kDa protein and culture filtrate protein 10. J. Infect. Dis. 181:1850– 54 Ulrichs T, Anding P, Porcelli S, Kaufmann SH, Munk ME. 2000. Increased numbers of ESAT-6- and purified protein derivative-specific gamma interferonproducing cells in subclinical and active tuberculosis infection. Infect. Immun. 68:6073–76 Lalvani A, Pathan AA, McShane H, Wilkinson RJ, Latif M, et al. 2001. Rapid detection of Mycobacterium tuberculosis infection by enumeration of antigenspecific T cells. Am. J. Respir. Crit. Care Med. 163:824–28 Chaisson RE, Schecter GF, Theueer CP, Rutherford GW, Echenberg DF, Hopewell PC. 1987. Tuberculosis in patients with the acquired immunodeficiency syndrome. Clinical features, response to therapy, and survival. Am. Rev. Respir. Dis. 136:570–74 Casanova J-L, Abel L. 2002. Genetic dissection of immunity to mycobacteria: the human model. Annu. Rev. Immunol. 20:581–620 Medina E, North RJ. 1998. Resistance ranking of some common inbred mouse strains to Mycobacterium tuberculosis and relationship to major histocompat-

17.

18.

19.

20.

21.

22.

23.

24.

25.

P1: IKH

619

ibility complex haplotype and Nramp1 genotype. Immunology 93:270–74 North RJ, Medina E. 1998. How important is Nramp1 in tuberculosis? Trends Microbiol. 6:441–43 Medina E, North RJ. 1996. Evidence inconsistent with a role for the Bcg gene (Nramp1) in resistance of mice to infection with virulent Mycobacterium tuberculosis. J. Exp. Med. 183:1045–51 North RJ, Ryan L, LaCourse R, Mogues T, Goodrich ME. 1999. Growth rate of mycobacteria in mice as an unreliable indicator of mycobacterial virulence. Infect. Immun. 67:5483–85 Schell RF, Ealey WF, Harding GE, Smith DW. 1974. The influence of vaccination on the course of experimental airborne tuberculosis in mice. J. Reticuloendothel. Soc. 16:131–38 Chackerian AA, Alt JM, Perera TV, Dascher CC, Behar SM. 2002. Dissemination of Mycobacterium tuberculosis is influenced by host factors and precedes the initiation of T-cell immunity. Infect. Immun. 70:4501–9 Mogues T, Goodrich ME, Ryan L, LaCourse R, North RJ. 2001. The relative importance of T cell subsets in immunity and immunopathology of airborne Mycobacterium tuberculosis infection in mice. J. Exp. Med. 193:271–80 Medina E, North RJ. 1999. Genetically susceptible mice remain proportionately more susceptible to tuberculosis after vaccination. Immunology 96:16–21 McCune RM, Tompsett R. 1956. Fate of Mycobacterium tuberculosis in mouse tissues as determined by the microbial enumeration technique. J. Exp. Med. 104:737–61 Mitsos L-M, Cardon LR, Ryan L, LaCourse R, North RJ, Gros P. 2003. Susceptibility to tuberculosis: a locus on mouse chromosome 19 (Trl-4) regulates Mycobacterium tuberculosis replication in the lungs. Proc. Natl. Acad. Sci. USA 100:6610–15

11 Feb 2004

22:13

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

620

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

26. Alsaadi A-I, Smith DW. 1973. The fate of virulent and attenuated mycobacteria in guinea pigs infected by the respiratory route. Am. Rev. Respir. Dis. 107:1041–46 27. Lurie MB, Zappasodi P, Tickner C. 1955. On the nature of genetic resistance to tuberculosis in the light of host-parasite relationships in natively resistant and susceptible rabbits. Am. Rev. Tuberc. Pulm. Dis. 72:297–329 28. Shi L, Jung Y-J, Tyagi S, Gennaro ML, North RJ. 2003. Expression of Th1-mediated immunity in mouse lungs induces a Mycobacterium tuberculosis transcription pattern characteristic of nonreplicating persistence. Proc. Natl. Acad. Sci. USA 100:241–46 29. Dunn PL, North RJ. 1995. Virulence ranking of some Mycobacterium tuberculosis and Mycobacterium bovis strains according to their ability to multiply in the lungs, induce lung pathology and cause mortality in mice. Infect. Immun. 63:3428–37 30. Dunn PL, North RJ. 1996. Persistent infection with virulent but not avirulent Mycobacterium tuberculosis in the lungs of mice causes progressive pathology. J. Med. Microbiol. 45:103–9 31. Mitsos L-M, Cardon LR, Fortin A, Ryan L, LaCourse R, et al. 2000. Genetic control of susceptibility to infection with Mycobacterium tuberculosis in mice. Genes Immun. 1:467–77 32. Kramnik I, Dietrich WF, Demant P, Bloom BR. 2000. Genetic control of resistance to experimental infection with virulent Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 97:8560–65 33. Lurie MB, 1932. The correlation between the histological changes and the fate of living tubercle bacilli in the organs of tuberculous rabbits. J. Exp. Med. 55:31–42 34. Smith DW, Harding GE. 1977. Experimental airborne tuberculosis in the guinea pig. Am. J. Pathol. 89:273–76 35. Gehr P, Mwangi DK, Ammann A, Mal-

36.

37.

38.

39.

40.

41.

42.

43.

oiy GM, Taylor CR, Weibel ER. 1981. Design of the mammalian respiratory system. V. Scaling morphometric pulmonary diffusion capacity to body mass: wild and domestic mammals. Respir. Physiol. 44:61–86 Sapoval B, Filoche M, Weibel ER. 2002. Smaller is better—but not too small: a physical scale for the design of mammalian pulmonary acinus. Proc. Natl. Acad. Sci USA 99:10411–16 Fisler JS, Warden CH. 1997. Mapping of mouse obesity genes: a generic approach to a complex trait. J. Nutr. 127(9):1909S– 1916S Sanchez F, Radaeva TV, Nikonenko BV, Persson AS, Sengul S, et al. 2003. Multigenic control of disease after virulent Mycobacterium tuberculosis infection in mice. Infect. Immun. 71:126–31 Mackaness GB. 1969. The influence of immunologically commited lymphoid cells on macrophage activity in vivo. J. Exp. Med. 129:973–92 Ragno S, Romano M, Howell S, Pappin DJ, Jenner PJ, Colston MJ. 2001. Changes in gene expression in macrophages infected with Mycobacterium tuberculosis: a combined transcriptomic and proteomic approach. Immunology 104:99–108 Ehrt S, Schnappinger D, Bekiranov S, Drenkow J, Shi S, et al. 2001. Reprogramming of the macrophage transcriptome in response to interferon-gamma and Mycobacterium tuberculosis: signaling roles of nitric oxide synthase-2 and phagocytic oxidase. J. Exp. Med. 194:1123–40 Rook GA, Steele J, Ainsworth M, Leveton C. 1987. A direct effect of glucocorticoid hormones on the ability of human and murine macrophages to control the growth of M. tuberculosis. Eur. J. Respir Dis. 71:286–91 Adcock IM. 2001. Glucocorticoidregulated transcription factors. Pulm. Pharmacol Ther. 14:211–19

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTITUBERCULOSIS IMMUNITY 44. Means TK, Wang S, Lien E, Yoshimura A, Golenbock DT, Fenton MJ. 1999. Human Toll-like receptors mediate cellular activation by Mycobacterium tuberculosis. J. Immunol. 163:3920–27 45. Means TK, Jones BW, Schromm AB, Shurtleff BA, Smith JA, et al. 2001. Differential effects of Toll-like receptor antagonist on Mycobacterium tuberculosis-induced macrophage responses. J. Immunol. 166:4074–82 46. Abel B, Thieblemont N, Quesniaux VJ, Brown N, Mpagi J, et al. 2002. Toll-like receptor 4 expression is required to control chronic Mycobacterium tuberculosis infection in mice. J. Immunol. 169:3155– 62 47. Kamath AB, Alt J, Debbabi H, Behar SM. 2003. Toll-like receptor 4-defective C3H/HeJ mice are not more susceptible than other C3H substrains to infection with Mycobacterium tuberculosis. Infect. Immun. 71:4112–18 48. Reiling N, Holscher C, Fehrenbach A, Kroger S, Kirschning CJ, et al. 2002. Toll-like receptor (TLR)2- and TLR4-mediated pathogen recognition in resistance to airborne infection with Mycobacterium tuberculosis. J. Immunol. 169:3480–84 49. Boom WH. 1996. The role of T cell subsets in Mycobacterium tuberculosis infection. Infect. Agents Dis. 5:73–81 50. Flynn JL, Chan J. 2001. Immunology of tuberculosis. Annu. Rev. Immunol. 19:93–129 51. Kaufmann SH. 2001. How can immunology contribute to the control of tuberculosis? Nat. Rev. Immunol. 1:20–30 52. Boom WH. 1999. γ δ T cells and Mycobacterium tuberculosis. Microbes Infect. 1:187–95 53. Flynn JL, Chan J, Triebold KJ, Dalton DK, Stewart TA, Bloom BR. 1993. An essential role for interferon-γ in resistance to Mycobacterium tuberculosis infection. J. Exp. Med. 178:2249–54 54. Cooper AM, Dalton DK, Stewart TA,

55.

56.

57.

58.

59.

60.

61.

62.

P1: IKH

621

Griffen JP, Russell DG, Orme IM. 1993. Disseminated tuberculosis in IFNγ gene-disrupted mice. J. Exp. Med. 178: 2243–47 Cooper AM, Magram J, Ferrante J, Orme IM. 1997. Interleukin 12 (IL-12) is crucial to the development of protective immunity in mice intravenously infected with Mycobacterium tuberculosis. J. Exp. Med. 186:39–45 Jung Y-J, LaCourse R, Ryan L, North RJ. 2002. Evidence inconsistent with a negative influence of T helper 2 cells on protection afforded by a dominant T helper 1 response against Mycobacterium tuberculosis lung infection in mice. Infect. Immun. 70:6436–43 Behar SM, Dascher CC, Grusby MJ, Wang CR, Brenner MB. 1999. Susceptibility of mice deficient in CD1D or TAP1 to infection with Mycobacterium tuberculosis. J. Exp. Med. 189:1973–80 Sousa AO, Mazzaccaro RJ, Russell RG, Lee FK, Turner OC, et al. 2000. Relative contribution of distinct MHC Class Idependent cell populations in protection to tuberculosis infection in mice. Proc. Natl. Acad. Sci. USA 97:4204–8 Rolph MS, Raupach B, Kobernick HH, Collins HL, Perarnau B, et al. 2001. MHC Class Ia-restricted T cells partially account for beta2-microglobulindependent resistance to Mycobacterium tuberculosis. Eur. J. Immunol. 31:1944– 49 Urdahl KB, Liggitt D, Bevan MJ. 2003. CD8+ T cells accumulate in the lungs of Mycobacterium tuberculosis-infected Kb−/−Db−/− mice, but provide minimal protection. J. Immunol. 170:1987–94 Schaible UE, Collins HL, Priem F, Kaufmann SH. 2002. Correction of the iron overload defect in beta-2-microglobulin knockout mice by lactoferrin abolishes their increased susceptibility to tuberculosis. J. Exp. Med. 196:1507–13 Kaufmann SH, Flesch IE. 1988. The role of T cell-macrophage interactions

11 Feb 2004

22:13

622

63.

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

64.

65.

66.

67.

68.

69.

70.

AR

NORTH

AR210-IY22-20.tex

¥

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

P1: IKH

JUNG

in tuberculosis. Springer Semin. Immunopathol. 10:337–58 Mason CM, Dobard E, Shellito J, Nelson S. 2001. CD4+ lymphocyte responses to pulmonary infection with Mycobacterium tuberculosis in naive and vaccinated BALB/c mice. Tuberculosis 81:327–34 Murray HW, Nathan CF. 1999. Macrophage microbicidal mechanisms in vivo: reactive nitrogen versus oxygen intermediates in the killing of intracellular visceral Leishmania donovani. J. Exp. Med. 189:741–46 MacMicking JD, North RJ, LaCourse R, Mudgett JS, Shah SK, Nathan CF. 1997. Identification of nitric oxide synthase as a protective locus against tuberculosis. Proc. Natl. Acad. Sci. USA 94:5243–48 Scanga CA, Mohan VP, Tanaka K, Alland D, Flynn JL, Chan J. 2001. The inducible nitric oxide synthase locus confers protection against aerogenic challenge of both clinical and laboratory strains of Mycobacterium tuberculosis in mice. Infect. Immun. 69:7711–17 Jung Y-J, LaCourse R, Ryan L, North RJ. 2002. Virulent but not avirulent Mycobacterium tuberculosis can evade the growth inhibitory action of a T helper 1-dependent, nitric oxide synthase 2independent defense in mice. J. Exp. Med. 196:991–98 Cooper AM, Segal BH, Frank AA, Holland SM, Orme IM. 2000. Transient loss of resistance to pulmonary tuberculosis in p47phox47−/− mice. Infect. Immun. 68:1231–34 Bodnar KA, Serbina NV, Flynn JL. 2001. Fate of Mycobacterium tuberculosis within murine dendritic cells. Infect. Immun. 69:800–9 Tascon RE, Soares CS, Ragno S, Stavopoulos E, Hirst EM, Colston MJ. 2000. Mycobacterium tuberculosisactivated dendritic cells induce protective immunity in mice. Immunology 99: 473–80

71. Goldmann O, Road M, Medina E. 2002. Phagocytosis of bacillus CalmetteGuerin-infected necrotic macrophages induces a maturation phenotype and evokes antigen-presentation function in dendritic cells. Immunology 107:500–6 72. Scanga CA, Mohan VP, Yu K, Joseoh H, Tanaka K, et al. 2000. Depletion of CD4 T cells causes reactivation of murine persistent tuberculosis despite continued expression of interferon gamma and nitric oxide synthase 2. J. Exp. Med. 192:347– 58 73. Flynn JL, Scanga CA, Tanaka KE, Chan J. 1998. Effects of aminoguanidine on latent murine tuberculosis. J. Immunol. 160:1796–803 74. Winslow GM, Roberts AD, Blackman MA, Woodland DL. 2003. Persistence and turnover of antigen-specific CD4 T cells during chronic tuberculosis infection in the mouse. J. Immunol. 170:2046– 52 75. Hernandez-Pando R, Orozcoe H, Sampieri A, Pavon L, Velasquillo C, et al. 1996. Correlation between the kinetics of Th1, Th2 cells and pathology in a murine model of experimental pulmonary tuberculosis. Immunology 89: 26–33 76. Roach DR, Martin E, Bean AG, Renick DM, Briscoe H, Britton WJ. 2001. Endogenous inhibition of antimycobacterial immunity by IL-10 varies between mycobacterial species. Scand. J. Immunol. 54:163–70 77. Jung Y-J, Ryan L, LaCourse R, North RJ. 2003. Increased interleukin-10 expression is not responsible for failure of T helper 1 immunity to resolve airborne Mycobacterium tuberculosis infection in mice. Immunology 109:295–99 78. Russell DG. 2001. Mycobacterium tuberculosis: here today, and here tomorrow. Nat. Rev. Mol. Cell. Biol. 2:569– 77 79. Via LE, Fratti RA, McFalone M, PaganRamos E, Deretic D, Deretic V. 1998.

11 Feb 2004

22:13

AR

AR210-IY22-20.tex

AR210-IY22-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ANTITUBERCULOSIS IMMUNITY Effects of cytokines on mycobacterial phagosome maturation. J. Cell Sci. 111:897–905 80. Ramachandra L, Noss E, Boom WH, Harding CV. 2001. Processing of Mycobacterium tuberculosis antigen 85B involves intraphagosomal formation of peptide-major histocompatibility complex II complexes and is inhibited by live bacilli that decrease phagosome maturation. J. Exp. Med. 194:1421–32 81. Cappelli G, Volpe P, Sanduzzi A, Sacchi A, Colizzi V, Mariani F. 2001. Human macrophage gamma interferon decreases gene expression but not replication of Mycobacterium tuberculosis: analysis of the host-pathogen reciprocal influence on transcription in a comparison of strains H37Rv and CMT97. Infect. Immun. 69:7262–70 82. Monahan IM, Betts J, Banerjee DK, Butcher PD. 2001. Differential expres-

83.

84.

85.

86.

P1: IKH

623

sion of mycobacterial proteins following phagocytosis by macrophages. Microbiology 147:459–71 Britton WJ, Palendira U. 2003. Improving vaccines against tuberculosis. Immunol. Cell Biol. 81:34–45 Wiegeshaus EH, McMurray DN, Grover AA, Harding GE, Smith DW. 1970. Host-parasite relationships in experimental tuberculosis. Am. Rev. Respir. Dis. 102:422–29 van Rie A, Warren R, Richardson M, Victor TC, Gie RP, et al. 1999. Exogenous reinfection as a cause of recurrent tuberculosis after curative treatment. N. Engl. J. Med. 341:1174–79 Caminero JA, Pena MJ, Campos-Herrero MI, Rodriguez JC, Afonso O, et al. 2001. Exogenous reinfection with tuberculosis on a European island with a moderate incidence of disease. Am. J. Respir. Crit. Care Med. 163:717–20

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:599-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104614

Annu. Rev. Immunol. 2004. 22:625–55 doi: 10.1146/annurev.immunol.22.012703.104614 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on December 12, 2003

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION Rebecca H. Buckley Departments of Pediatrics and Immunology, Duke University Medical Center, Durham, North Carolina 27710; email:[email protected]

Key Words cytokine receptor genes, antigen receptor rearrangement genes, bone marrow transplantation, gene therapy ■ Abstract Mutations in nine different genes have been found to cause the human severe combined immunodeficiency syndrome. The products of three of the genes— IL-2RG, Jak3, and IL-7Rα—are components of cytokine receptors, and the products of three more—RAG1, RAG2, and Artemis—are essential for effecting antigen receptor gene rearrangement. Additionally, a deficiency of CD3δ, a component of the T-cell antigen receptor, results in a near absence of circulating mature CD3+ T cells and a complete lack of γ /δ T cells. Adenosine deaminase deficiency results in toxic accumulations of metabolites that cause T cell apoptosis. Finally, a deficiency of CD45, a critical regulator of signaling thresholds in immune cells, also causes SCID. Approaches to immune reconstitution have included bone marrow transplantation and gene therapy. Bone marrow transplantation, both HLA identical unfractionated and T cell–depleted HLA haploidentical, has been very successful in effecting immune reconstitution if done in the first 3.5 months of life and without pretransplant chemotherapy. Gene therapy was highly successful in nine infants with X-linked SCID, but the trials have been placed on hold due to the development of a leukemic process in two of the children because of insertional oncogenesis.

INTRODUCTION Human severe combined immunodeficiency (SCID) was first reported by Swiss workers more than 50 years ago (1). Infants with the condition were profoundly lymphopenic and died of infection before their first or second birthdays. In the ensuing years, differences in inheritance patterns were noted, indicating that there was more than one cause for this condition. In many families there was clearly X-linked recessive inheritance, whereas in others there was autosomal recessive inheritance. The first discovered molecular cause of human SCID, adenosine deaminase deficiency, was reported in 1972 (2). However, it was not until 21 years later that a second fundamental cause of the condition was found, i.e., the molecular 0732-0582/04/0423-0625$14.00

625

12 Feb 2004

15:56

626

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 1 Abnormal genes known to cause SCID Chromosome

Disease

1q31–32

SCID caused by CD45 deficiency∗

5p13

SCID due to IL-7 receptor alpha chain deficiency∗

10p13

SCID (radiation sensitive; Athabascan) due to mutations in the Artemis gene∗

11p13

SCID caused by RAG1 or RAG2 deficiencies∗

11q23

SCID caused by CD3 delta chain deficiency

19p13.1

SCID caused by Jak3 deficiency∗

20q13.11

SCID caused by adenosine deaminase (ADA) deficiency∗

Xq13.1

X-linked SCID caused by common gamma-chain (γ c) deficiency∗

∗

Gene cloned and sequenced; gene product known.

basis of X-linked SCID (3, 4). Since then, remarkable progress has been made in elucidating several other causes of this syndrome. It is now known that SCID can be caused in humans by mutations in at least nine different genes (Table 1) (5–9a), and the likelihood is that there are other causes yet to be discovered. Regardless of the underlying defect, infants with this syndrome are lymphopenic (Figure 1) and have profound deficiencies of T and B cell numbers and function (Figures 2a, 3a). In certain types, there are also marked deficiencies of natural killer (NK) cells and NK cell function (Figure 2a).

CLINICAL PRESENTATION Affected infants begin to have problems with oral moniliasis, diarrhea, and failure to thrive in the first few months of life. Although the diagnosis of SCID can be made much earlier, it is frequently not made until serious infections develop, with the average age at referral for immune testing being approximately 6 months. Persistent infections with opportunistic organisms, such as Candida albicans, Pneumocystis carinii, varicella, adenovirus, respiratory syncitial virus, parainfluenza 3, cytomegalovirus, Epstein-Barr virus (EBV), and bacillus Calmette-Guerin (BCG), lead to death (Table 2) (10, 11). These infants also lack the ability to reject allografts, leaving them at risk for fatal GVHD. This condition is uniformly fatal in the first two years of life unless immune reconstitution can be accomplished (12–14). Recognition of the characteristic lymphopenia (Figure 1) can result in early diagnosis—even at birth (11, 13). Their lymphocytes fail to proliferate in vitro in response to mitogens, antigens,or allogeneic cells (Figure 3a). Serum immunoglobulins and antibodies are diminished to absent. The thymuses are very small (usually less than 1g) and lack thymocytes, corticomedullary distinction, and Hassall’s corpuscles. However, recent studies have shown that these thymuses are capable of

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

627

Figure 1 Means +/− SEM of absolute lymphocyte counts in 132 SCID infants at presentation, showing the characteristic of lymphopenia in all forms of SCID.

supporting T cell development when normal stem cells are provided (13, 15). Thymus-dependent areas of the spleen are devoid of lymphocytes, and lymph nodes and tonsils are absent. Flow cytometric studies have shown that there are unique lymphocyte phenotypes for the various genetic forms of SCID (Table 3), with some having B cells and no NK cells (so-called T−B+NK− SCID), others having no T cells but normal or elevated numbers of B and NK cells (T−B+NK+ SCID), others having no B cells but many NK cells (T−B−NK+ SCID), and others having extremely low numbers of all types of lymphocytes (T−B−NK− SCID) (Figure 2a) (6, 11, 12, 16). SCID is a pediatric emergency (11–13). Nearly all cases could be diagnosed at birth if routine blood counts and manual differentials were done and flow cytometry and T cell functional studies performed when lymphocyte counts are below the newborn normal range (2000–11,000/mm3) (11, 17). Treatment could then be given shortly after birth (13).

MOLECULAR CAUSES OF SCID Enormous strides have been made in identifying the molecular causes of SCID, with all but one of these discoveries having been made within the past 10 years (Table 1) (4–9, 18, 19). The products of three of the mutated genes that cause SCID

12 Feb 2004

15:56

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

628

AR

Figure 2 Means +/– SEM of CD20+ B cells, CD3+ T cells, and CD16+ NK cells in 132 SCID patients before (a) and most recently after transplantation (b) in the 102 survivors according to genetic type, as compared with ranges for normal controls.

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

629

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

Figure 3 Means +/− SEM cpm [3H] thymidine incorporation by proliferating lymphocytes from the 132 SCIDs before (a) and most recently after transplantation in the 102 survivors (b) according to genetic type in response to the mitogens, PHA, Con A, and PWM, as compared with means +/− SEM for normal controls.

12 Feb 2004

15:56

630

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY TABLE 2 Causes of death in 30 SCID infants post-transplantation Number of infants affected

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Cause of death Cytomegalovirus infection

7

Adenovirus infection

7

Epstein Barr virus infection/lymphoma

3

Enterovirus/Rotovirus infections

3

Parainfluenza 3 infection

2

Varicella infection

2

Herpesvirus infection

1

Respiratory syncitial virus infection

1

Candida sepsis

2

Mitochondrial defect

1

Nephrotic syndrome due to chemotherapy

1

Pulmonary hypertension

1

Veno-occlusive disease

1

are components of cytokine receptors. The products of three others are essential for rearrangement of antigen receptor genes, and the product of another is a component of the T-cell antigen receptor that appears essential for T cell development. The product of one gene is necessary to prevent toxic accumulation of metabolic wastes that lead to lymphocyte apoptosis. Finally, the product of the remaining mutated gene is the common leukocyte antigen, CD45, a phosphatase critical for regulating signaling thresholds in immune cells.

TABLE 3 Lymphocyte phenotypes of the different molecular types of SCID Lymphocyte phenotype

Type of SCID

T−B+NK−

X-linked (γ c deficiency) Jak 3 deficiency CD45 deficiency

T−B+NK+

IL-7R alpha chain deficiency CD3 delta chain deficiency

T−B−NK−

Adenosine deaminase deficiency

T−B−NK+

RAG1 or RAG2 deficiency Artemis deficiency

15 Mar 2004

15:34

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

631

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Adenosine Deaminase Deficiency An absence of the enzyme adenosine deaminase (ADA) was identified by Giblett and coworkers in 1972 as a cause of SCID (2); this defect accounts for approximately 17% of patients with the condition (Figure 4) (12, 18). The gene encoding ADA was mapped to chromosome 20q13.2–q13.11, cloned, and sequenced (20). The ADA deficiency caused by mutations in this gene results in marked accumulations of adenosine, 20 -deoxyadenosine and 20 -O-methyladenosine. The latter directly or indirectly lead to lymphocyte apoptosis, resulting in the absence of T cell function. There are certain distinguishing features of ADA deficiency, including multiple skeletal abnormalities of chondro-osseous dysplasia on radiographic examination. The latter include flaring of the costochondral junctions and a bone-in-bone anomaly in the vertebral bodies. ADA-deficient patients have a more profound lymphopenia than do infants with other types of SCID, with mean absolute lymphocyte counts of less than 500/mm3 and a deficiency of all three types of immune cells (T−B−NK− SCID) (Figure 2a, Table 3) (11, 12). Milder forms of this condition have been reported, leading to delayed diagnosis of immunodeficiency even to adulthood (21). The diagnosis should be suspected in any patient with recurrent infections who has severe lymphopenia.

Common Gamma Chain Deficiency X-linked recessive severe combined immunodeficiency (SCID-X1) is the most common form of SCID, accounting for approximately 46% of U.S. cases (Figure 4) (11, 12). The abnormal gene in SCID-X1 was mapped to the Xq13 region

Figure 4 Relative frequencies of the different genetic types of SCID among 170 patients seen consecutively by the author over 3.5 decades.

15 Mar 2004

15:34

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

632

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

and later identified as the gene encoding the common gamma chain (γ c) shared by cell surface receptors for various interleukin molecules (IL-2, IL-4, IL-7, IL-9, IL-15, and IL-21) (Table 1; Figure 5) (3, 4, 22, 23). Among the first 136 SCIDX1 patients studied, 95 distinct mutations were identified, resulting in abnormal γ c chains in two thirds of the cases and absent γ c protein in the remainder (24). The finding that the mutated gene results in faulty signaling through several cytokine receptors explains how T, B, and NK cells can be affected by a single mutation (19, 25). This was the initially discovered example of T−B+NK− SCID (Table 3). Humans with γ c deficiency differ from the γ c murine knockouts in that they do have B cells, whereas the mice do not (26). Even though B cells are the dominant lymphocyte type present in the circulation, the B cells do not undergo class switch recombination. An exception to SCID being invariably fatal without marrow transplantation or gene therapy was seen in a patient with SCID-X1 who underwent spontaneous clinical improvement and was found to have reversion of a documented mutation in the gene encoding γ c, presumably in a T cell precursor (27).

Janus Kinase 3 Deficiency (Jak3-Deficient SCID) SCID patients with autosomal recessive SCID caused by Jak3 deficiency resemble all other types in their susceptibility to infection and to graft-versus-host disease from allogeneic T cells. However, they have lymphocyte characteristics most closely resembling those of patients with X-linked SCID, including an elevated percentage of B cells and very low percentages of T and NK cells (Figure 2a; Table 3) (11, 12). Because Jak3 is the only signaling molecule known to be associated with γ c (28), it was a candidate gene for mutations leading to autosomal recessive T−B+NK− SCID, the identical lymphocyte phenotype seen in SCID-X1 (Figures 2a and 5). This proved to be the case and thus far more than 30 patients have been identified who lack Jak3 (5, 11, 29, 30). As in the case of γ c deficiency, B cells are present in humans with Jak3 deficiency but they do not undergo isotype switching, whereas B cells are absent in the Jak3 murine knockout (26). Naturally occurring Jak3 mutations that result in SCID can serve as valuable tools for delineation of the phenotype and clinical course of Jak3-deficient SCID following bone marrow transplantation, and for Jak3 structure-function analysis. A report of one such evaluation of Jak3 mutations has been made for a series of 23 Jak3-deficient patients from Europe, where it was noted that the clinical phenotype can vary greatly with this condition, ranging from classical SCID to almost normal (31). The author and her associates have found novel Jak3 mutations in 11 Jak3-deficient SCID patients, the first reported series from the United States, and have summarized the immunologic and clinical data as well as the results of bone marrow transplantation in these patients (Figure 4) (31a). Like SCID-X1 patients, Jak3-deficient SCID infants have very low numbers of NK cells, even after successful marrow transplantation (Figure 2a,b) (12). Moreover, in further similarity to SCID-X1 patients, they often fail to develop normal B cell function after transplantation despite their high

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

633

numbers of B cells. Their failure to develop NK cells or B cell function is believed to be due to these host B cells’ abnormal cytokine receptors. The impaired IL-4 and IL-21 cytokine receptor signaling is thought to contribute to the host B cell dysfunction even though adequate T cell help is provided by the donor-derived T cells (32). Based on findings from IL-15 deficient mice, the NK cell deficiency in both SCID-X1 and Jak3-deficient SCIDs is thought to be due to failure to signal through the IL-15 receptor (33).

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-7 Receptor Alpha Chain Deficiency (IL-7Rα-Deficient SCID) Several of the author’s SCID patients who had previously been shown not to have either γ c or Jak3 deficiency had a T−B+NK+ phenotype. Because mice whose genes for either the alpha chain of the IL-7 receptor (34) or of IL-7 itself (35) have been mutated are profoundly deficient in T and B cell function but have normal natural killer cell function, mutations in these genes were sought in human SCID (Figure 2a). Mutations in the gene for IL-7Rα on chromosome 5p13 have been found thus far in 17 of the author’s patients, as well as in 3 others (36), making it the third most common cause of human SCID in the United States (Figure 4) (6; J. Roberts, S. Brown, R. Buckley, submitted). Thus far, no humans who have SCID because of IL-7 deficiency have been found. The finding that mutations in IL-7Rα alone result in T cell deficiency but not B cell or NK cell deficiency implies that the T cell but not the NK cell defect in SCID-X1 and Jak3-deficient SCID results from an inability to signal through the IL-7 receptor (Figure 5) (6, 37). Unlike IL-7Rα mutant mice, humans with IL-7Rα mutations not only have B cells, but these B cells appear to function normally after T cell immune reconstitution is effected by allogeneic bone marrow transplantation.

Recombinase-Activating Gene Deficiencies (RAG1- or RAG2-Deficient SCID) Infants with autosomal recessive SCID caused by mutations in recombinaseactivating genes, RAG1 and RAG2, resemble all others in their infection susceptibility and complete absence of T or B cell function. However, their lymphocyte phenotype differs from those of patients with SCID caused by γ c, Jak3, IL-7Rα, or ADA deficiencies in that they lack both B and T lymphocytes and have primarily NK cells in their circulation (T−B−NK+ SCID) (Table 3; Figure 2a). This particular phenotype suggested a possible problem with their antigen receptor genes, leading to the discovery of mutations in RAG1 and RAG2 in approximately half of such SCID infants (7, 38, 39). These genes, on chromosome 11p13, encode proteins necessary for somatic rearrangement of antigen receptor genes on T and B cells. The proteins recognize recombination signal sequences (RSSs) and introduce a DNA double-stranded break, permitting V, D, and J gene rearrangements. RAG1 or RAG2 mutations result in a functional inability to form antigen receptors through genetic recombination. This genetic type of SCID is more common in Europe than in the United States. Only 5 such patients have been

12 Feb 2004

15:56

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

634

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

found among the 170 SCID patients evaluated by the author (Figure 4). RAG1or RAG2-deficient SCIDs frequently fail to develop B cells after bone marrow transplantation. In addition to causing the SCID phenotype, some mutations in RAG1 or RAG2 genes lead to partially impaired V(D)J recombinational activity resulting in Omenn’s syndrome (39, 40). Omenn’s syndrome is characterized by the development soon after birth of a generalized erythroderma and desquamation, diarrhea, hepatosplenomegaly, hypereosinophilia, and markedly elevated serum IgE levels but very low levels or absence of the other immunoglobulin isotypes. The absolute lymphocyte count is elevated due to circulating, activated, and oligoclonal T lymphocytes that do not respond normally to mitogens or antigens in vitro (41, 42). Circulating B cells are not found, and lymph node architecture is abnormal due to a lack of germinal centers (43). Omenn’s syndrome is fatal unless corrected by bone marrow transplantation. Unlike the situation for SCID infants, pretransplant chemotherapy is necessary for bone marrow graft acceptance in Omenn’s syndrome.

CD3δ Chain Deficiency (CD3δ–Deficient SCID) The most recently discovered cause of human autosomal recessive SCID is CD3δ chain deficiency (9a). Infants with mutations in the gene encoding the delta chain of CD3 resemble all others in their infection susceptibility and complete absence of T cell function. Mutations in the human genes encoding CD3ε and CD3γ chains result in only a partial arrest of T cell maturation and, therefore, only moderate immunodeficiency (43a, 43b). By contrast, a homozygous stop codon mutation in the region of CD3δ that encodes the extracellular domain of CD3δ resulted in a profound deficiency of mature circulating CD3+ T cells, no CD4+ or CD8+ T cells, and a total absence of γ /δ T cells in three Mennonite first cousins (9a). The number of B cells was either normal or increased, and NK cells were normal in all. Thus, their lymphocyte phenotype resembled that of IL-7Rα deficiency. Lymphocyte responses to mitogens were absent. In distinction from the other eight molecular types of human SCID, these infants with CD3δ deficiency each had a nearly normal sized thymus on chest radiography. Histopathologically, there were moderate populations of T cell precursors but no typical corticomedullary distinction and no Hassell’s corpuscles. These findings suggest that CD3δ is essential for human T cell development (9a).

CD45 Deficiency Another autosomal recessive cause of human SCID is a mutation in the gene encoding the common leukocyte surface protein CD45 (9, 44, 45). This hematopoieticcell-specific transmembrane protein tyrosine phosphatase functions to regulate Src kinases required for T- and B-cell antigen receptor signal transduction (46). A 2month-old male infant presented with a clinical picture of SCID and was found to have very low numbers of T and NK cells but an elevated number of B cells

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

635

(Table 3). The T cells failed to respond to mitogens, and serum immunoglobulins diminished with time. He was found to have a large deletion at one CD45 allele and a point mutation causing an alteration of the intervening sequence 13 donor splice site at the other (9). A second case of SCID due to CD45 deficiency has been reported (44, 45).

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Artemis Deficiency Another recently discovered cause of human SCID is a deficiency of a novel V(D)J recombination/DNA repair factor that belongs to the metallo-β−lactamase superfamily. It is encoded by a gene on chromosome 10p called Artemis (Table 1) (8, 47). A deficiency of this factor results in an inability to repair DNA after doublestranded cuts have been made by RAG1 or RAG2 gene products in rearranging antigen receptor genes from their germline configuration. Similar to RAG1- and RAG2deficient SCID, this defect results in another form of T−B−NK+ SCID (Table 3), also called Athabascan SCID (48). In addition, there is increased radiation sensitivity of both skin fibroblasts and bone marrow cells of those affected with this type of SCID.

APPROACHES TO IMMUNE RECONSTITUTION Bone Marrow Transplantation Shortly after the discovery of HLA in 1968, (49, 50) immune function was transferred to an infant with SCID by transplantation of bone marrow from his HLAidentical sister (51). Over the ensuing decade, however, lethal graft-versus-host disease (GVHD) was a major problem when marrow from HLA-mismatched donors was used (52). In the late 1970s, studies in rats (53) and mice (54) revealed that allogeneic marrow or spleen cells that were depleted of T cells rescued the recipient from lethal irradiation without causing fatal GVHD, despite differences in MHC antigens between donor and host. Techniques developed in the early 1980s for the depletion of T cells from human marrow made it possible to restore immune function in all forms of SCID by marrow transplantation (10–14, 55–65). Because the defect in infants with SCID leads to a complete absence of T cell function, they cannot reject allografts. Therefore, successful marrow transplantation in SCID does not require pretransplant chemotherapeutic conditioning. Moreover, prophylaxis for GVHD is also not necessary after transplantation of HLA-identical or even rigorously T cell–depleted HLA haploidentical marrow because in the latter case the T cells that cause GVHD have been removed. These circumstances provide a unique opportunity to study the development of T cells from donor hematopoietic stem cells in an unmanipulated recipient because less than 1 million T cells per kilogram are given when one uses rigorous T cell–depletion techniques (66). At the time of this writing the author and her colleagues had performed hematopoietic stem cell transplantation in 132

12 Feb 2004

15:56

636

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

consecutive infants with SCID at our institution over 21.3 years, and 102 of them survived. The outcome in all but 30 of these transplants has been previously reported (12, 13). The 132 patients ranged in age from newborn to 21 months at diagnosis. They fulfilled the criteria of the World Health Organization for diagnosis of SCID (67). The patients were from 117 families and fit eight categories of the disease based on family history, sex, other clinical features, and results of enzyme analyses or molecular studies (11). The largest number of patients—62 boys from 50 families or 47%—had x-linked SCID due to mutations in the gene encoding the common γ chain (γ c) (3, 4, 11, 68). Eight patients from 8 families had SCID due to mutations in the gene encoding Jak3 (5, 11, 29). Seventeen patients from 16 families had SCID caused by mutations in the gene encoding IL-7Rα (6). Twentytwo infants from 19 families had SCID due to a deficiency of ADA (69). Fifteen patients from 15 families had autosomal recessive inheritance but unknown mutations; and 4 boys with no family history had SCID of unknown type. Immunologic monitoring was done whenever feasible every 3 weeks until T cell function was established (usually at 3–4 months post-transplantation), then every 3 months for the next 9 months, every 6 months for the next 2 years, then once a year thereafter. The studies were done with the approval of the Duke University Committee on Human Investigations and written informed consent of the parents. All 117 HLA-haploidentical and 7 of the 15 HLA-identical transplants from related donors were T cell depleted (12, 55, 57). Thirty-three patients received from one to three additional T cell–depleted marrow transplants, from either the original donor or another haploidentical relative. None of the marrow recipients received any pretransplant conditioning or post-transplant prophylaxis against GVHD. Two patients were given cyclosporine for 1 month because of cutaneous GVHD from transplacentally transferred maternal T cells at presentation. Five of the infants who received haploidentical marrow transplants also received unrelated placental blood transplants. Four of the latter received pretransplant conditioning and were also given post-transplant prophylaxis against GVHD.

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PATIENTS

Results of Bone Marrow Transplantation at Duke FACTORS INFLUENCING SURVIVAL Of the 132 SCID patients, 102 (77%) are alive. None show any evidence of susceptibility to opportunistic infections and most are in good general health. The follow-up ranges from 2 months to 21.3 years after transplantation. Of these 102 patients, 96 have survived 1 or more years after transplantation, 68 have been alive 5 or more years, and 37 for 10 or more years. Median follow-up of surviving patients is 5.4 years. All 15 recipients of marrow from HLA-identical donors, 87 of the 117 recipients given T cell–depleted haploidentical bone marrow from a related donor, and 2 of 5 infants from the latter group also given unrelated placental blood transplants are among the survivors. The survival rates are similar for the different genetic types of SCID, except that only 1

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

637

Figure 6 Survival following bone marrow transplantation of 102 of 132 SCIDs according to genetic type of SCID.

of the small group of 4 male infants with SCID of unknown type survive (Figure 6). Influences on survival include race (more who were Caucasian survive, p < 0.001), sex (all but three of the females transplanted survive, p < 0.05), and age at the time of transplant. Of the 36 infants transplanted during the first 3.5 months of life, 35 (97%) survive, compared to 67/96 (70%) who were transplanted after that age (Figure 7). Twenty-four of the 30 deaths occurred from viral infections: CMV, 7; EBV, 3; adenovirus, 7; enteroviruses, 2; parainfluenza 3, 2; Herpes zoster, 2; and Herpes simplex, 1 (Table 2). Two infants died of candida sepsis. One died from an unrelated mitochondrial defect after stable marrow engraftment, and another died of the nephrotic syndrome after chemotherapy had been given for what had been mistakenly diagnosed elsewhere as a malignancy. One ADA-deficient patient who had been treated with polyethylene glycol modified bovine ADA for 18 years died of pulmonary hypertension. None of the patients died from GVHD, despite the fact that 87 of the surviving patients received haploidentical bone marrow transplants. GRAFT-VERSUS-HOST DISEASE GVHD occurred in 40/117 patients given T cell– depleted haploidentical parental marrow, 6/15 given unfractionated HLA-identical marrow, and 4/5 given placental blood. In 35/49 cases, this complication occurred when there was persistence of transplacentally transferred maternal T cells. Most of the GVHD after administration of T cell–depleted marrow was mild (Grade I or II) and required no treatment (70). Ten patients from the entire group had Grade III

12 Feb 2004

15:56

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

638

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

Figure 7 Kaplan Meier plot of 36 SCID infants transplanted in the first 3.5 months of life. Thirty-five survive from 3 months to 21.3 years posttransplantation; only 5 had HLA-identical donors.

GVHD involving the skin, gastrointestinal tract, and/or marrow; 9/10 were given corticosteroids and cyclosporine as treatment; one was given corticosteroids alone and two were given tacrolimus. One neonate with autosomal recessive SCID who received a 3-antigen mismatched transplant developed Grade 4 GVHD with autoimmune hemolytic anemia, bone marrow suppression, diarrhea, and cholestatic liver disease. He subsequently received a living-related liver transplant and remains on a low dose of tacrolimus 1.5 years later. His mother was the donor for both the bone marrow and the liver segment. He is a complete hematopoietic chimera and now has normal liver function. No patients died of GVHD, but one of the recipients of unrelated placental blood had severe acute GVHD and now has chronic GVHD, requiring continuous cyclosporine therapy. ENGRAFTMENT AND CHIMERISM Genetic analyses of blood lymphocytes at the survivors’ latest evaluations have shown that all of the T cells in 95/102 are of donor origin. In one γ c-deficient patient, four T cell–depleted transplants—two from each parent—failed to engraft, but the patient is still alive at 10 years of age despite very low T cell function. Of note, 13/18 ADA-deficient patients given T cell–depleted haploidentical marrow and all 4 of those given HLA identical marrow, all without pretransplant chemotherapy or post-transplant GVHD prophylaxis, are alive at 0.3–18.6 years after transplantation, with hematopoietic chimerism in 10. Two of the 6 infants who failed to become chimeric received successful gene therapy in Italy (71), and the other 5 are receiving polyethylene

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

639

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

glycol-modified bovine ADA after rejecting haploidentical parental hematopoietic stem cell transplants. By contrast to the development of donor T cells in all but the above 8 patients, the B cells in most cases remain those of the recipient. However, 5/15 recipients of HLA-identical marrow and 32/117 haploidentical marrow recipients have some donor B cells (range 3%–100% of all B cells, mean 55.8 +/− SEM 8.3%). LYMPHOCYTE PHENOTYPES Patients with the various genetic types of SCID had distinct lymphocyte phenotypes before transplantation (Figure 2a) (11, 12). All patients had a profound deficiency of T cells and, when T cells were present, they were usually transplacentally transferred maternal T cells. In one Jak3-deficient patient, there were 8268 circulating maternal T cells/mm3 at presentation (72). B cells were elevated in γ c-deficient, IL-7Rα-deficient, and Jak3-deficient SCIDs, normal in autosomal recessive SCIDs of unknown molecular type, and absent in RAG1- and RAG2-deficient SCIDs. Numbers of NK cells were lowest in γ cdeficient, Jak3-deficient (T−B+NK−) SCIDs, and in ADA-deficient (T−B−NK−) SCIDs, but were normal in IL-7Rα-deficient (T−B+NK+) SCIDs, RAG1- or RAG2deficient SCIDs (T−B−NK+), in other autosomal recessive SCIDs of unknown molecular type, and in the four males with unknown inheritance (Figure 2a) (11, 12). At the most recent evaluation following transplantation, the mean number of T cells in the 102 surviving patients was within the normal range for the γ cdeficient, ADA-deficient, IL-7Rα-deficient, and Jak3-deficient SCIDs and below the normal range in the RAG1- or RAG2-deficient SCIDs and in the autosomal recessive SCIDs of unknown molecular cause (Figure 2b). The mean number of B cells was still elevated in γ c-deficient and Jak3-deficient SCIDs, very low in the RAG1- or RAG2-deficient SCIDs, but normal in all others. The mean number of NK cells remained low in the γ c-deficient and Jak3-deficient groups but was normal in the others. T CELL FUNCTION Figure 3 shows in vitro responses to the mitogens (phytohemagglutinin, concanavalin A, and pokeweed mitogen) by T cells from patients with the various types of SCID, before (3a) and after (3b) transplantation, as compared to such responses by T cells from normal adults. Remarkably, mean responses to all three mitogens were normal in all groups after transplantation as compared with extremely low responses before transplantation. Moreover, T cells from all patients with SCID responded poorly to allogeneic cells before transplantation; however, after transplantation T cells from all groups responded normally to allogeneic cells, candida, and tetanus antigens (not shown). ROLE OF THE THYMUS Because very few T cells were present in the T cell– depleted donor marrow, it was initially surprising that genetically donor T cells emerged in these SCID infants at 90–120 days post-transplantation, considering that they all had vestigial thymi and no T cells prior to transplantation. To clarify whether the donor stem cells actually developed into T cells within the infants’

12 Feb 2004

15:56

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

640

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

thymuses, we analyzed signal joint T cell–receptor recombination excision circles (TRECs) in 83 of the patients with SCID given marrow transplants (15). We found that T cell numbers were low before and early after transplantation, with a predominance of CD45RO+ T cells (primarily due to transplacentally transferred maternal cells), and TRECs were undetectable in blood mononuclear cells. A majority of infants given either T cell–depleted marrow or unfractionated HLAidentical marrow developed genetically donor TREC+, CD45RA+, CD62L+ T cells (recent thymic emigrants) by 3–6 months post-transplantation. Thymic output peaked (5525 ± 1502 TRECs/µg PBMC DNA) 1–2 years after transplantation and declined to low (<100/µg DNA) levels over 14 years, compared to over 80 years in normal individuals. CD45RA+ T cell numbers peaked at year 1 after transplantation (1513 ± 324 cells/µl blood), but declined to 107 ± 35 cells/µl at 14 years after transplantation. From these studies, we concluded that the vestigial thymus of infants with SCID can produce enough genetically donor T cells to confer T cell function by 3–6 months after transplantation, regardless of whether the donor is HLA-identical or haploidentical (15). T CELL DIVERSITY Immunoscope analysis of the TCR Vβ repertoire was performed on 15 SCID patients given bone marrow transplants (BMT) (73). Before and within the first 100 days after BMT, patients’ PBMC displayed an oligoclonal or skewed T cell repertoire, low TREC values, and a predominance of CD45RO+ T cells. In contrast, the presence of high numbers of CD45RA+ cells in the circulation of SCID patients >100 days post-BMT correlated with active T cell output by the thymus, as revealed by high TREC values, and a polyclonal T cell repertoire demonstrated by a Gaussian distribution of Vβ-specific peaks (Figure 8). Ten years after BMT we observed a decrease of the normal polyclonal T cell repertoire and an increase of a more skewed T cell repertoire. A decline of TREC levels and a decrease in the number of CD45RA+ cells beyond 10 years of BMT was concomitant with the detection of oligoclonal CD3+CD8+CD45RO+ cells. The switch from a polyclonal to a more skewed repertoire, observed in the CD3+CD8+CD45RO+ T cell subset, is a phenomenon that occurs normally with decreased thymic output during aging, but not as rapidly as in this patient population. We conclude that a normal T cell repertoire develops in SCID patients as a result of thymic output, and the repertoire remains highly diverse for the first 10 years after BMT. The TCR diversity positively correlated with TREC levels (73).

−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→ Figure 8 Immunoscope profile of TCR Vβ families. Each Vβ family was examined by PCR amplification and run-off reaction. Results are shown for each Vβ family as a density peak histogram. CDR3 sizes are shown on the x-axis and the peak fluorescence intensity is shown on the y-axis. (A) Immunoscope profile of a normal subject. (B) Immunoscope profile of a Jak-3 deficient SCID patient (J-1) before BMT. (C) Immunoscope profile of TCR Vβ families in an X-SCID patient at 218 days and (D) at 1984 days after BMT. [Figure reprinted with permission from (73).]

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

12 Feb 2004 15:56 AR AR210-IY22-21.tex AR210-IY22-21.sgm LaTeX2e(2002/01/18)

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

641

12 Feb 2004

15:56

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

642

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY B CELL FUNCTION B cell function did not develop to the extent that T cell function did. Serum IgG prior to transplantation was in most cases maternal or from intravenous immunoglobulin, but paraproteins were present in some. One of the IL-7Rα-deficient patients had both IgG and IgA paraproteins prior to transplantation, as has been noted previously in SCID (74–76). At their latest evaluation, 58 patients have normal serum IgA, 80 have normal IgM, and 48 have isohemagglutinins appropriate for host red blood cell type. Sixty-two of the 102 (61%) patients are currently receiving immunoglobulin replacement to prevent bacterial and common viral infections. All patients who are not receiving immunoglobulin infusions have demonstrated the capacity to produce antibodies to one or more vaccine antigens (data not shown). The best B cell function is in the IL-7Rα-deficient SCIDs and in the ADA-deficient SCIDs, despite the fact that most of the children in those groups do not have donor B cells. This indicates that the molecular type of SCID is the principal determinant as to whether good B cell function will develop. NK CELL FUNCTION Before engraftment, NK cell numbers and function were lowest in γ c-deficient and Jak3-deficient SCIDs (p < 0.001) (Figure 2a), whereas they were higher than normal in all other types except ADA deficiency. Following transplantation, many γ c- and Jak3-deficient SCIDs continued to have low NK function, whereas this function was normal in all other types. The NK cells in the IL-7Rα-deficient SCID patients did not interfere with engraftment of T cell– depleted parental bone marrow stem cells despite no pretransplant chemotherapy. The NK cells in the RAG-and Artemis-deficient patients may have been responsible for the longer time course to development of T cell function but did not prevent engraftment even though no chemotherapy was given. TRANSPLANTS PERFORMED IN THE FIRST 3.5 MONTHS OF LIFE As already noted, survival of the SCID patients who were transplanted in the first 3.5 months of life was superior to that in those performed later (Figure 7). We hypothesized that the kinetics of immune reconstitution would be different for those transplanted in the neonatal period when compared to those transplanted after that time. We compared immune function in 21 SCID patients transplanted in the neonatal period with that in 70 SCIDs transplanted after that (13). We measured lymphocyte phenotypes, proliferative responses to mitogens, immunoglobulin levels, and Tcell antigen receptor excision circles (TRECs) pretransplantation and sequentially post-transplantation. Infants transplanted in the newborn period developed higher lymphocyte responses to phytohemagglutinin and higher numbers of CD3+ and CD45RA+ T cells in the first 3 years of life than those transplanted late (p < 0.05) (Figures 9 and 10). TRECs peaked earlier and with higher values (p < 0.01) in the neonatal transplants (181 days–1 year) than in the late transplants (1–3 years) (Figure 11). Thus, SCID recipients of allogeneic, related hematopoietic stem cells in the neonatal period had higher levels of T cell reconstitution and thymic output and a higher survival rate than those transplanted after 28 days of life (13).

15:56 AR AR210-IY22-21.tex AR210-IY22-21.sgm

Figure 9 Increased numbers of circulating na¨ıve T cells in the early (newborn) versus late transplantation groups. Shown are the mean (+/−SEM) numbers of CD3+, CD45RA+, and CD45RO+ cells in patients receiving transplants early (n = 20) compared with those receiving transplants late (n = 66). The early group had increased numbers of CD3+ cells at 271 days to 1 year, 1–2 years, and 2–3 years after transplantation (P < .05). These numbers gradually declined and were comparable to the late group by 6 years after transplantation. [Figure reprinted with permission from (13).]

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

12 Feb 2004 LaTeX2e(2002/01/18)

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

643

12 Feb 2004

15:56

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

644

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

Figure 10 Comparison of proliferative responses to PHA in the early and late transplantation groups. Shown are the mean (+/−SEM) counts per minute of [3H] thymidine incorporation. Infants receiving transplants within the first 28 days of life (n = 20) had increased T cell proliferation to PHA at 91 through 120 days, 121 through 180 days, and 181 through 270 days after transplantation compared with those receiving transplants late (n = 69) (P < .05); n = total number of individuals analyzed in each group over 19.2 years. [Figure reprinted with permission from (13).]

Figure 11 Faster and quantitatively higher thymic output in the early- versus latetransplantation groups. Shown are the mean (+/−SEM) number of TRECs for patients receiving early (n = 19) and late (n = 55) transplantations. Patients receiving transplants early had higher TREC values at 91 through 180 days and 181 days to 1 year after transplantation (P < .01). The mean TREC value peaked at 181 days to 1 year in those receiving transplants early and at 1 to 3 years in those receiving transplants late. [Figure reprinted with permission from (13).]

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

645

BOOSTER TRANSPLANTS In an attempt to overcome poor B or T cell function or resistance to engraftment, booster transplants were performed in 33 (25%) of the 132 patients, and 21 (64%) of them survived. All of these were done without pretransplant chemotherapy and with the same T cell depletion method as used for the initial transplants. Twenty-two patients received booster transplants from the same parental donor; and eight of them died of opportunistic viral infections. Nine received booster transplants from the other parent; four of those died. One patient received a blood transfusion from an identical twin SCID donor who had accepted a marrow transplant from their father, and another received a cord blood transplant. Immune function improved in all but four of the survivors who received booster transplants.

Overview The above studies clearly demonstrate that transplantation of T cell–depleted HLAidentical or HLA-haploidentical bone marrow is highly effective in reconstituting T cell immunity in all of the known genetic types of SCID. No chemotherapeutic conditioning is required to achieve engraftment because the recipient is virtually devoid of T cells at the time of transplantation. This advantage eliminates adverse effects caused by these toxic agents, including neutropenia, red cell and platelet transfusion-dependency, mucositis, veno-occlusive disease, busulfan lung disease, growth suppression, sterility, and a 15% risk of later malignancy (77). Although GVHD prophylaxis was not used except for 1 month of cyclosporine given to two infants who presented with GVHD and for the placental blood transplants, clinically significant GVHD was seldom seen. The omission of GVHD prophylaxis with cyclosporine permitted the infants to develop T cell function without hindrance. Normal T cell function appeared within 2 weeks after transplantation of unfractionated HLA-identical marrow because of the adoptive transfer of mature donor T cells. By contrast, it did not develop until 3–4 months after administration of T cell–depleted marrow, whether HLA-identical or haploidentical (12, 57). The latter is the average time required for the donor stem cells to become phenotypically and functionally mature T cells in the recipient (12, 57). T cell function often developed much earlier in neonatal recipients (13) and in patients in whom transplacental transfer of maternal T cells had occurred, but it developed later in some patients who had high numbers of NK cells at presentation. In the case of the unrelated placental blood transplants, T cells were present immediately, but T cell function was suppressed by the large doses of corticosteroids and cyclosporine needed to prevent or treat GVHD. As for B cells, only 6/12 survivors of HLA-identical and 21/76 survivors of haploidentical transplants have some donor B cells (2%–100% of total B cells); 62/102 are currently receiving immunoglobulin replacement therapy to prevent bacterial and common viral infections because the capacity to produce protective antibodies has not yet been demonstrated. However, several of these 62 patients may be able

12 Feb 2004

15:56

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

646

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

to discontinue immunoglobulin treatment because they are now producing IgA and isohemagglutinins. Recent progress in identifying the molecular causes of SCID permitted us to study SCID mutations in relation to the outcome of hematopoietic stem cell transplantation. Most of the γ c-deficient and Jak3-deficient patients who do not have any evidence of donor-derived B cells continue to have poor B cell function, as demonstrated by failure of isotype-switching following immunization with bacteriophage φX174 and the inability to produce IgA, IgM, and IgE or isohemagglutinins normally in vivo. Thus, normal stem cells that mature in γ c- and Jak3-deficient patients develop into normal T cells, but much less frequently into normal B cells; the host B cells in these patients most likely fail to function because they lack normal cytokine receptors. By contrast, a majority of the ADA- and IL-7Rα-deficient patients and those with autosomal recessive SCID of unknown molecular cause have good host B cell function, indicating that those mutations do not adversely affect B cell function. Before transplantation, NK cell numbers and function were also lowest in γ cand Jak3-deficient patients, whereas they were higher than normal in most other types of SCID. Following transplantation, profoundly low NK cell numbers and function persisted in most γ c- and Jak3-deficient patients, whereas these were normal in all other types of SCID. Although the ability to give half-matched T cell–depleted parental marrow to patients with SCID has been a remarkable therapeutic advance, it is not a perfect treatment. During the 3–4 months needed for donor stem cells to develop into mature functioning T cells, the infant is susceptible to viral infections. Pretransplant chemotherapy fails to accelerate immune reconstitution, heightens the susceptibility of the recipient to infection, and necessitates the use of cyclosporine, which prolongs the T cell deficiency (78). The poor B cell function in γ c-deficient and Jak3-deficient patients in whom donor B cells do not develop has led some to use pretransplant conditioning. However, chemotherapy does not guarantee that donor B cells will develop, and the risks outweigh the potential for development of B cell function. Indeed, in a study from Europe, there was no better B cell function in those who received pretransplant chemotherapy than in those who did not (78). Finally, resistance to engraftment is a problem that was overcome in all but six cases by booster or second parent T cell–depleted transplants, again without conditioning. Placental blood transplantation from unrelated donors is fraught with problems because of GVHD in patients with SCID. In most institutions performing placental blood transplants for SCID, pretransplant chemotherapy is given, and prolonged (6 months) GVHD prophylaxis with cyclosporine is required (79–81). All of this heightens the risk of infection. In utero stem cell transplants from related donors do not appear to offer any advantage over such transplants done soon after birth. The mother would probably not be used as a donor for an in utero transplant because of the risks of anesthesia during pregnancy. The invasive procedures required in

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

647

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

in utero stem cell administration carry risks, and one would also not be able to detect or treat either a graft-versus-graft reaction or GVHD in utero (82, 83). SCID is a pediatric emergency, and the potential exists to diagnose this condition routinely at birth (11). Cord blood white cell counts and manual differentials can detect the lymphopenia that is almost invariably present, and appropriate immunologic tests could then be done. Prenatal diagnosis can often be made when there is a family history of SCID. If a stem cell transplant from a relative can be done in the first 3.5 months of life, before infections develop, there is a high probability (>95%) of success.

Results of Bone Marrow Transplantation at Other Centers Most of the published results of bone marrow transplantation for SCID at other centers have come from the European Group for Blood and Marrow Transplantation and the European Society for Immunodeficiency (14, 78, 84). The transplants were done at 37 different centers in Europe and, therefore, the approaches were not all the same. In a report of a long-term retrospective study of immune reconstitution in 193 SCIDs transplanted in Europe with T cell–depleted HLA-nonidentical marrow between January 1, 1983 and December 31, 1993, the overall survival at the time of publication in 1998 was 92 patients, or 47%. Eighty-nine of the 116 (77%) patients who survived for at least 6 months had been given pretransplant chemotherapy. Seventy-seven of the deaths occurred within 6 months after transplantation, and 24 more occurred after 6 months (78). A more recent publication from that group covered 1082 transplants in 919 immunodeficient patients in Europe during the 32 year period from January 1968 through December of 1999 (14). Of the 475 SCID patients transplanted, pretransplant conditioning was given to 275 or 57% of the patients. Three-year survival rates with sustained engraftment were better for HLA-identical (77%) than for mismatched transplants (54%). There was significant overall improvement in the outcomes of SCID transplants over time; the improvement went from approximately 40% survival of mismatched transplants performed from 1968–1985 to approximately 78% in those performed from 1996–1999. The latter was attributed to the development of rigorous T cell– depletion techniques, more effective antibiotics, and perhaps earlier recognition of SCID. Another finding was that the outcome was significantly better in the B cell+ SCIDs than in the B cell− SCIDS (14, 84). The population reported was not characterized at a molecular level. In the United States, 16 infants with Athabascan SCID due to Artemis mutations were given bone marrow transplants at the University of California in San Francisco between 1984 and 1999 (85). Seven of them received HLA-identical sibling marrow, and nine received T cell–depleted parental marrow. All but two of the infants received pretransplant chemotherapy. All seven of those who received HLA-identical marrow survived, whereas only five of the nine who received parental marrow survived. Three of the four children who died received radiation or busulfan, and two of the eight long-term survivors who were also

12 Feb 2004

15:56

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

648

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

recipients of cytototoxic chemotherapy failed to develop secondary teeth. Thus, children with this radiation-sensitive form of SCID had a poor outcome if they were given pretransplant chemotherapy (85). At Children’s Hospital of Los Angeles, between November of 1984 and December of 1997, 48 SCID infants were given bone marrow transplants (86). Eleven received HLA-identical related bone marrow transplants, and 37 received T cell–depleted haploidentical parental marrow. All recipients received pretransplant conditioning except one who received HLAidentical sibling marrow. At the time of the report, all of the 11 who received HLA-identical marrow were surviving, but only 17 of the 37 (47%) who received T cell–depleted parental marrow were surviving (86). Although the approaches used to attempt immune reconstitution by bone marrow transplantation in SCID infants have differed from center to center, much has been learned about what factors influence success or failure. In general, the mortality has been much higher at centers that use pretransplant chemotherapy.

Gene Therapy Until the past year, there was great optimism that primary immunodeficiency diseases for which the molecular defects have been identified could be corrected by gene therapy. From 1999–2002, 11 patients with SCID-X1 were administered autologous bone marrow cells into which a normal γ c cDNA had been successfully transduced by retroviral gene transfer (87, 88). In 9 of the 11 patients, molecular studies demonstrated normal transgene expression in circulating T and NK cells by approximately 30–40 days after gene therapy, but it was minimal in B cells. Two of the patients did not express the transgene and were given allogeneic bonemarrow transplants. The nine infants with transgene expression developed normal T cell function at between 90 and 120 days after the treatments, similar to the kinetics after T cell–depleted allografts. Despite the fact that the transgene was minimally expressed in B cells, the nine who developed normal T cell function also did not require intravenous immunoglobulin infusions and were at home off of all medications. Thus, the efficacy of gene therapy in conferring immune function in those infants with SCID-X1 seemed to be far superior to that of allogeneic marrow stem-cell transplantation. Tragically, however, serious adverse events occurred in the fourth and fifth patients treated by the French group (88a). Both children developed leukaemic-like processes, with expanded clonal populations of T cells. The clones carry the inserted γ c cDNA, and the leukemias are considered to have been induced by the retroviral gene therapy by a process called insertional oncogenesis. The positions of insertion in both children are in or near a gene on chromosome 11 called LMO-2. The product of LMO-2 is crucial for normal hematopoiesis and serves a regulatory function (89). However, LMO-2 is also an oncogene that is aberrantly expressed in acute lymphoblastic leukemia of childhood. Both children were treated with chemotherapy and responded to it, but there was a recurrence in one, who was then given a matched unrelated donor transplant after chemoablation, thus destroying

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

P1: IKH

649

the gene corrected cells. The other patients who were given gene therapy are being monitored closely and one has been found to have a similar insertion near the LMO-2 gene but has not yet developed a clonal proliferation. In view of these serious adverse events, retroviral gene therapy trials are currently on hold. Because of the above events, the treatment of genetically determined immunodeficiency disorders remains a problem, with allogeneic stem-cell transplantation seeming to be the current best option for those defects that are invariably fatal early in life (12, 14). Efforts are being made to improve this therapy by giving higher numbers of affinity-purified allogeneic stem cells in preparations nearly devoid of T cells (90). If the imperfect results seen with allogeneic stem-cell therapy in the past were due to an insufficient number of stem cells, this approach should result in better immune reconstitution. The fact that such cell suspensions are virtually devoid of T cells should also circumvent the problem of GVHD (90). The only remaining obstacle would then be to ensure that diagnosis is made early before untreatable infections develop. However, this obstacle remains formidable because there is no currently no screening for any primary immunodeficiency disease at birth or during childhood or adulthood in any country. Thus, most patients are not diagnosed until they develop a serious infection, which will certainly adversely affect the ultimate outcome of definitive therapy (11, 12). In summary, T cell–depleted haplo-identical marrow transplantation provides life-saving therapy for all forms of SCID. The remaining prospect of gene therapy offers hope that the remaining defects in these chimeras will eventually be correctable by that means. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Glanzmann E, Riniker P. 1950. Essentielle lymphocytophtose. Ein neues krankeitsbild aus der Sauglingspathologie. Ann. Paediatr. 174:1–5 2. Giblett ER, Anderson JE, Cohen I. 1972. Adenosine deaminase deficiency in two patients with severely impaired cellular immunity. Lancet 2:1067–70 3. Noguchi M, Yi H, Rosenblatt HM, Filipovich AH, Adelstein S, et al. 1993. Interleukin-2 receptor gamma chain mutation results in X-linked severe combined immunodeficiency in humans. Cell 73:147–57 4. Puck JM, Deschenes SM, Porter JC, Dutra AS, Brown CJ, et al. 1993. The interleukin-2 receptor gamma chain

maps to Xq13.1 and is mutated in X-linked severe combined immunodeficiency, SCIDX1. Hum. Mol. Genet. 2: 1099–104 5. Russell SM, Tayebi N, Nakajima H, Riedy MC, Roberts JL, et al. 1995. Mutation of Jak3 in a patient with SCID: Essential role of Jak3 in lymphoid development. Science 270:797–800 6. Puel A, Ziegler SF, Buckley RH, Leonard WJ. 1998. Defective IL7R expression in T(−)B(+)NK(+) severe combined immunodeficiency. Nat. Genet. 20:394–97 7. Schwarz K, Gauss GH, Ludwig L, Pannicke U, Li Z, et al. 1996. RAG mutations in human B cell-negative SCID. Science 274:97–99

12 Feb 2004

15:56

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

650

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

8. Moshous D, Callebaut I, de Chasseval R, Corneo B, Cavazzana-Calvo M, et al. 2001. Artemis, a novel DNA double-strand break repair/V(D)J recombination protein, is mutated in human severe combined immune deficiency. Cell 105:177–86 9. Kung C, Pingel JT, Heikinheimo M, Klemola T, Varkila K, et al. 2000. Mutations in the tyrosine phosphatase CD45 gene in a child with severe combined immunodeficiency disease. Nat. Med. 6:343–45 9a. Dadi H, Simon A, Roifman C. 2003. Effect of CD3δ deficiency on maturation of α/β and γ /δ T-cell lineages in severe combined immunodeficiency. N. Engl. J. Med. 349:1821–28 10. Stephan JL, Vlekova V, Le Deist F, Blanche S, Donadieu J, et al. 1993. Severe combined immunodeficiency: a retrospective single-center study of clinical presentation and outcome in 117 cases. J. Pediatr. 123:564–72 11. Buckley RH, Schiff RI, Schiff SE, Markert ML, Williams LW, et al. 1997. Human severe combined immunodeficiency (SCID): Genetic, phenotypic and functional diversity in 108 infants. J. Pediatr. 130:378–87 12. Buckley RH, Schiff SE, Schiff RI, Markert ML, Williams LW, et al. 1999. Hematopoietic stem cell transplantation for the treatment of severe combined immunodeficiency. N. Engl. J. Med. 340: 508–16 13. Myers LA, Patel DD, Puck JM, Buckley RH. 2002. Hematopoietic stem cell transplantation for severe combined immunodeficiency in the neonatal period leads to superior thymic output and improved survival. Blood 99:872–78 14. Antoine C, Muller S, Cant A, CavazzanaCalvo M, Veys P, et al. 2003. Long-term survival and transplantation of haemopoietic stem cells for immunodeficiencies: report of the European experience 1968–99. Lancet 361:553–60 15. Patel DD, Gooding ME, Parrott RE, Curtis KM, Haynes BF, Buckley RH.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

2000. Thymic function after hematopoietic stem-cell transplantation for the treatment of severe combined immunodeficiency. N. Engl. J Med. 342:1325–32 Fischer A, Malissen B. 1998. Natural and engineered disorders of lymphocyte development. Science 280:237–53 Altman PL. 1961. Blood leukocyte values: man. In Blood and Other Body Fluids, ed. DS Dittmer, pp. 125–26. Washington, DC: Fed. Am. Soc. Exp. Biol. Hirschhorn R. 1995. Adenosine deaminase deficiency: molecular basis and recent developments. Clin. Immunol. Immunopathol. 76:S219–27 Noguchi M, Nakamura Y, Russell SM, Ziegler SF, Tsang M, et al. 1993. Interleukin-2 receptor gamma chain: a functional component of the interleukin-7 receptor. Science 262:1877–80 Valerio D, McIvor RS, Williams SR, Duyvesteyn MG, van Ormondt H Jr, et al. 1984. Cloning of human adenosine deaminase cDNA and expression in mouse cells. Gene 31:147–53 Shovlin CL, Simmonds HA, Fairbanks LD, Deacock SJ, Hughes JMB, et al. 1994. Adult onset immunodeficiency caused by inherited adenosine deaminase deficiency. J. Immunol. 153:2331–39 Sugamura K, Asao H, Kondo M, Tanaka N, Ishii N, et al. 1996. The interleukin-2 receptor gamma chain: its role in the multiple cytokine receptor complexes and T cell development in XSCID. Annu. Rev. Immunol. 14:179–205 Asao H, Okuyama C, Kumaki S, Ishii N, Tsuchiya S, et al. 2001. Cutting edge: the common gamma-chain is an indispensable subunit of the IL-21 receptor complex. J. Immunol. 167:1–5 Puck JM, Pepper AE, Henthorn PS, Candotti F, Isakov J, et al. 1997. Mutation analysis of IL2RG in human X-linked severe combined immunodeficiency. Blood 89:1968–77 Russell SM, Keegan AD, Harada N, Nakamura Y, Noguchi M, et al. 1993.

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

26.

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

27.

27a.

28.

29.

30.

31.

31a.

Interleukin-2 receptor gamma chain: a functional component of the interleukin-4 receptor. Science 262:1880–83 Kokron CM, Bonilla FA, Oettgen HC, Ramesh N, Geha RS, Pandolfi F. 1997. Searching for genes involved in the pathogenesis of primary immunodeficiency diseases: lessons from mouse knockouts. J. Clin. Immunol. 17:109–26 Stephan V, Wahn V, Le Deist F, Dirksen U, Broker B, et al. 1996. Atypical X-linked severe combined immunodeficiency due to possible spontaneous reversion of the genetic defect in T cells. N. Engl. J. Med. 335:1563–67 Buckley RH. 2002. Primary cellular immunodeficiencies. J. Allerg. Clin. Immunol. 109:747–757 Kawamura M, McVicar DW, Johnston JA, Blake TB, Chem Y, et al. 1994. Molecular cloning of L-Jak, a Janus family protein tyrosine kinase expressed in natural killer cells and activated leukocytes. Proc. Natl. Acad. Sci.USA 91:6374–78 Macchi P, Villa A, Gillani S, Sacco MG, Frattini A, et al. 1995. Mutations of Jak-3 gene in patients with autosomal severe combined immune deficiency (SCID). Nature 377:65–68 Bozzi F, Lefranc G, Villa A, Badolato R, Schumacher RF, et al. 1998. Molecular and biochemical characterization of JAK3 deficiency in a patient with severe combined immunodeficiency over 20 years after bone marrow transplantation: implications for treatment. Br. J. Haematol. 102:1363–66 Notarangelo LD, Mella P, Jones A, de Saint Basile G, Savoldi G, et al. 2001. Mutations in severe combined immune deficiency (SCID) due to JAK3 deficiency. Hum. Mutat. 18:255–63 Roberts J, Lengi A, Brown S, Chen M, Zhou Y-J, et al. 2004. Janus Kinase 3 (Jak3) deficiency: clinical, immunologic and molecular analyses of 10 patients and outcomes of stem cell transplantation. Blood. In press

P1: IKH

651

32. Ozaki K, Spolski R, Feng CG, Qi CF, Cheng J, et al. 2002. A critical role for IL-21 in regulating immunoglobulin production. Science 298:1630–34 33. Kennedy MK, Glaccum M, Brown SN, Butz EA, Viney JL, et al. 2000. Reversible defects in natural killer and memory CD8 T cell lineages in interleukin 15-deficient mice. J. Exp. Med. 191:771–80 34. Peschon JJ, Morrissey PJ, Grabstein KH, Ramsdell FJ, Maraskovsky E, et al. 1994. Early lymphocyte expansion is severely impaired in interleukin 7 receptor deficient mice. J. Exp. Med. 180:1955–60 35. Vonfreedenjeffry U, Burdach S, Murray R. 1995. Severe lymphopenia in interleukin-7 deficient mice show that interleukin-7 is a nonredundant cytokine. Exp. Hematol. 23:892 36. Roifman CM, Zhang JY, Chitayat D, Sharfe N. 2000. A partial deficiency of interleukin-7R alpha is sufficient to abrogate T-cell development and cause severe combined immunodeficiency. Blood 96:2803–7 37. Leonard WJ. 2001. Cytokines and immunodeficiency diseases. Nat. Rev. Immunol. 1:200–8 38. Schwarz K, Notarangelo L, Spanopoulou E, Vezzoni P, Villa A. 1999. Recombination defects. See Ref. 91, pp. 155–66 39. Corneo B, Moshous D, Gungor T, Wulffraat N, Philippet P, et al. 2001. Identical mutations in RAG1 or RAG2 genes leading to defective V(D)J recombinase activity can cause either T-B-severe combined immune deficiency or Omenn syndrome. Blood 97:2772–76 40. Villa A, Santagata S, Bozzi F, Giliani S, Frattini A, et al. 1998. Partial V(D)J recombination activity leads to Omenn syndrome. Cell 93:885–96 41. Rieux-Laucat F, Bahadoran P, Brousse N, Selz F, Fischer A, et al. 1998. Highly restricted human T cell repertoire in peripheral blood and tissue-infiltrating lymphocytes in Omenn’s syndrome. J. Clin. Invest. 102:312–21

12 Feb 2004

15:56

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

652

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY

42. Brooks EG, Filipovich AH, Padgett JW, Mamlock R, Goldblum RM. 1999. T-cell receptor analysis in Omenn’s syndrome: evidence for defects in gene rearrangement and assembly. Blood 93:242–50 43. Martin JV, Willoughby PB, Giusti V, Price G, Cerezo L. 1995. The lymph node pathology of Omenn’s syndrome. Am. J. Surg. Pathol. 19:1082–87 43a. Soudais C, De Villartay JP, Le Deist F, Fischer A, Lisowska-Grospierre B. 1993. Independent mutations of the human CD3-epsilon gene resulting in a T cell receptor/CD3 complex immunodeficiency. Nat. Genet. 3:77–81 43b. Arnaiz-Villena A, Timon M, Corell A, Perez-Aciego P, Martin-Villa JM, Regueiro JR. 1992. Brief report: primary immunodeficiency caused by mutations in the gene encoding the CD3-γ subunit of the T lymphocyte receptor. N. Engl. J. Med. 327:529–33 44. Cale CM, Klein NJ, Novelli V, Veys P, Jones AM, Morgan G. 1997. Severe combined immunodeficiency with abnormalities in expression of the common leucocyte antigen, CD45. Arch. Dis. Child 76:163–64 45. Tchilian EZ, Wallace DL, Wells RS, Flower DR, Morgan G, Beverley PC. 2001. A deletion in the gene encoding the CD45 antigen in a patient with SCID. J. Immunol. 166:1308–13 46. Hermiston ML, Xu Z, Weiss A. 2003. CD45: A critical regulator of signaling thresholds in immune cells. Annu. Rev. Immunol. 21:107–37 47. Nicolas N, Moshous D, Cavazzana-Calvo M, Papadopoulo D, de Chasseval R, et al. 1998. A human severe combined immunodeficiency (SCID) condition with increased sensitivity to ionizing radiations and impaired V(D)J rearrangements defines a new DNA recombination/repair deficiency. J. Exp. Med. 188:627–34 48. Li L, Moshous D, Zhou Y, Wang J, Xie G, et al. 2002. A founder mutation in Artemis, an SNM1-like protein, causes

49. 50.

51.

52.

53.

54.

55.

56.

57.

SCID in Athabascan-speaking Native Americans. J. Immunol. 168:6323–29 Dausset J. 1958. Iso-leuko anticorps. Acta Haematol. 20:156 Amos DB, Bach FH. 1968. Phenotypic expressions of the major histocompatibility locus in man (HL-A): leukocyte antigens and mixed leukocyte culture reactivity. J. Exp. Med. 128:623–37 Gatti RA, Meuwissen HJ, Allen HD, Hong R, Good RA. 1968. Immunological reconstitution of sex-linked lymphopenic immunological deficiency. Lancet 2:1366– 69 Bortin MM, Rimm AA. 1977. Severe combined immunodeficiency disease. Characterization of the disease and results of transplantation. JAMA 238:591– 600 Muller-Ruchholtz W, Wottge HU, MullerHermelink HK. 1976. Bone marrow transplantation in rats across strong histocompatibility barriers by selective elimination of lymphoid cells in donor marrow. Transplant Proc. 8:537–41 Reisner Y, Itzicovitch L, Meshorer A, Sharon N. 1978. Hematopoietic stem cell transplantation using mouse bone marrow and spleen cells fractionated by lectins. Proc. Natl. Acad. Sci. USA 75:2933–36 Reisner Y, Kapoor N, Kirkpatrick D, Pollack MS, Cunningham-Rundles S, et al. 1983. Transplantation for severe combined immunodeficiency with HLA-A, B, D, DR incompatible parental marrow cells fractionated by soybean agglutinin and sheep red blood cells. Blood 61:341–48 Friedrich W, Goldmann SF, Ebell W, Blutters-Sawatzki R, Gaedecke G, et al. 1985. Severe combined immunodeficiency: treatment by bone marrow transplantation in 15 infants using HLAhaploidentical donors. Eur. J. Pediatr. 144: 125–30 Buckley RH, Schiff SE, Sampson HA, Schiff RI, Markert ML, et al. 1986. Development of immunity in human severe primary T cell deficiency following

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

58.

59.

60.

61.

62.

63.

64.

haploidentical bone marrow stem cell transplantation. J. Immunol. 136:2398– 407 O’Reilly RJ, Brochstein J, Collins N, Keever C, Kapoor N, et al. 1986. Evaluation of HLA-haplotype disparate parental marrow grafts depleted of T lymphocytes by differential agglutination with a soybean lectin and E rosette depletion for the treatment of severe combined immunodeficiency. Vox Sang. 51:81–86 Moen RC, Horowitz SD, Sondel PM, Borcherding WR, Trigg ME, et al. 1987. Immunologic reconstitution after haploidentical bone marrow transplantation for immune deficiency disorders: treatment of bone marrow cells with monoclonal antibody CT-2 and complement. Blood 70:664–69 O’Reilly RJ, Keever CA, Small TN, Brochstein J. 1989. The use of HLA-nonidentical T cell depleted marrow transplants for correction of severe combined immunodeficiency disease. Immunodef. Rev. 1:273–309 Wijnaendts L, LeDeist F, Griscelli C, Fischer A. 1989. Development of immunologic functions after bone marrow transplantation in 33 patients with severe combined immunodeficiency. Blood 74:2212–19 Fischer A, Landais P, Friedrich W, Morgan G, Gerritsen B, et al. 1990. European experience of bone marrow transplantation for severe combined immunodeficiency. Lancet 336:850–54 Dror Y, Gallagher R, Wara DW, Colombe BW, Merino A, et al. 1993. Immune reconstitution in severe combined immunodeficiency disease after lectin-treated, T cell depleted haplocompatible bone marrow transplantation. Blood 81:2021– 30 Giri N, Vowels M, Ziegler JB, Ford D, Lam-Po-Tang R. 1994. HLA non-identical T cell depleted bone marrow transplantation for primary immunodeficiency diseases. Aust. NZ J. Med. 24:26–30

P1: IKH

653

65. Buckley RH. 2001. Bone marrow transplantation in primary immunodeficiency. In Clinical Immunology: Principles and Practice, RR Rich, pp. 101.1–16. St. Louis: Mosby. 2nd ed. 66. Schiff SE, Kurtzberg J, Buckley RH. 1987. Studies of human bone marrow treated with soybean lectin and sheep erythrocytes: stepwise analysis of cell morphology, phenotype and function. Clin. Exp. Immunol. 68:685–93 67. International Union of Immunological Societies. 1999. Primary immunodeficiency diseases. Report of an IUIS Scientific Committee. Clin. Exp. Immunol. 118(Suppl. 1):1–28 68. Puck JM. 1994. Molecular and genetic basis of X-linked immunodeficiency disorders. J. Clin. Immunol. 14:81–89 69. Hirschhorn R. 1999. Immunodeficiency diseases due to deficiency of adenosine deaminase. See Ref. 91, pp. 121–39 70. Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, et al. 1995. Consensus conference on acute GVHD grading. Bone Marrow Transplant. 15: 825–28 71. Aiuti A, Slavin S, Aker M, Ficara F, Deola S, et al. 2002. Correction of ADASCID by stem cell gene therapy combined with nonmyeloablative conditioning. Science 296:2410–13 72. Friedman NJ, Schiff SE, Ward FE, Schiff RI, Buckley RH. 1991. Graft-versusgraft and graft-versus-host reactions after HLA-identical bone marrow transplantation in a patient with severe combined immunodeficiency with transplacentally acquired lymphoid chimerism. Pediatr. Allerg. Immunol. 2:111–16 73. Sarzotti M, Patel DD, Li X, Ozaki DA, Cao S, et al. 2003. T cell repertoire development in humans with SCID after nonablative allogeneic marrow transplantation. J. Immunol. 170:2711–18 74. Ghory P, Schiff S, Buckley R. 1986. Appearance of multiple benign paraproteins during early engraftment of soy lectin

12 Feb 2004

15:56

654

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

75.

76.

77.

78.

79.

80.

81.

82.

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

P1: IKH

BUCKLEY T cell-depleted haploidentical bone marrow cells in severe combined immunodeficiency. J. Clin. Immunol. 6:161–69 Kent EF, Crawford J, Cohen HJ, Buckley RH. 1990. Development of multiple monoclonal serum immunoglobulins (multiclonal gammopathy) following both HLA-identical unfractionated and T cell-depleted haploidentical bone marrow transplantation in severe combined immunodeficiency. J. Clin. Immunol. 10: 106–14 Gerritsen EJA, van Tol MJD, Lankester AC, van der Weijden-Rajas CPM, Jol-van der Zjide CM, et al. 1993. Immunoglobulin levels and monoclonal gammopathies in children after bone marrow transplantation. Blood 82:3493–502 Clement-De Boers A, Oostdijk W, Van Weel-Sipman MH, Van den Broeck J, Wit JM, Vossen JM. 1996. Final height and hormonal function after bone marrow transplantation in children. J. Pediatr. 129:544–50 Haddad E, Landais P, Friedrich W, Gerritsen B, Cavazzana-Calvo M, et al. 1998. Long-term immune reconstitution and outcome after HLA-nonidentical T-celldepleted bone marrow transplantation for severe combined immunodeficiency: a European retrospective study of 116 patients. Blood 91:3646–53 Kurtzberg J, Laughlin M, Graham ML, Smith C, Olson JF, et al. 1996. Placental blood as a source of hematopoietic stem cells for transplantation into unrelated recipients. N. Engl. J. Med. 335:157–66 Gluckman E, Rocha V, Boyer-Chammard A, Locatelli F, Arcese W, et al. 1997. Outcome of cord blood transplantation from related and unrelated donors. N. Engl. J. Med. 337:373–81 Knutsen AP, Wall DA. 2000. Umbilical cord blood transplantation in severe T-cell immunodeficiency disorders: twoyear experience. J. Clin. Immunol. 20: 466–76 Flake AW, Roncarolo MG, Puck JM,

83.

84.

85.

86.

87.

88.

88a.

Almeida-Porada G, Evans MI, et al. 1996. Treatment of X-linked severe combined immunodeficiency by in utero transplantation of paternal bone marrow. N. Engl. J. Med. 335:1806–10 Wengler GS, Lanfranchi A, Frusca T, Verardi R, Neva A, et al. 1996. In-utero transplantation of parental CD34 haematopoietic progenitor cells in a patient with X-linked severe combined immunodeficiency (SCIDX1). Lancet 348:1484– 87 Bertrand Y, Landais P, Friedrich W, Gerritsen B, Morgan G, et al. 1999. Influence of severe combined immunodeficiency phenotype on the outcome of HLA non-identical T cell-depleted bone marrow transplantation. J. Pediatr. 134:740– 48 O’Marcaigh AS, DeSantes K, Hu D, Pabst H, Horn B, et al. 2001. Bone marrow transplantation for T-B- severe combined immunodeficiency disease in Athabascanspeaking native Americans. Bone Marrow Transplant. 27:703–9 Smogorzewska EM, Brooks J, Annett G, Kapoor N, Crooks GM, et al. 2000. T cell depleted haploidentical bone marrow transplantation for the treatment of children with severe combined immunodeficiency. Arch. Immunol. Ther. Exp. 48: 111–18 Cavazzana-Calvo M, Hacein-Bey S, de Basile G, Gross F, Yvon E, et al. 2000. Gene therapy of human severe combined immunodeficiency (SCID)-X1 disease. Science 288:669–72 Hacein-Bey-Abina S, Le Deist F, Carlier F, Bouneaud C, Hue C, et al. 2002. Sustained correction of X-linked severe combined immunodeficiency by ex vivo gene therapy. N. Engl. J. Med. 346:1185–93 Hacein-Bey-Abina S, Von Kalle C, Schmidt M, McCormack MP, Wulffraat N, et al. 2003. LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCID-X1. Science 302:415– 419

12 Feb 2004

15:56

AR

AR210-IY22-21.tex

AR210-IY22-21.sgm

LaTeX2e(2002/01/18)

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

89. Rabbitts TH. 1998. LMO T-cell translocation oncogenes typify genes activated by chromosomal translocations that alter transcription and developmental processes. Genes Dev. 12:2651–57 90. Handgretinger R, Klingebiel T, Lang P, Schumm M, Neu S, et al. 2001. Megadose transplantation of purified periph-

P1: IKH

655

eral blood CD34+ progenitor cells from HLA-mismatched parental donors in children. Bone Marrow Transplant. 27:777– 83 91. Ochs HD, Smith CIE, Puck JM, eds. 1999. Primary Immunodeficiency Diseases: A Molecular and Genetic Approach. New York/Oxford: Oxford Univ. Press

Buckley.qxd

3/12/2004

2:12 PM

Page 1

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HUMAN SEVERE COMBINED IMMUNODEFICIENCY

C-1

Figure 5 Diagram showing that Janus kinase 3 (Jak3) is the major signal transducer for the common gamma chain (γc) shared by multiple cytokine receptors. Mutations in the IL2RG gene cause X-linked SCID, whereas mutations in the Jak3 gene result in a form of autosomal recessive SCID that mimics X-SCID in lymphocyte phenotype (i.e., T–,B+ NK–). Mutations in the alpha chain of the IL-7 receptor also cause SCID, but unlike Xlinked and Jak3-deficient SCID, IL-7Rα chain–deficient SCID infants have both B and NK cells (i.e., are T–,B+,NK+). [Figure reprinted with permission from (27a).]

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:625-655. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104731

Annu. Rev. Immunol. 2004. 22:657–82 doi: 10.1146/annurev.immunol.22.012703.104731 First published online as a Review in Advance on December 12, 2003

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS∗ Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, Building 10, Room 11N256, National Institutes of Health, Bethesda, Maryland 20892; email: [email protected]

Key Words hypoxia, adenosine receptors, immunity, inflammation ■ Abstract Immune cell–mediated destruction of pathogens may result in excessive collateral damage to normal tissues, and the failure to control activated immune cells may cause immunopathologies. The search for physiological mechanisms that downregulate activated immune cells has revealed a critical role for extracellular adenosine and for immunosuppressive A2A adenosine receptors in protecting tissue from inflammatory damage. Tissue damage–associated deep hypoxia, hypoxia-inducible factors, and hypoxia-induced accumulation of adenosine may represent one of the most fundamental and immediate tissue-protecting mechanisms, with adenosine A2A receptors triggering “OFF” signals in activated immune cells. In these regulatory mechanisms, oxygen deprivation and extracellular adenosine accumulation serve as “reporters,” while A2A adenosine receptors serve as “sensors” of excessive tissue damage. The A2A receptor–triggered generation of intracellular cAMP then inhibits activated immune cells in a delayed negative feedback manner to prevent additional tissue damage. Targeting A2A adenosine receptors may have important clinical applications.

INTRODUCTION Trauma and infection are two of the major threats to healthy cells, and in response to these threats the immune system has developed the ability to rapidly heal wounds and defend against microbial pathogens. The destruction of pathogens and ∗ The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

657

12 Feb 2004

21:12

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

658

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

virus-infected cells by activated immune cells is mediated by the action of a variety of secreted proinflammatory cytokines and cytotoxic molecules (1) and/or by direct cell-mediated cytotoxicity (2). However, the action of toxic proinflammatory molecules and cytotoxic cells may also result in undesirable outcomes, largely because of collateral tissue damage. Uncontrolled inflammation plays an important role in the pathogenesis of major diseases including cancer, heart disease, atherosclerosis, and sepsis (3–8). Accordingly, the immune response must be tightly controlled by highly evolved and effective downregulating immunological mechanisms, as well as by “nonimmune” molecules, which include metabolites that may have the capacity to inhibit activated immune cells and thereby prevent collateral tissue damage (9). Studies of the role of such nonimmune factors in affecting activated immune cells in the local tissue microenvironment were mostly restricted to the pharmacological evaluation of substances that are released and generated during the neuroendocrine stress response (10), as well as to certain amino acids (11) and arachidonic acid metabolites (12). Many of these molecules elevate levels of intracellular cAMP via the activation of adenylyl cyclase. Among such pharmacologically immunosuppressive molecules are catecholamines, neuropetides, histamine, and prostaglandines of the E and I series (10, 12). The cAMP-elevating effects and anti-inflammatory actions of catecholamines could be further potentiated by the well-known permissive effects of cortisol, which is released upon activation of the hypothalamic pituitary adrenal axis (10). Interest in the mechanisms that may inhibit activated immune cells in the local tissue environment has also been motivated by one of the old mysteries of cancer immunology, the Hellstrom paradox (13). The still poorly understood in vivo resistance of larger solid tumors to the attack by in vitro highly lytic antitumor T cells has been explained by a “hostile” tumor environment, which leads to the paradoxical coexistence of tumors and antitumor immune cells in the same cancer patient and negatively impacts the outcome of adoptive immunotherapy (14). Our search for the physiological mechanisms (“OFF” signals) that limit collateral tissue damage by immune cells was in large degree driven by the possibility that cancerous tissues are protected from antitumor immune cells by the same mechanism that protects normal tissues from excessive collateral immune damage. Thus, revealing the molecular basis of normal tissue-protecting mechanisms during inflammation could be of utmost importance both for the understanding of innate and adoptive immunity and for the development of novel and more selective anti-inflammatory and anticancer treatments. This task may be facilitated by the recent demonstration (15) that the ubiquitous purine nucleoside, adenosine, is physiologically involved in inhibiting activated immune cells and in protecting tissue from acute inflammatory damage in vivo. It is believed that extracellular adenosine accumulates in inflamed areas with damaged microcirculation, diminished blood supply, and low oxygen tension. Under such conditions, adenosine first serves to “report” the excessive collateral immune damage; it then prevents additional damage by inhibiting activated immune cells. Immunosuppressive effects

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HYPOXIA AND ADENOSINE IN IMMUNITY

P1: IKH

659

Figure 1 Delayed negative feedback downregulation of activated immune cells in the inflamed local tissue environment. The inflammatory tissue damage is accompanied by local tissue hypoxia and direct cell injury, which are conducive to the accumulation of extracellular adenosine. The extracellular adenosine triggers high-affinity A2A adenosine receptors on activated immune cells and leads to an increase in intracellular cAMP. cAMP, an intracellular “OFF” signal (9), then inhibits intracellular signaling pathways, leading to the interruption of proinflammatory processes in immune cells in a delayed negative feedback manner. In this model, the extracellular adenosine, a “metabokine,” may act as both a “reporter” of tissue damage and a “retaliatory metabolite” (120) to counteract further immune cell activation and hence limit tissue inflammation.

of extracellular adenosine are mediated by A2A adenosine receptors—through elevation of intracellular cAMP—in a delayed negative feedback manner (Figure 1). In this review we describe hypoxia-induced processes and adenosine receptor signaling in immune cells. We also summarize evidence that extracellular adenosine plays a critical role in the physiological regulation of inflammation and the protection of tissue from excessive damage. These studies were made possible because of a series of studies on the effects of cAMP-mediated signaling by Gs protein–coupled receptors, including adenosine receptors (reviewed in 9, 16–18). Here, we focus on the effects of hypoxia as it relates to the development, function,

12 Feb 2004

21:12

660

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

and adaptation of immune cells, as well as on the known roles of hypoxia-inducible factors. We also discuss studies of extracellular adenosine formation and signaling through different subtypes of adenosine receptors.

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ROLE OF HYPOXIA AND HYPOXIA-INDUCIBLE FACTORS IN THE DEVELOPMENT AND FUNCTIONS OF IMMUNE CELLS Local tissue hypoxia in the inflamed and damaged tissue microenvironment is one of the immediate consequences of the disruption in the microcirculation (9, 16– 17). Oxygen deprivation, in turn, results in dramatic changes in cell metabolism, owing to activities of hypoxia-inducible factors (19, 20).

Hypoxic and Normoxic Pathways of Hypoxia-Inducible Factor Stabilization Oxygen is the most critical molecule for survival for most organisms. Accordingly, oxygen-dependent organisms have evolved mechanisms for detecting levels of oxygen (sensors). One such mechanism organisms use for adapting to a lack of oxygen involves the transcriptional activity of hypoxia-inducible factors (HIFs) (19, 21). HIFs are stabilized under hypoxic conditions and enhance the expression of genes that allow either increased oxygen delivery and/or improved cell survival in conditions of limited oxygen availability. Thus, HIF-regulated genes have an important role in vascularization (VEGF, PDGF), vasodilation (iNOS), erythropoesis (EPO), and anaerobic glycolytic pathways (GLUT1, PGK) (20). HIFs are heterodimers consisting of one of three alpha subunits (HIF-1α, HIF-2α, or HIF-3α) bound to HIF-1β, which is also known as the aryl hydrocarbon receptor nuclear translocator. While HIF-1β mRNA and protein expression do not change under hypoxic conditions (HIF-1β mRNA expression can be compared to that of housekeeping genes), hypoxia upregulates HIF-1α protein expression mostly on a posttranslational level, although hypoxia can lead to increased HIF-1α mRNA accumulation in some settings (22, 23). Hypoxic stabilization of HIF-1α protein results when a lack of molecular oxygen inhibits its degradation (Figure 2). Prolylhydroxylases of the EGLN family play a critical role in the regulation of HIF-1α degradation. This discovery both identified oxygen sensors and provided the molecular mechanism of HIF-1α stabilization in hypoxic conditions and its degradation under normoxic conditions (24). As its name suggests, HIF-1α is mainly expressed under hypoxic conditions, but there is also evidence for the accumulation of HIF-1α under some normoxic conditions. These include the stabilization and transactivation of HIF-1α by reactive nitrogen- or oxygen-derived radicals: RNS (25) and ROS (26), cytokines [TNF-α (27) and IL1-β (28)], growth factors [IGF (29)], and the postischemic phase of tissues [for instance, in the brain, where HIF-1α is likely to participate in the mediation of the phenomenon of late ischemic preconditioning (30)].

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HYPOXIA AND ADENOSINE IN IMMUNITY

P1: IKH

661

Figure 2 Oxygen tension–dependent regulation of HIF-1α. At high oxygen tensions, HIF1α is targeted for destruction by an E3 ubiquitin ligase complex that contains the von Hippel– Lindau tumor suppressor protein (pVHL). pVHL can bind to a HIF-1α protein only if the latter has been hydroxylated by oxygen tension–dependent prolyl hydroxylases on conserved prolyl residues within the oxygen-dependent degradation domain. HIF-1α is degraded through the ubiquitin-proteasome pathway. HREs are hypoxia-responsive elements in HIF-1α–regulated genes. VEC designates complex of pVHL, elongins B/C, Cul2, and RBX1. HyP, hydroxyproline; Ub, ubiquitin. For further explanation, see text.

Hypoxia-independent, normoxic upregulation of the HIF-1α subunit in antigen receptor–activated murine T lymphocytes provides insight into the immune response (31). This function was mainly accounted for by the selective upregulation of the alternatively spliced isoform I.1 mRNA that endcodes the HIF-1α protein without the first 12 N-terminal amino acids. The expression of the long I.2 isoform mRNA is constitutive, whereas the short I.1 isoform mRNA could be induced in the normoxic condition in T cells by T-cell receptor (TCR) activation (31). The rapid accumulation of the short I.1 HIF-1α isoform in activated T cells has also been observed in vivo during cytokine-mediated inflammation. This led to the conclusion that the short I.1 isoform of HIF-1α is expressed as an immediate early gene, thereby ensuring rapid adaptation of cells to changing demands in

12 Feb 2004

21:12

662

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

metabolism in activated or hypoxic conditions (31). Such a pattern of expression of HIF-1α suggests its physiological role in T-cell activation and is in agreement with the view that activated T cells derive their ATP almost completely from glycolysis instead of oxidative phosphorylation in the resting state (32–34).

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

In Vivo Evidence for Local Tissue Hypoxia as an Initiating Event in the Regulation of Immune Cells and Modulation of Tissue Damage According to the model that was the basis for our studies (Figure 1), local tissue hypoxia—which develops as a result of local tissue damage—may represent one of the first events that initiates the termination of inflammation by creating conditions that are conducive to the accumulation of extracellular adenosine. In addition, hypoxia may result in signaling events, which recruit an evolutionarily old and powerful physiological response and immediate early tissue-defense mechanism. The first genetic in vivo evidence that HIF-1α plays an important role in developing and regulating the immune system and in protecting tissue from damage was provided by selective deletion of HIF-1α in T and B lymphocytes (35). Using the RAG-2-deficient blastocyst complementation system to bypass embryonic lethality, we analyzed the effects of HIF-1α deficiency in T and B lymphocytes. In agreement with the view that HIF-1α is an important regulator of lymphocytes, the development of B lymphocytes was abnormal. Moreover, tissue damage caused by autoimmunity was observed in HIF-1α-deficient chimeric mice but not in control animals (35). No gross changes in T-cell development have been noticed in these mice. B cell–mediated autoimmunity in HIF-1α-deficient chimeric mice is of special interest for future studies because hypoxia blocks mitogen-induced proliferation due to HIF-1α-dependent cyclin-dependent kinase p21 and p27 upregulation, CDK2 activity downregulation, and Rb hypophosphorylation in B lymphocytes (36). It is likely that, while not crucial in T-cell development, the HIF-1a activities are very important in T-cell activation and effector functions. In support of this view are observations of higher lytic activities of cytotoxic T cells in vitro at hypoxic, but physiologically relevant, oxygen concentrations (37) as compared with the cytotoxicity of cells cultured at 21% oxygen, i.e., a condition routinely used in immunological experiments. It remains to be established whether effects of hypoxia on the development of more lytic CD8+ T cells (37) could be explained by changes in, e.g., CDK2 activity that have been observed in B lymphocytes (36). As compared with lymphocytes, phagocytes (granulocytes, monocytes) are relatively poor in mitochondria and their functions require an anaerobic glycolysis as the main source of ATP (38). Moreover, phagocytes exert their microbicidal effector functions mostly under metabolically hostile conditions, because levels of oxygen, glucose, and pH are low in inflamed areas (39, 40). Although in lymphocytes the combined blockage of both mitochondrial and glycolytic pathways is necessary to inhibit their cytolytic effector functions (41), phagocytic effector functions prove to be less susceptible to inhibitors of mitochondrial respiration (38).

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HYPOXIA AND ADENOSINE IN IMMUNITY

P1: IKH

663

By contrast, glycolytic inhibitors (e.g., 2-deoxyglucose) strongly decrease superoxide anion production of phagocytes (42), demonstrating that neutrophils and macrophages are highly dependent on the anaerobic glycolytic pathway for the generation of ATP (43). These observations strongly suggested that phagocytes are able to metabolically adapt to hypoxic environments. These data further suggest that hypoxia-inducible factors may play a key role in the functioning of phagocytes in vivo, and this prediction was confirmed in an important series of experiments where HIF-1α was selectively deleted in macrophages and neutrophils (44). It was shown that HIF-1α is essential for the upregulation of enzymes of the glycolytic pathway to supply phagocytes with sufficient levels of ATP. As a result of energy starvation, HIF-1α gene-depleted phagocytes were compromised in their ability to migrate, to aggregate, and to kill bacteria. Moreover, mice with conditional deletion of HIF-1α in phagocytes exhibited fewer signs of inflammatory tissue damage in acute in vivo models of phorbol ester–induced ear inflammation as well as in two chronic models of skin and joint inflammation. Together these results strongly support the view that HIF-1α is critically involved in the survival and function of phagocytes, especially at sites of inflammation (44). Because phagocytes are also responsible to a large extent for collateral inflammatory tissue damage, hypoxic stabilization of HIF-1α in myeloid cells may hence be considered as a proinflammatory process enabling macrophages and granulocytes to cause more inflammatory tissue damage under hypoxia. The role of HIF-1α in T cells during acute inflammation in vivo may be different and needs to be established.

ROLE OF EXTRACELLULAR ADENOSINE AND ADENOSINE RECEPTORS IN THE DEVELOPMENT AND FUNCTIONS OF IMMUNE CELLS The decades-long history of studies of immunosuppressive effects of extracellular adenosine and the many scientists who developed the conceptual framework and experimental tools have been discussed in recent reviews (9, 16–18, 45). In this chapter, we review the mechanisms of adenosine formation, signaling by adenosine receptors, in vitro pharmacological studies of adenosine receptors on immune cells, and in vivo genetic evidence for the physiological role of A2A receptors.

Adenosine Formation Under Hypoxic Conditions The tissue-protecting properties of pharmacologically activated A2A receptors may reflect the physiological role of both hypoxia and adenosine because the concentration of adenosine is strongly elevated in areas of hypoxic tissues (46, 47). Even short periods of hypoxia lead to the enhanced breakdown of adenine nucleotides to adenosine because of the decreased production of ATP and accumulation of AMP, which can be further metabolized to adenosine through dephosphorylation by the cytosolic-50 -nucleotidase (Figure 3). In addition to the substrate-dependent

12 Feb 2004

21:12

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

664

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

Figure 3 Adenine nucleotides and nucleoside-related metabolic pathways. Pathways of adenosine formation are represented by solid lines, those of its degradation or removal by dotted lines. The long dashed lines indicate pathways by which AMP can be degraded without the formation of adenosine. For further explanation, see text. ATP: adenosine triphosphate; ADP: adenosine diphosphate; AMP: adenosine monophosphate; ADO: adenosine; INO: inosine; HX: hypoxanthine; X: xanthine; IMP: inosine-monophosphate; SAH: s-adenosylhomocysteine; 1: cytosolic-50 -nucleotidase; 2: family of ecto-nucleotidases, including the endothelial ecto-nucleoside triphosphate diphosphohydrolase (NTPDase, ecto-Apyrase, CD39), which hydrolyzes ATP or ADP to form AMP; 3: ecto-50 -nucleotidase (CD73); 4: AMP deaminase; 5: adenosine deaminase; 6: purine nucleoside phosphorylase; 7: xanthine oxidase; 8: bidirectional nucleoside transporter; 9: adenosine kinase.

formation of adenosine via cytosolic- and ecto-50 -nucleotidases, the extracellular adenosine concentrations may be further potentiated by preventing its reutilization through the inhibition of salvage pathways, i.e., hypoxia-dependent inhibition of the enzyme adenosine kinase that rephosphorylates the nucleoside to AMP (48). This could be a significant source of extracellular adenosine in conditions of deep hypoxia, which are associated with tissue damage and inflammation. In the myocardium, for instance, adenosine formation is directly proportional to the AMP concentration. Under normoxic conditions, the adenosine formed was in part converted back into AMP by phosphorylation via the enzyme adenosine kinase. Because the metabolic cycle between AMP and adenosine usually has a high turnover rate, any decrease in the adenosine kinase activity will automatically translate into

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HYPOXIA AND ADENOSINE IN IMMUNITY

P1: IKH

665

enhanced adenosine formation even when AMP concentrations are only slightly increased, e.g., by hypoxia (48). It is interesting to note that hypoxia dramatically inhibits adenosine kinase activity down to only 6% of levels under basal normoxic conditions. This finding may explain the strongly amplified formation of adenosine under hypoxic conditions. Because both formation of adenosine and stabilization of HIFs occur in response to hypoxia, a link between hypoxia-inducible factor stabilization, adenosine formation, and adenosine receptor signaling may exist, although this possibility has yet to be addressed. It is important to test whether accumulated (not degraded) HIF-1α may affect adenosine kinase activity directly either via protein-protein interactions or via its transcriptional activities in hypoxic conditions. Such a test would provide an important mechanism of adenosine accumulation in damaged and hypoxic tissues. Hypoxia may also result in the upregulation of an adenine nucleotide–metabolizing ecto-enzyme cascade, comprising ecto-ATP apyrase (CD39) and ecto-50 AMP nucleotidase (CD73) (49). The de novo synthesis of functional active CD73 is dependent on binding of HIF-1α to the hypoxia response element–containing promotor region of the CD73 encoding genes. Because hypoxia causes the release of intracellular adenine nucleotides, subsequent degradation by hypoxia-induced upregulation of ATP- and AMP-metabolizing ecto-enzymes (CD39/CD73) will lead to an even more enhanced production of adenosine.

Signaling by Adenosine Receptors The interest of biochemists, physiologists, and drug designers in adenosine was mostly driven by the well-established hemodynamic actions of adenosine ever since Drury & Szent-Gyorgyi (50) crystallized AMP and its degradation product adenosine to describe the effects of the nucleoside on the cardiovascular system in different mammalian species. The realization that the psychostimulatory effects of caffeine are largely due to antagonism of brain-tissue adenosine receptor signaling (51) further stimulated interest in studies of the adenosine receptork system. A better understanding of the purinergic receptors and their intracellular signaling pathways led to the classification of adenosine receptors according to the rank order of potencies of agonists with respect to the intracellular production of cAMP (52). Extracellular adenosine signals through heterotrimeric G proteins that can either stimulate (Gs) or inhibit (Gi) adenylyl cyclase, the enzyme that catalyzes the formation of cAMP. The cloning of four adenosine receptors (A1, A2A, A2B, A3) (52) helped to prove that high-affinity A2A and low-affinity A2B receptors activate the adenylyl cyclase, whereas high-affinity A1 and low-affinity A3 receptors inhibit it (Figure 4). Accordingly, when immune cells acquire the expression of all types of these receptors, they will be recruited in a stepwise manner with Gi-coupled A1 receptors activated first at very low adenosine levels, followed by the stimulation of Gs-coupled A2A and A2B receptors, and finally by Gi-coupled A3 receptors (17).

12 Feb 2004

21:12

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

666

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

Figure 4 Classification of adenosine receptors and their coupling to the enzyme adenylyl cyclase. Adenosine receptors belong to the family of heptahelical transmembrane G-proteincoupled receptors. Extracellularly orientated adenosine receptors bind adenosine with high and low affinity and can either stimulate or inhibit the enzyme adenylyl cyclase. The adenylyl cyclase consists of two regulatory and two catalytic units and generates from ATP the potent second messenger, cAMP.

When activated pharmacologically, cAMP-elevating A2A and A2B receptors inhibit inflammatory immune responses (9, 16–18). Activation of A3 receptors exerts anti-inflammatory effects by still-unknown mechanisms (53). Very low concentrations of adenosine can bind to and activate A1 receptors that can inhibit the adenylyl cyclase–activating A2A and A2B receptors. Because higher levels of adenosine can overcome the inhibition mediated by A1 receptors, A1 receptors are thought to exert a tonic inhibitory effect on A2 receptor functions. This view is supported by observations of increases in effects of A2A receptor activation by pharmacological antagonism of A1 receptors (54). Predicting the outcome of individual adenosine receptor activation is complicated by the recent realization of the possibility of di- or heterodimerization of adenosine receptors (55–57). In addition, the dual coupling of, e.g., A2B receptors to Gs and Gq was found in different cells (58). Thus, although A2A and A2B receptors lead to accumulation of cAMP, A2B receptors may also induce the mobilization of calcium from intracellular stores. At least in neutrophils, both

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HYPOXIA AND ADENOSINE IN IMMUNITY

P1: IKH

667

cAMP-dependent and cAMP-independent effects of A2A receptors have been discussed (59–61). These findings further complicate the analysis of intracellular pathways that are triggered by extracellular adenosine, and the overall outcome of adenosine signaling could at least partially be determined by the proportion of receptors coupled to one versus two different types of G proteins on the same cell. Researchers also discussed whether adenosine receptors may be coupled to and signal through other G proteins. The abundance of Gs proteins in some tissues may determine the coupling of, e.g., A2A receptor to Gs versus another protein, e.g., Golf protein (62). Moreover, adenosine receptor subtypes are also coupled by distinct G proteins to several other effector signaling systems, and they can trigger different intracellular pathways in various cells (63–65). This suggests the need to carefully investigate the adenosine receptor signaling pathways in each individual type of immune cell. Because of these pleiotropic effects of adenosine receptors on intracellular signaling molecules, their effects in vivo should be characterized with respect to the distribution of the different adenosine receptor subtypes. This distribution is likely dependent on the state of proliferation, activation, and differentiation of immune cells. Detection of adenosine receptors would be greatly facilitated by the availability of specific antibodies to their extracellular domains, a need still not met in studies of murine adenosine receptor subtypes (16). The interpretation of studies of the expression of A2A receptors on human peripheral blood leukocytes using mAb and flow cytometry is limited by the need to permeabilize cells (66). Using quantitative real-time PCR (67), researchers compared levels of adenosine receptor subtype–specific mRNAs in tissues of healthy mice. These studies revealed that A2A receptor mRNA levels are the highest in the spleen and lymph nodes (Figure 5), supporting a functional role of this immunosuppressive adenosine receptor subtype in the regulation of the immune response in peripheral lymphoid tissues. Adenosine receptors are found on virtually all immune cells, including polymorphonuclear leukocytes, monocytes, macrophages, dendritic cells, lymphocytes, platelets, as well as endothelial cells (18). Studies of the mechanisms that govern expression of these receptors are still required to better understand the immunoregulation by adenosine receptors. The most commonly used approach was based on pharmacological tools, i.e., the study of selective agonists and antagonists of different adenosine receptors (9, 16–18, 45).

In Vitro Pharmacological Studies of Adenosine Receptors on Cells of the Innate and Adaptive Immune System POLYMORPHONUCLEAR LEUKOCYTES Polymorphonuclear leukocytes (PMN)– mediated host protection from invading microorganisms by highly potent bactericidal effector mechanisms is augmented by the abilities of PMN to adhere, phagocytose, and degrade engulfed microorganisms, as well as to destroy pathogens by toxic oxygen radicals and degranulation of bactericidal enzymes and proteins into phagolysosomes (68). However, under conditions of pathological stimulation, the same spectrum of PMN bactericidal effector functions could be nonspecifically directed against the host’s healthy tissue and result in extensive collateral damage.

21:12

668

AR AR210-IY22-22.tex AR210-IY22-22.sgm

Figure 5 Comparison of the relative expression of A2A receptor mRNA in lymphoid and nonlymphoid murine tissues. Total RNA was extracted from the organs of a C57Bl6 mouse, and A2A receptor mRNA were determined in relationship to that of a housekeeping gene (18S ribosomal RNA) by quantitative real-time PCR. Shown are the calculated expression levels normalized to that determined in spleen tissue.

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

12 Feb 2004 LaTeX2e(2002/01/18) P1: IKH

SITKOVSKY ET AL.

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HYPOXIA AND ADENOSINE IN IMMUNITY

P1: IKH

669

The collateral damage by activated PMN is most likely due to their strong adhesion to the vascular endothelium. This adhesion is mediated in part by the same receptors (CR3, CD11b/CD18) that ensure PMN attachment to pathogenic microorganisms opsonized by the complement fragment C3bi. In addition to their role in PMN adhesion, the integrins CD11b/CD18 also activate neutrophils by “outside-in cosignaling” and thereby dramatically enhance the release of toxic proteolytic enzymes and oxygen radicals into a protected microenvironment (69) in the microscopic space that is formed between PMN and the vascular endothelium to which PMN are attached. Physiologic inhibitors of toxic compounds seem to be too large to enter into this space, so that endothelial cells are damaged by the secreted radicals and enzymes (70). This space, however, could also contain elevated concentrations of adenosine either because of degradation of extracellular ATP by ectonucleotidases located on the endothelial surface (71) or because of the production of adenosine by PMN (72). The tissue-protecting role of extracellular adenosine during PMN activities was suggested by experiments where the removal of adenosine by the enzyme adenosine deaminase (ADA) resulted in strong enhancement of PMN-dependent cytotoxicity toward endothelial cells (73). The extracellular adenosine may protect the microvascular endothelium from PMN by inhibiting the expression of β2-integrins (54), adhesion (74), oxygen radical production (75), degranulation (76), and production of TNF-α (77). Adenosine A2A receptors on human neutrophils were implicated as mediators of protective effects of adenosine, although more recently it was suggested that adenosine A3 receptors are involved in the inhibition of oxygen radical production (78). The overall effects of adenosine on PMN depend on the interplay of highaffinity A1 and high-affinity A2 receptors and on the concentration of adenosine. Indeed, the anti-inflammatory effects of A2A receptors are in part prevented by A1 receptors, whereas the chemotaxis, adhesion, and oxygen radical production were stimulated by A1 and inhibited by A2A receptors (79). Because A2A receptors can also restore directed migration when neutrophil chemotaxis is decreased by TNF-α (80), adenosine at low concentrations, e.g., in still-healthy tissue during the beginning of immune attack, could serve to promote bacterial defense mechanisms by increasing chemotaxis to infected sites. In contrast, when inflammatory tissue damage and adenosine levels increase, adenosine is expected to inhibit the neutrophil-mediated cytotoxicity, and it may thereby limit tissue injury. Determining whether tissue-protecting activation of A2A receptors might also compromise the bactericidal functions of polymorphonuclear leukocytes was important. It appears that adenosine does not inhibit all functions of PMN with the same efficacy and to the same extent (81, 82). Although complement receptor type 3–mediated phagocytosis and the associated intracellular production of bactericidal oxygen radicals were not inhibited, adenosine strongly decreased the extracellular release of toxic superoxide anions (83). In agreement with these observations is the finding that exogenously applied adenosine even at 100 µM did not prevent human PMN from killing yeast (84).

12 Feb 2004

21:12

670

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

Thus, adenosine, acting via A2A receptors, represents a unique cAMP-elevating stimulus, which might have been evolutionarily evolved to selectively inhibit the unwanted release of tissue-toxic oxygen radicals without compromising the hostprotecting bactericidal effector mechanisms of PMN.

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Mononuclear Phagocytes and Dendritic Cells Macrophages and dendritic cells play a central role in the antigen nonspecific antimicrobial defense as well as in the regulation of a Th1 or a Th2 type–specific immune response (85). Activation of A1 receptors usually has stimulatory effects on macrophages, whereas A2 receptors are inhibitory. Recent pharmacological observations of A3 receptors as strong inhibitors of monocyte functions (86, 87) suggest that the A3 receptor may play a role in the regulation of the functions of circulating blood monocytes and macrophages that is more important than its role in neutrophils. Indeed, lipopolysaccharide (LPS)-induced production of IL-12 by macrophages was inhibited, whereas IL-10 secretion was enhanced following increased A2A receptor expression (88). Interestingly, the expression and function of A2A (88) and A2B (89) receptors on macrophages could be upregulated by inflammatory cytokines. The IFN-γ -induced upregulation of A2B receptors in murine bone marrow macrophages resulted in the suppression of TNF-α and IL-1β mRNA levels as well as in the inhibition of the expression of MHC class II genes (89). Dendritic cells are also affected by pharmacological activation of adenosine receptors. In LPS-stimulated mature human dendritic cells, A1 and A3 receptor mRNAs were downregulated, while A2A receptor mRNA was still expressed. As a result, stimulation of LPS-differentiated dendritic cells by adenosine or an A2A receptor agonist increased adenylyl cyclase activity, enhanced intracelluluar cAMP, and inhibited IL-12 production (90). In dendritic cells, cAMP-elevating substances other than adenosine increase IL-10 and lower expression of MHC type II (91). Therefore, adenosine’s effects on dendritic cells, which are similar to those reported for the nucleoside in macrophages (88), could be expected. Indeed, adenosine decreased the capacity of maturing dendritic cells to induce T helper (Th1) polarization of naive CD4+ T lymphocytes. Moreover, A2A receptor activation favored production of CCL17 over CXCL10 chemokines upon differentiation of dendritic cells with LPS (92). However, in contrast to other cAMP-elevating agents, chronic exposure of dendritic cells to adenosine did not diminish the expression of MHC class I and II molecules, or of costimulatory molecules, but rather further enhanced it (92). Taken together, these data suggest that changes in expression levels of adenosine receptors on antigen-presenting cells (macrophages, dendritic cells) may play an important role in the downregulation and polarization of the immune response. Adenosine receptors may either modulate MHC class I and II expression and/or decrease IL-12 and enhance IL-10 or IL-4 production to favor the initiation of a Th2 response over a Th1 response. Thus, regulation of cells of the innate immune system by extracellular adenosine may be critical in the modulation of the interactions of cells of the innate and the adaptive immune responses. In

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

HYPOXIA AND ADENOSINE IN IMMUNITY

P1: IKH

671

addition, T and B lymphocytes could also be affected directly by adenosine receptor signaling.

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T and B Lymphocytes Responses of cells of the adaptive immune system are modulated by the complex interplay of a variety of extracellular molecules, and most studies have focused on the effects of cytokines on the activation and differentiation of T lymphocytes. Much remains to be learned about how a nonimmune factor, like adenosine, can modulate the progression of na¨ıve T lymphocytes into effector cells and how it can alter effector cell functions (93).

Thymocytes The thymus is organized in a unique fashion, and this spatial and temporal arrangement of the complex interaction of different cell types (epithelial cells, dendritic cells, macrophages, myeloids, fibroblasts, and other stromal cells) and different factors (cytokines, hormones, neuropeptides, intercellular matrix molecules) determines the thymocyte maturation in the local microenvironment. When compared with other lymphoid and nonlymphoid tissues, the microenvironment of thymocytes is hypoxic even under normal physiologic conditions (94). Thus, the thymic environment could be conducive to the formation of adenosine. That the A2A receptor signaling in thymocytes results in the highest levels of cAMP compared with all other tested immune cells, including mature T cells, is of special interest (93, 95, 96). The multiple roles of adenosine in thymocyte and peripheral T-cell physiology were also elucidated by findings in patients and animals with ADA deficiency, which usually results in a severe combined immunodeficiency [reviewed in (96), (97)]. ADA severe combined immunodeficiency is characterized by functional and developmental defects of T and B cells and severe depletion of T and B lymphocytes. The defect of the adenosine-metabolizing enzyme causes an increase in the levels of both intracellular and extracellular adenosine, although researchers believed only the intracellular accumulation of adenosine and derived compounds causes lympho-cytotoxicity. Extracellular adenosine, however, was capable of directly inducing apoptosis in a subset of immature thymocytes, and cAMP-elevating effects of adenosine were considered in studies of thymocyte differentiation (95, 98). Using A2A receptor knockout mice, investigators demonstrated that the direct apoptotic effects of extracellular adenosine on a subset of CD4+CD8+ double-positive thymocytes were completely accounted for by signaling through A2A receptors with no contribution of intracellular lymphotoxicity or of compensating A2B receptors. Unexpectedly, the extracellular adenosine strongly inhibited TCR-triggered activation of both A2A receptor–positive and A2A receptor–deficient thymocytes in the presence of ADA inhibitors. This was confirmed with thymocytes from ADA gene– deficient mice, suggesting the existence also of A2A receptor–independent effects of

12 Feb 2004

21:12

672

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

extracellular adenosine on thymocytes (96, 99, 100). However, whether these effects of adenosine contribute to the selection of thymocytes in the thymus remains to be established conclusively.

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Peripheral T Cells Abundant evidence from in vitro pharmacological experiments strongly suggests that adenosine can inhibit peripheral T-cell activation, proliferation, and production of inflammatory cytokines while enhancing the production of anti-inflammatory cytokines in these cells. Extracellular adenosine triggers immunomodulatory, cAMP-mediated pathways in T cells by virtue of signaling through A2A adenosine receptors. The use of a selective agonist (CGS21680) and an antagonist (ZM241385) of A2A adenosine receptors as well as A2A receptor–deficient mice in functional assays and genetic probes for different subtypes of adenosine receptors led to the identification of A2A receptors as the predominantly expressed subtype of Gs-coupled adenosine receptors in murine peripheral T cells (95, 99, 101). Importantly, analysis of T cells isolated from mice that are heterozygous for the A2A receptor expression revealed a gene-dose effect and no Gs-coupled A2A adenosine receptor reserve (no spare receptors) (102). These in vitro data are in agreement with more recent observations that there is no compensatory increase of A1, A2B, or A3 receptors in different murine lymphoid or nonlymphoid tissues in healthy A2A receptor knockout mice (67). In further support of the biological significance of the regulatory role of the A2A receptor system is the “memory” of T cells exposed to adenosine, as evidenced by the long-lasting increase in intracellular cAMP on TCR effector functions even after brief contact of cells with the nucleoside (101). Activation of A2A adenosine receptors inhibits TCR-triggered IL-2 receptor upregulation, thereby explaining the inhibition of T cells’ proliferation by extracellular adenosine (93). The exposure to extracellular adenosine also inhibited all tested TCR-triggered effector functions of CD8+ cytotoxic T lymphocytes including their TCR-triggered proliferation (93), lethal hit delivery by granule exocytosis, as well as FasL mRNA upregulation (101). However, the hierarchy of susceptibility of different effector functions to inhibition by A2 adenosine receptor signaling remains to be established. Studies of the role of A2A receptors in the differentiation of naive CD4+ cells into Th1 or Th2 polarized effector and memory cells would also be interesting. The use of monoclonal antihuman A2A receptor antibody in a flow cytometric assay of human peripheral blood leukocytes demonstrated that more CD4+ than CD8+ T cells express A2A receptors, but activation of T cells increased A2A receptor expression, predominantly in CD8+ T cells. Studies of T helper–cell subsets (Th1 and Th2) revealed that lymphokine-producing cells are more likely to express A2A receptors than are cells that do not produce lymphokines. These results suggest that A2A receptors are variably expressed on T-cell subsets and may regulate cytokine production in activated T lymphocytes. In agreement with the latter conclusion

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

HYPOXIA AND ADENOSINE IN IMMUNITY

P1: IKH

673

is an inverse relationship between plasma levels of ADA and decreased ratios of IFN-γ to IL-4–producing CD4+ cells in pregnant patients (103) known to have elevated plasma concentrations of adenosine (104).

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

B Cells The use of antihuman A2A receptor mAb underestimates the number of receptors and revealed no A2A adenosine receptor expression on B cells (66), but the incubation of B cells with adenosine analogs was followed by the accumulation of intracellular cAMP (105). Despite the identification of A2A receptor expression on B lymphocytes, little is known about the effects of A2A receptor signaling in B-cell development, activation, antibody-production class switching, and cytokine secretion. These areas of B-cell physiology and immunology clearly require future studies, especially as observations of the effects of cAMP on activated B cells (106, 107) support the prediction that B cells could be a sensitive target of adenosine receptor signaling. The effects reviewed above of extracellular adenosine on immune cells have been observed mostly in pharmacological experiments, and it was not known whether there are sufficient levels of extracellular adenosine in vivo to signal through A2A receptors on immune cells. Another important caveat is that cAMP-elevating adenosine receptors represented only one example of many Gsprotein-coupled receptors, including receptors for prostaglandins, histamine, and β-adrenergic receptors, which also could inhibit immune cells pharmacologically and were discussed as potential immunoregulators (9, 10, 15, 18). Neither adenosine receptors nor any other known cell-surface receptors were demonstrated to function physiologically in vivo in downregulating the immune response. Therefore conclusive testing using a genetic approach and in vivo assays of the immune response and tissue damage was required to establish the physiological role of A2A receptors. In the next section, we review evidence that demonstrates that no other Gs protein–coupled receptors can compensate for the in vivo functions of A2A receptors in protecting tissues from acute inflammatory damage (15).

In Vivo Evidence for the Physiological and Nonredundant Role of A2A Adenosine Receptors in Downregulating Immune Cells and Protecting Tissue from Excessive Damage For adenosine receptors to affect activated immune cells, the local tissue levels of extracellular adenosine should be sufficiently high and immune cells should express sufficient numbers of adenosine receptors in a tightly coordinated manner. This ensures that the immune response is terminated by the A2A receptor → cAMP signaling only in areas of excessive tissue damage and that sufficient time is allowed for pathogen destruction by activated immune cells before they are shut off by this tissue-protecting mechanism (Figure 1).

12 Feb 2004

21:12

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

674

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

The conclusive proof of this model in biochemical and pharmacological experiments in vitro has been complicated by uncertainties in selectivity and tissue distribution of agonists and antagonists of adenosine receptors, while the reliable determination of local tissue levels of extracellular adenosine is limited by available methods. The development of A2A receptor gene–deficient mice provided the most valuable tool to address this issue and to establish the role of these receptors in mechanisms of pain, blood pressure regulation, and platelet aggregation (108). Analyses of these mice also clarified mechanisms of the effects of caffeine (109) and brain ischemia in mice (110), (111). The A2A receptor gene–deficient mice (110) were used in experiments that established the critical role of A2A adenosine receptors in several models of immunemediated tissue damage in vivo (15). In the absence of A2A adenosine receptors on immune cells, the extracellular adenosine that accumulates in inflamed areas will not be able to trigger the production of intracellular cAMP in activated immune cells. This, in turn, would allow for uninhibited and uninterrupted activities of activated immune cells and for continued tissue damage. Such an outcome has been the case in virtually all tested in vivo assays of inflammation, and the absence of A2A receptors resulted in the dramatic enhancement of tissue damage. The activation of T cells and NK cells by subthreshold doses of the polyclonal TCR activator, Con A, in the model of autoimmune and viral hepatitis caused only minimal tissue damage in wild-type mice but resulted in more extensive tissue damage, more prolonged and higher levels of proinflammatory cytokines, and even death of some littermates that were deficient in A2A adenosine receptors. In an important control experiment, the lymphoid cells from A2A receptor knockout mice had normal cAMP responses to stimulation with ligands of other Gs-protein-coupled receptors such as isoproterenol or prostaglandin E2. This established that the deficiency in the A2A receptor did not affect the immunosuppressive function of other Gs-protein-coupled receptors. Moreover, the exacerbated inflammation and tissue damage in A2A receptor–deficient mice could not be explained by an increased susceptibility of hepatocytes toward TNF-α, as injection of TNF-α was equally efficient in directly destroying hepatocytes in both A2A receptor–deficient and wild-type mice in vivo. The tissue-protecting properties of adenosine A2A receptors were further confirmed in another in vivo model of inflammatory liver injury elicited by the Pseudomonas exotoxin A and in studies of chemically induced hepatoxicity by carbon tetrachloride. A2A receptor deficiency also exacerbated excessive proinflammatory cytokine production and tissue damage in in vivo septic models after subcutaneous or systemic intravenous injection of a bacterial endotoxin, LPS. These observations of excessive tissue damage and prolonged cytokine presence in A2A receptor–deficient mice as compared with wild-type mice were confirmed in studies with A2A receptor antagonist–treated wild-type mice. Taken together, these experiments provide the first genetic and pharmacological evidence of a nonredundant role for endogenous adenosine and for A2A adenosine receptors in protecting tissue from acute inflammatory damage by the downregulation of activated immune cells in vivo.

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

HYPOXIA AND ADENOSINE IN IMMUNITY

P1: IKH

675

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONCLUSIONS AND PERSPECTIVES The use of well-established models of immune cell–mediated acute tissue damage in animals with genetically inactivated HIF-1α or A2A receptors firmly established the role of hypoxia-associated changes in cell metabolism in the regulation of the immune response in the local tissue microenvironment. Hypoxia and hypoxiaassociated accumulation of extracellular adenosine appear to belong to the most fundamental and immediate events that report tissue damage and initiate processes that minimize this damage by the inhibition of activated immune cells. The crucial roles that local tissue damage and hypoxia-associated extracellular adenosine and A2A adenosine receptor–triggered signaling play to regulate cells of the innate and the adaptive immune systems (15) highlight the need to evaluate the contribution of these mechanisms in the pathogenesis of a large spectrum of diseases. The recruitment of the A2A adenosine receptor pathway seems to be highly desirable in preventing acute tissue damage, e.g., during sepsis, where the use of antibactericidal drugs causes the release of huge amounts of bacterial toxins that lead to a potentially fatal systemic inflammatory response syndrome (16, 18). The combined use of selective A2A agonists with other anti-inflammatory and pathogendestroying drugs may have synergistic therapeutic effects, and this should further reinvigorate interest in adenosine-based compounds to pharmaceutically activate immunosuppressive adenosine receptors in many clinical applications (16, 61, 112–115). Targeting A2A receptors by pharmacologically enhanced levels of endogenously formed adenosine (116) represents another therapeutic alternative, as demonstrated by the finding that the anti-inflammatory drug methothrexate, a gold standard in therapy of rheumatoid arthritis, promotes the signaling through the adenosine receptor system (117). The observation of a dramatic exacerbation of immune cell–induced tissue damage following genetic or pharmacological inactivation of A2A receptors not only demonstrates the nonredundant role of A2A receptors in the downregulation of inflammatory tissue injury, but also provides the novel rationale for targeting (inactivating) A2A receptors to enhance and/or prolong the immune response. While activation of cAMP-elevating A2A receptors has been studied extensively with the goal of decreasing inflammation (16, 17, 60), it has now become possible to block the natural anti-inflammatory pathways by inactivation of A2A adenosine receptors. This may provide hitherto unavailable therapeutic options to develop more effective adjuvants for vaccines (118) or to accomplish much stronger and uninhibited destruction of cancerous tissues by enhancing the proinflammatory effector functions of immune antitumor cells (119). Thus, continuous efforts to reveal those endogenous mechanisms that terminate inflammation and thereby protect tissue from excessive damage may be rewarded by the development of novel drugs that antagonize or enhance the natural anti-inflammatory A2A adenosine receptor pathways and thereby allow the rational management of inflammatory processes in vivo (9).

12 Feb 2004

21:12

676

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

ACKNOWLEDGMENTS

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The authors thank Drs. Maria Abbracchio, Pier Giovanni Baraldi, Louis Belardinelli, Pier Andrea Borea, Jeff Burnstock, Jiang Fan Chen, Bruce Cronstein, Francesco DiVirgillo, Joel Linden, Bertil Fredholm, Kenneth A. Jacobson, Marlene A. Jacobson, Gregg L. Semenza, and Michael Schwartzshild, as well as William E. Paul and the members of the Laboratory of Immunology, NIAID, NIH, for advice and help. This work was supported in part by a grant from the Deutsche Forschungsgemeinschaft to M.T. (Th733/2–1). The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Rosenberg HF, Gallin JI. 1999. Inflammation. In Fundamental Immunology, ed. WE Paul, pp. 1151–71. Philadelphia: Lippincott, Williams and Wilkins 2. Henkart PA, Sitkovsky MV. 2003. Cytotoxic T Lymphocytes. In Fundamental Immunology, ed. WE Paul, pp. 1127–51. Philadelphia: Lippincott, Williams and Wilkins 3. Shacter E, Weitzman SA. 2002. Chronic inflammation and cancer. Oncology 16: 217–32 4. McPherson JA, Barringhaus KG, Bishop GG, Sanders JM, Rieger JM, et al. 2001. Adenosine A(2A) receptor stimulation reduces inflammation and neointimal growth in a murine carotid ligation model. Arterioscler. Thromb. Vasc. Biol. 21:791–96 5. Pockley AG. 2002. Heat shock proteins, inflammation, and cardiovascular disease. Circulation 105:1012–17 6. Shebuski RJ, Kilgore KS. 2002. Role of inflammatory mediators in thrombogenesis. J. Pharmacol. Exp. Ther. 300:729–35 7. Mulvihill NT, Foley JB. 2002. Inflammation in acute coronary syndromes. Heart 87:201–4 8. Cohen J. 2002. The immunopathogenesis of sepsis. Nature 420:885–91 9. Sitkovsky MV. 2003. Use of the A2A adenosine receptor as a physiological immunosuppressor and to engineer inflam-

10.

11.

12.

13.

14.

15.

16.

17.

mation in vivo. Biochem. Pharmacol. 65:493–501 Webster JI, Tonelli L, Sternberg EM. 2002. Neuroendocrine regulation of immunity. Annu. Rev. Immunol. 20:125–63 Mellor AL, Chandler P, Lee GK, Johnson T, Keskin DB, et al. 2002. Indoleamine 2,3-dioxygenase, immunosuppression and pregnancy. J. Reprod. Immunol. 57:143–50 Calder PC, Grimble RF. 2002. Polyunsaturated fatty acids, inflammation and immunity. Eur. J. Clin. Nutr. 56(Suppl. 3):S14–19 Hellstrom I, Hellstrom KE, Pierce GE, Yang JP. 1968. Cellular and humoral immunity to different types of human neoplasms. Nature 220:1352–54 Rosenberg SA. 2001. Progress in human tumour immunology and immunotherapy. Nature 411:380–84 Ohta A, Sitkovsky M. 2001. Role of G-protein-coupled adenosine receptors in downregulation of inflammation and protection from tissue damage. Nature 414:916–20 Linden J. 2001. Molecular approach to adenosine receptors: receptor-mediated mechanisms of tissue protection. Annu. Rev. Pharmacol. Toxicol. 41:775–87 Fredholm BB, Ijzerman AP, Jacobson KA, Klotz K-N, Linden J. 2001. International Union of Pharmacology. XXV.

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

HYPOXIA AND ADENOSINE IN IMMUNITY

18.

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

19.

20.

21.

22.

23.

24.

25.

26.

27.

Nomenclature and classification of adenosine receptors. Pharmacol. Rev. 53:527– 52 Thiel M, Caldwell CC, Sitkovsky MV. 2003. The critical role of adenosine A2A receptors in downregulation of inflammation and immunity in the pathogenesis of infectious diseases. Microbes Infect. 5:515–26 Semenza GL. 1998. Hypoxia-inducible factor 1: master regulator of O2 homeostasis. Curr. Opin. Genet. Dev. 8:588–94 Semenza GL. 2000. HIF-1: Mediator of physiological and pathophysiological responses to hypoxia. J. Appl. Physiol. 88:1474–80 Semenza GL. 1999. Regulation of mammalian O2 homeostasis by hypoxiainducible factor 1. Annu. Rev. Cell Dev. Biol. 15:551–78 Wiener CM, Booth G, Semenza GL. 1996. In vivo expression of mRNAs encoding hypoxia-inducible factor 1. Biochem. Biophys. Res. Commun. 225:485–88 Yu AY, Frid MG, Shimoda LA, Wiener CM, Stenmark K, Semenza GL. 1998. Temporal, spatial and oxygen-regulated expression of hypoxia-inducible factor 1 in the lung. Am. J. Physiol. Lung Cell Mol. Physiol. 275:L818–26 Masson N, Willam C, Maxwell PH, Pugh CW, Ratcliffe PJ. 2001. Independent function of two destruction domains in hypoxia-inducible factor-alpha chains activated by prolyl hydroxylation. EMBO J. 20:5197–206 Brune B, Zhou J. 2003. The role of nitric oxide (NO) in stability regulation of hypoxia inducible factor-1α (HIF-1α). Curr. Med. Chem. 10:845–55 Haddad JJ. 2002. Antioxidant and prooxidant mechanisms in the regulation of redox(y)-sensitive transcription factors. Cell Signal. 14:879–97 Jung YJ, Isaacs JS, Lee SM, Trepel J, Liu ZG, Neckers L. 2003. Hypoxia-inducible factor induction by tumour necrosis factor in normoxic cells requires receptor-

28.

29.

30.

31.

32.

33.

34.

35.

36.

P1: IKH

677

interacting protein-dependent nuclear factor kappa B activation. Biochem. J. 370:1011–17 Haddad JJ. 2002. Recombinant human interleukin (IL)-1beta-mediated regulation of hypoxia-inducible factor-1α (HIF-1α) stabilization, nuclear translocation and activation requires an antioxidant/reactive oxygen species (ROS)-sensitive mechanism. Eur. Cytokine Netw. 13:250– 60 Zelzer E, Levy Y, Kahana C, Shilo BZ, Rubinstein M, Cohen B. 1998. Insulin induces transcription of target genes through the hypoxia-inducible factor HIF1α/ARNT. EMBO J. 17:5085–94 Jones NM, Bergeron M. 2001. Hypoxic preconditioning induces changes in HIF-1 target genes in neonatal rat brain. J. Cereb. Blood Flow Metab. 21:1105–14 Lukashev D, Caldwell C, Ohta A, Chen P, Sitkovsky M. 2001. Differential regulation of two alternatively spliced isoforms of hypoxia-inducible factor-1 alpha in activated T lymphocytes. J. Biol. Chem. 276:48754–63 Brand KA, Hermfisse U. 1997. Aerobic glycolysis by proliferating cells: a protective strategy against reactive oxygen species. FASEB J. 11:388–95 Valdemarsson S, Monti M. 1994. Increased ratio between anaerobic and aerobic metabolism in lymphocytes from hyperthyroid patients. Eur. J. Endocrinol. 130:276–80 Cheung K, Morris B. 1984. The respiration and energy metabolism of sheep lymphocytes. Aust. J. Exp. Biol. Med. Sci. 62 (Pt. 6):671–85 Kojima H, Gu H, Nomura S, Caldwell CC, Kobata T, et al. 2002. Abnormal B lymphocyte development and autoimmunity in hypoxia-inducible factor 1α-deficient chimeric mice. Proc. Natl. Acad. Sci. USA 99:2170–74 Goda N, Ryan HE, Khadivi B, McNulty W, Rickert RC, Johnson RS. 2003. Hypoxia-inducible factor 1α is essential

12 Feb 2004

21:12

678

37.

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

38.

39.

40.

41.

42.

43.

44.

45.

46.

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL. for cell cycle arrest during hypoxia. Mol. Cell Biol. 23:359–69 Caldwell CC, Kojima H, Lukashev D, Armstrong J, Farber M, et al. 2001. Differential effects of physiologically relevant hypoxic conditions on T lymphocyte development and effector functions. J. Immunol. 167:6140–49 Fossati G, Moulding DA, Spiller DG, Moots RJ, White MR, Edwards SW. 2003. The mitochondrial network of human neutrophils: role in chemotaxis, phagocytosis, respiratory burst activation, and commitment to apoptosis. J. Immunol. 170:1964–72 Saadi S, Wrenshall LE, Platt JL. 2002. Regional manifestations and control of the immune system. FASEB J. 16:849– 56 Schor H, Vaday GG, Lider O. 2000. Modulation of leukocyte behavior by an inflamed extracellular matrix. Dev. Immunol. 7:227–38 MacDonald HR, Koch CJ. 1977. Energy metabolism and T cell-mediated cytolysis. I. Synergism between inhibitors of respiration and glycolysis. J. Exp. Med. 146:698–709 Simchowitz L, Mehta J, Spilberg I. 1979. Chemotactic factor-induced generation of superoxide radicals by human neutrophils: effect of metabolic inhibitors and antiinflammatory drugs. Arthritis Rheum. 22:755–63 Borregaard N, Herlin T. 1982. Energy metabolism of human neutrophils during phagocytosis. J. Clin. Invest. 70:550–57 Cramer T, Yamanishi Y, Clausen BE, Forster I, Pawlinski R, et al. 2003. HIF1α is essential for myeloid cell-mediated inflammation. Cell 112:645–57 Fredholm BB, Cunha RA, Svenningsson P. 2003. Pharmacology of adenosine A2A receptors and therapeutic applications. Curr. Top Med. Chem. 3:413– 26 Rudolphi KA, Schubert P, Parkinson FE, Fredholm BB. 1992. Neuroprotective role

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

of adenosine in cerebral ischemia. Trends Pharmacol. Sci. 13:439 Berne RM, Rubio R. 1974. Regulation of coronary blood flow. Adv. Cardiol. 12:303–17 Decking UKM, Schlieper G, Kroll K, Schrader J. 1997. Hypoxia-induced inhibition of adenosine kinase potentiates cardiac adenosine release. Circ. Res. 81:154– 64 Synnestvedt K, Furuta GT, Comerford KM, Louis N, Karhausen J, et al. 2002. Ecto-50 -nucleotidase (CD73) regulation by hypoxia-inducible factor-1 mediates permeability changes in intestinal epithelia. J. Clin. Invest. 110:993–1002 Drury AN, Szent-Gyorgy A. 1929. The physiological activity of adenine compounds with special reference to their actions upon the mammalian heart. J. Physiol. 68:213–37 Fredholm BB. 1995. Adenosine, adenosine receptors and the actions of caffeine. Pharmacol. Toxicol. 76:93–101 Olah ME, Stiles GL. 1995. Adenosine receptor subtypes: characterization and therapeutic regulation. Annu. Rev. Pharmacol. Toxicol. 35:581–606 Salvatore CA, Tilley SL, Latour AM, Fletcher DS, Koller BH, Jacobson MA. 2000. Disruption of the A(3) adenosine receptor gene in mice and its effect on stimulated inflammatory cells. J. Biol. Chem. 275:4429–34 Thiel M, Chambers JD, Chouker A, Fischer S, Zourelidis C, et al. 1996. Effect of adenosine on the expression of beta 2integrins and 1-selectin of polymorphonuclear leukocytes in vitro. J. Leukoc. Biol. 59:671–82 Franco R, Ferre S, Agnati L, Torvinen M, Gines S, et al. 2000. Evidence for adenosine/dopamine receptor interactions: indications for heteromerization. Neuropsychopharmacology 23:S50–59 Gines S, Hillion J, Torvinen M, Le Crom S, Casado V, et al. 2000. Dopamine D1 and adenosine A1 receptors form

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

HYPOXIA AND ADENOSINE IN IMMUNITY

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

57.

58.

59.

60.

61.

62.

63.

functionally interacting heteromeric complexes. Proc. Natl. Acad. Sci. USA 97: 8606–11 Fuxe K, Ferre S, Zoli M, Agnati LF. 1998. Integrated events in central dopamine transmission as analyzed at multiple levels. Evidence for intramembrane adenosine A2A/dopamine D2 and adenosine A1/dopamine D1 receptor interactions in the basal ganglia. Brain Res. Brain Res. Rev. 26:258–73 Linden J, Thai T, Figler H, Robeva AS. 1999. Characterization of human A2B adenosine receptors: radioligand binding, western blotting and coupling to Gq in HEK-293 and HMC-1 mast cells. Mol. Pharmacol. 56:705–13 Cronstein BN, Kramer SB, Rosenstein ED, Korchak HM, Weissmann G, Hirschorn R. 1988. Occupancy of adenosine receptors raises cyclic AMP alone and in synergy with occupancy of chemoattractant receptors and inhibits membrane depolarization. Biochem. J. 252:709–15 Thiel M, Bardenheuer H, Poch G, Madel C, Peter K. 1991. Pentoxifylline does not act via adenosine receptors in the inhibition of the superoxide anion production of human polymorphonuclear leukocytes. Biochem. Biophys. Res. Commun. 180:53–58 Sullivan GW, Rieger JM, Scheld WM, Macdonald TL, Linden J. 2001. Cyclic AMP-dependent inhibition of human neutrophil oxidative activity by substituted 2propynylcyclohexyl adenosine A(2A) receptor agonists. Br. J. Pharmacol. 132: 1017–26 Kull B, Svenningsson P, Fredholm BB. 2000. Adenosine A(2A) receptors are colocalized with and activate g(olf) in rat striatum. Mol. Pharmacol. 58:771–77 Fredholm BB, Arslan G, Halldner L, Kull B, Schulte G, Wasserman W. 2000. Structure and function of adenosine receptors and their genes. Naunyn Schmiedebergs Arch. Pharmacol. 362:364–74

P1: IKH

679

64. Sexl V, Mancusi G, Holler C, GloriaMaercker E, Schutz W, Freissmuth M. 1997. Stimulation of the mitogen-activated protein kinase via the A2A-adenosine receptor in primary human endothelial cells. J. Biol. Chem. 272:5792–99 65. Seidel MG, Klinger M, Freissmuth M, Holler C. 1999. Activation of mitogenactivated protein kinase by the A(2A)adenosine receptor via a rap1-dependent and via a p21(ras)-dependent pathway. J. Biol. Chem. 274:25833–41 66. Koshiba M, Rosin DL, Hayashi N, Linden J, Sitkovsky MV. 1999. Patterns of A2A extracellular adenosine receptor expression in different functional subsets of human peripheral T cells. Flow cytometry studies with anti-A2A receptor monoclonal antibodies. Mol. Pharmacol. 55:614–24 67. Lukashev DE, Smith PT, Caldwell CC, Ohta A, Apasov SG, Sitkovsky MV. 2003. Analysis of A2a receptor-deficient mice reveals no significant compensatory increases in the expression of A2b, A1, and A3 adenosine receptors in lymphoid organs. Biochem. Pharmacol. 65:2081– 90 68. Witko-Sarsat V, Rieu P, DescampsLatscha B, Lesavre P, Halbwachs-Mecarelli L. 2000. Neutrophils: molecules, functions and pathophysiological aspects. Lab. Invest. 80:617–53 69. Nathan C, Srimal S, Farber C, Sanchez E, Kabbash L, et al. 1989. Cytokine induced respiratory burst of human neutrophils: dependence on extracellular matrix proteins and cd11/cd18 integrins. J. Cell Biol. 109:1341–49 70. Arfors KE, Lundberg C, Lindbom L, Lundberg K, Beatty PG, Harlan JM. 1987. A monoclonal antibody to the membrane glycoprotein complex CD18 inhibits polymorphonuclear leukocyte accumulation and plasma leakage in vivo. Blood 69:338–40 71. Pearson JD, Gordon JL. 1985. Nucleotide metabolism by endothelium. Annu. Rev. Physiol. 47:617–27

12 Feb 2004

21:12

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

680

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

72. Newby AC, Holmquist CA. 1981. Adenosine production inside rat polymorphonuclear leucocytes. Biochem. J. 200:399– 403 73. Cronstein BN, Levin RI, Belanoff J, Weissmann G, Hirschhorn R. 1986. Adenosine: an endogenous inhibitor of neutrophil-mediated injury to endothelial cells. J. Clin. Invest. 78:760–70 74. Cronstein BN, Levin RI, Philips M, Hirschhorn R, Abramson SB, Weissmann G. 1992. Neutrophil adherence to endothelium is enhanced via adenosine A1 receptors and inhibited via adenosine A2 receptors. J. Immunol. 148:2201–6 75. Cronstein BN, Rosenstein ED, Kramer SB, Weissman G, Hirschhorn R. 1985. Adenosine: a physiologic modulator of superoxide anion generation by human neutrophils. Adenosine acts via an A2 receptor on human neutrophils. J. Immunol. 2:1366–71 76. Richter J. 1992. Effect of adenosine analogues and cAMP-raising agents on TNF, GM-CSF-, and chemotactic peptideinduced degranulation in single adherent neutrophils. J. Leukoc. Biol. 51:270–75 77. Thiel M, Chouker A. 1995. Acting via A2 receptors, adenosine inhibits the production of tumor necrosis factor-alpha of endotoxin-stimulated human polymorphonuclear leukocytes. J. Lab. Clin. Med. 124:275–82 78. Gessi S, Varani K, Merighi S, Cattabriga E, Iannotta V, Leung E, et al. 2002. A(3) adenosine receptors in human neutrophils and promyelocytic HL60 cells: a pharmacological and biochemical study. Mol. Pharmacol. 61:415–24 79. Cronstein B, Daguma L, Nichols D, Hutchison A, Williams M. 1990. The adenosine/neutrophil paradox resolved: human neutrophils possess both A1 and A2 receptors that promote chemotaxis and inhibit O2 generation, respectively. J. Clin. Invest. 85:1150–57 80. Sullivan GW, Linden J, Hewlett EL, Carper HT, Hylton JB, Mandell GL. 1990.

81.

82.

83.

84.

85.

86.

87.

88.

Adenosine and related compounds counteract tumor necrosis factor-alpha inhibition of neutrophil migration: implications of a novel cyclic amp-independent action on the cell surface. J. Immunol. 145:1537– 44 Cronstein BN, Kubersky SM, Weissmann G, Hirschhorn R. 1987. Engagement of adenosine receptors inhibits hydrogen peroxide (H2O2) release by activated human neutrophils. Clin. Immunol. Immunopathol. 42:76–85 Zalavary S, Bengtsson T. 1998. Modulation of the chemotactic peptide- and immunoglobulin G-triggered respiratory burst in human neutrophils by exogenous and endgenous adenosine. Eur. J. Pharmacol. 354:215–25 Thiel M, Holzer K, Kreimeier U, Mortiz S, Peter K, Messmer K. 1997. Effects of adenosine on the functions of circulating polymorphonuclear leukocytes during hyperdynamic endotoxemia. Infect. Immun. 65:2136–44 Smail EH, Cronstein BN, Meshulam T, Esposito AL, Ruggeri RW, Diamond RD. 1992. In vitro, Candida albicans releases the immune modulator adenosine and a second, high-molecular weight agent that blocks neutrophil killing. J. Immunol. 148:3588–95 Fearon DT, Locksley RM. 1996. The instructive role of innate immunity in the acquired immune response. Science 272:50– 54 Broussas M, Cornillet-Lefebvre P, Potron G, Nguyen P. 1999. Inhibition of fMLPtriggered respiratory burst of human monocytes by adenosine: involvement of A3 adenosine receptor. J. Leukoc. Biol. 66:495–501 Sajjadi FG, Takabayashi K, Foster AC, Domingo RC, Firestein GS. 1996. Inhibition of TNF-α expression by adenosine. Role of A3 adenosine receptors. J. Immunol. 156:3435–42 Khoa ND, Montesinos MC, Reiss AB, Delano D, Awadallah N, Cronstein BN.

12 Feb 2004

21:12

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

HYPOXIA AND ADENOSINE IN IMMUNITY

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

89.

90.

91.

92.

93.

94.

95.

96.

2001. Inflammatory cytokines regulate function and expression of adenosine A(2A) receptors in human monocytic THP-1 cells. J. Immunol. 167:4026–32 Xaus J, Mirabet M, Lloberas J, Soler C, Lluis C, et al. 1999. IFN-gamma upregulates the A(2B) adenosine receptor expression in macrophages: a mechanism of macrophage deactivation. J. Immunol. 162:3607–14 Panther E, Idzko M, Herouy Y, Rheinen H, Gebicke-Haerter PJ, et al. 2001. Expression and function of adenosine receptors in human dendritic cells. FASEB J. 15:1963–70 Kambayashi T, Wallin RP, Ljunggren HG. 2001. cAMP-elevating agents suppress dendritic cell function. J. Leukoc. Biol. 70:903–10 Panther E, Corinti S, Idzko M, Herouy Y, Napp M, et al. 2003. Adenosine affects expression of membrane molecules, cytokine and chemokine release, and the Tcell stimulatory capacity of human dendritic cells. Blood 101:3985–90 Huang S, Koshiba M, Apasov S, Sitkovsky M. 1997. Role of A2a adenosine receptor-mediated signaling in inhibition of T cell activation and expansion. Blood 90:1600–10 Hale LP, Braun RD, Gwinn WM, Greer PK, Dewhirst MW. 2002. Hypoxia in the thymus: role of oxygen tension in thymocyte survival. Am. J. Physiol. Heart Circ. Physiol. 282:H1467–77 Apasov SG, Chen J-F, Smith PT, Schwarzschild MA, Fink JS, Sitkovsky MV. 2000. Study of A2A adenosine receptor gene deficient mice reveals that adenosine analogue CGS 21680 possesses no A2A receptor-unrelated lymphotoxicity. Br. J. Pharmacol. 131:43–50 Apasov SG, Sitkovsky MV. 1999. The extracellular versus intracellular mechanisms of inhibition of TCR-triggered activation in thymocytes by adenosine under conditions of inhibited adenosine deaminase. Int. Immunol. 11:179–89

P1: IKH

681

97. Aldrich MB, Blackburn MR, Datta SK, Kellems RE. 2000. Adenosine deaminase-deficient mice: models for the study of lymphocyte development and adenosine signaling. Adv. Exp. Med. Biol. 486:57–63 98. Apasov SG, Koshiba M, Chused TM, Sitkovsky MV. 1997. Effects of extracellular ATP and adenosine on different thymocyte subsets: possible role of ATP-gated channels and G proteincoupled purinergic receptors. J. Immunol. 158:5095–105 99. Apasov S, Chen JF, Smith P, Sitkovsky MV. 2000. A(2A) receptor-dependent and A(2A) receptor-independent effects of extracellular adenosine on murine thymocytes in conditions of adenosine deaminase deficiency. Blood 95:3859–67 100. Apasov SG, Blackburn MR, Kellems RE, Smith PT, Sitkovsky MV. 2001. Adenosine deaminase deficiency increases thymic apoptosis and causes defective T cell receptor signaling. J. Clin. Invest. 108:131–41 101. Koshiba M, Kojima H, Huang S, Apasov S, Sitkovsky MV. 1997. Memory of extracellular adenosine/A2a purinergic receptor-mediated signalling in murine T cells. J. Biol. Chem. 272:25881–89 102. Armstrong JM, Chen J-F, Schwarzschild MA, Apasov S, Smith PT, et al. 2001. Gene dose effect reveals no Gs protein coupled A2A adenosine receptor reserve in murine T lymphocytes. Studies of cells from A2A receptor gene-deficient mice. Biochem. J. 354:123–30 103. Yoneyama Y, Sawa R, Suzuki S, Yoneyama K, Doi D, Araki T. 2002. Relationship between adenosine deaminase activity and cytokine-secreting T cells in normal pregnancy. Obstet. Gynecol. 100:754–58 104. Yoneyama Y, Suzuki S, Sawa R, Otsubo Y, Power GG, Araki T. 2000. Plasma adenosine levels increase in women with normal pregnancies. Am. J. Obstet. Gynecol. 182:1200–3

12 Feb 2004

21:12

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

682

AR

AR210-IY22-22.tex

AR210-IY22-22.sgm

LaTeX2e(2002/01/18)

P1: IKH

SITKOVSKY ET AL.

105. Bonnafous JC, Dornand J, Favero J, Mani JC. 1981. Lymphocyte membrane adenosine receptors coupled to adenylate cyclase: properties and occurrence in various lymphocyte subclasses. J. Recept. Res. 2:347–66 106. Pastan IH, Johnson GS, Anderson J. 1975. Role of cyclic nucleotides in growth control. Annu. Rev. Immunol. 44:491–522 107. Holte H, Torjesen P, Blomhoff HK, Ruud E, Funderud S, Smeland EB. 1988. Cyclic AMP has the ability to influence multiple events during B cell stimulation. Eur. J. Immunol. 18:1359–66 108. Ledent C, Vaugeois JM, Schiffmann SN, Pedrazzini T, El Yacoubi M, et al. 1997. Aggressiveness, hypoalgesia and high blood pressure in mice lacking the adenosine A2a receptor. Nature 388:674– 78 109. El Yacoubi M, Ledent C, Parmentier M, Ongini E, Costentin J, Vaugeois JM. 2001. In vivo labelling of the adenosine A2A receptor in mouse brain using the selective antagonist [3H]SCH 58261. Eur. J. Neurosci. 14:1567–70 110. Chen J-F, Huang ZH, Ma JY, Zhu JM, Moratalla R, et al. 1999. A2A adenosine receptor deficiency attenuates brain injury induced by transient focal ischemia in mice. J. Neurosci. 19:9192–9200 111. Aden U, Halldner L, Lagercrantz H, Dalmau I, Ledent C, Fredholm BB. 2003. Aggravated brain damage after hypoxic ischemia in immature adenosine A2A knockout mice. Stroke 34:739–44 112. Jacobson KA, Gao Z-G, Chen A, Barak D, Kim S, et al. 2001. Neoceptor concept based on molecular complementarity in GPCRs: a mutant adenosine A(3) receptor with selectively enhanced affinity for amine-modified nucleosides. J. Med. Chem. 44:4125–36

113. Murphree LJ, Marshall MA, Rieger JM, MacDonald TL, Linden J. 2002. Human A2A adenosine receptors: high-affinity agonist binding to receptor-G protein complexes containing Gbeta(4). Mol. Pharmacol. 61:455–62 114. Baraldi PG, Cacciari B, Moro S, Spalluto G, Pastorin G, et al. 2002. Synthesis, biological activity, and molecular modeling investigation of new pyrazolo[4,3-e]-1,2,4-triazolo[1,5-c]pyrimidine derivatives as human A(3) adenosine receptor antagonists. J. Med. Chem. 45:770–80 115. Cassada DC, Tribble CG, Kaza AK, Fiser SM, Long SM, et al. 2001. Adenosine analogue reduces spinal cord reperfusion injury in a time-dependent fashion. Surgery 130:230–35 116. Firestein GS, Boyle D, Bullough DA, Gruber HE, Sajjadi FG, et al. 1994. Protective effect of an adenosine kinase inhibitor in septic shock. J. Immunol. 152: 5853–59 117. Cronstein BN, Naime D, Ostad E. 1993. The antiinflammatory mechanism of methotrexate. Increased adenosine release at inflamed sites diminishes leukocyte accumulation in an in vivo model of inflammation. J. Clin. Invest. 92:2675– 82 118. Schijns VE. 2001. Induction and direction of immune responses by vaccine adjuvants. Crit. Rev. Immunol. 21:75–85 119. Chambers CA, Kuhns MS, Egen JG, Allison JP. 2001. CTLA-4-mediated inhibition in regulation of T cell responses: mechanisms and manipulation in tumor immunotherapy. Annu. Rev. Immunol. 19:565–94 120. Newby AC. 1984. Adenosine and the concept of retaliatory metabolites. Trends Biochem. Sci. 2:42–44

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:657-682. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104639

Annu. Rev. Immunol. 2004. 22:683–709 doi: 10.1146/annurev.immunol.22.012703.104639 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on December 12, 2003

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS Jaehyuk Choi,1 David R. Enis,1 Kian Peng Koh,1 Stephen L. Shiao,1 and Jordan S. Pober Interdepartmental Program in Vascular Biology and Transplantation, Boyer Center for Molecular Medicine, Yale University School of Medicine, New Haven, Connecticut 06536-0812; email: [email protected]

Key Words endothelial activation, endothelial dysfunction, antigen presentation, vascular biology, cytokines ■ Abstract Human vascular endothelial cells (EC) basally display class I and II MHC-peptide complexes on their surface and come in regular contact with circulating T cells. We propose that EC present microbial antigens to memory T cells as a mechanism of immune surveillance. Activated T cells, in turn, provide both soluble and contact-dependant signals to modulate normal EC functions, including formation and remodeling of blood vessels, regulation of blood flow, regulation of blood fluidity, maintenance of permselectivity, recruitment of inflammatory leukocytes, and antigen presentation leading to activation of T cells. T cell interactions with vascular EC are thus bidirectional and link the immune and circulatory systems.

INTRODUCTION Vascular endothelial cells (EC) form a one-cell-thick lining of the circulatory system, separating blood from tissues. T lymphocytes routinely come into contact with the surface of vascular EC under two circumstances. First, as T cells circulate in blood they pass through true capillaries where lumena are narrower than a T cell. Second, as circulating T cells enter the tissues, they attach to and transmigrate through EC of postcapillary venules. In humans, microvascular EC lining capillaries and venules express class I and class II MHC molecules, so whenever foreign (e.g., microbe-derived) peptides are present in endothelial MHC molecules, there is an opportunity to trigger T cell responses. We review the consequences of such antigen-specific encounters for both the T cell and EC. Specifically, we describe six regulatory functions of vascular EC, namely regulation of blood vessel formation and remodeling, blood flow, permselectivity, blood fluidity and hemostasis, 1

The first four authors are listed alphabetically and contributed equally to this work.

0732-0582/04/0423-0683$14.00

683

15 Mar 2004

15:36

684

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL.

leukocyte trafficking, and T cell activation and differentiation. In each instance, we describe effects of T cells and their products on these functions. Because EC in various vascular beds differ in their properties, we begin with a consideration of EC heterogeneity.

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

EC HETEROGENEITY Endothelial cells of all tissues display specific structural or functional specializations characteristic of arteries, capillaries, or veins (Figure 1). Arterial EC are efficient producers of vasoregulatory autacoids such as nitric oxide (NO), whereas venular EC express leukocyte adhesion molecules, such as P-selectin, E-selectin, and vascular adhesion molecule-1 (VCAM-1) (1–3). Venular EC also express higher levels of autacoid receptors (such as histamine receptors), rendering this portion of the microcirculation the primary site of fluid leak during inflammation (4). The basis of arterial versus venous EC differentiation remains unclear, but embryonic vessel remodeling requires expression of the arterial and venousspecific bidirectional signaling molecules Ephrin-B2 and EphB4 (5). Blood flow is undoubtedly another crucial determinant of vascular differentiation. Capillary EC form the major exchange surface between the blood and tissues and acquire tissue-specific features related to differences in the passage of macromolecules across the vasculature (6). These differences involve both intercellular junctions and transcellular pathways. For example, liver and (human) spleen have discontinuous capillaries, called sinusoids, that allow free passage of macromolecules from the blood into these tissues and vice versa. EC-lining glomerular capillaries (as well as liver sinusoidals) form patches of membrane devoid of cytoplasm, called fenestrae, which permit high levels of fluid exchange across the EC. In contrast, brain EC form electrically tight junctions that limit transendothelial passage of fluid and macromolecules except through a vesicular transport pathway originating in subplasmalemmal caveolae. Such tissue-specific specializations of EC are thought to be induced by signals provided by the local tissue. For example, astrocytes or astrocyte-conditioned medium causes EC to form tight junctions and acquire high electrical resistance characteristic of brain EC (7). Venular EC may also become specialized in different tissues. The most striking example of this occurs in lymph nodes and mucosal lymphoid tissue where certain venular EC acquire the high endothelial phenotype conducive to adherence and transmigration of na¨ıve lymphocytes. This subject has been extensively reviewed elsewhere and is not be discussed further (8). It has been proposed that flat venules from different tissues express unique collections of proteins, especially leukocyte adhesion molecules that signal a molecular address for tissue-specific homing of memory T cells (8). Aside from expression of MAdCAM in the EC of the gastrointestinal tract (9) and possibly E-selectin on chronically inflamed EC in the skin (10), there are few data to support this attractive idea.

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

P1: IKH

685

FUNCTIONS OF EC AND THEIR MODULATION BY T CELLS Although EC in different sites differ functionally as well as structurally, EC in all tissues must perform a number of common functions, several of which we review here.

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Blood Vessel Formation and Remodeling The growth and remodeling of the vascular system is governed by EC, which initiate new vessel growth through either vasculogenesis (aggregation of EC into tubes from dispersed precursor cells known as angioblasts) or angiogenesis (outgrowth of EC tubes from preexisting vessels) (11). Nascent EC tubes, formed by either mechanism, recruit investing cells (smooth muscle cells or pericytes) and remodel into an organized arterial-capillary-venous circulation using a tightly controlled process of selective vessel enlargement and pruning, involving EC growth and apoptosis. Vascular EC give rise to lymphatic EC through a process called lymphangiogenesis (11) and may also give rise to hematopoetic stem cells (12). Several growth factors mediate vasculogenesis, angiogenesis, or lymphangiogenesis during development. Foremost among these are the various isoforms of vascularendothelial cell growth factor (VEGF) and their receptors, especially VEGFR-1 (also called flt-1) and VEGFR-2 (also called flk-1 or KDR), which play critical roles in EC differentiation, survival, proliferation, and migration (13). Lymphatic EC, which initially develop from blood vessel EC, respond to a distinct VEGF isoform, VEGF-C and its receptor VEGFR-3 (flt-4) (14). In most tissues, the recruitment of smooth muscle cells and pericytes by developing vessels depends on EC secretion of platelet-derived growth factor B (PDGF-B), which binds to PDGF receptor β on mesenchymal cells (11). Angiopoietin-1, which binds the tie-2 receptor on EC, is required for the survival, maturation, and stabilization of mature blood vessels. The related factor angiopoietin-2 acts as a tie-2 antagonist under most circumstances, causing vessel destabilization that can result in angiogenic sprouting (in the presence of VEGF) or vessel regression (in the absence of VEGF). Transforming growth factor (TGF)-β generally acts to stabilize mature vessels, although the effect appears to be context specific (11, 15). Vessel formation normally ceases at birth but may be reactivated in several situations, including tissue injury, hypoxia, tumorogenesis, and chronic inflammation. Basic fibroblast growth factor (b-FGF or FGF-2) may be the principal mediator of angiogenesis in tissue repair (16) and can act cooperatively with VEGF (17). In hypoxic tissues, VEGF plays a primary role and its secretion is upregulated by the oxygen-sensitive transcription factor hypoxia-inducible factor (HIF)-1α (18). These responses occur in the absence of T cells (e.g., in SCID or Rag−/− mice), but T cells may play an important role in angiogenesis associated with chronic inflammatory conditions such as rheumatoid arthritis and atherosclerosis. Although T cells are not sources of classical vascular growth factors such as VEGFs, angiopoietins, or PDGFs, they may induce expression of these factors in mononuclear

12 Feb 2004

15:59

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

686

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL.

phagocytes. In addition, T cells can directly synthesize b-FGF and heparin-binding epidermal-like growth factor (HB-EGF), which are proangiogeneic (19). T cell– derived cytokines may also inhibit angiogenesis. For example, TGF-β, IFN-γ , and TNF are negative regulators of EC growth in culture (15, 20, 21), and the antiangiogenic properties of IFN-γ have been demonstrated in vivo (22). Paradoxically, TNF may stimulate angiogenesis in vivo (23), although this effect may be indirect, mediated through proangiogenic signals released by recruited leukocytes or by EC. For example, TNF may stimulate production of sphingosine-1-phosphate in EC (24), which is proangiogenic, through interactions with Edg family receptors (25). Much recent attention has focused on chemokines (26), especially CXC-family chemokines containing a Glu-Leu-Arg (“ELR”) tripeptide motif, which have been reported to be proangiogenic (27). CXC chemokines lacking this motif (especially the IFN-γ -inducible factors IP-10, Mig, and I-TAC) may have angiostatic action. Chemokine receptor expression and responses may vary among EC types (28, 29). EC death is closely associated with vessel remodeling and angiogenesis. TNF is the most potent inducer of apoptosis in many cell types, but EC are normally resistant to killing by TNF unless a cosignal is also delivered. TNF sensitizers may act by interfering with intracellular signaling/transcriptional activation pathways that mediate cell survival, and TNF killing can use either caspase-dependent or -independent mechanisms (30, 31). Activated T cells can also cause cell-contact-dependent apoptosis of EC through surface-bound proteins including TNF itself (which is synthesized as a membrane protein), FasL, and TRAIL. Similar to TNF-induced apoptosis, killing via FasL appears to require sensitization of the endothelium, e.g., by IFN-γ -induced increases of Fas and procaspase-8 (32). In contrast, TRAIL can induce apoptosis in unsensitized EC (33). Activated T cells can also kill EC via pathways independent of cytokine and cell surface ligands, e.g., cytolytic T lymphocytes (CTL) kill EC via a granule exocytosis pathway involving perforin and granzymes (34). T cell–mediated lysis of EC could play a role in the angiogenesis of immune inflammation or in a recently described form of limited microvascular remodeling “microangioectasia” associated with chronic T cell recruitment (35).

Control of Blood Flow Arterial and arteriolar EC control blood flow by reducing trophic vascular smooth muscle tone through release of autacoids such as NO, prostacyclin (PGI2) or endothelium-derived hyperpolarizing factor (EDHF) (36). NO plays a dominant role in vasorelaxation, especially in large conduit arteries, whereas EDHF may predominate in smaller vessels (37). The identity of EDHF varies depending on species and tissue; in human coronary arteries, EDHF appears to be a cytochrome P450-derived epoxyeicosatrienoic acid (EET) (38). Impaired EC-mediated vasodilatation, commonly referred to as endothelial dysfunction, is frequently associated with inflammation and elevated cytokine levels. TNF and IFN-γ rapidly induce cultured EC to synthesize tetrahydrobiopterin, a limiting cofactor for eNOS conversion of L-arginine to NO and L-citrulline, and increase NO production (39). At later times, these cytokines diminish NO release

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

P1: IKH

687

by destabilizing eNOS transcripts (40). Inflammatory cytokines may also reduce NO bioavailability by enhancing EC production of superoxide, which inactivates NO (41). TNF can further inhibit endothelial NO production by reducing protein kinase B (Akt)-mediated phosphorylation of eNOS (42). The clinical significance of TNF actions is underscored by recent findings that anti-TNF therapy can reverse endothelial dysfunction in heart failure patients (43). In contrast to the inhibitory effects on eNOS, TNF enhances PGI2 production in EC through induction of cyclooxygenase 2 (COX-2) and cytosolic phospholipase A2 (44). Increases in phospholipase A2 may also enhance EET production. In addition, human peripheral blood lymphocytes can stimulate PGI2 synthesis in EC by contact-dependent signaling independent of cytokine release (45). TRAIL can increase prostanoid production through NO-dependent activation of COX-1 (46).

Regulation of Permselectivity Capillary EC normally regulate passage of macromolecules between blood and tissues. In tissues where EC are continuous, movement of proteins across the endothelium is based on size and charge. This property, called permselectivity, is regulated by the tightness of EC-EC junctions, by the level of vesicular trafficking through the caveolar system, or possibly by specialized transcellular structures termed vesiculo-vacuolar organelles (VVOs) (47, 48). EC vesicles may also reverse transport molecules, e.g., chemokines, from tissue into the bloodstream (49). Endothelial permselectivity can be lost in the presence of cytokines, leading to tissue edema and deposition of fibrin and other plasma proteins in the interstitium. T cell–derived cytokines may modulate permselectivity directly by acting on EC receptors, or indirectly through recruitment of inflammatory cells (such as neutrophils or macrophages), which can damage EC (50). IFNγ and TNF directly induce cytoskeletal rearrangement in cultured EC, stimulating formation of gaps between cells, and loss of permselectivity (51). TNF or IL-1 induces an increase in permeability beginning within 1–3 h of cytokine exposure, which is reversed 12–36 h after withdrawal (52). This response is slower in onset but more prolonged than the wheal-and-flare induced by autacoids (such as histamine), and may underlie the vascular leak syndrome caused by therapeutic administrations of cytokines such as IL-2 (53). There is also evidence for T cell–contact-dependent vascular leak not reproduced by cytokines (54) of which the mechanism is unknown. During inflammation, fluid leakage is initially confined to postcapillary venules, but later involves the capillaries as well (50).

Regulation of Fluidity and Hemostasis EC normally prevent coagulation and maintain blood fluidity through several mechanisms: (A) binding and activation of antithrombin III on cell surface heparan sulfates (55); (B) inhibition of tissue factor activity by display of tissue factor pathway inhibitor (56); (C) conversion of thrombin, through binding to

12 Feb 2004

15:59

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

688

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL.

thrombomodulin, from a procoagulant activator of fibrinogen to an anticoagulant activator of protein C (57); and (D) initiation of fibrinolysis by secretion of plasminogen activators (PAs) (58). These properties may be enhanced by EC-derived autocoids—for example, NO and PGI2 inhibit platelet activation (59) and 11,12EET promotes tissue-type PA release (60). T cell–derived cytokines can convert EC from an antithrombotic to a prothrombotic state. In particular, TNF causes EC to synthesize procoagulant proteins such as tissue factor (TF) and plasminogen activator inhibitor-1 (61). At the same time, TNF diminishes the expression of thrombomodulin by suppressing gene transcription (62). These effects can also be mediated by T cell contact involving membrane TNF or CD154 (63, 64). (CD154, also referred to as CD40L, is expressed on activated CD4+ T cells and engages CD40 on EC.) Cytokines and contact-dependent signaling pathways initiate transcription of TF through AP-1 and NF-κB; CD154 also activates Egr-1 and is a more potent inducer of TF (65). The physiological synergy between activation responses, such as generation of TF, and dysfunctional responses such as loss of thrombomodulin, enhance fibrin deposition on the EC surface and may underlie pathophysiological processes such as intravascular thrombosis in the Shwartzman reaction (66). Apoptosis of EC may further promote coagulation by de-encrypting TF and by releasing membrane vesicles that can assemble a prothrombinase complex (67).

Regulation of Leukocyte Trafficking and Inflammation Resting EC generally do not interact with circulating leukocytes because they lack surface molecules that can initiate tethering, e.g., selectins or VCAM-1. Moreover, basal production of NO actively inhibits leukocyte adhesion and activation (68). EC-derived EETs may also have anti-inflammatory properties (69). Maintenance of adherens and of tight junctions in resting EC may further restrict leukocyte passage between EC (70). In the presence of cytokines or CD40 ligation by CD154+ T cells, EC acquire capacity to bind and activate leukocytes by enhanced expression of leukocyte adhesion molecules [E-selectin, intercellular adhesion molecule1 (ICAM-1), and VCAM-1] and of chemokines. This topic has been extensively reviewed elsewhere and is not discussed further here (71). EC that have been activated by TNF to promote nonspecific inflammation may be further modified by T cells and their cytokines (e.g., IL-6 or IL-4) to favor specific types of inflammation, e.g., those characteristic of Th1 or Th2 responses (72, 73).

Regulation of T Cell Activation and Differentiation Endothelial cells, which form the interface between the blood and the underlying tissue, are uniquely positioned to alert the peripheral pool of circulating memory T cells to the presence of a foreign pathogen. We hypothesize that infection leads to local acquisition and presentation of pathogen-derived peptides on both MHC class I and MHC class II molecules at the EC surface. Circulating memory T cells, which have frequent and intimate contact with endothelial membranes

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

P1: IKH

689

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

in the lumen of microvessels, can thus recognize the presence of pathogens in tissues before they leave the bloodstream. In other words, we propose that EC behave as sentinels, presenting antigen so as to initiate a secondary immune response. Consequently, in transplantation, graft EC, which express allogeneic MHC molecules and are viewed by the immune system as infected, may initiate rejection reactions. In the following sections we review the evidence that support this proposal. FORMATION OF PEPTIDE-MHC COMPLEXES ON EC MHC expression patterns show species differences. In humans, vascular EC basally express both MHC class I and MHC class II molecules (74), and in nonlymphoid tissues, the expression of MHC molecules on EC is much greater than on other cells, such that immunostaining for MHC class I and class II molecules primarily identifies vascular EC (75, 76). Rodent EC basally express high levels of class I molecules but not class II molecules (77, 78). MHC expression may vary among tissues. Class II MHC molecules are observed on EC throughout the human microvasculature as well as veins, but arterial EC expression varies with anatomic location (79). T cells regulate the level of MHC expression on EC. In fact, IFN-γ or some other activating signal underlies “basal” expression levels of these molecules. The evidence for this assertion derives from several observations: (A) cultured EC lose class II MHC expression completely and markedly reduce class I expression, although expression levels may be sustained or restored with IFN-γ treatment (80, 81); (B) similarly, transplantating human or pig coronary artery segments into immunodeficient mice (which makes mouse IFN-γ that does not act on human or pig cells) leads to disappearance of basal class II MHC and marked reduction in class I MHC expression that can be restored by species-appropriate IFN-γ injection (82); (C) treatment of dogs with cyclosporine, which inhibits IFN-γ production, can reduce MHC molecule expression on EC in vivo (83) but does not inhibit IFNγ -mediated induction on EC in vitro (84); and (D) genetic knockout of IFN-γ or IFN-γ receptors in mice reduces basal levels of class I MHC expression (85). In primates, inflammation or treatment with IFN-γ causes further upregulation of MHC molecules above basal levels (86). Following IFN-γ treatment, other cell types may also express MHC molecules but generally at lower levels than EC do (87). In mice (88) but not in rats (89), inflammation is sufficient to induce expression of class II MHC molecules on EC that do not otherwise express them. Some variant MHC molecules such as CD1a, b, and c may also be expressed by EC, but their regulation is less clear (90). Surface expression of both MHC class I and class II molecules requires the expression of a variety of accessory proteins that mediate assembly and peptide loading. Molecules responsible for generation of peptide antigens and peptide loading of class I molecules within the ER (e.g., LMP2, 7, and TAP1, 2, respectively) are induced in EC by the same cytokine signals (e.g., IFN-γ , TNF, IFN-α/β) as those that induce the class I structural chains, but induction of these accessory molecules is faster than that of class I chains because their synthesis is controlled

12 Feb 2004

15:59

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

690

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL.

by preexisting transcription factors, e.g., STATs or NF-κB, instead depending on transcription factors that must be newly synthesized (e.g., Interferon Regulatory Factor-1 (IRF-1), which controls class I structural genes) (91, 92). Proteins that act as accessory molecules for class II assembly, e.g., invariant chain or HLA-DM, are induced concomitantly with class II structural genes in response to IFN-γ following a several hour delay (93). This response depends on newly synthesized class II transactivator (CIITA) (94). The three polymorphic class I loci (HLA-A, B, and C) and the class II loci are quantitatively different in their responses to various cytokines (95), but each class I locus product is expressed at levels sufficient for recognition by CD8+ T cells in untreated as well as cytokine-treated EC (95, 96). In contrast, IFN-γ -treated EC express much greater levels of HLA-DR and HLA-DP than HLA-DQ (93), and it is unclear if EC can interact with HLA-DQ-restricted CD4+ T cells. Cultured human EC acquire protein antigens, process them in a chloroquineinhibitible manner, and present peptide-MHC complexes that can activate peptidespecific T cell clones (97). Interestingly, murine EC appear capable of performing cross presentation of exogenously acquired antigens on MHC class I molecules, a property usually ascribed solely to dendritic cells. Specifically, EC isolated from the pancreatic vasculature but not from elsewhere are able to activate insulinspecific CD8+ T cells (98). COSTIMULATION BY EC Costimulation necessary for activation of resting T cells may be mediated by three types of molecules: signaling molecules (true costimulators), adhesion molecules, and activating cytokines. B7 molecules are the major costimulatory molecules in mice (99). It is unclear whether LFA-3 (which binds CD2) or B7 (which binds CD28) is more important for human T cell responses. In vitro blocking experiments of these pathways in humans generally demonstrate quantitatively similar levels of inhibition of T cell responses (100), and experiments using LFA-3 versus B7-transduced fibroblasts indicate that LFA-3 is more potent for breaking T cell anergy (101). Immune inflammation in vivo may be reduced either with a CTLA4-Ig fusion protein that prevents CD28 engagement by B7-moelcules (102) or with an LFA-3-Ig fusion protein that prevents engagement of CD2 by LFA-3 (103). These clinical observations are complicated because CTLA4-Ig may deliver inhibitory signals to APC (rather than simply blocking engagement of CD28 by B7 molecules) (104), and the LFA-3-Ig reagent used can deplete CD2+ cells (105). Human EC constitutively express LFA-3 (106) but do not generally express either B7-1 or B7-2 (107). There are some possible exceptions: Cultured cardiac microvascular EC have been reported to express B7-2 (108), and cultured brain microvascular EC have been reported to express B7-1 and B7-2 following activation by cytokines (109, 110). To date however, B7 molecules have not been described on human EC in situ. Cultured human EC can enhance T cell activation via the CD2 but not the CD28 pathway unless they have been transfected or transduced to express exogenous B7 molecules (111).

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

P1: IKH

691

An additional important costimulatory pathway involves CD40 and its ligand CD154 (112). In this system, T cells express CD154, and the APC expresses the signaling receptor (CD40). Surprisingly, costimulation primarily affects the T cell, yet there are no biochemical data demonstrating that CD154 can deliver an “outside-in” signal. CD154 on T cells could activate APC, which in turn provide signals (e.g., B7 molecules) that costimulate T cells, but no other signaling system has been found to be necessary for the CD40 effect. EC from humans (113), mice (112), and pigs (114) express CD40; and CD40 expression in humans is increased by treatment with TNF, IL-1, and IFN-γ (113). EC, compared with other APC, accelerate and enhance the activation-dependent induction of CD154 on T cells, largely via LFA-3 (115). Interestingly, activated EC may also express CD154 in vitro (116) and in vivo (117, 118). This may contribute to costimulation, because T cells express low but functional levels of CD40 (119). Several new costimulatory molecules have been recently identified. These include three B7-related molecules that preferentially react with memory T cells, namely inducible costimulator ligand (ICOS-L, also called GL-50, B7RP-1, B7h, or B7-H2), which interacts with ICOS (also called AILIM) on T cells, as well as Programmed Death Ligand 1 (PDL-1, also called B7-H1) and PDL-2 (also called B7-DC), both of which interact with PD-1 on T cells. ICOS signals are thought to be costimulatory. EC express ICOS-L and the level of expression can be increased by TNF or IL-1 (120). On the other hand, PD-1 signals are thought to be inhibitory. PDL-1 and -2 are expressed on EC, and expression is increased by IFN-γ (121, 122; A. Meszaros and J. Pober, unpublished data). There is little known about the role of these molecules on EC in vivo. Several other novel costimulators appear to belong to the TNFR and TNF superfamilies. Examples expressed by EC include 4-1BB and TL-1 or its variant TL-1A (123–125). Several molecules involved in development may also provide costimulation to T cells. These include contact guidance molecules, such as neuropilins and semaphorins (126), members of the ephrin/Eph family (127), members of the Wnt pathway (128), and members of the Notch pathway (129, 130). EC express molecules in all of these families and therefore have the potential to use them to influence T cell activation. To date, their roles in EC-mediated T cell activation are less well established than those of classical costimulators. Adhesion molecules may play two roles in T cell activation. First, leukocyte integrins such as LFA-1 (also known as α Lβ 2 integrin) and very late activation antigen-4 (VLA-4, also known as α 4β 1 integrin), which bind to ligands on EC, can signal outside-in (131–133). In particular, engagement of LFA-1 plus TCR appears to activate the acidic sphingomyelinase and JNK signaling pathways more efficiently than does engagement of TCR alone (134) and may stabilize cytokine transcripts (135). However, costimulation via these ligands does not seem to be sufficient to replace the more traditional CD28/CD2-mediated costimulatory pathways (136). Second, adhesion molecule engagement may play a nonredundant role in optimal spatiotemporal positioning of the TCR and signaling molecules. In the absence of integrin ligands such as fibronectin or vitronectin, T cells cannot

12 Feb 2004

15:59

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

692

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL.

polarize properly and interact poorly with surfaces coated with anti-CD3 antibodies with or without anti-CD28 (137). We return to this topic when we address the role of the immune synapse in T cell activation. Non-integrin adhesion molecules may also mediate T cell/EC interactions. Na¨ıve and central memory T cells that home to lymph nodes express L-selectin, which can recognize carbohydrate determinants expressed on high endothelium of lymphoid organ venules. EC selectins, including E-selectin and P-selectin, can interact with ligands on T cells, which are largely attached to P-selectin glycoprotein ligand-1 (138). E-selectin ligands, sometimes called cutaneous lymphocyte antigen-1, have been noted to be enriched on skin-homing T cells or on TH1 T cells, and can be induced on T cells exposed to IL-12 (10, 139, 140). T cells and EC both express various isoforms of CD44, a receptor for hyaluronan, that can also influence T cell activation (141). Cytokines represent a third class of molecules that may contribute to T cell costimulation. EC make IL-1α when treated with proinflammatory molecules, e.g., LPS, TNF, or IL-1 itself (142). In conjunction with a TCR-directed signal, e.g., via the lectin Concanavalin A, EC-derived IL-1 enhances the proliferation of thymocytes and Th2 T cell clones (143, 144), although the role of endothelial-derived IL-1α activation of resting mature T cells remains unclear. EC may also secrete IL-15, a T cell and NK cell mitogen, whose production is generally attributed to stromal cells (145). EC have been reported to make a number of cytokines that have been shown to influence the differentiation of activated T cells while not necessarily acting as T cell mitogens or costimulators. These include IFNα/β (146), IL-6 (147), IL-11 (148), IL-12 (149), and IL-18 (150). Like most cells, EC can produce IFN-α/β when challenged by virus or double-stranded RNA (146), but it has not been shown whether they produce sufficient IFN-α/β to influence T cell responses (151). Cultured EC have been reported to make IL-11 (148) that may favor TH2 differentiation (152). EC may also synthesize IL-12 and IL-18, which promote TH1 differentiation (149, 150), and IL-6 (147) which may favor TH2 differentiation. These and other EC-derived cytokines may augment the T cell activation in response to MHC-peptide complexes presented by another cell (153). This could occur in the secondary lymphoid organs or in the periphery. For example, the TNF family member TL-1A secreted by EC may increase IL-2 responsiveness in T cells, in part by upregulating CD25 expression (125). In addition to being able to directly activate resting effector cells, EC also express molecules relevant for the downregulation of regulatory T cell (Treg) activity, i.e., glucocorticoid-inducible TNF receptor-ligand (GITR-L) (154) and IL-6 (147). Although the roles of endothelial-derived GITR-L and IL-6 are unclear, these molecules could inhibit Treg and facilitate the initiation of an effector response (155, 156). EC/T CELL CONTACT REGIONS Effective activation of T cells by conventional APCs may require the formation of a specialized attachment region called an immune synapse. On professional APC, the immune synapse evolves so that adhesion

15 Mar 2004

15:36

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

P1: IKH

693

molecules become concentrated at its edge (forming a peripheral supramolecular activation complex or p-SMAC) while TCR/MHC complexes and costimulator molecules are moved to the center (forming a central supramolecular activation complex or c-SMAC) (157). Immune synapses may serve to stabilize adhesion and extend the duration of bidirectional signaling between the APC and the T cell (158, 159). Some, but not all, APC types actively participate in forming the immune synapse (160, 161). For example, applying an actin depolymerization agent (i.e., cytochalasin) to B cells has little influence on either immune synapse formation or T cell activation (159, 161). In contrast, dendritic cells polarize their cytoskeleton during T cell activation, and blocking this polarization, i.e., by deleting a cytoskeletal component like fascin, abrogates subsequent T cell activation (162). The cell biology of the contact region between EC and T cells has unique features because EC have a specialized capacity to capture circulating T cells as part of the process of diapedesis. This capacity has been analyzed in detail for TNF-treated EC, which have increased levels of ICAM-1 and newly synthesized VCAM-1 (163). When placed in coculture, the EC start a process that resembles phagocytosis of the bound T cell. Micovilli-like structures emerge from the apical membrane of the EC and form a cup-like extension that appears to engulf the T cell. Both ICAM-1 and VCAM-1 are concentrated in this cup, which has been referred to as a “docking structure” and is enriched in ezrin-moesin linker proteins as well as cytoskeletal components, i.e., vinculin. The EC plays an active role in the formation of the docking structure; these membrane changes depend upon Rho and probably phosphoinositide signaling within the EC. The docking structure may stabilize adhesion in the face of shear stress and serve as a prelude for transendothelial migration, either through the junction or through the EC cytoplasm. Preliminary results from our laboratory indicate a mature immune synapse, involving segregation of the c-SMAC and p-SMAC, can develop within the region of contact formed by a docking structure when a relevant signal, e.g., superantigen, is present. A comparsion of the EC docking structure to an immune synapse is diagrammed in Figure 2. Studies using connexin inhibitors have suggested a role for gap junctions in T cell interactions with B cells (164). Interestingly, EC can form heterotypic gap junctions with T cells that can facilitate the intercellular transfer of ions and signaling molecules (165). Cytoplasmic dye transfer assays demonstrate that EC and T cells form heterotypic gap junctions in as little as 30 minutes up to a few hours (165, 166). Structures interpreted as gap junctions have been observed in vivo between T cells and EC during the course of autoimmune inflammation (167). Whether or not T cell/EC heterotypic gap junctions actually influence the capacity of class II MHC–expressing EC to activate T lymphocytes requires further elucidation. EVIDENCE OF EC IMMUNOLOGICAL FUNCTIONS IN VITRO It is difficult to assess the capacity of human APC to present antigens to unprimed T cells because the frequency of mature T cells that respond to any one specific antigen upon first

12 Feb 2004

15:59

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

694

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL.

encounter is very low, i.e., less than 1 in 106. In mice, this can be addressed by expressing a particular TCR as a transgene so that all, or almost all, na¨ıve T cells can be activated by the same antigen. In humans, there are several alternative experimental strategies to examine antigen responses. These include examining memory responses (e.g., in vaccinated individuals) in which the frequency of peptide-specific T cells reactive with the immunogen has greatly increased, i.e., up to 1 in 105 or more. Human EC that have been treated with IFN-γ to express MHC class II molecules can present recall antigens to memory CD4+ T cells (168). Human EC can also stimulate T cell clones (169). Similar experiments using rat EC have been unsuccessful, whereas mouse EC can activate some T cell clones but not others (170–172). A second strategy has been to examine direct allogeneic responses, taking advantage of the fact that TCRs specific for many different foreign peptides bound to self-MHC molecules can react with self-peptides bound to the same allelic form of a nonself-MHC molecule. Consequently, the frequency of T cells that can be activated by an allogeneic APC in a na¨ıve host is quite high and can exceed 1% of the total T cell population. In mice, this alloresponse is largely mediated by na¨ıve T cells, but in humans, this alloresponse is largely mediated by memory T cells. This is because mice live a short time and most na¨ıve T cells never encounter their cognate antigen to become memory cells, whereas adult humans encounter many nonself (largely microbial) antigens over decades such that more than half of circulating T cells are memory cells (173, 174). Significantly, human memory but not na¨ıve CD4+ T cells respond to allogeneic EC by secreting IL-2, expressing IL-2Rα (so as to form the high-affinity IL-2 receptor complex), and proliferating (111, 175). The number of T cells that respond to EC by producing IL-2 is less than that stimulated by professional APC expressing the same MHC molecules, and the proliferative response is smaller (111). It has been proposed that in the absence of CD80 or CD86, memory T cells may be induced to transmigrate rather than secrete cytokines and proliferate (176, 177), allowing recognition of antigen on EC to be a signal for extravasation. This conclusion has been disputed (177). The responses of human CD8+ T cells are also separable into unresponsive na¨ıve cells and responsive memory cells (178). In mice, CD8+ T cells but not CD4+ T cells respond to allogeneic EC cultures. This is true even when the mouse EC have been treated with IFN-γ to express class II MHC and CD86 (179). Two points warrant further elaboration. First, why do only memory cells respond to EC? It has been noted that memory T cells have less-stringent requirements for activation than do na¨ıve T cells (180). This makes teleological sense in that overreactive na¨ıve T cells would increase dangers of autoimmunity, whereas underractive memory cells would limit the advantage of a rapid and enhanced response to pathogens conferred by acquisition of memory. Thus, na¨ıve T cells could fail to respond to EC because EC provide inadequate costimulation. Alternatively, na¨ıve T cells may not be able to form stable conjugates with EC because, unlike memory T cells, they lack the adhesion molecules necessary to bind to peripheral (or cultured) EC.

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

P1: IKH

695

Second, why is the frequency of EC-responsive T cells lower than that of professional APC-responsive T cells even among the memory subset? It is possible that the EC simply express fewer cross-reactive allogeneic MHC molecules owing to a more restricted set of self peptides displayed on these molecules. However, we found that T cell clones stimulated by allogeneic BLC consistently respond to restimulation with EC from the same donor (181). An alternative explanation is that EC not only stimulate but also simultaneously suppress T cell activation so that only the most strongly stimulated T cells can respond. Indeed, inhibition can be directly observed in mixing experiments using EC and BLC as stimulators (181, 182). Unexpectedly, whereas BLC-stimulated clones recognize the allogeneic MHC molecules on both the EC and the BLC, EC-stimulated clones recognize allogeneic MHC molecules on EC but not on BLC autologous to the EC (182). In fact, adding EC to the BLC/T cell cocultures suppressed the anti-BLC response while promoting the anti-EC response (181). EC-stimulated CD8+ clones express a distinct panoply of cell surface molecules (including stable expression of CD69, CD25, and CD154) from BLC-stimulated clones. Functionally, these unusual T cells are still effector cells that express perforin and mediate cytolysis of targets (178). A final approach to address the problem of studying the low frequency of antigen-responsive human T cells is to use polyclonal T cell activators. These agents mimic antigen in that full T cell activation still depends upon accessory signals provided by the APC (183). Three polyclonal activators are commonly used: bacterial superantigen, anti-CD3 antibodies, and plant-derived mitogens. Bacterial superantigen is probably the most physiological because it works by binding both to MHC class II molecules on the APC and to specific germ-line encoded portions of TCR β chains on the T cell. Each bacterial superantigen reacts with approximately 5%–10% of the total T cell population, i.e., the subpopulations that utilize the relevant Vβ gene segments (184). Cultured human EC that express MHC class II molecules will present superantigens to T cells (183). In contrast, anti-CD3 antibodies generally cannot be used to activate T cells when EC are used as accessory cells. This is because anti-CD3 antibodies need to be anchored on some substrate or bound via FcR on the APC (185), and EC generally lack FcR. EC do support anti-CD3 modulated T cell activation if the antibody is presented bound to a bead (185), or if the EC are transduced to express FcR (T. Dengler and J. Pober, unpublished data). EC readily support activation and proliferation of T cells activated by mitogenic plant-derived lectins such as concanavalin A or (for human T cells) phytohemagglutinin A (PHA). PHA not only clusters TCRs, but also forms bridges between the T cell and the EC. Consequently, the PHA response does not require MHC molecule expression on the EC and cannot be inhibited by interfering with LFA-1 or ICAM-1 (186). In the presence of PHA, EC can also activate na¨ıve T cells (111). EC increase T cell proliferation at suboptimal doses of PHA and markedly augment cytokine production even at supraoptimal doses of PHA. Approximately 50% of the costimulation provided by EC to PHA-activated and superantigen-stimulated T cells is mediated via LFA-3/CD2 interactions (111, 183). The identities of the other significant costimulators in these systems remain

12 Feb 2004

15:59

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

696

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL.

unclear, although Wnt signals contribute to the capacity of EC to render PHAactivated T cells resistant to cyclosporine (128). PHA has been used to elucidate intracellular signals in T cells generated in response to EC-mediated costimulation. EC activate multiple transcription factors, stimulating a 2- to 3-fold upregulation of luciferase activity driven from AP-1, NFAT, and NF-κB response promoters (187). EC influence both the amount and composition of AP-1 that is formed, predominantly by augmenting de novo synthesis of c-Fos (188, 189). Furthermore, EC increase NFAT activity by calcineurinindependent inhibition of nuclear export (128). Endothelial cells also stabilize certain mRNA species in activated T cells (187). For example, whole EC augment the de novo synthesis of IL-2 and IFN-γ mRNA primarily by increasing transcription, whereas increased synthesis of IL-4 and CD154 is predominantly accomplished through mRNA stabilization (169, 190; W. Ma and J. Pober, unpublished data). PHA can also be used to study the effects of EC on CD4+ T cell differentiation. Human blood monocytes, through the secretion of IL-12, drive both na¨ıve CD4+ T cells and unpolarized memory CD4+ T cells to a Th1 phenotype, favoring IFN-γ production at the expense of IL-4 production. In contrast, human EC do not alter the Th1/Th2 balance, retaining both IFN-γ and IL-4 synthesis (111). Interestingly, mouse EC have been described to favor Th2 differentiation, consistent with their selective capacity to activate and expand Th2 clones (171). EVIDENCE OF EC ANTIGEN PRESENTATION IN VIVO The assessment of EC immunological functions in vivo is technically challenging, and it has been difficult to produce definitive evidence for antigen presentation. Nevertheless, transplant and autoimmunity models in mice have generally supported the hypothesis that EC can use antigen presentation to activate and recruit resting memory T cells in vivo. The absence of class II molecules on rat EC, even during inflammation, has been proposed as an explanation of why depletion of passenger leukocytes (dendritic cells or DC) permits transplantation of rat kidneys across MHC barriers (191, 192), whereas depletion of passenger leukocytes from human kidneys has no impact on allograft rejection (193). Even in rats, infusion of IFNγ -treated cultured EC (but not untreated EC) may substitute for DC in inducing the rejection of previously stable, leukocyte-depleted kidney grafts (89). In mice, vascularized cardiac allografts may be acutely rejected via direct recognition of foreign class I molecules by CD8+ cells, even when the grafts have been made chimeric such that hematopoetic cells (DC) are syngenic to the recipient (194). This shows that some nonhematopoetic allograft cells, most likely EC, effectively stimulate an allograft response. EC may also participate in indirect allorecognition, a phenomenon in which T cells respond to allogeneic proteins that have been processed into peptides and presented, similarly to microbial peptides, bound to self-MHC molecules on self-APC. In this case, CD8+ T cells have been found to target skin graft peptides displayed by host EC (195). There is also evidence of preferential recruitment of CD8+ T cells specific for an antigenic peptide in tissue, indirectly implicating EC as providing the homing signal (98).

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

P1: IKH

697

The evidence for antigen presentation by EC in humans is even more indirect. The best such evidence involves the kinetics of antigen recognition in sensitized hosts. In primates, intradermal injection of PPD antigen elicits a response that is detectable by microscopy within hours in sensitized but not na¨ıve test subjects (196). This finding implies either that antigen-specific T cells were resident at the injection site, that resting PPD-specific T cells are constantly fluxing through the injection site at a very high rate, or that PPD-specific memory T cells can be selectively activated within the microcirculation, e.g., by antigen-recognition on local EC. The last explanation seems to be the most likely, but no critical test of this idea has been designed. A final piece of evidence comes from transplantation. Both rejecting human kidneys and hearts harbor CTL that are specific for EC (197, 198), and it is likely that EC were involved in the initial activation and differentiation of these effector T cells. T CELL EFFECTS ON ENDOTHELIAL CELL IMMUNOGENICITY T cell–derived cytokines influence the expression levels of molecules relevant to their antigen presentation capacity, i.e., MHC-peptide complexes, costimulatory molecules, and adhesion molecules. As mentioned previously, T cell–derived cytokines, i.e., TNF and IFN-γ , upregulate the molecules involved in antigen processing, i.e., TAP-1, and both cytokines can upregulate the expression of MHC class I molecules (92, 199). Furthermore, IFN-γ is the only known cytokine inducer of MHC class II molecules on EC and is necessary for their constitutive and inducible expression (200). In addition, IFN-γ , TNF, and IL-1 have been shown to upregulate the costimulatory molecules of the B7 family, including ICOS-L (120) and the inhibitory molecule PD-L1 (201). These same T cell–derived cytokines also upregulate the costimulatory molecules of the TNF receptor superfamily, i.e., CD40 on EC (113). TNF and IL-1, but not IFNγ , specifically upregulate the expression of members of the TNF superfamily, including TL1 and its variant, TL-1A (125). Lastly, proinflammatory cytokines induce the expression of adhesion molecules on EC (i.e., E-selectin, VCAM-1, and ICAM-1) that may also contribute to memory T cell activation by costimulation of stabilizing adhesion. In addition to cytokine-mediated effects, there is some evidence that T lymphocytes also mediate EC immune responses through contact-dependent signaling pathways. CD8+ T cells (or NK cells) can induce class II expression on EC by a pathway that is independent of IFN-γ and of CIITA (202). T lymphocytes stimulated with a combination of PHA and PMA can through cell-cell contact upregulate human brain microvascular EC expression of ICAM-1, VCAM-1, and E-selectin, as well as release of IL-6 and IL-8 (203, 204). This may involve CD154 because CD40 signaling leads to upregulation of E-selectin, VCAM-1, and chemokines in both porcine and human EC, as well as CD86 in porcine but not human EC (114). Although many effects may be CD40 mediated, there is recent evidence that some of the immune capacity of EC is also mediated by some unknown, CD40-independent, contact-dependent signal from T cells (205).

15 Mar 2004

15:38

698

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL.

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONCLUDING REMARKS In their strategic position at the interface between blood and tissue, EC are involved in regulating several key physiological functions, both immune and nonimmune. EC are also ideally positioned to come into frequent contact with circulating T cells. In this review, we describe both how EC can activate T cells by presenting antigens (a function we believe contributes to immune surveillance for pathogens) and how T cells can modify crucial EC functions, including antigen presentation (Figure 3). We have also noted that EC are heterogeneous and show significant differences among species. This degree of variability should be considered when attempting to extrapolate results observed in one tissue or species to another. Despite this complexity, there are now considerable data supporting the conclusion that there are extensive bidirectional interactions between T cells and EC relevant for understanding immunity and pathophysiology. ACKNOWLEDGMENTS The authors are supported by grants from the NIH (HL-36003, HL-51014, HL62188, and HL-70295 to JSP), the Medical Scientist Training Program (GM-07205 to JC, DRE, and SLS), and a graduate scholarship from the Agency for Science, Technology, and Research in Singapore (to KPK). The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. McEver RP, Beckstead JH, Moore KL, Marshall-Carlson L, Bainton DF. 1989. GMP-140, a platelet alpha-granule membrane protein, is also synthesized by vascular endothelial cells and is localized in Weibel-Palade bodies. J. Clin. Invest. 84:92–99 2. Cotran RS, Gimbrone MA Jr., Bevilacqua MP, Mendrick DL, Pober JS. 1986. Induction and detection of a human endothelial activation antigen in vivo. J. Exp. Med. 164:661–66 3. Rice GE, Munro JM, Corless C, Bevilacqua MP. 1991. Vascular and nonvascular expression of INCAM-110. A target for mononuclear leukocyte adhesion in normal and inflamed human tissues. Am. J. Pathol. 138:385–93 4. Heltianu C, Simionescu M, Simionescu

N. 1982. Histamine receptors of the microvascular endothelium revealed in situ with a histamine-ferritin conjugate: characteristic high-affinity binding sites in venules. J. Cell Biol. 93:357–64 5. Wang HU, Chen ZF, Anderson DJ. 1998. Molecular distinction and angiogenic interaction between embryonic arteries and veins revealed by ephrin-B2 and its receptor Eph-B4. Cell 93:741–53 6. Petzelbauer P, Schechner JS, Pober JS. 2003. Endothelium. In Fitzpatrick’s Dermatology in General Medicine, ed. IM Freedberg, AZ Eisen, K Wolff, KF Austen, LA Goldsmith, SI Katz, pp. 216– 29. New York: McGraw-Hill Health Professions Division 7. Janzer RC, Raff MC. 1987. Astrocytes induce blood-brain barrier properties

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

8.

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

9.

10.

11. 12.

13.

14.

15.

16.

17.

in endothelial cells. Nature 325:253– 57 Butcher EC, Picker LJ. 1996. Lymphocyte homing and homeostasis. Science 272:60–66 Berlin C, Berg EL, Briskin MJ, Andrew DP, Kilshaw PJ, et al. 1993. Alpha 4 beta 7 integrin mediates lymphocyte binding to the mucosal vascular addressin MAdCAM-1. Cell 74:185–95 Berg EL, Yoshino T, Rott LS, Robinson MK, Warnock RA, et al. 1991. The cutaneous lymphocyte antigen is a skin lymphocyte homing receptor for the vascular lectin endothelial cell-leukocyte adhesion molecule 1. J. Exp. Med. 174:1461–66 Jain RK. 2003. Molecular regulation of vessel maturation. Nat. Med. 9:685–93 de Bruijn MF, Ma X, Robin C, Ottersbach K, Sanchez MJ, Dzierzak E. 2002. Hematopoietic stem cells localize to the endothelial cell layer in the midgestation mouse aorta. Immunity 16:673–83 Ferrara N. 1999. Vascular endothelial growth factor: molecular and biological aspects. Curr. Top. Microbiol. Immunol. 237:1–30 Makinen T, Jussila L, Veikkola T, Karpanen T, Kettunen MI, et al. 2001. Inhibition of lymphangiogenesis with resulting lymphedema in transgenic mice expressing soluble VEGF receptor-3. Nat. Med 7:199–205 Madri JA, Pratt BM, Tucker AM. 1988. Phenotypic modulation of endothelial cells by transforming growth factor-beta depends upon the composition and organization of the extracellular matrix. J. Cell Biol. 106:1375–84 Broadley KN, Aquino AM, Woodward SC, Buckley-Sturrock A, Sato Y, et al. 1989. Monospecific antibodies implicate basic fibroblast growth factor in normal wound repair. Lab. Invest. 61:571–75 Asahara T, Bauters C, Zheng LP, Takeshita S, Bunting S, et al. 1995. Synergistic effect of vascular endothelial growth factor and basic fibroblast growth

18.

19.

20.

21.

22.

23.

24.

25.

26.

P1: IKH

699

factor on angiogenesis in vivo. Circulation 92:II365–71 Semenza GL. 1999. Regulation of mammalian O2 homeostasis by hypoxiainducible factor 1. Annu. Rev. Cell Dev. Biol. 15:551–78 Blotnick S, Peoples GE, Freeman MR, Eberlein TJ, Klagsbrun M. 1994. T lymphocytes synthesize and export heparinbinding epidermal growth factor-like growth factor and basic fibroblast growth factor, mitogens for vascular cells and fibroblasts: differential production and release by CD4+ and CD8+ T cells. Proc. Natl. Acad. Sci. USA 91:2890– 94 Friesel R, Komoriya A, Maciag T. 1987. Inhibition of endothelial cell proliferation by gamma-interferon. J. Cell Biol. 104:689–96 Sato N, Goto T, Haranaka K, Satomi N, Nariuchi H, et al. 1986. Actions of tumor necrosis factor on cultured vascular endothelial cells: morphologic modulation, growth inhibition, and cytotoxicity. J. Natl. Cancer Inst. 76:1113–21 Fathallah-Shaykh HM, Zhao LJ, Kafrouni AI, Smith GM, Forman J. 2000. Gene transfer of IFN-gamma into established brain tumors represses growth by antiangiogenesis. J. Immunol. 164:217–22 Leibovich SJ, Polverini PJ, Shepard HM, Wiseman DM, Shively V, Nuseir N. 1987. Macrophage-induced angiogenesis is mediated by tumour necrosis factor-alpha. Nature 329:630–32 Xia P, Gamble JR, Rye KA, Wang L, Hii CS, et al. 1998. Tumor necrosis factor-alpha induces adhesion molecule expression through the sphingosine kinase pathway. Proc. Natl. Acad. Sci. USA 95:14196–201 Lee MJ, Thangada S, Claffey KP, Ancellin N, Liu CH, et al. 1999. Vascular endothelial cell adherens junction assembly and morphogenesis induced by sphingosine1-phosphate. Cell 99:301–12 Rossi D, Zlotnik A. 2000. The biology

12 Feb 2004

15:59

700

27.

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

28.

29.

30.

31.

32.

33.

34.

35.

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL. of chemokines and their receptors. Annu. Rev. Immunol. 18:217–42 Strieter RM, Polverini PJ, Kunkel SL, Arenberg DA, Burdick MD, et al. 1995. The functional role of the ELR motif in CXC chemokine-mediated angiogenesis. J. Biol. Chem. 270:27348–57 Watson CA, Camera-Benson L, PalmerCrocker R, Pober JS. 1995. Variability among human umbilical vein endothelial cultures. Science 268:447–48 Bernardini G, Ribatti D, Spinetti G, Morbidelli L, Ziche M, et al. 2003. Analysis of the role of chemokines in angiogenesis. J. Immunol. Methods 273:83–101 Slowik MR, Min W, Ardito T, Karsan A, Kashgarian M, Pober JS. 1997. Evidence that tumor necrosis factor triggers apoptosis in human endothelial cells by interleukin-1-converting enzymelike protease-dependent and -independent pathways. Lab. Invest. 77:257–67 Madge LA, Li JH, Choi J, Pober JS. 2003. Inhibition of phosphatidylinositol 3-kinase sensitizes vascular endothelial cells to cytokine-initiated cathepsindependent apoptosis. J. Biol. Chem. 278:21295–306 Li JH, Kluger MS, Madge LA, Zheng L, Bothwell AL, Pober JS. 2002. Interferongamma augments CD95(APO-1/Fas) and pro-caspase-8 expression and sensitizes human vascular endothelial cells to CD95-mediated apoptosis. Am. J. Pathol. 161:1485–95 Li JH, Kirkiles-Smith NC, McNiff JM, Pober JS. 2003. TRAIL induces apoptosis and inflammatory gene expression in human endothelial cells. J. Immunol. 171:1526–33 Zheng L, Ben LH, Pober JS, Bothwell AL. 2002. Porcine endothelial cells, unlike human endothelial cells, can be killed by human CTL via Fas ligand and cannot be protected by Bcl-2. J. Immunol. 169:6850–55 Secomb TW, Konerding MA, West CA, Su M, Young AJ, Mentzer SJ. 2003. Mi-

36.

37.

38.

39.

40.

41.

42.

43.

44.

croangiectasias: structural regulators of lymphocyte transmigration. Proc. Natl. Acad. Sci. USA 100:7231–34 Vanhoutte PM, Mombouli JV. 1996. Vascular endothelium: vasoactive mediators. Prog. Cardiovasc. Dis. 39:229–38 Urakami-Harasawa L, Shimokawa H, Nakashima M, Egashira K, Takeshita A. 1997. Importance of endothelium-derived hyperpolarizing factor in human arteries. J. Clin. Invest. 100:2793–99 Fisslthaler B, Popp R, Kiss L, Potente M, Harder DR, et al. 1999. Cytochrome P450 2C is an EDHF synthase in coronary arteries. Nature 401:493–97 Rosenkranz-Weiss P, Sessa WC, Milstien S, Kaufman S, Watson CA, Pober JS. 1994. Regulation of nitric oxide synthesis by proinflammatory cytokines in human umbilical vein endothelial cells. Elevations in tetrahydrobiopterin levels enhance endothelial nitric oxide synthase specific activity. J. Clin. Invest. 93:2236– 43 Yoshizumi M, Perrella MA, Burnett JC Jr, Lee ME. 1993. Tumor necrosis factor downregulates an endothelial nitric oxide synthase mRNA by shortening its halflife. Circ. Res. 73:205–9 Zhang DX, Yi FX, Zou AP, Li PL. 2002. Role of ceramide in TNF-alpha-induced impairment of endothelium-dependent vasorelaxation in coronary arteries. Am. J. Physiol. Heart Circ. Physiol. 283:H1785– 94 Kim F, Gallis B, Corson MA. 2001. TNFalpha inhibits flow and insulin signaling leading to NO production in aortic endothelial cells. Am. J. Physiol. Cell Physiol. 280:C1057–65 Fichtlscherer S, Rossig L, Breuer S, Vasa M, Dimmeler S, Zeiher AM. 2001. Tumor necrosis factor antagonism with etanercept improves systemic endothelial vasoreactivity in patients with advanced heart failure. Circulation 104:3023–25 Ristimaki A, Viinikka L. 1992. Modulation of prostacyclin production by

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

45.

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

46.

47. 48.

49.

50.

51.

52.

53.

cytokines in vascular endothelial cells. Prostaglandins Leukot. Essent. Fatty Acids 47:93–99 Merhi-Soussi F, Dominguez Z, Macovschi O, Dubois M, Savany A, et al. 2000. Human lymphocytes stimulate prostacyclin synthesis in human umbilical vein endothelial cells. Involvement of endothelial cPLA2. J. Leukoc. Biol. 68:881–89 Zauli G, Pandolfi A, Gonelli A, Di Pietro R, Guarnieri S, et al. 2003. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) sequentially upregulates nitric oxide and prostanoid production in primary human endothelial cells. Circ. Res. 92:732–40 Michel CC, Curry FE. 1999. Microvascular permeability. Physiol. Rev. 79:703–61 Dvorak AM, Feng D. 2001. The vesiculovacuolar organelle (VVO). A new endothelial cell permeability organelle. J. Histochem. Cytochem. 49:419–32 Middleton J, Patterson AM, Gardner L, Schmutz C, Ashton BA. 2002. Leukocyte extravasation: chemokine transport and presentation by the endothelium. Blood 100:3853–60 Joris I, Cuenoud HF, Doern GV, Underwood JM, Majno G. 1990. Capillary leakage in inflammation. A study by vascular labeling. Am. J. Pathol. 137:1353–63 Stolpen AH, Guinan EC, Fiers W, Pober JS. 1986. Recombinant tumor necrosis factor and immune interferon act singly and in combination to reorganize human vascular endothelial cell monolayers. Am. J. Pathol. 123:16–24 Yi ES, Ulich TR. 1992. Endotoxin, interleukin-1, and tumor necrosis factor cause neutrophil-dependent microvascular leakage in postcapillary venules. Am. J. Pathol. 140: 659–63 Cotran RS, Pober JS, Gimbrone MA Jr, Springer TA, Wiebke EA, et al. 1988. Endothelial activation during interleukin 2 immunotherapy. A possible mechanism for the vascular leak syndrome. J. Immunol. 140:1883–88

P1: IKH

701

54. Damle NK, Doyle LV. 1990. Ability of human T lymphocytes to adhere to vascular endothelial cells and to augment endothelial permeability to macromolecules is linked to their state of post-thymic maturation. J. Immunol. 144:1233–40 55. Bourin MC, Lindahl U. 1993. Glycosaminoglycans and the regulation of blood coagulation. Biochem. J. 289:313– 30 56. Broze GJ Jr. 1995. Tissue factor pathway inhibitor and the revised theory of coagulation. Annu. Rev. Med. 46:103–12 57. Esmon CT. 1995. Thrombomodulin as a model of molecular mechanisms that modulate protease specificity and function at the vessel surface. FASEB J. 9:946– 55 58. van Hinsbergh VW. 1988. Regulation of the synthesis and secretion of plasminogen activators by endothelial cells. Haemostasis 18:307–27 59. Radomski MW, Palmer RM, Moncada S. 1987. The anti-aggregating properties of vascular endothelium: interactions between prostacyclin and nitric oxide. Br. J. Pharmacol. 92:639–46 60. Node K, Ruan XL, Dai J, Yang SX, Graham L, et al. 2001. Activation of Galpha mediates induction of tissue-type plasminogen activator gene transcription by epoxyeicosatrienoic acids. J. Biol. Chem. 276:15983–89 61. Martin NB, Jamieson A, Tuffin DP. 1993. The effect of interleukin-4 on tumour necrosis factor-alpha induced expression of tissue factor and plasminogen activator inhibitor-1 in human umbilical vein endothelial cells. Thromb. Haemost. 70:1037–42 62. Conway EM, Rosenberg RD. 1988. Tumor necrosis factor suppresses transcription of the thrombomodulin gene in endothelial cells. Mol. Cell Biol. 8:5588–92 63. Reverdiau-Moalic P, Watier H, Iochmann S, Pouplard C, Rideau E, et al. 1998. Human allogeneic lymphocytes trigger endothelial cell tissue factor expression by a

12 Feb 2004

15:59

702

64.

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

65.

66.

67.

68.

69.

70.

71.

72.

73.

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL. tumor necrosis factor-dependent pathway. J. Lab. Clin. Med. 132:530–40 Miller DL, Yaron R, Yellin MJ. 1998. CD40L-CD40 interactions regulate endothelial cell surface tissue factor and thrombomodulin expression. J. Leukoc. Biol. 63:373–79 Bavendiek U, Libby P, Kilbride M, Reynolds R, Mackman N, Schonbeck U. 2002. Induction of tissue factor expression in human endothelial cells by CD40 ligand is mediated via activator protein 1, nuclear factor kappa B, and Egr-1. J. Biol. Chem. 277:25032–39 Nawroth PP, Handley DA, Esmon CT, Stern DM. 1986. Interleukin 1 induces endothelial cell procoagulant while suppressing cell-surface anticoagulant activity. Proc. Natl. Acad. Sci. USA 83:3460– 64 Bombeli T, Karsan A, Tait JF, Harlan JM. 1997. Apoptotic vascular endothelial cells become procoagulant. Blood 89:2429–42 Kubes P, Suzuki M, Granger DN. 1991. Nitric oxide: an endogenous modulator of leukocyte adhesion. Proc. Natl. Acad. Sci. USA 88:4651–55 Node K, Huo Y, Ruan X, Yang B, Spiecker M, et al. 1999. Anti-inflammatory properties of cytochrome P450 epoxygenasederived eicosanoids. Science 285:1276– 79 Luscinskas FW, Ma S, Nusrat A, Parkos CA, Shaw SK. 2002. Leukocyte transendothelial migration: a junctional affair. Semin. Immunol. 14:105–13 Springer TA. 1995. Traffic signals on endothelium for lymphocyte recirculation and leukocyte emigration. Annu. Rev. Physiol. 57:827–72 Hurst SM, Wilkinson TS, McLoughlin RM, Jones S, Horiuchi S, et al. 2001. IL-6 and its soluble receptor orchestrate a temporal switch in the pattern of leukocyte recruitment seen during acute inflammation. Immunity 14:705–14 Schleimer RP, Sterbinsky SA, Kaiser J, Bickel CA, Klunk DA, et al. 1992. IL-4

74.

75.

76.

77.

78.

79.

80.

81.

induces adherence of human eosinophils and basophils but not neutrophils to endothelium. Association with expression of VCAM-1. J. Immunol. 148:1086– 92 Pober JS. 1999. Immunobiology of human vascular endothelium. Immunol. Res. 19:225–32 Natali PG, De Martino C, Quaranta V, Nicotra MR, Frezza F, et al. 1981. Expression of Ia-like antigens in normal human nonlymphoid tissues. Transplantation 31:75–78 Hayry P, von Willebrand E, Andersson LC. 1980. Expression of HLA-ABC and -DR locus antigens on human kidney, endothelial, tubular, and glomerular cells. Scand. J. Immunol. 11:303–10 Choo JK, Seebach JD, Nickeleit V, Shimizu A, Lei H, et al. 1997. Species differences in the expression of major histocompatibility complex class II antigens on coronary artery endothelium: implications for cell-mediated xenoreactivity. Transplantation 64:1315–22 Hart DN, Fabre JW. 1981. Major histocompatibility complex antigens in rat kidney, ureter, and bladder. Localization with monoclonal antibodies and demonstration of Ia-positive dendritic cells. Transplantation 31:318–25 Turner RR, Beckstead JH, Warnke RA, Wood GS. 1987. Endothelial cell phenotypic diversity. In situ demonstration of immunologic and enzymatic heterogeneity that correlates with specific morphologic subtypes. Am. J. Clin. Pathol. 87:569–75 Pober JS, Gimbrone MA Jr, Cotran RS, Reiss CS, Burakoff SJ, et al. 1983. Ia expression by vascular endothelium is inducible by activated T cells and by human gamma interferon. J. Exp. Med. 157: 1339–53 McDouall RM, Batten P, McCormack A, Yacoub MH, Rose ML. 1997. MHC class II expression on human heart microvascular endothelial cells: exquisite sensitivity

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

82.

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

83.

84.

85.

86.

87.

88.

89.

to interferon-gamma and natural killer cells. Transplantation 64:1175–80 Tereb DA, Kirkiles-Smith NC, Kim RW, Wang Y, Rudic RD, et al. 2001. Human T cells infiltrate and injure pig coronary artery grafts with activated but not quiescent endothelium in immunodeficient mouse hosts. Transplantation 71:1622– 30 Belitsky P, Miller SM, Gupta R, Lee S, Ghose T. 1990. Induction of MHC class II expression in recipient tissues caused by allograft rejection. Transplantation 49:472–76 Groenewegen G, Buurman WA, Jeunhomme GM, van der Linden CJ. 1985. Effect of cyclosporine on MHC class II antigen expression on arterial and venous endothelium in vitro. Transplantation 40:21–25 Goes N, Sims T, Urmson J, Vincent D, Ramassar V, Halloran PF. 1995. Disturbed MHC regulation in the IFN-gamma knockout mouse. Evidence for three states of MHC expression with distinct roles for IFN-gamma. J. Immunol. 155:4559– 66 Munro JM, Pober JS, Cotran RS. 1989. Tumor necrosis factor and interferongamma induce distinct patterns of endothelial activation and associated leukocyte accumulation in skin of Papio anubis. Am. J. Pathol. 135:121–33 Messadi DV, Pober JS, Murphy GF. 1988. Effects of recombinant gamma-interferon on HLA-DR and DQ expression by skin cells in short-term organ culture. Lab. Invest. 58:61–67 de Waal RM, Bogman MJ, Maass CN, Cornelissen LM, Tax WJ, Koene RA. 1983. Variable expression of Ia antigens on the vascular endothelium of mouse skin allografts. Nature 303:426–29 Ferry B, Halttunen J, Leszczynski D, Schellekens H, van der Meide PH, Hayry P. 1987. Impact of class II major histocompatibility complex antigen expression on the immunogenic potential of isolated

90.

91.

92.

93.

94.

95.

96.

97.

98.

P1: IKH

703

rat vascular endothelial cells. Transplantation 44:499–503 Murphy GF, Bronstein BR, Knowles RW, Bhan AK. 1985. Ultrastructural documentation of M241 glycoprotein on dendritic and endothelial cells in normal human skin. Lab. Invest. 52:264–69 Min W, Pober JS, Johnson DR. 1996. Kinetically coordinated induction of TAP1 and HLA class I by IFN-gamma: The rapid induction of TAP1 by IFN-gamma is mediated by Stat1 alpha. J. Immunol. 156:3174–83 Johnson DR, Pober JS. 1994. HLA class I heavy-chain gene promoter elements mediating synergy between tumor necrosis factor and interferons. Mol. Cell Biol. 14:1322–32 Collins T, Korman AJ, Wake CT, Boss JM, Kappes DJ, et al. 1984. Immune interferon activates multiple class II major histocompatibility complex genes and the associated invariant chain gene in human endothelial cells and dermal fibroblasts. Proc. Natl. Acad. Sci. USA 81:4917– 21 Steimle V, Siegrist CA, Mottet A, Lisowska-Grospierre B, Mach B. 1994. Regulation of MHC class II expression by interferon-gamma mediated by the transactivator gene CIITA. Science 265:106–9 Johnson DR. 2003. Locus-specific constitutive and cytokine-induced HLA class I gene expression. J. Immunol. 170:1894– 902 Epperson DE, Pober JS. 1994. Antigenpresenting function of human endothelial cells. Direct activation of resting CD8 T cells. J. Immunol. 153:5402–12 Savage CO, Brooks CJ, Harcourt GC, Picard JK, King W, et al. 1995. Human vascular endothelial cells process and present autoantigen to human T cell lines. Int. Immunol. 7:471–79 Savinov AY, Wong FS, Stonebraker AC, Chervonsky AV. 2003. Presentation of antigen by endothelial cells and chemoattraction are required for homing of

12 Feb 2004

15:59

704

99.

100.

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

101.

102.

103.

104.

105.

106.

107.

108.

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL. insulin-specific CD8+ T cells. J. Exp. Med. 197:643–56 Lenschow DJ, Walunas TL, Bluestone JA. 1996. CD28/B7 system of T cell costimulation. Annu. Rev. Immunol. 14:233–58 Parra E, Wingren AG, Hedlund G, Kalland T, Dohlsten M. 1997. The role of B7-1 and LFA-3 in costimulation of CD8+ T cells. J. Immunol. 158:637–42 Boussiotis VA, Freeman GJ, Griffin JD, Gray GS, Gribben JG, Nadler LM. 1994. CD2 is involved in maintenance and reversal of human alloantigen-specific clonal anergy. J. Exp. Med. 180:1665–73 Sayegh MH, Turka LA. 1998. The role of T-cell costimulatory activation pathways in transplant rejection. N. Engl. J. Med. 338:1813–21 Ellis CN, Krueger GG. 2001. Treatment of chronic plaque psoriasis by selective targeting of memory effector T lymphocytes. N. Engl. J. Med. 345:248–55 Grohmann U, Orabona C, Fallarino F, Vacca C, Calcinaro F, et al. 2002. CTLA4-Ig regulates tryptophan catabolism in vivo. Nat. Immunol. 3:1097–101 Kaplon RJ, Hochman PS, Michler RE, Kwiatkowski PA, Edwards NM, et al. 1996. Short course single agent therapy with an LFA-3-IgG1 fusion protein prolongs primate cardiac allograft survival. Transplantation 61:356–63 Smith ME, Thomas JA. 1990. Cellular expression of lymphocyte function associated antigens and the intercellular adhesion molecule-1 in normal tissue. J. Clin. Pathol. 43:893–900 Denton MD, Geehan CS, Alexander SI, Sayegh MH, Briscoe DM. 1999. Endothelial cells modify the costimulatory capacity of transmigrating leukocytes and promote CD28-mediated CD4(+) T cell alloactivation. J. Exp. Med. 190:555– 66 Jollow KC, Zimring JC, Sundstrom JB, Ansari AA. 1999. CD40 ligation induced phenotypic and functional expression of CD80 by human cardiac microvas-

109.

110.

111.

112.

113.

114.

115.

116.

117.

cular endothelial cells. Transplantation 68:430–39 Prat A, Biernacki K, Becher B, Antel JP. 2000. B7 expression and antigen presentation by human brain endothelial cells: requirement for proinflammatory cytokines. J. Neuropathol. Exp. Neurol. 59:129–36 Omari KI, Dorovini-Zis K. 2001. Expression and function of the costimulatory molecules B7-1 (CD80) and B7-2 (CD86) in an in vitro model of the human blood– brain barrier. J. Neuroimmunol. 113:129– 41 Ma W, Pober JS. 1998. Human endothelial cells effectively costimulate cytokine production by, but not differentiation of, naive CD4+ T cells. J. Immunol. 161:2158–67 Grewal IS, Flavell RA. 1998. CD40 and CD154 in cell-mediated immunity. Annu. Rev. Immunol. 16:111–35 Karmann K, Hughes CC, Schechner J, Fanslow WC, Pober JS. 1995. CD40 on human endothelial cells: inducibility by cytokines and functional regulation of adhesion molecule expression. Proc. Natl. Acad. Sci. USA 92:4342–46 Rushworth SA, Bravery CA, Thompson S. 2001. Human CD154 induces activation of porcine endothelial cells and upregulation of MHC class II expression. Transplantation 72:127–32 Karmann K, Hughes CC, Fanslow WC, Pober JS. 1996. Endothelial cells augment the expression of CD40 ligand on newly activated human CD4+ T cells through a CD2/LFA-3 signaling pathway. Eur. J. Immunol. 26:610–17 Mach F, Schonbeck U, Sukhova GK, Bourcier T, Bonnefoy JY, et al. 1997. Functional CD40 ligand is expressed on human vascular endothelial cells, smooth muscle cells, and macrophages: implications for CD40-CD40 ligand signaling in atherosclerosis. Proc. Natl. Acad. Sci. USA 94:1931–36 Reul RM, Fang JC, Denton MD, Geehan C, Long C, et al. 1997. CD40 and CD40 ligand (CD154) are coexpressed on

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

118.

119.

120.

121.

122.

123.

124.

125.

microvessels in vivo in human cardiac allograft rejection. Transplantation 64:1765–74 Hakkinen T, Karkola K, Yla-Herttuala S. 2000. Macrophages, smooth muscle cells, endothelial cells, and T-cells express CD40 and CD40L in fatty streaks and more advanced human atherosclerotic lesions. Colocalization with epitopes of oxidized low-density lipoprotein, scavenger receptor, and CD16 (Fc gammaRIII). Virchows Arch. 437:396–405 Blotta MH, Marshall JD, DeKruyff RH, Umetsu DT. 1996. Cross-linking of the CD40 ligand on human CD4+ T lymphocytes generates a costimulatory signal that up-regulates IL-4 synthesis. J. Immunol. 156:3133–40 Khayyamian S, Hutloff A, Buchner K, Grafe M, Henn V, et al. 2002. ICOSligand, expressed on human endothelial cells, costimulates Th1 and Th2 cytokine secretion by memory CD4+ T cells. Proc. Natl. Acad. Sci. USA 99:6198–203 Freeman GJ, Long AJ, Iwai Y, Bourque K, Chernova T, et al. 2000. Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. J. Exp. Med. 192:1027–34 Latchman Y, Wood CR, Chernova T, Chaudhary D, Borde M, et al. 2001. PDL2 is a second ligand for PD-1 and inhibits T cell activation. Nat. Immunol. 2:261–68 Boussaud V, Soler P, Moreau J, Goodwin RG, Hance AJ. 1998. Expression of three members of the TNF-R family of receptors (4-1BB, lymphotoxin-beta receptor, and Fas) in human lung. Eur. Respir. J. 12:926–31 Broll K, Richter G, Pauly S, Hofstaedter F, Schwarz H. 2001. CD137 expression in tumor vessel walls. High correlation with malignant tumors. Am. J. Clin. Pathol. 115:543–49 Migone TS, Zhang J, Luo X, Zhuang L, Chen C, et al. 2002. TL1A is a TNF-like ligand for DR3 and TR6/DcR3 and func-

126.

127.

128.

129.

130.

131.

132.

133.

P1: IKH

705

tions as a T cell costimulator. Immunity 16:479–92 Kikutani H, Kumanogoh A. 2003. Semaphorins in interactions between T cells and antigen-presenting cells. Nat. Rev. Immunol. 3:159–67 Yu G, Luo H, Wu Y, Wu J. 2003. Ephrin B2 Induces T cell Costimulation. J. Immunol. 171:106–14 Murphy LL, Hughes CC. 2002. Endothelial cells stimulate T cell NFAT nuclear translocation in the presence of cyclosporin A: involvement of the wnt/glycogen synthase kinase-3 beta pathway. J. Immunol. 169:3717–25 Villa N, Walker L, Lindsell CE, Gasson J, Iruela-Arispe ML, Weinmaster G. 2001. Vascular expression of Notch pathway receptors and ligands is restricted to arterial vessels. Mech. Dev. 108:161–64 Yvon ES, Vigouroux S, Rousseau RF, Biagi E, Amrolia P, et al. 2003. Overexpression of the Notch ligand, Jagged-1 induces alloantigen-specific human regulatory T cells. Blood 102:3815–21 Van Seventer GA, Shimizu Y, Horgan KJ, Shaw S. 1990. The LFA-1 ligand ICAM-1 provides an important costimulatory signal for T cell receptor-mediated activation of resting T cells. J. Immunol. 144:4579– 86 Van Seventer GA, Shimizu Y, Horgan KJ, Luce GE, Webb D, Shaw S. 1991. Remote T cell co-stimulation via LFA-1/ICAM-1 and CD2/LFA-3: demonstration with immobilized ligand/mAb and implication in monocyte-mediated co-stimulation. Eur. J. Immunol. 21:1711–18 Van Seventer GA, Bonvini E, Yamada H, Conti A, Stringfellow S, et al. 1992. Costimulation of T cell receptor/CD3mediated activation of resting human CD4+ T cells by leukocyte functionassociated antigen-1 ligand intercellular cell adhesion molecule-1 involves prolonged inositol phospholipid hydrolysis and sustained increase of intracellular Ca2+ levels. J. Immunol. 149:3872–80

12 Feb 2004

15:59

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

706

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL.

134. Ni HT, Deeths MJ, Li W, Mueller DL, Mescher MF. 1999. Signaling pathways activated by leukocyte functionassociated Ag-1-dependent costimulation. J. Immunol. 162:5183–89 135. Geginat J, Clissi B, Moro M, Dellabona P, Bender JR, Pardi R. 2000. CD28 and LFA1 contribute to cyclosporin A-resistant T cell growth by stabilizing the IL-2 mRNA through distinct signaling pathways. Eur. J. Immunol. 30:1136–44 136. Green JM, Karpitskiy V, Kimzey SL, Shaw AS. 2000. Coordinate regulation of T cell activation by CD2 and CD28. J. Immunol. 164:3591–95 137. Sims TN, Dustin ML. 2002. The immunological synapse: integrins take the stage. Immunol. Rev. 186:100–17 138. Xia L, Sperandio M, Yago T, McDaniel JM, Cummings RD, et al. 2002. P-selectin glycoprotein ligand-1-deficient mice have impaired leukocyte tethering to E-selectin under flow. J. Clin. Invest. 109:939–50 139. Austrup F, Vestweber D, Borges E, Lohning M, Brauer R, et al. 1997. P- and Eselectin mediate recruitment of T-helper1 but not T-helper-2 cells into inflammed tissues. Nature 385:81–83 140. Xie H, Lim YC, Luscinskas FW, Lichtman AH. 1999. Acquisition of selectin binding and peripheral homing properties by CD4(+) and CD8(+) T cells. J. Exp. Med. 189:1765–76 141. Siegelman MH, DeGrendele HC, Estess P. 1999. Activation and interaction of CD44 and hyaluronan in immunological systems. J. Leukoc. Biol. 66:315–21 142. Kurt-Jones EA, Fiers W, Pober JS. 1987. Membrane interleukin 1 induction on human endothelial cells and dermal fibroblasts. J. Immunol. 139:2317–24 143. Greenbaum LA, Horowitz JB, Woods A, Pasqualini T, Reich EP, Bottomly K. 1988. Autocrine growth of CD4+ T cells. Differential effects of IL-1 on helper and inflammatory T cells. J. Immunol. 140:1555– 60 144. Lichtman AH, Chin J, Schmidt JA, Ab-

145.

146.

147.

148.

149.

150.

151.

152.

153.

154.

bas AK. 1988. Role of interleukin 1 in the activation of T lymphocytes. Proc. Natl. Acad. Sci. USA 85:9699–703 Waldmann TA, Tagaya Y. 1999. The multifaceted regulation of interleukin-15 expression and the role of this cytokine in NK cell differentiation and host response to intracellular pathogens. Annu. Rev. Immunol. 17:19–49 Muller U, Steinhoff U, Reis LF, Hemmi S, Pavlovic J, et al. 1994. Functional role of type I and type II interferons in antiviral defense. Science 264:1918–21 Loppnow H, Libby P. 1989. Adult human vascular endothelial cells express the IL6 gene differentially in response to LPS or IL1. Cell Immunol. 122:493–503 Suen Y, Chang M, Lee SM, Buzby JS, Cairo MS. 1994. Regulation of interleukin-11 protein and mRNA expression in neonatal and adult fibroblasts and endothelial cells. Blood 84:4125–34 Lienenluke B, Germann T, Kroczek RA, Hecker M. 2000. CD154 stimulation of interleukin-12 synthesis in human endothelial cells. Eur. J. Immunol. 30:2864– 70 Gerdes N, Sukhova GK, Libby P, Reynolds RS, Young JL, Schonbeck U. 2002. Expression of interleukin (IL)-18 and functional IL-18 receptor on human vascular endothelial cells, smooth muscle cells, and macrophages: implications for atherogenesis. J. Exp. Med. 195:245–57 Harcourt JL, Hagan MK, Offermann MK. 2000. Modulation of double-stranded RNA-mediated gene induction by interferon in human umbilical vein endothelial cells. J. Interferon. Cytokine. Res. 20:1007–13 Curti A, Ratta M, Corinti S, Girolomoni G, Ricci F, et al. 2001. Interleukin-11 induces Th2 polarization of human CD4(+) T cells. Blood 97:2758–63 Geppert TD, Lipsky PE. 1989. Antigen presentation at the inflammatory site. Crit. Rev. Immunol. 9:313–62 Gurney AL, Marsters SA, Huang RM,

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

155.

156.

157.

158.

159.

160.

161.

162.

163.

164.

Pitti RM, Mark DT, et al. 1999. Identification of a new member of the tumor necrosis factor family and its receptor, a human ortholog of mouse GITR. Curr. Biol. 9:215–18 Shimizu J, Yamazaki S, Takahashi T, Ishida Y, Sakaguchi S. 2002. Stimulation of CD25(+)CD4(+) regulatory T cells through GITR breaks immunological selftolerance. Nat. Immunol. 3:135–42 Pasare C, Medzhitov R. 2003. Toll pathway-dependent blockade of CD4+ CD25+ T cell-mediated suppression by dendritic cells. Science 299:1033–36 Krummel MF, Davis MM. 2002. Dynamics of the immunological synapse: finding, establishing and solidifying a connection. Curr. Opin. Immunol. 14:66–74 Tseng SY, Dustin ML. 2002. T-cell activation: a multidimensional signaling network. Curr. Opin. Cell Biol. 14:575–80 Wulfing C, Sumen C, Sjaastad MD, Wu LC, Dustin ML, Davis MM. 2002. Costimulation and endogenous MHC ligands contribute to T cell recognition. Nat. Immunol. 3:42–47 Hiltbold EM, Poloso NJ, Roche PA. 2003. MHC class II-peptide complexes and APC lipid rafts accumulate at the immunological synapse. J. Immunol. 170:1329–38 Wetzel SA, McKeithan TW, Parker DC. 2002. Live-cell dynamics and the role of costimulation in immunological synapse formation. J. Immunol. 169:6092–101 Al -Alwan MM, Rowden G, Lee TD, West KA. 2001. The dendritic cell cytoskeleton is critical for the formation of the immunological synapse. J. Immunol. 166:1452– 56 Barreiro O, Yanez-Mo M, Serrador JM, Montoya MC, Vicente-Manzanares M, et al. 2002. Dynamic interaction of VCAM-1 and ICAM-1 with moesin and ezrin in a novel endothelial docking structure for adherent leukocytes. J. Cell Biol. 157:1233–45 Oviedo-Orta E, Gasque P, Evans WH.

165.

166.

167.

168.

169.

170.

171.

172.

173.

P1: IKH

707

2001. Immunoglobulin and cytokine expression in mixed lymphocyte cultures is reduced by disruption of gap junction intercellular communication. FASEB J. 15:768–74 Guinan EC, Smith BR, Davies PF, Pober JS. 1988. Cytoplasmic transfer between endothelium and lymphocytes: quantitation by flow cytometry. Am. J. Pathol. 132:406–9 Oviedo-Orta E, Errington RJ, Evans WH. 2002. Gap junction intercellular communication during lymphocyte transendothelial migration. Cell Biol. Int. 26:253–63 Raine CS, Cannella B, Duijvestijn AM, Cross AH. 1990. Homing to central nervous system vasculature by antigen-specific lymphocytes. II. Lymphocyte/endothelial cell adhesion during the initial stages of autoimmune demyelination. Lab. Invest. 63:476–89 Johnson DR, Hauser IA, Voll RE, Emmrich F. 1998. Arterial and venular endothelial cell costimulation of cytokine secretion by human T cell clones. J. Leukoc. Biol. 63:612–19 Wagner CR, Vetto RM, Burger DR. 1984. The mechanism of antigen presentation by endothelial cells. Immunobiology 168:453–69 St Louis JD, Lederer JA, Lichtman AH. 1993. Costimulator deficient antigen presentation by an endothelial cell line induces a nonproliferative T cell activation response without anergy. J. Exp. Med. 178:1597–605 Fabry Z, Sandor M, Gajewski TF, Herlein JA, Waldschmidt MM, et al. 1993. Differential activation of Th1 and Th2 CD4+ cells by murine brain microvessel endothelial cells and smooth muscle/pericytes. J. Immunol. 151:38–47 Kawai T, Seki M, Watanabe H, Eastcott JW, Smith DJ, Taubman MA. 2000. T(h)1 transmigration anergy: a new concept of endothelial cell-T cell regulatory interaction. Int. Immunol. 12:937–48 DeBruyne LA, Ensley RD, Olsen SL,

12 Feb 2004

15:59

708

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

174.

175.

176.

177.

178.

179.

180.

181.

182.

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHOI ET AL. Taylor DO, Carpenter BM, et al. 1993. Increased frequency of alloantigen-reactive helper T lymphocytes is associated with human cardiac allograft rejection. Transplantation 56:722–27 Deacock SJ, Lechler RI. 1992. Positive correlation of T cell sensitization with frequencies of alloreactive T helper cells in chronic renal failure patients. Transplantation 54:338–43 Marelli-Berg FM, Hargreaves RE, Carmichael P, Dorling A, Lombardi G, Lechler RI. 1996. Major histocompatibility complex class II-expressing endothelial cells induce allospecific nonresponsiveness in naive T cells. J. Exp. Med. 183:1603–12 Marelli-Berg FM, Frasca L, Weng L, Lombardi G, Lechler RI. 1999. Antigen recognition influences transendothelial migration of CD4+ T cells. J. Immunol. 162:696–703 Tay SS, McCormack A, Lawson C, Rose ML. 2003. IFN-gamma reverses the stop signal allowing migration of antigenspecific T cells into inflammatory sites. J. Immunol. 170:3315–22 Dengler TJ, Pober JS. 2000. Human vascular endothelial cells stimulate memory but not naive CD8+ T cells to differentiate into CTL retaining an early activation phenotype. J. Immunol. 164:5146–55 Lodge PA, Haisch CE. 1993. T cell subset responses to allogeneic endothelium. Proliferation of CD8+ but not CD4+ lymphocytes. Transplantation 56:656–61 de Jong R, Brouwer M, Miedema F, van Lier RA. 1991. Human CD8+ T lymphocytes can be divided into CD45RA+ and CD45RO+ cells with different requirements for activation and differentiation. J. Immunol. 146:2088–94 Biedermann BC, Pober JS. 1998. Human endothelial cells induce and regulate cytolytic T cell differentiation. J. Immunol. 161:4679–87 Biedermann BC, Pober JS. 1999. Human vascular endothelial cells favor clonal ex-

183.

184.

185.

186.

187.

188.

189.

190.

pansion of unusual alloreactive CTL. J. Immunol. 162:7022–30 Murphy LL, Mazanet MM, Taylor AC, Mestas J, Hughes CC. 1999. Single-cell analysis of costimulation by B cells, endothelial cells, and fibroblasts demonstrates heterogeneity in responses of CD4(+) memory T cells. Cell Immunol. 194:150–61 Fraser J, Arcus V, Kong P, Baker E, Proft T. 2000. Superantigens—powerful modifiers of the immune system. Mol. Med. Today 6:125–32 Westphal JR, Tax WJ, Willems HW, Koene RA, Ruiter DJ, De Waal RM. 1992. Accessory function of endothelial cells in anti-CD3-induced T-cell proliferation: synergism with monocytes. Scand. J. Immunol. 35:449–57 Hughes CC, Savage CO, Pober JS. 1990. Endothelial cells augment T cell interleukin 2 production by a contact-dependent mechanism involving CD2/LFA-3 interaction. J. Exp. Med. 171:1453–67 Mestas J, Hughes CC. 2001. Endothelial cell costimulation of T cell activation through CD58-CD2 interactions involves lipid raft aggregation. J. Immunol. 167:4378–85 Hughes CC, Pober JS. 1993. Costimulation of peripheral blood T cell activation by human endothelial cells. Enhanced IL2 transcription correlates with increased c-fos synthesis and increased Fos content of AP-1. J. Immunol. 150:3148–60 Hughes CC, Pober JS. 1996. Transcriptional regulation of the interleukin-2 gene in normal human peripheral blood T cells. Convergence of costimulatory signals and differences from transformed T cells. J. Biol. Chem. 271:5369–77 Murakami K, Ma W, Fuleihan R, Pober JS. 1999. Human endothelial cells augment early CD40 ligand expression in activated CD4+ T cells through LFA-3mediated stabilization of mRNA. J. Immunol. 163:2667–73

12 Feb 2004

15:59

AR

AR210-IY22-23.tex

AR210-IY22-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS 191. Lechler RI, Batchelor JR. 1982. Restoration of immunogenicity to passenger celldepleted kidney allografts by the addition of donor strain dendritic cells. J. Exp. Med. 155:31–41 192. McKenzie JL, Beard ME, Hart DN. 1984. Depletion of donor kidney dendritic cells prolongs graft survival. Transplant Proc. 16:948–51 193. Brewer Y, Palmer A, Taube D, Welsh K, Bewick M, et al. 1989. Effect of graft perfusion with two CD45 monoclonal antibodies on incidence of kidney allograft rejection. Lancet 2:935–37 194. Kreisel D, Krupnick AS, Gelman AE, Engels FH, Popma SH, et al. 2002. Nonhematopoietic allograft cells directly activate CD8+ T cells and trigger acute rejection: an alternative mechanism of allorecognition. Nat. Med. 8:233–39 195. Valujskikh A, Lantz O, Celli S, Matzinger P, Heeger PS. 2002. Cross-primed CD8(+) T cells mediate graft rejection via a distinct effector pathway. Nat. Immunol. 3:844–51 196. Silber A, Newman W, Reimann KA, Hendricks E, Walsh D, Ringler DJ. 1994. Kinetic expression of endothelial adhesion molecules and relationship to leukocyte recruitment in two cutaneous models of inflammation. Lab. Invest. 70:163–75 197. Jutte NH, Knoop CJ, Heijse P, Balk AH, Mochtar B, et al. 1996. Human heart endothelial-cell-restricted allorecognition. Transplantation 62:403–6 198. Deckers JG, Daha MR, Van der Kooij SW, Van der Woude FJ. 1998. Epithelialand endothelial-cell specificity of renal graft infiltrating T cells. Clin. Transplant. 12:285–91

P1: IKH

709

199. Ma W, Lehner PJ, Cresswell P, Pober JS, Johnson DR. 1997. Interferon-gamma rapidly increases peptide transporter (TAP) subunit expression and peptide transport capacity in endothelial cells. J. Biol. Chem. 272:16585–90 200. Watson CA, Petzelbauer P, Zhou J, Pardi R, Bender JR. 1995. Contact-dependent endothelial class II HLA gene activation induced by NK cells is mediated by IFN-gamma-dependent and -independent mechanisms. J. Immunol. 154:3222–33 201. Mazanet MM, Hughes CC. 2002. B7-H1 is expressed by human endothelial cells and suppresses T cell cytokine synthesis. J. Immunol. 169:3581–88 202. Collinge M, Pardi R, Bender JR. 1998. Class II transactivator-independent endothelial cell MHC class II gene activation induced by lymphocyte adhesion. J. Immunol. 161:1589–93 203. Lou J, Dayer JM, Grau GE, Burger D. 1996. Direct cell/cell contact with stimulated T lymphocytes induces the expression of cell adhesion molecules and cytokines by human brain microvascular endothelial cells. Eur. J. Immunol. 26:3107– 13 204. Omari KM, Dorovini-Zis K. 2003. CD40 expressed by human brain endothelial cells regulates CD4+ T cell adhesion to endothelium. J. Neuroimmunol. 134:166– 78 205. Yarwood H, Mason JC, Mahiouz D, Sugars K, Haskard DO. 2000. Resting and activated T cells induce expression of E-selectin and VCAM-1 by vascular endothelial cells through a contactdependent but CD40 ligand-independent mechanism. J. Leukoc. Biol. 68:233–42

3/12/2004 2:13 PM

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

See legend on next page

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Choi.qxd Page 1

C-1

Choi.qxd

3/12/2004

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

2:13 PM

Page 2

CHOI ET AL.

Figure 1 Heterogeneity of endothelial cells. This figure highlights some specialized features of arterioles, capillaries, and venules, with characteristic surface molecules depicted on each endothelial type. Molecules induced during inflammation by TNF are shown in blue. In the center panel, three types of specialized capillary EC (fenestrated, sinusoidal, and continuous) are shown. Some molecules localized to continuous endothelial junctions are shown in the lower insert. Note that arterial EC are specialized to control blood flow (via NO and EDHF production), whereas capillary EC are specialized to control solute exchange (via formation of fenestrae or sinusoids), and venular EC are specialized to recruit leukocytes (via adhesion molecule expression).

Choi.qxd

3/12/2004

2:13 PM

Page 3

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS

C-3

Figure 2 Comparison of EC and DC T cell contact regions. On the left side, this figure illustrates some of the membrane-bound molecules on the EC surface that may participate in antigen presentation and subsequent T cell activation. Depicted is a docking structure, unique to T cell/EC adhesion, that forms a membrane cup around the base of the T cell. Preliminary data suggest that, in the presence of antigen, an immune synapse can form within the docking structure, differentiating into a central supramolecular activation complex (c-SMAC), consisting of MHC and costimulatory molecules, and a peripheral supramolecular activation complex (p-SMAC), consisting of adhesion molecules. Such immune synapses are well formed in contacts between a T cell and classical antigen presenting cell (DC), which is shown on the right side of the figure. The molecules that are thought to be unique to either APC type are colored differentially, EC in pink and DC in yellow.

Choi.qxd

3/12/2004

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-4

2:13 PM

Page 4

CHOI ET AL.

Figure 3 T cell–mediated effects on EC function. This figure illustrates some lymphocytemediated influences on EC functions described in the text. Via soluble and membranebound mediators, the T cells can induce EC activation and/or dysfunction in the processes depicted above.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:683-709. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18) P1: FDK 10.1146/annurev.immunol.22.012703.104527

Annu. Rev. Immunol. 2004. 22:711–43 doi: 10.1146/annurev.immunol.22.012703.104527 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on December 12, 2003

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS1 Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda Department of Pathology, University of Massachusetts Medical School, Worcester, Massachusetts 01655; email: [email protected], [email protected], [email protected]

Key Words T cells, B cells, antibody, cytotoxicity, vaccine ■ Abstract The purpose of immunological memory is to protect the host from reinfection, to control persistent infections, and, through maternal antibody, to protect the host’s immunologically immature offspring from primary infections. Immunological memory is an exclusive property of the acquired immune system, where in the presence of CD4 T cell help, T cells and B cells clonally expand and differentiate to provide effector systems that protect the host from pathogens. Here we describe how T and B cell memory is generated in response to virus infections and how these cells respond when the host is infected again by similar or different viruses.

INTRODUCTION Immunological memory to viral infections was acknowledged long before the discovery of either viruses or of the immune system. Nearly 2500 years ago, the ancient Greek Thucydides noted that patients recovering from the plague of Athens did not contract it again (The Peloponnesian Wars, Book 4). Over 1000 years ago, the Chinese practiced the art of variolation, whereby exudates from pustules of patients suffering from mild cases of smallpox were inoculated onto previously 1

Abbreviations: AICD, activation-induced cell death; AID, activation-induced cytidine deaminase; APC, antigen-presenting cells; BCR, B cell receptor; CFSE, 5,6-carboxyfluorescein; CMV, cytomegalovirus; CTL, cytotoxic T lymphocyte; DHF, dengue hemorrhagic fever; EBV, Epstein-Barr virus; FDC, follicular dendritic cell; GC, germinal center; HIV, human immunodeficiency virus; HSV, herpes simplex virus; IFN, interferon; IL, interleukin; LCMV, lymphocytic choriomeningitis virus; MCMV, murine cytomegalovirus; Mγ HV, mouse γ herpes virus; PC, plasma cell; PEC, peritoneal exudate cell; PV, Pichinde virus; PyV, polyomavirus; RSV, respiratory syncytial virus; SIV, simian immunodeficiency virus; TCR, T cell receptor; TNF, tumor necrosis factor; VSV, vesicular stomatitis virus; VV, vaccinia virus; VZV, varicella zoster virus. 0732-0582/04/0423-0711$14.00

711

19 Feb 2004

14:33

712

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

TABLE 1 Antiviral vaccines licensed in the United States

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Vaccine

Type of antigen used

Route

Adenovirus

Live virus

i.m.

Hepatitis A

Inactivated virus

i.m.

Hepatitis B

Recombinant protein

i.m.

Influenza

Inactivated virus, or viral antigens

i.m.

Japanese encephalitis

Inactivated virus

s.c.

Measles

Live virus

s.c.

Mumps

Live virus

s.c.

Rubella

Live virus

s.c.

Polio

Inactivated Virus

s.c.

Polio

Live virus

oral

Rabies

Inactivated virus

i.m.

Varicella

Live virus

s.c.

Yellow fever

Live virus

s.c.

Smallpox (Variola)

Live vaccinia virus

Intracutaneous

The routinely recommended vaccines are in bold letters. Based on the appendix published in Vaccines (235). i.m., intramuscular; s.c., subcutaneous.

uninfected individuals to protect them from the more virulent forms of smallpox (variola) virus (1). Over 200 years ago, Edward Jenner similarly inoculated subjects with cowpox exudates to protect them from contracting the antigenically related smallpox (1). That experiment in “vaccination” (from the Latin word, vacca, for cow)—an experiment in immunological memory to viral infections—may alone have initiated the fields of both immunology and virology. Today there are vaccines available for 13 human viruses (Table 1), with testing of many additional vaccines under way, all based on the concept of developing immunological memory. Immunological memory is an exclusive property of the “adaptive” or “acquired” immune system. Notably, antigen-specific clones of T cell receptor (TCR) αβexpressing T cells and B cells with help from T cells proliferate and differentiate in response to a primary infection and remain in the host at relatively high frequencies after resolution of the infection. Resistance to reinfection is a property of these expanded clones of T cells equipped with antiviral effector functions and of antibody-secreting plasma cells (PCs) and nonsecreting memory B cells that can be stimulated to rapidly differentiate into antibody-secreting cells. The cellular components of the “innate” or “natural” immune system, such as macrophages, granulocytes, and NK cells, may provide strong barriers to infection, but there is no evidence that they manifest memory (2, 3). Viral infections may also be controlled by antigen-specific TCR γ δ-expressing T cells (4) or by immunoglobulin

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IMMUNITY TO VIRUSES

P1: FDK

713

produced by B cells in T cell–independent responses (5), but the evidence that these cells can mount an effective memory response also is weak. The purpose of immunological memory is to protect a host from reinfection, to control persistent infection, and to protect immunologically immature fetuses and neonates by passive transfer of maternal antibody. It has been suggested that the evolutional pressure to protect the fetus may be the primary stimulus for the development of immune memory. The argument is that if a host has survived a viral infection, it does not need memory, as it has already demonstrated its ability to resist that virus without it (6). This argument, however, discounts the fact that a host might not survive a higher dose of that pathogen or that a host continually ill with acute viral infections would be at an evolutional disadvantage even if the infections do not kill it. Further, the “herd immunity” of communities reduces the incidence of virus and its opportunity to spread to others. Based on experimental models, we discuss here new concepts in T and B cell memory to viral infections.

INDUCTION OF THE IMMUNE RESPONSE Viruses most often enter the host through respiratory, gut, or genital mucosal surfaces, although they also may directly enter the blood stream via arthropod vectors, animal bite, or hypodermic syringe (7). Entry through the blood is likely to shunt the virus directly into the spleen, where T cells, B cells, and antigen presenting cells (APC) are readily available to initiate an immune response. In the more common peripheral routes the viruses will likely encounter dendritic cells, APC that develop from bone marrow progenitors, enter the circulation, and home to peripheral tissue, where they await activation by “danger” signals (8, 9). These danger signals may arise from a direct infection of the dendritic cell or from virusinduced death of neighboring cells. They also arise from virus-induced cytokines caused by the interaction of viral proteins with Toll receptors (10, 11) or by the activation of protein kinase R (PKR) by double-stranded viral RNA (3). These activated antigen-bearing dendritic cells express the chemokine receptor CCR7 and migrate into the draining lymph nodes, where they encounter T and B cells and initiate immune responses (12).

The Primary T Cell Response Na¨ıve T cells migrate throughout the body and collect in the secondary lymphoid organs, such as the spleen, lymph nodes, and Peyer’s patches (13, 14). Entry into the spleen is from the splenic artery. These T cells migrate to the white pulp and congregate around arterioles in periarteriolar lymphocyte sheaths. After a few hours they move to the red pulp and then exit the spleen in the venous blood. Entry of na¨ıve T cells into the lymph nodes is through high endothelial venules, expressing the addressin PNAd and the chemokine SLC, which are the respective ligands for the homing receptors CD62L and CCR7. These migrating na¨ıve T cells enter the

15 Mar 2004

15:39

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

714

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

lymph node paracortex and eventually leave the node by lymphatic vessels and re-enter the blood stream by the thoracic duct. This provides a mechanism for the continual transient deposition and then recirculation of T cells into the lymphoid organs that colocalize APC and B cells. The frequency of T cells recognizing any given peptide is usually quite low. Na¨ıve CD8 T cell precursor frequencies to lymphocytic choriomeningitis virus (LCMV)-infected targets are about 1/50,000 CD8 T cells (15, 16). It is thus important to have significant lymphocyte circulation to optimize the chances for these low-frequency encounters. The interaction of the T cell with the APC results in the triggering of the TCR by the antigen-presenting MHC and by an augmentation of this process by costimulatory molecules. CD28 on the na¨ıve T cell interacts with B7-1 or B7-2 on the APC, and this augments the synthesis of IL-2 (17). These newly activated T cells will be induced to express CD40L, which can interact with CD40 on APC, including dendritic cells, macrophages, and B cells, sending important messages between the interacting cell types (18). CD27, another member of the CD40 family, is important for acute and memory T cell responses with some viruses (19). This initial interaction may have long-term influences on the stability and recall of memory immune responses (20–24). There then follows a dramatic proliferative expansion and differentiation of the responding T cells, which up-regulate growth factor receptors, alter their expression of chemokine receptors and cell adhesion molecules, produce immunoregulatory or antiviral cytokines, and, in the case of CD8 T cells, develop differentiated granules containing cytotoxic molecules such as perforin and granzymes (25) (Figure 1). These T cells down-regulate CCR7 and CD62L, leave the spleen and lymph nodes, and migrate into peripheral tissue to encounter the virus (26, 27). The proliferative expansions of CD8 T cells are preprogrammed events requiring only a brief encounter with the APC (28–30). For example, 5,6-carboxy-fluorescein (CFSE)-labeled T cells exposed to APC in vitro for as little as 2 h underwent >7 rounds of division, as shown by loss of CFSE label, when transferred into an antigen-free mouse (29). Calculations by LCMV-specific limiting dilution assays (LDA) or by monitoring the increase in numbers of transgenic T cells seeded into infected mice have indicated that T cells may undergo 14 to 15 cycles of division, or as many as 3 divisions a day, before the peak of the immune response (15, 16, 31). CD4 T cells tend to undergo a more limited number of divisions on contact with antigen, possibly explaining why the CD4 to CD8 ratio increases during infections (32).

Conversion of the T Cell Response into the Memory State: Immune Silencing by Apoptosis and Dissemination Infections have often resolved by the time the immune response reaches its peak, and there follows a contraction or “silencing” phase, associated with cell loss and the conversion of an acute effector response into a memory response (Figure 1). This is accounted for by two phenomena: the apoptosis of the activated T cells

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IMMUNITY TO VIRUSES

P1: FDK

715

(33) and the emigration or “diaspora” of T cells out of the spleen and lymph nodes and into the peripheral tissue (34–38). Activated T cells paradoxically synthesize both proapoptotic (Fas, FasL) and antiapoptotic (Bcl-2, Bcl-XL) regulatory proteins (39, 40). Low-affinity T cell interactions may stimulate cells into cycle but not induce enough Bcl-XL to prevent cells from rapidly undergoing apoptosis (41). This favors the expansion of highaffinity antigen-specific T cells and makes for a more efficient immune response. Nevertheless, apoptosis occurs even in the high-affinity populations, and overstimulation of the TCR with antigen can lead to the “activation-induced cell death” (AICD) of antigen-specific T cells. This AICD is usually mediated by Fas/FasL interactions, where the ligated Fas molecule engages the Fas-activated death domain (FADD), which initiates the caspase cascade leading to apoptosis (39). The silencing of the T cell response as infections resolve, however, occurs normally in mice with mutations in Fas and FasL (42, 43), consistent with a process distinct from AICD. This is not surprising, as the antigen required to signal the TCR is mostly gone before the T cell response contracts. T cell silencing is delayed in mice lacking IFNγ (44), IFNγ R (45), or perforin (44), but the significance of this is unclear owing to the delayed clearance of antigen. Silencing also occurs somewhat normally in mice expressing either Bcl-2 or Bcl-XL transgenes in their T cells (33, 46). This is a bit of an enigma, as CD8 and CD4 memory T cells display enhanced expression of Bcl-2 (47, 48). LCMV-activated T cells from CD43-deficient mice exhibit a more modest silencing phase and have elevated Bcl-2 expression in their CD8 T cells, consistent with this surface molecule regulating apoptosis during the silencing phase (49). The silencing of the immune response is also associated with the diaspora of T cells from the lymph nodes to the periphery, where they are maintained at high frequency with minimal cell division (34–38). Of note is that, in LCMV-infected mice, a much higher frequency of CD8 T cells in the spleen and lymph nodes than in peripheral tissue [peritoneal cavity (PEC), lung, and fat] express a preapoptotic phenotype, i.e., reactivity with Annexin V, expression of Fas, and sensitivity to TCR-triggered AICD (50). Transfer of LCMV-specific transgenic T cells from the spleen into a PEC environment rendered them resistant to AICD, whereas transfer of transgenic T cells from the PEC into a spleen leukocyte environment rendered them sensitive to AICD (50). Many factors have been shown to regulate the AICD of T cells, including cytokines such as IL-2, IL-6, IFNα, IFNγ , and TGFβ, α1β2 integrins, and CD44 ligation (39, 50), and these could be responsible for some organ-dependent differences in lymphocyte survival. The apoptotic process at the end of infection represents the transition point between the acute and memory response. What then distinguishes cells that survive from cells that do not, and what is regulating this apoptosis? The evidence indicates that the apoptosis is at least in part stochastic and does not depend on the TCR. Analyses of the frequency of T cells specific for different LCMV epitopes showed that their distribution in respect to each other is similar at the peak of the acute response and in the long-term memory state (51, 52). Because of the stochastic

15 Mar 2004

15:40

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

716

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

nature of the TCR repertoire that leaves the thymus (53) and the chance encounter of antigen-specific T cells with APC, individuals use different TCR to make similar antigen-specific responses (54, 55). Longitudinal analyses of TCR usage in individual LCMV- or influenza virus–infected mice showed similar TCR distributions between the peak, silencing phase, and memory responses (54, 56, 57). Although survival of CD8 T cells into memory does not appear to be dependent on the TCR, it is associated with a subpopulation of cells expressing high levels of IL-7 receptor (R). These have elevated levels of Bcl-2, whereas IL-7Rlo–expressing cells display markers for caspase activation (58). Transfer of IL-7Rhi but not IL-7Rlo T cells into mice gave rise to memory cells, indicating that the IL-7Rhi cells were indeed precursors for memory.

The Primary B Cell Response Na¨ıve B cells first encounter antigen, CD4 T cells, and dendritic cells in the T cell–rich white pulp region in the spleen. The three-dimensional structures of nonprocessed viral proteins are recognized by the membrane-bound immunoglobulin molecules that function as B cell receptors (BCR), which are complexed to Igα and Igβ chains involved in transduction of the BCR-derived signals (59). Interaction of B cells with antigen and the appropriate costimulatory signals leads to B cell activation. The activated B cells subsequently follow one of two alternative paths. One path is to clonally expand and differentiate into effector PCs in the extrafollicular T cell–rich areas. These PCs may produce IgM or may undergo isotype switching and secrete IgG or IgA. The BCR of these extrafollicularly developing PCs are encoded by the unmutated germline repertoire, and the lifespan of these cells is short (3 to 4 days) (60). The second path is to enter the follicles and start to participate in the germinal center (GC) reaction (Figure 2). The formation of the specialized microenvironment of GC in secondary lymphoid tissues is crucial for the generation of long-term antibody responses and B cell memory. The extrafollicular differentiation into short-lived PCs can occur in a T cell–independent fashion, whereas the GC pathway resulting in long-term B cell responses in the form of antibody-secreting long-lived PCs and nonsecreting memory B cells requires CD4 T cell help (61, 62).

Conversion of the B Cell Response to the Memory State: The Germinal Center Reaction B cells activated by antigen and T cell help enter the follicles and become GC founders. This migration is driven by the expression of CXCR5 on B cells, allowing them to respond to the CXC chemokine BLC. CXCR5-deficient mice have greatly impaired lymphocyte migration into the splenic follicles and completely lack GC formation (63). The BLC ligand for CXCR5 is expressed in the follicles by nonlymphoid cells, most likely by follicular dendritic cells (FDC) (64). Of note is that about 75% of HIV-infected individuals have impaired expression of

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IMMUNITY TO VIRUSES

P1: FDK

717

CXCR5 on B cells (65). The expression of CD40L on activated helper T cells is essential for the establishment and maintenance of GC. CD40L-deficient or CD40deficient mice lack GC, and the administration of anti-CD40L-specific antibodies leads to the quick disappearance of established GC (66). Signaling of CD40 on B cells by CD40L on activated T cells rescues B cells in the GC from apoptotic death (66). B cells in the GC re-express some of the genes that were expressed at earlier developmental stages in the bone marrow, including Rag-1 and Rag-2, GL-7, and peanut agglutinin receptor (67). The GC founder B cells start to proliferate rapidly in the dark zone, which is the site for somatic hypermutation and isotype switching. Both of these processes require the expression of the RNA-editing enzyme AID (activation-induced cytidine deaminase) in B cells (68). Migration to the adjacent light zone is followed by an important selection step. Follicular dendritic cell (FDC)-bound antigen and costimulatory signals, most importantly CD40L, provide survival signals for cells that have a BCR of a given affinity. In the absence of rescue signals B cells are programmed to die by apoptosis (69, 70). Additional undefined signals may be important for the generation of GC-dependent longterm B cell responses. SAP is a small adaptor molecule associated with a variety of signal-transducing leukocyte receptors such as SLAM, 2B4, CD84, and Ly9 (71). SAP-deficient mice mounted nearly normal antibody responses to LCMV, up to day 10 post infection, but thereafter the number of antibody-secreting cells dropped, GC development declined, and B cells failed to differentiate into longlived PCs and memory B cells (72). This defect was due to the SAP deficiency in CD4 T cells. SAP deficiency in humans is manifested as X-linked lymphoproliferative disease, which is characterized with immunoglobulin abnormalities and fatal mononucleosis associated with Epstein-Barr virus (EBV) infection. B cells leaving the GC become more resistant to apoptosis, upregulate Bcl-2 and Bcl-XL, and downregulate Fas (73). They become either PCs, which are large antibody-secreting cells, or memory B cells. The decision between these two paths is influenced by the affinity of the BCR, costimulatory signals, and cytokines (67, 74). Viral antigens may persist on FDC for months after infections, and GC may be maintained for months (75). This may in part account for the continued elevation of antiviral antibody that occurs after an infection has been cleared.

The Kinetics of Antiviral Antibody Responses On days 3 or 4 after systemic infection of mice with LCMV, vesicular stomatitis virus (VSV), or polyomavirus (PyV), antiviral antibody-secreting cells become detectable in the spleen by ELISPOT assay, and virus-specific IgM can be measured in the serum by ELISA (76–78). This early IgM response is followed by isotype switching, resulting in the appearance of virus-specific IgG-secreting cells in the spleen starting on day 4–5 post infection in LCMV-infected and also in PyVinfected mice. Virus-specific serum IgG levels peak on day 15 in LCMV-infected and on day 21 in PyV-infected mice (5, 77), and these antibody levels remain high

19 Feb 2004

14:33

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

718

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

for the rest of the life of the mice (79–81). The time of appearance of antibodies with neutralizing activity varies in different virus infections. For many viruses, such as VSV, rotavirus, and influenza virus in mice or yellow fever virus in humans, neutralizing antibodies appear during the first week post infection (82–85). In contrast, neutralizing antibodies specific to LCMV, HIV, or hepatitis B virus only appear several months (50–150 days) after infection (79, 86). In most cases, the mechanisms causing the delay in the appearance of neutralizing antibodies are not known. Depletion of CD8+ T cells in LCMV-infected mice resulted in a much earlier appearance of neutralizing antibodies, leading to the hypothesis that B lymphocytes specific for neutralizing LCMV epitopes bind the virus and serve as targets for virus-specific CTL activity (87).

The Importance of Isotype Switching and Somatic Hypermutation in Antiviral Antibody Responses Somatic hypermutation and selection changes the immunoglobulin variable region and leads to an increase in the affinity of the antibodies. Class switch recombination, on the other hand, results in changes of the heavy chain constant region and the production of antibodies of different isotype classes. Isotype switching is largely influenced by cytokines. In many acute virus infections the production of IgG2a is predominant, because these infections induce IFNγ secretion by NK cells and T cells, and this cytokine is a potent switch factor for IgG2a (88). Changes in the conditions of infection, however, can influence the isotype responses. IgG2a is the predominant antibody during acute LCMV infection, but congenitally infected mice, whose T cells are mostly tolerant to infection, make primarily an IgG1 response during their lifelong persistent infection (89, 90). Many viruses can also act as T cell–independent antigens and elicit protective antibody responses in the absence of CD4 T cell help (91). For example, PyV-infected TCR αβγ δ knockout (KO) mice secrete virus-specific IgM and IgG antibodies and control the infection, which is lethal in B cell– and T cell–deficient SCID mice (5). These T cell–independent responses must occur extrafollicularly, as the PyV-infected T cell–deficient mice completely lack GC (5). Crosslinking of the BCR by repetitively organized virion epitopes may be important for the efficient induction of T cell–independent antibodies, but repetitiveness of the antigen may not be sufficient to induce isotype switching (76, 92). In the case of PyV, T cell–independent IgG or IgA are induced by live virus infection but not by noninfectious virus-like particles (76), suggesting that the replicating virus induces some helper functions. IgG secretion was also observed in LCMV- and vaccinia virus (VV)–infected CD40L-deficient mice (93) and in VSV-infected αβ TCR KO mice (94), and protective IgA responses were produced by T cell–deficient mice after rotavirus infection (95). Recent studies indicate that, on exposure to IFNα, IFNγ , or CD40L, dendritic cells upregulate TNF family members (Blys, APRIL) that act on B cell signaling receptors (TACI, BAFFR, and BCMA), and induce isotype switching in the presence of additional cytokines, like IL-10 or TGFβ

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IMMUNITY TO VIRUSES

P1: FDK

719

(96). The ability of viral infections to elicit the synthesis of IFNα and IFNγ may thus enable them to induce T cell–independent IgG and IgA. The isotype of antiviral antibodies alters their function, and the in vivo antiviral efficacy of recombinant antibodies that have identical variable regions, specificities, and affinities, but different isotypes, may vary. Passive transfer of influenza virus–specific IgG mediated both prophylactic and therapeutic effects in SCID mice, whereas transfer of virus-specific IgM and IgA with similar variable regions was effective only prophylactically (97). On the other hand, mice able to secrete antiviral IgG and IgA but not IgM also had increased susceptibility to influenza virus (98). As the immune response progresses, there is an increase in average affinity of the secreted antibodies as a result of somatic hypermutation and selection in the GC. Affinity maturation of antiviral antibodies has been demonstrated in influenza virus–infected mice (99). Influenza virus infection is significantly more pathogenic in AID-deficient mice, which could produce only unmutated IgM and not highaffinity isotype-switched antibody responses (100). Affinity maturation may not always be needed, however. The average affinity of IgG antibodies to VSV G glycoprotein is high at the early phase (day 6) of infection and does not show increased affinity due to somatic mutation at later phases (82). Perhaps there was selective pressure for mice in the past to encode germline high-affinity receptors for a pathogen with antigens related to those of VSV.

PROPERTIES AND ACTIVATION OF MEMORY T AND B CELLS Characteristics of Memory T Cells T cells can be characterized as na¨ıve, effector, or memory cells. Memory T cells can sometimes be distinguished from na¨ıve T cells by altered expression of integrins, activation molecules, and chemokine receptors (for reviews, see 13, 101). Only a few molecules, such as a glycosylated form of CD43 demarked by mAb 4B11, can differentiate memory cells from activated effector cells (49, 102). Part of the problem is the considerable heterogeneity of memory T cell populations and the fact that many cells in the body are masquerading as memory cells but have not gone through the memory cell differentiation program (103, 104). Bona fide memory T cells need to be distinguished from “pseudo” memory cells that arise as a consequence of homeostatic proliferation. The immune system senses lymphopenic conditions, and na¨ıve cells will up-regulate memory marker antigens, such as CD44, as they proliferate to fill up space (105). The homeostatic proliferation of na¨ıve CD4 or CD8 lymphocytes is initially dependent on IL-7 (106–108), and the cells respond to self, not foreign antigens (109, 110). Although they can acquire a memory cell phenotype, they do not undergo a foreign antigen and IL-2-dependent expansion and silencing phase, and hence are distinct from virus-specific memory cells.

19 Feb 2004

14:33

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

720

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

The heterogeneity of virus-specific memory CD8 T cells was first shown by cells analyzed by virus-specific limiting dilution assays (111, 112). CTL precursors were found at similar frequencies in both small 2N DNA-containing and in large 4N DNA-containing cells, indicating that a subfraction of the cells was proliferating (111). They also were heterogeneous for CD62L expression, where the in vivo blast cells were CD62Lhi, in contrast to the acute infection, where dividing T cells were mostly CD62Llo (111, 112). CD62L has subsequently been shown to colocalize on T cells with the lymph node–homing chemokine receptor CCR7 (26, 27), and the terms “central memory” and “effector memory” have been used to distinguish CD62Lhi CCR7hi from CD62Llo CCR7lo T cell populations, respectively, in humans and in mice (113, 114). Whether there is a uniform or linear differentiation pathway is not clear (113, 115–117), despite extensive study. There is, however, some consensus that the central memory cells cycle more rapidly than effector memory cells prior to antigen exposure and that they produce more IL-2 and proliferate faster when antigen is present. On antigen exposure, central memory cells will differentiate into effector cells, migrate into peripheral tissue, and mediate effector function (118, 119). Memory and effector cells are probably best defined by their activation status rather than phenotypic markers: Effector cells are antigen-experienced T cells acting in response to the antigenic load during primary infection, secondary infection or viral reactivation, whereas memory cells are resting antigen-experienced T cells in the absence of antigenic stimulation.

Activation Properties of Memory T Cells The efficacy of memory cells is a consequence of their increased frequency, their ease in activation, and their ability to reside within or migrate into peripheral tissues. Individual T cells do not undergo a somatic mutation-based affinity maturation like B cells, but there is a selection of T cell populations with higher affinity during the acute infection when antigen is present. Some studies on the affinity maturation of T cell populations are difficult to evaluate, as there can be a “functional avidity maturation” of the T cells without a change in the TCR, due in part to enhanced adhesion molecules and changes in cytokine-producing profiles (120, 121). LCMV-specific memory CD8 T cells have initial TCR signaling events similar to those of na¨ıve cells, but enhanced downstream signaling, including the phosphorylation of LAT (linker for activation of T cells), ERK (extracellular signal-related kinase), and JNK (c-Jun-n-terminal kinase) (122). This may be due, at least in part, to more extensive lipid raft formation on memory cells than in na¨ıve T cells. Distinctions in lipid raft formation have also been reported between na¨ıve and activated T cells specific for influenza virus (123). The chromatin structure needs to be loosened before naive T cells generate cytokines (124), but this modification has already occurred in memory cells, which constitutively express mRNA for some effector proteins, including IFNγ , MIP-1α, RANTES, perforin, FasL, and granzyme B (125), and which rapidly produce chemokines and cytokines after a brief TCR stimulation (126, 127).

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

IMMUNITY TO VIRUSES

P1: FDK

721

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Some memory cells appear to be in higher states of activation than others and can mediate cytotoxic functions. Blast-sized memory cells from LCMV-immune mice can lyse sensitive peptide-coated target cells directly ex vivo (128, 129), and studies using CSFE-labeled LCMV epitope-pulsed splenocytes as targets for in vivo cytotoxicity assays have shown a rapid lysis of those target cells within 4 h of transfer into LCMV-immune mice (130). Memory T cells therefore represent a highly effective and rapidly mobilizable effector system.

Stability and Maintenance of Memory T Cell Populations Studies in humans have shown that CD8 (131, 132) and CD4 (133) T cell responses to viruses may last for decades after infection without any obvious re-exposure, although more quantitative studies demonstrating memory T cell stability have been done in mice, which have much shorter life spans (52, 134). Antigen is not needed for the maintenance of CD4 or CD8 memory T cell populations. This was demonstrated by the survival of antigen-primed T cells transferred into mice lacking both antigen and the presenting class I or class II MHC molecules (135, 136). This contrasted with na¨ıve phenotype CD4 or CD8 T cells, which required MHC molecules for survival and homeostatic proliferation (109). It has been argued that few viruses, including LCMV, are ever truly cleared from the host, and their antigens may provide sufficient stimuli to maintain a memory cell population (6, 137). However, if antigen were driving the responses, one would expect some evolution of the TCR repertoire, and this does not happen where studied in the LCMV and influenza systems (54, 56, 57). The stability in CD8 T cell memory is achieved by continuous low-level division (111) that is dependent on IL-15, perhaps with IL-7 as an auxiliary back-up system (138–140). The long-term stability of CD8 memory cells specific to VSV (141) and LCMV (142) was decreased in IL-15 KO mice. IL-15 signaling of memory CD8 T cells elevates the expression of Bcl-2, and this may enhance the ability of these cells to survive (121). One study showed that IL-15α receptors need to be on cells in the host but not specifically on the memory cells themselves, arguing for indirect effects (143). The signals regulating cytokine production and the survival of memory CD8 T cells are not understood, but it has been suggested that the signaling of CD8αα on memory CD8 T cells by the class I thymic leukemia (TL) antigen may maintain memory in some instances (144, 145). Fewer studies have been done on the stability of virus-specific CD4 memory in animal models. LDA for IL-2-producing CD4 T cells in the LCMV system showed them to be very stable (146), but studies with peptide-induced intracellular IFNγ assays and with the binding of class II MHC tetramers showed modest declines under conditions of greater CD8 T cell stability (47, 147). Studies with Sendai virus in mice showed first a decline and then an unusual increase in memory CD4 T cells in mice over one year of age (148). The reason for this has not been determined, but a similar memory cell inflation for CD8 T cells has been seen in MCMV infection, where antigen persists (149).

19 Feb 2004

14:33

722

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Impact of Early Signaling Events on the Stability and Recall of Memory T Cell Populations The long-term stability of CD8 T cells and their ability to mount a secondary immune response requires signaling from CD4 T cells and CD40/CD40L interactions during the early stages of the primary response (20–24, 93). Studies in the LCMV, VV, and Listeria monocytogenes systems all showed that, whereas CD8 T cells are generated and enter the memory state in the absence of CD4 T cells, their ability to respond to recall antigens in vivo or in vitro is dramatically impaired in the absence of CD4 T cells. Notably, the CD4 T cells were required during the initial phase of the primary response and were not required at the time of the recall response (20). This means that different differentiation programs develop in CD8 T cells early after receiving CD4 T cell help and that they last for a long time in the absence of CD4 T cells. This might come in part through an interaction of CD40L on the CD4 T cell and CD40 on the CD8 T cell, as shown by experiments with transgenic CD8 T cells from CD40-deficient mice (23). Early T cell–dependent signaling events are also important for the differentiation program in B cells, but the belief is that, in contrast to CD8 T cells, helper CD4 T cells must be present at the time of memory B cell reactivation (150).

Role of Memory T Cells in Resistance to Secondary Challenge with Homologous Virus Memory T cells, like na¨ıve T cells, undergo a multicycled proliferative expansion when re-encountering antigen (151, 152). They nonspecifically migrate into peripheral sites of infection, and, if they find their cognate antigen, they will remain to combat the infection (153, 154). While it is well documented that resistance to homologous viral challenge is in large part mediated by neutralizing antibody, memory T cells undoubtedly can play a role. First, vaccination with T cell epitopes or vectors encoding such epitopes can protect mice and monkeys from infection with a variety of viruses, including LCMV and simian immunodeficiency virus (SIV), respectively (155–158). Second, the presence of memory T cells in peripheral tissue correlates with protective immunity. CD4 and CD8 memory T cells are initially at high levels in the lung after infection of mice with Sendai virus or influenza virus, but they gradually decline after a few months, in contrast to the spleen, where their frequencies are more stable (159, 160). Resistance of influenza-immune mice to an intranasal challenge with a neutralizing antibodynoncross-reactive serotype of influenza correlated with the duration of the memory T cells in the lung (159, 160), and intratracheal transfer of Sendai virus–specific memory CD4 T cells from an immune lung into a na¨ıve lung rendered resistance to infection (160). A limited number of studies have shown that the TCR repertoire changes only modestly after a homologous challenge. TCR analyses showed similarities between the peak in the primary response and the recall response, but some additional clones

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

IMMUNITY TO VIRUSES

P1: FDK

723

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

appear (54, 56, 57). Some of these new clones could be due to the fact that part of the secondary response to infection is a new primary T cell response. In influenzaor LCMV-infected mice adoptively reconstituted with Thy 1-congenic memory cells, a primary response can be seen in the presence of a memory response (161, 162). Sometimes, however, there can be more dramatic changes on secondary challenge. The NP366– and PA224 Db–restricted epitopes of influenza virus stimulate comparable primary T cell responses, but the NP366 epitope dominates the secondary response. This has been linked to the ability of NP366 to be presented more efficiently than PA224 by nondendritic APC (163).

Characteristics, Maintenance, and Reactivation of B Cell Memory Humoral immunity to systemic virus infections can last for several decades in humans and for the lifetime of mice (77). Antibodies are important in preventing reinfection of the host and in providing protection to the fetus and newborn by maternal antibodies. The maintenance of high serum levels of antibodies can be defined as serological memory. The half-life of serum immunoglobulins is 7– 21 days (164), so long-term maintenance of the serum antibody levels requires continuous secretion of these molecules. This can be achieved by sustained antibody production by long-lived plasma cells (PC) (even in the complete absence of antigen), by activation of memory B cells to differentiate into PC, or, when virus persists, by recruitment and differentiation of na¨ıve B cells, which can be activated continuously to replace dying PC (Figure 2). The life span of terminally differentiated antibody-secreting PC is determined by both B cell-intrinsic factors and environmental signals, but, like the survival of memory T cells, it is not dependent on the presence of antigen (165). Antibodysecreting cells accumulated in the spleen of mice during the acute LCMV infection but thereafter decreased in the spleen and increased in the bone marrow, somewhat like the diaspora of T cells after resolution of infection, but to a more discrete destination. The PC residing in the bone marrow maintained high antiviral serum antibody levels even when memory B cells were eliminated by irradiation (81). The migration of PC to the bone marrow is initiated by their surface expression of the CXCR4. CXCR4−/− fetal liver chimeric mice have a largely diminished (∼30%) PC population in the bone marrow (166). In contrast to antibody-secreting PC, memory B cells are in a resting state and do not secrete antibodies unless antigenically stimulated. Memory B cells can be detected by ELISPOT assays as cells that start to secrete antibody after antigen restimulation in culture (167). They accumulate in the spleen after virus is cleared and after the frequency of antibody-secreting cells in the spleen declines. Re-challenge with antigen leads to fast and robust anamnestic antibody responses, due to the CD4 T cell–dependent proliferation and differentiation of memory B cells. This activation results in massive expansion of B cells in the T cell zones of the spleen and lymph nodes, a secondary GC reaction, and a rapid and massive

19 Feb 2004

14:33

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

724

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

increase in immunoglobulin secretion by a new wave of PC. How the memory pool is changed after a series of infections and how it is restored after repeated encounters with the antigen are not well understood. However, if a different challenge virus has a cross-reactive epitope to a previously encountered virus, the cross-reactive response is usually dominant. This has been referred to as “original antigenic sin,” and sequential infections with closely related but not identical strains of influenza virus or dengue virus elaborate potent antibody responses to the cross-reactive epitopes (168, 169). Although long-lived PC in mice may account for the maintenance of serum antibody levels for many months and even for the lifetime of the animals, it is less likely that the same mechanism is sufficient to maintain human antibody levels for several decades. Antigen-antibody complexes attached to FDC have an average half-life of 8 weeks (170) and therefore cannot sustain lifelong serological memory in humans. It has been suggested that the polyclonal activation of memory B cells utilizing bystander T cell help may function to maintain the serological memory in the absence of antigen. For example, the activation of Toll receptor 9 by immunostimulatory CpG nucleotide sequences, together with IL-15, and bystander help from an allospecific CD4 T cell response led to a proliferative response of memory but not na¨ıve human B cells in vitro (171). The significance of this phenomenon in vivo is uncertain, but viral infections in humans have long been known to elevate antibody levels to self antigens, suggesting potent B cell stimulation (172).

MEMORY T CELL ATTRITION Virus-Induced Lymphopenia A common feature of many severe viral infections is a transient lymphopenia that occurs prior to the peak in the T cell response (Figure 1). This lymphopenia, which is associated with reductions in most lymphocyte classes, occurs after infections of mice or humans with influenza, measles, varicella zoster (VZV), Ebola, Venezuelan equine encephalitis, and LCM viruses, among many others (104). Although the causes of lymphopenia may vary with the virus, a common mechanism may be that induced by the type I IFN response to infection (173, 174). Memory CD8 T cells are particularly sensitive to apoptosis and cell loss in the early stages of LCMV infection or after stimulation of mice with the IFN inducer poly I:C. This cell loss parallels type I IFN levels during infection and does not occur in type I IFN R KO mice (174). Virus-induced cytokines thus create a lymphopenic environment, which might be advantageous by making room for the development of a vigorous virus-specific T cell response. Notably, virus- or tumor-specific T cell responses are often enhanced in lymphopenic conditions induced by irradiation or by treatment with immunosuppressive drugs such as cyclophosphamide (175–177). The preexisting memory cells must then compete with CD8 T cells responding to the virus and other T cells responding to homeostatic signals to restore the lymphodeficiency. Virus-specific bona fide memory CD8 T cells, however, compete poorly with other

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

IMMUNITY TO VIRUSES

P1: FDK

725

T cells in reconstituting lymphopenic environments, created by genetics (T cell KO and SCID mice), irradiation, poly I:C-treatment, or viral infection (104). Thus, after cytokines induce the apoptosis of memory CD8 T cells, their recovery is sluggish, and other T cells either undergoing homeostatic proliferation or else responding to the new antigenic challenge dilute them out.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Attrition of CD8 T Cell Memory by Heterologous Viral Infections Studies in the mouse have shown that as many as 5% of the CD8 T cells can be directed against a single viral epitope (51, 178), indicating that the immune system cannot accommodate an ever-increasing memory pool without either enlarging its lymph nodes, which it does not do, or deleting pre-existing memory T cells, which it does do (52, 178). In mice sequentially infected with LCMV, Pichinde virus (PV), VV, MCMV, and VSV, each infection led to reductions in memory CD8 T cells specific to previously encountered viruses (52, 178). After several infections there was a >90% reduction in memory cells in some cases. In another study, mouse γ herpes virus (Mγ HV) persistent infection led to a substantial decline (∼50%) in T cells specific for influenza virus (179). The losses in memory T cell frequencies were not identical for all CD8 T cell epitopes (178), in part due to T cell crossreactivity between epitopes of unrelated viruses. For example, LCMV and PV encode subdominant epitopes (LCMV-NP205 and PV-NP205) with 6 of 8 amino acids in common (Table 2). Infection of LCMV-immune mice with PV ultimately leads to a deletion in CD8 T cell frequencies to most of the LCMV-encoded epitopes, but not to NP205, which increases in frequency (180). Thus, memory T cell responses to cross-reactive epitopes are maintained, whereas responses to noncross-reactive epitopes are lost during virus infections in sequence (Figure 1). This phenomenon may resolve prior controversies regarding the need for antigen to maintain T cell memory (181). T cell memory seems very stable in the absence of any overt infections, but in nature the host experiences many infections that should reduce that memory pool. Under those natural conditions, a re-exposure to a homologous or a cross-reactive antigen would help to maintain memory. The loss of CD8 T cell memory after heterologous viral infections could be explained by a passive model, whereby the pre-existing memory T cells are displaced from a finite number of protective niches simply by competition with the newly arising memory cells, or by an active model, whereby memory T cells are directly killed off and fail to return to their prior numbers. There is some evidence to support the concept of an active model, as the studies on virus-induced lymphopenia discussed above show that memory CD8 T cells undergo apoptosis in the wake of virus-induced cytokines (174). BrdU incorporation studies have suggested that there is bystander division of CD8 memory T cells during infection (182), but this appears to be an IL-15-dependent homeostatic division that is at least partly in response to the IFN-induced apoptosis and may not be “proliferation” in the true sense of the word, meaning an increase in number (174). An assessment of bystander CD44 + CD8 T cells known not to be specific to or cross-reactive with

19 Feb 2004

14:33

726

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 2 Some cross-reactive viral class 1–restricted T cell epitopesa Virus/epitope

Sequence

Reference

Mouse LCMV NP-205 (Kb): PV NP-205 (Kb): VV P1 (DNA pol) (Kb) VV P10 (E7R) (Kb) VV P24 (16.7K put) (Kb)

Y T V KY P NL Y T V KF P NM YN S L Y P N V S T L N F N NL AIVNYANL

(180)

FLU A NP-147 (Kd): FLU A PB2-146 (Kd): FLU A PB2-185 (Kd):

T Y Q R T RA L V R T G V F P N E V G A R I LT S E ERELVRKTR

(188)

RSV M2-82 (Kd): RSV M2-71 (Kd):

SYIGSINNI EYALGVVGV

(189)

FLU A HA-515 (Kd): FLU A NS1-50 (Kd):

ILAIYSTVASSL LGLDIETATRAGKQIVERI

(190)

FLU A NP-366 (Db): FLU A NP-366 variant (Db):

ASNENMDAM ASNENMETM

(191)

Dengue 2 NS3-298 (Kd): Dengue 3 NS3-299 (Kd):

GYISTRVEM GYISTRVGM

(192)

Human FLU A M1-58 (A2): EBV BMLF1-280 (A2):

G I L G F VF T L GL C T LVA M L

Selin et al.c

FLU A NA-231 (A2): HCV NS3-1073 (A2):

C V N GS CF T V C V N GV CW T V

(197)

HPV16 E7-11 (A2): Coronavirus NS2-52 (A2):

Y M L DL Q P E T TMLDIQPED

(195)

HIVenvgp-120 (A2): M. tuberculosis 19 kDa (A2):

VPTDPNPPEV VLTDGNPPEV

(198)

Rotavirus VP4- 86(A2): FLU A M1-55 (A2):

CPTNQQVVLEGTNKTD LTKGILGFVFTLTVPSERG

(196)

Dengue 2 NS3-71 (B62) Dengue 3 NS3-71 (B62)

DVKKDLISY SVKKDLISY

(193)

Hantavirus Sin Nombre N-421 (A1): Hantavirus Seoul virus N-421 (A1):

ISNQEPLKL ISNQEPMKL

(194)

Selin et al.b

a

This table lists the virus, viral protein, epitope number, and class I MHC-restricted element for various mouse (top) and human (bottom) cross-reactive CD8 T cell epitopes. FLU = influenza virus.

b c

Unpublished epitope data from L. Selin, B. Sheridan & M. Cornberg.

Unpublished epitope data from L. Selin, S. Clute, Y. Naumov, K. Luzuriaga & J. Sullivan.

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IMMUNITY TO VIRUSES

P1: FDK

727

LCMV showed a decrease in their number and increase in their apoptosis during the acute infection (174). BrdU incorporation studies in the influenza virus and Mγ HV systems showed much more incorporation in antigen-specific cells than in putative “bystander” cells (183, 184). Adoptive transfers of CSFE-labeled LCMV-immune splenocytes into Thy 1 congenic mice showed that LCMV stimulated significant increases in number and >7 divisions of the CD8 and CD4 donor cells, but poly I:C stimulated limited CD8 T cell division that was offset by sufficient apoptosis to cause a loss in number (152). Thus, the combination of bystander T cell loss, their only moderate capacity to recover, and the competition from newly emerging antigen-specific T cells, could all contribute to loss of pre-existing memory T cells during viral infections.

Attrition of CD4 Memory Cells Less is known about the stability of memory CD4 T cell frequencies in the wake of other pathogens. One study showed that, compared to age-matched controls, sequential infection of LCMV-immune mice with PV, VV, MCMV, and VSV did not cause an attrition of CD4 T cell memory to LCMV (185). The curiosity here is that CD8 memory may be very stable but rapidly reduced by other infections, whereas CD4 memory may be less stable, but unfazed by other infections (47, 185). These differences in memory attrition, however, may have more to do with the fact that these viral infections cause a much greater proliferative expansion of CD8 than of CD4 T cells, putting a greater strain on the immune system to accommodate newly formed memory CD8 T cells along with the pre-existing ones. Bacteria tend to be stronger inducers of CD4 T cell responses, and CD4 T cell memory to L. monocytogenes in mice was reduced after infection with Mycobacterium bovis (186).

CROSS-REACTIVITY AND HETEROLOGOUS IMMUNITY Memory T cells specific to one virus can unexpectedly play roles in the pathogenesis of putatively unrelated viruses, and this is referred to as heterologous immunity. The nature of T cell recognition is degenerate in that a single T cell can often recognize more than one epitope, with one calculation suggesting as many as 106 peptideMHC combinations (187). CD8 T cells have been shown to be cross-reactive between two different epitopes on the same viral protein (188, 189), between two proteins within the same virus (188, 190), between similar proteins of closely related viruses (180, 191–194), and between different proteins of unrelated viruses (L.K. Selin, S. Clute, Y. Naumov, and M. Cornberg, unpublished; 195–197) (Table 2). These cross-reactive epitopes may or may not have significant amino acid homology. Cross-reactive epitopes have also been shown between proteins of viruses and intracellular bacteria (198). The mobilization of cross-reactive memory cells into a primary immune response can alter protective immunity, immunopathology, and the immunodominance of subsequent T cell responses (31, 180).

19 Feb 2004

14:33

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

728

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

In a systemic i.p. inoculation route model, several viruses (LCMV, PV, MCMV) were each shown to provide a level of protective immunity against VV by 3 days postinfection. Protective immunity to VV could be transferred by LCMV-immune CD8 and CD4 T cell populations and was dependent on IFNγ (199). Protection was also seen against lung VV titers in intranasally inoculated mice. Notably, by 3 days postinfection, some of the memory T cells of each epitope specificity to LCMV were making IFNγ in vivo in response to the VV infection (37). It is not clear whether this early IFNγ production was caused by cross-reactivity or by a nonselective cytokine-dependent activation, but VV elicited the expansion of selected specificities of LCMV-specific T cells later in infection, arguing for cross-reactivity, and putative cross-reactive epitopes between LCMV and VV were identified (Table 2). Dramatic differences in immunopathology were noted in these models. I.p.-inoculated mice developed severe T cell infiltration and necrosis of visceral fat, with a pathology similar to the pannicultis seen in human erthyema nodosum. Of interest is that erythema nodosum–like lesions are associated with viral and bacterial infections and sometimes seen during vaccinations (31, 200). The respiratory infection revealed dramatic differences in pathology associated with the accumulation of LCMV-specific T cells in BALT-like structures and the presentation of bronchiolitis obliterans, similar to a human disease sporadically associated with infections and transplant rejections (201). In several respiratory models, prior infections often predisposed the host to more dramatic infiltrates by IFNγ -producing lymphocytes, leading to protective immunity, but this was not always the case. A history of an LCMV infection protected mice against VV but inhibited the clearance of RSV (202, 203), whereas a history of influenza infection protected mice from VV but rendered them more susceptible to MCMV and LCMV (202). Patterns of heterologous immunity can therefore be complicated and hard to predict, but they are reproducible. Heterologous immunity may alter the Th1/Th2 response. RSV infections in children are sometimes associated with severe eosinophilia that is likely dependent on Th2 CD4 T cell cytokines (204). RSV will cause a comparable syndrome in mice if they are first immunized with VV encoding RSV G protein (205, 206). Immunization with G primes for a discrete population of Th2 CD4 memory cells that are elicited and do damage on RSV challenge (207). However, if mice receive an influenza virus infection prior to the recombinant VV-G immunization, the RSV-challenged mice fail to develop the eosinophilia and tolerate the infection (206). This shows how three viruses in sequence can alter the pathological picture of the third virus. T cell cross-reactivity can also alter imunodominance patterns (208). LCMV and PV encode cross-reactive subdominant epitopes (LCMV-NP205 and PVNP205) (Table 2). However, if mice immune to one of these viruses are then challenged with the other, the NP205 response becomes dominant and the normally dominant responses are much reduced (Figure 1) (180). This type of phenomenon has also been called original antigenic sin in T cells (209). This skewing away from the “normal” T cell response caused by the increased frequency of T cells

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IMMUNITY TO VIRUSES

P1: FDK

729

cross-reactive to an otherwise subdominant epitope may possibly lead to a less effective T cell response. Original antigenic sin in B cells has been described with dengue virus, which exists in four serotypes defined by neutralizing antibodies (169). Cases of dengue hemorrhagic fever (DHF) are often much more severe when an individual is infected with a second serotype. This has been proposed to be a consequence of “immune enhancement” antibodies, that is, when nonneutralizing antibodies to the first encountered virus bind to the second virus and enhance its ability to infect macrophages by way of Fc receptors (169). However, T cell responses against different dengue viral serotypes also are highly cross-reactive (Table 2) (192), and DHF is associated with high levels of T cell– dependent cytokines (210). Dengue virus-specific T cells generated during DHF have low affinity to the serotype causing infection and higher affinity to other serotypes (211). This may mean that, as a consequence of heterologous immunity, “bad” T cells are being generated and the “good” T cells that would be effective in clearing the virus are suppressed. Many viruses, including EBV, VZV, measles, mumps, and polio, cause more severe pathology in young adults than in children, and it has been speculated that the recruitment of memory T cells specific to previously encountered viruses may alter the pathology (31). For example, the difference between subclinical EBV infections and mononucleosis is the magnitude of the CD8 T cell response that develops (212). Our recent studies have shown that many of the EBV-specific T cells elicited during acute mononucleosis cross-react with influenza virus (S. Clute and L.K. Selin, manuscript in preparation). T cell cross-reactivity and heterologous immunity are important issues in the transplantation field. Many virus infections in mice and man induce virus-specific T cells that cross-react with allogeneic MHC molecules (213–215), and viral infections have often been found to precede the rejection of human allografts (216). Viral infections of mice leave them with much higher frequencies of allo-specific memory T cells than na¨ıve mice (213, 215, 217), and sequential infections with different viruses can further alter this allo-specific memory T cell repertoire (215, 217). Tolerization models have been developed to enable hosts to accept engraftment by allogeneic tissue. However, viral infections can break tolerance and stimulate the rejection of the graft (218), and a history of a viral infection, by virtue of increasing a memory pool of allospecific T cells, can render a host more difficult to tolerize (215, 217).

IMMUNE MEMORY DURING PERSISTENT INFECTIONS Modulation of Memory T Cells During Persistent Viral Infections Infection of mice with high doses of LCMV leads to a persistent infection with a continued evolution of the T cell response and an exhaustion or anergization of T cells (54, 107, 219, 220). High levels of persisting antigen can drive T cells either

19 Feb 2004

14:33

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

730

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

into apoptosis or into different states of anergy. In persistent LCMV infection, NP396-specific CD8 T cells, which probably receive the most signaling from this highly expressed epitope, are deleted, while GP33-specific T cells remain present but fail to secrete IFNγ (220). In fact, there is a progression of dysfunction of cells in this system, in that the ability to produce IL-2 and mediate cytotoxicity is lost first, followed by the ability to produce TNFα and then IFNγ (221, 222). Despite this dysfunction the cells continue to proliferate in vivo and maintain their numbers. Many forms of T cell dysfunction have been observed in other persistent infections, including (a) cytokine-producing poorly cytotoxic cells low in perforin, as shown with HIV (223), (b) cytokine-producing poorly cytotoxic cells high in perforin, as shown in murine polyomavirus infections (224), (c) poorly cytotoxic cells that do not produce cytokines, as shown in SIV infection in monkeys (225), (d) cytokine-producing cells not readily cytotoxic, as shown with HIV infection (226), (e) cells in which this dysfunction can be restored in vitro after stimulation with cytokines or IL-2 (224–226), and (f) cells in which the defect cannot be readily corrected in vitro (223). There undoubtedly are many mechanisms associated with overstimulation of T cells that cause deletion or dysfunction in an antigenexcess environment, including significant contributions by regulatory cytokines and growth factors. A unique mechanism in the PyV system is the acquisition by CD8 T cells of negatively signaling NK receptor molecules that inhibit their ability to lyse targets or control tumor development (227). Similar inhibitory receptors have also been found on HIV- and EBV-specific T cells (2). Persistent infections in humans have also been associated with different T cell phenotypes. For example, the predominant CD8 T cell population during EBV and HCV persistent infections is reported to be CD45RA− CD27+ CD28+ (116, 117), whereas that in HIV infection is CD45RA− CD27+ CD28− (117, 228), and that in CMV infection is CD45RA+ CD27− CD28− (116, 117). Of interest is that the phenotypes of T cells specific to different epitopes of the same virus can be different within the same persistent infection. For example, during persistent EBV infection, CD8 T cells specific to the “lytic” epitope BMLF-1 are mostly CCR7− and often CD45RA+, whereas those specific to latent epitopes, which may be expressed more during persistent infection, are CCR7+ CD45RA− (229, 230). These results suggest that memory T cell subsets are created by their initial and ongoing antigen experience and cytokine environment and may have flexibility in converting from one phenotype to another, as discussed above. Although T cells become selectively deleted or functionally exhausted during persistent infections when the antigen load is high, lower levels of persisting antigen may instead maintain or even increase T cell memory. Low-level reactivation of herpes simplex virus type 1 (HSV-1) is associated with a persistent activation of CD8+ T cells in the trigeminal ganglia, which house the putatively “latent” HSV-1 infection (231). MCMV causes a low-grade persistent infection in which there is a continuous increase in CD8 T cells specific to MCMV over time, this has been referred to as “memory inflation” (149). Similarly, in humans, very high frequencies of CD4 cells specific to CMV have been reported (232).

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

IMMUNITY TO VIRUSES

P1: FDK

731

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Importance of Memory T and B Cells in Controlling Persistent Infections The maintenance of viral persistence requires continual surveillance by the immune system. Immunosuppressive drugs reactivate a number of persistent human viruses regulated by T cells, including HSV, VZV, CMV, and EBV, arguing that these infections are continually under T cell surveillance, as shown in the HSV-1-infected ganglia (216, 231). Antibody, however, is an important requirement to maintain control of persistent infection of mice with either Mγ HV or mouse hepatitis virus (233, 234). Because so many viruses have the capacity to form low-level persistent infections, a very important role for memory T and B cells may thus be to control these smoldering infections such that they do not recrudesce and again cause a disease.

CONCLUSIONS Although armed with highly distinct effector functions, there are many parallels between the T cell and B cell components of the immune response and their conversion into memory. As a consequence of high-affinity receptor triggering, both cell types up-regulate Bcl-2 family proteins to resist apoptosis and then differentiate and proliferate. Primary CD8 T or B cell responses can occur in the absence of CD4 T cell help, but that help is absolutely necessary for the development of longterm CD8 T and B cell responses. Memory B and T cell responses are long-lived in the absence of antigen, but they are modulated and influenced by other infections, and memory B and T cells can either provide protective immunity or contribute to immunopathology on infection with homologous or heterologous viruses.

ACKNOWLEDGMENTS This work was supported by NIH research grants AI17672, AI46629, AI46578, AI49320, AR35506, CA34461, CA66644, and DK32520. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Fenner F. 1996. Poxviruses. In Virology, ed. B Fields, DM Knipe, PM Howley, pp. 2673–2702. Philadelphia: LippincottRaven 2. Cerwenka A, Lanier LL. 2001. Natural killer cells, viruses and cancer. Nat. Rev. Immunol. 1:41–49

3. Biron CA, Sen G.C. 2001. Interferons and other cytokines. In Fundamental Virology, ed. DM Knipe, PM Howley, pp. 321–351. Philadelphia: Lippincott, Williams & Wilkins 4. Selin LK, Santolucito PA, Pinto AK, Szomolanyi-Tsuda E, Welsh RM. 2001.

19 Feb 2004

14:33

732

5.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

6.

7.

8.

9. 10.

11.

12.

13. 14.

15.

16.

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

Innate immunity to viruses: control of vaccinia virus infection by gamma delta T cells. J. Immunol. 166:6784–94 Szomolanyi-Tsuda E, Welsh RM. 1996. T cell-independent antibody-mediated clearance of polyoma virus in T celldeficient mice. J. Exp. Med. 183:403–11 Zinkernagel RM. 2002. On differences between immunity and immunological memory. Curr. Opin. Immunol. 14:523– 36 Welsh RM, McFarland HI. 1993. Mechanisms of viral pathogenesis. In Viruses and Bone Marrow, ed. NS Young, pp. 3– 30. New York: Marcel Dekker Reis e Sousa C, Sher A, Kaye P. 1999. The role of dendritic cells in the induction and regulation of immunity to microbial infection. Curr. Opin. Immunol. 11:392– 99 Matzinger P. 2002. The danger model: a renewed sense of self. Science 296:301–5 Compton T, Kurt-Jones EA, Boehme KW, Belko J, Latz E, et al. 2003. Human cytomegalovirus activates inflammatory cytokine responses via CD14 and Toll-like receptor 2. J. Virol. 77:4588–96 Haynes LM, Moore DD, Kurt-Jones EA, Finberg RW, Anderson LJ, Tripp RA. 2001. Involvement of toll-like receptor 4 in innate immunity to respiratory syncytial virus. J. Virol. 75:10730–37 Sallusto F, Mackay CR, Lanzavecchia A. 2000. The role of chemokine receptors in primary, effector, and memory immune responses. Annu. Rev. Immunol. 18:593– 620 Sprent J, Surh CD. 2002. T cell memory. Annu. Rev. Immunol. 20:551–79 Butcher EC, Picker LJ. 1996. Lymphocyte homing and homeostasis. Science 272:60–66 Selin LK, Nahill SR, Welsh RM. 1994. Cross-reactivities in memory cytotoxic T lymphocyte recognition of heterologous viruses. J. Exp. Med. 179:1933–43 Blattman JN, Antia R, Sourdive DJ, Wang X, Kaech SM, et al. 2002. Esti-

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

mating the precursor frequency of naive antigen-specific CD8 T cells. J. Exp. Med. 195:657–64 Harris NL, Ronchese F. 1999. The role of B7 costimulation in T-cell immunity. Immunol. Cell Biol. 77:304–11 Grewal IS, Flavell RA. 1998. CD40 and CD154 in cell-mediated immunity. Annu. Rev. Immunol. 16:111–36 Hendriks J, Gravestein LA, Tesselaar K, van Lier RA, Schumacher TN, Borst J. 2000. CD27 is required for generation and long-term maintenance of T cell immunity. Nat. Immunol. 1:433–40 Janssen EM, Lemmens EE, Wolfe T, Christen U, von Herrath MG, Schoenberger SP. 2003. CD4+ T cells are required for secondary expansion and memory in CD8+ T lymphocytes. Nature 421:852– 56 Shedlock DJ, Shen H. 2003. Requirement for CD4 T cell help in generating functional CD8 T cell memory. Science 300:337–39 Sun JC, Bevan MJ. 2003. Defective CD8 T cell memory following acute infection without CD4 T cell help. Science 300:339–42 Bourgeois C, Rocha B, Tanchot C. 2002. A role for CD40 expression on CD8+ T cells in the generation of CD8+ T cell memory. Science 297:2060–63 Lauvau G, Vijh S, Kong P, Horng T, Kerksiek K, et al. 2001. Priming of memory but not effector CD8 T cells by a killed bacterial vaccine. Science 294:1735–39 Whitton JL, Zhang J. 1995. Principles of cytotoxic T lymphocyte induction and recognition. Curr. Top. Microbiol. Immunol. 202:247–59 Potsch C, Vohringer D, Pircher H. 1999. Distinct migration patterns of naive and effector CD8 T cells in the spleen: correlation with CCR7 receptor expression and chemokine reactivity. Eur. J. Immunol. 29:3562–70 Xie H, Lim YC, Luscinskas FW, Lichtman AH. 1999. Acquisition of selectin

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

IMMUNITY TO VIRUSES

28.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

29.

30.

31.

32.

33.

34.

35.

36.

37.

binding and peripheral homing properties by CD4+ and CD8+ T cells. J. Exp. Med. 189:1765–76 Kaech SM, Ahmed R. 2001. Memory CD8+ T cell differentiation: initial antigen encounter triggers a developmental program in naive cells. Nat. Immunol. 2:415– 22 van Stipdonk MJ, Lemmens EE, Schoenberger SP. 2001. Naive CTLs require a single brief period of antigenic stimulation for clonal expansion and differentiation. Nat. Immunol. 2:423–29 Mercado R, Vijh S, Allen SE, Kerksiek K, Pilip IM, Pamer EG. 2000. Early programming of T cell populations responding to bacterial infection. J. Immunol. 165:6833–39 Welsh RM, Selin LK. 2002. No one is naive: the significance of heterologous Tcell immunity. Nat. Rev. Immunol. 2:417– 26 Foulds KE, Zenewicz LA, Shedlock DJ, Jiang J, Troy AE, Shen H. 2002. Cutting edge: CD4 and CD8 T cells are intrinsically different in their proliferative responses. J. Immunol. 168:1528–32 Razvi ES, Jiang Z, Woda BA, Welsh RM. 1995. Lymphocyte apoptosis during the silencing of the immune response to acute viral infections in normal, lpr and Bcl-2transgenic mice. Am. J. Pathol. 147:79– 91 Marshall DR, Turner SJ, Belz GT, Wingo S, Andreansky S, et al. 2001. Measuring the diaspora for virus-specific CD8+ T cells. Proc. Natl. Acad. Sci. USA 98: 6313–18 Masopust D, Vezys V, Marzo AL, Lefrancois L. 2001. Preferential localization of effector memory cells in nonlymphoid tissue. Science 291:2413–17 Reinhardt RL, Khoruts A, Merica R, Zell T, Jenkins MK. 2001. Visualizing the generation of memory CD4 T cells in the whole body. Nature 410:101–5 Chen HD, Fraire AE, Joris I, Brehm MA, Welsh RM, Selin LK. 2001. Memory

38.

39.

40.

41.

42.

43.

44.

45.

46.

P1: FDK

733

CD8+ T cells in heterologous antiviral immunity and immunopathology in the lung. Nat. Immunol. 2:1067–76 Wiley JA, Hogan RJ, Woodland DL, Harmsen AG. 2001. Antigen-specific CD8+ T cells persist in the upper respiratory tract following influenza virus infection. J. Immunol. 167:3293–99 Lenardo M, Chan KM, Hornung F, McFarland H, Siegel R, et al. 1999. Mature T lymphocyte apoptosis–immune regulation in a dynamic and unpredictable antigenic environment. Annu. Rev. Immunol. 17:221–53 Rathmell JC, Thompson CB. 2002. Pathways of apoptosis in lymphocyte development, homeostasis, and disease. Cell 109(Suppl.):S97–107 Gett AV, Sallusto F, Lanzavecchia A, Geginat J. 2003. T cell fitness determined by signal strength. Nat. Immunol. 4:355– 60 Zimmermann C, Rawiel M, Blaser C, Kaufmann M, Pircher H. 1996. Homeostatic regulation of CD8+ T cells after antigen challenge in the absence of Fas (CD95). Eur. J. Immunol. 26:2903–10 Lohman BL, Razvi ES, Welsh RM. 1996. T-lymphocyte downregulation after acute viral infection is not dependent on CD95 (Fas) receptor-ligand interactions. J. Virol. 70:8199–203 Badovinac VP, Tvinnereim AR, Harty JT. 2000. Regulation of antigen-specific CD8+ T cell homeostasis by perforin and interferon-gamma. Science 290:1354–58 Lohman BL, Welsh RM. 1998. Apoptotic regulation of T cells and absence of immune deficiency in virus-infected IFNgamma receptor knock-out mice. J. Virol. 72:7815–21 Petschner F, Zimmerman C, Strasser A, Grillot D, Nunez G, Pircher H. 1998. Constitutive expression of Bcl-xL or Bcl-2 prevents peptide antigen-induced T cell deletion but does not influence T cell homeostasis after a viral infection. Eur. J. Immunol. 28:560–69

19 Feb 2004

14:33

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

734

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

47. Homann D, Teyton L, Oldstone MB. 2001. Differential regulation of antiviral T-cell immunity results in stable CD8+ but declining CD4+ memory. Nat. Med. 7:892–93 48. Grayson JM, Zajac AJ, Altman JD, Ahmed R. 2000. Cutting edge: increased expression of Bcl-2 in antigen-specific memory CD8+ T cells. J. Immunol. 164:3950–54 49. Onami TM, Harrington LE, Williams MA, Galvan M, Larsen CP, et al. 2002. Dynamic regulation of T cell immunity by CD43. J. Immunol. 168:6022–31 50. Wang XZ, Stepp SE, Brehm MA, Chen HD, Selin LK, Welsh RM. 2003. Virusspecific CD8 T cells in peripheral tissues are more resistant to apoptosis than those in lymphoid organs. Immunity 18:631– 42 51. Murali-Krishna K, Altman JD, Suresh M, Sourdive D, Zajac AJ, et al. 1998. Counting antigen-specific CD8 T cells: a reevaluation of bystander activation during viral infection. Immunity 8:177–87 52. Selin LK, Vergilis K, Welsh RM, Nahill SR. 1996. Reduction of otherwise remarkably stable virus-specific cytotoxic T lymphocyte memory by heterologous viral infections. J. Exp. Med. 183:2489–99 53. Maryanski JL, Attuil V, Hamrouni A, Mutin M, Rossi M, et al. 2001. Individuality of Ag-selected and preimmune TCR repertoires. Immunol. Res. 23:75–84 54. Lin MY, Welsh RM. 1998. Stability and diversity of T cell receptor (TCR) repertoire usage during lymphocytic choriomeningitis virus infection of mice. J. Exp. Med. 188:1993–2005 55. Maryanski JL, Jongeneel CV, Bucher P, Casanova JL, Walker PR. 1996. Singlecell PCR analysis of TCR repertoires selected by antigen in vivo: a high magnitude CD8 response is composed of very few clones. Immunity 4:47–55 56. Blattman JN, Sourdive DJ, MuraliKrishna K, Ahmed R, Altman JD. 2000. Evolution of the T cell repertoire during

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

primary, memory, and recall responses to viral infection. J. Immunol. 165:6081–90 Turner SJ, Diaz G, Cross R, Doherty PC. 2003. Analysis of clonotype distribution and persistence for an influenza virusspecific CD8+ T cell response. Immunity 18:549–59 Kaech SM, Wherry EJ, Konieczny BT, Ahmed R. 2003. Selective expression of the interleukin 7 receptor identifies effector CD8 T cells that give rise to long-lived memory cells. Nat. Immunol. 4:1191–99 Reth M, Wienands J. 1997. Initiation and processing of signals from the B cell antigen receptor. Annu. Rev. Immunol. 15:453–79 Ho F, Lortan JE, MacLennan IC, Khan M. 1986. Distinct short-lived and long-lived antibody-producing cell populations. Eur. J. Immunol. 16:1297–301 Sze DM, Toellner KM, Garcia de Vinuesa C, Taylor DR, MacLennan IC. 2000. Intrinsic constraint on plasmablast growth and extrinsic limits of plasma cell survival. J. Exp. Med. 192:813–21 O’Connor BP, Gleeson MW, Noelle RJ, Erickson LD. 2003. The rise and fall of long-lived humoral immunity: terminal differentiation of plasma cells in health and disease. Immunol. Rev. 194:61–76 Forster R, Kremmer E, Mattis E, Kremmer E, Wolf E, et al. 1996. A putative chemokine receptor, BLR1, directs B cell migration to defined lymphoid organs and specific anatomic compartments of the spleen. Cell 87:1037–47 Gunn MD, Ngo VN, Ansel KM, Ekland EH, Cyster JG, Williams LT. 1998. A B-cell-homing chemokine made in lymphoid follicles activates Burkitt’s lymphoma receptor-1. Nature 391:799–803 Forster R, Schweigard G, Johann S, Emrich T, Kremmer E, et al. 1997. Abnormal expression of the B-cell homing chemokine receptor BLR1 during the progression of acquired immunodeficiency syndrome. Blood 90:520–25 Liu Y, deBoutellier O, Fujii H. 1997.

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

IMMUNITY TO VIRUSES

67.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

68.

69.

70.

71.

72.

73.

74.

75.

76.

Mechanisms of selection and differentiation in germinal centers. Curr. Opin. Immunol. 9:256–62 Han S, Zheng B, Takahashi Y, Kelsoe G. 1997. Distinctive characteristics of germinal center B cells. Semin. Immunol. 9:255–60 Muramatsu M, Kinoshita K, Fagarasan S, Yamada S, Shinkai Y, Honjo T. 2000. Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme. Cell 102:553–63 vanEijk M, Defrance T, Hennino A, deGroot C. 2001. Death-receptor contribution to the germinal center reaction. Trends Immunol. 22:677–82 Han S, Hathcock K, Zheng B, Kepler TB, Hodes R, Kelsoe G. 1995. Cellular interaction in germinal centers. Roles of CD40 ligand and B7-2 in established germinal centers. J. Immunol. 155:556–67 Li C, Iosef C, Jia CY, Han VK, Li SS. 2003. Dual functional roles for the X-linked lymphoproliferative syndrome gene product SAP/SH2D1A in signaling through the signaling lymphocyte activation molecule (SLAM) family of immune receptors. J. Biol. Chem. 278:3852–59 Crotty S, Kersh EN, Cannons J, Schwartzberg PL, Ahmed R. 2003. SAP is required for generating long-term humoral immunity. Nature 421:282–87 Henderson A, Calame K. 1998. Transcriptional regulation during B cell development. Annu. Rev. Immunol. 16:163–200 Lin KI, Tunyaplin C, Calame K. 2003. Transcriptional regulatory cascades controlling plasma cell differentiation. Immunol. Rev. 194:19–28 Bachmann MF, Odermatt B, Hengartner H, Zinkernagel RM. 1996. Induction of long-lived germinal centers associated with persisting antigen after viral infection. J. Exp. Med. 183:2259–69 Szomolanyi-Tsuda E, Le QP, Garcea RL, Welsh RM. 1998. T cell-independent IgG responses in vivo are elicited by live virus

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

P1: FDK

735

infection but not by immunization with viral proteins or virus-like particles. J. Virol. 72:6665–70 Slifka MK, Ahmed R. 1996. Long-term humoral immunity against viruses: revisiting the issue of plasma cell longevity. Trends Microbiol. 4:394–400 Freer G, Burkhart C, Ciernik I, Bachmann MF, Hengartner H, Zinkernagel RM. 1994. Vesicular stomatitis virus Indiana glycoprotein as a T-cell-dependent and -independent antigen. J. Virol. 68:3650– 55 Hotchin J. 1962. The biology of lymphocytic choriomeningitis infection: virusinduced immune disease. Cold Spring Harbor Symp. Quant. Biol. 27:479–99 Buchmeier MJ, Welsh RM, Dutko FJ, Oldstone MBA. 1980. The virology and immunobiology of lymphocytic choriomeningitis virus infection. Adv. Immunol. 30:275–331 Slifka MK, Antia R, Whitmire JK, Ahmed R. 1998. Humoral immunity due to longlived plasma cells. Immunity 8:363–72 Roost H-P, Bachmann MF, Haag A, Kalinke U, Pliska V, et al. 1995. Early high-affinity neutralizing anti-viral lgG responses without further improvements of affinity. Proc. Natl. Acad. Sci. USA 92:1257–61 Franco MA, Greenberg HB. 1999. Immunity to rotavirus infection in mice. J. Infect. Dis. 179(Suppl. 3):S466–69 Gerhard W, Mozdzanowska K, Furchner M, Washko G, Maiese K. 1997. Role of the B-cell response in recovery of mice from primary influenza virus infection. Immunol. Rev. 159:95–103 Tomori O. 1999. Impact of yellow fever on the developing world. Adv. Virus Res. 53:5–34 Zinkernagel RM, LaMarre A, Ciurea A, Hunziker L, Ochsenbein AF, et al. 2001. Neutralizing antiviral antibody responses. Adv. Immunol. 79:1–53 Planz O, Seiler P, Hengartner H, Zinkernagel RM. 1996. Specific cytotoxic T

19 Feb 2004

14:33

736

88.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

89.

90.

91.

92.

93.

94.

95.

96.

97.

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

cells eliminate B cells producing virusneutralizing antibodies. Nature 382:726– 29 Collins JT, Dunnick WA. 1993. Germline transcripts of the murine immunoglobulin gamma 2a gene: structure and induction by IFN-gamma. Int. Immunol. 5:885–91 Courtelier JP, Van der Logt JT, Hessen FW, Warnier G, Van Snick J. 1987. lgG2a restriction of murine antibodies elicited by viral infections. J. Exp. Med. 1988:2373– 78 Thomsen AR, Volkert M, Marker O. 1985. Different isotype profiles of virus-specific antibodies in acute and persistent lymphocytic choriomeningitis virus infection in mice. Immunology 55:213–23 Szomolanyi-Tsuda E, Welsh RM. 1998. T cell-independent anti-viral antibody responses. Curr. Opin. Immunol. 10:431–35 Bachmann MF, Rohrer UH, Kundig TM, Burki K, Hengartner H, Zinkernagel RM. 1993. The influence of antigen organization on B cell responsiveness. Science 262:1448–51 Borrow P, Tishon A, Lee S, Xu J, Grewal IS, et al. 1996. CD40L-deficient mice show deficits in antiviral immunity and have an impaired memory CD8+ CTL response. J. Exp. Med. 183:2129–42 Maloy KJ, Odermatt B, Hengartner H, Zinkernagel RM. 1998. Interferon γ producing γ δ T cell-dependent antibody isotype switching in the absence of germinal center formation during virus infection. Proc. Natl. Acad. Sci. USA 95:1160– 65 Franco MA, Greenberg HB. 1997. Immunity to rotavirus in T cell deficient mice. Virology 238:169–79 Litinskiy MB, Nardelli B, Hilbert DM, He B, Schaffer A, et al. 2002. DCs induce CD40-independent immunoglobulin class switching through BLyS and APRIL. Nat. Immunol. 3:822–29 Palladino G, Mozdzanowska K, Washko G, Gerhard W. 1995. Virus-neutralizing antibodies of immunoglobulin G (IgG)

98.

99.

100.

101.

102.

103.

104.

105.

106.

but not of IgM or IgA isotypes can cure influenza virus pneumonia in SCID mice. J. Virol. 69:2075–81 Baumgarth N, Herman OC, Jager GC, Brown LE, Herzenberg LA, Che J. 2000. B-1 and B-2 cell-derived immunoglobulin M antibodies are nonredundant components of the protective response to influenza virus infection. J. Exp. Med. 192:271–80 Clarke SH, Staudt LM, Kavaler J, Schwartz D, Gerhard WU, Weigert MG. 1990. V region gene usage and somatic mutation in the primary and secondary responses to influenza virus hemagglutinin. J. Immunol. 144:2795–801 Harada Y, Muramatsu M, Shibata T, Honjo T, Kuroda K. 2003. Unmutated immunoglobulin M can protect mice from death by influenza virus infection. J. Exp. Med. 197:1779–85 Swain SL, Croft MC, Dubey C, Haynes L, Rogers P, Zhang X, Bradley LM. 1996. From naive to memory T cells. Immunol. Rev. 150:143–67 Harrington LE, Galvan M, Baum LG, Altman JD, Ahmed R. 2000. Differentiating between memory effector CD8 T cells by altered expression of cell surface Oglycans. J Exp. Med. 191:1241–46 Woodland DL, Dutton RW. 2003. Heterogeneity of CD4+ CD8+ T cells. Curr. Opin. Immunol. 15:336–42 Peacock CD, Kim S-K, Welsh RM. 2003. Attrition of virus-specific memory CD8+ T cells during reconstitution of lymphopenic environments. J. Immunol. 171:655–63 Kieper WC, Jameson SC. 1999. Homeostatic expansion phenotypic conversion of naive T cells in response to self peptide/MHC ligands. Proc. Natl. Acad. Sci. USA 96:13306–11 Schluns KS, Kieper WC, Jameson SC, Lefrancois L. 2000. Interleukin-7 mediates the homeostasis of naive memory CD8 T cells in vivo. Nat. Immunol. 1:426– 32

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IMMUNITY TO VIRUSES 107. Tan JT, Dudl E, LeRoy E, Murray R, Sprent J, Weinberg KI, Surh CD. 2001. IL-7 is critical for homeostatic proliferation survival of naive T cells. Proc. Natl. Acad. Sci. USA 98:8732–37 108. Goldrath AW, Sivakumar PV, Glaccum M, Kennedy MK, Bevan MJ, et al. 2002. Cytokine requirements for acute Basal homeostatic proliferation of naive and memory CD8+ T cells. J. Exp. Med. 195: 1515–22 109. Tanchot C, Lemonnier FA, Perarnau B, Freitas AA, Rocha B. 1997. Differential requirements for survival and proliferation of CD8 naive or memory T cells. Science 276:2057–62 110. Kassiotis G, Zamoyska R, Stockinger B. 2003. Involvement of avidity for major histocompatibility complex in homeostasis of naive and memory T cells. J. Exp. Med. 197:1007–16 111. Razvi ES, Welsh RM, McFarland HI. 1995. In vivo state of antiviral CTL precursors: characterization of a cycling population containing CTL precursors in immune mice. J. Immunol. 154:620–32 112. Tripp RA, Hou S, Doherty PC. 1995. Temporal loss of the activated L-selectin-low phenotype for virus-specific CD8+ memory T cells. J. Immunol. 154:5870–75 113. Sallusto F, Langenkamp A, Geginat J, Lanzavecchia A. 2000. Functional subsets of memory T cells identified by CCR7 expression. Curr. Top. Microbiol. Immunol. 251:167–71 114. Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. 1999. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401:708–12 115. Wherry EJ, Teichgraber V, Becker TC, Masopust D, Kaech SM, et al. 2003. Lineage relationship and protective immunity of memory CD8 T cell subsets. Nat. Immunol. 4:225–34 116. Tomiyama H, Matsuda T, Takiguchi M. 2002. Differentiation of human CD8+ T cells from a memory to memory/

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

P1: FDK

737

effector phenotype. J. Immunol. 168: 5538–50 Appay V, Dunbar PR, Callan M, Klenerman P, Gillespie GM, et al. 2002. Memory CD8+ T cells vary in differentiation phenotype in different persistent virus infections. Nat. Med. 8:379–85 Roman E, Miller E, Harmsen A, Wiley J, Von Andrian UH, et al. 2002. CD4 effector T cell subsets in the response to influenza: heterogeneity, migration, and function. J. Exp. Med. 196:957–68 van der Most RG, Murali-Krishna K, Ahmed R. 2003. Prolonged presence of effector-memory C8 T cells in the central nervous system after dengue virus encephalitis. Int. Immunol. 1:119–25 Slifka MK, Whitton JL. 2000. Activated and memory CD8+ T cells can be distinguished by their cytokine profiles and phenotypic markers. J. Immunol. 164:208–16 Slifka MK, Whitton JL. 2001. Functional avidity maturation of CD8+ T cells without selection of higher affinity TCR. Nat. Immunol. 2:711–17 Kersh EN, Kaech SM, Onami TM, Moran M, Wherry EJ, et al. 2003. TCR signal transduction in antigen-specific memory CD8 T cells. J. Immunol. 170:5455–63 Drake DR, III, Braciale TJ. 2001. Cutting edge: lipid raft integrity affects the efficiency of MHC class I tetramer binding and cell surface TCR arrangement on CD8+ T cells. J. Immunol. 166:7009–13 Murphy KM, Reiner SL. 2002. The lineage decisions of helper T cells. Nat. Rev. Immunol. 2:933–44 Kaech SM, Hemby S, Kersh E, Ahmed R. 2002. Molecular and functional profiling of memory CD8 T cell differentiation. Cell 111:837–51 Slifka MK, Rodriguez F, Whitton JL. 1999. Rapid on/off cycling of cytokine production by virus-specific CD8+ T cells. Nature 401:76–79 Walzer T, Marcais A, Saltel F, Bella C, Jurdic P, Marvel J. 2003. Cutting edge: immediate RANTES secretion by resting

19 Feb 2004

14:33

738

128.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

129.

130.

131.

132.

133.

134.

135.

136.

137.

138.

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

memory CD8 T cells following antigenic stimulation. J. Immunol. 170:1615–19 Selin LK, Welsh RM. 1997. Cytolytically active memory CTL present in lymphocytic choriomeningitis virus (LCMV)immune mice after clearance of virus infection. J. Immunol. 158:5366–73 Zimmermann C, Brduscha-Riem K, Blaser C, Zinkernagel RM, Pircher H. 1996. Visualization, characterization, and turnover of CD8+ memory T cells in virusinfected hosts. J. Exp. Med. 183:1367–75 Barber DL, Wherry EJ, Ahmed R. 2003. Cutting edge: rapid in vivo killing by memory CD8 T cells. J. Immunol. 171:27–31 Demkowicz WE, Littaua RA, Wang J, Ennis FA. 1996. Human cytotoxic T-cell memory: long-lived responses to vaccinia virus. J. Virol. 70:2627–31 Van Epps HL, Terajima M, Mustonen J, Arstila TP, Corey EA, et al. 2002. Longlived memory T lymphocyte responses after hantavirus infection. J. Exp. Med. 196:579–88 Sierra B, Garcia G, Perez AB, Morier L, Rodriguez R, et al. 2002. Long-term memory cellular immune response to dengue virus after a natural primary infection. Int. J. Infect. Dis. 6:125–28 Lau LL, Jamieson BD, Somasundaram T, Ahmed R. 1994. Cytotoxic T-cell memory without antigen. Nature 369:648–52 Murali-Krishna K, Lau LL, Sambhara S, Lemonnier F, Altman J, Ahmed R. 1999. Persistence of memory CD8 T cells in MHC class I-deficient mice. Science 286:1377–81 Swain SL, Hu H, Huston G. 1999. Class II-independent generation of CD4 memory T cells from effectors. Science 286:1381–83 Ciurea A, Klenerman P, Hunziker L, Horvath E, Odermatt B, et al. 1999. Persistence of lymphocytic choriomeningitis virus at very low levels in immune mice. Proc. Natl. Acad. Sci. USA 96:11964–69 Prlic M, Lefrancois L, Jameson SC. 2002.

139.

140.

141.

142.

143.

144.

145.

146.

Multiple choices: regulation of memory CD8 T cell generation and homeostasis by interleukin (IL)-7 and IL-15. J. Exp. Med. 195:F49–52 Tan JT, Ernst B, Kieper WC, LeRoy E, Sprent J, Surh CD. 2002. Interleukin (IL)-15 and IL-7 jointly regulate homeostatic proliferation of memory phenotype CD8+ cells but are not required for memory phenotype CD4+ cells. J. Exp. Med. 195:1523–32 Kieper WC, Tan JT, Bondi-Boyd B, Gapin L, Sprent J, et al. 2002. Overexpression of interleukin (IL)-7 leads to IL15-independent generation of memory phenotype CD8+ T cells. J. Exp. Med. 195:1533–39 Schluns KS, Williams K, Ma A, Zheng XX, Lefrancois L. 2002. Cutting edge: requirement for IL-15 in the generation of primary and memory antigen-specific CD8 T cells. J. Immunol. 168:4827–31 Becker TC, Wherry EJ, Boone D, MuraliKrishna K, Antia R, et al. 2002. Interleukin 15 is required for proliferative renewal of virus-specific memory CD8 T cells. J. Exp. Med. 195:1541–48 Burkett PR, Koka R, Chien M, Chai S, Chan F, et al. 2003. IL-15R alpha expression on CD8+ T cells is dispensable for T cell memory. Proc. Natl. Acad. Sci. USA 100:4724–29 Leishman AJ, Naidenko OV, Attinger A, Koning F, Lena CJ, et al. 2001. T cell responses modulated through interaction between CD8αα and the nonclassical MHC class I molecule, TL. Science 294:1936–39 Liu Y, Xiong Y, Naidenko OV, Liu JH, Zhang R, et al. 2003. The crystal structure ˚ resoluof a TL/CD8αα complex at 2.1 A tion: implications for modulation of T cell activation and memory. Immunity 18:205– 15 Varga SM, Welsh RM. 1998. Stability of virus-specific CD4+ T cell frequencies from acute infection into long term memory. J. Immunol. 161:367–74

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IMMUNITY TO VIRUSES 147. Varga SM, Welsh RM. 1998. Cutting edge: detection of a high frequency of virus-specific CD4+ T cells during acute infection with lymphocytic choriomeningitis virus. J. Immunol. 161:3215–18 148. Topham DJ, Doherty PC. 1998. Longitudinal analysis of the acute Sendai virusspecific CD4 T cell response and memory. J. Immunol. 161:4530–35 149. Karrer U, Sierro S, Wagner M, Oxenius A, Hengel H, et al. 2003. Memory inflation: continuous accumulation of antiviral CD8+ T cells over time. J. Immunol. 170:2022–29 150. Vieira P, Rajewsky K. 1990. Persistence of memory B cells in mice deprived of T cell help. Int. Immunol. 2:487–94 151. Veiga-Fernandes H, Walter U, Bourgeois C, McLean A, Rocha B. 2000. Response of naive and memory CD8 T cells to antigen stimulation in vivo. Nat. Immunol. 1:47–53 152. Kim SK, Brehm MA, Welsh RM, Selin LK. 2002. Dynamics of memory T cell proliferation under conditions of heterologous immunity and bystander stimulation. J. Immunol. 169:90–98 153. Topham DJ, Castrucci M, Wingo FS, Belz GT, Doherty PC. 2001. The role of antigen in the localization of naive, acutely activated, and memory CD8+ T cells to the lung during influenza pneumonia. J. Immunol. 167:6983–90 154. Ely KH, Cauley LS, Roberts AD, Brennan JW, Cookenham T, Woodland DL. 2003. Nonspecific recruitment of memory CD8+ T cells to the lung airways during respiratory virus infections. J. Immunol. 170:1423–29 155. Robinson HL, Pertmer TM. 2000. DNA vaccines for viral infections: basic studies and applications. Adv. Virus Res. 55:1– 74 156. An LL, Whitton JL. 1997. A multivalent minigene vaccine, containing B-cell, cytotoxic T-lymphocyte, and Th epitopes from several microbes, induces appropriate responses in vivo and confers protec-

157.

158.

159.

160.

161.

162.

163.

164.

165.

P1: FDK

739

tion against more than one pathogen. J. Virol. 71:2292–302 Oxenius A, Martinic MM, Hengartner H, Klenerman P. 1999. CpG-containing oligonucleotides are efficient adjuvants for induction of protective antiviral immune responses with T-cell peptide vaccines. J. Virol. 73:4120–26 Ciupitu A-MT, Petersson M, O’Donnell CL, Williams K, Jindal S, et al. 1998. Immunization with a lymphocytic choriomeningitis virus peptide mixed with heat shock protein 70 results in protective antiviral immunity and specific cytotoxic T lymphocytes. J. Exp. Med. 187:685– 91 Hogan RJ, Usherwood EJ, Zhong W, Roberts AA, Dutton RW, et al. 2001. Activated antigen-specific CD8+ T cells persist in the lungs following recovery from respiratory virus infections. J. Immunol. 166:1813–22 Hogan RJ, Zhong W, Usherwood EJ, Cookenham T, Roberts AD, Woodland DL. 2001. Protection from respiratory virus infections can be mediated by antigen-specific CD4+ T cells that persist in the lungs. J. Exp. Med. 193:981–86 Turner SJ, Cross R, Xie W, Doherty PC. 2001. Concurrent naive and memory CD8+ T cell response to a influenza virus. J. Immunol. 167:2753–58 Badovinac VP, Messingham KA, Hamilton SE, Harty JT. 2003. Regulation of CD8+ T cells undergoing primary and secondary responses to infection in the same host. J. Immunol. 170:4933–42 Crowe S, Turner SJ, Miller SC, Roberts AD, Rappolo RA, et al. 2003. Differential antigen presentation regulates the changing patterns of CD8+ T cell immunodominance in primary and secondary influenza virus infections. J. Exp. Med. 198:399– 410 Vieira P, Rajewsky K. 1988. The half-lives of serum immunoglobulins in adult mice. Eur. J. Immunol 18:313–16 Manz RA, Lohning M, Cassese G, Thiel

19 Feb 2004

14:33

740

166.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

167.

168.

169.

170.

171.

172.

173.

174.

175.

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

A, Radbruch A. 1998. Survival of longlived plasma cells is independent of antigen. Int. Immunol. 10:1703–11 Hargreaves DC, Hyman PL, Lu TT, Ngo VN, Bidgol A, et al. 2001. A coordinated change in chemokine responsiveness guides plasma cell movements. J. Exp. Med. 194:45–56 Slifka MK, Ahmed R. 1996. Limiting dilution analysis of virus-specific memory B cells by an ELISPOT assay. J. Immunol. Methods 199:37–46 Fazekas de St. Groth S, Webster RG. 1966. Disquisitions on original antigenic sin. II. Proof in lower creatures. J. Exp. Med. 124:347–61 Halstead SB. 1989. Antibody, macrophages, dengue virus infection, shock and hemorrhage: a pathogenetic cascade. Rev. Infect. Dis. 11:S830–39 Tew JG, Mandel TE. 1979. Prolonged antigen half-life in the lymphoid follicles of specifically immunized mice. Immunology 37:69–76 Bernasconi NL, Traggiai E, Lanzavecchia A. 2002. Maintenance of serological memory by polyclonal activation of human memory B cells. Science 298:2199– 202 Mims CA. 1983. Immunopathology in virus disease. Philos. Trans. R. Soc. London Ser. B 303:189–98 Binder D, Fehr J, Hengartner H, Zinkernagel RM. 1997. Virus-induced transient bone marrow aplasia: major role of interferon-alpha/beta during acute infection with the noncytopathic lymphocytic choriomeningitis virus. J. Exp. Med. 185:517–30 McNally JM, Zarozinski CC, Lin MY, Brehm MA, Chen HD, Welsh RM. 2001. Attrition of bystander CD8 T cells during virus-induced T cell and interferon responses. J. Virol. 75:5965–76 Pfizenmaier K, Jung H, Starzinski-Powitz A, Rollinghoff M, Wagner H. 1977. The role of T cells in anti-herpes simplex virus immunity. I. Induction of antigen-

176.

177.

178.

179.

180.

181.

182.

183.

184.

specific cytotoxic T lymphocytes. J. Immunol. 119:939–44 Dummer W, Niethammer AG, Baccala R, Lawson BR, Wagner N, et al. 2002. T cell homeostatic proliferation elicits effective antitumor autoimmunity. J. Clin. Invest. 110:185–92 Oehen S, Brduscha-Riem K. 1999. Naive cytotoxic T lymphocytes spontaneously acquire effector function in lymphocytopenic recipients: a pitfall for T cell memory studies? Eur. J. Immunol. 29:608–14 Selin LK, Lin MY, Kraemer KA, Schneck JP, Pardoll D, et al. 1999. Attrition of T cell memory: selective loss of lymphocytic choriomeningitis virus (LCMV) epitopespecific memory CD8 T cells following infections with heterologous viruses. Immunity 11:733–42 Liu H, Andreansky S, Diaz G, Turner SJ, Wodarz D, Doherty PC. 2003. Quantitative analysis of long-term virus-specific CD8+-T-cell memory in mice challenged with unrelated pathogens. J. Virol. 77:7756–63 Brehm MA, Pinto AK, Daniels KA, Schneck JP, Welsh RM, Selin LK. 2002. T cell immunodominance and maintenance of memory regulated by unexpectedly crossreactive pathogens. Nat. Immunol. 3:627– 34 Oehen S, Waldner H, Kundig TM, Hengarter H, Zinkernagel RM. 1992. Antivirally protective cytotoxic T cell memory to lymphocytic choriomeningitis virus is governed by persisting antigen. J. Exp. Med. 176:1273–81 Tough DL, Borrow P, Sprent J. 1996. Induction of bystander T cell proliferation by viruses and type I interferon. Science 272:1947–50 Flynn KJ, Riberdy JM, Christensen JP, Altman JD, Doherty PC. 1999. In vivo proliferation of naive and memory influenza-specific CD8+ T cells. Proc. Natl. Acad. Sci. USA 96:8597–602 Belz GT, Doherty PC. 2001. Virusspecific and bystander CD8+ T cell

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

IMMUNITY TO VIRUSES

185.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

186.

187.

188.

189.

190.

191.

192.

proliferation in the persistent phases of a gammaherpesvirus infection. J. Virol. 75:4435–38 Varga SM, Selin LK, Welsh RM. 2001. Independent regulation of lymphocytic choriomeningitis virus-specific T cell memory pools: relative stability of CD4 memory under conditions of CD8 memory T cell loss. J. Immunol. 166:1554–61 Smith DK, Dudani R, Pedras-Vasconcelos JA, Chapdelaine Y, van Faassen H, Sad S. 2002. Cross-reactive antigen is required to prevent erosion of established T cell memory and tumor immunity: a heterologous bacterial model of attrition. J. Immunol. 169:1197–206 Mason D. 1998. A very high level of crossreactivity is an essential feature of the T cell repertoire. Immunol. Today 19:395– 404 Anderson RW, Bennick JR, Yewdell JW, Maloy WL, Coligan JE. 1992. Influenza basic polymerase 2 peptides are recognized by influenza nucleoprotein-specific cytotoxic T lymphocytes. Mol. Immunol. 29:1089–96 Kulkarni AB, Morse HC, III, Bennink JR, Yewdell JW, Murphy BR. 1993. Immunization of mice with vaccinia virus-M2 recombinant induces epitope-specific and cross-reactive Kd-restricted CD8+ cytotoxic T cells. J. Virol. 67:4086–92 Kuwano K, Reyes RE, Humphreys RE, Ennis FA. 1991. Recognition of disparate HA and NS1 peptides by an H-2kdrestricted, influenza specific CTL clone. Mol. Immunol. 28:1–7 Haanen JBAG, Wolkers MC, Kruisbeek AM, Schumacher TNM. 1999. Selective expansion of cross-reactive CD8+ memory T cells by viral variants. J. Exp. Med. 190:1319–28 Spaulding AC, Kurane I, Ennis FA, Rothman AL. 1999. Analysis of murine CD8+ T-cell clones specific for the dengue virus NS3 protein: flavivirus cross-reactivity and influence of infecting serotype. J. Virol. 73:398–403

P1: FDK

741

193. Zivny J, DeFronzo M, Jarry W, Jameson J, Cruz J, et al. 1999. Partial agonist effect influences the CTL response to a heterologous dengue virus serotype. J. Immunol. 163:2754–60 194. Van Epps HL, Schmaljohn CS, Ennis FA. 1999. Human memory cytotoxic Tlymphocyte (CTL) responses to Hantaan virus infection: identification of virusspecific and cross-reactive CD8+ CTL epitopes on nucleocapsid protein. J. Virol. 73:5301–8 195. Nilges K, Hohn H, Pilch H, Neukirch C, Freitag K, et al. 2003. Human papillomavirus type 16 E7 peptide-directed CD8+ T cells from patients with cervical cancer are cross-reactive with the coronavirus NS2 protein. J. Virol. 77:5464– 74 196. Shimojo N, Maloy WL, Anderson RW, Biddison WE, Coligan JE. 1989. Specificity of peptide binding by the HLA-A2.1 molecule. J. Immunol. 143:2939–47 197. Wedemeyer H, Mizukoshi E, Davis AR, Bennink JR, Rehermann B. 2001. Crossreactivity between hepatitis C virus and influenza A virus determinant-specific cytotoxic T cells. J. Virol. 75:11392–400 198. Hohn H, Kortsik C, Tully G, Nilges K, Necker A, et al. 2003. Longitudinal analysis of Mycobacterium tuberculosis 19kDa antigen-specific T cells in patients with pulmonary tuberculosis: association with disease activity and cross-reactivity to a peptide from HIVenv gp120. Eur. J. Immunol. 33:1613–23 199. Selin LK, Varga SM, Wong IC, Welsh RM. 1998. Protective heterologous antiviral immunity and enhanced immunopathogenesis mediated by memory T cell populations. J. Exp. Med. 188: 1705–15 200. Di Giusto CA, Bernhard JD. 1986. Erythema nodosum provoked by hepatitis B vaccine. Lancet 2:1042 201. Schlesinger C, Meyer CA, Veeraraghavan S, Koss MN. 1998. Constrictive (obliterative) bronchiolitis: diagnosis, etiology, a

19 Feb 2004

14:33

742

202.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

203.

204.

205.

206.

207.

208.

209.

210.

AR

WELSH

AR210-IY22-24.tex

¥

SELIN

¥

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

P1: FDK

SZOMOLANYI-TSUDA

critical review of the literature. Ann. Diagn. Pathol. 2:321–34 Chen HD, Fraire AE, Joris I, Welsh RM, Selin LK. 2003. Specific history of heterologous virus infections determines antiviral immunity and immunopathology in the lung. Am. J. Pathol. 163:1341–55 Ostler T, Pircher H, Ehl S. 2003. “Bystander” recruitment of systemic memory T cells delays the immune response to respiratory virus infection. Eur. J. Immunol. 33:1839–48 Kapikian AZ, Mitchell RH, Chanock RM, Shvedoff RA, Stewart CE. 1969. An epidemiological study of altered clinical reactivity to respiratory syncitial (RS) virus infection in children previously vaccinated with an inactivated RS virus vaccine. Am. J. Epidemiol. 89:405–21 Graham BS, Bunton LA, Wright PF, Karzon DT. 1991. Role of T lymphocyte subsets in the pathogenesis of primary infection rechallenge with respiratory syncytial virus in mice. J. Clin. Invest. 88:1026–33 Walzl G, Tafuro S, Moss P, Openshaw PJ, Hussell T. 2000. Influenza virus lung infection protects from respiratory syncitial virus-induced immunopathology. J. Exp. Med. 192:1317–26 Varga SM, Wang X, Welsh RM, Braciale TJ. 2001. Immunopathology in RSV infection is mediated by a discrete oligoclonal subset of antigen-specific CD4+ T cells. Immunity 15:637–46 Yewdell JW, Bennink JR. 1999. Immunodominance in major histocompatibility complex class I-restricted T lymphocyte responses. Annu. Rev. Immunol. 17:51–88 Klenerman P, Zinkernagel RM. 1998. Original antigenic sin impairs cytotoxic T lymphocyte responses to viruses bearing variant epitopes. Nature 394:421–22 Libraty DH, Young PR, Pickering D, Endy TP, Kalayanarooj S, et al. 2002. High circulating levels of the dengue virus nonstructural protein NS1 early in dengue illness correlate with the development of

211.

212.

213.

214.

215.

216.

217.

218.

219.

dengue hemorrhagic fever. J. Infect. Dis. 186:1165–68 Mongkolsapaya J, Dejnirattisai W, Xu XN, Vasanawathana S, Tangthawornchaikul N, et al. 2003. Original antigenic sin and apoptosis in the pathogenesis of dengue hemorrhagic fever. Nat. Med. 9:921–27 Moss DJ, Burrows SR, Silins SL, Misko I, Khanna R. 2001. The immunology of Epstein-Barr virus infection. Philos. Trans. R. Soc. London Ser. B 356:475–88 Nahill SR, Welsh RM. 1993. High frequency of cross-reactive cytotoxic T lymphocytes elicited during the virus-induced polyclonal cytotoxic T lymphocyte response. J. Exp. Med. 177:317–27 Burrows SR, Khanna R, Silins SL, Moss DJ. 1999. The influence of antiviral T-cell responses on the alloreactive repertoire. Immunol. Today 20:203–7 Brehm MA, Markees TG, Daniels KA, Greiner DL, Rossini AA, Welsh RM. 2003. Direct visualization of crossreactive effector and memory allo-specific CD8 T cells generated in response to viral infections. J. Immunol. 170:4077–86 Gaston JSH, Waer M. 1985. Virus-specific MHC-restricted T lymphocytes may initiate allograft rejection. Immunol. Today 6:237–39 Adams AB, Williams MA, Jones TR, Shirasugi N, Durham MM, et al. 2003. Heterologous immunity provides a potent barrier to transplantation tolerance. J. Clin. Invest 111:1887–95 Welsh RM, Markees TG, Woda BA, Daniels KA, Brehm MA, et al. 2000. Virus-induced abrogation of transplantation tolerance induced by donor-specific transfusion and anti-CD154 antibody. J. Virol. 74:2210–18 Moskophidis D, Lechner F, Pircher H, Zinkernagel RM. 1993. Virus persistence in acutely infected immunocompetent mice by exhaustion of antiviral cytotoxic effector T cells. Nature 362: 758–61

19 Feb 2004

14:33

AR

AR210-IY22-24.tex

AR210-IY22-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IMMUNITY TO VIRUSES 220. Zajac AJ, Blattman JN, Murali-Krishna K, Sourdive DJD, Suresh M, et al. 1998. Viral immune evasion due to persistence of activated T cells without effector function. J. Exp. Med. 188:2205–13 221. Fuller MJ, Zajac AJ. 2003. Ablation of CD8 and CD4 T cell responses by high viral loads. J. Immunol. 170:477– 86 222. Wherry EJ, Blattman JN, Murali-Krishna K, van der MR, Ahmed R. 2003. Viral persistence alters CD8 T-cell immunodominance and tissue distribution and results in distinct stages of functional impairment. J. Virol. 77:4911–27 223. Appay V, Nixon DF, Donohoe SM, Gillespie GMA, Dong T, et al. 2000. HIVspecific CD8+ T cells produce antiviral cytokines but are impaired in cytolytic function. J. Exp. Med. 192:63–75 224. Moser JM, Altman JD, Lukacher AE. 2001. Antiviral CD8+ T cell responses in neonatal mice: susceptibility to polyoma virus-induced tumors is associated with lack of cytotoxic function. J. Exp. Med. 193:595–606 225. Xiong Y, Luscher MA, Altman JD, Hulsey M, Robinson HL, et al. 2001. Simian immunodeficiency virus (SIV) infection of rhesus macaque induces SIV-specific CD8+ T cells with a defect in effector function that is reversible on extended IL2 incubation. J. Virol. 75:3028–33 226. Gray CM, Lawrence J, Schapiro JM, Altman JD, Winters MA, et al. 1999. Frequency of class I HLA-restricted antiHIV CD8+ T cells in individuals receiving highly active antiretroviral therapy (HAART). J. Immunol. 162:1780– 88 227. Moser JM, Gibbs J, Jensen PE, Lukacher AE. 2002. CD94-NKG2A receptors regu-

228.

229.

230.

231.

232.

233.

234.

235.

P1: FDK

743

late antiviral CD8+ T cell responses. Nat. Immunol. 3:189–95 Champagne P, Ogg GS, King AS, Knabenhans C, Ellefsen K, et al. 2001. Skewed maturation of memory HIVspecific CD8 T lymphocytes. Nature 410:106–11 Catalina MD, Sullivan JL, Brody RM, Luzuriaga K. 2002. Phenotypic and functional heterogeneity of EBV epitopespecific CD8+ T cells. J. Immunol. 168:4184–91 Hislop AD, Annels NE, Gudgeon NH, Leese AM, Rickinson AB. 2002. Epitopespecific evolution of human CD8+ T cell responses from primary to persistent phases of Epstein-Barr virus infection. J. Exp. Med. 195:893–905 Khanna KM, Bonneau RH, Kinchington PR, Hendricks RL. 2003. Herpes simplex virus-specific memory CD8+ T cells are selectively activated and retained in latently infected sensory ganglia. Immunity 18:593–603 Sester M, Sester U, Gartner B, Kubuschok B, Girndt M, et al. 2002. Sustained high frequencies of specific CD4 T cells restricted to a single persistent virus. J. Virol. 76:3748–55 Virgin HW, Speck SH. 1999. Unraveling immunity to gamma-herpesviruses: a new model for understanding the role of immunity in chronic virus infection. Curr. Opin. Immunol. 11:371–79 Ramakrishna C, Stohlman SA, Atkinson RD, Shlomchik MJ, Bergmann CC. 2002. Mechanisms of central nervous system viral persistence: the critical role of antibody and B cells. J. Immunol. 168:1204– 11 Plotkin S, Orestein W. 1999. Vaccines. Philadelphia: Saunders

3/12/2004 2:15 PM

Figure 1 Dynamics of the CD8 T cell response during sequential heterologous virus infections. This figure demonstrates the kinetics of virus growth, IFN synthesis, virus-induced lymphopenia, T cell expansion and then apoptotic decline, and stability of CD8 memory in mice infected with LCMV. It shows the enhancement of T cells specific to cross-reactive (cxr) and the attrition against non-cxr epitopes after challenge with a heterologous virus, such as PV.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Welsh.qxd Page 1

IMMUNITY TO VIRUSES C-1

C-2

3/12/2004

WELSH

2:15 PM

■

SELIN ■

Figure 2 Generation and maintenance of antiviral antibodies. 1. Naïve B cells (B) in the extrafollicular T cell zone of the secondary lymphoid organs differentiate into short-lived PC in response to viral antigens (V). The short-lived PC produce IgM (light blue) or isotype-switched IgG or IgA (purple) that have unmutated germ-line sequences. 2. B cells activated by the virus and CD4 T helper (Th) cells undergo the germinal center (GC) reaction and differentiate into long-lived PC that secrete IgG or IgA. 3. B cells activated by the virus and Th undergo the GC reaction and become nonsecreting memory B cells (BM). The BM in response to antigen and Th become PC or may return to the GC to continue the GC reaction before terminal differentiation into PC.

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Welsh.qxd Page 2

SZOMOLANYI-TSUDA

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:711-743. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

12 Feb 2004

16:10

AR

AR210-IY22-25.tex

AR210-IY22-25.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104702

Annu. Rev. Immunol. 2004. 22:745–63 doi: 10.1146/annurev.immunol.22.012703.104702 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on December 12, 2003

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: Function, Generation, and Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Maintenance Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia Institute for Research in Biomedicine, CH-6500 Bellinzona, Switzerland; email: [email protected]; [email protected]; [email protected]

Key Words TH1/TH2, T lymphocyte activation, chemokine receptors, lymphocyte migration, effector function ■ Abstract The memory T cell pool functions as a dynamic repository of antigenexperienced T lymphocytes that accumulate over the lifetime of the individual. Recent studies indicate that memory T lymphocytes contain distinct populations of central memory (TCM) and effector memory (TEM) cells characterized by distinct homing capacity and effector function. This review addresses the heterogeneity of TCM and TEM, their differentiation stages, and the current models for their generation and maintenance in humans and mice.

THE CELLULAR BASIS OF IMMUNOLOGICAL MEMORY Memory is the hallmark of the acquired immune system. It results from the clonal expansion and differentiation of antigen-specific lymphocytes that ultimately persist for a lifetime. Memory lymphocytes confer immediate protection in peripheral tissues and mount recall responses to antigens in secondary lymphoid organs. In the B cell system these functions are carried out by distinct cell types. Protective memory is mediated by plasma cells that secrete antibodies, whereas reactive memory is mediated by memory B cells that proliferate and differentiate to plasma cells in response to secondary antigenic stimulation (1–3). A similar division of labor has recently emerged in the T cell system (4–6). According to the model proposed (7), protective memory is mediated by effector memory T cells (TEM) that migrate to inflamed peripheral tissues and display immediate effector function, whereas reactive memory is mediated by central memory T cells (TCM) that home to T cell areas of secondary lymphoid organs, have little or no effector function, but readily proliferate and differentiate to effector cells in response to antigenic stimulation. In this review we first describe the properties of TCM and TEM in terms of their heterogeneity, effector functions, and responsiveness to antigen or cytokines. 0732-0582/04/0423-0745$14.00

745

12 Feb 2004

16:10

746

AR

AR210-IY22-25.tex

SALLUSTO

¥

GEGINAT

¥

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

P1: IKH

LANZAVECCHIA

We then discuss a pathway of progressive T cell differentiation along which TCM are arrested at an intermediate stage preceding that of TEM. Finally, we consider current models for generation and maintenance of memory T cell subsets.

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PHENOTYPIC AND FUNCTIONAL PROPERTIES OF TCM AND TEM Definition of Human TCM and TEM TCM and TEM were initially defined in the human system based on two distinct criteria: (a) the absence or presence of immediate effector function and (b) the expression of homing receptors that allow cells to migrate to secondary lymphoid organs versus nonlymphoid tissues (4). Human TCM are CD45R0+ memory cells that constitutively express CCR7 and CD62L, two receptors that are also characteristic of na¨ıve T cells, which are required for cell extravasation through high endothelial venules (HEV) and migration to T cell areas of secondary lymphoid organs (8, 9). When compared with na¨ıve T cells, TCM have higher sensitivity to antigenic stimulation, are less dependent on costimulation, and upregulate CD40L to a greater extent, thus providing more effective stimulatory feedback to dendritic cells (DC) and B cells. Following TCR triggering, TCM produce mainly IL-2, but after proliferation they efficiently differentiate to effector cells and produce large amounts of IFN-γ or IL-4. Human TEM are memory cells that have lost the constitutive expression of CCR7, are heterogeneous for CD62L expression, and display characteristic sets of chemokine receptors and adhesion molecules that are required for homing to inflamed tissues. When compared with TCM, TEM are characterized by rapid effector function. CD8 TEM carry large amounts of perforin, and both CD4 and CD8 produce IFN-γ , IL-4, and IL-5 within hours following antigenic stimulation. Some CD8 TEM express CD45RA (here defined as TEMRA) and carry the largest amount of perforin. Thus in humans, the TEM pool contains bona fide TH1, TH2, and CTL. The relative proportions of TCM and TEM in blood vary in the CD4 and CD8 compartments; TCM is predominant in CD4 and TEM in CD8. Within the tissues, however, TCM and TEM show characteristic patterns of distributions. TCM are enriched in lymph nodes and tonsils, whereas lung, liver, and gut contain greater proportions of TEM (10). In antigen-primed individuals, tetanus toxoid-specific CD4 T cells can be detected in circulating TCM and TEM up to 10 years after antigenic stimulation, and their frequencies increase in both subsets following booster immunization (4). The same is true for the CD8 compartment: Antigen-specific T cells, detected by HLA/peptide multimer staining, can be found in TCM and TEM subsets, although the relative proportion can be highly variable. For example, HIVspecific T cells are largely TEM (CD45RA–CCR7−), whereas CMV-specific T cells are predominantly TEMRA (CCR7–CD45RA+) (11). In some melanoma patients, melan-A tetramer+ T cells are mainly CCR7+CD45RA+ and do not respond to

12 Feb 2004

16:10

AR

AR210-IY22-25.tex

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MEMORY T CELL SUBSETS

P1: IKH

747

antigenic stimulation ex vivo, whereas in other patients melan-A tetramer+ cells have a TEM or TEMRA phenotype (CCR7− CD45R0+ or CCR7−CD45RA+) and responses to melan-A peptide can be readily demonstrated (12, 13). Interestingly, a longitudinal study of EBV-infected patients revealed differences in lytic versus latent epitope-specific composition of T cell populations in the chronic carrier stage of infection (14). Thus while tetramer-positive cells in the infectious stage have a CCR7−CD45RA− phenotype, cells specific for lytic but not latent epitopes acquire CD45RA in the chronic carrier stage. A detailed TCR repertoire analysis was performed by spectra-typing in CD45R0+ memory CD8 T cells using CD62L to discriminate between TCM and TEM (15). The analysis of six influenza-specific T cell clones showed that two clonotypes were shared between TCM and TEM, whereas four were detected only in TCM. A similar picture was obtained after a nine-month period. These results indicate that the same expanded clone can be present in both TCM and TEM subsets and that, within the influenza-specific CD8 memory T cell pool, the clonotype distribution is remarkably stable with no evidence of conversion from CD62L+ and CD62L− memory subsets or vice versa. The presence of memory T cells with different migratory capacity and effector function was also documented in mice. Two populations of memory CD4 T cells survive for months after immunization with antigen in adjuvant: One, found primarily in the lymph nodes, produces IL-2; the other larger population found in nonlymphoid tissues produces IFN-γ (5). Similarly, two populations of antigenspecific memory CD8 T cells are present following viral or bacterial infection (6). Whereas CD8 memory T cells isolated from nonlymphoid tissues exhibit lytic activity directly ex vivo, their splenic counterparts do not. These results extended the TCM/TEM paradigms to the mouse system. In summary, there is now convincing evidence that antigen-specific CD4 and CD8 memory T cells persist as TCM and TEM populations. However, since the first description of TCM and TEM, it became evident that these two broad subsets were heterogeneous in expression of chemokine receptors, adhesion, and costimulatory molecules (4).

Heterogeneity of TCM and TEM Subsets of TCM and TEM with distinct functional programs can be identified according to the expression of surface molecules. Costimulatory molecules have been the first markers used to dissect the heterogeneity of memory T cells. CD27 and CD28, which are expressed on na¨ıve T cells, are also expressed on some memory T cells but are absent in a subset of CD8 memory T cells characterized by high effector function and expression of CD45RA (16, 17). The CD27−CD45RA+ CD8 T cell population largely overlaps with TEMRA. However, some cells within TEMRA express CD27 and display phenotypic and functional features that are intermediate between na¨ıve and effector T cells (18). A sizable proportion of circulating TCM expresses CXCR5, the receptor for CXCL13, a chemokine produced in B cell follicles (19, 20). These CXCR5+ T

12 Feb 2004

16:10

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

748

AR

AR210-IY22-25.tex

SALLUSTO

¥

GEGINAT

¥

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

P1: IKH

LANZAVECCHIA

cells, which have been defined as follicular helper T cells (TFH), are nonpolarized and upon activation produce IL-2 and some IL-10. They are also present in tonsils where they express CD40L and ICOS and provide spontaneous help to B cells (21–23). The TEM pool can be subdivided according to the expression of chemokine receptors characteristic of TH1 or TH2 cells (24). Thus within TEM, CCR5 and CXCR6 discriminate TH1 and CTL, whereas CCR3 and CRTh2 identify TH2 cells (25–28). CXCR3 and CCR4 are expressed on TH1 and TH2 (29, 30), respectively, but are also expressed on distinct subsets of TCM. As is discussed below, CXCR3+ and CCR4+ TCM subsets represent pre-effector cells (pre-TH1 and pre-TH2, respectively). The combinatorial expression of adhesion molecules and chemokine receptors allows tissue-specific targeting of T cell and leukocyte subsets (31, 32). Thus the simultaneous expression of CLA and CCR4 identifies skin homing T cells (33), whereas the expression of α4β7 and CCR9 is characteristic of gut-homing T cells (34). Some skin-homing and gut-homing T cells express CCR7 (10), suggesting that they may be capable of homing to lymphoid as well as nonlymphoid tissues. In addition, while most TCM simultaneously express CCR7 and CD62L, there are several TEM, especially within CD4, that lack CCR7 but express CD62L (4). This finding is consistent with the ability of CD62L+ TEM to enter lymph nodes through HEV using other receptors that bind to chemokines capable of mediating arrest under flow (35). This may be particularly relevant in inflammatory conditions when chemokines produced in peripheral tissues may be transported and displayed on the luminal face of endothelial cells (36, 37). In summary, functional subsets can be discriminated within TCM and TEM using chemokine receptors and other markers (Figure 1). It should be noted, however, that some of these markers are rapidly and transiently modulated upon cell activation. For instance, following antigenic stimulation TEM transiently upregulate CCR7 and CXCR5 while downregulating CCR5 (38). Furthermore, CD62L is rapidly shed after TCR triggering or following lymph node immigration (39, 40). Thus the phenotypic characterization of TCM and TEM applies to only resting cells, i.e., those that are not engaged in an antigen-driven response.

Response of TCM and TEM to Antigenic Stimulation When stimulated in vitro, memory T cells show low-activation threshold and vigorous proliferation. Although both TCM and TEM have a high responsiveness to antigenic stimulation, the expansion potential decreases from TCM to TEM and is very low in CD8 TEMRA (4, 41). The reduced proliferative capacity correlates with a decrease in telomere length and with an increased propensity to undergo apoptosis. These intrinsic constraints can be overridden by costimulation, which induces telomerase activity and upregulation of antiapoptotic molecules (42, 43). Indeed, whereas memory T cells (especially TEM) have short telomeres (4), a small subset

12 Feb 2004

16:10

AR

AR210-IY22-25.tex

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MEMORY T CELL SUBSETS

P1: IKH

749

Figure 1 Phenotypic heterogeneity of human memory T cells. Note that the percentages are only indicative of those found in a healthy human adult because there is considerable individual variability.

of CD8 memory T cells, which have undergone several rounds of cell division in vivo (as detected by low levels of T cell receptor excision circles), has long telomeres, possibly owing to induction of telomerase activity (18). When analyzed at the clonal level, TCM stimulated in vitro appear to be heterogeneous in their ability to differentiate (44). Some cells can be propagated in a noneffector state by stimulation under neutral conditions and can be induced to differentiate to TH1 and TH2 upon stimulation in the presence of IL-12 or IL-4, respectively. Others, however, spontaneously differentiate to IFN-γ - or IL4-producing cells, even if stimulated in the absence of polarizing cytokines. Therefore, these cells represent pre-TH1 and pre-TH2 and can be identified according to the expression of CXCR3 and CCR4, respectively (41; J. Geginat, unpublished data). When stimulated under neutral conditions, TEM retain their TH1 or TH2 phenotype, demonstrating that the pattern of cytokine gene expression that is imprinted at priming can be stably maintained (44). However, when stimulated under opposite polarizing conditions, most TEM can acquire the ability to express the alternative cytokine. Thus TH1 maintain the IFN-γ -producing ability but are also able to produce IL-4. Similarly, most TH2 maintain IL-4 production and acquire IFN-γ -producing ability. An exception is represented by CRTh2+ T cells, which are irreversibly committed to the TH2-lineage and are unable to acquire IFN-γ -producing ability when stimulated with IL-12. The ability of most human TEM to undergo divergent

12 Feb 2004

16:10

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

750

AR

AR210-IY22-25.tex

SALLUSTO

¥

GEGINAT

¥

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

P1: IKH

LANZAVECCHIA

TH1 and TH2 differentiation is a property that is not shared by mouse T cells, which become rapidly committed following stimulation (45). Considering that in the human system a substantial number of TCM are uncommitted and most memory T cells maintain cytokine flexibility, it is possible that expression of the opposite cytokine can be enforced by delivering recall antigens together with the appropriate polarizing signals. In summary, TCM are heterogeneous and contain both uncommitted and precommitted noneffector cells. Importantly, the CCR7+ phenotype is rapidly lost upon antigenic stimulation concomitant with the differentiation of TCM to effector cells. In addition, with the exception of a transient CCR7 upregulation induced by TCR stimulation, TEM remain CCR7− and maintain both memory and flexibility of cytokine gene expression. These results are consistent with the notion that CCR7 expression is irreversibly lost upon differentiation and marks the TCM to TEM transition. How do TCM and TEM achieve antigenic stimulation in vivo? TCM, with access to secondary lymphoid organs, can be stimulated by antigen presented by mature DC, whereas TEM, which are largely excluded from these areas, can be stimulated by antigen presented by nonprofessional APC in a milieu that does not favor stable cell-cell interactions (46, 47). How then can the quality of the primary response be perpetuated in the secondary response? We consider three possibilities. First, the T cell stimulatory conditions provided by the innate immunity through DC activation may be similar in both primary and secondary responses and therefore generate the same type of effector cells from uncommitted precursors. Second, precommitted T cells may have been generated in the primary response, and upon subsequent antigen exposure, the cells rapidly differentiate to TH1 or TH2. A third possibility is that TEM may condition DC in the peripheral tissues or may reach the antigen-stimulated lymph nodes by upregulating CCR7 or through alternative pathways. In both of these cases the local production of IFN-γ and IL-4 by TEM may further direct a TH1 or TH2 response. It is interesting to consider the differentiation stage of memory T cells within the framework of the hygiene hypothesis (48). Individuals that have undergone multiple TH1 responses may be protected from allergic diseases not only by memory TH1 cells present in nonlymphoid tissues but also by the large repertoire of potentially cross-reactive pre-TH1 cells present in secondary lymphoid organs.

Response of TCM and TEM to Homeostatic Cytokines Proliferation of memory T cells can be driven not only by antigenic stimulation but also by cytokines. Indeed, under steady-state conditions, memory T cells slowly turn over in the absence of antigen (49, 50). In vivo studies using gene-targeted mice demonstrated that IL-7 and IL-15, which are constitutively produced by a variety of cells, play an essential role for maintenance of both CD4 (51) and CD8 T cells (52–54). Early studies in the human system showed that T cells proliferate in response to γ -common dependent cytokines in a TCR-independent

12 Feb 2004

16:10

AR

AR210-IY22-25.tex

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MEMORY T CELL SUBSETS

P1: IKH

751

fashion (55). More recent studies demonstrated that TCM not only proliferate in response to homeostatic cytokines but also differentiate to effector cells expressing receptors for inflammatory chemokines and producing large amounts of effector cytokines (41, 56). In response to IL-7 and IL-15, CXCR3+ TCM differentiate to TH1, whereas most CCR4+ TCM differentiate to TH2, consistent with the notion that these subsets are precommitted. Under the same culture conditions, CD8 TCM generate a variety of effector cells that include CD45RA+ and CD45RA− CTL (41, 57). The findings that TEMRA appear only late during the immune response (14) and that CD45RA+ can be upregulated on memory T cells following cytokinedriven but not antigen-driven proliferation (41) suggest that the TEMRA subset is generated primarily through homeostatic rather than antigen-dependent pathways.

T CELL DIFFERENTIATION: EFFECTOR CELLS AND INTERMEDIATES The molecular definition of TH1 and TH2 differentiation provides a powerful paradigm to understand where, along the T cell differentiation pathway, TCM and TEM are positioned and to identify the signals involved in their generation. The evidence that we discuss below indicates that TCM are cells arrested at intermediate stages of differentiation preceding TEM and suggests that TCM are generated by subthreshold stimulation.

Imprinting of TH1 and TH2 Differentiation Upon antigenic stimulation, na¨ıve T cells can enter a TH1 or TH2 differentiation program, which involves the coordinated expression of genes controlling tissue homing and effector function. Signals emanating from the TCR and from the IL-12 or the IL-4 receptor act in synergy to induce specific transcription factors (58, 59) that mediate chromatin remodeling events at target genes (60–64). In particular in differentiating TH1 and TH2 cells, T-bet and GATA-3 induce histone modifications and DNA methylation of the IFN-γ and IL-4 genes, respectively, increasing their accessibility to the transcriptional machinery (65). Other genes characteristic of TEM, such as receptors for inflammatory chemokines and perforin, appear to be controlled by similar chromatin remodeling events (66, 67). Histone acetylation at a locus can be inherited through mitosis and can therefore contribute to the maintenance of specific states of gene activity from one generation to the next. Circulating CCR5+ and CRTh2+ TEM show patterns and levels of histone acetylation at cytokine genes characteristic of in vitro cultured TH1 and TH2 cells, whereas TCM have an acetylation state similar to that of na¨ıve T cells (44). The cytokine gene flexibility characteristic of human TEM described above can be explained by the capacity of these cells to upregulate the relevant transcription factor. Indeed, upon stimulation under TH2 conditions, CCR5+ TH1 cells upregulate GATA-3 and acquire Il4 acetylation and expression. In contrast, when

12 Feb 2004

16:10

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

752

AR

AR210-IY22-25.tex

SALLUSTO

¥

GEGINAT

¥

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

P1: IKH

LANZAVECCHIA

stimulated in the presence of IL-12, CRTh2+ TH2 cells are unable to upregulate Tbet and therefore fail to acquire IFN-γ -producing capacity (44). These results are in agreement with transfection experiments demonstrating that T-bet and GATA-3 are necessary and sufficient to induce IFN-γ and IL-4 gene expression (58, 68). Recent data on the expression of T-bet and GATA-3 in TCM and TEM suggest a progressive loss of flexibility as T helper cells proceed from na¨ıve to the effector memory stages of differentiation (69). Chromatin remodeling at the Ifng and Il4 loci is acquired progressively as a function of TCR and cytokine stimulation (70). Regulation of cytokine gene accessibility allows for a wide range of expression levels in response to antigenic stimulation. For instance, IFN-γ and IL-4 are produced in very small amounts by na¨ıve T cells (45) and at increased levels in some TCM, whereas a rapid and greater production is characteristic of highly differentiated TH1 and TH2 present within TEM (44). The levels and kinetics of cytokine gene expression are obviously relevant parameters of effector function and, in this context, it is worth considering that in vitro and in vivo read outs can differ significantly and that the nature of the stimulus (i.e., antigenic versus pharmacological stimulation) may lead to divergent results (68, 71). T cell differentiation to TH1 or TH2 can be inhibited either by neutralizing IL12 and IL-4 or by addition of TGF-β (30, 72). This results in the generation of nonpolarized T cells that retain constitutive CCR7 expression, have low levels of GATA-3 and T-bet, and carry hypoacetylated IFN-γ and IL-4 genes (44, 73, 74). These cells have all the characteristics of TCM, i.e., being either uncommitted or precommitted to TH1 or TH2.

Signal Strength and Progressive T Cell Differentiation In addition to the role cytokines play, it is worth emphasizing that of TCR signal strength as a major factor in determining T cell differentiation. The strength of signaling that T cells receive can vary widely depending not only on the concentration of antigen and costimulatory molecules (determining the rate of TCR triggering and signal amplification) (75, 76) but also on the duration of the interaction between T cells and APC (determining the duration of signaling) (77). Recent studies revealed that DC/T cell interactions are highly dynamic and can last from a few minutes to several hours (46, 78). Importantly, a prolonged stimulation can compensate for a lower level of antigen or costimulatory molecules, whereas an increase in antigen or costimulatory molecules can compensate for a shorter stimulation. With use of an in vitro priming system in which the strength of antigenic stimulation is tightly controlled, na¨ıve CD4 T cells primed by a weak stimulus proliferate but do not develop effector function (79). Upon in vivo transfer these cells migrate to lymph nodes where they rapidly proliferate and differentiate in response to antigen. In contrast, T cells primed by a strong stimulus in the presence of IL-12 or IL-4 differentiate to TH1 or TH2 that, upon in vivo transfer, are excluded from lymph nodes and migrate to inflamed peripheral tissues. After an initial

12 Feb 2004

16:10

AR

AR210-IY22-25.tex

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MEMORY T CELL SUBSETS

P1: IKH

753

stimulation, CD8 T cell differentiation to peripheral tissue homing effector cells is promoted by IL-2, whereas IL-15 expands nonpolarized T cells that, upon in vivo transfer, home to both lymphoid and nonlymphoid tissues (80, 81). Signal strength can also be modulated in vitro by the number and type of DC used (82). Whereas low DC/T cell ratio preferentially generate TCM-like cells, higher ratios favor priming of effector cells and eventually lead to T cell deletion by activationinduced cell death. The expression of costimulatory molecules, such as ICOS-L and 4-1BBL, plays a role in reinforcing the signals by APC and is required for effective T cell responses (83–86). Initial studies on the requirements for CD8 T cell priming suggested that their differentiation follows an autonomous program after brief stimulation with highly stimulatory engineered APC (87). However, further studies showed that signal strength also plays a key role in CD8 T cell differentiation (88, 89). Priming by a signal of suboptimal strength, such as that provided by immature DC, induces CD8 T cell proliferation, but the cells fail to upregulate antiapoptotic molecules and receptors for homeostatic cytokines (88). Upon in vivo transfer these cells, defined as unfit, die by neglect. Thus both CD4 and CD8 T cells show a signal strength–dependent differentiation despite the more readily inducible proliferation and differentiation of CD8 T cells. Taken together the available data are consistent with the concept that the strength of the signals delivered by TCR and cytokine receptors drives T cells through hierarchical thresholds of differentiation (90). The progression follows a sequence of proliferation, preceding the acquisition of fitness, effector function and, eventually, death (Figure 2). Because TCR and cytokine stimulation are stochastic events, not all proliferating T cells receive the same strength of signal. Consequently, primed T cells reach a variety of differentiation stages that contain effector cells as well as cells that have been arrested at intermediate levels of differentiation. These intermediates retain lymph node homing receptors and have initiated, but not completed, the remodeling events of genes involved in effector functions. This spectrum can be simply resolved into distinct subsets of TCM and TEM surviving after antigen clearance. The question remains—how can different levels of signal strength be delivered in the course of the same immune response? We can envision two possibilities. The first is that in highly stimulatory conditions some activated T cells may prematurely detach from DC and fail to receive further stimulation consistent with observations of dynamic DC/T cell interactions in vivo (78). Second, different levels of stimulation may be delivered at different times. At early stages, large numbers of mature DC carrying high doses of antigen and secreting large amounts of polarizing cytokines would promote a massive proliferation of antigen-specific T cells and drive them into effector cells, some of which will persist as TEM. At later time points, the stimulatory conditions may change substantially with arrival of fewer DC, which furthermore carry small amounts of antigen and have exhausted their cytokine-producing capacity (91). Such conditions may lead to the expansion of noneffector cells, of which the fit ones will persist as TCM. According

12 Feb 2004

16:10

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

754

AR

AR210-IY22-25.tex

SALLUSTO

¥

GEGINAT

¥

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

P1: IKH

LANZAVECCHIA

Figure 2 A signal strength model for T cell differentiation and memory T cell generation. The duration and intensity of antigenic stimulation is indicated by the length and thickness of the solid arrows. Antigen-independent events leading to T cell proliferation and differentiation are indicated by the dotted lines. AICD, activation-induced cell death.

to this model (7), precursors of TEM would be preferentially generated early on, whereas precursors of TCM would be preferentially generated at late stages of the immune response. The model proposed above is consistent with the observation that memory precursors are not present at the peak of the immune response but progressively appear during the contraction phase coincident with global changes in transcriptional profiling (92).

MONITORING THE GENERATION OF TCM AND TEM The signal strength model discussed above is consistent with several studies that established a relationship between the nature of antigenic stimulation in vivo and the fate of T cells generated. Strong stimuli, such as live bacteria or antigen formulated with microbial products, induce fully differentiated CTL and TH1 cells; killed bacteria or antigen associated with the bacterial cell wall–derived Ribi adjuvant induce the expansion of noneffector CD8 or CD4 T cells (93, 94). Furthermore,

12 Feb 2004

16:10

AR

AR210-IY22-25.tex

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MEMORY T CELL SUBSETS

P1: IKH

755

antigens cross-presented by immature DC induce abortive proliferation or expansion of noneffector cells depending on the level of antigen expression (95–99). In these conditions, in vivo administration of IL-12 or T cell help increases the generation of effector cells. Reintroduction of antigen, but not inflammatory stimuli, during primary infection reinforces the generation of memory T cells, indicating that a sustained antigen presentation is critical for effective memory T cell generation (100). Several studies examined the heterogeneity of responding T cells in the course of antigen-driven immune responses. Following immunization with a protein antigen, two types of responding T cells can be identified as early as day five: B helper T cells expressing the follicular homing receptor CXCR5+ and tissue homing inflammatory T cells expressing CXCR3 (101). CD4 T cells responding to influenza virus were found to be a heterogeneous continuum of cells in terms of number of cell divisions, phenotype, and function (102). Interestingly, those migrating to the lung are characterized by loss of CD62L, downregulation of CCR7, secretion of large amounts of IFN-γ , and reduced IL-2 levels relative to those in the secondary lymphoid organs. While the response declines with viral clearance, a range of resting cell subsets reflecting the pattern at the peak of response is retained, suggesting that heterogeneous effector populations may give rise to corresponding memory populations (102). Furthermore, the same CD8 T cell clonotypes responding to influenza are present in the lymphoid tissue and in the virus-infected lung (103), demonstrating that the heterogeneity in tissue localization can be generated within the same T cell clone. In the mouse system it has been possible to directly address the relationship between population of primed T cells and memory subsets. A study on CD4 T cells showed that cytokine-secreting cells fail to generate memory, whereas nonpolarized cells, with the characteristic of pre-TH1, have the ability to transfer strong secondary TH1 responses (104). In a recent study the precursor-product relationship between TCM and TEM has been analyzed in the CD8 system by adoptive transfer experiments (105). The results point to three main conclusions. First, TEM are present only transiently, and, upon transfer, they all convert to TCM by reacquiring CCR7 and CD62L expression. Second, the rate of conversion from TEM to TCM is inversely proportional to the strength of stimulation. Third, TCM convert to TEM upon antigenic restimulation but not under steady-state conditions. These findings led to the proposition that in mice TCM and TEM do not necessarily represent distinct subsets, but are part of a continuum in a linear nave → effector → TEM → TCM differentiation pathway (106). These results are in apparent contrast with the persistence and stability of both TCM and TEM subsets in humans (15) and with the failure of TEM to reacquire the constitutive CCR7 expression in vitro (107), suggesting a relevant difference between these two species. Recent findings indicate that T cell help is an absolute requirement for generation of functional CD8 T cell memory. Indeed, animals primed in the absence of T cell help mount normal primary CTL responses but fail to mount secondary responses to the same antigen in spite of normal numbers of memory T cells

12 Feb 2004

16:10

756

AR

AR210-IY22-25.tex

SALLUSTO

¥

GEGINAT

¥

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

P1: IKH

LANZAVECCHIA

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

(108–110). The nature of the T cell defect it is not clear, but it is tempting to speculate that it may be due to a qualitative or quantitative defect of TCM. The mechanism of T cell help is also unclear. Helper T cells can boost CTL responses not only through DC activation (111–113) and provision of IL-2 but also through direct T-T interactions mediated by CD40L (114). Positive signals controlling the generation of memory T cells are provided by other members of the TNF receptor family such as OX40 and CD27 (115–117).

OPEN QUESTIONS AND PERSPECTIVES TCM and memory B cells represent memory stem cells because they respond to chronic or repeated antigenic stimulation by self-renewing and generation of effector T cells and plasma cells (118). Memory cells divide constitutively, even in the absence of antigen (50), and in response to cytokines proliferate and differentiate to effector T cells. We therefore propose that TCM and memory B cells may also function as stem cells under homeostatic conditions. Accordingly, TCM may divide asymmetrically in response to cytokines, maintaining their number and, at the same time, differentiating and replenishing TEM that are lost in peripheral tissues, such as lung or gut. A similar mechanism was recently proposed for human memory B cells, which maintain serological memory through continuous polyclonal activation (119). Cell tracking experiments and detailed kinetics and repertoire analysis of memory T cell subsets should be performed in order to address this important issue in humans. Mouse experiments showed that TCM have much higher capacity to reconstitute the memory T cell pool than effector cells (104, 105). These observations are highly relevant for immunotherapy. Injection of large numbers of effector cells may provide immediate protection but fail to reconstitute long-term memory (120). It would be important to find out the optimal conditions to prime and expand TCM-like cells in humans. Although the existence of TCM and TEM is now well documented, there are significant differences between mouse and human that need to be clarified. One aspect relates to the extent of differentiation of the TCM subset and the mechanisms responsible for cytokine memory and flexibility in both CD4 and CD8 compartments. In addition, human TCM and TEM populations appear to be stable with no evidence of interconversion, whereas they appear to be dynamic and plastic in the mouse system (15, 105). This difference may be related to the experimental systems or may reflect differences between the species. A particularly striking difference between experimental animals and humans relates to number of specificities that need to be accomodated and maintained in the memory pool. Since the total pool does not substantially increase with aging, new specificities that progressively accumulate lead to dilution of pre-existing memory T cells. The same mechanism will favor the expansion of cross-reacting T cells, which become enriched in the memory pool. Both attrition and cross-reactivity have been documented in mice (121) and may

12 Feb 2004

16:10

AR

AR210-IY22-25.tex

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

MEMORY T CELL SUBSETS

P1: IKH

757

play an even more important role in humans because of the much longer lifespan and antigenic exposure. The prevalence of cross-reactive memory T cells in elderly should be tested experimentally. Solving any of these questions will have not only theoretical but also practical implications for vaccination and immunotherapy.

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ACKNOWLEDGMENTS We thank Amanda Gett for critical reading and comments. The Institute for Research in Biomedicine is supported by the Helmut Horten Foundation. The work on immunological memory in the authors’ laboratories is supported by grants from the European Community (QLK2-CT-2001–01250) and the Swiss National Funds (3100–63885 and 3100–101962). The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Manz RA, Thiel A, Radbruch A. 1997. Lifetime of plasma cells in the bone marrow. Nature 388:133–34 2. Slifka MK, Antia R, Whitmire JK, Ahmed R. 1998. Humoral immunity due to longlived plasma cells. Immunity 8:363–72 3. Ochsenbein AF, Pinschewer DD, Sierro S, Horvath E, Hengartner H, Zinkernagel RM. 2000. Protective long-term antibody memory by antigen-driven and T help-dependent differentiation of longlived memory B cells to short-lived plasma cells independent of secondary lymphoid organs. Proc. Natl. Acad. Sci. USA 97:13263–68 4. Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. 1999. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401:708–12 5. Reinhardt RL, Khoruts A, Merica R, Zell T, Jenkins MK. 2001. Visualizing the generation of memory CD4 T cells in the whole body. Nature 410:101–5 6. Masopust D, Vezys V, Marzo AL, Lefrancois L. 2001. Preferential localization of effector memory cells in nonlymphoid tissue. Science 291:2413–17 7. Lanzavecchia A, Sallusto F. 2000. Dy-

8.

9.

10.

11.

12.

namics of T lymphocyte responses: intermediates, effectors, and memory cells. Science 290:92–97 Campbell JJ, Bowman EP, Murphy K, Youngman KR, Siani MA, et al. 1998. 6-C-kine (SLC), a lymphocyte adhesiontriggering chemokine expressed by high endothelium, is an agonist for the MIP3beta receptor CCR7. J. Cell Biol. 141:1053–59 Forster R, Schubel A, Breitfeld D, Kremmer E, Renner-Muller I, et al. 1999. CCR7 coordinates the primary immune response by establishing functional microenvironments in secondary lymphoid organs. Cell 99:23–33 Campbell JJ, Murphy KE, Kunkel EJ, Brightling CE, Soler D, et al. 2001. CCR7 expression and memory T cell diversity in humans. J. Immunol. 166:877–84 Champagne P, Ogg GS, King AS, Knabenhans C, Ellefsen K, et al. 2001. Skewed maturation of memory HIVspecific CD8 T lymphocytes. Nature 410:106–11 Dunbar PR, Smith CL, Chao D, Salio M, Shepherd D, et al. 2000. A shift in the phenotype of melan-A-specific CTL identifies melanoma patients with an active

12 Feb 2004

16:10

758

13.

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

14.

15.

16.

17.

18.

19.

20.

21.

AR

AR210-IY22-25.tex

SALLUSTO

¥

GEGINAT

¥

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

P1: IKH

LANZAVECCHIA

tumor-specific immune response. J. Immunol. 165:6644–52 Valmori D, Scheibenbogen C, Dutoit V, Nagorsen D, Asemissen AM, et al. 2002. Circulating tumor-reactive CD8+ T cells in melanoma patients contain a CD45RA+CCR7− effector subset exerting ex vivo tumor-specific cytolytic activity. Cancer Res. 62:1743–50 Hislop AD, Annels NE, Gudgeon NH, Leese AM, Rickinson AB. 2002. Epitopespecific evolution of human CD8+ T cell responses from primary to persistent phases of Epstein-Barr virus infection. J. Exp. Med. 195:893–905 Baron V, Bouneaud C, Cumano A, Lim A, Arstila TP, et al. 2003. The repertoires of circulating human CD8+ central and effector memory T cell subsets are largely distinct. Immunity 18:193–204 Hamann D, Baars PA, Rep MH, Hooibrink B, Kerkhof-Garde SR, et al. 1997. Phenotypic and functional separation of memory and effector human CD8+ T cells. J. Exp. Med. 186:1407–18 Appay V, Dunbar PR, Callan M, Klenerman P, Gillespie GM, et al. 2002. Memory CD8+ T cells vary in differentiation phenotype in different persistent virus infections. Nat. Med. 8:379–85 Rufer N, Zippelius A, Batard P, Pittet MJ, Kurth I, et al. 2003. Ex vivo characterization of human CD8+ T subsets with distinct replicative history and partial effector functions. Blood 102:1779–87 Gunn MD, Ngo VN, Ansel KM, Ekland EH, Cyster JG, Williams LT. 1998. A B-cell-homing chemokine made in lymphoid follicles activates Burkitt’s lymphoma receptor-1. Nature 391:799–803 Legler DF, Loetscher M, Roos RS, ClarkLewis I, Baggiolini M, Moser B. 1998. B cell-attracting chemokine 1, a human CXC chemokine expressed in lymphoid tissues, selectively attracts B lymphocytes via BLR1/CXCR5. J. Exp. Med. 187:655– 60 Breitfeld D, Ohl L, Kremmer E, Ellwart J,

22.

23.

24.

25.

26.

27.

28.

29.

30.

Sallusto F, et al. 2000. Follicular B helper T cells express CXC chemokine receptor 5, localize to B cell follicles, and support immunoglobulin production. J. Exp. Med. 192:1545–52 Schaerli P, Willimann K, Lang AB, Lipp M, Loetscher P, Moser B. 2000. CXC chemokine receptor 5 expression defines follicular homing T cells with B cell helper function. J. Exp. Med. 192:1553– 62 Kim CH, Rott LS, Clark-Lewis I, Campbell DJ, Wu L, Butcher EC. 2001. Subspecialization of CXCR5+ T cells: B helper activity is focused in a germinal centerlocalized subset of CXCR5+ T cells. J. Exp. Med. 193:1373–81 Sallusto F, Lanzavecchia A, Mackay CR. 1998. Chemokines and chemokine receptors in T-cell priming and Th1/Th2- mediated responses. Immunol. Today 19:568– 74 Loetscher P, Uguccioni M, Bordoli L, Baggiolini M, Moser B, et al. 1998. CCR5 is characteristic of Th1 lymphocytes. Nature 391:344–45 Kim CH, Kunkel EJ, Boisvert J, Johnston B, Campbell JJ, et al. 2001. Bonzo/CXCR6 expression defines type 1polarized T-cell subsets with extralymphoid tissue homing potential. J. Clin. Invest. 107:595–601 Sallusto F, Mackay CR, Lanzavecchia A. 1997. Selective expression of the eotaxin receptor CCR3 by human T helper 2 cells. Science 277:2005–7 Nagata K, Tanaka K, Ogawa K, Kemmotsu K, Imai T, et al. 1999. Selective expression of a novel surface molecule by human Th2 cells in vivo. J. Immunol. 162:1278–86 Bonecchi R, Bianchi G, Bordignon PP, D’Ambrosio D, Lang R, et al. 1998. Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s. J. Exp. Med. 187:129–34 Sallusto F, Lenig D, Mackay CR,

12 Feb 2004

16:10

AR

AR210-IY22-25.tex

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

MEMORY T CELL SUBSETS

31.

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

32.

33.

34.

35.

36.

37.

38.

Lanzavecchia A. 1998. Flexible programs of chemokine receptor expression on human polarized T helper 1 and 2 lymphocytes. J. Exp. Med. 187:875–83 Butcher EC, Picker LJ. 1996. Lymphocyte homing and homeostasis. Science 272:60–66 von Andrian UH, Mackay CR. 2000. Tcell function and migration. Two sides of the same coin. N. Engl. J. Med. 343:1020– 34 Campbell JJ, Haraldsen G, Pan J, Rottman J, Qin S, et al. 1999. The chemokine receptor CCR4 in vascular recognition by cutaneous but not intestinal memory T cells. Nature 400:776–80 Zabel BA, Agace WW, Campbell JJ, Heath HM, Parent D, et al. 1999. Human G protein-coupled receptor GPR9-6/CC chemokine receptor 9 is selectively expressed on intestinal homing T lymphocytes, mucosal lymphocytes, and thymocytes and is required for thymusexpressed chemokine-mediated chemotaxis. J. Exp. Med. 190:1241–56 Campbell JJ, Hedrick J, Zlotnik A, Siani MA, Thompson DA, Butcher EC. 1998. Chemokines and the arrest of lymphocytes rolling under flow conditions. Science 279:381–84 Middleton J, Neil S, Wintle J, Clark-Lewis I, Moore H, et al. 1997. Transcytosis and surface presentation of IL-8 by venular endothelial cells. Cell 91:385–95 Stein JV, Rot A, Luo Y, Narasimhaswamy M, Nakano H, et al. 2000. The CC chemokine thymus-derived chemotactic agent 4 (TCA-4, secondary lymphoid tissue chemokine, 6Ckine, exodus-2) triggers lymphocyte function-associated antigen 1-mediated arrest of rolling T lymphocytes in peripheral lymph node high endothelial venules. J. Exp. Med. 191:61– 76 Sallusto F, Kremmer E, Palermo B, Hoy A, Ponath P, et al. 1999. Switch in chemokine receptor expression upon TCR stimulation reveals novel homing poten-

39.

40.

41.

42.

43.

44.

45.

46.

47.

P1: IKH

759

tial for recently activated T cells. Eur. J. Immunol. 29:2037–45 Kishimoto TK, Jutila MA, Butcher EC. 1990. Identification of a human peripheral lymph node homing receptor: a rapidly down-regulated adhesion molecule. Proc. Natl. Acad. Sci. USA 87:2244–48 Chao CC, Jensen R, Dailey MO. 1997. Mechanisms of L-selectin regulation by activated T cells. J. Immunol. 159:1686– 94 Geginat J, Lanzavecchia A, Sallusto F. 2003. Proliferation and differentiation potential of human CD8+ memory T-cell subsets in response to antigen or homeostatic cytokines. Blood 101:4260–66 Son NH, Murray S, Yanovski J, Hodes RJ, Weng N. 2000. Lineage-specific telomere shortening and unaltered capacity for telomerase expression in human T and B lymphocytes with age. J. Immunol. 165:1191–96 Boise LH, Minn AJ, Noel PJ, June CH, Accavitti MA, et al. 1995. CD28 costimulation can promote T cell survival by enhancing the expression of Bcl-XL. Immunity 3:87–98 Messi M, Giacchetto I, Nagata K, Lanzavecchia A, Natoli G, Sallusto F. 2003. Memory and flexibility of cytokine gene expression as separable properties of human TH1 and TH2 lymphocytes. Nat. Immunol. 4:78–86 Grogan JL, Mohrs M, Harmon B, Lacy DA, Sedat JW, Locksley RM. 2001. Early transcription and silencing of cytokine genes underlie polarization of T helper cell subsets. Immunity 14:205–15 Gunzer M, Schafer A, Borgmann S, Grabbe S, Zanker KS, et al. 2000. Antigen presentation in extracellular matrix: interactions of T cells with dendritic cells are dynamic, short lived, and sequential. Immunity 13:323–32 Harris NL, Watt V, Ronchese F, Le Gros G. 2002. Differential T cell function and fate in lymph node and nonlymphoid tissues. J. Exp. Med. 195:317–26

12 Feb 2004

16:10

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

760

AR

AR210-IY22-25.tex

SALLUSTO

¥

GEGINAT

¥

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

P1: IKH

LANZAVECCHIA

48. Yazdanbakhsh M, Kremsner PG, van Ree R. 2002. Allergy, parasites, and the hygiene hypothesis. Science 296:490–94 49. Michie CA, McLean A, Alcock C, Beverley PC. 1992. Lifespan of human lymphocyte subsets defined by CD45 isoforms. Nature 360:264–65 50. Tough DF, Sprent J. 1994. Turnover of naive- and memory-phenotype T cells. J. Exp. Med. 179:1127–35 51. Seddon B, Tomlinson P, Zamoyska R. 2003. Interleukin 7 and T cell receptor signals regulate homeostasis of CD4 memory cells. Nat. Immunol. 4:680–86 52. Zhang X, Sun S, Hwang I, Tough DF, Sprent J. 1998. Potent and selective stimulation of memory-phenotype CD8+ T cells in vivo by IL-15. Immunity 8:591– 99 53. Lodolce JP, Boone DL, Chai S, Swain RE, Dassopoulos T, et al. 1998. IL-15 receptor maintains lymphoid homeostasis by supporting lymphocyte homing and proliferation. Immunity 9:669–76 54. Schluns KS, Kieper WC, Jameson SC, Lefrancois L. 2000. Interleukin-7 mediates the homeostasis of naive and memory CD8 T cells in vivo. Nat. Immunol. 1:426–32 55. Unutmaz D, Pileri P, Abrignani S. 1994. Antigen-independent activation of naive and memory resting T cells by a cytokine combination. J. Exp. Med. 180:1159– 64 56. Geginat J, Sallusto F, Lanzavecchia A. 2001. Cytokine-driven proliferation and differentiation of human naive, central memory, and effector memory CD4+ T cells. J. Exp. Med. 194:1711–19 57. Alves NL, Hooibrink B, Arosa FA, Van Lier RA. 2003. IL-15 induces antigenindependent expansion and differentiation of human naive CD8+ T cells in vitro. Blood 102:2541–46 58. Zheng W, Flavell RA. 1997. The transcription factor GATA-3 is necessary and sufficient for Th2 cytokine gene expression in CD4 T cells. Cell 89:587–96

59. Szabo SJ, Kim ST, Costa GL, Zhang X, Fathman CG, Glimcher LH. 2000. A novel transcription factor, T-bet, directs Th1 lineage commitment. Cell 100:655– 69 60. Mullen AC, High FA, Hutchins AS, Lee HW, Villarino AV, et al. 2001. Role of T-bet in commitment of TH1 cells before IL-12-dependent selection. Science 292:1907–10 61. Agarwal S, Rao A. 1998. Modulation of chromatin structure regulates cytokine gene expression during T cell differentiation. Immunity 9:765–75 62. Takemoto N, Kamogawa Y, Jun-Lee H, Kurata H, Arai KI, et al. 2000. Cutting edge: chromatin remodeling at the IL4/IL-13 intergenic regulatory region for Th2-specific cytokine gene cluster. J. Immunol. 165:6687–91 63. Lee GR, Fields PE, Flavell RA. 2001. Regulation of IL-4 gene expression by distal regulatory elements and GATA-3 at the chromatin level. Immunity 14:447–59 64. Iezzi G, Scotet E, Scheidegger D, Lanzavecchia A. 1999. The interplay between the duration of TCR and cytokine signaling determines T cell polarization. Eur. J. Immunol. 29:4092–101 65. Ansel KM, Lee DU, Rao A. 2003. An epigenetic view of helper T cell differentiation. Nat. Immunol. 4:616–23 66. Scotet E, Schroeder S, Lanzavecchia A. 2001. Molecular regulation of CCchemokine receptor 3 expression in human T helper 2 cells. Blood 98:2568–70 67. Lu Q, Wu A, Ray D, Deng C, Attwood J, et al. 2003. DNA methylation and chromatin structure regulate T cell perforin gene expression. J. Immunol. 170:5124–32 68. Afkarian M, Sedy JR, Yang J, Jacobson NG, Cereb N, et al. 2002. T-bet is a STAT1-induced regulator of IL-12R expression in naive CD4+ T cells. Nat. Immunol. 3:549–57 69. Sundrud MS, Grill SM, Ni D, Nagata K, Alkan SS, et al. 2003. Genetic reprogramming of primary human T cells reveals

12 Feb 2004

16:10

AR

AR210-IY22-25.tex

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

MEMORY T CELL SUBSETS

70.

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

functional plasticity in Th cell differentiation. J. Immunol. 171:3542–49 Lee DU, Agarwal S, Rao A. 2002. Th2 lineage commitment and efficient IL-4 production involves extended demethylation of the IL-4 gene. Immunity 16:649–60 Barber DL, Wherry EJ, Ahmed R. 2003. Cutting edge: rapid in vivo killing by memory CD8 T cells. J. Immunol. 171:27–31 Sad S, Mosmann TR. 1994. Single IL2-secreting precursor CD4 T cell can develop into either Th1 or Th2 cytokine secretion phenotype. J. Immunol. 153:3514–22 Gorelik L, Fields PE, Flavell RA. 2000. Cutting edge: TGF-beta inhibits Th type 2 development through inhibition of GATA3 expression. J. Immunol. 165:4773–77 Gorelik L, Constant S, Flavell RA. 2002. Mechanism of transforming growth factor beta-induced inhibition of T helper type 1 differentiation. J. Exp. Med. 195:1499– 505 Valitutti S, Muller S, Cella M, Padovan E, Lanzavecchia A. 1995. Serial triggering of many T-cell receptors by a few peptideMHC complexes. Nature 375:148–51 Viola A, Schroeder S, Sakakibara Y, Lanzavecchia A. 1999. T lymphocyte costimulation mediated by reorganization of membrane microdomains. Science 283:680–82 Iezzi G, Karjalainen K, Lanzavecchia A. 1998. The duration of antigenic stimulation determines the fate of naive and effector T cells. Immunity 8:89–95 Stoll S, Delon J, Brotz TM, Germain RN. 2002. Dynamic imaging of T celldendritic cell interactions in lymph nodes. Science 296:1873–76 Iezzi G, Scheidegger D, Lanzavecchia A. 2001. Migration and function of antigenprimed nonpolarized T lymphocytes in vivo. J. Exp. Med. 193:987–93 Manjunath N, Shankar P, Wan J, Weninger W, Crowley MA, et al. 2001. Effector differentiation is not prerequisite for genera-

81.

82.

83.

84.

85.

86.

87.

88.

89.

P1: IKH

761

tion of memory cytotoxic T lymphocytes. J. Clin. Invest. 108:871–78 Weninger W, Crowley MA, Manjunath N, von Andrian UH. 2001. Migratory properties of naive, effector, and memory CD8+ T cells. J. Exp. Med. 194:953–66 Langenkamp A, Casorati G, Garavaglia C, Dellabona P, Lanzavecchia A, Sallusto F. 2002. T cell priming by dendritic cells: thresholds for proliferation, differentiation and death and intraclonal functional diversification. Eur. J. Immunol. 32:2046– 54 Hutloff A, Dittrich AM, Beier KC, Eljaschewitsch B, Kraft R, et al. 1999. ICOS is an inducible T-cell co-stimulator structurally and functionally related to CD28. Nature 397:263–66 Coyle AJ, Lehar S, Lloyd C, Tian J, Delaney T, et al. 2000. The CD28-related molecule ICOS is required for effective T cell-dependent immune responses. Immunity 13:95–105 Kopf M, Coyle AJ, Schmitz N, Barner M, Oxenius A, et al. 2000. Inducible costimulator protein (ICOS) controls T helper cell subset polarization after virus and parasite infection. J. Exp. Med. 192:53–61 Diehl L, van Mierlo GJ, den Boer AT, van der Voort E, Fransen M, et al. 2002. In vivo triggering through 4-1BB enables Th-independent priming of CTL in the presence of an intact CD28 costimulatory pathway. J. Immunol. 168:3755–62 van Stipdonk MJ, Lemmens EE, Schoenberger SP. 2001. Naive CTLs require a single brief period of antigenic stimulation for clonal expansion and differentiation. Nat. Immunol. 2:423–29 Gett AV, Sallusto F, Lanzavecchia A, Geginat J. 2003. T cell fitness determined by signal strength. Nat. Immunol. 4:355– 60 van Stipdonk MJ, Hardenberg G, Bijker MS, Lemmens EE, Droin NM, et al. 2003. Dynamic programming of CD8+ T lymphocyte responses. Nat. Immunol. 4:361– 65

12 Feb 2004

16:10

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

762

AR

AR210-IY22-25.tex

SALLUSTO

¥

GEGINAT

¥

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

P1: IKH

LANZAVECCHIA

90. Lanzavecchia A, Sallusto F. 2002. Opinion-decision making in the immune system: progressive differentiation and selection of the fittest in the immune response. Nat. Rev. Immunol. 2:982–87 91. Langenkamp A, Messi M, Lanzavecchia A, Sallusto F. 2000. Kinetics of dendritic cell activation: impact on priming of TH1, TH2 and nonpolarized T cells. Nat. Immunol. 1:311–16 92. Kaech SM, Hemby S, Kersh E, Ahmed R. 2002. Molecular and functional profiling of memory CD8 T cell differentiation. Cell 111:837–51 93. Lauvau G, Vijh S, Kong P, Horng T, Kerksiek K, et al. 2001. Priming of memory but not effector CD8 T cells by a killed bacterial vaccine. Science 294:1735–39 94. Wang X, Mosmann T. 2001. In vivo priming of CD4 T cells that produce interleukin (IL)-2 but not IL-4 or interferon (IFN)gamma, and can subsequently differentiate into IL-4- or IFN-gamma-secreting cells. J. Exp. Med. 194:1069–80 95. Kurts C, Kosaka H, Carbone FR, Miller JF, Heath WR. 1997. Class I-restricted cross-presentation of exogenous selfantigens leads to deletion of autoreactive CD8+ T cells. J. Exp. Med. 186:239– 45 96. Hawiger D, Inaba K, Dorsett Y, Guo M, Mahnke K, et al. 2001. Dendritic cells induce peripheral T cell unresponsiveness under steady state conditions in vivo. J. Exp. Med. 194:769–79 97. Hernandez J, Aung S, Redmond WL, Sherman LA. 2001. Phenotypic and functional analysis of CD8+ T cells undergoing peripheral deletion in response to cross-presentation of self-antigen. J. Exp. Med. 194:707–17 98. Bonifaz L, Bonnyay D, Mahnke K, Rivera M, Nussenzweig MC, Steinman RM. 2002. Efficient targeting of protein antigen to the dendritic cell receptor DEC-205 in the steady state leads to antigen presentation on major histocompatibility complex class I products and peripheral CD8+

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

T cell tolerance. J. Exp. Med. 196:1627– 38 Hernandez J, Aung S, Marquardt K, Sherman LA. 2002. Uncoupling of proliferative potential and gain of effector function by CD8+ T cells responding to self-antigens. J. Exp. Med. 196:323–33 Busch DH, Kerksiek KM, Pamer EG. 2000. Differing roles of inflammation and antigen in T cell proliferation and memory generation. J. Immunol. 164:4063–70 Campbell DJ, Kim CH, Butcher EC. 2001. Separable effector T cell populations specialized for B cell help or tissue inflammation. Nat. Immunol. 2:876–81 Roman E, Miller E, Harmsen A, Wiley J, Von Andrian UH, et al. 2002. CD4 effector T cell subsets in the response to influenza: heterogeneity, migration, and function. J. Exp. Med. 196:957–68 Turner SJ, Diaz G, Cross R, Doherty PC. 2003. Analysis of clonotype distribution and persistence for an influenza virusspecific CD8+ T cell response. Immunity 18:549–59 Wu CY, Kirman JR, Rotte MJ, Davey DF, Perfetto SP, et al. 2002. Distinct lineages of TH1 cells have differential capacities for memory cell generation in vivo. Nat. Immunol. 3:852–58 Wherry EJ, Teichgraber V, Becker TC, Masopust D, Kaech SM, et al. 2003. Lineage relationship and protective immunity of memory CD8 T cell subsets. Nat. Immunol. 4:225–34 Seder RA, Ahmed R. 2003. Similarities and differences in CD4+ and CD8+ effector and memory T cell generation. Nat. Immunol. 4:835–42 Langenkamp A, Nagata K, Murphy K, Wu L, Lanzavecchia A, Sallusto F. 2003. Kinetics and expression patterns of chemokine receptors in human CD4+ T lymphocytes primed by myeloid or plasmacytoid dendritic cells. Eur. J. Immunol. 33:474–82 Janssen EM, Lemmens EE, Wolfe T, Christen U, von Herrath MG,

12 Feb 2004

16:10

AR

AR210-IY22-25.tex

AR210-IY22-25.sgm

LaTeX2e(2002/01/18)

MEMORY T CELL SUBSETS

109.

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

110.

111.

112.

113.

114.

115.

Schoenberger SP. 2003. CD4+ T cells are required for secondary expansion and memory in CD8+ T lymphocytes. Nature 421:852–56 Sun JC, Bevan MJ. 2003. Defective CD8 T cell memory following acute infection without CD4 T cell help. Science 300:339–42 Shedlock DJ, Shen H. 2003. Requirement for CD4 T cell help in generating functional CD8 T cell memory. Science 300:337–39 Ridge JP, Di Rosa F, Matzinger P. 1998. A conditioned dendritic cell can be a temporal bridge between a CD4+ T-helper and a T-killer cell. Nature 393:474–78 Schoenberger SP, Toes RE, van der Voort EI, Offringa R, Melief CJ. 1998. T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L interactions. Nature 393:480–83 Bennett SR, Carbone FR, Karamalis F, Flavell RA, Miller JF, Heath WR. 1998. Help for cytotoxic-T-cell responses is mediated by CD40 signalling. Nature 393:478–80 Bourgeois C, Rocha B, Tanchot C. 2002. A role for CD40 expression on CD8+ T cells in the generation of CD8+ T cell memory. Science 297:2060–63 Maxwell JR, Campbell JD, Kim CH, Vella

116.

117.

118.

119.

120.

121.

P1: IKH

763

AT. 1999. CD40 activation boosts T cell immunity in vivo by enhancing T cell clonal expansion and delaying peripheral T cell deletion. J. Immunol. 162:2024–34 Maxwell JR, Weinberg A, Prell RA, Vella AT. 2000. Danger and OX40 receptor signaling synergize to enhance memory T cell survival by inhibiting peripheral deletion. J. Immunol. 164:107–12 Hendriks J, Gravestein LA, Tesselaar K, van Lier RA, Schumacher TN, Borst J. 2000. CD27 is required for generation and long-term maintenance of T cell immunity. Nat. Immunol. 1:433–40 Fearon DT, Manders P, Wagner SD. 2001. Arrested differentiation, the selfrenewing memory lymphocyte, and vaccination. Science 293:248–50 Bernasconi NL, Traggiai E, Lanzavecchia A. 2002. Maintenance of serological memory by polyclonal activation of human memory B cells. Science 298:2199– 202 Dudley ME, Rosenberg SA. 2003. Adoptive-cell-transfer therapy for the treatment of patients with cancer. Nat. Rev. Cancer 3:666–75 Welsh RM, Selin LK. 2002. No one is naive: the significance of heterologous Tcell immunity. Nat. Rev. Immunol. 2:417– 26

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:745-763. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104554

Annu. Rev. Immunol. 2004. 22:765–87 doi: 10.1146/annurev.immunol.22.012703.104554 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on December 12, 2003

CONTROL OF T CELL VIABILITY

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Philippa Marrack1 and John Kappler2 Howard Hughes Medical Institute and Integrated Department of Immunology, National Jewish Medical and Research Center, and Departments of Medicine, 1Biochemistry and Molecular Genetics, and 2Pharmacology, University of Colorado Health Sciences Center, Denver, Colorado 80206; email: [email protected]; [email protected]

Key Words death, survival, na¨ıve, activated, memory ■ Abstract The factors affecting T cell viability vary depending on the type and status of the T cell involved. Na¨ıve T cells die via a Bcl-2/Bim dependent route. Their deaths are prevented in animals by IL-7 and contact with MHC. Activated T cells die in many different ways. Among these is a pathway involving signals that come from outside the T cell and affect it via surface receptors such as Fas. Activated T cells also die through a pathway driven by signals generated within the T cell itself, a cell autonomous route. This pathway involves members of the Bcl-2 family, in particular Bcl-2, Bcl-xl, Bim, and probably Bak. The viability of CD8+ and CD4+ memory T cells is controlled in different ways. CD8+ memory T cells are maintained by IL-15 and IL-7. The control of CD4+ memory T cells is more mysterious, with roles reported for IL-7 and/or contact via the TCR.

INTRODUCTION Numerous experiments show that the total number of T cells in the animal is fairly tightly regulated. This is in spite of the fact that, at least in young animals, T cells are continuously being created. Moreover, during infections antigen-specific T cells divide rapidly such that up to 50% of the CD8+ T cells in the animal are responding to that single infectious agent (1). Common sense and experimental evidence therefore indicate that T cells must be disappearing at about the same rate as they are being produced, so that their total numbers remain constant. There are, of course, many different kinds of T cells distinguished by the types of receptors and coreceptors they bear: αβTCR versus γ δ TCR and CD4+ versus CD8+. Each of these broad categories contains subpopulations. For example, some CD4+ αβ TCR-bearing T cells, the NK T cells, bear a special subset of αβ TCRs and also the NK1.1 marker. Different populations of γ δ T cells are designed to react with different antigens and are found in particular parts of the body. To complicate matters further each of these subpopulations probably occurs in various states depending on its prior experience with antigen. Na¨ıve T cells certainly behave differently than their activated counterparts, and likewise the 0732-0582/04/0423-0765$14.00

765

19 Feb 2004

12:10

766

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 1 Summary of major factors that affect T cell viability Status of αβTCR+ T cell

In animals, what makes them:

Na¨ıve

Die? Ratio of Bcl-2/Bim

Live? IL-7 selecting MHC selecting peptide?

Activated

Fas TNFRs CD8+ cytotoxic T cells ratio of Bcl-2/Bim

IL-2 family cytokines type 1 IFNs? Signals from innate immunity NF-κB transcribed genes

CD8+ memory

Ratio of Bcl-2/Bim type 1 interferons

IL-15 IL-7

CD4+ memory

Ratio of Bcl-2/Bim

IL-7 TCR signals?

Question marks within the table indicate conclusions that are less well established in vivo.

rules that govern the life and death of memory T cells are not the same as those that control the existence of other T cells. In addition, not all T cells with the same exposure to antigen are the same. T cells that have been activated with antigen while the innate immune system is responding survive better than those activated with antigen alone. Memory T cells exist in different parts of the body, and the factors that control their survival in lymphoid organs may not be the same as those that are effective outside the lymphatic system. Given these complexities, and the fact that what we really want to understand is how T cell numbers are controlled in that most mysterious of objects, the intact animal, it is not surprising that we do not understand T cell homeostasis and that almost every discovery is controversial. None of these facts allow easy writing of this review. However, fortunately for the authors, relatively little is known about the life and death of some T cell subsets, for example, γ δ T cells, and this review can concentrate on the αβ TCR+ T cells that have received most experimental attention in this regard. The discussion is broken up according to the experience with antigen, or lack of it, of the T cell involved, since it is antigen experience that defines in large part the fate of the T cell. A summary of the conclusions drawn in this review is given in Table 1.

A NOTE ON ASSAYS Ideally we would like to understand how T cells live and die in the intact animal. To a certain extent this is possible in mice. Administration of inhibiting or stimulating antibodies, cytokines, and chemokines, the ability to transfer stem cells and T cells, and the existence of transgenic and knockout mice and retroviruses allow all kinds of experiments to be done in these animals. However, some experiments, even with mouse cells, have to be done in vitro and are therefore subject to artifacts owing to all the changes that occur when cells are cultured at relatively low densities outside

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL SURVIVAL

P1: IKH

767

the body where they are no longer affected by the architecture of the tissue in which they exist, where they exist in different concentrations of potentially important metabolites such as glucose and oxygen, and where the cells are often kept alive by artificial means, i.e., the addition of factors such as IL-2, which may not normally be involved in maintenance of the cells in vivo. Even in animals conditions may not always be as straightforward as they may seem. Experiments are sometimes done in lymphopenic mice, animals in which T cells expand by the process known as homeostatic expansion. Although the factors that govern homeostatic expansion may be the same as those that control survival in the normal animal, this is not necessarily the case, and experiments in which factors that promote survival and/or expansion of T cells in animals that lack other lymphocytes may not necessarily reflect the truth of T cell survival in normal mice. The factors that control T cell survival and expansion in lymphopenic animals are often the same as those that affect the survival of the same cells in lymphocyte-sufficient animals, however, and the former experiments are usually over much more quickly than the latter. Therefore, experiments with mice lacking lymphocytes are often done in lieu of experiments in normal animals. Interpretation of such experiments should also be made with caution, however. As far as human T cells are concerned, of course, techniques to assess the cells in vivo are few and far between. Occasional genetic mutations allow insights about what happens to cells in man, but otherwise ideas must be inferred either from experiments in vitro, or by analogy to experiments done in mice. In this review we consider na¨ıve, activated, and memory T cells separately, and discuss experiments done in vitro and in mice lacking or containing normal populations of lymphocytes, with greatest faith placed in the last of these. Parallel conclusions drawn in man, when possible, are also discussed.

THE VIABILITY OF NA¨IVE T CELLS Effects of Cytokines The fact that the numbers of na¨ıve T cells do not change very much between one individual and another suggests that na¨ıve cells are maintained by some limiting factor(s) for which they compete. Experiments in vitro were the first to indicate that cytokines were among these factors. The survival of both human and mouse na¨ıve T cells is promoted in vitro by IL-7 (2–6), and experiments show that mouse na¨ıve T cells are also maintained by IL-4 (2, 6). Interleukin-6 also helps maintain na¨ıve T cells in vitro, but it acts on a smaller proportion of na¨ıve cells than IL-7 and IL-4 do, for reasons that are not clear (7). Likewise, survival potentiating effects of IL-12 have also been reported (8). Indeed, there may be additional factors made by stromal cells that have yet to be identified (9). Of these cytokines, IL-6 probably does not contribute to the survival of na¨ıve T cells in animals because IL-6-deficient mice contain normal numbers of T cells (10), and T cells harvested from normal animals do not contain high levels of

19 Feb 2004

12:10

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

768

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

phosphorylated Stat 1 and Stat3, the two Stats that are downstream of the IL-6 receptor (7). Likewise, IL-12- and IL-4-deficient animals have normal numbers of T cells, although they lack the ability to convert these cells efficiently to the TH1 or TH2 phenotypes, respectively (11, 12). Moreover, IL-4 is not produced constitutively in animals, and although the cytokine may be retained, once produced, on extracellular matrix, it seems unlikely that a cytokine that is produced in large part only during antigen challenge, and only some antigen challenges at that, would be required for maintenance of na¨ıve T cells. However, in mice IL-4 may sometimes affect T cell survival because in mice deprived of IL-7 (see below), removal of IL-4 as well definitely reduced the numbers of na¨ıve T cells. However, mice lacking the gamma chain that is common to IL-2-related cytokine receptors, or its essential downstream signaling molecule, Jak3, strongly indicated that some IL-2-related protein was involved in na¨ıve T cell survival in animals (13–15). Interleukin 7 is probably the most important of these cytokines in vivo, although some results are confused by the fact that IL-7 or its high-affinity receptor is required for efficient development of T cells bearing αβTCRs (16, 17). The first experiments to show clearly that IL-7 is involved in the maintenance of the na¨ıve T cell population in mice were done by Boursalian & Bottomly (18). These investigators thymectomized adult mice and then treated the animals with anti-IL-7 antibodies. The na¨ıve but not the memory T cells in the treated mice dropped in number. The fall in these cells was more marked if the animals lacked the ability to make IL-4. The investigators concluded that IL-7 was involved in the maintenance of naive T cells in mice, although its activity could be replaced to some extent by IL-4. Similar results with regard to the requirement for IL-7, using anti-IL-7 receptor antibodies, have been reported by others (19). These studies were both subject to the objection that the investigators did not monitor T cell division, and therefore there was a fear that the results were distorted by T cell proliferation, or lack of it. Other investigators have therefore tackled the problem using CFSE to measure T cell division. Out of necessity, many of these experiments have involved lymphopenic mice and interpreted results involving T cell proliferation in such animals as parallels to T cell survival in normal mice. Also, many of these experiments have ensured the na¨ıve status of their T cells by using TCR transgenic T cells. For example, Schluns et al. showed that CD4+ or CD8+ na¨ıve T cells do not divide in animals that cannot make IL-7. They supported their argument that proliferation in T cell–deficient animals is equivalent to survival in normal mice by showing that TCR transgenic T cells need to express the IL-7 receptor in order to survive in T cell–sufficient animals (20). Others have reached similar conclusions (21). How does IL-7 affect the survival of T cells? The fact that IL-4 and IL-7 are active, but other members of the IL-2 family are not, even when they are made available to the cells, suggests that the process involves some intracellular pathway that is stimulated by the receptors for IL-7 and IL-4, but not by those for IL-2, IL-15, etc. Na¨ıve T cells bear high-affinity receptors for IL-4 and IL-7 (22, 23), however, and not for IL-2 or IL-15 (24, 25), so the difference in sensitivity is probably

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL SURVIVAL

P1: IKH

769

the consequence of differential receptor expression rather than a requirement for an intracellular pathway that is not shared by all these cytokines. It is likely that the relevant pathway involves induction of the antiapoptotic protein Bcl-2. Bcl-2 expression is induced by all members of the IL-2 family (26, 27). Survival via increased Bcl-2 is probably not the only positive attribute that IL-7 induces in resting T cells, however. Rathmell et al. are interested in the loss of metabolic activity and reduction in cell size that accompanies culture of na¨ıve T cells in the absence of stimulants, and they have recently shown that induction of Bcl-2 by IL-7 is sufficient to keep the T cells alive, but that maintenance of their metabolic activity requires signaling via phosphatidylinositol 3-kinase and the mammalian target of rapamycin, pathways that are not required for induction of Bcl-2 (6). Thus, not surprisingly, IL-7 induces more than one type of event to help keep T cells alive and functional.

Effects of T Cell Receptor Engagement During the last 10 years a series of papers have reported that engagement of TCRs is involved in keeping na¨ıve T cells alive in animals. On the whole this phenomenon has not been demonstrated in vitro. In lymphopenic animals, however, it is generally agreed that the homeostatic expansion and/or survival of T cells requires recognition of at least the MHC protein that was involved in positive selection of the T cell in question (28–35). However, there is disagreement over whether the need for MHC for homeostatic proliferation indicates that na¨ıve T cells in normal mice have this same requirement. Some investigations suggest that it is (18, 36–38). For example, T cells bearing the AND TCR, specific for IEk + a pigeon/moth cytochrome c peptide, survive best in mice expressing IEk (18). However, in other cases a match between the TCR and its selecting MHC is not needed to keep na¨ıve T cells alive in normal mice (39–42). What causes all this confusion? One possibility is that the effects of MHC have something to do with the affinity/avidity of the TCR on the T cell being tested for its self-MHC/peptide ligand. In support of this idea is the finding that only some transgenic TCRs allow T cells to proliferate in T cell–deficient mice (32). Alternatively, it is possible that homeostatic expansion in T cell–deficient animals is a completely different matter from extended survival in normal mice. In support of this, it has been shown that T cells generated such that, once mature, they lack tyrosine kinases needed to signal via their TCRs (and CD4 or CD8), have a normal life span but cannot proliferate in T cell–deficient hosts (43). Dorfman & Germain (41) have suggested that the discrepancies are due to differences in maturation in the T cells being tested. Some T cells may have emerged from the thymus incompletely mature. These cells would need to complete positive selection in the periphery. Other cells may leave the thymus in a fully mature state, and these cells would not need further exposure to self-MHC. This is a clever idea, supported only by the fact that homeostatic expansion occurs for T cells bearing only some transgenic TCRs (32). Finally, part of the problem may be due to differential needs

19 Feb 2004

12:10

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

770

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

for engagement of molecules such as CD4 or CD8 on the T cells (44, 45). Some, all, or none of these explanations may eventually turn out to be valid. What is the signal delivered by interaction with MHC that helps to keep na¨ıve T cells alive, if indeed it exists? There are two possibilities. One is that the signal is actually delivered by a weak interaction with the TCR itself. The other is that the interaction with MHC is only needed to bring the T cell close to some other cell, perhaps a dendritic cell, and that the survival signal is delivered from this other cell via a receptor on the T cell that is not its TCR. The literature strongly suggests that at least the first of these possibilities is correct. Components of CD3 are phosphorylated in normal T cells in animals, and this phosphorylation disappears rapidly after interaction between the TCRs on the T cells and MHC in the animals is interrupted (46, 47). The downstream molecules affected by this signaling are not known, although some of them may be members of the NF-κB family (48). T cells that are not continuously exposed to self-MHC signal less well through their TCRs when they encounter antigen (47), suggesting that downstream phenomena are induced in the cells by this continuous contact with self, but the nature of these downstream phenomena involved in either survival or function has yet to be discovered.

Why Do Na¨ıve T Cells Die? Another way to approach the question of what keeps na¨ıve T cells alive is to find out what causes their death, with the idea that knowledge of this pathway might help us understand how it is interrupted and controlled. The literature contains descriptions of two broadly defined pathways that drive the death of T cells, involving the engagement of death receptors such as Fas and the receptors for TNFα on the surface of the cells (extrinsic pathways) or that are internal to the cell and involve changes in signaling within the cell itself and are not dependent on extrinsic signals. There is much evidence that external signals via Fas ligand or TNFα are not involved in the death of na¨ıve T cells. Na¨ıve T cells do not express the receptors for these ligands. The survival of normal CD4+ and CD8+ T cells is not affected in animals that lack these proteins or their ligands. On the other hand, na¨ıve T cells certainly do die, both in culture and in animals, so some phenomenon must mediate this turnover. Evidence suggests that na¨ıve cells may actually die via a route that is quite similar to one used by activated T cells, i.e., via a mitochondrially dependent pathway involving Bcl-2 and Bim. Na¨ıve T cells, like activated T cells, are protected from death in vitro by superoxide scavengers such as MnTBAP (49). The half-life in vitro of na¨ıve T cells lacking Bim is dramatically longer than that of normal T cells (50, 51; P. Marrack, personal observation). Bim-deficient or Bcl-2 overexpressing mice accumulate naive T cells (50–52). However, considering that the thymus size and presumably thymus output is quite normal in these animals, it is surprising that these mice do not contain even greater numbers of T cells than they do. Perhaps survival of these cells depends on a balance between other life and death factors. For example, lymphocytes

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

T CELL SURVIVAL

P1: IKH

771

in Bim-deficient animals contain much less Bcl-2 and Bcl-xl than their normal counterparts (51; P. Marrack, personal observation). It could be that these lower levels of antiapoptotic factor allow some death signaling protein to replace Bim and act to control the balance between life and death of these cells. Therefore, probably, in normal mice na¨ıve T cells die because their balance of Bcl-2 (controlled by the availability of IL-7) and Bim drops below acceptable levels.

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Control of Na¨ıve T Cell Numbers In theory numbers of na¨ıve T cells can be controlled by input, from the thymus and other sources, or output, by recruitment into activated and memory T cell pools or by death of the cells. Control by thymus production is not the subject of this review and will not be further discussed here although it is of considerable interest. In animals that are not exposed to massive arrays of infectious agents, or superantigens or lymphopenic, the effects of displacement into other pools of T cells are relatively minor. Thus for the purposes of this review we will here consider only those factors that may affect survival of the cells and hence affect homeostatic numbers of na¨ıve T cells in the animal. It is likely that competition for IL-7 is a major factor that affects the numbers of na¨ıve T cells because animals that express more or less IL-7 contain more or less na¨ıve T cells (53, 54). However, other factors may also be limiting and affect the final numbers. For example, microscopy shows that na¨ıve T cells contact, from time to time, dendritic cells, or some other component in lymph nodes (55, 56). This may be related to the need of na¨ıve T cells for engagement of MHC (see above). One could imagine that na¨ıve T cells might compete for access to antigen presenting cells. In fact such is the case for na¨ıve T cells undergoing homeostatic expansion in lymphopenic animals because administration of syngeneic dendritic cells markedly increases the numbers of T cells that divide rapidly in such hosts (A. de Souza, J. Robinson, J. Bender, J. Kappler, and P. Marrack, unpublished observations). Thus, to the extent that homeostatic expansion is a parallel to normal lymphocyte survival, access to certain types of antigen presenting cells may be a controlling factor for T cell survival in animals.

Some Unsolved Problems about Na¨ıve T Cells There are still many unknowns about the survival of na¨ıve T cells in animals, and they are listed as follows: Why does the same signal (IL-7) induce na¨ıve T cell proliferation in T cell– deficient mice and survival in normal animals? Is this simply a matter of the amount of IL-7 or are other factors involved? Do other cells that detect IL-7 compete with αβTCR+ T cells? B cells and γ δTCR+ T cells both depend on IL-7 for their development and for their maintenance (although the latter is not well established for γ δTCR+ T cells) (17, 57–59). If all these cells compete for IL-7, then loss of one population

19 Feb 2004

12:10

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

772

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

should allow increases in the size of the others. On the whole this has not been observed. B cell–deficient mice do not have markedly more T cells and vice versa (60; P. Marrack, personal observation). Mice lacking αβTCR+ T cells do have an increase in certain populations of γ δTCR+ cells (61), so perhaps some competition between the populations does occur. More likely, however, the populations are to some extent independent because, although they may all compete for IL-7, each population may require some additional factor, which is truly limiting and population specific. Such factors may include self-MHC for αβTCR+ T cells (see discussion above), and BAFF/BlyS/TALL-1/THANK for B cells (62, 63). Are as yet undiscovered factors involved in the survival of the cells? Is self MHC really required for survival of the cells. If so, what changes does MHC evoke in the cell to effect its survival?

THE VIABILITY OF ACTIVATED T CELLS Once they encounter antigen T cells divide rapidly in animals, and then almost as quickly most of their progeny die. The expansion phase is to some extent accompanied by exposure to IL-2 family cytokines. Perhaps these serve to keep dividing T cells alive in the animals because they certainly do so very well in vitro. Good proliferation of antigen-stimulated T cells also requires engagement of CD28 on their surfaces (64, 65). CD28 ligation induces Bcl-xl production (27, 66) so expanding T cells may also be protected from death by higher levels of Bcl-xl induced by this route. It has been difficult to study activation-induced death of T cells in culture because on the whole T cells that have been activated (via anti-TCR, for example) in vitro do not show the same tendency to die as their in vivo counterparts. Several factors may account for the difference. In vitro cells are in less intimate contact with each other, so cell-to-cell interactions that trigger death (see below) may be less frequent in culture than in animals. In vitro stimulated T cells make lots of IL-2, an antiapoptotic factor, whereas Il-2 is produced only transiently in vivo. In fact, turnover and regulation of all the IL-2 family of cytokines may be less profound in culture than in mice. All these points have led most individuals who are interested in the maintenance and death of activated T cells to focus on the behavior of the cells in animals. The factors that cause the rapid death of activated T cells have been the subject of extensive studies for many years. Many mechanisms have been proposed; however, it is likely that the phenomenon is dominated by two major routes.

Signaling Death Via Surface Receptors Some time ago two groups discovered Fas (67, 68), a protein that appears on the surfaces of activated T cells. Subsequently many experiments have shown that activated T cells die less readily if they lack Fas, or their hosts lack Fas ligand

15 Mar 2004

15:43

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL SURVIVAL

P1: IKH

773

(69–71). Other experiments have implicated another pair of members of the Fas/Fas ligand family, TNF and the p50 TNF receptor, as well as Fas in T cell death (72). In many ways the idea that the death of activated T cells is caused by Fas or Fas relatives is attractive. Activated but not resting T cells express Fas and TNF receptors (73, 74). These proteins certainly activate a death pathway, via caspases, within the cell (75–77). Other cells in the body certainly express the ligands for Fas and TNF receptors (78). Therefore there is at first thought no reason why these proteins should not be involved in activated T cell death (Figure 1). There are, however, some problems with the idea. Animals and human beings lacking Fas or its ligand do not accumulate activated CD4+ or CD8+ T cells, or their progeny, memory T cells bearing CD4 or CD8. Rather, these individuals acquire large numbers of CD4- CD8- B220+ T cells (79, 80). The extent to which these cells appear in mice depends on the strain in which the Fas/Fas ligand deficiency is expressed. On the MRL background many double negative cells appear. In C57BL/6 mice, however, the deficiencies allow accumulation of relatively few such cells (81). These cells may be derived from autoreactive T cells that originally bore CD4 or CD8 and were not killed in these death receptor–deficient mice (82, 83). However, this conclusion is not certain, and the double negative cells that appear in Fas-deficient animals may have other provenances (84–86). There is also the issue of FLIP expression. To varying degrees activated T cells express FLIP and this protein protects the cells against Fas-mediated death (87, 88), thus the Fas that is expressed on activated T cells may not always be able to kill them.

Signaling Death Via Cell Autonomous Pathways We and others have suggested that many activated T cells die via a pathway that does not involve signals from other cells. Rather, the pathway is intrinsic to the activated T cell itself and is automatically set into effect by activation of the cell (89). Evidence suggests that this pathway involves members of the Bcl-2 family and is very focused on the mitochondrion. The Bcl-2-related proteins fall into three subfamilies (90, 91). One of these, typified by Bcl-2 and Bcl-xl, is antiapoptotic and includes all four of the short peptide regions, BH1–4, that characterize the family. Another subfamily, composed in T cells of Bax and Bak, are the actual killers of the cell (92, 93) and include only three of the four homology domains, namely BH1–3. A third group of these proteins, which contain only BH3 domains, are thought to be messengers of death, located at various organelles and key checkpoints throughout the cell and its metabolism, to monitor the health of these various components (94). It is thought that failure in one of these components, for example, failure of Akt to maintain the BH3 protein, BAD, in its harmless state, unleashes the messenger of death that then migrates to the mitochondria and perhaps elsewhere where it directly or indirectly induces the activity of one of the executioners and precipitates cell death (95, 96). It is not known exactly how death is induced. However, it is likely that the process involves a disturbance in the mitochondrial membranes, perhaps formation

15 Mar 2004

15:43

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

774

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

of a pore, by Bax or Bak, although many other mechanisms have been proposed (97–103). Experiments by ourselves and others strongly suggest that a pathway involving Bcl-2 proteins is often involved in activated T cell death. Overexpression of Bcl-2 increases T cell responses in vivo, perhaps by protecting activated T cells against death (52). However, there is a difference of opinion on this point (104), so perhaps the protection is only transient (see below). T cells lacking the messenger of death, Bim, are also protected from death (50, 51, 94, 105). T cells lacking both Bak and Bax likewise do not die (92, 93), suggesting that either of these two executioners can kill the cell. In vivo, at the time activated T cells approach the point at which they will begin to die, their levels of Bcl-2 and Bcl-xl fall (51, 106–108). In some cell types Bim is usually held away from mitochondria on microtubules, via high-affinity binding to a component of the dynein motor complex (109). In healthy T cells, however, Bim is not on microtubules. Rather, it is always located on mitochondria bound to Bcl-2 and Bcl-xl (Y. Zhu, B.J. Swanson, M. Wang, D.A. Hildeman, B.C. Schaefer, X. Liu, H. Suzuki, K. Mihara, J. Kappler, and P. Marrack, unpublished data). In activated T cells, levels of Bim may rise transiently by a small amount. This, together with the fact that Bcl-2 levels fall, leads to a reduction in the proportion of the Bim that can be complexed with Bcl-2. Levels of the Bim/Bcl-xl complex rise, but levels of free Bim may also increase (Y. Zhu, B.J. Swanson, M. Wang, D.A. Hildeman, B.C. Schaefer, X. Liu, H. Suzuki, K. Mihara, J. Kappler, and P. Marrack, unpublished data). Directly or indirectly this leads to loss of mitochondrial integrity. Diagrams of this sequence of events are shown in Figures 2A,B,C. Of the executioners, we believe that Bak and not Bax is the crucial entity because Bak changes its shape to the active, aggregating configuration well before the activated T cells die, and well before Bax does (Y. Zhu, J. Kappler, and P. Marrack, unpublished data). We have recently solved the structure of an extended BH3 peptide from Bim bound to Bcl-xl (109a). Transposition of the BH3 Bim peptide into the known structure of Bax indicates that in order for the Bim peptide to bind Bax it would have to profoundly distort the Bax protein and fix it in the configuration known to destroy mitochondrial function and kill the cell. Our experimental data confirm this idea. Because Bax and Bak are very similar we believe that engagement of the extended Bim BH3 domain might catalyze a similar aggregation of Bak, as diagrammed in Figure 3. Interestingly, the three Bcl-2-related proteins, Bcl-2, Bim, and Bak, that appear to be most intimately involved in activated T cell death are always located on mitochondria, suggesting that the death of activated T cells via this method is very mitochondria-centric.

What Saves Activated T Cells from Death? Even under the most drastic conditions not all antigen-activated T cells die in animals. It is also clear that the conditions under which T cells encounter antigen

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL SURVIVAL

P1: IKH

775

affect their propensity to die. A smaller proportion of the antigen-specific T cells die if antigen is given to the animals with adjuvant (107, 110, 111), a phenomenon upon which all successful vaccines may depend. Adjuvant can be replaced by inflammatory cytokines such as TNFα (perversely, given its potential role in T cell death, see above), interferons, IL-1, and so on. It is not known whether any of these cytokines act directly or via antigen presenting cells. The idea that antigen presenting cells may have some role to play here is supported by the observation that anti-CD40, which presumably acts by binding to and stimulating dendritic cells, also reduces the numbers of activated T cells that die after an immune response (112–114). Likewise, antibodies that could substitute by binding to receptors on T cells for many of the ligands that appear on activated dendritic cells, anti-Ox40, anti-4-1BB ligand, or anti-CD27 all also help to keep target T cells alive (115–121). Ox40, 4-1BB ligand, and CD27 are all related to TNF receptors. They lack death domains in their intracellular tails but all act via TRAFs, usually TRAF2 and/or TRAF 5, to activate NF-κB family members for transcriptional activity (121–125). NF-κB and its relatives have been implicated in improved survival of many cells, so these findings have been taken together to suggest that the products of genes induced by NF-κB act to preserve activated T cells. In a search for such genes we recently compared gene expression in T cells that had been activated with superantigen in vivo and the absence and presence of adjuvants, LPS + anti-CD40, or vaccinia virus infection. Both of these sets of adjuvants profoundly improve the survival of superantigen activated T cells in mice. Our comparison came up with gratifyingly few genes. Among these was Bcl-3 (126), a protein that is related to IκB and that binds to NF-κB (127–129). However, instead of inhibiting NF-κB activity by retaining the transcription factor in the cytoplasm, Bcl-3 binds NF-κB in the nucleus and acts as a transcriptional transactivater (130). To study this phenomenon further we have examined activated T cell survival in Bcl-3-deficient and Bcl-3 transgenic mice. Adjuvants such as LPS continue to reduce activated T cell death in Bcl-3-deficient mice. However, the rates of death of both resting and activated T cells are drastically reduced in Bcl-3 transgenic mice (J. White, M. Bassetti, J. Kappler, and P. Marrack, unpublished observations). Thus Bcl-3 is sufficient but not necessary for rescue of activated T cells, indicating that there are probably several routes that can contribute to this important biological function. Further experiments are in progress to find out how Bcl-3 and other adjuvantinduced genes protect activated T cells from death. Bcl-3 has been reported to induce Bcl-2 in some cell lines (131). However, Bcl-2 mRNA was not induced on our T cell adjuvant chips (126). Likewise, the mRNAs for some of the proteins we think are involved in the death of T cells, such as those for Bim, Bax, and Bak, were not reduced in T cells exposed to adjuvant. There were interesting increases in the mRNAs for some other proteins that might help the mitochondria in T cells defend themselves against attack. These included mRNAs for the voltagedependent anion channels and adenine nucleotide transporters that are involved in movement of materials in and out of mitochondria. Interestingly, both these sets

19 Feb 2004

12:10

776

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

of proteins have been implicated in Bcl-2-related malfunction of mitochondria, and both have been shown to bind to the Bcl-2-related executioners, Bax and Bak (99, 101, 132). Perhaps T cells defend themselves against death by increasing their content of the proteins that might be inactivated by Bax and Bak.

Some Unsolved Problems about Activated T Cells

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

There are of course many unknowns about the life and death of activated cells. Problems of particular interest in the context of this article include the following: What determines which cells will live or die? Is it a matter of location, predisposition or chance? What receptors on activated T cells detect signals from oinnate immunity to promote survival of the cells? What proteins does the innate immune system affect inside T cells to promote the survival of the cells? Can this knowledge be harnessed to improve vaccines and/or treatment of autoimmune diseases?

THE VIABILITY OF MEMORY T CELLS The problem of immunological memory has been a hotly contested field for many years. Explanations for the fact that animals have immunological memory, in some cases for their lifetimes, range from the idea that, once exposed, the animal is never completely free of antigen, to the notion that memory T cells are extremely long-lived. The real truth probably lies between these two extremes. One of the difficulties in understanding memory T cells is that, by their very nature, it is hard to study such cells in vitro. For example, the memory T cells that do not divide but simply survive for a long time will be very hard to distinguish in vitro from some other kind of cell that has simply acquired the ability to withstand the harsh conditions of a culture dish.

The Viability of CD8+ Memory T Cells EFFECTS OF CYTOKINES There has recently been some progress in this regard for CD8+ αβTCR+ T cells, the category of T cells whose memory population is best understood. In this case, however, success in vitro depended on in vivo experiments that indicated which factors might be needed for memory cell survival. Some years ago Tough & Sprent showed that a population of memory T cells in mice was dividing slowly and that this slow rate of division could be increased by increases in type I interferons. Later the same group realized that the type I interferons were not acting directly on the memory cells but rather via induction of IL-15 (133– 135). Subsequently, other groups showed that CD8+ memory T cells are in a state of constant but slow division in animals (136–138). This division is driven by

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL SURVIVAL

P1: IKH

777

IL-15 and, to a lesser extent, IL-7 (20, 136–140), two cytokines that are produced constitutively in animals (141, 142). During inflammation IL-15 production by macrophages increases (135, 141–143), although this has been difficult to establish because this cytokine has been notoriously difficult to assay in vivo. Perhaps this is because of another unexpected property of IL-5, the fact that it can bind with high affinity to the alpha chain of its receptor and thus be retained on IL-15Rα+ cells, where it can then engage IL-5Rβγ receptors on other cells (144). Additional support for the idea that IL-15 is the crucial factor for CD8+ memory T cell survival in animals comes from mice that are deficient in IL-15 or the IL-15specific chain (IL-15Ra) of its receptor (145, 146). These mice have low numbers of CD8+ memory T cells. Also, the realization that IL-15 keeps CD8+ memory T cells alive in animals allows creation of similar cells in vitro, by growth in IL-15. Such cells behave like central CD8+ memory T cells (147) in every respect when transferred back in to animals (148). Culture in IL-2 creates T cells that migrate to nonlymphoid organs (148). These cells may be equivalent to peripheral memory cells (147); however, IL-2 is probably not a supporting hormone for CD8+ memory T cells in animals (see below), so this attribution may not be appropriate. EFFECTS OF TCR ENGAGEMENT It is clear that CD8+ memory T cells do not rely on class I MHC interactions to keep themselves alive in animals (136). However, retained antigen and/or reexposure to antigen may help keep up the numbers of antigen-specific CD8+ memory cells in certain cases.

The Viability of CD4+ Memory T Cells Maintenance of CD4+ memory T cells has been much more difficult to study, and there is still considerable confusion about this problem. This is probably because, in mice, two populations of such cells exist, one dividing fairly rapidly, perhaps in response to environmental or self-antigens, and another that does not divide at all (149, 150). These two populations probably have different requirements to keep themselves alive and at constant levels. The rapidly dividing population should require TCR engagement and therefore class II exposure and signaling via the TCR. The nondividing population requires other factors, probably IL-7 (149–152).

Why Do Memory T Cells Die? Very little is known about this feature of memory T cells. One might predict that animals lacking the major death routes for activated T cells, Fas and/or Bim, would accumulate memory T cells in large numbers. On the whole, however, this does not happen. A population of T cells does increase in Fas-deficient mice and human beings. However, it lacks both CD4 and CD8, and its origin is still debated (79, 80; see above). Bim-deficient mice do contain more lymphocytes than their wildtype litter mates. However, the different types of T cells accumulate proportionately with no particularly marked-over accumulation of T cells bearing memory markers

19 Feb 2004

12:10

778

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

(M. Wang, P. Oliver, unpublished data). We believe this is because the loss of the death-dealing protein, Bim, is offset by a drop in levels of antiapoptotic proteins in the cells Bcl-2 and Bcl-xl, leaving the cells as open to death as their Bim+/+ counterparts.

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

How Are Memory T Cells Controlled? Memory cells probably control themselves to a large extent by competing with each other for their survival factors. This idea is suggested by elegant experiments in which infection by a series of different viruses established a series of memory T cell populations specific for each virus and experiments in which each new memory population reduced the numbers of memory cells for prior infections proportionately (153). Memory populations are controlled in other ways as well. For example, removal of IL-2 from animals has the unexpected consequence of dramatically increasing the numbers of memory T cells (137). Type 1 interferons, while perhaps promoting the survival of actively responding cells (154), reduce the numbers of memory T cells for previous infections (155). Other cell types consume IL-15 and IL-7, so removal of these competing populations allows greater numbers of memory αβTCR+ T cells to appear (20). Greatest and most important of all is the problem of how activated T cells are spared from death and selected to become memory T cells. As discussed above, activation of innate immunity has a big impact on this event. Evidence suggests that the process involves a slow conversion of the survivors of activation induced cell death into properly differentiated memory cells, probably under the influence of IL-7 (20, 156).

Some Unsolved Problems about Memory T Cells Why do memory T cells die? How are memory T cell numbers affected by other cells in the body that may use the same molecules to promote their survival? Are all CD8+ or CD4+ memory cells equivalent?

CONCLUSION The past 10 years of work on the viability of different kinds of T cells in animals has been remarkably productive and satisfying. Each type of T cell follows its own rules, but there are enough similarities between the different kinds of cells to indicate some generalities. Many of the cells depend on some member of the IL-2 family of cytokines for their survival. However, T cells with different statuses rely most heavily on different IL-2 relatives. It would be extremely interesting to understand the evolution of this complex state of affairs. For example, one could imagine that the primitive immune system had only one IL-2-like cytokine, resembling to a great degree, IL-7. Then the diaspora of lymphocyte types led to a need for diversification of the IL-7 ancestor, causing the appearance of the many members of this family that exist today.

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

T CELL SURVIVAL

P1: IKH

779

ACKNOWLEDGMENTS This work was supported by USPHS grants AI-17,134, A-18,785, AI-52,225, and AI-22,295. The Annual Review of Immunology is online at http://immunol.annualreviews.org

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

LITERATURE CITED 1. Murali-Krishna K, Altman JD, Suresh M, Sourdive DJ, Zajac AJ, et al. 1998. Counting antigen-specific CD8 T cells: a reevaluation of bystander activation during viral infection. Immunity 8:177–87 2. Vella A, Teague TK, Ihle J, Kappler J, Marrack P. 1997. Interleukin 4 (IL-4) or IL-7 prevents the death of resting T cells: stat6 is probably not required for the effect of IL-4. J. Exp. Med. 186:325–30 3. Hassan J, Reen DJ. 1998. IL-7 promotes the survival and maturation but not differentiation of human postthymic CD4+ T cells. Eur. J. Immunol. 28:3057–65 4. Soares MV, Borthwick NJ, Maini MK, Janossy G, Salmon M, Akbar AN. 1998. IL-7-dependent extrathymic expansion of CD45RA+ T cells enables preservation of a naive repertoire. J. Immunol. 161:5909–17 5. Webb LM, Foxwell BM, Feldmann M. 1999. Putative role for interleukin-7 in the maintenance of the recirculating naive CD4+ T-cell pool. Immunology 98:400–5 6. Rathmell JC, Farkash EA, Gao W, Thompson CB. 2001. IL-7 enhances the survival and maintains the size of naive T cells. J. Immunol. 167:6869–76 7. Teague TK, Schaefer BC, Hildeman D, Bender J, Mitchell T, et al. 2000. Activation-induced inhibition of interleukin 6-mediated T cell survival and signal transducer and activator of transcription 1 signaling. J. Exp. Med. 191:915– 26 8. Ben-Sasson SZ, Makedonski K, Hu-Li J, Paul WE. 2000. Survival and cytokine

9.

10.

11.

12.

13.

14.

15.

16.

polarization of naive CD4(+) T cells in vitro is largely dependent on exogenous cytokines. Eur. J. Immunol. 30:1308–17 Zhou YW, Aritake S, Tri Endharti A, Wu J, Hayakawa A, et al. 2003. Murine lymph node-derived stromal cells effectively support survival but induce no activation/proliferation of peripheral resting T cells in vitro. Immunology 109:496– 503 Kopf M, Baumann H, Freer G, Freudenberg M, Lamers M, et al. 1994. Impaired immune and acute-phase responses in interleukin-6-deficient mice. Nature 368:339–42 Magram J, Connaughton SE, Warrier RR, Carvajal DM, Wu CY, et al. 1996. IL-12-deficient mice are defective in IFN gamma production and type 1 cytokine responses. Immunity 4:471–81 Kuhn R, Rajewsky K, Muller W. 1991. Generation and analysis of interleukin-4 deficient mice. Science 254:707–10 DiSanto JP, Guy-Grand D, Fisher A, Tarakhovsky A. 1996. Critical role for the common cytokine receptor gamma chain in intrathymic and peripheral T cell selection. J. Exp. Med. 183:1111–18 Thomis DC, Berg LJ. 1997. Peripheral expression of Jak3 is required to maintain T lymphocyte function. J. Exp. Med. 185:197–206 Nakajima H, Shores EW, Noguchi M, Leonard WJ. 1997. The common cytokine receptor gamma chain plays an essential role in regulating lymphoid homeostasis. J. Exp. Med. 185:189–95 Peschon JJ, Morrissey PJ, Grabstein KH, Ramsdell FJ, Maraskovsky E,

19 Feb 2004

12:10

780

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

17.

18.

19.

20.

21.

22.

23.

24.

25.

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

et al. 1994. Early lymphocyte expansion is severely impaired in interleukin 7 receptor-deficient mice. J. Exp. Med. 180:1955–60 von Freeden-Jeffry U, Vieira P, Lucian LA, McNeil T, Burdach SE, Murray R. 1995. Lymphopenia in interleukin (IL)7 gene-deleted mice identifies IL-7 as a nonredundant cytokine. J. Exp. Med. 181:1519–26 Boursalian TE, Bottomly K. 1999. Survival of naive CD4 T cells: roles of restricting versus selecting MHC class II and cytokine milieu. J. Immunol. 162:3795–801 Vivien L, Benoist C, Mathis D. 2001. T lymphocytes need IL-7 but not IL-4 or IL-6 to survive in vivo. Int. Immunol. 13:763–68 Schluns KS, Kieper WC, Jameson SC, Lefrancois L. 2000. Interleukin-7 mediates the homeostasis of naive and memory CD8 T cells in vivo. Nat. Immunol. 1:426–32 Tan JT, Dudl E, LeRoy E, Murray R, Sprent J, et al. 2001. IL-7 is critical for homeostatic proliferation and survival of naive T cells. Proc. Natl. Acad. Sci. USA 98:8732–37 Barner M, Mohrs M, Brombacher F, Kopf M. 1998. Differences between IL4R alpha-deficient and IL-4-deficient mice reveal a role for IL-13 in the regulation of Th2 responses. Curr. Biol. 8:669– 72 Webb LM, Foxwell BM, Feldmann M. 1997. Interleukin-7 activates human naive CD4+ cells and primes for interleukin-4 production. Eur. J. Immunol. 27:633–40 Vuillier F, Scott-Algara D, Dighiero G. 1991. Extensive analysis of lymphocyte subsets in normal subjects by threecolor immunofluorescence. Nouv. Rev. Fr. Hematol. 33:31–38 Kanegane H, Tosato G. 1996. Activation of naive and memory T cells by interleukin-15. Blood 88:230–35

26. Akbar AN, Borthwick NJ, Wickremasinghe RG, Panayoitidis P, Pilling D, et al. 1996. Interleukin-2 receptor common gamma-chain signaling cytokines regulate activated T cell apoptosis in response to growth factor withdrawal: selective induction of anti-apoptotic (bcl-2, bcl-xL) but not pro-apoptotic (bax, bclxS) gene expression. Eur. J. Immunol. 26:294–99 27. Mueller DL, Seiffert S, Fang W, Behrens TW. 1996. Differential regulation of bcl2 and bcl-x by CD3, CD28, and the IL-2 receptor in cloned CD4+ helper T cells. A model for the long-term survival of memory cells. J. Immunol. 156:1764– 71 28. Takeda S, Rodewald HR, Arakawa H, Bluethmann H, Shimizu T. 1996. MHC class II molecules are not required for survival of newly generated CD4+ T cells, but affect their long-term life span. Immunity 5:217–28 29. Tanchot C, Lemonnier FA, Perarnau B, Freitas AA, Rocha B. 1997. Differential requirements for survival and proliferation of CD8 naive or memory T cells. Science 276:2057–62 30. Kirberg J, Berns A, von Boehmer H. 1997. Peripheral T cell survival requires continual ligation of the T cell receptor to major histocompatibility complex-encoded molecules. J. Exp. Med. 186:1269–75 31. Beutner U, MacDonald HR. 1998. TCRMHC class II interaction is required for peripheral expansion of CD4 cells in a T cell-deficient host. Int. Immunol. 10:305–10 32. Ernst B, Lee DS, Chang JM, Sprent J, Surh CD. 1999. The peptide ligands mediating positive selection in the thymus control T cell survival and homeostatic proliferation in the periphery. Immunity 11:173–81 33. Kieper WC, Jameson SC. 1999. Homeostatic expansion and phenotypic conversion of naive T cells in response to self

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

T CELL SURVIVAL

34.

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

35.

36.

37.

38.

39.

40.

41.

42.

43.

peptide/MHC ligands. Proc. Natl. Acad. Sci. USA 96:13306–11 Bender J, Mitchell T, Kappler J, Marrack P. 1999. CD4+ T cell division in irradiated mice requires peptides distinct from those responsible for thymic selection. J. Exp. Med. 190:367–74 Goldrath AW, Bevan MJ. 1999. Lowaffinity ligands for the TCR drive proliferation of mature CD8+ T cells in lymphopenic hosts. Immunity 11:183– 90 Nesic D, Vukmanovic S. 1998. MHC class I is required for peripheral accumulation of CD8+ thymic emigrants. J. Immunol. 160:3705–12 Brocker T. 1999. The role of dendritic cells in T cell selection and survival. J. Leukoc. Biol. 66:331–35 Muranski P, Chmielowski B, Ignatowicz L. 2000. Mature CD4+ T cells perceive a positively selecting class II MHC/peptide complex in the periphery. J. Immunol. 164:3087–94 Clarke SR, Rudensky AY. 2000. Survival and homeostatic proliferation of naive peripheral CD4+ T cells in the absence of self peptide: MHC complexes. J. Immunol. 165:2458–64 Dorfman JR, Stefanova I, Yasutomo K, Germain RN. 2000. CD4+ T cell survival is not directly linked to self-MHCinduced TCR signaling. Nat. Immunol. 1:329–35 Dorfman JR, Germain RN. 2002. MHCdependent survival of naive T cells? A complicated answer to a simple question. Microb. Infect. 4:547–54 Rodriguez-Barbosa JI, Zhao Y, Zhao G, Ezquerra A, Sykes M. 2002. Murine CD4 T cells selected in a highly disparate xenogeneic porcine thymus graft do not show rapid decay in the absence of selecting MHC in the periphery. J. Immunol. 169:6697–710 Seddon B, Legname G, Tomlinson P, Zamoyska R. 2000. Long-term survival but impaired homeostatic proliferation

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

P1: IKH

781

of naive T cells in the absence of p56lck. Science 290:127–31 Konig R, Shen X, Maroto R, Denning TL. 2002. The role of CD4 in regulating homeostasis of T helper cells. Immunol. Res. 25:115–30 Shen X, Konig R. 2001. Post-thymic selection of peripheral CD4+ Tlymphocytes on class II major histocompatibility antigen-bearing cells. Cell Mol. Biol. (Noisy-le-grand) 47:87–96 Seddon B, Zamoyska R. 2002. TCR signals mediated by Src family kinases are essential for the survival of naive T cells. J. Immunol. 169:2997–3005 Stefanova I, Dorfman JR, Germain RN. 2002. Self-recognition promotes the foreign antigen sensitivity of naive T lymphocytes. Nature 420:429–34 Zheng Y, Vig M, Lyons J, Van Parijs L, Beg AA. 2003. Combined deficiency of p50 and cRel in CD4+ T cells reveals an essential requirement for nuclear factor kappaB in regulating mature T cell survival and in vivo function. J. Exp. Med. 197:861–74 Hildeman DA, Mitchell T, Teague TK, Henson P, Day BJ, et al. 1999. Reactive oxygen species regulate activationinduced T cell apoptosis. Immunity 10:735–44 Bouillet P, Metcalf D, Huang DC, Tarlinton DM, Kay TW, et al. 1999. Proapoptotic Bcl-2 relative Bim required for certain apoptotic responses, leukocyte homeostasis, and to preclude autoimmunity. Science 286:1735–38 Hildeman DA, Zhu Y, Mitchell TC, Bouillet P, Strasser A, et al. 2002. Activated T cell death in vivo mediated by proapoptotic bcl-2 family member bim. Immunity 16:759–67 Strasser A, Harris AW, Cory S. 1991. bcl-2 transgene inhibits T cell death and perturbs thymic self-censorship. Cell 67:889–99 Mertsching E, Burdet C, Ceredig R. 1995. IL-7 transgenic mice: analysis of

19 Feb 2004

12:10

782

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

54.

55.

56.

57.

58.

59.

60.

61.

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

the role of IL-7 in the differentiation of thymocytes in vivo and in vitro. Int. Immunol. 7:401–14 Williams IR, Rawson EA, Manning L, Karaoli T, Rich BE, Kupper TS. 1997. IL-7 overexpression in transgenic mouse keratinocytes causes a lymphoproliferative skin disease dominated by intermediate TCR cells: evidence for a hierarchy in IL-7 responsiveness among cutaneous T cells. J. Immunol. 159:3044–56 Ingulli E, Mondino A, Khoruts A, Jenkins MK. 1997. In vivo detection of dendritic cell antigen presentation to CD4(+) T cells. J. Exp. Med. 185:2133– 41 Miller MJ, Wei SH, Cahalan MD, Parker I. 2003. Autonomous T cell trafficking examined in vivo with intravital twophoton microscopy. Proc. Natl. Acad. Sci. USA 100:2604–9 Namen AE, Lupton S, Hjerrild K, Wignall J, Mochizuki DY, et al. 1988. Stimulation of B-cell progenitors by cloned murine interleukin-7. Nature 333:571– 73 Hayashi S, Kunisada T, Ogawa M, Sudo T, Kodama H, et al. 1990. Stepwise progression of B lineage differentiation supported by interleukin 7 and other stromal cell molecules. J. Exp. Med. 171:1683– 95 Tomana M, Ideyama S, Iwai K, Gyotoku J, Germeraad WT, et al. 1993. Involvement of IL-7 in the development of gamma delta T cells in the thymus. Thymus 21:141–57 Kitamura D, Roes J, Kuhn R, Rajewsky K. 1991. A B cell-deficient mouse by targeted disruption of the membrane exon of the immunoglobulin mu chain gene. Nature 350:423–26 Lahn M, Kanehiro A, Takeda K, Terry J, Hahn YS, et al. 2002. MHC class I-dependent Vgamma4+ pulmonary T cells regulate alpha beta T cell-independent airway responsiveness. Proc. Natl. Acad. Sci. USA 99:8850–55

62. Shu HB, Johnson H. 2000. B cell maturation protein is a receptor for the tumor necrosis factor family member TALL1. Proc. Natl. Acad. Sci. USA 97:9156– 61 63. Batten M, Groom J, Cachero TG, Qian F, Schneider P, et al. 2000. BAFF mediates survival of peripheral immature B lymphocytes. J. Exp. Med. 192:1453–66 64. Sharpe AH. 1995. Analysis of lymphocyte costimulation in vivo using transgenic and ‘knockout’ mice. Curr. Opin. Immunol. 7:389–95 65. Noel PJ, Boise LH, Green JM, Thompson CB. 1996. CD28 costimulation prevents cell death during primary T cell activation. J. Immunol. 157:636–42 66. Boise LH, Minn AJ, Noel PJ, June CH, Accavitti MA, et al. 1995. CD28 costimulation can promote T cell survival by enhancing the expression of Bcl-XL. Immunity 3:87–98 67. Trauth BC, Klas C, Peters AM, Matzku S, Moller P, et al. 1989. Monoclonal antibody-mediated tumor regression by induction of apoptosis. Science 245:301– 5 68. Itoh N, Yonehara S, Ishii A, Yonehara M, Mizushima S, et al. 1991. The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis. Cell 66:233–43 69. Alderson MR, Tough TW, Davis-Smith T, Braddy S, Falk B, et al. 1995. Fas ligand mediates activation-induced cell death in human T lymphocytes. J. Exp. Med. 181:71–77 70. Mogil RJ, Radvanyi L, GonzalezQuintial R, Miller R, Mills G, et al. 1995. Fas (CD95) participates in peripheral T cell deletion and associated apoptosis in vivo. Int. Immunol. 7:1451–58 71. Lynch DH, Ramsdell F, Alderson MR. 1995. Fas and FasL in the homeostatic regulation of immune responses. Immunol. Today 16:569–74 72. Sytwu HK, Liblau RS, McDevitt HO. 1996. The roles of Fas/APO-1 (CD95)

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

T CELL SURVIVAL

73.

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

74.

75.

76.

77.

78.

79.

80.

81.

82.

and TNF in antigen-induced programmed cell death in T cell receptor transgenic mice. Immunity 5:17–30 Miyawaki T, Uehara T, Nibu R, Tsuji T, Yachie A, et al. 1992. Differential expression of apoptosis-related Fas antigen on lymphocyte subpopulations in human peripheral blood. J. Immunol. 149:3753– 58 Ryffel B, Mihatsch MJ. 1993. TNF receptor distribution in human tissues. Int. Rev. Exp. Pathol. 34(Pt. B):149–56 Cohen GM. 1997. Caspases: the executioners of apoptosis. Biochem. J. 326 (Pt. 1):1–16 Peter ME, Krammer PH. 1998. Mechanisms of CD95 (APO-1/Fas)-mediated apoptosis. Curr. Opin. Immunol. 10:545– 51 Lenardo M, Chan KM, Hornung F, McFarland H, Siegel R, et al. 1999. Mature T lymphocyte apoptosis–immune regulation in a dynamic and unpredictable antigenic environment. Annu. Rev. Immunol. 17:221–53 Kavurma MM, Khachigian LM. 2003. Signaling and transcriptional control of Fas ligand gene expression. Cell Death Differ. 10:36–44 Groghan TW, Davignon JL, Evans J, Allison JP, Eisenberg RA, et al. 1989. Diminished expression of the T cell receptor on the expanded lymphocyte population in MRL/Mp-lpr/lpr mice. Autoimmunity 2:97–111 Fisher GH, Rosenberg FJ, Straus SE, Dale JK, Middleton LA, et al. 1995. Dominant interfering Fas gene mutations impair apoptosis in a human autoimmune lymphoproliferative syndrome. Cell 81:935–46 Kakkanaiah VN, Pyle RH, Nagarkatti M, Nagarkatti PS. 1990. Evidence for major alterations in the thymocyte subpopulations in murine models of autoimmune diseases. J. Autoimmun. 3:271–88 Sainis K, Datta SK. 1988. CD4+ T cell lines with selective patterns of au-

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

P1: IKH

783

toreactivity as well as CD4− CD8− T helper cell lines augment the production of idiotypes shared by pathogenic antiDNA autoantibodies in the NZB x SWR model of lupus nephritis. J. Immunol. 140:2215–24 Singer PA, Theofilopoulos AN. 1990. Novel origin of lpr and gld cells and possible implications in autoimmunity. J. Autoimmun. 3:123–35 Herron LR, Eisenberg RA, Roper E, Kakkanaiah VN, Cohen PL, Kotzin BL. 1993. Selection of the T cell receptor repertoire in Lpr mice. J. Immunol. 151:3450–59 Giese T, Davidson WF. 1995. In CD8+ T cell-deficient lpr/lpr mice, CD4+ B220+ and CD4+B220− T cells replace B220+ double-negative T cells as the predominant populations in enlarged lymph nodes. J. Immunol. 154:4986–95 Pestano GA, Zhou Y, Trimble LA, Daley J, Weber GF, Cantor H. 1999. Inactivation of misselected CD8 T cells by CD8 gene methylation and cell death. Science 284:1187–91 Irmler M, Thome M, Hahne M, Schneider P, Hofmann K, et al. 1997. Inhibition of death receptor signals by cellular FLIP. Nature 388:190–95 Refaeli Y, Van Parijs L, London CA, Tschopp J, Abbas AK. 1998. Biochemical mechanisms of IL-2-regulated Fasmediated T cell apoptosis. Immunity 8:615–23 Hildeman DA, Mitchell T, Kappler J, Marrack P. 2003. T cell apoptosis and reactive oxygen species. J. Clin. Invest. 111:575–81 Adams JM, Cory S. 1998. The Bcl-2 protein family: arbiters of cell survival. Science 281:1322–26 Opferman JT, Korsmeyer SJ. 2003. Apoptosis in the development and maintenance of the immune system. Nat. Immunol. 4:410–15 Lindsten T, Ross AJ, King A, Zong WX, Rathmell JC, et al. 2000. The combined

19 Feb 2004

12:10

784

93.

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

94.

95.

96.

97.

98.

99.

100.

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

functions of proapoptotic Bcl-2 family members bak and bax are essential for normal development of multiple tissues. Mol. Cell 6:1389–99 Wei MC, Zong WX, Cheng EH, Lindsten T, Panoutsakopoulou V, et al. 2001. Proapoptotic BAX and BAK: a requisite gateway to mitochondrial dysfunction and death. Science 292:727–30 Strasser A, Puthalakath H, Bouillet P, Huang DC, O’Connor L, et al. 2000. The role of bim, a proapoptotic BH3-only member of the Bcl-2 family in cell-death control. Ann. NY Acad. Sci. 917:541– 48 Zha J, Harada H, Yang E, Jockel J, Korsmeyer SJ. 1996. Serine phosphorylation of death agonist BAD in response to survival factor results in binding to 14–3–3 not BCL-X(L). Cell 87:619–28 Datta SR, Dudek H, Tao X, Masters S, Fu H, et al. 1997. Akt phosphorylation of BAD couples survival signals to the cellintrinsic death machinery. Cell 91:231– 41 Ishibashi Y, Nishimaki K, Asoh S, Nanbu-Wakao R, Yamada T, Ohta S. 1998. Pore formation domain of human pro-apoptotic Bax induces mammalian apoptosis as well as bacterial death without antagonizing anti-apoptotic factors. Biochem. Biophys. Res. Commun. 243:609–16 Nouraini S, Six E, Matsuyama S, Krajewski S, Reed JC. 2000. The putative pore-forming domain of Bax regulates mitochondrial localization and interaction with Bcl-X(L). Mol. Cell Biol. 20:1604–15 Brenner C, Cadiou H, Vieira HL, Zamzami N, Marzo I, et al. 2000. Bcl-2 and Bax regulate the channel activity of the mitochondrial adenine nucleotide translocator. Oncogene 19:329–36 Basanez G, Nechushtan A, Drozhinin O, Chanturiya A, Choe E, et al. 1999. Bax, but not Bcl-xL, decreases the lifetime of planar phospholipid bilayer membranes

101.

102.

103.

104.

105.

106.

107.

108.

109.

at subnanomolar concentrations. Proc. Natl. Acad. Sci. USA 96:5492–97 Tsujimoto Y, Shimizu S. 2002. The voltage-dependent anion channel: an essential player in apoptosis. Biochimie 84:187–93 Epand RF, Martinou JC, Montessuit S, Epand RM, Yip CM. 2002. Direct evidence for membrane pore formation by the apoptotic protein Bax. Biochem. Biophys. Res. Commun. 298:744–49 Kuwana T, Mackey MR, Perkins G, Ellisman MH, Latterich M, et al. 2002. Bid, Bax, and lipids cooperate to form supramolecular openings in the outer mitochondrial membrane. Cell 111:331–42 Kaipia A, Hsu SY, Hsueh AJ. 1997. Expression and function of a proapoptotic Bcl-2 family member Bcl-XL/Bcl2-associated death promoter (BAD) in rat ovary. Endocrinology 138:5497–504 Davey GM, Kurts C, Miller JF, Bouillet P, Strasser A, et al. 2002. Peripheral deletion of autoreactive CD8 T cells by cross presentation of self-antigen occurs by a Bcl-2-inhibitable pathway mediated by Bim. J. Exp. Med. 196:947–55 Salmon M, Pilling D, Borthwick NJ, Viner N, Janossy G, et al. 1994. The progressive differentiation of primed T cells is associated with an increasing susceptibility to apoptosis. Eur. J. Immunol. 24:892–99 Mitchell T, Kappler J, Marrack P. 1999. Bystander virus infection prolongs activated T cell survival. J. Immunol. 162:4527–35 Grayson JM, Zajac AJ, Altman JD, Ahmed R. 2000. Cutting edge: increased expression of Bcl-2 in antigen-specific memory CD8+ T cells. J. Immunol. 164:3950–54 Puthalakath H, Huang DC, O’Reilly LA, King SM, Strasser A. 1999. The proapoptotic activity of the Bcl-2 family member Bim is regulated by interaction with the dynein motor complex. Mol. Cell 3:287–96

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL SURVIVAL 109a. Liu X, Dai S, Zhu Y, Marrack P, Kappler JW. 2003. The structure of a Bcl-xL/Bim fragment complex: implications for Bim function. Immunity 19(3):341–52 110. Kearney ER, Pape KA, Loh DY, Jenkins MK. 1994. Visualization of peptidespecific T cell immunity and peripheral tolerance induction in vivo. Immunity 1:327–39 111. Vella AT, McCormack JE, Linsley PS, Kappler JW, Marrack P. 1995. Lipopolysaccharide interferes with the induction of peripheral T cell death. Immunity 2:261–70 112. Croft M, Joseph SB, Miner KT. 1997. Partial activation of naive CD4 T cells and tolerance induction in response to peptide presented by resting B cells. J. Immunol. 159:3257–65 113. Sakata K, Sakata A, Kong L, Vela-Roch N, Talal N, Dang H. 1999. Monocyte rescue of human T cells from apoptosis is CD40/CD154 dependent. Scand J. Immunol. 50:479–84 114. Whitmire JK, Murali-Krishna K, Altman J, Ahmed R. 2000. Antiviral CD4 and CD8 T-cell memory: differences in the size of the response and activation requirements. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 355:373–79 115. Watts TH, DeBenedette MA. 1999. T cell co-stimulatory molecules other than CD28. Curr. Opin. Immunol. 11:286–93 116. Gramaglia I, Jember A, Pippig SD, Weinberg AD, Killeen N, Croft M. 2000. The OX40 costimulatory receptor determines the development of CD4 memory by regulating primary clonal expansion. J. Immunol. 165:3043–50 117. Maxwell JR, Weinberg A, Prell RA, Vella AT. 2000. Danger and OX40 receptor signaling synergize to enhance memory T cell survival by inhibiting peripheral deletion. J. Immunol. 164:107–12 118. Takahashi C, Mittler RS, Vella AT. 1999. Cutting edge: 4-1BB is a bona fide CD8 T cell survival signal. J. Immunol. 162:5037–40

P1: IKH

785

119. Cannons JL, Lau P, Ghumman B, DeBenedette MA, Yagita H, et al. 2001. 4-1BB ligand induces cell division, sustains survival, and enhances effector function of CD4 and CD8 T cells with similar efficacy. J. Immunol. 167:1313– 24 120. Lee HW, Park SJ, Choi BK, Kim HH, Nam KO, Kwon BS. 2002. 4-1BB promotes the survival of CD8+ T lymphocytes by increasing expression of Bcl-xL and Bfl-1. J. Immunol. 169:4882–88 121. Gravestein LA, Amsen D, Boes M, Calvo CR, Kruisbeek AM, Borst J. 1998. The TNF receptor family member CD27 signals to Jun N-terminal kinase via Traf-2. Eur. J. Immunol. 28:2208–16 122. Arch RH, Thompson CB. 1998. 4-1BB and Ox40 are members of a tumor necrosis factor (TNF)-nerve growth factor receptor subfamily that bind TNF receptor-associated factors and activate nuclear factor kappaB. Mol. Cell Biol. 18:558–65 123. Kawamata S, Hori T, Imura A, TakaoriKondo A, Uchiyama T. 1998. Activation of OX40 signal transduction pathways leads to tumor necrosis factor receptorassociated factor (TRAF) 2- and TRAF5mediated NF-kappaB activation. J. Biol. Chem. 273:5808–14 124. Akiba H, Nakano H, Nishinaka S, Shindo M, Kobata T, et al. 1998. CD27, a member of the tumor necrosis factor receptor superfamily, activates NF-kappaB and stress-activated protein kinase/c-Jun Nterminal kinase via TRAF2, TRAF5, and NF-kappaB-inducing kinase. J. Biol. Chem. 273:13353–58 125. Speiser DE, Lee SY, Wong B, Arron J, Santana A, et al. 1997. A regulatory role for TRAF1 in antigen-induced apoptosis of T cells. J. Exp. Med. 185:1777–83 126. Mitchell TC, Hildeman D, Kedl RM, Teague TK, Schaefer BC, et al. 2001. Immunological adjuvants promote activated T cell survival via induction of Bcl3. Nat. Immunol. 2:397–402

19 Feb 2004

12:10

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

786

AR

AR210-IY22-26.tex

MARRACK

¥

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

P1: IKH

KAPPLER

127. Ohno H, Takimoto G, McKeithan TW. 1990. The candidate proto-oncogene bcl3 is related to genes implicated in cell lineage determination and cell cycle control. Cell 60:991–97 128. Bours V, Villalobos J, Burd PR, Kelly K, Siebenlist U. 1990. Cloning of a mitogen-inducible gene encoding a kappa B DNA-binding protein with homology to the rel oncogene and to cellcycle motifs. Nature 348:76–80 129. Kerr LD, Duckett CS, Wamsley P, Zhang Q, Chiao P, et al. 1992. The protooncogene bcl-3 encodes an I kappa B protein. Genes Dev. 6:2352–63 130. Fujita T, Nolan GP, Liou HC, Scott ML, Baltimore D. 1993. The candidate protooncogene bcl-3 encodes a transcriptional coactivator that activates through NFkappa B p50 homodimers. Genes Dev. 7:1354–63 131. Viatour P, Bentires-Alj M, Chariot A, Deregowski V, de Leval L, et al. 2003. NF-kappa B2/p100 induces Bcl-2 expression. Leukemia 17:1349–56 132. Cheng EH, Sheiko TV, Fisher JK, Craigen WJ, Korsmeyer SJ. 2003. VDAC2 inhibits BAK activation and mitochondrial apoptosis. Science 301:513– 17 133. Tough DF, Sprent J. 1994. Turnover of naive- and memory-phenotype T cells. J. Exp. Med. 179:1127–35 134. Tough DF, Borrow P, Sprent J. 1996. Induction of bystander T cell proliferation by viruses and type I interferon in vivo. Science 272:1947–50 135. Zhang X, Sun S, Hwang I, Tough DF, Sprent J. 1998. Potent and selective stimulation of memory-phenotype CD8+ T cells in vivo by IL-15. Immunity 8:591– 99 136. Murali-Krishna K, Lau LL, Sambhara S, Lemonnier F, Altman J, Ahmed R. 1999. Persistence of memory CD8 T cells in MHC class I-deficient mice. Science 286:1377–81 137. Ku CC, Murakami M, Sakamoto A, Kap-

138.

139.

140.

141.

142.

143.

144.

145.

146.

pler J, Marrack P. 2000. Control of homeostasis of CD8+ memory T cells by opposing cytokines. Science 288:675–88 Becker TC, Wherry EJ, Boone D, Murali-Krishna K, Antia R, et al. 2002. Interleukin 15 is required for proliferative renewal of virus-specific memory CD8 T cells. J. Exp. Med. 195:1541–48 Goldrath AW, Sivakumar PV, Glaccum M, Kennedy MK, Bevan MJ, et al. 2002. Cytokine requirements for acute and Basal homeostatic proliferation of naive and memory CD8+ T cells. J. Exp. Med. 195:1515–22 Tan JT, Ernst B, Kieper WC, LeRoy E, Sprent J, Surh CD. 2002. Interleukin (IL)-15 and IL-7 jointly regulate homeostatic proliferation of memory phenotype CD8+ cells but are not required for memory phenotype CD4+ cells. J. Exp. Med. 195:1523–32 Grabstein KH, Eisenman J, Shanebeck K, Rauch C, Srinivasan S, et al. 1994. Cloning of a T cell growth factor that interacts with the beta chain of the interleukin-2 receptor. Science 264:965–68 Waldmann T, Tagaya Y, Bamford R. 1998. Interleukin-2, interleukin-15, and their receptors. Int. Rev. Immunol. 16:205–26 Liew FY. 2003. The role of innate cytokines in inflammatory response. Immunol. Lett. 85:131–34 Dubois S, Mariner J, Waldmann TA, Tagaya Y. 2002. IL-15Ralpha recycles and presents IL-15 in trans to neighboring cells. Immunity 17:537–47 Kennedy MK, Glaccum M, Brown SN, Butz EA, Viney JL, et al. 2000. Reversible defects in natural killer and memory CD8 T cell lineages in interleukin 15-deficient mice. J. Exp. Med. 191:771–80 Lodolce JP, Boone DL, Chai S, Swain RE, Dassopoulos T, et al. 1998. IL-15 receptor maintains lymphoid homeostasis by supporting lymphocyte homing and proliferation. Immunity 9:669–76

19 Feb 2004

12:10

AR

AR210-IY22-26.tex

AR210-IY22-26.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL SURVIVAL 147. Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. 1999. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401:708–12 148. Manjunath N, Shankar P, Wan J, Weninger W, Crowley MA, et al. 2001. Effector differentiation is not prerequisite for generation of memory cytotoxic T lymphocytes. J. Clin. Invest. 108:871– 78 149. Seddon B, Tomlinson P, Zamoyska R. 2003. Interleukin 7 and T cell receptor signals regulate homeostasis of CD4 memory cells. Nat. Immunol. 4:680–86 150. de Souza AR, Swanson B, Robertson J, Bender J, Kappler J, Marrack P. 2002. Some properties of T cells in animals. Adv. Exp. Med. Biol. 512:121–28 151. Swain SL. 2000. CD4 T-cell memory can persist in the absence of class II. Philos. Trans. R Soc. Lond. Ser. B Biol. Sci. 355:407–11

P1: IKH

787

152. Swain SL, Agrewala JN, Brown DM, Roman E. 2002. Regulation of memory CD4 T cells: generation, localization and persistence. Adv. Exp. Med. Biol. 512:113–20 153. Selin LK, Lin MY, Kraemer KA, Pardoll DM, Schneck JP, et al. 1999. Attrition of T cell memory: selective loss of LCMV epitope-specific memory CD8 T cells following infections with heterologous viruses. Immunity 11:733–42 154. Marrack P, Kappler J, Mitchell T. 1999. Type I interferons keep activated T cells alive. J. Exp. Med. 189:521–30 155. McNally JM, Zarozinski CC, Lin MY, Brehm MA, Chen HD, Welsh RM. 2001. Attrition of bystander CD8 T cells during virus-induced T-cell and interferon responses. J. Virol. 75:5965–76 156. Kaech SM, Hemby S, Kersh E, Ahmed R. 2002. Molecular and functional profiling of memory CD8 T cell differentiation. Cell 111:837–51

Marrack.qxd

3/12/2004

2:16 PM

Page 1

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL SURVIVAL

Figure 1 Extrinsic signals kill activated T cells via a caspase-dependent pathway.

C-1

Marrack.qxd

2:16 PM

MARRACK

■

Page 2

KAPPLER

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

3/12/2004

Figure 2 Intrinsic signals kill T cells via a pathway that destroys mitochondrial function. (A) In resting T cells the messenger of death, Bim, is bound to Bcl-2 and Bcl-xl and does not precipitate Bak or Bak aggregation and death of the cell. (B) T cell activation causes a fall in Bcl-2 mRNA and protein. (C) In activated T cells levels of Bcl-2 fall, and consequently less of the Bim protein is bound to Bcl-2. This may release Bim, leading to activation of Bak, Bak aggregation, and death of the cell.

Marrack.qxd

3/12/2004

2:16 PM

Page 3

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL SURVIVAL

Figure 3 An extended BH3-containing peptide from Bim may bind to Bak, freezing the Bak protein in its death-causing configuration.

C-3

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:765-787. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104716

Annu. Rev. Immunol. 2004. 22:789–815 doi: 10.1146/annurev.immunol.22.012703.104716 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on December 12, 2003

ASTHMA: Mechanisms of Disease Persistence and Progression

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp Yale University School of Medicine, Section of Pulmonary and Critical Care Medicine, New Haven, Connecticut; email: [email protected], [email protected], [email protected]

Key Words Th2, tolerance, memory, epithelium, eosinophils ■ Abstract When asthma is diagnosed, eosinophilic inflammation and airway remodeling are established in the bronchial airways and can no longer be separated as cause and effect because both processes contribute to persistence and progression of disease, despite anti-inflammatory therapy. Th2 cells are continually active in the airways, even when disease is quiescent. IL-13 is the key effector cytokine in asthma and stimulates airway fibrosis through the action of matrix metalloproteinases on TGF-β and promotes epithelial damage, mucus production, and eosinophilia. The production of IL-13 and other Th2 cytokines by non-T cells augments the inflammatory response. Inflammation is amplified by local responses of the epithelium, smooth muscle, and fibroblasts through the production of chemokines, cytokines, and proteases. Injured cells produce adenosine that enhances IL-13 production. We review human and animal data detailing the cellular and molecular interactions in established allergic asthma that promote persistent disease, amplify inflammation, and, in turn, cause disease progression.

INTRODUCTION In the past 20 years, the prevalence of asthma has almost doubled, such that asthma now affects approximately 8% to 10% of the population in the United States. It is the leading cause of hospitalization among young children. This epidemic increase in asthma has been attributed to aspects of Western culture, including outdoor and indoor air pollution, childhood immunizations, and cleaner living conditions, but no single cause has been identified. The high prevalence rate has markedly increased the cost of this disease as measured in health care dollars, time away from work and school, and mortality. Asthma has been the focus of media, public health, and research initiatives to improve awareness and compliance with medications and to understand the causes and course of disease. Asthma is a chronic inflammatory disease of the airways characterized by recurrent episodes of airway obstruction and wheezing. The presence of eosinophilic 0732-0582/04/0423-0789$14.00

789

12 Feb 2004

16:14

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

790

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

infiltration in the airways has been known for almost 100 years, having first been reported in 1908 in a patient who died of asthma (1). Yet, inflammation was not believed to be a cause of asthma until recently. In the early 1980s, the development of flexible, fiberoptic bronchoscopy permitted evaluation of the airways in groups of asthmatics with well-defined lung function during periods of disease inactivity. These studies revealed that even when patients were asymptomatic, inflammation of the airways was present. Around the same time, the seminal papers describing CD4 Th subsets and their functional effects were published (2, 3), and subsequent studies showed that Th2 cells were present in the airways of asthmatics (4). These findings, together with the association of Th2 cytokines and allergic diseases, led to the theory that Th2 cells promote asthma. For the past 15 years, asthma research has almost exclusively focused on inflammation as a cause of disease. Asthma has many causes, including inhaled exposure to allergens and isocyantes, and in many cases the inciting agents are not known. For the most part, independent of the cause, the pathology is identical and the course of disease similar, indicating that the inflammatory response and the airway pathology are common responses of the respiratory tract to injury. In this review, we focus on allergic asthma, the type experienced by approximately 80% of asthmatics. Many of the observations also apply to nonallergic asthmatics. We review some of the mechanisms by which inflammatory and structural cells in the airways communicate to intensify disease and ultimately promote persistence and progression of asthma.

Pathophysiology In asthma, the airway wall is infiltrated with mononuclear cells, which are mostly CD4 T cells, and with eosinophils. Mast cells, macrophages, plasma cells, and neutrophils are variably increased in the airways of asthmatics compared with those of controls, but electron microscopy studies show that mast cells are activated as they have evidence of degranulation (5). In the airway lumen, mucus is mixed with activated macrophages, lymphocytes, eosinophils, and sloughed epithelial cells. In some asthmatics, especially severe cases, neutrophils are increased. Structural changes of the airway wall, collectively referred to as airway remodeling, may be a result either of the interaction of inflammatory mediators with stromal cells or of tissue injury. Local factors, including the structural cells in the airway and the matrix, respond to inflammation in a characteristic, coordinated fashion that may be an attempt to repair the damage caused by local inflammation—an effort to keep the airway intact. Airway wall thickening ranges from 10% to 300% of normal, leading to a reduction in the airway luminal diameter (6, 7). The small airways (2– 4 mm) are commonly involved, and in fatal asthma, all the airways except the largest are affected. In addition to inflammatory cells, most of the elements in the airway wall contribute to the increased thickness. Mucous glands hypertrophy, and there is metaplasia of the airway epithelium into mucus-secreting cells. The subepithelial layer, which is 4–5 microns thick in normal subjects, ranges from 7–23 microns in

15 Mar 2004

15:46

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ASTHMA: PERSISTENCE AND PROGRESSION

P1: IKH

791

asthmatics as a result of deposition of collagen (Types I, III, and V), fibronectin, and tenascin just below the basement membrane in the lamina reticularis (7, 8). The matrix also contains increased proteoglycans and glycosaminoglycans, including hyaluronan. Myofibroblasts, which produce collagens, are hyperplastic. Smooth muscle mass is increased and may occupy up to three times the normal area, predominantly because of cell hyperplasia (9). There is also vascular dilatation and angiogenesis, increased vascular permeability, and airway wall edema (6). Airway remodeling and inflammation result in airway hyperresponsiveness (AHR) and airway obstruction, which causes breathlessness and wheezing. AHR is defined as an increased bronchoconstrictor response to a nonspecific stimulus (10) and is sometimes referred to as “twitchy” airways. The response can be measured in the laboratory using nonselective stimuli that provoke bronchoconstriction in all asthmatics and is generally performed using dose-response curves by inhalation of these agents, such as methacholine. The precise mechanisms that control AHR are poorly understood. The magnitude of AHR correlates with the level of airway inflammation (11). But other factors, including the reduced airway diameter, an increase in smooth muscle contractility, the degree of epithelial injury, dysfunctional neuronal regulation, an increase in microvascular permeability, and many inflammatory mediators, have been associated with AHR (12). When the bronchial airways are narrowed by structural changes with mucus and inflammatory cells within the airway lumen, it is easy to imagine how any stimulus that increases smooth muscle constriction will result in airway obstruction. Inflammation and the structural changes described have been observed in airway biopsies of children years before the symptoms of asthma become manifest (13); therefore it is believed that symptomatic airway obstruction and wheezing occur when a critical degree of airway remodeling has occurred. In some children and adults with asthma, disease symptoms disappear during life. In the absence of symptoms, they have persistently increased airway responsiveness to methacholine, and airway biopsies and bronchoalveolar lavage (BAL) show elevated numbers of eosinophils, suggesting that this is asymptomatic disease rather than remission (14–17). In fact, a majority in this group develop symptoms of asthma again, indicating that asthma rarely resolves. Anti-inflammatory treatment of asthma with inhaled or oral corticosteroids reduces inflammation, but upon withdrawal of treatment, inflammation recurs. Corticosteroid treatment has shown limited, if any, benefit in reducing airway remodeling (18–21). Although earlier intervention with anti-inflammatory drugs may limit disease, to date we are poor at identifying asthmatics before manifestation of symptoms. Over time, lung function in asthmatics declines more rapidly compared with that of normal individuals, indicating disease progression (22). At the time asthma is diagnosed, airway inflammation and remodeling can no longer be treated as cause and effect because they are intertwined in their contributions to disease (Figure 1). Likewise, it is impossible to determine the initiating steps of the effector response that caused inflammation and remodeling. Here, our discussion begins with Th2 cells, but it is currently not clear that this is where asthma starts. This review

12 Feb 2004

16:14

792

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

focuses on human and animal data detailing the cellular and molecular interactions in established allergic asthma that promote persistent disease, amplify inflammation, and in turn cause disease progression.

CD4 Th CELLS IN ASTHMA

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD4 Th2 Cells In asthma, CD4 Th2 cells are believed to initiate and perpetuate disease. Lymphocytes make up a small percentage of total leukocytes in the lung. Yet, CD4 T cells are increased in the airways of asthmatics, and they express activation markers including CD25 and Class II MHC (23, 24). IL-4, IL-5, and IL-13 protein and mRNA levels are increased in BAL fluid, BAL cells, and airway biopsies of asthmatics (4, 25). The transcription factor GATA-3 is expressed at high levels in CD4 T cells from asthmatic airways (26), whereas t-bet mRNA is undetectable (27), indicating an unambiguous Th2 lymphocyte phenotype in the respiratory tract in asthma. Animal studies corroborate the hypothesis that Th2 cells promote allergic airways disease and have helped to define the pathways by which Th2 cells and Th2 cytokines affect these changes. CD4 T cells are required for antigen-induced allergic airway inflammation and AHR (28). When Th2 cells cannot develop in response to antigen, such as in IL-4Rα−/− or Stat6−/− mice, allergic airway inflammation is not induced. Th2 cells, when activated in the respiratory tract with inhaled antigen, can stimulate an acute allergic inflammatory response with eosinophils, AHR, and mucus hypersecretion (29, 30). Transgenic mice expressing a dominant-negative mutant of GATA-3 in an inducible and T cell–specific fashion were developed and analyzed in a murine model of allergic inflammation (31). These studies demonstrated that inhibition of GATA-3 activity after sensitization and Th2 induction but before inhaled antigen challenge caused blunting of Th2 effects, including eosinophilic inflammation and AHR. These animal studies have proved critical in defining the importance of Th2 cells in early disease pathogenesis, yet because these models examine allergic phenomena after days of inhaled antigen exposure, they do not exhibit chronic remodeling of the airways that defines asthma at diagnosis. Chronic overexpression of individual Th2 cytokines in the respiratory tract has been achieved using lung-specific promoters. The Clara cell 10-kDa (CC10) promoter driving the Th2 cytokines, IL-4, IL-5, IL-9, or IL-13, in the airway epithelium exhibited characteristic allergic inflammatory features in the airways, including eosinophilia and mucus overproduction (32–35). Transgenic mice that overexpress IL-13, IL-9, and IL-5 showed AHR and collagen deposition in the airways, indicating that chronic exposure to Th2 cytokines could also induce airway remodeling. Mice deficient in t-bet, a transcription factor that directs Th1 cell development, spontaneously developed eosinophilic airway inflammation, mucus metaplasia, airway collagen deposition, myofibroblast hypertrophy, and AHR.

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

ASTHMA: PERSISTENCE AND PROGRESSION

P1: IKH

793

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD4 Th1 cells cannot be generated in these mice; therefore, activated CD4 T cells from t-bet−/− mice produce high levels of Th2 cytokines and minimal IFN-γ . IL-4, IL-5, and IL-13 levels were increased in the BAL in naive t-bet−/− mice, and there were increased numbers of effector/memory CD4 T cells in the lungs of these mice (27; L. Cohn, unpublished observations), indicating that CD4 Th2 cells are chronically activated in the respiratory tract and can induce an asthmalike syndrome in mice. Together, these animal models support the hypothesis that chronic activation of Th2 cells is sufficient for the induction of inflammation, the physiologic and the chronic pathologic changes associated with asthma.

CD4 Th1 cells IFN-γ is also increased in the serum and BAL during acute asthma exacerbations, and CD4 Th1 cells have been identified in the respiratory tract in some asthmatics (23). These findings have raised the possibility that Th1 cells may contribute to disease pathology. In mice, Th1 cells activated in the respiratory tract with inhaled antigen cause a neutrophil-predominant inflammatory response without mucus metaplasia or AHR (29). Th1 cells alone do not produce any of the characteristic features of asthma. If Th2 cell development is inhibited and shifted toward Th1 induction, then allergic airway inflammation is reduced, perhaps recapitulating the importance of Th2 cells in development of disease and once again showing that Th1 cells alone do not cause asthma (36). The factors that influence which CD4 Th subsets are generated, i.e., the balance of Th1/Th2 and regulatory T (Treg) cells, during initial exposure to allergen is the basis of the hygiene hypothesis and is not a focus of this review. Studies of Th1 cells and IFN-γ in regulating disease, once Th2 cell priming has occurred, are complex. A number of studies show that IFN-γ or IFN-γ -producing Th1 cells inhibit Th2-induced eosinophilia, mucus production, and AHR. These effects appear to be a result of IFN-γ on effector responses, rather than direct inhibition of Th2 cells (37–42). Yet, other studies have shown that Th1 cells enhance pulmonary inflammatory responses and AHR (43–45). Viral infections, which typically activate CD4 and CD8 IFN-γ -producing cells, commonly exacerbate asthma. CD4 Th1 cells can recruit and activate Th2 cells in the absence of specific antigen, indicating a possible mechanism of Th1-driven augmentation of allergic inflammation (46). The conflicting results of the role of Th1 cells in these models may indicate that the timing of activation of Th1 cells in the setting of ongoing Th2-induced inflammation determines the net outcome. Further complicating our interpretation of these data is the problem that none of these animal studies tests the effects of Th1 cells in models of chronic airway inflammation. A small clinical trial of inhaled recombinant IFN-γ administered to mild asthmatics for three weeks led to a reduction in the percentage of airway eosinophils and did not increase other inflammatory cells in the BAL. Although very limited in scope, this study suggests that, over a short period, IFN-γ does not exacerbate chronic disease. The long-term effects of elevated IFN-γ in the airways, as has been observed in some asthmatics, is not known.

15 Mar 2004

15:46

794

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Th2 Cell Development Although not the primary focus of this review, how Th2 responses develop in the respiratory tract in asthmatics is important and warrants a brief outline as this dysregulation may continue to promote Th2 cell development once disease is established. Exposure to foreign protein antigens in the respiratory tract should not induce active inflammation. Mucosal surfaces encounter nontoxic proteins continually, and vigorous immune responses do not generally develop. The respiratory tract must maintain its essential gas exchange function and therefore has evolved to limit access of proteins to the immune system with barriers like the mucus layer and intercellular tight junctions (Figure 2A). The respiratory mucosa is not totally impenetrable, and active mechanisms suppress pulmonary immune responses. Inhaled antigens induce immune unresponsiveness in naive T cells. Animal studies have shown that the tolerogenic effects of inhaled antigen are both antigen specific and nonspecific by a “bystander” effect, require the local lymph nodes (LN) that drain the respiratory tract, yet act systemically (47–50). Inhaled antigen can render CD4 T cells anergic to antigen (51). CD4, CD8, and γ /δ T cells can be induced by inhaled antigen to become Treg (40, 48, 52). Treg effects may depend on IL-10 (53, 54), TGF-β (55), and ICOS-ICOS-L costimulation (56). Antigen-presenting cells (APCs) at mucosal surfaces have unique features that stimulate the development of tolerance. The generation of Treg appears to be predominantly under the control of mucosal dendritic cells (DCs). DCs that line the mucosal surfaces produce IL-10 and also promote the generation of Tr1/Th3 cells and Th2 cells (57, 58). Once a population of effector/suppressor T cells producing IL-10, IL-4, and TGF-β is established in the lymphoid tissue, the cytokine environment will influence the cell-cell interactions that lead to immune tolerance. Low expression of Class II MHC and costimulatory molecules on DCs at mucosal surfaces also play a role in the deviation of mucosal immune responses toward Treg or anergy (59, 60). IL-10 decreases MHC Class II and CD80 expression (61). In addition, Treg cells have recently been shown to induce tolerogenic DCs that have decreased costimulatory activity and can induce CD4 T cell anergy, a feedback mechanism to enhance immune unresponsiveness (Figure 2A) (62, 63). In the lung, alveolar macrophages are extremely potent in suppressing immune responses, possibly owing to a regulatory effect on the DC (64). If inhaled protein antigens do not induce tolerance, CD4 T cells are more commonly directed down a Th2 pathway. As noted above, myeloid DCs preferentially skew immune responses toward Th2 and suppressor T-cell populations (58). Innate immune responses by mast cells, γ /δ T cells, and NKT cells may help to promote Th2 cell development by their secretion of IL-4 and IL-13 (65–67). Many aeroallergens possess protease activity that may allow them to overcome the protective mucosal barrier and induce mast-cell degranulation and IL-4 production (68, 69). The genetics of an individual likely play a key role in biasing CD4 T-cell responses. Asthma traits have been linked strongly with flanking markers of a number of different genes including the human cytokine gene cluster on chromosome

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ASTHMA: PERSISTENCE AND PROGRESSION

P1: IKH

795

5q31, which contains the genes for IL-4, IL-5, IL-9, IL-13, CD14, and GM-CSF. Polymorphisms in a number of these genes have been associated with a higher incidence of asthma and atopy. For example, nucleotide polymorphisms associated with atopy lead to increased binding of transcription factors, including NFAT-1 and AP-1, and enhanced IL-4 transcription (70–72). Two genetic variants of IL-13 were associated with asthma. One variant in the promoter region was associated with increased nuclear protein binding and may affect IL-13 regulation. The polymorphism in the IL-13 coding region alters the cytokine charge and may alter the ligand-receptor interaction (73, 74). These genetic factors could influence effector pathways in asthma by enhancing Th2 cell generation or increasing the production or signaling of Th2 cytokines.

Th2 Cell Persistence In asthma, even during periods of quiescent disease, Th2 cytokines are produced in the airways. In the absence of recent antigen exposure, a typical Th2 response should die down because normal regulatory functions eliminate effector CD4 T cells after activation. When activation is continual, these mechanisms should also eliminate the responsive cells. Th2 cell persistence, therefore, must be due to one or a combination of (a) increased generation from naive CD4 T cell precursors, (b) increased recruitment and/or proliferation of effector/memory CD4 T cells, or (c) reduced elimination of effector/memory CD4 T cells. The remodeled, inflamed airways may promote continued Th2 cell activation in asthma because they can no longer function as an efficient barrier to limit local immune responses (Figure 2B). When exposed to allergens, the damaged airway epithelium may allow more soluble proteins to cross from the luminal to the apical surface, potentially increasing antigen presentation. Injured airway epithelial cells produce GM-CSF, which increases DC maturation. Allergen challenge induces DC progenitors in the bone marrow (75). Increased vascular permeability should allow rapid migration of inflammatory cells from the vascular space into the airway. In airway biopsies of asthmatics, DCs were increased in number and activated, expressing more cell-surface Class II MHC molecules (76, 77). DC activation is likely induced by locally secreted mediators such as GM-CSF and by interactions with activated T cells via CD40-CD40 ligand (78–80). Memory CD4 Th2 cells that reside in the airways can be activated locally by DC that provide essential costimulatory signals between B7-RP-1-ICOS and OX-40-OX-40 ligands, thereby leading to rapid induction of cytokines and AHR (79, 81–83). Antigen presentation may be prolonged owing to a small population of airway APC that can present antigen for up to eight weeks following inhalational exposure (81). In this setting, administration of inhaled antigen induces activation rather than tolerance. In mice with active Th2-induced pulmonary inflammation or in mice previously sensitized to an antigen, inhaled exposure that had induced protective tolerance in naive mice led to T-cell activation (51, 84–87). Naive T cells may also be activated in LNs to become Th2 effector cells. Small numbers of activated DCs

12 Feb 2004

16:14

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

796

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

that present antigen were observed to migrate to the local LN where the environment and features of the DC favor Th2 generation (79, 88). Airway inflammation and changes in airway structure promote continued activation of memory CD4 Th2 cells and generation of Th2 effector cells from naive precursors, thus maintaining the pool of activated Th2 cells in the airway (Figure 2B). Continued T-cell activation leads to clonal deletion from antigen-induced cell death in many systems. Then, why does repeated stimulation with allergens cause T-cell activation? Studies of CD4+ T lymphocytes isolated from the peripheral blood of asthmatics and stimulated in vitro showed a diminished capacity to undergo fas-mediated apoptosis as compared with that of controls (89, 90). In both of these instances, the limited ability to undergo apoptosis could merely be a marker of T-cell activation, since recently stimulated CD4 T cells are resistant to fas-induced cell death and have increased levels of antiapoptotic Bcl proteins (91). Alternatively, asthmatic Th2 cells may have specific pathways activated for long-term survival. Th1 cells have reduced survival in vitro after antigenic stimulation owing to enhanced apoptosis (92). In airway biopsies of asthmatics and in peripheral blood of patients with atopic dermatitis, IFN-γ -producing T lymphocytes had higher rates of apoptosis than did IL-4-producing T cells (93, 94). While this may explain how a skewed Th2 response becomes more polarized over time, it does not explain how Th2 cells become resistant to programmed cell death. Recent studies define a critical role for IFN- γ in activation-induced cell death. IFN-γ acting through its receptor and Stat-1, stimulates apoptosis by activating caspase-8 just downstream of the death receptor (95). Therefore, a highly polarized population of Th2 cells may persist longer in the absence of IFN-γ . A lack of IFN-γ production also leads to the generation of more memory CD4 cells, suggesting a possible link between CD4 Th2 cell survival and memory-cell generation (96). In the respiratory tract of IFN-γ -receptor (R)-deficient mice that were immunized and challenged with inhaled ovalbumin, there was prolonged eosinophilia and increased Th2 cytokines, indicating that a lack of IFN-γ signaling enhances effector Th2 responses (97). IL-4 also prolongs survival of activated T cells by inducing expression of Bcl-2, a pathway that is independent of caspase-8 (98). Last, cross-regulation of IL-4 and IFN-γ may enhance these responses, as increasing IL-4 will reduce IFN-γ production and vice versa. In Th2 cells that lacked production of IL-4, IFN-γ production was only minimally increased, yet BAL eosinophilia was markedly diminished (99). And, in IFN-γ R−/− mice that received a polarized population of Th2 cells and exposure to inhaled antigen, eosinophilia was five times higher than in IFN-γ R+/+ mice that received the same cells (37). These concepts may help to explain the presence of chronic airway inflammation in t-bet-deficient mice, in which defective Th1 cell development leads to more effector/memory CD4 T cell in the airways and BAL fluid with high levels of IL-4 and low levels of IFNγ (27). IFN-γ is clearly not the only pathway by which CD4 Th cells undergo apoptosis, as many other mechanisms have been described. Yet a subtle increase in Th2 cell survival may be important in asthma. If, over time, these effects lead to

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

ASTHMA: PERSISTENCE AND PROGRESSION

P1: IKH

797

a more polarized Th2 cell population in the respiratory tract and IFN-γ production is reduced, allergic inflammation may become more profound. The presence of IFN-γ in the airways of some asthmatics suggests that this is not the mechanism of Th2 persistence in all asthmatics.

EFFECTOR PATHWAYS THAT ENHANCE DISEASE

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Th2 Cytokines CD4 Th2 lymphocytes contribute to the inflammatory response and to airway remodeling by producing cytokines. Each cytokine has distinct functional effects in induction of disease, but IL-13 predominates in its contribution to the pathophysiology in asthma (100). CD4 lymphocytes produce a majority of the Th2 cytokines in the respiratory tract in asthma. In airway biopsies and BAL cells from asthmatics, IL-4, IL-5, and IL-13 colocalize with T-cell markers in a majority of, but not all, cells (101). Yet, CD8, γ /δ and NKT cells, eosinophils, mast cells, basophils, NK cells, and subsets of Class II MHC-expressing accessory cells can produce Th2 cytokines (101–105). IL-4, IL-5, and IL-13 produced by non-CD4 T cells may be essential for the development and perpetuation of asthma. The period of cytokine secretion from this array of cells will vary. Thus, as lymphocytes die or cease production of cytokines after their activation, waves of cytokines produced by non-CD4 T cells may follow. For example, IL-4 production activates eosinophils, and eosinophils then produce IL-4 (106). Th2 cell production of IL-25 also enhances cytokine production (105). IL-25 is a cytokine from the IL-17 family that is produced by Th2 cells and mast cells and promotes Th2-induced lung pathology (105, 107). In mice infected via the airway with adenovirus-expressing IL-25, there were increased airway mononuclear cells and eosinophils; high serum levels of IgE, IgG1, and IgA; and increased mucus production. These effects were a result of IL-4, IL-5, and IL-13, and the histological changes were still observed in RAGdeficient mice, indicating production was not by T or B cells. A CD11c- and Class II MHC-expressing accessory cell that expressed mRNA for IL-13 and IL-5 was responsible. These results concur with studies from our laboratory indicating that RAG−/− mice possess an IL-13-producing cell that can stimulate mucus production (L. Cohn, unpublished observations). These studies highlight the significance of effector cytokines produced by nonlymphocytes. In asthmatics, enhancing Th2 cytokines will increase the magnitude of and/or prolong the immune response.

Eosinophilia Airway eosinophilia is the most characteristic finding in asthma and has been considered central in the pathogenesis of disease. Yet, to date, researchers have not established a precise role of eosinophils in disease development and persistence. Animal studies have detailed the pathways by which eosinophils are mobilized from the bone marrow and recruited to the respiratory tract. These studies have

12 Feb 2004

16:14

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

798

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

also defined how individual Th2 cytokines control these steps. Airway eosinophilia depends on both IL-5 and Stat6 signaling. In the absence of IL-5, blood and BAL eosinophils are not increased in response to Th2 cell activation, yet eosinophils are present in the airway tissue at approximately one third the level observed in IL-5+/+ mice (106, 108–110). In mice lacking effective signaling for IL-4 and IL-13, few eosinophils were observed in the airway tissue or in BAL in response to Th2 cell activation in the respiratory tract. This occurred despite the ability of these mice to generate and release eosinophils from the bone marrow (111, 112). Associated with these findings was a marked reduction in eosinophil-recruiting chemokines including eotaxin-1. Thus, CD4 T cells provide essential signals for eosinophil mobilization, activation, and recruitment to the respiratory tract. There are also other strain-specific factors in mice that lead to different levels of eosinophils in the respiratory tract in response to comparable Th2 stimuli (113). Once eosinophils have been recruited to the respiratory tract, their activation leads to secretion of major basic protein, eosinophil cationic protein and eosinophil peroxidase; cytokines such as TNF-α, GM-CSF, IL-4, IL-13, and IL-5; chemokines including RANTES and eotaxin; and platelet-derived growth factor (106). These factors damage airway epithelial cells, stimulate mucus secretion and fibrosis, and induce bronchospasm and AHR. In addition, eosinophil numbers in the sputum and airway wall correlate with disease severity (24, 114). Some murine models of acute allergic airway inflammation have shown that AHR is dependent on airway eosinophilia, although other models have not (115, 116). One addition to these debates is the issue of localization of eosinophils. Although past investigations routinely sampled the BAL as a measure of airway leukocytes, more recent studies have revealed differential regulation of eosinophils in the lung and airway wall compared with the airway lumen (99, 117, 118). Thus, some animal studies may have missed critical features of the airway wall by only assessing the BAL. In the established asthmatic with eosinophils in both locations, this is not likely to be of great relevance. Yet, when eosinophils are manipulated, differential regulation in the respiratory tract must be considered. This was highlighted in recent clinical trials using a humanized anti-IL-5 antibody to treat asthma. The initial studies showed that a single dose of antibody had no significant effect on lung function over time (119). Whereas a single dose or multiple doses of anti-IL-5 almost eliminated blood and BAL or sputum eosinophils, multiple doses had only a modest effect on eosinophils in mucosal biopsies (120, 121). Tissue eosinophils appear to have reduced levels of IL-5R on the cell surface, indicating less dependence on IL-5 (122, 123). Therefore, testing the role of eosinophils in maintenance of symptoms in asthma must wait until eosinophils can be eliminated from tissues, and this may require blocking local chemokines, GATA-3, or Stat6 activation. In the meantime, we can speculate that eosinophils play key roles in maintenance and progression of asthma by (a) enhancing inflammation through the production of cytokines, including IL-13, and presenting antigen to and activating Th2 cells (109, 124) and by (b) increasing remodeling by stimulating subepithelial fibrosis (39). IL-4, IL-5, and IL-13 produced by eosinophils maintain their

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

ASTHMA: PERSISTENCE AND PROGRESSION

P1: IKH

799

recruitment to the airways. IL-5, GM-CSF, and IL-3 may enhance eosinophil survival by increasing antiapoptotic Bcl proteins bcl-2 and bcl-xL (125–127). Thus, the eosinophil, by its number and location, is a prime candidate to amplify local immune responses, increase tissue damage and fibrosis, and in so doing sustain itself in the airways.

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Adenosine Adenosine is generated by dephosphorylation of adenine nucleotides released from inflammatory and injured cells. In asthma, adenosine is increased in the bloodstream and airways (128, 129). Adenosine has long been known to induce bronchoconstriction in asthmatics, but not in normal subjects. It can engage cell-surface adenosine G-protein-coupled receptors on mast cells, eosinophils, macrophages, neurons, epithelial cells, and smooth muscle cells. Adenosine can be removed by metabolism by adenosine deaminase (ADA) or adenosine kinase. Recent studies in CC10 IL-13 Tg mice show that IL-13 induced high levels of adenosine and decreased ADA activity (130). When ADA was administered to IL-13 Tg mice and adenosine levels were returned close to normal, airway inflammation was reduced and there was minimal subepithelial fibrosis. Parallel to these findings, ADA-deficient mice had high levels of adenosine and a lung phenotype strikingly similar to CC10 IL-13 Tg mice with high levels of IL-13. This pathology was largely blocked with an IL-13 inhibitor. Thus, adenosine stimulates IL-13, and IL-13 stimulates adenosine. This link is critical for expression of allergic airway inflammation and remodeling in mice. In asthma adenosine is likely generated from inflammatory cells, damaged airway epithelium, and other structural cells in the lung. IL-13 produced in the lung may inhibit ADA, which increases adenosine levels and stimulates release of more IL-13. This in turn activates effector pathways that damage tissue, leading to the generation of more adenosine. In these studies, blocking the increase in adenosine by administering ADA relieved the cycle, which suggests that adenosine stimulates a nonredundant pathway to increase IL-13 and promote disease.

Monocyte Chemotactic Protein-1 Monocyte chemotactic protein-1 (MCP-1) also appears to be involved in a chain of inflammatory events that promotes airway remodeling in asthma. MCP-1 is elevated in the airways of asthmatics (131) and is produced by bronchial epithelial cells, macrophages, and smooth muscle cells. In mice, IL-13-induced inflammation is associated with an increase in MCP-1 release (132). Recent studies indicate that adenosine, through its effects on the A2B receptor, increases the release of MCP-1 from bronchial smooth muscle cells (133). This may be one pathway by which IL-13 increases MCP-1. MCP-1 may control multiple different aspects of the Th2 inflammatory response. First, it attracts effector/memory Th cells (134). Therefore, circulating Th2 cells will be recruited to the airways. MCP-1 has been shown, in some studies, to enhance Th2 cell differentiation. Naive CD4 T cells stimulated

12 Feb 2004

16:14

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

800

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

with antigen and MCP-1 induced IL-4 but not IFN-γ (135), and MCP-1-deficient mice could not be induced to generate Th2 cells (136). The mechanism by which MCP-1 influences Th2 cell generation has not been defined. In other studies, a deficiency in MCP-1 did not polarize strictly along Th1/Th2 lines (137), and mice with a deficiency in CCR2, the primary receptor for MCP-1, exhibited enhanced eosinophilia and a defect in IFN-γ production (138), suggesting that MCP-1 effects may not be Th2 cell specific. MCP-1 stimulates mast-cell mediator release (139), including Th2 cytokines, which will promote further Th2 cell generation. MCP-1 also acts to induce remodeling of the airways by stimulating TGF-β1 release from macrophages and fibroblasts (140, 141) and by promoting airway fibrosis (132). The induction of MCP-1 in a Th2 inflammatory response can amplify Th2-induced effects in a number of ways and promote subepithelial fibrosis. Upon stimulation with MCP-1, fibroblasts produce more MCP-1, which recruits Th2 cells, which in turn produce more IL-13, which stimulates more MCP-1 release, and the cycle continues.

Matrix Metalloproteinases Matrix metalloproteinases (MMPs) are a large family of proteinases that are elevated in lung diseases characterized by airway remodeling, including asthma. MMP catalytic activity depends on cleavage of a prodomain and the presence of a zinc-binding site. MMPs are categorized according to the types of molecules they break down—including collagenases, gelatinases, or stromeolysins—and their location, e.g., cell-surface MMPs. A disintegrin and metalloproteinase (ADAM) compose another class of metalloproteinase with two functional domains: a disintegrin that promotes adhesion of the molecule to cell-surface integrins; and a metalloproteinase that can cleave cell-surface cytokines into their active forms, release receptors, and act on numerous other cell-surface molecules. MMPs are responsible for extracellular matrix remodeling during normal physiologic events, like embryogenesis and wound repair. Their ability to break down collagen and elastin, the major structural proteins in the lung, imply that they play a major role in the development of lung pathology. Yet, MMPs also have essential roles in leukocyte migration from the vascular space to sites of inflammation and cleavage of molecules, such as adhesion receptors and membrane-bound and inactive cytokines (142). MMP-2 (Collagenase A), MMP-3 (Stromeolysin-1), MMP-9 (Gelatinase B), and MMP-12 (metalloelastase) are all elevated in asthmatics, although MMP-9 is the predominant MMP in asthma (143). In CC10 IL-13 Tg mice with airway inflammation and remodeling, MMP-2, -9, -12, -13, and -14 levels were increased (144), and in models of acute allergic inflammation, MMP-2 and MMP-9 were elevated (145). MMP-9 is consistently increased in blood, BAL fluid, sputum, and biopsies of asthmatics, and it is increased in the BAL fluid 24 h after antigen challenge (146, 147). Tissue inhibitors of metalloproteinase (TIMPs) are the specific inhibitors of MMPs, and TIMP-1, which covalently binds MMP-9, is also increased in asthma.

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ASTHMA: PERSISTENCE AND PROGRESSION

P1: IKH

801

During acute asthma exacerbations, the ratio of MMP-9 to TIMP-1 increases, and this has been hypothesized to promote airway remodeling (148). In the lung, MMP-9 is produced by neutrophils, macrophages, eosinophils, lymphocytes, mast cells, and DCs. When activated, epithelial cells, fibroblasts, smooth muscle cells, and endothelial cells produce MMP-9 (147). In mice, a deficiency in MMP-9 reduces allergic airway inflammation. DC migration requires MMP-9 for antigen uptake in the airway, and this role of MMP-9 may explain the effect on airway inflammation (149, 150). Once the DC has acquired antigen, MMP-9 is not required for DC migration to the local LNs. When CC10 IL-13 Tg mice are crossed to MMP-9-deficient mice, there is an increase in BAL neutrophils and enhanced expression in the lung of the neutrophil-attracting chemokines KC and MIP-2 (151). These studies bypass the need for T cells to produce IL-13, thus eliminating the requirement of MMP-9 in DC-induced lymphocyte activation. MMP-9 plays a role in the suppression of neutrophilia, but it does not have a direct effect on eosinophil recruitment. Subepithelial fibrosis observed in CC10 IL-13 Tg mice was due, in part, to MMP-9-dependent activation of TGF-β1 (152). TGF-β1 plays a central role in the pathogenesis of a variety of fibrotic disorders (153). Thus, in mice, MMP-9 appears to play a critical role in DC-mediated T-cell activation and in subepithelial fibrosis, likely by activation of latent TGF-β. Mast cells and lung tissue macrophages produce MMP-9 when stimulated by contact with activated T cells (154, 155), indicating another potential positive feedback mechanism to provide local DCs with what they need for immune responsiveness. Other MMPs modulate inflammation and remodeling in some models of asthma. MMP-12 regulates subepithelial fibrosis and eosinophilia in CC10 IL-13 Tg mice. Similar to mice deficient in MMP-9, MMP-12-deficient mice bred to CC10 IL13 Tg animals have less active TGF-β1 and a reduction in airway fibrosis, but these mice have fewer BAL eosinophils. MMP-12, therefore, plays an important role in eosinophil recruitment to the respiratory tract and subepithelial fibrosis in this asthma model. Elevation in MMP-2 activity in mice with allergic airway inflammation regulates inflammatory cell migration within the lung (156). When MMP-2 effects are blocked, antigen-induced inflammation, including lymphocytes and eosinophils, is markedly reduced in the airway lumen (BAL), but enhanced in the lung parenchyma and around the airways, indicating a defect in migration out of the airway tissue. These findings are associated with reduced eotaxin levels in the BAL, suggesting that MMP-2 modulates leukocyte chemoattractants in asthma. The potential mechanisms of MMP-2 activity include acting directly or indirectly on the epithelium to release chemokines, or degrading of chemokines in the airway tissue to generate a gradient (157, 158). These studies show that MMPs play multiple important roles in inflammation and remodeling via their effects on leukocyte activation and recruitment to and through the respiratory tract, on inflammatory mediators, and on airway fibrosis. The recent finding of an association of asthma and AHR with ADAM33, a metalloproteinase with unknown function (159), helps to focus on the diversity of MMP effects and their importance in disease. ADAM33 is expressed in smooth muscle cells, fibroblasts, and myofibroblasts, but not in T lymphocytes or other

12 Feb 2004

16:14

802

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

inflammatory cells, suggesting ADAM33 plays a role in airway remodeling. Mutations of the ADAM33 gene may affect its function or concentration. ADAM33, if it functions like other ADAMs, will act on cell-surface molecules, possibly by shedding growth factors or cytokines (160). Therefore, ADAM33 mutations could affect smooth muscle or fibroblast growth, increase subepithelial fibrosis, and/or enhance inflammation. This molecule has the potential to expand our understanding of the local factors produced by structural cells in the lung that contribute to remodeling and inflammation, development, and persistence of asthma.

Epithelial Damage and Activation In asthma, the airway epithelium is abnormal. Epithelial damage and shedding is commonly observed; the sloughed cells can be found in the sputum and the patches of denuded epithelium observed in airway biopsies. Mast cells, DCs, lymphocytes, and eosinophils are readily identified abutting the basal cell layer. Epithelial activation increases mucus-secreting goblet cells, and increased expression of genes is involved in epithelial secretion (161–163). Epithelial damage and activation may in part be the result of a chronic inflammatory stimulus, but the enhanced responses observed in asthmatics may also reflect host susceptibility factors. Although it is not known if Th2 cytokines, eosinophils, or a viral or bacterial infection first injure the epithelium prior to asthma development, by the time asthma is diagnosed, the epithelium is bathed in inflammatory mediators from eosinophils, mast cells, and lymphocytes, many of which increase release of cytokines from and cause damage to epithelial cells (164). Asthmatic bronchial epithelial cells constitutively secrete more IL-8 and GM-CSF when compared with bronchial epithelial cells from nonasthmatics (165, 166). IL-4 and IL-13 increase release of IL-8 and GMCSF, RANTES, and CCL20 (MIP-3α) (167–169). These mediators activate several early components in the immune response: IL-8 stimulates neutrophil recruitment, CCL20 induces migration of immature DCs to the airway, and GM-CSF activates DCs and macrophages and promotes eosinophil survival. Mechanical stress comparable to the force generated from bronchoconstriction stimulates cultured epithelial cells to release TGF-β1, endothelin-1, and endothelin-2 (170). The release of endothelins may further enhance bronchospasm, stimulate smooth muscle proliferation, and activate the epithelium (171, 172). Protease-activated receptor2 (PAR-2) is a member of the PAR family of G-protein-coupled receptors that are activated by proteolysis by trypsin-like molecules. Among a number of cell types in asthma that have increased levels of PAR-2 staining, asthmatic epithelium has increased PAR-2 expression compared with that of normal epithelium (173). PAR-2 has a number of protective functions, including stimulating growth and repair of the epithelium, yet activation of PAR-2 also increases release of MMP-9 and GM-CSF from airway epithelial cells (174–176). All these responses of epithelial cells are designed to protect the host, and the injured epithelium should repair by inducing the growth of new cells. IL-13 stimulated epithelial proliferation in normal human bronchial epithelial cells (177), but it appears that damaged

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ASTHMA: PERSISTENCE AND PROGRESSION

P1: IKH

803

asthmatic epithelial cells may be restricted in their ability to proliferate in part because of increased expression of p21 (waf), a cyclin-dependent kinase inhibitor that negatively regulates cell growth (178). All these differences observed in the asthmatic epithelial cells may be a result of chronic stimulation, an effect of certain inflammatory mediators, or a property of the host that predisposes to disease. In an attempt to protect the epithelium, IL-13 converts it from an absorptive state to a secretory phenotype (179) and stimulates mucus production and secretion. Increased mucus production should provide more volume in which to bind and/or dissolve mediators, inhaled substances, and cellular debris. Yet IL-13 reduces ciliary beat frequency (180), and ciliated cells are sloughed. In asthma, a poorly functioning mucociliary escalator and increased mucus production result in mucus pooling, cough, and increased airway obstruction. As noted, increases in airway stress may enhance release of inflammatory mediators and further damage the epithelium, and these effects will further increase mucus production. In mice, Th2 cells stimulate airway epithelial mucus production, although Th1 cells activated in the airways do not (29), suggesting that increased mucus secretion evolved to assist with Th2 immunity. Yet mucus has become another effector pathway that is dysregulated in asthma because it leads to worse airway obstruction and promotes disease. CC10 Tg mice overexpressing IL-4, IL-5, IL-13, IL-10, and IL-9 all exhibit a marked increase in airway epithelial mucus staining (32– 35, 181). Mucus is not induced in IL-4Rα- or Stat6-deficient mice when Th2 cells are transferred and activated in the respiratory tract with inhaled specific antigen or when IL-13 is administered in the airway. In IL-13−/− mice that were immunized or received adoptive transfer of IL-13−/− Th2 cells and challenged with inhaled antigen, mucus was not induced (112, 182–184). Furthermore, in the overexpression models tested, mucus induction in all cases was IL-13 dependent (181, 185). Together, these studies show that in mice IL-13 is essential for mucus induction and that it is highly potent, with low levels inducing mucus effectively. Studies also show that IL-13 mediates its effects directly on airway epithelial cells and not by activating an intermediate hematopoietic cell (184, 186). Although Th2 cells that produce IL-4 (but not IL-13) were unable to induce mucus or an increase in mucin gene expression in IL-13−/− mice, there was epithelial cell hypertrophy, suggesting that other Th2 factors can activate the epithelium (184). In humans, IL-13 may not have an exclusive role in inducing mucus. In cultured human airway epithelial cells, it appears that IL-9, IL-4, and IL-13 can increase mucus staining and/or mucin gene induction (187–189). Some of the variability in these results may stem from different methods of epithelial culture and from the state of differentiation and activation of the epithelial cells, but ultimately, the difficulty in mimicking an in vivo environment may not permit identification of the precise pathways to mucus induction in humans. Holgate et al. (190) termed the interaction of the proinflammatory epithelium with smooth muscle and myofibroblasts as the “epithelial mesenchymal trophic unit.” This terminology emphasizes the importance of the interconnections between elements in the airway and their continued dialogue. The effects of the

12 Feb 2004

16:14

804

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

epithelium in asthma extend in many directions. Layered on the interactions of the epithelium with smooth muscle cells and fibroblasts are extensive signals between inflammatory, epithelial, and mesenchymal cells, and from epithelial and mesenchymal cells back to inflammatory cells, both in the respiratory tract and at other sites including the bone marrow, spleen, and peripheral blood.

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONCLUSIONS In a group of asthmatics given a two-week course of oral corticosteroids, there was a reduction in airway inflammatory cells and some inflammatory cytokines, but the treatment had no effect on TGF-β levels or collagen deposition (19). Thus, in this syndrome, control of inflammatory cells does not imply control over the profibrotic pathway. Asthma is a patterned response of the airways that requires both Th2 cells and specific host factors to establish disease. These host factors may range from signaling molecules and transcription factors to effector cytokines. Together, these elements lead to the generation of a memory response in the airways that will not relinquish. We expect that a reduction in inflammation should limit disease over time, but host factors are dysregulated in established asthma, and data that suggest why disease persists during anti-inflammatory therapy are accumulating. We have presented a number of possible scenarios to elucidate how CD4 Th2 cells, inflammation, and remodeling persist through the life of an asthmatic. It is not clear to us if these explanations are sufficient to support chronicity. Asthma is a serious and common condition for which there is rarely resolution, only control. For many patients, this is adequate and safe. But, for the severe asthmatic whose condition cannot be controlled with traditional anti-inflammatory therapy, we should pursue these questions further. A better understanding of the interactions between the components that amplify and promote persistent inflammation and progressive structural changes in the airway may be the key to limiting disease. ACKNOWLEDGMENTS The authors thank Drs. Anuradha Ray, Ann Haberman, and Robert Homer for critical review of the manuscript, Susan Ardito for administrative assistance, and Sarah Whitaker for artwork. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Ellis AG. 1908. The pathologic anatomy of bronchial asthma. Am. J. Med. Sci. 136:407–10 2. Kim J, Woods A, Becker-Dunn E, Bot-

tomly K. 1985. Distinct functional phenotypes of cloned Ia-restricted helper T cells. J. Exp. Med. 162:188–201 3. Mosmann TR, Cherwinski H, Bond MW,

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

ASTHMA: PERSISTENCE AND PROGRESSION

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

Giedlin MA, Coffman RL. 1986. Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins. J. Immunol. 136:2348–57 Robinson DS, Hamid Q, Ying S, Tsicopoulos A, Barkans J, et al. 1992. Predominant TH2-like bronchoalveolar Tlymphocyte population in atopic asthma. N. Engl. J. Med. 326:298–304 Laitinen A, Laitinen LA, Virtanen IT. 1997. Bronchial Biopsies. In Asthma, ed. PJ Barnes, MM Grunstein, AR Leff, AJ Woolcock, pp. 209–24. Philadelphia: Lippencott-Raven Homer RJ, Elias JA. 2000. Consequences of long-term inflammation. Airway remodeling. Clin. Chest Med. 21:331–43, ix Roberts CR, Okazawa M, Wiggs B, Pare P. 1997. Airway Wall Thickening. In Asthma, ed. PJ Barnes, MM Grunstein, AR Leff, AJ Woolcock, pp. 925–36. Philadelphia: Lippencott-Raven Roche WR, Beasley R, Williams JH, Holgate ST. 1989. Subepithelial fibrosis in the bronchi of asthmatics. Lancet 1:520–24 Black J. 1997. Airway Smooth Muscle in Asthma. In Asthma, ed. PJ Barnes, MM Grunstein, AR Leff, AJ Woolcock, pp. 809–22. Philadelphia: Lippencott-Raven Boushey HA. 1982. Bronchial hyperreactivity to sulfur dioxide: physiologic and political implications. J. Allergy Clin. Immunol. 69:335–38 Jeffery PK, Wardlaw AJ, Nelson FC, Collins JV, Kay AB. 1989. Bronchial biopsies in asthma. An ultrastructural, quantitative study and correlation with hyperreactivity. Am. Rev. Respir. Dis. 140: 1745–53 Cockcroft DW. 1997. Airway Responsiveness. In Asthma, ed. PJ Barnes, MM Grunstein, AR Leff, AJ Woolcock, pp. 1253–66. Philadelphia: LippencottRaven Pohunek P, Roche WR, Trzikova J, Kurdmann J, Warner JO. 2000. Eosinophlic

14.

15.

16.

17.

18.

19.

20.

21.

P1: IKH

805

inflammation in the bronchial mucosa in children with bronchial asthma. Eur. Respir. J. 11:160s Panhuysen CI, Vonk JM, Koeter GH, Schouten JP, van Altena R, et al. 1997. Adult patients may outgrow their asthma: a 25-year follow-up study. Am. J. Respir. Crit. Care Med. 155:1267–72 Warke TJ, Fitch PS, Brown V, Taylor R, Lyons JD, et al. 2002. Outgrown asthma does not mean no airways inflammation. Eur. Respir. J. 19:284–87 van den Toorn LM, Overbeek SE, de Jongste JC, Leman K, Hoogsteden HC, Prins JB. 2001. Airway inflammation is present during clinical remission of atopic asthma. Am. J. Respir. Crit. Care Med. 164:2107–13 Boulet LP, Turcotte H, Brochu A. 1994. Persistence of airway obstruction and hyperresponsiveness in subjects with asthma remission. Chest 105:1024–31 Olivieri D, Chetta A, Del Donno M, Bertorelli G, Casalini A, et al. 1997. Effect of short-term treatment with lowdose inhaled fluticasone propionate on airway inflammation and remodeling in mild asthma: a placebo-controlled study. Am. J. Respir. Crit. Care Med. 155:1864– 71 Chakir J, Shannon J, Molet S, Fukakusa M, Elias J, et al. 2003. Airway remodeling-associated mediators in moderate to severe asthma: effect of steroids on TGF-β, IL-11, IL-17, and type I and type III collagen expression. J. Allergy Clin. Immunol. 111:1293–98 Lundgren R, Soderberg M, Horstedt P, Stenling R. 1988. Morphological studies of bronchial mucosal biopsies from asthmatics before and after ten years of treatment with inhaled steroids. Eur. Respir. J. 1:883–89 Sont JK, Willems LN, Bel EH, van Krieken JH, Vandenbroucke JP, Sterk PJ. 1999. Clinical control and histopathologic outcome of asthma when using airway hyperresponsiveness as an additional

12 Feb 2004

16:14

806

22.

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

23.

24.

25.

26.

27.

28.

29.

30.

31.

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

guide to long-term treatment. The AMPUL Study Group. Am. J. Respir. Crit. Care Med. 159:1043–51 Lange P, Parner J, Vestbo J, Schnohr P, Jensen G. 1998. A 15-year follow-up study of ventilatory function in adults with asthma. N. Engl. J. Med. 339:1194–200 Corrigan CJ, Hartnell A, Kay AB. 1988. T lymphocyte activation in acute severe asthma. Lancet 1:1129–32 Walker C, Kaegi MK, Braun P, Blaser K. 1991. Activated T cells and eosinophilia in bronchoalveolar lavages from subjects with asthma correlated with disease severity. J. Allergy Clin. Immunol. 88:935– 42 Huang SK, Xiao HQ, Kleine-Tebbe J, Paciotti G, Marsh DG, et al. 1995. IL-13 expression at the sites of allergen challenge in patients with asthma. J. Immunol. 155:2688–94 Nakamura Y, Ghaffar O, Olivenstein R, Taha RA, Soussi-Gounni A, et al. 1999. Gene expression of the GATA-3 transcription factor is increased in atopic asthma. J. Allergy Clin. Immunol. 103:215–22 Finotto S, Neurath MF, Glickman JN, Qin S, Lehr HA, et al. 2002. Development of spontaneous airway changes consistent with human asthma in mice lacking T-bet. Science 295:336–38 Wills-Karp M. 1999. Immunologic basis of antigen-induced airway induced hyperresponsiveness. Annu. Rev. Immunol. 17:255–81 Cohn L, Homer RJ, Marinov A, Rankin J, Bottomly K. 1997. Induction of airway mucus production by T helper 2 (Th2) cells: a critical role for interleukin 4 in cell recruitment but not mucus production. J. Exp. Med. 186:1737–47 Cohn L, Tepper JS, Bottomly K. 1998. IL4-independent induction of airway hyperresponsiveness by Th2, but not Th1, cells. J. Immunol. 161:3813–16 Zhang DH, Yang L, Cohn L, Parkyn L, Homer R, et al. 1999. Inhibition of allergic inflammation in a murine model of asthma

32.

33.

34.

35.

36.

37.

38.

39.

by expression of a dominant-negative mutant of GATA-3. Immunity 11:473–82 Lee JJ, McGarry MP, Farmer SC, Denzler KL, Larson KA, et al. 1997. Interleukin-5 expression in the lung epithelium of transgenic mice leads to pulmonary changes pathognomonic of asthma. J. Exp. Med. 185:2143–56 Rankin JA, Picarella D, Tarallo A, DiCosmo B, Whitsett JA, et al. 1994. In vivo effects of the overexpression of IL-4 in the lungs of transgenic mice. Am. J. Resp. Crit. Care Med. 149:1071a. (Abstr.) Temann UA, Geba GP, Rankin JA, Flavell RA. 1998. Expression of interleukin 9 in the lungs of transgenic mice causes airway inflammation, mast cell hyperplasia, and bronchial hyperresponsiveness. J. Exp. Med. 188:1307–20 Zhu Z, Homer R, Wang Z, Chen Q, Geba G, et al. 1999. Transgenic expression of IL-13 in murine lung causes airway inflammation, mucus hypersecretion, subendothelial fibrosis, eotaxin production and airways hyperresponsiveness to methacholine. J. Clin. Investig. 103: 779–88 Kuperman D, Schofield B, Wills-Karp M, Grusby MJ. 1998. Signal transducer and activator of transcription factor 6 (Stat6)deficient mice are protected from antigeninduced airway hyperresponsiveness and mucus production. J. Exp. Med. 187:939– 48 Cohn L, Homer RJ, Niu N, Bottomly K. 1999. T helper 1 cells and interferon gamma regulate allergic airway inflammation and mucus production. J. Exp. Med. 190:1309–18 Huang TJ, MacAry PA, Eynott P, Moussavi A, Daniel KC, et al. 2001. Allergenspecific Th1 cells counteract efferent Th2 cell-dependent bronchial hyperresponsiveness and eosinophilic inflammation partly via IFN-γ . J. Immunol. 166:207–17 Blyth DI, Wharton TF, Pedrick MS, Savage TJ, Sanjar S. 2000. Airway

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

ASTHMA: PERSISTENCE AND PROGRESSION

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

40.

41.

42.

43.

44.

45.

46.

subepithelial fibrosis in a murine model of atopic asthma: suppression by dexamethasone or anti-interleukin-5 antibody. Am. J. Respir. Cell Mol. Biol. 23:241–46 Renz H, Lack G, Saloga J, Schwinzer R, Bradley K, et al. 1994. Inhibition of IgE production and normalization of airways responsiveness by sensitized CD8 T cells in a mouse model of allergen-induced sensitization. J. Immunol. 152:351–60 Iwamoto I, Nakajima H, Endo H, Yoshida S. 1993. Interferon γ regulates antigeninduced eosinophil recruitment into the mouse airways by inhibiting the infiltration of CD4+ T cells. J. Exp. Med. 177:573–76 Hofstra CL, Van Ark I, Hofman G, Nijkamp FP, Jardieu PM, Van Oosterhout AJ. 1998. Differential effects of endogenous and exogenous interferon-gamma on immunoglobulin E, cellular infiltration, and airway responsiveness in a murine model of allergic asthma. Am. J. Respir. Cell Mol. Biol. 19:826–35 Randolph DA, Carruthers CJ, Szabo SJ, Murphy KM, Chaplin DD. 1999. Modulation of airway inflammation by passive transfer of allergen-specific Th1 and Th2 cells in a mouse model of asthma. J. Immunol. 162:2375–83 Holtzman MJ, Sampath D, Castro M, Look DC, Jayaraman S. 1996. The onetwo of T helper cells: does interferongamma knock out the Th2 hypothesis for asthma? Am. J. Respir. Cell Mol. Biol. 14: 316–18 Hansen G, Berry G, DeKruyff RH, Umetsu DT. 1999. Allergen-specific Th1 cells fail to counterbalance Th2 cellinduced airway hyperreactivity but cause severe airway inflammation. J. Clin. Investig. 103:175–83 Stephens R, Randolph DA, Huang G, Holtzman MJ, Chaplin DD. 2002. Antigen-nonspecific recruitment of Th2 cells to the lung as a mechanism for viral infection-induced allergic asthma. J. Immunol. 169:5458–67

P1: IKH

807

47. Seymour BWP, Gershwin LJ, Coffman RL. 1998. Aerosol-induced immunoglobulin (Ig)-E unresponsiveness to ovalbumin does not require CD8+ or T cell receptor (TCR)-γ /δ + T cells or interferon (IFN)-γ in a murine model of allergen sensitization. J. Exp. Med. 187:721–31 48. McMenamin C, Pimm C, McKersey M, Holt PG. 1994. Regulation of IgE responses to inhaled antigen in mice by antigen-specific gamma-delta T cells. Science 265:1869–71 49. Wolvers DA, Coenen-de Roo CJ, Mebius RE, van der Cammen MJ, Tirion F, et al. 1999. Intranasally induced immunological tolerance is determined by characteristics of the draining lymph nodes: studies with OVA and human cartilage gp-39. J. Immunol. 162:1994–98 50. McMenamin C, Girn B, Holt PG. 1992. The distribution of IgE plasma cells in lymphoid and non-lymphoid tissues of high-IgE responder rats: differential localization of antigen-specific and ‘bystander’ components of the IgE response to inhaled antigen. Immunology 77:592–96 51. Tsitoura DC, DeKruyff RH, Lamb JR, Umetsu DT. 1999. Intranasal exposure to protein antigen induces immunological tolerance mediated by functionally disabled CD4+ T cells. J. Immunol. 163:2592–600 52. McMenamin C, Holt PG. 1993. The natural immune response to inhaled soluble protein antigens involves major histocompatibility complex (MHC) class I-restricted CD8+ T cell-mediated but MHC class II-restricted CD4+ T celldependent immune deviation resulting in selective suppression of immunolobulin E production. J. Exp. Med. 178:889–99 53. Kaya Z, Dohmen KM, Wang Y, Schlichting J, Afanasyeva M, et al. 2002. Cutting edge: a critical role for IL-10 in induction of nasal tolerance in experimental autoimmune myocarditis. J. Immunol. 168:1552–56 54. Akbari O, DeKruyff RH, Umetsu DT.

12 Feb 2004

16:14

808

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

55.

56.

57.

58.

59. 60.

61.

62.

63.

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

2001. Pulmonary dendritic cells producing IL-10 mediate tolerance induced by respiratory exposure to antigen. Nat. Immunol. 2:725–31 Haneda K, Sano K, Tamura G, Shirota H, Ohkawara Y, et al. 1999. Transforming growth factor-beta secreted from CD4(+) T cells ameliorates antigeninduced eosinophilic inflammation. A novel high-dose tolerance in the trachea. Am. J. Respir. Cell Mol. Biol. 21:268–74 Akbari O, Freeman GJ, Meyer EH, Greenfield EA, Chang TT, et al. 2002. Antigenspecific regulatory T cells develop via the ICOS-ICOS-ligand pathway and inhibit allergen-induced airway hyperreactivity. Nat. Med. 8:1024–32 Iwasaki A, Kelsall BL. 1999. Freshly isolated Peyer’s patch, but not spleen, dendritic cells produce interleukin 10 and induce the differentiation of T helper type 2 cells. J. Exp. Med. 190:229–39 Stumbles PA, Thomas JA, Pimm CL, Lee PT, Venaille TJ, et al. 1998. Resting respiratory tract dendritic cells preferentially stimulate T helper cell type 2 (Th2) responses and require obligatory cytokine signals for induction of Th1 immunity. J. Exp. Med. 188:2019–31 Strober W, Kelsall B, Marth T. 1998. Oral tolerance. J. Clin. Immunol. 18:1–30 Janeway CA, Bottomly K. 1994. Signals and signs for lymphocyte responses. Cell 76:275–85 Koppelman B, Neefjes JJ, de Vries JE, de Waal Malefyt R. 1997. Interleukin-10 down-regulates MHC class II αβ peptide complexes at the plasma membrane of monocytes by affecting arrival and recycling. Immunity 7:861–71 Chang CC, Ciubotariu R, Manavalan JS, Yuan J, Colovai AI, et al. 2002. Tolerization of dendritic cells by T(S) cells: the crucial role of inhibitory receptors ILT3 and ILT4. Nat. Immunol. 3:237–43 Martin E, O’Sullivan B, Low P, Thomas R. 2003. Antigen-specific suppression of a primed immune response by dendritic

64.

65.

66.

67.

68.

69.

70.

71.

72.

cells mediated by regulatory T cells secreting interleukin-10. Immunity 18:155– 67 Holt PG. 2000. Antigen presentation in the lung. Am. J. Respir. Crit. Care Med. 162:S151–56 Kung TT, Stelts D, Zurcher JA, Jones H, Umland SP, et al. 1995. Mast cells modulate allergic pulmonary eosinophilia in mice. Am. J. Respir. Cell Mol. Biol. 12:404–9 Zuany-Amorim C, Ruffie C, Haile S, Vargaftig BB, Pereira P, Pretolani M. 1998. Requirement for γ δ T cells in allergic airway inflammation. Science 280:1265–67 Akbari O, Stock P, Meyer E, Kronenberg M, Sidobre S, et al. 2003. Essential role of NKT cells producing IL-4 and IL-13 in the development of allergen-induced airway hyperreactivity. Nat. Med. 9:582– 88 Hewitt CR, Brown AP, Hart BJ, Pritchard DI. 1995. A major house dust mite allergen disrupts the immunoglobulin E network by selectively cleaving CD23: innate protection by antiproteases. J. Exp. Med. 182:1537–44 Machado DC, Horton D, Harrop R, Peachell PT, Helm BA. 1996. Potential allergens stimulate the release of mediators of the allergic response from cells of mast cell lineage in the absence of sensitization with antigen-specific IgE. Eur. J. Immunol. 26:2972–80 Song Z, Casolaro V, Chen R, Georas SN, Monos D, Ono SJ. 1996. Polymorphic nucleotides within the human IL-4 promoter that mediate overexpression of the gene. J. Immunol. 156:424–29 Rosenwasser LJ, Borish L. 1997. Genetics of atopy and asthma: the rationale behind promoter-based candidate gene studies (IL-4 and IL-10). Am. J. Respir. Crit. Care Med. 156:S152–55 Kawashima T, Noguchi E, Arinami T, Yamakawa-Kobayashi K, Nakagawa H, et al. 1998. Linkage and association of an interleukin 4 gene polymorphism with

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

ASTHMA: PERSISTENCE AND PROGRESSION

73.

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

74.

75.

76.

77.

78.

79.

80.

atopic dermatitis in Japanese families. J. Med. Genet. 35:502–4 Heinzmann A, Mao XQ, Akaiwa M, Kreomer RT, Gao PS, et al. 2000. Genetic variants of IL-13 signalling and human asthma and atopy. Hum. Mol. Genet. 9:549–59 van der Pouw Kraan TC, van Veen A, Boeije LC, van Tuyl SA, de Groot ER, et al. 1999. An IL-13 promoter polymorphism associated with increased risk of allergic asthma. Genes Immun. 1:61–65 Lambrecht BN, Carro-Muino I, Vermaelen K, Pauwels RA. 1999. Allergeninduced changes in bone-marrow progenitor and airway dendritic cells in sensitized rats. Am. J. Respir. Cell Mol. Biol. 20:1165–74 Moller GM, Overbeek SE, Van HeldenMeeuwsen CG, Van Haarst JM, Prens EP, et al. 1996. Increased numbers of dendritic cells in the bronchial mucosa of atopic asthmatic patients: downregulation by inhaled corticosteroids. Clin. Exp. Allergy 26:517–24 Tunon-De-Lara JM, Redington AE, Bradding P, Church MK, Hartley JA, et al. 1996. Dendritic cells in normal and asthmatic airways: expression of the alpha subunit of the high affinity immunoglobulin E receptor (Fc epsilon RI-alpha). Clin. Exp. Allergy 26:648–55 Broide DH, Lotz M, Cuomo AJ, Coburn DA, Federman EC, Wasserman SI. 1992. Cytokines in symptomatic asthma airways. J. Allergy Clin. Immunol. 89:958– 67 Huh JC, Strickland DH, Jahnsen FL, Turner DJ, Thomas JA, et al. 2003. Bidirectional interactions between antigenbearing respiratory tract dendritic cells (DCs) and T cells precede the late phase reaction in experimental asthma: DC activation occurs in the airway mucosa but not in the lung parenchyma. J. Exp. Med. 198:19–30 Cella M, Scheidegger D, PalmerLehmann K, Lane P, Lanzavecchia A,

81.

82.

83.

84.

85.

86.

87.

88.

P1: IKH

809

Alber G. 1996. Ligation of CD40 on dendritic cells triggers production of high levels of interleukin-12 and enhances T cell stimulatory capacity: T-T help via APC activation. J. Exp. Med. 184:747– 52 Julia V, Hessel EM, Malherbe L, Glaichenhaus N, O’Garra A, Coffman RL. 2002. A restricted subset of dendritic cells captures airborne antigens and remains able to activate specific T cells long after antigen exposure. Immunity 16:271–83 Salek-Ardakani S, Song J, Halteman BS, Jember AG, Akiba H, et al. 2003. OX40 (CD134) controls memory T helper 2 cells that drive lung inflammation. J. Exp. Med. 198:315–24 Gonzalo JA, Tian J, Delaney T, Corcoran J, Rottman JB, et al. 2001. ICOS is critical for T helper cell-mediated lung mucosal inflammatory responses. Nat. Immunol. 2:597–604 Hurst SD, Seymour BW, Muchamuel T, Kurup VP, Coffman RL. 2001. Modulation of inhaled antigen-induced IgE tolerance by ongoing Th2 responses in the lung. J. Immunol. 166:4922–30 Holt PG, Batty JE, Turner KJ. 1981. Inhibition of specific IgE responses in mice by pre-exposure to inhaled antigen. Immunology 42:409–17 Schwarze J, Hamelmann E, Bradley KL, Takeda K, Gelfand EW. 1997. Respiratory syncytial virus infection results in airway hyperresponsiveness and enhanced airway sensitization to allergen. J. Clin. Investig. 100:226–33 Swirski FK, Sajic D, Robbins CS, Gajewska BU, Jordana M, Stampfli MR. 2002. Chronic exposure to innocuous antigen in sensitized mice leads to suppressed airway eosinophilia that is reversed by granulocyte macrophage colony-stimulating factor. J. Immunol. 169:3499–506 Constant SL, Brogdon JL, Piggott DA, Herrick CA, Visintin I, et al. 2002. Resident lung antigen-presenting cells have the capacity to promote Th2 T cell

12 Feb 2004

16:14

810

89.

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

90.

91.

92.

93.

94.

95.

96.

97.

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

differentiation in situ. J. Clin. Investig. 110:1441–48 Jayaraman S, Castro M, O’Sullivan M, Bragdon MJ, Holtzman MJ. 1999. Resistance to Fas-mediated T cell apoptosis in asthma. J. Immunol. 162:1717–22 Spinozzi F, Fizzotti M, Agea E, Piattoni S, Droetto S, et al. 1998. Defective expression of Fas messenger RNA and Fas receptor on pulmonary T cells from patients with asthma. Ann. Intern. Med. 128:363– 69 Van Parijs L, Abbas AK. 1998. Homeostasis and self-tolerance in the immune system: turning lymphocytes off. Science 280:243–48 Varadhachary AS, Perdow SN, Hu C, Ramanarayanan M, Salgame P. 1997. Differential ability of T cell subsets to undergo activation-induced cell death. Proc. Natl. Acad. Sci. USA 94:5778–83 Cormican L, O’Sullivan S, Burke CM, Poulter LW. 2001. IFN-γ but not IL-4 T cells of the asthmatic bronchial wall show increased incidence of apoptosis. Clin. Exp. Allergy 31:731–39 Akdis M, Trautmann A, Klunker S, Daigle I, Kucuksezer UC, et al. 2003. T helper (Th) 2 predominance in atopic diseases is due to preferential apoptosis of circulating memory/effector Th1 cells. FASEB J. 17:1026–35 Refaeli Y, Van Parijs L, Alexander SI, Abbas AK. 2002. Interferon γ is required for activation-induced death of T lymphocytes. J. Exp. Med. 196:999–1005 Wu CY, Kirman JR, Rotte MJ, Davey DF, Perfetto SP, et al. 2002. Distinct lineages of TH1 cells have differential capacities for memory cell generation in vivo. Nat. Immunol. 3:852–58 Coyle AJ, Tsuyuki S, Bertrand C, Huang S, Aguet M, et al. 1996. Mice lacking the IFN-γ receptor have impaired ability to resolve a lung eosinophilic inflammatory response associated with a prolonged capacity of T cells to exhibit a Th2 cytokine profile. J. Immunol. 156:2680–85

98. Vella AT, Dow S, Potter TA, Kappler J, Marrack P. 1998. Cytokine-induced survival of activated T cells in vitro and in vivo. Proc. Natl. Acad. Sci. USA 95:3810– 15 99. Cohn L, Herrick C, Niu N, Homer R, Bottomly K. 2001. IL-4 promotes airway eosinophilia by suppressing IFN-γ production: defining a novel role for IFN-γ in the regulation of allergic airway inflammation. J. Immunol. 166:2760–67 100. Elias JA, Lee CG, Zheng T, Ma B, Homer RJ, Zhu Z. 2003. New insights into the pathogenesis of asthma. J. Clin. Investig. 111:291–97 101. Ying S, Durham SR, Corrigan CJ, Hamid Q, Kay AB. 1995. Phenotype of cells expressing mRNA for TH2-type (interleukin 4 and interleukin 5) and TH1-type (interleukin 2 and interferon γ ) cytokines in bronchoalveolar lavage and bronchial biopsies from atopic asthmatic and normal control subjects. Am. J. Respir. Cell Mol. Biol. 12:477–87 102. Ferrick DA, Schrenzel MD, Mulvania T, Hsieh B, Ferlin WG, Lepper H. 1995. Differential production of interferon-γ and interleukin 4 in response to Th1- and Th2stimulating pathogens by γ -δ T cells. Nature 373:255–57 103. Yoshimoto T, Bendelac A, Watson C, HuLi J, Paul WE. 1995. Role of NK1.1+ T cells in a TH2 response and in immunoglobulin E production. Science 270:1845–47 104. Hoshino T, Winkler-Pickett RT, Mason AT, Ortaldo JR, Young HA. 1999. IL-13 production by NK cells: IL-13-producing NK and T cells are present in vivo in the absence of IFN-γ . J. Immunol. 162:51–59 105. Fort MM, Cheung J, Yen D, Li J, Zurawski SM, et al. 2001. IL-25 induces IL-4, IL-5, and IL-13 and Th2-associated pathologies in vivo. Immunity 15:985–95 106. Rothenberg ME. 1998. Eosinophilia. N. Engl. J. Med. 338:1592–600 107. Ikeda K, Nakajima H, Suzuki K, Kagami S, Hirose K, et al. 2003. Mast cells

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

ASTHMA: PERSISTENCE AND PROGRESSION

108.

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

109.

110.

111.

112.

113.

114.

115.

produce interleukin-25 upon Fc epsilon RI-mediated activation. Blood 101:3594– 96 Nakajima H, Iwamoto I, Tomoe S, Matsumara R, Tomioka H, et al. 1992. CD4+ T-lymphocytes and interleukin-5 mediate antigen-induced eosinophil infiltration into mouse trachea. Am. Rev. Respir. Dis. 146:374–77 Mattes J, Yang M, Mahalingam S, Kuehr J, Webb DC, et al. 2002. Intrinsic defect in T cell production of interleukin (IL)-13 in the absence of both IL-5 and eotaxin precludes the development of eosinophilia and airways hyperreactivity in experimental asthma. J. Exp. Med. 195:1433–44 Mould AW, Matthaei KI, Young IG, Foster PS. 1997. Relationship between interleukin-5 and eotaxin in regulating blood and tissue eosinophilia in mice. J. Clin. Investig. 99:1064–71 Mathew A, MacLean JA, DeHaan E, Tager AM, Green FH, Luster AD. 2001. Signal transducer and activator of transcription 6 controls chemokine production and T helper cell type 2 cell trafficking in allergic pulmonary inflammation. J. Exp. Med. 193:1087–96 Mattes J, Yang M, Siqueira A, Clark K, MacKenzie J, et al. 2001. IL-13 induces airways hyperreactivity independently of the IL-4R alpha chain in the allergic lung. J. Immunol. 167:1683–92 Morokata T, Ishikawa J, Ida K, Yamada T. 1999. C57BL/6 mice are more susceptible to antigen-induced pulmonary eosinophilia than BALB/c mice, irrespective of systemic T helper 1/T helper 2 responses. Immunology 98:345–51 Bousquet J, Chanez P, LaCoste JY, Barneon G, Ghavanian N, et al. 1990. Eosinophilic inflammation in asthma. N. Engl. J. Med. 323:1033–39 Corry DB, Folkesson HG, Warnock ML, Erle DJ, Matthay MA, et al. 1996. Interleukin 4, but not interleukin 5 or eosinophils, is required in a murine model of acute airway hyperreactivity. J. Exp.

116.

117.

118.

119.

120.

121.

122.

123.

P1: IKH

811

Med. 183:109–17. Erratum. 1997. J. Exp. Med. 185(9):1715 Foster PS, Hogan SP, Ramsey AJ, Matthaei KI, Young IG. 1996. Interleukin 5 deficiency abolishes eosinophilia, airways hyperreactivity, and lung damage in a mouse model. J. Exp. Med. 183:195–201 Gerwin N, Gonzalo JA, Lloyd C, Coyle AJ, Reiss Y, et al. 1999. Prolonged eosinophil accumulation in allergic lung interstitium of ICAM-2 deficient mice results in extended hyperresponsiveness. Immunity 10:9–19 Hogan SP, Mould A, Kikutani H, Ramsay AJ, Foster PS. 1997. Aeroallergeninduced eosinophilic inflammation, lung damage, and airways hyperreactivity in mice can occur independently of IL-4 and allergen-specific immunoglobulins. J. Clin. Investig. 99:1326–39 Leckie MJ, ten Brinke A, Khan J, Diamant Z, O’Connor BJ, et al. 2000. Effects of an interleukin-5 blocking monoclonal antibody on eosinophils, airway hyper-responsiveness, and the late asthmatic response. Lancet 356:2144–48 Kips JC, O’Connor BJ, Langley SJ, Woodcock A, Kerstjens HA, et al. 2003. Effect of SCH55700, a humanized antihuman interleukin-5 antibody, in severe persistent asthma: a pilot study. Am. J. Respir. Crit. Care Med. 167:1655–59 Flood-Page PT, Menzies-Gow AN, Kay AB, Robinson DS. 2003. Eosinophil’s role remains uncertain as anti-interleukin5 only partially depletes numbers in asthmatic airway. Am. J. Respir. Crit. Care Med. 167:199–204 Gregory B, Kirchem A, Phipps S, Gevaert P, Pridgeon C, et al. 2003. Differential regulation of human eosinophil IL-3, IL5, and GM-CSF receptor alpha-chain expression by cytokines: IL-3, IL-5, and GM-CSF down-regulate IL-5 receptor alpha expression with loss of IL-5 responsiveness, but up-regulate IL-3 receptor alpha expression. J. Immunol. 170:5359–66 Liu LY, Sedgwick JB, Bates ME, Vrtis RF,

12 Feb 2004

16:14

812

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

124.

125.

126.

127.

128.

129.

130.

131.

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

Gern JE, et al. 2002. Decreased expression of membrane IL-5 receptor alpha on human eosinophils: I. Loss of membrane IL-5 receptor alpha on airway eosinophils and increased soluble IL-5 receptor alpha in the airway after allergen challenge. J. Immunol. 169:6452–58 Shi HZ, Humbles A, Gerard C, Jin Z, Weller PF. 2000. Lymph node trafficking and antigen presentation by endobronchial eosinophils. J. Clin. Investig. 105:945–53 Dibbert B, Daigle I, Braun D, Schranz C, Weber M, et al. 1998. Role for Bcl-xL in delayed eosinophil apoptosis mediated by granulocyte-macrophage colony-stimulating factor and interleukin5. Blood 92:778–83 Dewson G, Walsh GM, Wardlaw AJ. 1999. Expression of Bcl-2 and its homologues in human eosinophils. Modulation by interleukin-5. Am. J. Respir. Cell Mol. Biol. 20:720–28 Tsuyuki S, Bertrand C, Erard F, Trifilieff A, Tsuyuki J, et al. 1995. Activation of the Fas receptor on lung eosinophils leads to apoptosis and the resolution of eosinophilic inflammation of the airways. J. Clin. Investig. 96:2924–31 Mann JS, Holgate ST, Renwick AG, Cushley MJ. 1986. Airway effects of purine nucleosides and nucleotides and release with bronchial provocation in asthma. J. Appl. Physiol. 61:1667–76 Driver AG, Kukoly CA, Ali S, Mustafa SJ. 1993. Adenosine in bronchoalveolar lavage fluid in asthma. Am. Rev. Respir. Dis. 148:91–97 Blackburn MR, Lee CG, Young HW, Zhu Z, Chunn JL, et al. 2003. Adenosine mediates IL-13-induced inflammation and remodeling in the lung and interacts in an IL-13-adenosine amplification pathway. J. Clin. Investig. 112:332–44 Sousa AR, Lane SJ, Nakhosteen JA, Yoshimura T, Lee TH, Poston RN. 1994. Increased expression of the monocyte chemoattractant protein-1 in bronchial tis-

132.

133.

134.

135.

136.

137.

138.

139.

140.

sue from asthmatic subjects. Am. J. Respir. Cell Mol. Biol. 10:142–47 Zhu Z, Ma B, Zheng T, Homer RJ, Lee CG, et al. 2002. IL-13-induced chemokine responses in the lung: role of CCR2 in the pathogenesis of IL-13-induced inflammation and remodeling. J. Immunol. 168:2953–62 Zhong H, Belardinelli L, Maa T, Feoktistov I, Biaggioni I, Zeng D. 2003. A2B Adenosine Receptors Increase Cytokine Release by Bronchial Smooth Muscle Cells. Am. J. Respir. Cell Mol. Biol. In press Carr MW, Roth SJ, Luther E, Rose SS, Springer TA. 1994. Monocyte chemoattractant protein 1 acts as a T-lymphocyte chemoattractant. Proc. Natl. Acad. Sci. USA 91:3652–56 Karpus WJ, Lukacs NW, Kennedy KJ, Smith WS, Hurst SD, Barrett TA. 1997. Differential CC chemokine-induced enhancement of T helper cell cytokine production. J. Immunol. 158:4129–36 Gu L, Tseng S, Horner RM, Tam C, Loda M, Rollins BJ. 2000. Control of TH2 polarization by the chemokine monocyte chemoattractant protein-1. Nature 404:407–11 Lu B, Rutledge BJ, Gu L, Fiorillo J, Lukacs NW, et al. 1998. Abnormalities in monocyte recruitment and cytokine expression in monocyte chemoattractant protein 1-deficient mice. J. Exp. Med. 187:601–8 Traynor TR, Kuziel WA, Toews GB, Huffnagle GB. 2000. CCR2 expression determines T1 versus T2 polarization during pulmonary Cryptococcus neoformans infection. J. Immunol. 164:2021–27 Conti P, Boucher W, Letourneau R, Feliciani C, Reale M, et al. 1995. Monocyte chemotactic protein-1 provokes mast cell aggregation and [3H]5HT release. Immunology 86:434–40 Smith RE, Strieter RM, Phan SH, Kunkel SL. 1996. C-C chemokines: novel mediators of the profibrotic inflammatory

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

ASTHMA: PERSISTENCE AND PROGRESSION

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

141.

142.

143.

144.

145.

146.

147.

148.

149.

150.

response to bleomycin challenge. Am. J. Respir. Cell Mol. Biol. 15:693–702 Hogaboam CM, Bone-Larson CL, Lipinski S, Lukacs NW, Chensue SW, et al. 1999. Differential monocyte chemoattractant protein-1 and chemokine receptor 2 expression by murine lung fibroblasts derived from Th1- and Th2-type pulmonary granuloma models. J. Immunol. 163:2193–201 Madri JA, Graesser D. 2000. Cell migration in the immune system: the evolving inter-related roles of adhesion molecules and proteinases. Dev. Immunol. 7:103–16 Kelly EA, Jarjour NN. 2003. Role of matrix metalloproteinases in asthma. Curr. Opin. Pulm. Med. 9:28–33 Zheng T, Zhu Z, Wang Z, Homer RJ, Ma B, et al. 2000. Inducible targeting of IL-13 to the adult lung causes matrix metalloproteinase- and cathepsindependent emphysema. J. Clin. Investig. 106:1081–93 Kumagai K, Ohno I, Okada S, Ohkawara Y, Suzuki K, et al. 1999. Inhibition of matrix metalloproteinases prevents allergeninduced airway inflammation in a murine model of asthma. J. Immunol. 162:4212– 19 Kelly EA, Busse WW, Jarjour NN. 2000. Increased matrix metalloproteinase-9 in the airway after allergen challenge. Am. J. Respir. Crit. Care Med. 162:1157–61 Atkinson JJ, Senior RM. 2003. Matrix metalloproteinase-9 in lung remodeling. Am. J. Respir. Cell Mol. Biol. 28:12–24 Mautino G, Capony F, Bousquet J, Vignola AM. 1999. Balance in asthma between matrix metalloproteinases and their inhibitors. J. Allergy Clin. Immunol. 104:530–33 Vermaelen K, Pauwels R. 2003. Accelerated airway dendritic cell maturation, trafficking, and elimination in a mouse model of asthma. Am. J. Respir. Cell Mol. Biol. 29:405–9 Vermaelen KY, Cataldo D, Tournoy K, Maes T, Dhulst A, et al. 2003. Ma-

151.

152.

153.

154.

155.

156.

157.

158.

159.

P1: IKH

813

trix metalloproteinase-9-mediated dendritic cell recruitment into the airways is a critical step in a mouse model of asthma. J. Immunol. 171:1016–22 Lanone S, Zheng T, Zhu Z, Liu W, Lee CG, et al. 2002. Overlapping and enzyme-specific contributions of matrix metalloproteinases-9 and -12 in IL-13induced inflammation and remodeling. J. Clin. Investig. 110:463–74 Lee CG, Homer RJ, Zhu Z, Lanone S, Wang X, et al. 2001. Interleukin-13 induces tissue fibrosis by selectively stimulating and activating transforming growth factor β(1). J. Exp. Med. 194:809–21 Clark DA, Coker R. 1998. Transforming growth factor-β (TGF-β). Int. J. Biochem. Cell Biol. 30:293–98 Baram D, Vaday GG, Salamon P, Drucker I, Hershkoviz R, Mekori YA. 2001. Human mast cells release metalloproteinase9 on contact with activated T cells: juxtacrine regulation by TNF-α. J. Immunol. 167:4008–16 Ferrari-Lacraz S, Nicod LP, Chicheportiche R, Welgus HG, Dayer JM. 2001. Human lung tissue macrophages, but not alveolar macrophages, express matrix metalloproteinases after direct contact with activated T lymphocytes. Am. J. Respir. Cell Mol. Biol. 24:442–51 Corry DB, Rishi K, Kanellis J, Kiss A, Song L-z, et al. 2002. Decreased allergic lung inflammatory cell egression and increased susceptibility to asphyxiation in MMP2-deficiency. Nat. Immunol. 3:347– 53 Sheppard D. 2002. Killed by leukocytes that don’t know when to leave. Nat. Immunol. 3:337–38 McQuibban GA, Gong JH, Wong JP, Wallace JL, Clark-Lewis I, Overall CM. 2002. Matrix metalloproteinase processing of monocyte chemoattractant proteins generates CC chemokine receptor antagonists with anti-inflammatory properties in vivo. Blood 100:1160–67 Van Eerdewegh P, Little RD, Dupuis J, Del

12 Feb 2004

16:14

814

160.

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

161.

162.

163.

164.

165.

166.

167.

AR

COHN

AR210-IY22-27.tex

¥

ELIAS

¥

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHUPP

Mastro RG, Falls K, et al. 2002. Association of the ADAM33 gene with asthma and bronchial hyperresponsiveness. Nature 418:426–30 Shapiro SD, Owen CA. 2002. ADAM-33 surfaces as an asthma gene. N. Engl. J. Med. 347:936–38 Hoshino M, Morita S, Iwashita H, Sagiya Y, Nagi T, et al. 2002. Increased expression of the human Ca2+-activated Clchannel 1 (CaCC1) gene in the asthmatic airway. Am. J. Respir. Crit. Care Med. 165:1132–36 Zhou Y, Dong Q, Louahed J, Dragwa C, Savio D, et al. 2001. Characterization of a calcium-activated chloride channel as a shared target of Th2 cytokine pathways and its potential involvement in asthma. Am. J. Respir. Cell Mol. Biol. 25:486–91 Nikolaidis NM, Zimmermann N, King NE, Mishra A, Pope SM, et al. 2003. Trefoil factor-2 is an allergen-induced gene regulated by TH2 cytokines and STAT6 in the lung. Am. J. Respir. Cell Mol. Biol. 29:458–64 White SR. 1997. Epithelium as a target. In Asthma, ed. PJ Barnes, MM Grunstein, AR Leff, AJ Woolcock, pp. 875– 900. Philadelphia: Lippencott-Raven Davies RJ, Wang JH, Trigg CJ, Devalia JL. 1995. Expression of granulocyte/ macrophage-colony-stimulating factor, interleukin-8 and RANTES in the bronchial epithelium of mild asthmatics is down-regulated by inhaled beclomethasone dipropionate. Int. Arch. Allergy Immunol. 107:428–29 Sousa AR, Poston RN, Lane SJ, Nakhosteen JA, Lee TH. 1993. Detection of GMCSF in asthmatic bronchial epithelium and decrease by inhaled corticosteroids. Am. Rev. Respir. Dis. 147:1557–61 Nakamura Y, Azuma M, Okano Y, Sano T, Takahashi T, et al. 1996. Upregulatory effects of interleukin-4 and interleukin13 but not interleukin-10 on granulocyte/macrophage colony-stimulating factor production by human bronchial ep-

168.

169.

170.

171.

172.

173.

174.

175.

176.

ithelial cells. Am. J. Respir. Cell Mol. Biol. 15:680–87 Lordan JL, Bucchieri F, Richter A, Konstantinidis A, Holloway JW, et al. 2002. Cooperative effects of Th2 cytokines and allergen on normal and asthmatic bronchial epithelial cells. J. Immunol. 169:407–14 Reibman J, Hsu Y, Chen LC, Bleck B, Gordon T. 2003. Airway epithelial cells release MIP-3α/CCL20 in response to cytokines and ambient particulate matter. Am. J. Respir. Cell Mol. Biol. 28:648– 54 Tschumperlin DJ, Shively JD, Kikuchi T, Drazen JM. 2003. Mechanical stress triggers selective release of fibrotic mediators from bronchial epithelium. Am. J. Respir. Cell Mol. Biol. 28:142–49 Murlas CG, Gulati A, Singh G, Najmabadi F. 1995. Endothelin-1 stimulates proliferation of normal airway epithelial cells. Biochem. Biophys. Res. Commun. 212:953–59 Fagan KA, McMurtry IF, Rodman DM. 2001. Role of endothelin-1 in lung disease. Respir. Res. 2:90–101 Knight DA, Lim S, Scaffidi AK, Roche N, Chung KF, et al. 2001. Protease-activated receptors in human airways: upregulation of PAR-2 in respiratory epithelium from patients with asthma. J. Allergy Clin. Immunol. 108:797–803 Vliagoftis H, Befus AD, Hollenberg MD, Moqbel R. 2001. Airway epithelial cells release eosinophil survivalpromoting factors (GM-CSF) after stimulation of proteinase-activated receptor 2. J. Allergy Clin. Immunol. 107:679– 85 Vliagoftis H, Schwingshackl A, Milne CD, Duszyk M, Hollenberg MD, et al. 2000. Proteinase-activated receptor-2mediated matrix metalloproteinase-9 release from airway epithelial cells. J. Allergy Clin. Immunol. 106:537–45 Cocks TM, Fong B, Chow JM, Anderson GP, Frauman AG, et al. 1999. A protective

12 Feb 2004

16:14

AR

AR210-IY22-27.tex

AR210-IY22-27.sgm

LaTeX2e(2002/01/18)

ASTHMA: PERSISTENCE AND PROGRESSION

177.

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

178.

179.

180.

181.

182.

183.

role for protease-activated receptors in the airways. Nature 398:156–60 Booth BW, Adler KB, Bonner JC, Tournier F, Martin LD. 2001. Interleukin-13 induces proliferation of human airway epithelial cells in vitro via a mechanism mediated by transforming growth factor-α. Am. J. Respir. Cell Mol. Biol. 25:739–43 Puddicombe SM, Torres-Lozano C, Richter A, Bucchieri F, Lordan JL, et al. 2003. Increased expression of p21(waf) cyclin-dependent kinase inhibitor in asthmatic bronchial epithelium. Am. J. Respir. Cell Mol. Biol. 28:61–68 Danahay H, Atherton H, Jones G, Bridges RJ, Poll CT. 2002. Interleukin-13 induces a hypersecretory ion transport phenotype in human bronchial epithelial cells. Am. J. Physiol. Lung Cell Mol. Physiol. 282: L226-36 Laoukili J, Perret E, Willems T, Minty A, Parthoens E, et al. 2001. IL-13 alters mucociliary differentiation and ciliary beating of human respiratory epithelial cells. J. Clin. Investig. 108:1817–24 Lee CG, Homer RJ, Cohn L, Link H, Jung S, et al. 2002. Transgenic overexpression of interleukin (IL)-10 in the lung causes mucus metaplasia, tissue inflammation, and airway remodeling via IL-13dependent and -independent pathways. J. Biol. Chem. 277:35466–74 Grunig G, Warnock M, Wakil AE, Venkayya R, Brombacher F, et al. 1998. Requirement for IL-13 independently of IL-4 in experimental asthma. Science 282:2261–63 Walter DM, McIntire JJ, Berry G, McKenzie AN, Donaldson DD, et al. 2001. Critical role for IL-13 in the development of

184.

185.

186.

187.

188.

189.

190.

P1: IKH

815

allergen-induced airway hyperreactivity. J. Immunol. 167:4668–75 Whittaker L, Niu N, Temann UA, Stoddard A, Flavell RA, et al. 2002. Interleukin-13 mediates a fundamental pathway for airway epithelial mucus induced by CD4 T cells and interleukin-9. Am. J. Respir. Cell Mol. Biol. 27:593–602 Temann UA, Ray P, Flavell RA. 2002. Pulmonary overexpression of IL-9 induces Th2 cytokine expression, leading to immune pathology. J. Clin. Investig. 109:29– 39 Kuperman DA, Huang X, Koth LL, Chang GH, Dolganov GM, et al. 2002. Direct effects of interleukin-13 on epithelial cells cause airway hyperreactivity and mucus overproduction in asthma. Nat. Med. 8:885–89 Longphre M, Li D, Gallup M, Drori E, Ordonez CL, et al. 1999. Allergen-induced IL-9 directly stimulates mucin transcription in respiratory epithelial cells. J. Clin. Investig. 104:1375–82 Dabbagh K, Takeyama K, Lee HM, Ueki IF, Lausier JA, Nadel JA. 1999. IL-4 induces mucin gene expression and goblet cell metaplasia in vitro and in vivo. J. Immunol. 162:6233–37 Atherton HC, Jones G, Danahay H. 2003. IL-13-induced changes in the goblet cell density of human bronchial epithelial cell cultures: MAP kinase and phosphatidylinositol 3-kinase regulation. Am. J. Physiol. Lung Cell Mol. Physiol. 285:L730–39 Holgate ST, Davies DE, Lackie PM, Wilson SJ, Puddicombe SM, Lordan JL. 2000. Epithelial-mesenchymal interactions in the pathogenesis of asthma. J. Allergy Clin. Immunol. 105:193–204

Cohn.qxd

3/12/2004

2:17 PM

Page 1

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ASTHMA: PERSISTENCE AND PROGRESSION

C-1

Figure 1 Asthma is a result of airway inflammation and remodeling. Interactions between cells and molecules in the respiratory tract and between Th2 and other inflammatory cells cause remodeling and inflammation in the airways. The continued positive feedback between these elements promotes persistent disease.

3/12/2004

C-2 COHN

2:17 PM

■

ELIAS ■

See legend on next page

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Cohn.qxd Page 2

CHUPP

Cohn.qxd

3/12/2004

2:17 PM

Page 3

ASTHMA: PERSISTENCE AND PROGRESSION

C-3

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Figure 2 Fate of the airways after exposure to inhaled allergen. (A) In the absence of inflammation, inhaled allergen does not induce an inflammatory response because protective features of the respiratory tract insure immune tolerance. (B) In asthma, the inflamed airways promote immune responsiveness. Inhaled allergen stimulates further Th2 cell activation, activation of inflammatory cells, release of inflammatory mediators, and epithelial damage, thus leading to persistent inflammation and airway remodeling.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:789-815. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104608

Annu. Rev. Immunol. 2004. 22:817–90 doi: 10.1146/annurev.immunol.22.012703.104608 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on January 26, 2004

CD1: Antigen Presentation and T Cell Function Manfred Brigl and Michael B. Brenner Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Division of Rheumatology, Immunology, and Allergy, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts 02115; email: [email protected], [email protected]

Key Words lipid antigens, infection, cancer, autoimmunity, NKT cells ■ Abstract This review summarizes the major features of CD1 genes and proteins, the patterns of intracellular trafficking of CD1 molecules, and how they sample different intracellular compartments for self- and foreign lipids. We describe how lipid antigens bind to CD1 molecules with their alkyl chains buried in hydrophobic pockets and expose their polar lipid headgroup whose fine structure is recognized by the TCR of CD1-restricted T cells. CD1-restricted T cells carry out effector, helper, and adjuvantlike functions and interact with other cell types including macrophages, dendritic cells, NK cells, T cells, and B cells, thereby contributing to both innate and adaptive immune responses. Insights gained from mice and humans now delineate the extensive range of diseases in which CD1-restricted T cells play important roles and reveal differences in the role of CD1a, CD1b, and CD1c in contrast to CD1d. Invariant TCRα chains, self-lipid reactivity, and rapid effector responses empower a subset of CD1d-restricted T cells (NKT cells) to have unique effector functions without counterpart among MHCrestricted T cells. This review describes the function of CD1-restricted T cells in antimicrobial responses, antitumor immunity, and in regulating the balance between tolerance and autoimmunity.

CD1 GENES CD1 molecules were first identified using monoclonal antibodies that bound to human thymocytes (1), and were designated Cluster of Differentiation 1 (CD1) based upon their leukocyte staining characteristics at the First International Workshop on Human Leukocyte Differentiation Antigens (2).1 CD1 genes have an intron/exon 1

Abbreviations used: ACAID, anterior chamber-associated immune deviation; AHR, allergen-induced airway hyperreactivity; APC, antigen-presenting cell; ANS, 1-anilinonapthalene-8-sulfonic acid; bp, base pairs; CD1, Cluster of Differentiation 1; DC, dendritic cell; DN, double negative; DTH, delayed type hypersensitivity; EAE, experimental autoimmune encephalomyelitis; EMCV-D, diabetogenic encephalomyocarditis virus; ER, endoplasmic reticulum; GM-CSF, granulocyte/macrophage colony-stimulating factor; GPI, glycosylphosphatidylinositol; HEV, high endothelial venule; HSV-1/-2, herpes simplex virus type 1 and 2; IEL, intestinal intraepithelial lymphocytes; IFN-γ , interferon-γ ;

0732-0582/04/0423-0817$14.00

817

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

818

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Figure 1 A genomic map of the human and mouse CD1 locus and the syntenic regions between human chromosome 1 (upper part) and mouse chromosomes 3 and 1 (lower part) indicates a 750,000 bp gap in synteny between the two species. This suggests that a chromosomal translocation may have led to the deletion of CD1A, B, and C in rodents. CD1 genes are depicted as black boxes, and the direction of transcription is indicated by arrowheads. Orthologous genes are connected by dashed lines. The hypothetical break point (wavy line) for the ancestral mouse species is arbitrarily placed between the group 1 and group 2 gene clusters within the human CD1 locus.

structure comparable to that of major histocompatibility complex (MHC) class I genes and encode type 1 integral membrane proteins consisting of α1, α2, and α3 domains, similar to MHC class I molecules (3, 4). The α3 domains are the most homologous among CD1 family members (71%–88%), showing clear but limited homology (21%) to MHC class I α3 domains (5). The CD1 genes are not human orthologs of murine Tla genes nor non-classical MHC-encoded class I genes, but rather form a distinct locus of tightly clustered genes located in a 170,000 bp region of chromosome 1q23.1 in human and on chromosome 3 in mouse, regions that are unlinked to the mammalian MHC (6–9) (Figures 1 and 2). The five CD1 genes that have been identified in man are designated CD1A, CD1B, CD1C, CD1D, and CD1E, corresponding to five CD1 proteins: CD1a, CD1b, CD1c, CD1d, and CD1e (3, 5, 6, 10). We adopt the use of upper case letters for genes and lower case for proteins. The nomenclature used during active identification of the CD1 genes made assignments based on the nominal size of LAM, lipoarabinomannan; LPS, lipopolysaccharide; MCA, methylchlorantrene; MDL, mature DC lysomomes; MHC, major histocompatibility complex; NK, natural killer; NKT, Natural Killer T cell; ODN, oligodinucleotide; PIM, phosphatidylinositol mannoside; RSV, respiratory syncytial virus; SLE, systemic lupus erythematosus; PPD, purified protein derivative; TCR, T cell receptor; TNF-α, tumor necrosis factor-α; TRAIL, TNF-related apoptosis-inducing ligand.

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

819

Figure 2 Similarity and proposed evolutionary relationship between known CD1, MHC class I, and non-classical MCH class I protein sequences. All known full-length CD1 proteins from humans, mice, and other mammals (h, human; m, mouse; gp, guinea pig; rab, rabbit; sh, sheep; rat; pig) and examples of classical and non-classical human MHC class I proteins are shown (15). The unrooted dendrogram was created using the Neighbor-Joining method of Satori and Nei.

their genomic EcoRI fragments and is now replaced by their CD designations (R4 = CD1A, R1 = CD1B, R7 = CD1C, R3 = CD1D, and R2 = CD1E). Based on sequence identities in the α1 and α2 domains, the CD1 isoforms were separated into two sets, the CD1a, b, and c set (group 1) and the CD1d set (group 2) and CD1e intermediate (10) (Figure 2). Mice only contain CD1D orthologs. Two highly homologous CD1D genes (CD1D1 and CD1D2) are located on chromosome 3 in tail-to-tail orientation ∼6 kb apart (11, 12) (Figure 1). The existence of two CD1D genes in mice is likely the result of a relatively recent duplication event. Rats contain only one CD1 gene that is an ortholog of the human CD1D (13, 14). A chromosomal break event is now

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

820

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

commonly believed to be the reason for the absence of group 1 CD1 from rodents (9, 11, 15) (Figure 1). Unlike the characteristic polymorphism of MHC class I and II genes, allelic variation of CD1 genes is extremely limited (16, 17). Studies of 110 (16) or 166 (18) unrelated donors from various ethnic groups revealed polymorphisms in CD1A, B, C, D, and E genes. The nucleotide substitutions in CD1B and CD1C were silent, whereas two CD1A, two CD1E, and CD1D alleles resulted in amino acid changes. Two alleles of CD1A, designated CD1A∗ 01 and CD1A∗ 02, have sequence differences that are predicted to be outside of the antigen binding groove, and no differences in expression or antibody reactivity were noted (18, 19). These alleles occur with different frequencies in ethnic populations (17). CD1E alleles, CD1E∗ 01 and CD1E∗ 02, were noted and could potentially influence ligand binding (16, 20, 21). However, to date the few studies carried out found no correlation between CD1 allelic polymorphisms and mycobacterial infection (22; C. Dascher, unpublished data). The reasons for the overall lack of polymorphism of CD1 alleles are a matter of speculation. One possibility is that the opportunity for variation in the structure of lipid tails of microbial species may be markedly less than the potential for variation in the sequence of microbial antigenic peptides. Single nucleotide changes can readily alter peptide anchor residues needed for peptide binding to MHC alleles. On the other hand, lipid tails are synthesized by multiple enzymatic steps, and the modifications that microorganisms might make are limited by the structural constraints needed for lipid organization in microbial membranes and cell walls. Thus, if the lipid tails of antigens binding CD1 are more constrained in their potential structural diversity, correspondingly little polymorphism of CD1 grooves might be needed to accommodate their binding. Besides humans and rodents, CD1 genes have now been reported in sheep (23, 24), cow (25), rabbit (26, 27), guinea pig (28), and rhesus macaque (29), suggesting their preservation as a gene family in evolution (Figure 2). Whereas humans express a single example of CD1A, B, C, D, and E, these isoforms are expanded or deleted in particular species. Rabbits express two CD1A genes and one CD1E gene (27) as well as CD1B and CD1D orthologs (26). Guinea pigs also possess an extended family of CD1 genes (28). More than 11 potential CD1 genes were identified by cross-hybridization to a conserved α3 probe, and among these 11 genes were 4 CD1B orthologs and 3 CD1C orthologs, but no genes cross-hybridizing with CD1A or CD1D could be found. Many of the guinea pig CD1B and C genes contain tyrosine-based sorting motifs in their cytoplasmic tails. Interestingly, one of the guinea pig CD1B orthologs, CD1B3, lacked the typical tyrosine-based sorting motif that in human CD1b mediates binding to the AP3 adaptor protein complex and localization in late endosomes and lysosomes (30), and instead displayed a pattern of intracellular trafficking like that of human CD1a (31). Given that guinea pig CD1A genes have not yet been found, evolutionary pressure may have resulted in modifying a CD1B ortholog into a gene that codes for a CD1 molecule that can traffic through and sample the early endosome recycling pathway like human CD1a (15, 31). Evolutionary pressure to develop CD1 isoforms that sample

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

821

relevant endosomal compartments may also account for the fact that murine CD1d associates with the adaptor protein AP3 that mediates its localization to lysosomes (32, 33). Although mice lack CD1b, they have evolved CD1d to traffic along the endosomal route followed by CD1b in other species. These examples have led to a “traffic hypothesis” of CD1 evolution: Each species evolves CD1 isoforms that follow distinct intracellular trafficking routes enabling the sampling of antigens, even if it is not the same isoform in each species (15, 31).

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD1 PROTEINS CD1 polypeptides are expressed as heterodimers composed of the CD1 heavy chain noncovalently paired with β 2-microglobulin. CD1 heavy chains have a Mr of 49, 45, and 43 kDa for CD1a, b, and c, respectively, with the differences mainly accounted for by the number of N-linked glycan additions, since the peptide backbones of each are similar (∼33 kDa) (34–38). Most CD1d protein complexes are expressed on the cell surface as a 49 kD heavy chain associated with β 2-microglobulin (39, 40). In general, association with β 2-microglobulin appears to be necessary for cell surface expression of various CD1 isoforms, as shown by lack of surface localization in β 2-microglobulin-deficient cells (38, 41–43). The ratios of β 2-microglobulin associated with CD1a, b, and c are not always similar, suggesting that in some cases the CD1 heterodimers may be more prone to dissociate than the MHC class I/β 2microglobulin complexes (34, 36). Moreover, the expression of human and murine CD1d at the cell surface without β 2-microglobulin has been observed (44). The fully glycosylated CD1d heavy chains were associated with β 2-microglobulin, but a subset of β 2-microglobulin-free CD1d molecules either contain only immature high mannose (endoglycosidase H-sensitive) N-linked glycans or completely lack glycan additions (45). Both forms may occur on the same cell types. In the case of intestinal epithelial cells, the form lacking glycosylation was restricted to the apical surface (46). The functional significance of the β 2-microglobulin-free forms of CD1d is not known. CD1 mRNA splicing complexity occurs and may generate secreted forms of CD1 proteins (47). A secreted form of CD1a results from unspliced transcripts, and intrathymic splicing results in a secreted CD1c product. CD1e mRNA that lack transmembrane domains or that lack α1 or α2 domains as a result of alternative splicing also have been observed and may correspond to protein products found in Golgi or late endosomal compartments, or lead to secreted products (48).

CD1 ATOMIC STRUCTURES CD1d The atomic structures have now been determined for murine CD1d and human CD1a and b. The CD1d structure revealed the striking similarity in secondary, tertiary, and quaternary structure of CD1 to MHC class I in which the α-chain

15 Mar 2004

15:50

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

822

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

folds into three domains, α1, α2, and α3 associated noncovalently with β 2-microglobulin (49). A putative antigen-binding superdomain is composed of the α1 and α2 helices that sit atop and traverse an eight-stranded antiparallel β-pleated sheet platform (Figure 3a,b). Compared to MHC class I, the α2 helix is kinked upward ˚ higher above the β-sheet, resulting in a deeper and the α1 helix is raised 4–6 A groove. In addition, the β-strand platform displays substantially greater curvature and is more bowl shaped compared with MHC class I or class II platforms. The relative positions of the α1 and α2 helices are closer together along their longi˚ compared with MHC class I tudinal axes, resulting in a narrower groove (14 A) or II. Despite the overall similarity in domain organization, striking differences between CD1d and MHC class I and II are found in the topography and molecular surfaces of the antigen-binding grooves. For CD1d, the overall volume and surface area of the groove is larger than any MHC class I molecule, and the smaller pockets characteristic of MHC molecules are coalesced into two large pockets, designated A0 and F0 (49) (Figure 3c). The CD1d groove is closed at both ends but is accessible at the top of the molecule through a narrow opening extending from the center of the groove to a point over the center of the F0 pocket. Few amino acid sidechains that line the groove of CD1d are capable of hydrogen bonding. As a result, the likelihood of forming an extensive hydrogen bonding network at the ends of a peptide (as in class I) or along the sides of the longitudinal axis of the groove (as in class II) is not apparent in CD1d.

CD1b The structure of ligand-bound CD1b complexes [ganglioside GM2/CD1b and phosphatidylinositol (PI)/CD1b] showed the lipid tails to be buried in hydrophobic channels and the polar head of the lipid ligand oriented at the surface of the molecule between the CD1 α-helices (50) (Figure 3). A space-filling surface view illustrates how only the polar head of the glycolipid is exposed at the CD1b surface (Figure 3d). The amount of surface-exposed hydrophilic lipid head group covers only a small surface of the CD1 molecule, compared with MHC class I, where the presented peptide covers the central plane, and MHC class II, where the peptide extends over the whole surface (Figure 3d). Compared with CD1d, unexpected differences were noted in the antigen-binding domain of CD1b. Instead of the single opening for the ligand-binding groove providing access to the two large A0 and F0 pockets, the CD1b structure revealed a network of four hydrophobic channels (designated A0 , C0 , F0 , and T0 ) in the α1 and α2 superdomain, each of which contained a hydrocarbon chain of 11–22 carbons in length (50) (Figure 3c). Three of these channels, A0 , C0 , and F0 , communicate with the top surface of the CD1b molecule through openings between the α1 and α2 helices and are interconnected at their other termini with a fourth distinct channel, designated the T0 tunnel. Furthermore, the C0 channel extends to a portal opening under the α2 helix and may allow egress of longer chains out through its under-the-helix portal. These remarkable structural features suggest a mechanism by which the hydrophobic binding capacity of the

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

823

channels may be utilized to accommodate lipids containing several hydrocarbon tails or long hydrocarbon chains that sequentially traverse interconnected channels or exit the C0 portal under the α2 helix (Figure 3c).

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD1a The structure of the human CD1a antigen-binding groove revealed two large hydrophobic pockets extending out from the center of the groove and running between the α1 and α2 helices (51) (Figure 3c). This hydrophobic cavity pattern was more like CD1d than the four-channel structure of CD1b. The CD1a cocrystal with a sphingolipid, 30 -sulfated β-D-galactosylceramide, showed that the two hydrocarbon binding regions each accommodated one alkyl chain (Figure 3c). The A0 pocket runs deep into the CD1a groove, with its terminus clearly defined as it ends in a semicircular curve. It perfectly accommodated the 18 carbons of the sphingosine base. In contrast, the F0 pocket is long and straight and widened as it approached the surface of CD1a between the helices. The sulfogalactosyl polar head of the antigen was positioned at the entrance to the F0 pocket with both the sulfate and the galactose moieties anchored by hydrogen bonding to polar sidechains in the α−helical groove at the junction of the A0 and F0 pockets (51) (Figure 3c). This positioned the polar lipid headgroup to partially protrude from the surface of the CD1a binding groove for TCR recognition, while at the same time being nestled closely in the groove (Figure 3d). Thus, for CD1a, the intersection of the A0 and F0 pockets contains polar residues that can interact with and provide specificity for binding the polar moieties of amphipathic lipids. Moreover, unlike the CD1b structure in which the acyl chains were enclosed in the hydrophobic channels, in the CD1a structure the amide-linked acyl chain was exposed and protruded at the TCR recognition surface, making van der Waals contacts with hydrophobic amino acid sidechains of the F0 pocket (Figure 3c,d).

Comparison of CD1d, CD1b, and CD1a Of the three known structures, the CD1b groove is the largest, with an estimated ˚ compared with that of CD1a (1300 A) ˚ or CD1d (1650 A). ˚ total volume of 2200 A Moreover, the four interconnected channels of CD1b might accommodate up to 76 carbons whereas the two channels of CD1a are predicted to accommodate 36 carbons (49–51). Of the binding channels, the A0 pockets are more conserved among the isoforms than are the other channels (Figure 3c). The T0 tunnel was observed only in CD1b as it runs across the top of the β-strand platform and is made possible by the small side chains Gly 98 and Gly 116. In the CD1a and CD1d structures, larger sidechains block a separate T0 tunnel (51) (Figure 3c). The C0 channel of CD1b also does not occur in CD1a or CD1d, as bulky hydrophobic residues block any possible C0 portals (Figure 3c). Thus the ligandbinding capacities and the structural organization of the hydrophobic channels are distinct for CD1a, b, and d and may allow binding of an array of structurally diverse lipid-containing antigens.

13 Feb 2004

18:42

824

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

ANTIGENS

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Foreign Microbial Lipid and Lipopeptide Antigens Identification of the first antigen presented by a CD1 molecule to T cells indicated that the antigens were lipids (52). Thus far, the foreign antigens presented by CD1 molecules that have been characterized include a range of diverse lipids found in the cell walls of mycobacteria. Mycolic acid, an α-branched β-hydroxy long chain fatty acid (52), glucose-monomycolate (GMM), which consists of mycolic acid esterified to a single glucose moiety (53), phosphatidylinositol mannoside (PIM) and lipoarabinomannan (LAM), which both consist of a phosphatidylinositol anchor linked to additional glycans (54), are all presented by CD1b molecules (Figure 4). A CD1c-restricted T cell line was shown to recognize mycobacterial mannosylβ-1-phosphoisoprenoid, a glycophospholipid with only a single short lipid tail (55) (Figure 4). CD1a-restricted T cells have been recently shown to recognize a lipopeptide (didehydroxymycobactin) from Mycobacterium tuberculosis, defining a new biochemical class of antigens for CD1-restricted T cells (56) (Figure 4). Thus far, microbial lipid and lipopeptide antigens presented by CD1 have been confirmed only for CD1a, b, and c. There have been conflicting reports as to whether murine CD1d-restricted T cells recognize glycosylphosphatidylinositol (GPI)-anchored glycoproteins from Plasmodium or Trypanosoma spp. in vitro or affect the IgG response to GPI-linked proteins in vivo (57–61). There is no apparent reason why CD1d molecules could not also present foreign lipid antigens. However, at present, foreign antigen presentation has been limited to CD1a, b, and c.

Self-Lipid Antigens CD1-restricted T cells can be stimulated by exposure to CD1+ antigen-presenting cells (APCs) in the absence of foreign antigens, resulting in their activation as measured by proliferation, cytokine secretion, or cytolysis. This was first observed for CD1 group 1–restricted T cells (62) and later for CD1d-restricted T cells (63, 64). Examples of CD1-presented self-lipids that occur in normal mammalian tissues include ubiquitous phospholipids, such as phosphatidylinositol, phosphatidylethanolamine, and phosphatidylglycerol, which can be recognized by murine CD1d-restricted T cell hybridomas (65) (Figure 4). Cellular phosphatidylinositol and GPI have been eluted from CD1d proteins, indicating that these phospholipids may be naturally loaded into CD1d (66, 67). However, recognition of self-GPIs that results in activation of CD1d-restricted T cells has not been demonstrated to date, and CD1d+ APCs lacking the ability to synthesize GPIs retain the ability to stimulate autoreactive CD1d-restricted T cell hybridomas (58, 67). Therefore, it is not yet clear whether GPIs are self-antigens that activate CD1drestricted T cells. Several self-sphingolipids have been found to activate human or murine CD1-restricted T cells. Sphingolipids containing complex oligosaccharides, such as the ganglioside GM1 which is mainly present in neural tissues, can be presented by CD1b to T cells (68). The ganglioside GD3, which constitutes

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

825

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

Figure 4 Examples of antigens presented by CD1. Structures of CD1-presented antigens are grouped according to lipid classes. Phosphatidylinositol mannoside (PIM) containing two mannosyl residues is depicted, but antigenic PIMs typically have additional mannosyl residues.

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

826

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

a major fraction of the glycolipids of tumors of neuroectodermal origin, such as melanomas, and which also occurs in normal mammalian tissues (69), was recognized by murine CD1d-restricted T cells (70). Sulfatide, another sphingolipid and main constituent of mammalian brain lipids, is a sulfate ester of β-D-galactosylceramide and could be recognized by CD1a-, CD1b-, or CD1crestricted T cells, indicating that the same self-lipid antigen can, in some cases, be presented by different CD1 molecules (71) (Figure 4). It is not known whether self-lipid antigens are constitutively expressed at the cell surface or in intracellular membranes, or if alterations in synthesis, processing, or trafficking control their stimulatory capacity. The frequent demonstration of self-reactivity, however, highlights this as a characteristic feature of many CD1-restricted T cells.

α-Galactosylceramide (αGalCer) Most mouse and human CD1d-restricted T cells that use TCRα-invariant TCRs recognize α-galactosylceramide (αGalCer), a glycosphingolipid found in marine sponges (72) (Figure 4). The αGalCer structure resembles mammalian ceramides in that it contains a sphinosine-like base, an amide-linked acyl chain, and an O-linked pyranose. However, the anomeric carbon of the sugar is in the α-linkage to the oxygen, whereas in mammals, the corresponding linkage is of the β-anomeric type. αGalCer has no known physiological function in mammalian immunity but has been widely used as a pharmacological agent to study activation of CD1drestricted T cells.

FUNCTIONAL ANALYSIS OF LIPID BINDING TO CD1 Biophysical analyses of LAM binding to CD1b revealed an equilibrium binding constant (KD = kdiss/kass) of 3.2 × 10−8 M using evanescence wave sensor analyses and a KD of 5.3 × 10−8 M calculated by Scatchard analysis, values that are in a range similar to those calculated for high-affinity peptide binding to MHC molecules (73). Separately, a KD of 6.6 × 10−7 M was calculated using surface plasmon resonance for binding of a phosphatidylinositol mannoside (PIM2, with two mannose and two C:18 moieties) to CD1b. Since CD1b is likely to be loaded in acidic endosomal compartments (30, 74, 75), pH-dependent binding of LAM and glucose monomycolate to CD1b was examined and found to occur at pH 4. This correlated with partial unfolding of the α-helical segments of CD1b as measured by circular dichroism and with greater binding of a hydrophobic environmentsensitive fluorescent probe, 1-anilino-napthalene-8-sulfonic acid (ANS) (73). The studies above utilized immobilized soluble human CD1b with lipids in the fluid phase. Because lipids are expected to associate into micelles in aqueous solutions, this may have unknown biophysical effects on antigen-loading analysis. Using another approach, biotinylated lipids were immobilized and soluble CD1d molecules in the fluid phase were examined for their binding characteristics using surface

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

827

plasmon resonance (76). These studies revealed a KD of 3.4 × 10−7 M for the binding of soluble murine CD1d to immobilized acyl-biotin-substituted αGalCer (76). In contrast, studies using an isoelectric focusing equilibrium binding assay and isothermal titration calorimetry showed a KD in the range of 1–9.7 × 10−6 M for the binding of soluble CD1d to αGalCer (77).

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TCR SPECIFICITY IN RECOGNITION OF CD1-ANTIGEN COMPLEXES TCRs of CD1a-, b-, and c-Restricted T Cells Molecular cloning and transfection of cDNAs encoding the TCRαβ or γ δ chains isolated from CD1a-, b-, and c-restricted glycolipid antigen-specific T cells confirmed that T cell recognition of lipid antigens was mediated by the TCR (78–80). The sequences of the αβ TCRs that recognize CD1a-, b-, and c-presented microbial lipid antigens revealed diverse TCRα and β chain rearrangements and junctional diversity including evidence for D segment usage and N-nucleotide additions at the V-D and J joining ends (78). Hence, in primary structure, the αβ TCRs that recognize CD1a, b, and c are indistinguishable from those that recognize MHC class I or II complexes with peptides. Similar to TCR recognition of peptide-MHC complexes, a role for the CDR3 loop sequences of the TCR in recognition of specific CD1-lipid antigen complexes was confirmed by a combination of TCR mutagenesis and transfection (81).

TCRs of CD1d-Restricted T Cells A remarkable feature of many human and murine T cells that recognize CD1d molecules is their use of an invariantly rearranged TCRα chain. In mice, the Vα14 gene segment is rearranged in a germline configuration with Jα18 (formerly called Jα281), and this Vα14Jα18 TCRα chain is paired predominantly with TCRβ chains that use Vβ8, Vβ7, or Vβ2 gene segments (82, 83). Human CD1d-restricted T cells use Vα24 gene segments invariantly rearranged with Jα18 (formerly JαQ), which are highly homologous to the murine Vα14 and Jα18 gene segments, respectively (83–85). The human Vα24Jα18 invariant TCRα chain is preferentially paired with Vβ11, which is homolgous to murine Vβ8. In both humans and mice, Vα24+ and Vα14+ invariant TCRα chains, respectively, have been observed that contain noncoding nucleotide substitutions near the V/J junctional region, presumably as a result of trimming followed by N-region additions (64, 82, 83). CD1d-restricted T cells using Vα24Jα18+ and Vα14Jα18+ TCRs with single amino acid substitutions at the end of the V gene segment, a region corresponding to the beginning of the CDR3 loop, have also been found, but the significance of this limited sequence variation is not known (83, 86). Hence, it appears that the invariant, germline-configured TCRα chains of CD1d-restricted T cells are generated by the same mechanisms that give rise to diverse TCRs, but that there

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

828

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

is strong selection pressure to maintain the germline amino acid sequence. The VDJ junctional regions of the TCRβ chains of TCRα-invariant CD1d-restricted T cells are extremely diverse, with little evidence of biased J chain usage or of conserved amino acid motifs (83, 87). By analyzing Vβ CDR3 sequences from individual CD1d-restricted TCRα-invariant T cell clones in mice, a clonal size of 5–10 T cells per clone was estimated, which is similar to that of na¨ıve conventional T cells, suggesting that CD1d-restricted T cells in previously unchallenged mice are not clonally expanded (88, 89). Most murine and human TCRα-invariant CD1d-restricted T cells recognize αGalCer (72, 90, 91) and stain with αGalCer-loaded CD1d tetramers (92–95). However, thus far, no single microbial or self-antigen has been found with similar antigenic capacity for TCRα-invariant CD1d-restricted T cells. Tail-deleted murine CD1d molecules, which lack the tyrosine-based endosomal targeting motif and traffic differently than normal CD1d, do not stimulate Vα14+ CD1d-restricted hybridomas, suggesting that these T cells recognize self-lipids loaded in endosomal compartments (96, 97). In contrast, CD1d-recognition by Vα24+/Vβ11+ CD1d-restricted T cell clones was not dependent on the endosomal targeting motif of human CD1d (64). Interestingly, CD1d-restricted T cells reactive with the ganglioside GD3, which have been generated in mice by immunization with a human melanoma cell line, comprised a subpopulation of αGalCer/CD1d-tetramer+ T cells, suggesting that the TCRβ chain may influence the antigen-specificity of Vα14+ TCRα-invariant TCRs (70). Other studies also concluded that the TCRβ chain can influence the binding of Vα14Jα18+ TCRs to CD1d/antigen complexes (98, 99). CD1d-restricted T cells that use diverse Vα gene segments, in contrast to invariant TCRα chains, have been identified in both humans and mice (97, 100–104). Sequencing of the TCRs of a large panel of Vα14– CD1d-restricted T cell hybridomas derived from MHC class II–deficient mice showed that they contained expanded T cell populations that use Vα3.2Jα9 or Vα8 TCRα chains paired preferentially with Vβ8 TCRβ chains (105). Other analyses have identified additional Vα14– CD1d-restricted T cells that do not use Vα3.2Jα9 or Vα8 TCRα chains, but these studies also found frequent use of genes from the Vβ8 gene family (106). Given the lack of specific markers for Vα14– CD1d-restricted T cells, this population is less studied and its relative frequency remains controversial (100, 103, 105). But Vα14– CD1d-restricted T cells are clearly not rare and may prove to be numerically substantial and functionally important. Human CD1d-restricted T cell lines using TCRs other than Vα24/Vβ11 have been isolated from bone marrow and livers of hepatitis C–infected patients (107, 108) and Vα24– T cells that stain with αGalCer/CD1d-tetramers have been observed in peripheral blood of humans (104). Further, murine T cells using Vγ 4+ γ δ TCRs recognized myocytes infected with coxsackie virus B3 in a CD1d-dependent manner (109). In mice, the Vα14– CD1d-restricted T cells resemble the Vα14-invariant subset in demonstrating autoreactivity to CD1d+ APCs; however, Vα14– CD1d-restricted T cell hybridomas were able to respond to tail-deleted CD1d molecules, in contrast

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

829

to Vα14+ CD1d-restricted T cell hybridomas that did not recognize such CD1d molecules (96, 97). The Vα14-invariant CD1d-restricted T cells tested thus far did not respond to αGalCer (65, 110). Thus, Vα14+ and Vα14– CD1d-restricted T cells in mice appear to have different antigen specificities.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TCR-Recognition of CD1-Antigen Complexes Although cocrystals of CD1-antigen complexes with the TCRs that recognize them have not yet been solved, optimized molecular modeling and mutagenesis of CD1b and CD1d α1- and α2-helical residues predict possible TCR contacts and suggest a diagonal orientation of the molecular footprint of the TCR across the longitudinal axis of the CD1 α-helices, similar to that observed for TCR-MHC complexes (78, 111, 112). Surface plasmon resonance has revealed a relatively high-affinity binding of soluble Vα14+ TCR to the αGalCer/CD1d complex, compared with that of αβ TCRs with MHC/peptide complexes (77, 113). CD1-restricted TCRs that recognize foreign antigens can distinguish even small changes in the structure of the hydrophilic head group of the antigens. For example, studies of mycobacterial mycolic acid antigens presented by CD1b reveal that changes to any of the hydrophilic functions of the fatty acids or their attached glycans alter recognition. For example, substitutions of the carboxylic acid (52, 81), the type of the esterified sugar moiety (79), the stereochemical arrangement of the glycan and its linkage to the acyl chain (53, 79), or changes in the hydrophilic R-group attached to the main acyl chain (81) all determine the specificity of recognition by the TCR. TCRα-invariant CD1d-restricted T cells recognize either glucose or galactose in the α-anomeric linkage, but mannose (a stereoisomer of glucose and galactose) in this linkage is not recognized, and neither are a variety of other related sugar head groups (72, 91). In addition, the α-anomeric linkage is absolutely essential for recognition since the β-anomeric forms of these ceramide-like antigens are not recognized. Specificity for the sugar head group was also shown by the finding that an αGalCer containing an additional galactose [αGal(1–2)GalCer] was not recognized by TCRα-invariant CD1d-restricted T cells, unless cleaved in lysosomes to generate αGalCer (114). These studies emphasize that foreign, self-, and synthetic lipid antigen recognition by the TCR of CD1-restricted T cells involves fine specificity in distinguishing the hydrophilic moieties of the amphipathic lipid antigens. Functional studies have also examined the bioactivity of CD1-presented lipids with differences in the acyl chains. αGalCer molecules with acyl chains of 26 carbons were optimal for stimulating CD1d-restricted T cell responses, whereas acyl chains even two carbons shorter were significantly less active. Similarly, the optimum sphingosine base was C18, and again, those tails that were even a few carbons shorter displayed significantly less bioactivity (72, 115). CD1c-restricted T cells reactive with mycobacterial mannosyl-β-1-phosphoisoprenoid also recognize structurally related semisynthetic mannosyl-β-1-phosphodolichols (MPD) that are similar to phosphodolichols found in eukaryotes and prokaryotes (55).

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

830

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Strong T cell responses were seen with semisynthetic MPDs containing C55 dolichols typically found in protozoan pathogens, but responses to MPDs containing long C95 dolichols, which can be found in eukaryotic cells and human tissues, were very weak or absent (55). In these studies of ceramide and dolichol antigens, it was not possible to distinguish changes in TCR recognition of the CD1/lipid complexes from the effect of hydrocarbon length on direct binding to CD1 or from influences resulting from trafficking and delivery of the glycolipids to relevant antigen-loading compartments, since these studies were carried out in live cells rather than using immobilized molecules. For example, differences in recognition of phosphatidylethanolamine by a CD1d-restricted T cell hybridoma due to differences in the saturation of its alkyl chains correlated with the ability of lipid binding to CD1d (116). Another study using glucose monomycolate glycolipid antigens of varying hydrocarbon tail lengths concluded that longer alkyl chain containing antigens were delivered to late endosomes whereas shorter ones were loaded at the cell surface (117).

ASSEMBLY AND INTRACELLULAR TRAFFICKING OF CD1 MOLECULES The intracellular trafficking of antigen-presenting molecules is central to their ability to intersect and bind antigens that have entered cells and then deliver them to the cell surface where they can be recognized by T cells. For MHC class I and class II molecules, the trafficking routes and loading compartments are essential for sampling cytosolic and endosomal antigens, respectively (118, 119) (Figure 5a). The intracellular trafficking routes followed by CD1 molecules are distinct from those of their MHC counterparts.

CD1 Assembly CD1 heavy chains are translocated into the ER where N-linked glycans are attached, interactions with ER chaperones occur, and association with β 2-microglobulin takes place (38, 43, 120). Studies on CD1d revealed that it associates in the ER with both calnexin and calreticulin and the thiol oxidoreductase ERp57 in a manner dependent on glucose trimming of its N-linked glycans, a process coupled to disulfide bond formation in the CD1d heavy chain (120). Similarly, the β 2-microglobulin-free CD1b heavy chain was found to associate with calnexin and calreticulin (38, 43). Engineered forms of CD1d tagged with a KDEL sequence that were retained in the ER bound phosphatidylinositol, suggesting that lipid ligand binding to CD1d can occur in the ER (67). Based on pulse chase experiments, it was determined that newly synthesized CD1b and CD1d molecules are then rapidly delivered from the Golgi to the plasma membrane, presumably along the secretory pathway (121, 122) (Figure 5b,c,d).

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

831

CD1a, b, and c Trafficking

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

After reaching the plasma membrane, CD1b molecules appear to follow a major pathway of internalization into clathrin-coated pits (74, 122, 123) (Figure 5b). Deletion of the tail of CD1b results in failure to localize in late endosomes and lysosomes, leads to its accumulation on the plasma membrane, and compromises its capacity to present foreign antigens (74, 124). Adaptor proteins mediate cargoselective transport by interacting with tyrosine-based sorting motifs or dileucine −−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→ Figure 5 Intracellular trafficking of CD1 and MHC class I and II molecules. (a) MHC class I and II intracellular trafficking. MHC class I and II molecules are assembled in the endoplasmic reticulum (ER). MHC class I molecules associate with β 2-microglobulin (β 2m), acquire peptides delivered from the cytosol and follow the secretory route to the plasma membrane. MHC class II molecules assemble into invariant chain (Ii) complexes in the ER that prevent peptide binding and direct their trafficking to endosomal compartments. In lysosomes, the Ii is cleaved, peptides are loaded and MHC class II molecules traffic to the cell surface. These pathways allow peptide sampling from the cytosol and endosomal compartments by MHC class I and class II, respectively. Nuc, nucleus; HC, heavy chain; G, Golgi; TGN, trans Golgi network; TAP, transporter associated with antigen processing. (b) CD1a and CD1b intracellular trafficking. Newly synthesized CD1 molecules assemble with β 2-microglobulin in the ER, acquire selflipids (open circle with zigzag tail), and traffic to the plasma membrane along the secretory route. Cell surface CD1b molecules interact with adaptor protein AP2 and are internalized in clathrin-coated pits. CD1a molecules are internalized by an unknown mechanism. CD1a is then sorted into recycling endosomes and traffics back to the plasma membrane in an AFR6-dependent manner. It is largely excluded from lysosomes. In contrast, CD1b is sorted to late endosomes and directed to lysosomes by binding the adaptor protein AP3. CD1a and CD1b then acquire distinct self- or foreign lipid antigens (filled circle and zigzag tail) in the recycling endosomes or lysosomes, respectively. (c) CD1c and CD1d intracellular trafficking. Assembly is as above for CD1a and b. However, CD1c broadly traffics in both early recycling endosomes as well as in late endosomes and lysosomes. CD1d traffics mainly in early and late endosomes but not in recycling endosomes and only partially localizes in lysosomes. Neither CD1c nor CD1d associates with the adaptor protein AP3. CD1c and CD1d then acquire distinct self- or foreign lipid antigens (filled circle and zigzag tail) in the intracellular compartments shown. (d) Murine CD1d intracellular trafficking. Newly synthesized CD1d molecules may or may not assemble with β 2-microglobulin in the ER, acquire self-lipids (open circle with zigzag tail), and traffic to the plasma membrane along the secretory route. A subset of CD1d molecules associate in MHC class II/Ii chain complexes and may be directed to endosomal compartments from the trans Golgi network (TGN). CD1d molecules are internalized from the plasma membrane and traffic through the early and late endosomal compartment and are delivered to lysosomes in an AP3-dependent manner. CD1d then acquires distinct self-lipid antigens (filled circle and zigzag tail) in late endosomal/lysosomal compartments.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

13 Feb 2004 18:42

832

AR

BRIGL

AR210-IY22-28.tex

¥

BRENNER

Figure 5 (Continued)

AR210-IY22-28.sgm LaTeX2e(2002/01/18) P1: IKH

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

Figure 5 (Continued)

P1: IKH

833

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

834

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

motifs in the cytoplasmic tails of integral membrane proteins. The internalization of CD1b and its subsequent delivery to deep endocytic compartments is determined by tyrosine-based sorting motifs in its cytoplasmic tail (YXXZ, where Y = tyrosine, X = spacer, and Z = bulky hydrophobic residue). Binding to the adaptor protein AP2 was demonstrated using surface plasmon resonance (122). The mechanism responsible for delivery to lysosomes, the main steady-state intracellular location of CD1b (74, 125) (Figure 5b), was shown to be the preferential association of the CD1b cytoplasmic tail tyrosine-based sorting motif with another adaptor protein, AP3, based on yeast two-hybrid assays (30) and surface plasmon resonance measurements (122). In AP3-deficient cells, CD1b was mislocated to the plasma membrane and early endosomes, failed to localize to lysosomes and failed to present CD1b-restricted glycolipid antigens, whereas localization and the antigen-presenting function of other CD1 molecules were not affected (30). Delivery of antigens to lysosomal compartments may be essential for efficient loading and presentation by CD1b. In the case of LAM, the antigen is taken up by the macrophage mannose receptor (MR) at the cell surface and subsequently delivered to the lysosomes of dendritic cells (DCs) where it colocalized with CD1b (75). 14C-labeled GMM with long lipid tails (C80) was also shown to be taken up by cells and delivered to the lysosomes where it colocalized with CD1b (117). In contrast, short-chain (C32) GMM was less efficiently presented by cells and was found to localize less efficiently to lysosomes. Instead, it appeared that the C32 GMM was able to load into CD1b on the plasma membrane (117). In contrast to the steady-state intracellular localization of CD1b in lysosomes, CD1a was found to be largely excluded from lysosomes and instead localized to early recycling endosomes, using an ARF6-dependent pathway (125) (Figure 5b). Consistent with the differences in intracellular trafficking of CD1a and CD1b, antigen presentation by CD1b was critically dependent on vesicular acidification, whereas presentation by CD1a was independent of acidification blockade (125). Studies in human Langerhans cells confirm CD1a trafficking in Rab 11+ recycling endosomes and in Birbeck granules (126). Thus, the intracellular trafficking routes followed by CD1a and CD1b represent largely separate localization in early recycling endosomes and lysosomes, respectively (Figure 5b). In contrast to their largely nonoverlapping and restricted steady-state localization, the intracellular localization of CD1c partly overlapped both CD1a and CD1b, suggesting that it might broadly survey throughout the endocytic system (127, 128) (Figure 5c).

CD1d Trafficking The intracellular localization of human CD1d was partly in LAMP-1+ late endosomes and lysosomes; however, unlike CD1b whose localization in lysosomes was nearly absolute, a substantial fraction of human CD1d molecules did not colocalize with LAMP-1 and appeared to be in more proximal endosomal compartments (30) (Figure 5c). The less efficient localization of human CD1c and CD1d to

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

835

lysosomes, compared with CD1b, was likely based on the inability of the cytoplasmic tail tyrosine-based sorting motif of CD1c and CD1d to bind AP3 (30). After assembly in the ER, murine CD1d is rapidly transported to the plasma membrane with a time course consistent with trafficking along the secretory route (121). Internalization from the plasma membrane is dependent on the CD1d cytoplasmic tail, following which traffic through early and late endosomes to lysosomes occurs (121) (Figure 5d). The steady-state localization of murine CD1d revealed a pattern of abundant colocalization with LAMP-1 (97, 129). The CD1d cytoplasmic tail tyrosine-based sorting motif controlled its localization to LAMP-1+ compartments (97, 129). Like the role of AP3 in the selective delivery of human CD1b to lysosomes (30), it was shown using yeast two-hybrid binding and surface plasmon resonance binding that the murine CD1d cytoplasmic tail also mediates binding to AP3, which controls its delivery to lysosomes (32, 33). The striking localization of murine CD1d in lysosomes proved to be essential for antigen presentation. For example, a cytoplasmic tail-truncated CD1d that does not traffic through lysosomes fails to stimulate Vα14+ CD1d-restricted T cells, but is still capable of stimulating Vα14– CD1d-reactive T cells (97, 129). Further, AP3-deficient Pearl, Mocha, and AP-3b1LN mice displayed altered CD1d trafficking and a marked reduction in Vα14+ CD1d-restricted T cells in thymus, spleen, and liver (32, 33). These studies suggest that CD1d acquires the self-lipid antigens that stimulate Vα14+ CD1d-restricted T cells in lysosomes. Intracellular trafficking and antigen presentation by CD1d have been found to be influenced by MHC class II molecules, the invariant chain, cathepsin S, and cathepsin L proteases. Murine CD1d was noted to coimmunoprecipitate with the invariant chain (121), and human CD1d was noted to coimmunoprecipitate with MHC class II complexes (130). These findings suggest that at least for a fraction of CD1d molecules, their trafficking to endosomal compartments could be influenced by MHC class II/invariant chain interactions. This was evident in invariant chain–deficient mice where B cells and DCs, which normally express invariant chain, revealed increased surface expression of CD1d compared to wildtype mice, whereas cells that do not express invariant chain, like thymocytes, did not have altered surface expression (121). Invariant chain–deficient DCs were inefficient at stimulating self-reactive Vα14+ CD1d-restricted T cells (121). However, the lack of invariant chain did not alter the percentage or total number of Vα14+ CD1d-restricted T cells in the thymus, spleen, or liver and did not alter the ability of splenocytes to present αGalCer or Gal(α1–2)αGalCer (33). Mice lacking cathepsin S, an endosomal protease critical in invariant chain processing, revealed altered trafficking of CD1d, with striking accumulation in endosomes colocalized with invariant chain (131). The antigen-presenting function of CD1d by APCs from cathepsin S–deficient (catS−/−) mice was abnormal, with a reduced capacity to stimulate autoreactive hybridomas directly or to stimulate fresh CD1d-restricted T cells with αGalCer. In vivo, catS−/− mice failed to select or expand normal numbers of CD1d-restricted T cells (131). Cathepsin L, another lysosomal cysteine protease that can participate in invariant chain cleavage,

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

836

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

was found to influence CD1d-restricted T cell recognition and development (132). Unlike cathepsin S, which appeared to alter CD1d trafficking, cathepsin L–deficient (catL−/−) mice did not show altered intracellular CD1d localization or cell surface expression. Yet, catL−/− mice lacked the Vα14+ CD1d-restricted T cells, while maintaining normal numbers of other CD1d-restricted T cells, such as the Vα3.2+ subset. CatL−/− thymocytes expressing CD1d failed to stimulate autoreactive Vα14+ CD1d-restricted T cell hybridomas, whereas Vα3.2+ CD1d-restricted T cell hybridomas were responsive. The effect appeared to be limited to thymocytes, in that neither catL−/− nor catS−/− (in contrast to the above study) splenocytes showed alteration in their ability to stimulate any CD1d-restricted T cell population (132). These studies emphasize that cathepsin L plays a critical role in the presentation of self-ligands by CD1d in thymocytes that are recognized by the Vα14+ subset of CD1d-restricted T cells (132). Thus, the association of CD1d with MHC class II/invariant chain complexes and a role for endosomal proteases that may affect invariant chain processing reveal the involvement of molecules typically thought to participate only in antigen presentation by MHC class II. However, whether the effects by which these molecules influence CD1d intracellular trafficking and antigen-presenting function are direct or whether they affect other proteins that more directly impact antigen presentation by CD1d is not known.

CELLULAR EXPRESSION OF CD1a, b, AND c AND CD1a-, b-, AND c-RESTRICTED T CELLS Expression of CD1a, b, and c on Antigen-Presenting Cells The most striking expression of CD1 molecules is on dendritic cells and other professional APCs. CD1a, b, and c are widely used as DC markers in humans. Langerhans cells express CD1a as a key marker and also express CD1c, but appear to lack expression of CD1b (Figure 6). Freshly isolated immature and mature dermal Langerhans cells efficiently presented antigens to CD1a-restricted T cells (133). Dermal DCs and interdigitating DCs in lymph nodes express CD1b (134). Expression of CD1c is characteristic of human DC populations. However, CD1c is unique among group 1 CD1 antigens in its expression on subsets of B cells. CD1c is expressed in lymph node mantle zones and germinal centers (134), in marginal zone B cells of spleen (135), and on a subpopulation of circulating B cells in fetal and adult human peripheral blood (136, 137). The expression of CD1 isoforms on DCs has also been determined in studies using in vitro differentiation systems. CD34+ hematopoietic progenitor cells isolated from either umbilical cord blood or bone marrow cultured with granulocyte/ macrophage colony-stimulating factor (GM-CSF) and tumor necrosis factor-α (TNF-α) develop into two separate DC lineages (Figure 6). One subset is CD1a+ and lacks expression of CD14. As is characteristic of Langerhans cells in the

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

837

Figure 6 CD1 expression during myeloid DC ontogeny. CD34+ hematopoietic progenitorand blood monocyte-derived DCs acquire CD1 expression early in their development. The DC precursor intermediates derived from CD34+ progenitors can be distinguished by the absence or expression of CD1a, and both populations develop expression of CD1b and CD1c. These precursors further differentiate into two CD1a+ DC populations with distinct features. The DCs that differentiate from the CD1a+ precursors acquire Langerin, E-cadherin, and Birbeck granules, which are characteristic features of Langerhans cells. The DCs that develop from the CD1a–CD14+ precursors acquire CD1a expression and phenotypic characteristics of interstitial DCs, including CD9, CD68, and factor XIIIa. Blood monocytes, which express CD1d, give rise to DCs that acquire high levels of expression of the CD1a, b, and c isoforms. After activation with exogenous stimuli such as TNF-α, expression of CD1b and CD1c on monocyte-derived DCs increases slightly, whereas CD1a expression decreases. This is in contrast to MHC class II, which is strikingly redistributed to the cell surface from its intracellular pool upon activation.

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

838

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

epidermis, these precursors develop into DCs that express the Langerin protein, E-cadherin, Birbeck granules, CD1a, and CD1c, but not CD1b (138, 139) (Figure 6). Another subset of CD34+ progenitors develops into DCs that pass through a CD14+CD1a– intermediary population. These precursors develop into DCs that express CD1a, b, and c and lack Langerin, E-cadherin, and Birbeck granules and that appear to correspond to lymph node interdigitating DCs and DCs in many nonlymphoid tissues (140) (Figure 6). Human blood monocytes also can be induced to differentiate into DCs by culture with GM-CSF and interleukin-4 (IL-4) (141, 142) (Figure 6). Culture in the presence of fetal calf serum enhances CD1a, b, and c expression after monocyte culture with GM-CSF, whereas human serum inhibits expression of CD1a on these in vitro cultured DCs (143–145). The monocyte-derived CD1a+ and CD1a– DC subpopulations may have different functional capabilities (146).

CD1 Trafficking During DC Maturation Recent studies show that CD1 molecules do not follow the same course as MHC class II molecules during DC maturation (123, 147) (Figure 6). In immature DCs, both CD1b and MHC class II are located in multilamellar lysosomes, but CD1b is found on the limiting membrane, whereas MHC class II is located in the internal membranes (74, 123). During DC maturation, the surface levels of CD1b and c either remain constant or increase only slightly, whereas surface levels of MHC class II increase strikingly (123, 147). The expression of CD1a molecules typically decreases and CD1d levels remain barely detectable. During DC maturation, unraveling of multilamellar lysosomes and tubulation of multivesicular bodies are part of the process that results in delivery of MHC class II proteins to the plasma membrane (123, 148–150). During the first hours after DC maturation is initiated, MHC class II molecules and CD1b and c molecules were found to segregate from one another and localize in different intracellular vesicles (123). After maturation has occurred, CD1b and c molecules maintained their prematuration steady-state distribution, whereas MHC class II molecules redistributed to the plasma membrane. The intracellular lysosomes containing CD1b and c were markedly altered compared with lysosomes in immature DCs. These new lysosomes (called mature DC lysomomes or MDL) lack the multilamellar structure of immature DCs and appear as electron-dense single-membrane vesicles that still contain CD63 and LAMP1, but are nearly devoid of MHC class II (123). The localization of CD1 molecules during DC maturation appears to result from their redelivery to lysosomes via continued active internalization from the plasma membrane, since the internalization rate for CD1b from the plasma membrane does not change during DC maturation, whereas internalization is markedly reduced for MHC class II (123). Interestingly, glycolipid antigens were efficiently presented to CD1-restricted T cells by immature DCs, whereas MHC class II-restricted responses occurred more efficiently when stimulated by mature DCs (147). The ability of CD1 antigen-presentation to function in immature DCs likely relates to

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

839

the fact that CD1 trafficking to present antigens at the cell surface is not dependent on DC maturation in the same manner as for MHC class II molecules (123, 147).

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD1a-, b-, and c-Restricted T Cells in Antimicrobial Immunity Using mycobacterial species, a large number of CD1a-, b-, or c-reactive T cell lines and clones were generated, all of which recognized lipid antigens. The first M. tuberculosis–specific T cells were restricted by CD1b and were double negative (CD4–CD8–) (141). Subsequently, CD1-restricted T cells expressing CD8 (151) and CD4 (152) were found, making it clear that CD1-restricted T cells are a component in all of the major phenotypic groups previously thought to be MHC class I– or class II–restricted. Expression of CD1a, b, and c by dendritic cells in infectious lesions has been studied and correlates with an effective immune response in leprosy lesions (153, 154). T cell clones and lines that recognize microbial lipid antigens presented by CD1a, b, and c have been isolated from the blood of healthy donors and from tuberculoid leprosy lesions, suggesting that such T cells are part of the normal peripheral T cell repertoire and can be recruited to sites of inflammation (52–55, 141, 151, 152, 155, 156). Most of the CD1-restricted microbial antigen-specific T cells described so far recognize lipid antigens from mycobacteria, but two CD1restricted T cell lines recognizing antigens from Haemophilus influenza type B have been described, suggesting that CD1a, b, and c molecules may present antigens from a wider variety of microorganisms to T cells (157). The effector capabilities of the mycobacteria-specific T cell clones were all of the TH1 type that produce high levels of interferon-γ (IFN-γ ) and TNF-α (151, 152). Most are potent cytotoxic T cells that contain perforin, and both CD8+ and CD4+ CD1-restricted T cells showed direct microbicidal activity attributed to granulysin (151, 158–160). Antigen-specific CD1b-restricted human T cells were able to kill macrophages infected with M. tuberculosis in a CD1b-dependent manner (159). CD1a, b, and c molecules have been shown to localize in phagosomes of some DCs infected with mycobacteria, suggesting that infected DCs may also be able to stimulate CD1-restricted T cell responses (161). However, the expression of CD1 molecules by DCs was found to be downregulated following infection of monocyte-derived DCs with live M. tuberculosis, but not following phagocytosis of heat-killed bacteria (162, 163). This resulted in complete loss of CD1a, b, and c from the cell surface by 48 h after infection in vitro and was associated with the corresponding disappearance of mRNA for all three CD1 isoforms (162). M. bovis bacillus Calmette Guerin (BCG) infection of blood monocytes was also found to diminish the upregulation of CD1b that would otherwise be induced by culture in the presence of GM-CSF (164). Phagocytosis of several types of heat-killed bacteria, but not of inert particles, by the THP-1 macrophage cell line resulted in down-modulation of CD1b expression on the cell surface (165). These findings suggest that downregulation of CD1 antigen-presenting molecules may represent a microbial evasion strategy. One host mechanism that may overcome these problems is the delivery of

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

840

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

mycobacterial antigens to uninfected DCs via apoptotic bodies that are released from macrophages infected with M. tuberculosis (166). This mechanism may provide a means of delivering microbial antigens to professional APCs that are capable of efficiently stimulating CD1-restricted T cell responses. Responses to M. tuberculosis isoprenoid lipids, such as mannosyl-β-1phosphodolichols, were observed in patients recently infected with M. tuberculosis but not in normal controls (55). CD1-restricted T cell responses to lipid antigens were found to be increased in individuals who recently converted to PPD (purified protein derivative) skin test positive following contact with infected individuals or in individuals with active tuberculosis who started antibiotic therapy, compared with normal uninfected individuals. The predominant proliferative and IFN-γ producing responses were in the CD4+ T cell pool of peripheral blood leukocytes (167). Additionally, the major CD8+ T cell response occurring after M. bovis BCG immunization was CD1-restricted (168). M. tuberculosis-specific CD1-restricted responses are also seen in guinea pigs, a species that has several CD1b and CD1c isoforms (169). Vaccination of guinea pigs with mycobacterial lipids has a protective effect against infection with virulent M. tuberculosis resulting in both reduced bacterial burden and pulmonary pathology, which suggests that CD1-restricted T cells may also contribute to protective memory responses in appropriate animal models (170). Together, these findings show that CD1-mediated presentation of microbial antigens occurs during microbial infection and that CD1-restricted T cells contribute to TH1 biased cell-mediated antimicrobial responses (Figure 7). They may

Figure 7 CD1a-, CD1b-, and CD1c-restricted T cells in antimicrobial immunity. DCs that are infected with intracellular bacteria present foreign bacterial lipid antigens on the cell surface bound to CD1 molecules. CD1-restricted T cells that are specific for the foreign microbial lipids are stimulated to carry out effector functions, including the secretion of cytolytic granules containing perforin and granulysin, which lyse the infected cells and have direct antimicrobial effects, respectively, and the production of IFN-γ and TNF-α, which activate the microbicidal functions of macrophages.

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

841

respond earlier to foreign antigens than MHC-restricted T cells by recognizing antigens presented by immature DCs and then participate as part of the memory response that follows infection (147, 171–173). In addition to recognizing foreign microbial lipid antigens as described above, many CD1-restricted T cells appear to recognize CD1a, b, or c in the absence of foreign lipid antigens. Such CD1 self-reactive T cell clones express either αβ or γ δ TCRs, are cytolytic, and secrete mainly IFN-γ and other TH1-type cytokines (62, 80, 172, 174). CD1a, b, c, and d self-reactive T cell clones induced immature monocyte-derived DC maturation in vitro and were critical in determining the IL12-producing capacity of DCs (172, 175). T cell factors appear to be important in developing critical IL-12 production by DCs, but MHC-restricted foreign antigenspecific T cells are rare at the initiation of a new immune response. Thus, CD1 selfreactive T cells may play a critical role in providing DC instruction at early points in the immune response and could thereby influence the subsequent adaptive CD1 and MHC-restricted foreign antigen-specific T cell response. CD1 self-reactive T cells might bridge the gap between immediate innate and delayed adaptive immune responses (80, 176).

Expression of CD1a, b, and c in Noninfectious Lesions CD1 expressing APCs have also been observed in a variety of chronic inflammatory conditions associated with autoimmune diseases. Expression of CD1a and CD1b was observed on CD68+ endoneurial macrophages in acute and chronic inflammatory demyelinating polyneuropathies and in vasculitic neuropathies, conditions characterized by T cell infiltration of peripheral nerves (177, 178). In the central nervous system, CD1b was expressed in chronic acute plaques of multiple sclerosis lesions, especially on astrocytes (179). In a guinea pig model of experimental autoimmune encephalomyelitis (EAE), CD1b and CD1c expression was detected on astrocytes as well as infiltrating cells in affected lesions (180). Interestingly, CD1-restricted T cells specific for gangliosides and sulfatide, lipids that are enriched in neural tissues, have been isolated from individuals with multiple sclerosis, suggesting that such T cells may participate in the pathogenesis of organ-specific autoimmune diseases (68, 71). CD1 bearing DCs have been noted in several other diseases. For example, DCs expressing CD1a and CD1c were noted in rheumatoid synovium, with preferential localization to the synovial lining (181). Slightly increased expression of CD1b and CD1c was observed in psoriasis lesions (182). CD1a bearing DCs were found in salivary glands of Sj¨ogren’s syndrome and correlated with the presence of lymphocytic infiltration (183). CD1a, b, c, and d were all noted in macrophagederived lipid-laden foam cells in atherosclerotic plaques (184). In tumors found in humans, infiltrating DCs often express CD1 isoforms, suggesting that they may play a role in presenting tumor glycolipid antigens and could also be of prognostic value (185–187). CD1a was expressed by the main cells of histiocytosis X, a malignancy of Langerhans cells (188, 189), and in the cellular infiltrate of oral

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

842

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

lichen planus, a lesion considered to be precancerous (190). CD1b and CD1c expression was significantly increased in mycosis fungoides, a form of cutaneous T cell lymphoma (182). The expression of CD1 isoforms by B and T cell leukemia and lymphoma cells may be a potential target for immune recognition (191). However, CD1c and CD1d are two of the genes with the most profoundly depressed transcripts on chronic lymphocytic leukemia cells, suggesting that downregulation of CD1 molecules may represent a means of tumor immune evasion (192). Together, these studies on the expression of CD1a, b, and c suggest that CD1restricted T cells are likely to participate in infectious, inflammatory, autoimmune, and malignant conditions. Rather than having only a restricted role in mammalian immunity, all evidence points to CD1a-, b-, and c-restricted T cells participating like MHC-restricted T cells in an extensive array of host defense and immunopathologic processes.

CELLULAR EXPRESSION OF CD1d CD1d Expression in Humans In humans, CD1d expression on myeloid lineage cells appears to be regulated independently from CD1a, b, and c. Low levels of CD1d can be detected on most monocytes, and this expression decreases during culture with GM-CSF and IL4, a condition that strongly induces CD1a, b, and c expression (193) (Figure 6). Maturation of monocyte-derived DCs with lipopolysaccharide (LPS), TNF-α, or CD40/CD40L interaction does not alter the expression of CD1d significantly (193, 194). Despite these low expression levels, monocytes, monocyte-derived immature DCs, mature DCs, and macrophages can potently stimulate the proliferation and cytokine secretion of Vα24+ CD1d-restricted T cell clones when loaded with αGalCer (193). This suggests that CD1d may be present at functional levels on resting monocytes and tissue macrophages at all times, allowing CD1d and CD1ddependent T cells to function at early points during host response to infection or other challenges. CD1d expression has also been detected by immunohistochemistry on dermal DCs, but not on Langerhans cells (195). In contrast to the low levels of surface expression of CD1d on myeloid lineage cells, many circulating and splenic human B cells express CD1d at substantial levels (194, 196). In human lymph node, mantle zone B cells were strongly CD1d+ by immunohistochemical analysis, whereas CD1d+ cells in germinal centers were rare (194). Expression of CD1d has also been reported at high levels on cortical thymocytes, but it is not detectable on resting mature T cells (194). However, after activation with PHA low levels of CD1d surface expression have been observed on human T cells (194), and CD1d proteins appear to accumulate intracellularly (197). Whether CD1d expression on activated human T cells has functional relevance is unknown. In humans, CD1d is also expressed on epithelial cells, parenchymal cells, and vascular smooth muscle cells in nonlymphoid tissues including gut and liver (196, 198). Freshly isolated human intestinal epithelial cells were able to present

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

843

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

αGalCer to a murine CD1d-restricted T cell hybridoma, suggesting that intestinal epithelial cells may present lipid antigens to intestinal CD1d-restricted T cells (199). IFN-γ and heat shock protein 110 upregulated CD1d expression on intestinal epithelial cell lines, although this has not been shown for professional APCs (200, 201). Thus, in humans, CD1d expression is much broader than CD1a, b, and c expression, in that it is expressed on most monocytes, macrophages, DCs, and B cells, as well as on certain nonlymphoid cells. However, in contrast to CD1a, b, and c expression, CD1d expression is characteristically low and is not clearly upregulated on professional APCs during maturation.

CD1d Expression in Mice Mice have two CD1 genes, CD1D1 and CD1D2. Expression of CD1d2 protein in mice has been reported only on thymocytes, and its expression was not sufficient for the development of CD1d-restricted T cells (202, 203). Thus, due to the limited expression of CD1d2, most of the CD1d expression in mice outside of the thymus seems to be expression of CD1d1. In mice, CD1d is expressed on professional APCs, including splenic DCs, macrophages, and B cells (204–206). Similar to human CD1a, b, and c proteins, exposure to IL-4 and GM-CSF has been shown to increase surface expression of murine CD1d on bone marrow–derived macrophages and DCs, although only moderately (206, 207). IFN-γ did not upregulate CD1d expression on bone marrow– derived macrophages (206). Bone marrow–derived Flt3-ligand stimulated murine DCs cultured in the presence of LPS or IFN-α increased their surface expression of CD1d in parallel with MHC class II, CD80, CD86, and CD40 (207). Furthermore, increased CD1d expression has been reported on DCs in inflamed colonic lamina propria (208). Thus, in contrast to human CD1d, expression of murine CD1d on myeloid lineage cells appears to be upregulated by culture in GM-CSF/IL-4 and under inflammatory conditions. Splenic CD21hiCD23loIgMhiIgDlo marginal zone B cells appear to have the highest levels of CD1d expression among B cells in mice (205). In mice, CD1d has been reported to be expressed on immature and mature thymocytes and on peripheral T cells (205, 206). A subset of splenic CD161– Vα14Jα18+ T cells has been shown to express high levels of CD1d, and these T cells were able to autopresent αGalCer (209). However, whether CD1d expression on peripheral murine T cells has a physiologic role is not known. Expression of murine CD1d outside the lymphoid system has been observed in liver and on gastrointestinal epithelium (39, 204). A murine intestinal epithelial cell line presented αGalCer but not Gal(α1–2)αGalCer to a CD1d-restricted T cell hybridoma, suggesting that intestinal epithelial cells in mice may be able to present lipid antigens to intestinal CD1d-restricted T cells, but that antigen-processing may not be efficient (199). However, in situ hybridization techniques detected CD1d mRNA in intestinal tissue only in Paneth cells, but not in epithelial cells (210), and others also have not detected expression of CD1d in murine intestinal tissue by immunohistochemistry (204, 206) or western blot (211).

13 Feb 2004

18:42

844

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD1d-RESTRICTED T CELLS CD1d-restricted T cells have been shown to contribute to antimicrobial responses, antitumor immunity, and the balance between tolerance and autoimmunity. The divergent roles of CD1-restricted T cells in promoting both inflammatory and tolerogenic responses can be understood on the basis of their ability to promptly secrete cytokines and cytotoxic granules, and to regulate the function of DCs, NK cells, T cells, and B cells. The existence of functionally distinct subsets of CD1drestricted T cells and modification of their effector functions by costimulatory pathways may explain the different outcomes following their activation.

Phenotype of CD1d-Restricted T Cells CD1d-reactive T cells were initially identified among NK1.1+ thymocytes from normal mice and CD4+ T cells in MHC class II–deficient animals (63, 100), and subsequently shown to be present also in peripheral blood of humans (64). Many CD1d-restricted T cells in mice and humans coexpress CD161, a cell surface molecule usually observed on NK cells that corresponds to the NK1.1 antigen, and therefore are often referred to as Natural Killer T (NKT) cells. However, it has become clear that CD3+CD161+ T cells are heterogeneous and that many do not recognize CD1d molecules but instead recognize other restriction elements, including MHC class I and II, TL, Qa-1, and H2-M3 (171, 212, 213). In uninfected C57BL/6 mice, 20%–80% of CD161+ T cells stained with αGalCer/CD1d tetramers, with spleen and bone marrow containing larger populations of NK1.1+αGalCer/CD1d tetramer– T cells than thymus and liver (92, 214). In humans, 20%–25% of T cells from PBMCs were CD161+, but only less than 1% of the CD161+ T cells are stained with αGalCer/CD1d tetramers (94). Furthermore, many αGalCer/CD1d-tetramer+ T cells do not express CD161. Importantly, following activation, expression of the CD161 antigen on T cells can be upregulated on a majority of conventional T cells (215, 216) and downregulated on CD1d-restricted T cells (217, 218). Thus, CD1d-restricted T cells and CD161+CD3+ T cells (NKT cells) are frequently not identical T cell subpopulations, and therefore use of the term NKT cells has become confusing and inaccurate. It is preferable to refer to them as CD1d-restricted T cells or CD1d tetramer+ T cells. CD1d-restricted T cells in mice and humans express additional receptors commonly found on NK cells. CD1d-restricted T cells in C57BL/6 mice express intermediate levels of IL-2 receptor β (CD122), and expression of the inhibitory NK receptors Ly49A, C/I and Ly49G2 has been reported on a subset of CD161+CD3+ T cells (219, 220). Many Vα24+/Vβ11+ T cells in human PBMCs express CD122 (221). Expression of CD94, a C-type lectin, has been reported on approximately 50% of αGalCer/CD1d tetramer+ T cells in human PBMCs (94), whereas the immunoglobulin superfamily killer inhibitory receptors (KIRs) CD158a/b appeared to be absent on human CD1d-restricted T cells (95).

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

845

A striking feature of most murine and human CD1d-restricted T cells is their expression of markers associated with recently activated or memory T cells. In mice, most CD1d-restricted T cells are CD44hiCD69intCD45RBhiCD62LloCCR7neg, and in humans, most CD1d-restricted T cells are CD45RO+CD45RA–CD25+CD62L– CCR7–, but only 5%–15% express CD69 (221–224). Interestingly, CD1d-restricted T cells in human cord blood and in germ-free mice display this activated/memory surface phenotype, suggesting that previous exposure to foreign microbial antigens is not the reason for this phenotype and that stimulation by CD1d-presented selfligands is likely to be sufficient (223–225). In addition, CD1d-restricted T cells in mice, and to a lesser extent in humans, express intermediate levels of TCR at the cell surface, which may be the consequence of continuous low-level TCR stimulation provided by recognition of self-antigens that are constitutively presented by CD1d. CD1d-restricted T cells are usually CD4+ or double negative (DN). In mice, 60%–90% of CD1d-restricted T cells have been reported to be CD4+ and 10%– 40% to be DN, and only very few appear to express CD8α or CD8β (92, 93, 214). In humans, a mean of 50% of αGalCer/CD1d tetramer+ T cells are CD4+, with high donor-to-donor variability, and CD8α expression is common, but only very few CD8β + CD1d-restricted T cells exist (<1% in unstimulated PBMCs) (94, 104, 226). The surface phenotype of murine and human CD1d-restricted T cells that do not use Vα14Jα18 or Vα24Jα18 TCRs, respectively, has been difficult to determine on freshly isolated cells owing to the lack of specific reagents for their detection. Analysis of the surface phenotype of murine Vα14Jα18– CD1d-restricted T cell clones or T cells from mice with transgenic expression of such TCRs showed expression of markers of the NK lineage and of activation/memory markers similar to that of Vα14Jα18+ CD1d-restricted T cells (102, 105, 227).

Development and Selection of CD1d-Restricted T Cells Like MHC class I– and II–restricted T cells, most CD1d-restricted T cells appear to originate in the thymus (228–230). There also exists evidence for the additional origin of CD1d-restricted T cells in developing liver and bone marrow (231, 232). Recent studies using αGalCer/CD1d tetramers have identified HSAlowCD44lowCD161– Vα14+ T cells as precursors of Vα14+CD161+ CD1drestricted T cells (228–230). The ultimate precursors of Vα14+ CD1d-restricted T cells appear to be CD4+CD8+ (DP) thymocytes that can also give rise to MHC class I– and II–restricted CD8+ and CD4+ T cells (228–230). CD1d-restricted T cells can first be detected in thymus, spleen, and liver 5–8 days after birth, and their numbers in spleen and liver reach a plateau at 4–6 weeks of age (230). Positive selection of TCRα-invariant CD1d-restricted T cells appears to be mediated by CD1d-expressing CD4+CD8+ DP immature cortical thymocytes and not by thymic epithelial cells (233). However, the structures of CD1d-presented ligands that are used for positive selection of CD1d-restricted T cells are not known.

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

846

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Reduced numbers of αGalCer/CD1d tetramer+ T cells have been observed in fetal thymic organ cultures exposed to αGalCer, in mice repeatedly injected with αGalCer from day 3 after birth on, and in transgenic mice overexpressing CD1d, suggesting that CD1d-restricted T cells undergo negative selection when engaged by high-avidity antigens or abundant self-antigens during development (234, 235). Negative selection could be mediated by bone marrow–derived dendritic cells overexpressing CD1d, but not by thymic epithelial cells (234). After completion of their TCRα and β chain rearrangement, CD1d-restricted T cells detected in thymus with αGalCer/CD1d tetramers do not yet appear to express NK receptors, such as CD161 or members of the Ly-49 inhibitory NK receptor family (228– 230). Subsequently, NK receptor– CD4+/DN thymic precursors have been shown to divide rapidly and to acquire expression of NK receptors. The delayed expression of inhibitory NK receptors on CD1d-restricted T cells during development has been proposed as an evolutionarily conserved mechanism to terminate their expansion and to control the autoreactivity of these cells (236). Immature thymic NK receptor−αGalCer/CD1d tetramer+ T cells appear to produce much more IL-4 than IFN-γ mRNA and protein, whereas in mature NK receptor+ CD1d-restricted T cells this balance reverts in favor of IFN-γ (229, 230, 264). Export of most CD1d-restricted T cells from the thymus seems to occur before the expression of NK receptors, indicating that complete maturation of CD1d-restricted T cells occurs mainly post-thymically. Several gene mutations selectively disturb the development of CD161+CD3+ T cells or Vα14Jα18+ CD1d-restricted T cells, but do not affect the development of conventional CD4+ and CD8+ T cells, indicating additional differences between the developmental pathways of T cells restricted by either CD1d or MHC class I and II. These gene mutations include the tyrosine kinases Fyn (237, 238) and Itk (239); the transcription factors Ets-1 (240), Irf-1 (241), and RelB (242, 243); IL-15 (244) and the cytokine receptors IL-15Rα (245), IL-2/IL-15Rβ but not IL-2 (246) and GM-CSFRβ (247); and membrane lymphotoxin (248, 249).

Tissue Distribution, Localization, and Recruitment of CD1d-Restricted T Cells In mice, of the total lymphocytes per organ, αGalCer/CD1d tetramer+ T cells comprise approximately 0.5%–1% in thymus, 1%–2% in spleen, 0.5% in lymph nodes, 10%–50% in liver, 0.5% in bone marrow, and 1% of intestinal intraepithelial lymphocytes (IEL) (92, 214). In humans, Vα24+ CD1d-restricted T cells account for a mean of 0.2% of peripheral blood T cells, as determined by use of both Vα24/Vβ11 antibodies and αGalCer/CD1d tetramers (85, 94, 104). Their numbers in liver of humans also are lower than they are in liver of mice (250–252). The chemokine receptor and homing molecule profile of many CD1d-restricted T cells in mice and humans is more like that of effector (memory) T cells than that of na¨ıve T cells. In humans, for example, 50%–90% of Vα24+/Vβ11+

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

847

CD1d-restricted T cells expressed CCR5, CXCR3, and CXCR6, chemokine receptors associated with TH1 responses and migration to sites of inflammation, and only about 20% of Vα24+/Vβ11+ CD1d-restricted T cells expressed CCR7 and almost no peripheral blood Vα24+/Vβ11+ CD1d-restricted T cells expressed CXCR5, both receptors required for migration into T and B cell zones of lymphoid organs (94, 95, 253, 254). In mice, most Vα14+ CD1d-restricted T cells expressed CXCR3 and CXCR6 (255). Only murine NK1.1– Vα14+ CD1d-restricted T cells that are considered to be immature and recently released from the thymus (229, 230), but not NK1.1+ Vα14+ CD1d-restricted T cells, migrated in response to the CCR7 ligand SLC/CCL21 (255). In mice, Vα14+ CD1d-restricted T cells from spleen, but not liver, bone marrow, or blood, migrated in response to the CXCR5 ligand BCA-1/CXCL13, suggesting that these cells could be recruited to B cell–rich areas in the spleen and thus modulate the follicular B cell response. Interestingly, some chemokine receptors appear to be differentially expressed by CD4+ and CD4– CD1d-restricted T cell subsets. For example, in humans, CCR4 was preferentially expressed by CD4+ CD1d-restricted T cells, whereas the chemokine receptors CCR6 and CXCR6 were preferentially expressed on CD4– CD1d-restricted T cells, and CD4+ and CD4– CD1d-restricted T cell subsets preferentially migrated in response to the CCR4 ligand TARC and the CCR6 ligand LARC, respectively (253, 254). Similarly, differences in the chemotactic response of CD4+ and DN Vα14+ CD1d-restricted T cells have been observed in mice (255). Thus, particular subsets of CD1d-restricted T cells with differential expression of chemokine receptors may be recruited to different sites. Up to 20% of human Vα24+ CD1drestricted T cells expressed cutaneous lymphocyte antigen (CLA), associated with homing to skin, and 30%–75% stained positive for the integrin α 4β 7, associated with homing to gut (94, 253). In contrast, only a small proportion of CD1drestricted T cells in humans and mice expressed high levels of L-selectin (CD62L) that would allow entry into secondary lymphoid organs via high endothelial venules (HEVs). Together, these features suggest that most CD1d-restricted T cells recirculate through peripheral tissues and enter lymph nodes most likely through the afferent lymphatics rather than through HEVs. CD1d-restricted T cells can be recruited rapidly to inflamed and infected tissues in vivo. Injection of mycobacterial lipids into skin of mice leads to a granulomatous reaction that depends on Vα14Jα18+ T cells (256–258). Vα14Jα18+ T cells could be detected at the injection site within 6 h. The mechanism of recruitment of Vα14Jα18+ T cells to these lipid-induced granulomas is unclear, although it has been shown that CD1d expression is not required and that Vα14Jα18+ T cells can be recruited by subcutaneous injection of TNF-α alone (258). Vα14+ T cells accumulated in the lungs of mice during pulmonary infection with Cryptococcus neoformans, and this recruitment was dependent on the CCR2 ligand MCP-1 (259). During anterior chamber associated immune deviation (ACAID), CD161+ CD3+ T cells are recruited to the spleen in a CXCR2- and MIP-2-dependent manner (260).

13 Feb 2004

18:42

848

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Activation and Effector Functions of CD1d-Restricted T Cells CYTOKINE SECRETION AND CYTOTOXICITY In mice and humans, TCRα-invariant CD1d-restricted T cells secrete large amounts of IFN-γ , IL-4, IL-2, IL-5, IL-10, IL-13, GM-CSF, and TNF-α within minutes after TCR stimulation (94, 261, 262), a property that distinguishes these T cells from na¨ıve MHC class I– and II–restricted T cells that acquire their ability to secrete cytokines during proliferation after primary stimulation. IL-4 and IFN-γ proteins could only be detected in activated, but not in resting, αGalCer/CD1d-tetramer+ T cells (92). However, abundant mRNA transcripts for IL-4 and IFN-γ were detected in resting αGalCer/CD1d-tetramer+ T cells (263, 264). Unlike conventional T cells, TCRα-invariant CD1d-restricted T cells activate IL-4 and IFN-γ transcription already during thymic development and populate the periphery with both cytokine loci previously modified by histone acetylation (264). This constitutive cytokine mRNA expression may allow the rapid cytokine production and secretion by TCRα-invariant CD1d-restricted T cells. The events following activation of CD1d-restricted T cells by αGalCer involve TCR ligation by the CD1d/αGalCer complex and interactions of costimulatory molecules that lead to T cell activation, resulting in cytokine secretion and increased CD40L expression (265) (Figure 8). In mice, CD1d-restricted T cells constitutively express CD28, and its interaction with CD80 and CD86 expressed by APCs augmented both IFN-γ and IL-4 secretion by the T cells after TCR-mediated activation (262, 266). Through CD40/CD40L interaction and IFN-γ stimulation, DCs are then activated and secrete IL-12 (172, 265, 267) (Figure 8). Vα14Jα18+ T cells showed a constitutively high expression of IL-12 receptor (IL-12R), and IL12R expression on these cells increased in an IL-12- and IFN-γ -dependent manner (265, 275). Although IL-12Rβ2 and IL-4Rα expression on CD1d-restricted T cells and CD40 expression on APCs were not required for the immediate IL-4 and IFN-γ production by CD1d-restricted T cells following αGalCer stimulation (263), available IL-12 and IFN-γ can amplify the activation of CD1d-restricted T cells and augment their IFN-γ secretion (265–268) (Figure 8). Inhibitory NK receptors, such as Ly49A, C/I, and Ly49G2, expressed on CD161+CD3+ T cells appear to be capable of inhibiting TCR-mediated activation, suggesting that downregulatory mechanisms may counterbalance the activation of CD1d-restricted T cells (219, 220). In addition to the prompt cytokine secretion, CD1d-restricted T cells are also potently cytolytic and release perforin and granzymes and express membranebound members of the TNF family [FasL or TNF-related apoptosis-inducing ligand (TRAIL)] (269–271). The cytotoxicity of CD1d-restricted T cells is augmented by IL-2 and IL-12 and is likely to be important in antimicrobial and antitumor immune responses (272, 273). Furthermore, human CD1d-restricted T cell clones expressed granulysin after activation with αGalCer, suggesting that they can contribute directly to bacterial killing (274).

TH1/TH2 BALANCE The in vivo responses of CD1d-restricted T cells may be influenced by a variety of local factors present during their activation. TH1 responses

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

849

Figure 8 Effects of αGalCer recognition. αGalCer is taken up by APCs and presented to CD1d-restricted T cells. This antigenic stimulus potently elicits production of both TH1 cytokines, such as IFN-γ , and TH2 cytokines, such as IL-4 and IL-13, from CD1drestricted T cells. Recognition of αGalCer induces CD1d-restricted T cells to secrete IFN-γ and upregulate CD40L, which stimulates the maturation of immature DCs and the production of IL-12 from DCs. IFN-γ and IL-12 activate NK cells and T cells, whereas TH2 cytokines activate B cells. IFN-γ released by CD1d-restricted T cells, NK cells, and T cells activates macrophages. NK, NK cell; T, T cell; B, B cell; Mø, macrophage.

were favored by cytokines such as IL-12, which augmented IFN-γ production, or by signaling through CD161 on murine T cells, which induced IFN-γ secretion and polarized the cytokine response of activated CD1d-restricted T cells to TH1 (275, 276). The TH1/TH2 bias of the cytokine response of CD1d-restricted T cells during cerebral malaria infection in mice was influenced by genes located in the NK complex, further indicating a role for NK receptors on cytokine production (277). TH2-type cytokine responses were favored when CD1d-restricted T cells were activated in the presence of IL-7 or IL-18 (278–280). Neonatal, and to a lesser extent adult, Vα24+/Vβ11+ T cells differentiated preferentially into TH1-type T cells when stimulated with PHA in the presence of monocyte-derived DCs. Alternatively, when stimulated by CD4+CD3–CD11c– plasmacytoid cell-derived DCs, they differentiated into TH2-type T cells, suggesting that distinct subsets of DCs influence their cytokine production (281). An αGalCer derivative that lacks one alkyl chain also stimulated TH2-biased cytokine secretion (282). In mice, administration of a single dose of αGalCer elicits IFN-γ and IL-4 secretion, whereas repeated αGalCer applications tend to induce a TH2 bias in CD1d-restricted T cell cytokine secretion (263, 283, 284). In contrast, altering the antigen dose or slowly delivering antigen by an osmotic pump, manipulations that polarize na¨ıve CD4+

13 Feb 2004

18:42

850

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC class II–restricted T cells, did not alter the ratio of intracellular IFN-γ to IL-4 staining of CD1d-restricted T cells in response to αGalCer (263). FUNCTIONALLY DISTINCT SUBSETS CD4 coreceptor expression on human CD1drestricted T cells correlates with their effector functions. The CD4– subset produced mainly TH1 cytokines and prominently expressed perforin after activation, whereas the CD4+ subset potently produced both TH1 and TH2 cytokines (94, 95, 285). This suggests that different activities of CD1d-restricted T cells could result from activation of such functionally distinct subsets. Interestingly, as discussed above, analysis of CD4+ and CD4– subsets of CD1d-restricted T cells revealed differences in their expression of chemokine receptors and in their chemotactic responses to chemokines. Thus, CD1d-restricted T cell subsets with distinct functional capabilities may be recruited to different sites. INTERACTION WITH OTHER CELL TYPES One major attribute of CD1d-restricted T cells is their ability to stimulate many other cell types. After the initial activation of CD1d-restricted T cells, the IFN-γ and IL-2 that they secrete in combination with IL-12 produced by DCs can lead to activation of NK cells and memory CD4+ and CD8+ T cells (284, 286–289) (Figure 8). This activation seems to increase IFN-γ secretion and cytotoxicity by these cell types. IFN-γ from all these sources may contribute to the activation of macrophages (290) (Figure 8). IL-4 secretion, and likely CD40/CD40L interactions, activate B cells, which can lead to their proliferation and production of immunoglobulins (284, 286, 291, 292) (Figure 8). αGalCer-induced activation of CD1d-restricted T cells has many effects on DCs, including differentiation and maturation (172, 293, 293a), induction of IL-12 secretion (172, 265, 267), killing of DCs (294, 295), and release of myeloid progenitors from bone marrow to the periphery (296). The maturation of DCs as a consequence of αGalCer administration in mice led to an adjuvant-like induction of the CD4+ and CD8+ T cell response to a coadministered protein (293, 293a, 293b). This enhancement of the CD8+ and CD4+ T cell response required CD40 signaling but not signaling through the IFN-γ receptor, indicating that IFN-γ was not required for this adjuvant-like effect of αGalCer (293a). FATE FOLLOWING ACTIVATION A number of studies have found that murine CD161+CD3+ T cells become undetectable in vivo soon after stimulation with anti-CD3 antibodies (297), αGalCer (298) or IL-12 (297), and during viral (299) and bacterial (300) infections. Subsequent analyses that used αGalCer/CD1d tetramers to more specifically detect CD1d-restricted T cells appeared to confirm their disappearance after stimulation with αGalCer in vivo (92, 301). A fraction of CD161+CD3+ T cells and αGalCer/CD1d tetramer+ T cells express markers of cells undergoing apoptotic cell death following activation in vivo, suggesting that the disappearance of CD1d-restricted T cells is due to cell death. However, the extent to which this activation-induced cell death occurs is controversial (298, 301, 302). In contrast, other studies have found that rather than dying, CD1d-restricted

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

851

T cells persist and/or alter expression of their key surface markers. Downregulation of CD161 has been observed on CD161+CD3+ T cells after anti-CD3 stimulation in vitro, on αGalCer/CD1d tetramer+ T cells after stimulation with αGalCer or IL-12, and during Salmonella typhimurium infection in vivo (217, 218, 303). CD1d-restricted T cells appear to lose αGalCer/CD1d tetramer staining gradually within hours after i.p. administration of αGalCer (218, 303a), suggesting that CD1d-restricted T cells, like MHC-restricted T cells, downregulate their TCR surface levels following activation. Using αGalCer/CD1d tetramer staining, CD1d-restricted T cells were observed to expand after αGalCer administration (218, 303a, 303b) and during infection with Plasmodium berghei (277). Following S. typhimurium infection, the total numbers of CD1d-restricted T cells in spleen and liver did not change significantly during the first three days of infection (218, 304). Thus, following activation, a fraction of CD1d-restricted T cells appears to undergo activation-induced cell death. However, a number of CD1d-restricted T cells seem to survive, downregulate their TCR and CD161, and appear capable of robust in vivo proliferation and expansion. The population dynamics of CD1drestricted T cells at sites of infection or inflammation are therefore likely the result of a combinatin of cell death, proliferation, and recruitment.

CD1d-RESTRICTED T CELLS IN ANTIMICROBIAL IMMUNITY CD1d-restricted T cells contribute to antimicrobial host responses in bacterial, parasitic, viral, and fungal infections. Depending on the infection studied, CD1drestricted T cells have been found to control growth of microorganisms, influence antibody production and adaptive immune responses against them, and contribute to immunopathologic tissue injury.

Bacterial Infections In Pseudomonas aeruginosa infection, CD1d-deficient animals and mice treated with anti-CD1d mAb had impaired bacterial clearance from their lungs and decreased numbers of neutrophils in the broncho-alveolar lavage during early infection (290). Decreased levels of MIP-2, a chemokine secreted by activated alveolar macrophages and able to recruit neutrophils into infected lungs, were noted in broncho-alveolar lavage of infected CD1d-deficient animals. Administration of αGalCer prior to P. aeruginosa infection improved the clearance of bacteria from the lungs through enhanced phagocytosis of bacteria by activated alveolar macrophages. Thus, CD1d-restricted T cells may contribute to bacterial clearance by activation of macrophages and recruitment of neutrophils (290). In a model of Lyme disease, mouse strains normally resistant to the spirochete Borrelia burgdorferi developed arthritis and had increased spirochete DNA in tissues when CD1drestricted T cells were absent (305). Infected CD1d-deficient animals had enhanced production of spirochete-specific IgG2a antibodies that are commonly induced in

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

852

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

disease-susceptible mice. Hence, CD1d-restricted T cells appear able to modulate B cell activation and antibody production during infection. In a noninfectious model, injection of lipids from M. tuberculosis cell walls results in the rapid accumulation of Vα14Jα18+ T cells in subcutaneous granulomas (256–258). CD1d expression has been observed in tuberculoid granulomas in humans (274), and Vα24Jα18 transcripts have been detected in T cell–reactive lepromatous lesions together with CD1d expression on CD83+ cells (306). Together, these studies suggest that CD1d-restricted T cells also may contribute to antimycobacterial immunity. However, mice deficient in CD1d-restricted T cells do not show impaired protective immunity to respiratory or i.v. infection with M. tuberculosis or M. bovis BCG (307–311). In contrast, administration of antiCD1d mAb impaired early immunity to M. tuberculosis infection (312), and in one study Jα18-deficient animals were marginally more susceptible to infection with M. tuberculosis (313). Following i.v. infection with M. bovis BCG, Jα18deficient mice developed more granulomas that had signs of caseation and larger cellular infiltrates, suggesting that Vα14+ CD1d-restricted T cells may play a regulatory role by limiting lymphocyte influx and tissue pathology (311). Thus, although not absolutely required for protective immunity, CD1d-restricted T cells appear to participate in the immune response to M. tuberculosis and M. bovis BCG infection. Most examples suggest that CD1d-restricted T cells are beneficial in bacterial infection. However, anti-CD1d mAb blocking during infection with Listeria monocytogenes improved survival and the pathogen-specific TH1 immune response (314). Activation of CD1d-restricted T cells during the immune response to microbial infection can also lead to increased tissue damage, as suggested by the finding that Jα18-deficient mice had less severe liver damage during Salmonella choleraesuis infection compared with controls (315).

Parasitic Infections Jα18-deficient animals that lack TCRα-invariant CD1d-restricted T cells have increased parasite burden during the course of Leishmania major infection (316). A role for CD1d-restricted T cells in the humoral immune response to GPI-linked antigens from Plasmodium spp. and Trypanosoma cruzi has been observed in mice deficient in CD1d or the Jα18 gene, but these effects remain controversial and an influence on survival has not been found by all investigators (57–61, 317, 317a). The role of CD1d-restricted T cells during infection with a strain of Plasmodium berghei that causes cerebral malaria was dependent on the genetic background of infected mice. For example, CD1d-deficient BALB/c mice were more susceptible to infection and had more severe cerebral pathology than wild-type BALB/c mice, whereas CD1d-deficient and Jα18-deficient C57BL/6 mice were partially protected against disease compared with controls (277). This study showed further that the functional properties of CD1d-restricted T cells appear to vary according to the expression of loci of the NK complex, suggesting that NK receptors may

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

853

influence the cytokine response of CD1d-restricted T cells during infection (277). After inoculation with Schistosoma mansoni eggs, CD1d-deficient animals had a reduced protective TH2 response and developed less marked fibrotic granuloma pathology (318). CD4+CD161+ T cells from MHC class II–deficient animals augmented the protective CD8+ T cell response to Toxoplasma gondii infection after vaccination with attenuated T. gondii, suggesting that CD1d-restricted T cells can help in priming a pathogen-specific CD8+ T cell response (319). Thus, for a variety of parasitic infections in mouse models, CD1d-restricted T cells have been implicated in protective TH2 immunity and in modulating antiparasite antibody responses.

Viral Infections Mice deficient in CD1d-restricted T cells were more susceptible to infection with herpes simplex virus type 1 and 2 (HSV-1/-2) (320, 321) and diabetogenic encephalomyocarditis virus (EMCV-D) (323). Following respiratory syncytial virus (RSV) infection, CD1d-deficient C57BL/6 mice developed more severe disease, whereas CD1d-deficient BALB/c and 129 × C57BL/6 mice were less ill compared with controls (322). In CD1d-deficient BALB/c mice, numbers of virusspecific CD8+ MHC class I–restricted T cells were reduced, suggesting that CD1drestricted T cells influence the adaptive immune response to RSV infection (322). In many infection models, the protective effect of CD1d-restricted T cells can be mediated by Vα14+ TCRα-invariant CD1d-restricted T cells. A role for CD1drestricted T cells using diverse TCRα chains, however, was observed in the immune response of mice to hepatitis B antigens and EMCV-D infection and in hepatits C–infected patients (108, 323, 324). In contrast to protective effects, a CD1dmediated immune response appeared to be reponsible for the tissue pathology following infection with coxsackie virus B3 (109). Wild-type and Jα18-deficient, but not CD1d-deficient animals, developed myocarditis that was mediated by CD1drestricted Vγ 4+ T cells. Thus, CD1d-restricted T cells that use TCRα-invariant and diverse TCRs, as well as γ δTCRs, are implicated in reponses to viral infections.

Fungal Infections Jα18-deficient animals had delayed clearance of Cryptococcus neoformans from lung and an impaired in vitro recall response and delayed type hypersensitivity reaction to cryptococcal antigen, suggesting that CD1d-restricted T cells contribute to the development of an adaptive immune response to this pathogen (259).

Mechanism of the Antimicrobial Effect Studies with CD1d- and Jα18-deficient mice demonstrate a role for CD1d-restricted T cells during the normal host response to microbial infection that can be mediated either directly by CD1d-restricted T cells themselves through both killing of infected cells and microbicidal effects, or indirectly through recruitment and

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

854

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Figure 9 Role of CD1d-restricted T cells during microbial infection. Upon exposure to microorganisms, APCs become activated through Toll-like receptors (TLR) and secrete proinflammatory cytokines, such as IL-12. Signals provided by IL-12 amplify the weak TCR-mediated responses of CD1d-restricted T cells to self-lipid antigens presented by CD1d on the APCs, and lead to productive activation and cytokine secretion by CD1d-restricted T cells. This mechanism of activation may explain how TCRαinvariant CD1d-restricted T cells with their restricted TCR-specificity can become rapidly activated early in many different microbial infections. The early cytokine secretion by CD1d-restricted T cells during infection leads to activation of macrophages, NK cells, T cells, and B cells. Chemokines secreted by CD1d-restricted T cells contribute to the recruitment of other immune cells including granulocytes. Activated CD1d-restricted T cells can activate or kill macrophages directly. IFN-γ secretion and CD40L stimulation lead to maturation of immature DCs. TLR, Toll-like receptor; NKT, CD1d-restricted T cell; Gr, granulocyte; Mø, macrophage; NK, NK cell; T, T cell; B, B cell.

activation of cells of the innate immune system or through modulation of the adaptive immune response (Figure 9). A major unresolved question is whether CD1d-restricted T cells directly recognize foreign microbial antigens. An alternative mechanism has been proposed in which CD1d-restricted T cells can be activated by microbial products through stimulation by the inflammatory cytokine IL-12, secreted by DCs after exposure to bacterial products, in combination with recognition of CD1d-presented self-lipids (304) (Figure 9). This suggests that the activation of CD1d-restricted T cells during microbial infection may not require TCR-mediated recognition of foreign microbial antigens and would allow

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

855

their stimulation early and to virtually any infectious agent that stimulates IL-12 secretion.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Activation of CD1d-Restricted T Cells with αGalCer in Microbial Infection In addition to their role during the natural course of infection, many studies have investigated the effects of pharmacological activation of CD1d-restricted T cells by αGalCer and have demonstrated dramatic protective effects against infection with P. aeruginosa, M. tuberculosis, Plasmodium spp., T. cruzi, hepatitis B, EMCV-D, RSV, mCMV, and C. neoformans (290, 322, 323, 325–330). As described earlier, the specific pharmacologic activation of CD1d-restricted T cells by αGalCer may improve host control of microbial infections by widespread immune activation (Figure 8). This “adjuvant” effect seems also to be reponsible for the improved efficacy of vaccines when administered together with αGalCer (331). Thus, activation of CD1d-restricted T cells by pharmacological compounds like αGalCer may have therapeutic potential in humans for the treatment of infectious diseases and the improvement of vaccine efficacy.

CD1d-RESTRICTED T CELLS IN ANTITUMOR IMMUNITY CD1d-restricted T cells potently promote tumor rejection in murine models that use exogenously administered IL-12 or αGalCer as a stimulus, but they also contribute to the natural antitumor immune response in the absence of exogenous stimulation.

IL-12-Mediated Antitumor Immunity Studies in Jα18-deficient animals and studies using adoptive transfer of IL-12activated Vα14+ T cells indicated that CD1d-restricted T cells were crucially important for the antitumor and antimetastatic activity of IL-12 (332, 333). However, the relative contribution of CD1d-restricted T cells and NK cells is controversial (303, 334–338). IL-12-induced antitumor immunity seems to be mediated by IFN-γ , secreted by CD1d-restricted T cells and NK cells, and perforin-mediated killing by NK cells. The antitumor effects mediated by IFN-γ appear to include direct inhibition of tumor growth and angiogenesis and activation of other effector cell types (339, 340). In addition, IFN-γ can function to upregulate tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) on NK cells, thus allowing killing of TRAIL-sensitive tumor targets (341).

αGalCer-Mediated Antitumor Immunity αGalCer was initially purified from marine sponges on the basis of its antitumor properties, and was later shown to be presented by CD1d to Vα14Jα18+ T cells (72,

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

856

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

342). Antitumor effects of αGalCer have been observed against various tumors of different origins and their metastasis, including melanomas, colon, lung, prostate, breast and renal cell carcinomas, and lymphomas (341–344). As described earlier, activation of Vα14Jα18+ CD1d-restricted T cells by αGalCer leads to initial IFN-γ secretion and CD40L upregulation on CD1d-restricted T cells, which subsequently promotes DC production of IL-12 (Figure 8). The secretion of IL-12 by DCs is required for the antitumor and antimetastatic effect of αGalCer, and together with the IFN-γ secreted by CD1d-restricted T cells, secondarily activates NK cells and CD8+ cytotoxic T lymphocytes (CTLs) that function as direct antitumor effectors (302, 343–345). The antitumor activity of these effector cells is dependent on their IFN-γ secretion and TRAIL expression on NK cells, but, surprisingly, is independent of perforin and FasL expression (302, 341, 344, 345). Improved efficiency of tumor-rejection and a prolonged IFN-γ response were observed when DCs pulsed with αGalCer were used instead of αGalCer administration alone (301, 346). Thus, although activated Vα14Jα18+ T cells display direct NK-like nonspecific tumor cell lysis in vitro (270), the in vivo antitumor and antimetastasis responses of CD1d-restricted T cells activated by αGalCer are mainly mediated by IFN-γ - and IL-12-dependent subsequent activation of NK cells and CD8+ CTLs.

Natural Tumor Immunosurveillance Studies using chemical mutagenesis with methylchlorantrene (MCA) suggest that CD1d-restricted T cells contribute to natural tumor immunosurveillance, in the absence of exogenous stimulation of Vα14Jα18+ T cells by αGalCer or IL-12 (347). Natural host immunity against MCA-induced sarcoma required endogenous IL-12 production, early IFN-γ production by CD1d-restricted T cells, and IFN-γ expression by non-CD1d-restricted T cells, possibly NK cells and CD8+ T cells (348). Tumor rejection was dependent on both NK cells and CD8+ T cells and was perforin-dependent, whereas CD1d-restricted T cells were required but mediated their antitumor effects in a perforin-independent manner. This suggests that NK cells and CD8+ T cells were the direct antitumor effectors (Figure 10). Recognition of CD1d-presented antigens may be required for the natural antitumor activity of CD1d-restricted T cells, since adoptive transfer experiments showed that a CD1d-positive host environment was essential (349). Whether recognition of CD1d-presented self-antigens—unaltered or altered—by CD1d-restricted T cells is simply required in addition to other, CD1d-independent, tumor-induced “danger signals” (e.g., IL-12) or whether CD1d-restricted T cell activation is caused by the presentation of tumor-derived antigens in the context of CD1d remains to be determined. Recently, CD1d-mediated recognition of GD3, a ganglioside enriched in melanoma cells, has been reported and supports the latter possibility (70). Such tumor-derived lipid-antigens may activate CD1d-restricted T cells when presented by CD1d+ APCs that have taken up tumor cell fragments (70) or by CD1d+ tumors themselves.

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES

P1: IKH

857

Figure 10 Model of CD1d-restricted T cell function in natural tumor immunosurveillance. CD1d-restricted T cells can mediate tumor rejection in the absence of exogenous stimulation through activation by endogenous or tumor-derived lipid antigens (upper part of figure). This tumor rejection depends on CD1d-recognition, IL-12, and IFN-γ , and the final antitumor effector cells are NK cells and CD8+ CTLs that recognize tumor-derived peptide antigens. NK cells and CTLs kill tumor cells directly in a perforin-dependent manner and secrete IFN-γ , which has direct antitumor effects, including inhibition of tumor angiogenesis, and may activate other effector cells. CD1d-restricted T cells can also suppress antitumor immunity (lower part of figure). IL-13 produced by CD1d-restricted T cells acts on CD11b+Gr-1+ myeloid cells and induces the secretion of inhibitory cytokines, such as TGF-β, which in turn leads to inhibition of tumor-specific CTLs and suppression of antitumor immunity. NKT, CD1d-restricted T cell; NK, NK cell; CD8+, tumor-specific CD8+ T cell.

Cancer Vaccination CD1d-restricted T cells have also been shown to participate in antitumor responses induced by cancer vaccination. Antitumor immunity against early tumor growth, induced by adoptive transfer of lymphocytes from draining lymph nodes of mice immunized with tumor-specific antigen, was abrogated in Jα18-deficient mice (350). Antitumor immunity stimulated by vaccination with irradiated melanoma cells engineered to secrete GM-CSF was abrogated in CD1d- and Jα18-deficient mice. DCs from immunized CD1d-deficient mice showed compromised maturation and impaired ability to stimulate T cells, suggesting that DCs may be critical for regulating vaccine-induced antitumor immunity (351).

13 Feb 2004

18:42

858

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Suppression of Antitumor Immunity Paradoxically, CD1d-restricted T cells can also suppress natural antitumor responses. This was shown in a mouse tumor model in which the tumor regressed after initial growth because of a tumor-specific CD8+ CTL response, but then recurred because of subsequent suppression of the CTL response mediated by CD1d-restricted T cells (352, 353). The suppression of the antitumor CTL response appears to require IL-13 production by CD1d-restricted T cells and subsequent IL-13-induced TGF-β secretion by CD11b+Gr-1+ myeloid cells (354) (Figure 10). Additional evidence for suppression of antitumor immune responses by CD1d-restricted T cells comes from studies in CD1d-deficient animals where CpG oligodinucleotide (ODN)-induced antitumor activity is enhanced compared with wild-type animals (355). Interestingly, the CpG-ODN-induced antitumor activity was present in CD1d-deficient but not in Jα18-deficient or wild-type animals, suggesting that CD1d-restricted T cells with TCRs other than the invariant Vα14Jα18+ TCR were mediators of the suppressive effect. Also, a CD4+DX5+ NKT cell population induced by UV-irradiation that suppressed skin cancer development and delayed type hypersensitivity (DTH) responses has been described, although CD1d-restriction has so far only been demonstrated for DTH suppression but not for the antitumor function of these NKT cells (356). Thus, in natural tumor immunosurveillance CD1d-restricted T cells can promote or inhibit the development of protective antitumor responses (Figure 10).

Antitumor Immunity in Humans Several studies document that human Vα24+/Vβ11+ T cells activated by stimulation with αGalCer can directly kill CD1d+ tumor cells in vitro in a predominantly perforin- and granzyme B-dependent manner, with additional TNF-α-, FasL-, and TRAIL-mediated killing (191, 289, 357–361). Human CD1d-restricted T cells can also mediate their antitumor activity by activating NK cells, an effect that can be caused by IL-2 or IFN-γ secretion by CD1d-restricted T cells (289, 362). Moreover, human CD1d-restricted T cells stimulated with αGalCer can induce IL-12 secretion by DCs, which may contribute to their antitumor activity (172). Thus, human CD1d-restricted T cells in vitro display direct CD1d-dependent cytotoxicity against tumor cells and also can activate NK cells to kill tumor targets. Numerical deficiencies of CD1d-restricted T cells, and in some cases their loss of IFN-γ production, have been reported for patients with solid tumors, multiple myeloma, or myelodysplastic syndrome, suggesting a correlation between advanced cancer and impaired CD1d-restricted T cell function (361, 363–367). A phase I clinical trial using i.v. αGalCer therapy to stimulate CD1d-restricted T cells in patients with solid tumors showed that αGalCer was well tolerated over a wide range of doses and cytokine responses were observed in several patients with relatively high pretreatment numbers of Vα24+/Vβ11+ T cells; however, no clinical responses were recorded (365).

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

859

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Thus, CD1d-restricted T cells function in tumor immunity when activated by αGalCer or IL-12 and are capable of modulating natural tumor immunosurveillance. They may be directly cytolytic against tumor cells or influence the activation of cytolytic NK cells and CD8+ CTLs, and their antitumor activity critically depends on IFN-γ . Recognition of tumor-derived glycolipids presented by CD1d may contribute to the activation of CD1d-restricted T cells in the absence of exogenous stimulation. Reconstitution and/or pharmacological activation of CD1d-restricted T cells with αGalCer may have therapeutic potential for certain human cancers.

CD1d-RESTRICTED T CELLS IN AUTOIMMUNITY, TOLERANCE, AND ALLERGY CD1d-restricted T cells help maintain tolerance to self-antigens and thereby prevent autoimmune disease. On the other hand, they also can mediate tissue damage and play a pathogenic role in autoimmunity.

Type 1 Diabetes Early reports using CD161+CD3+ or other surrogate markers for NKT cells have suggested numerical and functional deficiencies of CD1d-restricted T cells in nonobese diabetic (NOD) mice, a mouse strain that spontaneously develops a form of autoimmune diabetes that resembles human type 1 diabetes (368– 371). More recent studies using CD1d-tetramers have confirmed a deficiency of CD1d-restricted T cells (214, 372). The genetic control of Vα14+ CD1d-restricted T cell numbers in NOD mice was recently mapped to genomic regions implicated in conferring susceptibility to type 1 diabetes and lupus on mouse chromosomes 1 and 2 (373). An additional locus that controls Vα14+ CD1d-restricted T cell numbers in mice was recently identified on chromosome 18 (373a). CD1d-deficient NOD mice had earlier diabetes onset, greater disease penetrance, and more severe disease, suggesting that the deficiency of CD1d-restricted T cells is linked to the development of diabetes in NOD mice (374–376). In addition, the development of diabetes in NOD mice was ameliorated by interventions aimed at correcting the CD1d-restricted T cell deficiency such as adoptive transfer of CD1d-restricted T cells and transgenic expression of the invariant Vα14Jα18 TCR (369, 371, 377, 378). The functional defects observed in the remaining NKT cell population in NOD mice were a reduced ability to produce IL-4 (368) and a defect in both proliferation and differentiation towards an IFN-γ -secreting phenotype upon TCR engagement and IL-12 stimulation (378a). However, the functional defects of CD1d-restricted T cells in NOD mice warrant reexamination with more specific reagents for the characterization of these cells. Protection from diabetes conferred by CD1d-restricted T cells was associated with a TH2 shift within pancreatic islets, and IL-4 was implicated as a key mediator of the immunoregulation induced by CD1d-restricted T cells (377, 379). However, the mechanism of action and the

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

860

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Figure 11 Model for the function of CD1d-restricted T cells in preventing type 1 diabetes. Activation of CD1d-restricted T cells leads to tolerizing DCs that suppress the autoantigen-specific T cell response, possibly by secretion of TGF-β or IL-10, thus preventing tissue damage. The mechanism of action and the target cells of IL-4 have not been determined.

target cells of IL-4 have not been determined. Interestingly, when na¨ıve T cells expressing a transgenic TCR derived from a diabetogenic β cell-specific MHC class II–restricted T cell were stimulated with their auto-antigen in the presence of CD1d-restricted T cells, the initial activation of the β cell-specific T cells was not blocked, but both their production of IL-2 and IFN-γ and later proliferation were inhibited (380). The resulting β cell-specific T cells did not induce significant insulitis, and they were unable to destroy β cells. These findings suggest that CD1d-restricted T cells may prevent and ameliorate type 1 diabetes by preventing the differentiation of autoreactive T cells into effector cells (Figure 11). The inhibition of autoimmune inflammation by CD1d-restricted T cells may occur through the induction and/or recruitment of tolerogenic DCs (376) (Figure 11). In addition to a role in the natural course of type 1 diabetes, stimulation of CD1d-restricted T cells by administration of αGalCer in NOD mice prevented the onset and recurrence of diabetes and prolonged the survival of pancreatic islets transplanted into newly diabetic NOD mice (372, 374, 376, 381). αGalCer-induced protection from diabetes was associated with suppression of both T and B cell autoimmunity and IFN-γ production, and the generation of tolerogenic islet autoantigen-specific T cells with a protective cytokine production profile (372, 381). There are also reports of numeric and functional deficiencies of CD1d-restricted T cells in human type 1 diabetics (382, 383), but one study found no difference in αGalCer/CD1d tetramer+ cells between diabetics and control donors (384), and another study

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

861

found evidence of increased numbers of Vα24+/Vβ11+ T cell in patients with type 1 diabetes (385).

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Transplant Tolerance CD1d-restricted T cells can contribute to transplant graft acceptance. Long-term survival of corneal allografts was impaired in Jα18-deficient compared with wildtype animals and correlated with reduced generation of allospecific T regulatory cells, whereas the initial rejection was comparable to that in wild-type animals (386). Induction of long-term acceptance of cardiac allografts induced by blockade of LFA-1/ICAM-1 or CD28/B7 interactions was dependent on Vα14Jα18+ T cells (387), and immune tolerance to combined heart and bone marrow MHCmismatched transplants after fractionated lymphoid irradiation failed to develop in CD1d-deficient animals (388). Vα14Jα18+ T cells are also required for the acceptance of rat islet xenografts in mice treated with anti-CD4 antibody (389). Thus, CD1d-restricted T cells can function as immunosuppressive regulatory T cells that contribute to the induction of tolerance to transplant antigens.

Anterior Chamber-Associated Immune Deviation Another example of a role for CD1d-restricted T cells in the induction of tolerance is the anterior chamber-associated immune deviation (ACAID), where systemic tolerance is induced to antigens introduced to the anterior chamber of the eye. Tolerance, as evidenced by a deficiency in the antigen-specific delayed-type hypersensitivity response at peripheral sites, was abrogated in CD1d-deficient mice (390). During tolerance induction, CD1d-restricted T cells were recruited to the spleen in a MIP-2-dependent manner, whereas the chemokine RANTES, secreted by CD1d-restricted T cells, was required for the recruitment of APCs and CD8+ T cells to the spleen (260, 391, 392). IL-10, produced by CD1d-restricted T cells, was required for the induction of tolerance in ACAID, but the ultimate suppressor cell appears to be a non-CD1d-restricted antigen-specific CD8+ T cell (393).

Experimental Autoimmune Encephalomyelitis An important role for CD1d-restricted T cells has been shown in EAE, a mouse model for multiple sclerosis. SJL mice, which have the tendency to develop chronic EAE, show numerical deficiencies in CD161+ T cells similar to those observed in NOD mice (394). During EAE, CD1d expression has been observed on microglia and on macrophages in the central nervous system (395). Transgenic expression of the Vα14Jα18+ TCR in NOD mice, another mouse strain that is highly susceptible to EAE induction, protected mice from EAE and was associated with a striking inhibition of auto-antigen-specific IFN-γ secretion (396). This protection was independent of IL-4. Studies of the EAE model using αGalCer administration confirmed the capacity of CD1d-restricted T cell activation to modulate disease. αGalCer administration can protect mice from EAE (397, 398), although this has not been observed

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

862

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

consistently (399). Activation of CD1d-restricted T cells by αGalCer in mice expressing a myelin-reactive transgenic TCR prevented EAE in an IL-4-dependent manner when αGalCer was administered prior to the activation of myelin-reactive T cells. However, when simultaneous activation of CD1d-reactive T cells occurred together with that of myelin-reactive T cells, αGalCer increased the severity of EAE due to an excessive TH1 response (400). This suggests that the temporal relationship between the activation of CD1d-restricted T cells by αGalCer and disease-specific T cells is important in disease development. Manipulations that enhance the TH2 bias of CD1d-restricted T cells strikingly reduce the development of EAE. For example, coadministration of blocking antibodies against CD86 together with αGalCer biased the αGalCer response to TH2 and suppressed EAE (399). In addition, EAE was profoundly suppressed by the administration of a modified form of αGalCer, which lacks one alkyl chain, that selectively elicited IL-4 production by CD1d-restricted T cells (282). Thus, the ability of CD1d-restricted T cells that are activated by αGalCer to modulate the course of EAE in mice appears to depend on the balance of IL-4 versus IFN-γ secretion that is induced. In contrast, the natural protective effect of CD1d-restricted T cells during EAE seems to be mediated by inhibition of auto-antigen-specific IFN-γ secretion in an IL-4-independent manner.

Systemic Lupus Erythematosus CD1d-reactive T cells expressing a Vα4.4+/Vβ9+ TCR were found to modulate a lupus-like syndrome in mice that was characterized by increased serum immunoglobulin concentrations, appearance of anti-double-stranded DNA antibodies, proteinuria, and immune complex glomerulonephritis, and that was dependent on IFN-γ (102). Transgenic T cells isolated from the spleen and those that were CD4+ or CD8+ from the bone marrow–induced disease. Interestingly, however, DN Vα4.4+/Vβ9+ TCR transgenic T cells from bone marrow dominantly suppressed the development of disease, suggesting that the microenvironment and/or coreceptor expression may influence the effector functions of autoreactive CD1d-restricted T cells. A CD1d-restricted T cell population with suppressor capabilities using diverse TCRs has also been isolated from human bone marrow (107). In addition, a pathogenic role for CD1d-restricted T cells in the development of systemic lupus erythematosus (SLE) is supported by experiments showing that administration of αGalCer to NZB/NZW mice, a strain that develops lupus spontaneously, exacerbated lupus disease activity as measured by earlier onset of proteinuria and increased mortality (401), by studies showing that transfer of activated Vα14+ CD1d-restricted T cells in young NZB/NZW mice induced an autoimmune-like inflammation (402), and by studies showing that administration of anti-CD1d blocking antibodies ameliorated SLE and reduced the production of IgM and IgG2a/2banti-dsDNA antibodies in NZB/NZW mice (403, 401). Interestingly, CD1c-restricted autoreactive T cells from lupus patients stimulated antibody production by CD1c+ B cells and induced isotype switching in vitro, suggesting that CD1-restricted T cells may also play a pathogenic role in the development of lupus in humans (406).

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

863

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

In contrast, MRL/lpr mice, another model for SLE, showed decreased numbers and functions of CD1d/αGalCer tetramer+ T cells. Administration of αGalCer resulted in the expansion of CD1d/αGalCer tetramer+ T cells and alleviated the inflammatory dermatitis (303b). CD1d deficiency exacerbated lupus nephritis and increased anti-dsDNA antibodies induced by the hydrocarbon oil pristane (405). These studies suggest that CD1d-restricted T cells can also play a protective role in the development of lupus. However, CD1d-deficient MRL/lpr mice did not develop lupus skin disease or nephritis differently from wild-type MRL/lpr mice (404).

Colitis and Hepatitis Other examples showing the capability of CD1d-restricted T cells to either induce or suppress the development of inflammatory conditions come from studies of murine models of intestinal inflammation. In a model using oxazolone to induce colitis by intrarectal administration subsequent to skin sensitization, IL-13producing CD1d-restricted T cells were required for the development of colitis (407). In contrast, in a murine model of colitis induced by dextran sodium sulfate, activation of CD1d-restricted T cells or adoptive transfer of αGalCer-activated CD1d-restricted T cells resulted in reduced colitis, indicating that CD1d-restricted T cells can also downregulate intestinal inflammation (408). TCRα-deficient mice develop spontaneous intestinal inflammation, a process that appears to be suppressed by B cells. B cells in mesenteric lymph nodes of TCRα-deficient mice showed strikingly increased CD1d-expression and CD1d/TCRα-double-deficient mice developed more severe colonic inflammation at later time points, suggesting that CD1d expression on B cells may be important for their regulatory function in this model (409). However, whether CD1d-restricted T cells are required for the downregulation of intestinal inflammation induced by these regulatory B cells is not known. Autoimmune hepatitis induced by administration of concavalin A to mice depended on the presence of CD1d-restriced T cells (410). The cytotoxicity of CD1drestricted T cells against hepatocytes induced by concavalin A required perforin and FasL expression on the activated T cells and could be augmented by an IL-4dependent autocrine mechanism (411).

Allergic and Hypersensitivity Reactions The ability to promptly produce large amounts of IL-4 upon primary stimulation and the ability to influence IgE production induced by anti-IgD antibodies in mice suggested that CD161+ T cells may be important in allergic reactions (412). However, CD1d-restricted T cells were not required for IgE production in vivo after stimulation with anti-IgD mAbs (203, 413, 414), or in response to immunizations with ovalbumin or Nippostrongylus brasiliensis infection (415, 416), but appeared to be required, together with conventional CD4+ T cells, for IL-18-induced IgE antibody production (280). Indeed, CD1d-deficient and Jα18deficient mice displayed defects in the development of allergen-induced airway

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

864

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

hyperreactivity (AHR) (417, 418). However, the extent to which CD1d-restricted T cells influence airway eosinophilia and the development of allergen-specific IgE antibodies in this model remains controversial (417–419). Contact sensitivity is a T cell–dependent immune response to skin sensitization with reactive haptens that appears to require IgM antibodies produced by B-1 B cells to initiate T cell recruitment (419a). Contact sensitivity was defective in Jα18-deficient mice, and adoptive transfer experiments showed that it was dependent on IL-4 produced by Vα14+ CD1d-restricted T cells (419b), showing that CD1d-restricted T cells are required for B cell activation and IgM antibody production in this form of hypersensitivity reaction.

SUMMARY CD1 antigen presentation provides a pathway of T cell stimulation independent of MHC class I and II. Antigen presentation by CD1 molecules differs from that by MHC molecules by having distinct requirements for intracellular trafficking, processing, and loading of lipid antigens. CD1-reactive T cells are found among the CD4, CD8, and DN T cell subsets previously assumed to be MHC-restricted. Among CD1-reactive T cells, several clearly different subpopulations exist, based on the CD1 isoform that they recognize, the diversity of their TCRs, and their effector functions. CD1a-, b-, and c-restricted T cells that recognize foreign microbial lipids appear to behave much like MHC-restricted T cells that recognize microbial protein antigens. They use clonally specific TCRs and may contribute to memory responses. In contrast, self-lipid reactive CD1-restricted T cells affect immune responses by mechanisms that lie outside the paradigm of foreign antigen-specific MHC-restricted adaptive immunity. The self-reactivity of CD1-restricted T cells may be regulated by the strength of the TCR signal received and by the microenvironment of cytokines and costimulation evoked by microbial interactions with APCs. In a variety of circumstances, early activation of CD1-restricted T cells occurs in advance of the MHC-restricted adaptive immune response. The rapid response of some CD1restricted T cells (TCRα-invariant CD1d-restricted T cells, and γ δ T cells) may be activated at the same time as cells of the innate phase of the immune response. CD1-reactive T cells can stimulate DC differentiation and instruct DC maturation toward IL-12-producing inflammatory DCs, and they can potently activate NK cells, other T cells, and B cells to amplify the innate immune response and influence the subsequent adaptive immune response to infection. CD1-reactive T cells can mediate tumor rejection via IL-12 production, NK and T cell activation, or by direct cytolysis. Alternatively, they may help to suppress tumor rejection, prevent autoimmunity and generate tolerance. The mechanisms responsible for the tolerogenic actions of CD1-restricted T cells involve TH2 cytokine secretion and CD1-dependent effects on other regulatory cells. Thus, divergent patterns of CD1-reactive T cells resulting in inflammatory and tolerogenic responses can be

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

P1: IKH

865

understood on the basis of their potential to produce IFN-γ and to stimulate IL-12 production, or alternatively, to produce TH2 cytokines and to generate tolerogenic DCs. Studies using αGalCer to activate CD1d-restricted T cells pharmacologically show their therapeutic potential in microbial immunity, antitumor immunity, and autoimmunity. Thus, CD1-reactive T cells participate in the full range of T cell responses previously thought to be mediated only by peptide-specific MHCrestricted T cells.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ACKNOWLEDGMENTS We thank J. Gumperz for her essential contribution to the content of the manuscript and figures. In addition, we thank D. Zajonc, I. Wilson, M. Cernadas, C. Dascher, M. Sugita, and B. Moody for their help in preparing figures and for their comments, as well as S. Moskowitz for graphic art contributions. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. McMichael AJ, Pilch JR, Galfre G, Mason DY, Fabre JW, Milstein C. 1979. A human thymocyte antigen defined by a hybrid myeloma monoclonal antibody. Eur. J. Immunol. 9:205–10 2. Bernard A, Boumsell L. 1984. The clusters of differentiation (CD) defined by the First International Workshop on Human Leucocyte Differentiation Antigens. Hum. Immunol. 11:1–10 3. Martin LH, Calabi F, Lefebvre FA, Bilsland CA, Milstein C. 1987. Structure and expression of the human thymocyte antigens CD1a, CD1b, and CD1c. Proc. Natl. Acad. Sci. USA 84:9189–93 4. Longley J, Kraus J, Alonso M, Edelson R. 1989. Molecular cloning of CD1a (T6), a human epidermal dendritic cell marker related to class I MHC molecules. J. Invest. Dermatol. 92:628–31 5. Martin LH, Calabi F, Milstein C. 1986. Isolation of CD1 genes: a family of major histocompatibility complex-related differentiation antigens. Proc. Natl. Acad. Sci. USA 83:9154–58 6. Calabi F, Milstein C. 1986. A novel family of human major histocompatibility

7.

8.

9.

10.

11.

12.

complex-related genes not mapping to chromosome 6. Nature 323:540–43 Albertson DG, Fishpool R, Sherrington P, Nacheva E, Milstein C. 1988. Sensitive and high resolution in situ hybridization to human chromosomes using biotin labelled probes: assignment of the human thymocyte CD1 antigen genes to chromosome 1. EMBO J. 7:2801–5 Yu CY, Milstein C. 1989. A physical map linking the five CD1 human thymocyte differentiation antigen genes. EMBO J. 8:3727–32 Moseley WS, Watson ML, Kingsmore SF, Seldin MF. 1989. CD1 defines conserved linkage group border between human chromosomes 1 and mouse chromosomes 1 and 3. Immunogenetics 30:378– 82 Calabi F, Jarvis JM, Martin L, Milstein C. 1989. Two classes of CD1 genes. Eur. J. Immunol. 19:285–92 Bradbury A, Belt KT, Neri TM, Milstein C, Calabi F. 1988. Mouse CD1 is distinct from and co-exists with TL in the same thymus. EMBO J. 7:3081–86 Balk SP, Bleicher PA, Terhorst C. 1991.

13 Feb 2004

18:42

866

13.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Isolation and expression of cDNA encoding the murine homologues of CD1. J. Immunol. 146:768–74 Ichimiya S, Matsuura A, Takayama S, Kikuchi K. 1993. Molecular cloning of a cDNA encoding the rat homologue of CD1. Transplant Proc. 25:2773–74 Matsuura A, Hashimoto Y, Kinebuchi M, Kasai K, Ichimiya S, et al. 1997. Rat CD1 antigen: structure, expression and function. Transplant Proc. 29:1705–6 Dascher CC, Brenner MB. 2003. Evolutionary constraints on CD1 structure: insights from comparative genomic analysis. Trends Immunol. 24:412–18 Han M, Hannick LI, DiBrino M, Robinson MA. 1999. Polymorphism of human CD1 genes. Tissue Antigens 54:122–27 Dascher CD, Brenner MB. 2003. Genetics and Immunobiology of CD1. HLA 2002. In press Oteo M, Parra JF, Mirones I, Gimenez LI, Setien F, Martinez-Naves E. 1999. Single strand conformational polymorphism analysis of human CD1 genes in different ethnic groups. Tissue Antigens 53:545–50 Oteo M, Arribas P, Setien F, Parra JF, Mirones I, Gomez del Moral M, Martinez-Naves E. 2001. Structural characterization of two CD1A allelic variants. Hum. Immunol. 62:1137–41 Mirones I, Oteo M, Parra-Cuadrado JF, Martinez-Naves E. 2000. Identification of two novel human CD1E alleles. Tissue Antigens 56:159–61 Tamouza R, Sghiri R, Ramasawmy R, Neonato MG, Mombo LE, et al. 2002. Two novel CD1 E alleles identified in black African individuals. Tissue Antigens 59:417–20 Jones DC, Gelder CM, Ahmad T, Campbell IA, Barnardo MC, et al. 2001. CD1 genotyping of patients with Mycobacterium malmoense pulmonary disease. Tissue Antigens 58:19–23 Rhind SM, Dutia BM, Howard CJ, Hopkins J. 1996. Discrimination of two sub-

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

sets of CD1 molecules in the sheep. Vet. Immunol. Immunopathol. 52:265–70 Ferguson ED, Dutia BM, Hein WR, Hopkins J. 1996. The sheep CD1 gene family contains at least four CD1B homologues. Immunogenetics 44:86–96 Parsons KR, Howard CJ, Sopp P. 1991. Immunohistology of workshop monoclonal antibodies to the bovine homologue of CD1. Vet. Immunol. Immunopathol. 27:201–6 Calabi F, Belt KT, Yu CY, Bradbury A, Mandy WJ, Milstein C. 1989. The rabbit CD1 and the evolutionary conservation of the CD1 gene family. Immunogenetics 30:370–77 Hayes SM, Knight KL. 2001. Group 1 CD1 genes in rabbit. J. Immunol. 166: 403–10 Dascher CC, Hiromatsu K, Naylor JW, Brauer PP, Brown KA, et al. 1999. Conservation of a CD1 multigene family in the guinea pig. J. Immunol. 163:5478–88 Angenieux C, Salamero J, Fricker D, Wurtz JM, Maitre B, et al. 2003. Common characteristics of the human and rhesus macaque CD1e molecules: conservation of biochemical and biological properties during primate evolution. Immunogenetics 54:842–49 Sugita M, Cao X, Watts GF, Rogers RA, Bonifacino JS, Brenner MB. 2002. Failure of trafficking and antigen presentation by CD1 in AP-3-deficient cells. Immunity 16:697–706 Dascher CC, Hiromatsu K, Xiong X, Sugita M, Buhlmann JE, et al. 2002. Conservation of CD1 intracellular trafficking patterns between mammalian species. J. Immunol. 169:6951–58 Cernadas M, Sugita M, Van Der Wel N, Cao X, Gumperz JE, et al. 2003. Lysosomal localization of murine CD1d mediated by AP-3 is necessary for NK T cell development. J. Immunol. 171:4149–55 Elewaut D, Lawton AP, Nagarajan NA, Maverakis E, Khurana A, et al. 2003. The adapter protein AP-3 is required

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

34.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

35.

36.

37.

38.

39.

40.

41.

42.

for CD1d-mediated antigen presentation of glycosphingolipids and development of Valpha14i NKT cells. J. Exp. Med. 198:1133–46 Knowles RW, Bodmer WF. 1982. A monoclonal antibody recognizing a human thymus leukemia-like antigen associated with beta 2-microglobulin. Eur. J. Immunol. 12:676–81 van de Rijn M, Lerch PG, Knowles RW, Terhorst C. 1983. The thymic differentiation markers T6 and M241 are two unusual MHC class I antigens. J. Immunol. 131:851–55 Olive D, Dubreuil P, Mawas C. 1984. Two distinct TL-like molecular subsets defined by monoclonal antibodies on the surface of human thymocytes with different expression on leukemia lines. Immunogenetics 20:253–64 Lerch PG, van de Rijn M, Smart JE, Knowles RW, Terhorst C. 1986. Isolation and purification of the human thymocyte antigens T6 and M241. Mol. Immunol. 23:131–39 Sugita M, Porcelli SA, Brenner MB. 1997. Assembly and retention of CD1b heavy chains in the endoplasmic reticulum. J. Immunol. 159:2358–65 Bleicher PA, Balk SP, Hagen SJ, Blumberg RS, Flotte TJ, Terhorst C. 1990. Expression of murine CD1 on gastrointestinal epithelium. Science 250:679–82 Bilsland CA, Milstein C. 1991. The identification of the beta 2-microglobulin binding antigen encoded by the human CD1D gene. Eur. J. Immunol. 21:71–78 Brutkiewicz RR, Bennink JR, Yewdell JW, Bendelac A. 1995. TAP-independent, beta 2-microglobulin-dependent surface expression of functional mouse CD1.1. J. Exp. Med. 182:1913–19 Bauer A, Huttinger R, Staffler G, Hansmann C, Schmidt W, et al. 1997. Analysis of the requirement for beta 2-microglobulin for expression and formation of human CD1 antigens. Eur. J. Immunol. 27:1366–73

P1: IKH

867

43. Huttinger R, Staffler G, Majdic O, Stockinger H. 1999. Analysis of the early biogenesis of CD1b: involvement of the chaperones calnexin and calreticulin, the proteasome and beta 2-microglobulin. Int. Immunol. 11:1615–23 44. Balk SP, Burke S, Polischuk JE, Frantz ME, Yang L, et al. 1994. Beta 2microglobulin-independent MHC class Ib molecule expressed by human intestinal epithelium. Science 265:259–62 45. Kim HS, Garcia J, Exley M, Johnson KW, Balk SP, Blumberg RS. 1999. Biochemical characterization of CD1d expression in the absence of beta2microglobulin. J. Biol. Chem. 274:9289– 95 46. Somnay-Wadgaonkar K, Nusrat A, Kim HS, Canchis WP, Balk SP, et al. 1999. Immunolocalization of CD1d in human intestinal epithelial cells and identification of a beta2-microglobulin-associated form. Int. Immunol. 11:383–92 47. Woolfson A, Milstein C. 1994. Alternative splicing generates secretory isoforms of human CD1. Proc. Natl. Acad. Sci. USA 91:6683–87 48. Angenieux C, Salamero J, Fricker D, Cazenave JP, Goud B, et al. 2000. Characterization of CD1e, a third type of CD1 molecule expressed in dendritic cells. J. Biol. Chem. 275:37757–64 49. Zeng Z, Castano AR, Segelke BW, Stura EA, Peterson PA, Wilson IA. 1997. Crystal structure of mouse CD1: an MHC-like fold with a large hydrophobic binding groove. Science 277:339–45 50. Gadola SD, Zaccai NR, Harlos K, Shepherd D, Castro-Palomino JC, et al. 2002. Structure of human CD1b with bound ˚ a maze for alkyl chains. ligands at 2.3 A, Nat. Immunol. 3:721–26 51. Zajonc DM, Elsliger MA, Teyton L, Wilson IA. 2003. Crystal structure of CD1a in complex with a sulfatide self antigen ˚ Nat. Immunol. at a resolution of 2.15 A. 4:808–15 52. Beckman EM, Porcelli SA, Morita CT,

13 Feb 2004

18:42

868

53.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

54.

55.

56.

57.

58.

59.

60.

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Behar SM, Furlong ST, Brenner MB. 1994. Recognition of a lipid antigen by CD1-restricted alpha beta+ T cells. Nature 372:691–94 Moody DB, Reinhold BB, Guy MR, Beckman EM, Frederique DE, et al. 1997. Structural requirements for glycolipid antigen recognition by CD1brestricted T cells. Science 278:283–86 Sieling PA, Chatterjee D, Porcelli SA, Prigozy TI, Mazzaccaro RJ, et al. 1995. CD1-restricted T cell recognition of microbial lipoglycan antigens. Science 269:227–30 Moody DB, Ulrichs T, Muhlecker W, Young DC, Gurcha SS, et al. 2000. CD1c-mediated T-cell recognition of isoprenoid glycolipids in Mycobacterium tuberculosis infection. Nature 404:884–88 Moody DB, Young DC, Cheng TY, Rosat JP, Roura-Mir C, et al. 2004. T cell activation by lipopeptide antigens. Science 303:527–31 Schofield L, McConville MJ, Hansen D, Campbell AS, Fraser-Reid B, et al. 1999. CD1d-restricted immunoglobulin G formation to GPI-anchored antigens mediated by NKT cells. Science 283:225–29 Molano A, Park SH, Chiu YH, Nosseir S, Bendelac A, Tsuji M. 2000. Cutting edge: the IgG response to the circumsporozoite protein is MHC class IIdependent and CD1d-independent: exploring the role of GPIs in NK T cell activation and antimalarial responses. J. Immunol. 164:5005–9 Duthie MS, Wleklinski-Lee M, Smith S, Nakayama T, Taniguchi M, Kahn SJ. 2002. During Trypanosoma cruzi infection CD1d-restricted NK T cells limit parasitemia and augment the antibody response to a glycophosphoinositolmodified surface protein. Infect. Immun. 70:36–48 Romero JF, Eberl G, MacDonald HR, Corradin G. 2001. CD1d-restricted NK T cells are dispensable for specific anti-

61.

62.

63.

64.

65.

66.

67.

68.

69.

body responses and protective immunity against liver stage malaria infection in mice. Parasite Immunol. 23:267–69 Procopio DO, Almeida IC, Torrecilhas AC, Cardoso JE, Teyton L, et al. 2002. Glycosylphosphatidylinositol-anchored mucin-like glycoproteins from Trypanosoma cruzi bind to CD1d but do not elicit dominant innate or adaptive immune responses via the CD1d/NKT cell pathway. J. Immunol. 169:3926–33 Porcelli S, Brenner MB, Greenstein JL, Balk SP, Terhorst C, Bleicher PA. 1989. Recognition of cluster of differentiation 1 antigens by human CD4−CD8– cytolytic T lymphocytes. Nature 341: 447–50 Bendelac A, Lantz O, Quimby ME, Yewdell JW, Bennink JR, Brutkiewicz RR. 1995. CD1 recognition by mouse NK1+ T lymphocytes. Science 268:863– 65 Exley M, Garcia J, Balk SP, Porcelli S. 1997. Requirements for CD1d recognition by human invariant Valpha24+ CD4−CD8− T cells. J. Exp. Med. 186: 109–20 Gumperz JE, Roy C, Makowska A, Lum D, Sugita M, et al. 2000. Murine CD1drestricted T cell recognition of cellular lipids. Immunity 12:211–21 Joyce S, Woods AS, Yewdell JW, Bennink JR, De Silva AD, et al. 1998. Natural ligand of mouse CD1d1: cellular glycosylphosphatidylinositol. Science 279:1541–44 De Silva AD, Park JJ, Matsuki N, Stanic AK, Brutkiewicz RR, et al. 2002. Lipid protein interactions: The assembly of CD1d1 with cellular phospholipids occurs in the endoplasmic reticulum. J. Immunol. 168:723–33 Shamshiev A, Donda A, Carena I, Mori L, Kappos L, De Libero G. 1999. Self glycolipids as T-cell autoantigens. Eur. J. Immunol. 29:1667–75 Pukel CS, Lloyd KO, Travassos LR, Dippold WG, Oettgen HF, et al. 1982.

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

70.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

71.

72.

73.

74.

75.

76.

77.

78.

GD3, a prominent ganglioside of human melanoma. Detection and characterisation by mouse monoclonal antibody. J. Exp. Med. 155:1133–47 Wu DY, Segal NH, Sidobre S, Kronenberg M, Chapman PB. 2003. Crosspresentation of disialoganglioside GD3 to natural killer T cells. J. Exp. Med. 198:173–81 Shamshiev A, Gober HJ, Donda A, Mazorra Z, Mori L, De Libero G. 2002. Presentation of the same glycolipid by different CD1 molecules. J. Exp. Med. 195:1013–21 Kawano T, Cui J, Koezuka Y, Toura I, Kaneko Y, et al. 1997. CD1d-restricted and TCR-mediated activation of Valpha14 NKT cells by glycosylceramides. Science 278:1626–29 Ernst WA, Maher J, Cho S, Niazi KR, Chatterjee D, et al. 1998. Molecular interaction of CD1b with lipoglycan antigens. Immunity 8:331–40 Sugita M, Jackman RM, van Donselaar E, Behar SM, Rogers RA, et al. 1996. Cytoplasmic tail-dependent localization of CD1b antigen-presenting molecules to MIICs. Science 273:349–52 Prigozy TI, Sieling PA, Clemens D, Stewart PL, Behar SM, et al. 1997. The mannose receptor delivers lipoglycan antigens to endosomes for presentation to T cells by CD1b molecules. Immunity 6:187–97 Naidenko OV, Maher JK, Ernst WA, Sakai T, Modlin RL, Kronenberg M. 1999. Binding and antigen presentation of ceramide-containing glycolipids by soluble mouse and human CD1d molecules. J. Exp. Med. 190:1069–80 Cantu C 3rd, Benlagha K, Savage PB, Bendelac A, Teyton L. 2003. The paradox of immune molecular recognition of alpha-galactosylceramide: low affinity, low specificity for CD1d, high affinity for alphabeta TCRs. J. Immunol. 170:4673–82 Grant EP, Degano M, Rosat JP, Stenger S,

79.

80.

81.

82.

83.

84.

85.

86.

P1: IKH

869

Modlin RL, et al. 1999. Molecular recognition of lipid antigens by T cell receptors. J. Exp. Med. 189:195–205 Moody DB, Guy MR, Grant E, Cheng TY, Brenner MB, et al. 2000. CD1bmediated T cell recognition of a glycolipid antigen generated from mycobacterial lipid and host carbohydrate during infection. J. Exp. Med. 192:965–76 Spada FM, Grant EP, Peters PJ, Sugita M, Melian A, et al. 2000. Selfrecognition of CD1 by gamma/delta T cells: implications for innate immunity. J. Exp. Med. 191:937–48 Grant EP, Beckman EM, Behar SM, Degano M, Frederique D, et al. 2002. Fine specificity of TCR complementarity-determining region residues and lipid antigen hydrophilic moieties in the recognition of a CD1-lipid complex. J. Immunol. 168:3933–40 Koseki H, Imai K, Nakayama F, Sado T, Moriwaki K, Taniguchi M. 1990. Homogenous junctional sequence of the V14+ T-cell antigen receptor alpha chain expanded in unprimed mice. Proc. Natl. Acad. Sci. USA 87:5248–52 Lantz O, Bendelac A. 1994. An invariant T cell receptor alpha chain is used by a unique subset of major histocompatibility complex class I-specific CD4+ and CD4−8− T cells in mice and humans. J. Exp. Med. 180:1097–106 Porcelli S, Yockey CE, Brenner MB, Balk SP. 1993. Analysis of T cell antigen receptor (TCR) expression by human peripheral blood CD4−8− alpha/beta T cells demonstrates preferential use of several V beta genes and an invariant TCR alpha chain. J. Exp. Med. 178:1– 16 Dellabona P, Padovan E, Casorati G, Brockhaus M, Lanzavecchia A. 1994. An invariant V alpha 24-J alpha Q/V beta 11 T cell receptor is expressed in all individuals by clonally expanded CD4−8− T cells. J. Exp. Med. 180:1171–76 Kent SC, Hafler DA, Strominger JL,

13 Feb 2004

18:42

870

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

87.

88.

89.

90.

91.

92.

93.

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Wilson SB. 1999. Noncanonical Valpha 24JalphaQ T cells with conservative alpha chain CDR3 region amino acid substitutions are restricted by CD1d. Hum. Immunol. 60:1080–89 Porcelli S, Gerdes D, Fertig AM, Balk SP. 1996. Human T cells expressing an invariant V alpha 24-J alpha Q TCR alpha are CD4− and heterogeneous with respect to TCR beta expression. Hum. Immunol. 48:63–67 Ronet C, Mempel M, Thieblemont N, Lehuen A, Kourilsky P, Gachelin G. 2001. Role of the complementaritydetermining region 3 (CDR3) of the TCR-beta chains associated with the V alpha 14 semi-invariant TCR alphachain in the selection of CD4+ NK T cells. J. Immunol. 166:1755–62 Matsuda JL, Gapin L, Fazilleau N, Warren K, Naidenko OV, Kronenberg M. 2001. Natural killer T cells reactive to a single glycolipid exhibit a highly diverse T cell receptor beta repertoire and small clone size. Proc. Natl. Acad. Sci. USA 98:12636–41 Burdin N, Brossay L, Koezuka Y, Smiley ST, Grusby MJ, et al. 1998. Selective ability of mouse CD1 to present glycolipids: alpha-galactosylceramide specifically stimulates V alpha 14+ NK T lymphocytes. J. Immunol. 161:3271–81 Spada FM, Koezuka Y, Porcelli SA. 1998. CD1d-restricted recognition of synthetic glycolipid antigens by human natural killer T cells. J. Exp. Med. 188:1529–34 Matsuda JL, Naidenko OV, Gapin L, Nakayama T, Taniguchi M, et al. 2000. Tracking the response of natural killer T cells to a glycolipid antigen using CD1d tetramers. J. Exp. Med. 192:741– 54 Benlagha K, Weiss A, Beavis A, Teyton L, Bendelac A. 2000. In vivo identification of glycolipid antigen-specific T cells using fluorescent CD1d tetramers. J. Exp. Med. 191:1895–903

94. Gumperz JE, Miyake S, Yamamura T, Brenner MB. 2002. Functionally distinct subsets of CD1d-restricted natural killer T cells revealed by CD1d tetramer staining. J. Exp. Med. 195:625–36 95. Lee PT, Benlagha K, Teyton L, Bendelac A. 2002. Distinct functional lineages of human V(alpha)24 natural killer T cells. J. Exp. Med. 195:637–41 96. Brossay L, Tangri S, Bix M, Cardell S, Locksley R, Kronenberg M. 1998. Mouse CD1-autoreactive T cells have diverse patterns of reactivity to CD1+ targets. J. Immunol. 160:3681–88 97. Chiu YH, Jayawardena J, Weiss A, Lee D, Park SH, et al. 1999. Distinct subsets of CD1d-restricted T cells recognize self-antigens loaded in different cellular compartments. J. Exp. Med. 189:103– 10 98. Gui M, Li J, Wen LJ, Hardy RR, Hayakawa K. 2001. TCR beta chain influences but does not solely control autoreactivity of Valpha 14J281T cells. J. Immunol. 167:6239–46 99. Schumann J, Voyle RB, Wei BY, MacDonald HR. 2003. Cutting edge: influence of the TCR Vbeta domain on the avidity of CD1d:alphagalactosylceramide binding by invariant Valpha14 NKT Cells. J. Immunol. 170:5815–19 100. Cardell S, Tangri S, Chan S, Kronenberg M, Benoist C, Mathis D. 1995. CD1restricted CD4+ T cells in major histocompatibility complex class II-deficient mice. J. Exp. Med. 182:993–1004 101. Park SH, Roark JH, Bendelac A. 1998. Tissue-specific recognition of mouse CD1 molecules. J. Immunol. 160:3128– 34 102. Zeng D, Dick M, Cheng L, Amano M, Dejbakhsh-Jones S, et al. 1998. Subsets of transgenic T cells that recognize CD1 induce or prevent murine lupus: role of cytokines. J. Exp. Med. 187:525–36 103. Behar SM, Podrebarac TA, Roy CJ, Wang CR, Brenner MB. 1999. Diverse

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

104.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

105.

106.

107.

108.

109.

110.

111.

112.

TCRs recognize murine CD1. J. Immunol. 162:161–67 Gadola SD, Dulphy N, Salio M, Cerundolo V. 2002. Valpha24-JalphaQindependent, CD1d-restricted recognition of alpha-galactosylceramide by human CD4+ and CD8alphabeta+ T lymphocytes. J. Immunol. 168:5514–20 Park SH, Weiss A, Benlagha K, Kyin T, Teyton L, Bendelac A. 2001. The mouse CD1d-restricted repertoire is dominated by a few autoreactive T cell receptor families. J. Exp. Med. 193:893–904 Behar SM, Cardell S. 2000. Diverse CD1d-restricted T cells: diverse phenotypes, and diverse functions. Semin. Immunol. 12:551–60 Exley MA, Tahir SM, Cheng O, Shaulov A, Joyce R, et al. 2001. A major fraction of human bone marrow lymphocytes are Th2-like CD1d-reactive T cells that can suppress mixed lymphocyte responses. J. Immunol. 167:5531–34 Exley MA, He Q, Cheng O, Wang RJ, Cheney CP, et al. 2002. Cutting edge: compartmentalization of Th1-like noninvariant CD1d-reactive T cells in hepatitis C virus-infected liver. J. Immunol. 168:1519–23 Huber S, Sartini D, Exley M. 2003. Role of CD1d in coxsackievirus b3-induced myocarditis. J. Immunol. 170:3147–53 Makowska A, Kawano T, Taniguchi M, Cardell S. 2000. Differences in the ligand specificity between CD1d-restricted T cells with limited and diverse T-cell receptor repertoire. Scand. J. Immunol. 52:71–79 Melian A, Watts GF, Shamshiev A, De Libero G, Clatworthy A, et al. 2000. Molecular recognition of human CD1b antigen complexes: evidence for a common pattern of interaction with alpha beta TCRs. J. Immunol. 165:4494– 504 Burdin N, Brossay L, Degano M, Iijima H, Gui M, Wilson IA, Kronenberg M. 2000. Structural requirements for anti-

113.

114.

115.

116.

117.

118.

119.

120.

121.

P1: IKH

871

gen presentation by mouse CD1. Proc. Natl. Acad. Sci. USA 97:10156–61 Sidobre S, Naidenko OV, Sim BC, Gascoigne NR, Garcia KC, Kronenberg M. 2002. The V alpha 14 NKT cell TCR exhibits high-affinity binding to a glycolipid/CD1d complex. J. Immunol. 169:1340–48 Prigozy TI, Naidenko O, Qasba P, Elewaut D, Brossay L, et al. 2001. Glycolipid antigen processing for presentation by CD1d molecules. Science 291:664– 67 Brossay L, Naidenko O, Burdin N, Matsuda J, Sakai T, Kronenberg M. 1998. Structural requirements for galactosylceramide recognition by CD1-restricted NK T cells. J. Immunol. 161:5124–28 Rauch J, Gumperz JE, Robinson C, Skold M, Roy C, et al. 2003. Structural features of the acyl chain determine selfphospholipid antigen recognition by a CD1d-restricted invariant NKT (iNKT) cell. J. Biol. Chem. 278:47508–15 Moody DB, Briken V, Cheng TY, RouraMir C, Guy MR, et al. 2002. Lipid length controls antigen entry into endosomal and nonendosomal pathways for CD1b presentation. Nat. Immunol. 3:435– 42 Pamer E, Cresswell P. 1998. Mechanisms of MHC class I–restricted antigen processing. Annu. Rev. Immunol. 16:323–58 Cresswell P. 1994. Assembly, transport, and function of MHC class II molecules. Annu. Rev. Immunol. 12:259–93 Kang SJ, Cresswell P. 2002. Calnexin, calreticulin, and ERp57 cooperate in disulfide bond formation in human CD1d heavy chain. J. Biol. Chem. 277:44838– 44 Jayawardena-Wolf J, Benlagha K, Chiu YH, Mehr R, Bendelac A. 2001. CD1d endosomal trafficking is independently regulated by an intrinsic CD1d-encoded tyrosine motif and by the invariant chain. Immunity 15:897–908

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

872

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

122. Briken V, Jackman RM, Dasgupta S, Hoening S, Porcelli SA. 2002. Intracellular trafficking pathway of newly synthesized CD1b molecules. EMBO J. 21:825–34 123. van der Wel NN, Sugita M, Fluitsma DM, Cao X, Schreibelt G, et al. 2003. CD1 and major histocompatibility complex II molecules follow a different course during dendritic cell maturation. Mol. Biol. Cell 14:3378–88 124. Jackman RM, Stenger S, Lee A, Moody DB, Rogers RA, et al. 1998. The tyrosine-containing cytoplasmic tail of CD1b is essential for its efficient presentation of bacterial lipid antigens. Immunity 8:341–51 125. Sugita M, Grant EP, van Donselaar E, Hsu VW, Rogers RA, et al. 1999. Separate pathways for antigen presentation by CD1 molecules. Immunity 11:743–52 126. Salamero J, Bausinger H, Mommaas AM, Lipsker D, Proamer F, et al. 2001. CD1a molecules traffic through the early recycling endosomal pathway in human Langerhans cells. J. Invest. Dermatol. 116:401–8 127. Sugita M, van Der Wel N, Rogers RA, Peters PJ, Brenner MB. 2000. CD1c molecules broadly survey the endocytic system. Proc. Natl. Acad. Sci. USA 97:8445–50 128. Briken V, Jackman RM, Watts GF, Rogers RA, Porcelli SA. 2000. Human CD1b and CD1c isoforms survey different intracellular compartments for the presentation of microbial lipid antigens. J. Exp. Med. 192:281–88 129. Chiu YH, Park SH, Benlagha K, Forestier C, Jayawardena-Wolf J, et al. 2002. Multiple defects in antigen presentation and T cell development by mice expressing cytoplasmic tail-truncated CD1d. Nat. Immunol. 3:55–60 130. Kang SJ, Cresswell P. 2002. Regulation of intracellular trafficking of human CD1d by association with MHC class II molecules. EMBO J. 21:1650–60

131. Riese RJ, Shi GP, Villadangos J, Stetson D, Driessen C, et al. 2001. Regulation of CD1 function and NK1.1+ T cell selection and maturation by cathepsin S. Immunity 15:909–19 132. Honey K, Benlagha K, Beers C, Forbush K, Teyton L, et al. 2002. Thymocyte expression of cathepsin L is essential for NKT cell development. Nat. Immunol. 3:1069–74 133. Pena-Cruz V, Ito S, Dascher CC, Brenner MB, Sugita M. 2003. Epidermal Langerhans cells efficiently mediate CD1a-dependent presentation of microbial lipid antigens to T cells. J. Invest. Dermatol. 121:517–21 134. Cattoretti G, Berti E, Mancuso A, D’Amato L, Schiro R, et al. 1987. An MHC class I related family of antigens with widespread distribution on resting and activated cells. In Leucocyte Typing III, ed. AJ McMichael, pp. 89–91. Oxford: Oxford Univ. Press 135. Smith ME, Thomas JA, Bodmer WF. 1988. CD1c antigens are present in normal and neoplastic B-cells. J. Pathol. 156:169–77 136. Small TN, Knowles RW, Keever C, Kernan NA, Collins N, et al. 1987. M241 (CD1) expression on B lymphocytes. J. Immunol. 138:2864–68 137. Plebani A, Proserpio AR, Guarneri D, Buscaglia M, Cattoretti G. 1993. B and T lymphocyte subsets in fetal and cord blood: age-related modulation of CD1c expression. Biol. Neonate. 63:1–7 138. Caux C, Dezutter-Dambuyant C, Schmitt D, Banchereau J. 1992. GM-CSF and TNF-alpha cooperate in the generation of dendritic Langerhans cells. Nature 360:258–61 139. Caux C, Vanbervliet B, Massacrier C, Dezutter-Dambuyant C, de Saint-Vis B, et al. 1996. CD34+ hematopoietic progenitors from human cord blood differentiate along two independent dendritic cell pathways in response to GM-CSF + TNF alpha. J. Exp. Med. 184:695–706

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES 140. Banchereau J, Briere F, Caux C, Davoust J, Lebecque S, et al. 2000. Immunobiology of dendritic cells. Annu. Rev. Immunol. 18:767–811 141. Porcelli S, Morita CT, Brenner MB. 1992. CD1b restricts the response of human CD4−8− T lymphocytes to a microbial antigen. Nature 360:593–97 142. Kasinrerk W, Baumruker T, Majdic O, Knapp W, Stockinger H. 1993. CD1 molecule expression on human monocytes induced by granulocytemacrophage colony-stimulating factor. J. Immunol. 150:579–84 143. Duperrier K, Eljaafari A, DezutterDambuyant C, Bardin C, Jacquet C, et al. 2000. Distinct subsets of dendritic cells resembling dermal DCs can be generated in vitro from monocytes, in the presence of different serum supplements. J. Immunol. Methods 238:119–31 144. Pietschmann P, Stockl J, Draxler S, Majdic O, Knapp W. 2000. Functional and phenotypic characteristics of dendritic cells generated in human plasma supplemented medium. Scand. J. Immunol. 51:377–83 145. Xia CQ, Kao KJ. 2002. Heparin induces differentiation of CD1a+ dendritic cells from monocytes: phenotypic and functional characterization. J. Immunol. 168:1131–38 146. Chang CC, Wright A, Punnonen J. 2000. Monocyte-derived CD1a+ and CD1a− dendritic cell subsets differ in their cytokine production profiles, susceptibilities to transfection, and capacities to direct Th cell differentiation. J. Immunol. 165:3584–91 147. Cao X, Sugita M, Van Der Wel N, Lai J, Rogers RA, et al. 2002. CD1 molecules efficiently present antigen in immature dendritic cells and traffic independently of MHC class II during dendritic cell maturation. J. Immunol. 169:4770– 77 148. Kleijmeer M, Ramm G, Schuurhuis D, Griffith J, Rescigno M, et al. 2001. Reor-

149.

150.

151.

152.

153.

154.

155.

156.

P1: IKH

873

ganization of multivesicular bodies regulates MHC class II antigen presentation by dendritic cells. J. Cell Biol. 155:53– 63 Chow A, Toomre D, Garrett W, Mellman I. 2002. Dendritic cell maturation triggers retrograde MHC class II transport from lysosomes to the plasma membrane. Nature 418:988–94 Barois N, de Saint-Vis B, Lebecque S, Geuze HJ, Kleijmeer MJ. 2002. MHC class II compartments in human dendritic cells undergo profound structural changes upon activation. Traffic 3:894– 905 Rosat JP, Grant EP, Beckman EM, Dascher CC, Sieling PA, et al. 1999. CD1-restricted microbial lipid antigenspecific recognition found in the CD8+ alpha beta T cell pool. J. Immunol. 162:366–71 Sieling PA, Ochoa MT, Jullien D, Leslie DS, Sabet S, et al. 2000. Evidence for human CD4+ T cells in the CD1-restricted repertoire: derivation of mycobacteriareactive T cells from leprosy lesions. J. Immunol. 164:4790–96 Sieling PA, Jullien D, Dahlem M, Tedder TF, Rea TH, et al. 1999. CD1 expression by dendritic cells in human leprosy lesions: correlation with effective host immunity. J. Immunol. 162:1851–58 Narayanan RB, Girdhar A, Girdhar BK. 1990. CD1-positive epidermal Langerhans cells in regressed tuberculoid and lepromatous leprosy lesions. Int. Arch. Allergy Appl. Immunol. 92:94–96 Beckman EM, Melian A, Behar SM, Sieling PA, Chatterjee D, et al. 1996. CD1c restricts responses of mycobacteria-specific T cells. Evidence for antigen presentation by a second member of the human CD1 family. J. Immunol. 157:2795–803 Gong J, Stenger S, Zack JA, Jones BE, Bristol GC, et al. 1998. Isolation of mycobacterium-reactive CD1-restricted T cells from patients with human

13 Feb 2004

18:42

874

157.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

158.

159.

160.

161.

162.

163.

164.

165.

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

immunodeficiency virus infection. J. Clin. Invest. 101:383–89 Fairhurst RM, Wang CX, Sieling PA, Modlin RL, Braun J. 1998. CD1 presents antigens from a gram-negative bacterium, Haemophilus influenzae type B. Infect. Immun. 66:3523–26 Stenger S, Hanson DA, Teitelbaum R, Dewan P, Niazi KR, et al. 1998. An antimicrobial activity of cytolytic T cells mediated by granulysin. Science 282:121–25 Stenger S, Mazzaccaro RJ, Uyemura K, Cho S, Barnes PF, et al. 1997. Differential effects of cytolytic T cell subsets on intracellular infection. Science 276:1684–87 Ochoa MT, Stenger S, Sieling PA, Thoma-Uszynski S, Sabet S, et al. 2001. T-cell release of granulysin contributes to host defense in leprosy. Nat. Med. 7:174– 79 Schaible UE, Hagens K, Fischer K, Collins HL, Kaufmann SH. 2000. Intersection of group I CD1 molecules and mycobacteria in different intracellular compartments of dendritic cells. J. Immunol. 164:4843–52 Stenger S, Niazi KR, Modlin RL. 1998. Down-regulation of CD1 on antigenpresenting cells by infection with Mycobacterium tuberculosis. J. Immunol. 161:3582–88 Mariotti S, Teloni R, Iona E, Fattorini L, Giannoni F, et al. 2002. Mycobacterium tuberculosis subverts the differentiation of human monocytes into dendritic cells. Eur. J. Immunol. 32:3050–58 Giuliani A, Prete SP, Graziani G, Aquino A, Balduzzi A, et al. 2001. Influence of Mycobacterium bovis bacillus Calmette Guerin on in vitro induction of CD1 molecules in human adherent mononuclear cells. Infect. Immun. 69:7461–70 De Lerma Barbaro A, Tosi G, Valle MT, Megiovanni AM, Sartoris S, et al. 1999. Distinct regulation of HLA class II and class I cell surface expression in

166.

167.

168.

169.

170.

171.

172.

173.

174.

the THP-1 macrophage cell line after bacterial phagocytosis. Eur. J. Immunol. 29:499–511 Schaible UE, Winau F, Sieling PA, Fischer K, Collins HL, et al. 2003. Apoptosis facilitates antigen presentation to T lymphocytes through MHC-I and CD1 in tuberculosis. Nat. Med. 9:1039–46 Ulrichs T, Moody DB, Grant E, Kaufmann SH, Porcelli SA. 2003. T-cell responses to CD1-presented lipid antigens in humans with Mycobacterium tuberculosis infection. Infect. Immun. 71:3076– 87 Kawashima T, Norose Y, Watanabe Y, Enomoto Y, Narazaki H, et al. 2003. Cutting edge: major CD8 T cell response to live bacillus Calmette-Guerin is mediated by CD1 molecules. J. Immunol. 170:5345–48 Hiromatsu K, Dascher CC, LeClair KP, Sugita M, Furlong ST, et al. 2002. Induction of CD1-restricted immune responses in guinea pigs by immunization with mycobacterial lipid antigens. J. Immunol. 169:330–39 Dascher CC, Hiromatsu K, Xiong X, Morehouse C, Watts G, et al. 2003. Immunization with a mycobacterial lipid vaccine improves pulmonary pathology in the guinea pig model of tuberculosis. Int. Immunol. 15:915–25 Gumperz JE, Brenner MB. 2001. CD1specific T cells in microbial immunity. Curr. Opin. Immunol. 13:471–78 Vincent MS, Leslie DS, Gumperz JE, Xiong X, Grant EP, Brenner MB. 2002. CD1-dependent dendritic cell instruction. Nat. Immunol. 3:1163–68 Dascher CC, Brenner MB. 2003. CD1 antigen presentation and infectious disease. Contrib. Microbiol. 10:164–82 Faure F, Jitsukawa S, Miossec C, Hercend T. 1990. CD1c as a target recognition structure for human T lymphocytes: analysis with peripheral blood gamma/delta cells. Eur. J. Immunol. 20:703–6

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES 175. Leslie DS, Vincent MS, Spada FM, Das H, Sugita M, et al. 2002. CD1-mediated gamma/delta T cell maturation of dendritic cells. J. Exp. Med. 196:1575–84 176. Vincent MS, Gumperz JE, Brenner MB. 2003. Understanding the function of CD1-restricted T cells. Nat. Immunol. 4:517–23 177. Khalili-Shirazi A, Gregson NA, Londei M, Summers L, Hughes RA. 1998. The distribution of CD1 molecules in inflammatory neuropathy. J. Neurol. Sci. 158:154–63 178. Van Rhijn I, Van den Berg LH, Bosboom WM, Otten HG, Logtenberg T. 2000. Expression of accessory molecules for T-cell activation in peripheral nerve of patients with CIDP and vasculitic neuropathy. Brain 123:2020–29 179. Battistini L, Fischer FR, Raine CS, Brosnan CF. 1996. CD1b is expressed in multiple sclerosis lesions. J. Neuroimmunol. 67:145–51 180. Cipriani B, Chen L, Hiromatsu K, Knowles H, Raine CS, et al. 2003. Upregulation of group 1 CD1 antigen presenting molecules in guinea pigs with experimental autoimmune encephalomyelitis: an immunohistochemical study. Brain Pathol. 13:1–9 181. Page G, Lebecque S, Miossec P. 2002. Anatomic localization of immature and mature dendritic cells in an ectopic lymphoid organ: correlation with selective chemokine expression in rheumatoid synovium. J. Immunol. 168:5333– 41 182. Fivenson DP, Nickoloff BJ. 1995. Distinctive dendritic cell subsets expressing factor XIIIa, CD1a, CD1b and CD1c in mycosis fungoides and psoriasis. J. Cutan. Pathol. 22:223–28 183. van Blokland SC, Wierenga-Wolf AF, van Helden-Meeuwsen CG, Drexhage HA, Hooijkaas H, et al. 2000. Professional antigen presenting cells in minor salivary glands in Sjogren’s syndrome: potential contribution to the histopatho-

184.

185.

186.

187.

188.

189.

190.

191.

192.

P1: IKH

875

logical diagnosis? Lab. Invest. 80:1935– 41 Melian A, Geng YJ, Sukhova GK, Libby P, Porcelli SA. 1999. CD1 expression in human atherosclerosis. A potential mechanism for T cell activation by foam cells. Am. J. Pathol. 155:775–86 Coventry BJ. 1999. CD1a positive putative tumour infiltrating dendritic cells in human breast cancer. Anticancer Res. 19:3183–87 Bell D, Chomarat P, Broyles D, Netto G, Harb GM, et al. 1999. In breast carcinoma tissue, immature dendritic cells reside within the tumor, whereas mature dendritic cells are located in peritumoral areas. J. Exp. Med. 190:1417–26 Iwamoto M, Shinohara H, Miyamoto A, Okuzawa M, Mabuchi H, et al. 2003. Prognostic value of tumor-infiltrating dendritic cells expressing CD83 in human breast carcinomas. Int. J. Cancer 104:92–97 Emile JF, Fraitag S, Leborgne M, de Prost Y, Brousse N. 1994. Langerhans’ cell histiocytosis cells are activated Langerhans’ cells. J. Pathol. 174:71– 76 Hage C, Willman CL, Favara BE, Isaacson PG. 1993. Langerhans’ cell histiocytosis (histiocytosis X): immunophenotype and growth fraction. Hum. Pathol. 24:840–45 Villarroel Dorrego M, Correnti M, Delgado R, Tapia FJ. 2002. Oral lichen planus: immunohistology of mucosal lesions. J. Oral Pathol. Med. 31:410–14 Metelitsa LS, Weinberg KI, Emanuel PD, Seeger RC. 2003. Expression of CD1d by myelomonocytic leukemias provides a target for cytotoxic NKT cells. Leukemia 17:1068–77 Zheng Z, Venkatapathy S, Rao G, Harrington CA. 2002. Expression profiling of B cell chronic lymphocytic leukemia suggests deficient CD1-mediated immunity, polarized cytokine response, altered adhesion and increased intracellular

13 Feb 2004

18:42

876

193.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

194.

195.

196.

197.

198.

199.

200.

201.

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

protein transport and processing of leukemic cells. Leukemia 16:2429–37 Spada FM, Borriello F, Sugita M, Watts GF, Koezuka Y, Porcelli SA. 2000. Low expression level but potent antigen presenting function of CD1d on monocyte lineage cells. Eur. J. Immunol. 30:3468– 77 Exley M, Garcia J, Wilson SB, Spada F, Gerdes D, et al. 2000. CD1d structure and regulation on human thymocytes, peripheral blood T cells, B cells and monocytes. Immunology 100:37–47 Gerlini G, Hefti HP, Kleinhans M, Nickoloff BJ, Burg G, Nestle FO. 2001. CD1d is expressed on dermal dendritic cells and monocyte-derived dendritic cells. J. Invest. Dermatol. 117:576–82 Blumberg RS, Terhorst C, Bleicher P, McDermott FV, Allan CH, et al. 1991. Expression of a nonpolymorphic MHC class I-like molecule, CD1D, by human intestinal epithelial cells. J. Immunol. 147:2518–24 Salamone MC, Rabinovich GA, Mendiguren AK, Salamone GV, Fainboim L. 2001. Activation-induced expression of CD1d antigen on mature T cells. J. Leukoc. Biol. 69:207–14 Canchis PW, Bhan AK, Landau SB, Yang L, Balk SP, Blumberg RS. 1993. Tissue distribution of the nonpolymorphic major histocompatibility complex class I-like molecule, CD1d. Immunology 80:561–65 van de Wal Y, Corazza N, Allez M, Mayer LF, Iijima H, et al. 2003. Delineation of a CD1d-restricted antigen presentation pathway associated with human and mouse intestinal epithelial cells. Gastroenterology 124:1420–31 Colgan SP, Morales VM, Madara JL, Polischuk JE, Balk SP, Blumberg RS. 1996. IFN-gamma modulates CD1d surface expression on intestinal epithelia. Am. J. Physiol. 271:C276–83 Colgan SP, Pitman RS, Nagaishi T, Mizoguchi A, Mizoguchi E, et al. 2003.

202.

203.

204.

205.

206.

207.

208.

209.

210.

Intestinal heat shock protein 110 regulates expression of CD1d on intestinal epithelial cells. J. Clin. Invest. 112:745– 54 Chen YH, Wang B, Chun T, Zhao L, Cardell S, et al. 1999. Expression of CD1d2 on thymocytes is not sufficient for the development of NK T cells in CD1d1-deficient mice. J. Immunol. 162:4560–66 Mendiratta SK, Martin WD, Hong S, Boesteanu A, Joyce S, Van Kaer L. 1997. CD1d1 mutant mice are deficient in natural T cells that promptly produce IL-4. Immunity 6:469–77 Brossay L, Jullien D, Cardell S, Sydora BC, Burdin N, et al. 1997. Mouse CD1 is mainly expressed on hemopoieticderived cells. J. Immunol. 159:1216–24 Roark JH, Park SH, Jayawardena J, Kavita U, Shannon M, Bendelac A. 1998. CD1.1 expression by mouse antigen-presenting cells and marginal zone B cells. J. Immunol. 160:3121–27 Mandal M, Chen XR, Alegre ML, Chiu NM, Chen YH, Castano AR, Wang CR. 1998. Tissue distribution, regulation and intracellular localization of murine CD1 molecules. Mol. Immunol. 35:525–36 Brasel K, De Smedt T, Smith JL, Maliszewski CR. 2000. Generation of murine dendritic cells from flt3-ligandsupplemented bone marrow cultures. Blood 96:3029–39 Krajina T, Leithauser F, Moller P, Trobonjaca Z, Reimann J. 2003. Colonic lamina propria dendritic cells in mice with CD4+ T cell-induced colitis. Eur. J. Immunol. 33:1073–83 Hameg A, Apostolou I, Leite-DeMoraes M, Gombert JM, Garcia C, et al. 2000. A subset of NKT cells that lacks the NK1.1 marker, expresses CD1d molecules, and autopresents the alphagalactosylceramide antigen. J. Immunol. 165:4917–26 Lacasse J, Martin LH. 1992. Detection of CD1 mRNA in Paneth cells of the mouse

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

211.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

212.

213.

214.

215.

216.

217.

218.

219.

intestine by in situ hybridization. J. Histochem. Cytochem. 40:1527–34 Mosser DD, Duchaine J, Martin LH. 1991. Biochemical and developmental characterization of the murine cluster of differentiation 1 antigen. Immunology 73:298–303 Joyce S, Negishi I, Boesteanu A, DeSilva AD, Sharma P, et al. 1996. Expansion of natural (NK1+) T cells that express alpha beta T cell receptors in transporters associated with antigen presentation-1 null and thymus leukemia antigen positive mice. J. Exp. Med. 184:1579– 84 Kronenberg M, Gapin L. 2002. The unconventional lifestyle of NKT cells. Nat. Rev. Immunol. 2:557–68 Hammond KJ, Pellicci DG, Poulton LD, Naidenko OV, Scalzo AA, et al. 2001. CD1d-restricted NKT cells: an interstrain comparison. J. Immunol. 167: 1164–73 Slifka MK, Pagarigan RR, Whitton JL. 2000. NK markers are expressed on a high percentage of virus-specific CD8+ and CD4+ T cells. J. Immunol. 164:2009–15 Assarsson E, Kambayashi T, Sandberg JK, Hong S, Taniguchi M, et al. 2000. CD8+ T cells rapidly acquire NK1.1 and NK cell-associated molecules upon stimulation in vitro and in vivo. J. Immunol. 165:3673–79 Chen H, Huang H, Paul WE. 1997. NK1.1+ CD4+ T cells lose NK1.1 expression upon in vitro activation. J. Immunol. 158:5112–19 Wilson MT, Johansson C, OlivaresVillagomez D, Singh AK, Stanic AK, et al. 2003. The response of natural killer T cells to glycolipid antigens is characterized by surface receptor downmodulation and expansion. Proc. Natl. Acad. Sci. USA 100:10913–18 Skold M, Cardell S. 2000. Differential regulation of Ly49 expression on CD4+ and CD4−CD8− (double nega-

220.

221.

222.

223.

224.

225.

226.

227.

228.

229.

P1: IKH

877

tive) NK1.1+ T cells. Eur. J. Immunol. 30:2488–96 Maeda M, Lohwasser S, Yamamura T, Takei F. 2001. Regulation of NKT cells by Ly49: analysis of primary NKT cells and generation of NKT cell line. J. Immunol. 167:4180–86 Prussin C, Foster B. 1997. TCR V alpha 24 and V beta 11 coexpression defines a human NK1 T cell analog containing a unique Th0 subpopulation. J. Immunol. 159:5862–70 Bendelac A, Rivera MN, Park SH, Roark JH. 1997. Mouse CD1-specific NK1 T cells: development, specificity, and function. Annu. Rev. Immunol. 15:535–62 D’Andrea A, Goux D, De Lalla C, Koezuka Y, Montagna D, et al. 2000. Neonatal invariant Valpha24+ NKT lymphocytes are activated memory cells. Eur. J. Immunol. 30:1544–50 van Der Vliet HJ, Nishi N, de Gruijl TD, von Blomberg BM, et al. 2000. Human natural killer T cells acquire a memoryactivated phenotype before birth. Blood 95:2440–42 Park SH, Benlagha K, Lee D, Balish E, Bendelac A. 2000. Unaltered phenotype, tissue distribution and function of Valpha14+ NKT cells in germ-free mice. Eur. J. Immunol. 30:620–25 Takahashi T, Chiba S, Nieda M, Azuma T, Ishihara S, et al. 2002. Cutting edge: analysis of human V alpha 24+CD8+ NK T cells activated by alphagalactosylceramide-pulsed monocytederived dendritic cells. J. Immunol. 168: 3140–44 Skold M, Faizunnessa NN, Wang CR, Cardell S. 2000. CD1d-specific NK1.1+ T cells with a transgenic variant TCR. J. Immunol. 165:168–74 Gapin L, Matsuda JL, Surh CD, Kronenberg M. 2001. NKT cells derive from double-positive thymocytes that are positively selected by CD1d. Nat. Immunol. 2:971–78 Benlagha K, Kyin T, Beavis A, Teyton L,

13 Feb 2004

18:42

878

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

230.

231.

232.

233.

234.

235.

236. 237.

238.

239.

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Bendelac A. 2002. A thymic precursor to the NK T cell lineage. Science 296:553– 55 Pellicci DG, Hammond KJ, Uldrich AP, Baxter AG, Smyth MJ, Godfrey DI. 2002. A natural killer T (NKT) cell developmental pathway involving a thymus-dependent NK1.1−CD4+ CD1ddependent precursor stage. J. Exp. Med. 195:835–44 Makino Y, Yamagata N, Sasho T, Adachi Y, Kanno R, et al. 1993. Extrathymic development of V alpha 14-positive T cells. J. Exp. Med. 177:1399–408 Shimamura M, Miura-Ohnuma J, Huang YY. 2001. Major sites for the differentiation of V alpha 14+ NKT cells inferred from the V-J junctional sequences of the invariant T-cell receptor alpha chain. Eur. J. Biochem. 268:56–61 Bendelac A. 1995. Positive selection of mouse NK1+ T cells by CD1expressing cortical thymocytes. J. Exp. Med. 182:2091–96 Chun T, Page MJ, Gapin L, Matsuda JL, Xu H, et al. 2003. CD1d-expressing dendritic cells but not thymic epithelial cells can mediate negative selection of NKT cells. J. Exp. Med. 197:907–18 Pellicci DG, Uldrich AP, Kyparissoudis K, Crowe NY, Brooks AG, et al. 2003. Intrathymic NKT cell development is blocked by the presence of alphagalactosylceramide. Eur. J. Immunol. 33: 1816–23 MacDonald HR. 2002. T before NK. Science 296:481–82 Gadue P, Morton N, Stein PL. 1999. The Src family tyrosine kinase Fyn regulates natural killer T cell development. J. Exp. Med. 190:1189–96 Eberl G, Lowin-Kropf B, MacDonald HR. 1999. Cutting edge: NKT cell development is selectively impaired in Fyndeficient mice. J. Immunol. 163:4091– 94 Gadue P, Stein PL. 2002. NK T cell precursors exhibit differential cytokine reg-

240.

241.

242.

243.

244.

245.

246.

247.

ulation and require Itk for efficient maturation. J. Immunol. 169:2397–406 Walunas TL, Wang B, Wang CR, Leiden JM. 2000. Cutting edge: the Ets1 transcription factor is required for the development of NK T cells in mice. J. Immunol. 164:2857–60 Ohteki T, Yoshida H, Matsuyama T, Duncan GS, Mak TW, Ohashi PS. 1998. The transcription factor interferon regulatory factor 1 (IRF-1) is important during the maturation of natural killer 1.1+ T cell receptor-alpha/beta+ (NK1+ T) cells, natural killer cells, and intestinal intraepithelial T cells. J. Exp. Med. 187:967–72 Elewaut D, Shaikh RB, Hammond KJ, De Winter H, Leishman AJ, et al. 2003. NIK-dependent RelB activation defines a unique signaling pathway for the development of Vα14i NKT cells. J. Exp. Med. 197:1623–33 Sivakumar V, Hammond KJ, Howells N, Pfeffer K, Weih F. 2003. Differential requirement for Rel/nuclear factor κB family members in natural killer T cell development. J. Exp. Med. 197:1613–21 Kennedy MK, Glaccum M, Brown SN, Butz EA, Viney JL, et al. 2000. Reversible defects in natural killer and memory CD8 T cell lineages in interleukin 15-deficient mice. J. Exp. Med. 191:771–80 Lodolce JP, Boone DL, Chai S, Swain RE, Dassopoulos T, et al. 1998. IL-15 receptor maintains lymphoid homeostasis by supporting lymphocyte homing and proliferation. Immunity 9:669–76 Ohteki T, Ho S, Suzuki H, Mak TW, Ohashi PS. 1997. Role for IL-15/IL-15 receptor beta-chain in natural killer 1.1+ T cell receptor-alpha beta+ cell development. J. Immunol. 159:5931–35 Sato H, Nakayama T, Tanaka Y, Yamashita M, Shibata Y, et al. 1999. Induction of differentiation of pre-NKT cells to mature Valpha14 NKT cells by granulocyte/macrophage colony-stimulating

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

248.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

249.

250.

251.

252.

253.

254.

255.

factor. Proc. Natl. Acad. Sci. USA 96: 7439–44 Elewaut D, Brossay L, Santee SM, Naidenko OV, Burdin N, et al. 2000. Membrane lymphotoxin is required for the development of different subpopulations of NK T cells. J. Immunol. 165:671–79 Iizuka K, Chaplin DD, Wang Y, Wu Q, Pegg LE, et al. 1999. Requirement for membrane lymphotoxin in natural killer cell development. Proc. Natl. Acad. Sci. USA 96:6336–40 Norris S, Doherty DG, Collins C, McEntee G, Traynor O, Hegarty JE, O’Farrelly C. 1999. Natural T cells in the human liver: cytotoxic lymphocytes with dual T cell and natural killer cell phenotype and function are phenotypically heterogenous and include Valpha24JalphaQ and gammadelta T cell receptor bearing cells. Hum. Immunol. 60:20–31 Karadimitris A, Gadola S, Altamirano M, Brown D, Woolfson A, et al. 2001. Human CD1d-glycolipid tetramers generated by in vitro oxidative refolding chromatography. Proc. Natl. Acad. Sci. USA 98:3294–98 Kita H, Naidenko OV, Kronenberg M, Ansari AA, Rogers P, et al. 2002. Quantitation and phenotypic analysis of natural killer T cells in primary biliary cirrhosis using a human CD1d tetramer. Gastroenterology 123:1031–43 Kim CH, Johnston B, Butcher EC. 2002. Trafficking machinery of NKT cells: shared and differential chemokine receptor expression among Valpha24+ Vbeta11+ NKT cell subsets with distinct cytokine-producing capacity. Blood 100:11–16 Thomas SY, Hou R, Boyson JE, Means TK, Hess C, et al. 2003. CD1d-restricted NKT cells express a chemokine receptor profile indicative of Th1-type inflammatory homing cells. J. Immunol. 171:2571–80 Johnston B, Kim CH, Soler D, Emoto

256.

257.

258.

259.

260.

261.

262.

263.

P1: IKH

879

M, Butcher EC. 2003. Differential chemokine responses and homing patterns of murine TCRalphabeta NKT cell subsets. J. Immunol. 171:2960–69 Apostolou I, Takahama Y, Belmant C, Kawano T, Huerre M, et al. 1999. Murine natural killer cells contribute to the granulomatous reaction caused by mycobacterial cell walls. Proc. Natl. Acad. Sci. USA 96:5141–46 Gilleron M, Ronet C, Mempel M, Monsarrat B, Gachelin G, Puzo G. 2001. Acylation state of the phosphatidylinositol mannosides from Mycobacterium bovis bacillus Calmette Guerin and ability to induce granuloma and recruit natural killer T cells. J. Biol. Chem. 276:34896– 904 Mempel M, Ronet C, Suarez F, Gilleron M, Puzo G, et al. 2002. Natural killer T cells restricted by the monomorphic MHC class 1b CD1d1 molecules behave like inflammatory cells. J. Immunol. 168:365–71 Kawakami K, Kinjo Y, Uezu K, Yara S, Miyagi K, et al. 2001. Monocyte chemoattractant protein-1-dependent increase of V alpha 14 NKT cells in lungs and their roles in Th1 response and host defense in cryptococcal infection. J. Immunol. 167:6525–32 Faunce DE, Sonoda KH, Stein-Streilein J. 2001. MIP-2 recruits NKT cells to the spleen during tolerance induction. J. Immunol. 166:313–21 Arase H, Arase N, Nakagawa K, Good RA, Onoe K. 1993. NK1.1+ CD4+ CD8thymocytes with specific lymphokine secretion. Eur. J. Immunol. 23:307–10 Yoshimoto T, Paul WE. 1994. CD4pos, NK1.1pos T cells promptly produce interleukin 4 in response to in vivo challenge with anti-CD3. J. Exp. Med. 179:1285–95 Matsuda JL, Gapin L, Baron JL, Sidobre S, Stetson DB, et al. 2003. Mouse V alpha 14i natural killer T cells are resistant to cytokine polarization in vivo.

13 Feb 2004

18:42

880

264.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

265.

266.

267.

268.

269.

270.

271.

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

Proc. Natl. Acad. Sci. USA 100:8395– 400 Stetson DB, Mohrs M, Reinhardt RL, Baron JL, Wang ZE, et al. 2003. Constitutive cytokine mRNAs mark natural killer (NK) and NK T cells poised for rapid effector functions. J. Exp. Med. 198:1069–76 Kitamura H, Iwakabe K, Yahata T, Nishimura S, Ohta A, et al. 1999. The natural killer T (NKT) cell ligand alphagalactosylceramide demonstrates its immunopotentiating effect by inducing interleukin (IL)-12 production by dendritic cells and IL-12 receptor expression on NKT cells. J. Exp. Med. 189:1121–28 Hayakawa Y, Takeda K, Yagita H, Van Kaer L, Saiki I, Okumura K. 2001. Differential regulation of Th1 and Th2 functions of NKT cells by CD28 and CD40 costimulatory pathways. J. Immunol. 166:6012–18 Tomura M, Yu WG, Ahn HJ, Yamashita M, Yang YF, et al. 1999. A novel function of Valpha14+CD4+NKT cells: stimulation of IL-12 production by antigenpresenting cells in the innate immune system. J. Immunol. 163:93–101 Yang YF, Tomura M, Ono S, Hamaoka T, Fujiwara H. 2000. Requirement for IFNgamma in IL-12 production induced by collaboration between vα14+ NKT cells and antigen-presenting cells. Int. Immunol. 12:1669–75 Arase H, Arase N, Kobayashi Y, Nishimura Y, Yonehara S, Onoe K. 1994. Cytotoxicity of fresh NK1.1+ T cell receptor alpha/beta+ thymocytes against a CD4+8+ thymocyte population associated with intact Fas antigen expression on the target. J. Exp. Med. 180:423–32 Kawano T, Cui J, Koezuka Y, Toura I, Kaneko Y, et al. 1998. Natural killerlike nonspecific tumor cell lysis mediated by specific ligand-activated Valpha14 NKT cells. Proc. Natl. Acad. Sci. USA 95:5690–93 Nieda M, Nicol A, Koezuka Y, Kikuchi

272.

273.

274.

275.

276.

277.

278.

279.

A, Lapteva N, et al. 2001. TRAIL expression by activated human CD4+V alpha 24NKT cells induces in vitro and in vivo apoptosis of human acute myeloid leukemia cells. Blood 97:2067–74 Arase H, Arase-Fukushi N, Good RA, Onoe K. 1993. Lymphokine-activated killer cell activity of CD4−CD8−TCR alpha beta+ thymocytes. J. Immunol. 151:546–55 Hashimoto W, Takeda K, Anzai R, Ogasawara K, Sakihara H, et al. 1995. Cytotoxic NK1.1 Ag+ alpha beta T cells with intermediate TCR induced in the liver of mice by IL-12. J. Immunol. 154:4333–40 Gansert JL, Kiessler V, Engele M, Wittke F, Rollinghoff M, et al. 2003. Human NKT cells express granulysin and exhibit antimycobacterial activity. J. Immunol. 170:3154–61 Kawamura T, Takeda K, Mendiratta SK, Kawamura H, Van Kaer L, et al. 1998. Critical role of NK1+ T cells in IL-12induced immune responses in vivo. J. Immunol. 160:16–19 Arase H, Arase N, Saito T. 1996. Interferon gamma production by natural killer (NK) cells and NK1.1+ T cells upon NKR-P1 cross-linking. J. Exp. Med. 183:2391–96 Hansen DS, Siomos MA, Buckingham L, Scalzo AA, Schofield L. 2003. Regulation of murine cerebral malaria pathogenesis by CD1d-restricted NKT cells and the natural killer complex. Immunity 18:391–402 Hameg A, Gouarin C, Gombert JM, Hong S, Van Kaer L, et al. 1999. IL-7 up-regulates IL-4 production by splenic NK1.1+ and NK1.1− MHC class Ilike/CD1-dependent CD4+ T cells. J. Immunol. 162:7067–74 Leite-De-Moraes MC, Hameg A, Pacilio M, Koezuka Y, Taniguchi M, et al. 2001. IL-18 enhances IL-4 production by ligand-activated NKT lymphocytes: a pro-Th2 effect of IL-18 exerted through NKT cells. J. Immunol. 166:945–51

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES 280. Yoshimoto T, Min B, Sugimoto T, Hayashi N, Ishikawa Y, et al. 2003. Nonredundant roles for CD1d-restricted natural killer T cells and conventional CD4+ T cells in the induction of immunoglobulin E antibodies in response to interleukin 18 treatment of mice. J. Exp. Med. 197:997–1005 281. Kadowaki N, Antonenko S, Ho S, Rissoan MC, Soumelis V, et al. 2001. Distinct cytokine profiles of neonatal natural killer T cells after expansion with subsets of dendritic cells. J. Exp. Med. 193:1221–26 282. Miyamoto K, Miyake S, Yamamura T. 2001. A synthetic glycolipid prevents autoimmune encephalomyelitis by inducing TH2 bias of natural killer T cells. Nature 413:531–34 283. Singh N, Hong S, Scherer DC, Serizawa I, Burdin N, et al. 1999. Cutting edge: activation of NK T cells by CD1d and alpha-galactosylceramide directs conventional T cells to the acquisition of a Th2 phenotype. J. Immunol. 163:2373–77 284. Burdin N, Brossay L, Kronenberg M. 1999. Immunization with alpha-galactosylceramide polarizes CD1-reactive NK T cells towards Th2 cytokine synthesis. Eur. J. Immunol. 29:2014– 25 285. Sandberg JK, Bhardwaj N, Nixon DF. 2003. Dominant effector memory characteristics, capacity for dynamic adaptive expansion, and sex bias in the innate Valpha24 NKT cell compartment. Eur. J. Immunol. 33:588–96 286. Carnaud C, Lee D, Donnars O, Park SH, Beavis A, et al. 1999. Cutting edge: cross-talk between cells of the innate immune system: NKT cells rapidly activate NK cells. J. Immunol. 163:4647– 50 287. Eberl G, MacDonald HR. 2000. Selective induction of NK cell proliferation and cytotoxicity by activated NKT cells. Eur. J. Immunol. 30:985–92

P1: IKH

881

288. Eberl G, Brawand P, MacDonald HR. 2000. Selective bystander proliferation of memory CD4+ and CD8+ T cells upon NKT or T cell activation. J. Immunol. 165:4305–11 289. Metelitsa LS, Naidenko OV, Kant A, Wu HW, Loza MJ, et al. 2001. Human NKT cells mediate antitumor cytotoxicity directly by recognizing target cell CD1d with bound ligand or indirectly by producing IL-2 to activate NK cells. J. Immunol. 167:3114–22 290. Nieuwenhuis EE, Matsumoto T, Exley M, Schleipman RA, Glickman J, et al. 2002. CD1d-dependent macrophagemediated clearance of Pseudomonas aeruginosa from lung. Nat. Med. 8:588– 93 291. Kitamura H, Ohta A, Sekimoto M, Sato M, Iwakabe K, et al. 2000. Alphagalactosylceramide induces early B-cell activation through IL-4 production by NKT cells. Cell. Immunol. 199:37–42 292. Galli G, Nuti S, Tavarini S, GalliStampino L, De Lalla C, et al. 2003. CD1d-restricted help to B cells by human invariant natural killer T lymphocytes. J. Exp. Med. 197:1051–57 293. Fujii S, Shimizu K, Smith C, Bonifaz L, Steinman RM. 2003. Activation of natural killer T cells by alphagalactosylceramide rapidly induces the full maturation of dendritic cells in vivo and thereby acts as an adjuvant for combined CD4 and CD8 T cell immunity to a coadministered protein. J. Exp. Med. 198:267–79 293a. Hermans IF, Silk JD, Gileadi U, Salio M, Mathew B, et al. 2003. NKT cells enhance CD4+ and CD8+ T cell responses to soluble antigen in vivo through direct interaction with dendritic cells. J. Immunol. 171:5140–47 293b. Stober D, Jomantaite I, Schirmbeck R, Reimann J. 2003. NKT cells provide help for dendritic cell-dependent priming of MHC class I-restricted CD8(+) T cells in vivo. J. Immunol. 170:2540–48

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

882

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

294. Nicol A, Nieda M, Koezuka Y, Porcelli S, Suzuki K, et al. 2000. Dendritic cells are targets for human invariant Valpha24+ natural killer T-cell cytotoxic activity: an important immune regulatory function. Exp. Hematol. 28:276–82 295. Yang OO, Racke FK, Nguyen PT, Gausling R, Severino ME, et al. 2000. CD1d on myeloid dendritic cells stimulates cytokine secretion from and cytolytic activity of Valpha24JalphaQ T cells: a feedback mechanism for immune regulation. J. Immunol. 165:3756– 62 296. Leite-de-Moraes MC, Lisbonne M, Arnould A, Machavoine F, Herbelin A, et al. 2002. Ligand-activated natural killer T lymphocytes promptly produce IL-3 and GM-CSF in vivo: relevance to peripheral myeloid recruitment. Eur. J. Immunol. 32:1897–904 297. Eberl G, MacDonald HR. 1998. Rapid death and regeneration of NKT cells in anti-CD3epsilon- or IL-12-treated mice: a major role for bone marrow in NKT cell homeostasis. Immunity 9:345–53 298. Osman Y, Kawamura T, Naito T, Takeda K, Van Kaer L, et al. 2000. Activation of hepatic NKT cells and subsequent liver injury following administration of alpha-galactosylceramide. Eur. J. Immunol. 30:1919–28 299. Hobbs JA, Cho S, Roberts TJ, Sriram V, Zhang J, et al. 2001. Selective loss of natural killer T cells by apoptosis following infection with lymphocytic choriomeningitis virus. J. Virol. 75:10746–54 300. Kirby AC, Yrlid U, Wick MJ. 2002. The innate immune response differs in primary and secondary Salmonella infection. J. Immunol. 169:4450–59 301. Fujii S, Shimizu K, Kronenberg M, Steinman RM. 2002. Prolonged IFNgamma-producing NKT response induced with alpha-galactosylceramideloaded DCs. Nat. Immunol. 3:867–74 302. Hayakawa Y, Takeda K, Yagita H, Kakuta S, Iwakura Y, et al. 2001. Crit-

303.

303a.

303b.

304.

305.

306.

307.

ical contribution of IFN-gamma and NK cells, but not perforin-mediated cytotoxicity, to anti-metastatic effect of alphagalactosylceramide. Eur. J. Immunol. 31:1720–27 Park SH, Kyin T, Bendelac A, Carnaud C. 2003. The contribution of NKT cells, NK cells, and other gammachain-dependent non-T non-B cells to IL-12-mediated rejection of tumors. J. Immunol. 170:1197–201 Crowe NY, Uldrich AP, Kyparissoudis K, Hammond KJ, Hayakawa Y, et al. 2003. Glycolipid antigen drives rapid expansion and sustained cytokine production by NK T cells. J. Immunol. 171:4020–27 Yang JQ, Saxena V, Xu H, Van Kaer L, Wang CR, Singh RR. 2003. Repeated alpha-galactosylceramide administration results in expansion of NK T cells and alleviates inflammatory dermatitis in MRL-lpr/lpr mice. J. Immunol. 171:4439–46 Brigl M, Bry L, Kent SC, Gumperz JE, Brenner MB. 2003. Mechanism of CD1d-restricted natural killer T cell activation during microbial infection. Nat. Immunol. 4:1230–37 Kumar H, Belperron A, Barthold SW, Bockenstedt LK. 2000. Cutting edge: CD1d deficiency impairs murine host defense against the spirochete, Borrelia burgdorferi. J. Immunol. 165:4797– 801 Mempel M, Flageul B, Suarez F, Ronet C, Dubertret L, et al. 2000. Comparison of the T cell patterns in leprous and cutaneous sarcoid granulomas. Presence of Valpha24-invariant natural killer T cells in T-cell-reactive leprosy together with a highly biased T cell receptor Valpha repertoire. Am. J. Pathol. 157:509–23 Behar SM, Dascher CC, Grusby MJ, Wang CR, Brenner MB. 1999. Susceptibility of mice deficient in CD1D or TAP1 to infection with Mycobacterium tuberculosis. J. Exp. Med. 189:1973–80

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUNCTIONS OF CD1 MOLECULES 308. Sousa AO, Mazzaccaro RJ, Russell RG, Lee FK, Turner OC, et al. 2000. Relative contributions of distinct MHC class Idependent cell populations in protection to tuberculosis infection in mice. Proc. Natl. Acad. Sci. USA 97:4204–8 309. D’Souza CD, Cooper AM, Frank AA, Ehlers S, Turner J, Bendelac A, Orme IM. 2000. A novel nonclassic beta2microglobulin-restricted mechanism influencing early lymphocyte accumulation and subsequent resistance to tuberculosis in the lung. Am. J. Respir. Cell Mol. Biol. 23:188–93 310. Kawakami K, Kinjo Y, Uezu K, Yara S, Miyagi K, et al. 2002. Minimal contribution of Valpha14 natural killer T cells to Th1 response and host resistance against mycobacterial infection in mice. Microbiol. Immunol. 46:207–10 311. Dieli F, Taniguchi M, Kronenberg M, Sidobre S, Ivanyi J, et al. 2003. An anti-inflammatory role for Valpha14 NK T cells in Mycobacterium bovis bacillus Calmette-Guerin-infected mice. J. Immunol. 171:1961–68 312. Szalay G, Zugel U, Ladel CH, Kaufmann SH. 1999. Participation of group 2 CD1 molecules in the control of murine tuberculosis. Microbes Infect. 1:1153–57 313. Sugawara I, Yamada H, Mizuno S, Li CY, Nakayama T, Taniguchi M. 2002. Mycobacterial infection in natural killer T cell knockout mice. Tuberculosis 82:97– 104 314. Szalay G, Ladel CH, Blum C, Brossay L, Kronenberg M, Kaufmann SH. 1999. Cutting edge: anti-CD1 monoclonal antibody treatment reverses the production patterns of TGF-beta 2 and Th1 cytokines and ameliorates listeriosis in mice. J. Immunol. 162:6955–58 315. Ishigami M, Nishimura H, Naiki Y, Yoshioka K, Kawano T, et al. 1999. The roles of intrahepatic Valpha14+ NK1.1+ T cells for liver injury induced by Salmonella infection in mice. Hepatology 29:1799–808

P1: IKH

883

316. Ishikawa H, Hisaeda H, Taniguchi M, Nakayama T, Sakai T, et al. 2000. CD4+ Vα14 NKT cells play a crucial role in an early stage of protective immunity against infection with Leishmania major. Int. Immunol. 12:1267–74 317. Miyahira Y, Katae M, Takeda K, Yagita H, Okumura K, et al. 2003. Activation of natural killer T cells by alpha-galactosylceramide impairs DNA vaccine-induced protective immunity against Trypanosoma cruzi. Infect. Immun. 71:1234–41 317a. Hansen DS, Siomos MA, De KoningWard T, Buckingham L, Crabb BS, Schofield L. 2003. CD1d-restricted NKT cells contribute to malarial splenomegaly and enhance parasite-specific antibody responses. Eur. J. Immunol. 33: 2588–98 318. Faveeuw C, Angeli V, Fontaine J, Maliszewski C, Capron A, et al. 2002. Antigen presentation by CD1d contributes to the amplification of Th2 responses to Schistosoma mansoni glycoconjugates in mice. J. Immunol. 169:906–12 319. Denkers EY, Scharton-Kersten T, Barbieri S, Caspar P, Sher A. 1996. A role for CD4+ NK1.1+ T lymphocytes as major histocompatibility complex class II independent helper cells in the generation of CD8+ effector function against intracellular infection. J. Exp. Med. 184:131– 39 320. Grubor-Bauk B, Simmons A, Mayrhofer G, Speck PG. 2003. Impaired clearance of herpes simplex virus type 1 from mice lacking CD1d or NKT cells expressing the semivariant V alpha 14-J alpha 281 TCR. J. Immunol. 170:1430–34 321. Ashkar AA, Rosenthal KL. 2003. Interleukin-15 and natural killer and NKT cells play a critical role in innate protection against genital herpes simplex virus type 2 infection. J. Virol. 77:10168– 71 322. Johnson TR, Hong S, Van Kaer L, Koezuka Y, Graham BS. 2002. NK

13 Feb 2004

18:42

884

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

323.

324.

325.

326.

327.

328.

329.

330.

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

T cells contribute to expansion of CD8+ T cells and amplification of antiviral immune responses to respiratory syncytial virus. J. Virol. 76:4294–303 Exley MA, Bigley NJ, Cheng O, Tahir SM, Smiley ST, et al. 2001. CD1dreactive T-cell activation leads to amelioration of disease caused by diabetogenic encephalomyocarditis virus. J. Leukoc. Biol. 69:713–18 Baron JL, Gardiner L, Nishimura S, Shinkai K, Locksley R, Ganem D. 2002. Activation of a nonclassical NKT cell subset in a transgenic mouse model of hepatitis B virus infection. Immunity 16:583–94 Kakimi K, Guidotti LG, Koezuka Y, Chisari FV. 2000. Natural killer T cell activation inhibits hepatitis B virus replication in vivo. J. Exp. Med. 192:921–30 Chackerian A, Alt J, Perera V, Behar SM. 2002. Activation of NKT cells protects mice from tuberculosis. Infect. Immun. 70:6302–9 Gonzalez-Aseguinolaza G, de Oliveira C, Tomaska M, Hong S, Bruna-Romero O, et al. 2000. α-galactosylceramideactivated Valpha 14 natural killer T cells mediate protection against murine malaria. Proc. Natl. Acad. Sci. USA 97: 8461–66 Duthie MS, Kahn SJ. 2002. Treatment with alpha-galactosylceramide before Trypanosoma cruzi infection provides protection or induces failure to thrive. J. Immunol. 168:5778–85 van Dommelen SL, Tabarias HA, Smyth MJ, Degli-Esposti MA. 2003. Activation of natural killer (NK) T cells during murine cytomegalovirus infection enhances the antiviral response mediated by NK cells. J. Virol. 77:1877–84 Kawakami K, Kinjo Y, Yara S, Koguchi Y, Uezu K, et al. 2001. Activation of Valpha14+ natural killer T cells by alphagalactosylceramide results in development of Th1 response and local host resistance in mice infected with Cryp-

331.

332.

333.

334.

335.

336.

337.

338.

339.

tococcus neoformans. Infect. Immun. 69:213–20 Gonzalez-Aseguinolaza G, Van Kaer L, Bergmann CC, Wilson JM, Schmieg J, et al. 2002. Natural killer T cell ligand alpha-galactosylceramide enhances protective immunity induced by malaria vaccines. J. Exp. Med. 195:617–24 Cui J, Shin T, Kawano T, Sato H, Kondo E, et al. 1997. Requirement for Valpha14 NKT cells in IL-12-mediated rejection of tumors. Science 278:1623–26 Shin T, Nakayama T, Akutsu Y, Motohashi S, Shibata Y, et al. 2001. Inhibition of tumor metastasis by adoptive transfer of IL-12-activated Valpha14 NKT cells. Int. J. Cancer 91:523–28 Anzai R, Seki S, Ogasawara K, Hashimoto W, Sugiura K, et al. 1996. Interleukin-12 induces cytotoxic NK1+ alpha beta T cells in the lungs of euthymic and athymic mice. Immunology 88:82–89 Kodama T, Takeda K, Shimozato O, Hayakawa Y, Atsuta M, et al. 1999. Perforin-dependent NK cell cytotoxicity is sufficient for anti-metastatic effect of IL-12. Eur. J. Immunol. 29:1390–96 Nakui M, Ohta A, Sekimoto M, Sato M, Iwakabe K, et al. 2000. Potentiation of antitumor effect of NKT cell ligand, alpha-galactosylceramide by combination with IL-12 on lung metastasis of malignant melanoma cells. Clin. Exp. Metastasis 18:147–53 Smyth MJ, Taniguchi M, Street SE. 2000. The anti-tumor activity of IL-12: mechanisms of innate immunity that are model and dose dependent. J. Immunol. 165:2665–70 Takeda K, Hayakawa Y, Atsuta M, Hong S, Van Kaer L, et al. 2000. Relative contribution of NK and NKT cells to the anti-metastatic activities of IL-12. Int. Immunol. 12:909–14 Hayakawa Y, Takeda K, Yagita H, Smyth MJ, Van Kaer L, et al. 2002. IFNgamma-mediated inhibition of tumor

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

340.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

341.

342.

343.

344.

345.

346.

347.

angiogenesis by natural killer T-cell ligand, alpha-galactosylceramide. Blood 100:1728–33 Smyth MJ, Crowe NY, Hayakawa Y, Takeda K, Yagita H, Godfrey DI. 2002. NKT cells—conductors of tumor immunity? Curr. Opin. Immunol. 14:165–71 Smyth MJ, Cretney E, Takeda K, Wiltrout RH, Sedger LM, et al. 2001. Tumor necrosis factor-related apoptosisinducing ligand (TRAIL) contributes to interferon gamma-dependent natural killer cell protection from tumor metastasis. J. Exp. Med. 193:661–70 Kobayashi E, Motoki K, Uchida T, Fukushima H, Koezuka Y. 1995. KRN7000, a novel immunomodulator, and its antitumor activities. Oncol. Res. 7:529–34 Nakagawa R, Motoki K, Ueno H, Iijima R, Nakamura H, et al. 1998. Treatment of hepatic metastasis of the colon26 adenocarcinoma with an alphagalactosylceramide, KRN7000. Cancer Res. 58:1202–7 Smyth MJ, Crowe NY, Pellicci DG, Kyparissoudis K, Kelly JM, et al. 2002. Sequential production of interferon-gamma by NK1.1+ T cells and natural killer cells is essential for the antimetastatic effect of alpha-galactosylceramide. Blood 99:1259–66 Nakagawa R, Nagafune I, Tazunoki Y, Ehara H, Tomura H, et al. 2001. Mechanisms of the antimetastatic effect in the liver and of the hepatocyte injury induced by alpha-galactosylceramide in mice. J. Immunol. 166:6578–84 Toura I, Kawano T, Akutsu Y, Nakayama T, Ochiai T, Taniguchi M. 1999. Cutting edge: inhibition of experimental tumor metastasis by dendritic cells pulsed with alpha-galactosylceramide. J. Immunol. 163:2387–91 Smyth MJ, Thia KY, Street SE, Cretney E, Trapani JA, et al. 2000. Differential tumor surveillance by natural killer (NK) and NKT cells. J. Exp. Med. 191:661–68

P1: IKH

885

348. Smyth MJ, Crowe NY, Godfrey DI. 2001. NK cells and NKT cells collaborate in host protection from methylcholanthrene-induced fibrosarcoma. Int. Immunol. 13:459–63 349. Crowe NY, Smyth MJ, Godfrey DI. 2002. A critical role for natural killer T cells in immunosurveillance of methylcholanthrene-induced sarcomas. J. Exp. Med. 196:119–27 350. Stewart TJ, Smyth MJ, Fernando GJ, Frazer IH, Leggatt GR. 2003. Inhibition of early tumor growth requires J alpha 18-positive (natural killer T) cells. Cancer Res. 63:3058–60 351. Gillessen S, Naumov YN, Nieuwenhuis EE, Exley MA, Lee FS, et al. 2003. CD1d-restricted T cells regulate dendritic cell function and antitumor immunity in a granulocyte-macrophage colony-stimulating factor-dependent fashion. Proc. Natl. Acad. Sci. USA 100:8874–79 352. Terabe M, Matsui S, Noben-Trauth N, Chen H, Watson C, et al. 2000. NKT cellmediated repression of tumor immunosurveillance by IL-13 and the IL-4RSTAT6 pathway. Nat. Immunol. 1:515– 20 353. Ostrand-Rosenberg S, Clements VK, Terabe M, Park JM, Berzofsky JA, Dissanayake SK. 2002. Resistance to metastatic disease in STAT6-deficient mice requires hemopoietic and nonhemopoietic cells and is IFN-gamma dependent. J. Immunol. 169:5796–804 354. Terabe M, Matsui S, Park JM, Mamura M, Noben-Trauth N, et al. 2003. Transforming growth factor-{beta} production and myeloid cells are an effector mechanism through which CD1drestricted T cells block cytotoxic T lymphocyte-mediated tumor immunosurveillance: abrogation prevents tumor recurrence. J. Exp. Med. 198:1741–52 355. Sfondrini L, Besusso D, Zoia MT, Rodolfo M, Invernizzi AM, et al. 2002. Absence of the CD1 molecule

13 Feb 2004

18:42

886

356.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

357.

358.

359.

360.

361.

362.

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

up-regulates antitumor activity induced by CpG oligodeoxynucleotides in mice. J. Immunol. 169:151–58 Moodycliffe AM, Nghiem D, Clydesdale G, Ullrich SE. 2000. Immune suppression and skin cancer development: regulation by NKT cells. Nat. Immunol. 1:521–25 Nicol A, Nieda M, Koezuka Y, Porcelli S, Suzuki K, et al. 2000. Human invariant valpha24+ natural killer T cells activated by alpha-galactosylceramide (KRN7000) have cytotoxic anti-tumour activity through mechanisms distinct from T cells and natural killer cells. Immunology 99:229–34 Hagihara M, Gansuvd B, Ueda Y, Tsuchiya T, Masui A, et al. 2002. Killing activity of human umbilical cord blood-derived TCRValpha24+ NKT cells against normal and malignant hematological cells in vitro: a comparative study with NK cells or OKT3 activated T lymphocytes or with adult peripheral blood NKT cells. Cancer Immunol. Immunother. 51:1–8 Gansuvd B, Hagihara M, Yu Y, Inoue H, Ueda Y, et al. 2002. Human umbilical cord blood NK T cells kill tumors by multiple cytotoxic mechanisms. Hum. Immunol. 63:164–75 Takahashi T, Haraguchi K, Chiba S, Yasukawa M, Shibata Y, Hirai H. 2003. Valpha24+ natural killer T-cell responses against T-acute lymphoblastic leukaemia cells: implications for immunotherapy. Br. J. Haematol. 122:231–39 Dhodapkar MV, Geller MD, Chang DH, Shimizu K, Fujii S, et al. 2003. A reversible defect in natural killer T cell function characterizes the progression of premalignant to malignant multiple myeloma. J. Exp. Med. 197:1667–76 Ishihara S, Nieda M, Kitayama J, Osada T, Yabe T, et al. 2000. Alphaglycosylceramides enhance the antitumor cytotoxicity of hepatic lymphocytes obtained from cancer patients by activat-

363.

364.

365.

366.

367.

368.

369.

370.

ing CD3−CD56+ NK cells in vitro. J. Immunol. 165:1659–64 Tahir SM, Cheng O, Shaulov A, Koezuka Y, Bubley GJ, et al. 2001. Loss of IFN-gamma production by invariant NK T cells in advanced cancer. J. Immunol. 167:4046–50 Yanagisawa K, Seino K, Ishikawa Y, Nozue M, Todoroki T, Fukao K. 2002. Impaired proliferative response of V alpha 24 NKT cells from cancer patients against alpha-galactosylceramide. J. Immunol. 168:6494–99 Giaccone G, Punt CJ, Ando Y, Ruijter R, Nishi N, et al. 2002. A phase I study of the natural killer T-cell ligand alphagalactosylceramide (KRN7000) in patients with solid tumors. Clin. Cancer Res. 8:3702–9 Fujii S, Shimizu K, Klimek V, Geller MD, Nimer SD, Dhodapkar MV. 2003. Severe and selective deficiency of interferon-gamma-producing invariant natural killer T cells in patients with myelodysplastic syndromes. Br. J. Haematol. 122:617–22 Kenna T, Mason LG, Porcelli SA, Koezuka Y, Hegarty JE, et al. 2003. NKT cells from normal and tumorbearing human livers are phenotypically and functionally distinct from murine NKT cells. J. Immunol. 171:1775– 79 Gombert JM, Herbelin A, TancredeBohin E, Dy M, Carnaud C, Bach JF. 1996. Early quantitative and functional deficiency of NK1+-like thymocytes in the NOD mouse. Eur. J. Immunol. 26:2989–98 Baxter AG, Kinder SJ, Hammond KJ, Scollay R, Godfrey DI. 1997. Association between alphabetaTCR+CD4− CD8−T-cell deficiency and IDDM in NOD/Lt mice. Diabetes 46:572–82 Godfrey DI, Kinder SJ, Silvera P, Baxter AG. 1997. Flow cytometric study of T cell development in NOD mice reveals a deficiency in

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

371.

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

372.

373.

373a.

374.

375.

376.

377.

378.

alphabetaTCR+CDR−CD8− thymocytes. J. Autoimmun. 10:279–85 Poulton LD, Smyth MJ, Hawke CG, Silveira P, Shepherd D, et al. 2001. Cytometric and functional analyses of NK and NKT cell deficiencies in NOD mice. Int. Immunol. 13:887–96 Hong S, Wilson MT, Serizawa I, Wu L, Singh N, et al. 2001. The natural killer T-cell ligand alpha-galactosylceramide prevents autoimmune diabetes in nonobese diabetic mice. Nat. Med. 7:1052– 56 Esteban LM, Tsoutsman T, Jordan MA, Roach D, Poulton LD, et al. 2003. Genetic control of NKT cell numbers maps to major diabetes and lupus loci. J. Immunol. 171:2873–78 Zhang F, Liang Z, Matsuki N, Van Kaer L, Joyce S, et al. 2003. A murine locus on chromosome 18 controls NKT cell homeostasis and Th cell differentiation. J. Immunol. 171: 4613–20 Wang B, Geng YB, Wang CR. 2001. CD1-restricted NK T cells protect nonobese diabetic mice from developing diabetes. J. Exp. Med. 194:313–20 Shi FD, Flodstrom M, Balasa B, Kim SH, Van Gunst K, et al. 2001. Germline deletion of the CD1 locus exacerbates diabetes in the NOD mouse. Proc. Natl. Acad. Sci. USA 98:6777–82 Naumov YN, Bahjat KS, Gausling R, Abraham R, Exley MA, et al. 2001. Activation of CD1d-restricted T cells protects NOD mice from developing diabetes by regulating dendritic cell subsets. Proc. Natl. Acad. Sci. USA 98:13838–43 Hammond KJ, Poulton LD, Palmisano LJ, Silveira PA, Godfrey DI, Baxter AG. 1998. alpha/beta-T cell receptor (TCR)+CD4−CD8− (NKT) thymocytes prevent insulin-dependent diabetes mellitus in nonobese diabetic (NOD)/Lt mice by the influence of interleukin (IL)4 and/or IL-10. J. Exp. Med. 187:1047– 56 Lehuen A, Lantz O, Beaudoin L, Laloux

378a.

379.

380.

381.

382.

383.

384.

P1: IKH

887

V, Carnaud C, et al. 1998. Overexpression of natural killer T cells protects Valpha14-Jalpha281 transgenic nonobese diabetic mice against diabetes. J. Exp. Med. 188:1831–39 Falcone M, Yeung B, Tucker L, Rodriguez E, Sarvetnick N. 1999. A defect in interleukin 12-induced activation and interferon gamma secretion of peripheral natural killer T cells in nonobese diabetic mice suggests new pathogenic mechanisms for insulin-dependent diabetes mellitus. J. Exp. Med. 190:963–72 Laloux V, Beaudoin L, Jeske D, Carnaud C, Lehuen A. 2001. NKT cell-induced protection against diabetes in V alpha 14J alpha 281 transgenic nonobese diabetic mice is associated with a TH2 shift circumscribed regionally to the islets and functionally to islet autoantigen. J. Immunol. 166:3749–56 Beaudoin L, Laloux V, Novak J, Lucas B, Lehuen A. 2002. NKT cells inhibit the onset of diabetes by impairing the development of pathogenic T cells specific for pancreatic Beta cells. Immunity 17:725–36 Sharif S, Arreaza GA, Zucker P, Mi QS, Sondhi J, et al. 2001. Activation of natural killer T cells by alphagalactosylceramide treatment prevents the onset and recurrence of autoimmune type 1 diabetes. Nat. Med. 7:1057–62 Wilson SB, Kent SC, Patton KT, Orban T, Jackson RA, et al. 1998. Extreme TH1 bias of invariant Valpha24JalphaQ T cells in type 1 diabetes. Nature 391:177–81. Erratum. 1999. Nature 399: 84 Kukreja A, Cost G, Marker J, Zhang C, Sun Z, et al. 2002. Multiple immunoregulatory defects in type-1 diabetes. J. Clin. Invest. 109:131–40 Lee DJ, Abeyratne A, Carson DA, Corr M. 1998. Induction of an antigenspecific, CD1-restricted cytotoxic T lymphocyte response in vivo. J. Exp. Med. 187:433–38

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

888

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

385. Oikawa Y, Shimada A, Yamada S, Motohashi Y, Nakagawa Y, et al. 2002. High frequency of Valpha24+ Vbeta11+ Tcells observed in type 1 diabetes. Diabetes Care 25:1818–23 386. Sonoda KH, Taniguchi M, SteinStreilein J. 2002. Long-term survival of corneal allografts is dependent on intact CD1d-reactive NKT cells. J. Immunol. 168:2028–34 387. Seino KI, Fukao K, Muramoto K, Yanagisawa K, Takada Y, et al. 2001. Requirement for natural killer T (NKT) cells in the induction of allograft tolerance. Proc. Natl. Acad. Sci. USA 98:2577–81 388. Higuchi M, Zeng D, Shizuru J, Gworek J, Dejbakhsh-Jones S, et al. 2002. Immune tolerance to combined organ and bone marrow transplants after fractionated lymphoid irradiation involves regulatory NK T cells and clonal deletion. J. Immunol. 169:5564–70 389. Ikehara Y, Yasunami Y, Kodama S, Maki T, Nakano M, et al. 2000. CD4+ Valpha14 natural killer T cells are essential for acceptance of rat islet xenografts in mice. J. Clin. Invest. 105:1761– 67 390. Sonoda KH, Exley M, Snapper S, Balk SP, Stein-Streilein J. 1999. CD1-reactive natural killer T cells are required for development of systemic tolerance through an immune-privileged site. J. Exp. Med. 190:1215–26 391. Sonoda KH, Stein-Streilein J. 2002. CD1d on antigen-transporting APC and splenic marginal zone B cells promotes NKT cell-dependent tolerance. Eur. J. Immunol. 32:848–57 392. Faunce DE, Stein-Streilein J. 2002. NKT cell-derived RANTES recruits APCs and CD8+ T cells to the spleen during the generation of regulatory T cells in tolerance. J. Immunol. 169:31–38 393. Sonoda KH, Faunce DE, Taniguchi M, Exley M, Balk S, Stein-Streilein J. 2001. NK T cell-derived IL-10 is essential for the differentiation of antigen-specific

394.

395.

396.

397.

398.

399.

400.

401.

T regulatory cells in systemic tolerance. J. Immunol. 166:42–50 Yoshimoto T, Bendelac A, Hu-Li J, Paul WE. 1995. Defective IgE production by SJL mice is linked to the absence of CD4+, NK1.1+ T cells that promptly produce interleukin 4. Proc. Natl. Acad. Sci. USA 92:11931–34 Busshoff U, Hein A, Iglesias A, Dorries R, Regnier-Vigouroux A. 2001. CD1 expression is differentially regulated by microglia, macrophages and T cells in the central nervous system upon inflammation and demyelination. J. Neuroimmunol. 113:220–30 Mars LT, Laloux V, Goude K, Desbois S, Saoudi A, et al. 2002. Cutting edge: V alpha 14-J alpha 281 NKT cells naturally regulate experimental autoimmune encephalomyelitis in nonobese diabetic mice. J. Immunol. 168:6007–11 Singh AK, Wilson MT, Hong S, Olivares-Villagomez D, Du C, et al. 2001. Natural killer T cell activation protects mice against experimental autoimmune encephalomyelitis. J. Exp. Med. 194:1801–11 Furlan R, Bergami A, Cantarella D, Brambilla E, Taniguchi M, et al. 2003. Activation of invariant NKT cells by alphaGalCer administration protects mice from MOG35-55-induced EAE: critical roles for administration route and IFNgamma. Eur. J. Immunol. 33:1830–38 Pal E, Tabira T, Kawano T, Taniguchi M, Miyake S, Yamamura T. 2001. Costimulation-dependent modulation of experimental autoimmune encephalomyelitis by ligand stimulation of V alpha 14 NK T cells. J. Immunol. 166:662–68 Jahng AW, Maricic I, Pedersen B, Burdin N, Naidenko O, et al. 2001. Activation of natural killer T cells potentiates or prevents experimental autoimmune encephalomyelitis. J. Exp. Med. 194:1789– 99 Zeng D, Liu Y, Sidobre S, Kronenberg

13 Feb 2004

18:42

AR

AR210-IY22-28.tex

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

FUNCTIONS OF CD1 MOLECULES

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

402.

403.

404.

405.

406.

407.

408.

409.

M, Strober S. 2003. Activation of natural killer T cells in NZB/W mice induces Th1-type immune responses exacerbating lupus. J. Clin. Invest. 112:1211–22 Morshed SR, Mannoor K, Halder RC, Kawamura H, Bannai M, et al. 2002. Tissue-specific expansion of NKT and CD5+B cells at the onset of autoimmune disease in (NZBxNZW)F1 mice. Eur. J. Immunol. 32:2551–61 Zeng D, Lee MK, Tung J, Brendolan A, Strober S. 2000. Cutting edge: a role for CD1 in the pathogenesis of lupus in NZB/NZW mice. J. Immunol. 164:5000–4 Chan OT, Paliwal V, McNiff JM, Park SH, Bendelac A, Shlomchik MJ. 2001. Deficiency in beta2-microglobulin, but not CD1, accelerates spontaneous lupus skin disease while inhibiting nephritis in MRL-Fas(lpr) mice: an example of disease regulation at the organ level. J. Immunol. 167:2985–90 Yang JQ, Singh AK, Wilson MT, Satoh M, Stanic AK, et al. 2003. Immunoregulatory role of CD1d in the hydrocarbon oil-induced model of lupus nephritis. J. Immunol. 171:2142–53 Sieling PA, Porcelli SA, Duong BT, Spada F, Bloom BR, et al. 2000. Human double-negative T cells in systemic lupus erythematosus provide help for IgG and are restricted by CD1c. J. Immunol. 165:5338–44 Heller F, Fuss IJ, Nieuwenhuis EE, Blumberg RS, Strober W. 2002. Oxazolone colitis, a TH2 colitis model resembling ulcerative colitis, is mediated by IL-13-producing NK-T cells. Immunity 17:629–38 Saubermann LJ, Beck P, De Jong YP, Pitman RS, Ryan MS, et al. 2000. Activation of natural killer T cells by alphagalactosylceramide in the presence of CD1d provides protection against colitis in mice. Gastroenterology 119:119–28 Mizoguchi A, Mizoguchi E, Takedatsu H, Blumberg RS, Bhan AK. 2002.

410.

411.

412.

413.

414.

415.

416.

417.

P1: IKH

889

Chronic intestinal inflammatory condition generates IL-10-producing regulatory B cell subset characterized by CD1d upregulation. Immunity 16:219–30 Takeda K, Hayakawa Y, Van Kaer L, Matsuda H, Yagita H, Okumura K. 2000. Critical contribution of liver natural killer T cells to a murine model of hepatitis. Proc. Natl. Acad. Sci. USA 97:5498–503 Kaneko Y, Harada M, Kawano T, Yamashita M, Shibata Y, et al. 2000. Augmentation of Valpha14 NKT cellmediated cytotoxicity by interleukin 4 in an autocrine mechanism resulting in the development of concanavalin A-induced hepatitis. J. Exp. Med. 191:105–14 Yoshimoto T, Bendelac A, Watson C, Hu-Li J, Paul WE. 1995. Role of NK1.1+ T cells in a TH2 response and in immunoglobulin E production. Science 270:1845–47 Smiley ST, Kaplan MH, Grusby MJ. 1997. Immunoglobulin E production in the absence of interleukin-4-secreting CD1-dependent cells. Science 275:977– 79 Chen YH, Chiu NM, Mandal M, Wang N, Wang CR. 1997. Impaired NK1+ T cell development and early IL-4 production in CD1-deficient mice. Immunity 6:459–67 Cui J, Watanabe N, Kawano T, Yamashita M, Kamata T, et al. 1999. Inhibition of T helper cell type 2 cell differentiation and immunoglobulin E response by ligand-activated Valpha14 natural killer T cells. J. Exp. Med. 190:783–92 Brown DR, Fowell DJ, Corry DB, Wynn TA, Moskowitz NH, et al. 1996. Beta 2-microglobulin-dependent NK1.1+ T cells are not essential for T helper cell 2 immune responses. J. Exp. Med. 184:1295–304 Akbari O, Stock P, Meyer E, Kronenberg M, Sidobre S, et al. 2003. Essential role of NKT cells producing IL-4 and IL-13 in the development of allergen-induced

13 Feb 2004

18:42

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

890

AR

BRIGL

AR210-IY22-28.tex

¥

AR210-IY22-28.sgm

LaTeX2e(2002/01/18)

P1: IKH

BRENNER

airway hyperreactivity. Nat. Med. 9:582– 88 418. Lisbonne M, Diem S, de Castro Keller A, Lefort J, Araujo LM, et al. 2003. Cutting edge: invariant Valpha14 NKT cells are required for allergen-induced airway inflammation and hyperreactivity in an experimental asthma model. J. Immunol. 171:1637–41 419. Korsgren M, Persson CG, Sundler F, Bjerke T, Hansson T, et al. 1999. Natural killer cells determine development of allergen-induced eosinophilic airway inflammation in mice. J. Exp. Med. 189:553–62 419a. Tsuji RF, Szczepanik M, Kawikova I, Paliwal V, Campos RA, et al. 2002. B cell-dependent T cell responses: IgM an-

tibodies are required to elicit contact sensitivity. J. Exp. Med. 196:1277–90 419b. Campos RA, Szczepanik M, Itakura A, Akahira-Azuma M, Sidobre S, et al. 2003. Cutaneous immunization rapidly activates liver invariant V alpha 14 NKT cells stimulating B-1 B cells to initiate T cell recruitment for elicitation of contact sensitivity. J. Exp. Med. 198: 1785–96 420. Garcia KC, Degano M, Pease LR, Huang M, Peterson PA, et al. 1998. Structural basis of plasticity in T cell receptor recognition of a self peptide-MHC antigen. Science 279:1166–72 421. Wilson IA. 1996. Another twist to MHCpeptide recognition. Science 272:973– 74

3/12/2004 2:18 PM

FUNCTIONS OF CD1 MOLECULES

Figure continued on next page

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Brigl.qxd Page 1

C-1

C-2

3/12/2004

BRIGL

2:18 PM

■

Figure 3 CD1 atomic structures. (a) Front and (b) top view of the CD1a-sulfatide and CD1b-phosphatidylinositol (PI) complexes and the mouse CD1d (mCD1d) structure without ligand. α1-α3 domains and β2-microglobulin (β2M) are shown in light gray ribbons, α1 and α2 helices in brown; yellow, carbon atoms of the ligands; red, oxygen; green, sulfur; blue, nitrogen; purple, phosphate; and cyan, detergent molecules. (c) Comparison of the antigen-binding grooves of CD1a, CD1b, and mCD1d. Molecular surfaces are shown as transparent binding pockets with or without bound ligand from a view directly into the groove, similar to (b). The A9 and F9 pockets, the C9 channel, and the T9 tunnel are labeled accordingly. Important protein residues are shown in green. (d) Space filling representations of the view as shown in (b) for CD1a and CD1b with bound ligand in comparison to MHC class I (H-2Kb) and MHC class II (I-Ak) with bound peptide. Atoms are shown as spheres. Gray, protein residues. Figures were prepared using Molscript, GRASP, and Raster3D software with the coordinates of CD1a (51) (1ONQ), CD1b (50) (1GZQ), mCD1d (49) (1CD1), H-2Kb (420) (2CKB), and I-Ak (421) (1D9K).

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Brigl.qxd Page 2

BRENNER

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:817-890. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104543

Annu. Rev. Immunol. 2004. 22:891–928 doi: 10.1146/annurev.immunol.22.012703.104543 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on December 16, 2003

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: Basic Chemokinese Grammar Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

for Immune Cells Antal Rot1 and Ulrich H. von Andrian2 1

Novartis Institute for Biomedical Research, Vienna, A-1235 Austria; email: [email protected] 2 CBR Institute for Biomedical Research and Department of Pathology, Harvard Medical School, Boston, Massachusetts 02115; email: [email protected]

Key Words chemokine receptors, chemotaxis, endothelial cells, lymphocyte trafficking ■ Abstract Chemokines compose a sophisticated communication system used by all our cell types, including immune cells. Chemokine messages are decoded by specific receptors that initiate signal transduction events leading to a multitude of cellular responses, leukocyte chemotaxis and adhesion in particular. Critical determinants of the in vivo activities of chemokines in the immune system include their presentation by endothelial cells and extracellular matrix molecules, as well as their cellular uptake via “silent” chemokine receptors (interceptors) leading either to their transcytosis or to degradation. These regulatory mechanisms of chemokine histotopography, as well as the promiscuous and overlapping receptor specificities of inflammation-induced chemokines, shape innate responses to infections and tissue damage. Conversely, the specific patterns of homeostatic chemokines, where each chemokine is perceived by a single receptor, are charting lymphocyte navigation routes for immune surveillance. This review presents our current understanding of the mechanisms that regulate the cellular perception and pathophysiologic meaning of chemokines.

INTRODUCTION The immune system consists of billions of motile cells that roam at large in the body. In need of communication signals to navigate these cells along their convoluted pathways in time and space, to orchestrate their interactions and enable them to form architecturally complex organs of their own, the immune system turned to chemokines. Chemokines are building blocks of the most versatile, coherently functioning system of intercellular communication signals. Chemokine messages are decoded through specific cell-surface receptors. Upon chemokine

0732-0582/04/0423-0891$14.00

891

15 Mar 2004

16:7

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

892

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

binding, these receptors unleash cascades of intracellular secondary mediators that turn on cell-specific intrinsic functional programs, including directional migration, the best-known effect of chemokines. The sum of individual “cell reflexes” results in complex system responses during both steady-state conditions and innate and adaptive immune reactions. Like any other communication system, the chemokine network bears several noteworthy attributes of a language, which we call here “Chemokinese.” Like human languages, which follow general rules of grammar, Chemokinese also has rules. In this review, we describe the basic grammar of Chemokinese and how it applies to cell communication in the process of inflammation and immune responses.

CHEMOKINESE ALPHABET Chemokines are defined independently of their function, based on their amino acid composition, specifically on the presence of a conserved tetra-cysteine motif (1–3). The relative position of the first two consensus cysteines (either separated by a non-conserved amino acid or next to each other) provides the basis for division of chemokines into the two major subclasses, CXC and CC chemokines, respectively (Figure 1). Three homologous molecules are also regarded as chemokines. These are CX3CL1, with three intervening amino acids between the first cysteines, and XCL1 and XCL2, which lack two out of four canonical cysteines. To date, the official nomenclature accounts for 43 human chemokines (4, 5) (Figure 1). However, isoforms and polymorphisms, splice variants and enzymatically processed forms, as well as chemokines encoded in viral genomes enlarge substantially the number of distinct chemokine molecules that can naturally affect human cells. Additional rare chemokines may be stashed away in the human genome waiting to be discovered. But even at their presently known diversity, chemokines, through almost infinite combinatorial possibilities of their simultaneous or sequential appearance, can convey the most complex cellular messages. Chemokines carry encrypted messages to cells that have the means to decode them, and hence are operating as a communication medium, a cell language. Individual chemokines may be compared to distinct ideograms, symbols representing a particular idea, word, or morpheme as, for example, those in Sumerian cuneiform scripts, Egyptian hieroglyphs, or some characters in Chinese and Japanese languages. Chemokinese dispatches are prompted by diverse stimuli from many different cell types. Perceived and decoded, these messages can influence the most profound aspects of a cell’s life, including not only eponymous chemotaxis and adhesion but also proliferation, maturation, differentiation, apoptosis, malignant transformation, and dissemination (6). Chemokines on the whole can affect any cell type. However, notable specificity patterns of responsiveness to different chemokines have been identified, determined by the expression of the apposite chemokine receptors.

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

P1: IKH

893

HOW CELLS GRASP CHEMOKINESE

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Chemokine Receptors The recognition of chemokine-encoded messages is mediated by specific cellsurface G-protein-coupled receptors (GPCRs) with seven transmembrane domains. The human chemokine receptor system at present consists of 19 different GPCRs (7) (Figure 1). The GPCR family is the most diverse class of cell-surface receptors. Specific GPCRs perceive signals from ligands as varied as hormones; biogenic amines; neuropeptides; neurotransmitters; classical chemoattractants; lipid mediators; pheromones; sweet, bitter, and umami tasting molecules; odorants; and light photons. Chemokine GPCRs carry out three fundamental parts of the chemokine-induced “cellular reflex”: message acquisition, semantic extraction (decoding), and initiation of cell responses. These three functions are not interdependent. In this respect, Chemokinese is similar to human languages where word perception, interpretation of meaning, and induction of a response (or lack thereof) can also be separated. The message acquisition, i.e., chemokine binding to a particular receptor, is determined by receptor affinity, which varies greatly between ligands and does not necessarily translate into functional potency (8, 9). The meaning of a chemokine is influenced by its position on the agonist-antagonist scale for a particular receptor. The ability either to shift the receptor into its “productive” signaling conformation or to prevent such a shift determines a chemokine’s agonist and antagonist potency, respectively (10, 11). In addition, a chemokine’s meaning depends on the magnitude of the elicited response (efficacy) and ability to induce various regulatory and signal transduction pathways, which determine the actual response. The efficacy and the quality of the response also depend on the chemokine-induced shift in receptor configuration, which, however, may be at variance with that determining chemokine potency (10). For the same receptor the elicited functions may differ, depending on the ligand (11) as well as the microenvironment, including available G-protein isotypes, signal regulatory pathways, etc. It will take some time before the molecular interactions responsible for matching 43 chemokines and 19 receptors will become clear because these interactions are likely to vary for each receptor and ligand. Also, the determinants of agonism and antagonism are different for each chemokine and its receptor(s). The multiple signaling pathways (12) and cell responses induced by chemokines vary, too, as they depend on the responding cells and a multitude of other factors. However, there is one response, chemotaxis, that is elicited by (almost) all chemokines in (almost) all receptor-bearing cells. Experimental work of the past few years and analogies with the mode of action of other GPCRs, especially those for classical chemoattractants, have identified many of the molecular events that may lead from chemokine GPCRs to cell locomotion.

15 Mar 2004

16:9

894

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

From GPCR to Chemotaxis Leukocytes stimulated by chemoattractants assume a polarized morphology (13). They become elongated and develop a wide lammellipod (pseudopod) at the leading edge and a tail-like projection at the trailing end (uropod). The generation of filamentous actin at the pseudopod leads to its extension, which together with uropod retraction drives cell locomotion (13, 14). However, polar leukocyte morphology is not a response to a chemotactic gradient, as it can be observed during spontaneous random walk and chemokinesis, agonist-stimulated migration that lacks directionality (15). Chemotaxis combines enhanced-rate locomotion with a gradient-imposed, unidirectional character of response. When the gradient is reversed during chemotaxis, leukocytes either retract the pseudopod and form a new one at the tail end (16) or, more often, make a U-turn while maintaining their original pseudopod (15, 17). This suggests that in moving leukocytes the pseudopod is specialized for gradient perception. The relative increase in concentration of some chemokine receptors at the leading edge of moving leukocytes (18) may be the first step in the generation of intracellular mediator cascades at the pseudopod. Chemokine receptors function as allosteric molecular relays where chemokine binding to the extracellular portions modifies the tertiary structure of the receptor, allowing the intracellular part to bind and activate heterotrimeric G-proteins. In response, the activated G-proteins exchange guanosine diphosphate for guanosine triphosphate (GTP) and dissociate into α- and βγ -subunits. Chemokine receptors can couple to several different Gα isotypes (19, 20). The βγ -subunits that dissociate from the pertussis toxin-sensitive Giα mediate chemokine-induced signals (21, 22) by activating the phosphoinositide-3 kinases (PI3K) (Figure 2), which lead to generation of phosphotidylinositol-3,4,5-trisphosphate (PIP3). The importance of this reaction is evidenced by the deficient migration of myeloid leukocytes to chemokines in mice lacking PI3Kγ (23). Lymphocyte chemotaxis is not severely compromised in PI3Kγ -deficient mice (23). This implies that either different signal transduction mechanisms prevail in chemokine-induced lymphocyte migration or other PI3K isotypes, recently suggested to amplify PI3Kγ mediated reactions, could play a more important role in lymphocytes than in myeloid cells (24, 25). PI3K and its product PIP3 translocate to the pseudopod where they colocalize with the small GTPase Rac (26, 27). PIP3 activates Rac through specific guanine nucleotide exchange factors (GEFs) (28, 29). In addition, PIP3 serves as a docking site for protein kinase B, which also translocates to the pseudopod and can induce actin polymerization (30). Rac is indispensable for leukocyte migration (31). Its downstream effectors include p21-activated kinase (PAK) and the Wiskott-Aldrich Syndrome protein (WASp) homologue, WAVE, which stimulate actin-related protein (Arp) 2/3. This induces focal actin polymerization, responsible for the development and forward extension of the pseudopod. When exogenous PIP3 is introduced into a cell, it induces endogenous PIP3, suggesting a self-amplification loop that establishes intracellular gradients of

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CHEMOKINES IN IMMUNITY

P1: IKH

895

effectors in the pseudopod (32). This also implies that the PIP3-induced activation of Rac can signal back to one of the PI3K family members, possibly PI3Kδ (24). The events that lead from GPCR signal to Rac-initiated actin polymerization explain how cells that read Chemokinese gradients can migrate in response. However, reading and understanding are not the same. The signals centering on Rac are sufficient to activate the actin-based engine that propels the cells, but they do not contribute to the “comprehension” of a gradient’s direction and, therefore, cannot support the unidirectional bias of locomotion, the hallmark of chemotaxis. The latter requires another Rho GTPase, Cdc42. Without Cdc42, leukocytes exhibit a random walk, rather than directed migration, when placed in a chemotactic gradient (33–37). Curiously, the target of Cdc42, PAK1, acts upstream of it. Activated directly by Gβγ , PAK1 binds Pixα, a GEF that is specific for Cdc42, to form a complex that recruits and activates Cdc42 (38). This signaling complex is responsible for F-actin nucleation via activated Arp 2/3 and exclusion of PI3-phosphatases, the negative regulators of PIP3, from the leading edge of the cell (38, 39). Gradient sensing is also dependent on functional PI3K, although it is not entirely clear how (38). The mechanism that couples Cdc42-mediated cell orientation and steering to the Rac-driven actin motor has not been determined. Additionally, the events that lead to Cdc42- and Rac-mediated actin polymerization (Figure 2) explain only what happens in the pseudopod, i.e., the front part of cells, following the beckoning of chemokines.

Dual Signaling Pathways in Moving Leukocytes Formyl peptide induces the activation of different sets of mediators at the pseudopod and the uropod of moving neutrophils (40). In parallel to the sequence leading to Rac activation at the pseudopod, another pathway leads through pertussis toxin insensitive Gα 12/13 to the activation of Rho by Rho GEFs at the uropod (40). Activated Rho, through its effector p160ROCK, a serine/threonine kinase, induces focal activation of myosin II, formation and contraction of actin-myosin complexes, thereby leading to retraction of the uropod (41, 42). In addition, Rho prevents the formation of lamellipodia at the trailing end (43). These two alternative signaling pathways involving either Rac or Rho are mutually inhibitory, which explains why they are spatially confined to the pseudopod and uropod, respectively. Together they induce and maintain functional and morphological cell polarity and drive locomotion (40). The equilibrium between GPCR signaling pathways through Gi and Gα12/13 may also control the balance between cellular movement toward or away from the source of a gradient. For example, sphingosine-1-phosphate (S1P) can induce either chemoattraction or repulsion depending on the receptor it is using. S1P3 (Edg3) is an “attractant” receptor, which upon stimulation with S1P gives rise to both Gi/Rac and Gα12/13/Rho activation pathways. It can be converted to a “repellant” receptor by simply blocking the Gi/Rac pathway with pertussis toxin (44). Determining whether these findings also apply to chemokine-driven leukocyte

12 Feb 2004

16:20

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

896

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

migration is important. Leukocytes do not bear S1P3, and whether chemokinetriggered GPCRs are coupled to the Rho-activating G12/13 or Gq/11 pathways is still unknown (45). Nevertheless, it is attractive to speculate that similar mechanisms may influence chemokine-induced leukocyte migration, at least in some settings. For example, CXCL12 and CCL26 can induce chemorepulsion through CXCR4 and CCR2, respectively (46, 47). Chemokine-driven migration away from the source of a chemokine may promote leukocyte departure from tissues, as was suggested for CXCL12-mediated expulsion of newly generated T cells from the thymus (48) and prevention of T-cell entry into bone marrow (46). By analogy with the paradigm observed for S1P3, it is possible that the ability of CXCL12 to induce either chemoattraction or chemorepulsion of T cells is determined by the balance between alternative signal transduction pathways. The intracellular events described here, which lead from GPCRs through PI3K to small GTPases, appear to be necessary and sufficient to mediate chemokineinduced leukocyte chemotaxis at least in some leukocyte types. By contrast, the chemokine-induced intracellular signals leading to rapid integrin activation, which enables firm leukocyte–endothelial cell (EC) adhesion, are not yet fully characterized. Potential effectors include Rho (49) and the small GTPase Rap1 (50–52). In addition, chemokine GPCRs activate many other mediators and pathways that induce other cell responses including exocytosis, proliferation, gene transcription, and apoptosis. Inevitably, these mediators, through known or still-to-be-discovered pathways and feedbacks of tightly woven intracellular signaling networks, may also influence cell migration. The apparent ability of chemokine receptors to dimerize and heterodimerize introduces a further dimension of complexity into chemokine-induced signaling (53, 54). For more details on chemokine signaling, we refer the reader to recent reviews on this subject (36, 55–57).

Shutting Off the GPCR Signal Chemokine receptor signaling is transient because the mechanism of activation contains a built-in shut-off sequence. The Gα subunits have intrinsic GTPase activity to hydrolyze GTP and reunite with Gβγ to return to the initial conformation of inactive heterotrimers. Molecules called regulators of G-protein signaling (RGS) can modify the GTPase activity of the Gα subunits (58, 59). RGS act bimodally, enhancing the GTPase activity of some Gα isotypes while inhibiting it in others (60), and show specificity for individual chemokine receptors (61, 62). Two additional mechanisms of chemokine receptor silencing include (a) desensitization, caused by steric hindrance of G-protein activation due to receptor phosphorylation by G-protein-coupled receptor kinases (GRK), and (b) downregulation caused by βarrestin- or adaptin-2-mediated receptor sequestration and internalization through clathrin-coated pits or caveolae (63–65). Desensitization and internalization are regulated separately and require the phosphorylation of different serine residues in the C-terminal tail of chemokine receptors (66). Internalization, apparently, can even be uncoupled from signaling through Gαi (67). However, as suggested by

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

P1: IKH

897

the lack of CXCL12-induced lymphocyte chemotaxis in β2-arrestin and GRK6deficient mice, these molecules may also play yet unknown positive regulatory roles in chemokine signaling (68). The pathways leading to either degradation or recycling of chemokine receptors following their internalization are determined by the guardians of all vesicular machinations, Rab GTPases (69), which also influence the surface reexpression of internalized chemokine receptors (67, 70).

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Alternative Signaling Mechanisms Intriguingly, the cell nucleus may be an additional site for chemokine signaling. Specific nuclear localization leading to changes in morphological and functional cell parameters has been observed with the natural nonsecreted variant of CCL27 called PESKY (71). The secreted CCL27, as well as other chemokines, may be targeted to the nucleus, following internalization through their GPCRs (71, 72). It is not clear what the molecular nature of putative nuclear chemokine receptors is and too early to judge how important their function may be. Sulfated sugars of the glycosaminoglycan (GAG) family, such as heparan sulfate and chondroitin sulfate, bind chemokines and in response can mediate outside-in signaling through the proteins they are decorating. For example, CCL5 signals in HeLa-CD4 cells that express GAGs and are completely devoid of GPCRs, but does not signal in mutant cells that lack GAGs (73). This GPCR-independent signal is mediated by GAG-decorated CD44 (74). CXCL4, in addition to using CXCR3B (75), may also signal through chondroitin sulfate–decorated membrane molecules on neutrophils, monocytes, and lymphocytes (76). GAGs on a target cell membrane may also influence chemokine signals by another mechanism. GAGbound chemokines, owing either to their presentation to GPCRs in cis-geometry or to GAG binding-induced changes in their secondary structure (77), may gain enhanced agonistic properties (78, 79). However, the best known and arguably most important role of GAGs is their presentation of chemokines in trans-geometry.

WRITTEN CHEMOKINESE Notes on Sugar To carry their message to remote target cells, chemokines have to diffuse away from their source, hence remain soluble. However, in vivo chemokines interact with GAGs, which can limit their dissemination. GAGs decorate proteins in the extracellular matrix and on cell surfaces and not only interfere with chemokine diffusion, but also provide highly specific substrates onto which Chemokinese messages are “written.” All chemokines bind GAGs, yet they show remarkable disparity in affinities for various GAGs and vice versa (80, 81). This is owing, in part, to the diverse functional domains that different chemokines use for GAG binding (82). For example, CXCL8 uses its C-terminal α-helix to bind heparan sulfate (83), CCL3—three noncontiguous basic amino acids at positions 18, 46,

12 Feb 2004

16:20

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

898

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

and 48 (84)—and CCL5—two amino acid clusters, 44–47 and 55–59 (85). This means that chemokines not only encode distinctive meanings perceived through their cognate receptors, but also contain within their structure clues as to where the message should be posted. Written Chemokinese is read by contact, like the Braille language of the blind, whereby leukocytes adhere to surfaces onto which chemokines are immobilized. Chemokines written on GAGs are not like words written in stone, impressed into clay or printed on paper. Because of the changing equilibria of chemokine production, diffusion, GAG-immobilization, mobilization, and consumption, Chemokinese writing is similar to the way text changes during electronic word processing, where words on the computer screens are written, rearranged, deleted, and written again. It has not been determined if GAG-bound chemokines can signal as complexes through chemokine GPCRs or if chemokines must first dissociate from GAGs to interact with their receptors. GAG-immobilized chemokines may be deposited in domains directly accessible to leukocytes or in concealed microenvironments. Therefore, the regulation of in vivo activity of chemokines may take place at the level of their GAG immobilization and mobilization. For example, the digestion of GAG-bearing molecules by specific enzymes may either downregulate chemokine activity as a result of chemokine removal from microanatomical sites that are “visible” to leukocytes (86) or expose the previously concealed activity by releasing chemokines from sequestered stores (87). Also, chemokine immobilization may be influenced by GAG synthesis and sulfation (88) and by soluble GAGs (89, 90). Complexing with GAGs may serve to neutralize chemokines stored in T-cell granules (91). Other cell types can also store and release chemokines, demonstrating that the information storage function of written languages is also a feature of Chemokinese. Intracellular chemokine depots serve as small souvenirs from the past or crib notes for the future. For example, following an inflammatory insult, ECs store some of the chemokines they produce in Weibel-Palade bodies (92). The stored chemokines can be released instantaneously upon repeated inflammatory stimulation. Neutrophils, eosinophils, cytotoxic T cells, and mast cells carry stored chemokine in case they have to communicate in a hurry (91, 93, 94). Because they store chemokines, including CXCL1, CXCL4, CXCL5, CXCL7, CXCL8, CCL1, CCL5, and CCL7, platelets, handicapped by their inability to synthesize proteins de novo, are by no means inarticulate (95, 96). Chemokines not only can be immobilized by GAGs, but also can attach to other natural and artificial substrates, including polycarbonate membranes used in Boyden-type chambers or transwells. This allows chemokines to induce haptotaxis, directed migration to substrate gradients, the mechanism responsible for in vitro leukocyte migration to CXCL8 and CCL5 (97, 98). In the context of migration induced by immobilized chemokines, the term haptotaxis is used with considerable corruption of the original denotation, which stands for movement along gradients of adhesion molecules (99). With the exception of CX3CL1 and CXCL16, chemokines are not adhesive per se, but they induce leukocyte adhesion

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

P1: IKH

899

by activating integrins (100), which allows chemokines to operate at the blood-EC interface and induce leukocyte adhesion and emigration.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Writing Is on the (Vessel) Wall Chemokine immobilization by GAGs on the luminal surface of ECs is a prerequisite for optimal progression of leukocyte-EC adhesion (101, 102) during both chemokine-induced inflammatory recruitment and constitutive homing of leukocytes to tissues. The contribution of GAG binding to the in vivo proemigratory activity of chemokines has been formally demonstrated first for CXCL8 (103) and recently also for CCL2, CCL4, and CCL5 (104). Leukocytes adhere to the endothelium stepwise, in a process that takes place in the following sequence only. First, leukocytes establish a loose tethering interaction with ECs. If this interaction persists, owing to the shear stress exerted by the bloodstream, it manifests as leukocyte rolling. Tethering and rolling are mediated by selectins and their ligands (105, 105a) as well as by integrins of the α 4 subfamily (106, 107). Rolling leukocytes become activated by EC-bound chemokines that induce a rapid increase in the affinity of α 4 and/or β 2 integrins (108–110). Next, adherent leukocytes spread and transmigrate across the EC barrier in a process that, at least in vitro, can be induced by apically immobilized chemokines and the presence of fluid shear stress (111– 113). Thus, a chemokine immobilized on the apical EC surface, but not soluble chemokines, may accomplish (a) stimulation of leukocyte tethering and rolling, (b) induction of firm adhesion, and (c) initiation of lymphocyte transmigration (111, 112). Other chemokines may act sequentially and divide the roles in induction of firm adhesion and transmigration (114) or use two alternative receptors for these two steps, as apparently happens with CCL5, which signals to monocytes through CCR1 and CCR5 to induce adhesion and transmigration, respectively (115). Irrespective of its exact function, the key aspect of GAG-mediated chemokine presentation on the EC surface is to provide a mechanism of directing the chemokine signal to leukocytes that have already initiated interactions with the EC and away from freely circulating leukocytes. The latter, when stimulated by chemokines, lose their ability to tether and emigrate (116, 117). Thus, an immobilized chemokine on ECs induces leukocyte adhesion and emigration, whereas a blood-borne chemokine functions as an adhesion inhibitor. There are indications that specific mechanisms select tissue chemokines to appear on the luminal EC surface. For example, of the two CCR4 ligands capable of in vivo T-cell recruitment into the skin, CCL17, but not the more potent CCL22, preferentially appears on the EC surface (118, 119).

Editor of Chemokinese Graffiti Tissue-derived chemokines can cross the EC barrier passively through intercellular junctions. In addition, venular ECs can internalize and transport chemokines in an abluminal-to-luminal direction (103). These two alternative routes are mutually nonexclusive. However, the pericellular route may bypass the EC membrane

15 Mar 2004

16:9

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

900

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

domains capable of efficient chemokine presentation (Figure 3). The chemokine specificity fingerprint of in situ binding to venular ECs (120) suggests that Duffy antigen, which is expressed by these cells (121), may contribute to chemokine transport by ECs (103, 120). Duffy antigen (also known as Duffy antigen receptor for chemokines, or DARC), originally described on erythrocytes, is a serpentine seven-spanner that lacks a critical tripeptide DRY motif that is required for GPCR function (122). Accordingly, Duffy antigen does not signal when it binds one of its numerous ligands of the CC and CXC families. Duffy antigen was recently grouped with D6, another nonsignaling seven-spanner that binds multiple CC chemokines (123). Their suggested name, “interceptors,” [which stands for chemokine internalizing (pseudo)receptors, abbreviated CIPR-1 and CIPR-2, respectively], reflects their main function in nucleated cells (124). As Duffy antigen is expressed on both erythrocytes and ECs, the initial experimental interrogation of the Duffy knockouts revealed no specific clues regarding its function on ECs. Luo et al. (125) described a reduction of lipopolysaccharide (LPS)-induced neutrophil infiltration into the peritoneal cavity, lungs, and gut of Duffy antigen–knockout mice. Conversely, Dawson et al. (126) observed an increase in neutrophils in the lungs and hepatic sinusoids after LPS application. Curiously, these conflicting findings may both result from lack of Duffy antigen on erythrocytes. Apparently, in addition to its accepted function as a chemokine sink, Duffy antigen may serve also as a chemokine reservoir. Injected chemokines disappear from plasma of Duffy antigen–deficient mice, whereas wild types maintain their chemokine plasma levels much longer (124, 127). Consequently, Duffy antigen on erythrocytes may act as a chemokine buffer, reducing in some settings the plasma levels of chemokines while maintaining them in others, thus indirectly influencing leukocyte emigration. However, Duffy antigen expressed by ECs may also directly impact chemokineinduced leukocyte emigration. Chemokine binding and uptake by venular ECs is a function of their Duffy antigen expression (124). Duffy antigen supports transport of chemokines by ECs and epithelial cells in vitro, leading to enhanced leukocyte transmigration across the barriers formed by these cells (124, 128). Importantly, in contrast to the chemokine-scavenging function proposed for D6 (see below), chemokine internalization by Duffy antigen does not result in their degradation (124). Finally, in mice that lack Duffy antigen on ECs, but not on erythrocytes, neutrophils exhibit diminished emigration to suboptimal concentrations of chemokines (124). Cumulatively, these data suggest that by selecting chemokines for EC internalization and transcytosis, Duffy antigen may serve as an editor of Chemokinese messages destined to appear on the luminal surface. It is not clear if endothelial Duffy antigen is responsible for these functions alone or in cooperation with heparan sulfate. The latter possibility is supported by reduced CXCL8 association with ECs in heparatinase-treated human skin (103). There may also be other yet unknown interceptors involved in EC transcytosis of chemokines. For example, CCL19, which does not bind to Duffy antigen, is effectively transcytosed and luminally presented in high endothelial venules (HEV) (129).

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

P1: IKH

901

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Chemokinese Telecommunication Channels and Their Censor In addition to distribution within their tissue of origin, chemokines can disseminate via preformed channels, e.g., afferent lymphatics into the draining lymph nodes (LN). Here they are channeled by conduit structures (130) to HEV, transcytosed, and presented on the luminal surface to induce leukocyte recruitment (Figure 3). This mechanism was demonstrated for CCL2 produced in skin during inflammation as well as for several other chemokines injected into skin (131, 132). The purpose of such a long-distance broadcast of formulated Chemokinese messages is possibly to boost in the LNs the number of cells involved in innate pathogen containment (see below). However, excessive transmission of inflammatory chemokines may lead to uncontrolled recruitment of potentially histotoxic leukocytes. A possible safeguard against this may be the chemokine interceptor D6 (CIPR-2), which is expressed by lymphatic ECs (133). As D6 internalizes many CC chemokines and presumably targets them for lysosomal degradation (134), it may function at this microanatomical location as a censor of the Chemokinese broadcast to the LNs. This allows only selected chemokines from the periphery to reach LNs unaltered, whereas others may arrive only after mitigation by D6. Additionally, D6 may serve as a gatekeeper of leukocyte trafficking through the lymphatics by modifying local chemokine gradients (124).

CHEMOKINES IN INNATE IMMUNITY Few functions of Chemokinese are as self-evidently important as its role in orchestrating leukocyte responses during pathogen containment. Infectious microorganisms can directly stimulate chemokine production by tissue dendritic cells (DCs) and macrophages as well as by many parenchymal and stromal cells. Conserved microbial pathogen-associated molecular patterns induce chemokines through pattern recognition receptors, such as Toll-like receptors, or the nucleotide-binding site leucine-rich repeat proteins NOD1 and NOD2 (135, 136). Endogenous molecules associated with infection and injury, e.g., fibrinogen, elastase, and defensins, can also signal through pattern recognition receptors to induce chemokine production (137–139). However, classically the major inflammatory and immunomodulatory cytokines such as IL-1, TNF-α, IFNγ , IL-4, IL-5, IL-6, IL-13, and IL-17, induced in injury or infection, stimulate through their respective receptors the production of many different chemokines (1–3). Why do these eloquent exogenous and endogenous alarm signals need to be paraphrased in Chemokinese? First, such translation amalgamates different signals into one coherent and universally understood message. Also, the translating cells may wield their interpretational bias and provide chemokine-encoded clues to reveal themselves and their anatomical residence (140, 141). Chemokines produced in response to pathogens typically are ligands of CCR1, CCR2, CCR3, CCR5, CXCR2, and CXCR3. All these receptors engage multiple chemokines, which may be considered “synonymous” insofar as the message they convey through their promiscuous

12 Feb 2004

16:20

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

902

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

receptors is similar. In addition, many of these chemokines can signal through several different receptors (Figure 1) and therefore are comparable to homonyms, words with multiple meanings. Synonymy and homonymy of chemokines have euphemistically been branded “chemokine redundancy.” However, chemokines and their receptors are not truly redundant. Few chemokines carry identical meanings or are fully interchangeable. This is because of their unique homonymy profiles (as illustrated by the color bar code in Figure 1) and because of different agonistic properties synonymous chemokines have on the same receptor, i.e., potency, efficacy, and ability to induce receptor desensitization and internalization. What is the reason for chemokine synonymy and homonymy, and what function may such overlapping specificities carry in host defense? We attempt to answer these questions by comparing the chemokine-driven “cell reflex” in response to infection with the sensory functions of the CNS.

Sensory and Effector Roles of Chemokinese Evolution selected GPCR-based systems to serve three of our sensory functions. Vision, taste, and smell all rely on GPCRs. In case of olphactory detection, single receptors can engage several different odorants, and a single odorant can bind several different receptors. By combinatorial engagement of approximately 400 different odorant receptors, humans can distinguish several thousand different odors (142). Similarly, chemokine receptors may fill a sensory function for the immune system. Different pathogens lead to the production of characteristic fingerprints of chemokines (143), which are projected onto the luminal surface of venular ECs where the pathogen is “described” in Chemokinese to the patroling leukocytes. These telling abstractions of a virus, bacterium, fungus, protozoa, or helmith are read through the GPCRs, which immediately trigger the “cell reflex,” leukocyte adhesion, and emigration. The overlapping ligand and receptor specificities within the chemokine system may not only serve the robustness of the leukocyte response (144), but also, by analogy with olfactory perception, be required for increasing the combinatorial finesse of pathogen portrayal and generation of diversity in the leukocyte response to it. Further diversity may be provided by classical chemoattractants and migration modifiers derived from other endogenous mediator systems (e.g., activated complement and pro- or anti-inflammatory lipids) or directly from pathogens. The leukocytes recruited by this cocktail of agonists divide among them the tasks of pathogen containment by different microbicidal means (143): amplification and fine-tuning of leukocyte recruitment by producing or inducing further chemokines (145); pathogen acquisition, processing, and initiation of the most pertinent adaptive immune response (146); and setting the stage for tissue repair. As reviewed recently (143, 147), mice that are deficient for different chemokines and chemokine receptors may have compromised innate immunity to pathogens. Conversely, animals overexpressing chemokines may, in some settings, be protected from infection (148). There are many fine nuances in the assortment of leukocytes that become mobilized in response to different pathogens (143), but with considerable

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CHEMOKINES IN IMMUNITY

P1: IKH

903

oversimplification the reactions can be categorized as either type 1 or type 2, depending on the cytokines and chemokines produced and the involvement of polar sets of effector lymphocytes and leukocytes. CCR1, CCR5, CXCR3, and CXCR6 and their ligands are associated with type 1 responses, whereas CCR3, CCR4, and CCR8 and their ligands are associated with type 2 responses. These dichotomous chemokine pathways are also reflected in two main routes of pathogenesis in chronic inflammatory diseases. The discovery of Toll-like receptors made it clear how microorganisms can directly induce type 1 chemokines (135). Recent results showed that innate type 2 responses are also directly stimulated by pathogens (149). Thus, the amount and composition of chemokines that are generated by infected individuals in response to certain pathogens can determine the type of innate immune response. This may happen, at least in some settings, without the involvement of adaptive immunity, which, however, may further amplify the preexisting type 1 or type 2 bias. This differentiation can also be influenced by chemokines. The presence of CCL3 or CCL2 at the time of antigen challenge of naive T cells induces Th1 and Th2 differentiation, respectively (150). Such chemokine influence was confirmed in CCL2-deficient mice, which have reduced Th2 responses (151). Conversely, mice deficient for CCR2, the only known CCL2 receptor, have impaired Th1 responses (152), owing to reduced monocytes trafficking to sites of inflammation (153). How this apparent discrepancy between the Th1/Th2 profiles in CCL2 and CCR2 knockout mice may be reconciled and several alternative hypothetical mechanisms of chemokines determining the T helper bias are discussed in a recent review (154). Teleologically, chemokines induced by a pathogen should select a panel of leukocyte types with effector functions that are tailored to eliminate the invader. This is not always the case. Because pathogens evolve rapidly under the selection pressure from our defenses, they may escape the effector functions of leukocytes recruited by chemokines. Some infectious microorganisms even exploit the chemokine-driven defenses in their pathogenesis strategies, and the effector functions of leukocytes may contribute to the unfavorable disease outcome (155). The recruited leukocytes may be used as safe havens, vehicles to disseminate, or Trojan horses to get access to organs that are difficult to reach. No other pathogen class uses and abuses Chemokinese as brazenly as the viruses.

Chemokinese as a Foreign Language: The Virokinese Successful warfare requires the communication language to remain secret. Higher organisms failed this mission during the billion years’ war they have waged against viruses. Endowed with a remarkable ability to attain genetic information, some viruses have broken the code of Chemokinese and learned how to use it in their diverse survival stratagies. For example, HIV masquerades as a “chemokine” and uses chemokine receptors to promote its fusion with target cells. Pox and herpes viruses encode in their genomes chemokine receptors that are expressed by the infected cells. This allows the host chemokines to drive the infected cells into

12 Feb 2004

16:20

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

904

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

remote sites for undisturbed cycles of replication and infectivity. Alternatively, viral chemokine receptors may induce uncontrolled cell proliferation, providing new cell targets for virus replication. Herpes viruses produce chemokines with agonist and antagonist features. Agonists may lure their potential targets or skew the host responses toward Th2, which renders antiviral defenses futile. Antagonists may simply derail the migration of immune cells. Finally, viruses can produce chemokine binding proteins that interfere with the Chemokinese communication of the host by inhibiting either receptor binding of chemokines or their immobilization by GAGs, or both. For further details on the fascinating world of Virokinese, we refer the reader to several recent reviews (156–158).

Unique Meanings of Chemokines In contrast to synonymous and homonymous chemokines, there are chemokinereceptor pairs that appear to be monogamous, including CXCR4-CXCL12, CXCR5-CXCL13, CCR6-CCL20, CCR8-CCL1, and CCR9-CCL25. The majority of these include the constitutively expressed “homeostatic” chemokines. There are also several homeostatic two ligands/one receptor “relationships”: CCL19 and CCL21 bind to CCR7, CCL17 and CCL22 to CCR4, and CCL27 and CCL28 to CCR10. These pathways are mainly involved in directing the tissue-specific migration of leukocytes between and within primary, secondary, and tertiary lymphoid organs. They are also required for the development of lymphoid organs and the establishment of functional microenvironments within them. The contribution of homeostatic chemokines to lymphorganogenesis has been reviewed recently (159). However, the division into inducible and homeostatic chemokines is by no means absolute. Homeostatic chemokines are upregulated in inflammation and are responsible for the recruitment of leukocytes, specifically into chronic lesions where they can promote the formation of ectopic lymphoid structures (160–163). Many inducible chemokines are also produced homeostatically, and some are secreted into normal colostrum, milk, eccrine sweat, saliva, and tears (164–166). Such secretion may take advantage of defensin-like direct antimicrobial properties that have been described for several chemokines, including CXCL4, CXCL7, CXCL9, CXCL10, CXCL11, CCL5, and CCL28. Also, several β-defensins are full agonists of CCR6. For more on this subject, see the review by Yang et al. (166a) in this volume. As discussed in detail below, homeostatic chemokines and their apposite receptors determine and restrict the migratory paths of lymphoid cells. However, a single signal can elicit diverse functional responses in different cells. For example, CXCL12 induces the migration of primordial germ cells and lateral sensory organ cells in developing zebrafish and critically contributes to embryonal hematopoeisis, vasculogenesis, and cardiogenesis, as well as the development of the cerebellum and dentate gyrus in mice. Postnatally, CXCL12 drives the migration of hematopoietic and EC progenitors required for bone marrow homing and neovasculogenesis, respectively. It is also involved in inflammatory recruitment of diverse leukocytes. In tumors of various origins, CXCL12 promotes cell survival and

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

P1: IKH

905

induces metastasis. In the brain, where CXCL12 has apparently originated (167), it attracts microglia, stimulates astrocytes, elicits postsynaptic currents in Purkinje cells, and promotes pain perception. Given all these diverse effects, CXCL12 might mean “Just do it!” in Chemokinese.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Chemokines in Inflammatory Diseases To a hammer every problem looks like a nail; to an immune system every problem looks like an infection. The downside of carrying stencil templates for highly effective alarm messages is that they may be applied to innocuous situations that are interpreted as danger, such as chronic inflammatory and autoimmune diseases (168). Whatever the primary triggers may be, chemokines appear in the lesions of debilitating human diseases such as rheumatoid arthritis, bronchial asthma, multiple sclerosis, type I diabetes, arteriosclerosis, and inflammatory bowel disease. The mechanisms of chemokine induction are probably similar to those in response to pathogens. Chemokines recruit leukocytes that, through their individual effector functions, perpetuate the vicious disease cycles by causing tissue damage, providing further inflammatory signals, and propagating the molecular sequences of autoimmunity (169). A new generation of anti-inflammatory drugs is expected to emerge from current pharmacological efforts to target chemokines and their receptors (170–172a). But which chemokines or receptors to target? For some diseases the targets may be as self-evident as CCR9 is in inflammatory bowel disease. However, in rheumatoid arthritis at least 12 different chemokines have been suggested as playing pathoetiological roles. This implicates the involvement of at least nine distinct receptors. The chemokine involvement is similarly complex in other inflammatory diseases. Curiously, the alternative elimination of a single chemokine or the silencing of a single receptor may scramble the Chemokinese message in an inflammatory disease sufficiently to break the backbone of pathogenesis. This has been demonstrated by the alleviation of disease-specific parameters in experimental and clinical arteriosclerosis in mice and humans deficient for either CCL2, CCR2, CXCR2, or CX3CR1 (173–176), and for CCR2, CCR5, or CX3CR1 (177, 178), respectively. Yet in theory, removing nonessential chemokines may leave the message intact, similar to the way that in Semitic languages the meaning of a word is clear without registering the vowels. Chemokine- or chemokine receptor–deficient mice and human polymorphisms that result in chemokine pathway deficiencies, e.g., CCR5132, provide possibilities to explore both of these scenarios and have been reviewed in detail recently (147, 170–172).

CHEMOKINES IN ADAPTIVE IMMUNITY The ability to use language is a skill that keeps evolving throughout a person’s life. As we accumulate experiences and our frame of reference changes, so does our interpretation of language. Many immune cells, particularly T cells, undergo analogous changes in their comprehension of Chemokinese (172a, 178a). These

12 Feb 2004

16:20

906

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

changes, reflected by characteristic expression of chemokine receptors, accompany every major T-cell differentiation event. The following examples highlight how different immunological tasks determine the capacity of T-cell subsets to read and interpret chemokine signals.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Naive Lymphocyte Seeking Mature DC: Matchmaking in Chemokinese The single-minded mission pursued by naive T cells is to find a cognate antigen. However, a primary T-cell response is only induced when the antigen is presented by mature DCs (179). This necessitates that these two dissimilar cells unite in an event orchestrated by chemokines, but not before DCs have acquired antigen, a process efficient only prior to their maturation. Immature DCs and their precursors develop in the bone marrow and migrate into peripheral tissues and secondary lymphoid organs. Normal mouse LNs contain at least six distinct DC subsets with different immunological activities, tissue distribution, and trafficking properties (180). Some DCs (e.g., plasmacytoid DCs) are thought to home into LNs from the blood (see below), whereas others originate in peripheral sites and take residence in LNs by way of afferent lymph vessels. Peyer’s patches (PPs) and the spleen do not have afferent lymph vessels, so all resident DCs have probably homed in there from the blood. Immature DCs in LNs, PPs, and the spleen collect and process lymph-borne, intestinal, or blood-borne antigens, respectively. Their counterparts in peripheral tissues acquire antigens that arise locally or enter through the body’s surfaces. Another source of DCs is circulating monocytes, which express numerous receptors for inflammatory chemokines and other chemoattractants and which accumulate at sites of acute inflammation and in LNs that receive afferent lymph from inflamed tissues (132, 181, 182). Irrespective of the site where they capture antigens, DCs must migrate into the T-cell area of lymphoid tissues and assume a fully mature phenotype before they can “educate” naive T cells (172a, 178a, 179). DC maturation induces several critical changes, including the pattern of chemokine receptor expression (179, 182–184). Maturing DCs must downregulate most of their initial chemokine receptors and upregulate CCR7 to present antigens to T cells. CCR7 allows both tissueresident and monocyte-derived DCs in peripheral sites to enter afferent lymphatics, which express CCL21 and possibly CCL19 (185). DC responsiveness to lymphatic CCL19 is supported by a second autocrine signal in the form of cysteinylated leukotrienes (186). There is still uncertainty about the precise step(s) at which CCR7 ligands function. They may assist DCs in finding lymph vessels and/or enable DCs to enter these vessels. CCR7 ligands are also essential in allowing DCs that have entered the subcapsular sinus or reside in the superficial cortex in LNs to migrate into the T-cell area in the deep cortex (187). Here, they form a welcoming committee that presents antigen to T cells that continuously enter this region via HEV. Recent measurements indicate that a single DC in an LN gets in

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CHEMOKINES IN IMMUNITY

P1: IKH

907

touch with as many as 500 different T cells per hour (188). However, in the vast majority of their visits to lymphoid organs, T cells fail to find “their” antigen, and after a few hours they return to the bloodstream to try their luck elsewhere. Circulating naive T cells express several essential traffic molecules, including the adhesion molecules L-selectin, LFA-1, and α4ß7 integrin, and two chemokine receptors, CCR7 and CXCR4. This restricted repertoire of traffic molecules is tailored for interactions with the specialized microvessels in lymphoid tissues and keeps naive T cells from straying into other regions of the body. For the purpose of lymphocyte trafficking, CCR7 and its ligands CCL19 and CCL21 are essential (189). CCL21 is synthesized in HEVs (131, 190), the principal port of lymphocyte entry into LNs and PPs (191), and by lymphatic ECs and some interstitial cells in T-cell areas of lymphoid organs. CCL19 has also been claimed to be generated by lymphatic endothelium (186) and is produced by DCs and probably other cells in T-cell areas, but not in HEVs. As discussed above, lymph-borne CCL19 can be transcytosed from the extravascular compartment to the luminal surface of HEVs, where it is presented to lymphocytes together with CCL21 (129). The genes for CCL19 and the HEV-expressed form of CCL21 are deleted in a strain of mice called paucity in lymph node T cells (plt/plt) (187). Consequently, naive T cells undergo rolling via L-selectin, the chemoattractant-independent first step in the homing cascade, but they do not undergo activated integrin-dependent arrest in LN HEV of plt/plt mice (131). However, intracutaneous injection into plt/plt mice of either CCL19 or CCL21 leads to chemokine translocation to HEV in draining LNs and reconstitution of integrin activation and T-cell homing (129, 131). Thus, either of the two CCR7 agonists is sufficient for integrin activation on rolling T cells in HEV, but HEV-expressed CCL21 probably provides the predominant signal. The situation is somewhat more complex for naive B-cell homing. Compared with the migration of T cells, B-cell migration to LNs and PPs of plt/plt mice is much less compromised (192). This is because rolling B cells in HEVs can activate integrins and arrest not only in response to CCR7 agonists, but also upon CXCR4 triggering by CXCL12, which is presented in some HEVs in the outer cortex (193). In PPs, another B cell–specific chemokine pathway can induce moderate sticking via CXCR5 on B cells and CXCL13, which is selectively expressed in PP HEV segments that are in contact with B follicles (193). In contrast, CCL21 is only presented by HEVs in the interfollicular T-cell area of PPs (194). The physiological role of CXCR4 and its ligand CXCL12 in trafficking of lymphocytes, especially of naive T cells, is unresolved. Mice with a genetic deficiency in either molecule are not viable (195, 196), but using bone marrow chimeras, researchers have performed a few studies with CXCR4-deficient hematopoietic cells (193, 197). One study found a modest role for CXCR4 in T-cell sticking in LN HEV of plt/plt mice (193). This work was carried out with unfractionated T cells, leaving open the possibility that the T cells that responded to CXCL12 in HEV were LN-homing memory T cells. Indeed, studies that have specifically analyzed naive T cells found that, although CXCL12 induced chemotaxis in vitro, it failed to trigger integrin-mediated arrest in vivo or to reconstitute naive T-cell homing

12 Feb 2004

16:20

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

908

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

in plt/plt mice (131, 161). Thus, there appears to be a dissociation in the type of response that is elicited by CXCL12 in different CXCR4-expressing lymphocyte subsets. B cells (and probably some memory T cells) respond with chemotaxis and rapid integrin activation, whereas naive T cells undergo only chemotaxis. An additional level of complexity is introduced by the finding (discussed above) that most T cells are attracted by low to modest concentrations of CXCL12, but a subset migrates away when exposed to a high concentration of CXCL12 (46). Lymphocyte homing to the spleen is a special case because no specific adhesive interactions with ECs appear to be required (198). However, lymphocytes must migrate into the peri-arteriolar lymphoid sheath (PALS) after extravasating from marginal-zone sinuses in the red pulp. To find their proper place in the PALS, B- and T-cells use integrins (62) and depend on CXCR5 and CCR7 to sense chemokine gradients that guide them into B follicles and the T-cell area, respectively. A genetic deficiency in either receptor or their respective ligands or inhibition of chemokine signaling by pertussis toxin inhibits lymphocyte migration into the PALS and results in severely disorganized white pulp cords (189, 199, 200). B cells that diapedese across HEV in LNs must traverse the T-cell area to reach B follicles. This directed migration is mediated by CXCR5 on the migrating B cell and CXCL13, which is generated in B follicles (200, 201). Once they have reached the follicles, most B cells continue to wander in random directions within the B-cell compartment (202). Similarly, T cells that home into LNs display chemokinesis when viewed both by multiphoton microscopy in freshly excised LNs in vitro and by intravital microscopy in vivo (202, 203). The factor(s) that stimulate Band T-cell chemokinesis in their respective compartments remain to be identified. Mature DCs express several chemokines that might attract T cells, and in vitro studies have shown that DCs can stimulate random T-cell migration in collagen gels (204). Whether DCs or DC-derived secreted factors are required for interstitial lymphocyte migration in LNs has not been determined.

Effector and Memory Cell Migration: Staying on Track in the Tower of Babel Full-fledged activation induces T-cell proliferation followed by acquisition of effector functions. Whether effector cells differentiate to support primarily cytotoxic, interferon-γ -driven (type 1) responses or humoral, IL-4-dependent (type 2) responses depends on a variety of factors that have been reviewed elsewhere (205, 206). Irrespective of their specific flavor, effector cells are entrusted with the critical task of eliminating the source of their cognate antigen. To do this, T cells must develop a sophisticated and diversified understanding of Chemokinese. One of the earliest changes is seen on activated CD4 T cells, which upregulate CXCR5 and reduce CCR7 expression. This allows them to migrate toward the edges of B follicles to provide help to B cells and promote germinal center formation (207–210). Conversely, antigen-stimulated B cells acquire responsiveness to CCL19, which drives them toward the T-cell area.

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CHEMOKINES IN IMMUNITY

P1: IKH

909

An equally important requirement is that effector cells gain efficient access to nonlymphoid tissues that contain their cognate antigen. Typically, these tissues experience some form of distress due to, for example, an infection. The ensuing innate immune response leads to local production of inflammatory chemokines and enhanced expression of adhesion molecules on venular ECs (211, 212). Thus, effector cells upregulate adhesion molecules and chemokine receptors that allow them to migrate to sites of inflammation (213, 214). As discussed above, some pathogens can only be eliminated by type 1 responses, whereas others require a type 2 response. The cytokines released by either effector subset suppress each other’s effects. To avoid an immunological stalemate, Th1 and Th2 cells obey different traffic signals and express distinct patterns of chemokine receptors (reviewed in 215). In particular, CCR5 and, less conspicuously, CXCR3 predominate on Th1 cells, whereas CCR4 and CCR8 are primarily found on Th2 cells. It has also been reported that CCR3 and CXCR4 are preferentially expressed on Th2 cells, but this is somewhat controversial. The chemokines that stimulate Th1 cells are induced by IFN-γ and suppressed by IL-4, which induces a chemokine milieu that preferentially attracts Th2 cells. While distinct chemokine receptor patterns are clearly induced on in vitro– polarized effector Th cells, the situation in vivo is often less clear-cut. One major modulating factor during the induction of effector cells in lymphoid organs is whether an antigen is encountered in a cutaneous or intestinal context (216). In a sense, the skin and the intestine use different dialects of Chemokinese, even in the absence of inflammation. Thus, dermal postcapillary venules present CCL17 and CCL27 on their luminal surface. Effector cells that are generated in response to cutaneous antigens express the respective counter-receptors, CCR4 and/or CCR10, and expression of at least one of these is required for homing into the skin (217). The analogous constitutive chemokine in the small intestine is CCL25, whose receptor, CCR9, is expressed on thymocytes and naive CD8 T cells [CCL25 is also expressed in the thymus, where it may play a role in the generation of naive T cells (218, 219)]. Effector T cells that are stimulated by DCs from gut-associated lymphoid tissues maintain CCR9 expression and responsiveness to CCL25, whereas T cells lose responsiveness to CCL25 when activated by DCs from other lymphoid organs (220). Indeed, DCs in PPs and mesenteric LNs acquire a lymphoid organ-specific flavor that allows them to target effector cells to the small intestine, the site most likely to contain a cognate antigen (220, 221). This tissue-specific imprinting is mediated, in part, by the DCs’ ability to make T cells perceptive to regional chemokines. Whereas CCR9 or CCR4/CCR10 expression is a necessary component for effector cell recruitment to the small bowel or skin, respectively, their presence alone is not sufficient for homing. T cells must additionally express appropriate adhesion molecules that interact with organ-specific endothelial addressins (212). Recent work has shown that antibody-secreting plasma cells also consist of different subsets that are characterized by distinct chemokine receptor patterns and differential regional distribution (222). Most long-lived IgG-producing plasma

12 Feb 2004

16:20

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

910

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

cells reside in the bone marrow. These cells depend on CXCR4-CXCL12 to reach this site. In contrast, IgA-producing plasma cells are targeted to mucous membranes. A subset of these cells expresses CCR9 and, like small bowel–tropic effector T cells, migrates to the small intestine. However, CCR9 is not involved in targeting lymphocytes to the colon because CCL25 is not expressed in the large bowel. IgA-producing plasma cells use CCR10 to home into the colon and other mucosal sites, which express the CCR10 ligand CCL28 (223). However, T cells do not appear to use CCR10 to enter the colon. The chemokine pathways for effector T-cell migration to this and other anatomic regions remain to be identified. For the purpose of this discussion, effector cells are defined as arising in response to a recent antigenic stimulus, exerting immediate effector activity upon T-cell receptor stimulation, and being short-lived; whereas memory cells have been exposed to antigen in the more distant past, are long-lived, and confer enhanced protection against recall antigens. Memory cells are not a homogenous population, in particular with regard to chemokine receptor expression. One important distinction is the presence or absence of CCR7 (224). Most CCR7+ memory cells also express L-selectin and are thus able to home into LNs, whereas CCR7– memory cells do not express L-selectin and cannot home into any secondary lymphoid organ, except the spleen (214, 225). The CCR7+ and CCR7– subsets have been termed central memory (TCM) cells and effector memory (TEM) cells, respectively (224). TEM cells owe their name to their ability to mount rapid effector responses upon rechallenge and to their repertoire of traffic molecules that resemble effector cells (224). Given the diversity of effector cells’ migratory specificities, it is not suprising that TEM cells also fall into several subcategories that express homing receptors for skin, gut, or neither (226). Indeed, following antigen challenge in vivo, memory cells can be found lodging in multiple nonlymphoid sites throughout the body (227, 228). TCM cells have arguably the broadest migratory spectrum of any T-cell subset because they express both CCR7 and receptors for inflammatory chemokines (214, 224). Thus, TCM cells migrate efficiently to lymphoid tissues as well as to sites of inflammation (214). Compared with TEM cells, they also have the highest proliferative potential and confer superior immune protection (229, 230). TCM cells can arise directly from naive T cells (225, 229, 231) or from TEM cells (230). In the latter case, CCR7 expression is biphasic, i.e., high expression is seen in naive T cells, which become CCR7– during the effector stage and retain this phenotype while standing guard as TEM cells in peripheral tissues. At some stage, TEM cells begin to re-express CCR7, which might allow them to enter local lymphatics and migrate to LNs (232). It is not clear whether these cells continue their journey by recirculating exclusively through lymphoid organs or retain peripheral homing capacity as well. However, the latter scenario is favored by the observation that many circulating memory cells with skin- and gut-homing phenotypes also express CCR7 (233).

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

P1: IKH

911

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SEMANTICS OF CHEMOKINESE Like words in human languages, chemokines have no inherent semantic value. Their meaning is decoded by the compulsory presence of the apposite receptors and depends largely on the responding cell’s nature and nurture, i.e., its genetically governed functional program and environmental signals. It is clear from the example with CXCL12 that the same chemokine may have dissimilar meanings for cells of different lineages. Curiously, the meaning of a chemokine for cells of the same lineage may also vary. For example, CXCL12 signaling through its receptor on T cells may result, depending on the dominance of either p38 MAPK or Akt in signal transduction, in a pro- or antiapoptotic signal, respectively (234). We have already discussed the dramatic change of chemokine meaning depending if the chemokine is either immobilized on the EC surface (pro-emigratory) or soluble in plasma (anti-emigratory). The following examples show how chemokine meaning is changed by the simultaneous presence of different chemokines, by chemokine processing in limbo between their secretion and perception, and by concomitant activation of the responding cells.

Semantic Shift by the Presence of Different Chemokines Individual chemokines may dramatically modify each other’s effects in several different ways and at different levels. Many cells possess multiple chemokine receptors and are exposed to their diverse ligands simultaneously or sequentially. How do combinations of dissimilar chemokines work and what is the outcome? How may leukocyte responses be influenced by different spatial and temporal sequences of exposure to chemokines or by the direction and steepness of their gradients? The answers to these questions of Chemokinese syntax are still sketchy and are largely based on extrapolations of the pioneering work by Foxman et al. (235, 236). Apparently, leukocytes can maintain a memory of earlier chemokine exposure, and by comparing it with a newly encountered gradient, they can navigate in simultaneous and sequential gradients of homologous and heterologous attractants (235, 236). One result of a cell’s exposure to simultaneous conflicting gradients may be an amalgamation of signals and compromise responses along the new integrated vector gradient (idiomatic gradient perception). The other possible outcome is hierarchical prioritization of gradients with the prospect of subdominant signals being silenced, perhaps owing to a heterologous receptor desensitization and downregulation (237). However, there are also examples of chemokines synergistically potentiating each other’s effects by priming the responding cells (238) through heterodimerization of their receptors (57) or by forming heteroaggregates, which may create a new, more powerful meaning. In addition, chemokines can directly influence each other’s receptor binding because chemokines with agonistic activity for one receptor may be antagonists for another. Such chemokine “antonyms” include CCL7, that acts as an antagonist

12 Feb 2004

16:20

912

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

on CCR5 (239); all three CXCR3 agonists that function as CCR3 antagonists (9, 240); CCL18, an agonist for an unidentified receptor on T cells that is also a CCR3 antagonist (241); and two eotaxins, CCL11 and CCL26, that are CCR2 antagonists (47, 242). In addition, CCL11 shows high-affinity binding to CXCR3 but does not signal and does not compete for binding with other ligands. Thus, CXCR3 may function as a CCL11-sequestering decoy (9). Undoubtedly, more examples of chemokine antagonism will emerge in the future.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Proteolytic Processing of Chemokines Many chemokines are processed enzymatically after their secretion, leading to changes in their agonist properties. For example, dipeptidyl aminopeptidase IV (CD26) can cleave the two N-terminal amino acids of chemokines with proline, hydroxyproline, or alanine in the second N-terminal position, and it can dramatically change the perception of these chemokines by their receptors (243). Other proteases may also specifically cleave chemokines with either up- or downregulation of their agonistic properties and potential induction of antagonism. Examples of semantic shifts induced by proteolytic digestion of chemokines are listed in Table 1.

Semantic Shift by Functional Coupling and Uncoupling of Chemokine Receptors Another way of modulating chemokine perception is through cytokine-dependent regulation of receptor expression or its coupling to signal transduction. This circumstance was not appreciated at the time of the discovery of the first chemokines. Initially it was postulated that there was a profound functional dichotomy between CC chemokines on one hand and CXCL8 and related chemokines that share the ELR tripeptide motif on the other hand. The latter were thought to affect neutrophils but not monocytes, whereas the former stimulate monocytes but not neutrophils. However, subsequent work has shown that monocytes express both functional CXCL8 receptors (244, 245), can migrate in response to this chemokine (246), adhere firmly under shear flow in response to CXCL8 and other CXCR2 ligands (247–249), and localize to inflammatory lesions by means of CXCR2 (175, 250). The apparent disagreement of these data with the initial observations of the exclusive neutrophil activity of CXCL8 was elegantly resolved by a report showing that cytokines, IL-4 and IL-13 in particular, can increase the expression of both CXCR1 and CXCR2 on the surface of monocytes and couple them to productive signal transduction pathways (251). Similarly, IFNγ stimulation of neutrophils upregulates CCR3 and renders these cells responsive to CCR3 ligands, which are commonly thought of as classical eosinophil agonists (252). Conversely, the induction of CCR3 on B cells by IL-2 and IL-4 makes these cells susceptible to apoptotic signals through CCL11 (253). There are many other cases of downregulation or upregulation of chemokine receptor expression and function by cytokines and other stimuli. For example, as discussed above, immature “inflammatory” DCs lose CCR1, CCR2, and CCR5 in the process of maturation and start to express

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

CHEMOKINES IN IMMUNITY

913

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 1 Examples of semantic shift induced by proteolytic digestion of chemokines Chemokine

Enzyme

Processed forms

Receptor/Outcome

CXCL7

Elastase Chymotrypsin

CXCL7 (25–94) CXCL7(23–94)

CXCR2 agonism↑

CXCL8

Many

CXCL8(3–79) CXCL8(8–79) CXCL8(9–79) CXCL8(10–79) CXCL8(11–79)

With each step CXCR1 and CXCR2 agonism↑

CXCL9

CD26

CCL9(3–103)

CXCR3 agonism↓ CXCR3B(?)agonism→

CXCL10

CD26

CCL10(3–77)

CXCR3 agonism↓ CXCR3B(?)agonism→

CXCL11

CD26

CCL11(3–73)

CXCR3 agonism ↓ CXCR3B(?) agonism→

CXCL12

CD26

CXCL12(3–68)

CXCR4 agonism↓ desensitization→

CXCL12

Elastase MMP-1, -2, -3, -9, -13, -14 Cathepsin G

CXCL12(4–67) CXCL12(5–67) CXCL12(6–67)

CXCR4 agonism↓

CCL2

MMP-1, -3

CCL2(5–76)

CCR2 agonism↓ antagonism↑

CCL3L1

CD26

LD78β(3–70)

CCR1 and CCR5 agonism↑

CCL4

CD26 (?)

CCL4(3–69)

CCR1agonism↑ CCR5 agonism→

CCL5

CD26

CCL5(3–68)

CCR1 and CCR3 agonism↓ CCR5 agonism→

CCL7

MMP-1, -2, -3, -13, -14

CCL7(5–76)

CCR1, CCR2, CCR3 agonism ↓, antagonism ↑

CCL8

MMP-3

CCL8(5–76)

CCR2 agonism↓, antagonism ↑

CCL11

CD26

CCL11(3–74)

CCR3 agonism↓ desensitization →

CCL13

MMP-1, -3

CCL13(4–75) CCL13(5–75) CCL13(8–75)

CCR2, CCR3 agonism↓, antagonism ↑

CCL14

Serine proteases

CCL14(9–74)

CCR1, CCR5, CCR3 agonism ↑

CCR7 (254). LPS treatment can accelerate this process. In contrast, simultaneous LPS plus IL-10 treatment retains the inflammatory chemokine receptors and inhibits CCR7 induction (8). In addition, the combination of IL-10 plus LPS changes the signal coupling of CCR1, CCR2, and CCR5 and renders them nonsignaling in response to their cognate ligands, which, however, are internalized and sequestered. Thus, these receptors have transformed into chemokine-scavenging decoys (8).

12 Feb 2004

16:20

914

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONCLUSIONS Chemokines are molecular symbols that make up cell messages and, depending on the biological context, can have different meanings. Chemokines are perceived and acted upon by many different cell types. Each interpretation, each reaction, mirrors as much of the responding cell’s nature as it reflects on the chemokine itself. Our efforts to understand Chemokinese exemplify the difficulties faced by the researchers in the postgenomic era. The human genome, “the book of life,” is open; we can read it and one day will surely understand. Decoding the complete meaning of Chemokinese is going to be a similarly monumental task as has been the deciphering of the ancient Egyptian hieroglyphs. Our success will open the door for new disease remedies that, unlike our current drugs, which silence cells by gag order, will work more as cell speech therapy. ACKNOWLEDGMENTS We thank Pius Loetscher, Rob Nibbs, Dieter Maurer, Tam´as Schweighoffer, Charles Mackay, and Lesley Silberstein for carefully reading the manuscript and Joe Moore for superb editorial assistance. Research was supported by NIH grants AR42689, HL62524, HL54936, and HL56949 to U.H.v.A. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Baggiolini M, Dewald B, Moser B. 1997. Human Chemokines: an update. Annu. Rev. Immunol. 15:675–705 2. Rollins BJ. 1997. Chemokines. Blood 90:909–28 3. Luster AD. 1998. Chemokines—chemotactic cytokines that mediate inflammation. N. Engl. J. Med. 338:436–45 4. Zlotnik A, Yoshie O. 2000. Chemokines: a new classification system and their role in immunity. Immunity 12:121–27 5. Bacon K, Baggiolini M, Broxmeyer H, Horuk R, Lindley I, et al. 2002. Chemokine/chemokine receptor nomenclature. J. Interferon Cytokine Res. 22:1067–68 6. Rossi D, Zlotnik A. 2000. The biology of chemokines and their receptors. Annu. Rev. Immunol. 18:217–42 7. Murphy PM. 2002. International Union of Pharmacology. XXX. Update on

8.

9.

10.

11.

chemokine receptor nomenclature. Pharmacol. Rev. 54:227–29 D’Amico G, Frascaroli G, Bianchi G, Transidico P, Doni A, et al. 2000. Uncoupling of inflammatory chemokine receptors by IL-10: generation of functional decoys. Nat. Immunol. 1:387–91 Xanthou G, Duchesnes CE, Williams TJ, Pease JE. 2003. CCR3 functional responses are regulated by both CXCR3 and its ligands CXCL9, CXCL10 and CXCL11. Eur. J. Immunol. 33:2241–50 Kenakin T, Onaran O. 2002. The ligand paradox between affinity and efficacy: Can you be there and not make a difference? Trends Pharmacol. Sci. 23:275– 80 Kenakin T. 2003. Ligand-selective receptor conformations revisited: the promise and the problem. Trends Pharmacol. Sci. 24:346–54

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CHEMOKINES IN IMMUNITY 12. Ganju RK, Brubaker SA, Meyer J, Dutt P, Yang Y, et al. 1998. The alphachemokine, stromal cell-derived factor1α, binds to the transmembrane Gprotein-coupled CXCR-4 receptor and activates multiple signal transduction pathways. J. Biol. Chem. 273:23169–75 13. Stossel TP. 1994. The E. Donnall Thomas Lecture, 1993. The machinery of blood cell movements. Blood 84:367– 79 14. Weiner OD, Servant G, Welch MD, Mitchison TJ, Sedat JW, Bourne HR. 1999. Spatial control of actin polymerization during neutrophil chemotaxis. Nat. Cell Biol. 1:75–81 15. Zigmond SH, Levitsky HI, Kreel BJ. 1981. Cell polarity: an examination of its behavioral expression and its consequences for polymorphonuclear leukocyte chemotaxis. J. Cell Biol. 89:585–92 16. Ramsey WS. 1972. Analysis of individual leucocyte behavior during chemotaxis. Exp. Cell. Res. 70:129–39 17. Zigmond SH. 1977. Ability of polymorphonuclear leukocytes to orient in gradients of chemotactic factors. J. Cell Biol. 75:606–16 18. Nieto M, Frade JMR, Sancho D, Mario M, Martinez-A C, Sanchez-Madrid F. 1997. Polarization of chemokine receptors to the leading edge during lymphocyte chemotaxis. J. Exp. Med. 186:153– 58 19. Kuang Y, Wu Y, Jiang H, Wu D. 1996. Selective G protein coupling by C-C chemokine receptors. J. Biol. Chem. 271:3975–78 20. al-Aoukaty A, Schall TJ, Maghazachi AA. 1996. Differential coupling of CC chemokine receptors to multiple heterotrimeric G proteins in human interleukin-2-activated natural killer cells. Blood 87:4255–60 21. Arai H, Tsou C-L, Charo IF. 1997. Chemotaxis in a lymphocyte cell line transfected with C-C chemokine receptor 2B: evidence that directed migration

22.

23.

24.

25.

26.

27.

28.

29.

30.

P1: IKH

915

is mediated by βg dimers released by activation of Gai-coupled receptors. Proc. Natl. Acad. Sci. USA 94:14495–99 Neptune ER, Bourne HR. 1997. Receptors induce chemotaxis by releasing the βg subunit of Gi, not by activating Gq or Gs. Proc. Natl. Acad. Sci. USA 94:14489–94 Hirsch E, Katanaev VL, Garlanda C, Azzolino O, Pirola L, et al. 2000. Central role for G protein-coupled phosphoinositide 3-kinase gamma in inflammation. Science 287:1049–53 Sadhu C, Masinovsky B, Dick K, Sowell CG, Staunton DE. 2003. Essential role of phosphoinositide 3-kinase delta in neutrophil directional movement. J. Immunol. 170:2647–54 Curnock AP, Sotsios Y, Wright KL, Ward SG. 2003. Optimal chemotactic responses of leukemic T cells to stromal cell-derived factor-1 requires the activation of both class IA and IB phosphoinositide 3-kinases. J. Immunol. 170:4021–30 Huang YE, Iijima M, Parent CA, Funamoto S, Firtel RA, Devreotes P. 2003. Receptor-mediated regulation of PI3Ks confines PI(3,4,5)P3 to the leading edge of chemotaxing cells. Mol. Biol. Cell 14:1913–22 Gardiner EM, Pestonjamasp KN, Bohl BP, Chamberlain C, Hahn KM, Bokoch GM. 2002. Spatial and temporal analysis of Rac activation during live neutrophil chemotaxis. Curr. Biol. 12:2029–34 Welch HC, Coadwell WJ, Ellson CD, Ferguson GJ, Andrews SR, et al. 2002. P-Rex1, a PtdIns(3,4,5)P3- and Gβγ regulated guanine-nucleotide exchange factor for Rac. Cell 108:809–21 Innocenti M, Frittoli E, Ponzanelli I, Falck JR, Brachmann SM, et al. 2003. Phosphoinositide 3-kinase activates Rac by entering in a complex with Eps8, Abi1, and Sos-1. J. Cell Biol. 160:17–23 Servant G, Weiner OD, Herzmark P, Balla T, Sedat JW, Bourne HR. 2000.

12 Feb 2004

16:20

916

31.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

32.

33.

34.

35.

36.

37.

38.

39.

40.

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

Polarization of chemoattractant receptor signaling during neutrophil chemotaxis. Science 287:1037–40 Roberts AW, Kim C, Zhen L, Lowe JB, Kapur R, et al. 1999. Deficiency of the hematopoietic cell-specific Rho family GTPase Rac2 is characterized by abnormalities in neutrophil function and host defense. Immunity 10:183–96 Weiner OD, Neilsen PO, Prestwich GD, Kirschner MW, Cantley LC, Bourne HR. 2002. A PtdInsP(3)- and Rho GTPasemediated positive feedback loop regulates neutrophil polarity. Nat. Cell. Biol. 4:509–13 Weber KS, Klickstein LB, Weber PC, Weber C. 1998. Chemokine-induced monocyte transmigration requires Cdc42-mediated cytoskeletal changes. Eur. J. Immunol. 28:2245–51 Haddad E, Zugaza JL, Louache F, Debili N, Crouin C, et al. 2001. The interaction between Cdc42 and WASP is required for SDF-1-induced T-lymphocyte chemotaxis. Blood 97:33–38 Allen WE, Zicha D, Ridley AJ, Jones GE. 1998. A role for Cdc42 in macrophage chemotaxis. J. Cell Biol. 141:1147–57 Jones GE. 2000. Cellular signaling in macrophage migration and chemotaxis. J. Leukoc. Biol. 68:593–602 Etienne-Manneville S, Hall A. 2002. Rho GTPases in cell biology. Nature 420:629–35 Li Z, Hannigan M, Mo Z, Liu B, Lu W, et al. 2003. Directional sensing requires Gβγ -mediated PAK1 and PIX α-dependent activation of Cdc42. Cell 114:215–27 Comer FI, Parent CA. 2002. PI 3-kinases and PTEN: how opposites chemoattract. Cell 109:541–44 Xu J, Wang F, Van Keymeulen A, Herzmark P, Straight A, et al. 2003. Divergent signals and cytoskeletal assemblies regulate self-organizing polarity in neutrophils. Cell 114:201–14

41. Worthylake RA, Lemoine S, Watson JM, Burridge K. 2001. RhoA is required for monocyte tail retraction during transendothelial migration. J. Cell Biol. 154:147–60 42. Alblas J, Ulfman L, Hordijk P, Koenderman L. 2001. Activation of RhoA and ROCK are essential for detachment of migrating leukocytes. Mol. Biol. Cell. 12:2137–45 43. Worthylake RA, Burridge K. 2003. RhoA and ROCK promote migration by limiting membrane protrusions. J. Biol. Chem. 278:13578–84 44. Sugimoto N, Takuwa N, Okamoto H, Sakurada S, Takuwa Y. 2003. Inhibitory and stimulatory regulation of Rac and cell motility by the G12/13-Rho and Gi pathways integrated downstream of a single G protein-coupled sphingosine-1phosphate receptor isoform. Mol. Cell. Biol. 23:1534–45 45. Vogt S, Grosse R, Schultz G, Offermanns S. 2003. Receptor-dependent RhoA activation in G12/G13-deficient cells: genetic evidence for an involvement of Gq/G11. J. Biol. Chem. 278:28743–49 46. Poznansky MC, Olszak IT, Foxall R, Evans RH, Luster AD, Scadden DT. 2000. Active movement of T cells away from a chemokine. Nat. Med. 6:543– 48 47. Ogilvie P, Paoletti S, Clark-Lewis I, Uguccioni M. 2003. Eotaxin-3 is a natural antagonist for CCR2 and exerts a repulsive effect on human monocytes. Blood 102:789–94 48. Poznansky MC, Olszak IT, Evans RH, Wang Z, Foxall RB, et al. 2002. Thymocyte emigration is mediated by active movement away from stroma-derived factors. J. Clin. Invest. 109:1101–10 49. Laudanna C, Campbell JJ, Butcher EC. 1996. Role of Rho in chemoattractantactivated leukocyte adhesion through integrins. Science 271:981–83 50. Katagiri K, Maeda A, Shimonaka M, Kinashi T. 2003. RAPL, a Rap1-binding

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

51.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

52.

53.

54.

55. 56.

57.

58.

59.

60.

molecule that mediates Rap1-induced adhesion through spatial regulation of LFA-1. Nat. Immunol. 4:741–48 Shimonaka M, Katagiri K, Nakayama T, Fujita N, Tsuruo T, et al. 2003. Rap1 translates chemokine signals to integrin activation, cell polarization, and motility across vascular endothelium under flow. J. Cell Biol. 161:417–27 Sebzda E, Bracke M, Tugal T, Hogg N, Cantrell DA. 2002. Rap1A positively regulates T cells via integrin activation rather than inhibiting lymphocyte signaling. Nat. Immunol. 3:251–58 Rodriguez-Frade JM, Mellado M, Martinez AC. 2001. Chemokine receptor dimerization: two are better than one. Trends Immunol. 22:612–17 Mellado M, Rodriguez-Frade JM, VilaCoro AJ, Fernandez S, Martin de Ana A, et al. 2001. Chemokine receptor homoor heterodimerization activates distinct signaling pathways. EMBO J. 20:2497– 507 Thelen M. 2001. Dancing to the tune of chemokines. Nat. Immunol. 2:129–34 Curnock AP, Logan MK, Ward SG. 2002. Chemokine signalling: pivoting around multiple phosphoinositide 3-kinases. Immunology 105:125–36 Mellado M, Rodriguez-Frade JM, Manes S, Martinez AC. 2001. Chemokine signaling and functional responses: the role of receptor dimerization and TK pathway activation. Annu. Rev. Immunol. 19:397– 421 Kehrl JH. 1998. Heterotrimeric G protein signaling: Roles in immune function and fine-tuning by RGS proteins. Immunity 8:1–10 Bowman EP, Campbell JJ, Druey KM, Scheschonka A, Kehrl JH, Butcher EC. 1998. Regulation of chemotactic and proadhesive responses to chemoattractant receptors by RGS (regulator of Gprotein signaling) family members. J. Biol. Chem. 273:28040–48 Moratz C, Kang VH, Druey KM, Shi CS,

61.

62.

63.

64.

65.

66.

67.

P1: IKH

917

Scheschonka A, et al. 2000. Regulator of G protein signaling 1 (RGS1) markedly impairs Gi alpha signaling responses of B lymphocytes. J. Immunol. 164:1829– 38 Shi GX, Harrison K, Wilson GL, Moratz C, Kehrl JH. 2002. RGS13 regulates germinal center B lymphocytes responsiveness to CXC chemokine ligand (CXCL)12 and CXCL13. J. Immunol. 169:2507–15 Lippert E, Yowe DL, Gonzalo JA, Justice JP, Webster JM, et al. 2003. Role of regulator of G protein signaling 16 in inflammation-induced T lymphocyte migration and activation. J. Immunol. 171:1542–55 Aragay AM, Mellado M, Frade JM, Martin AM, Jimenez-Sainz MC, et al. 1998. Monocyte chemoattractant protein-1induced CCR2B receptor desensitization mediated by the G protein-coupled receptor kinase 2. Proc. Natl. Acad. Sci. USA 95:2985–90 Barlic J, Khandaker MH, Mahon E, Andrews J, DeVries ME, et al. 1999. βarrestins regulate interleukin-8-induced CXCR1 internalization. J. Biol. Chem. 274:16287–94 Fan GH, Yang W, Wang XJ, Qian Q, Richmond A. 2001. Identification of a motif in the carboxyl terminus of CXCR2 that is involved in adaptin 2 binding and receptor internalization. Biochemistry 40:791–800 Huttenrauch F, Nitzki A, Lin FT, Honing S, Oppermann M. 2002. β-arrestin binding to CC chemokine receptor 5 requires multiple C-terminal receptor phosphorylation sites and involves a conserved Asp-Arg-Tyr sequence motif. J. Biol. Chem. 277:30769–77 Feniger-Barish R, Belkin D, Zaslaver A, Gal S, Dori M, et al. 2000. GCP2-induced internalization of IL-8 receptors: hierarchical relationships between GCP-2 and other ELR(+)-CXC chemokines and mechanisms regulating

12 Feb 2004

16:20

918

68.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

69.

70.

71.

72.

73.

74.

75.

76.

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

CXCR2 internalization and recycling. Blood 95:1551–59 Fong AM, Premont RT, Richardson RM, Yu YR, Lefkowitz RJ, Patel DD. 2002. Defective lymphocyte chemotaxis in βarrestin 2- and GRK6-deficient mice. Proc. Natl. Acad. Sci. USA 99:7478–83 Somsel Rodman J, Wandinger-Ness A. 2000. Rab GTPases coordinate endocytosis. J. Cell Sci. 113 (Pt 2):183– 92 Fan GH, Lapierre LA, Goldenring JR, Richmond A. 2003. Differential regulation of CXCR2 trafficking by Rab GTPases. Blood 101:2115–24 Gortz A, Nibbs RJ, McLean P, Jarmin D, Lambie W, et al. 2002. The chemokine ESkine/CCL27 displays novel modes of intracrine and paracrine function. J. Immunol. 169:1387–94 Nibbs RJ, Graham GJ. 2003. CCL27/ PESKY: a novel paradigm for chemokine function. Expert Opin. Biol. Ther. 3:15– 22 Chang TL, Gordon CJ, Roscic-Mrkic B, Power C, Proudfoot AE, et al. 2002. Interaction of the CC-chemokine RANTES with glycosaminoglycans activates a p44/p42 mitogen-activated protein kinase-dependent signaling pathway and enhances human immunodeficiency virus type 1 infectivity. J. Virol. 76:2245– 54 Roscic-Mrkic B, Fischer M, Leemann C, Manrique A, Gordon CJ, et al. 2003. RANTES (CCL5) uses the proteoglycan CD44 as an auxiliary receptor to mediate cellular activation signals and HIV-1 enhancement. Blood 102:1169–77 Lasagni L, Francalanci M, Annunziato F, Lazzeri E, Giannini S, et al. 2003. An alternatively spliced variant of CXCR3 mediates the inhibition of endothelial cell growth induced by IP-10, Mig, and I-TAC, and acts as functional receptor for platelet factor 4. J. Exp. Med. 197:1537– 49 Brandt E, Ludwig A, Petersen F,

77.

78.

79.

80.

81.

82.

83.

84.

85.

Flad HD. 2000. Platelet-derived CXC chemokines: old players in new games. Immunol. Rev. 177:204–16 Goger B, Halden Y, Rek A, Mosl R, Pye D, et al. 2002. Different affinities of glycosaminoglycan oligosaccharides for monomeric and dimeric interleukin8: a model for chemokine regulation at inflammatory sites. Biochemistry 41: 1640–46 Wang D, Sai J, Richmond A. 2003. Cell surface heparan sulfate participates in CXCL1-induced signaling. Biochemistry 42:1071–77 Ali S, Palmer AC, Banerjee B, Fritchley SJ, Kirby JA. 2000. Examination of the function of RANTES, MIP-1α, and MIP1β following interaction with heparinlike glycosaminoglycans. J. Biol. Chem. 275:11721–27 Witt DP, Lander AD. 1994. Differential binding of chemokines to glycosaminoglycan subpopulations. Curr. Biol. 4:394–400 Patel DD, Koopmann W, Imai T, Whichard LP, Yoshie O, Krangel MS. 2001. Chemokines have diverse abilities to form solid phase gradients. Clin. Immunol. 99:43–52 Lortat-Jacob H, Grosdidier A, Imberty A. 2002. Structural diversity of heparan sulfate binding domains in chemokines. Proc. Natl. Acad. Sci. USA 99:1229–34 Webb LMC, Ehrengruber MU, ClarkLewis I, Baggiolini M, Rot A. 1993. Binding to heparan sulfate or heparin enhances neutrophil responses to interleukin 8. Proc. Natl. Acad. Sci. USA 90: 7158–62 Koopmann W, Krangel MS. 1997. Identification of a glycosaminoglycanbinding site in chemokine macrophage inflammatory protein-1α. J. Biol. Chem. 272:10103–9 Proudfoot AE, Fritchley S, Borlat F, Shaw JP, Vilbois F, et al. 2001. The BBXB motif of RANTES is the principal site for heparin binding and

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

86.

87.

88.

89.

90.

91.

92.

93.

controls receptor selectivity. J. Biol. Chem. 276:10620–26 Marshall LJ, Ramdin LS, Brooks T, DPhil PC, Shute JK. 2003. Plasminogen activator inhibitor-1 supports IL-8mediated neutrophil transendothelial migration by inhibition of the constitutive shedding of endothelial IL-8/heparan sulfate/syndecan-1 complexes. J. Immunol. 171:2057–65 Li Q, Park PW, Wilson CL, Parks WC. 2002. Matrilysin shedding of syndecan1 regulates chemokine mobilization and transepithelial efflux of neutrophils in acute lung injury. Cell 111:635–46 Carter NM, Ali S, Kirby JA. 2003. Endothelial inflammation: the role of differential expression of N-deacetylase/Nsulphotransferase enzymes in alteration of the immunological properties of heparan sulphate. J. Cell Sci. 116:3591–600 Kuschert GS, Coulin F, Power CA, Proudfoot AE, Hubbard RE, et al. 1999. Glycosaminoglycans interact selectively with chemokines and modulate receptor binding and cellular responses. Biochemistry 38:12959–68 Mack M, Pfirstinger J, Weber C, Weber KS, Nelson PJ, et al. 2002. Chondroitin sulfate A released from platelets blocks RANTES presentation on cell surfaces and RANTES-dependent firm adhesion of leukocytes. Eur. J. Immunol. 32:1012– 20 Wagner L, Yang OO, Garcia-Zepeda EA, Ge Y, Kalams SA, et al. 1998. β-chemokines are released from HIV1-specific cytolytic T-cell granules complexed to proteoglycans. Nature 391:908–11 Wolff B, Burns AR, Middleton J, Rot A. 1998. Endothelial cell “memory” of inflammatory stimulation: human venular endothelial cells store interleukin-8 in Weibel-Palade bodies. J. Exp. Med. 188:1757–62 Gibbs BF, Wierecky J, Welker P, Henz BM, Wolff HH, Grabbe J. 2001. Hu-

94.

95.

96.

97.

98.

99.

100.

101.

102.

103.

P1: IKH

919

man skin mast cells rapidly release preformed and newly generated TNF-α and IL-8 following stimulation with anti-IgE and other secretagogues. Exp. Dermatol. 10:312–20 Persson-Dajotoy T, Andersson P, Bjartell A, Calafat J, Egesten A. 2003. Expression and production of the CXC chemokine growth-related oncogenealpha by human eosinophils. J. Immunol. 170:5309–16 Kameyoshi Y, Dorschner A, Mallet AI, Christophers E, Schroder JM. 1992. Cytokine RANTES released by thrombinstimulated platelets is a potent attractant for human eosinophils. J. Exp. Med. 176:587–92 Boehlen F, Clemetson KJ. 2001. Platelet chemokines and their receptors: what is their relevance to platelet storage and transfusion practice? Transfus. Med. 11:403–17 Rot A. 1993. Neutrophil attractant/ activation protein-1 (interleukin-8) induces in vitro neutrophil migration by haptotactic mechanism. Eur. J. Immunol. 23:303–6 Wiedermann CJ, Kowald E, Reinisch N, Kaehler CM, von Luettichau I, et al. 1993. Monocyte haptotaxis induced by the RANTES chemokine. Curr. Biol. 3:735–39 Carter SB. 1967. Haptotaxis and the mechanism of cell motility. Nature 213:256–60 Laudanna C, Kim JY, Constantin G, Butcher E. 2002. Rapid leukocyte integrin activation by chemokines. Immunol. Rev. 186:37–46 Rot A. 1992. Endothelial cell binding of NAP-1/IL-8: role in neutrophil emigration. Immunol. Today 13:291–94 Tanaka Y, Adams DH, Shaw S. 1993. Proteoglycans on endothelial cells present adhesion-inducing cytokines to leukocytes. Immunol. Today 14:111–15 Middleton J, Neil S, Wintle J, ClarkLewis I, Moore H, et al. 1997.

12 Feb 2004

16:20

920

104.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

105.

105a.

106.

107.

108.

109.

110.

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

Transcytosis and surface presentation of IL-8 by venular endothelial cells. Cell 91:1001–11 Proudfoot AE, Handel TM, Johnson Z, Lau EK, LiWang P, et al. 2003. Glycosaminoglycan binding and oligomerization are essential for the in vivo activity of certain chemokines. Proc. Natl. Acad. Sci. USA 100:1885–90 Lawrence MB, Springer TA. 1991. Leukocytes roll on a selectin at physiologic flow rates: distinction from and prerequisite for adhesion through integrins. Cell 65:859–73 von Andrian UH, Chambers JD, McEvoy LM, Bargatze RF, Arfors KE, Butcher EC. 1991. Two-step model of leukocyteendothelial cell interaction in inflammation: distinct roles for LECAM-1 and the leukocyte beta 2 integrins in vivo. Proc. Natl. Acad. Sci. USA 88:7538–42 Berlin C, Bargatze RF, von Andrian UH, Szabo MC, Hasslen SR, et al. 1995. α4 integrins mediate lymphocyte attachment and rolling under physiologic flow. Cell 80:413–22 Alon R, Kassner PD, Carr MW, Finger EB, Hemler ME, Springer TA. 1995. The integrin VLA-4 supports tethering and rolling in flow on VCAM-1. J. Cell Biol. 128:1243–53 Campbell JJ, Hedrick J, Zlotnik A, Siani MA, Thompson DA, Butcher EC. 1998. Chemokines and the arrest of lymphoyctes rolling under flow conditions. Science 279:381–84 Chan JR, Hyduk SJ, Cybulsky MI. 2001. Chemoattractants induce a rapid and transient upregulation of monocyte α4 integrin affinity for vascular cell adhesion molecule 1 which mediates arrest: an early step in the process of emigration. J. Exp. Med. 193:1149–58 Constantin G, Majeed M, Giagulli C, Piccio L, Kim JY, et al. 2000. Chemokines trigger immediate β2 integrin affinity and mobility changes: differential regulation and roles in lympho-

111.

112.

113.

114.

115.

116.

117.

118.

cyte arrest under flow. Immunity 13:759– 69 Cinamon G, Grabovsky V, Winter E, Franitza S, Feigelson S, et al. 2001. Novel chemokine functions in lymphocyte migration through vascular endothelium under shear flow. J. Leukoc. Biol. 69:860–66 Cinamon G, Shinder V, Alon R. 2001. Shear forces promote lymphocyte migration across vascular endothelium bearing apical chemokines. Nat. Immunol. 2:515–22 Cuvelier SL, Patel KD. 2001. Sheardependent eosinophil transmigration on interleukin 4-stimulated endothelial cells: a role for endothelium-associated eotaxin-3. J. Exp. Med. 194:1699–709 Weber KS, von Hundelshausen P, ClarkLewis I, Weber PC, Weber C. 1999. Differential immobilization and hierarchical involvement of chemokines in monocyte arrest and transmigration on inflamed endothelium in shear flow. Eur. J. Immunol. 29:700–12 Weber C, Weber KS, Klier C, Gu S, Wank R, et al. 2001. Specialized roles of the chemokine receptors CCR1 and CCR5 in the recruitment of monocytes and T(H)1-like/CD45RO(+) T cells. Blood 97:1144–46 Hechtman DH, Cybulsky MI, Fuchs HJ, Baker JB, Gimbrone MA Jr. 1991. Intravascular IL-8: Inhibitor of polymorphonuclear leukocyte accumulation at sites of acute inflammation. J. Immunol. 147:883–92 Luscinskas FW, Kiely J-M, Ding H, Obin MS, H´ebert CA, et al. 1992. In vitro inhibitory effect of IL-8 and other chemoattractants on neutrophilendothelial adhesive interactions. J. Immunol. 149:2163–71 D’Ambrosio D, Albanesi C, Lang R, Girolomoni G, Sinigaglia F, Laudanna C. 2002. Quantitative differences in chemokine receptor engagement generate diversity in integrin-dependent

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

119.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

120.

121.

122.

123.

124.

125.

126.

127.

lymphocyte adhesion. J. Immunol. 169: 2303–12 Biedermann T, Schwarzler C, Lametschwandtner G, Thoma G, CarballidoPerrig N, et al. 2002. Targeting CLA/ E-selectin interactions prevents CCR4mediated recruitment of human Th2 memory cells to human skin in vivo. Eur. J. Immunol. 32:3171–80 Hub E, Rot A. 1998. Binding of RANTES, MCP-1, MCP-3, and MIP-1α to cells in human skin. Am. J. Pathol. 152:749–57 Peiper SC, Wang Z-X, Neote K, Martin AW, Showell HJ, et al. 1995. The Duffy antigen/receptor for chemokines (DARC) is expressed in endothelial cells of Duffy negative individuals who lack the erythrocyte receptor. J. Exp. Med. 181:1311–17 Hadley TJ, Peiper SC. 1997. From malaria to chemokine receptor: the emerging physiologic role of the Duffy blood group antigen. Blood 89:3077– 91 Nibbs RJ, Wylie SM, Yang J, Landau NR, Graham GJ. 1997. Cloning and characterization of a novel promiscuous human beta-chemokine receptor D6. J. Biol. Chem. 272:32078–83 Nibbs RJ, Graham GJ, Rot A. 2003. Chemokines on the move: Control by the chemokine “interceptors” Duffy blood group antigen and D6. Semin. Immunol. 15: 287–94 Luo H, Chaudhuri A, Zbrzezna V, He Y, Pogo AO. 2000. Deletion of the murine Duffy gene (Dfy) reveals that the Duffy receptor is functionally redundant. Mol. Cell. Biol. 20:3097–101 Dawson TC, Lentsch AB, Wang Z, Cowhig JE, Rot A, et al. 2000. Exaggerated response to endotoxin in mice lacking the Duffy antigen/receptor for chemokines (DARC). Blood 96:1681– 84 Fukuma N, Akimitsu N, Hamamoto H, Kusuhara H, Sugiyama Y, Sekimizu K.

128.

129.

130.

131.

132.

133.

P1: IKH

921

2003. A role of the Duffy antigen for the maintenance of plasma chemokine concentrations. Biochem. Biophys. Res. Commun. 303:137–39 Lee JS, Frevert CW, Wurfel MM, Peiper SC, Wong VA, et al. 2003. Duffy antigen facilitates movement of chemokine across the endothelium in vitro and promotes neutrophil transmigration in vitro and in vivo. J. Immunol. 170:5244–51 Baekkevold ES, Yamanaka T, Palframan RT, Carlsen HS, Reinholt FP, et al. 2001. The CCR7 ligand ELC (CCL19) is transcytosed in high endothelial venules and mediates T cell recruitment. J. Exp. Med. 193:1105–12 Gretz JE, Norbury CC, Anderson AO, Proudfoot AE, Shaw S. 2000. Lymphborne chemokines and other low molecular weight molecules reach high endothelial venules via specialized conduits while a functional barrier limits access to the lymphocyte microenvironments in lymph node cortex. J. Exp. Med. 192:1425–40 Stein JV, Rot A, Luo Y, Narasimhaswamy M, Nakano H, et al. 2000. The CC chemokine thymus-derived chemotactic agent 4 (TCA-4, secondary lymphoid tissue chemokine, 6Ckine, exodus-2) triggers lymphocyte functionassociated antigen 1-mediated arrest of rolling T lymphocytes in peripheral lymph node high endothelial venules. J. Exp. Med. 191:61–76 Palframan RT, Jung S, Cheng G, Weninger W, Luo Y, et al. 2001. Inflammatory chemokine transport and presentation in HEV: A remote control mechanism for monocyte recruitment to lymph nodes in inflamed tissues. J. Exp. Med. 194:1361–74 Nibbs RJ, Kriehuber E, Ponath PD, Parent D, Qin S, et al. 2001. The betachemokine receptor D6 is expressed by lymphatic endothelium and a subset of vascular tumors. Am. J. Pathol. 158:867– 77

12 Feb 2004

16:20

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

922

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

134. Fra AM, Locati M, Otero K, Sironi M, Signorelli P, et al. 2003. Cutting edge: scavenging of inflammatory CC chemokines by the promiscuous putatively silent chemokine receptor D6. J. Immunol. 170:2279–82 135. Janeway CA, Jr., Medzhitov R. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197–216 136. Girardin SE, Boneca IG, Carneiro LA, Antignac A, Jehanno M, et al. 2003. Nod1 detects a unique muropeptide from gram-negative bacterial peptidoglycan. Science 300:1584–87 137. Smiley ST, King JA, Hancock WW. 2001. Fibrinogen stimulates macrophage chemokine secretion through toll-like receptor 4. J. Immunol. 167:2887–94 138. Devaney JM, Greene CM, Taggart CC, Carroll TP, O’Neill SJ, McElvaney NG. 2003. Neutrophil elastase up-regulates interleukin-8 via toll-like receptor 4. FEBS Lett. 544:129–32 139. Biragyn A, Ruffini PA, Leifer CA, Klyushnenkova E, Shakhov A, et al. 2002. Toll-like receptor 4-dependent activation of dendritic cells by betadefensin 2. Science 298:1025–29 140. Alferink J, Lieberam I, Reindl W, Behrens A, Weiss S, et al. 2003. Compartmentalized production of CCL17 in vivo: strong inducibility in peripheral dendritic cells contrasts selective absence from the spleen. J. Exp. Med. 197:585–99 141. Maxion HK, Kelly KA. 2002. Chemokine expression patterns differ within anatomically distinct regions of the genital tract during Chlamydia trachomatis infection. Infect. Immun. 70:1538–46 142. Kajiya K, Inaki K, Tanaka M, Haga T, Kataoka H, Touhara K. 2001. Molecular bases of odor discrimination: Reconstitution of olfactory receptors that recognize overlapping sets of odorants. J. Neurosci. 21:6018–25 143. Chensue SW. 2001. Molecular machi-

144.

145.

146.

147.

148.

149.

150.

151.

152.

153.

nations: chemokine signals in hostpathogen interactions. Clin. Microbiol. Rev. 14:821–35 Mantovani A. 1999. The chemokine system: redundancy for robust outputs. Immunol. Today 20:254–57 Salazar-Mather TP, Hamilton TA, Biron CA. 2000. A chemokine-to-cytokine-tochemokine cascade critical in antiviral defense. J. Clin. Invest. 105:985–93 Huang Q, Liu D, Majewski P, Schulte LC, Korn JM, et al. 2001. The plasticity of dendritic cell responses to pathogens and their components. Science 294:870– 75 Power CA. 2003. Knock out models to dissect chemokine receptor function in vivo. J. Immunol. Methods 273:73–82 Mehrad B, Wiekowski M, Morrison BE, Chen SC, Coronel EC, et al. 2002. Transient lung-specific expression of the chemokine KC improves outcome in invasive aspergillosis. Am. J. Respir. Crit. Care Med. 166:1263–68 Shinkai K, Mohrs M, Locksley RM. 2002. Helper T cells regulate type-2 innate immunity in vivo. Nature 420:825– 29 Karpus WJ, Lukacs NW, Kennedy KJ, Smith WS, Hurst SD, Barrett TA. 1997. Differential CC chemokine-induced enhancement of T helper cell cytokine production. J. Immunol. 158:4129–36 Gu L, Tseng S, Horner RM, Tam C, Loda M, Rollins BJ. 2000. Control of TH2 polarization by the chemokine monocyte chemoattractant protein-1. Nature 404:407–11 Boring L, Gosling J, Chensue SW, Kunkel SL, Farese RV Jr., et al. 1997. Impaired monocyte migration and reduced type 1 (Th1) cytokine responses in C-C chemokine receptor 2 knockout mice. J. Clin. Invest. 100:2552–61 Peters W, Dupuis M, Charo IF. 2000. A mechanism for the impaired IFN-gamma production in C-C chemokine receptor 2 (CCR2) knockout mice: role of CCR2 in

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

154.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

155.

156.

157.

158.

159.

160.

161.

162.

163.

linking the innate and adaptive immune responses. J. Immunol. 165:7072–77 Luther SA, Cyster JG. 2001. Chemokines as regulators of T cell differentiation. Nat. Immunol. 2:102–7 Yoneyama H, Harada A, Imai T, Baba M, Yoshie O, et al. 1998. Pivotal role of TARC, a CC chemokine, in bacteriainduced fulminant hepatic failure in mice. J. Clin. Invest. 102:1933–41 Lalani AS, Barrett JW, McFadden I. 2000. Modulating chemokines: more lessons from viruses. Immunol. Today 21:100–6 Murphy PM. 2001. Viral exploitation and subversion of the immune system through chemokine mimicry. Nat. Immunol. 2:116–22 Alcami A. 2003. Viral mimicry of cytokines, chemokines and their receptors. Nat. Rev. Immunol. 3:36–50 Ansel KM, Cyster JG. 2001. Chemokines in lymphopoiesis and lymphoid organ development. Curr. Opin. Immunol. 13:172–79 Hjelmstrom P, Fjell J, Nakagawa T, Sacca R, Cuff CA, Ruddle NH. 2000. Lymphoid tissue homing chemokines are expressed in chronic inflammation. Am. J. Pathol. 156:1133–38 Weninger W, Carlsen HS, Goodarzi M, Moazed F, Crowley MA, et al. 2003. Na¨ıve T cell recruitment to nonlymphoid tissues: a role for endotheliumexpressed CCL21 in autoimmune disease and lymphoid neogenesis. J. Immunol. 170:4638–48 Grant AJ, Goddard S, AhmedChoudhury J, Reynolds G, Jackson DG, et al. 2002. Hepatic expression of secondary lymphoid chemokine (CCL21) promotes the development of portal-associated lymphoid tissue in chronic inflammatory liver disease. Am. J. Pathol. 160:1445–55 Shi K, Hayashida K, Kaneko M, Hashimoto J, Tomita T, et al. 2001. Lymphoid chemokine B cell-attracting

164.

165.

166.

166a.

167.

168. 169.

170.

171.

172.

P1: IKH

923

chemokine-1 (CXCL13) is expressed in germinal center of ectopic lymphoid follicles within the synovium of chronic arthritis patients. J. Immunol. 166:650– 55 Jones AP, Webb LM, Anderson AO, Leonard EJ, Rot A. 1995. Normal human sweat contains interleukin-8. J. Leukoc. Biol. 57:434–37 Michie CA, Tantscher E, Schall T, Rot A. 1998. Physiological secretion of chemokines in human breast milk. Eur. Cytokine Netw. 9:123–29 Hieshima K, Ohtani H, Shibano M, Izawa D, Nakayama T, et al. 2003. CCL28 has dual roles in mucosal immunity as a chemokine with broadspectrum antimicrobial activity. J. Immunol. 170:1452–61 Yang D, Biragyn A, Hoover DM, Lubkowski J, Oppenheim J. 2004. Multiple roles of antimicrobial defensins, cathelicidins, and eosinophil-derived neurotoxin in host defense. Annu. Rev. Immunol. 22: In press Huising MO, Stet RJ, Kruiswijk CP, Savelkoul HF, Lidy Verburg-van Kemenade BM. 2003. Molecular evolution of CXC chemokines: extant CXC chemokines originate from the CNS. Trends Immunol. 24:307–13 Gerard C, Rollins BJ. 2001. Chemokines and disease. Nat. Immunol. 2:108–15 Godessart N, Kunkel SL. 2001. Chemokines in autoimmune disease. Curr. Opin. Immunol. 13:670–75 Onuffer JJ, Horuk R. 2002. Chemokines, chemokine receptors and small-molecule antagonists: recent developments. Trends Pharmacol. Sci. 23:459–67 Schwarz MK, Wells TN. 2002. New therapeutics that modulate chemokine networks. Nat. Rev. Drug Discov. 1:347–58 Carter PH. 2002. Chemokine receptor antagonism as an approach to antiinflammatory therapy: ‘just right’ or plain wrong? Curr. Opin. Chem. Biol. 6:510–25

12 Feb 2004

16:20

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

924

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

172a. Mackay CR. 2001. Chemokines: immunology’s high impact factors. Nat. Immunol. 2:95–101 173. Gu L, Okada Y, Clinton SK, Gerard C, Sukhova GK, et al. 1998. Absence of monocyte chemoattractant protein-1 reduces atherosclerosis in low density lipoprotein receptor-deficient mice. Mol. Cell 2:275–81 174. Boring L, Gosling J, Cleary M, Charo IF. 1998. Decreased lesion formation in CCR2−/− mice reveals a role for chemokines in the initiation of atherosclerosis. Nature 394:894– 97 175. Boisvert WA, Santiago R, Curtiss LK, Terkeltaub RA. 1998. A leukocyte homologue of the IL-8 receptor CXCR-2 mediates the accumulation of macrophages in atherosclerotic lesions of LDL receptor-deficient mice. J. Clin. Invest. 101:353–63 176. Combadiere C, Potteaux S, Gao JL, Esposito B, Casanova S, et al. 2003. Decreased atherosclerotic lesion formation in CX3CR1/apolipoprotein E double knockout mice. Circulation 107:1009– 16 177. Szalai C, Duba J, Prohaszka Z, Kalina A, Szabo T, et al. 2001. Involvement of polymorphisms in the chemokine system in the susceptibility for coronary artery disease (CAD). Coincidence of elevated Lp(a) and MCP-1 -2518 G/G genotype in CAD patients. Atherosclerosis 158:233– 39 178. McDermott DH, Fong AM, Yang Q, Sechler JM, Cupples LA, et al. 2003. Chemokine receptor mutant CX3CR1M280 has impaired adhesive function and correlates with protection from cardiovascular disease in humans. J. Clin. Invest. 111:1241–50 178a. Moser B, Loetscher P. 2001. Lymphocyte traffic control by chemokines. Nat. Immunol. 2:123–28 179. Banchereau J, Briere F, Caux C, Davoust J, Lebecque S, et al. 2000. Immunobi-

180.

181.

182.

183.

184.

185.

186.

187.

188.

ology of dendritic cells. Annu. Rev. Immunol. 18:767–811 Cavanagh LL, von Andrian UH. 2002. Travellers in many guises: The origins and destinations of dendritic cells. Immunol. Cell Biol. 80:448–62 Janatpour MJ, Hudak S, Sathe M, Sedgwick JD, McEvoy LM. 2001. Tumor necrosis factor-dependent segmental control of MIG expression by high endothelial venules in inflamed lymph nodes regulates monocyte recruitment. J. Exp. Med. 194:1375–84 Randolph GJ. 2001. Dendritic cell migration to lymph nodes: cytokines, chemokines, and lipid mediators. Semin. Immunol. 13:267–74 Sallusto F, Palermo B, Lenig D, Miettinen M, Matikainen S, et al. 1999. Distinct patterns and kinetics of chemokine production regulate dendritic cell function. Eur. J. Immunol. 29:1617– 25 Sozzani S, Allavena P, Vecchi A, Mantovani A. 1999. The role of chemokines in the regulation of dendritic cell trafficking. J. Leukoc. Biol. 66:1–9 Martin-Fontecha A, Sebastiani S, Hopken UE, Uguccioni M, Lipp M, et al. 2003. Regulation of dendritic cell migration to the draining lymph node: impact on T lymphocyte traffic and priming. J. Exp. Med. 198:615–21 Robbiani DF, Finch RA, Jager D, Muller WA, Sartorelli AC, Randolph GJ. 2000. The leukotriene C(4) transporter MRP1 regulates CCL19 (MIP-3β, ELC)-dependent mobilization of dendritic cells to lymph nodes. Cell 103: 757–68 Gunn MD, Kyuwa S, Tam C, Kakiuchi T, Matsuzawa A, et al. 1999. Mice lacking expression of secondary lymphoid organ chemokine have defects in lymphocyte homing and dendritic cell localization. J. Exp. Med. 189:451–60 Bousso P, Robey E. 2003. Dynamics of CD8(+) T cell priming by dendritic cells

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

CHEMOKINES IN IMMUNITY

189.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

190.

191.

192.

193.

194.

195.

196.

197.

in intact lymph nodes. Nat. Immunol. 4:579–85 Forster R, Schubel A, Breitfeld D, Kremmer E, Renner-Muller I, et al. 1999. CCR7 coordinates the primary immune response by establishing functional microenvironments in secondary lymphoid organs. Cell 99:23–33 Gunn MD, Tangemann K, Tam C, Cyster JG, Rosen SD, Williams LT. 1998. A chemokine expressed in lymphoid high endothelial venules promotes the adhesion and chemotaxis of naive T lymphocytes. Proc. Natl. Acad. Sci. USA 95:258–63 Girard J-P, Springer TA. 1995. High endothelial venules (HEVs): Specialized endothelium for lymphocyte migration. Immunol. Today 16:449–57 Nakano H, Tamura T, Yoshimoto T, Yagita H, Miyasaka M, et al. 1997. Genetic defect in T lymphocyte-specific homing into peripheral lymph nodes. Eur. J. Immunol. 27:215–21 Okada T, Ngo VN, Ekland EH, Forster R, Lipp M, et al. 2002. Chemokine requirements for B cell entry to lymph nodes and Peyer’s patches. J. Exp. Med. 196:65– 75 Warnock RA, Campbell JJ, Dorf ME, Matsuzawa A, McEvoy LM, Butcher EC. 2000. The role of chemokines in the microenvironmental control of T versus B cell arrest in Peyer’s patch high endothelial venules. J. Exp. Med. 191:77– 88 Ma Q, Jones D, Borghesan PR, Segal RA, Nagasawa T, et al. 1998. Impaired B-lymphopoiesis, myelopoiesis, and derailed cerebellar neuron migration in CXCR4- and SDF-1-deficient mice. Proc. Natl. Acad. Sci. USA 95:9448–53 Tachibana K, Hirota S, Lizasa H, Yoshida H, Kawabata K, et al. 1998. The chemokine receptor CXCR4 is essential for vascularization of the gastrointestinal tract. Nature 393:591–94 Ma Q, Jones D, Springer TA. 1999.

198.

199.

200.

201.

202.

203.

204.

205.

206.

207.

P1: IKH

925

The chemokine receptor CXCR4 is required for the retention of B lineage and granulocytic precursors within the bone marrow microenvironment. Immunity 10:463–71 Nolte MA, Hamann A, Kraal G, Mebius RE. 2002. The strict regulation of lymphocyte migration to splenic white pulp does not involve common homing receptors. Immunology 106:299–307 Chaffin KE, Perlmutter RM. 1991. A pertussis toxin-sensitive process controls thymocyte emigration. Eur. J. Immunol. 21:2565–73 Forster R, Mattis AE, Kremmer E, Wolf E, Brem G, Lipp M. 1996. A putative chemokine receptor, BLRI, directs B cell migration to defined lymphoid organs and specific anatomic compartments of the spleen. Cell 87:1037–47 Ansel KM, Ngo VN, Hyman PL, Luther SA, Forster R, et al. 2000. A chemokinedriven positive feedback loop organizes lymphoid follicles. Nature 406:309–14 Miller MJ, Wei SH, Parker I, Cahalan MD. 2002. Two-photon imaging of lymphocyte motility and antigen response in intact lymph node. Science 296:1869–73 Miller MJ, Wei SH, Cahalan MD, Parker I. 2003. Autonomous T cell trafficking examined in vivo with intravital twophoton microscopy. Proc. Natl. Acad. Sci. USA 100:2604–9 Gunzer M, Schafer A, Borgmann S, Grabbe S, Zanker KS, et al. 2000. Antigen presentation in extracellular matrix: interactions of T cells with dendritic cells are dynamic, short lived, and sequential. Immunity 13:323–32 O’Garra A, Arai N. 2000. The molecular basis of T helper 1 and T helper 2 cell differentiation. Trends Cell. Biol. 10:542– 50 Glimcher LH, Murphy KM. 2000. Lineage commitment in the immune system: the T helper lymphocyte grows up. Genes Dev. 14:1693–711 Garside P, Ingulli E, Merica RR, Johnson

12 Feb 2004

16:20

926

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

208.

209.

210.

211.

212.

213.

214.

215.

216.

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

JG, Noelle RJ, Jenkins MK. 1998. Visualization of specific B and T lymphocyte interactions in the lymph node. Science 281:96–99 Breitfeld D, Ohl L, Kremmer E, Ellwart J, Sallusto F, et al. 2000. Follicular B helper T cells express CXC chemokine receptor 5, localize to B cell follicles, and support immunoglobulin production. J. Exp. Med. 192:1545–52 Schaerli P, Willimann K, Lang AB, Lipp M, Loetscher P, Moser B. 2000. CXC chemokine receptor 5 expression defines follicular homing T cells with B cell helper function. J. Exp. Med. 192:1553– 62 Ingulli E, Ulman DR, Lucido MM, Jenkins MK. 2002. In situ analysis reveals physical interactions between CD11b+ dendritic cells and antigen-specific CD4 T cells after subcutaneous injection of antigen. J. Immunol. 169:2247–52 Springer TA. 1994. Traffic signals for lymphocyte recirculation and leukocyte emigration: The multi-step paradigm. Cell 76:301–14 von Andrian UH, Mackay CR. 2000. Tcell function and migration. Two sides of the same coin. N. Engl. J. Med. 343:1020–34 Xie H, Lim YC, Luscinskas FW, Lichtman AH. 1999. Acquisition of selectin binding and peripheral homing properties by CD4(+) and CD8(+) T cells. J. Exp. Med. 189:1765–76 Weninger W, Crowley MA, Manjunath N, von Andrian UH. 2001. Migratory properties of naive, effector, and memory CD8(+) T cells. J. Exp. Med. 194:953– 66 Syrbe U, Siveke J, Hamann A. 1999. Th1/Th2 subsets: distinct differences in homing and chemokine receptor expression? Springer Semin. Immunopathol. 21:263–85 Campbell DJ, Butcher EC. 2002. Rapid acquisition of tissue-specific homing phenotypes by CD4(+) T cells activated

217.

218.

219.

220.

221.

222.

223.

224.

225.

in cutaneous or mucosal lymphoid tissues. J. Exp. Med. 195:135–41 Kunkel EJ, Butcher EC. 2002. Chemokines and the tissue-specific migration of lymphocytes. Immunity 16:1–4 Wurbel MA, Philippe JM, Nguyen C, Victorero G, Freeman T, et al. 2000. The chemokine TECK is expressed by thymic and intestinal epithelial cells and attracts double- and single-positive thymocytes expressing the TECK receptor CCR9. Eur. J. Immunol. 30:262–71 Carramolino L, Zaballos A, Kremer L, Villares R, Martin P, et al. 2001. Expression of CCR9 beta-chemokine receptor is modulated in thymocyte differentiation and is selectively maintained in CD8(+) T cells from secondary lymphoid organs. Blood 97:850–57 Mora JR, Bono MR, Manjunath N, Weninger W, Cavanagh LL, et al. 2003. Selective imprinting of gut-homing T cells by Peyer’s patch dendritic cells. Nature 424:88–93 Stagg AJ, Kamm MA, Knight SC. 2002. Intestinal dendritic cells increase T cell expression of α4β7 integrin. Eur. J. Immunol. 32:1445–54 Luther SA, Ansel KM, Cyster JG. 2003. Overlapping roles of CXCL13, interleukin 7 receptor alpha, and CCR7 ligands in lymph node development. J. Exp. Med. 197:1191–98 Kunkel EJ, Kim CH, Lazarus NH, Vierra MA, Soler D, et al. 2003. CCR10 expression is a common feature of circulating and mucosal epithelial tissue IgA Absecreting cells. J. Clin. Invest. 111:1001– 10 Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. 1999. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401:708–12 Iezzi G, Scheidegger D, Lanzavecchia A. 2001. Migration and function of antigen-primed nonpolarized T lymphocytes in vivo. J. Exp. Med. 193:987–93

12 Feb 2004

16:20

AR

AR210-IY22-29.tex

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CHEMOKINES IN IMMUNITY 226. Campbell JJ, Butcher EC. 2000. Chemokines in tissue-specific and microenvironment-specific lymphocyte homing. Curr. Opin. Immunol. 12:336–41 227. Reinhardt RL, Bullard DC, Weaver CT, Jenkins MK. 2003. Preferential accumulation of antigen-specific effector CD4 T cells at an antigen injection site involves CD62E-dependent migration but not local proliferation. J. Exp. Med. 197:751–62 228. Masopust D, Vezys V, Marzo AL, Lefrancois L. 2001. Preferential localization of effector memory cells in nonlymphoid tissue. Science 291:2413–17 229. Manjunath N, Shankar P, Wan J, Weninger W, Crowley MA, et al. 2001. Effector differentiation is not prerequisite for generation of memory cytotoxic T lymphocytes. J. Clin. Invest. 108:871– 78 230. Wherry EJ, Teichgraber V, Becker TC, Masopust D, Kaech SM, et al. 2003. Lineage relationship and protective immunity of memory CD8 T cell subsets. Nat. Immunol. 4:225–34 231. Lauvau G, Vijh S, Kong P, Horng T, Kerksiek K, et al. 2001. Priming of memory but not effector CD8 T cells by a killed bacterial vaccine. Science 294:1735–39 232. Mackay CR, Marston WL, Dudler L. 1990. Naive and memory T cells show distinct pathways of lymphocyte recirculation. J. Exp. Med. 171:801–17 233. Campbell JJ, Murphy KE, Kunkel EJ, Brightling CE, Soler D, et al. 2001. CCR7 expression and memory T cell diversity in humans. J. Immunol. 166:877– 84 234. Vlahakis SR, Villasis-Keever A, Gomez T, Vanegas M, Vlahakis N, Paya CV. 2002. G protein-coupled chemokine receptors induce both survival and apoptotic signaling pathways. J. Immunol. 169:5546–54 235. Foxman EF, Campbell JJ, Butcher EC. 1997. Multistep navigation and the com-

236.

237.

238.

239.

240.

241.

242.

243.

244.

P1: IKH

927

binatorial control of leukocyte chemotaxis. J. Cell Biol. 139:1349–60 Foxman EF, Kunkel EJ, Butcher EC. 1999. Integrating conflicting chemotactic signals. The role of memory in leukocyte navigation. J. Cell Biol. 147:577– 88 Ali H, Richardson RM, Haribabu B, Snyderman R. 1999. Chemoattractant receptor cross-desensitization. J. Biol. Chem. 274:6027–30 Hauser CJ, Fekete Z, Goodman ER, Kleinstein E, Livingston DH, Deitch EA. 1999. CXCR2 stimulation primes CXCR1 [Ca2+]i responses to IL-8 in human neutrophils. Shock 12:428–37 Blanpain C, Migeotte I, Lee B, Vakili J, Doranz BJ, et al. 1999. CCR5 binds multiple CC-chemokines: MCP-3 acts as a natural antagonist. Blood 94:1899–905 Loetscher P, Pellegrino A, Gong JH, Mattioli I, Loetscher M, et al. 2001. The ligands of CXC chemokine receptor 3, I-TAC, Mig, and IP10, are natural antagonists for CCR3. J. Biol. Chem. 276:2986–91 Nibbs RJ, Salcedo TW, Campbell JD, Yao XT, Li Y, et al. 2000. CC chemokine receptor 3 antagonism by the beta-chemokine macrophage inflammatory protein 4, a property strongly enhanced by an amino-terminal alanine-methionine swap. J. Immunol. 164:1488–97 Ogilvie P, Bardi G, Clark-Lewis I, Baggiolini M, Uguccioni M. 2001. Eotaxin is a natural antagonist for CCR2 and an agonist for CCR5. Blood 97:1920–24 Van Damme J, Struyf S, Wuyts A, Van Coillie E, Menten P, et al. 1999. The role of CD26/DPP IV in chemokine processing. Chem. Immunol. 72:42–56 Walz A, Meloni F, Clark-Lewis I, von Tscharner V, Baggiolini M. 1991. [Ca2+] changes and respiratory burst in human neutrophils and monocytes induced by NAP-1/interleukin-8, NAP-2, and gro/MGSA. J. Leukoc. Biol. 50:279–86

12 Feb 2004

16:20

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

928

AR

ROT

AR210-IY22-29.tex

¥

AR210-IY22-29.sgm

LaTeX2e(2002/01/18)

P1: IKH

VON ANDRIAN

245. Chuntharapai A, Lee J, Hebert CA, Kim KJ. 1994. Monoclonal antibodies detect different distribution patterns of IL-8 receptor A and IL-8 receptor B on human peripheral blood leukocytes. J. Immunol. 153:5682–88 246. Bishayi B, Samanta AK. 1996. Identification and characterization of specific receptor for interleukin-8 from the surface of human monocytes. Scand. J. Immunol. 43:531–36 247. Weber KS, von Hundelshausen P, ClarkLewis I, Weber PC, Weber C. 1999. Differential immobilization and hierarchical involvement of chemokines in monocyte arrest and transmigration on inflamed endothelium in shear flow. Eur. J. Immunol. 29:700–12 248. Huo Y, Weber C, Forlow SB, Sperandio M, Thatte J, et al. 2001. The chemokine KC, but not monocyte chemoattractant protein-1, triggers monocyte arrest on early atherosclerotic endothelium. J. Clin. Invest. 108:1307–14 249. Gerszten RE, Garcia-Zepeda EA, Lim YC, Yoshida M, Ding HA, et al. 1999. MCP-1 and IL-8 trigger firm adhesion of monocytes to vascular endothelium

250.

251.

252.

253.

254.

under flow conditions. Nature 398:718– 23 Zernecke A, Weber KS, Erwig LP, Kluth DC, Schroppel B, et al. 2001. Combinatorial model of chemokine involvement in glomerular monocyte recruitment: role of CXC chemokine receptor 2 in infiltration during nephrotoxic nephritis. J. Immunol. 166:5755–62 Bonecchi R, Facchetti F, Dusi S, Luini W, Lissandrini D, et al. 2000. Induction of functional IL-8 receptors by IL-4 and IL-13 in human monocytes. J. Immunol. 164:3862–69 Bonecchi R, Polentarutti N, Luini W, Borsatti A, Bernasconi S, et al. 1999. Upregulation of CCR1 and CCR3 and induction of chemotaxis to CC chemokines by IFN-gamma in human neutrophils. J. Immunol. 162:474–79 Jinquan T, Jacobi HH, Jing C, Millner A, Sten E, et al. 2003. CCR3 expression induced by IL-2 and IL-4 functioning as a death receptor for B cells. J. Immunol. 171:1722–31 Sozzani S, Allavena P, Vecchi A, Mantovani A. 2000. Chemokines and dendritic cell traffic. J. Clin. Immunol. 20:151–60

Rot.qxd

3/12/2004

2:19 PM

Page 1

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CHEMOKINES IN IMMUNITY

C-1

Figure 1 Human chemokines and their receptors. Solid and dashed lines connect receptors with their agonists and antagonists, respectively, and are color coded to correspond with the colors of individual receptor hubs. The dashed-dotted line that connects CXCR3 with CCL11 represents a nonagonist-nonantagonist binding interaction. The bars next to individual chemokine numbers reflect the colors assigned to their apposite receptors.

3/12/2004

C-2 ROT

2:19 PM

■

Figure 2 Signaling from chemokine receptors, through activation of four small GTPases, to leukocyte chemotaxis and integrin activation. Red arrows lead to the activation of Rac, required for the actin polymerization at the pseudopod; green arrows lead to the activation of Cdc42, required for the establishment of orientation machinery at the pseudopod; blue arrows lead to the activation of Rho, required for the retraction of the uropod; brown arrows lead to the activation of Rap, required for inside-out signaling of integrins. All abbreviations and individual steps are described in the text.

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rot.qxd Page 2

VON ANDRIAN

3/12/2004 2:19 PM

See legend on next page

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rot.qxd Page 3

CHEMOKINES IN IMMUNITY C-3

Rot.qxd

3/12/2004

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-4

ROT

2:19 PM

■

Page 4

VON ANDRIAN

Figure 3 Suggested involvement of interceptors Duffy antigen and D6 in chemokine interactions with ECs of tissue venules, afferent lymphatic vessels, and high endothelial venules (HEV) of draining lymph nodes (LN). Chemokines (green and blue kernels) diffuse into the bloodstream through the junctions between ECs (a). Alternatively, CXC and CC chemokines are internalized and transcytosed by interceptor Duffy antigen (CIPR-1) on ECs (b). This optimally supports their luminal presentation by GAGs to the chemokine receptors (CKR) on rolling leukocytes (c), resulting in activation of integrins, firm leukocyte adhesion, and emigration. Conversely, chemokines in plasma stimulate the free-floating leukocytes (d), resulting in deactivation of their emigratory responsiveness. Plasma chemokines bind to Duffy antigen on erythrocytes (e), which serves both as a sink as well as a long-term depot for chemokines in the blood. In addition, chemokines may diffuse through the afferent lymphatics into the draining LNs (f) where, via a conduit system, they reach the HEV, are transcytosed and presented on the luminal surface (g), and can recruit pertinent leukocytes into the LN. Chemokine interceptor D6 (CIPR-2), expressed by lymphatic ECs, binds a subset of CC chemokines, internalizes and targets them for degradation (h), and thus offsets the remote control of leukocyte recruitment into the LN by the chemokines derived at the periphery.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:891-928. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18) P1: IKH 10.1146/annurev.immunol.22.012703.104622

Annu. Rev. Immunol. 2004. 22:929–79 doi: 10.1146/annurev.immunol.22.012703.104622 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on December 16, 2003

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS Sidney Pestka,1,2,3 Christopher D. Krause,1 Devanand Sarkar,4 Mark R. Walter,5 Yufang Shi,1,3 and Paul B. Fisher4 1

Department of Molecular Genetics, Microbiology, and Immunology, Robert Wood Johnson Medical School, University of Medicine and Dentistry of New Jersey, Piscataway, New Jersey 08854; email: [email protected], [email protected], [email protected] 2 PBL Biomedical Laboratories, Piscataway, New Jersey 08854 3 Cancer Institute of New Jersey, New Brunswick, New Jersey 08901 4 Department of Pathology, College of Physicians and Surgeons, Columbia University, New York, NY 10032; email: [email protected], [email protected] 5 Department of Microbiology and Center for Biophysical Sciences and Engineering, University of Alabama at Birmingham, Birmingham, Alabama 35294; email: [email protected]

Key Words Class 2 cytokines, interferons, signal transduction, immune regulation, receptor structure, cytokine structure, cytokine evolution, receptor evolution ■ Abstract The Class 2 α-helical cytokines consist of interleukin-10 (IL-10), IL-19, IL-20, IL-22, IL-24 (Mda-7), and IL-26, interferons (IFN-α, -β, -ε, -κ, -ω, -δ, -τ , and -γ ) and interferon-like molecules (limitin, IL-28A, IL-28B, and IL-29). The interaction of these cytokines with their specific receptor molecules initiates a broad and varied array of signals that induce cellular antiviral states, modulate inflammatory responses, inhibit or stimulate cell growth, produce or inhibit apoptosis, and affect many immune mechanisms. The information derived from crystal structures and molecular evolution has led to progress in the analysis of the molecular mechanisms initiating their biological activities. These cytokines have significant roles in a variety of pathophysiological processes as well as in regulation of the immune system. Further investigation of these critical intercellular signaling molecules will provide important information to enable these proteins to be used more extensively in therapy for a variety of diseases.

INTRODUCTION This review provides a comprehensive analysis of IL-10 and related cytokines and receptors. Comparison of the interferons (Type I and Type II), the first cytokines discovered, with the rest of the Class 2 cytokines and receptors provides a 0732-0582/04/0423-0929$14.00

929

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

930

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

fundamental base for understanding the functions, pathways, evolution, and structure of all of these critical intercellular signaling molecules. Compared to the first few cytokines discovered, cloned, purified, and characterized by laborious methodology governed by the technology before DNA sequences began to fill databases and before the genomes of humans, Caenorhabditis elegans, and numerous other organisms became available, more recent discoveries have proceeded rapidly, partly in silico, by searching the cDNA and genome databases. The latest discoveries of these cytokines and receptors have outpaced our ability to determine their functions and mechanisms of action. Owing to their significance to biology and medicine, the Class 2 cytokines have been intensively studied by scientists in many fields, and a wealth of information has been accumulated. The main points about their biological functions, receptor usage, signaling pathways, genetic background, and homology are summarized in tabular form. The Class 2 cytokines will be discussed in the following order: IL-10, interferons and interferon-like molecules, IL-19, IL-20, IL-22, IL-24 (Mda-7), IL-26 (AK155), IL-28A, IL-28B, and IL-29. The sections on evolution and crystal structures of these molecules fit into the goal of integrating this group of related molecules despite the fact that they exhibit very different functions, sometimes even antagonistic functions. The evolutionary analysis and the crystal structures provide insights and perspective into the immunology of these molecules. Surely, in addition to increasing our basic understanding of their mechanisms of actions and biological functions, further exploration of this important group of cytokines will lead to better clinical applications for the treatment of autoimmunity, viral infections, cancer, and many other diseases.

INTERLEUKIN-10 Interleukin-10 (IL-10) was initially characterized as cytokine synthesis inhibitory factor (CSIF) (1) from con A-stimulated Th2 cells that inhibited the production of cytokines such as IL-2, TNF-α, IFN-γ , and GM-CSF by Th1 cells, but not Th2 cells, in response to antigens presented by APC. Immediately after this initial report, B cells and B cell lymphoma were also found to produce IL-10 (2), and a homologue of IL-10, BCRFI, was discovered from the Epstein-Barr virus genome (3). It was therefore proposed that IL-10 and its viral homologue play a key role in suppressing the host immune system, allowing the growth of lymphoma and the survival of virus in the host. As elevated levels of cytokines were associated with immune diseases, there quickly developed and has since remained overwhelming interest in this cytokine.

Biological Activities IL-10 initiates a wide variety of activities when it binds to its cellular receptor complex. The mechanism of IL-10 inhibition of cytokine production was initially

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

931

believed to be an inhibition of the antigen-presentation capacity of macrophages (4) and dendritic cells (5–8). Further investigations revealed that IL-10 also plays important roles in blocking cytokine production (5), the expression of costimulatory molecules, including CD80, CD86, and MHC Class II (9, 10), and chemokine secretion (11, 12). It also modifies chemokine receptor expression (13, 14), increases β2-integrin ligand (e.g., ICAM-1) expression (10), and increases radical generation (15). This therefore limits the magnitude of activation and concurrent cytokine release by T cells during the a specific immune response. IL-10 partially inhibits the induction of activities initiated by other cytokines, notably IFN-γ , IL-2, TNF-α, and IL-4. Except affecting T cell responses, IL-10 has also been shown to promote the development of B1 cells (16), to enhance the survival of B cells, T cells, and tumor cells (17, 18), and, paradoxically, to induce apoptosis in chronic B cell leukemia cells (19). Recently, IL-10 was discovered to promote the activity of NK cells (20). The principle function of IL-10, however, is to limit the magnitude of an immune response, as mice lacking IL-10 exhibit spontaneous enterocolitis and other symptoms akin to Crohn’s disease (21). More relevant to peripheral immunity, mice lacking IL-10 by gene deletion have increased Th1 responses (22) and, consequentially, enhanced clearance of bacterial (23, 24), fungal (25), or toxoplasmic (26) infection. These mice also show exaggerated asthmatic (27) and allergic responses (28, 29). A recent study showed that IL-10 promotes the differentiation of CD11clowCD45RBhigh dendritic cells that lead to the differentiation of Type 1 regulatory cells both in vitro and in vivo (30). This observation may also explain the immunosuppressive effect of IL-10.

Viral Interleukin-10 Homologs Immediately after the cDNA’s encoding murine and human IL-10 were cloned, an Epstein-Barr viral homolog of IL-10, BCRF1, was discovered (3, 31). Since then, a number of homologs of IL-10 have been found in the genomes of several herpesviruses (32–34), a poxvirus (35), and primate cytomegaloviruses (36, 37). Repeated independent capture of cellular IL-10 genes or mRNAs by viral genomes (38) strongly suggests evolutionary advantages to viruses producing their own sources of IL-10 during host infection; however, the in vivo roles of viral IL-10 homologs in the viral life cycle, in immune evasion and/or in helping virus-infected cells to survive immune surveillance remain to be defined. It is nevertheless sensible to speculate that IL-10 and its viral homologs play a key role in suppressing the host immune system, allowing the growth of virus-induced tumors and survival of viruses in the host. Generally, viral IL-10 (vIL-10) proteins all seem to interact with the same receptor as does cellular IL-10 (cIL-10), and induce a subset of activities induced by cIL-10. Interestingly, ovIL-10 (orf poxvirus IL-10) seems to have maintained nearly all cIL-10 activities (39, 40). In general, viruses produce and release vIL-10 from their host cells during the lytic phase of their life cycle (41, 42), when the chance that the infection would be detected by the immune system is greatest.

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

932

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

The viruses encoding vIL-10 homologs are more likely to persist in the host. Interestingly, with the exception of cytomegalovirus-encoded IL-10 (cmvIL-10), all vIL-10 homologs have high similarity to cIL-10, especially in their COOHtermini and are encoded within a single exon, suggesting recent acquisition of the IL-10 mRNAs into viral genomes. On the other hand, cmvIL-10 appears to have been captured much earlier as a gene, as its sequence diverges greatly from that of IL-10 and contains introns, some of which are conserved among cIL-10 genes. One can speculate that, because cmvIL-10 homologs can be found in several primate CMV genomes (36, 37), the capture of IL-10 by CMV preceded the coevolution of primates and their respective CMVs. Comparing the sequences of vIL-10 and the corresponding cIL-10s, the greatest divergence occurs in the NH2 terminus of the molecules (38, 43).

Genes Encoding IL-10, IL-10R1, and IL-10R2 The structure of the gene encoding IL-10 is expected to be highly conserved for three reasons: (a) the IL-10 gene in the pufferfish Fugu rubripes and Tetraodon nigroviridis also has four introns and five exons (44, 214), (b) the genes encoding human IL-10-like homologs all have similar exon-intron structures (45), and (c) the exon-exon borders occur in similar parts of the encoded proteins (45). The IL-10 gene locus spans about 2 kb in the human and piscine genomes. The first exon encodes the signal peptide and the A helix. The second exon encodes the AB loop and Helix B. The third exon encodes the C and D helices. The fourth exon encodes the DE loop and the E helix. Finally, the fifth exon encodes the F helix, the COOH tail of IL-10, and an untranslated segment containing AUUUA response elements, which, when bound by AUF1, reduce the stability of the IL-10 mRNA (46). Interestingly, exon-intron borders occur along the COOH-termini of the helices. The structures of the genes encoding IL-10R1 and IL-10R2 are highly similar to each other and to other members of the Class 2 cytokine receptor family (45), and even to members of the Class 1 cytokine receptor and growth hormone receptor families (47). Both encompass about 30–40 kDa of genomic DNA in mammalian genomes. The first exon encodes the signal peptide and the beginning amino acids of the mature peptide. The second and third exons encode the first SD100 domain. The fourth and fifth exons encode the second SD100 domain. The sixth exon encodes primarily the transmembrane domain and (in IFN-γ R2) most of the residues encoding the Jak association site (48). The seventh exon encodes the remaining intracellular domain (including Stat recruitment sites) and untranslated sequence, which may play a role in mRNA stability of the receptor chains. It is reasonable to speculate that the tandem duplication of SD100 domains in the Class 2 cytokine receptors derived from duplication of two exons encoding a single SD100 domain. So far no splice variants are known of IL-10 mRNA, IL-10R1 mRNA, or IL10R2 mRNA. One could speculate however, that, like the IL-22BP, a variant of

15 Mar 2004

16:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

933

either IL-10R1 or IL-10R2, elimination of exons 6 and/or 7 would encode a soluble receptor. Splice variants of IL-10 receptors could play a physiological role, as splice variants of Type I IFN receptors have been observed and may play functional roles. One splice variant of IFN-αR2 with exon 6 missing encodes a soluble receptor that has been found in human blood and urine (49) and may function as a receptor antagonist. Another splice variant of IFN-αR2 employs an alternate exon 7 missing the Jak1 association site (50–52) and may function as a membrane-bound receptor antagonist. Two splice variants missing the transmembrane domain of IFN-αR1 also exist and may function as soluble receptors (53). Furthermore, variants in which one or more exons of the extracellular domain were removed could encode receptors that are nonfunctional or have altered abilities to interact with their ligands. This has been observed for a splice variant of IFN-αR1 that lacks the second of four SD100 domains (47, 54). Finally, splice variants have been noted for IL-28R1 that delete the transmembrane domain (exon 6) and delete exons 5–7 (55, 56). These latter IL-28R1 variants may functions as soluble IL-28 receptors.

Signal Transduction The activation of the IL-10 receptor complex results not only in the inhibition of the synthesis of several cytokine genes, but it also prevents several cytokines from inducing their biological activities on their target cells, stifling activities normally induced by these cytokines. Similarly, pretreatment with IFN-γ of IL-10responsive cells causes them to mount a weaker response to IL-10. The mechanism for these activities are similar and have recently been elucidated in the context of the current model of the activation of the IL-10 receptor complex described below. The IL-10 receptor complex on cells is composed of four transmembrane polypeptides: two chains of IL-10R1 that bind ligand and two chains of IL-10R2 that initiate signal transduction (Figure 1). Owing to its structural homology to the IFN-γ receptor complex, shown to be preassembled (57), the IL-10 receptor complex was predicted to be preassembled as well. Fluorescence resonance energy transfer studies support this prediction (C. Krause and S. Pestka, unpublished data). Jak1 is constitutively bound to IL-10R1. This constitutive interaction is mediated by the peptide sequence SVLLFKK at the NH2-terminus of the intracellular domain of IL-10R1 with the FERM domain of Jak1 (51, 52). Tyk2 constitutively associates with the IL-10R2 chain (58). Upon activation of the IL10 receptor complex by IL-10, Jak1 and Tyk2 are activated (59), presumably by cross-phosphorylation of two tyrosine residues on the intracellular domain of IL10R1 (60). These phosphorylated tyrosines (at amino acids 427 and 477) mediate the direct interaction of Stat3 to the IL-10 receptor complex. Subsequently, phosphorylation of Stat3, Stat1, and Stat5 by Jak1 and Tyk2 occurs, and homodimers and heterodimers of these Stat proteins form (61) and translocate to the nucleus to drive transcription of Stat3-responsive genes. Among genes activated by IL-10 are SOCS-1 and SOCS-3 (62). SOCS-1 is known to be the major physiological inhibitor of IFN-γ -, IL-10-, and IL-4-induced signal transduction (62–64), and it

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

934

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

has been shown that SOCS-1 silences signaling by binding to the activation loop of Jak kinases and physically occluding the kinase active site, preventing it from phosphorylating its substrates (65). This inhibition is presumably relieved when the Elongin BC complex, recruited to the SOCS box of SOCS-1 and SOCS-3, targets itself and SOCS-1 or SOCS-3 for proteolytic degradation (66–68). This conveniently explains how IL-10 inhibits the effects of other cytokine signaling systems, and how IFN-γ can inhibit IL-10 signaling as well. Furthermore, induction of SOCS-3 by IL-10 inhibits many aspects of gp130 signaling initiated by IL-6 (69). SOCS induction and subsequent cytokine receptor inhibition may mediate many anti-inflammatory activities of IL-10. Other anti-inflammatory activities of IL-10 could be explained by its ability to stabilize IκB∀ in the cytoplasm (70) and thereby prevent LPS-induced NF-κB activation (71) by inhibiting the IκB kinase (72), which, by phosphorylating the NF-κB inhibitor IκBα, induces the proteolysis of IκBα and the release of NF-κB. Recently, the mechanism by which IL-10 inhibits the activity of metalloproteases has begun to be elucidated in cancer cell lines. IL-10 induces the tyrosine phosphorylation and activation of a transcription factor called IL-10E1 (73), which is recruited to the receptor complex at the sites of IL-10R1 tyrosine phosphorylation and is presumably also activated by the Jak kinases (74). IL-10E1 binds to an element upstream of a metalloprotease inhibitor gene and drives its transcription. Thus, there may be several types of transcription factors activated by IL-10, consistent with the concept of multiple signaling pathways initiated by IL-10. Some of these pathways work in concert to elicit IL-10 activities, and some of these pathways may exhibit their activities independently of other activities. This has already been observed with IFN-γ signaling, when different subsets of activities are observed in response to IFN-γ signaling even when Stat1 is not present in cells (75–78) or is not able to be activated by IFN-γ (77). In support of this, a region at the COOH terminus of the intracellular domain (containing a functionally important serine residue) that is not involved in Stat3 recruitment is known to be required for anti-inflammatory activity of IL-10 (79). Deletion of this area does not influence Stat3 phosphorylation but eliminates observed anti-inflammatory activities. However, Stat3 must also be phosphorylated for anti-inflammatory activities of IL-10 to be observed, suggesting that this particular pathway must cooperate with Stat3 activation for anti-inflammatory activities in response to IL-10. Recently, a flurry of work concerning the cellular immunology of IL-10mediated immunosuppression has occurred, especially with respect to the maternalfetal interface during human and murine pregnancy. Expression of IL-10 and IL10R1 increases during the early stages of healthy pregnancy (80), suggesting an involvement of IL-10 activity in the establishment and maintenance of pregnancy. Reduced levels of IL-10 in endometrial or placental tissues are associated with recurrent spontaneous abortions (81) and other pregnancy-associated immune complications such as preeclampsia (82), suggesting that IL-10 plays an important role in maintenance of pregnancy. IL-10 is expressed and constitutively released within endometrial tissues by γ δT cells (83) and decidual macrophages (84). IL-10 from

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

935

these sources presumably helps to inhibit elements of the immune system that hone into the endometrial tissue. IL-10 could be secreted from fetal macrophages found in the placenta, possibly to promote the immune refractory state of the trophoblast cells by activities such as induction of the inhibitory HLA-G (85) to inhibit natural killer cell activity, reduction of Class II MHC on the trophoblast cell surface, and inhibition of effects of maternal cytokines on the trophoblast cells, by mechanisms similar to those described above and reviewed elsewhere (43). IL-10 is recognized as a key cytokine released by T cells with regulatory activities. As research progresses, it is becoming evident that regulatory T cells may be as diverse as the effector T cells they regulate. Just as many subpopulations of T cells secrete common groups of cytokines, so do several subpopulations [but possibly not all of them (86)] of regulatory αβT cells release IL-10 and use it to preform a primary immunosuppressive capacity. CD4+CD25+ T regulatory cells are well-established sources of IL-10, and their involvement and their use of IL-10 have been shown to play a key role in immune regulation. IL-10 also comes from γ δT cells, a population of T cells known to line areas of the body that come into constant contact with flora, such as the intestine, colon, skin, mouth, nasal passage, lung, and uterus. IL-10 from these sources undoubtedly play a role in maintaining the immune homeostasis in these tissues. Indeed, IL-10 is a required component of γ δT cell release in inhibiting the immune response in a number of scenarios such as maintenance of pregnancy (83), juvenile idiopathic arthritis (87), heart transplants (88), and tumor survival in model systems (89). It has been established that the immune response in women tends to be more TH2-oriented than that in men, in which it is more TH1-oriented (89a,b). Steroids have been long recognized as modulators of immune function, and corticosteroids in general are known to exhibit immunosuppresant activities. It is becoming clear that steroids influence the level of cytokine gene transcription, and higher levels of testosterone are known to increase IL-10 synthesis in T cells (90, 91). Thus, testosterone and dihydrotestosterone, used in some clinical trials to inhibit autoimmune disease, may exhibit their effects through IL-10 to weaken and shift the immune system toward humoral immunity in men.

Clinical Applications Owing to its potent anti-inflammatory and immunosuppressive effects, IL-10 has attracted much attention for potential clinical applications. Because IL-10-deficient mice develop chronic ileocolitis that can be prevented by IL-10 administration, various studies have examined its use in Crohn’s disease. Unfortunately, only modest clinical responses were reported, possibly owing to polymorphism of the IL-10R1 gene or the instability of IL-10 in patients. IL-10 therapy was also tested for psoriasis, but only temporary clinical improvement was achieved (92). Also, clinical use of IL-10 is often accompanied by increased IFN-γ production (93). In contrast to its anti-inflammatory effect, neutralization of IL-10 has shown promise in the treatment of systemic lupus erythematosus (SLE). SLE patients

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

936

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

often have hyperproduction of IL-10, which is proposed to be involved in autoantibody production and lupus development (94, 95). Furthermore, administration of anti-IL-10 can prevent lupus development in the NZB × NZW mouse model (96). In open clinical trials, three weeks of murine anti-IL-10 therapy in patients resulted in rapid and long-lasting amelioration of lupus symptoms (97). Although this study is preliminary, involving few patients and lacking proper controls, future studies using humanized antibodies hold great promise for lupus treatment. One major concern in clinically manipulating the levels of IL-10 is its critical role in immune homeostasis. Long-term application of IL-10 could cause immunodeficiency, whereas continuous use of anti-IL-10 may lead to hyperimmune reactivities. Nevertheless, investigations in the past decade have clearly demonstrated that modulation of the availability of IL-10 as a potential immune therapy holds great promise for better treatment of many diseases such as malignancy, autoimmunity, and allergic reactions.

INTERFERONS AND INTERFERON-LIKE CYTOKINES In 1957 Isaacs & Lindenmann (98) discovered a substance, released by cells exposed to a virus, that protected cells from viral infection—they called it interferon. The interferons represent proteins with antiviral activity that are secreted from cells in response to a variety of stimuli (99–101). There are two types of interferons, Type I and Type II, and interferon-like cytokines. Type I interferons consist of seven classes—IFN-α, IFN-β, IFN-ε, IFN-κ, IFN-ω, IFN-δ, and IFN-τ ; four interferon-like cytokines have been reported: limitin (found only in mice), IL-28A, IL-28B, and IL-29 found in human and other mammals (Table 1). Type II interferon consists of IFN-γ only. IFN-α, IFN-β, IFN-ε, IFN-κ, IFN-ω, IL-28A, IL-28B, and IL-29 are found in humans, whereas IFN-δ (102), IFN-τ (103), and limitin (104) are not. IFN-τ was described first as ovine trophoblast protein-1 and is found in ungulates where it is required for implantation of the ovum (103), but there is no direct human homologue. Hu-IFN-κ, although it exhibits low specific antiviral activity, is expressed in human keratinocytes (105). IL-28A, IL-28B, and IL-29 are found in humans and other mammals (55, 56) and function like Type I interferons. Human IFN-ε (106) has not been characterized in significant detail; however, it appears to play a role in reproductive function in placental mammals. Murine and human IFN-ε promoters possess reproductive endocrine regulatory elements (106a). Its expression is constitutive in murine placental and ovarian tissue (106a). Finally, there have been reports of constitutively produced IFN species isolated from murine placental tissue, murine peri-implantation concepti, and human placental amniotic membrane that possess antiviral activity that is immunologically distinct from IFN-α or IFN-β, and is expressed from mRNA unrelated to IFN-α or IFN-β (106b–e). Interferons were the first cytokines discovered and the first to be used therapeutically (99, 107–109). This section will provide a short overview.

CSIF (43) Zcyto10 (157) IL-TIF (159) C49A (199), Mob-5 (181), Mda-7 (160), FISP (200) AK155 (161) IFN-λ2, -λ3, λ1 (55)

IL-10 IL-19 IL-20

IL-22

IL-24

Unknown IL-10R2

IL-10R2 IL-20R2 IL-20R2 IL-20R2 IL-10R2 None predicted IL-20R2 IL-20R2

Unknown Jak1 + Tyk2, Stat 2 + 1 + 3 + 5 (55, 56, 266)

Jak1 +Jak2, Stat 1 + 3 + 5, Akt, MAPK (177) Jak1 + Tyk2, Stat 3 + 1 + 5 (43, 177) Jak1 + Jak2, Stat 3 + 1 (151) Jak1 + Jak2, Stat 3 + 1 (151, 157) Jak1 + Jak2, Stat 3 + 1 Jak1 + Tyk2, Stat 3 + 1 + 5 (163, 164) None known Jak1 + Jak2, Stat 3 + 1 (152) Jak1 + Jak2, Stat 3 + 1 (152)

Gαβγ , akt, MAPK, src, PI3K (265) Jak1 + Tyk2, Stat 2 + 1 + 3 + 5, MAPK, PRMT1 (177)

Signal transduction pathways employed

This table delineates the genome-encoded ligands (or their close orthologs) known to interact with members of the Class 2 cytokine receptor family. IL-28A, IL-28B, and IL-29 were clustered together. Also, Type I IFN members have been clustered together. The columns from left to right indicate standard names for the ligand or ligand family, alternate names for the ligand (or major subgroups of the Type I IFN family), the location of its gene(s) in the human genome, the components of its receptor complex (the receptor chain predicted or known to bind Jak1 is listed first), and biological activities initiated by the ligand when it interacts with its receptor. If a ligand is known to interact with more than one type of receptor complex, the row for that particular ligand is split and each type of receptor complex is indicated. Location of only the human genes are indicated because murine and rat homologs of these human ligand genes are generally found on the respective syntenic chromosome locations in the murine and rat genomes, respectively. (C) indicates transcription occurs toward the centromere, whereas (T) indicates transcription occurs away from the centromere. Although most IFN-α genes are transcribed toward the telomere, some are transcribed toward the centromere. Data were obtained from the following Web site: http://www.ncbi.nlm.nih.gov/mapview/map search.cgi?taxid=9606. The Jak kinase associating with IL-20R2 is not known; Jak2 is speculated based on sequence similarity of IL-20R2 with other Class 2 cytokine receptors.

Unknown IL-28R1

IL-10R1 IL-20R1 IL-20R1 IL-22R1 IL-22R1 IL-22BP IL-20R1 IL-22R1

IFN-γ R2

IFN-αR2

Receptor complex chain 2

AR210-IY22-30.sgm

12q14 + 3(C) 19q13 + 2(T,C,T)

1q32 + 2(T)

12q14 + 3(C)

1q32 + 1(C) 1q32 + 1(T) 1q32 + 1(T)

IFN-γ R1

TF IFN-αR1

Receptor complex chain 1

AR210-IY22-30.tex

IL-26 IL-28A, -28B, -29

IFN-γ

13q34(T) 9p21 + 3(T)(α, ω), 9p21 + 3(T)(β), 9p21 + 1(T)(κ), 9p21 + 3(C)(ε) 12q14 + 3(C)

Gene locus (human)

AR

Type II IFN

IFN-α, -β, -δ, -ε, -κ, -τ , -ω, limitin

Common alternate names/members

12:12

Factor VIIa Type I IFN

Ligand

TABLE 1 Ligands of the Class 2 cytokine receptor family

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

19 Feb 2004 LaTeX2e(2002/01/18)

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

937

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

938

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

In general, exposure of human cells to viruses or double-stranded RNAs induces the production of IFN-α, IFN-β, and IFN-ω species. The ratio of IFN-α, IFN-β, and IFN-ω produced by cells varies with the tissue of origin and the species of the organism. T-lymphocytes produce IFN-γ when stimulated with specific antigens or when stimulated with mitogens such as staphylococcal enterotoxin A, B, or the combination of phytohemagglutinin and phorbol esters or ionophores. Most of the interferons in therapeutic use are produced through genetic engineering (99–101, 107). As the first biotherapeutic approved, IFN-α paved the way for the many new biotherapeutics to follow. Nevertheless, we have just touched the surface of understanding the multitude of human interferons. Many procedures have been described for the partial purification of human and animal interferons (99–101, 107, 110). It was not until 1978 that interferons were purified to homogeneity (111–115). The introduction of reverse phase and normal phase high performance liquid chromatography to the purification of proteins (112–115) led to the first successful purification of these proteins so that sufficient amounts were available for their chemical, biological, and immunological studies. Various affinity purification techniques, particularly antibody affinity chromatography, were utilized to purify human leukocyte (α) and fibroblast (β) interferons. During the purification of leukocyte interferon, it became evident that multiple species existed (107), confirmed after cloning 12 members of the IFN-α family (107, 108). It was astonishing at the time to find that this family of proteins exhibited different relative activities (116). Twelve human IFN-α species were identified (107, 108, 117–120) and many allelic variants. In addition, many new interferons have been obtained from cells from patients with various diseases where mutations led to interferons with novel activities (121–125). Each IFN-α species seems to exhibit a distinct profile of activities (antiviral, antiproliferative, and stimulation of cytotoxic activities of natural killer cells and Tcells) (126). There is no clear correlation among the activities so that the members of the IFN-α family exhibit unique activity profiles. For the most part, the IFN-α species are not glycosylated, although some contain carbohydrates. Predominantly, only one recombinant IFN-α protein is used therapeutically (IFN-α2a, IFN-α2b, and IFN-α2c, allelic variants) so that the remaining IFN-α species remain an untapped reservoir of opportunity. Why the body produces so many of these interferons is an unanswered question. As our understanding of the mechanism of their receptor interactions develops, some of these answers should be forthcoming. Overall, IFN-β and IFN-ω have similar activities as IFN-α, but many differences have been found. IL-28A, IL-28B, and IL-29 are induced by virus infection (56), exhibit antiviral activity, and use the Jak-Stat pathway and interferon-stimulated response element (ISRE) (55, 56). The receptor for the Type I interferons consists of two chains, IFN-αR1 and IFNαR2 (Table 2, Figure 1). The interferon-like molecules IL-28A, IL-28B, and IL29, however, use a different receptor complex consisting of IL-10R2 and IL-28R1 chains (Tables 1 and 2, Figure 1). The interferons and interferon-like molecules signal through the Jak-Stat pathway (109, 127, 128). IFN-α, IFN-β, IFN-ε, IFN-κ,

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

IL-10, RELATED CYTOKINES, AND RECEPTORS

939

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TABLE 2 The Class 2 cytokine receptors Standard name

Common alternate names/members

Tissue factor

Gene locus

Ligands

TF (177, 265), Factor III, coagulation factor III

1p22 + 1(T)

Factor VIIa

IFN-γ R1

IFN-γ R (267), IFN-γ Rα (268)

6q23 + 2(C)

IFN-γ

IFN-γ R2

AF-1 (251), IFN-γ Rβ (269)

21q22 + 11(T)

IFN-γ

IFN-αR1

IFN-αR (270)

21q22 + 11(T)

IFN-α, IFN-β, IFN-ω, IFN-δ, IFN-ε, IFN-κ, IFN-τ , limitin

IFN-αR2

IFN-α/βR (186), IFN-αR2a (271), IFN-αR2b, IFN-αR2c, IFN-αR2L (50), IFN-αR2S, IFN-αR2-1 (185), IFN-αR2-2, IFN-αR2-3

21q22 + 11(T)

IFN-α, IFN-β, IFN-ω, IFN-δ, IFN-ε, IFN-κ, IFN-τ , limitin

IL-10R1

IL-10Rα, IL-10R (176, 272)

11q23 + 3(T)

Cellular and viral IL-10s

IL-10R2

IL-10Rβ (175), CRFB4 (273), CRF2-4 (274)

21q22 + 11(T)

Cellular and viral IL-10, IL-22, IL-28A, IL-28B, IL-29

IL-20R1

CRF2-8 (275), CRFB8, IL-20Rα (157), ZcytoR7, IL20RA (152)

6q23 + 2(C)

IL-19, IL-20, IL-24

IL-20R2

CRF2-11 (45), DIRS1 (157), IL-20Rβ, IL-20RB (152)

3q22 + 3(T)

IL-19, IL-20, IL-24

IL22R1

CRF2-9 (275), IL-TIFR1 (164), ZcytoR11, IL22R (163), IL-22RA1

1p36 + 11(T)

IL-22, IL-20, IL-24

IL22BP

CRF2-10 (45), ZcytoR16, IL-22RA2 (184), CRF2-s1 (183)

6q23 + 2(C)

IL-22

IL28R1

CRF2-12 (45), IL-28Rα (56), IFN-λR1 (55), LICRII (266)

1p36 + 11(T)

IL-28A, IL-28B, IL-29

This table describes the various members of the Class 2 cytokine receptor family. The columns from left to right indicate the currently accepted name for the receptor, alternate names or physiologically relevant splice variants for the receptor, the location of its gene in the human genome, and ligands that signal through that receptor chain when it is part of an intact receptor complex. Location of only the human genes are indicated because murine and rat homologs of these human receptor genes are generally found on the respective syntenic chromosome locations in the murine and rat genome, respectively. (C) indicates transcription occurs toward the centromere, whereas (T) indicates transcription occurs away from the centromere. Data were obtained from the following Web site: http://www.ncbi.nlm.nih.gov/mapview/map search.cgi?taxid=9606.

IFN-ω, IFN-δ, IFN-τ , and limitin signal through the Type I receptors IFN-αR1 and IFN-αR2c, although the pathway of IFN-ε has not been defined (Tables 1 and 2). All the Type I interferons activate Stat1, Stat2, Tyk2, and Jak1 and induce genes that have the ISRE in the promoter (129, 130). IFN-γ uses the unique receptor chains IFN-γ R1 and IFN-γ R2 (Tables 1 and 2), and activates Jak1 and Jak2, Stat1

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

940

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

that in turn induces genes containing the gamma activation sequence (GAS) in the promoter (75, 129). In addition to the Jak-Stat pathway, the interferons utilize other pathways, some of which are noted in Table 1. Other pathways independent of the Jak-Stat pathway can initiate signal transduction by the Type I interferons (Table 1). For example, cells from mice lacking Stat1 can still respond to Type I interferons: IFN-α and IFN-β can modulate proliferative responses in phagocytes from Stat1deficient mice (78). Understanding of these events is further complicated by the multiplicity of the Type I interferons that exhibit different activities although they interact with the same receptor. This is an area that needs to be explored in great detail to understand this family of proteins and their mechanisms of action. The Type I interferons exhibit a wide breadth of biological activities: antiviral, antiproliferative, stimulation of cytotoxic activity (126, 131–136) of a variety of cells of the immune system (T-cells, natural killer cells, monocytes, macrophages, and dendritic cells), increasing expression of tumor-associated surface antigens (137–139), stimulation of other surface molecules such as MHC Class I antigens (137, 140, 141), induction and/or activation of proapoptotic genes and proteins (e.g., TRAIL, caspases, Bak, and Bax) (142), repression of antiapoptotic genes [e.g., Bcl-2, inhibitor of apoptosis protein (IAP)] (142), modulation of differentiation (143–145) and antiangiogenic activity (146, 147). All these actions make interferon a most promising agent to treat various diseases. The challenge is to be able to use this enormous potential of the interferons without the debilitating side effects. Appropriate technology to locally deliver the interferons in tumors could overcome the problem of systemic side effects. Overall, it is highly likely that interferons will play a major role in the next generation of novel antitumor and antiviral therapies.

Interleukin-19 IL-19 was cloned by searching EST databases for IL-10 homologues (148). The complete cDNA was cloned by screening an Epstein-Barr virus (EBV)-transformed B cell library that contains a 534-bp open reading frame encoding a protein of 177 amino acids. The 30 UTR contains only one AUUUA mRNA instability motif. IL-19 is a secreted protein with an 18-amino acid N-terminal signal peptide. The putative molecular weight is ∼21 kDa. However, there are two potential N-linked glycoslyation sites at positions 56 and 135 that might explain multiple IL-19 proteins in the region of 35–40 kDa in SDS-PAGE analysis. Crystal structure of IL-19 revealed that only the former glycosylation site bound to oligosaccharide, although this source of IL-19 was isolated from a Drosophila cell line (149). Although the amino acid sequence shows ∼20% identity to that of IL-10, they are very similar molecules having helical structure. However, unlike IL-10 and like IL-20, IL-19 contains six conserved cysteine residues and functions as a monomer (149). IL-19 is located in chromosome 1q32 (Tables 1 and 2, Figure 1) in an IL-10 cytokine cluster (148, 150). The gene contains seven exons and six introns, and the coding sequence lies in exon 3 to 7 (150). The presence of the additional noncoding exons

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

941

allows the generation of two cDNAs having different 50 UTR sequences possibly owing to alternative splicing or promoter usage. Interestingly, the cDNA having a long 50 UTR harbors an in-frame ATG-codon 114 bp upstream of the first ATG of the IL-19 cDNA. If translation starts from the upstream ATG this will extend the protein an additional 38 amino acids, resulting in a long signal peptide sequence of 52 amino acids. It was shown that inclusion of such a long signal peptide severely limited the amount of secreted IL-19, indicating that the translation most probably starts from the latter ATG (148). The human IL-19 promoter region has been cloned (150). However, the transcription factor(s) involved in the regulation of expression remain to be identified. Mouse IL-19 shows 71% amino acid identity to human IL-19 (150). It contains 176 amino acids with a predicted signal peptide sequence of 24 amino acids. There are three potential N-linked glycoslysation sites in the mouse IL-19 gene. The genomic structure of mouse IL-19 is similar to that of human IL-19. Although structurally IL-10 and IL-19 are similar molecules, the homologous region of IL-10 that interacts with IL-10R1 is much less conserved in IL-19, indicating that IL-19 does not interact with IL-10 receptors (148). Indeed, it was demonstrated by independent studies that IL-19 selectively binds to IL-20 receptor complexes and activates STAT proteins (151, 152). IL-19 mRNA expression can be detected principally in human monocytes under basal conditions (148, 153). Very low levels of IL-19 mRNA could be detected in B cells and it could not be detected in T cells (153). Activation of monocytes with LPS resulted in induction of IL-19 after 4 h (148). This induction was delayed in comparison to that of IL-10, which was induced as early as 2 h after treatment. Pretreatment of monocytes with IL-4 or IL-13, but not with IFN-γ , resulted in a significant increase in IL-19 mRNA induction by LPS. Among a variety of cytokines that include other interleukins, TNF-α, and IFN-γ , only GM-CSF could induce IL-19 mRNA expression. Treatment of mouse monocytes with mouse IL-19 or human monocytes with human IL-19 resulted in the induction of IL-6 and TNF-α mRNA and protein in a dose-dependent manner (150). This induction did not require de novo protein synthesis. Interestingly, mouse IL-19 was active on human monocytes, but not vice versa. Pretreatment of monocytes with IL-10 abolished both IL-6 and TNF-α induction by IL-19. Treatment of monocytes with IL-19 resulted in production of reactive oxygen species (ROS) and induction of apoptosis in these cells (150). These effects might be due to the production of TNF-α by IL-19 because TNF-α antibody treatment could prevent ROS production and protect cells from IL-19mediated apoptosis. However, the significance of these findings to the in vivo setting remains to be elucidated. The finding that IL-19 binds to IL-20 receptor complexes indicates that like IL20 and its receptors, IL-19 might also be involved in the pathogenesis of chronic inflammatory diseases, such as psoriasis. One characteristic feature of psoriatic skin is infiltration of IFN-γ -producing Th1 cells that are responsible for mediating inflammation and tissue destruction (154). Treatment with IL-4 can shift the

19 Feb 2004

12:12

942

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

differentiation of na¨ıve and memory T cells toward a Th2 phenotype, thus suppressing Th1 action and alleviating psoriasis (155). Administration of human IL-4 to 20 patients with severe psoriasis significantly improved their condition (156). This therapeutic approach did not affect the mRNA level of IL-20 or its receptors in skin biopsies, but there was a significant reduction in the level of IL-19 mRNA in all but one patient (156). This indirect correlation between IL-19 expression and psoriasis, in parallel to the psoriatic-like condition seen in IL-20 transgenic mice, needs to be corroborated by direct studies, including transgenic mice or IL-19 inhibition experiments using soluble receptors or neutralizing antibodies.

Interleukin-20 The discovery of IL-20 is an example of successful database mining. EST databases were screened using an algorithm that identifies translated sequences containing both a signal sequence and one or more amphipathic helices commonly found in helical cytokines, such as IL-10 (157). One EST from a human keratinocyte library showed the highest homology and the cDNA was cloned by RACE. The cDNA is 925 bp long with an open reading frame of 531 bp encoding a putative polypeptide of 176 amino acids. Sequence comparison analysis revealed that this protein shows 40%, 33%, and 28% amino acid identity to IL-19, IL-24/Mda-7, and IL-10, respectively, and therefore it was named IL-20 (Tables 1 and 2). The 30 UTR of IL-20 cDNA contains seven AUUUA and one precise UUAUUUAUU mRNA instability motif often found in cytokine mRNAs. Northern blot analyses showed that IL-20 mRNA is expressed at very low levels in skin, trachea, and in other tissues. Mouse IL-20 cDNA also encodes a protein of 176 amino acids and shows 76% amino acid identity to human IL-20. Whereas IL-10 and IL-26 have four conserved cysteine residues, IL-19 and IL-20 contain six cysteine residues. The formation of disulfide bonds between the extra cysteines in IL-20 would bring the hinge region of the molecule into close contact with the amino-terminus helix, thereby preventing the formation of an intercalating dimer. Analysis of the recombinant protein revealed that IL-20 functions as a monomer, while IL-10 and IL-26 function as intercalating dimers. Radiation hybrid mapping demonstrated that IL-20 maps to chromosome 1q32 in human and to the syntenic region on chromosome 1 in mouse (157). In human chromosome 1 a region of 195 kb includes IL-10, IL-19, IL-20, and IL-24/Mda-7 in a sequential order (Tables 1 and 2). The function of IL-20 was elucidated using a transgenic animal approach (157). Both human and mouse IL-20 were overexpressed in transgenic mice using different promoters that would allow the production of circulating protein, specific expression in epithelium or cells of lymphoid origin, and a broad expression pattern in all tissues. In all cases, the transgenic mice were smaller than their normal littermates, had a shiny appearance with tight, wrinkled skin, and died within the first few days after birth. Histologically, only skin showed abnormal morphology in transgenic mice having thickened epidermis, hyperkeratosis, and a compact

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

943

stratum corneum and had ultramicroscopic features similar to psoriatic skin. The transgenic mice showed characteristic changes in the distribution of some of the differentiation and proliferation markers in skin. The expression of K5 and K14 differentiation markers and K6 and PCNA proliferation markers could be detected both in the suprabasal and basal layers of epidermis, whereas in the normal littermates the expression was detected only in the basal layer. These changes were observed in all types of transgenic mice, including the one in which IL-20 is produced in liver through an albumin promoter, indicating the involvement of circulating IL-20 in modifying differentiation and proliferation of keratinocytes. To elucidate the mechanism of action of IL-20, studies were initiated to identify receptors for this molecule (Tables 1 and 2, Figure 1). Because IL-20 is structurally similar to IL-10, different Class 2 cytokine receptors alone or in pair-wise combination were transiently transfected into COS-7 cells, and their ability to bind to biotin-labeled IL-20 was assayed. Only the pair-wise transfection of two orphan receptors, zcytor7 and DIRS1, showed positive IL-20 binding, and therefore they were named IL-20R1 and IL-20R2 (Tables 1 and 2, Figure 1), respectively. In separate studies, it was shown that IL-20 also binds to IL-22R1 and IL-20R2 (151, 152). Binding of IL-20 to its receptors leads to the activation of the JAK-STAT pathway (151, 157). The activity of a STAT-responsive luciferase reporter was robustly augmented when the cells were treated with a combination of IL-20 and EGF, IL-1β, or TNF-α (157). IL-20 treatment resulted in nuclear translocation of STAT3, but not STAT1. Microarray analysis demonstrated that IL-20 treatment of human keratinocyte HaCaT cells resulted in induction of genes involved in inflammation such as TNF-α, MRP-14, and MCP-1, and IL-1 synergized with IL-20 in these inductions (157). Thus, it was inferred that IL-20 has proinflammatory properties for keratinocytes. Overexpression of IL-20 receptor subunits was detected in psoriatic skin, especially in keratinocytes, immune cells, and endothelial cells, indicating the involvement of IL-20 and its receptors in the pathogenesis of this disease (157). In these contexts IL-20 blockade by soluble receptors, monoclonal antibodies, or other approaches involving inhibitors or small molecules has been proposed as a potential therapeutic modality for chronic inflammatory skin disease (158).

Interleukin-22 DISCOVERY AND CHARACTERIZATION OF IL-22 IL-22 is one of several cytokines with limited homology to IL-10 (148, 157, 159–161). The cDNA for the mouse IL-22 ortholog was identified as a gene specifically induced by IL-9 in mouse T cells, and the protein was originally designated IL-10-related T cell–derived inducible factor (IL-TIF) (159). Induction of IL-22 by IL-9 was obtained in thymic lymphoma cells, T helper cells, and mast cells, and also by IL-9 or Con A in freshly isolated splenocytes. Constitutive expression of the IL-22 was detected in thymus and brain. IL-22 with 22% identity to IL-10 consists of 179 amino acids, including a potential signal peptide. Human IL-22 was cloned based on its homology to

19 Feb 2004

12:12

944

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

the mouse ortholog (162) and as a novel secreted protein (163). Recombinant human IL-22 expressed in COS cells appears to be a glycosylated secreted protein migrating as several bands in the region of about 25–40 kDa (164). IL-22 mRNA expression was induced in T cells by anti-CD3-stimulation and further upregulated by the addition of Con A (162, 163). Activated human and mouse Th1 rather than Th2 CD4+ cells produce IL-22 (165). Injection of LPS in vivo stimulated IL-22 mRNA expression in various organs (162). It was shown that stimulation of HepG2 human hepatoma cells with IL-22 upregulated the production of acute phase reactants such as serum amyloid A, α1-antichymotrypsin, and haptoglobin. IL-22 induces the production of reactive oxygen species in resting B cells. Injection of IL-22 in mice induced production of serum amyloid A in mouse liver. These observations suggest involvement of IL-22 in the inflammatory response (162, 166). IL-22, in contrast to IL-10, had no effect on LPS-induced production of TNF, IL-1, and IL-6 from freshly isolated human monocytes and did not inhibit the action of IL-10 in these assays (163). IL-22 had no effect on IFN-γ production from in vitro polarized Th1 cells; however, modest inhibitory effects of IL-22 on IL-4 production from Th2 cells was observed (163). Administration of murine IL-22 either by intravenous injection or by adenovirus resulted in systemic effects, including decreased red cell and serum albumin levels, but increased platelet, serum amyloid A, and fibrinogen levels, and lowered body weight (165). Basophilia in the proximal renal tubules was observed (165). All these observations indicate involvement of IL-22 in inflammatory and perhaps immune responses in various organs.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

BIOLOGICAL ACTIVITIES

Whereas there is a single copy of the IL-22 gene in the human genome and in genomes of BALB/c and DBA/2 mice, genomes of other mouse strains such as C57Bl/6, FVB, and 129 possess two IL-22 genes that were designated IL-22α and IL-22β (162). The two copies were well conserved, demonstrating 98% nucleotide identity in the coding region. The murine genes were mapped on chromosome 10 in the region of the IFN-γ gene. The human IL-22 gene is located on chromosome 12 about 90 Kb from the IFN-γ gene (Table 1). The human IL-22 gene and the murine IL-22α gene consist of six exons where exon 1 contains a part of 50 UTR and exons 2–5 cover the coding region of the IL-22 mRNA. The murine IL-22β gene contains several single nucleotide changes and a 658 nucleotide deletion covering the first noncoding exon and a segment of a putative promoter, suggesting that the IL-22β gene may not be expressed (162). Some genetic data suggest a link of IL-22 to allergy and asthma. IL-22 is induced by IL-9, a Th2 cytokine active on T and B lymphocytes, mast cells, and eosinophils, and potentially involved in allergy and asthma (167, 168). The IL-22 gene (and also the ak155 gene) is located on human chromosome 12q14 + 3 (Table 1), where several loci potentially linked to asthma and atopy have been identified by genetic studies, particularly in the 12q13.12–q23.3 region (169). The strongest evidence for linkage is in a region near the gene encoding IFN-γ (170–173). THE IL-22 GENE

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

945

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

However, the gene for IFN-γ appears to be highly conserved (no sequence variations were detected in 265 individuals), suggesting that mutations of the IFN-γ gene are unlikely to be a significant cause of inherited asthma (174). The IL-22 and IL-26 genes are positioned next to the IFN-γ gene and, thus, are possible candidates for linkage to asthma. THE IL-22 RECEPTOR COMPLEX The functional IL-10 receptor complex consists of two chains (175), the ligand-binding IL-10R1 subunit (176) and the second IL-10R2 subunit that supports signaling through the IL-10R1 chain (175). Both chains belong to the Class 2 cytokine receptor family (177) (Tables 1 and 2). One orphan receptor CRF2-9 (cytokine receptor family Class 2 member) was capable of binding IL-22 when expressed in monkey COS cells or in Chinese hamster cells as demonstrated by flow cytometry or by crosslinking (163, 164). However, only in COS cells was CRF2-9 able to sustain IL-22 signaling as measured by Stat activation, whereas CRF2-9 expressed in hamster cells did not render them responsive to IL-22 (164), suggesting an additional receptor subunit is required. This second subunit was the IL-10R2 chain. When IL-10R2 and IL022R1 were coexpressed in hamster cells, the cells become responsive to IL-22 (164). In addition, anti-IL-10R2 antibody completely blocked IL-22-mediated expression of serum amyloid A and luciferase reporter activity in HepG2 cells (162). Thus, IL22 requires expression of IL-22R1 and the ubiquitously-expressed IL-10R2 for signaling and induction of biological activities (Tables 1 and 2). SIGNAL TRANSDUCTION Rapid Stat activation following treatment with IL-22 was observed in various cell lines (159, 162–164, 178). Stat3, to a lesser extent Stat1, and in certain cells Stat5, were activated in cells after IL-22 treatment. Analysis of amino acids surrounding Tyr residues within the IL-22R1 intracellular domain revealed the presence of four potential Stat3 recruitment sites, phospho-Tyr-XX-Gln sequence (YXXQ motif), consistent with Stat3 activation by IL-22. In addition, when the chimeric IL-22R1/γ R1 chain, where the IL-22R1 intracellular domain was replaced by the intracellular domain of the human IFN-γ R1 chain, was expressed in COS cells, the pattern of IL-22-induced Stat activation changed to that characteristic of IFN-γ signaling—only Stat1 DNA-binding complexes were observed (164). Thus, the activation of Stats is indeed mediated by the IL-22R1 intracellular domain, because the substitution of the IL-22R1 intracellular domain by the IFN-γ R1 intracellular domain caused a change in the pattern of Stat activation. The use of chimeric receptors also allowed evaluation of responsiveness of cells to IL-22 treatment without knowing IL-22 specific biological activities. In cells expressing the chimeric IL-22R1/γ R1 and intact IL-10R2 chains, IL-22 induced IFN-γ -like biological responses. Thus, in these cells not only IL-22-induced signaling events (Stat activation) could be measured but also IL-22 induced IFN-γ -like biological activities, such as MHC Class I antigen expression (36, 164, 175, 179). The IL-22 receptor (Figure 1) is likely to be structurally homologous to the IFN-γ , IL-10, and other IL-10 family cytokines receptor complexes (177). The

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

946

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

IL-22 receptor complex demonstrated a distinct feature: Not only was the IL22R1 chain able to bind IL-22, but also IL-10R2 was capable of binding the ligand independently of IL-22R1 in heterologous cell lines (163, 164). Thus, both chains of the IL-22 receptor complex can independently bind ligand, whereas in the IL10, IFN-α, and IFN-γ receptor complexes only one chain (the R1 chain, which also recruits Stat proteins) can bind ligand in the absence of the other. In all of these receptors, the second (R2) chains are necessary for signaling (177). The ability of the IL-10R2 alone to bind IL-22 was particularly surprising because this chain is unable to bind IL-10 (36, 175). Because IL-10R2 is ubiquitously expressed but unable to transduce a signal without an additional chain (IL-10R1 or IL-22R1), it is possible that secreted IL-22 will be retained at the site of secretion by being bound to the IL-10R2 chain, providing local action but preventing its action at remote sites. However, this “exception” could become a rule for receptor complexes for other IL-10-related cytokines, where only the combination of both chains provide receptor complexes with high affinity for a ligand, whereas each chain alone provides low binding affinity (180, 181). Clearly both IL-22R1 and IL-10R2 chains are required for reconstitution of the functional IL-22 receptor complex, and the IL-10R2 chain functions as the IL22R2 chain. Sharing of receptor chains between the family of Class 2 cytokines is relatively common (Tables 1 and 2): IL-10, IL-22, IL-28A, IL-28B, and IL-29 use the IL-10R2 chain in their receptor complexes; IL-19, IL-20, and IL-24 share chain IL-20R1; IL-19, IL-20, and IL-24 share IL-20R2; and IL-20, IL-22, and IL-24 share IL-22R1 (Figure 1, Table 1). EXPRESSION OF THE IL-22R1 mRNA By Northern blotting the expression of the IL22R1 mRNA was detected in several normal tissues including kidney, liver, and pancreas, where the level of expression was the highest (164). The fact that the IL-22R1 gene is expressed in normal liver tissue correlates with its functions as a hepatocyte-stimulating factor (162). Several solid tumor cell lines constitutively express the IL-22R1 mRNA: colorectal adenocarcinoma SW480, lung carcinoma A549, melanoma G361, hepatoma HepG2, and renal carcinoma Caki-1 and TK10 cell lines, HT29 intestinal epithelial cell line, and HaCaT keratinocyte cell line (162–164, 178, 180). It is interesting to note that all cell lines expressing the IL22R1 mRNA are nonhematopoietic tumor cell lines. Because IL-22 induces Stat3 activation in IL-22-responsive cells and expression of constitutively active Stat3 was shown to cause cellular transformation (182), it is possible that in certain cells IL-22 could act as an autocrine or paracrine factor driving cells to malignancy. NATURALLY OCCURRING IL-22 ANTAGONIST The fact that IL-22 is capable of binding to the singly expressed IL-10R2 suggests that the activity of IL-22 can be regulated on the local level by sequestering locally expressed IL-22. However, the regulation of IL-22 action is not limited to the local level. Systemic regulation is provided by the soluble receptor IL-22-binding protein (IL-22BP), originally designated CRF2-10 or CRF2-X (cytokine receptor family Class 2 member 10 or X) (164, 178), also named IL-22RA2 and CRF2-s1 (183, 184). The use of decoy or

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

947

soluble receptors for cytokines utilizing members of the Class 2 cytokine receptor family for signaling is rather uncommon. It has been demonstrated only for Type I interferons where the human IFN-αR2c, one of the functional chains of the IFN-α receptor complex, has two splice variants, the membrane-bound IFN-αR2b with a short intracellular domain that appears unable to transduce the signal and the soluble IFN-αR2a form (50, 58, 179, 185–188). These receptors are the result of splicing events. The other soluble receptor, IL-22BP (164, 178), functions as an antagonist of IL-22 (164, 178). IL-22BP demonstrates 34% identity to the IL-22R1 extracellular domain. The purified protein appears on SDS-PAGE gels as a broad band in the region of about 35–45 kDa, suggesting glycosylation of the protein. Indeed, there are five potential sites for N-linked glycosylation (Asn-X-Thr/Ser) in CRF2-10 (164). IL-22BP binds IL-22 as demonstrated by crosslinking and direct binding (164, 178). IL-22BP blocked IL-22 binding to its receptor, Stat activation, and luciferase reporter activity; neutralized IL-22 activity in a proliferation assay; and inhibited the ability of IL-22 to induce the expression of the SOCS-3 gene (164, 178, 184). The affinity of the IL-22 receptor complex when both IL-22R1 and IL-10R2 chains were present was higher than the affinity of each chain alone (164). However, it seems that IL-22BP has even higher affinity for IL-22 than the membrane-bound IL-22 receptor complex because incubation of cells expressing the functional IL22 receptor complex with IL-22 in the presence of IL-22BP inhibited binding of IL-22 to the cellular receptors very effectively (164). IL-22BP binds specifically to IL-22, but not to IL-10, IL-19, IL-20, or IL-24 (151, 164, 178). Thus, the production of IL-22BP may be one of the mechanisms to precisely regulate IL-22 function and might modulate local inflamation. In this light, it is interesting to note that IL-22BP expression was detected in mononuclear cells of inflammatory sites, plasma cells, epithelial cells, placenta, skin, inflamed appendix, lung, gastrointestinal tract, lymph node, thymus, and spleen (184). The regulation of IL-22BP expression is not known; induction of IL-22BP expression in response to LPS or anti-CD3 Abs in peripheral mononuclear cells was not detected (178). IL-22 did not induce IL-22BP production (178). The IL-22BP expression profile by Northern blots and PCR analysis indicated that IL-22BP is highly expressed in human placenta, mammary gland, breast, spleen, skin, and lung, with lower expression in a variety of other tissues such as heart, pancreas, testis, thymus, prostate, digestive system, breast, lymph nodes, lung, skin, parotid, bladder, and kidney (164, 178, 183, 184). It is noteworthy that IL-22BP is expressed in some tumors such as ovarian, uterine, and rectal cancers, but not in the corresponding normal tissues (184). Considering that IL-22 shares receptors with other members of the IL-10 family, this decoy receptor adds another level of regulation to this complicated network. IL-22R1, IL-10R2, AND IL-22BP GENES Genes encoding receptors of the Class 2 cytokine receptor family are clustered on several human chromosomes (177, 189). The IL-10R2 gene maps to chromosome 21 together with IFN-αR1, IFN-αR2,

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

948

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

and IFN-γ R2 genes (Tables 1 and 2). The IL-22R1 gene is on chromosome 1. The IL-22BP gene is positioned on chromosome 6 in the vicinity of the IFN-γ R1 gene in a head-to-tail orientation and is in close proximity to the IL-20R1 gene. IL-10R2, IL-22R1, and IL-22BP genes have similar structural architecture, with two exceptions for the IL-22BP gene: the presence of an additional exon (exon 1) encoding the 50 UTR and the lack of two terminal exons encoding transmembrane and intracellular domains of IL-22R1 and IL-10R2 chains. Thus, IL-10R2 and IL22R1 genes contain 7 exons. The IL-22BP gene is composed of 6 exons (164). In addition, there are alternative splicing events that either insert the exon 4a between exons 4 and 5 creating a longer CRF2-10 splice variant (CRF2-10L) or eliminate the exon 5 creating a prematurely terminated short protein (CRF2-10S). It remains to be determined whether products of alternative slicing events are functional.

Interleukin-24 (Mda-7/IL-24) Melanoma differentiation-associated gene-7 (Mda-7) was identified using subtraction hybridization as a novel gene upregulated during terminal differentiation of human melanoma cells by interferon (IFN)-β and the protein kinase C activator mezerein (160). The Mda-7 cDNA is 1718 bp in length encoding a polypeptide of 206 amino acids with a predicted size of 23.8 kDa (160). The 30 UTR contains three consensus elements (AUUUA) presumably involved in mRNA instability and three polyadenylation sequences (AAUAAA). The Mda-7 gene is composed of seven exons and six introns and is located on chromosome 1q32 within an IL-10-related gene cluster containing four genes, including IL-10, IL-19, IL-20, and Mda-7 in linear order spanning 195 kb of genomic DNA (157, 190). Mda-7 has a 49 amino acid N-terminal hydrophobic signal peptide that allows the molecule to be cleaved and secreted. Mda-7 has three putative glycosylation sites at amino acids 95, 109, and 126 resulting in several molecular sizes of the secreted product. Structural analysis indicates that Mda-7 is a member of the four-helix bundle family of cytokine molecules with greatest homology to the IL-10 subfamily (Tables 1 and 2), which now includes IL-19, IL-20, IL-22 (IL-TIF), IL-26/AK155, IL-28A, IL-28B, and IL-29 (148, 157, 161, 163). IL-10 and Mda-7 share only ∼23% sequence identity, however the presence of an IL-10 signature sequence in Mda-7 indicates that it belongs to the IL-10 subfamily. Analysis of the expression profile of Mda-7 in different human tissues documents restricted expression to those tissues associated with the immune system, such as spleen, thymus, and peripheral blood leukocytes and also in normal melanocytes (160, 190, 191), suggesting cytokine-like properties of the Mda-7 molecule. Based on these observations, Mda-7 has been renamed IL-24 by the HUGO nomenclature system. The regulation of expression of Mda-7/IL-24 during terminal differentiation occurs predominantly at a post-transcriptional level (192, 193). Treatment with IFN-β plus mezerein promotes stability of Mda-7/IL-24 mRNA (192). At the level of transcription, Mda-7/IL-24 expression is predominantly regulated by two transcription factors, AP-1 and C/EBP (193).

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

949

Mda-7/IL-24 displays two distinct functional attributes. At low concentration, which might be considered physiological, Mda-7/IL-24 functions as a cytokine (194). However, when Mda-7/IL-24 is overexpressed via an adenovirus-mediated gene delivery system it induces apoptosis selectively in cancer cells but not in normal cells (195–198). On the basis of these interesting properties, Mda-7/IL-24 has become the focus of intense scrutiny. Cloning of Mda-7/IL-24-like molecules in other species indicates that the function of Mda-7/IL-24 is complex. Two groups have separately cloned rat Mda-7/IL-24, as c49a and mob-5 (181, 199). Rat Mda7/IL-24 encodes a secreted protein containing 183 amino acid residues that display 58.7% identity to human Mda-7/IL-24 (181, 199). C49A expression is upregulated in wounded rat dermal cells during the process of wound repair, and MOB-5 is induced as a function of induction of the oncogenic ras pathway, indicating that the rat homologue of Mda-7/IL-24 is involved in proliferation rather than growth inhibition. The mouse Mda-7/IL-24-like molecule, termed FISP (IL-4-induced secreted protein), encodes a secreted protein of 220 amino acids, which displays 93% identity with MOB-5/C49A and 69% identity with Mda-7/IL-24 (200). FISP expression is induced during Type-2 T helper lymphocyte differentiation, and its expression is induced by TCR signaling involving protein kinase C activation and STAT-6-dependent IL-4R signaling. Mda-7/IL-24 and FISP share a number of properties, including expression predominantly in the immune system. Additionally, both Mda-7/IL-24 and FISP are induced during differentiation and by combinatorial treatment with a cytokine and a protein kinase C activator. Although the function of FISP is not known it appears that the function of this gene is more closely related to Mda-7/IL-24 than c49a/mob-5. Because Mda-7/IL-24 belongs to the IL-10 family of cytokines, it was predicted to exert its actions through cell-surface receptors (Tables 1 and 2, Figure 1). The IL-10 family of cytokines signal through receptors that are dimers of an R1 type of receptor (with a long cytoplasmic domain) and an R2 type of receptor (with a short cytoplasmic domain) (175, 176). For example, IL-10 signals through IL-10R1 and IL-10R2; IL-20 signals through IL-20R1 and IL-20R2; and IL-22 signals through IL-22R1 and IL-10R2 (157, 164). Transfection with specific receptor subunits and binding assays with Mda-7/IL-24 protein identified receptors that can bind Mda-7/IL-24 protein (151, 152, 180). Maximum binding activity for Mda-7/IL24 could be detected only when cells were transfected with a combination of IL-20R1/IL-20R2 or IL-22R1/IL-20R2, suggesting that these receptor complexes mediate Mda-7/IL-24 signal transduction (151, 152, 180). Electrophoretic mobility shift assays in human keratinocyte cell lines revealed that, like other IL-10 family members, Mda-7/IL-24 induces the activation of signal transducers and activators of transcription (STAT), notably STAT-1 and STAT-3 (180). Phosphorylation of STAT3 was also demonstrated by Western blot analysis following Mda-7/IL-24 treatment (151). A potential role of Mda-7/IL-24 as a cytokine and its involvement in the immune system has recently been highlighted (153, 194, 201). Mda-7/IL-24 expression could be induced in human peripheral blood mononuclear cells (PBMC) upon

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

950

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

treatment with PHA or LPS and in rat alveolar macrophages stimulated by LPS or IL-4 (194, 201). In addition, Mda-7/IL-24 mRNA could be detected by real-time RT-PCR in human monocytes and its expression was upregulated in monocytes by treatment with LPS or in T cells, especially in CD4+ na¨ıve and memory cells, following activation by anti-CD3 monoclonal antibody (153). The mouse counterpart of Mda-7/IL-24, FISP, was shown to have exclusive expression in Th2 lymphocytes (200). However, in humans the expression pattern in T cells is different from that in mouse. At earlier time points (6 h) Mda-7 expression was downregulated in cells undergoing Th1 differentiation (mediating cellular immunity) and slightly upregulated in cells undergoing Th2 differentiation (regulating humoral immunity). At 66 h, the expression increased in cells committed to Th1 differentiation (153). Treatment of PBMC with Mda-7/IL-24, purified from the conditioned media of a 293 cell line stably expressing Mda-7/IL-24, resulted in the induction of IL-6, IFN-γ , TNF-α, IL-1β, IL-12, and GM-CSF (194). Induction could be blocked, either completely or partially, by simultaneous administration of IL-10, which might be due to the sharing of receptor subunits by IL-10 family members and to a tenfold higher affinity of IL-10 for its receptor. It is worth noting that although IL-10 and Mda-7/IL-24 belong to the same family, functionally they elicit opposite effects, the former being a major suppressor of the immune response and inflammation (43), whereas the latter is immunomodulatory. However, treatment with Mda-7/IL-24 does not modulate the proliferative functions of PBMC (194). It was hypothesized that the secondary cytokines induced by Mda-7/IL-24 might activate antigen presenting cells to present tumor antigen, thereby triggering an antitumor immune response (194). Mda-7/IL-24 expression could be detected in melanocytes, but this expression was gradually lost with the progression of radial and vertical growth phases of melanoma and completely absent in metastatic melanomas (160, 191, 202). The loss of Mda-7/IL-24 expression in progressing melanoma might contribute to the ineffective immune response to melanoma. Another interesting finding is that Mda-7/IL-24 mRNA expression could be induced in human monocytes by infection with influenza A virus A/PR/8, indicating that Mda-7/IL-24 might contribute to antiviral immune responses (201). Studies designed to elucidate the functional properties of Mda-7/IL-24 by employing overexpression analyses uncovered a remarkable phenomenon. When Mda-7/IL-24 is delivered to cancer cells via a replication-incompetent adenovirus (Ad.Mda-7), it induces apoptosis in virtually all cancer cell types including melanoma, glioblastoma multiforme, and osteosarcoma and carcinomas of the breast, prostate, colon, liver, lung, nasopharynx, cervix, and ovary, irrespective of their tumor suppressor (such as p53, p16INK4A, or Rb) status (198, 203). In malignant gliomas, Ad.Mda-7 induces apoptosis in both wild-type and mutated p53-containing cells, whereas an adenovirus expressing wild-type p53 induces apoptosis only when p53 is mutated, thus indicating that Ad.Mda-7 might be more beneficial for the gene-based therapy of this particular cancer (204). Infection with Ad.Mda-7 or treatment with purified GST-Mda-7 protein also sensitizes glioma cells to the growth inhibitory effects of ionizing radiation (204). One notable

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

951

exception in Ad.Mda-7-mediated cancer-specific killing is pancreatic cancer cells that have frequent genetic mutations leading to the activation of the K-ras oncogene. However, a combination of Ad.Mda-7 and antisense K-ras oligonucleotides resulted in apoptosis in pancreatic cancer cells having mutated K-ras but not wild-type K-ras (205). In contrast, infection with Ad.Mda-7 did not influence the growth of a diverse array of normal cells including human fibroblasts, mammary, bronchial, renal and prostate epithelial cells, melanocytes, astrocytes, and vascular endothelial cells (198, 203). Infection with Ad.Mda-7 results in downregulation of antiapoptotic proteins, Bcl-2 and/or Bcl-xL, and upregulation of proapoptotic proteins, Bax and/or Bak, specifically in several cancer cell types, but not in normal cells (206). Interestingly, overexpression of either Bcl-2 or Bcl-xL in prostate cancer cells differentially protected cells against Ad.Mda-7-mediated apoptosis, indicating that this change in equilibrium between pro- and antiapoptotic molecules might serve as a major determinant for apoptosis mediated by Ad.Mda-7 (207). In melanoma and glioma cells, but not in their normal counterparts, induction of apoptosis by Ad.Mda-7 is mediated by the activation of p38 MAPK pathway that results in upregulation of growth arrest and DNA damage (GADD)-inducible genes such as GADD153, GADD45 and GADD34 (204, 208). In a lung cancer model it was shown that Ad.Mda-7-induced apoptosis is mediated by upregulation and activation of double-stranded RNA-dependent protein kinase PKR (209). Activation of PKR also activated p38 MAPK, indicating that p38 MAPK might be an essential molecule in mediating Ad.Mda-7-induced apoptosis. In addition, Ad.Mda-7 infection suppresses the expression of inducible nitric oxide synthetase that is essential for survival of melanoma cells but not of normal melanocytes (210). These findings indicate that Ad.Mda-7 can induce apoptosis in cancer cells by multiple mechanisms. Moreover, in addition to its direct effect on cancer cells resulting in apoptosis, Ad.Mda-7 also inhibits tumor growth by inhibiting angiogenesis (211). One interesting observation is that the apoptosis-inducing effect of Ad.Mda-7 does not require activation of the JAK-STAT pathway. There is also no correlation between induction of apoptosis and the expression of the IL-20/IL-22 receptor subunits (212). Thus, it might be concluded that there is a discrete disengagement between the cytokine function and apoptosis-inducing properties of Mda-7/IL-24.

Interleukin-26 (IL-26/AK155) Subtraction hybridization coupled with representational difference analysis identified IL-26/AK155 as a gene upregulated in human T cells following infection with herpesvirus saimiri (HVS), with the capacity to transform these cells in culture (161). The cDNA consists of 1076 bp containing an open reading frame of 513 bp encoding a putative polypeptide of 171 amino acids that contains a predicted hydrophobic signal sequence 21 aa long. The IL-26/AK155 protein has 24.7% amino acid identity and 47% amino acid similarity to human IL-10. Structural

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

952

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

analysis revealed that IL-26/AK155 (hereafter referred to as IL-26 only) contains six helices with four highly conserved cysteine residues that are assumed to be relevant for dimer formation as is the case in IL-10. Indeed, it was demonstrated that while under denaturating conditions the protein migrates at a position of ∼19 kDa, in nondenaturating conditions it migrates at ∼36 kDa indicating spontaneous dimer formation. The IL-26 gene is located on chromosome 12q14 + 3 in a cluster where IFNγ and IL-22 are also positioned (Table 1). This gene consists of five exons that are interrupted by four introns. Two microsatellite polymorphisms at the IL-26 locus have been identified, one in the third intron (D12S2511) and one in the 30 region of the gene (D12S2510) (213). Apart from being useful as markers in association or linkage studies of polygenic inflammatory or allergic diseases, no other significance of these polymorphisms has been documented. Under basal conditions, IL-26 mRNA could not be detected in any cell line by Northern blot analysis (161). Upon infection of T cells with various strains of HVS, expression was induced. However, IL-26 mRNA could be detected in unstimulated fresh peripheral blood cells of 10 healthy donors by RT-PCR analysis. Although the IL-10 transcript was detected in most cell lines of T or B lineage, the IL-26 transcript was rather specific for T cells. The level of activity of T cells did not alter the IL-26 transcript level and neither stimulation by phorbol esters nor inhibition by cyclosporine altered IL-26 mRNA level. It was determined that IL-26 mRNA is specifically overexpressed by T cells after HVS transformation. In another study, under basal conditions no IL-26 mRNA could be detected in monocytes, NK cells, and B and T cells (153). The expression could be induced only in NK cells and T cells upon stimulation with IL-2/IL-12 and anti-CD3 monoclonal antibody, respectively. This induction of IL-26 mRNA was observed specifically in activated memory cells (CD4+; CD45RO+) and during polarization toward Type 1 T helper cells. The observations that IL-26 is induced by viral infection and in Th1 cells upon stimulation indicate that this gene may play an immune-protective role, especially against viral infection. However, no comprehensive studies have been conducted to determine the precise function(s) of IL-26. More importantly, the receptors for IL-26 have not been identified and also the role of IL-26 in regulating cytokine production by other immune cells has also not been investigated. Thus, the role of IL-26 in the regulation of HVS-mediated transformation and its function in regulation of the immune system remain to be determined.

EVOLUTION OF CLASS 2 CYTOKINES AND THEIR RECEPTORS The recent discovery of several IL-10-like ligands, Type I IFN ligands, and receptors binding these ligands raises the issue of how the ligand and receptor families evolved and when each family originated during evolution. An analysis of the phylogeny of the known human members of the Class 2 cytokine receptor family

15 Mar 2004

16:13

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

953

has been reported (45, 55, 56, 177, 214). However, researchers require a more thorough analysis of receptor evolutionary relationships to obtain a better understanding of how the family originated and radiated. Phylogenetic analyses are highly sensitive to which subdomain of a multi-domain protein is being analyzed and from which species the protein sequences originate. To stabilize the analysis of the receptors, receptors originating from several vertebrate species were included, and alignments were restricted to the most conserved domain, encoded by the first 150 amino acids of each receptor. Gene and cDNA sequences from pufferfish (Tetraodon nigroviridis, Takifugu rubripes), zebrafish (Danio rerio), African frogs (Xenopus laevis, Xenopus tropicalis), chicken (Gallus gallus), mouse (Mus musculus), cow (Bos taurus), and human (Homo sapiens) species were the predominant sources for these analyses. Results from these analyses demonstrate that some receptors in the Class 2 cytokine receptor family had evolved to forms homologous to those in humans prior to vertebrate speciation, whereas a few piscine receptors, appearing to code for IFN-like receptors, do not have mammalian homologs (177, 214). Also, although one would expect a clear separation between ligand-binding receptors and accessory receptors, this was not observed (214). Because the two types of receptor chain (ligand-binding and accessory) can be discriminated by three properties (length of intracellular domain, association with Jak1 versus Jak2 or Tyk2, and association of Stat proteins), and because these discriminating properties also apply to some members of the Class 1 cytokine receptor family (such as IL-2 receptors), the ligand-binding receptor chains and accessory receptor chains were assessed in separate phylogenetic analyses. Separate analyses have not been reported. Additionally, all the ligands that bind receptors of the Class 2 cytokine receptor family have been aligned only recently (214), although alignments of Type I IFNs and IL-10-like cytokines have been done previously (55). Published phylogenetic analysis of the ligands vary depending on how many members are being aligned (45, 55, 214), suggesting a multi-species alignment could improve the accuracy of the analysis. Results from these analyses demonstrate that several ligands in the Class 2 cytokine receptor family had evolved to forms homologous to those in humans prior to vertebrate speciation (214). We refined our analysis by enforcing equal representation among different members of each family and by maximizing the species divergence of a given protein. Furthermore, we chose tissue factor as an outgroup for the ligand-binding chain phylogenetic analysis because it is the only member of the Class 2 receptor family that binds ligand in an atypical manner and has no significant intracellular domain. In agreement with the hypothesis that ligands and their receptors coevolve over time (234, 235), the evolution of Class 2 ligand-binding receptors and their respective ligands follow a similar phylogenetic relationship (Figure 2). Both Type I and Type II IFNs and their receptors appear to have diverged prior to the remaining members of the family. Piscine homologs of Type I IFN, unlike avian and mammalian homologs of Type I IFN, are encoded by single genes containing five exons and four introns with splice sites in similar positions as other members of the Class

15 Mar 2004

16:13

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

954

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

2 ligands, suggesting that a retrotransposition event during the speciation of the tetrapod lineages yielded the single-exon Type I IFNs (237). Piscine Type I IFN may interact with piscine CRFB1 or CRFB2, as these receptors diverge between IFN-αR2 and IFN-γ R1. The next to diverge is the interferon-like IL-10 homolog IL-28 and its receptor. Piscine CRFB7 diverges at a similar point. Afterward, a piscine cytokine tentatively designated IL-24 diverges. The next to diverge are IL-10R1 and IL-22R1, of which no clear piscine homolog can been identified. Similarly, IL-22, IL-10, and IL-26 diverge from the remaining members of the Class 2 ligands; no obvious piscine homolog of IL-22 can be identified, but clear piscine homologs of IL-10 have been identified. After divergence of amphibian and mammalian IL-24 proteins, IL-19 and IL-20 species diverge from each other. Fish and amphibians encode IL-20 homologs, which appear to be homologs of the predecessor of both IL-19 and IL-20. Finally, piscine CRFB8 and CRFB9 diverge from each other. The nearest mammalian homologs of these receptors are IL-20R1 and IL-22BP, respectively, which are neighbors in the human genome. One can speculate that CRFB9 may bind CRFB8 ligands, such as piscine IL-20 or piscine IL-24, and (as hypothesized for IL-22BP) antagonize their actions in fish and amphibians. Interestingly, it appears that members of the Class 2 cytokine receptor and ligand families that diverge first possess antiviral activity, whereas those that diverge later do not possess this ability. In accordance with this fact, IL-28, which diverged between IFN and IL-10-like subfamilies, has 1/100 the specific antiviral activity of Type I IFN (55, 56), and 1/10 the specific antiviral activity of IFN-γ . Phylogeny of the accessory chains, in contrast, did not match that of the ligands or the ligand-binding receptor chains (Figure 2). Piscine CRFB6 diverged first and encodes a receptor with no obvious mammalian homolog. A split occurs between receptors that bind Tyk2 (IFN-αR1 and IL-10R2) and the IL-20R2/IFNγ R2 subgroup. It is tempting to speculate that IL-20R2, like IFN-γ R2, associates with Jak2. If so, IL-22R1 could interact with either IL-10R2 or IL-20R2 to allow the activation of Jak1 and either Tyk2 or Jak2, respectively, in addition to restricting ligand signaling to either IL-22 or IL-20/IL-24, respectively. IFN-αR1, whose mammalian and avian homologs are composed of four SD100 domains, can be split into the first two SD100 domains and the second two SD100 domains, with each pair of domains having homology to other Class 2 cytokine receptors. The phylogeny here, as reported elsewhere (177, 214), suggests that early in the development of IFN-αR1 there was a duplication of the extracellular domain, followed by divergent evolution of each subdomain. For unknown reasons the second and third subdomains, instead of the first and second or third and fourth subdomains, seem to be the ones principally involved in IFN binding in humans (238, 239). These two IFN-αR1 subdomains form a group distinct from that formed by piscine CRFB5, tetrapod IL-10R2, and piscine CRFB4. Although CRFB5 does not appear to have a direct mammalian homolog, piscine CRFB4 appears to be the piscine homolog of IL-10R2. Within the CRFB3/IFN-γ R2/IL-20R2 subgroup, piscine CRFB3 diverges prior to the IL-20R2/IFN-γ R2 split. Finally, IFN-γ R2

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

955

and IL-20R2 split from each other. A homolog to IL-20R2 can be identified in the Takifugu rubripes genome (accession number CAAB01003251.1) as well as other piscine genomes (C. Krause and S. Pestka, unpublished data). The extensive sharing of IL-10R2 (and to a lesser extent IL-20R2) with several cytokine ligand-binding chains suggests that as ligands and their ligand-binding receptor chains evolved and radiated, they generally did not require the equivalent radiation and evolution of an accessory receptor chain. The sole exceptions appear to be IFN-γ R1 and IFN-αR2, which are the first two receptors to diverge from the Class 2 cytokine receptor family, and they may have required monogamous accessory receptor chains as other ligand-binding chains and their ligands evolved and competed more frequently for a common pool of accessory receptor chains. Interestingly, although a definite interaction is presumed between IFN-γ R1 and IFN-γ R2 in intact cells as determined by fluorescence resonance energy transfer (FRET), a weak interaction may exist between IFN-γ R1 and IL-10R2 because a small amount of FRET can be observed between that receptor pair (57). Interestingly, a similar but weak interaction is observed between IL-10R1 and IFN-γ R2, yet no FRET is seen between IFN-γ R2 and IL-10R2 (C. Krause and S. Pestka, unpublished data). This potential for polygamy between an accessory receptor for a ligand:ligand-binding chain complex during evolution is concordant with the accessory receptor contributing only the Jak kinase and no other specific function to signal transduction elicited by a ligand (58). All major classes of vertebrates (fish, amphibians, birds, and mammals— reptiles are not represented in these phylogenies but are presumed to be included as well) encode receptors and/or ligands with IFN-like activities and receptors with IL-10-like activities (214). Thus, it appears that the entire range of Class 2 cytokine receptor chains and their ligands evolved prior to vertebrate speciation. Interestingly, searches in the Ciona intestinalis genome database yielded weak homologs to only small portions of the IFN-αR1, Tetraodon CRFB4 and CRFB8 receptors, and no obvious ligand homolog (C. Krause and S. Pestka, unpublished data). Ciona intestinalis, a sea squirt, is an invertebrate chordate and is a somewhat distant relative to vertebrates, but it is the closest vertebrate relative with the entire genome sequenced. Thus, the Class 2 cytokine receptor and ligand families appear to have emerged during vertebrate evolution. Furthermore, no components of the vertebrate adaptive immune system (B cells, T cells, MHC antigens, lymphocytes, immunoproteosome subunits) are found in Ciona, yet most elements of the mammalian adaptive immune system appear to be present in all forms of vertebrates. This suggests that the Class 2 cytokine receptor family evolved at the same time as the adaptive immune system—during the evolution of the vertebrate lineage from invertebrate chordates. All Class 2 ligands are known to be secreted predominantly by members of the immune system. This suggests that these ligands as well as their receptors evolved as the immune system required new ways to communicate among its various elements as well as with nonhematopoietic tissues. Just as there is coevolution of ligands and receptor pairs, so might there be coevolution of cell systems and their signaling networks.

15 Mar 2004

16:18

956

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

STRUCTURE OF THE CLASS 2 CYTOKINES AND RECEPTORS A total of 12 different Class 2 α-helical cytokines have been defined by X-ray and NMR methods (Table 3). The structures are best described by six secondary structure elements (A–F), where helices A, C, D, and F adopt a classic four-helix bundle topology as first described in the structure of porcine growth hormone (240) (Figure 3). Helix pairs AC and DF are linked by long overhand connections (B and E) that pack against one another on one edge of the bundle. Although all Class 2 cytokines exhibit the same domain architecture, structural comparisons reveal a clear division of the molecules into two subgroups designated as the Type I IFN and IL-10 family folds (Figure 3).

The IL-10 Fold Two key structural features uniquely define the IL-10 fold. The first is conformation of helix F which is bent at an angle of ∼50◦ (Figures 3 and 4). The second feature TABLE 3 Summary of Class 2 Cytokines, Receptors, and Receptor Complex Structures Class/molecule

Species

Method

Resolution (A)

References

Type I IFNs 1. IFN-β 2. IFN-α2b 3. IFN-α2a 4. IFN-β 5. IFN-τ

Murine Human Human Human Opine

X-ray X-ray NMR X-ray X-ray

2.2 2.9 — 2.2 2.1

(276) (217) (277) (216) (215)

Type II IFNs 1. IFN-γ 2. IFN-γ

Human Bovine

X-ray X-ray

2.5, 2.9 2.0

(218, 252) (221)

IL-10 Family Members 1. IL-10 2. IL-10 3. IL-10 4. IL-22 5. IL-19

Human ebv cmv Human Human

X-ray X-ray X-ray X-ray X-ray

2.0–1.6 1.9 2.7 2.0 2.0

(224, 225, 278) (247) (248) (226) (149)

Receptors 1. TF 2. TF 3. IFN-αR2

Human Rabbit Human

X-ray X-ray NMR

2.4–1.7 2.4 —

(258, 279, 280) (281) (230)

Receptor complexes 1. IFN-γ /IFN-γ R1 2. TF/FVIIa 3. IL-10/IL-10R1 4. cmvIL-10/IL-10R1

Human Human Human cmv/Human

X-ray X-ray X-ray X-ray

2.9–2.0 2.0–2.1 2.9 2.7

(220, 222, 223) (259, 282) (229) (248)

15 Mar 2004

16:18

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

957

of the fold is the conserved location of the AB loop that encircles helix F at the position of the bend. When this structural motif was first characterized in the structure of IL-10, five residues were identified (Leu-47, Phe-71, Tyr-72, Ala139, and Glu-142) that were conserved between IFN-γ and IL-10 and appeared to be critical for stabilizing the bend in helix F and properly orienting the AB loop (225). These five residues were predicted to be useful as a “fingerprint” for detecting other molecules with the same structural fold. With few exceptions, the three-dimensional arrangement of these residues are conserved in the recently identified IL-10 family members IL-19, IL-20, IL-22, IL-24, and IL-26 (148, 157, 159–161, 163). The importance of this structural feature is further emphasized by crystal structures of IFN-γ /IFN-γ R1 and IL-10/IL-10R1 complexes that show this motif forms the majority of the high-affinity receptor-binding sites (site 1a) for each ligand (220, 229).

The Type I Interferon Fold Helix F is essentially straight in the Type I IFNs (Figures 3 and 5). However, the AB loop in the IFNs still encircles helix F as described for the IL-10 fold. In fact, superposition of IL-10 family members and the Type I IFNs reveals Glu-142 is conserved in helix F of both families (215–217). Consistent with subtle differences in their structures, each glutamate residue participates in different hydrogen bonding schemes. In IL-10 family members, Glu-142 hydrogen bonds to the main chain of the AB loop, but in the Type I IFNs it interacts with Tyr-122 located on helix D. Although these interactions are related, the helix F/AB loop motif in the Type I IFNs is distinguished by a conserved disulfide bond that covalently links the AB loop and helix F (215–217). Mutagenesis studies confirm that this site is important for Type I IFN/IFN-αR2 interactions (241–243). Thus, the straight helix F that is disulfide bonded to the AB loop provides a structurally conserved site1a-binding site for interaction with IFN-αR2 that is very distinct from the IL-10 family.

Quaternary Structure of IL-10 Family Members IL-10 family members may exist as either monomers (IL-19, IL-20, and IL-22) or as domain-swapped or intercalated dimers, as shown for IFN-γ and the cellular and viral IL-10s (Figure 3 and 6) (244). This topology occurs when each domain of the dimer is formed from secondary structural elements A-D from one peptide chain and E-F from the other. Thus, the domain architecture remains the same although the helical segments are contributed from two different peptide chains. At least for IL-10, crystal structure analysis and protein engineering studies suggest the length of the DE loop plays a critical role in defining the final quaternary structure of the molecule (245). Except for cmvIL-10, which exists as a disulfide-linked dimer, the interaction between the intercalated peptide chains is noncovalent. Biophysical studies have shown the IL-10 dimer is quite unstable at low protein concentrations (246). This may limit the potency of the molecule to the immediate vicinity of the cells that secrete it where its

19 Feb 2004

12:12

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

958

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

local concentration is high enough to ensure the conformational integrity of the dimer. To date, the intercalated dimer topology has been identified in the crystal structures of IFN-γ , cIL-10, and the vIL-10 homologs from Epstein Barr (ebvIL-10) and cytomegaloviruses (cmvIL-10) (218, 224, 225, 247, 248). The structures reveal a great diversity in the orientation of the twofold related domains of the dimers. The interdomain angles for IFN-γ , cIL-10, and cmvIL-10 are approximately 60◦ , 90◦ , and 130◦ , respectively (Figure 6). These domain angles significantly alter the extent of interaction between the domains. For example, the twofold related C helices of IFN-γ form extensive hydrophobic interdomain contacts that prevent the surface from interacting with solvent and other receptor chains. In contrast, the IL-10 dimers exhibit very minimal domain interfaces where all of the helices in each domain are fully solvent accessible.

Quaternary Structure and Receptor Oligomerization Changes in the domain angles of the dimers significantly alter the geometry of the receptors in the complexes (Figure 6). This may modify the orientation of the intracellular domains (ID) of the high-affinity R1 receptors, leading to changes in the accessibility of the IDs to different signaling molecules. Support for this hypothesis is provided by fragment complementation studies and peptide mimetic structure-function studies with erythropoietin that show changes in the orientations of the extracellular cytokine-binding domains (CBDs) changes the ID positions and alters biological activity (249, 250). In contrast to erythropoietin, IFN-γ and IL-10 each require a second receptor chain (IFN-γ R2 or IL-10R2) to induce their biological activities (175, 251). Thus, the different orientations of the high-affinity binary complexes also change the orientation of the R2-binding sites and may modulate the R2 ID orientations as previously described for the R1s. It is anticipated that the yet-to-be-determined dimeric structures of IL-24 and IL-26 along with the monomeric IL-10 family members will provide additional quaternary structure diversity that will influence the activation of their respective signaling complexes. Although much of our structure-function knowledge about the Class 2 cytokines is from the dimeric cytokines, a complete understanding of their functional significance has remained elusive. For example, Josephson et al. found that an engineered IL-10 monomer is sufficient to stimulate biological activity (245). Similar studies have been reported with mutants of covalent IFN-γ dimers that bind only one receptor chain (252, 253). Although monomeric forms of the dimers are sufficient for biological activity, recent studies have shown that subtle differences in cell surface receptor levels (e.g., IL-10R1) can have dramatic effects on biological activity (254, 255). The dimers certainly have an energetic advantage because they can simultaneously bind two receptors. This advantage may allow certain cell types to express very low receptor levels, making them selectively sensitive to dimeric ligands that can engage the receptors and induce biological responses, whereas

15 Mar 2004

16:20

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

959

monomeric ligands cannot. The larger receptor-ligand complexes formed by the dimers may also recruit more or different molecules into the intracellular signaling complexes that cannot be accomplished by the monomeric ligands.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Class 2 Cytokine Receptors The division of α-helical cytokine receptors into the Class 1 and Class 2 families was first made by sequence comparisons of the extracellular cytokine-binding domains (CBDs). Specifically, the Class 2 receptors did not contain a WSXWS motif and had different cysteine sequence positions compared with the Class 1 receptors (256, 257). Since these first observations, the structures of four Class 2 CBDs have been determined that provide new insight into their structure and functional properties (Table 1). Crystal structures of the high-affinity receptors for IFN-γ (IFN-γ R1) and IL-10 (IL-10R1) were determined bound to their respective ligands, while the free form of IFN-αR2 was recently determined by NMR methods (220, 229, 230). Finally, crystal structures of human tissue factor (TF) by itself and bound to factor VIIa (FVIIa) are almost identical, suggesting that the CBDs are essentially rigid structures that do not undergo large conformational changes upon ligand binding (258, 259). Based on these four structures, the Class 2 cytokine receptor CBDs are comprised of two fibronectin type III domains (FBN-III). The FBN-III domains consist of two antiparallel β-sheets formed from seven β-strands A, B, E and G, F, C, and C0 (Figure 7a). The N-terminal FBN-III is labeled D1, and the C-terminal membrane proximal domain is labeled D2. This simple architecture is displayed by all Class 2 cytokine receptor CBDs except IFNAR1, which contains two tandem CBDs or four FBN-III domains (257). The structures of the IFN-γ /IFNγ R1 and IL-10/IL-10R1 complexes revealed the cytokine-binding site is located at the interface between the domains. Five different receptor segments (L2–L6) contact the ligand (Figure 7b). They are formed from residues connecting the CC’ (L2) and EF (L3) β-strands and the domain linker (L4) in D1 and BC (L5) and FG (L6) loops in D2.

Differences Between Class 1 and Class 2 Cytokine Receptors The structures of the Class 2 cytokine receptor CBDs differ in two main ways from the related Class 1 receptors that include the growth hormone receptor (GHR). The first is the architecture of the C0 β-strand in D1 and its subsequent linker into β-strand E (Figures 7b, 8). For Class 2 receptors, β-strand C0 forms normal antiparallel hydrogen bonds with β-strand C followed by a short turn leading into β-strand E from the other sheet. In Class 1 receptors, β-strand C0 forms several hydrogen bonds with β-strand C but then crosses over and interacts with β-strand E in an antiparallel manner before a tight turn connects the peptide chain to β-strand E. A more obvious and functionally significant difference in the structures is the relative orientation of the D1 and D2 domains in the receptors (Figure 8). Initial

15 Mar 2004

16:22

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

960

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

comparisons of GHR, TF, and IFN-γ R1 CBD structures suggested the interdomain angles for Class 1 and Class 2 receptors would cluster into two fairly tight groups around 90◦ and 120◦ , respectively. However, IL-10R1 and IFN-αR2 structures reveal large variations in the rotation and translation of the D1 and D2 domains that are similar to those observed for Class 1 receptors. Thus, domain orientation alone does not correlate with the classification of a receptor into the Class 1 or Class 2 cytokine receptor family. Comparison of IFN-γ /IFNγ R1, IL-10/IL-10R1 complexes revealed that the L1 loop in Class 2 CBDs is buried in the domain interface where it cannot interact with the ligand (Figure 7b). In contrast, the domain orientations of GHR and IL-4R in the GH/GHR and IL-4/IL-4R complexes expose residues in L1 for interaction with the ligand (228, 260). This functional distinction in L1 is influenced by several factors, including differences in the domain orientations of the receptors as well as the amino acid sequence and secondary structures of the L1 and L5 loops and finally the orientation of the ligand on the receptor. L5 plays an important role in the functionality of L1 because it is adjacent to L1 in the three-dimensional structure of the CBDs where it can pack against and effectively bury L1 in the core of the receptor (Figure 8). Although this paradigm appears to hold true for all Class 2 receptors that bind IL-10 family members, the recent NMR structure of IFN-αR2 reveals a partially exposed L1 loop (230). Whether these residues participate in Type I IFN binding remains to be determined. However, this difference in receptor structure is consistent with the separation of the Type I IFN and IL-10 family ligands into two distinct structural classes that may recognize their receptors in unique ways.

The High-Affinity Receptor Binding Interface Currently, our understanding of Class 2 receptor-ligand interactions is limited to the IFN-γ and IL-10 receptor complexes (220, 229). Despite considerable sequence and structure differences between IL-10R1 and IFN-γ R1, each contact their respective ligands in very similar ways (229). The receptors each bury ap˚ 2 of surface area upon complex formation. In both recepproximately 1000 A tor complex structures, the L2-L4 and L5-L6 loops of the receptors contact distinct surfaces of their respective ligands, called site 1a and site 1b, respectively (Figure 9). Site 1a, the larger of the two contact sites (∼70% of the buried surface area in the interface), is centered on the bend in helix F and AB loop motif, while site 1b is located near the N-terminus of helix A and the C-terminus of helix F. Although the structure of the IFN-α/IFN-αR2 complex has not been reported, mutagenesis studies have shown IFN-γ R2 also interacts with the AB loop and helix F residues equivalent to site 1a in IL-10 (241, 242). Interestingly, receptor binding and NMR studies have not identified an interaction between IFN-αR2 and the analogous site 1b (261, 262). These studies are consistent with the previously described distinct structural features of the Type I IFNs and IFN-αR2. The importance of structure in the site 1–binding site is further emphasized in the structure of the cmvIL-10/IL-10R1 complex (248). cmvIL-10 shares only

15 Mar 2004

16:23

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

961

27% sequence identity with cellular IL-10 (36). However, the domain structure of cmvIL-10 closely mimics the cellular IL-10s IL-10R1-binding site. The energetic contributions of residues in IL-10/IL-10R1 interface have not been determined. However, conserved interactions in the interfaces of the IL-10 and cmvIL-10 receptor complexes identified eight IL-10 residues (Gln-24, Arg-27, Lys-34, Gln-38, Asp-44, Ser-141, Asp-144, and Glu-151) that are predicted to form an energetic hotspot for receptor binding. Several of these residues are positioned adjacent to one another where they form ion pairs (Arg-27/Glu-151 and Lys-34/Asp-144) on helices A and F. Thus, these residues appear to have a dual role in stabilizing the tertiary structure of the molecule and participating in high-affinity interactions with IL-10R1.

Mechanisms of Ligand Receptor Specificity Consistent with their low amino acid sequence identity, the IL-10 family members are much more structurally diverse than the Type I IFNs. Despite this diversity, several of the ligands in the IL-10 family bind more than one receptor, while others appear to be very specific (151, 180). The greatest differences between the IL-10 family members occur at the NH2- and COOH-termini of the molecules and in helix A (Figure 4). For example, helix A is 2 helical turns shorter in IFN-γ than cmvIL-10 or IL-19, which both have elaborate and long N-termini. Because these changes occur in the 1b receptor-binding site, this region is predicted to play an important role in the specificity of the high-affinity receptors. Superposition of the recent crystal structures of IL-19 and IL-22 onto IL-10 in the IL-10/IL-10R1 complex identifies steric clashes in the site 1b region that could influence the specificity of the ligands (149, 226). However, comparison of the IFN-γ and IL10 receptor complexes reveals significant changes in the orientation of the ligands on the receptor surfaces suggesting the mechanisms of ligand specificity may be more complex (Figure 10). These differences are largely due to conformational changes in the IFN-γ R1 and IL-10R1 L2 loops that make intimate mainchainmainchain hydrogen bonds with the AB loop of the ligands. Because structural differences occur in the AB loops of IL-19 and IL-22, simple superpositions of newly determined ligands onto a known receptor complex model may result in considerable errors in the orientations of the ligands. Overcoming this modeling challenge will be an important future research area. As described above for the IL-10 family, the structural and sequence differences observed in the AB loops and the C-terminal ends of helix A of the Type I IFNs (site 1a) are sufficient to change the affinity and/or the orientation of the molecules on IFN-αR2. Additional structural differences are observed for helix F, which is slightly rotated in each Type I IFN structure (215–217). In contrast, the bent conformation and the sequence register of helix F in the IL-10 family is absolutely conserved. Excluding structural differences that may impact IFN-αR2 interactions, the largest structural changes between the Type I IFNs occur in the CD loop. In IFN-α2, the C-terminal end of helix C bends at an angle of ∼70◦ , while this same

15 Mar 2004

16:24

962

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL.

region exists in structurally distinct extended conformations in IFN-β and IFN-τ . The importance of this region in IFNAR1 binding remains to be determined.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Tissue Factor/FVIIa Interactions Despite the structural similarity of human TF with the Class 2 cytokine receptors, the binding paradigm used by FVIIa is completely different. As shown in Figure 11, ˚ 2) contacts with surfaces of TF that are distinct FVIIa makes extensive (∼1800 A from the high-affinity cytokine-binding site described above. The first epidermal growth factor (EGF)-like domain of FVIIa forms tight associations with the D1/D2 crevice created by the interdomain angle of TF. EGF contacts occur with TF βstrands B, C0 , and E in D1 and strands G and the L5 loop in D2. Additional contacts between the protease domain and TF occur on β-strands F and G of D1. Despite this extensive interaction surface, the cytokine-binding site is located on the opposite side of the receptor where it is completely solvent accessible. At this time it is unclear whether the recruitment of this receptor in blood clotting occurred because it allowed a primordial receptor to coevolve cytokine and FVIIa-binding sites largely independently of one another or whether there is a yet unidentified ligand for TF.

CONCLUDING REMARKS AND PERSPECTIVE The discovery, isolation, and successful clinical application of interferons marked a new era of biotechnology and medicine. The interferons were the first of the proteins we now recognize as the Class 2 cytokine family that suggests a new paradigm for the investigations of the mechanisms controlling critical biological processes such as immune responses, skin biology, and pregnancy. Interestingly, the biological effects of these cytokines are exerted largely through modulating the immune system directly and indirectly. Genetic and crystal structural studies seem to indicate that these cytokines evolved as part of unique adaptation of the immune system to combat microbial pathogens. In addition to playing a critical role in innate immune responses, these cytokines also bridge the innate and the adaptive immune responses and participate in acquired antigen-specific immune reactivities. Studies of genetically modified animals and clinical investigations have clearly revealed the yin-yang effects of different members of these cytokines in the maintenance of immune homeostasis. Various immune disorders are currently being evaluated for being treated with Class 2 cytokines. Because these cytokines also significantly modulate cell growth and viability in addition to their dramatic effects on immune responses, studies employing these cytokines for immune therapy offer new opportunities to treat many diseases that have been fairly intractable to leaps in treatment comparable to the initiation of antibiotic treatment of infectious diseases. Although many positive steps continue to advance therapy for some autoimmune diseases, cancer, and viral diseases, it seems that many of the potent natural intercellular signaling molecules of the body as these Class 2

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

IL-10, RELATED CYTOKINES, AND RECEPTORS

P1: IKH

963

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

cytokines should be able to be harnessed to correct the pathophysiology that leads to these devastating conditions. Developing methods to use these molecules to regenerate normal immune systems, physiology, and homeostasis will be a challenge for the next years. Nevertheless, these potent molecules have the innate power to accomplish this alone and in combinations, but it will take new challenging initiatives and technologies (e.g., delivery systems, eliminating side effects) to apply them safely and effectively. The enormous progress in understanding how these molecules function, however, paves the way for their future use in combating many maladies of humanity. ACKNOWLEDGMENTS We thank current and past members of our laboratories and many colleagues who have participated in studies related to this review. Special thanks go to Gary Brewer, William Clark, Dolly Philopova, Jeffry Cook, Gianni Garotta, Robin Hochstrasser, Lara Izotova, Sergei Kotenko, Srijata Sarkar (Pestka and Krause), Rahul V. Gopalkrishnan, Zao-zhong Su, Irina V. Lebedeva, Moira Sauane, Hongping Jiang, Paul Dent, Shinichi Kitada, John C. Reed and Cy A. Stein (Fisher and Sarkar), Arthur I. Roberts (Shi), and Ken Beaudrie. Work in the Pestka laboratory was supported by NIH grants CA46465, AI36450, and AI43369. Work from the Fisher laboratory was supported by NIH grants CA35675, CA87170, CA097318, and CA098712, the Lustgarten Foundation for Pancreatic Cancer Research, the Department of Defense Army Initiative in Prostate Cancer DAMD 024-0041, the Samuel Waxman Cancer Research Foundation, and the Michael and Stella Chernow Endowment. Work from the Shi laboratory was supported by NIH grants AI43384, CA76492, AI50222, and IIH00208 from the National Space Biomedical Research Institute. Work from the Sarkar laboratory was supported by the Department of Defense Army Initiative in Breast Cancer DAMD 17–03-1–0290. Work from the Walter laboratory was supported by NIH grant AI47300. For the preparation of the final text we are indebted to the enormous and thorough efforts of Nancy Ravena, the constant support and advice from Ellen Feibel, and assistance from Catherine Kuo. The Annual Review of Immunology is online at http://immunol.annualreviews.org

LITERATURE CITED 1. Fiorentino DF, Bond MW, Mosmann TR. 1989. Two types of mouse T helper cell. IV. Th2 clones secrete a factor that inhibits cytokine production by Th1 clones. J. Exp. Med. 170:2081– 95 2. O’Garra A, Stapleton G, Dhar V, Pearce M, Schumacher J, et al. 1990. Produc-

tion of cytokines by mouse B cells: B lymphomas and normal B cells produce interleukin 10. Int. Immunol. 2:821– 32 3. Moore KW, Vieira P, Fiorentino DF, Trounstine ML, Khan TA, Mosmann TR. 1990. Homology of cytokine synthesis inhibitory factor (IL-10) to the

19 Feb 2004

12:12

964

4.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

5.

6.

7.

8.

9.

10.

11.

12.

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL. Epstein-Barr virus gene BCRFI. Science 248:1230–34 Ding L, Shevach EM. 1992. IL-10 inhibits mitogen-induced T cell proliferation by selectively inhibiting macrophage costimulatory function. J. Immunol. 148:3133–39 Fiorentino DF, Zlotnik A, Vieira P, Mosmann TR, Howard M, et al. 1991. IL-10 acts on the antigen-presenting cell to inhibit cytokine production by Th1 cells. J. Immunol. 146:3444–51 Macatonia SE, Doherty TM, Knight SC, O’Garra A. 1993. Differential effect of IL-10 on dendritic cell-induced T cell proliferation and IFN-γ production. J. Immunol. 150:3755–65 Enk AH, Angeloni VL, Udey MC, Katz SI. 1993. Inhibition of Langerhans cell antigen-presenting function by IL-10. A role for IL-10 in induction of tolerance. J. Immunol. 151:2390–98 Peguet-Navarro J, Moulon C, Caux C, Dalbiez-Gauthier C, Banchereau J, Schmitt D. 1994. Interleukin-10 inhibits the primary allogeneic T cell response to human epidermal Langerhans cells. Eur. J. Immunol. 24:884–91 Ding L, Linsley PS, Huang LY, Germain RN, Shevach EM. 1993. IL-10 inhibits macrophage costimulatory activity by selectively inhibiting the up-regulation of B7 expression. J. Immunol. 151:1224– 34 Willems F, Marchant A, Delville JP, Gerard C, Delvaux A, et al. 1994. Interleukin-10 inhibits B7 and intercellular adhesion molecule-1 expression on human monocytes. Eur. J. Immunol. 24:1007–9 Jinquan T, Larsen CG, Gesser B, Matsushima K, Thestrup-Pedersen K. 1993. Human IL-10 is a chemoattractant for CD8+ T lymphocytes and an inhibitor of IL-8-induced CD4+ T lymphocyte migration. J. Immunol. 151:4545–51 Kasama T, Strieter RM, Lukacs NW, Burdick MD, Kunkel SL. 1994. Regu-

13.

14.

15.

16.

17.

18.

19.

20.

lation of neutrophil-derived chemokine expression by IL-10. J. Immunol. 152: 3559–69 Sozzani S, Ghezzi S, Iannolo G, Luini W, Borsatti A, et al. 1998. Interleukin 10 increases CCR5 expression and HIV infection in human monocytes. J. Exp. Med. 187:439–44 Takayama T, Morelli AE, Onai N, Hirao M, Matsushima K, et al. 2001. Mammalian and viral IL-10 enhance C-C chemokine receptor 5 but down-regulate C-C chemokine receptor 7 expression by myeloid dendritic cells: impact on chemotactic responses and in vivo homing ability. J. Immunol. 166:7136– 43 Dokka S, Shi X, Leonard S, Wang L, Castranova V, Rojanasakul Y. 2001. Interleukin-10-mediated inhibition of free radical generation in macrophages. Am. J. Physiol. Lung Cell Mol. Physiol. 280:L1196–202 Ishida H, Hastings R, Kearney J, Howard M. 1992. Continuous anti-interleukin 10 antibody administration depletes mice of Ly-1 B cells but not conventional B cells. J. Exp. Med. 175:1213–20 Levy Y, Brouet JC. 1994. Interleukin-10 prevents spontaneous death of germinal center B cells by induction of the bcl-2 protein. J. Clin. Invest. 93:424–28 Cohen SB, Crawley JB, Kahan MC, Feldmann M, Foxwell BM. 1997. Interleukin-10 rescues T cells from apoptotic cell death: association with an upregulation of Bcl-2. Immunology 92: 1–5 Fluckiger AC, Durand I, Banchereau J. 1994. Interleukin 10 induces apoptotic cell death of B-chronic lymphocytic leukemia cells. J. Exp. Med. 179:91– 99 Cai G, Kastelein RA, Hunter CA. 1999. IL-10 enhances NK cell proliferation, cytotoxicity and production of IFN-γ when combined with IL-18. Eur. J. Immunol. 29:2658–65

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS 21. Kuhn R, Lohler J, Rennick D, Rajewsky K, Muller W. 1993. Interleukin10-deficient mice develop chronic enterocolitis. Cell 75:263–74 22. Yang X, Gartner J, Zhu LH, Wang SH, Brunham RC. 1999. IL-10 gene knockout mice show enhanced Th1-like protective immunity and absent granuloma formation following Chlamydia trachomatis lung infection. J. Immunol. 162:1010–17 23. Murray PJ, Young RA. 1999. Increased antimycobacterial immunity in interleukin-10-deficient mice. Infect. Immun. 67:3087–95 24. Sewnath ME, Olszyna DP, Birjmohun R, ten Kate FJ, Gouma DJ, van Der PT. 2001. IL-10-deficient mice demonstrate multiple organ failure and increased mortality during Escherichia coli peritonitis despite an accelerated bacterial clearance. J. Immunol. 166:6323–31 25. Vazquez-Torres A, Jones-Carson J, Wagner RD, Warner T, Balish E. 1999. Early resistance of interleukin-10 knockout mice to acute systemic candidiasis. Infect. Immun. 67:670–74 26. Gazzinelli RT, Wysocka M, Hieny S, Scharton-Kersten T, Cheever A, et al. 1996. In the absence of endogenous IL-10, mice acutely infected with toxoplasma gondii succumb to a lethal immune response dependent on CD4+ T cells and accompanied by overproduction of IL-12, IFN- γ and TNF-α. J. Immunol. 157:798–805 27. Tournoy KG, Kips JC, Pauwels RA. 2000. Endogenous interleukin-10 suppresses allergen-induced airway inflammation and nonspecific airway responsiveness. Clin. Exp. Allergy 30:775– 83 28. Berg DJ, Leach MW, Kuhn R, Rajewsky K, Muller W, et al. 1995. Interleukin 10 but not interleukin 4 is a natural suppressant of cutaneous inflammatory responses. J. Exp. Med. 182:99–108 29. Grunig G, Corry DB, Leach MW, Sey-

30.

31.

32.

33.

34.

35.

36.

37.

P1: IKH

965

mour BW, Kurup VP, Rennick DM. 1997. Interleukin-10 is a natural suppressor of cytokine production and inflammation in a murine model of allergic bronchopulmonary aspergillosis. J. Exp. Med. 185:1089–99 Wakkach A, Fournier N, Brun V, Breittmayer JP, Cottrez F, Groux H. 2003. Characterization of dendritic cells that induce tolerance and T regulatory 1 cell differentiation in vivo. Immunity 18:605–17 Hsu DH, de Waal MR, Fiorentino DF, Dang MN, Vieira P, et al. 1990. Expression of interleukin-10 activity by Epstein-Barr virus protein BCRF1. Science 250:830–32 Rode HJ, Janssen W, Rosen-Wolff A, Bugert JJ, Thein P, et al. 1993. The genome of equine herpesvirus type 2 harbors an interleukin 10 (IL10)-like gene. Virus Genes 7:111–16 Rivailler P, Jiang H, Cho YG, Quink C, Wang F. 2002. Complete nucleotide sequence of the rhesus lymphocryptovirus: genetic validation for an Epstein-Barr virus animal model. J. Virol. 76:421–16 Rivailler P, Cho YG, Wang F. 2002. Complete genomic sequence of an Epstein-Barr virus-related herpesvirus naturally infecting a New World primate: a defining point in the evolution of oncogenic lymphocryptoviruses. J. Virol. 76:12055–68 Fleming SB, McCaughan CA, Andrews AE, Nash AD, Mercer AA. 1997. A homolog of interleukin-10 is encoded by the poxvirus orf virus. J. Virol. 71:4857– 61 Kotenko SV, Saccani S, Izotova LS, Mirochnitchenko OV, Pestka S. 2000. Human cytomegalovirus harbors its own unique IL-10 homolog (cmvIL-10). Proc. Natl. Acad. Sci. USA 97:1695– 700 Lockridge KM, Zhou SS, Kravitz RH, Johnson JL, Sawai ET, et al. 2000. Primate cytomegaloviruses encode and

19 Feb 2004

12:12

966

38.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

39.

40.

41.

42.

43.

44.

45.

46.

47.

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL. express an IL-10-like protein. Virology 268:272–80 Hughes AL. 2002. Origin and evolution of viral interleukin-10 and other DNA virus genes with vertebrate homologues. J. Mol. Evol. 54:90–101 Fleming SB, Haig DM, Nettleton P, Reid HW, McCaughan CA, et al. 2000. Sequence and functional analysis of a homolog of interleukin-10 encoded by the parapoxvirus orf virus. Virus Genes 21:85–95 Haig DM, Thomson J, McInnes CJ, Deane DL, Anderson IE, et al. 2002. A comparison of the anti-inflammatory and immuno-stimulatory activities of orf virus and ovine interleukin-10. Virus Res. 90:303–16 Stewart JP, Behm FG, Arrand JR, Rooney CM. 1994. Differential expression of viral and human interleukin-10 (IL-10) by primary B cell tumors and B cell lines. Virology 200:724–32 Touitou R, Cochet C, Joab I. 1996. Transcriptional analysis of the Epstein-Barr virus interleukin-10 homologue during the lytic cycle. J. Gen. Virol. 77(Pt. 6):1163–68 Moore KW, de Waal MR, Coffman RL, O’Garra A. 2001. Interleukin-10 and the interleukin-10 receptor. Annu. Rev. Immunol. 19:683–765 Zou J, Clark MS, Secombes CJ. 2003. Characterisation, expression and promoter analysis of an interleukin 10 homologue in the puffer fish, Fugu rubripes. Immunogenetics 55:325–35 Kotenko SV. 2002. The family of IL10-related cytokines and their receptors: related, but to what extent? Cytokine Growth Factor Rev. 13:223–40 Brewer G, Saccani S, Sarkar S, Lewis A, Pestka S. 2003. Increased interleukin-10 mRNA stability in melanoma cells is associated with decreased levels of A + U-rich element binding factor AUF1. J. Interferon Cytokine Res. 23:553–64 Lutfalla G, Gardiner K, Proudhon D,

48.

49.

50.

51.

52.

53.

54.

55.

56.

Vielh E, Uze G. 1992. The structure of the human interferon α/β receptor gene. J. Biol. Chem. 267:2802–9 Bach EA, Tanner JW, Marsters S, Ashkenazi A, Aguet M, et al. 1996. Ligandinduced assembly and activation of the γ interferon receptor in intact cells. Mol. Cell Biol. 16:3214–21 Novick D, Cohen B, Rubinstein M. 1992. Soluble interferon-α receptor molecules are present in body fluids. FEBS Lett. 314:445–48 Domanski P, Witte M, Kellum M, Rubinstein M, Hackett R, et al. 1995. Cloning and expression of a long form of the β subunit of the interferon α β receptor that is required for signaling. J. Biol. Chem. 270:21606–11 Usacheva A, Kotenko S, Witte MM, Colamonici OR. 2002. Two distinct domains within the N-terminal region of Janus kinase 1 interact with cytokine receptors. J. Immunol. 169:1302–8 Usacheva A, Sandoval R, Domanski P, Kotenko SV, Nelms K, et al. 2002. Contribution of the Box 1 and Box 2 motifs of cytokine receptors to Jak1 association and activation. J. Biol. Chem. 277:48220–26 Abramovich C, Ratovitski E, Lundgren E, Revel M. 1994. Identification of mRNAs encoding two different soluble forms of the human interferon αreceptor. FEBS Lett. 338:295–300 Cook JR, Cleary CM, Mariano TM, Izotova L, Pestka S. 1996. Differential responsiveness of a splice variant of the human type I interferon receptor to interferons. J. Biol. Chem. 271:13448–53 Kotenko SV, Gallagher G, Baurin VV, Lewis-Antes A, Shen M, et al. 2003. IFN-λs mediate antiviral protection through a distinct class II cytokine receptor complex. Nat. Immunol. 4:69–77 Sheppard P, Kindsvogel W, Xu W, Henderson K, Schlutsmeyer S, et al. 2003. IL-28, IL-29 and their class II cytokine receptor IL-28R. Nat. Immunol. 4:63–68

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS 57. Krause CD, Mei E, Xie J, Jia Y, Bopp MA, et al. 2002. Seeing the light: preassembly and ligand-induced changes of the interferon γ receptor complex in cells. Mol. Cell Proteomics 1:805–15 58. Kotenko SV, Izotova LS, Pollack BP, Muthukumaran G, Paukku K, Silvennoinen O, et al. 1996. Other kinases can substitute for Jak2 in signal transduction by interferon-γ . J. Biol. Chem. 271:17174–82 59. Finbloom DS, Winestock KD. 1995. IL10 induces the tyrosine phosphorylation of tyk2 and Jak1 and the differential assembly of STAT1 alpha and STAT3 complexes in human T cells and monocytes. J. Immunol. 155:1079–90 60. Weber-Nordt RM, Riley JK, Greenlund AC, Moore KW, Darnell JE, Schreiber RD. 1996. Stat3 recruitment by two distinct ligand-induced, tyrosinephosphorylated docking sites in the interleukin-10 receptor intracellular domain. J. Biol. Chem. 271:27954–61 61. Wehinger J, Gouilleux F, Groner B, Finke J, Mertelsmann R, Weber-Nordt RM. 1996. IL-10 induces DNA binding activity of three STAT proteins (Stat1, Stat3, and Stat5) and their distinct combinatorial assembly in the promoters of selected genes. FEBS Lett. 394:365–70 62. Ding Y, Chen D, Tarcsafalvi A, Su R, Qin L, Bromberg JS. 2003. Suppressor of cytokine signaling 1 inhibits IL-10-mediated immune responses. J. Immunol. 170:1383–91 63. Alexander WS, Starr R, Fenner JE, Scott CL, Handman E, et al. 1999. SOCS1 is a critical inhibitor of interferon γ signaling and prevents the potentially fatal neonatal actions of this cytokine. Cell 98:597– 608 64. Naka T, Tsutsui H, Fujimoto M, Kawazoe Y, Kohzaki H, et al. 2001. SOCS1/SSI-1-deficient NKT cells participate in severe hepatitis through dysregulated cross-talk inhibition of IFN-γ and IL-4 signaling in vivo. Immunity 14:535–45

P1: IKH

967

65. Yasukawa H, Misawa H, Sakamoto H, Masuhara M, Sasaki A, et al. 1999. The JAK-binding protein JAB inhibits janus tyrosine kinase activity through binding in the activation loop. EMBO J. 18:1309– 20 66. Kamura T, Sato S, Haque D, Liu L, Kaelin WG, et al. 1998. The Elongin BC complex interacts with the conserved SOCS-box motif present in members of the SOCS, ras, WD-40 repeat, and ankyrin repeat families. Genes Dev. 12:3872–81 67. Zhang JG, Farley A, Nicholson SE, Willson TA, Zugaro LM, et al. 1999. The conserved SOCS box motif in suppressors of cytokine signaling binds to elongins B and C and may couple bound proteins to proteasomal degradation. Proc. Natl. Acad. Sci. USA 96:2071–76 68. Hanada T, Yoshida T, Kinjyo I, Minoguchi S, Yasukawa H, et al. 2001. A mutant form of JAB/SOCS1 augments the cytokine-induced JAK/STAT pathway by accelerating degradation of wild-type JAB/CIS family proteins through the SOCS-box. J. Biol. Chem. 276:40746–54 69. Niemand C, Nimmesgern A, Haan S, Fischer P, Schaper F, et al. 2003. Activation of STAT3 by IL-6 and IL10 in primary human macrophages is differentially modulated by suppressor of cytokine signaling 3. J. Immunol. 170:3263–72 70. Shames BD, Selzman CH, Meldrum DR, Pulido EJ, Barton HA Jr, et al. 1998. Interleukin-10 stabilizes inhibitory κB-α in human monocytes. Shock 10:389–94 71. Wang P, Wu P, Siegel MI, Egan RW, Billah MM. 1995. Interleukin (IL)-10 inhibits nuclear factor κ B (NFκB) activation in human monocytes. IL-10 and IL-4 suppress cytokine synthesis by different mechanisms. J. Biol. Chem. 270: 9558–63 72. Schottelius AJ, Mayo MW, Sartor RB, Baldwin AS Jr. 1999. Interleukin-10

19 Feb 2004

12:12

968

73.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

74.

75.

76.

77.

78.

79.

80.

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL. signaling blocks inhibitor of κB kinase activity and nuclear factor κB DNA binding. J. Biol. Chem. 274:31868–74 Wang M, Hu Y, Stearns ME. 2003. A novel IL-10 signalling mechanism regulates TIMP-1 expression in human prostate tumour cells. Br. J. Cancer 88:1605–14 Stearns ME, Hu Y, Wang M. 2003. IL10 signaling via IL-10E1 is dependent on tyrosine phosphorylation in the IL-10Rα chain in human primary prostate cancer cell lines. Oncogene 22:3781–91 Shuai K, Schindler C, Prezioso VR, Darnell JE Jr. 1992. Activation of transcription by IFN-γ : tyrosine phosphorylation of a 91-kD DNA binding protein. Science 258:1808–12 Ramana CV, Grammatikakis N, Chernov M, Nguyen H, Goh KC, et al. 2000. Regulation of c-myc expression by IFN-γ through Stat1-dependent and -independent pathways. EMBO J. 19:263–72 Ramana CV, Gil MP, Han Y, Ransohoff RM, Schreiber RD, Stark GR. 2001. Stat1-independent regulation of gene expression in response to IFN-γ . Proc. Natl. Acad. Sci. USA 98:6674–79 Gil MP, Bohn E, O’Guin AK, Ramana CV, Levine B, et al. 2001. Biologic consequences of Stat1-independent IFN signaling. Proc. Natl. Acad. Sci. USA 98:6680–85 Riley JK, Takeda K, Akira S, Schreiber RD. 1999. Interleukin-10 receptor signaling through the JAK-STAT pathway. Requirement for two distinct receptorderived signals for anti-inflammatory action. J. Biol. Chem. 274:16513– 21 Vigano P, Somigliana E, Mangioni S, Vignali M, Vignali M, Di Blasio AM. 2002. Expression of interleukin-10 and its receptor is up-regulated in early pregnant versus cycling human endometrium. J. Clin. Endocrinol. Metab. 87:5730– 36

81. Ng SC, Gilman-Sachs A, Thaker P, Beaman KD, Beer AE, Kwak-Kim J. 2002. Expression of intracellular Th1 and Th2 cytokines in women with recurrent spontaneous abortion, implantation failures after IVF/ET or normal pregnancy. Am. J. Reprod. Immunol. 48:77–86 82. Orange S, Horvath J, Hennessy A. 2003. Preeclampsia is associated with a reduced interleukin-10 production from peripheral blood mononuclear cells. Hypertens. Pregnancy 22:1–8 83. Nagaeva O, Jonsson L, MinchevaNilsson L. 2002. Dominant IL-10 and TGF-β mRNA expression in γ δT cells of human early pregnancy decidua suggests immunoregulatory potential. Am. J. Reprod. Immunol. 48:9–17 84. Heikkinen J, Mottonen M, Komi J, Alanen A, Lassila O. 2003. Phenotypic characterization of human decidual macrophages. Clin. Exp. Immunol. 131:498–505 85. Moreau P, Adrian-Cabestre F, Menier C, Guiard V, Gourand L, et al. 1999. IL-10 selectively induces HLA-G expression in human trophoblasts and monocytes. Int. Immunol. 11:803–11 86. Levings MK, Sangregorio R, Sartirana C, Moschin AL, Battaglia M, et al. 2002. Human CD25+CD4+ T suppressor cell clones produce transforming growth factor β, but not interleukin 10, and are distinct from type 1 T regulatory cells. J. Exp. Med. 196:1335–46 87. de Kleer IM, Kamphuis SM, Rijkers GT, Scholtens L, Gordon G, et al. 2003. The spontaneous remission of juvenile idiopathic arthritis is characterized by CD30+ T cells directed to human heatshock protein 60 capable of producing the regulatory cytokine interleukin-10. Arthritis Rheum. 48:2001–10 88. Hara M, Kingsley CI, Niimi M, Read S, Turvey SE, et al. 2001. IL-10 is required for regulatory T cells to mediate tolerance to alloantigens in vivo. J. Immunol. 166:3789–96

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS 89. Ke Y, Kapp LM, Kapp JA. 2003. Inhibition of tumor rejection by γ δT cells and IL-10. Cell Immunol. 221:107–14 89a. Seymour BW, Friebertshauser KE, Peake JL, Pinkerton KE, Coffman RL, Gershwin LJ. 2002. Gender differences in the allergic response of mice neonatally exposed to environmental tobacco smoke. Dev. Immunol. 9(1):47–54 89b. Wilder RL. 1998. Hormones, pregnancy, and autoimmune diseases. Ann. NY Acad. Sci. 840:45–50 90. Dalal M, Kim S, Voskuhl RR. 1997. Testosterone therapy ameliorates experimental autoimmune encephalomyelitis and induces a T helper 2 bias in the autoantigen-specific T lymphocyte response. J. Immunol. 159:3–6 91. Liva SM, Voskuhl RR. 2001. Testosterone acts directly on CD4+ T lymphocytes to increase IL-10 production. J. Immunol. 167:2060–67 92. Kimball AB, Kawamura T, Tejura K, Boss C, Hancox AR, et al. 2002. Clinical and immunologic assessment of patients with psoriasis in a randomized, doubleblind, placebo-controlled trial using recombinant human interleukin 10. Arch. Dermatol. 138:1341–46 93. Tilg H, van Montfrans C, van den Ende A, Kaser A, van Deventer SJ, et al. 2002. Treatment of Crohn’s disease with recombinant human interleukin 10 induces the proinflammatory cytokine interferon γ . Gut 50:191–95 94. Llorente L, Zou W, Levy Y, RichaudPatin Y, Wijdenes J, et al. 1995. Role of interleukin 10 in the B lymphocyte hyperactivity and autoantibody production of human systemic lupus erythematosus. J. Exp. Med. 181:839–44 95. Park YB, Lee SK, Kim DS, Lee J, Lee CH, Song CH. 1998. Elevated interleukin-10 levels correlated with disease activity in systemic lupus erythematosus. Clin. Exp. Rheumatol. 16:283– 88 96. Ishida H, Muchamuel T, Sakaguchi

97.

98.

99.

100.

101.

102.

103.

104.

105.

106. 106a.

P1: IKH

969

S, Andrade S, Menon S, Howard M. 1994. Continuous administration of antiinterleukin 10 antibodies delays onset of autoimmunity in NZB/W F1 mice. J. Exp. Med. 179:305–10 Llorente L, Richaud-Patin Y, GarciaPadilla C, Claret E, Jakez-Ocampo J, et al. 2000. Clinical and biologic effects of anti-interleukin-10 monoclonal antibody administration in systemic lupus erythematosus. Arthritis Rheum. 43:1790–800 Isaacs A, Lindenmann J. 1957. Virus interference: I. The interferon. Proc. R. Soc. London Ser. B 147:258–67 Pestka S, Langer JA, Zoon KC, Samuel CE. 1987. Interferons and their actions. Annu. Rev. Biochem. 56:727–77 Pestka S. 1981. Cloning of human interferons. In Methods in Enzymology, ed. S Pestka, pp. 599–601. New York: Academic Pestka S. 1986. Interferon from 1981 to 1986. In Methods in Enzymology, ed. S Pestka, pp. 3–14. New York: Academic Lefevre F, Guillomot M, D’Andrea S, Battegay S, La Bonnardiere C. 1998. Interferon-δ: the first member of a novel type I interferon family. Biochimie 80:779–88 Roberts RM, Cross JC, Leaman DW. 1991. Unique features of the trophoblast interferons. Pharmacol. Ther. 51:329–45 Oritani K, Medina KL, Tomiyama Y, Ishikawa J, Okajima Y, et al. 2000. Limitin: an interferon-like cytokine that preferentially influences B-lymphocyte precursors. Nat. Med. 6:659–66 LaFleur DW, Nardelli B, Tsareva T, Mather D, Feng P, et al. 2001. Interferonκ, a novel type I interferon expressed in human keratinocytes. J. Biol. Chem. 276:39765–71 Chen J, Wood WI. 2003. Patent No. 6,300,475 Hardy MP, Owczarek CM, Jermiin L, Ejdeb¨ack M, Hertzog PJ. 2004. Characterization of the type I interferon locus

19 Feb 2004

12:12

970

106b.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

106c.

106d.

106e.

107.

108.

109.

110.

111.

112.

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL. and identification of novel genes. Genomics. In press Cross JC, Farin CE, Sharif SF, Roberts SM. 1990. Characterization of the antiviral activity constitutively produced by murine conceptuses: Absence of placental mRNAs for interferon alpha and beta. Mol. Reprod. Dev. 26:122–28 Bakhutashvili VI, Merabishvili DG, Gogitashvili RV, Korsantiia BM, Dzotsenidze LL. 1985. Physicochemical and biological properties of human placental amniotic interferon. Vopr. Virusol. 30:693–97 (In Russian) Duc-Gorran P, Chany C, Doly J. 1989. Unusually large interferon-alphalike mRNAs and high expression of interleukin-6 in human fetal annexes. J. Biol. Chem. 264:16507–11 Duc-Goiran P, Lebon P, Chany C. 1986. Measurement of interferon in human amniotic fluid and placental blood extract. Methods Enzymol. 119:541–51 Pestka S. 1983. The human interferons– from protein purification and sequence to cloning and expression in bacteria: before, between, and beyond. Arch. Biochem. Biophys. 221:1–37 Pestka S. 2000. The human interferon α species and receptors. Biopolymers 55:254–87 Stark GR, Kerr IM, Williams BR, Silverman RH, Schreiber RD. 1998. How cells respond to interferons. Annu. Rev. Biochem. 67:227–64 Berg K. 1982. Purification and characterization of murine and human interferons. A review of the literature of the 1970s. Acta Pathol. Microbiol. Immunol. Scand. Suppl. 279:1–136 Rubinstein M, Rubinstein S, Familletti PC, Miller RS, Waldman AA. 1978. Human leukocyte interferon purified to homogeneity. Science 202:1289–90 Rubinstein M, Rubinstein S, Familletti PC, Miller RS, Waldman AA, Pestka S. 1979. Human leukocyte interferon: production, purification to homogeneity,

113.

114.

115.

116.

117.

118. 119.

120.

121. 122. 123. 124. 125. 126.

and initial characterization. Proc. Natl. Acad. Sci. USA 76:640–44 Rubinstein M, Levy WP, Moschera JA, Lai C-Y, Hershberg RD, et al. 1981. Human leukocyte interferon: isolation and characterization of several molecular forms. Arch. Biochem. Biophys. 210:307–18 Stein S, Kenny C, Friesen HJ, Shively J, Del Valle U, Pestka S. 1980. NH2terminal amino acid sequence of human fibroblast interferon. Proc. Natl. Acad. Sci. USA 77:5716–19 Friesen HJ, Stein S, Pestka S. 1981. Purification of human. Fibroblast interferon by high-performance liquid chromatography. See Ref. 100, pp. 430–35 Evinger M, Rubinstein M, Pestka S. 1981. Antiproliferative and antiviral activities of human leukocyte interferons. Arch. Biochem. Biophys. 210:319–29 Maeda S, McCandliss R, Gross M, Sloma A, Familletti PC, et al. 1980. Construction and identification of bacterial plasmids containing nucleotide sequence for human leukocyte interferon. Proc. Natl. Acad. Sci. USA 77:7010–13 Goeddel DV, Pestka S. 1989. U.S. Patent No. 4,801,685 Henco K, Brosius J, Fujisawa A, Fujisawa J-I, Haynes JR, et al. 1985. Structural relationship of human interferon alpha genes and pseudogenes. J. Mol. Biol. 185:227–60 Pestka S. 1997. The human interferon alpha species and hybrid proteins. Semin. Oncol. 24(Suppl. 9):S9-4–9-17 Pestka S. 1998. U.S. Patent No. 5,789, 551 Pestka S. 1999. U.S. Patent No. 5,869, 293 Pestka S. 1999. U.S. Patent No. 6,001, 589 Pestka S. 2001. U.S. Patent No. 6,299, 870 Pestka S. 2001. U.S. Patent No. 6,300, 474 Ortaldo JR, Herberman RB, Harvey C,

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

IL-10, RELATED CYTOKINES, AND RECEPTORS

127.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

128.

129. 130.

131.

132.

133.

134.

135.

Osheroff P, Pan YC, et al. 1984. A species of human α interferon that lacks the ability to boost human natural killer activity. Proc. Natl. Acad. Sci. USA 81:4926–29 Pestka S. 1997. The interferon receptors. Semin. Oncol. 24(Suppl. 9):S9-18–9-40 Pestka S, Kotenko SV, Muthukumaran G, Izotova LS, Cook JR, Garotta G. 1997. The interferon γ (IFN-γ ) receptor: a paradigm for the multichain cytokine receptor. Cytokine Growth Factor Rev. 8:189–206 Darnell JE Jr. 1997. STATs and gene regulation. Science 277:1630–35 Fu XY, Schindler C, Improta T, Aebersold R, Darnell JE Jr. 1992. The proteins of ISGF-3, the interferon α-induced transcriptional activator, define a gene family involved in signal transduction. Proc. Natl. Acad. Sci. USA 89:7840–43 Ortaldo JR, Mason A, Rehberg E, Moschera J, Kelder B, et al. 1983. Effects of recombinant and hybrid recombinant human leukocyte interferons on cytotoxic activity of natural killer cells. J. Biol. Chem. 258:15011–15 Ortaldo JR, Mantovani A, Hobbs D, Rubinstein M, Pestka S, Herberman RB. 1983. Effects of several species of human leukocyte interferon on cytotoxic activity of NK cells and monocytes. Int. J. Cancer 31:285–89 Rehberg E, Kelder B, Hoal EG, Pestka S. 1982. Specific molecular activities of recombinant and hybrid leukocyte interferons. J. Biol. Chem. 257:11497– 502 Herberman RB, Ortaldo JR, Mantovani A, Hobbs DS, Kung H-F, Pestka S. 1982. Effect of human recombinant interferon on cytotoxic activity of natural killer (NK) cells and monocytes. Cell Immunol. 67:160–67 Ortaldo JR, Herberman RB, Pestka S. 1982. Augmentation of human natural killer cells with human leukocyte and human recombinant leukocyte interferon. In NK Cells and Other Natural Effector

136.

137.

138.

139.

140.

141.

142.

143.

144.

P1: IKH

971

Cells, ed. RB Herberman, pp. 1279–83. New York: Academic Ortaldo JR, Mason A, Rehberg E, Kelder B, Harvey C, et al. 1983. Augmentation of NK activity with recombinant and hybrid recombinant human leukocyte interferons. In The Biology of the Interferon System, ed. E De Maeyer, H Schellekens, pp. 353–58. Amsterdam: Elsevier Sci. Greiner JW, Hand PH, Noguchi P, Fisher PB, Pestka S, Schlom J. 1984. Enhanced expression of surface tumor-associated antigens on human breast and colon tumor cells after recombinant human leukocyte α-interferon treatment. Cancer Res. 44:3208–14 Greiner JW, Schlom J, Pestka S, Langer JA, Giacomini P, et al. 1985. Modulation of tumor associated antigen expression and shedding by recombinant human leukocyte and fibroblast interferons. Pharmacol. Ther. 31:209–36 Greiner JW, Guadagni F, Noguchi P, Pestka S, Colcher D, et al. 1987. Recombinant interferon enhances monoclonal antibody-targeting of carcinoma lesions in vivo. Science 235:895–98 Basham TY, Bourgeade MF, Creasey AA, Merigan TC. 1982. Interferon increases HLA synthesis in melanoma cells: interferon-resistant and -sensitive cell lines. Proc. Natl. Acad. Sci. USA 79:3265–69 Dolei A, Capobianchi MR, Ameglio F. 1983. Human interferon-γ enhances the expression of class I and class II major histocompatibility complex products in neoplastic cells more effectively than interferon-α and interferon-β. Infect. Immun. 40:172–76 Clemens MJ. 2003. Interferons and apoptosis. J. Interferon Cytokine Res. 23:277–92 Keay S, Grossberg SE. 1980. Interferon inhibits the conversion of 3T3-L1 mouse fibroblasts into adipocytes. Proc. Natl. Acad. Sci. USA 77:4099–103 Fisher PB, Hermo H Jr, Prignoli DR,

19 Feb 2004

12:12

972

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

145.

146.

147.

148.

149.

150.

151.

152.

153.

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL. Weinstein IB, Pestka S. 1984. Hybrid recombinant human leukocyte interferon inhibits differentiation in murine B-16 melanoma cells. Biochem. Biophys. Res. Commun. 119:108–15 Fisher PB, Prignoli DR, Hermo H Jr, Weinstein IB, Pestka S. 1985. Effects of combined treatment with interferon and mezerein on melanogenesis and growth in human melanoma cells. J. Interferon Res. 5:11–22 Fidler IJ. 2001. Regulation of neoplastic angiogenesis. J. Natl. Cancer Inst. Monogr. 28:10–14 Kerbel R, Folkman J. 2002. Clinical translation of angiogenesis inhibitors. Nat. Rev. Cancer 2:727–39 Gallagher G, Dickensheets H, Eskdale J, Izotova LS, Mirochnitchenko OV, et al. 2000. Cloning, expression and initial characterization of interleukin-19 (IL-19), a novel homologue of human interleukin-10 (IL-10). Genes Immun. 1:442–50 Chang C, Magracheva E, Kozlov S, Fong S, Tobin G, et al. 2003. Crystal structure of interleukin-19 defines a new subfamily of helical cytokines. J. Biol. Chem. 278:3308–13 Liao YC, Liang WG, Chen FW, Hsu JH, Yang JJ, Chang MS. 2002. IL-19 induces production of IL-6 and TNF-α and results in cell apoptosis through TNF-α. J. Immunol. 169:4288–97 Dumoutier L, Leemans C, Lejeune D, Kotenko SV, Renauld JC. 2001. Cutting edge: STAT activation by IL-19, IL-20 and mda-7 through IL-20 receptor complexes of two types. J. Immunol. 167:3545–49 Parrish-Novak J, Xu W, Brender T, Yao L, Jones C, et al. 2002. Interleukins 19, 20, and 24 signal through two distinct receptor complexes. Differences in receptor-ligand interactions mediate unique biological functions. J. Biol. Chem. 277:47517–23 Wolk K, Kunz S, Asadullah K, Sabat

154.

155.

156.

157.

158.

159.

160.

161.

162.

R. 2002. Cutting edge: immune cells as sources and targets of the IL-10 family members? J. Immunol. 168:5397–402 Nestle FO, Turka LA, Nickoloff BJ. 1994. Characterization of dermal dendritic cells in psoriasis. Autostimulation of T lymphocytes and induction of Th1 type cytokines. J. Clin. Invest. 94:202–9 Rocken M, Urban J, Shevach EM. 1994. Antigen-specific activation, tolerization, and reactivation of the interleukin 4 pathway in vivo. J. Exp. Med. 179:1885–93 Ghoreschi K, Thomas P, Breit S, Dugas M, Mailhammer R, et al. 2003. Interleukin-4 therapy of psoriasis induces Th2 responses and improves human autoimmune disease. Nat. Med. 9:40–46 Blumberg H, Conklin D, Xu WF, Grossmann A, Brender T, et al. 2001. Interleukin 20: discovery, receptor identification, and role in epidermal function. Cell 104:9–19 Rich BE. 2003. IL-20: a new target for the treatment of inflammatory skin disease. Expert Opin. Ther. Targets 7:165– 74 Dumoutier L, Louahed J, Renauld JC. 2000. Cloning and characterization of IL-10-related T cell-derived inducible factor (IL-TIF), a novel cytokine structurally related to IL-10 and inducible by IL-9. J. Immunol. 164:1814–19 Jiang H, Lin JJ, Su ZZ, Goldstein NI, Fisher PB. 1995. Subtraction hybridization identifies a novel melanoma differentiation associated gene, mda7, modulated during human melanoma differentiation, growth and progression. Oncogene 11:2477–86 Knappe A, Hor S, Wittmann S, Fickenscher H. 2000. Induction of a novel cellular homolog of interleukin-10, AK155, by transformation of T lymphocytes with herpesvirus saimiri. J. Virol. 74:3881–17 Dumoutier L, Van Roost E, Colau D, Renauld JC. 2000. Human interleukin-10related T cell-derived inducible factor:

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

IL-10, RELATED CYTOKINES, AND RECEPTORS

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

163.

164.

165.

166.

167.

168.

169.

170.

molecular cloning and functional characterization as an hepatocyte-stimulating factor. Proc. Natl. Acad. Sci. USA 97: 10144–49 Xie MH, Aggarwal S, Ho WH, Foster J, Zhang Z, et al. 2000. Interleukin (IL)22, a novel human cytokine that signals through the interferon receptor-related proteins CRF2-4 and IL-22R. J. Biol. Chem. 275:31335–39 Kotenko SV, Izotova LS, Mirochnitchenko OV, Esterova E, Dickensheets H, et al. 2001. Identification of the functional interleukin-22 (IL-22) receptor complex: the IL-10R2 chain (IL-10Rβ) is a common chain of both the IL-10 and IL-22 (IL-10-related T cell-derived inducible factor, IL-TIF) receptor complexes. J. Biol. Chem. 276:2725–32 Pittmann DD, Goad B, Lambert AJ, Clark E, Tan XY, Spaulding V, et al. 2001. IL-22 is a tightly-regulated IL-10like molecule that induces as acute-phase response and renal tubular basophilia. Genes Immun. 2:172 (Abstr.) Wei CC, Ho TW, Liang WG, Chen GY, Chang MS. 2003. Cloning and characterization of mouse IL-22 binding protein. Genes Immun. 4:204–11 Levitt RC, McLane MP, MacDonald D, Ferrante V, Weiss C, et al. 1999. IL-9 pathway in asthma: new therapeutic targets for allergic inflammatory disorders. J. Allergy Clin. Immunol. 103:S485–91 McLane MP, Haczku A, van de Rijn M, Weiss C, Ferrante V, MacDonald D, et al. 1998. Interleukin-9 promotes allergen-induced eosinophilic inflammation and airway hyperresponsiveness in transgenic mice. Am. J. Respir. Cell Mol. Biol. 19:713–20 Cookson W. 1999. The alliance of genes and environment in asthma and allergy. Nature 402:B5–11 Barnes KC, Neely JD, Duffy DL, Freidhoff LR, Breazeale DR, et al. 1996. Linkage of asthma and total serum

171.

172.

173.

174.

175.

176.

177.

178.

179.

P1: IKH

973

IgE concentration to markers on chromosome 12q: evidence from AfroCaribbean and Caucasian populations. Genomics 37:41–50 Nickel R, Wahn U, Hizawa N, Maestri N, Duffy DL, et al. 1997. Evidence for linkage of chromosome 12q15-q24.1 markers to high total serum IgE concentrations in children of the German Multicenter Allergy Study. Genomics 46:159– 62 Barnes KC, Freidhoff LR, Nickel R, Chiu YF, Juo SH, et al. 1999. Dense mapping of chromosome 12q13.12-q23.3 and linkage to asthma and atopy. J. Allergy Clin. Immunol. 104:485–91 Wilkinson J, Thomas NS, Morton N, Holgate ST. 1999. Candidate gene and mutational analysis in asthma and atopy. Int. Arch. Allergy Immunol. 118:265–67 Hayden C, Pereira E, Rye P, Palmer L, Gibson N, et al. 1997. Mutation screening of interferon-gamma (IFNgamma) as a candidate gene for asthma. Clin. Exp. Allergy 27:1412–16 Kotenko SV, Krause CD, Izotova LS, Pollack BP, Wu W, Pestka S. 1997. Identification and functional characterization of a second chain of the interleukin-10 receptor complex. EMBO J. 16:5894–903 Liu Y, Wei SH, Ho AS, de Waal MR, Moore KW. 1994. Expression cloning and characterization of a human IL-10 receptor. J. Immunol. 152:1821–29 Kotenko SV, Pestka S. 2000. Jak-Stat signal transduction pathway through the eyes of cytokine class II receptor complexes. Oncogene 19:2557–65 Dumoutier L, Lejeune D, Colau D, Renauld JC. 2001. Cloning and characterization of IL-22 binding protein, a natural antagonist of IL-10-related T cell-derived inducible factor/IL-22. J. Immunol. 166:7090–95 Kotenko SV, Izotova LS, Mirochnitchenko OV, Lee C, Pestka S. 1999. The intracellular domain of interferonα receptor 2c (IFN-αR2c) chain is

19 Feb 2004

12:12

974

180.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

181.

182.

183.

184.

185.

186.

187.

188.

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL. responsible for Stat activation. Proc. Natl. Acad. Sci. USA 96:5007–12 Wang M, Tan Z, Zhang R, Kotenko SV, Liang P. 2002. Interleukin 24 (MDA7/MOB-5) signals through two heterodimeric receptors, IL-22R1/IL-20R2 and IL-20R1/IL-20R2. J. Biol. Chem. 277:7341–47 Zhang R, Tan Z, Liang P. 2000. Identification of a novel ligand-receptor pair constitutively activated by ras oncogenes. J. Biol. Chem. 275:24436–43 Bromberg JF, Wrzeszczynska MH, Devgan G, Zhao Y, Pestell RG, et al. 1999. Stat3 as an oncogene. Cell 98:295–303 Gruenberg BH, Schoenemeyer A, Weiss B, Toschi L, Kunz S, et al. 2001. A novel, soluble homologue of the human IL-10 receptor with preferential expression in placenta. Genes Immun. 2:329–34 Xu W, Presnell SR, Parrish-Novak J, Kindsvogel W, Jaspers S, et al. 2001. A soluble class II cytokine receptor, IL22RA2, is a naturally occurring IL-22 antagonist. Proc. Natl. Acad. Sci. USA 98:9511–16 Lutfalla G, Holland SJ, Cinato E, Monneron D, Reboul J, et al. 1995. Mutant U5A cells are complemented by an interferon-α β receptor subunit generated by alternative processing of a new member of a cytokine receptor gene cluster. EMBO J. 14:5100–8 Novick D, Cohen B, Rubinstein M. 1994. The human interferon α/β receptor: characterization and molecular cloning. Cell 77:391–400 Pfeffer LM, Basu L, Pfeffer SR, Yang CH, Murti A, et al. 1997. The short form of the interferon α/β receptor chain 2 acts as a dominant negative for type I interferon action. J. Biol. Chem. 272: 11002–5 Hardy MP, Owczarek CM, Trajanovska S, Liu X, Kola I, Hertzog PJ. 2001. The soluble murine type I interferon receptor Ifnar-2 is present in serum, is independently regulated, and has both ago-

189.

190.

191.

192.

193.

194.

195.

196.

nistic and antagonistic properties. Blood 97:473–82 Reboul J, Gardiner K, Monneron D, Uze G, Lutfalla G. 1999. Comparative genomic analysis of the interferon/ interleukin-10 receptor gene cluster. Genome Res. 9:242–50 Huang EY, Madireddi MT, Gopalkrishnan RV, Leszczyniecka M, Su Z, et al. 2001. Genomic structure, chromosomal localization and expression profile of a novel melanoma differentiation associated (mda-7) gene with cancer specific growth suppressing and apoptosis inducing properties. Oncogene 20:7051–63 Ekmekcioglu S, Ellerhorst J, Mhashilkar AM, Sahin AA, Read CM, et al. 2001. Down-regulated melanoma differentiation associated gene (mda-7) expression in human melanomas. Int. J. Cancer 94:54–59 Madireddi MT, Dent P, Fisher PB. 2000. Regulation of mda-7 gene expression during human melanoma differentiation. Oncogene 19:1362–68 Madireddi MT, Dent P, Fisher PB. 2000. AP-1 and C/EBP transcription factors contribute to mda-7 gene promoter activity during human melanoma differentiation. J. Cell Physiol. 185:36–46 Caudell EG, Mumm JB, Poindexter N, Ekmekcioglu S, Mhashilkar AM, et al. 2002. The protein product of the tumor suppressor gene, melanoma differentiation-associated gene 7, exhibits immunostimulatory activity and is designated IL-24. J. Immunol. 168:6041– 46 Jiang H, Su ZZ, Lin JJ, Goldstein NI, Young CS, Fisher PB. 1996. The melanoma differentiation associated gene mda-7 suppresses cancer cell growth. Proc. Natl. Acad. Sci. USA 93: 9160–65 Madireddi MT, Su ZZ, Young CS, Goldstein NI, Fisher PB. 2000. Mda-7, a novel melanoma differentiation associated gene with promise for cancer gene

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

IL-10, RELATED CYTOKINES, AND RECEPTORS

197.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

198.

199.

200.

201.

202.

203.

204.

205.

therapy. Adv. Exp. Med. Biol. 465:239– 61 Mhashilkar AM, Schrock RD, Hindi M, Liao J, Sieger K, et al. 2001. Melanoma differentiation associated gene-7 (mda7): a novel anti-tumor gene for cancer gene therapy. Mol. Med. 7:271–82 Sarkar D, Su ZZ, Lebedeva IV, Sauane M, Gopalkrishnan RV, et al. 2002. mda7 (IL-24): signaling and functional roles. BioTechniques (Suppl.), pp. 30–39 Soo C, Shaw WW, Freymiller E, Longaker MT, Bertolami CN, et al. 1999. Cutaneous rat wounds express c49a, a novel gene with homology to the human melanoma differentiation associated gene, mda-7. J. Cell Biochem. 74:1– 10 Schaefer G, Venkataraman C, Schindler U. 2001. Cutting edge: FISP (IL4-induced secreted protein), a novel cytokine-like molecule secreted by Th2 cells. J. Immunol. 166:5859–63 Garn H, Schmidt A, Grau V, Stumpf S, Kaufmann A, et al. 2002. IL-24 is expressed by rat and human macrophages. Immunobiology 205:321–34 Ellerhorst JA, Prieto VG, Ekmekcioglu S, Broemeling L, Yekell S, et al. 2002. Loss of MDA-7 expression with progression of melanoma. J. Clin. Oncol. 20:1069–74 Sauane M, Gopalkrishnan RV, Sarkar D, Su ZZ, Lebedeva IV, et al. 2003. MDA7/IL-24: novel cancer growth suppressing and apoptosis inducing cytokine. Cytokine Growth Factor Rev. 14:35–51 Su ZZ, Lebedeva IV, Sarkar D, Gopalkrishnan RV, Sauane M, et al. 2003. Melanoma differentiation associated gene-7, mda-7/IL-24, selectively induces growth suppression, apoptosis and radiosensitization in malignant gliomas in a p53-independent manner. Oncogene 22:1164–80 Su ZZ, Lebedeva IV, Gopalkrishnan RV, Goldstein NI, Stein CA, et al. 2001. A combinatorial approach for selectively

206.

207.

208.

209.

210.

211.

212.

P1: IKH

975

inducing programmed cell death in human pancreatic cancer cells. Proc. Natl. Acad. Sci. USA 98:10332–37 Lebedeva IV, Su ZZ, Chang Y, Kitada S, Reed JC, Fisher PB. 2002. The cancer growth suppressing gene mda-7 induces apoptosis selectively in human melanoma cells. Oncogene 21:708–18 Lebedeva IV, Sarkar D, Su ZZ, Kitada S, Dent P, et al. 2003. Bcl-2 and BclxL differentially protect human prostate cancer cells from induction of apoptosis by melanoma differentiation associated gene-7, mda-7/IL-24. Oncogene. 22:8758–73 Sarkar D, Su ZZ, Lebedeva IV, Sauane M, Gopalkrishnan RV, et al. 2002. mda7 (IL-24) mediates selective apoptosis in human melanoma cells by inducing the coordinated overexpression of the GADD family of genes by means of p38 MAPK. Proc. Natl. Acad. Sci. USA 99:10054–59 Pataer A, Vorburger SA, Barber GN, Chada S, Mhashilkar AM, et al. 2002. Adenoviral transfer of the melanoma differentiation-associated gene 7 (mda7) induces apoptosis of lung cancer cells via up-regulation of the double-stranded RNA-dependent protein kinase (PKR). Cancer Res. 62:2239–43 Ekmekcioglu S, Ellerhorst JA, Mumm JB, Zheng M, Broemeling L, et al. 2003. Negative association of melanoma differentiation-associated gene (mda-7) and inducible nitric oxide synthase (iNOS) in human melanoma: MDA-7 regulates iNOS expression in melanoma cells. Mol. Cancer Ther. 2:9–17 Saeki T, Mhashilkar A, Swanson X, ZouYang XH, Sieger K, et al. 2002. Inhibition of human lung cancer growth following adenovirus-mediated mda-7 gene expression in vivo. Oncogene 21: 4558–66 Sauane M, Gopalkrishnan RV, Lebedeva I, Mei MX, Sarkar D, et al. 2003. mda7/IL-24 induces apoptosis of diverse

19 Feb 2004

12:12

976

213.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

214.

215.

216.

217.

218.

219. 220.

221.

222.

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL. cancer cell lines through JAK/STATindependent pathways. J. Cell Physiol. 196:334–45 Goris A, Marrosu MG, Vandenbroeck K. 2001. Novel polymorphisms in the IL-10 related AK155 gene (chromosome 12q15). Genes Immun. 2:284–86 Lutfalla G, Crollius HR, StangeThomann N, Jaillon O, Mogensen K, Monneron D. 2003. Comparative genomic analysis reveals independent expansion of a lineage-specific gene family in vertebrates: the class II cytokine receptors and their ligands in mammals and fish. BMC Genomics 4:29 Radhakrishnan R, Walter LJ, Subramaniam PS, Johnson HM, Walter MR. 1999. Crystal structure of ovine ˚ resolution. J. Mol. interferon-τ at 2.1 A Biol. 286:151–62 Karpusas M, Nolte M, Benton CB, Meier W, Lipscomb WN, Goelz S. 1997. The crystal structure of human interferon β at ˚ resolution. Proc. Natl. Acad. Sci. 2.2-A USA 94:11813–18 Radhakrishnan R, Walter LJ, Hruza A, Reichert P, Trotta PP, et al. 1996. Zinc mediated dimer of human interferonα2b revealed by X-ray crystallography. Structure 4:1453–63 Ealick SE, Cook WJ, Vijay-Kumar S, Carson M, Nagabhushan TL, et al. 1990. Three dimensional structure of recombinant human interferon-γ . Science 252:698–702 Deleted in proof Walter MR, Windsor WT, Nagabhushan TL, Lundell DJ, Lunn CA, et al. 1995. Crystal structure of a complex between interferon-γ and its soluble high-affinity receptor. Nature 376:230–35 Randal M, Kossiakoff AA. 2000. The ˚ structure of bovine interferon-γ ; 2.0 A assessment of the structural differences between species. Acta Crystallogr. D 56 (Pt. 1):14–24 Thiel DJ, le Du MH, Walter RL, D’Arcy A, Chene C, et al. 2000. Observation of

223.

224.

225.

226.

227. 228.

229.

230.

231. 232. 233. 234.

235.

236.

an unexpected third receptor molecule in the crystal structure of human interferonγ receptor complex. Structure 8:927–36 Randal M, Kossiakoff AA. 2001. The structure and activity of a monomeric interferon-γ :α-chain receptor signaling complex. Structure 9:155–63 Zdanov A, Schalk-Hihi C, Gustchina A, Tsang M, Weatherbee J, Wlodawer A. 1995. Crystal structure of interleukin-10 reveals the functional dimer with an unexpected topological similarity to interferon γ . Structure 3:591–601 Walter MR, Nagabhushan TL. 1995. Crystal structure of interleukin 10 reveals an interferon gamma-like fold. Biochemistry 34:12118–25 Nagem RA, Colau D, Dumoutier L, Renauld JC, Ogata C, Polikarpov I. 2002. Crystal structure of recombinant human interleukin-22. Structure 10:1051–62 Deleted in proof de Vos AM, Ultsch M, Kossiakoff AA. 1992. Human growth hormone and extracellular domain of its receptor: crystal structure of the complex. Science 255:306–12 Josephson K, Logsdon NJ, Walter MR. 2001. Crystal structure of the IL-10/IL10R1 complex reveals a shared receptor binding site. Immunity 15:35–46 Chill JH, Quadt SR, Levy R, Schreiber G, Anglister J. 2003. The human type I interferon receptor: NMR structure reveals the molecular basis of ligand binding. Structure 11:791–802 Deleted in proof Deleted in proof Deleted in proof Moyle WR, Campbell RK, Myers RV, Bernard MP, Han Y, Wang X. 1994. Co-evolution of ligand-receptor pairs. Nature 368:251–55 Koretke KK, Lupas AN, Warren PV, Rosenberg M, Brown JR. 2000. Evolution of two-component signal transduction. Mol. Biol. Evol. 17:1956–70 Deleted in proof

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS 237. Altmann SM, Mellon MT, Distel DL, Kim CH. 2003. Molecular and functional analysis of an interferon gene from the zebrafish, Danio rerio. J. Virol. 77:1992– 2002 238. Goldman LA, Cutrone EC, Dang A, Hao X, Lim JK, Langer JA. 1998. Mapping human interferon-α (IFN-α2) binding determinants of the type I interferon receptor subunit IFNAR-1 with human/bovine IFNAR-1 chimeras. Biochemistry 37:13003–10 239. Cutrone EC, Langer JA. 2001. Identification of critical residues in bovine IFNAR-1 responsible for interferon binding. J. Biol. Chem. 276:17140–48 240. Abdel-Meguid SS, Shieh HS, Smith WW, Dayringer HE, Violand BN, Bentle LA. 1987. Three-dimensional structure of a genetically engineered variant of porcine growth hormone. Proc. Natl. Acad. Sci. USA 84:6434–37 241. Piehler J, Schreiber G. 1999. Mutational and structural analysis of the binding interface between type I interferons and their receptor Ifnar2. J. Mol. Biol. 294:223–37 242. Piehler J, Roisman LC, Schreiber G. 2000. New structural and functional aspects of the type I interferon-receptor interaction revealed by comprehensive mutational analysis of the binding interface. J. Biol. Chem. 275:40425–33 243. Runkel L, deDios C, Karpusas M, Betzenhauser M, Muldowney C, et al. 2000. Systematic mutational mapping of sites on human interferon-β-1a that are important for receptor binding and functional activity. Biochemistry 39:2538–51 244. Bennett MJ, Schlunegger MP, Eisenberg D. 1995. 3D domain swapping: a mechanism for oligomer assembly. Protein Sci. 4:2455–68 245. Josephson K, DiGiacomo R, Indelicato SR, Iyo AH, Nagabhushan TL, et al. 2000. Design and analysis of an engineered human interleukin-10 monomer. J. Biol. Chem. 275:13552–57

P1: IKH

977

246. Syto R, Murgolo NJ, Braswell EH, Mui P, Huang E, Windsor WT. 1998. Structural and biological stability of the human interleukin 10 homodimer. Biochemistry 37:16943–51 247. Zdanov A, Schalk-Hihi C, Menon S, Moore KW, Wlodawer A. 1997. Crystal structure of Epstein-Barr virus protein BCRF1, a homolog of cellular interleukin-10. J. Mol. Biol. 268:460–67 248. Jones BC, Logsdon NJ, Josephson K, Cook J, Barry PA, Walter MR. 2002. Crystal structure of human cytomegalovirus IL-10 bound to soluble human IL-10R1. Proc. Natl. Acad. Sci. USA 99:9404–9 249. Livnah O, Johnson DL, Stura EA, Farrell FX, Barbone FP, et al. 1998. An antagonist peptide-EPO receptor complex suggests that receptor dimerization is not sufficient for activation. Nat. Struct. Biol. 5:993–1004 250. Remy I, Wilson IA, Michnick SW. 1999. Erythropoietin receptor activation by a ligand-induced conformation change. Science 283:990–93 251. Soh J, Donnelly RJ, Kotenko S, Mariano TM, Cook JR, et al. 1994. Identification and sequence of an accessory factor required for activation of the human interferon-γ receptor. Cell 76:796– 803 252. Landar A, Curry B, Parker MH, DiGiacomo R, Indelicato SR, et al. 2000. Design, characterization, and structure of a biologically active single-chain mutant of human IFN-gamma. J. Mol. Biol. 299:169–79 253. Krause CD, Lunn CA, Izotova LS, Mirochnitchenko O, Kotenko SV, et al. 2000. Studies on signaling of covalent heterodimers of interferon-γ : evidence for one-sided signaling in the active tetrameric receptor complex. J. Biol. Chem. 275:22995–3004 254. Ding Y, Qin L, Zamarin D, Kotenko SV, Pestka S, et al. 2001. Differential IL-10R1 expression plays a critical role

19 Feb 2004

12:12

978

255.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

256.

257.

258.

259.

260.

261.

262.

263.

264.

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

P1: IKH

PESTKA ET AL. in IL-10-mediated immune regulation. J. Immunol. 167:6884–92 Crepaldi L, Gasperini S, Lapinet JA, Calzetti F, Pinardi C, et al. 2001. Upregulation of IL-10R1 expression is required to render human neutrophils fully responsive to IL-10. J. Immunol. 167:2312–22 Bazan JF. 1990. Structural design and molecular evolution of a cytokine receptor superfamily. Proc. Natl. Acad. Sci. USA 87:6934–38 Bazan JF. 1990. Shared architecture of hormone binidng in type I and type II interferon receptors. Cell 61:753–54 Harlos K, Martin DM, O’Brien DP, Jones EY, Stuart DI, et al. 1994. Crystal structure of the extracellular region of human tissue factor. Nature 370:662–66 Banner DW, D’Arcy A, Chene C, Winkler FK, Guha A, et al. 1996. The crystal structure of the complex of blood coagulation factor VIIa with soluble tissue factor. Nature 380:41–46 Hage T, Sebald W, Reinemer P. 1999. Crystal structure of the interleukin4/receptor α chain complex reveals a mosaic binding interface. Cell 97:271–81 Roisman LC, Piehler J, Trosset JY, Scheraga HA, Schreiber G. 2001. Structure of the interferon-receptor complex determined by distance constraints from double-mutant cycles and flexible docking. Proc. Natl. Acad. Sci. USA 98:13231–36 Chill JH, Nivasch R, Levy R, Albeck S, Schreiber G, Anglister J. 2002. The human interferon receptor: NMR-based modeling, mapping of the IFN-α2 binding site, and observed ligand-induced tightening. Biochemistry 41:3575–85 Logsdon NJ, Jones BC, Josephson K, Cook J, Walter MR. 2002. Comparison of interleukin-22 and interleukin-10 soluble receptor complexes. J. Interferon Cytokine Res. 22:1099–112 Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. 1997. The

265.

266.

267.

268.

269.

270.

271. 272.

273.

274.

CLUSTAL X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876– 82 Versteeg HH, Peppelenbosch MP, Spek CA. 2003. Tissue factor signal transduction in angiogenesis. Carcinogenesis 24:1009–13 Dumoutier L, Lejeune D, Hor S, Fickenscher H, Renauld JC. 2003. Cloning of a new type II cytokine receptor activating signal transducer and activator of transcription (STAT)1, STAT2 and STAT3. Biochem. J. 370:391–96 Aguet M, Dembic Z, Merlin G. 1988. Molecular cloning and expression of the human interferon-γ receptor. Cell 55:273–80 Bach EA, Aguet M, Schreiber RD. 1997. The IFNγ receptor: a paradigm for cytokine receptor signaling. Annu. Rev. Immunol. 15:563–91 Hemmi S, Bohni R, Stark G, Di Marco F, Aguet M. 1994. A novel member of the interferon receptor family complements functionality of the murine interferon γ receptor in human cells. Cell 76:803– 10 Uze G, Lutfalla G, Gresser I. 1990. Genetic transfer of a functional human interferon α receptor into mouse cells: cloning and expression of its cDNA. Cell 60:225–34 Pestka S. 1997. The interferon receptors. Semin. Oncol. 24: S9-18–S9-40 Ho AS, Liu Y, Khan TA, Hsu DH, Bazan JF, Moore KW. 1993. A receptor for interleukin 10 is related to interferon receptors. Proc. Natl. Acad. Sci. USA 90:11267–71 Lutfalla G, McInnis MG, Antonarakis SE, Uze G. 1995. Structure of the human CRFB4 gene: comparison with its IFNAR neighbor. J. Mol. Evol. 41:338– 44 Lutfalla G, Gardiner K, Uze G. 1993. A new member of the cytokine receptor

19 Feb 2004

12:12

AR

AR210-IY22-30.tex

AR210-IY22-30.sgm

LaTeX2e(2002/01/18)

IL-10, RELATED CYTOKINES, AND RECEPTORS

275.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

276.

277.

278.

gene family maps on chromosome 21 at less than 35 kb from IFNAR. Genomics 16:366–73 Pestka S, Kotenko SV, Muthukumaran G, Izotova L, Cook JR, Garotta G. 1997. The interferon γ (IFN-γ ) receptor: a paradigm for the multichain cytokine receptor. Cytokine Growth Factor Rev. 8:189–206 Senda T, Saitoh S, Mitsui Y. 1995. Refined crystal structure of recombinant ˚ resolution. murine interferon-β at 2.15 A J. Mol. Biol. 253:187–207 Klaus W, Gsell B, Labhardt AM, Wipf B, Senn H. 1997. The three-dimensional high resolution structure of human interferon α-2a determined by heteronuclear NMR spectroscopy in solution. J. Mol. Biol. 274:661–75 Josephson K, Jones BC, Walter LJ, DiGiacomo R, Indelicato SR, Walter MR. 2002. Non-competitive antibody neutralization of IL-10 revealed by protein engi-

279.

280.

281.

282.

P1: IKH

979

neering and x-ray crystallography. Structure 10:981–87 Muller YA, Ultsch MH, Kelley RF, de Vos AM. 1994. Structure of the extracellular domain of human tissue factor: location of the factor VIIa binding site. Biochemistry 33:10864–70 Muller YA, Ultsch MH, de Vos AM. 1996. The crystal structure of the extracellular domain of human tissue factor ˚ resolution. J. Mol. Biol. refined to 1.7 A 256:144–59 Muller YA, Kelley RF, de Vos AM. 1998. Hinge bending within the cytokine re˚ ceptor superfamily revealed by the 2.4 A crystal structure of the extracellular domain of rabbit tissue factor. Protein Sci. 7:1106–15 Zhang E, St Charles R, Tulinsky A. 1999. Structure of extracellular tissue factor complexed with factor VIIa inhibited with a BPTI mutant. J. Mol. Biol. 285:2089–104

Pestka.qxd

3/12/2004

2:27 PM

Page 1

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

C-1

Figure 1 Schematic model of the resting and activated Class 2 cytokine receptor complex. The IFN-γ receptor complex has been shown to be preassembled (57) (A), with interactions observed between two accessory receptor chains (green) and between the accessory receptor chain and the ligand-binding chain (red). Addition of ligand (light blue) induces a conformational change in the receptor complex that appears to separate the intracellular domains (B). Transmembrane helices are shown as either straight or kinked depending on the presence of conserved glycine residues in the transmembrane domains of IFN-γR1 and IFN-γR2. We speculate that other Class 2 cytokine receptor complexes will also be preassembled and undergo conformational changes when bound by ligand. If the ligand is known to be a monomer in solution, it is possible that two monomers could be found in a single four-chain receptor complex (B) or that a single monomer could signal from a two-chain receptor complex (C). IL-22 is shown as a monomer in this figure because it exists in solution as a monomer and forms a 1:1:1 complex with soluble IL-22R1 and soluble IL-10R2 (263). Because all Class 2 cytokine receptor ligands have a similar sixhelix core monomer structure, they are shown with similarly shaped ellipses that are either dimeric or monomeric.

Pestka.qxd

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-2

3/12/2004

2:27 PM

Page 2

PESTKA ET AL.

Figure 2 Evolution and radiation of the ligands and receptors of the Class 2 cytokine receptor family. Dendograms are shown for ligands (left), ligand-binding receptor chains (center), and accessory receptor chains (right). Forty-seven ligands, 66 ligand-binding receptor chains, and 38 accessory chains were aligned in each dendogram. The first 150 amino acids of each receptor chain (encoding the signal peptide, the entire first SD100 domain, the inter-SD100 domain linker, and the beginning of the second SD100 subdomain) and the entire ligand sequence were selected for these analyses. The first and third SD100 subdomains of IFN-αR1 are called IFN-αR1 N and IFN-αR1 C, respectively. Proteins of mammalian, avian, amphibian, or piscine origin are indicated with a “m,” “b,” “a,” or “f” respectively. “Hsa” stands for Homo sapiens, whereas “Fru” stands for Takifugu rubripes. Branches between nodes are drawn to scale, whereas branches from the final node to the ligand itself are not. Phylogenetic analysis was done with ClustalX software (264). Ligands or receptor chains known to induce antiviral activity have their lines to the tree drawn in red. Ligands or receptor chains known to induce intermediate amounts of antiviral activity have their lines to the tree drawn in green. IL-10-like ligands and their receptor chains have their lines to the tree drawn in blue. Branches with bootstrap values above 750 are shown with a thick line, while those with lower bootstrap values are shown with a thin line.

Pestka.qxd

3/12/2004

2:27 PM

Page 3

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORS

C-3

Figure 3 Structural classification of the Class 2 α-helical cytokines. IFN-α2b, IL22, and IL-10 are shown as the prototypic Type I IFN, monomeric, and dimeric IL10 family members, respectively. Helices A, B, and F are colored cyan, while helices C, D, and E are colored green; 310 helices are colored blue. Predicted quaternary structures of IL-20, IL-24, and IL-26 are noted by italics and underlined.

Figure 4 Structures of IL-10 family members. Ribbon diagrams are shown for cellular IL-10 (a), ebvIL-10 (b), cmvIL-10 (c), IFN-γ (d), IL-19 (e), and IL-22 (f ) . The color scheme is the same as Figure 3.

Pestka.qxd

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-4

3/12/2004

2:27 PM

Page 4

PESTKA ET AL.

Figure 5 Structures of the Type I IFNs. Ribbon diagrams are shown for the NMR structure of human IFN-α2a (a) and the crystal structures of human IFN-β (b) and ovine IFN-τ (c). The color code is the same as for Figure 3.

Figure 6 IL-10 family intercalated dimer receptor complexes. Structures shown are the IFN-γ/IFN-γR1 complex (a), cellular IL-10/IL-10R1 complex (b), and the cmvIL-10/IL-10R1 complex (c). Each complex is shown in two orthogonal views. A black line represents the putative position of the cell membrane.

3/12/2004 2:27 PM

Figure 7 Organization of a Class 2 cytokine receptor and its ligand-binding site. A schematic diagram (a) and ribbon diagram (b) of IL10R1 bound to IL-10 is presented. Receptor β-strands are gold, the loops are white, and 310 helices are blue. The binding loops that contact site 1a and site 1b are colored red and green, respectively. Black dots represent putative N-linked glycosylation sites.

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Pestka.qxd Page 5

IL-10, RELATED CYTOKINES, AND RECEPTORS C-5

Pestka.qxd

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-6

3/12/2004

2:27 PM

Page 6

PESTKA ET AL.

Figure 8 Structural comparison of the Class 2 cytokine receptors. Each receptor is labeled as described in the text. SUP1 is a structural super position of the D1 domains of IL-10R1 in gold with GHR in cyan. SUP2 is a structural superposition of IL-10R1 (gold), GHR (cyan), IFN-γR1 (magenta), IFN-αR2 (green), and TF (blue).

Pestka.qxd

3/12/2004

2:27 PM

Page 7

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-10, RELATED CYTOKINES, AND RECEPTORSE

C-7

Figure 9 The IFN-γ and IL-10 receptor binding sites. Ribbon diagram of IFN-γ (a) depicts one domain of IFN-γ helices A-D colored gold and helices E-F colored cyan. Residues that contact IFN-γR1 are colored red. Adjacent to the ribbon diagram is a space-filling model of the domain with ball and stick receptor residues (yellow bonds) that contact site 1a and site 1b shown. The same figures are shown for IL-10 in part (b) of the diagram. The IL-10 helices A-D are colored cyan and helices E-F are colored yellow. The site 1a-binding site is colored red and the site 1b site is colored green. The spacefilling model shows key IL-10 receptor residues in ball and stick representations (white bonds) showing interactions site 1a and site 1b.

Pestka.qxd

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

C-8

3/12/2004

2:28 PM

Page 8

PESTKA ET AL.

Figure 10 Distinct ligand orientations in the IFN-γ and IL10 receptor complexes. The D1 domains of IL-10R1 and IFN-γR1 are superimposed and shown as green and cyan surfaces, respectively. For clarity, only the segments of IL-10 (purple) and IFNγ (magenta) that contribute residues to the interface are shown.

Figure 11 Comparison of cytokine- and FVIIa-binding paradigms. The IFN-γ /IFNγR1 (a) and TF/FVIIa (b) complexes are shown. One IFN-γR1 chain and TF are shown in gold as molecular surfaces. Each receptor is shown in the same orientation. IFN-γ and FVIIa are shown as ribbons color coded by secondary structure. The second IFN-γR1 chain is also shown as a ribbon in Figure 11a.

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

Annual Review of Immunology Volume 22, 2004

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONTENTS FRONTISPIECE—Leonard A. Herzenberg and Leonore A. Herzenberg GENETICS, FACS, IMMUNOLOGY, AND REDOX: A TALE OF TWO LIVES INTERTWINED, Leonard A. Herzenberg and Leonore A. Herzenberg SELF- AND NONSELF-RECOGNITION BY C-TYPE LECTINS ON DENDRITIC CELLS, Teunis B.H. Geijtenbeek, Sandra J. van Vliet, Anneke Engering, Bert A. ’t Hart, and Yvette van Kooyk

x 1

33

TRANSCRIPTIONAL CONTROL OF EARLY B CELL DEVELOPMENT, Meinrad Busslinger

UBIQUITIN LIGASES AND THE IMMUNE RESPONSE, Yun-Cai Liu LIGANDS FOR L-SELECTIN: HOMING, INFLAMMATION, AND BEYOND, Steven D. Rosen

55 81 129

INTEGRINS AND T CELL–MEDIATED IMMUNITY, Jonathan T. Pribila, Angie C. Quale, Kristen L. Mueller, and Yoji Shimizu

157

MULTIPLE ROLES OF ANTIMICROBIAL DEFENSINS, CATHELICIDINS, AND EOSINOPHIL-DERIVED NEUROTOXIN IN HOST DEFENSE, De Yang, Arya Biragyn, David M. Hoover, Jacek Lubkowski, and Joost J. Oppenheim

STARTING AT THE BEGINNING: NEW PERSPECTIVES ON THE BIOLOGY OF MUCOSAL T CELLS, Hilde Cheroutre THE BCR-ABL STORY: BENCH TO BEDSIDE AND BACK, Stephane Wong and Owen N. Witte

181 217 247

CD40/CD154 INTERACTIONS AT THE INTERFACE OF TOLERANCE AND IMMUNITY, Sergio A. Quezada, Lamis Z. Jarvinen, Evan F. Lind, and Randolph J. Noelle

307

THE THREE ES OF CANCER IMMUNOEDITING, Gavin P. Dunn, Lloyd J. Old, and Robert D. Schreiber

AUTOIMMUNE AND INFLAMMATORY MECHANISMS IN ATHEROSCLEROSIS, Georg Wick, Michael Knoflach, and Qingbo Xu THE DYNAMIC LIFE OF NATURAL KILLER CELLS, Wayne M. Yokoyama, Sungjin Kim, and Anthony R. French

329 361 405

v

P1: FRK

February 27, 2004

vi

22:25

Annual Reviews

AR210-FM

CONTENTS

THE ROLE OF COMPLEMENT IN THE DEVELOPMENT OF SYSTEMIC LUPUS ERYTHEMATOSUS, Anthony P. Manderson, Marina Botto, and Mark J. Walport

DROSOPHILA: THE GENETICS OF INNATE IMMUNE RECOGNITION AND RESPONSE, Catherine A. Brennan and Kathryn V. Anderson RAGS AND REGULATION OF AUTOANTIBODIES, Mila Jankovic,

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rafael Casellas, Nikos Yannoutsos, Hedda Wardemann, and Michel C. Nussenzweig

431 457

485

THE ROLE OF SUPPRESSORS OF CYTOKINE SIGNALING (SOCS) PROTEINS IN REGULATION OF THE IMMUNE RESPONSE, Warren S. Alexander and Douglas J. Hilton

503

NATURALLY ARISING CD4+

REGULATORY T CELLS FOR IMMUNOLOGIC SELF-TOLERANCE AND NEGATIVE CONTROL OF IMMUNE RESPONSES, Shimon Sakaguchi PHOSPHOINOSITIDE 3-KINASE: DIVERSE ROLES IN IMMUNE CELL ACTIVATION, Jonathan A. Deane and David A. Fruman IMMUNITY TO TUBERCULOSIS, Robert J. North and Yu-Jin Jung MOLECULAR DEFECTS IN HUMAN SEVERE COMBINED IMMUNODEFICIENCY AND APPROACHES TO IMMUNE RECONSTITUTION, Rebecca H. Buckley PHYSIOLOGICAL CONTROL OF IMMUNE RESPONSE AND INFLAMMATORY TISSUE DAMAGE BY HYPOXIA-INDUCIBLE FACTORS AND ADENOSINE A2A RECEPTORS, Michail V. Sitkovsky, Dmitriy Lukashev, Sergey Apasov, Hidefumi Kojima, Masahiro Koshiba, Charles Caldwell, Akio Ohta, and Manfred Thiel

531 563 599

625

657

T LYMPHOCYTE–ENDOTHELIAL CELL INTERACTIONS, Jaehyuk Choi, David R. Enis, Kian Peng Koh, Stephen L. Shiao, and Jordan S. Pober

683

IMMUNOLOGICAL MEMORY TO VIRAL INFECTIONS, Raymond M. Welsh, Liisa K. Selin, and Eva Szomolanyi-Tsuda

711

CENTRAL MEMORY AND EFFECTOR MEMORY T CELL SUBSETS: FUNCTION, GENERATION, AND MAINTENANCE, Federica Sallusto, Jens Geginat, and Antonio Lanzavecchia CONTROL OF T CELL VIABILITY, Philippa Marrack and John Kappler

745 765

ASTHMA: MECHANISMS OF DISEASE PERSISTENCE AND PROGRESSION, Lauren Cohn, Jack A. Elias, and Geoffrey L. Chupp

789

CD1: ANTIGEN PRESENTATION AND T CELL FUNCTION, Manfred Brigl and Michael B. Brenner

817

CHEMOKINES IN INNATE AND ADAPTIVE HOST DEFENSE: BASIC CHEMOKINESE GRAMMAR FOR IMMUNE CELLS, Antal Rot and Ulrich H. von Andrian

891

P1: FRK

February 27, 2004

22:25

Annual Reviews

AR210-FM

CONTENTS

vii

INTERLEUKIN-10 AND RELATED CYTOKINES AND RECEPTORS, Sidney Pestka, Christopher D. Krause, Devanand Sarkar, Mark R. Walter, Yufang Shi, and Paul B. Fisher

929

INDEXES

Annu. Rev. Immunol. 2004.22:929-979. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Subject Index Cumulative Index of Contributing Authors, Volumes 12–22 Cumulative Index of Chapter Titles, Volumes 12–22

ERRATA An online log of corrections to Annual Review of Immunology chapters may be found at http://immunol.annualreviews.org/errata.shtml

981 1011 1018