Annual Review of Immunology Volume 17 1999

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

P1: H

February 12, 1999 17:20 Annual Reviews AR078-00 AR78-FrontisP

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

9:49

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Annu. Rev. Immunol. 1999. 17:1–17 c 1999 by Annual Reviews. All rights reserved Copyright °

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

DISCOVERING THE ORIGINS OF IMMUNOLOGICAL COMPETENCE Jacques F. A. P. Miller The Walter and Eliza Hall Institute of Medical Research, P.O. Royal Melbourne Hospital, Victoria 3050, Australia; e-mail: [email protected] KEY WORDS:

lymphocytes, B cells, T cells, thymus, thymectomy

ABSTRACT Work done in the late 1950s and in the 1960s revealed the role of the thymus in virus-induced leukemia in mice. Thymectomizing mice at birth to test whether the virus first multiplied in thymus tissue and then spread elsewhere ultimately led to the conclusion that the thymus was essential to the normal development of the immune system. Subsequent testing to try to understand how the thymus contributes to the pool of immunocompetent lymphocytes opened a new chapter in immunology and required a reappraisal of many immunological phenomena and an understanding of the molecular interactions that take place during cell-tocell interactions.

Early Years and How I Ended Up in Australia Both my father and mother were born in Paris in 1896. My paternal grandfather, Francis Meunier, was the headmaster of Lyc´ee Henri IV in Paris, a learned man who had written books on the Greek and Latin languages. During the first World War, my father, Maurice Meunier, acted as interpreter of English for the British troops that came to France. In 1919, he married and left for China, having found a job in a French bank in Peking (now known as Beijing). He spent some 22 years in China and Japan, eventually becoming Manager of the Franco-Chinese Bank in Shanghai. Apart from English, he spoke Spanish fluently and learned Mandarin Chinese, which he could write, and also Japanese, which he wrote and spoke. In 1930, my mother, who had followed my father to China, returned to France for health reasons. Finding that she was pregnant, she decided to have the baby in France, and so, having been conceived in China, I was born in France, in 1 0732-0582/99/0410-0001$08.00

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

2

QC: NBL/tkj

9:49

T1: NBL

Annual Reviews

AR078-01

MILLER

Nice, in April 1931. In 1932, she went back to China with her three children, Jacqueline, the eldest, Jeanine, her second, and me, but she was back in France in 1935, both for her health and to allow Jacqueline to receive a “good” education at a boarding school. A year later, when we were just on the verge of returning to China, Jacqueline was diagnosed with pulmonary tuberculosis. Because of this, the family decided to go to Switzerland which, at that time, was the place where tuberculosis was supposed to be cured. We spent three years near Lausanne, and I do remember very well the doctor explaining to my mother what tuberculosis was and how little was known about the body’s resistance to such types of infections. In March 1939, my father joined us on a long service leave, but when the second world war broke out six months later, he was recalled to China. We hurriedly left Lausanne by car, crossing Northern Italy to Trieste, where we managed to get the last passenger boat out of Trieste. When France capitulated, the French concession in Shanghai was taken over by Vichy officials, but my father, who did not accept France’s surrender, rallied to the Gaullists and became active politically. He smuggled young Frenchmen, who wanted to join the British forces, out of the French concession onto British ships leaving for Britain. In 1940, he was actually invited by the British War Office to join the London Headquarters as a link between the French and British Treasury. But in December of that year, only a few years before the discovery of streptomycin, Jacqueline died, aged 17. So my father finally declined the offer from London for the family’s sake, and especially because these were the months of the blitz. Nevertheless it was obvious that he had to leave Shanghai for he was next on the list of Gaullists to be arrested by the Vichy officials. He also knew that the Japanese would enter the war very soon and that he would be at great risk, as he spoke and wrote their language fluently. So some kind of deal was made with the British authorities in Shanghai: We were given British passports and our surname was translated into English–hence Miller. We left in August 1941, taking the last cargo boat out of Shanghai bound for Batavia (now known as Jakarta). There we boarded a passenger ship and arrived in Sydney on the 25th of September 1941, barely three months before the bombing of Pearl Harbor. The Australians in Sydney did not recognize French banking credentials and thus would not employ my father on an equal footing. So he founded, together with another Frenchman, the Free French Delegation which took over the activities of the previous consulate, at that time defunct because Australia did not recognize Vichy. He offered his services to the Australian Government to translate any Japanese documents or information as required. He was also active in the war effort and helped with the taking in of supplies to New Caledonia. Prior to arriving in Australia, Jeanine and I had never been to school. We had teachers at home, wherever we lived. The last one in Shanghai was a 36-year

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

9:49

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ORIGINS OF IMMUNOLOGICAL COMPETENCE

3

old Viennese with a PhD who had escaped a Nazi prison. We had great fun with him and thought he must be very clever and wise to have a PhD at such a young age! Because my father had been impressed by the knowledge, culture, and broadmindedness of the Jesuits in Shanghai, with whom any subject could be discussed, he decided that I should go to a Jesuit college. He also thought that going to such a school would help me to get a better English accent and at least good manners! So I went to St Aloysius’ College, and as there was a convent next to this, Jeanine went there. Although we did have a month’s course in English before leaving Shanghai, we knew and understood so few words that we failed most of our exams during the first year, but we topped them all afterwards! At St Aloysius I met and frequented a brilliant young Austrian boy (a refugee from Vienna) who was a year ahead of me (fortunately, because he also topped his class). His name was Gus Nossal and we became life-long friends. I have followed one year behind in his footsteps first at school, then at the Sydney Medical School, and finally at the Royal Prince Alfred Hospital in Sydney. Because I had witnessed Jacqueline’s illness and had a great distaste for violence and war, I had wanted from an early age to study medicine and, if possible, to go into medical research. And so I was pleased to interrupt my medical studies for a year’s research as a BScMed science student in the laboratory of Professor de Burgh, again following in the footsteps of Gus Nossal. I too was given the task of deciphering how ectromelia virus multiplied in liver cells, but rather than continuing on the line of work that previous BScMed students had performed with normal liver, I thought it more interesting to determine whether the virus might interfere with some crucial biochemical events during liver regeneration after partial hepatectomy. Two papers resulted from this work (1, 2). After receiving my medical degree and doing an internship in Sydney, I applied for what was called a Gaggin Research Fellowship, advertised in the Medical Journal of Australia. It was given by the University of Queensland, Brisbane, and offered a return fare to the United Kingdom and a salary for two years in a Research Institute of the candidate’s own choosing. I was lucky to get this Fellowship, and I applied to many Institutes in England. Most were unable to take me, but one, the Chester Beatty Research Institute, an Institute of Cancer Research, in South Kensington, London, accepted me as a postdoctoral student for the PhD degree of the University of London. I arrived in the United Kingdom in 1958, not knowing exactly what I was going to do. Many of the scientists at the Institute were heavily involved in searching for new chemical carcinogenic compounds. Adding more compounds to an evergrowing list of carcinogenic agents did not interest me, as I would rather have used the experience I gained in my B.Med.Sci. year to work on some model in which pathogenetic mechanisms had to be investigated. Hence, I felt

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

4

QC: NBL/tkj

9:49

T1: NBL

Annual Reviews

AR078-01

MILLER

rather frustrated and did not really want to work in any of those laboratories. I was then told that the Institute had two satellites outside greater London, one being at a place called Pollards Wood, at Chalfont St Giles in Buckinghamshire. There, Dr. RJC Harris was working on the development of sarcomas in turkeys caused by the Rous sarcoma virus, and this line of investigation I thought might perhaps interest me. So I visited him. Instead of joining his group and working on some aspects of the Rous virus, he suggested that I might be willing to investigate the pathogenesis of lymphocytic leukemia induced in mice by what was presumed to be a virus that had recently been discovered by Ludwik Gross in the United States (3). This suited me perfectly. As a PhD student I was under the supervision of Dr. Harris, although my official supervisor had to be a full Professor of the University of London, which in my case was Professor Sir Alexander Haddow, the director of the Chester Beatty Research Institute. Pollards Wood was a large estate that had previously belonged to Bertram Mills, the circus owner. It had a magnificent Tudor-style mansion sitting in the middle of beautiful gardens and woods. The rooms had been refurbished as well-equipped laboratories and offices. There was also a kitchen and a dining room. All the buildings scattered throughout the estate that had previously housed animals such as horses, dogs, and elephants had also been converted to laboratories or animal quarters. A van from the main Institute in South Kensington came every day to bring mail and whatever supplies were required. Even though I was given only a small amount of space in one of the converted horse stables, it was a delight to work in such pleasant surroundings, away from the crowd, the noise, and the pollution of greater London.

The Thymus in Mouse Leukemia In the late 1950s, many scientists, including Haddow, did not believe that cancer in mammals could be caused by viruses. Gross had in fact not isolated a virus as such, but had been able to induce leukemia in so-called “low-leukemia” strains of mice, such as C3H, that normally did not develop the disease. He had done this by simply inoculating newborn mice with filtered extracts of leukemic tissues from “high-leukemia” strain mice, such as Ak [3]. Furthermore, not all low-leukemia strains developed the disease after such inoculation, e.g. C57BL mice were highly resistant, and even C3H mice in some laboratories were not as susceptible as mice of Gross’s own C3Hf/Gs strain. Repeating Gross’s observations using the strains of mice available at Pollards Wood might have taken months or years, and so I decided to write to Gross asking whether he would be kind enough to send his virus or mice harboring it. To my great relief, Gross accepted. My first experiment was just to repeat Gross’s original findings using his own C3H strain, and I soon confirmed the results he had described. It was

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

9:49

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ORIGINS OF IMMUNOLOGICAL COMPETENCE

5

known at that time that lymphocytic leukemia in mice involved the thymus and that adult thymectomy prevented spontaneous leukemia developing in high leukemic strain mice. It also prevented leukemia in low-leukemic strain mice otherwise induced by ionizing radiation and chemical carcinogens. As no one had yet investigated the role of the thymus in virus-induced leukemia, I thought this would be a good starting point for my PhD studies. Many questions had to be answered. Could thymectomy impair the leukemogenic process in virusinoculated mice? Could the virus multiply anywhere or only in thymus tissue? What would implantation of a normal syngeneic thymus achieve in mice whose own thymus had been surgically removed? Would thymectomized mice from high leukemic strains implanted with thymus tissue from mice of low leukemic strains develop leukemia? These questions seemed to me to be relevant, and I immediately set out to investigate them. But to do so, I required large numbers of mice of different inbred strains and hence considerable animal space that was not available. Six months after my arrival, however, Harris was offered the directorship of the Division of Virology of the Imperial Cancer Research Funds at Mill Hill, London. I was therefore left without an immediate supervisor, but I was really very fortunate to acquire his animal space and a small shack. I inoculated C3Hf/Gs mice with leukemic extracts immediately after birth and thymectomized these mice about 4 to 5 weeks later. None developed leukemia (4). Implantation, in such thymectomized inoculated mice of neonatal thymus tissue taken from uninoculated syngeneic mice restored the potential for leukemia development (5). A similar effect of thymectomy and thymus grafting had previously been observed in high leukemic strain mice or in low leukemic strain mice given irradiation or chemical carcinogenic agents (6). What would happen, however, if a high leukemic strain mouse (Ak) were thymectomized and implanted with thymus tissue from a low leukemic strain donor (C3H)? Since donors and hosts were allogeneic, this could be studied only in mice immunologically tolerant to the donor tissues. Although this aspect of my unsupervised work now appears quite naive, it did pave the way for me to perform experiments with immunologically tolerant mice, the phenomenon of tolerance having fascinated me since 1953 when, as a medical student, I had read in Nature the first report of its existence by Medawar and collaborators (7). Needing to learn how to inject newborn mice intravenously and how to skin graft to check whether tolerance had been established, I had therefore an excuse to approach Sir Peter Medawar and his group. An opportunity arose when Medawar delivered the Tercentenary Lecture of the Royal Society in London. It was an inspiring and stimulating talk given with clarity and wit. I learned how foreign tissue grafts were rejected and how tolerance to these might be induced experimentally by the inoculation of foreign hemopoietic cells into embryos or newborn animals. I

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

6

QC: NBL/tkj

9:49

T1: NBL

Annual Reviews

AR078-01

MILLER

approached Medawar with my problem, and he kindly asked his collaborator Leslie Brent to show me the techniques of intravenous injections into newborn mice and skin grafting. Leslie was a very nice teacher, and I am most grateful to him for spending so much time with me in those early days. Medawar, knowing that I was thymectomizing adult mice, said jokingly: “now that we have shown you how to inject newborn mice intravenously, perhaps one day you will show us how to thymectomize newborn mice.” What a prophetic statement! The results I obtained in immunologically tolerant Ak and C3H mice were clear-cut: Neoplasms developed in Ak thymuses grafted to thymectomized C3H mice. Some of the tumors arising in the thymus graft regressed after injection of lymphoid cells from C3H mice immunized against Ak tissues (8). Although I was immensely pleased with these results, they were not breaking new ground, being generally in accordance with previous work on leukemogenesis and on the induction and breaking of immunological tolerance. But the experience I gained in all this work was invaluable.

Effects of Neonatal Thymectomy As mentioned above, for leukemia to develop in those early days, the leukemic extracts had to be given at birth. Furthermore, inoculated C3Hf/Gs mice failed to develop the neoplasm when thymectomized after weaning but did so when subsequently grafted with syngeneic thymus tissue. What was most fascinating was the finding that grafting the thymus as late as 6 months after thymectomy still allowed leukemic transformation (5). The virus must clearly have remained latent, and it was indeed recoverable from the healthy nonleukemic tissues of neonatally inoculated thymectomized mice (9). Why should the virus be given at birth and not later? Where could it have multiplied? One possibility that I entertained, was that leukemogenic transformation occurred only if the virus could first multiply in the developing thymus. Thymectomy performed at weaning would remove the source of the malignant cells but not the virus, which would have spread to other sites and which would thus be available to transform cells whenever a neonatal thymus was grafted. If this were true, neonatal mice lacking a thymus from birth should no longer be susceptible to virus infection and would not develop leukemia when later grafted with thymus tissue. To test such a hypothesis, I had of course to teach myself the technique of neonatal thymectomy. After numerous attempts, I finally worked it out and thus had little immediate surgical mortality. Cannibalism by the mothers was, however, a major problem, and so I had to thymectomize large numbers of baby mice. I am most grateful to my wife Margaret who, although working as a technician in the Drosophila Laboratory at Pollards Wood, gave up much of her spare time at nights and weekends helping me thymectomize the babies and coaxing their mothers into not eating them! The survivors grew well at

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

9:49

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ORIGINS OF IMMUNOLOGICAL COMPETENCE

7

first but, after weaning, many wasted and died prematurely whether inoculated with virus or not. Adult thymectomy, on the other hand, had never shown any untoward effects such as weight loss or obvious pathology. This led me to conclude “that the thymus at birth may be essential to life” (10). Histological examination of the tissues of neonatally thymectomized mice revealed a marked deficiency of lymphocytes in the circulation and the lymphoid tissues, and many wasted mice had liver lesions suggesting infection by some hepatitis virus. Perhaps I might not have followed up these findings, had I not heard of the brilliant work of the famous immunologist Jim Gowans. He had recently shown that small lymphocytes were not short-lived cells, as had been thought before. On the contrary, they were immunologically competent cells with a long lifespan, recirculating from blood through lymphoid tissues into lymph and able to initiate immunological reactions when appropriately stimulated by antigen (11). Clearly, my neonatally thymectomized mice must have been immunodeficient, which accounted for their susceptibility to virus infections. I therefore tested their immune competence by grafting skin from allogeneic mice and from rats. The results were incredibly spectacular. The mice failed to reject such skin and failed to do so even when grafted before the onset of wasting. The grafts grew luxuriant tufts of hair and, to convince myself, I even transplanted some mice with four grafts, each from a different strain with a different color, some strains differing from the recipients at the strong histocompatibility locus, H-2. None of the grafts were rejected, and the recipients looked like having a patchwork quilt on their back! Since both Gowans and Medawar had firmly established that rejection of foreign skin grafts was mediated by lymphocytes, and since my mice were deficient in lymphocytes following neonatal thymectomy, it was logical for me to postulate that the thymus was the source of immunologically competent lymphocytes, at least during the neonatal period. At that time, a thymus immune function was unlikely to be accepted by the immunological community. There were many reasons for this. Unlike small lymphocytes taken by thoracic duct cannulation and unlike spleen and lymph node cells, thymus lymphocytes were generally poor in their ability to initiate immune reactions after adoptive transfer to appropriate recipients. Thoracic duct lymphocytes could home from blood into lymphoid tissues, “the only exception” being “the thymus in which very few small lymphocytes” appeared “to lodge” (11). The production of antibody-forming plasma cells and the formation of germinal centers, so prominent in spleen and lymph nodes, were not seen in normal thymus tissue. Defects in immune responsiveness had never been documented in mice whose thymus had been removed during adult life, a fact that had led some groups to conclude that “the thymus gland does not participate in the control of the immune response” (12). At a symposium

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

8

QC: NBL/tkj

9:49

T1: NBL

Annual Reviews

AR078-01

MILLER

on Cellular Aspects of Immunity (13), published in 1960, in which worldrenowned immunologists, including Burnet, Good, Lederberg, Medawar, and Mitchison, took part, not a single reference was made to the thymus or to its cells. Immunologists believed that, as a predominantly epithelial organ, the thymus had become vestigial during evolution and was just a graveyard for dying lymphocytes. Even in literary circles, the thymus seemed to have influenced writers with strange ideas. Lawrence Durrell in the Alexandria Quartet, for example, spoke of “the satiny skin that is given only to the thymus-dominated” (14), and as recently as 1971, one medical dictionary stated: “The function of the thymus gland is still obscure. One theory concerning its function is that it is concerned with general sexual maturity” (15). Faced with so much evidence against an immunological function for the thymus, I hesitated to publish my results immediately, but wished to receive some feedback from well-known immunologists. A summary was sent to some of them and I also spoke about my findings at various meetings in London, Oxford, and Perugia. When slides of my neonatally thymectomized mice bearing four different skin grafts, looking like a patchwork quilt, were shown at the British Society for Immunology Meeting in 1961, people were stunned, but my conclusions were regarded with skepticism. For example, Medawar was not convinced as was evident from a letter to me in which he wrote: “I take it that the thymic tissue seen in fishes is wholly or predominantly epithelial, as its phylogenetic origin suggests. It is a matter of some interest that many organs which seem to become redundant in the course of evolution undergo a sort of lymphocytic transformation” (16). Trivial criticisms abounded: What I had observed must surely have occurred only in the strain of mice that I had been using; my mice must have been in such poor health that any surgical trauma would prejudice their ability to reject skin grafts; whatever the thymus might have been doing in my mice, it could not possibly do in humans! At a Ciba Foundation Symposium on Tumour Viruses of Murine Origin held in Perugia in June 1961, the first international meeting where I presented results, my former mentor, RJC Harris, claimed the following: “Dr. Delphine Parrott in our laboratory has been thymectomizing day-old mice and there is at present no evidence that these animals are immunologically weaker than normal animals. They do not retain skin grafts, they are living and breeding quite normally. They do not die of laboratory infections” (17). All these comments and criticisms worried me a lot, and I decided again to repeat my work on even larger numbers of mice of different strains. But at that time, Sir Alexander Haddow, who had also attended the Perugia Meeting, urged me to immediately submit my initial results for publication. He suggested the medical journal, The Lancet, as I had already five papers in Nature and that journal might not have accepted a paper on such a controversial topic. I therefore sent a brief report to The Lancet

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

9:49

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ORIGINS OF IMMUNOLOGICAL COMPETENCE

9

and, contrary to the prevailing opinion, I postulated that “during embryogenesis the thymus would produce the originators of immunologically competent cells many of which would have migrated to other sites at about the time of birth. This would suggest that lymphocytes leaving the thymus are specially selected cells” (18). I had therefore proposed the bold postulate that the thymus was the site responsible for the development of immunologically competent small lymphocytes. This was the very first publication showing data supporting the immunological function of the thymus. Soon after this, I sent an application to present a paper at the New York Academy of Sciences meeting which was to be held in February 1962. This was accepted. It was my very first visit to the United States, in the middle of a harsh winter the likes of which I had never experienced. I gave my results in great detail, emphasizing that mice thymectomized at birth failed to reject skin both from totally unrelated strains (“H-2-incompatible”) and from other species such as rats (19). In the ensuing discussion, Martinez from Good’s group bluntly stated, without providing any data, that they also had shown that neonatally thymectomized mice were somewhat immunodeficient but, in contrast to my findings, prolonged skin graft survival occurred only in mice identical at the H-2 histocompatibility locus but differing at other weaker histocompatibility genes. Their mice did reject skin from H-2 incompatible strains. It seems strange that this group who later claimed to have had such results in 1961 (20), gave at this New York Meeting, in February 1962, a paper which was not on the thymus and in which the word thymus did not appear (21). They did, however, publish their findings later in 1962, again emphasizing the ability of their neonatally thymectomized mice to reject H-2-incompatible grafts (22). Such a discrepancy between their results and mine was later explained by their admission that they had not completely thymectomized their mice: “Careful autopsies performed in the thymectomized animals often revealed minute amounts of residual thymic tissue in these animals. With perfection of our technique a large proportion of neonatally thymectomized mice accepted H-2 incompatible grafts in contrast to partially thymectomized mice” (23). At the end of 1961, I had accumulated a large amount of data on the effects of thymectomy in newborn mice and on their rescue by normal syngeneic lymphocytes or by implanting thymus grafts. Although implantation of syngeneic thymus tissue allowed these mice to develop a normal immune system, grafting a thymus derived from a foreign strain induced specific immune tolerance to the histocompatibility antigens of the donor. Thus, lymphocytes developing in the thymus in the presence of foreign cells must have been deleted [i.e. “selectively thymectomized” as I suggested (24)]. Hence, by implication, the thymus should be the site where self-tolerance is imposed and where discrimination between self and nonself takes place. Sir Alexander Haddow again urged me

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

10

QC: NBL/tkj

9:49

T1: NBL

Annual Reviews

AR078-01

MILLER

to send all these detailed results for publication, and he kindly communicated them on my behalf to the Proceedings of the Royal Society Series B in late December 1961 [received by the Journal on January 5, 1962, and published later that year (24)]. I was also invited to present my work at the Royal Society in May 1962. My suggestion that the thymus could be involved in tolerance induction received strong support from Sir Macfarlane Burnet who stated in a lecture given at the University of London in June 1962: “If, as I believe, the thymus is the site where proliferation and differentiation of lymphocytes into clones with definable immunological functions occurs, we must also endow it with another function–the elimination or inhibition of self-reactive clones” (25). Burnet, having read the results I had obtained with neonatally thymectomized mice, was in fact one of the rare immunologists who believed in an immunological function of the thymus. It was during his 1962 visit to London that I had my first chance to speak to him. In 1963, I was awarded an Eleanor Roosevelt International Fellowship that allowed me to work for one year at the National Institutes of Health in Bethesda, in Dr. Lloyd Law’s department. There I neonatally thymectomized germfree mice and proved that these remained healthy after weaning, but yet were still unable to reject foreign skin grafts (26). With Law and collaborators, I consolidated my earlier observations that mice lacking a thymus were much more prone to develop neoplasms (27), thus adding weight to Burnet’s hypothesis of immunological surveillance. As mentioned before, Good’s group in the United States, working independently from me, soon confirmed the results I had already published. Furthermore, another group led by Waksman also obtained in rats similar results, which appeared in the scientific literature in 1962 (28).

The Thymus in the Adult In adult mice, thymectomy had for long been known not to produce any immune defects (29). Since total body irradiation destroyed lymphoid tissues, I reasoned that recovery of immune function following irradiation might be thymus-dependent. Mice were thymectomized after weaning and subjected to a sublethal dose of total irradiation. Whereas sham-thymectomized controls eventually fully recovered immune functions, the thymectomized mice remained immunocompromised. These results were sent to Nature and published in 1962 (30). At that time, a group at the Chester Beatty Research Institute in South Kensington, headed by Professor Koller, was studying the effects of heavy doses of irradiation on the hemopoietic system and its regeneration following an intravenous injection of bone marrow cells. Having just demonstrated the importance of the adult thymus in the recovery of immune function after

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

9:49

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

ORIGINS OF IMMUNOLOGICAL COMPETENCE

11

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

irradiation, I approached Koller’s co-workers and persuaded them to collaborate with me to test the hypothesis that, in “lethally” irradiated mice lacking a thymus, only the hemopoietic tissues but not the lymphoid tissues would regenerate after bone marrow injection. A very close collaboration thus ensued between Pollards Wood and the main Institute, and the results were exactly as predicted (31, 32). Since then, the technique of adult thymectomy, irradiation, and marrow protection has been used continuously for numerous experiments in cellular immunology.

Two Major Lymphocyte Subsets The work of Gowans, in the early 1960s, had shown that the recirculating small lymphocytes, in mammalian species, appeared to belong to a homogeneous population able to give rise to cells involved in both cellular and humoral immunity (11, 33). There was no reason to believe in the existence of separate subsets. If this were so, must all lymphocytes be thymus-derived? Of course neonatally thymectomized mice still had some lymphocytes, but these might have migrated from the thymus prior to birth. In birds, however, preventing the development of the other thymus-like organ, the bursa of Fabricius, by testosterone injection was known since 1956 to be associated with defects in antibody production in the mature bird (34). Burnet and his colleagues repeated and extended this work; they documented a division of labor among chicken lymphocytes, early bursectomy being associated with defects in antibody formation and early thymectomy with defects in some cellular immune responses (35). Since I had shown, however, that neonatal thymectomy in mice prevented both cellular and humoral immune responses (24, 36), Burnet was led to conclude that in “mammals it is highly probable that the thymus also carries out the function performed by the bursa of Fabricius in the chicken, which is to feed into the body the cells whose descendants will produce antibody” (37). But then, why did neonatally thymectomized mice show a deficiency of lymphocytes limited to those areas of lymph nodes and spleen known to be associated with changes induced by cellular immune responses, but not those areas where antibody-producing cells appeared (38)? A clue to this mystery came from a totally different line of investigation. Claman and his collaborators in Denver showed that irradiated mice receiving a mixture of marrow and thymus cells produced more antibody than controls given either cell source alone (39). As their model lacked genetic markers, it could not determine the origin of the antibody-forming cells. Thus the function of bone marrow cells might simply have been to protect the irradiated mice, thus allowing cells in the thymus inoculum to produce antibody. Davies and his collaborators, in Koller’s department, attempted to follow up this work by using adult thymectomized irradiated mice given bone marrow and thymus

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

12

QC: NBL/tkj

9:49

T1: NBL

Annual Reviews

AR078-01

MILLER

grafts from donors that had slight immunogenetic differences (40). These chimeras were challenged with sheep erythrocytes and their spleen cells transferred into irradiated recipients presensitized against either the thymus or the marrow donor. Those able to reject cells with the immunogenetic markers of the thymus donor produced antibody. Those immunized against the marrow donor produced much less. Since these transfer experiments were performed 30 days after irradiation and thymus grafting, at a time when the lymphoid cell population of the thymus graft had been entirely replaced by cells of marrow origin, as I had previously shown (41), the results were difficult to interpret. Thus hemolysins in the irradiated recipients presensitized against the thymus donor might well have been produced by marrow-derived cells that had first repopulated and then emigrated from the thymus graft. The antibody-producing cells would then have had the immunogenetic characteristics of the marrow donor and yet be thymus derived. Davies himself concluded: “It may be that thymus-derived cells can produce antibody, but only in the presence of cells of bone marrow origin. Equally cells of bone marrow origin may be the cells whose immunological potential is enhanced by association with cells of thymic origin. These are not problems which the present analysis can resolve” (42). In 1965, I was invited back to Australia by Professor Gustav Nossal, who had just been appointed director of the Walter and Eliza Hall Institute of Medical Research in Melbourne, to succeed Burnet. I was to lead a new laboratory at the Institute, and Gus had kindly chosen for me, as my first PhD student, a brilliant young man, Graham Mitchell, who had just graduated with first class honors from the University of Sydney Veterinary School. Graham was a delightful person to work with, and we too became life-long friends. We wanted to understand how the thymus contributed to the pool of immunocompetent recirculating small lymphocytes, and to achieve this we first investigated the ability of various cell types to restore immune functions in thymectomized mice. We inoculated cells from F1 hybrid mice into neonatally thymectomized or thymectomized-irradiated recipients of parental genotype, so as to have genetic markers. Since thymus cells were poor at initiating antibody responses, we used thoracic duct lymphocytes or thymus cells that had twice been serially transferred with antigen into two sets of irradiated mice (we even named these “educated thymus cells”). Bone marrow cells, “educated” thymus cells, or thoracic duct cells, when given alone to irradiated mice, produced only little or no antibody in response to sheep erythrocytes, as measured by a plaque technique on erythrocyte coated agar plates (each plaque representing a single antibodyforming cell). When, however, the “educated” thymus cells or thoracic duct cells were injected together with bone marrow cells, the plates were crowded with plaques. By using anti-H-2 sera, we were now able to determine which cell type produced the antibody. While the plates were incubating with specific

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

9:49

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ORIGINS OF IMMUNOLOGICAL COMPETENCE

13

antibodies able to eliminate either thymus-derived or bone-marrow derived cells, we waited anxiously for the results. My bet was that the antibodyforming cells would be thymus derived, but I have always bet the wrong horse! The results could not have been more spectacular: One set of plates had just nothing on it, the other was as crowded with plaques as before. Now, which set was which? We decoded the experiment and, of course, I lost my bet. But I was elated by results so convincing and so exciting that I felt just as “the lark at break of day arising from sullen earth, sings hymns at Heaven’s gate” (43)! Our work established, definitely, unequivocally, and for the first time, the existence, in species other than birds, of two major subsets of lymphocytes: antibody-forming cell precursors derived from lymphocytes in bone marrow, and thymus-derived cells essential to allow the former to respond to antigen by producing antibody. We sent a note to Nature (44) and a polished paper to the Proceedings of the National Academy of Science of the United States of America (45). The latter was unfortunately delayed by the 1967 Christmas mail. We of course performed a great deal of work that gave results as convincing as the early ones, and we proved beyond doubt that some interaction took place between these two major subsets of lymphocytes in antibody responses. We sent four papers to the Journal of Experimental Medicine, and these were accepted and published back to back (46–49). As a light exercise, I tried to find acronyms for the clumsy words thymus-derived, bone marrow–derived or antibody-forming cells precursor, but nothing pleased me. It was left to Ivan Roitt, the immunologist responsible for the world’s most popular textbook, to coin several years later the simple and minimalist terms, T and B cells (50)! How did the immunological community react to our findings? There was complete surprise, of course, but there was also disbelief when I presented these results at meetings held in the United States and Canada in 1968. At Val Morin, I was accused of “complicating things.” But the commonest and quite valid criticism of our view of how T and B cells collaborated was that two rare cells would never find each other. At a meeting in Brook Lodge, held in 1968, Gowans, who had clearly shown that recirculating small lymphocytes could initiate both cellular and humoral immune responses, stated: “Had it not been for Dr. Miller’s experiments I would have assumed that a single variety of small lymphocyte was involved in each of our experiments” (51). At the same meeting, Good was “concerned at separating thymus-derived from marrow-derived cells” since the former “are in fact, marrow-derived cells” (52). Even Burnet, despite his own work with chickens, expressed doubts “about the significance of results obtained in such biological monstrosities as pure line mice thymectomized, lethally irradiated, and salvaged by injection of bone marrow from another mouse” (53). The most sarcastic criticism came from Professor Bede Morris, the then Professor of Immunology at the John Curtin School of Medical

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

14

QC: NBL/tkj

9:49

T1: NBL

Annual Reviews

AR078-01

MILLER

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Research in Canberra, Australia, who likened B and T cells to the first and last letters of the word “bullshit”. In spite of all these criticisms, Graham and I persevered in our work. It was urgent to re-examine a multitude of immunological phenomena in terms of the two cell system: tolerance, memory, the carrier effect, autoimmunity, immune deficiency, genetically determined unresponsive states, original antigenic sin, etc. Within two to three years, the entire immunological community jumped on the band wagon, and since then, hardly an article has appeared in any immunological journal without mentioning the words T cells or B cells.

Conclusions Younger investigators working in cellular immunology are probably quite surprised at the account I have given. For today, the immunological function of the thymus is taken for granted, as if it had never ever been in doubt, and T and B cells have become household words. Research has progressed so fast in the last three decades that we can now probe the molecular basis of the interactions between T cells and their ligands and between T cells and other cells such as B cells and dendritic cells. Yet we have a long way to go, and after 40 years in medical research, I am still keen to find out exactly how lymphocyte homeostasis is maintained (e.g. 54), how T cells, B cells, and other cells interact (e.g. 55), and why the immune system fails to respond to self under normal conditions (e.g. 56, 57). In 1971 Burnet (58) stated: “None of my juniors seem to be worried as I am by the fact that the contribution of laboratory science to medicine has virtually come to an end.” A similar outlook on the future of surgery was held in 1930 by the famous surgeon Lord Moynihan (59): “We can surely never hope to see the craft of surgery made much more perfect than it is today. We are at the end of a chapter.” Yet would not both Burnet and Moynihan be greatly surprised and pleased by the technological breakthroughs and novel experimental approaches that have given us so much new knowledge both in surgery and immunology? And although we can employ numerous strategies to allow better survival of transplants, to deal with various forms of immunological aberrations, and to produce new vaccines, we still have to learn a great deal, in particular how to apply clinically the fundamental knowledge obtained from our bench work. I am thus in full agreement with the late Sir Karl Popper that “the deeper our learning, the more conscious, specific and articulate will be our knowledge of what we do not know, our knowledge of our ignorance” (60). As Sir Winston Churchill once said: “So much accomplished, so much still to be done.” ACKNOWLEDGMENTS I am indebted to the University of Queensland for the award of the Gaggin Fellowship which allowed me to pursue my early studies on thymus function

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

9:49

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ORIGINS OF IMMUNOLOGICAL COMPETENCE

15

in London. I thank the many colleagues who encouraged me during these early years, in particular the late Sir Alexander Haddow, who gave me complete freedom in choosing whatever line of work I thought was essential, and also the late Sir Peter Medawar, who gave me tremendous encouragement to continue what I was doing. I am of course most grateful to Professor Sir Gustav Nossal for having invited me to head a Unit in his Institute and for his enthusiastic encouragement and support, and I thank all my colleagues, assistants and students who have worked in close contact with me throughout my career. I am of course highly indebted to the various benefactors and granting agencies who have funded my work and last, but not least, to the many mice without which this work could not have been done. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Miller JFAP, de Burgh PM. 1957. Ectromelia virus multiplication in regenerating mouse liver. Aust. J. Exp. Biol. Med. Sci. 35:115–22 2. de Burgh PM, Miller JFAP. 1957. Cellular control in virus infection. Nature 175:550 3. Gross L. 1951. Pathogenic properties and “vertical” transmission of the mouse leukemia agent. Proc. Soc. Exp. Biol. Med. 78:342–48 4. Miller JFAP. 1959. Role of the thymus in murine leukaemia. Nature 183:1069 5. Miller JFAP. 1959. Fate of subcutaneous thymus grafts in thymectomized mice inoculated with leukaemic filtrates. Nature 184:1809–10 6. Miller JFAP. 1961. Aetiology and pathogenesis of mouse leukaemia. Adv. Cancer Res. 6:291–368 7. Billingham RE, Brent L, Medawar PB. 1953. “Actively acquired tolerance” of foreign cells. Nature 172:603–6 8. Miller JFAP. 1960. Studies on mouse leukaemia. III. The fate of thymus homografts in immunologically tolerant mice. Br. J. Cancer 14:244–55 9. Miller JFAP. 1960. Recovery of leukaemogenic agent from non-leukaemic tissues of thymectomized mice. Nature 187: 703 10. Miller JFAP. 1961. Analysis of the thymus influence in leukaemogenesis. Nature 191:248–49 11. Gowans JL, Gesner BM, McGregor DD. 1961. The immunological activity of lymphocytes. In: Biological Activity of the Leu-

12.

13. 14. 15. 16.

17.

18. 19. 20.

cocyte, Ciba Fdn. Study Group, ed. GEW Wolstenholme, M O’Connor, 10:32–44. London: Churchill MacLean LD, Zak SJ, Varco RL, Good RA. 1956. The role of the thymus in antibody production: an experimental study of the immune response in thymectomized rabbits. Transpl. Bull. 4:21–22 Wolstenholme GEW, O’Connor M. eds. 1960. Cellular Aspects of Immunity. Ciba Fdn. Symp. London: Churchill. 495 pp. Durrell L. 1957. Justine. London: Faber Medical Dictionary. 1971. London: Adam, Black. 29th ed. Miller JFAP. 1995. The discovery of thymus function. In Immunology: The Making of a Modern Science, ed. RB Gallagher, J Gilder, GJV Nossal, G Salvatore, pp. 75– 84. London: Academic Press Harris RJC. 1962. Discussion after Miller JFAP. Role of the thymus in virusinduced leukaemia. In Tumour Viruses of Murine Origin, ed. GEW Wolstenholme, M O’Connor, pp. 262–83. London: Churchill Miller JFAP. 1961. Immunological function of the thymus. Lancet 2:748–49 Miller JFAP. 1962. Role of the thymus in transplantation immunity. Ann. N. Y. Acad. Sci. 99:340–54 Good RA. 1994. The Minnesota scene: a crucial portal of entry to modern cellular immunology. In The Immunologic Revolution: Facts and Witnesses, ed. H Friedman, A Szentivanyi, pp. 105–68. Boca Raton, FL: CRC Press

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

16

QC: NBL/tkj

9:49

T1: NBL

Annual Reviews

AR078-01

MILLER

21. Warwick J, Archer OK, Good RA. 1962. Effect of previous homograft on passive transfer of delayed allergy. Ann. N. Y. Acad. Sci. 99:620–28 22. Martinez C, Kersey J, Papermaster BW, Good RA. 1962. Skin homograft survival in thymectomized mice. Proc. Soc. Exp. Biol. Med. 109:193–96 23. Martinez C, Dalmasso AP, Good RA. 1964. Homotransplantation of normal and neoplastic tissue in thymectomized mice. In The Thymus in Immunobiology, ed. RA Good, AE Gabrielsen, pp. 465–77. New York: Harper, Row 24. Miller JFAP. 1962. Effect of neonatal thymectomy on the immunological responsiveness of the mouse. Proc. R. Soc. London 156B:410–28 25. Burnet FM. 1962. The role of the thymus and related organs in immunity. Br. Med. J. 2:807–11 26. McIntire KR, Sell S, Miller JFAP. 1964. Pathogenesis of the post–neonatal thymectomy wasting syndrome. Nature 204:151– 55 27. Miller JFAP, Law LW, Ting RC. 1964. Influence of thymectomy on tumor induction by polyoma virus in C57BL mice. Proc. Soc. Exp. Biol. Med. 116:323–27 28. Arnason BG, Jankovic BD, Waksman BH. 1962. Effect of thymectomy on “delayed” hypersensitive reactions. Nature 194:99– 100 29. Miller JFAP, Osoba D. 1967. Current concepts of the immunological function of the thymus. Physiol. Rev. 47:437–520 30. Miller JFAP. 1962. Immunological significance of the thymus of the adult mouse. Nature 195:1318–19 31. Miller JFAP, Doak SMA, Cross AM. 1963. Role of the thymus in the recovery of the immune mechanism in the irradiated adult mouse. Proc. Soc. Exp. Biol. Med. 112:785–92 32. Cross AM, Leuchars E, Miller JFAP. 1964. Studies on the recovery of the immune response in irradiated mice thymectomized in adult life. J. Exp. Med. 119:837– 50 33. Howard JC, Gowans JL. 1972. The role of lymphocytes in antibody formation. III. The origin from small lymphocytes of cells forming direct and indirect haemolytic plaques to sheep erythrocytes in the rat. Proc. R. Soc. London 182B:193–209 34. Glick B, Chang TS, Jaap RG. 1956. The bursa of Fabricius and antibody production. Poultry Sci. 35:224–25 35. Warner NL, Szenberg A, Burnet FM. 1962. The immunological role of different lymphoid organs in the chicken. I. Dissociation

36.

37. 38.

39.

40.

41.

42.

43.

44. 45.

46.

47.

48.

of immunological responsiveness. Aust. J. Exp. Biol. Med. Sci. 40:373–88 Miller JFAP. 1963. Tolerance in the thymectomized animal. In La Tol´erance Acquise et la Tol´erance Naturelle a` l’´egard de Substances Antig´eniques D´efinies. pp. 47–75. Paris: CNRS Burnet FM. 1962. The thymus gland. Sci. Am. 207:50–57 Parrott DMV, de Sousa MAB, East J. 1966. Thymus-dependent areas in the lymphoid organs of neonatally thymectomized mice. J. Exp. Med. 123:191–204 Claman HN, Chaperon EA, Triplett RF. 1966. Thymus–marrow cell combinations –synergism in antibody production. Proc. Soc. Exp. Biol. Med. 122:1167–71 Davies AJS, Leuchars E, Wallis V, Marchant R, Elliott EV. 1967. The failure of thymus-derived cells to produce antibody. Transplantation 5:222–31 Dukor P, Miller JFAP, House W, Allman V. 1965. Regeneration of thymus grafts. I. Histological and cytological aspects. Transplantation 3:639–68 Davies AJS, Leuchars E, Wallis V, Marchant R, Sinclair NRStC, Elliott EV. 1968. The selective transfer test. An analysis of the primary response to sheep red cells. In Advance in Transplantation, ed. J Dausset, J Hamburger, G Math´e, pp. 97–100. Copenhagen: Munkshaard Shakespeare W. 1591. Sonnet No 29. In The Complete Oxford Shakespeare, eds. S Wells, G Taylor, I:375. Oxford: Oxford Univ. Press Miller JFAP, Mitchell GF. 1967. The thymus and the precursors of antigen-reactive cells. Nature 216:659–63 Mitchell GF, Miller JFAP. 1968. Immunological activity of thymus and thoracic duct lymphocytes. Proc. Natl. Acad. Sci. USA 59:296–303 Miller JFAP, Mitchell GF. 1968. Cell to cell interaction in the immune response. I. Hemolysin–forming cells in neonatally thymectomized mice reconstituted with thymus or thoracic duct lymphocytes. J. Exp. Med. 128:801–20 Mitchell GF, Miller JFAP. 1968. Cell to cell interaction in the immune response. II. The source of hemolysin–forming cells in irradiated mice given bone marrow and thymus or thoracic duct lymphocytes. J. Exp. Med. 128:821–37 Nossal GJV, Cunningham AJ, Mitchell GF, Miller JFAP. 1968. Cell to cell interaction in the immune response. III. Chromosomal marker analysis of single antibody– forming cells in reconstituted, irradiated,

P1: SAT/VKS

P2: NBL/plb

January 21, 1999

9:49

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

ORIGINS OF IMMUNOLOGICAL COMPETENCE

49.

50.

Annu. Rev. Immunol. 1999.17:1-17. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

51.

52.

53. 54.

or thymectomized mice. J. Exp. Med. 128: 839–53 Martin WJ, Miller JFAP. 1968. Cell to cell interaction in the immune response. IV. Site of action of anti–lymphocyte globulin. J. Exp. Med. 128:855–74 Roitt IM, Greaves MF, Torrigiani G, Brostoff J, Playfair JHL. 1969. The cellular basis of immunological responses. Lancet 2:367–69 Gowans JL. 1969. Discussion after Miller JFAP. In Immunological Tolerance. A Reassesment of Mechanisms of the Immune Response, ed. M Landy, W Braun, p. 169. New York: Academic Press Good RA. Discussion. 1969. Discussion after Miller JFAP. In Immunological Tolerance. A Reassesment of Mechanisms of the Immune Response, ed. M Landy, W Braun, p. 136. New York: Academic Press Burnet FM. 1972. Auto–immunity and Auto–immune Disease, 45. Lancaster, UK: MTP Berzins SP, Boyd RL, Miller, JFAP. 1998. The role of the thymus and recent thymus migrants in the maintenance of the adult

55.

56.

57.

58. 59. 60.

17

peripheral lymphocyte pool. J. Exp. Med. 187:1839–48 Bennett SRM, Carbone FR, Karamalis F, Flavell RA, Miller JFAP, Heath WR. 1998. Help for inducing cytotoxic–T–cell responses by cross–priming is mediated via CD40 signalling. Nature 393:478–80 Kurts C, Kosaka H, Carbone F, Miller JFAP, Heath WR. 1997. Class I-restricted cross-presentation of exogenous self antigens leads to deletion of autoreactive CD8+ T cells. J. Exp. Med. 186:239–45 Kurts C, Carbone F, Barnden M, Blanas E, Allison J, Heath WR, Miller JFAP. 1997. CD4+ T cell help impairs CD8+ T cell deletion induced by cross presentation of self antigens and favors autoimmunity. J. Exp. Med. 186:2057–62 Burnet FM. 1971. Genes, Dreams and Realities. Oxford, UK: MTP Moynihan W. 1930. Surgery in the immediate future. Br. Med. J. ii:612–14 Popper KR. 1972. Conjectures and Refutations. The Growth of Scientific Knowledge, p. 28. London: Routledge & Keagan Paul. 4th ed.

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:19–49

THE MULTIFACETED REGULATION OF INTERLEUKIN-15 EXPRESSION AND THE ROLE OF THIS CYTOKINE IN NK CELL DIFFERENTIATION AND HOST RESPONSE TO INTRACELLULAR PATHOGENS1 T. A. Waldmann and Y. Tagaya Metabolism Branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892-1374; e-mail: [email protected] KEY WORDS:

lymphokine, interleukin-15, IL-15 receptors, mRNA translation, NK cell development, autoimmunity

ABSTRACT Interleukin-15 (IL-15) is a 14- to 15-kDa member of the 4 α-helix bundle family of cytokines. IL-15 expression is controlled at the levels of transcription, translation, and intracellular trafficking. In particular, IL-15 protein is posttranscriptionally regulated by multiple controlling elements that impede translation, including 12 upstream AUGs of the 50 UTR, 2 unusual signal peptides, and the C-terminus of the mature protein. IL-15 uses two distinct receptor and signaling pathways. In T and NK cells the IL-15 receptor includes IL-2/15Rβ and γc subunits, which are shared with IL-2, and an IL-15-specific receptor subunit, IL-15Rα. Mast cells respond to IL-15 with a receptor system that does not share elements with the IL-2 receptor but uses a novel 60- to 65-kDa IL-15RX subunit. In mast cells IL-15 signaling involves Jak2/STAT5 activation rather than the Jak1/Jak3 and STAT5/STAT3 system used in activated T cells. In addition to its other functional activities in immune and nonimmune cells, IL-15 plays a pivotal role in the 1 The US government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

19

P1: APR/MBG

P2: NBL/ary

January 21, 1999

20

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA development, survival, and function of NK cells. Abnormalities of IL-15 expression have been described in patients with rheumatoid arthritis or inflammatory bowel disease and in diseases associated with the retroviruses HIV and HTLV-I. New approaches directed toward IL-15, its receptor, or its signaling pathway may be of value in the therapy of these disorders.

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

INTRODUCTION Intercellular communications involved in immune responses are often mediated by cytokines that show a high degree of redundancy and pleiotropy. The redundancy is explained in part by the sharing of common receptor subunits among the members of the cytokine receptor superfamily (1–5). In the case of the interleukin-2 receptor (IL-2R) system, the common gamma (γc ) subunit is shared by IL-2, IL-4, IL-7, and IL-9 (4–5). Recently, two groups simultaneously reported the recognition of an additional cytokine in this family, now known as IL-15, based on the ability of culture supernatants from two cell lines, CV-1/EBNA and the human T cell lymphotropic virus type I (HTLV-I)associated HuT-102, to stimulate the proliferation of the cytokine-dependent murine T-cell line CTLL-2 (6–8). The active materials in the two supernatants shared many characteristics such as an apparent molecular mass of 14–15 kDa and membrane-proximal signaling components in T and natural killer (NK) cells that consist of the IL-2Rβ and γc subunits of the IL-2 receptor (6–14). An appropriate anticytokine antibody was used to show that the two groups identified the same interleukin, which is now termed IL-15 (15). IL-2 and IL-15 share many features. They are both members of the 4 α-helix bundle cytokine family, they use both IL-2Rβ and γc for their action in T cells, and they have similar functional activities in these cells. Nevertheless, dramatic differences exist between these two cytokines in terms of their cellular sites of expression and the levels of control of their synthesis and secretion. IL-2 is produced by activated T cells and is controlled predominantly at the levels of mRNA transcription and stabilization, whereas control of IL-15 expression is much more complex, with regulation at the levels of transcription, translation, and intracellular trafficking and translocation (8, 10–11, 14–20). Furthermore, there are differences in the receptor and signaling pathways used by IL-2 and IL-15 in diverse cells (21, 22). Each cytokine has its own private receptor in T and NK cells: IL-15Rα and IL-2Rα for IL-15 and IL-2, respectively (21). IL-15 uses a novel IL-15RX receptor system and signal transduction pathway in select nonlymphoid cells including mast cells (22). As predicted from their sharing of receptor subunits, IL-2 and IL-15 have a number of redundant functions such as induction of T-cell proliferation and the costimulation of immunoglobulin synthesis (6, 8, 23–25). However, IL-15 also

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IL-15 AND ITS RECEPTORS

21

plays a pivotal role in the differentiation of NK cells from their progenitors, in the maintenance of the survival of such cells, and in their activation (26–39). IL-15 also has unique functions on nonlymphoid cells, including actions on muscle, brain microglia, and mast cells (11, 22, 40, 41). Abnormalities of IL-15 expression have been reported in inflammatory and neoplastic diseases (42–45). In particular, abnormally high levels of IL-15 transcription and translation were observed in HTLV-I-associated diseases such as adult T-cell leukemia (ATL) and the neurological disorder tropical spastic paraparesis (TSP) (45). Furthermore, abnormalities of IL-15 expression may occur in patients with inflammatory autoimmune diseases, such as rheumatoid arthritis and inflammatory bowel disease, with IL-15 at the apex of a cascade of inflammatory factors that includes TNF-α, which induces the expression of other inflammatory cytokines and chemokines involved in the pathogenesis of these diseases (42–44). Therapeutic agents are being developed to target the receptor and signaling elements shared by IL-2 and IL-15 to provide effective treatment for such disorders (10, 46–48).

THE STRUCTURE AND GENOMIC ARCHITECTURE OF IL-15 IL-15 is a 14- to 15-kDa glycoprotein whose mature form consists of 114 amino acids (aa) (8). It has two cystine disulfide cross-linkages at positions Cys42Cys88 (homologous to IL-2) and Cys35-Cys85, and three asparagine residues (119, 127, and 160) that in two cases are sites for N-linked glycosylation (8). IL-15 is a member of the 4 α-helix bundle cytokine family, which includes such cytokines as IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, and IL-9 (1, 8). The predicted folding topology of IL-15 suggests three loops connecting the four helices in an up-up down-down configuration (8). IL-15 shares no sequence homology with IL-2 or with other members of the cytokine superfamily; however, structural homology among these members is clearly conserved. There is 97% sequence identity between human and simian IL-15 and 82% sequence identity between human and porcine IL-15 (49–50). The IL-15 gene was mapped on chromosome 4q31 (human) and to the central region of chromosome 8 (mouse) by fluorescence in situ hybridization (50). The IL-15 gene consists of nine exons (exons 1–8 and a newly discovered exon 4a) and eight introns spanning at least 35 kb. This exon-intron organization contrasts with the four exon–three intron architectural pattern observed in IL-2, IL-4, and IL-5 (51). One form of human IL-15 mRNA contains a 50 untranslated region (UTR) of at least 352 nucleotides (nt), a coding sequence of 486 nt, and a 30 UTR of at least 400 nt (8). There are two alternative leader peptides, one with 48 aa and one with 21 aa (8, 19, 20, 52). In contrast to most signal peptides that are encoded in

P1: APR/MBG

P2: NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

22

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

a single or at most two exons, the IL-15 leader sequences are encoded in more than two exons (Figure 1) (19, 52). The classical long (48 aa) signal peptide associated with all secreted IL-15 is present in a 1.6 kb mRNA. It is encoded by exons 3, 4, and 5 of the human IL-15 gene (Figure 1) (50). The short 21-aa signal peptide is encoded by a 1.2-kb cDNA that lacks the elements encoded by exon 1. This signal peptide is encoded by exon 5 and by an additional 119-nt sequence inserted between exons 4 and 5 (new exon 4a) (19, 20, 52). The two signal peptides share 11 identical amino acids encoded by exon 5. The introduction of the 119 nt of exon 4a disrupts the 48-aa signal sequence by inserting a premature termination codon and then provides an alternative initiation codon with a poor Kozak context (TTCATGG) (19, 20, 52). The lack of exon 1 and the presence of a 219-nt fragment that originates from intron 1 in the 50 UTR of the 21-aa IL-15 signal peptide transcript suggests that this isoform may be regulated transcriptionally by an intronic enhancer/promoter in intron 1. As noted below, IL-15 associated with a short 21-aa signal peptide is not secreted but rather is stored intracellularly, appearing in nuclear and cytoplasmic components.

THE MULTIFACETED REGULATORY CONTROL OF IL-15 EXPRESSION IL-2 and IL-15 exhibit major differences in terms of their sites of synthesis and their levels of control of synthesis and secretion. IL-2 is produced by activated T cells, and its expression is regulated predominantly at the levels of mRNA transcription and message stabilization (17, 18). In contrast, Northern blot analysis indicated widespread constitutive expression of IL-15 mRNA in a variety of tissues such as placenta, skeletal muscle, kidney, lung, heart, fibroblasts, epithelial cells, and monocytes (8, 15). IL-15 mRNA could not be demonstrated by Northern blot analysis in normal resting or phytohemagglutinin-activated T cells, although the more sensitive RNase protection assay (RPA) indicated the presence of IL-15 mRNA in normal T cells obtained ex vivo and in the T-cell lines examined (8, 45).

Regulation of IL-15 Transcription The regulation of IL-15 expression is multifaceted. Modest control occurs at the level of transcription, and a dominant control occurs posttranscriptionally at the levels of translation and intracellular trafficking. In terms of transcriptional control, freshly isolated monocytes expressed only low levels of IL-15 mRNA that was upregulated when the monocytes were activated with LPS/IFN-γ (8, 15). In addition, infection of monocytes with herpesvirus 6, herpesvirus 7, Bacillus Calmette-Gu´erin (BCG), Mycobacterium tuberculosis, Toxoplasma

P2: NBL/ary

January 21, 1999 9:56

QC: NBL/tkj

Annual Reviews

IL-15 AND ITS RECEPTORS

Figure 1 There are two isoforms of IL-15 that contain two leader peptides, one with 48 amino acids (aa) and one with 21 aa. The 48aa signal peptide is encoded by exons 3, 4, and 5 of the human IL-15 gene. The short 21-aa signal peptide is encoded by exon 5 and an additional 119-nt sequence inserted between exons 4 and 5 (new exon 4 a). The introduction of the 119 nt of exon 4 a disrupts the 48-aa signal sequence by inserting a premature termination codon and then provides an alternative initiation codon. LSP, long signaling peptide; SSP, short signaling peptide.

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

P1: APR/MBG T1: NBL

AR078-02

23

P1: APR/MBG

P2: NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

24

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

gondii, Salmonella choleraesuis, Mycobacterium leprae, Cryptococcus neoformans, or Candida albicans was associated with an upregulation of IL-15 mRNA expression (31, 32, 39, 53–57). Cloning of the human and murine 50 flanking region of the IL-15 gene has permitted the study of the mechanisms underlying the constitutive and induced expression of IL-15 mRNA (45, 58). A series of conserved motifs between mouse and human IL-15 50 regulatory regions has been identified; these motifs include GCF, NF-κB, IRF-E, myb, γ IRE, NF-IL-6, and αINF-2. The IRF response element IRF-E and NF-κB sites are involved in the induced up-regulation of IL-15 mRNA expression (45, 58). As noted below, the development of NK cells requires IL-15 induction. Mice lacking the expression of the transcription factor IRF-1 (IRF-1−/− mice) also exhibit a severe NK-cell deficiency (35, 36). The IRF-1 deficiency does not affect the NK-cell progenitors themselves but rather affects the function of radio-resistant cells constituting the microenvironment required for NK-cell development. IRF-1−/− bone marrow cells can generate functional NK cells when cultured with IL-15 (35). As noted above, within the 50 upstream region of the mouse IL-15 gene, a 9-bp sequence, TTCACTTTC, spanning from −278 to −270 relative to the transcription initiation site, is in perfect concordance with the consensus IRF response element (IRF-E). This sequence motif binds IRF-1/2 proteins specifically. The importance of IRF-E for the activation of the IL-15 promoter was determined through the use of a series of reporter assays using IL-15 promoter deletion mutants (35, 45). These results support the view that IRF-1 in bone marrow stromal cells is pivotally involved in the up-regulation of IL-15 gene expression. The IL-15 generated acts on NK-cell precursors, stimulating their development into mature NK cells. In parallel studies, Ohteki and coworkers (36) demonstrated that IRF-1-induced IL-15 expression is important for the development of NK, NK-T cells, and intestinal intraepithelial T cells. Another transcription factor that appears to play an important role in IL-15 transcription is NF-κB. The HTLV-I-encoded tax protein transactivates IL-15 gene transcription through this site (45). IL-15 mRNA expression is increased in HTLV-I-infected T cells and T-cell lines. Using reporter constructs bearing the 50 regulatory region of the IL-15 gene, we found a positive correlation between HTLV-I tax protein expression and IL-15 promoter activity (45). Additionally, using a Jurkat T-cell transfectant that expressed tax under an inducible promoter, it was shown that the expression of IL-15 mRNA increased when tax was expressed. Mutations in the NF-κB motif or deletion of this sequence in the IL15 50 regulatory region eliminated the promoter activity in tax-transfected cells. These data represent evidence for transactivation of the IL-15 gene by the HTLVI tax protein through an NF-κB motif in HTLV-I-transformed T cells (45).

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

IL-15 AND ITS RECEPTORS

25

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Role of Translation in the Regulation of IL-15 Expression IL-15 is predominantly regulated posttranscriptionally at the level of translation and translocation. In particular, although IL-15 mRNA is widely expressed constitutively, it has been difficult to demonstrate IL-15 in supernatants of many cells that express such mRNA (8, 15). Although monocytes activated with LPS/IFN-γ expressed high levels of IL-15 mRNA, the culture supernatants and cell lysates from these cells contained little or no IL-15 protein as assessed by either an IL-15-specific ELISA or a CTLL-2 proliferation assay (15). This discordance between IL-15 mRNA expression and IL-15 protein production led us to examine normal IL-15 mRNA for posttranscriptional controls, particularly for features that could inhibit IL-15 production at the level of mRNA translation. We demonstrated that IL-15 expression is posttranscriptionally regulated by multiple elements including 12 upstream AUGs of the 50 UTR, a 48-aa signal peptide, and the C-terminus of the mature protein (15, 16). Our initial studies focused on the 50 UTR of IL-15 mRNA (15). In general, the 0 5 UTRs of effectively translated messages are short, simple, and unencumbered by AUGs upstream of the initiation AUG (59–61). In contrast to this pattern, the 50 UTR of IL-15 mRNA is long (at least 465 nt in mice and 352 nt in humans) and includes multiple upstream AUGs (5 in mice, 12 in humans) (8, 15). Kozak has emphasized that the presence of such AUGs in the 50 UTR may dramatically reduce the efficiency of mRNA translation into proteins (59–61). In general, the rare mRNAs with 50 AUG-burdened sequences include those encoding many protooncogenes, transcription factors, growth factors, receptor proteins, and signal transduction components. Among the 4 α-helix bundle cytokines, no upstream AUGs are present in the 50 UTRs of IL-2, IL-3, IL-4, IL-5, IL-10, IL-13, or IFN-γ , but they are seen in IL-7, IL-11, IL-13, IL-2Rα, IL-2Rβ, IL-5Rα, and the IL-9Rα receptor. IL-2Rα expression is controlled at both the transcriptional and translational levels (62). Inhibition of translation by upstream AUGs has been confirmed experimentally by deletion of upstream AUGs or by site-directed mutagenesis of the AUG triplets in TGF-β3, FGF-5, and IL-7 mRNAs (63–65). Such upstream AUG codons may represent a ploy by the cell to yield poorly translated mRNAs that encode critical regulatory proteins whose efficient expression might be dangerous to the cell or the organism. To define the effect of the upstream AUGs of the IL-15 50 UTR on normal IL-15 mRNA translation, three IL-15 constructs were transfected into COS cells, one representing the full wild-type IL-15 mRNA with an early hairpin and 12 AUGs, the second retaining the 10 distal upstream AUGs, and the third lacking upstream AUGs (15). This latter construct produced 4- to 5-fold more IL-15 than did cells transfected with the construct retaining 10 AUGs and 12to 15-fold more IL-15 than cells with the full wild-type construct. A number

P1: APR/MBG

P2: NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

26

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

of mechanisms have been reported to remove the 50 -UTR-AUG-mediated upstream impediments to translation. These mechanisms include ribosome shunting and internal initiation of translation, a process that bypasses the 50 CAPdependent scanning mechanism and translational inhibition of upstream AUGs through the use of an internal ribosome entry sequence (IRES) (66–69). Although COS cells transfected with the expression construct lacking the 50 UTR produced more IL-15 than did cells transfected with the wild-type construct (16), the quantity of IL-15 protein was very low (360 pg per 200,000 cells), 3 logs less than the quantity of IL-2 obtained (350,000 pg) with a comparable IL-2 construct. There were virtually equal levels of transcript for the two cytokines throughout the time-course despite the extreme disparity in protein production observed, suggesting differences in translational efficiency. Additionally, although IL-15 transcripts were readily translated in a wheat-germ in vitro translation system, they were poorly translated in a rabbit reticulocyte in vitro translation system. These data suggested that the mammalian translation systems (i.e., COS cells and rabbit reticulocytes) provided evidence for inhibitory and regulatory factors in addition to those in the 50 UTR that interfered with efficient synthesis of IL-15. We next examined the IL-15 mRNA for specific elements that might impede IL-15 expression and focused on the unusually long 48-aa isoform of the signal peptide (16). Because of the unusual length of this peptide, we considered the possibility that it might function as a negative regulator of IL-15 generation. To test this hypothesis we prepared expression constructs that exchanged the signal peptide coding sequences of IL-2 and IL-15 so that they were linked to the alternative mature protein coding sequence. The resulting chimeric cDNAs were transiently transfected into COS cells and the quantity of IL-15 determined. The total quantity of IL-15 generated (the sum of IL-15 retained within the cells and that secreted) increased 17- to 20-fold when the IL-15 signal peptide was replaced by that of IL-2 (16). In parallel studies, the quantity of IL-2 secreted was reduced 40- to 50-fold when COS cells were transfected with the reciprocal construct that had the IL-2 signal peptide replaced by that of IL-15. We demonstrated that the IL-15 and IL-2 protein expression differences observed could not be explained by differences in mRNA stability or by instability of the processed protein but rather were the result of a major impediment at the level of mRNA translation. In parallel studies, Onu and coworkers (20) demonstrated that wild-type IL-15 mRNA transcription was not associated with efficient secretion of IL-15 protein. Furthermore, after replacing the IL-15 signal peptide with that of CD33, translation and secretion increased, supporting the view that IL-15 expression is controlled mainly posttranscriptionally at the levels of translation and secretion. When the IL-15 isoform with the alternative short 21-aa signal

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

IL-15 AND ITS RECEPTORS

27

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

peptide was examined, there was no secretion of IL-15 by the transfected cells. However, there was a four- to fivefold increase in translation of the isoform containing exon 4a compared to that with the long signal peptide, at least when assessed in a rabbit reticulocyte in vitro translation system (52, 70). The data support the hypothesis that the IL-15 signal peptide or its coding sequence are important factors in the negative regulation of IL-15 protein expression. As stated by Bamford et al (16): The mechanisms underlying the [signal peptide]-mediated regulation of IL-15 translation have not been defined. However, with preliminary in vitro translation studies, we observed that the addition of canine microsomal membranes did not result in IL-15 chain completion and translocation into microsomes in contrast to the situation with the prototypical secretory protein, pre/prolactin, which was fully translocated and processed. . .. Therefore, a number of events or factors may be required for efficient IL-15 mRNA translation/translocation. It is possible that a translational activator(s) for chain elongation and translocation may be needed. Alternatively, a translational repressor or a stable secondary structure in the mRNA may prevent efficient IL-15 mRNA elongation and translocation. Furthermore, inefficient initiation of translation may contribute partially to the low levels of IL-15 protein generated in transfected COS cells. This stems from the observation that the start codon for the IL-15 coding sequence has a weak Kozak context (GTA ATGA). . .. In fact, modifying the start codon to a higher context (ACCATGG or GCCGCCATGA) increased IL-15 protein production fourto fivefold in transfected COS cells.

In additional studies we discovered that a third negative element may exist in the C-terminus of the IL-15 mature protein coding sequence or protein (16). Specifically for the purposes of antibody detection, we added the artificial epitope tag FLAG to the 30 end of the IL-15 protein. We noted that the presence of FLAG increased total IL-15 protein production 5- to 10-fold, suggesting that this modification disrupted an inhibitory cis-element in the coding sequence of the IL-15 mature protein C-terminus. When the three IL-15 mRNA modifications (elimination of the 50 UTR, switch of IL-15 signal peptide coding sequence with that of IL-2, and FLAG modification of the 30 coding sequence) were combined in a single construct and introduced into COS cells, at least 250-fold more IL-15 was produced than was observed with the wild-type IL-15 construct with an intact 50 UTR (16). These findings suggest that IL-15 mRNA, unlike IL-2 mRNA, may exist in translationally inactive pools. Control of translation has been observed in a variety of proteins, and this regulation can occur at all levels of translation (e.g., initiation, elongation, and termination). Most mRNA-specific translational regulation has involved cis-acting RNA sequence elements that mediate regulation. Such regulatory sequences in the 50 or 30 UTR of the mRNAs and mature coding sequence have been observed in transcripts for ferritin, erythroid 5-aminolevulinate synthase, thymidylate synthase, and murine p53 (71–73). Furthermore, regulation at the level of translation has been demonstrated for the 70-kDa heat shock protein

P1: APR/MBG

P2: NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

28

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

mRNA in chicken reticulocytes IL-1β, TNF-α, TGF-β3, TGF-β1, and GMCSF (74–75). In addition, researchers observed that one of the multiple levels of insulin biosynthesis regulation includes a glucose-dependent signal recognition particle-mediated translational arrest (76). The studies of IL-15 indicate that the translational control of IL-15, like that of insulin, occurs at multiple distinct levels (76). The removal of these negative control mechanisms in an integrated fashion may lead to a major increase in IL-15 synthesis. The variety of negative regulatory features controlling IL-15 expression may be required because of the potency of IL-15 as an inflammatory cytokine. If indiscriminantly expressed, IL-15, with its capacity to induce the expression of TNF-α, IL-1, IFN-γ , and other cytokines and chemokines involved in the inflammatory response, could be associated with serious disorders such as autoimmune diseases (43). In terms of a more positive role for IL-15, we propose that by maintaining a pool of translationally inactive IL-15 mRNA, diverse cells may respond rapidly to an intracellular infection or other stimuli by transforming IL-15 mRNA into a form of mRNA that can be translated effectively. The IL-15 protein produced and secreted could convert T and NK cells into activated killer cells that might provide an effective host response to an invading infectious agent.

Intracellular Trafficking of IL-15 As noted above, two isoforms of human IL-15 exist. One isoform has a short 21-aa putative signal peptide, whereas the other isoform has an unusually long 48-aa signal peptide (8, 19, 20, 52, 70). In addition to their role in the regulation of IL-15 translation, these signal peptides influence intracellular trafficking of IL-15. The 21-aa IL-15 isoform is translated, but IL-15 is not secreted (19, 20, 52). Experiments using different combinations of signal peptides and mature proteins (IL-2, IL-15, and green fluorescent protein) showed that the short signal peptide regulates the fate of the mature protein by controlling the intracellular trafficking to non–endoplasmic reticulum sites such as the cytoplasm and the nucleus (52). The production of an intracellular lymphokine is not typical of other soluble interleukin systems, suggesting a biological function for IL-15 as an intracellular molecule. The IL-15 associated with the long 48-aa signal peptide presents a more complex pattern and exists as multiple distinct molecular species within transfected COS cells. This multiplicity of protein species is partly the result of glycosylation because human IL-15 has two functional glycosylation sites. However, evidence from lysates using tunicamycin-treated IL-15-transfected cells indicates that the 48-aa signal peptide of IL-15 can be cleaved at two separate sites, yielding both partial and complete processing of the signal peptide. No evidence

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IL-15 AND ITS RECEPTORS

29

of stepwise processing could be demonstrated, suggesting that the partially and fully processed forms may traffic to different cellular compartments. The form of IL-15 that retains the full signal peptide but without evidence of glycosylation was demonstrable in the cytoplasm and nucleus. The other forms entered the endoplasmic reticulum where they were glycosylated. The passage of IL-15 through the ER was much slower than that of IL-2 (16). Nevertheless, IL-15 was secreted after trafficking through the Golgi, yielding a cytokine with a fully processed signal peptide. Evidence for trafficking through the Golgi includes the inhibition of secretion by culture with brefeldin A and the endoglycosidase H–resistant nature of the secreted IL-15. The two isoforms of IL-15 generated by usage of alternative signal peptides have different intracellular trafficking patterns. Sorting of the same protein to different cellular compartments by modifying the regulatory sequence also has been observed in other systems. Examples include proteins such as stem cell factor and Int-2, a fibroblast growth-factor-related oncoprotein (77, 78). In the case of Int-2, two different signal peptides are generated by the usage of different start codons in-frame, resulting in the alternative transport of the protein either to the secretory pathway or to the nucleus (78). The IL-15 case seems very similar to that of Int-2.

IL-15 RECEPTOR AND SIGNAL TRANSDUCTION PATHWAYS IL-15 Type-1 Receptors in T and NK Cells IL-15 uses two receptor and signaling pathways (7–10, 22). Cytokines such as IL-15 manifest considerable pleiotropy and redundancy controlling a wide range of functions in various cell types. The redundancy is explained in part by the sharing of common receptor subunits among members of the cytokine receptor family. Each cytokine has its own private receptor, but it usually shares one or more public receptors with other cytokines. Receptor elements are shared within the IL-2/15R system. In particular, the high-affinity IL-15R system in T and NK cells (type-1 IL-15 receptor) is made up of three distinct membrane components. Two of these components, IL-2/15Rβ and IL-2Rγ or γc are shared with the IL-2R system (7–9). In addition, the two cytokines have their own private α chains: IL-2Rα for IL-2 and IL-15Rα for IL-15 (21). The γc chain is also shared by IL-4, IL-7, and IL-9 (4, 5). The human IL-2/15Rβ mRNA predicts a primary translation product of 551 aa (79, 80). The receptor contains a 26-aa signal peptide, and the mature human IL-2/15Rβ is composed of 525 aa with an extracellular segment of

P1: APR/MBG

P2: NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

30

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

214 aa, a hydrophobic transmembrane stretch of 25 aa, and a 286-aa cytoplasmic domain. IL-2/15Rβ is expressed constitutively by NK cells and to a lesser extent by monocytes and CD8 cells. The human γc cDNA contains an open reading frame for a 369-aa residue polypeptide (81). This protein contains a 22-aa signal peptide, a 233-aa extracellular domain, a 28-aa hydrophobic transmembrane domain, and an 86-aa C-terminal cytoplasmic domain. IL-2/15Rβ and γc are members of the hematopoietin or cytokine superfamily of receptors that contain four conserved cysteines and the canonical WSXWS (trp-ser-X-trp-ser) motif. A novel IL-15-specific binding protein termed IL-15Rα was identified and its cDNA cloned by Giri and coworkers (21). IL-15Rα is a type-1 membrane protein with a predicted signal peptide of 32 aa, a 173-aa extracellular domain, a single membrane-spanning region of 21 aa, and a 37-aa cytoplasmic domain. In contrast to IL-2/15Rβ and γc , IL-15Rα is not a member of the cytokine receptor superfamily. However, a comparison of IL-2Rα and IL-15Rα revealed the shared presence of a conserved motif known as a GP-1 motif, or a SUSHI domain (21). Another factor linking IL-2Rα and IL-15Rα is the demonstration that IL-2Rα and IL-15Rα genes have a similar intron-exon organization. Moreover, they are closely linked in both human (10q14-15) and murine genomes (chromosome 2 linked to Vim-2 and Spna-2) (81). IL-15Rα binds IL-15 with a Ka of 1011/M, a 1000-fold higher affinity than that of IL-2Rα for IL-2. IL2/15Rβ in association with γc is able to bind IL-15 at a lower affinity (a Ka of ∼109/M) and in select cells can transduce an IL-15 signal in the absence of IL-15Rα. IL-15Rα has a wide cellular distribution. Its expression is observed in T cells, B cells, macrophages, and in thymic and bone marrow stromal cell lines (82). In addition, IL-15Rα mRNA is widespread in such tissues as liver, heart, spleen, lung, skeletal muscle, and activated vascular endothelial cells (21). IL-15Rα mRNA expression is increased in T cells after the addition of IL-2, an antiCD3 antibody, or phorbol-myristate acetate (PMA) (21). Furthermore, IL15Rα expression is augmented in macrophage cell lines after treatment with interferon-γ . Thus the widespread distribution of the IL-15Rα, IL-2/15Rβ, and γc elements of the IL-15R system is one of the mechanisms underlying the pleiotropy of IL-15.

IL-15 Signal Transduction Pathway in Activated T Cells When analyzed in activated T cells, Jak3 and Jak1 were shown to be coupled functionally to the receptor systems involving γc , including the receptors for IL-15, IL-2, IL-4, IL-7, and IL-9 (83–85). Furthermore, the addition of IL-15 or IL-2 to such receptor-expressing T cells led to the tyrosine phosphorylation and nuclear translocation of STAT3 and STAT5 (84–85). The IL-2 and IL-15

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

IL-15 AND ITS RECEPTORS

31

signaling pathways in T cells also involve the phosphorylation of the Src-related cytoplasmic tyrosine kinases p56lck and p72syk, the induction of the expression of the bcl-2 antiapoptotic protein, and the stimulation of the Ras/Raf/MAP kinase pathway leading to fos/jun activation (86).

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IL-15 Uses a Distinct Type-2 Receptor/Signal Transduction Pathway in Mast Cells IL-15 stimulates the proliferation of murine mast cell lines and normal bone marrow mast cells, whereas these cells do not respond to IL-2 (11, 22). This disparity in response to the two cytokines suggested the existence of a novel IL15-specific receptor system in mast cells not shared by IL-2. The mast cell lines PT-18 and Mc/9 did not express mRNA-encoding IL-2Rα or IL-2/15Rβ, which explained the failure of IL-2 signaling in such cells. Furthermore, transfection of these cells with a cytoplasmic-truncated transdominant mutant form of γc demonstrated that IL-15-signaling in mast cells does not require this receptor element. In addition, evidence indicated that IL-15Rα is not a critical element of the mast cell IL-15R and signaling pathway. We used 125I-IL-15 in a cross-linking study with the mast cell line PT-18, using disuccinimidyl suberate to identify a possible mast cell–specific IL-15R. The IL-15/IL-15R complex in mast cells migrated approximately 75–80 kDa, implying a cytokine receptor size of 60–65 kDa. These results suggested that mast cells express a novel 60- to 65-kDa IL-15R molecule (type-2 IL-15 receptor), which was provisionally designated IL-15RX. The lack of involvement in mast cells of the IL-2/15Rβ and γc chains used by IL-15 in T and NK cells and the possibility that a novel receptor (IL-15RX) is involved in IL-15 signaling in these cells prompted us to examine the membraneproximal events of IL-15 signal transduction in mast cells to determine whether they are different from those in T cells. IL-15 addition to the mast cell line PT-18 caused the phosphorylation of Jak2 kinase rather than Jak1 or Jak3 as observed with the type-1 IL-15R in T cells (22). IL-15 also stimulated Jak2 phosphorylation in bone marrow mast cells. In further contrast to T cells, the addition of IL-15 to mast cells led to a tyrosine phosphorylation and nuclear translocation process limited to STAT5 rather than the STAT3/STAT5 activation observed with the type-1 receptor in T cells (22). These results indicate an IL15 function distinct from that of IL-2 given that mast cells appear to use a novel IL-15 receptor and signaling pathway to stimulate the proliferation of these cells. However, the addition of IL-15 to bone marrow hematopoietic precursor cells does not result in the propagation of mast cells but rather induces NK cells (see below). Furthermore, these observations identify a second mechanism underlying the pleiotropy manifested by IL-15 in addition to the widespread tissue distribution of IL-15Rα.

P1: APR/MBG

P2: NBL/ary

January 21, 1999

32

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

BIOLOGICAL ACTIONS OF IL-15

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Role of IL-15 in T and B Cell Function As might be anticipated by their sharing of IL-2/15Rβ and γc receptor subunits in T, B, and NK cells, IL-15 and IL-2 have some common biological activities (6, 8, 12–14, 23–25). However, IL-15 also has unique functions, reflecting the much broader tissue distribution of its private receptor, IL-15Rα (21, 81). Furthermore, in select cells IL-15 uses a second receptor (IL-15RX) and signaling pathway not shared with IL-2 (22). Finally, the major differences between IL-2 and IL-15 in terms of their sites of synthesis and the regulation of their expression may also lead to different actions. For example, the expression of IL-15 mRNA in thymic and bone marrow stromal and epithelial cells suggests that IL-15 may play a role in the development of NK and T lymphocytes. IL-15 stimulates the proliferation of CTLL lines; antigen-dependent T-cell clones; activated CD4−8−, CD4+8+, CD4+, and CD8+ cells; and dendritic epidermal T cells (6, 8, 87–91). Dendritic cells, a class of potent antigen-presenting cells, are producers of IL-15 and induce activation and chemotactic activity for Th1, the subset of helper T cells (92, 93). As IL-2 is not produced by dendritic cells, this observation suggests that IL-15 is involved in normal immune responses that are distinguishable from those involving IL-2. IL-15, like IL-2 and IL-7, appears to be involved in the development of thymic-independent gut intraepithelial lymphocytes that do not develop normally in IL-2/15Rβ- or IRF-1 deficient mice (35). IL-15 stimulates and augments the proliferation of T cells from HIV-positive individuals and of T cells derived from primary human tumor cell cultures (94, 95). Furthermore, IL-15 synergizes with IL-12 to induce proliferation of murine Th1 clones (96, 97). IL-15 addition also promotes the induction of cytotoxic lymphocyte effector cells and lymphokine-activated killer cells. The addition of IL-15 to T cells leads to the induction of their expression of IL-2Rα (CD25), IL-2Rβ (CD122), and Fas (CD95) whereas it downregulates CD27 expression (98–101). IL-15 is a chemoattractant for T cells but not B cells, monocytes, or neutrophils (25, 42). Finally, IL-15 inhibits cytokine-deprivation-induced apoptosis and apoptosis induced by anti-Fas, dexamethasone anti-CD3, or anti-IgM in activated T and B cells (102). In light of its antiapoptotic effect in CD4 cells, Kanegane & Tosato (103) have suggested that IL-15 acts as a memory-facilitating factor for helper T cells. Furthermore, IL-15 stimulates memory-phenotype CD8+ cells in vivo (87). Although IL-15 does not have an effect on resting B cells, it induces proliferation and immunoglobulin synthesis by human tonsilar B cells costimulated by PMA or by an immobilized antibody to immunoglobulin M (24). Furthermore, when used in concert with CD40 ligand (CD40L), it is an inducer of polyclonal IgM, IgG1, and IgA but not IgG4 or IgE (24). The effect of IL-15 on Ig secretion

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

IL-15 AND ITS RECEPTORS

33

can be modulated by IL-10 (increased secretion) or by IL-4 (decreased secretion) (104). The effect of IL-4 may be caused by its diversion of the shared γc receptor, a conclusion supported by fluorescence resonance energy transfer (FRET) analysis of receptor subunit association after cytokine addition (105). The action of IL-15 on T, B, and NK cells could be blocked by select antibodies (e.g., Mikβ1) to the IL-2/15Rβ chain (8).

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IL-15 Plays a Pivotal Role in the Development, Survival, and Activation of NK Cells NK cells are bone marrow–derived CD2+, CD16+, and CD56+ human large granular lymphocytes (LGLs) that lack CD3 but express the ζ chain of the T-cell receptor. Observations suggest that IL-15 is important in the differentiation, survival, and function of NK cells, indeed that IL-15 may be essential for their development (26–39). For example, mice made deficient in IL-2/15Rβ by homologous recombination or through the use of an antibody to this receptor subunit are markedly deficient in NK cells (27). IL-2/15Rβ is required for the actions of IL-2 and IL-15 but is not used by other growth factors. In contrast, mice deficient in IL-2 or IL-2Rα, the private receptor used by IL-2, have a normal number of NK cells, suggesting that IL-15 may be required for NK cell development (27). Alternatively, the expression of either of these two cytokines might be sufficient for the maturation of NK cell progenitors. NK cells are also virtually absent in mice deficient in the signaling molecules required for IL-15 expression (e.g., there is no NK-cell development in IRF-1−/− mice) or in receptors or signaling molecules required for IL-15 action (35, 36). In particular, NK cells are markedly deficient in both human and mouse cells that do not express the normal γc chain used by IL-2, IL-4, IL-7, IL-9, and IL-15, whereas IL-2-, IL-4-, and IL-7- deficient mice express NK cells (106–108). Similarly, mice deficient in Jak3, which is required for IL-15 action, are also deficient in NK cells (109–110). The role of cytokines in NK-cell development has been studied directly using in vitro stromal-independent cultures of hematopoietic precursors. IL-2 addition in the presence of other cytokines, such as IL-7 or stem cell factor, leads to NK-cell differentiation (29, 38). However, IL-15 is even more effective in inducing bone marrow progenitor differentiation into NK cells. Furthermore, it is the one factor capable of inducing CD34+ CD7− bone marrow or cord blood cells to undergo such differentiation (29). In a similar way, the addition of IL-15 and to a lesser extent IL-2 to immature postnatal thymocytes or to fetal thymic organ cultures led to the development of NK cells that express the CD3−, CD56+, CD94 HLA Class I–specific inhibitory receptor phenotype (30). Nevertheless, in this study IL-15 alone was not sufficient for the induction of CD16 or the other HLA Class I–specific inhibitory receptors, implying that

P1: APR/MBG

P2: NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

34

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

other factors are required for the expression of these elements (30). A similar IL-15-mediated propagation of NK cells from their progenitors was observed in NK-cell-deficient IRF−/− mice (35). This latter observation supports the view that the defect in such mice is not in the NK-cell progenitors themselves but in the induction of the required IL-15 expression within radio-resistant bone marrow stromal cells. In various culture systems, stem cell factor, IL-7, and flt-3 ligand enhanced the IL-15-mediated expansion of NK cells (29). The data suggest that IL-15 is a dominant factor in the differentiation of NK cells from uncommitted progenitors. In terms of NK function, resting NK cells express IL-15Rα, IL-2/15Rβ, and γc , which are required for a response to picamolar concentrations of IL-15 (33). IL-15 facilitates the survival of NK cells ex vivo (33). In particular, the addition of IL-15 to human blood leukocytes led to the survival of NK cells in the absence of serum for the eight-day observation period (33). This IL-15-supported survival was associated with the prevention of programmed cell death, an inhibitory action that required bcl-2 expression (33). IL-2 was modestly effective in the intradiction of cell death, whereas other cytokines that use γc , such as IL-4, IL-7, and IL-9, were not effective nor were the monocytederived factors TNF-α, IL-1β, IL-10, or IL-12. IL-15 was effective as an NK-cell chemoattractant and activator. Moreover, IL-15 synergized with IL-12 to stimulate the production by NK cells of IFN-γ , TNF-α, and GM-CSF (32). NK cytotoxicity mediated by IL-15 was induced by a variety of infectious agents such as herpesvirus 6 and herpesvirus 7 (31, 39). In the cases examined, the upregulation of NK activity by these infectious agents was markedly reduced by the addition of monoclonal antibodies to IL-15 but not by antibodies to other cytokines such as IFN-α, IFN-γ , TNF-α, TGF-β, or IL-2, suggesting that IL-15 secreted in response to the infectious agents was responsible for the observed NK-cell activation. Patients lacking NK cells are subject to multiple infections with herpesviruses (111). Moreover, NK activation induced by agents such as herpesvirus 6 and herpesvirus 7 was blocked by the addition of an antibody to IL-2/15Rβ that blocks the action of IL-15 (31, 39). This evidence supports the view that IL-15 plays a pivotal role in the development, survival, and activation of NK cells.

Role of IL-15 in the Host Defense Against Intracellular Pathogens IL-15 as assessed by a specific ELISA assay or as an activity interpreted to be IL-15 based on inhibition by an anti-IL-15 antibody has been demonstrated in the supernatants of monocyte and macrophage preparations treated with various intracellular infectious agents (31, 32, 39, 53–57). IL-15 enhanced superoxide function and antifungal activity of human monocytes (56). Similarly,

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IL-15 AND ITS RECEPTORS

35

IL-15 contributed to the anticryptococcal activity of macrophages (54). IL-15 induces IL-8 and monocyte chemotactic protein-1 production in human monocytes (112). Furthermore, although neither IL-12 nor IL-15 addition resulted in the induction of IFN-γ production by either NK cells or γ δ T cells, they acted synergistically on these cells, inducing the production of IFN-γ , TNF-α, and GM-CSF (53). Finally, IL-15 mRNA and protein were most highly expressed in patients with immunologically resistant tuberculoid leprosy but not in unresponsive and susceptible lepromatous patients (57). These data taken together with those demonstrating the constitutive expression of IL-15 mRNA in various tissues support our view that IL-15 may act as an “alarmin” wherein by maintaining a pool of translationally inactive IL-15 mRNA diverse cells, such as macrophages and dendritic cells, may respond rapidly to an intracellular pathogen by converting impeded IL-15 mRNA into an effectively translatable form (92, 93). Thus the IL-15 response to infectious agents, such as viruses and other intracellular organisms, may represent a critical element in the host defense against these pathogens.

IL-15 Action in Nonimmunological Cells The broad tissue expression of mRNA encoding IL-15 and the IL-15Rα subunit suggests that IL-15 has activities beyond the immune system. An example of a nonimmunological action was observed in skeletal muscles that express IL-15 and IL-15Rα mRNA (40, 41). The addition of IL-15 to a cultured myoblast line did not induce proliferation but affected parameters associated with skeletal muscle fiber hypertrophy, especially when insulin-like growth factor levels were low, suggesting that IL-15 may be an anabolic agent that increases skeletal muscle mass (41). In another system unlike IL-2, IL-15 bound to vascular endothelial cells with a high affinity through the IL-15Rα subunit. Moreover, IL-15 promoted angiogenesis in a murine system with the induction of neovascularization of Matregel plugs after IL-15 addition (113). Mouse brain microglia and human fetal astrocytes and microglia express IL-15 mRNA and its trimeric receptor complex functionally coupled to Jak kinase activity (114, 115). The levels of IL-15 mRNA increased upon addition of IL-1β, IFN-γ , or TNF-α. IL-15 affected the functional properties of microglia such as their production of nitric oxide and their growth in culture. Thus IL-15 may participate in certain central nervous system and neuroendocrine functions previously ascribed to IL-2, which is expressed in only very minute concentrations in the central nervous system. As discussed above, mast cells express a type-2 IL-15 receptor and signaling pathway using Jak2 that is distinct from the trimeric receptors and signaling pathways (Jak1, Jak3) used by both IL-2 and IL-15 in activated T cells (22). IL-15, but not IL-2, stimulates mast cell proliferation in vitro and ex vivo.

P1: APR/MBG

P2: NBL/ary

January 21, 1999

36

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Moreover, the intermediate affinity type-2 IL-15 receptor may have a functional role in mast cell biology in that IL-15 is a mast cell growth and activation factor. Intestinal epithelial cells both express and respond to IL-15. IL-15 signals in T84 colonic epithelial cells in the absence of the IL-2/15Rβ chain, suggesting that IL-15 uses a receptor in these cells other than the classical type-1 trimeric IL-15 receptor (116). Finally, as noted above, in contrast to most cytokines, intracellular cytoplasmic and nuclear forms of IL-15 have been demonstrated. Such intracellular IL-15 may play a novel and as yet undefined role within the cells that produce it.

ABNORMALITIES OF IL-15 EXPRESSION IN DISEASE Abnormalities of IL-15 in Inflammatory Autoimmune Diseases Feldmann and coworkers have proposed that TNF-α is at the apex of a cytokine cascade that includes IL-1β, IL-6, GM-CSF, and a series of inflammatory chemokines, including Mip1α, Mip1β, and IL-8, that are intimately involved in the development and progression of rheumatoid arthritis (RA) (117). McInnes and coworkers have reported abnormalities of IL-15 in this disease and have suggested that IL-15 may precede TNF-α in the cytokine cascade (42, 43). In particular, IL-15-activated T cells can induce TNF synthesis by macrophages in RA via a cell-contact-dependent mechanism (43). They reported the presence of high concentrations of IL-15 in RA synovial fluid and showed that IL-15 is expressed by synovial-membrane-lining cells. Nevertheless, the presence of rheumatoid factor in the fluids may yield specious high estimates for IL-15 assessed by an ELISA. RA synovial fluids contain chemotactic and Tcell stimulatory activities attributable in part to IL-15. Oppenheimer-Marks and coworkers (118) demonstrated that IL-15 is produced by endothelial cells in rheumatoid tissues and that this cytokine markedly increases transendothelial migration of both CD4 and CD8 cells. Furthermore, they showed that IL-15 leads to T-cell accumulation in RA synovial tissues engrafted into severe combined immune deficiency (SCID) mice in vivo. In a parallel murine model the intra-articular injection of IL-15 induced a local tissue inflammatory infiltrate consisting predominantly of T lymphocytes (42). These data suggest that IL-15 can recruit and activate T cells into the synovial membrane, possibly contributing to the pathogenesis of RA. In support of this view, the injection of an IL-15 antagonist, the soluble form of IL-15Rα, into DBA/1 mice suppressed their development of collagen-induced arthritis (119). In summary, these reports suggest a role for IL-15 in the development of inflammatory RA and imply that antagonists to IL-15 action may have therapeutic potential in this disease.

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IL-15 AND ITS RECEPTORS

37

Abnormalities of IL-15 have also been reported in other inflammatory disorders. For example, increased numbers of IL-15-expressing cells are present in the circulation of patients with active ulcerative colitis, or Crohn’s disease (119). Furthermore, elevated levels of IL-15 correlated with disease activity and may reflect the degree of inflammation in the liver in type-C chronic liver disease (120). In addition, IL-15 triggers the growth of T cells in sarcoidosis through the IL-2/15Rβ/γc complex and may deliver proliferative signals leading to the development of the T-cell alveolitis observed in this disorder (121). Furthermore, IL-15 mRNA expression is upregulated in blood and cerebrospinal fluid mononuclear cells in multiple sclerosis (122). Finally, the observation that IL-15 stimulates mast cell proliferation suggests a potential role for this cytokine in mastocytosis (22).

IL-15 Action in Retroviral Diseases and Neoplasia HTLV-I infects CD4 T cells and is associated with a series of diseases including ATL, HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP), arthritis, uveitis, inflammatory lung disease, and infectious dermatitis (123– 125). HTLV-I-infected T-cells of patients with HAM/TSP expressed the HTLVI-encoded transactivator p40tax and exhibited abnormal spontaneous proliferation when studied ex vivo (124). This T-cell proliferation could be inhibited partially by the addition of an antibody to IL-2 or to the IL-2-specific receptor, IL-2Rα, suggesting a role for IL-2 in this process (125). Nevertheless, the addition of Mikβ1, an antibody to the IL-2/15Rβ chain, also reduced this proliferation (125). This antibody does not inhibit the action of IL-2 on the highaffinity IL-2R but does inhibit IL-15 function, an observation that suggests a role for IL-15 in the abnormal T-cell proliferation observed in HAM/TSP. IL-15 mRNA expression is increased in HTLV-I-infected T-cells (45). By using reporter constructs bearing the 50 regulatory region of the IL-15 gene, we observed a positive correlation between HTLV-I tax protein expression and IL-15 promoter activity in HTLV-I-infected T cells (45). We observed increased IL-15 mRNA expression not only in HAM/TSP but also in T-cell lines and ex vivo leukemic cells of patients with HTLV-I-associated ATL. Furthermore, IL-15 can replace the IL-2 signal in IL-15Rα-expressing IL-2-dependent ATL cell lines (126). Thus the production of IL-15 in this leukemia might be one factor in the constitutive activation of Jak3 observed with select lines and cells from patients with ATL (127, 128). The production of IL-15 by the HTLV-I-associated ATL cell line HuT-102 that permitted the identification of IL-15 was especially dramatic. This phenomenon is explained by our observation that IL-15 production by the ATL line HuT102 is associated with a human T-cell lymphotropic virus I R region/IL-15 fusion message (15). In particular, the predominant IL-15 mRNA expressed by HuT-102 cells is a chimeric RNA with a 118-nt segment of the R region

P1: APR/MBG

P2: NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

38

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

of the long terminal repeat (LTR) of HuT-102 joined to the 50 UTR of IL-15. Normally by alternative splicing this 118-nt element of R represents the most 50 region of several HTLV-I transcripts including those encoding tax, rex, and env. However, in HuT-102 this element derived from the R region is aberrantly spliced to the 50 UTR of IL-15. The high-level expression of IL-15 mRNA by HuT-102 appears to result from the transcription of a large quantity of a fusion message with the IL-15 allele under the regulatory control of the HTLVI LTR element. In addition, the introduction of the R segment eliminates over 200 nt of the IL-15 50 UTR, including all but two of the upstream AUGs that, as indicated by our other studies, appear to behave as impediments to translation (15). Thus the effective synthesis of IL-15 protein by the ATL cell line HuT102 appears to involve a marked increase in IL-15 mRNA transcription and translation secondary to the integration of HTLV-I provirus with a consequent production of a fusion message involving the HTLV-I R segment linked to the truncated 50 UTR of IL-15. IL-15 serum levels are elevated significantly in HIV-I-infected individuals (129). This cytokine enhances immune functions during HIV infection (130). Moreover, a positive correlation exists between IL-15 and serum immunoglobulin levels in this disorder, suggesting that this B-cell costimulatory cytokine may contribute to the pathogenesis of HIV-associated hypergammaglobulinemia (129). A series of tumor cell lines have been evaluated for the expression of IL-15 mRNA by RT-PCR analysis. Several of these cell lines, including lung, ovarian, melanoma, some leukemia, osteosarcoma, and especially rhabdomyosarcoma cell lines, expressed IL-15 mRNA (19, 131). In select situations Barzegar et al (131) showed that this mRNA represented the short signal peptide IL-15 mRNA isoform. Moreover, in most cases it was difficult to demonstrate IL-15 in the culture supernatants. If IL-15 plays a role in these neoplasias, it might be through its intracellular action (131).

OPPORTUNITIES FOR THERAPY DIRECTED TOWARD IL-15, ITS RECEPTOR, OR ITS SIGNALING TRANSDUCTION SYSTEM IL-2 is effective in the treatment of renal cell carcinoma and malignant melanoma, and in the therapy of patients with AIDS. In parallel, IL-15 was shown to help correct the impaired proliferative response of CD4+ lymphocytes from HIV-I-infected individuals without the mitogenic effect of IL-2 that might also induce HIV expression (130). Thus IL-15 could provide an alternative therapeutic option in the treatment of patients with select tumors or AIDS. The majority of therapeutic efforts involving the IL-15/IL-15R system are being directed toward inhibiting IL-15 action. The scientific basis for this approach

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IL-15 AND ITS RECEPTORS

39

was discussed in part above where it was suggested that IL-15 might contribute to the pathogenesis of RA, inflammatory bowel disease, and sarcoidosis. Furthermore, select malignant cells, for example, multiple myeloma and large granular lymphocytic leukemia cells, express IL-15 receptors (132). In addition, intragraft IL-15 transcripts were increased in patients rejecting renal allografts (133). A correlation existed between IL-15 transcripts within grafts being rejected as compared to nonrejected renal allografts, suggesting that IL15 may play a role in T- and NK-cell-mediated rejection (130). Moreover, IL-15 transcripts were present in the allografts in association with rejection of pancreatic islet allografts in wild-type mice and in IL-2 knockout mice, again suggesting that the IL-15/IL-15Rα system may be a valuable therapeutic target in organ transplantation protocols (134). Blocking the IL-15Rα with a receptor antagonist enhanced acceptance of islet cell allografts (135). Most IL-2/15 receptor-directed therapeutic approaches have targeted the IL2Rα chain (136–138). To exploit the difference in IL-2Rα expression between normal resting cells that do not express this receptor and IL-2Rα-expressing abnormal T cells in leukemia, select autoimmune disorders, and allograft rejection, clinical trials have been performed using unmodified murine anti-IL-2Rα, humanized antibodies, and antibodies armed with toxins and α- and β-emitting radionuclides (136–138). On the basis of two extensive randomized placebocontrolled trials, the humanized anti-IL-2Rα monoclonal antibody (Zenapax) received marketing approval by the FDA for the therapy of patients receiving renal allografts (139). Although IL-2Rα-directed therapy has met with considerable success, approaches directed toward this receptor subunit have limitations. In particular, antibodies to IL-2Rα do not inhibit the action of IL-15, a cytokine that does not use this subunit. Therefore, a number of approaches are being developed that focus on the IL-15 receptor and its signaling pathway. A diphtheria toxin IL-15 fusion protein DAB389 sIL-15 has been constructed that is directed toward the cytotoxic elimination of IL-15R-expressing cells (48). However, most approaches have been directed toward inhibiting IL-15 action. As noted above, the administration of the IL-15 inhibitor, the soluble IL-15Rα chain, prevented the development of murine collagen-induced arthritis (46). Furthermore, an IL-15 receptor antagonist produced by mutating glutamine residues within the C-terminus of IL-15 to aspartic acid competitively inhibited IL-15-triggered cell proliferation (135). This IL-15 mutant protein markedly attenuated antigen-specific delayed hypersensitivity responses in BALB/c mice and enhanced the acceptance of islet cell allografts (135). Our own therapeutic approaches have focused on the cytokine receptor subunits and signaling pathways shared among multiple cytokines, including IL-15, in an effort to yield more profound immunosuppression than can be achieved by inhibition of the synthesis or action of a single cytokine such as IL-2 or by an antibody directed toward a private receptor subunit such as IL-2Rα that

P1: APR/MBG

P2: NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

40

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

binds only a single cytokine. Our initial trials used Mikβ1, an antibody directed toward IL-2/15Rβ that is shared by IL-2 and IL-15. A humanized version of this antibody prolongs renal allograft survival in cynomolgus monkeys (140). In our initial clinical trial, we evaluated this antibody in the therapy of patients with T-cell-type large granular lymphocytic leukemia associated with hematocytopenias. The monoclonal LGLs involved in this disease express IL-2/15Rβ and γc but not IL-2Rα (141, 142). These cells respond by proliferation and cytokine induction to the IL-15 produced by associated monocytes (132). Additional therapeutic efforts focusing on the IL-15/IL-15R system are directed toward the development of an inhibitor of Jak3, the signaling molecule used by IL-2, IL-4, IL-7, and IL-15 as an agent to yield controlled immunosuppression. Deficiency of Jak3 in the autosomal form of severe combined immunodeficiency disease (SCID) in humans or in mice made deficient in Jak3 by homologous recombination exhibit a lack of NK cells and T- and B-cell abnormalities but do not develop disorders in nonimmunological systems, suggesting that Jak3 is a rational target to yield a controlled immunosuppression of value in the treatment of autoimmune diseases or in the prevention of allograft rejection (109, 110). In addition, Jak3 is activated constitutively in select leukemias such as HTLV-I-associated ATL cell lines (127–128). These observations suggest that drugs that inhibit Jak3 activation may be of value as immunosuppressive and antileukemic agents. In summary, our present understanding of the IL-15/IL-15R system and its signaling pathways opens new possibilities for more specific immune intervention.

CONCLUSIONS AND FUTURE DIRECTIONS IL-15 is a 14–15 kDa member of the 4 α-helix bundle family of cytokines. In contrast to the regulation of IL-2, which is controlled at the level of message transcription and stabilization, the regulation of IL-15 is much more complex with multifaceted controls at the levels of message transcription, message translation, and protein translocation and secretion. IL-15 is regulated in part at the level of transcription induced in association with infection of monocytes by intracellular pathogens. This upregulation of IL-15 mRNA expression involves both an IRF-1/IRF-E response element and an NF-κB signaling pathway. Nevertheless, IL-15 is controlled predominantly at the level of translation and translocation. IL-15 mRNA includes a number of elements that impede its translation. In particular, the 50 AUGs of human IL-15 mRNA are burdened with 12 upstream AUGs that interfere with effective IL-15 translation. Furthermore, the 48-aa signal peptide and a cis-acting negative regulatory element in the C-terminus of the mature protein impede translation. The removal of these

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IL-15 AND ITS RECEPTORS

41

negative control mechanisms in an integrated fashion may give rise to a major increase in IL-15 synthesis. The broad array of negative regulatory features controlling IL-15 expression may be required because of the potency of IL-15 in inducing the expression of TNF-α, IL-1, IFN-γ , and other inflammatory cytokines and chemokines that if indiscriminantly expressed would be associated with serious disorders such as autoimmune inflammatory diseases. In terms of a more positive role for IL-15, we propose that by maintaining a pool of translationally inactive IL-15 mRNA, diverse cells might respond rapidly to an intracellular infection by transforming IL-15 mRNA into a transcript that can be effectively translated. Despite our progress in understanding the multifaceted control of IL-15 translation, many questions must be answered before we can understand the molecular mechanisms underlying this translational control. There is considerable precedence for an impediment to translation manifested by AUGs upstream of an authentic initiation codon as seen with IL-15. Some of these transcripts may not use CAP-dependent scanning mechanisms to initiate translation; rather they may recruit ribosomes to an internal ribosome entry site (IRES), bypassing the impediments associated with the upstream 50 UTRs. Infections by intracellular organisms may lead to the induction of specific cytoplasmic proteins that bind to such a putative IRES, thereby facilitating translation. The mechanisms underlying the signal peptide coding sequence– or mature protein coding sequence–mediated regulations of IL-15 translation also have not been defined. A translational activator such as a specific RNA-binding protein might have to be produced to facilitate chain elongation and translocation. A more likely possibility given the effective translation of IL-15 mRNAs in the wheat-germ but not mammalian systems is that a mammalian translational repressor or a stable secondary structure in the IL-15 mRNA might be present that until released prevents efficient elongation of the translated message or translocation of the protein generated. Precedence for a required release from translational repressors before effective translation is possible is provided by transcripts for ferritin, erythroid 5-aminolevulinate synthase, and thymidylate synthase. IL-15 uses two distinct receptor and signaling pathways. In T and NK cells the type-1 IL-15 receptor includes IL-2/15Rβ and γc subunits, which are shared with IL-2, and an IL-15-specific receptor subunit, IL-15Rα. However, mast cells respond to IL-15 using another receptor system (type-2) that does not share elements with the IL-2R system but uses a novel 60- to 65-kDa IL-15RX element. IL-15 signaling involves activation of Jak1 and Jak3 as well as STAT3 and STAT5 in T and NK cells, whereas in mast cells IL-15-signaling through its specific IL-15RX receptor leads to Jak2 and STAT5 activation. The future molecular cloning of the gene encoding IL-15RX, the type-2 IL-15 receptor,

P1: APR/MBG

P2: NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

42

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA

would be of value in defining the tissue distribution of this receptor, and in delineating its involvement in mast cell and other cell biology. The IL-15 signal peptides are not only involved in the regulation of IL-15 translation but also direct its intracellular trafficking. Two isoforms of human IL-15 exist: one with a short (21-aa) signal peptide and another with a longer (48-aa) signal peptide. The IL-15 linked to the short signal peptide and some of that associated with the longer signal peptide is not secreted but is stored intracellularly, appearing in the nucleus and cytoplasmic components. Production of an intracellular lymphokine is not typical of other soluble interleukin systems. The possibility that IL-15 has biological functions as an intracellular molecule should be explored. IL-15 and IL-2 share some biological activities including the induction of T-cell proliferation, the activation of cytotoxic effector cells, the costimulation of immunoglobulin synthesis by B cells, and the activation of monocytes. In addition, IL-15 appears to play pivotal roles in the differentiation of NK cells from their progenitors, the maintenance of their survival, and their activation. Furthermore, IL-15 acts on an array of nonimmunological cells including mast cells, skeletal muscle cells, and microglia. The generation of an IL-15 knockout mouse that is now under way should assist in the definition of the unique nonredundant IL-15 functions. Abnormalities of IL-15 expression caused by P40 tax–mediated transactivation of IL-15 have been demonstrated in abnormal T cells in HTLV-I-associated ATL and in TSP/HSM. Abnormalities of IL-15 expression may also be involved in the pathogenesis of inflammatory autoimmune disorders such as RA and inflammatory bowel disease. The clinical application of new therapeutic agents that target IL-15 or the receptor and signaling elements shared by IL-2 and IL-15 may provide a new perspective for the treatment of such disorders. ACKNOWLEDGMENTS The authors acknowledge the excellent editorial assistance of Barbara Holmlund in the preparation of this manuscript. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Bazan JF. 1990. Structural design and molecular evolution of a cytokine receptor superfamily. Proc. Natl. Acad. Sci. USA 87:6934–38 2. Sato N, Miyajima A. 1994. Multimeric cytokine receptors: common versus specific functions. Curr. Opin. Cell Biol. 6: 174–79

3. Kishimoto T, Taga T, Akira S. 1994. Cytokine signal transduction. Cell 76:253– 62 4. Kondo M, Takeshita T, Ishii N, Nakamura M, Watanabe S, Arai K-I, Sugamura K. 1993. Sharing of the interleukin-2 (IL-2) receptor γ chain between receptors for IL2 and IL-4. Science 262:1874–77

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IL-15 AND ITS RECEPTORS 5. Noguchi M, Nakamura Y, Russell SM, Ziegler SF, Tsang M, Cao X, Leonard WJ. 1993. Interleukin 2 receptor γ chain: a functional component of the interleukin7 receptor. Science 262:1877–80 6. Burton JD, Bamford RN, Peters C, Grant AJ, Kurys G, Goldman CK, Brennan J, Roessler E, Waldmann TA. 1994. A lymphokine, provisionally designated interleukin-T and produced by a human adult T-cell leukemia line, stimulates Tcell proliferation and the induction of lymphokine-activated killer cells. Proc. Natl. Acad. Sci. USA 91:4935–39 7. Bamford RN, Grant AJ, Burton JD, Peters C, Kurys G, Goldman CK, Brennan J, Roessler E, Waldmann TA. 1994. The interleukin (IL) 2 receptor β chain is shared by IL-2 and a cytokine, provisionally designated IL-T, that stimulates T-cell proliferation and the induction of lymphokineactivated killer cells. Proc. Natl. Acad. Sci. USA 91:4940–44 8. Grabstein KH, Eisenman J, Shanebeck K, Rauch C, Srinivasan S, Fung V, Beers C, Richardson J, Schoenborn MA, Ahdieh M, Johnson L, Alderson MR, Watson JD, Anderson DM, Giri JG. 1994. Cloning of a T cell growth factor that interacts with the β chain of the interleukin-2 receptor. Science 264:965–68 9. Giri JG, Ahdieh M, Eisenman J, Shanebeck K, Grabstein K, Kumaki S, Namen A, Park LS, Cosman D, Anderson D. 1994. Utilization of the β and γ chains of the IL-2 receptor by the novel cytokine IL-15. EMBO J. 13:2822–30 10. Waldmann TA, Tagaya Y, Bamford R. 1998. Interleukin-2, interleukin-15, and their receptors. Int. Rev. Immunol. 16:205–26 11. Tagaya Y, Bamford RN, DeFilippis A, Waldmann TA. 1996. IL-15: a pleiotropic cytokine with diverse receptor/signaling pathways whose expression is controlled at multiple levels. Immunity 4:329–36 12. Kennedy MK, Park LS. 1996. Characterization of interleukin-15 (IL-15) and the IL-15 receptor complex. J. Clin. Immunol. 16:134–43 13. DiSanto JP. 1997. Cytokines: shared receptors, distinct functions. Curr. Biol. 7:R424–26 14. Bamford RN, Tagaya Y, Waldmann TA. 1997. Interleukin 15—What it does and how it is controlled. Immunologist 5:52– 56 15. Bamford RN, Battiata AP, Burton JD, Sharma H, Waldmann TA. 1996. Interleukin (IL) 15/IL-T production by the adult T-cell leukemia cell line HuT-102

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

43

is associated with a human T-cell lymphotrophic virus type I R region/IL-15 fusion message that lacks many upstream AUGs that normally attenuate IL15 mRNA translation. Proc. Natl. Acad. Sci. USA 93:2897–902 Bamford RN, DeFilippis AP, Azimi N, Kurys G, Waldmann TA. 1998. The 50 UTR, signal peptide and 30 coding sequences of IL-15 participate in its multifaceted translational control. J. Immunol. 160:4418–26 Fraser JD, Irving BA, Crabtree GR, Weiss A. 1991. Regulation of interleukin-2 gene enhancer activity by the T cell accessory molecule CD28. Science 251:313–16 Lindstein T, June CH, Ledbetter JA, Stella G, Thompson CB. 1989. Regulation of lymphokine messenger RNA stability by a surface-mediated T cell activation pathway. Science 244:339–43 Meazza R, Verdiani S, Biassoni R, Coppolecchia M, Gaggero A, Orengo AM, Colombo MP, Azzarone B, Ferrini S. 1996. Identification of a novel interleukin-15 (IL-15) transcript isoform generated by alternative splicing in human small cell lung cancer cell lines. Oncogene 12:2187–92 Onu A, Pohl T, Krause H, Bulfone-Paus S. 1997. Regulation of IL-15 secretion via the leader peptide of two IL-15 isoforms. J. Immunol. 158:255–62 Giri JG, Kumaki S, Ahdieh M, Friend DJ, Loomis A, Shanebeck K, DuBose R, Cosman D, Park LS, Anderson DM. 1995. Identification and cloning of a novel IL15 binding protein that is structurally related to the α chain of the IL-2 receptor. EMBO J. 14:3654–63 Tagaya Y, Burton JD, Miyamoto Y, Waldmann TA. 1996. Identification of a novel receptor/signal transduction pathway for IL-15/T in mast cells. EMBO J. 15:4928– 39 Trentin L, Zambello R, Facco M, Sancetta R, Agostini C, Semenzato G. 1997. Interleukin-15: a novel cytokine with regulatory properties on normal and neoplastic B lymphocytes. Leuk. Lymph. 27:35–42 Armitage RJ, Macduff BM, Eisenman J, Paxton R, Grabstein KH. 1995. IL-15 has stimulatory activity for the induction of B cell proliferation and differentiation. J. Immunol. 154:483–90 Wilkinson PC, Liew FY. 1995. Chemoattraction of human blood T lymphocytes by interleukin-15. J. Exp. Med. 181:1255– 59 Carson WE, Giri JG, Lindermann MJ, Linett ML, Ahdieh M, Paxton R,

P1: APR/MBG

P2: NBL/ary

January 21, 1999

44

27.

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

28.

29.

30.

31.

32.

33.

34.

35.

36.

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA Anderson D, Eisenmann J, Grabstein K, Caligiuri MA. 1994. Interleukin (IL) 15 is a novel cytokine that activates human natural killer cells via components of the IL-2 receptor. J. Exp. Med. 180:1395–403 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 β chain. J. Exp. Med. 185:499–505 Mr´ozek E, Anderson P, Caligiui MA. 1996. Role of interleukin-15 in the development of human CD56+ hematopoietic progenitor cells. Blood 87:2632–40 Cavazzana-Calvo M, Hacein-Bev S, de Saint Basile G, DeCoene C, Selz F, LeDeist F, Fischer A. 1996. Role of interleukin-2 (IL-2), IL-7, and IL-15 in natural killer cell differentiation from cord blood hematopoietic progenitor cells and from γc transduced severe combined immunodeficiency X1 bone marrow cells. Blood 88:3901–9 Mingari MC, Vitale C, Cantoni C, Bellomo R, Ponte M, Schiavetti F, Bertone S, Moretta A, Moretta L. 1997. Interleukin15 induced maturation of human natural killer cells from early thymic precursors, selective expression of CD94/NKG2-A as the only HLA class I-specific inhibitory receptor. Eur. J. Immunol. 27:1374–80 Flamand L, Stefanescu I, Menezes J. 1996. Human herpesvirus-6 enhances natural killer cell cytotoxicity via IL-15. J. Clin. Invest. 97:1373–81 Carson WE, Ross ME, Baiocchi RA, Marien MJ, Boiani N, Grabstein K, Caligiuri MA. 1995. Endogenous production of interleukin 15 by activated human monocytes is critical for optimal production of interferon-γ by natural killer cells in vitro. J. Clin Invest. 96:2578–82 Carson WE, Fehniger TA, Haldar S, Eckhert K, Lindemann MJ, Lai C-F, Croce CM, Baumann H, Caligiuri MA. 1997. Potential role for interleukin-15 in the regulation of human natural killer cell survival. J. Clin. Invest. 99:937–43 Ohteki T, Ho S, Suzuki H, Mak TW, Ohashi P. 1997. Role for IL-15/IL-15 receptor β-chain in natural killer 1. 1+ T cell receptor-αβ + cell development. J. Immunol. 159:5931–35 Ogasawara K, Hida S, Azimi N, Tagaya Y, Sato T, Yokochi-Fukuda Y, Waldmann TA, Taniguchi T, Taki S. 1998. Requirement for IRF-1 in the microenvironment supporting development of natural killer cells. Nature 391:700–3 Ohteki T, Yoshida H, Matsuyama T, Dun-

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

can 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-α/β + (NK1+T) cells, natural killer cells, and intestinal intraepithelial T cells. J. Exp. Med. 187:967–72 Leclercq G, Debacker V, De Smedt M, Plum J. 1996. Differential effects of interleukin-15 and interleukin-2 on differentiation of bipotential T/natural killer progenitor cells. J. Exp. Med. 184:325– 36 Puzanov IJ, Williams NS, Schatzle J, Sivakumar PV, Bennet M, Kumar V. 1997. Ontogeny of NK cells and the bone marrow microenvironment: Where does IL15 fit in? Res. Immunol. 148:195–201 Atedzoe BN, Ahmad A, Menezes J. 1997. Enhancement of natural killer cell cytotoxicity by the human herpesvirus-7 via IL-15 induction. J. Immunol. 159:4966– 72 Quinn LS, Haugk KL, Grabstein KH. 1995. Interleukin-15: a novel anabolic cytokine for skeletal muscle. Endocrinology 136:3669–72 Quinn LS, Haugk KL, Damon SE. 1997. Interleukin-15 stimulates C2 skeletal myoblast differentiation. Biochem. Biophys. Res. Commun. 239:6–10 McInnes IB, Al-Mughales J, Field M, Leung BP, Huang F-P, Dixon R, Sturrock RD, Wilkinson PC, Liew FY. 1996. The role of interleukin 15 in T-cell migration and activation in rheumatoid arthritis. Nature Med. 2:175–82 McInnes IB, Leung BP, Sturrock RD, Field M, Liew FY. 1997. Interleukin15 mediates T cell–dependent regulation of tumor necrosis factor-α production in rheumatoid arthritis. Nature Med. 3:189– 95 Reinecker HC, MacDermott RP, Mirau S, Dignass A, Podolsky DK. 1996. Intestinal epithelial cells both express and respond to interleukin-15. Gastroenterology 111:1706–13 Azimi N, Brown K, Bamford RN, Tagaya Y, Siebenlist U, Waldmann TA. 1998. Human T cell lymphotropic virus type I tax protein trans-activates interleukin 15 gene transcription through an NF-κB site. Proc. Natl. Acad. Sci. USA 95:2452–57 Ruchatz H, Leung BP, Wei XQ, McInnes IB, Liew FY. 1998. Soluble IL-15 receptor α-chain administration prevents murine collagen-induced arthritis: a role for IL15 in development of antigen-induced immunopathology. J. Immunol. 160:5654– 60

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IL-15 AND ITS RECEPTORS 47. Pilson RS, Levin W, Desai B, Reik LM, Lin P, Korkmaz-Duffy E, Campbell E, Tso JY, Kerwin JA, Hakimi J. 1997. Bispecific humanized anti-IL-2 receptor αβ antibodies inhibitory for both IL-2and IL-15-mediated proliferation. J. Immunol. 159:1543–56 48. vanderSpek JC, Sutherland J, Sampson E, Murphy JR. 1995. Genetic construction and characterization of the diphtheria toxin-related interleukin 15 fusion protein DAB389 sIL-15. Protein Eng. 8:1317–21 49. Canals A, Grimm DR, Gasbarre LC, Lunney JK, Zarlenga DS. 1997. Molecular cloning of cDNA encoding porcine interleukin-15. Gene 195:337–39 50. Anderson DM, Johnson L, Glaccum MB, Copeland NG, Gilbert DJ, Jenkins NA, Valentine V, Kirstein MN, Shapiro DN, Morris SW, Grabstein K, Cosman D. 1995. Chromosomal assignment and genomic structure of IL-15. Genomics 25:701–6 51. Seigel LJ, Harper ME, Wong-Staal F, Gallo RC, Nash WG, O’Brien SJ. 1984. Gene for T-cell growth factor: location on human chromosome 4q and feline chromosome BI. Science 223:175–78 52. Tagaya Y, Kurys G, Thies TA, Losi JM, Azimi N, Hanover JA, Bamford RN, Waldmann TA. 1997. Generation of secretable and non-secretable interleukin 15 isoforms through alternate usage of signal peptides. Proc. Natl. Acad. Sci. USA 94:14444–49 53. Doherty TM, Seder RA, Sher A. 1996. Induction and regulation of IL-15 expression in murine macrophages. J. Immunol. 156:735–41 54. Mody CH, Spurrell JCL, Wood CJ. 1998. Interleukin 15 induces antimicrobial activity after release by Cryptococcus neoformans-stimulated monocytes. J. Infect. Dis. 178:803–14 55. Nishimura H, Hiromatsu K, Kobayashi N, Grabstein KH, Paxton R, Sugamura K, Bluestone JA, Yoshikai Y. 1996. IL-15 is a novel growth factor for murine γ δ T cells induced by Salmonella infection. J. Immunol. 156:663–69 56. V´azquez N, Walsh TJ, Friedman D, Chanock SJ, Lyman CA. 1998. Interleukin-15 augments superoxide production and microbicidal activity of human monocytes against Candida albicans. Infect. Immun. 66:145–50 57. Jullien D, Sieling PA, Uyemura K, Mar ND, Rea TH, Modlin RL. 1997. IL15, an immunomodulator of T cell responses in intracellular infection. J. Immunol. 158:800–6

45

58. Washizu J, Nishimura H, Nakamura N, Nimura Y, Yoshikai Y. 1998. The NFkappa B binding site is essential for translational activation of the IL-15 gene. Immunogenetics 48:1–7 59. Kozak M. 1987. An analysis of 50 noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res. 15:8125–48 60. Kozak M. 1991. An analysis of vertebrate mRNA sequences: intimations of translational control. J. Cell Biol. 115:887–903 61. Kozak M. 1992. A consideration of alternative models for the initiation of translation in eukaryotes. Crit. Rev. Biochem. Mol. Biol. 27:385–402 62. Weinberg AD, Swain SL. 1990. IL-2 receptor (Tac antigen) protein expression is down-regulated by the 50 -untranslated region of the mRNA. J. Immunol. 144: 4712–20 63. Arrick BA, Lee AL, Grendell RL, Derynck R. 1991. Inhibition of translation of transforming growth factor-beta 3 mRNA by its 50 untranslated region. Mol. Cell. Biol. 11:4306–13 64. Bates B, Hardin J, Zhan X, Drickamer K, Goldfarb M. 1991. Biosynthesis of human fibroblast growth factor-5. Mol. Cell. Biol. 11:1840–45 65. Namen AE, Lupton S, Hjerrild K, Wignall J, Mochizuki DY, Schmierer A, Mosley B, March CJ, Urdal D, Gillis S, Cosman D, Goodwin RG. 1988. Stimulation of B-cell progenitors by cloned murine interleukin7. Nature 333:571–73 66. McBratney S, Chen CY, Sarnow P. 1993. Internal initiation of translation. Curr. Opin. Cell. Biol. 5:961–65 67. Belsham GJ, Sonenberg N, Svitkin YV. 1995. The role of the La autoantigen in internal initiation. Curr. Top. Microbiol. Immunol. 203:85–98 68. Macejak DG, Sarnow P. 1991. Internal initiation of translation mediated by the 50 leader of a cellular mRNA. Nature 353:90–94 69. Yueh A, Schneider RJ. 1996. Selective translation initiation by ribosome jumping in adenovirus-infected and heatshocked cells. Genes Dev. 10:1557–67 70. Nishimura H, Washizu J, Nakamura N, Enomoto A, Yoshikai Y. 1998. Translational efficiency is up-regulated by alternative exon in murine IL-15 mRNA. J. Immunol. 160:936–42 71. Hentze MW, Rouault TA, Caughman SW, Dancis W, Harford JB, Klausner RD. 1987. A cis-acting element is necessary and sufficient for translational regulation of human ferritin expression in

P1: APR/MBG

P2: NBL/ary

January 21, 1999

46

72.

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA response to iron. Proc. Natl. Acad. Sci. USA 84:6730–34 Chu E, Voeller D, Koeller DM, Drake JC, Takimoto CH, Maley GF, Maley F, Allegra CJ. 1993. Identification of an RNA binding site for human thymidylate synthase. Proc. Natl. Acad. Sci. USA 90:517– 21 Mosner J, Mummenbrauer T, Bauer C, Sczakiel G, Grosse F, Deppert W. 1995. Negative feedback regulation of wildtype p53 biosynthesis. EMBO J. 14:4442– 49 Theodorakis NG, Barerji SS, Morimoto RI. 1988. HSP70 mRNA translation in chicken reticulocytes is regulated at the level of elongation. J. Biol. Chem. 263:14579–85 Kaspar RL, Gehrke L. 1994. Peripheral blood mononuclear cells stimulated with C5a or lipopolysaccharide to synthesize equivalent levels of IL-1β mRNA show unequal IL-1β protein accumulation but similar polyribosome profiles. J. Immunol. 153:277–86 Welsh M, Scherberg N, Gilmore R, Steiner DF. 1986. Translational control of insulin biosynthesis. Evidence for regulation of elongation, initiation and signal recognition-particle-mediated translation arrest by glucose. Biochem. J. 235:459– 67 Flanagan JG, Chan DC, Leder P. 1991. Transmembrane form of kit ligand growth factor is determined by alternative splicing and is missing in the Sld mutant. Cell 64:1025–35 Acland P, Dixon M, Peters G, Dickson C. 1990. Subcellular fate of the Int-2 oncoprotein is determined by choice of initiation codon. Nature 343:662–65 Hatakeyama M, Tsudo M, Minamoto S, Kono T, Doi T, Miyata T, Miyasaka M, Taniguchi T. 1989. Interleukin-2 receptor β chain gene: generation of three receptor forms by cloned human α and β chain cDNAs. Science 244:551–56 Goldsmith MA, Greene WC. 1994. Interleukin-2 and the interleukin-2 receptor. In The Cytokine Handbook, ed. AW Thomson, pp. 57–80. London: Academic. 2nd ed. Takeshita T, Asao H, Ohtani K, Ishii N, Kumaki S, Tanaka N, Munakata H, Nakamura M, Sugamura K. 1992. Cloning of the γ chain of the human IL-2 receptor. Science 257:379–82 Anderson DM, Kumaki S, Abdieh M, Bertles J, Tometsko M, Loomis A, Giri J, Copeland NG, Gilbert DJ, Jenkins NA, Valentine V, Shapiro DN, Morris SW,

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

Park LS, Cosman D. 1995. Functional characterization of human interleukin 15 receptor α chain and close linkage of IL15RA and IL-2RA genes. J. Biol. Chem. 270:29862–69 Witthuhn BA, Silvennoinen O, Miura O, Lai KS, Cwik C, Liu ET, Ihle JN. 1994. Involvement of the Jak-3 Janus kinase in signaling by interleukins 2 and 4 in lymphoid and myeloid cells. Nature 370:153– 57 Johnston JA, Bacon CM, Finbloom DS, Rees RC, Kaplan D, Shibuya K, Ortaldo JR, Gupta S, Chen YQ, Giri JD, O’Shea JJ. 1995. Tyrosine phosphorylation and activation of STAT5, STAT3, and Janus kinases by interleukin 2 and 15. Proc. Natl. Acad. Sci. USA 92:8705–9 Lin J-X, Migone T-S, Tsang M, Friedmann M, Weatherbee JA, Zhou L, Yamauchi A, Bloom ET, Mietz J, John S, Leonard WJ. 1995. The role of shared receptor motifs and common Stat proteins in the generation of cytokine pleiotropy and redundancy by IL-2, IL-4, IL-7, IL13, and IL-15. Immunity 2:331–39 Miyazaki T, Liu Z-J, Kawahara A, Minami Y, Yamada K, Tsujimoto Y, Barsoumian EL, Perlmutter RM, Taniguchi T. 1995. Three distinct IL-2 signaling pathways mediated by bcl–2, c-myc and lck cooperate in hematopoietic cell proliferation. Cell 81:223–31 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 Carson WE, Caligiuri MA. 1996. Interleukin 15: a potential player during the innate immune response to infection. Exp. Parasitol. 84:291–94 Inagaki-Ohara K, Nishimura H, Mitani A, Yoshikai Y. 1997. Interleukin-15 preferentially promotes the growth of intestinal intraepithelial lymphocytes bearing γ δ T cell receptor in mice. Eur. J. Immunol. 27:2885–91 Garcia VE, Jullien D, Song M, Uyemura K, Shuai K, Morita CT, Modlin RL. 1998. IL-15 enhances the response of human γ δ T cells to nonpeptide microbial antigens. J. Immunol. 160:4322–29 Edelbaum D, Mohamadzadeh M, Bergstresser PR, Sugamura K, Takashima A. 1995. Interleukin (IL)-15 promotes the growth of murine epidermal γ δ T cells by a mechanism involving the β- and γc -chains of the IL-2 receptor. J. Invest. Dermatol. 105:837–43 Jonuleit H, Wiedemann K, Muller G, Degwert J, Hoppe U, Knop J, Enk AH.

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

IL-15 AND ITS RECEPTORS

93.

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

94.

95.

96.

97.

98.

99.

100.

101.

102.

103.

1997. Induction of IL-15 messenger RNA and protein in human blood-derived dendritic cells. J. Immunol. 158:2610–15 Blauvelt A, Asada H, Klaus-Kovtun V, Altman DJ, Lucey DR, Katz SI. 1996. Interleukin-15 mRNA is expressed by human keratinocytes, Langerhans cells, and blood-derived dendritic cells and is downregulated by ultraviolet B radiation. J. Invest. Dermatol. 106:1047–52 Patki AH, Quinones-Mateu ME, Dorazio D, Yen-Lieberman B, Boom WH, Thomas EK, Lederman MM. 1996. Activation of antigen-induced lymphocyte proliferation by interleukin-15 without the mitogenic effect of interleukin-2 that may induce human immunodeficiency virus-1 expression. J. Clin. Invest. 98:616–21 Lewko WM, Smith TL, Bowman DJ, Good RW, Oldham RK. 1995. Interleukin-15 and the growth of tumor derived activated T-cells. Cancer Biother. 10:13– 20 Seder RA. 1995. IL-2 and IL-15 induce a TH 1 response from naive CD4+ transgenic T cells following short term in vitro stimulation. In Progr. Immunol. 9th Int. Congr. Immunol., San Francisco, p. 75 Seder RA. 1996. High-dose IL-2 and IL15 enhance the in vitro priming of naive CD4+ T cells for IFN-γ but have differential effects on priming for IL-4. J. Immunol. 156:2413–22 Ye W, Young JD, Liu, C-C. 1996. Interleukin-15 induces the expression of mRNAs of cytolytic mediators and augments cytotoxic activities in primary murine lymphocytes. Cell. Immunol. 174:54– 62 Treiber-Held S, Stewart DM, Kurman CC, Nelson DL. 1996. IL-15 induces the release of soluble IL-2Rα from human peripheral blood mononuclear cells. Clin. Immunol. Immunopathol. 79:71–78 Treiber-Held S, Stewart DM, Barraclough HA, Kurman CC, Nelson DL. 1996. Release of sIL-2Rα from and activation of native human peripheral blood mononuclear cells by recombinant IL-15. Clin. Immunol. Immunopathol. 80:67–75 Bulfone-Paus S, D¨orkop H, Paus R, Krause, H, Pohl T, Onu A. 1997. Differential regulation of human T lymphoblast functions by IL-2 and IL-15. Cytokine 9:507–13 Bulfone-Paus S, Ungureanu D, Pohl T, Lindner G, Paus R, Ruckert R, Krause H, Kunzendorf U. 1997. Interleukin-15 protects from lethal apoptosis in vivo. Nat. Med. 3:1124–28 Kanegane H, Tosato G. 1996. Activa-

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

47

tion of naive and memory T cells by interleukin-15. Blood 88:230–35 Salvucci O, Mami-Chouaib F, Moreau JL, Theze J, Chehimi J, Chouaib S. 1996. Differential regulation of interleukin-12and interleukin-15-induced natural killer cell activation by interleukin-4. Eur. J. Immunol. 26:2736–41 Damjanovich S, Bene L, Matko J, Alileche A, Goldman CK, Sharrow S, Waldmann TA. 1997. Preassembly of interleukin 2 (IL-2) receptor subunits on resting Kit 225 K6 T cells and their modulation by IL-2, IL-7, and IL-15: a fluorescence resonance energy transfer study. Proc. Natl. Acad. Sci. USA 94:13134– 39 Cao X, Shores EW, Hu-Li J, Anver MR, Kelsall BL, Russell SM, Drago J, Noguchi M, Grinberg A, Bloom ET, Paul WE, Katz SI, Love PE, Leonard WJ. 1995. Defective lymphoid development in mice lacking expression of the common cytokine receptor γ chain. Immunity 2:223–38 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 γ chain. Proc. Natl. Acad. Sci. USA 92:377–81 Kundig TM, Schorle H, Bachmann MF, Hengartner H, Zinkernagel RM, Horak I. 1993. Immune responses in interleukin2-deficient mice. Science 262:1059– 61 Russell SM, Tayebi N, Nakajima H, Riedy MC, Roberts JL, Aman MJ, Migone TS, Nogouchi M, Markert NL, Buckley RH, O’Shea JJ, Leonard WJ. 1995. Mutation of Jak3 in a patient with SCID: essential role of Jak3 in lymphoid development. Science 270:797–800 Macchi P, Villa A, Gillani S, Sacco MG, Frattini A, Porta F, Vgazio AG, Johnston JA, Candotti F, O’Shea JJ, Vezzoni P, Notarangelo LD. 1995. Mutations of Jak-3 gene in patients with autosomal severe combined immunodeficiency disease (SCID). Nature 377:65–68 Biron CA, Byron KS, Sullivan JL. 1989. Severe herpesvirus infections in an adolescent without natural killer cells. New Engl. J. Med. 320:1731–35 Badolato R, Ponzi AN, Millesimo M, Notarangelo LD, Musso T. 1997. Interleukin-15 (IL-15) induces IL-8 and monocyte chemotactic protein 1 production in human monocytes. Blood 90:2804–9 Angiolillo AL, Kanegane H, Sgadari C, Reaman GH, Tosato G. 1997. Interleukin-15 promotes angiogenesis in

P1: APR/MBG

P2: NBL/ary

January 21, 1999

48

114.

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

WALDMANN & TAGAYA vivo. Biochem. Biophys. Res. Commun. 233:231–37 Lee YB, Satoh, J, Walker DG, Kim SU. 1996. Interleukin-15 gene expression in human astrocytes and microglia in culture. Neuroreport 7:1062–66 Hanisch U-K, Lyons SA, Prinz M, Nolte C, Weber JR, Kettenmann H, Kirchhoff F. 1997. Mouse brain microglia express interleukin-15 and its multimeric receptor complex functionally coupled to Janus kinase activity. J. Biol. Chem. 272:28853– 60 Stevens AC, Matthews J, Andres P, Baffis V, Zheng XX, Chae D-W, Smith J, Strom TB, Maslinski W. 1997. Interleukin-15 signals T84 colonic epithelial cells in the absence of the interleukin-2 receptor βchain. Am. J. Physiol. 272:G1201–8 Feldmann M, Brennan FM, Maini RN. 1996. Role of cytokines in rheumatoid arthritis. Annu. Rev. Immunol. 14:397– 440 Oppenheimer-Marks N, Brezinschek RI, Mohamadzadeh M, Vita R, Lipsky PE. 1998. Interleukin 15 is produced by endothelial cells and increases the transendothelial migration of T cells in vitro and in the SCID mouse-human rheumatoid arthritis model in vivo. J. Clin. Invest. 101:1261–72 Kirman I, Nielsen OH. 1996. Increased numbers of interleukin-15-expressing cells in active ulcerative colitis. Am. J. Gastroenterol. 91:1789–94 Kakumu S, Okumura A, Ishikawa T, Yano M, Enomoto A, Nishimura H, Yoshioka K, Yoshikai Y. 1997. Serum levels of IL10, IL-15 and soluble tumour necrosis factor-alpha (TNF-α) receptors in type C chronic liver disease. Clin. Exp. Immunol. 109:458–63 Agostini C, Trentin L, Facco M, Sancetta R, Cerutti A, Tassinari C, Cimarosto L, Adami F, Cipriani A, Zambello R, Semenzato G. 1996. Role of IL-15, IL-2 and their receptors in the development of T cell alveolitis in pulmonary sarcoidosis. J. Immunol. 157:910–18 Kivisakk P, Matusevicius D, He B, Soderstrom M, Fredrikson S, Link H. 1998. IL15 mRNA expression is up-regulated in blood and cerebrospinal fluid mononuclear cells in multiple sclerosis (MS). Clin. Exp. Immunol. 111:193–97 Yodoi J, Takatsuki K, Masuda T. 1974. Two cases of T-cell chronic lymphotropic leukemia in Japan. New Engl. J. Med. 290:572–73 Jacobson S, Zaninovic V, Mora O, Rodgers-Johnson P, Sheremata WA,

125.

126.

127.

128.

129.

130.

131.

132.

Gibbs CJ Jr, Gajdusek C, McFarlin DE. 1988. Immunological findings in neurological diseases associated with antibodies to HTLV-I: activated lymphocytes in tropical spastic paraparesis. Ann Neurol 23:196s–200s (Suppl.) Tendler CL, Greenberg SJ, Blattner WA, Manns A, Murphy E, Fleisher T, Hanchard B, Morgan O, Burton JD, Nelson DL, Waldmann TA. 1990. Transactivation of interleukin-2 and its receptor induces immune activation in HTLV-I associated myelopathy: pathogenic implications and a rationale for immunotherapy. Proc. Natl. Acad. Sci. USA 87:5218–22 Yamada Y, Sugawara K, Hata T, Tsuruta K, Moriuchi R, Maeda T, Atogami S, Murata K, Fujimoto K, Kohno T, Tsukasaki K, Tomonaga M, Hirakata Y, Kamihira S. 1998. Interleukin-15 (IL-15) can replace the IL-2 signal in IL-2-dependent adult Tcell leukemia (ATL) cell lines: expression of IL-15 receptor α on ATL cells. Blood 91:4265–72 Migone TS, Lin JX, Cereseto A, Mulloy JC, O’Shea JJ, Franchini G, Leonard WL. 1995. Constitutively activated JakSTAT pathway in T cells transformed with HTLV-I. Science 269:79–81 Xu X, Kang SH, Heidenreich O, Okerholm M, O’Shea JJ, Nerenberg MI. 1995. Constitutive activation of different Jak tyrosine kinases in human T cell leukemia virus type 1 (HTLV–1) tax protein or virus-transformed cells. J. Clin. Invest. 96:1548–55 Kacani L, Stoiber H, Dierich MP. 1997. Role of IL-15 in HIV-1-associated hypergammaglobulinemia. Clin. Exp. Immunol. 108:14–18 Chehimi J, Marshall JD, Salvucci O, Frank I, Chehimi S, Kawecki S, Bacheller D, Rifat S, Chouaib S. 1997. IL-15 enhances immune functions during HIV infection. J. Immunol. 158:5978–87 Barzegar C, Meazza R, Pereno R, PottinClemenceau C, Scudeletti M, BroutyBoye D, Doucet C, Taoufik Y, Ritz J, Musselli C, Mishal Z, Jasmin C, Indiveri F, Ferrini S, Azzarone B. 1998. IL15 is produced by a subset of human melanomas, and is involved in the regulation of markers of melanoma progression through juxtacrine loops. Oncogene 16:2503–12 Zambello R, Facco M, Trentin L, Sancetta R, Tassinari C, Perin A, Milani A, Pizzolo G, Rodeghiero F, Agostini C, Meazza R, Ferrini S, Semenzato G. 1997. Interleukin-15 triggers the proliferation and cytotoxicity of granular lymphocytes

P1: APR/MBG

P2: NBL/ary

January 21, 1999

9:56

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-02

IL-15 AND ITS RECEPTORS

133.

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

134.

135.

136.

137.

138.

in patients with lymphoproliferative disease of granular lymphocytes. Blood 89: 201–11 Pavlakis M, Strehlau J, Lipman M, Shapiro M, Maslinski W, Strom TB. 1996. Intragraft IL-15 transcripts are increased in human renal allograft rejection. Transplantation 62:543–45 Manfro RC, Roy-Chaudhury P, Zheng XX, Steiger J, Nickerson PW, Li Y, Maslinski W, Strom TB. 1997. Interleukin15 gene transcripts are present in rejecting islet allografts. Transplantation Proc. 29:1077–78 Kim YS, Maslinski W, Zheng XX, Stevens AC, Li XC, Tesch GH, Kelly VR, Strom TB. 1998. Targeting the IL15 receptor with an antagonist IL-15 mutant/Fcγ 2a protein blocks delayed-type hypersensitivity. J. Immunol. 160:5742– 48 Waldmann TA. 1993. The IL-2/IL-2 receptor system: a target for rational immune intervention. Immunol. Today 14: 264–70 Waldmann TA, White JD, Goldman CK, Top L, Grant A, Bamford R, Roessler E, Horak ID, Zaknoen S, Kasten-Sport`es C, England R, Horak E, Mishra B, Dipre M, Hale P, Fleisher TA, Junghans RP, Jaffe ES, Nelson DL. 1993. The interleukin2 receptor: a target for monoclonal antibody treatment of human T-cell lymphotropic virus I–induced adult T-cell leukemia. Blood 82:1701–12 Waldmann TA, White JD, Carrasquillo JA, Reynolds JC, Paik CH, Gansow OA, Brechbiel MW, Jaffe ES, Fleisher TA,

139.

140.

141.

142.

49

Goldman CK, Top L, Bamford RN, Zaknoen S, Roessler E, Kasten-Sport`es C, England R, Litou H, Johnson JA, JacksonWhite T, Manns A, Hanchard B, Junghans RP, Nelson DL. 1995. Radioimmunotherapy of interleukin-2Rα-expressing adult T-cell leukemia with yttrium-90-labeled anti-Tac. Blood 86:4063–75 Vincenti F, Kirkman R, Light S, Bumgardner G, Pescovitz M, Halloran P, Neylan J, Wilkinson A, Ekberg H, Gaston R, Backman L, Burdick J. 1998. Interleukin2-receptor blockade with daclizumab to prevent acute rejection in renal transplantation. Daclizumab Triple Therapy Study Group. New Engl. J. Med. 338:161–65 Tinubu SA, Hakimi J, Kondas JA, Bailon P, Familletti PC, Spence C, Crittenden MD, Parenteau GL, Dirbas FM, Tsudo M, Bacher JD, Kasten-Sport`es C, Martinucci JL, Goldman CK, Clark RE, Waldmann TA. 1994. Humanized antibody directed to the IL-2 receptor β chain prolongs primate cardiac allograft survival. J. Immunol. 153:4330–38 Tsudo M, Goldman CK, Bongiovanni KF, Chan WC, Winton EF, Yagita M, Grim EA, Waldmann TA. 1987. The P75 peptide is the receptor for interleukin 2 expressed on large granular lymphocytes and is responsible for the interleukin 2 activation of these cells. Proc. Natl. Acad. Sci. USA 84:5394–98 Yoon HJ, Sugamura K, Loughran TP Jr. 1990. Activation of leukemic large granular lymphocytes by interleukin-2 via the p75 interleukin-2 receptor. Leukemia 4:848–50

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:19-49. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:51–88

IMMUNODOMINANCE IN MAJOR HISTOCOMPATIBILITY COMPLEX CLASS I–RESTRICTED T LYMPHOCYTE RESPONSES1 Jonathan W. Yewdell and Jack R. Bennink Laboratory of Viral Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland 20892-0440; e-mail: [email protected], [email protected] KEY WORDS:

antigen processing, CTL, immunodominance, MHC

ABSTRACT Of the many thousands of peptides encoded by a complex foreign antigen that can potentially be presented to CD8+ T cells (TCD8+), only a small fraction induce measurable responses in association with any given major histocompatibility complex class I allele. To design vaccines that elicit optimal TCD8+ responses, a thorough understanding of this phenomenon, known as immunodominance, is imperative. Here we review recent progress in unraveling the molecular and cellular basis for immunodominance. Of foremost importance is peptide binding to class I molecules; only ∼1/200 of potential determinants bind at greater than the threshold affinity (K d > 500 nM) associated with immunogenicity. Limitations in the TCD8+ repertoire render approximately half of these peptides nonimmunogenic, and inefficient antigen processing further thins the ranks by approximately four fifths. As a result, only ∼1/2000 of the peptides in a foreign antigen expressed by an appropriate antigen presenting cell achieve immunodominant status with a given class I allele. A roughly equal fraction of peptides have subdominant status, i.e. they induce weak-to-nondetectable primary TCD8+ responses in the context of their natural antigen. Subdominant determinants may be expressed at or above levels of immunodominant determinants, at least on antigen presenting cells in vitro. The immunogenicity of subdominant determinants is often limited by immunodomination: suppression mediated by TCD8+ specific for immunodominant 1 The US government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

51

P1: PKS/SPD/MBG

January 25, 1999

52

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK determinants. Immunodomination is a central feature of TCD8+ responses, as it even occurs among clones responding to the same immunodominant determinant. Little is known about how immunodominant and subdominant determinants are distinguished by the TCD8+ repertoire, or how (and why) immunodomination occurs, but new tools are available to address these questions.

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

INTRODUCTION Discovery of Major Histocompatibility Complex Class I–Restricted Immunodominance In 1974, Zinkernagel & Doherty clearly established the biological importance of major histocompatibility complex class I molecules (hereafter referred to as class I molecules) by showing that cytotoxic T lymphocytes [now recognized as CD8+ T cells (TCD8+)] induced by viral infection recognize cells in a virusspecific, class I–restricted manner (1). Not long after, it was found that TCD8+ responses to even complex viruses expressing >100 gene products were often dominated by TCD8+ restricted to only one of the class I allomorphs expressed by the mice (allomorph refers to any of the alleles of the two to three class Ia genes expressed by a species) (2). When it became possible to examine responses to individual viral gene products with individual class I allomorphs, it was shown in mice (3) and humans (4) that few (if any) of the 10 influenza virus gene products were recognized in association with TCD8+ restricted by any given allomorph. Following the discovery that class I molecules bound only a small fragment of viral gene products (5), later appreciated to generally consist of 8–10 residues (6), it was shown that TCD8+ specific for individual gene products in association with a given allomorph often focused on a single peptide (6a). These findings echoed prior reports that TCD4+ responses to proteins frequently focused on one or a few peptides, termed immunodominant determinants (7). Other peptides, subdominant determinants, could elicit TCD4+ that recognize antigen presenting cells (APCs) exposed to either peptides or intact proteins, but they were only weakly immunogenic in the context of the intact protein. A third group of peptides, cryptic determinants, were immunogenic and antigenic only as synthetic peptides or larger proteolytic fragments. These terms apply equally to TCD8+ responses. Caution is in order, however. First, immunologists differ in their definitions of these terms and care must be taken to divine the usage in any given publication. Second, and crucially, these terms are defined strictly on a functional basis, and the classification of any given determinant depends entirely on the experimental conditions used to elicit T cells and gauge their numbers or activity. Although immunodominance plays a role in TCD8+ responses to tumor and histocompatibility antigens, the determinants recognized by these responses

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

IMMUNODOMINANCE IN TCD8+ RESPONSES

53

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

are already severely limited by self-tolerance. Therefore, we focus on TCD8+ responses to utterly foreign antigens, in most cases viruses, because only a few industrious souls have ventured to the far more antigenically complex bacteria and parasites that, when present intracellularly, often induce TCD8+ responses. As most of the recent advances in understanding immunodominance have been in the area of antigen presentation, the bulk of the review deals with these findings: It is important to emphasize right from the beginning, however, that TCD8+ repertoire and regulation play large, if not as well explored, roles in immunodominance.

Antigen Processing and Presentation in a Nutshell For a peptide to be immunogenic, it must do the following: 1. Be generated by “afferent” APCs from its precursor polypeptide and delivered to peptide-receptive class I molecules. (Afferent APCs trigger quiescent TCD8+ activation and proliferation. Under many circumstances, this task is accomplished by bone marrow–derived cells dedicated to the task, i.e. dendritic cells and macrophage/monocytes, referred collectively to as “professional” APCs). 2. Bind with sufficient affinity to class I molecules to produce enough cellsurface peptide–class I complexes to activate na¨ıve TCD8+. 3. Produce a complex with class I molecules on afferent APCs that is capable of triggering the activation and proliferation of a TCD8+ with a complementary T cell receptor (TCR). The interplay of these interdependent factors determines the strength of the immune response to a peptide; deficiencies in one area can be offset by gains in the other. This latter point is critical because it precludes identifying a single factor that is “responsible” for immunodominance. Moreover, for the same peptide to be biologically relevant, similar criteria must be met in the recognition of “efferent” APCs by activated TCD8+. (Efferent APCs trigger TCD8+ effector functions and are the raison d’ˆetre of the TCD8+). Before discussing the relative contributions of these factors to immunodominance, it is necessary to recapitulate current understanding of antigen processing and presentation (to present the maximal amount of material in the space allotted, we refer to only those original research papers that directly impact immunodominance; for ancillary information we refer readers to recent reviews). NATURE OF CLASS I–ASSOCIATED PEPTIDE LIGANDS The nature of class I– associated peptide ligands has been reviewed by several authors (8–10). The heart of major histocompatibility complex restriction is the interaction of the TCR with the peptide binding region of class I molecules. The free energy of

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

54

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

the interaction derives from contacts between the TCR with the α-helices of class I molecules and those residues in the bound peptide oriented away from the class I molecule. Peptide binding to class I molecules is principally due to two types of interactions. First, the amino and carboxy peptide termini interact with groove residues highly conserved between different class I allomorphs. Extrusion of the peptide past these residues interferes with this interaction, which accounts for the observations that 90% or more of peptides recovered from class I molecules are between 8 and 11 residues in length, and that synthetic peptides of this length nearly always bind to class I molecules with the highest affinity and are optimally antigenic. For some allomorphs, upward of 70% of the peptide ligands are of uniform length (most often nine residues), and for the other allomorphs, ∼80% of the peptides can be accounted for by including an additional residue (e.g. both 9mers and 10mers). Second, the antigen binding groove has two (or less commonly three) pockets that display a marked preference for one to five (most often one or two) of the 20 possible amino side chains. One of these pockets always accommodates the COOH terminus of peptide; the residues accommodated by the other(s) varies, depending on the allomorph, but are nearly always the second, third, or fifth residue from the amino terminus. The residues that comprise the pockets are highly variable between allomorphs. This, with a less important but still significant contribution from other variable residues in the binding pocket, results in each allomorph binding a unique set of peptides. A given peptide may bind to more than one allomorph; the odds of this happening are proportional to the degree of similarity between the binding grooves of the allomorphs. The influence of the pockets in peptide binding has an extremely important practical application: It enables the reasonably accurate prediction of peptides that may bind to a given class I allomorph based on the presence of the appropriate dominant anchor residues. As the number of known ligands for class I molecules grows, the more subtle effects of nonanchor residues on binding, and the cooperative (and noncooperative) effects of peptide residues on each other, can be computed by increasingly accurate algorithms that predict binding affinities. As described below, this has led to the ability to rapidly and reasonably inexpensively identify determinants present in proteins known to be recognized by TCD8+. GENERATION OF CLASS I–ASSOCIATED PEPTIDE LIGANDS As above, the generation of class I–associated peptide ligands has been reviewed by several authors (11–14). Most antigenic peptides presented by nonprofessional APCs (the exclusion of professional APCs is explained below) are derived from a cytosolic pool of proteins biosynthesized by the cells (endogenous antigens). The mechanism of targeting proteins to the cytosolic proteases that initiate

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

55

the production of antigenic peptides is largely undefined. It is generally the case that increasing protein degradation enhances antigenic peptide production, but most antigenic peptides originate from gene products that exhibit very low rates of degradation. To what extent peptides are derived from native proteins versus defective forms that never achieve a native state remains in question. The major cytosolic protease responsible for the production of antigenic peptides is the proteasome, but other cytosolic proteases probably contribute to antigen processing. Cytosolic peptides are delivered to the endoplasmic reticulum (ER) by the transporter associated with antigen processing (TAP). To be efficiently transported, peptides must be between 8 and 16 residues long and have the proper COOH-terminal residue. Mouse TAP prefers a hydrophobic residue, whereas human TAP prefers a hydrophobic or positively charged residue. These preferences match those exhibited by mouse and human class I molecules for COOH-terminal residues, which suggests that class I–binding peptides can either be produced in final form in the cytosol or possess amino terminal extensions of up to eight residues. In the latter case, trimming of NH2-terminal extensions would be needed. It has been demonstrated that TAP-deficient cells trim NH2-terminal residues from ER-targeted peptides, but whether TAP-transported peptides are similarly trimmed remains to be established. Peptides can associate with class I molecules in the ER in at least two distinct ways. First, peptides can bind to class I molecules associated with TAP. Class I molecules are recruited to TAP by binding to tapasin, a molecular chaperone apparently devoted to class I biosynthesis. The simple idea is that binding of class I molecules to TAP enhances the effective concentration of the peptide, thereby favoring loading. There may be additional complications, as tapasin has been reported to bind TAP-transported peptides (15). Second, as demonstrated by the ability of TAP-deficient cells to present peptides targeted to the ER by signal sequences, peptides can bind to class I molecules that are not bound to TAP. This latter process may also involve tapasin, which binds class I molecules prior to their association with TAP. Tapasin is not, however, required for class I assembly because tapasin-deficient cells demonstrate only a variable degree of impaired assembly of class I molecules, ranging from ∼20% to undetectable depending on the allomorph. One or both of these peptide-loading pathways may involve the participation of general-purpose ER chaperones because TAP-transported peptides can be recovered from numerous ER chaperones. Cytosolic chaperones also bind antigenic peptides. Although there is no experimental evidence, molecular chaperones could potentially play a role in immunodominance, either by virtue of their specificity for peptides or by their ability to properly transfer the peptide

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

56

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

to a relevant acceptor (16). A negative role can also be envisaged, if peptides are trapped in nonproductive association with chaperones. Class I molecules with bound peptides are stable on the cell surface for many hours. The loss of peptide destabilizes the molecule, which denatures with a half time of ∼15 min at 37◦ C. During this period, class I molecules can bind exogenous peptides. Such peptide-receptive molecules are stable for prolonged periods at 27C◦ or below. The existence of cell-surface peptide–receptive class I molecules has important practical applications. First, it enables sensitization of target cells by synthetic peptides (and probably accounts for the immunogenicity of synthetic peptides as well). Second, 37◦ C-induced dissociation of surface class I molecules (detected cytofluorographically by the loss of class I–specific mAb binding) accumulated by incubating cells at 27◦ C serves as the basis for the melting assay, a simple, highly informative method of determining peptide affinity for class I molecules. TCD8+ ACTIVATION The immunogenicity of a peptide-class I complex depends on the presence of responsive T cells with a complementary TCR. For this to occur, the TCR repertoire must be capable of generating an appropriate receptor from the pool of variable Vα and Vβ genes with their corresponding D and J genes and the amino acids that can be added at the V-D-J junctions. T cells bearing a complementary receptor must then pass the thymic Goldilocks test, binding self class I molecule-peptide complexes with just the right affinity to enable positive selection and disable negative selection. The TCR must also avoid binding to complexes with self peptides in the periphery that result in deletion or anergy (although the latter could possibly be overcome during the course of an immune response and its attendant inflammation). The triggering of TCD8+ activation by binding to peptide-class I complexes on the surface of an appropriate APC depends on the affinity of the TCR for the complex and the abundance of the peptide-class I complex, in principle according to the law of mass action (17, 18). TCD8+ activation is also greatly influenced by the interactions of accessory molecules (some expressed predominantly by professional APCs) that increase the avidity of cell-cell interaction and contribute in complex ways to signaling events in both cells. It is uncertain whether na¨ıve TCD8+ clones (or members of a single clone, for that matter) behave uniformly regarding the amount of TCR ligation required for activation. If there is clonal variation (almost to be expected), then two clones expressing TCRs with identical affinities for their respective peptide-class I complexes will require different numbers of peptide-class I complexes for activation. Thus, lacking precise immunochemical information regarding the affinity of isolated TCRs for peptide-class I complexes (even this is subject to vagaries regarding the affinity of soluble, isolated TCR versus TCR in its natural state in a

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

57

membrane teeming with other proteins, lipids, and saccharides), it is risky to rank relative affinities of the TCR–class I interactions based solely on the number of complexes required for stimulation. Intimately related to the sensitivity of T cell clones is the phenomenon of peptide antagonism, in which peptide–class I complexes of one type block the agonistic effects of the nominal antigen (19, 20). To date, antagonism has been observed using nonself peptides, but given the existence of positive selection, it is certainly possible that for some TCRs, self peptides provide antagonistic signals (indeed, this may play a physiological role in maintaining tolerance to peripheral antigens). Of particular relevance to the preceding paragraph, the presence of antagonistic self peptides can increase the number of agonist complexes needed for stimulation, leading to a false impression of low TCR affinity. Activation of na¨ıve TCD8+ results in the generation of primary armed effector cells and memory cells. As described below, there is often not a simple, direct relationship in the primary and secondary TCD8+ responses to different determinants in complex antigen. The extent to which this reflects the independent generation of secondary and primary TCD8+ as opposed to alterations in activity during the differentiation of primary effector cells into memory cells is just now being sorted out. NATURE OF THE APC A crucial question for immunodominance is the nature of the afferent APC after infection with different agents. Current dogma dictates that na¨ıve TCD8+ require multiple signaling events for activation: one transmitted through the TCR-CD3-CD8 complex as a result of binding to peptide-class I complexes, the other(s) transmitted by costimulatory TCD8+ cell-surface proteins, most often CD28. As the activating ligand for CD28, B7-1 (CD82) is expressed predominantly by professional APCs, it is thought that these cells (particularly dendritic cells) do most of the heavy lifting in stimulating na¨ıve TCD8+. Expression of costimulatory molecules can be induced in numerous cell types by the types of cytokines secreted early in the inflammatory process, however, raising the possibility of nonprofessional APC-induced TCD8+ activation in some immune responses. The “quality” of the APC involved in TCD8+ activation may be related to the variation in the requirement for TCD4+ in generating TCD8+ to different viruses. Possibly, viruses that utilize nonoptimal APCs (due to the effects of the virus on APC function or inability to infect professional APCs) are those that require more TCD4+-mediated help. A related, equally important question is the nature of antigens presented to na¨ıve T cells. In most situations in which TCD8+ are stimulated by virusinfected cells, the antigen is presumably endogenously synthesized by the APC. There must be instances, however, in which viruses of limited host cell range

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

58

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

are incapable of infecting effective APCs. In these circumstances, professional APCs must present exogenous antigens. This was initially discovered as the cross-priming phenomenon, viz mice mounted self class I–restricted TCD8+ responses to minor histocompatibility antigens presented by cells lacking the appropriate self-class I molecules (21). It has been repeatedly demonstrated that such cross-priming is dependent on histocompatible bone marrow–derived cells, presumably the professional APCs. Precisely what form of exogenous antigen is processed by these professional APCs is uncertain. The major issue is whether the APCs acquire antigen in a form requiring major proteolysis (e.g. a full-length protein) or whether a preprocessed (or nearly so) peptide is provided. The latter is suggested by findings that na¨ıve TCD8+ can be activated by immunization of mice with molecular chaperones containing viral or tumor peptides (22). In situations in which proteolysis is required, it is uncertain whether proteins are delivered to the cytosol or whether the peptides are generated by endosomal proteases. Although there is evidence that supports the loading of class I molecules from endosomally generated peptides (23), the biological relevance of these findings awaits confirmation. The importance to immunodominance of understanding how peptides are generated from exogenous antigens stems from the requirement that under most circumstances, efferent APCs present peptides derived from endogenous antigens. If endosome-generated peptides provide a significant source of ligands for stimulating na¨ıve TCD8+, then a safe prediction is that TCD8+ specific for a subset of these peptides will be of little use in the immune response because the rules for generating peptides in the cytosol and endosome cannot be identical. Conversely, the absence of such TCD8+ would suggest that the endosome is not a physiological source of class I ligands. ANTIGENIC SUBTERFUGE In response to ability of TCD8+ to interfere with their propagation, replicating antigens such as tumors or viruses have the potential to evolve mechanisms to interfere with TCD8+ activation or effector functions. To cite some known examples, tumor cells may secrete cytokines that interfere with TCD8+ activation or function, and viruses can produce proteins that interfere with peptide generation or class I biosynthesis. If the interference is selective for a subset of TCD8+ clones, antigenic peptides, or class I molecules, it can contribute to immunodominance in a given system. GETTING TECHNICAL Understanding natural phenomena depends entirely on the means used to observe and measure the phenomena. Until relatively recently, TCD8+ responses were assessed almost exclusively by their capacity to lyse target cells (as measured by 51Cr release), using TCD8+ populations ex vivo

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

59

(i.e. assessing lytic activity of cells without in vitro culturing) or following short-term stimulation in bulk or, more quantitatively, under limiting dilution conditions. In the past few years, three novel methods of measuring TCD8+ responses were introduced that will greatly expand understanding of TCD8+ responses: ELISPOT, which quantitates individual armed effector TCD8+ based on cytokine release; intracellular cytokine staining, which cytofluorographically identifies and quantitates armed effector TCD8; and tetrameric peptide–class I– avidin/streptavidin complexes, which cytofluorographically identify and quantitate resting or active TCD8+ bearing TCRs specific for a given peptide–class I complex. A number of recent studies utilizing these methods have revealed that the 51 Cr release assay grossly underestimates the numbers of TCD8+ that respond to viral or bacterial antigens (24). These new data will not, however, negate immunodominance-related findings made using the 51Cr release assay, which— limited as they may be to a subset of responding TCD8+—are probably reasonably representative of the entire response. Of greater concern to interpreting immunodominance-related findings are methodological differences in stimulating TCD8+ whose activation is assessed by 51Cr release assay. There is a particularly wide gulf in the methods used for studying mouse and human TCD8+. Because of ethical/medical constraints and the 1000-fold difference in body mass, human TCD8+ are almost always derived from peripheral blood lymphocytes (PBLs), whereas mouse TCD8+ are derived from lymphatic organs (routinely spleen, occasionally lymph nodes). Additionally, exposure of mice to antigens can be rigorously controlled whereas the antigenic history of humans is always subject to some uncertainty. It is important to recognize that even within a single system, seemingly minor variations in methods used to induce TCD8+ can produce major variations in apparent immunodominance. This is particularly true in the numerous studies in which memory TCD8+ (both mouse and human) are expanded in vitro prior to assay. If the conditions are suboptimal (which to some extent they will always be), there is a good chance that TCD8+ specific for “weaker” determinants will not be activated sufficiently to drive cell division to the point of distinguishing lytic activity from background values. Although it remains important to determine how the TCD8+ against the weaker determinants differ from TCD8+ specific for the immunodominant determinant (IDD) (whether simply in quantity or quality), it is never safe to conclude that the failure to detect a response against a given determinant means that such a response is completely absent. Extreme caution must be exercised in the assignation of cryptic to a determinant, as this is strictly dependent on the assay conditions utilized. For example, even an IDD may appear to be cryptic if virus-infected APCs fail to express sufficient levels of peptide–class I complexes because of low levels of viral

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

60

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

protein synthesis, or if the TCD8+ used for detection require an excessive amount of peptide–class I complexes. The solution to this problem is obvious but not simple, as it entails quantitation of peptide–class I complexes expressed on the cell surface. This is an arduous biochemical task, but the development of mAbs specific for peptide–class I complexes offers some hope for the future (25, 26). Finally, a cautionary note regarding the use of synthetic peptides. Only in few instances have the structures of naturally processed determinants been definitively established by structural methods, i.e. mass spectroscopy. More frequently, although still relatively uncommonly, the naturally processed peptide is shown to co-elute with a synthetic peptide in high-pressure liquid chromatography (HPLC). Most commonly, the identity of the natural peptide is inferred by identifying a synthetic peptide that activates TCD8+ optimally in vitro. In the latter two cases, it must always be considered that the natural peptide is not identical to the “optimal” synthetic peptide. The natural peptide may possess an extension or may be posttranslationally modified. Cys-containing peptides may cause considerable difficulties because Cys can dimerize the peptide or react with either sulfhydryl groups in serum or cellular proteins or with heavy metals (27). The bottom line is that one may be led astray by qualitative differences between optimal peptides and the genuine article that result in very large errors in quantitation of peptide–class I complexes expressed on APCs.

IMMUNODOMINANCE: CONTRIBUTION OF ANTIGEN PRESENTATION Affinity for Class I Molecules: The Highest Hurdle The discovery that cellular peptides recovered from a given class I allomorph exhibit highly conserved residues in two or three positions was a major advance in the study of immunodominance because it enabled the identification of upward of ∼80% of potentially antigenic peptides in a given antigen (9). Importantly, antigenic peptides identified independently of such motifs exhibit the same bias as pooled cellular peptides bound to the same allomorph, affirming the validity of using the motifs for prediction of antigenic determinants. The predictive value of peptide motifs and the role of peptide affinity for class I molecules in immunodominance have been most thoroughly examined by Sette and colleagues (28, 29). Initially, they synthesized a series of viral peptides conforming to the HLA-A∗ 0201 motif and correlated the affinity of the peptides for A∗ 0201 with their immunogenicity in mice expressing a chimeric Kb transgene in which the α1α2 domains are replaced by those of HLA-A∗ 0201 [enabling partial (30, 31) CD8 interaction with A∗ 0201 via the α3 domain].

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

61

Affinity was determined by the ability of the peptide to compete with the binding of a radiolabeled standard peptide to purified A∗ 0201 molecules in solution. Immunogenicity was assessed by the ability of the peptide to stimulate in vitro splenocytes derived from mice immunized with the same peptide (in adjuvant with a peptide that induces a TCD4+ response). Peptide-specific TCD8+ were induced by five out of five of the highest-affinity peptides (K d < 50 nM), three out of five of the intermediate affinity peptides (K d 50–500 nM), and none out of 13 of the lowest-affinity peptides (K d > 500 nM). Measuring the affinities of 11 defined A∗ 0201-restricted viral IDDs and 30 sequenced cellular peptides recovered from HLA-A2 indicated that 90% were high affinity, 7% intermediate affinity, and 4% low affinity, affirming similar findings by Parker and colleagues (32). Moving to TCD8+ responses in human PBLs, Team Sette examined the ability of 91 hepatitis B virus (HBV)-derived, A∗ 0201 conforming nonamers (affinity breakdown: 22 high, 21 intermediate, 48 low) to restimulate PBLs derived from A2-positive individuals acutely infected with HBV (29). Responses were induced by 45% of the high-affinity peptides, 14% of the intermediate-affinity peptides and 6% of the low-affinity peptides. A similar analysis was performed using synthetic peptides (from mostly viral sources) conforming to the HLAA11 binding motif (33). Of the 45 motif-containing peptides synthesized from viral and cellular peptides, 41 bound with intermediate or high affinity. This is a much higher percentage than was observed with A2 motif peptides, making the point that the predictive values of available peptide motifs can vary considerably between allomorphs. All the known viral IDDs bound to A11 with high affinity. When the immunogenicity of motif-containing peptides was examined in primary human PBL cultures (i.e. from virus seronegative donors), responses were elicited by 21 of 28 peptides with high affinity, 7 of 13 with intermediate affinity, and 1 of 4 with low affinity. In collaboration with the Ahmed laboratory, Sette and coworkers applied this approach to mouse TCD8+ responses to lymphocytic choriomeningitis virus (LCMV) (Kd-, Dd-, Kb-, and Db-restricted) (34–36), and influenza virus (IV) (Kb-and Db-restricted) (37). Of four previously defined LCMV IDDs, two bound to restricting molecules with high affinity and two with intermediate affinity. Searching for new determinants in two LCMV proteins using defined motifs, 2.2% of the potential number of peptides conformed to the peptide binding motif for any given allomorph. Approximately one quarter of these peptides bound with high or intermediate affinity to their respective allomorph (0.5% of all possible peptides in the two proteins). In contrast to the defined IDDs, none of the 21 intermediate or high-affinity binding peptides consistently sensitized target cells for ex vivo lysis by TCD8+ derived from LCMV-infected mice. Six peptides (all of intermediate affinity) were subdominant determinants (SDDs),

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

62

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

however, as they induced in vitro secondary responses in splenocytes from infected animals. SDD-specific TCD8+ were detected by limiting dilution assay (LDA) at ∼2%–5% the frequency of IDD-specific TCD8+. The even more extensive analysis with IV studied 47 Db and 151 Kb motifbearing peptides (1% and 3% of the potential number of IV-encoded peptides), of which 7 and 16 bound with high or intermediate affinities (0.15% and 0.35% of IV-encoded peptides, respectively). Following peptide immunization and restimulation, 12 of 14 high-affinity peptides and 4 of 9 intermediate-affinity peptides induced peptide-specific responses. Of the 16 peptide-specific TCD8+ populations generated, only two were capable of lysing IV-infected cells. Taking a crucial experimental step forward, TCD8+ specific for 13 of the peptides were tested for their ability to lyse cells exposed to decreasing amounts of peptide. Among the 10 high-affinity peptides tested, there was a ∼10,000-fold difference in the amount of peptide required to achieve an arbitrary level of lysis. TCD8+ raised to the previously defined IDD required the least amount of peptide, but TCD8+ specific for two other high-affinity determinants demonstrated a similar sensitivity (one of these was able to lyse IV-infected cells). Factoring in TCD8+ sensitivity, the two newly defined SDDs appeared to be expressed at levels at or above the IDD on IV-infected cells. Only one of these peptides could induce secondary in vitro responses in splenocytes from IVprimed animals, pointing to a possible difference between in vitro and in vivo presentation of the determinant. We have made similar findings regarding the Kd-restricted response to IV (38). Of the 27 nonameric peptides that conformed to the Kd binding motif, 10 (including the two known IDD) bound to Kd, as detected by the melting assay. The IDDs were not the most avid binders, ranking second and even fifth for the most dominant determinant [but both are of high affinity according to the classification of Sette et al (29)]. Of the eight novel peptides with low to high affinity, only the three high-affinity peptides stimulated TCD8+ from IV-primed mice in vitro. Genes encoding each of the 10 peptides were inserted into vaccinia virus (VV) and expressed as ER-targeted peptides. When TAP-deficient cells were infected with the rVVs, only rVVs encoding high- or intermediate-affinity peptides enhanced Kd cell-surface expression. This provides a direct correlation between endogenous and exogenous peptide binding to class I molecules and offers an explanation for the sharp cutoff between immunogenicity of intermediate- and low-affinity peptides. Immunization with the rVVs followed by restimulation with homologous peptide in vitro revealed that only the six most avid binders (including the two IDDs) were able to prime for peptide-specific TCD8+ responses. TCD8+ raised against the four novel determinants were able to lyse IV-infected cells, but at lower levels than IDD-specific TCD8+, demonstrating that antigen processing from viral gene products is more

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

63

limiting for the SDDs. Similarly, when IV-infected cells were used to stimulate splenocytes from IV- or rVV-infected mice, responses to the SDDs were at low levels (rVV-primed mice) or undetected (IV-primed mice). Altogether, these findings provide crucial insight into the relative contributions of the factors that contribute to immunodominance. First and foremost is the role of peptide binding to class I molecules. Based on the body of work by Sette and colleagues, it appears that 90+% of peptides recognized by TCD8+ bind to their respective class I molecules with an affinity constant of 500 nM or better. What are the odds of a given 8- to 11-residue stretch of a protein binding to any given allomorph with this affinity? Using the published peptide binding motifs for 17 human and 6 mouse class I allomorphs, the odds of a peptide of a given length randomly possessing anchor residues for a given allomorph can be calculated to be ∼1/132 on average for the different allomorphs. This calculation is based on the overall frequency of individual residues in proteins and assumes a random distribution of amino acids in anchor positions. The simple motifs are, of course, imperfect predictors of peptide binding. Accounting for the flexibility that class I molecules demonstrate in accommodating extended peptides would increase the odds to ∼1/100 and, accounting for those peptides that do not possess the canonical dominant anchor motif, to ∼1/70 (since approximately one third of defined IDDs do not fit their respective motif). The results of Sette et al suggest that approximately one third of motif-conforming peptides bind to class I molecules with a Kd of 500 nM or better, making the odds ∼1/200 for the binding of random peptide binding to a given class I allomorph with an immunologically significant affinity. This estimate is supported by studies that have examined the binding of randomly generated peptides to class I molecules (39–41). Thus, the possession of the proper sequence accounts (literally) for 99.5% for the immunodominance phenomenon. Of the 0.5% of peptides that bind to class I molecules with biologically significant affinity, evidence suggests that approximately half or more of these can induce TCD8+ responses as synthetic peptides (or virus-encoded minigenes), and that of these, approximately four fifths are expressed in quantities that relegate them to subdominant or cryptic status, because of low sensitivity of the TCD8+, poor antigen processing, or both. Multiplying these odds (1/200 × 1/2 × 1/5) results in the estimate that ∼1/2000 peptides in foreign antigens achieve IDD status in association with a given class I allomorph, with perhaps twice as many SDDs—at least in mice, where much of the evidence has been accumulated [or more accurately, in two inbred mouse strains maintained under germ-free (more or less) conditions]. The evidence for the degree of immunodominance in human anti-viral responses varies among viruses. The number of A∗ 0201-restricted determinants defined in responses to HBV, human immunodeficiency virus (HIV), and

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

64

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

hepatitis C virus is clearly greater than 1/2000. This discrepancy may be related to the chronic nature of these infections, or to the outbred nature of the population, because response to determinants can vary greatly among individuals (see below). Moreover, for many of these determinants, their IDD versus SDD status has not been established. In the case of HIV, where a very large number of determinants have been identified, evidence that TCD8+ exert strong selective pressure for determinant loss variants argues strongly for immunodominance, at least in some individuals (42), as does the oligoclonal expansion of TCD8+ (43). In the other relatively well-characterized human TCD8+ anti-viral responses [Epstein-Barr virus (EBV) (44) and cytomegalovirus (CMV) (45)], the frequency of IDDs seems similar to that observed in mice. There are a number of other important points to made from these findings: 1. TCD8+ responses to viruses encompass more SDDs than has generally been appreciated. This is potentially of great practical importance because TCD8+ specific for viral SDDs can afford protection to subsequent infection (36, 46–49) and enhance protection afforded by IDD-specific TCD8+ (48). TCD8+ specific for tumor SDDs can prevent tumor growth when induced by tumor cell (50, 51) or synthetic peptide (51) immunization. 2. The striking correlation between peptide affinity (measured by the binding of optimally sized synthetic peptides to either purified soluble class I molecules or class I molecules on the surface of TAP-deficient cells) and the immunogenicity of the peptide in the context of its natural antigen demonstrates that the association of TAP-transported peptides with class I molecules in the chaperone-rich ER must largely recapitulate the hierarchy in binding as measured in the absence of any facilitating factors. “Largely” is used advisedly, as the immunogenic peptides that score low in affinity measurements may actually bind with higher affinity in the ER. Two lines of evidence support this conclusion. First, it has been reported that immunogenic peptides of low or intermediate measured affinity are more likely to exhibit Koff values characteristic of high-affinity peptides than are nonimmunogenic peptides of similar affinity (52). This implies that the decreased affinity of the peptides measured reflects a diminished Kon value. It is not difficult to imagine mechanisms operative in the ER that could enhance the Kon values of a subset of peptides. Second, and more directly, there is at least one example in the literature of a peptide that is more antigenic as a biosynthesized minigene product than as a synthetic peptide (53). 3. The generation of natural anti-viral TCD8+ responses to determinants that score as cryptic on virus-infected APCs in vitro appears to be an infrequent event (see 54, 55 for a possible exception). This strongly implies that afferent

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

IMMUNODOMINANCE IN TCD8+ RESPONSES

65

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

APCs in vivo present viral determinants in a manner quantitatively similar to virus-infected APCs used in vitro. As mentioned above, this argues against the endosome as a significant source of immunogenic peptides in vivo. The extent to which this applies to tumor or minor histocompatibility antigens (or infections with other viruses) remains to be established. 4. Although peptide binding to class I molecules is the major factor in immunodominance, IDDs are frequently not simply the most avid binding peptides encoded by the virus. In some cases, IDD-specific TCD8+ are clearly very sensitive, requiring low levels of peptide–class I complexes for target cell lysis. In other cases, however, IDD-specific TCD8+ may require more complexes than SDD-specific TCD8. Only in a fraction of the latter cases is the determinant clearly present in sub-limiting amounts. In the subsequent sections, we discuss the three factors that combine to cause the poor immunogenicity of non-IDD class I–binding determinants: production of insufficient amounts of peptide–class I complexes, low numbers or sensitivity of TCD8+, and interference by IDD-specific TCD8+. It is important to recognize that these first two factors can be considered only in combination. Thus, for a peptide–class I complex expressed at a given level by APCs, this level may or may not be limiting, depending on the number of complexes required by TCD8+ that recognize the complex. Given the ability of T cells to recognize vanishingly small numbers of peptide–class I complexes [there is even a description of a T cell that recognizes cells calculated to express a single complex (56)], antigen processing can only safely be said to be absolutely limiting in cases in which APCs cannot produce a single determinant (also the definition of true crypticity).

Generation of Peptide–Class I Complexes QUANTITY OR QUALITY? Assessing the contribution of antigen processing to immunodominance requires quantitation of the levels of peptide–class I complexes expressed on the surface of APCs. Ideally, the APC would be the cell that actually presents the antigen to primary TCD8+ in vivo. Even if the identity of this cell were established (it is not), obtaining sufficient quantities of representative cells for analysis would be a considerable technical achievement. In practice, studies have been limited to tissue culture cells infected with viruses or bacteria. Quantitation of peptide–class I complexes can be performed in three ways. The first two methods described depend heavily on the assumption that the naturally processed peptide is identical to the synthetic peptide thought to represent the determinant. The simplest method is to determine the amount of synthetic peptides required to obtain a similar degree of lysis obtained with infected cells expressing levels

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

66

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

of complexes that do not saturate the TCD8+ used. If different peptides are restricted by the same class I molecule and bind with a similar affinity, the relative ratios of complexes can be estimated from peptide titration curves. Better, but more difficult, real numbers of complexes can be estimated by quantitating peptide binding; this also enables comparisons between peptides of different affinities or peptides that bind different allomorphs. More rigorously, peptides are HPLC purified from acid extracts of cells or isolated class I molecules. The amount of acid-soluble peptide present in HPLC fractions is determined by target cell sensitization using a synthetic peptide standard curve. This method is both arduous and expensive when dealing with peptides from infectious organisms, and it also suffers from uncertainties regarding the efficiency of peptide recovery and the presence of co-eluting peptides that compete for binding to class I molecules. It also cannot distinguish whether peptides were derived from intracellular or cell-surface class I molecules. The most elegant method for quantitation is the use of T-AGs, mAbs specific for individual peptide–class I complexes (25, 26). This method is both simple and precise, but it suffers from relatively low sensitivity (at least several hundred complexes are needed for detection) and requires the production of the T-AG, which to date has been a hit or (mostly) miss proposition. One possible solution to the difficulty in producing T-AGs is the use of soluble TCRs. Although the affinities of monovalent TCRs are usually too low for use in standard sandwich assays, the avidities of TCRs can be increased to useful levels by chemical or genetic cross-linking (57). Given the availability of a TCD8+ clone for a given determinant, this strategy offers a good chance of obtaining a reagent suitable for quantitating the complex on the APC surface. Only a limited number of studies have examined the number of foreign peptide–class I complexes generated by APCs. Rammensee and colleagues first showed that IV Kd- and Db-restricted nucleoprotein (NP) IDDs NP147–155 and NP366–374 were present in HPLC fractions of acid extracts at ∼300 copies per IV-infected cell (6). Using this method, we found that only ∼30 copies of NP147–155 are recovered per cell following infection with a rVV-expressing IV NP (58). The same cells expressed 1800 copies of the Kk-restricted IDD NP50–57. Following infection of cells with a rVV-encoding NP1–168 [this 168residue fragment is degraded with a t1/2 of 30 min; full-length NP (498 residues) is essentially stable], 105 copies of NP147–155 and 9300 copies of NP50–57 were recovered. Expressing either of the determinants as VV-encoded cytosolic or ERtargeted minigene products resulted in the recovery of an astounding ∼55,000 complexes per cell. The minigene-enhanced generation of complexes was associated with greatly enhanced primary anti-peptide responses, as assessed by ex vivo cytotoxic activity (particularly for NP147–155), yet only a slight increase

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

IMMUNODOMINANCE IN TCD8+ RESPONSES

67

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

in the generation of memory TCD8+ (59). Even the threefold enhancement of NP147–155 generation by NP1–168 was associated with an enhanced primary TCD8+ response. The use of a T-AG specific for the Kb-Ova257–264 complex revealed that ∼3500 complexes were expressed on the surface of cells infected with a rVV-expressing chicken ovalbumin (OVA) whereas rVVs expressing cytosolic or ER-targeted minigene products expressed more than 65,000 complexes per cell (26). The abilities of these rVVs to elicit primary and secondary TCD8+ responses were similar (59). These findings lead to several conclusions: 1. The enormous increase in peptide–class I complex formation obtained with cytosolic minigene products relative to full-length proteins demonstrates that the liberation of antigenic peptides from full-length gene products is probably always a limiting factor in the generation of peptide–class I complexes. 2. The extent to which this limits TCD8+ responses depends on exactly how inefficient peptide liberation is and on how many complexes are required to obtain maximal responses. For NP147–155, 30 complexes are limiting for primary TCD8+ responses to VV-NP, and immunogenicity is enhanced by even a threefold increase in complex formation, with further gains coming from an additional 50-fold increase. In the case of Ova257–264, 3000 complexes are sufficient to obtain maximal responses, whereas for NP50–57, 9300 complexes/cell are insufficient, and primary responses are enhanced by a sixfold increase in complex number. Similarly, it was found that mouse primary and secondary TCD8+ responses to rVV- or plasmid DNA-encoded HIV proteins were increased if peptide generation was enhanced by producing rapidly degraded forms of the protein (60). 3. In responding to the same antigen in different contexts, primary TCD8+ responses need not parallel memory TCD8+ response in magnitude. Thus, there was much less difference in the abilities of rVVs encoding NP and NP147–155 minigene product to prime for memory versus primary TCD8+ responses. This also demonstrates the conditional nature of immunodominance: Expressed by IV or as a VV-encoded minigene product, NP147–155 is a IDD, whereas when expressed as a VV-encoded full-length protein it is a SDD (because primary in vivo responses are low to undetectable by ex vivo 51Crrelease assays). The great variation in the abundance of IDDs appears to be common. Using a leukemic T cell line transfected with the HIV genome, the A∗ 0201-restricted viral peptides gag77–85 (group antigen) and RT476–484 (reverse transcriptase)

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

68

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

were present, respectively, at ∼400 and 12 copies per cell (61). Unlike the situation in inbred mice, where there is little variation in which peptides are immunodominant, individuals usually respond to one or the other of these peptides. In any event, in some individuals, the less prevalent determinant is preferred over the more prevalent (42). Similarly, in the course of a remarkable series of experiments characterizing Kd-restricted responses to the intracellular bacterium Listeria monocytogenes, Pamer and colleagues (62–66) have shown that the IDD is the leastabundant determinant. By Elispot analysis, primary TCD8+ responded to three peptides—LLO91–99 (listeriolysin O), p60217–225, and p60449–457—at ratios of 20:10:1 (62). This ratio remained unchanged over the subsequent 6 weeks, providing an example in which memory TCD8+ responses parallel primary responses. The TCD8+ response to each peptide was also measured using soluble strepavidin-tetramerized, peptide–class I complexes (63). This detected specific TCD8+ at levels similar to those of the Elispot analysis and revealed that at the peak of primary responses, 1.4% of all TCD8+ recognized LLO91–99, with 5-fold and 20-fold fewer cells, respectively, recognizing p60217–225 and p60449–457. During secondary in vivo responses, the number of cells responding to each determinant increased approximately 10-fold. Quantitation of peptides recovered from a macrophage cell line harboring endosomal bacteria revealed that these frequencies were actually inversely related to peptide abundance, with 700 LLO91–99, 2700 p60217–225, and 9000 p60449–457 determinants recovered per cell. By functional competition assay using Kd-restricted T cells specific for a fourth party peptide, the three peptides blocked recognition with a similar molar efficiency, which suggests that they bound to Kd with equal affinity. They were also equally effective at sensitizing target cells for lysis by their respective TCD8+. These last two findings suggest that similar amounts of the three complexes are required for triggering TCD8+ responses. When the stabilities of endogenously produced or synthetic peptide–induced complexes were examined (64), complexes containing either of the two dominant peptides were stable (t1/2 > 6 h) whereas the Kd-p60449–457 complex disappeared with a t1/2 of 1 h, which is consistent with the correlation of van der Burg et al (52) between immunodominance and complex stability. The great abundance of p60449–457 in the face of its rapid turnover is explained by its amazing efficiency of formation. Taking full advantage of unique features of the bacterial system to quantitate the turnover of precursor proteins delivered to the APC cytosol, the efficiency of peptide generation per degraded protein molecule was calculated to be 5%–10% for LLO91–99, 2.5%–3% for p60217–225, and 25%–30% for p60449–457 (65). The specificity of TAP may contribute to these figures because p60449–457 is a sixfold more efficient competitor than p60217–225 for TAP-mediated transport of a reporter peptide. The relationship between TCD8+ responses and abundance

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

69

of p60217–225-Kd complexes was examined by mutating p60216 to residues that modify peptide generation (66). Reducing peptide generation efficiency 10-fold prevented generation of both primary and secondary TCD8+ responses, whereas a twofold reduction or twofold enhancement had no effect on the magnitude of primary or secondary responses. In contrast to some of the viral systems described above, in which graded responses can be observed, this provides a more quantal example of TCD8+ responses to a determinant within a complex pathogen. In humans, A∗ 1101-restricted TCD8+ responses to EBV may also be influenced by the stability of peptide–class I complexes. Two EBV determinants are frequently recognized in association with A∗ 1101, EBNA3B416–424 (Epstein Barr nuclear antigen) and EBNA3B399–408. EBNA3B416–424-specific TCD8+ dominate both primary and secondary responses, being present at up to 20-fold higher frequencies in both primary (67) and secondary responses as measured by LDA (68). The peptides bind to surface class I molecules with similar affinities (determined by blocking lysis of cells sensitized for lysis with a third party peptide) and sensitize target cells with similar efficiency. This suggests that similar amounts of the two complexes are required for stimulation of their respective TCD8+. Quantitation of peptides from different EBV-transformed B cells revealed that the immunodominant peptide is present at 5- to 40-fold higher levels, depending on the cell line. This difference probably stems, at least in part, from the low stability of complexes formed with the SDD, whose t1/2 on the cell surface (assessed by biochemical recovery of detergent-solubilized complexes formed by viable cells incubated with peptide at 26◦ C) was measured to be at least threefold less than the dominant peptide. This effect may not be intrinsically related to the binding of the subdominant peptide to A∗ 1101 because with soluble class I molecules, the two complexes demonstrate similar stability at 37◦ C and similar resistance to acid treatment. This may mean that membrane and soluble class I molecules can exhibit selective differences in their binding to certain peptides. Alternatively, as cells are capable of internalizing peptide–class I complexes into endosomal compartments (where they are destroyed), it may mean that APCs can preferentially internalize class I molecules bearing certain peptides. Given the limited number of quantitative studies that have appeared, and the uncertainty regarding the relevance of in vitro APCs to in vivo APCs, it is not possible to reach firm conclusions regarding the relative abundance of IDDs, SDDs, and “cryptic determinants” (CD). These studies do, however, confirm estimates based on peptide titration curves that IDDs are not always the most abundantly expressed complexes and may, in fact, be expressed in very low numbers. Obviously, much more work is needed in this area, which given the increasing use of T-AGs and multivalent TCRs will surely be forthcoming.

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

70

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

FACTORS THAT AFFECT PEPTIDE GENERATION Limiting steps in antigen processing As discussed above, the liberation of a determinant from its fulllength gene product can greatly limit its immunogenicity. This does not necessarily mean that the determinant is liberated less efficiently than other more immunogenic determinants; only that peptide liberation is a limiting step in creating the number of complexes required to optimally activate na¨ıve TCD8+. Peptide liberation in the cytosol is, of course, just one step in the process of creating the trimolecular complex, which includes transport of determinant to the ER, where the peptide may be trimmed to a higher affinity form, and loading onto class I molecules (the last two steps could occur in reverse order). Because it is not yet possible to measure the rate of peptide liberation in the cytosol, to conclude that peptide liberation in the cytosol is the limiting step requires knowledge that the other steps in antigen processing occur at “normal” levels. Although it is possible to measure the efficiency of TAP-mediated transport using synthetic peptides in semi-intact cells, a potential drawback to this assay is its uncertain relevance to the situation in living cells, where peptide delivery to TAP may be facilitated by molecular chaperones and possibly even coupled to peptide generation. There is now evidence, however, that the efficiency of TAP-mediated transport of VV-encoded minigene products with short flanking residues correlates with the number of peptide–class I complexes generated (although the latter is inferred from indirect methods) (69). A more serious problem is that because peptide trimming can occur in the ER, the composition of the peptides transported by TAP that are generated from physiological antigens (as opposed to minigene products) is uncertain. Measuring the efficiency of the two remaining steps in the pathway, peptide trimming and assisted loading onto class I molecules, is an even more difficult problem. There is as yet only indirect evidence that TAP-transported peptides are trimmed in the ER, and the current evidence for facilitated loading amounts to little more than reasonable, if inspired, speculation. With the available technology it is, therefore, not possible to precisely separate the contribution of individual steps in antigen processing to the compromised immunogenicity of SDDs and CDs. There are a number of studies, however, that have laid the foundation for the future understanding of this question. The first (and still the most compelling) demonstration that the regions immediately flanking a determinant (flanking sequence) in a full-length protein can influence its immunogenicity is the work of Del Val et al (70). Inserting a murine CMV (MCMV) determinant into VV-encoded HBV core antigen, it was shown that flanking residues greatly influenced the in vitro presentation of the determinant (up to a 16-fold difference was detected in the amount of peptide recovered from acid extracts of infected cells) and its ability

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

71

to induce a protective TCD8+ response to a challenge with a dose of MCMV lethal to nonimmunized animals. Importantly, it was shown that this was not simply a consequence of increasing the overall degradation rate of chimeric protein. The effects of flanking sequences on immunogenicity have been most thoroughly characterized in Kb-restricted TCD8+ responses to OVA. The IDD Ova257–264 binds to Kb with high affinity (Kd ∼ 1 nM). TCD8+ responses to a SDD (Ova55–62) can be elicited if animals are immunized with cells osmotically loaded with amounts of OVA in excess of the minimal amount required to obtain responses to Ova257–264; in fact, the number of memory TCD8+ generated for the two determinants is similar under these conditions (71). Clones specific for each determinant were obtained that require similar amounts of the respective peptides for half-maximal activation. Because Ova55–62 binds to Kb with a Ka of ∼50 nM, this implies that the Ova55–62-specific clone is far more sensitive than the Ova257–264-specific clone. Despite this, 50-fold greater amounts of electroporated OVA were required to achieve a similar degree of stimulation, which implies that processing of OVA results in a ratio of Kb-Ova257–264 complexes to Kb-Ova55–62 complexes of ∼2500. Obviously, the lower affinity of Ova55–62 for Kb could contribute to its poor presentation. On the other hand, this affinity is well within the range observed for IDDs restricted by other class I molecules. The role of flanking sequences in the difference in efficiency of producing these two determinants was examined by measuring TCD8+ recognition of cells cytosolically loaded with 22-mer synthetic peptides composed of one of the determinants with natural flanking residues or flanking residues from the alternative peptide (72). For Ova257–264, substitution with the flanking residues of Ova55–62 decreased its processing efficiency more than twofold; the converse manipulation only slightly enhanced the presentation of Ova55–62. Most importantly, the efficiency of presentation correlated with the efficiency at which purified 20S proteasomes were able to liberate the determinant from the respective synthetic substrates. Ova55–62 was cleaved internally by proteasomes, and this could not be rectified by substitution with the Ova257–264 flanking sequences. Conversely, the Ova55–62 flanking sequences created cleavage sites within the Ova257–264 determinant. The idea that proteasomal destruction of potential determinants is a frequent contributor to immunodominance is supported by two additional studies. Tevethia and colleagues (73) examined a Db-restricted determinant from simian virus 40 tumor antigen (Tag), Tag489–497, that is nonimmunogenic in the context of Tag produced by simian virus 40 transformed cells or by VV-Tag. TCD8+ are easily elicited by a rVV encoding the ER-targeted peptide, but not a rVV encoding the cytosolic peptide. This correlates with the efficiency of

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

72

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

presentation in vitro by rVV-infected cells: Cells expressing the cytosolic peptide are not lysed. This is not due to inefficient transport of Tag489–497 by TAP, as inferred from its ability to block TAP-mediated transport of an indicator peptide in a biochemical assay using permeabilized cells. Rather, the problem appears to be proteasomal destruction because presentation of the cytosolic minigene product is enabled by treating cells with a proteasome inhibitor. The addition of flanking sequences to the cytosolic minigene product as simple as (Ala)2 at either end of the peptide, or inclusion in a full-length protein, also enhanced its presentation—presumably by enhancing the generation of a precursor that was not destroyed by proteasomes and could be transported by TAP. Ossendorp et al studied responses to the p15E574–581, the Kb-restricted IDD in the AKV/MCF type murine leukemia virus (MuLV) (74). The homologous protein in FMR type MuLV has six amino acid substitutions, one (Lys to Arg) at the NH2-terminus of the peptide and the other five located at least 10 residues from the peptide (it was assumed that these were too distant to affect peptide generation). The mutation within the determinant does not affect peptide affinity for Kb and does not affect recognition by TCD8+ raised to the AKV/MCF peptide. The two peptides are similarly immunogenic as synthetic peptides. Despite this, mice fail to mount a p15E574–581-specific TCD8+ response to FMR MuLV, and cells expressing FMR p15 are not recognized by p15-specific TCD8+. In vitro 20S proteasome digestion of synthetic 26mer peptides corresponding to the respective AKV/MCF and FMR sequences liberates the AKV/MCF peptide with a two-residue amino terminal extension and destroys the FMR peptide. The AKV/MCF 10mer peptide was an efficient competitor of TAP-mediated indicator peptide transport, whereas the 8mer was at least 30-fold less efficient. Based on these findings, the authors concluded that the poor antigenicity and immunogenicity of AKV/MCF p15 peptide was due to the Lys to Arg substitution resulting in proteasomal destruction and that cells naturally produced the 10mer, which was then trimmed in the secretory pathway. Although these conclusions are reasonable, they depend on the assumptions that the activities of purified 20S proteasomes accurately reflect proteasome cleavages within the cell and that the other amino acid differences between AKV/MCF and FMR p15E do not affect antigen processing. The latter possibility is more than pure conjecture because in extensive studies regarding the Kd-restricted presentation of rVV-encoded fragments of IV NP147–155, Yellen-Shaw et al found that the effects of flanking residues on peptide presentation can extend more than 50 residues from the determinant (75). The findings with Tag489–497 and p15E574–581 are consistent with the idea that peptide trimming in the ER is an obligate step in the presentation of

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

73

some determinants. In this eventuality, flanking sequences can also potentially play a role in immunodominance by influencing TAP-mediated transport. In thorough studies of TAP-specificity, Neisig et al identified three other optimal class I–binding peptides whose synthetic versions were extremely inefficient competitors of TAP-mediated transport (76). As with p15E574–581, extension of one of these poorly transported peptides by inclusion of one or two naturally flanking NH2-terminal residues greatly improved its interaction with TAP. It is likely that role of TAP in immunodominance in mice and humans has been obscured by the use of canonical motifs for choosing determinants for study and the concordance between the specificities of TAP and class I molecules for COOH-terminal residues. Because class I molecules are capable of binding some peptide with noncanonical COOH-terminal anchors, it is likely that TAP acts a formidable barrier against the presentation of these peptides. Just how efficient this barrier can be has been elegantly shown by Powis et al (77). Originally investigating the differential reactivity of a rat class I (RT1.Aa)specific mAb with cells expressing the relevant class I molecule but derived from disparate rat strains, they found that mAb reactivity segregated with what turned out to be the TAP locus. The strains in question express TAP alleles that differ greatly in specificity; one of the alleles is unable to transport peptides with a COOH-terminal Arg, a residue that is a (very) dominant anchor for RT1.Aa. The supply of peptides provided by the nonpermissive transporter is sufficient to enable the normal assembly and surface expression of other rat class I molecules, but it is so poor for RT1.Aa that export from the ER occurs at 10% the rate of cells expressing permissive TAP. Despite the prolonged period in the ER, which should enable binding of the odd Arg-terminating peptide transported by TAP, peptides recovered from RT1.Aa were nearly completely devoid of COOH-terminal Arg. This latter finding fits neatly with evidence of inefficient ER trimming of COOH-terminal extensions and suggests that the lack of such trimming activity contributes to determinant crypticity by preventing the generation of class I binding peptides from COOH-terminally extended precursors. Although it has yet to be formally shown that TAP polymorphism in the rat effects the TCD8+ response to foreign antigens, there is little doubt that this will be so, particularly because differences in TAP alleles result in alloreactivity. Thus, TAP clearly has the potential to play a role in immunodominance. In humans, TAP may exert a relatively subtle effect on immunodominance because human TAP is relatively promiscuous in its peptide binding: Only a few types of residues are strongly disfavored in the various positions. TAP is polymorphic in humans, however, and there is indirect evidence linking TAP

P1: PKS/SPD/MBG

January 25, 1999

74

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

alleles to the progression of HIV infection (78). Whether this is due to allelic differences in peptide transport resulting in altered TCD8+ responses remains to be demonstrated. Effects of cytokines on antigen processing Each of the steps in the antigen processing pathway can be affected by exposure of APCs to cytokines. The best-defined effects are induced by interferon γ and tumor necrosis factor α, which increase synthesis of TAP, class I molecules, and several molecular chaperones. Additionally, these cytokines modify proteasomes in two ways (11, 13, 14): first, by enhancing synthesis of three of proteasome subunits that displace homologous subunits during proteasome biogenesis, thereby producing “immunoproteasomes”; and second, by enhancing synthesis of a subset of the regulatory proteins that bind to the ends of proteasomes. These alterations in proteasomes potentially have the most profound effects on immunodominance because the specificity of proteasomes is altered by these modifications. The nature of these changes is controversial, however, and the in vivo effects on the immunogenicity of individual determinants remain largely unexplored. It is important to note that the biological relevance of cytokine-induced qualitative modifications in antigen processing requires that similar alterations in peptide generation occur in both afferent and efferent APCs. Features of proteins that contribute to immunodominance We have discussed the specific features of potential determinants and flanking sequences that can influence their immunogenicity. A more general issue is whether there are features of individual gene products that favor/disfavor the generation of immunogenic peptides. There are two reasons why this question cannot presently be answered with any degree of precision. First, the least understood portion of the antigen processing pathway is how biosynthesized proteins enter the pathway. Second, as emphasized throughout this review, there is precious little information regarding in vivo presentation of antigens to na¨ıve TCD8+. It is possible, however, to identify some properties of gene products that will influence the generation of peptides by infected APCs. First, and most obviously, for any gene product, the rate of peptide generation will be proportional to the rate of translation (this governs the abundance of the gene product and its byproducts; further, increased synthesis will also result in enhanced crosspriming). This is not to say that IDDs always come from the most abundant viral proteins: Indeed, they don’t. Rather, that given a certain inherent efficiency of peptide generation from a protein, expressing more of the protein will result in a concomitant increase in peptide generation. Second, in these circumstances, increased protein turnover favors peptide generation. Third, targeting of the protein to the ER can have positive and negative effects. On the plus side, if

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

75

the protein has a determinant in its signal sequence, there is a good chance that the peptide will be generated very efficiently; indeed, many of the most abundant peptides recovered from class I molecules are derived from signal sequences. On the minus side, if the determinant is present in the lumenal domain of the protein, its presentation will most likely be compromised. Fourth, for proteins expressed by bacteria and simple eukaryotes whose life cycle includes an intracellular phase, a special rule applies. Proteins that are targeted to the cytosol of cells will be preferentially presented. Finally, it has been observed that HIV determinants restricted by different allomorphs cluster in certain regions of several viral proteins (42), which suggests that regions of proteins can have properties that favor efficient peptide liberation. Features of pathogens that contribute to immunodominance The temporal sequence of viral gene expression can greatly influence immunodominance. In some cases, only a small subset of viral genes may be expressed by APCs. For example, responses to EBV in chronically infected individuals are largely limited to the few gene products that are constitutively expressed in latently infected B cells (79). For rVVs, expression of recombinant proteins expressed under the control of late viral promoters (i.e. after the initiation of viral DNA synthesis) can decrease immunogenicity (and antigenicity), possibly related to viral interference with host protein synthesis (80). Viruses may also influence immunodominance in more specific ways. Herpesviridae are the undisputed champions in this realm. One of the EBV proteins is abundantly expressed in latently infected cells but is infrequently immunogenic, probably because of a region that interferes with proteasomedependent peptide generation. CMV expresses numerous viral gene products that interfere with antigen processing, including one that specifically blocks the generation of peptides from an abundant viral structural protein, and others that target class I molecules in the ER for destruction, retain class I molecules intracellularly, or interfere with TAP function. HSV expresses a protein that prevents TAP-mediated peptide transport. Adenoviruses interfere with antigen processing by decreasing transcription of class I molecules and accessory antigen processing components, or by expressing a protein that retains class I molecules in the ER. The effects on immunodominance of such global interference with antigen processing have yet to be investigated in detail, but a simple prediction is that TCD8+ responses will focus on those determinants that for whatever reason are less affected by the strategy employed by the virus. In the case of human TCD8+ responses to CMV, this mechanism has been proposed to account for the immunodominance of a virion protein of such abundance that a sufficient number of copies are delivered to APCs from input virus to produce immunogenic

P1: PKS/SPD/MBG

January 25, 1999

76

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

quantities of peptide–class I complexes before viral gene expression can interfere with antigen presentation (45, 81).

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

The Other Side of the Coin: Contribution of TCD8+ Responses to Immunodominance TCD8+ REGULATION: IMMUNODOMINATION A major contributor to immunodominance is immunodomination: the suppression of SDD-specific responses by IDDs. This was one of the initial observations of immunodominance (82, 83), and it is detected as enhanced responses to nondominant determinants under conditions when responses to the IDD are prevented by altering or removing the determinant, its class I restriction element, or IDD-specific TCD8+. Immunodomination occurs in TCD8+ responses to virus-transformed cells (84), tumor antigens (50), minor H antigens (85), DNA vaccines (46), and viruses (49, 86). In some circumstances, immunodomination is limited largely to primary responses, the IDD having little effect on the priming of SDD-specific memory TCD8+ (87). It is probably more frequent that immunodomination occurs in both primary and memory TCD8+ responses. In gauging the effect of immunodomination in secondary responses, it is crucial to stimulate TCD8+ with nonlimiting amounts of SDDs in the absence of the potential IDDs because domination can occur in vitro. There are two general explanations for immunodomination. The first is that the IDD interferes with the generation of the SDD in APCs. Although this possibility has yet to be rigorously eliminated in any system by peptide quantitation, it has been repeatedly observed that TCD8+ recognition of SDDs is not affected by the coexpression of the IDD (84, 88). Most IDDs are of such low abundance as to make competition for binding to class I molecules extremely unlikely (in fact, it is difficult to observe such competition except under extreme overexpression of determinants from minigenes). In the unusual circumstance where one peptide (or two overlapping peptides) may be presented by more than one class I allomorph, this mechanism can contribute to the dominance of a response restricted to the allomorph that selectively acquires the peptide (89). Finally, immunodomination is frequently observed between peptides that bind to different class I allomorphs and would, therefore, be unlikely to compete for binding. The second, far more likely explanation is that TCD8+ specific for dominant peptides suppress responses to other peptides. This could occur by multiple mechanisms operating alone or in conjunction, including: reduction of antigen load through the actions of rapidly responding IDD-specific TCD8+ such that SDDs are expressed at suboptimal levels for TCD8+ activation; competition at the level of APCs for TCD8+ activation; and systemic suppression of responses to SDDs by IDD-specific TCD8+.

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

77

Immunodomination has been most extensively studied in LCMV infection of H-2d mice, where Ld-restricted, NP118–126-specific TCD8+ dominate responses to other determinants, including Kd-restricted GP283–291-specific TCD8+. The latter determinant is dominant if mice are infected with a LCMV mutant that was TCD8+-selected for loss of the NP118–126 determinant (88), or following infection with wild-type LCMV if mice express NP in the thymus from a transgene, resulting in the deletion of high-affinity NP118–126-specific TCD8+ (90). Under normal circumstances, TCD8+ responses to GP283–291 are delayed relative to the response to NP118–126, and it appears that clearance of virus by NP118–126-specific TCD8+ reduces the antigen load to a point that prevents stimulation of GP283–291-specific TCD8+. Consistent with this explanation, the response to GP283–291 was also suppressed in H-2d × H-2b F1 mice infected with the variant LCMV, presumably because of the presence of TCD8+ specific for the major H-2b-restricted determinant. This last finding points to a consistent feature of immunodomination: The ability of a determinant to dominate (or be dominated) is relative, and determinants restricted by the same or different allomorphs can be ordered in a hierarchy of domination. This has long been known to occur in TCD8+ responses to minor H antigens (91). The hierarchy can be altered by the prior experience of the immune system. This has been neatly shown with TCD8+ responses to Sendai virus in H-2b × H-2k F1 mice (87). The IDD in H-2b mice (NP324–332) is rendered subdominant in the F1 mice by the H-2k–restricted response to undefined (but non-NP) viral determinants. If, however, mice are infected with a rVV expressing NP prior to infection with Sendai virus, NP324–332-specific TCD8+ now dominate the response. Such a reversal of immunodominance has also been shown in responses to tumor cells (50). The ability of responding memory TCD8+ to suppress responses by na¨ıve TCD8+ was first demonstrated in experiments in which lymphocytes from virus-infected mice were adoptively transferred into na¨ıve mice (92, 93). The hierarchical nature of immunodomination has been examined by immunizing mice with mixtures of synthetic peptides that represent IDDs in their respective systems. Secondary peptide-restimulated TCD8+ from mice immunized with a mixture of five Kb-restricted peptides focused on two of the peptides (94). This could not be attributed to either peptide affinity for Kb or the numbers of peptide–class I complexes needed to trigger TCD8+ lysis. Rather, the hierarchy correlated with a 3.5-fold difference in the numbers of TCD8+ that responded to individual peptides, as determined by LDA. If, however, animals were immunized with dendritic cells pulsed with the synthetic peptide mixture, TCD8+ responses were distributed equally among the four most immunogenic peptides. These findings led the authors to propose that the number of APCs was limiting following peptide immunization and that immunodomination reflected TCD8+

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

78

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

competition for APCs. In a prior study (95), it was shown that immunizing mice with five Kd-restricted peptides also led to a hierarchy of responses that could not be attributed to peptide affinity for Kd. Indeed, when two peptides differing 30-fold in their affinity for Kd were coimmunized, the response was dominated by the weaker binding peptide. This could not be ascribed to differences in destruction by serum proteases or to recognition of either of the peptides by TCD4+. Most interestingly, the immunodomination could be eliminated if mice were treated with interleukin-12, pointing to cytokine-mediated regulation of TCD8+ responses. Together, these studies support the concept that activated TCD8+ cells can exert a suppressive effect on nonactivated (or possibly less activated) TCD8+ and that this can play a role in immunodomination. How localized this effect is remains to be determined (individual APC versus regions within a node, versus entire node etc) and may well differ, depending on the nature of the antigen. It is likely that in some circumstances this suppression would act synergistically with a reduction in antigen load to enhance the domination phenomenon. It is also plausible that in some instances suppression of TCD8+ responses is due to other components of the immune response to complex pathogens. Of particular relevance are the remarkable findings that prior immunization with tumor cells, allogeneic splenocytes, or even xenogeneic erythrocytes suppresses TCD8+ responses to unrelated alloantigens (96). Suppression is transferable by serum, and the suppressive factor appears to be antibody-bound TGF-β acting in a process that requires its Fc-receptor–mediated binding to macrophages (97, 98). As viruses induce robust antibody responses, it is plausible that this mechanism contributes to the immunodominance of determinants able to induce the most rapid TCD8+ responses. The possible role of antibody in immunodominance can be easily examined using knockout mice unable to produce antibody. It should be noted that although immunodomination is commonly observed in mouse TCD8+ responses to diverse antigens, its contribution to immunodominance in human TCD8+ responses has not been extensively examined. There is, however, at least one clear example of immunodomination involving B8-restricted, EBV-specific responses that is described below. The findings presented in the previous section prompt a question central to understanding immunodominance: Why do IDD-specific TCD8+ respond better than SDD-specific TCD8+? The simplest explanation would be that IDDs are the most abundant peptides expressed on the surface of the relevant APC and, consequently, the corresponding TCD8+ are most rapidly and vigorously activated. Although this may account for some IDDs, evidence presented above strongly suggests that there is no simple correlation between immunodominance and abundance (again with the caveat regarding the TCD8+ REPERTOIRE

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

IMMUNODOMINANCE IN TCD8+ RESPONSES

79

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

unknown properties of the relevant APC in vivo). Rather, it is probable that that in the eyes of the TCD8+ repertoire, some peptide–class I complexes are created more equal than others. There are two major questions to be addressed. First, to what extent is this Orwellian bias intrinsic to the TCR repertoire as opposed to being imposed by thymic or peripheral selection (nature versus nurture)? Second, how is the bias executed: increased numbers of TCD8+ (either in the number of clones or the average population of each clone) versus more rapid proliferation of an equivalent number of TCD8+ (size versus speed)? Issue 1: nature vs nurture The most direct approach to examine the extent to which the TCR repertoire is intrinsically biased toward IDDs is to produce TCRs independently of positive and negative selection (in a phage display library for example) and determine the frequencies and avidities of TCRs for IDDs and SDDs. This poses numerous and formidable technical difficulties and remains a distant goal. A less direct (but more accessible) approach to examining this question is to compare immunodominance in responses to a common antigen in organisms expressing a common restriction element but with distinct self antigens and TCR genes. Two such organisms are humans and transgenic mice expressing human class I molecules. Team Sette found a considerable overlap in peptide immunogenicity in humans and transgenic HLA A∗ 0201 or A11 mice (33, 99). Most importantly, immunization of such transgenic mice with viruses has been found to elicit TCD8+ that recognize the same IDD determinants as human TCD8+ (100–102). In the single study that compared TCR usage in man and mouse TCD8+ specific for the same peptide–class I complex, mouse and human TCRs were found to utilize nonhomologous Vα and Vβ segments (103). These findings indicate that the dominance of many determinants occurs independently of TCR genes and the precise nature of the self peptides operative in thymic and postthymic selection. As described below, within an individual, IDDs are often recognized by TCD8+ bearing TCRs composed of different Vβ and Vα chains. Together, these findings strongly suggest that there are special features of some IDDs (possibly also some non-IDDs recognized with similar high affinity by T cells) that enable them to interact favorably with TCRs. This could result from one or a combination of the following factors: (a) the orientation or nature of side chains available for interaction; (b) the induction of unique conformational alterations in the α helices of the class I binding groove by peptide binding; (c) the conformational flexibility (increased or decreased) of the peptide after binding to the groove. This is not to say that positive and negative selection have no effects on immunodominance. The potential of negative selection for influencing immunodominance has been elegantly shown in human TCD8+ responses to EBV

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

80

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

(104, 105). In most B8-positive individuals, responses are dominated by B8restricted, EBNA3A325–333-specific TCD8+ expressing a highly conserved TCR. These TCD8+ are strongly alloreactive to HLA B∗ 4402. EBV-infected individuals expressing B8 and B∗ 4402 make a less vigorous B8-restricted TCD8+ response to EBNA3A325–333 (threefold decrease in CTL number by LDA). Remarkably, these TCD8+ do not recognize B∗ 4402 and do not express the conserved oligoclonal TCRs typical of TCD8+ in B8-negative individuals. Presumably, immunodomination by B∗ 4402-reactive TCD8+ prevents the activation of these alternative TCD8+ in B∗ 4402-negative individuals. These findings demonstrate that tolerance to self class I molecules can influence the repertoire of virusspecific TCD8 and also that immunodomination even occurs within responses to individual IDDs. The contribution of tolerance to individual self peptides in shaping the TCR repertoire to foreign antigens is nearly completely undefined. It has been shown that TCD8+ specific for IV HA210–219 (hemagglutinin) cross-react with a peptide from an immunoglobulin VH gene (106, 107). HA210–219 is a subdominant peptide, but this self reactivity is probably not the critical factor limiting HA210–219 immunogenicity because other HA210–219-specific TCD8+ do not recognize the VH peptide. Issue 2: size vs speed Until recently, it was not possible to accurately enumerate TCD8+ and measure the diversity of their TCRs. There have been three recent developments that will accelerate progress in this area. First is the use of peptide–class I tetramers to enumerate and isolate TCD8+ specific for individual determinants. Second is the commercial availability of fluorochromeconjugated, Vβ segment specific–mAbs for nearly each of the mouse Vβ segments. Used in conjunction with peptide–class I tetramers, these mAbs provide a broad but extremely useful measure of TCR diversity. As panels of mAbs specific for mouse Vα and human Vα and Vβ segments become available, the discrimination of this method will increase and enable its widespread application to studies of human TCD8+ responses. Third are improvements in PCR-based methods and DNA sequencing efficiency that enable sequencing of TCR genes from individual TCD8+ isolated using peptide–class I tetramers. Assessing size verus speed entails determining the numbers of na¨ıve TCD8+ capable of responding to a given determinant. This is now theoretically feasible using peptide–class I tetramers, but it remains challenging because of the low frequency of na¨ıve cells in TCD8+ populations. The present discussion is limited to studies of activated primary TCD8+ or secondary TCD8+, which have provided useful information but leave the major issue largely unresolved. The diversity of TCR usage in T cell responses was first examined in TCD4+ responses to IDDs present in “Sigma” antigens (cytochrome, lysozyme, etc),

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES

81

revealing a highly restricted Vβ and Vα chain usage (108). The close evolutionary relationship of these antigens to self proteins, however, no doubt limits the diversity of these responses relative to responses to antigens in pathogens. Such self-tolerance may also contribute to the pauciclonal Kdrestricted TCD8+ primary response to a HLA CW3 determinant, which is composed of TCD8+ exclusively expressing Vβ10, most often in association with Jβ1.2 segments (109). The repertoire size in different individuals was found to consist of between 15 and 20 clonotypes. This landmark study was the first in which TCD8+ fresh from a responding animal were cytofluorographically sorted (based on Vβ10 expression) and the sequence of the TCR expressed by individual cells determined by PCR-based methodology. No doubt, many similar studies will follow in which peptide–class I tetramers are used to isolate TCD8+ whose TCRs are sequenced using primers that encompass all possible V segments. The diversity of TCRs in mouse and human responses to viral IDD determinants have been examined in a number of systems using somewhat less sophisticated and precise methods. The findings fall fairly evenly into two camps: those in which responses are dominated by cells of the same Vβ (or less often Vα) chain (45, 110–113), and those in which responses are composed of cells expressing multiple Vβ and Vα chains (87, 114–116). It is obviously premature to draw any firm conclusions regarding the contribution of TCR diversity to immunodominance. It may be reasonable to conclude, however, that the relationship will not be simple. A poignant example of the complexity possible is provided by the Kb-restricted response to HSV gB498–505 (glycoprotein B) (110). In C57BL/6 mice and other strains with similar TCR genes, TCD8+ expressing TCRs with Vβ10 or Vβ8S1 genes dominate the response. In C57/L mice, however, which lack the genes encoding these regions, the response is more diverse. Based on the amount of peptide required to sensitize target cells for TCD8+ lysis, the new clones are of the same sensitivity as the original clones and are present among memory cells at only slightly lower frequencies. The TCRs utilized are present in C57BL/6 mice but are presumably dominated by the oligoclonal responders. The domination of these clones [and in the case of EBV (105), as discussed above] supports the possibility of differential proliferation of na¨ıve TCD8+ expressing TCRs of similar affinity for same peptide–class I complex. There are several possible mechanisms that may apply. (a) TCRs containing certain V regions may signal better than other TCRs upon binding to a given peptide–class I complex, and these differences may be most (or only) apparent in stimulating na¨ıve TCD8+. (b) A subset of na¨ıve IDD-specific TCD8+ expressing a given TCR may require increased numbers of peptide–class I complexes because of the presence of tolerizing self peptides specific for the TCR. Once activated,

P1: PKS/SPD/MBG

January 25, 1999

82

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

however, the cells may be as sensitive as the IDD-specific cells expressing an alternative TCR nonreactive with self peptides.

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Questions, Questions, Questions We have strived to emphasize that immunodominance results from the complex interplay of three major factors: the quantities of peptide–class I complexes expressed on APCs, the repertoire of TCD8+ awaiting the complexes, and the ability of IDD-specific TCD8+ to suppress SDD-specific responses. It is to be expected that the contribution of these factors to immunodominance varies considerably in an antigen- and allomorph-dependent manner. Ultimately, understanding immunodominance will require answers to the following questions. What is the relevant APC for activating na¨ıve TCD8+, and under what circumstances does it present exogenous versus endogenous peptides? What is the relationship between the abundance of a peptide–class I complex on the relevant APCs and its selection as an IDD versus a SDD? Are some/many IDDs that are intrinsically more immunogenic because of an innate propensity to interact with TCRs, and if so, what is the structural basis for the interaction? How does the TCD8+ repertoire contribute to immunodominance and to what extent is this based on clonal diversity, or the size or proliferative capacity of individual clones? What controls the proliferative capacity of individual clones and how much is TCR related (tolerance/peptide antagonism) versus other factors (differences in internal signal transduction pathways or cytokine responsiveness)? What mechanisms underlie immunodomination and what are the roles of cytokines in this process? Why does immunodomination exist and is this an inevitable byproduct of the workings of the immune system or is there an evolutionary edge to using a minimal number of clonotypes to respond to a given antigen? [We previously suggested one possible advantage: minimization of the chance of self reactivity (38). Were this true, vaccines designed to elicit responses to the maximal number of target determinants would result in an increased incidence of autoimmunity.] Most of these questions address fundamental aspects of T cell biology, fitting final testimony to the central place that immunodominance occupies in T cell responses. Although reasonably complete answers will come neither easily nor rapidly, understanding of immunodominance is poised to increase logarithmically in the next few years because of recent technical advances and newfound interest attendant with the increased urgency to develop vaccines that induce effective TCD8+ responses to HIV and other organisms resistant to humoral immunity. Visit the Annual Reviews home page at http://www.AnnualReviews.org

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

IMMUNODOMINANCE IN TCD8+ RESPONSES

83

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Literature Cited 1. Zinkernagel RM, Doherty PC. 1974. Restriction of in vitro T cell-mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system. Nature 248:701–2 2. Zinkernagel RM, Doherty PC. 1979. MHC-restricted cytotoxic T cells: studies on the biological role of polymorphic major transplantation antigens determining T-cell restriction-specificity, function, and responsiveness. Adv. Immunol. 27:52–180 3. Bennink JR, Yewdell JW, Smith GL, Moss B. 1987. Anti-influenza virus cytotoxic T lymphocytes recognize the three viral polymerases and a nonstructural protein: responsiveness to individual viral antigens is major histocompatibility complex controlled. J. Virol. 61:1098–102 4. Gotch F, McMichael A, Smith G, Moss B. 1987. Identification of viral molecules recognized by influenza-specific human cytotoxic T lymphocytes. J. Exp. Med. 165:408–16 5. Townsend ARM, Gotch FM, Davey J. 1985. Cytotoxic T cells recognize fragments of the influenza nucleoprotein. Cell 42:457–67 6. Falk K, R¨otzschke O, Deres K, Metzger J, Jung G, Rammensee H-G. 1991. Identification of naturally processed viral nonapeptides allows their quantification in infected cells and suggests an allelespecific T cell epitope forecast. J. Exp. Med. 174:425–34 6a. Townsend ARM, Rothbard J, Gothch FM, Bahadur G, Wraith D, McMichael AF. 1986. The epitopes of influenza virus nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides. Cell 44:959–68 7. Sercarz EE, Lehmann PV, Ametani A, Benichou G, Miller A, Moudgil K. 1993. Dominance and crypticity of T cell antigenic determinants. Annu. Rev. Immunol. 11:729–66 8. Madden DR. 1995. The three-dimensional structure of peptide-MHC complexes. Annu. Rev. Immunol. 13:587–622 9. Rammensee H-G, Bachmann J, Stevanovic S. 1997. MHC Ligands and Peptide Motifs. Austin, TX: Landes Biosci. 10. Engelhard VH. 1994. Structure of peptides associated with MHC class I molecules. Curr. Op. Immunol. 6:13–23 11. Pamer E, Cresswell P. 1998. Mechanisms of MHC class I–restricted antigen processing. Annu. Rev. Immunol. 16:323–58

12. Elliott T. 1997. Transporter associated with antigen processing. Adv. Immunol. 65:47–109 13. Momburg F, Hammerling GJ. 1998. Generation and TAP-mediated transport of peptides for major histocompatibility complex class I molecules. Adv. Immunol. 68:191–256 14. Tanaka K, Tanahashi N, Tsurumi C, Yokota K-Y, Shimbara N. 1997. Proteasomes and antigen processing. Adv. Immunol. 64:1–38 15. LI S, Sjogren HO, Hellman U, Pettersson RF, Wang P. 1997. Cloning and functional characterization of a subunit of the transporter associated with antigen processing. Proc. Natl. Acad. Sci. USA 94:8708–13 16. Srivastava PK. 1993. Peptide-binding heat shock proteins in the endoplasmic reticulum: role in immune response to cancer and in antigen presentation. Adv. Cancer Research. 62:153–77 17. Sykulev Y, Cohen RJ, Eisen HN. 1995. The law of mass action governs antigenstimulated cytolytic activity of CD8+ cytotoxic lymphocytes. Proc. Natl. Acad. Sci. USA 92:11990–92 18. Kageyama S, Tsomides TJ, Sykulev Y, Eisen HN. 1995. Variations in the number of peptide-MHC class I complexes required to activate cytotoxic T cell responses. J. Immunol. 154:567–76 19. Sette A, Alexander J, Snoke K, Grey HM. 1996. Antigen analogs as tools to study T-cell activation function and activation. Semin. Immunol. 8:103–8 20. Sloan-Lancaster J, Allen PM. 1996. Altered peptide ligand-induced partial T cell activation: molecular mechanisms and role in T cell biology. Annu. Rev. Immunol. 14:1–27 21. Bevan MJ. 1995. Antigen presentation to cytotoxic T lymphocytes in vivo. J. Exp. Med. 182:639–41 22. Srivastava PK, Heike M. 1991. Tumorspecific immunogenicity of stress-induced proteins: convergence of two evolutionary pathways of antigen presentation? Semin. Immunol. 3:57–64 23. Jondal M, Schirmbeck R, Reimann J. 1996. MHC class I-restricted CTL responses to exogenous antigens. Immunity 5:295–302 24. McMichael AJ, O’Callaghan CA. 1998. A new look at T cells. J. Exp. Med. 187:1367–71 25. Andersen PS, Stryhn A, Hansen BE, Fugger L, Engberg L, Buus S. 1996.

P1: PKS/SPD/MBG

January 25, 1999

84

26.

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

27.

28.

29.

30.

31.

32.

33.

34.

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK A recombinant antibody with the antigen-specific major histocompatibility complex-restricted specificity of T cells. Proc. Natl. Acad. Sci USA 93:1820–24 Porgador A, Yewdell JW, Deng Y, Bennink JR, Germain RN. 1997. Localization, quantitation, and in situ detection of specific peptide-MHC class I complexes using a monoclonal antibody. Immunity 6:715–26 Meadows L, Wang W, den Haan JMM, Blokland E, Reinhardus C, Drijfhout JW, Shabanowitz J, Pierce R, Agulnik AI, Bishop CE, Hunt DF, Goulmy E, Engelhard VH. 1997. The HLAA∗ 0201-restricted H-Y antigen contains a posttranslationally modified cysteine that significantly affects T cell recognition. Immunity 6:273–81 Sette A, Sidney J, Del Guercio M-F, Southwood S, Ruppert J, Dahlberg C, Grey HM, Kudo RT. 1994. Peptide binding to the most frequent HLA-A class I alleles measured by quantitative molecular binding assays. Mol. Immunol. 31:813– 22 Sette A, Vitiello A, Reherman B, Fowler P, Nayersina R, Kast WA, Melief CJM, Oseroff C, Yuan L, Ruppert J, Sidney J, Guercio MFd, Southwood S, Kubo RT, Chesnut RW, Grey HM, Chisari FV. 1994. The relationship between class I binding affinity and immunogenicity of potential cytotoxic T cell epotopes. J. Immunol. 153:5586–92 LaFace DM, Vestberg M, Yang Y, Srivastava R, DiSanto J, Flomenberg N, Brown S, Sherman LA, Peterson PA. 1995. Human CD8 transgene regulation of HLA recognition by murine T cells. J. Exp. Med. 182:1315–25 Sun J, Leahy DJ, Kavathas PB. 1995. Interaction between CD8 and major histocompatibility complex (MHC) class I mediated by multiple contact surfaces that include the alpha 2 and alpha 3 domains of MHC class I. J. Exp. Med. 182:1275–80 Parker KC, Bednarek MA, Coligan JE. 1994. Scheme for ranking potential HLAA2 binding peptides based on independent binding of individual peptide sidechains. J. Immunol. 152:163–75 Alexander J, Oseroff C, Sidney J, Wentworth P, Keogh E, Hermanson G, Chisari FV, Kubo RT, Grey HM, Sette A. 1997. Derivation of HLA-A11/Kb transgenic mice: functional CTL repertoire and recognition of human A11-restricted CTL epitopes. J. Immunol. 159:4753–61 van der Most RG, Murali-Krishna K, Whitton JL, Oseroff C, Alexander J,

35.

36.

37.

38.

39.

40.

41.

42.

Southwood S, Sidney J, Chesnut RW, Sette A, Ahmed R. 1998. Identification of Db- and Kb-restricted subdominant cytotoxic T-cell responses in lymphocytic choriomeningitis virus-infected mice. Virology 240:158–67 van der Most RG, Sette A, Oseroff C, Alexander J, Murali-Krishna K, Lau LL, Southwood S, Sidney J, Chestnut RW, Matloubian M, Ahmed R. 1996. Analysis of cytotoxic T cells responses to dominant and subdominant epitopes during acute and chronic lymphocytic choriomeningitis virus infection. J. Immunol. 157:5543– 54 van der Most RG, Murali-Krishna K, Whitton JL, Oseroff C, Alexander J, Southwood S, Sidney J, Chesnut RW, Sette A, Ahmed R. 1998. Identification of Db- and Kb-restricted subdominant cytotoxic T-cell responses in lymphocytic choriomeningitis virus-infected mice. Virology 240:158–67 Vitiella A, Yuan L, Chestnut RW, Sidney J, Southwood S, Farness P, Jackson MR, Peterson PA, Sette A. 1996. Immunodominance analysis of CTL responses to influenza PR8 virus reveals two new dominant and subdominant Kb-restricted epitopes. J. Immunol. 157:5555–62 Deng Y, Yewdell JW, Eisenlohr LC, Bennink JR. 1997. MHC affinity, peptide liberation, T cell repertoire, and immunodominance all contribute to the paucity of MHC class I-restricted peptides recognized by antiviral CTL. J. Immunol. 158:1507–15 Udaka K, Wiesm¨uller K-H, Kienle S, Jung G, Walden P. 1995. Decrypting the structure of major histocompatibility complex class I-restricted cytotoxic T lymphocyte epitopes with complex peptide libraries. J. Exp. Med. 181:2097–108 Stryhn A, Petersen LO, Romme T, Holm CB, Holm A, Buus S. 1996. Peptide binding specificity of major histocompatibility complex class I resolved into an array of apparently independent subspecificities: quantitation by peptide libraries and improved prediction of binding. Eur. J. Immunol. 26:1911–18 Gavin MA, Dere B, Grandea AG, Hogquist KA, Bevan MJ. 1994. Major histocompatibility complex class I allelespecific peptide libraries: identification of peptides that mimic an H-Y T cell epitope. Eur. J. Immunol. 24:2124–33 Rowland-Jones S, Tan R, McMichael A. 1997. Role of cellular immunity in protection against HIV infection. Adv. Immunol. 65:277–346

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

IMMUNODOMINANCE IN TCD8+ RESPONSES 43. Pantaleo G, Fauci AS. 1995. New concepts in the pathogenesis of HIV infection. Annu. Rev. Immunol. 13:487–512 44. Rickinson AB, Moss DJ. 1997. Human cytotoxic T lymphocyte responses to Epstein-Barr virus infection. Annu. Rev. Immunol. 15:405–31 45. Wills MR, Carmichael AJ, Mynard K, Jin X, Weekes MP, Plachter B, Sissons JG. 1996. The human cytotoxic T-lymphocyte (CTL) response to cytomegalovirus is dominated by structural protein pp65: frequency, specificity, and T-cell receptor usage of pp65-specific CTL. J Virol. 70:7569–79 46. Fu T-M, Friedman A, Ulmer JB, Liu MA, Donnelly JJ. 1997. Protective cellular immunity: cytotoxic T-lymphocyte responses against dominant and recessive epitopes of influenza virus nucleoprotein induced by DNA immunization. J. Virol. 71:2715–21 47. Chen Y, Webster RG, Woodland DL. 1998. Induction of CD8+ T cell responses to dominant and subdominant epitopes and protective immunity to Sendai virus infection by DNA vaccination. J. Immunol. 160:2425–32 48. Oukka M, Manuguerra JC, Livaditis N, Tourdot S, Riche N, Vergnon I, Cordopatis P, Kosmatopoulos K. 1996. Protection against lethal viral infection by vaccination with nonimmunodominant peptides. J. Immunol. 157:3039–45 49. Lewicki HA, Von Herrath G, Evans CF, Whitton JL, Oldstone MB. 1995. CTL escape viral variants. II. Biologic activity in vivo. Virology 211:443–50 50. Van WC, Monach PA, Urban JL, Wortzel RD, Schreiber H. 1996. Immunodominance deters the response to other tumor antigens thereby favoring escape: prevention by vaccination with tumor variants selected with cloned cytolytic T cells in vitro. Tissue Antigens 47:399–407 51. Johnston JV, Malacko AR, Mizuno MT, McGowan P, Hellstrom I, Hellstrom KE, Marquardt H, Chen L. 1996. B7-CD28 costimulation unveils the hierarchy of tumor epitopes recognized by major histocompatibility complex class I-restricted CD8+ cytolytic T lymphocytes. J. Exp. Med. 183:791–800 52. van der Burg SH, Visseren MJ, Brandt RM, Kast WM, Melief CJ. 1996. Immunogenicity of peptides bound to MHC class I molecules depends on the MHCpeptide complex stability. J. Immunol. 156:3308–14 53. Hahn YS, Hahn CS, Braciale TJ. 1996. Endogenous presentation of a nascent

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

85

antigenic epitope to CD8+ CTL is more efficient than exogenous presentation. Immunol. Cell Biol. 74:394–400 Shi Y, Smith KD, Kurilla MG, Lutz CT. 1997. Cytotoxic CD8+ T cells recognize EBV antigen but poorly kill autologous EBV-infected B lymphoblasts: immunodominance is elicited by a peptide epitope that is presented at low levels in vitro. J. Immunol. 159:1844–52 Hill AB, Lee SP, Haurum JS, Murray N, Yao QY, Rowe M, Signoret N, Rickinson AB, McMichael AJ. 1995. Class I major histocompatibility complex-restricted cytotoxic T lymphocytes specific for Epstein-Barr virus (EBV)-transformed B lymphoblastoid cell lines against which they were raised. J. Exp. Med. 181:2221– 28 Sykulev Y, Joo M, Vturina I, Tsomides TJ, Eisen HN. 1996. Evidence that a single peptide-MHC complex on a target cell can elicit a cytolytic T cell response. Immunity 4:565–71 O’Herrin SM, Lebowitz MS, Bieler JG, al-Ramadi BK, Utz U, Bothwell AL, Schneck JP. 1997. Analysis of the expression of peptide-major histocompatibility complexes using high affinity soluble divalent T cell receptors. J. Exp. Med. 186:1333– 45 Ant´on LA, Yewdell JW, Bennink JR. 1997. MHC class I-associated peptides produced from endogenous gene products with vastly different efficiencies. J. Immunol. 158:2535–42 Restifo NP, Bac´ık I, Irvine KR, Yewdell JW, McCabe B, Anderson RW, Eisenlohr LC, Rosenberg SA, Bennink JR. 1995. Antigen processing in vivo and the elicitation of primary CTL responses. J. Immunol. 154:4414–22 Tobery TW, Siliciano RF. 1997. Targeting of HIV-1 antigens for rapid intracellular degradation enhances cytotoxic T lymphocyte (CTL) recognition and the induction of de novo CTL responses in vivo after immunization. J. Exp. Med. 185: 909–20 Tsomides TJ, Aldovini A, Johnson RP, Walker BD, Young RA, Eisen HN. 1994. Naturally processed viral peptides recognized by cytotoxic T lymphocytes on cells chronically infected by human immunodeficiency virus type 1. J. Exp. Med. 180:1283–93 Vijh S, Pamer EG. 1997. Immunodominant and subdominant CTL responses to Listeria monocytogenes infection. J. Immunol. 158:3366–71 Busch DH, Pilip IM, Vijh S, Pamer EG.

P1: PKS/SPD/MBG

January 25, 1999

86

64.

65.

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

66.

67.

68.

69.

70.

71.

72.

73.

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK 1998. Coordinate regulation of complex T cell populations responding to bacterial infection. Immunity 8:353–62 Busch DH, Pamer EG. 1998. MHC class I/peptide stability: implications for immunodominance, in vitro proliferation, and diversity of responding CTL. J. Immunol. 160:4441–48 Villanueva MS, Fischer P, Feen K, Pamer EG. 1994. Efficiency of MHC class I antigen processing: a quantitative analysis. Immunity 1:479–89 Vijh S, Pilip IM, Pamer EG. 1998. Effect of antigen-processing efficiency on in vivo T cell response magnitudes. J. Immunol. 160:3971–77 Steven NM, Leese AM, Annels NE, Lee SP, Rickinson AB. 1996. Epitope focusing in the primary cytotoxic T cell response to Epstein-Barr virus and its relationship to T cell memory. J. Exp. Med. 184:1801–13 Levitsky V, Zhang Q-J, Levitskaya J, Masucci MG. 1996. The life span of major histocompatibility complex-peptide complexes influences the efficiency of presentation and immunogenicity of two class I-restricted cytotoxic T lymphocyte epitopes in the Epstein-Barr virus nuclear antigen 4. J. Exp. Med. 183:915–26 Yellen-Shaw AJ, Laughlin CE, Metrione RM, Eisenlohr LC. 1997. Murine transporter associated with antigen presentation (TAP) preferences influence class Irestricted T cell responses. J. Exp. Med. 186:1655–62 Del Val M, Schlicht H-J, Ruppert T, Reddehase MJ, Koszinowski UH. 1991. Efficient processing of an antigenic sequence for presentation by MHC class I molecules depends on its neighboring residues in the protein. Cell 66:1145– 53 Chen W, Khilko S, Fecondo J, Margulies DH, McCluskey J. 1994. Determinant selection of major histocompatibility complex class I-restricted antigenic peptides is explained by class I-peptide affinity and is strongly influenced by nondominant anchor residues. J. Exp. Med. 180:1471– 83 Niedermann G, Butz S, Ihlenfeldt HG, Grimm R, Lucchiari M, Hosch¨utzky H, Jung G, Maier B, Eichmann K. 1995. Contribution of proteasome-mediated proteolysis to the hierarchy of epitopes presented by major histocompatibility complex class I molecules. Immunity 2:289–99 Fu TM, Mylin LM, Schell TD, Bacik I, Russ G, Yewdell JW, Bennink JR,

74.

75.

76.

77.

78.

79.

80.

81.

82.

Tevethia SS. 1998. An endoplasmic reticulum-targeting signal sequence enhances the immunogenicity of an immunorecessive simian virus 40 large T antigen cytotoxic T-lymphocyte epitope. J. Virol. 72:1469–81 Ossendorp F, Eggers M, Neisig A, Ruppert T, Groettrup M, Sijts A, Mengede E, Kloetzel PM, Neefjes J, Koszinowski U, Melief C. 1996. A single residue exchange within a viral CTL epitope alters proteasome-mediated degradation resulting in lack of antigen presentation. Immunity 5:115–24 Yellen-Shaw AJ, Wherry EJ, Dubois GC, Eisenlohr LC. 1997. Point mutation flanking a CTL epitope ablates in vitro and in vivo recognition of a full-length viral protein. J. Immunol. 158:3227–34 Neisig A, Roelse J, Sijts AJAM, Ossendorp F, Feltkamp MCW, Kast WM, Melief CJM, Neefjes JJ. 1995. Major differences in transporter associated with antigen presentation (TAP)-dependent translocation of MHC class I-presentable peptides and the effect of flanking sequences. J. Immunol. 154:1273–79 Powis SJ, Young LL, Joly E, Barker PJ, Richardson L, Brandt RP, Melief CJ, Howard JC, Butcher GW. 1996. The rat cim effect: TAP allele-dependent changes in a class I MHC anchor motif and evidence against C-terminal trimming of peptides in the ER. Immunity 4:159–65 Kaslow RA, Carrington M, Apple R, Park L, Munoz A, Saah AJ, Goedert JJ, Winkler C, O’Brien SJ, Rinaldo C, Detels R, Blattner W, Phair J, Erlich H, Mann DL. 1996. Influence of combinations of human major histocompatibility complex genes on the course of HIV-1 infection. Nat. Med. 2:405–11 Rickinson AB, Lee SP, Steven NM. 1996. Cytotoxic T lymphocyte responses to Epstein-Barr virus. Curr. Op. Immunol. 8:492–97 Coupar BEH, Andrew ME, Both GW, Boyle DB. 1986. Temporal regulation of influenza hemagglutinin expression in vaccinia virus recombinants and effects on the immune response. Eur. J. Immunol. 16:1479–87 Riddell SR, Rabin M, Geballe AP, Britt WJ, Greenberg PD. 1991. Class I MHCrestricted cytotoxic T lymphocyte recognition of cells infected with human cytomegalovirus does not require endogenous viral gene expression. J. Immunol. 146:2795–804 Zinkernagel RM, Althage A, Cooper S, Kreeb G, Klein PA, Sefton B, Flaherty L,

P1: PKS/SPD/MBG

January 25, 1999

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

IMMUNODOMINANCE IN TCD8+ RESPONSES

83.

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

84.

85.

86.

87.

88.

89.

90.

91.

92.

Stimpfling J, Shreffler D, Klein J. 1978. Ir-genes in H-2 regulate generation of anti-viral cytotoxic T cells. Mapping to K or D and dominance of unresponsiveness. J. Exp. Med. 148:592–606 Doherty PC, Biddison WE, Bennink JR, Knowles BB. 1978. Cytotoxic T-cell responses in mice infected with influenza and vaccinia viruses vary in magnitude with H-2 genotype. J. Exp. Med. 148:534– 43 Mylin LM, Bonneau RH, Lippolis JD, Tevethia SS. 1995. Hierarchy among multiple H-2b-restricted cytotoxic Tlymphocyte epitopes within simian virus 40 T antigen. J. Virol. 69:6665–77 Pion S, Fontaine P, Desaulniers M, Jutras J, Filep JG, Perreault C. 1997. On the mechanisms of immunodominance in cytotoxic T lymphocyte responses to minor histocompatibility antigens. Eur. J. Immunol. 27:421–30 van der Most RG, Concepcion RJ, Oseroff C, Alexander J, Southwood S, Sidney J, Chesnut RW, Ahmed R, Sette A. 1997. Uncovering subdominant cytotoxic T-lymphocyte responses in lymphocytic choriomeningitis virus-infected BALB/c mice. J. Virol. 71:5110–14 Cole GA, Hogg TL, Coppola MA, Woodland DL. 1997. Efficient priming of CD8+ memory T cells specific for a subdominant epitope following Sendai virus infection. J. Immunol. 158:4301–9 Weidt G, Utermohlen O, Heukeshoven J, Lehmann-Grubbe F, Deppert W. 1998. Relationship among immunodominance of single CD8+ T cell epitopes, virus load, and kinetics of primary antiviral CTL response. J. Immunol. 160:2923–31 Tussey LG, Rowland Jones S, Zheng TS, Androlewicz MJ, Cresswell P, Frelinger JA, McMichael AJ. 1995. Different MHC class I alleles compete for presentation of overlapping viral epitopes. Immunity 3:65–77 Von Herrath G, Dockter J, Nerenberg M, Gairin JE, Oldstone MB. 1994. Thymic selection and adaptability of cytotoxic T lymphocyte responses in transgenic mice expressing a viral protein in the thymus. J. Exp. Med. 180:1901–10 Wettstein PJ. 1986. Immunodomnance in the T-cell response to multiple non-H2 histocompatibity antigens. II. Observation of a hierarcy among dominant antigens. Immunogenetics 24:24–31 Bennink JR, Doherty PC. 1981. The response to H-2-different virus-infected cells is mediated by long-lived T lymphocytes and is diminished by prior virus

93.

94.

95.

96.

97.

98.

99.

100.

101.

102.

87

priming in a syngeneic environment. Cell Immunol. 61:220–24 Jamieson BD, Ahmed R. 1989. T cell memory. Long-term persistence of virusspecific cytotoxic T cells. J. Exp. Med. 169:1993–2005 Sandberg JK, Grufman P, Wolpert EZ, Franksson L, Chambers BJ, Karre K. 1998. Superdominance among immunodominant H-2Kb-restricted epitopes and reversal by dendritic cell-mediated antigen delivery. J. Immunol. 160:3163–69 Eberl G, Kessler B, Eberl LP, Brunda MJ, Valmori D, Corradin G. 1996. Immunodominance of cytotoxic T lymphocyte epitopes co-injected in vivo and modulation by interleukin-12. Eur. J. Immunol. 26:2709–16 Rowley DA, Stach RM. 1993. A first or dominant immunization. I. Suppression of simultaneous cytolytic T cell responses to unrelated alloantigens. J. Exp. Med. 178:835–40 Stach RM, Rowley DA. 1993. A first or dominant immunization. II. Induced immunoglobulin carries transforming growth factor beta and suppresses cytolytic T cell responses to unrelated alloantigens. J. Exp. Med. 178:841–52 Rowley DA, Stach RM. 1998. B lymphocytes secreting IgG linked to latent transforming growth factor-beta prevent primary cytolytic T lymphocyte responses. Int. Immunol. 10:355–63 Wentworth PA, Vitiello A, Sidney J, Keogh E, Chesnut RW, Grey H, Sette A. 1996. Differences and similarities in the A2. 1-restricted cytotoxic T cell repertoire in humans and human leukocyte antigentransgenic mice. Eur. J. Immunol. 26:97– 101 Shirai M, Arichi T, Nishioka M, Nomura T, Ikeda K, Kawanishi K, Engelhard VH, Feinstone SM, Berzofsky JA. 1995. CTL responses of HLA-A2. 1-transgenic mice specific for hepatitis C viral peptides predict epitopes for CTL of humans carrying HLA-A2.1. J. Immunol. 154:2733–42 Engelhard VH, Lacy E, Ridge JP. 1991. Influenza A-specific, HLA-A2.1restricted cytotoxic T lymphocytes from HLA-A2.1 transgenic mice recognize fragments of the M1 protein. J. Immunol. 146:1226–32 Man S, Newberg MH, Crotzer VL, Luckey CJ, Williams NS, Chen Y, Huczko EL, Ridge JP, Engelhard VH. 1995. Definition of a human T cell epitope from influenza A non-structural protein 1 using HLA-A2.1 transgenic mice. Int. Immunol. 7:597–605

P1: PKS/SPD/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

88

P2: PKS/ARY

11:57

QC: PKS

Annual Reviews

AR078-03

YEWDELL & BENNINK

103. Man S, Ridge JP, Engelhard VH. 1994. Diversity and dominance among TCR recognizing HLA-A2. 1+ influenza matrix peptide in human MHC class I transgenic mice. J. Immunol. 153:4458–67 104. Burrows SR, Silins SL, Moss DJ, Khanna R, Misko IS, Argaet VP. 1995. T cell receptor repertoire for a viral epitope in humans is diversified by tolerance to a background major histocompatibility complex antigen. J. Exp. Med. 182:1703–15 105. Burrows SR, Khanna R, Burrows JM, Moss DJ. 1994. An alloresponse in humans is dominated by cytotoxic T lymphocytes (CTL) cross-reactive with a single Epstein-Barr virus CTL epitope: implications for graft-versus-host disease. J. Exp. Med. 179:1155–61 106. Cao W, Myers-Powell BA, Braciale TJ. 1994. Recognition of an immunoglobulin Vh epitope by Influenza virus-specific class I major histocompatibility complexrestricted cytolytic T lymphocytes. J. Exp. Med. 179:195–202 107. Cao W, Myers-Powell BA, Braciale TJ. 1996. The weak CD8+ CTL response to an influenza hemagglutinin epitope reflects limited T cell availability. J. Immunol. 157:505–11 108. Davis MM, McHeyzer-Williams M, Chien YH. 1995. T-cell receptor V-region usage and antigen specificity. The cytochrome c model system. Ann NY Acad. Sci. 756:1–11 109. 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 comprised of very few clones. Immunity 4:47–55 110. Jones CM, Cose SC, Carbone FR. 1997. Evidence for cooperation between TCR V region and junctional sequences in

111.

112.

113.

114.

115.

116.

determining a dominant cytotoxic T lymphocyte response to herpes simplex virus glycoprotein B. Int. Immunol. 9:1319–28 Campos-Lima PO, Levitsky V, Imreh MP, Gavioli R, Masucci MG. 1997. Epitopedependent selection of highly restricted or diverse T cell receptor repertoires in response to persistent infection by EpsteinBarr virus. J. Exp. Med. 186:83–89 Silins SL, Cross SM, Elliott SL, Pye SJ, Burrows SR, Burrows JM, Moss DJ, Argaet VP, Misko IS. 1996. Development of Epstein-Barr virus-specific memory T cell receptor clonotypes in acute infectious mononucleosis. J. Exp. Med. 184:1815–24 Callan MF, Steven N, Krausa P, Wilson JD, Moss PA, Gillespie GM, Bell JI, Rickinson AB, McMichael AJ. 1996. Large clonal expansions of CD8+ T cells in acute infectious mononucleosis. Nat. Med. 2:906–11 Deckhut AM, Allan W, McMickle A, Eichelberger M, Blackman MA, Doherty PC, Woodland DL. 1993. Prominent usage of Vβ8.3 T cells in the H-2Dbrestricted response to an influenza A virus nucleoprotein epitope. J. Immunol. 151:2658–66 Horwitz MS, Yanagi Y, Oldstone MB. 1994. T-cell receptors from virus-specific cytotoxic T lymphocytes recognizing a single immunodominant nine-amino-acid viral epitope show marked diversity. J. Virol. 68:352–57 Kalams SA, Johnson RP, Trocha AK, Dynan MJ, Ngo HS, D’Aquila RT, Kurnick JT, Walker BD. 1994. Longitudinal analysis of T cell receptor (TCR) gene usage by human immunodeficiency virus 1 envelope-specific cytotoxic T lymphocyte clones reveals a limited TCR repertoire. J. Exp. Med. 179:1261–71

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:51-88. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

12:53

QC: NBL/UKS

T1: NBL

Annual Reviews

AR078-04

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:89–108 c 1999 by Annual Reviews. All rights reserved Copyright °

INTEGRATION OF T CELL RECEPTOR–DEPENDENT SIGNALING PATHWAYS BY ADAPTER PROTEINS James L. Clements,1 Nancy J. Boerth,1 Jong Ran Lee,1 and Gary A. Koretzky1,2 1Department

of Internal Medicine and 2Departments of Internal Medicine and Physiology and Biophysics, and the Interdisciplinary Graduate Programs in Molecular Biology and Immunology, University of Iowa College of Medicine, Iowa City, Iowa 52242; e-mail: [email protected], [email protected], [email protected], [email protected]

KEY WORDS:

T lymphocyte, second messengers, protein tyrosine kinase, signal transduction, ITAM

ABSTRACT The initiation of biochemical signal transduction following ligation of surface receptors with intrinsic cytoplasmic tyrosine kinase activity is common for many cell types. T lymphocytes also require activation of tyrosine kinases following T cell receptor (TCR) ligation for maximal stimulation. However, the TCR has no intrinsic tyrosine kinase activity. Instead, the TCR must rely on cytoplasmic tyrosine kinases that localize to the TCR complex and initiate TCR-mediated signaling events. Although much has been learned regarding how these cytosolic tyrosine kinases are activated and recruited to the TCR complex, relatively little is understood about how these initial events are translated into transcriptional activation of genes that regulate cytokine production, cell proliferation, and cell death. Recently, it has become clear that the class of intracellular molecules known collectively as adapter proteins, molecules with modular domains capable of recruiting additional proteins but that exhibit no intrinsic enzymatic activity, serve to couple proximal biochemical events initiated by TCR ligation with more distal signaling pathways.

89 0732-0582/99/0410-0089$08.00

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

90

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

CLEMENTS ET AL

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

INTRODUCTION The majority of T lymphocytes express a disulfide-linked T cell receptor (TCR) heterodimer composed of α and β polypeptide chains. The loci encoding the α and β chains belong to the immunoglobulin supergene family and are composed of multiple gene segments that encode the constant and variable subunits of the receptor (1). The potential for numerous rearrangements between these gene segments, when combined with the addition and removal of nucleotides at the intervening joining regions, generates the amazing capacity for recognition of diverse antigens (2). The cytoplasmic tails of the TCR chains contain no known signaling motifs, precluding the TCR heterodimer from initiating intracellular signaling. These signaling shortcomings are overcome through a noncovalent association with an invariant polypeptide complex consisting of CD3 γ , δ, ε chains, and a ζ-ζ homodimers or ζ -η heterodimer (3, 4). A direct link between TCR ligation and initiation of intracellular signal transduction was revealed when conserved sequences within the cytoplasmic chains of the CD3 complex and the ζ chains were identified that contain tyrosine residues in specific motifs (Y-X-X-I/L-spacer-Y-X-X-I/L, where X represents any amino acid) (5). These domains have been designated immunoreceptor tyrosine–based activation motifs (ITAMs) and are found in single copies within each chain of the CD3 complex and in triplicate in the ζ chains (6). Experimental evidence supports a model whereby TCR ligation activates the Src family protein tyrosine kinases (PTK) Lck and/or Fyn, resulting in the rapid phosphorylation and activation of the Syk family PTK ZAP-70 (7, 8). Another potential substrate of the Src family tyrosine kinases are the ITAMs themselves, thus providing an inducible mechanism of ZAP-70 recruitment to the TCR/CD3 signaling complex (9, 10). The end result of TCR-coupled tyrosine kinase activation is the initiation of numerous signaling pathways within the cell. The required nature of inducible tyrosine kinase activity following TCR ligation has been highlighted by numerous studies in which pharmacological inhibition of tyrosine kinase activity (11, 12) or generation of cell lines and mouse strains deficient for these kinases results in a marked inhibition of TCR-induced activation (13–17). In addition to our growing understanding of the molecular basis behind the proximal signaling events initiated by TCR ligation, much has been learned regarding the transcriptional activation of numerous genes following T cell activation. Perhaps the best-characterized example to date is the regulation of the interleukin 2 (IL-2) gene. TCR ligation in the context of appropriate costimuli results in the transcriptional activation of the IL-2 gene. Transcription of IL-2 mRNA is dependent on the formation and activation of a number of transcription factors, including AP-1, NF-κB, and the nuclear factor of activated T cells (NF-AT) (18). Transcriptionally active NF-AT consists of AP-1 complexed with an NF-AT component that is translocated to the nucleus in a calcium-dependent,

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ADAPTER PROTEINS IN TCR SIGNALING

91

cyclosporin A–sensitive manner following TCR ligation (19). Thus, formation of the nuclear NF-AT complex and subsequent activation of the IL-2 gene depends on the integration of a number of signals emanating from the plasma membrane. Interestingly, the products of the growing NF-AT gene family have been implicated in the regulation of many additional genes, including those encoding numerous cytokines as well as the pro-apototic protein CD95 ligand (19–22). Despite these advances in our understanding of the proximal and more distal biochemical events that occur following TCR ligation, it is not clear in many cases how initial signaling events are translated into transcriptional activation and cell proliferation. For example, although TCR-dependent activation of the small GTP-binding protein Ras requires cytosolic tyrosine kinase function (23), how these TCR activated tyrosine kinases are coupled with Ras activation remains poorly understood. Potential mechanisms for TCR-dependent Ras activation have been suggested based on observations made in other cell types. As mentioned, many growth factor receptors exhibit intrinsic tyrosine kinase activity that is induced following ligand binding and receptor dimerization. In a search for molecules that are recruited to the cytoplasmic portion of the activated epidermal growth factor receptor, Lowenstein et al identified Grb2, an adapter protein comprised of a single SH2 domain flanked by two SH3 domains (24). It was subsequently demonstrated that Grb2 associates constitutively with the guanine nucleotide exchange factor Sos, thus providing a mechanism for epidermal growth factor–mediated Ras activation in fibroblasts following recruitment of the Grb2/Sos complex to the plasma membrane via the phosphorylated epidermal growth factor receptor (25, 26). Additional studies have defined a highly conserved signaling pathway beginning with Ras activation and involving a dual-specific tyrosine/threonine kinase cascade that ultimately regulates the transcriptional activation of a number of genes (27, 28). Together, these studies have led to the elucidation of a complete biochemical circuit that is initiated by a surface receptor, requires PTK activation, induces Ras activation, and culminates in transcriptional activation.

ADAPTER PROTEINS THAT PROMOTE TCR SIGNALING Grb2 and Shc Given the function of Grb2 in coupling growth factor receptor ligation with Ras activation, several groups addressed a role for this adapter protein in mediating TCR-coupled Ras activation. Because TCR ligation activates Ras in a PTK-dependent manner, it was hypothesized that inducibly phosphorylated tyrosine residues in the ITAMs of the CD3 complex and ζ chains could recruit the Grb2-Sos complex to the membrane, resulting in Ras activation. To date,

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

92

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

CLEMENTS ET AL

the direct recruitment of Grb2-Sos to phosphorylated ζ-chain ITAMs following TCR ligation has not been demonstrated formally. However, Grb2 has served as a useful focal point for the identification of additional proteins that may function to couple TCR-dependent tyrosine kinases with downstream signaling events (Table 1). In this capacity, a complex consisting of the adapter molecules Shc and Grb2 has been found in association with ζ following TCR engagement (29). Shc consists of a single COOH-terminal SH2 domain, a central proline/glycine rich collagen homology domain (CH1), and an amino-terminal phosphotyrosine binding domain (PTB) (30–32). Following TCR ligation and the subsequent phosphorylation of ITAMs, Shc is recruited to the ITAM in an SH2 domain–dependent manner (29). Shc itself is a substrate of the TCRactivated tyrosine kinases, resulting in the recruitment of Grb2 via the Grb2 SH2 domain (33). Formation of a ζ /Shc/Grb2 complex provides an attractive model for the recruitment of Sos to the plasma membrane and subsequent activation of Ras following TCR ligation. However, several studies suggest that Shc does not mediate TCR-inducible Ras activation and that Shc may play a more important role in other receptor-mediated signaling events, including those initiated by ligation of CD4 and the IL-2 receptor (34–37). Additionally, Grb2 has been found in association with phosphorylated ζ in the absence of Shc, which suggests that Shc is not required to mediate Grb2 recruitment (35). To address further the role of Shc in TCR-mediated signaling, the SH2 domain of Grb2 or Shc was transfected into a T cell line, and the impact on TCRdependent transcriptional activation of NF-AT and AP-1 was determined (34). Although overexpression of the Grb2 SH2 domain significantly interferes with

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ADAPTER PROTEINS IN TCR SIGNALING

93

TCR-dependent activation of these factors, the SH2 domain of Shc has little effect. It should be noted that although the SH2 domain of Shc fails to interfere with TCR-dependent signaling, expression of the isolated PTB domain of Shc diminishes TCR-dependent activation of NF-AT (38). This effect is likely due to the ability of the Shc PTB domain to bind to and sequester ZAP-70 (38). Together, these observations suggest that Shc-dependent regulation of TCR signaling is complex and that different domains of the Shc adapter protein may mediate distinct signaling events. A functional relationship between the formation of a Shc/Grb2/Sos complex and TCR-dependent activation of Ras has not been determined. However, a number of laboratories have provided data regarding the formation and stability of the Grb2/Sos complex in T cells. Although Grb2 and Sos are present in a constitutive complex in fibroblasts, TCR ligation appears to enhance the association between Grb2 and Sos (33). This effect may be mediated by the binding of Shc to Grb2, which can enhance the association between Grb2 and Sos (33). It is not known whether Shc is required to mediate the association of Grb2 with Sos in T cells, but it is possible that Shc serves to increase the pool of Grb2-associated Sos following TCR ligation. Another difference between T lymphocytes and fibroblasts is found at the level of down-regulating Ras activity. It has been well established that both growth factor receptor and TCR ligation lead to a transient activation of Ras. The state of Ras activation is determined by the opposing activity of guanine nucleotide exchange factors, such as Sos, and GTPase activating proteins, which catalyze the endogenous hydrolyzing activity of Ras (39). One additional mechanism whereby Ras activity may be down-regulated following growth factor receptor ligation involves the phosphorylation-dependent dissociation of Sos from Grb2, which effectively removes Sos from the membrane (40, 41). In fibroblasts, this mechanism has been shown to be dependent on the MEK family of tyrosine/threonine kinases (42). Interestingly, TCR ligation does not lead to dissociation of the Grb2-Sos complex (33, 43), despite the fact that Sos is rapidly phosphorylated following TCR engagement (43). This suggests that phosphorylation of Sos is not sufficient for its dissociation from Grb2. However, it has been reported that MAP kinase–dependent phosphorylation of Sos following phorbol ester treatment of peripheral blood T lymphoblasts modulates the ability of the Grb2/Sos complex to associate with other phosphoproteins, including membrane-associated pp36 (44). Thus, in T lymphocytes, it is possible that phosphorylation of Sos, although having no direct effect on Grb2 binding, prevents proper recruitment and localization of the Grb2/Sos complex.

Grap and Shb Several recent reports describe the characterization of two additional adapter proteins that may function to mediate proximal TCR signals. Grap is a recently

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

94

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

CLEMENTS ET AL

cloned adapter protein that contains a single SH2 domain flanked by SH3 domains and exhibits 58% sequence identity to Grb2 (45, 46) (Table 1). Grap is expressed predominantly in lymphoid tissues and binds to tyrosinephosphorylated Shc following TCR ligation in a manner analogous to Grb2. Grap also interacts via its SH3 domains with Sos, dynamin (a GTPase involved in membrane protein localization) (47), and Sam 68 (a nuclear RNA–binding protein that serves as a substrate of Src kinases during mitosis) (48). Given the known functions of the various Grap-associated proteins identified to date, Grap-based complexes may participate in antigen-stimulated endocytosis of the TCR complex, cellular proliferation, and cell cycling. It will be of interest to determine whether Grb2 and Grap function in redundant TCR-dependent signaling pathways or recruit unique signaling molecules that promote distinct distal signals. Shb was originally identified as a serum-responsive gene in an insulinproducing cell line (49). Shb mRNA is expressed ubiquitously and contains two potential initiating methionine codons, giving rise to two Shb isoforms. Both isoforms contain a single COOH-terminal SH2 domain and a NH2-terminal proline-rich sequence, but they display no apparent effector domains (Table 1). Thus, Shb has been proposed to function as an adapter molecule. Following TCR ligation in the Jurkat T cell line, Shb binds to tyrosine phosphorylated ζ chain and a tyrosine phosphoprotein of 36/38 kDa, presumably LAT (see below) (50). Interestingly, although interaction with ζ is dependent on the SH2 domain of Shb, binding to the 36/38-kDa phosphoprotein is thought to be mediated by a central, PTB-like domain of Shb. Interestingly, overexpression of Shb in Jurkat cells potentiates the basal tyrosine phosphorylation levels of Shbassociated pp36/38 and pp70, whereas a point mutation within the Shb SH2 domain (R522K) diminishes TCR-dependent protein phosphorylation (50). These data suggest that Shb may function to couple TCR-dependent tyrosine kinases with potential downstream substrates.

pp36/LAT In addition to the complexes described above, it is clear that Grb2 mediates the formation of several additional distinct signaling complexes following TCR ligation, each of which may couple the TCR with different downstream signaling cascades. Other Grb2-associated molecules identified following TCR ligation include tyrosine phosphoproteins of approximately 36/38 kDa, 76 kDa, and 116 kDa (51–53). The 36/38-kDa Grb2-associated phosphoprotein was initially characterized as a membrane-associated protein capable of binding the SH2 domain of phospholipase Cγ 1 (PLCγ 1) or Grb2 following TCR ligation (53, 54). The functional relevance of the pp36/38–PLCγ 1 association was suggested by experiments in which a chimeric tyrosine phosphatase was

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ADAPTER PROTEINS IN TCR SIGNALING

95

engineered to selectively recruit pp36/38 and maintain this protein in a nonphosphorylated state (55). The generation of soluble inositol phosphates and intracellular calcium flux was significantly impaired following TCR ligation in Jurkat cells expressing this chimeric protein, despite induced tyrosine phosphorylation of PLCγ 1. This suggests that pp36/38 may play a role in modulating PLCγ 1 activity following TCR ligation. More definitive evidence for the involvement of pp36/38 in TCR-dependent recruitment of effector molecules to the membrane came following the purification, cloning, and further characterization of pp36/38. The pp36/38 cDNA encodes a 233–amino acid protein containing a putative transmembrane domain and numerous potential tyrosine phosphorylation sites but no SH2, SH3, or enzymatic domains (56) (Table 1). Transcripts specific for pp36/38 are restricted to tissues of hematopoietic origin and are found in T cell, NK cell, and mast cell lines but not in B cells or monocytes. Given the potential role for pp36/38 in TCR-dependent signaling, this molecule was named linker of activated T cells, or LAT. Antiserum raised against LAT specifically stains the membrane of Jurkat T cells, which confirms the membrane localization of pp36/38. When coexpressed in fibroblasts with the Syk tyrosine kinase, LAT is phosphorylated on tyrosines. Likewise ZAP-70, when cotransfected with Lck, is also capable of phosphorylating LAT, which suggests that LAT is a substrate of ZAP-70 following TCR ligation in T cells. Mutation of LAT tyrosines at position 171 and 191 to phenylalanine has a dominant negative effect, as evidenced by the ability of the mutant protein to inhibit TCR-dependent activation of the NF-AT and AP-1 transcription factors. These tyrosine residues appear to mediate the recruitment of a number of effector molecules, including Grb2, PLCγ 1, and the p85 subunit of phosphatidylinositol 30 -kinase (Figure 1). In turn, a number of additional effector molecules, including Sos, may also be recruited to the membrane by virtue of association with proteins that bind directly to phosphorylated LAT (51, 56). The recruitment of signaling complexes to LAT provides an attractive potential mechanism for signal amplification, second messenger generation, and Ras activation following TCR ligation.

SLP-76 The 76-kDa Grb2-associated phosphoprotein has also been purified and its cDNA cloned. The cDNA encodes a protein with no obvious enzymatic activity but which contains several distinct domains, including an amino-terminal acidic domain, a proline-rich central region, and a COOH-terminal SH2 domain (57) (Table 1). Northern blot analysis reveals restricted expression to thymus, lymph node, spleen, and cells of hematopoietic origin. Given the structure and expression pattern of this molecule, it was designated SH2 domain containing leukocyte protein of 76 kD, or SLP-76. Initial studies with SLP-76

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

96

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

CLEMENTS ET AL

Figure 1 Potential roles for the adapter proteins Grb2, linker of activated T cells (LAT), and SH2domain leukocyte protein of 76 kDa (SLP-76) in mediating positive T cell receptor (TCR) signals. Following TCR ligation, ZAP-70 is activated and phosphorylates a number of substrates, including LAT and SLP-76. Phosphorylated LAT then recruits a number of molecules to the cell surface, including Grb2 and phospholipase-Cγ 1 (PLCγ 1). Recruitment of Grb2 to LAT may localize Grb2-associated Sos to Ras, resulting in Ras activation and the initiation of distal signaling events. Phosphorylated SLP-76 recruits Vav, although the functional consequence of this association remains unknown. (Stripes) SH2 domains; (stipples) SH3 domains. PKC, Protein kinase C; DAG, diacylglycerol; PIP2, phosphatidylinositol-4,5-bisphosphate; IP3, inositol trisphosphate.

focused on a potential role in TCR-mediated signal transduction. To this end, it was demonstrated that transient overexpression of SLP-76 in the Jurkat T cell line markedly potentiates TCR signaling, as measured by AP-1– or NF-AT–dependent reporter systems (58, 59). Analysis of more proximal signals demonstrated that overexpression of SLP-76 expression enhances TCR-coupled extracellular regulated kinase (ERK) activation, as determined by monitoring the phosphorylation and enzymatic activity of a cotransfected, epitope-tagged ERK molecule (59). A more definitive role for SLP-76 has been defined by the generation of a SLP-76 deficient Jurkat T cell line. These cells manifest a marked reduction in PLCγ 1 tyrosine phosphorylation, calcium flux, and ERK activation following TCR ligation, implicating SLP-76 in each of these biochemical events (60). As a result, TCR-dependent activation of AP-1– and NF-AT–dependent reporter genes is deficient in this cell line. Together, these data strongly suggest that SLP-76 plays an important role in mediating both ERK activation and the release of intracellular calcium following TCR ligation (Figure 1).

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ADAPTER PROTEINS IN TCR SIGNALING

97

Given the observation that SLP-76 is not an enzyme, it seems likely that additional SLP-76–associated proteins are required for coupling SLP-76 with downstream signaling events. Two additional tyrosine phosphoproteins, with apparent molecular weights of 62,000 and 130,000, are precipitated from stimulated T cell lysates with the SH2 domain of SLP-76 (58). The cDNA encoding the 130-kDa protein has since been cloned and characterized and is discussed in more detail below. To date, the 62-kDa protein remains unidentified. The SH2 domain of SLP-76 also precipitates serine/threonine kinase activity from TCR-stimulated T cell lysates, although the identity of the kinase remains unknown (58). Several groups have reported the inducible association of the SH2 domain of Vav (a guanine nucleotide exchange factor for the Rac/Rho family of small GTP-binding proteins) with the acidic NH2-terminus of SLP-76 in a phosphotyrosine-dependent manner (61–64). The notion that these and/or additional SLP-76–associated proteins are important for mediating SLP-76 function is supported by the observation that deletion of the acidic NH2-terminus, the Grb2 binding site, or the SH2 domain of SLP-76 abrogates the ability of overexpressed SLP-76 to augment TCR-dependent NF-AT activation (59). The rapid tyrosine phosphorylation of SLP-76 following TCR engagement and the subsequent recruitment of Vav focused attention on mapping the specific tyrosines phosphorylated following TCR ligation and their role in mediating SLP-76 function. The NH2-terminal acidic domain of SLP-76 harbors two tyrosine residues (Y113, Y128) in a repeated motif (D-Y-E-S-P) and a third tyrosine (Y145) in a similar motif (D-Y-E-P-P) (57). Substitution with phenylalanine at positions 113 and 128 significantly diminishes tyrosine phosphorylation of SLP-76 following TCR ligation, suggesting that these tyrosines are the primary targets of TCR-coupled tyrosine kinases (65). Interestingly, this dual mutation (Y113/128F) or mutation of all three tyrosines (Y113/128/145F) also diminishes the ability of SLP-76 to augment NF-AT or the IL-2 promoter in response to TCR ligation, which implies that tyrosine phosphorylation is required for optimal SLP-76 function (65, 66). Although mutation of either tyrosine alone does not significantly reduce the ability of SLP-76 to activate NF-AT (65), Vav binding is markedly diminished in the context of these single tyrosine substitutions, demonstrating that both tyrosines are required for this association (N Fang, G Koretzky, submitted). These data suggest additionally that Vav binding is not required for the SLP-76–dependent activation of NF-AT. Given the demonstrated role of Vav in catalyzing the activation of Rac-1 (67), the SLP-76–Vav complex may function in the regulation of this enzyme and subsequent downstream signals, including those that may lead to changes in actin polymerization and cytoskeletal organization. A number of investigators are currently exploring these possibilities.

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

98

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

CLEMENTS ET AL

Numerous studies have shown that SLP-76 plays an important role in translating TCR-coupled tyrosine kinase activation into more distal signaling events. The generation of a SLP-76–deficient mouse strain has now provided direct evidence that SLP-76 is also required for T lymphocyte development. The expression of SLP-76 is regulated during thymocyte development, with highest expression detected at (a) an early stage of maturation that correlates with pre–TCR-dependent activation and (b) a late stage of thymic development just following the onset of mature TCR-dependent selection events (68). In SLP76–deficient mice, thymocyte development is arrested at the CD25+CD44− stage of early thymic development, resulting in the absence of T lymphocytes in the peripheral lymphoid compartments of these mice (69, 69a). A similar maturational block is also observed in mice with recombinase-activating gene (RAG)-1−/− (70). However, unlike RAG-1−/− mice, SLP-76–deficient thymocytes retain the capacity to rearrange TCRβ genes and express preTα mRNA. These observations suggest that the developmental block arises as a consequence of failed pre-TCR signaling, which is normally required for transition to the CD25−CD44− stage of development and further maturation. Null mutations of either the Lck or ZAP-70 tyrosine kinases, although having more severe effects on later stages of thymic development, do not block maturation at the CD25+CD44− stage of maturation (14, 71). This suggests that functional redundancy exists at the level of tyrosine kinase activation following pre-TCR ligation. In support of this notion, mice deficient either for both Src family tyrosine kinases, Lck and Fyn, or for both Syk family members, ZAP70 and Syk, manifest a developmental arrest at the CD25+CD44− stage of development (72, 73). SLP-76 has been implicated as a substrate of ZAP-70 following ligation of the TCR (64, 66). It is likely that a similar biochemical event occurs following ligation of the pre-TCR complex that can be catalyzed by either Syk or ZAP-70 and is required for further thymocyte maturation. Interestingly, unlike the Src and Syk family tyrosine kinases, there appears to be no functional redundancy at the level of SLP-76 following engagement of the pre-TCR. Additional evidence for SLP-76 functioning to couple Syk family tyrosine kinases with more distal signaling events has come from platelets isolated from SLP-76–deficient mice. The collagen receptor glycoprotein VI is coupled with Syk through the ITAM-containing FcR γ -chain (74). Collagendependent platelet aggregation and granule release is lost in the absence of SLP-76 (J Clements, J R-Lee, G Koretzky, submitted). Thus, SLP-76 is a likely substrate of Syk and a required component of collagen-induced signaling pathways in platelets. It will be of interest to determine the precise role of SLP-76 in mediating pre-TCR and collagen receptor–dependent signals and to identify additional adapter proteins that may also regulate these processes.

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

ADAPTER PROTEINS IN TCR SIGNALING

99

ADAPTER PROTEINS THAT INTERFERE WITH TCR SIGNALING

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Cbl and Crk In addition to serving as positive regulators of TCR-dependent signaling pathways, there is increasing evidence that adapter proteins may also serve to interfere with TCR signaling and may function in some cases to mediate signals that promote anergy (Table 2). The 116-kDa Grb2-associated tyrosine phosphoprotein visualized following TCR ligation has been identified as the product of the proto-oncogene c-cbl (75). Cbl was originally characterized as the cellular homologue of a B cell lymphoma inducing retroviral oncogene (v-Cbl) (76). Cbl contains an NH2-terminal PTB domain, a proline-rich region, and several tyrosine residues that are phosphorylated following TCR ligation (75, 77). The observation that the Caenorhabditis elegans homologue of Cbl (Sli-1) prevents Ras activation following Let-23 ligation suggested that Cbl may play a negative regulatory role in additional cell types as well (78, 79). In support of this notion, overexpression of Cbl in a T cell line diminishes AP-1 activation following TCR ligation (80). However, the mechanism(s) by which Cbl may impact negatively upon TCR-dependent signaling remains unclear. Cbl binds to the NH2-terminal SH3 domain of Grb2 in a constitutive manner, with experimental evidence suggesting that the association of Grb2 with Cbl precludes binding of Sos (81). Thus, sequestration of Grb2 by Cbl may prevent recruitment of Sos to the plasma membrane and the subsequent activation of Ras. Alternatively, Cbl may play a more direct role in modulating TCR-dependent signals by recruiting distinct effector molecules that function as negative regulators of TCR signaling. In support of this notion, TCR ligation and subsequent tyrosine phosphorylation of Cbl promotes dissociation from Grb2 and the recruitment of additional molecules to

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

100

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

CLEMENTS ET AL

Figure 2 Potential roles for the adapter proteins Crk and Cbl in mediating T cell anergy or negative signals following T cell receptor (TCR) ligation. In anergic T cells, Cbl preferentially associates with Crk and the Rap-1–specific guanine nucleotide exchange factor C3G. Activated Rap-1 may sequester Raf, preventing propagation of Ras-mediated signals. Cbl also associates with the p85 subunit of phosphatidylinositol 3-kinase (PI-3 kinase) following TCR ligation. Given the demonstrated role of PI-3 kinase as a negative regulator of TCR signaling, the Cbl/PI-3 kinase complex may function to down-regulate TCR-dependent activation. SH2 and SH3 domains are indicated as in Figure 1.

Cbl, including the p85 subunit of phosphatidylinositol 30 -kinase (PI-3 kinase) (81, 82) (Figure 2). Although PI-3 kinase has been described as a positive regulator of IL-2 receptor–dependent signaling (83), transfection of a T cell line with a constitutively active PI-3 kinase mutant diminished TCR-dependent activation of NF-AT (84). In contrast, a dominant negative isoform of PI-3 kinase potentiated NF-AT activation in response to TCR ligation. However, the role of Cbl in mediating these negative effects of PI-3 kinase is not known. It has been well established that ligation of the TCR, in the absence of appropriate costimuli, induces a state of nonresponsiveness or anergy (85). A role for the adapter protein Cbl in the maintenance of T cell anergy has also been described. In addition to association with Grb2, it is now evident that Cbl can bind to Crk family adapter proteins following TCR ligation (82, 86, 87). Like Grb2, CrkL is composed of a single SH2 domain flanked by two SH3 domains (88). In contrast to Grb2, CrkL recruits tyrosine-phosphorylated Cbl following TCR ligation via the Crk SH2 domain. In anergic T cells, CrkL appears to associate preferentially with Cbl and the guanine nucleotide exchange factor C3G (89). C3G has been demonstrated to catalyze GDP exchange on the Ras family

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

12:53

QC: NBL/UKS

T1: NBL

Annual Reviews

AR078-04

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ADAPTER PROTEINS IN TCR SIGNALING

101

member Rap-1 (90). One current model suggests that the preferential formation of a CrkL-Cbl-C3G complex and the subsequent activation of Rap-1 would sequester Raf-1, the kinase immediately distal to Ras (91), short-circuiting Rasdependent signaling and ultimately promoting an anergic state (89) (Figure 2). In support of this notion, several studies have demonstrated that TCR ligation in tolerized T cell clones fails to couple with Ras activation and Ras-dependent distal signaling pathways (92, 93). The Crk-associated protein Cas-L has also been implicated as a potential regulator of Crk-C3G–mediated signaling in T lymphocytes (94). Cas-L was initially identified as a 105-kDa protein that is phosphorylated following β1 integrin cross-linking in T cells (95). Cas-L binds to the COOH-terminal domain of focal adhesion kinase (Fak) via the Cas-L NH2-terminal SH3 domain and is a substrate of Fak and Src family tyrosine kinases following β1 integrin cross-linking (96). Thus, Cas-L may function to promote signals generated at focal contact points between lymphocytes and the extracellular matrix. Following TCR/CD3 ligation, CasL is tyrosine phosphorylated and binds to the SH2 domain of Crk (97). C3G can also be found in a complex with Crk and Cas-L via association with the SH3 domains of Crk (94). Thus, in a manner analogous to the Crk/Cbl complex, Crk/Cas-L complexes may regulate Rap-1 activity via C3G, although a direct role for Cas-L in promoting anergy has not been described. Alternatively, these distinct Crk-based signaling complexes may have different effects on T cell activation and anergy. Recent evidence suggests that Cas-L may be phosphorylated by different families of tyrosine kinases in a receptor-dependent fashion (94). It will be of interest to determine if the more distal signals mediated by Cas-L are conserved or diverge following ligation of different T lymphocyte surface receptors.

SLAP-13O/Fyb and SKAP-55 In a manner analogous to Grb2, SLP-76 has been used as a probe for the identification of proteins that may function to regulate TCR signaling. Two phosphoproteins bind inducibly to the SLP-76 SH2 domain following TCR ligation: an unidentified 62-kDa protein and the SLP-76–associated phosphoprotein of 130 kDa, or SLAP-130 (58, 98). Although the primary sequence of SLAP-130 fails to reveal any obvious enzymatic function, it likely serves an adapter role because it contains several domains that may mediate associations with other proteins. SLAP-130 contains 16 putative tyrosine phosphorylation sites, a central proline-rich region, a potential nuclear localization sequence, and an SH3-like domain (98, 99). Unlike SLP-76, overexpression of SLAP-130 does not augment TCR-induced activation of NF-AT in Jurkat T cells. Additionally, cotransfection of SLAP-130 and SLP-76 cDNAs abrogates the ability of SLP-76 to augment TCR-dependent NF-AT activation. These data suggest

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

102

12:53

QC: NBL/UKS

Annual Reviews

T1: NBL

AR078-04

CLEMENTS ET AL

that SLAP-130 may serve as a negative regulator of TCR-mediated signaling in this model system. A cDNA clone encoding the murine 130-kDa SLP-76–associated protein was independently isolated following purification of a protein that associates with the SH2 domain of Fyn (99). Interestingly, although this Fyn-binding protein (Fyb) is identical to SLAP-130, overexpression of the Fyb cDNA increases TCR-induced IL-2 secretion approximately three- to fivefold in DC27.10 murine hybridoma cells. The mechanism(s) by which SLAP-130 interferes with SLP76–dependent NF-AT activation in one experimental system and potentiates TCR signaling in another is not clear. It is possible that these differences arise as a consequence of the stimulation conditions or functional assays utilized in the two studies. It is also possible that the level of overexpression obtained following transient transfection of SLAP-130 in Jurkat T cells, which was significantly greater than that obtained following stable transfection of Fyb in the murine hybridoma, may impact TCR signaling in different ways. In support of this notion, transfection of increasing concentrations of Fyb cDNA correlates with decreased IL-2 production in the DC27.10 model system (99). Another intriguing potential explanation for the differences observed in the studies utilizing SLAP-130 or Fyb cDNA is that differential expression of SLAP-130/Fyb–associated proteins in the different cell types used may be responsible for dictating the signaling outcome. In addition to the inducible association of SLAP-130 with SLP-76, SLAP-130 constitutively associates with another adapter molecule, SKAP-55 (for Src kinase-associated phosphoprotein of 55 kDa) (100), via the COOH-terminal SKAP-55 SH3 domain and the SLAP-130 proline-rich region (100a). SKAP-55 also possesses an NH2terminal pleckstrin homology (PH) domain that mediates constitutive association with Fyn and several tyrosines in motifs that are predicted to bind to the SH2 domain of src family PTKs. Although it is likely that SKAP-55 functions as an adapter protein, the precise function of SKAP-55 remains unclear. Interestingly, a recently identified homologue of SKAP-55 (SKAP-HOM) also binds to SLAP-130 but not to Fyn (100b). SKAP-HOM expression is restricted to thymocytes and peripheral blood lymphocytes but is not found in transformed cells such as Jurkat, whereas SKAP-55 is found in all these cell types (B Schraven, personal communication). Thus, it is possible that the different signaling outcomes obtained with SLAP-13O/Fyb cDNAs are dictated by differential expression of SKAP-55, SKAP-HOM, or other SLAP-130/Fyb binding proteins.

AP-1 and AP-2 In addition to directing the assembly of specific intracellular signaling complexes, a role for adapter proteins in down-regulating cell activation via the internalization of lymphocyte signaling receptors has been described recently.

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

12:53

QC: NBL/UKS

T1: NBL

Annual Reviews

AR078-04

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ADAPTER PROTEINS IN TCR SIGNALING

103

Serine phosphorylation of the CD3 γ chain is associated with internalization of the TCR/CD3 receptor complex (101). These phosphorylation events have been proposed to induce a conformational change that exposes a DxxxLL motif in the cytoplasmic domain of CD3 γ that then recruits the multimeric adapter proteins AP-1 and AP-2 (102). AP-1 and AP-2 have been shown to regulate the internalization and the subsequent termination of cell-surface receptor signaling by promoting the translocation of the receptor cluster and targeting this complex to clathrin-coated vesicles (103, 104). It is thought that AP-1 and AP-2 binding to CD3 would likewise target the TCR/CD3 receptor complex to these vesicles. In addition to regulation of CD3 internalization, AP-2 has also been implicated in the regulation of the cell-surface expression of T cell coreceptors such as CTLA-4 (105). The association of the TCR signaling complex and additional immune receptors with the AP-1 and/or AP-2 adapter proteins may promote desensitization of the response via removal of the these receptor complexes from the surface of the cell.

CONCLUDING REMARKS Adapter proteins function as key signaling intermediates in coupling proximal TCR-dependent signals with more distal biochemical pathways. The inducible formation of multimeric adapter protein–based signaling complexes provides an attractive mechanism for the generation of diverse signaling pathways and the amplification of receptor-initiated signals. To date, significant progress has been made in the elucidation of TCR-coupled, adapter protein–mediated signaling that has filled gaps in our understanding of how proximal events are translated into distal signaling pathways. Although the identification and biochemical characterization of novel adapter proteins has provided insight into how TCR-activated tyrosine kinases are coupled with distal signaling pathways, the nature of adapter protein function requires consideration of the dynamic assembly of diverse signaling complexes, each of which may be capable of mediating separate signaling pathways. Thus, understanding how signaling cascades are integrated will require the identification of additional adapter protein–coupled effector molecules as well as defining how adapter protein– based signaling complexes are assembled and the subcellular localization of these complexes following TCR ligation. ACKNOWLEDGMENTS We thank Dr. Kevin Latinis for critical discussion and Drs. Burkhart Schraven and Arthur Weiss for communicating results prior to publication. Visit the Annual Reviews home page at http://www.AnnualReviews.org

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

104

12:53

QC: NBL/UKS

T1: NBL

Annual Reviews

AR078-04

CLEMENTS ET AL

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Literature Cited 1. Chan A, Mak TW. 1989. Genomic organization of the T cell receptor. Cancer Detect. Prev. 14:261–67 2. Willerford DM, Swat W, Alt FW. 1996. Developmental regulation of V(D)J recombination and lymphocyte differentiation. Curr. Opin. Genet. Dev. 6:603–9 3. de la Hera A, Muller U, Olsson C, Isaaz S, Tunnacliffe A. 1991. Structure of the T cell antigen receptor (TCR): two CD3 epsilon subunits in a functional TCR/ CD3 complex. J. Exp. Med. 173:7–17 4. Manolios N, Letourneur F, Bonifacino JS, Klausner RD. 1991. Pairwise, cooperative and inhibitory interactions describe the assembly and probable structure of the T-cell antigen receptor. EMBO J. 10:1643–51 5. Reth M. 1989. Antigen receptor tail clue. Nature 338:383–84 6. Chan AC, Shaw AS. 1996. Regulation of antigen receptor signal transduction by protein tyrosine kinases. Curr. Opin. Immunol. 8:394–401 7. Gauen LK, Zhu Y, Letourneur F, Hu Q, Bolen JB, et al. 1994. Interactions of p59fyn and ZAP-70 with T-cell receptor activation motifs: defining the nature of a signalling motif. Mol. Cell. Biol. 14:3729–41 8. Weiss A, Littman DR. 1994. Signal transduction by lymphocyte antigen receptors. Cell 76:263–74 9. Qian DI, Griswold-Prenner I, Rosner MR, Fitch FW. 1993. Multiple components of the T cell antigen receptor complex become tyrosine-phosphorylated upon activation. J. Biol. Chem. 268:4488–93 10. Wange RL, Malek SN, Desiderio S, Samelson LE. 1993. Tandem SH2 domains of ZAP-70 bind to T cell antigen receptor zeta and CD3 epsilon from activated Jurkat T cells. J. Biol. Chem. 268: 19797–801 11. June CH, Fletcher MC, Ledbetter JA, Schieven GL, Siegel JN, Phillips AF, Samelson LE. 1990. Inhibition of tyrosine phosphorylation prevents T-cell receptor-mediated signal transduction. Proc. Natl. Acad. Sci. USA 87:7722–26 12. Graber M, June CH, Samelson LE, Weiss A. 1992. The protein tyrosine kinase inhibitor herbimycin A, but not genistein, specifically inhibits signal transduction by the T cell antigen receptor. Int. Immunol. 4:1201–10 13. Straus DB, Weiss A. 1992. Genetic

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

evidence for the involvement of the lck tyrosine kinase in signal transduction through the T cell antigen receptor. Cell 70:585–93 Molina T, Kishihara K, Siderovski D, van Ewijk W, Narendran A, Timms E, Wakeham A, Paige CJ, Hartmann KU, Veillette A, Davidson D, Mak T. 1992. Profound block in thymocyte development in mice lacking p56lck. Nature 357:161–64 Stein PL, Lee HM, Rich S, Soriano P. 1992. pp59fyn mutant mice display differential signaling in thymocytes and peripheral T cells. Cell 70:741–50 Chan AC, Kadlecek TA, Elder ME, Filipovich AH, Kuo WL, Iwashima M, Parslow TG, Weiss A. 1994. ZAP-70 deficiency in an autosomal recessive form of severe combined immunodeficiency. Science 264:1599–601 Gelfand EW, Weinberg K, Mazer BD, Kadlecek TA, Weiss A. 1995. Absence of ZAP-70 prevents signaling through the antigen receptor on peripheral blood T cells but not on thymocytes. J. Exp. Med. 182:1057–65 Rothenberg EV, Ward SB. 1996. A dynamic assembly of diverse transcription factors integrates activation and celltype information for interleukin 2 gene regulation. Proc. Natl. Acad. Sci. USA 93:9358–65 Rao A, Luo C, Hogan PG. 1997. Transcription factors of the NFAT family: regulation and function. Annu. Rev. Immunol. 15:707–47 Hodge MR, Ranger AM, de la Brousse FC, Hoey T, Grusby MJ, Glimcher LH. 1996. Hyperproliferation and dysregulation of IL-4 expression in NF-ATpdeficient mice. Immunity 4:397–405 Latinis KM, Norian LA, Eliason SL, Koretzky GA. 1997. Two NFAT transcription factor binding sites participate in the regulation of CD95 (Fas) ligand expression in activated human T cells. J. Biol. Chem. 272:31427–34 Holtz-Heppelmann CJ, Algeciras A, Badley AD, Paya CV. 1998. Transcriptional regulation of the human FasL promoter-enhancer region. J. Biol. Chem. 273:4416–23 Izquierdo MS, Leevers J, Marshall CJ, Cantrell D. 1993. p21ras couples the T cell antigen receptor to extracellular signal-regulated kinase 2 in T lymphocytes. J. Exp. Med. 178:1199–208

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

12:53

QC: NBL/UKS

T1: NBL

Annual Reviews

AR078-04

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ADAPTER PROTEINS IN TCR SIGNALING 24. Lowenstein EJ, Daly RJ, Batzer AG, Li W, Margolis B, Lammers R, Ullrich A, Skolnik EY, Bar-Sagi D, Schlessinger J. 1992. The SH2 and SH3 domaincontaining protein GRB2 links receptor tyrosine kinases to ras signaling. Cell 70:431–42 25. Buday L, Downward J. 1993. Epidermal growth factor regulates p21ras through the formation of a complex of receptor, Grb2 adapter protein, and Sos nucleotide exchange factor. Cell 73:611–20 26. Egan SE, Giddings BW, Brooks MW, Buday L, Sizeland AM, Weinberg RA. 1993. Association of Sos Ras exchange protein with Grb2 is implicated in tyrosine kinase signal transduction and transformation. Nature 363:45–51 27. Blenis J. 1993. Signal transduction via the MAP kinases: proceed at your own RSK. Proc. Natl. Acad. Sci. USA 90: 5889–92 28. Davis RJ. 1995. Transcriptional regulation by MAP kinases. Mol. Reprod. Dev. 42:459–67 29. Ravichandran KS, Lee KK, Songyang Z, Cantley LC, Burn P, Burakoff SJ. 1993. Interaction of Shc with the zeta chain of the T cell receptor upon T cell activation. Science 262:902–5 30. Kavanaugh WM, Williams LT. 1994. An alternative to SH2 domains for binding tyrosine-phosphorylated proteins. Science 266:1862–65 31. Pelicci GL, Lanfrancone L, Grignani F, McGlade J, Cavallo F, Farni G, Nicoletti I, Grignans F, Pawson T, Pelicci PG. 1992. A novel transforming protein (SHC) with an SH2 domain is implicated in mitogenic signal transduction. Cell 70:93–104 32. Blaikie P, Immanuel D, Wu J, Li N, Yajnik V, Margolis B. 1994. A region in Shc distinct from the SH2 domain can bind tyrosine-phosphorylated growth factor receptors. J. Biol. Chem. 269:32031–34 33. Ravichandran KS, Lorenz U, Shoelson SE, Burakoff SJ. 1995. Interaction of Shc with Grb2 regulates association of Grb2 with mSOS. Mol. Cell. Biol. 15:593–600 34. Northrop JP, Pustelnik MJ, Lu AT, Grove JR. 1996. Characterization of the roles of SH2 domain-containing proteins in T-lymphocyte activation by using dominant negative SH2 domains. Mol. Cell. Biol. 16:2255–63 35. Osman N, Lucas SC, Turner H, Cantrell D. 1995. A comparison of the interaction of Shc and the tyrosine kinase ZAP70 with the T cell antigen receptor zeta

36.

37.

38.

39. 40.

41.

42.

43.

44.

45.

46.

47.

105

chain tyrosine-based activation motif. J. Biol. Chem. 270:13981–86 Baldari CT, Pelicci G, Di Somma MM, Milia E, Giuli S, Pelissi PG, Telford JL. 1995. Inhibition of CD4/p56lck signaling by a dominant negative mutant of the Shc adaptor protein. Oncogene 10:1141–47 Ravichandran KS, Igras V, Shoelson SE, Fesik SW, Burakoff SJ. 1996. Evidence for a role for the phosphotyrosinebinding domain of Shc in interleukin 2 signaling. Proc. Natl. Acad. Sci. USA 93:5275–80 Milia E, Di Somma MM, Baldoni F, Chiari R, Lanfrancone L, Pelicci PG, Telford JL, Baldari CT. 1996. The aminoterminal phosphotyrosine binding domain of Shc associates with ZAP-70 and mediates TCR dependent gene activation. Oncogene 13:767–75 Boguski MS, McCormick F. 1993. Proteins regulating Ras and its relatives. Nature 366:643–54 Langlois WJ, Sasaoka T, Saltiel AR, Olefsky JM. 1995. Negative feedback regulation and desensitization of insulinand epidermal growth factor-stimulated p21ras activation. J. Biol. Chem. 270: 25320–23 Waters SB, Holt KH, Ross SE, Syu LJ, Guan KL, Saltiel AR, Karetzky GA, Pessin JE. 1995. Desensitization of Ras activation by a feedback disassociation of the SOS-Grb2 complex. J. Biol. Chem. 270:20883–86 Holt KH, Kasson BG, Pessin JE. 1996. Insulin stimulation of a MEK-dependent but ERK-independent SOS protein kinase. Mol. Cell. Biol. 16:577–83 Zhao H, Li YY, Fucini RV, Ross SE, Pessin JE, Koretzky GA. 1997. T cell receptor-induced phosphorylation of Sos requires activity of CD45, Lck, and protein kinase C, but not ERK. J. Biol. Chem. 272:21625–34 Buday L, Warne PH, Downward J. 1995. Downregulation of the ras activation pathway by MAP kinase phosphorylation of Sos. Oncogene 11:1327–31 Feng GS, Ouyang YB, Hu DP, Shi ZQ, Gentz R, Ni J. 1996. Grap is a novel SH3SH2-SH3 adaptor protein that couples tyrosine kinases to the Ras pathway. J. Biol. Chem. 271:12129–32 Trub T, Frantz JD, Miyazaki M, Band H, Shoelson SE. 1997. The role of a lymphoid-restricted, Grb2-like SH3SH2-SH3 protein in T cell receptor signaling. J. Biol. Chem. 272:894–902 Urrutia R, Henley JR, Cook T, McNiven

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

106

48.

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

49.

50.

51.

52.

53.

54.

55.

56.

57.

12:53

QC: NBL/UKS

T1: NBL

Annual Reviews

AR078-04

CLEMENTS ET AL MA. 1997. The dynamins: redundant or distinct functions for an expanding family of related GTPases? Proc. Natl. Acad. Sci. USA 94:377–84 Lock P, Fumagalli S, Polakis P, McKormick F, Courtmeige SA. 1996. The human p62 cDNA encodes Sam68 and not the RasGAP-associated p62 protein. Cell 84:23–34 Welsh M, Mares J, Karlsson T, Lavergne C, Breant B, Claesson-Welsh L. 1994. Shb is a ubiquitously expressed Src homology 2 protein. Oncogene 9:19–27 Welsh M, Songyang Z, Frantz JD, Trub T, Reedquist KA, Karlsson T, Miyazki M, Cantley LC, Band H, Schoelson SE. 1998. Stimulation through the T cell receptor leads to interactions between SHB and several signaling proteins. Oncogene 16:891–901 Reif K, Buday L, Downward J, Cantrell DA. 1994. SH3 domains of the adapter molecule Grb2 complex with two proteins in T cells: The guanine nucleotide exchange protein Sos and a 75-kDa protein that is a substrate for T cell antigen receptor-activated tyrosine kinases. J. Biol. Chem. 269:14081–87 Motto DG, Ross SE, Jackman JK, Sun Q, Olson AL, Findell PR, Koretzky GA. 1994. In vivo association of Grb2 with pp116, a substrate of the T cell antigen receptor-activated protein tyrosine kinase. J. Biol. Chem. 269:21608–13 Sieh M, Batzer A, Schlessinger J, Weiss A. 1994. GRB2 and phospholipase Cγ 1 associate with a 36- to 38-kilodalton phosphotyrosine protein after T-cell receptor stimulation. Mol. Cell. Biol. 14: 4435–42 Buday L, Egan SE, Viciana PR, Cantrell DA, Downward J. 1994. A complex of Grb2 adaptor protein, Sos exchange factor, and a 36-kDa membrane-bound tyrosine phosphoprotein is implicated in ras activation in T cells. J. Biol. Chem. 269: 9019–23 Motto DG, Musci MA, Ross SE, Koretzky GA. 1996. Tyrosine phosphorylation of Grb2-associated proteins correlates with phospholipase C gamma 1 activation in T cells. Mol. Cell. Biol. 16: 2823–29 Zhang W, Sloan-Lancaster J, Kitchen J, Trible RP, Samelson LE. 1998. LAT: the ZAP-70 tyrosine kinase substrate that links T cell receptor to cellular activation. Cell 92:83–92 Jackman J, Motto D, Sun Q, Tanemoto M, Turck C, Peltz GA, Koretzky GA, Findell PR. 1995. Molecular cloning of

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

SLP-76, a 76kDa tyrosine phosphoprotein associated with Grb2 in T cells. J. Biol. Chem. 270:7029–32 Motto DG, Ross SE, Wu J, HendricksTaylor LR, Koretzky GA. 1996. Implication of the GRB2-associated phosphoprotein SLP-76 in T cell receptor-mediated interleukin 2 production. J. Exp. Med. 183:1937–43 Musci MA, Motto DG, Ross SE, Fang N, Koretzky GA. 1997. Three domains of SLP-76 are required for its optimal function in a T cell line. J. Immunol. 159: 1639–47 Yablonski D, Kuhne MR, Kadlecek T, Weiss A. 1998. Uncoupling of nonreceptor tyrosine kinases from PLCγ 1 in a SLP-76-deficient T cell. Science 281: 413–16 Wu J, Motto DG, Koretzky GA, Weiss A. 1996. Vav and SLP-76 interact and functionally cooperate in IL-2 gene activation. Immunity 4:593–602 Onodera H, Motto DG, Koretzky GA, Rothstein DM. 1996. Differential regulation of activation-induced tyrosine phosphorylation and recruitment of SLP-76 to vav by distinct isoforms of the CD45 protein tyrosine phosphatase. J. Biol. Chem. 271:22225–30 Tuosto L, Michel F, Acuto O. 1996. p95vav associates with tyrosine-phosphorylated SLP-76 in antigen-stimulated T cells. J. Exp. Med. 184:1161–67 Raab M, da Silva AJ, Findell PR, Rudd CE. 1997. Regulation of Vav-SLP-76 binding by ZAP-70 and its relevance to TCR zeta/CD3 induction of interleukin2. Immunity 6:155–64 Fang N, Motto DG, Ross SE, Koretzky GA. 1996. Tyrosines 113, 128, and 145 of SLP-76 are required for optimal augmentation of NFAT promoter activity. J. Immunol. 157:3769–73 Wardenburg JB, Fu C, Jackman JK, Flotow H, Wilkinson SE, et al. 1996. Phosphorylation of SLP-76 by the ZAP-70 protein-tyrosine kinase is required for T-cell receptor function. J. Biol. Chem. 271:19641–44 Crespo P, Schuebel KE, Ostrom AA, Gutkind JS, Bustelo XR. 1997. Phosphotyrosine-dependent activation of Rac-1 GDP/GTP exchange by the vav proto-oncogen product. Nature 385: 169–72 Clements JL, Ross-Barta SE, Tygrett LT, Waldschmidt TJ, Koretzky GA. 1998. SLP-76 expression is restricted to hematopoietic cells of monocyte, granulocyte, and T lymphocyte lineage and is

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

12:53

QC: NBL/UKS

T1: NBL

Annual Reviews

AR078-04

ADAPTER PROTEINS IN TCR SIGNALING

69.

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

69a.

70.

71.

72.

73.

74.

75.

76.

77.

78.

regulated during T cell maturation and activation. J. Immunol. In press Clements JL, Yang B, Ross-Barta SE, Eliason SL, Hrstka RF, Williamson RA, Koretzky GA. 1998. Requirement for the leukocyte specific adapter protein SLP76 for normal T cell development. Science 281:416–19 Pivniouk V, Tsitsikov E, Swinton P, Rathbun G, Alt FW, Geha R. 1998. Impaired viability and profound block in thymocyte development in mice lacking the adaptor protein SLP-76. Cell 94:229–38 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 Negishi I, Motoyama N, Nakayama K, Nakayama K, Senju S, Hatakeyama S, Zhang Q, Chan AC, Loh DY. 1995. Essential role for ZAP-70 in both positive and negative selection of thymocytes. Nature 376:435–38 Cheng AM, Negisihi I, Anderson SJ, Chan AC, Bolen J, Loh DY, Pawson T. 1997. The Syk and ZAP-70 SH2containing tyrosine kinases are implicated in pre-T cell receptor signaling. Proc. Natl. Acad. Sci. USA 94:9797–801 van Oers NS, Lowin-Kropf B, Finlay D, Connolly K, Weiss A. 1996. Alpha beta T cell development is abolished in mice lacking both Lck and Fyn protein tyrosine kinases. Immunity 5:429–36 Watson SP, Gibbins J. 1998. Collagen receptor signalling in platelets: extending the role of the ITAM. Immunol. Today 19:260–64 Donovan JA, Wange RL, Langdon WY, Samelson LE. 1994. The protein product of the c-cbl protooncogene is the 120kDa tyrosine-phosphorylated protein in Jurkat cells activated via the T cell antigen receptor. J. Biol. Chem. 269:22921– 24 Langdon WY, Hyland CD, Grumont RJ, Morse HC. 1989. The c-cbl protooncogene is preferentially expressed in thymus and testis tissue and encodes a nuclear protein. J. Virol. 63:5420–24 Lupher ML, Reedquist KA, Miyake S, Langdon WY, Band H. 1996. A novel phosphotyrosine-binding domain in the N-terminal transforming region of Cbl interacts directly and selectively with ZAP-70 in T cells. J. Biol. Chem. 271:24063–68 Yoon CH, Lee J, Jongeward GD, Sternberg PW. 1995. Similarity of sli-1, a

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

107

regulator of vulval development in C. elegans, to the mammalian protooncogene c-cbl. Science 269:1102–5 Jongeward GD, Clandinin TR, Sternberg PW. 1995. sli-1, a negative regulator of let-23-mediated signaling in C. elegans. Genetics 139:1553–66 Rellahan BL, Graham LJ, Stoica B, DeBell KE, Bonvini E. 1997. Cbl-mediated regulation of T cell receptor-induced AP1 activation. Implications for activation via the Ras signaling pathway. J. Biol. Chem. 272:30806–11 Meisner H, Conway BR, Hartley D, Czech MP. 1995. Interactions of Cbl with Grb2 and phosphatidylinositol 30 kinase in activated Jurkat cells. Mol. Cell. Biol. 15:3571–78 Buday L, Khwaja A, Sipeki S, Farago A, Downward J. 1996. Interactions of Cbl with two adapter proteins, Grb2 and Crk, upon T cell activation. J. Biol. Chem. 271:6159–63 Reif K, Burgering BM, Cantrell DA. 1997. Phosphatidylinositol 3-kinase links the interleukin-2 receptor to protein kinase B and p70 S6 kinase. J. Biol. Chem. 272:14426–33 Reif K, Lucas S, Cantrell D. 1997. A negative role for phosphoinositide 3kinase in T-cell antigen receptor function. Curr. Biol. 7:285–93 Jenkins MK, Mueller D, Schwartz RH, Carding S, Bottomley K, Stadecker MJ, Urdahl KB, Norton SD. 1991. Induction and maintenance of anergy in mature T cells. Adv. Exp. Med. Biol. 292:167– 76 Reedquist KA, Fukazawa T, Panchamoorthy G, Langdon WY, Shoelson SE, Druker BS, Band H. 1996. Stimulation through the T cell receptor induces Cbl association with Crk proteins and the guanine nucleotide exchange protein C3G. J. Biol. Chem. 271:8435–42 Sawasdikosol S, Chang JH, Pratt JC, Wolf G, Shoelson SE, Burakoff SJ. 1996. Tyrosine-phosphorylated Cbl binds to Crk after T cell activation. J. Immunol. 157:110–16 ten Hoeve J, Morris C, Heisterkamp N, Groffen J. 1993. Isolation and chromosomal localization of CRKL, a human crk-like gene. Oncogene 8:2469–74 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 Gotoh T, Hattori S, Nakamura S, Kitayama H, Noda M, Takai Y, Kaibuchi

P1: SAT/PLB

P2: ARS/NBL/ary

January 21, 1999

108

91.

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

92. 93.

94.

95.

96.

97.

98.

12:53

QC: NBL/UKS

T1: NBL

Annual Reviews

AR078-04

CLEMENTS ET AL K, Matsui H, Hatase O, Takahashi H, Takeshi K, Matsude M. 1995. Identification of Rap1 as a target for the Crk SH3 domain-binding guanine nucleotide-releasing factor C3G. Mol. Cell. Biol. 15:6746–53 Moodie SA, Willumsen BM, Weber MJ, Wolfman A. 1993. Complexes of Ras.GTP with Raf-1 and mitogenactivated protein kinase kinase. Science 260:1658–61 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 Ohashi Y, Tachibana K, Kamiguchi K, Fujita H, Morimoto C. 1998. T cell receptor-mediated tyrosine phosphorylation of Cas-L, a 105-kDa Crkassociated substrate-related protein, and its association of Crk and C3G. J. Biol. Chem. 273:6446–51 Minegishi M, Tachibana K, Sato T, Iwata S, Nojima Y, Morimoto C. 1996. Structure and function of Cas-L, a 105-kD Crk-associated substrate-related protein that is involved in beta 1 integrinmediated signaling in lymphocytes. J. Exp. Med. 184:1365–75 Tachibana K, Urano T, Fujita H, Ohashi Y, Kamiguchi K, Iwata S, Hirai H, Morimoto C. 1997. Tyrosine phosphorylation of Crk-associated substrates by focal adhesion kinase. A putative mechanism for the integrin-mediated tyrosine phosphorylation of Crk-associated substrates. J. Biol. Chem. 272:29083–90 Kanda H, Mimura T, Morino T, Hamasaki K, Nakamoto T, Hirai H, Morimoto C, Yazaki Y, Nojima Y. 1997. Ligation of the T cell antigen receptor induces tyrosine phosphorylation of p105CasL, a member of the p130Casrelated docking protein family, and its subsequent binding to the Src homology 2 domain of c-Crk. Eur. J. Immunol. 27:2113–17 Musci M, Hendricks-Taylor L, Motto D, Paskind M, Kamens J, Turck CW, Koretzky GA. 1997. Molecular cloning of SLAP-130, an SLP-76-associated substrate of the T cell antigen receptorstimulated protein tyrosine kinases. J. Biol. Chem. 272:11674–77

99. da Silva A, Li Z, De Vera C, Canto E, Findell P, Rudd C. 1997. Cloning of a novel T-cell protein FYB that binds FYN and SH2-domain-containing leukocyte protein 76 and modulates interleukin 2 production. Proc. Natl. Acad. Sci. USA 94:7493–98 100. Marie-Cardine A, Bruyns E, Eckerskorn C, Kirchgessner H, Meuer SC, Schraven B. 1997. Molecular cloning of SKAP55, a novel protein that associates with the protein tyrosine kinase p59fyn in human T-lymphocytes. J. Biol. Chem. 272:16077–80 100a. Marie-Cardine A, Hendricks-Taylor LR, Boerth NJ, Zhao H, Schraven B, Koretzky GA. 1998. Molecular interaction between the fyn-associated protein, SKAP55 and the SLP-76-associated protein SLAP-130. J. Biol. Chem. 273: 25789–95 100b. Marie-Cardine A, Verhagen AM, Eckerskorn C, Schraven B. 1998. SKAPHOM, a novel adaptor protein homologous to the FYN-associated protein SKAP-55. FEBS Lett. 435:55–60 101. Luton F, Legendre V, Gorvel JP, SchmittVerhulst AM, Boyer C. 1997. Tyrosine and serine protein kinase activities associated with ligand-induced internalized TCR/CD3 complexes. J. Immunol. 158:3140–47 102. Dietrich J, Kastrup J, Nielsen BL, Odum N, Geisler C. 1997. Regulation and function of the CD3gamma DxxxLL motif: a binding site for adaptor protein-1 and adaptor protein-2 in vitro. J. Cell Biol. 138:271–81 103. Boll W, Gallusser A, Kirchhausen T. 1995. Role of the regulatory domain of the EGF-receptor cytoplasmic tail in selective binding of the clathrin-associated complex AP-2. Curr. Biol. 5:1168– 78 104. Ohno H, Aguilar RC, Fournier MC, Hennecke S, Cosson P, Bonifacino JS. 1997. Interaction of endocytic signals from the HIV-1 envelope glycoprotein complex with members of the adaptor medium chain family. Virology 238: 305–15 105. Chuang E, Alegre ML, Duckett CS, Noel PJ, Vander Heiden MG, Thompson CB. 1997. Interaction of CTLA-4 with the clathrin-associated protein AP50 results in ligand-independent endocytosis that limits cell surface expression. J. Immunol. 159:144–51

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:89-108. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999. 17:109–47 c 1999 by Annual Reviews. All rights reserved Copyright °

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

EVOLUTION OF ANTIGEN BINDING RECEPTORS Gary W. Litman Department of Pediatrics, University of South Florida College of Medicine, All Children’s Hospital, St. Petersburg, Florida 33701; e-mail: [email protected]

Michele K. Anderson and Jonathan P. Rast Division of Biology, California Institute of Technology, Pasadena, CA 91125; e-mail: [email protected], [email protected] KEY WORDS:

vertebrate phylogeny, immunoglobulin, T cell antigen receptors, adaptive immunity, gene rearrangement

ABSTRACT This review addresses issues related to the evolution of the complex multigene families of antigen binding receptors that function in adaptive immunity. Advances in molecular genetic technology now permit the study of immunoglobulin (Ig) and T cell receptor (TCR) genes in many species that are not commonly studied yet represent critical branch points in vertebrate phylogeny. Both Ig and TCR genes have been defined in most of the major lineages of jawed vertebrates, including the cartilaginous fishes, which represent the most phylogenetically divergent jawed vertebrate group relative to the mammals. Ig genes in cartilaginous fish are encoded by multiple individual loci that each contain rearranging segmental elements and constant regions. In some loci, segmental elements are joined in the germline, i.e. they do not undergo genetic rearrangement. Other major differences in Ig gene organization and the mechanisms of somatic diversification have occurred throughout vertebrate evolution. However, relating these changes to adaptive immune function in lower vertebrates is challenging. TCR genes exhibit greater sequence diversity in individual segmental elements than is found in Ig genes but have undergone fewer changes in gene organization, isotype diversity, and mechanisms of diversification. As of yet, homologous forms of antigen binding receptors have not been identified in jawless vertebrates; however, acquisition of large amounts of structural data for the antigen binding receptors that are found in a variety of jawed vertebrates has defined shared characteristics that provide unique insight into the distant origins of the

109 0732-0582/99/0410-0109$08.00

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

110

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

rearranging gene systems and their relationships to both adaptive and innate recognition processes.

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

INTRODUCTION The adaptive immune system is enormously complex. Its primary effector molecules are the antigen binding receptors [immunoglobulin (Ig) and T cell receptors (TCR)], which are members of the Ig gene superfamily that encompasses large numbers of related gene families serving diverse functions in separate cell lineages (1). The evolutionary acquisition of the complex set of diversification mechanisms found in the antigen binding receptors of present day mammals can be traced in part by comparisons to homologous genes of progressively more divergent vertebrate species. Superimposition of characters related to both sequence and genomic organization of Ig and TCR on the relatively well-established vertebrate phylogeny has proven critical in achieving our current understanding of their evolution. From both structural and functional perspectives, the evolution of Ig and TCR, which appear to be confined to the jawed vertebrates (Figure 1a), is unique and remarkably complex, as it is associated with particularly large families of diverse genes. Furthermore, the somatic diversification of the antigen binding regions of Ig and TCR, which occurs in all jawed vertebrates, is typically associated with a unique form of genetic rearrangement. In higher vertebrates, a second, unrelated form of somatic gene rearrangement takes place during Ig heavy-chain class switching and is associated with additional somatic diversification of germline genes. Antigen binding receptors also share signaling pathways with the nonrearranging mediators of innate immunity (2, 3). Thus, a broad interpretation of the evolution of antigen binding receptors needs to be framed within the boundaries of a highly diversified system in which effector molecules are selected on the basis of both templated germline diversity and untemplated somatic differences. Although comparisons between Ig-type receptors in widely divergent groups have revealed extraordinary diversity in recombining systems, they simultaneously have defined common features that possibly represent constraining elements in the immune system. Broadscale comparisons can be used to generate hypotheses regarding the evolution of this diversity; however, only comparisons of more closely related species can be used to test their validity. As becomes apparent, the degree of plasticity in Ig gene organization and the mechanisms of diversification represent confounding aspects of a pure sequence comparison approach to the evolution of antigen binding receptors. This variation could not have been predicted (reconstructed in an evolutionary sense) by examining structure (and function) in the mammals alone. Although relatively

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

ANTIGEN BINDING RECEPTORS

111

few phylogenetic groups have been examined, within-group variation can be extreme. Every effort has been made in this review to emphasize common features in the Ig and TCR of diverse vertebrates and to relate differences in gene structure and organization that influence function.

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

TWO MAJOR ANTIGEN BINDING RECEPTOR CLASSES ARE ENCODED BY REARRANGING GENES In addition to Ig and TCR, various members of the Ig superfamily have defined roles in the adaptive immune response, including major histocompatibility complex (MHC) I and MHC II, which recognize and bind portions of antigen and share C1 domain types with Ig and TCR (1). All four classes of genes are present in jawed vertebrates (Figures 1a and 2). The structurally related but functionally distinct Igs and TCRs are considered to be the principal antigen binding receptors (4–6) and are the principal focus of this review. Ig and TCR are both structurally and organizationally related, function at the surface of cells, transduce signals along with coreceptors, and are the products of developmentally programmed rearrangement of segmental elements. Although somatic rearrangement and diversification are distinguishing features of Ig and TCR, even genetic rearrangement is not characteristic of all Ig loci (7, 8). In some species, specialized genetic processes exist for diversifying more limited numbers of recombining elements, whereas other species possess extremely large numbers of recombining elements. Rearranged Ig loci undergo extensive somatic hypermutation, whereas TCR genes tend to be mutated only in specialized cellular microenvironments (9, 10). The overall extent of mutation, of forces driving selection of mutations (i.e. antigen dependence versus independence), and of temporal dependence of the mutation process likewise exhibit considerable interspecies variation (10, 11). Considering the relatively few species that have been characterized, the number of different gene diversification themes that have been identified is extraordinary.

IMMUNOGLOBULIN GENES IN CARTILAGINOUS FISH Heavy-Chain Genes The living cartilaginous fish, comprised of the two major radiations, 1. sharks, skates, and rays and 2. chimaeras and ratfish, are the most phylogenetically distant vertebrate group relative to mammals in which Ig genes have been identified (12–14) (Figures 1a and 1b). The cartilaginous fish diverged from a common

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

112

11:2

QC: ARS/Uks

Figure 1 Phylogenetic relationships among the vertebrates. (a) Relationships among major vertebrate taxa; (b) relationships among the cartilaginous fishes (class Chondrichthyes); (c) relationships among the bony fishes: the ray-finned fishes (Actinopterygii) and the lobe-finned fishes (Sarcopterygii); and (d) relationships among the tetrapods. (Ovals) Approximate divergence times, in millions of years ago, based on paleontologic evidence (15) or molecular comparisons (16); the accuracy of these estimates vary.

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

Figure 1 (Continued )

113

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

114

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

Figure 2 Major events in the evolution of rearranging antigen binding receptors superimposed on a phylogenetic tree of the major vertebrate groups. The placement of events relative to major vertebrate group divergence points is inferred from the data described in this review. Ig, immunoglobulin; IgH, Ig heavy chain; IgL, Ig light chain; NAR, novel antigen receptor; DH, Ig heavy chain diversity region; JH, Ig heavy chain joining region; VH, Ig heavy chain variable region; RAG, recombination activating gene; TCR, T cell receptor; MHC, major histocompatibility complex.

ancestor with the other living jawed vertebrates between 450 and 575 million years ago (MYA) (15, 16). In addition to IgM-type genes (hereafter referred to as IgM), other classes of Ig genes have been identified in these species. Certain features of IgM gene organization, structure, and function have been shown to be shared with higher vertebrate Ig, including the following: segmental rearrangement of V, D, and J elements (12); amino acid identity with the corresponding rearranging elements found in higher vertebrates (7, 17, 18); separate Ig domains in single C region exons; differential processing of secretory and transmembrane (TM) forms (14, 19); and somatic mutation of rearranged genes (18) (see below). However, major differences are evident: (a) Multiplicity of loci, i.e. Ig heavy-chain genes in horned shark (Heterodontus francisci) are encoded in as many as 100 independently functional clusters in the genomic forms V-D1-D2-J-C-TM (20); (b) there is restricted intracluster rearrangement (18); (c) short (typically <350 bp) intersegmental distances exist between rearranging elements (7, 13, 21–23); (d ) an additional D segment exists within clusters

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

115

that increases the potential for junctional diversification (7, 18); (e) there is germline joining of segmental elements in significant numbers of clusters (7, 8, 13) (see below); and ( f ) there is an absence of octamer binding sites and a presence of TCRβ-type CRE elements (24). Furthermore, although the VH genes of tandemly organized loci, i.e. mammals, appear to evolve in a birthand-death turnover process (25), it is unclear whether cluster-type Ig genes are subject to the same diversifying process. A single IgM VH family (VH1) in the horned shark in which the members are ∼90% related at the peptide level accounts for all but one gene cluster (7, 18). The monotypic VHII gene cluster, which is ∼60% identical to VH1 at the predicted peptide level, presumably diverged from a VHI-type gene by a process in which the intervening sequence (IVS) between D1 and D2 was excised and recombined in an inverted orientation, equivalent to a hybrid signal join. This results in a different utilization pattern for D1 and D2, and possible relief from any gene correction processes, thus allowing greater divergence. Analyses of the rearrangements of specific Ig gene clusters in horned shark have revealed typical patterns of junctional diversity and ruled out intercluster recombination of segmental elements. Junctional diversity presumably preceded combinatorial diversity in the evolution of mechanisms that diversify Ig genes (18). The phenomenon of germline joining is unique to the cartilaginous fish. Germline-joined horned shark IgM heavy-chain genes (VD-J-C; VDJ-C), which comprise 50% of the clusters analyzed, resemble their unjoined counterparts in terms of overall exon organization and sequence relatedness (7, 18). Both VD-J-C– and VD-DJ-C–joined germline genes also are found in the IgM genes in little skate (Raja erinacea), which similarly possess nonjoined forms (13). Various lines of evidence indicate that these genes are derived from typical unrearranged ancestral genes (7, 8, 26). IgM gene clusters also have been identified in a separate lineage of cartilaginous fish represented by the ratfish (14). These are found in unjoined and joined forms and are related to the shark and skate IgMs. A variant IgM-type cluster recovered from a ratfish cosmid library possesses a first constant region exon that lacks significant sequence identity with CH1 from the IgM genes found in this species. Duplication of a CH2 exon and loss of a typical CH1 exon accounts for this difference (14). The IgM variant lacks the cysteine residue in CH1 that participates in interchain disulfide bonding with light chain, which suggests that it could function either as a single-folded chain or as a homodimer. This presumably derived character also has been reported in the nurse shark (Ginglymostoma cirratum) novel antigen receptor (NAR) (22, 27) and in camel IgG (28, 29). A completely germline-joined IgM, which has been shown to be expressed in both secretory and TM forms, and an extensive family of pseudogenes, comprising ∼90% of all VH hybridizing segments, also are present in this species (14).

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

116

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

IgX, a second distinct class of cluster-type Ig heavy-chain gene, has been described from skates. VX is ∼60% identical at the nucleotide and amino acid levels to the VH of the cartilaginous fish IgM genes (30). IgX clusters contain two D and one J segment as well as two CX exons followed by a single cysteine-rich exon lacking a TM character (and by inference designated secretory) (21, 30). Multiple IgX (as well as IgM) clusters are present in skate (30). In situ hybridization studies with interphase nuclei demonstrate that both IgM and IgX clusters are found on multiple chromosomes and do not appear to be linked (21). The overall regulation, including possible allelic exclusion, of multiple cluster–type Ig loci, is not understood (24, 26); however, other multigene families encoding diverse specificities are also organized in multiple potentially independently functional loci (31). The complete genomic nucleotide sequence of an IgX locus up to the secretory exon has been reported (21). Although the structures of short regulatory motifs in the J-C IVS of both shark IgM and skate IgX are related closely to those important in the transcriptional regulation of mammalian Ig expression, their relative positions are not conserved (32). The functional significance of these motifs is not yet established (G Warr, D Ross, M Anderson, unpublished observations). Longer IgX-related transcripts have been identified (30) and shown to consist of VX, CX1, and CX2 as well as four additional Ig domains (21). IgW described in sandbar shark (Carcharhinus plumbeus) (33) and NARC (new antigen receptor from cartilaginous fish) described in nurse shark (34) consist of IgX-like V, C1, and C2 regions as well as four additional C terminal regions (C3–C6) (34), which are homologous to the last four exons of the long IgX transcript and exhibit significant identity with the corresponding regions of the nurse shark NAR (22) (see below). The relationship of IgX to IgW, NARC, and NAR in these three phylogenetically disparate species was examined further by characterizing both short and long forms of IgX in the skate; the short form has not been identified in any shark species at the molecular genetic level. The coding regions of the long and short forms of IgX share ∼90% nucleotide identity, with differences concentrated in complementarity determining regions (CDRs) (23). Differential processing of the same locus most likely accounts for the extraordinary identities, which extend to include IVSs between DX elements (23). NAR, a third class of Ig in cartilaginous fish, was identified in the nurse shark using polymerase chain reaction (PCR) that targeted shared sequence motifs in C1-type Ig domains (22). NAR possesses an unusual V region, which is not related closely to those of any particular Ig or TCR V regions, a truncated C1 domain, and four C-terminal Ig domains. Furthermore, NAR V regions, unlike those of conventional Ig and TCR, do not form dimers but presumably function as independent domains (22, 27). NAR possesses the same cluster-type

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

117

genomic organization as IgM and IgX but each cluster contains three versus two D elements. Only four NAR loci have been identified in nurse shark, of which two are expressed and two presumably are pseudogenes. Two of the three D genes are germline-joined in one of the pseudogenes. Two V types are defined on the basis of differences in noncanonical cysteine residues that may stabilize the functional V monomer against variation (35). Cysteine residues in CDR3 are encoded typically by preferred reading frames of rearranged D segments (27). NAR has been identified in another species of shark and in sting ray (M Flajnik, personal communication). NAR undergoes extensive hypermutation (see below). Several, but not all, of these features of NAR, i.e. its function as a single chain and the intermediate TCR-Ig V–like sequences are potential properties of a primordial antigen binding receptor (22, 27); however, the three D segments and extensive hypermutation may be derived characters. Phylogenetic analyses indicate that the last four C-region domains of NAR diverged from an IgX/IgW/NARC/NAR common ancestor after the IgM/IgX divergence but prior to the divergence of the sharks and skates. Presumably, NAR would have existed in a common ancestor of both sharks and rays/skates (23). Alternatively, NAR could represent a chimeric structure, reflecting C terminal elements of IgX and a C1-type domain as well as a V region of divergent origin.

Light-Chain Genes Light-chain genes have been identified in both major radiations of cartilaginous fish. Type I light-chain genes in horned shark and skate cannot be classified as either κ or λ chains (36). In horned shark, light-chain genes exist in multiple, independently functioning clusters consisting of closely linked single VL and JL segments, and a CL exon (37). The type I light-chain genes in horned shark possess a typical regulatory octamer (37), and germline joining has not been detected. In contrast, a multisite PCR priming strategy and analyses of genomic clones in the little skate failed to identify any unjoined type I light-chain genes. Several of these germline-joined genes are expressed and appear to undergo somatic mutation (8, 38). Type II light-chain genes have been characterized in sandbar shark (39–41), two different skate species (42), and ratfish (14). Genomic cloning and PCR analyses suggest that type II light-chain genes in the sandbar shark are germline-joined (40) and are encoded in clusters that can be classified in two sequence groups (41). All type II light-chain genes in skate are also germline-joined (42). Type II light-chain genes have been shown at the protein level to associate with IgM heavy chains in clearnose skate (Raja eglanteria) (M Anderson, unpublished observations). Unlike the heavy-chain genes in the cartilaginous fish, light-chain genes of a given type within a species are either

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

118

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

completely joined or unjoined. Although type I genes can be joined (skate) or unjoined (shark), type II genes are joined in sharks, skates, and ratfish, confounding definition of any simple pattern for the acquisition of the joined character. However, the observations are consistent overall with germline joining being a derived characteristic. Attempts to classify type I and II light-chain genes relative to κ- or λ-type are complicated by the large phylogenetic distances between the cartilaginous fish and the higher vertebrate prototypes, as well as the relatively rapid rates of evolution of antigen binding receptors. However, the type I genes certainly are unlike those of higher vertebrates. The classification of a third type of light chain found in the nurse shark (43) and horned shark (42) as κ-like is well supported by phylogenetic analyses (43, 44). The κ-like genes are not germline-joined. The distinctive character of the three families of Ig light-chain genes will facilitate studies of potential isotypic exclusion.

IMMUNOGLOBULIN GENES IN RAY-FINNED FISH Heavy-Chain Genes Unlike the organization of heavy-chain genes in the cartilaginous fish, an IgM heavy-chain gene is encoded at a single mammalian-like locus in representative species of the chondrostean, holostean, and teleost radiations of the ray-finned fish (Figure 1c). The IgM loci in teleosts such as ladyfish (Elops saurus), channel catfish (Ictalurus punctatus), Southern pufferfish (Spheroides nephelus), and rainbow trout (Oncorhynchus mykiss) consist of multiple VH, DH, and JH elements linked to IgM CH exons that are linked within ∼100 kb. Shorter intersegmental (VH–DH) distances are found in teleost fish Igs than are found in mammalian Igs (45, 46). The constant regions of IgM heavy chains in teleosts are encoded by four CH and two TM exons that are related to IgM heavy-chain genes in higher vertebrates. However, differences have been noted in cysteine distributions and potential glycosylation patterns (47, 48) as well as in the relative hydropathicity of the CH3 domain(s) relative to CH3 in IgM heavy chains of other vertebrates (47). TM exons have been shown to be spliced directly into CH3, producing a shorter membrane-bound form of heavy chain (48, 49). The phenomenon is tissue independent and may be generalized in the teleost fish (50). Both the CH3to-TM mode of differential processing found in teleost fish and the CH4-to-TM pattern found in sharks, amphibians, and mammals (14) are found in the bowfin (Amia calva), a representative of the holostean lineage that diverged prior to the lineage that included the latest common ancestors of the teleosts (Figure 1c) (51). Ig accessory molecules, which possibly relate to signal transduction molecules found in higher vertebrates, are associated with the membrane form

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

119

of Ig in catfish (52). Studies of enhancer function in catfish B cell lines have revealed that the Ig heavy-chain enhancer lies downstream of CH4 and not in the J-C intron. It is possible that enhancer function had to arise in the J-C intron as a precondition to the emergence of class switching in order to avoid deletion of the enhancer during the switch recombination event (53). The potential for generating VH gene diversity by combinatorial joining in ray-finned fish is considerable. In catfish, seven different VH families, multiple D segments, and at least nine JH segments, which are linked within 1.9 kb to CH1, have been described (54). Trout encode at least 11 VH gene families and 10–20 DH segments (55, 56); rearrangement joins in CDR3 are relatively short by comparison with mammals. Although the reduced length complicates assessment of DH contributions, there is no direct evidence that the short CDR3 lengths compromise antibody diversity (57). A second class of Ig heavy-chain gene, which encodes an IgM CH1 domain plus seven additional CH exons, is contiguous with the IgM heavy-chain gene in catfish. A similar situation has been observed in the pufferfish; however, the second isotype contains only six additional CH exons (T Ota, C Amemiya, personal communication). The extent of amino acid sequence identity between the second fish isotype and corresponding mouse and human IgD exons is quite low; however, identity between some CH exons in mouse and human IgD is also very low (58). The chromosomal location of the IgM and IgDlike exons, the predicted alternative splicing, the separate secretory and TM domains, and the coexpression with IgM (µ-type) heavy chains in some B cells support a relationship of the catfish gene to IgD but a hinge region, such as that found in IgD, is not present. The difference in gene organization between the cartilaginous and ray-finned fish is extreme. Although subsequent radiations of vertebrates preserved the overall mammalian-like organization of the Ig heavychain gene locus seen in the ray-finned fish, changes have taken place in its form and function that underscore the unusually fluid character of Ig gene evolution.

Light-Chain Genes The chondrostean radiation, represented by the sturgeon (Acipenser), is the most distant lineage of the ray-finned fish relative to the teleosts (Figure 1c). In this species, multiple VL genes are encoded upstream of JL sequences and a single (or relatively few) CL region(s), a pattern that is distinct from the organization patterns described in the teleosts (59) and reminiscent of the κ light-chain gene locus in mammals. In contrast, Southern blot analyses suggest that one form of Ig light-chain (IgLC1) gene in two teleost species, trout and Atlantic cod (Gadus morhua), is organized in a cluster-type organization, similar to the cluster–type I (unjoined) light-chain genes in horned shark (60); however, in teleost fish, the individual clusters are closely linked (60, 61). A second

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

120

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

light-chain isotype in trout (IgLC2), which is only distantly related to IgLC1, also exhibits cluster-type organization. Two repeat sequences that are found upstream of mammalian Jκ and may represent transcription factor binding sites were identified in sterile IgLC2 transcripts (62). In catfish, two classes of light chains designated G and F, which share 52% and 34% amino acid identity in the V and C regions, respectively, have been identified. The VF region is present in multiple families and is homologous to higher vertebrate Vκ, and VG is closely related to IgLC1. CG and CF domains cannot be unequivocally assigned to other vertebrate light-chain classes (60, 61, 63). Southern blot analyses are consistent with at least 15 G clusters and more than 50 F clusters. The G loci consist of at least two VG elements upstream of two JG elements, one of which is likely to be a pseudogene, and a single CG exon. F loci resemble basic VL-JL-CL cartilaginous fish-like light-chain clusters. As in the G cluster, the VF and JF segments are in opposite transcriptional polarity, indicating rearrangement by inversion rather than by deletion (61, 63). Neither type of cluster is associated with germline joining. The close linkage of the individual clusters could permit intercluster recombination, although this has not been demonstrated. The relationships among light-chain gene organization patterns in the cartilaginous fish, the sturgeon, and several different teleost lineages cannot be reconciled in a simple model (59).

IMMUNOGLOBULIN GENES IN THE LOBE-FINNED FISH Heavy-Chain Genes in a Living Fossil Different forms of organization of Ig heavy-chain genes are associated with different mechanisms of somatic diversification and regulation of gene expression. Specifically, combinatorial rearrangement of segmental elements does not appear to be used to diversify gene clusters in cartilaginous fish (18). Furthermore, unlike Ig gene loci in higher vertebrates, gene rearrangement per se is not an obligate requirement (7, 8, 39, 40). The living coelacanth (Latimeria chulumnae) is a surviving member of an ancient superorder that was long thought to be extinct (Figure 1c). VH and DH segments in coelacanth Ig heavy-chain genes are linked closely (∼190 bp), a similar character to that described in the cartilaginous fish (64). The V-D pairs themselves are in close tandem linkage (3–4 kb), presumably upstream of JH and CH. The putative coding regions of DH contain recombination signal sequences (RSSs) with 12-bp spacers and are related to a “join” of shark-type DH1 and DH2 (7). The apparent absence of additional DH segments (and RSSs) downstream of V-D is consistent with precommitted joining of V-D pairs. Southern blotting of coelacanth DNA using a VH-specific

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

ANTIGEN BINDING RECEPTORS

121

probe reveals a large number of hybridizing elements. All of the genomic VH genes thus far sequenced are members of a single family; cDNA analyses will be required to further assess VH family divergence and to characterize JH and CH contributions.

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Heavy-Chain Gene Isotype Diversity Modern lungfish are a critical group in terms of understanding possible distinctions between Ig genes in ray-finned fish and in representatives of the phylogenetic radiation that includes the ancestors of the amphibian and reptiles (Figure 1d ). Three distinct heavy-chain cDNAs have been identified in African lungfish (Protopterus aethiopicus): (a) an IgM gene encoding four CH region domains, (b) a second type encoding seven CH domains, and (c) a third type encoding two CH domains and a 30 cysteine-rich domain, similar to that in an avian IgY gene (see below). The seven CH domain-containing genes resemble the IgD-like gene in the ray-finned fish discussed above; however, CH1 is not IgM-like. Six distinct VH families have been identified in lungfish (65) (T Ota, C Amemiya, personal communication). By analogy to avians and through general phylogenetic inference, it is possible that the heavy-chain gene class switch arose prior to the divergence of the lungfish and tetrapods. Light-chain gene structures have not been described for either the coelacanth or lungfish.

IMMUNOGLOBULIN GENES IN AMPHIBIANS Heavy-Chain Genes In addition to their phylogenetic position, amphibians are of particular significance as an immunological model system that undergoes metamorphosis (66). The Ig heavy-chain gene locus in the African clawed frog (Xenopus laevis), an anuran amphibian, encodes three types of C regions: IgM, IgX, and IgY. The TM regions of IgX and IgY are more closely related to the TM regions of Xenopus IgM than to mammalian forms, supporting a hypothesis that IgM duplicated to give rise to IgY and IgX (67), which is not related to IgX in cartilaginous fish but rather is considered to be an IgA functional equivalent (68). No evidence has been found, in either developing or mature lymphocytes, for IgD-like genes or for alternative splicing of CH regions. Amphibians are the most divergent species relative to the mammals (Figure 1a), in which a mammalian-type heavy-chain class-switching mechanism has been documented, although this characteristic could be present in lobe-finned fish (Figure 2). The µ-type switch region (S) in Xenopus is ∼5 kb and consists of 23 repeats of ∼150 bp; each repeat consists of shorter internal repeats and palindromic sequences. Class switching occurs after gene rearrangement and is greatly accelerated in the secondary immune response. In IgX-expressing B

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

122

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

cells, the µ-type heavy-chain gene is deleted and Sµ and Sx, which are not related at the sequence level, are joined. As in mammals, switching recombination occurs at microsites formed through repetitive palindromes and is associated with tissue-specific gene expression (69). Eleven different VH gene families, extensively diversified DH genes, and a high level of JH diversity are found in Xenopus (70–72). Certain VH families have multiple members and all but the VHIII and VHVII families have been found to be interspersed (73). Multiple 50 Ig regulatory octamer sequences and octamer-like motifs are found upstream of Xenopus VH genes, which suggests that their regulation could have unique features (70, 73). Restricted utilization of DH sequences and limited N region diversity are associated with early development in Xenopus (74). Axolotol (Ambystoma mexicanum) is a neotenic amphibian, i.e. the mature form of the animal retains larval features. Both IgM and IgY have been identified in axolotol. In axolotol IgY, additional cysteines, as have been found in Xenopus Cν , avian Cν , and human Cε, are found in the Cν 1 and Cν 2 regions. The IgY molecule in axolotol is secretory (75). Three axolotol Ig VH families as well as four DH and six JH elements have been identified (75, 76). Analyses of spleen Ig cDNAs reveal greater numbers of out-of-frame sequences in early embryos than at later developmental stages. A slight increase in the length of CDR3 between subadult and adult axolotols has been reported (77). The CDR3 loop of axolotol VH genes has been interpreted to be less somatically diversified than CDR3 in any developmental stage of mouse or Xenopus. The discovery that VH regions in axolotol have both shorter junctional sequences and extensive conservation of the germline JH sequences possibly relates to the decreased affinities in antibody responses in this species compared with Xenopus and teleost fish (78).

Light-Chain Genes Three classes of light-chain genes have been identified in Xenopus (79–81). The ρ locus contains at least four Vρ and five Jρ elements as well as a single Cρ gene and is classified as κ-like on the basis of both Vρ sequences and organizational features of the ρ locus (44, 80). The σ -type light chain consists of multiple Vσ 1 and Vσ 2 as well as Cσ 1 and Cσ 2 segments (79); both types of genes rearrange to their respective Jσ types. The σ and ρ genes are encoded at separate loci. Expressed Vσ sequences are highly conserved and diversity of σ and ρ light-chain genes in Xenopus may be markedly restricted, as opposed to the heavy-chain genes, which are as diversified, if not more so, than their mammalian counterparts (72). A third light-chain type in Xenopus represents a higher vertebrate λ lightchain homolog. Six distinct Vλ families, two Jλ segments, and two distinct Cλ

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

ANTIGEN BINDING RECEPTORS

123

exons have been characterized. Southern blotting of DNA as well as sequence analysis from individual animals indicates that the Vλand Cλ sequences are diverse and polymorphic (81). The complex diversity in the λ-type light-chain genes argues against diversity restriction in the overall light-chain repertoire in Xenopus.

IMMUNOGLOBULIN GENES IN REPTILES Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Heavy-Chain Genes Southern blotting of reptilian [caiman (Caiman crocodylus) and turtle (Chelydra serpentina)] DNAs indicates large numbers of VH genes that cross hybridize with murine and species-specific VH probes (82) (G Litman, unpublished observation). Screening of a single genomic equivalent of a turtle genomic λ DNA library using homologous VH probes resulted in hybridization to >500 individual clones. At the stringency used, the hybridizing elements could be as little as 60% related to the probe used for screening. Each λ genomic clone typically consists of multiple, linked VH genes; however, many of the hybridizing bands may represent pseudogenes. Overrepresentation of individual clones does not account for the large number of positives. In situ chromosomal hybridization mapped turtle genes to four separate linkage groups (26) (C Amemiya, G Litman, unpublished observation). More recent studies in another species of turtle (Pseudemys scripta) defined four families of VH genes, which are encoded at a single µ-type locus (83). Northern blot analyses using a CH4 probe identified two µ-type transcripts and at least two non–µ-type transcripts. Analyses of rearranged Ig heavy-chain gene transcripts from a single animal suggest large numbers of JH segments and/or extensive somatic mutation in FR4. In terms of the generation of antibody diversity, it is apparent that reptiles contain particularly large numbers of VH genes. Although avians technically are members of the reptilian radiation of vertebrates, their Ig genes represent a marked departure from the typical reptilian organization, as defined in caiman and turtles.

IMMUNOGLOBULIN GENES IN AVIANS Heavy-Chain genes The chicken (Gallus domesticus) is the most comprehensively characterized nonmammalian immunological model system (84–86). Three classes of heavychain genes are present in this species. Class switching occurs from IgM to either IgY or IgA; IgM to IgY switching occurs during the secondary antibody response and is associated with increased antibody affinity (see below). In addition to IgM and IgA (87), two forms of IgY, possessing nearly identical CH1

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

124

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

and CH2 domains, are present in the duck Anas platyrhynchos (88). The shorter transcript encodes CH1 and CH2 plus a unique short exon encoded between CH2 and CH3. The longer transcript encodes CH1–CH4 in either secretory or TM forms. On the basis of inferred intrachain disulfides and the number of secretory exons, the full-length form of IgY resembles IgE; however, TM sequences and similar RNA processing are more characteristic of IgG, which suggests that IgG and IgE diverged from a common ancestral molecule resembling IgY (89, 90). Chickens possess a >60–80 kb heavy-chain gene locus consisting of a single functional VH gene, 16 relatively similar DH segments, and a single JH element, which are arrayed upstream of the CH region genes (84, 91). Only limited combinatorial diversity and little junctional diversity is achieved through differential rearrangement of DH segments. The principal means of diversification is achieved through gene conversion that utilizes a pool of 80–100 50 -truncated upstream VH9 (pseudogenes), which lack RSSs and are in different relative transcriptional polarities. Certain VH9 possess DH-like and JH-like sequences at the 30 ends, and gene conversion can extend into these regions; preferred VH9 exchanges have also been described (92). Furthermore, a high incidence of DH-DH joining and P nucleotide addition also diversifies the VH locus (92). Given the phylogenetic positions of avians as having diverged from a common ancestor with the reptiles after the divergence of the mammals, the avian Ig heavy-chain gene presumably is a derived evolutionary character. Gene conversion also has been documented unequivocally in one mammal (93) and presumably occurs in others (94–96).

Light-Chain Genes The somatic hyperconversion process described above was characterized initially for the single λ-like light-chain locus of the chicken (97, 98). Twenty-six VL9 are upstream of single, closely linked VL and JL segments (97, 99). A variant of this arrangement, in which the pseudogenes are upstream of two functional VL segments, has been described in another avian species (100). The frequency of VL9 use depends on the proximity of the pseudogene and target genes as well as the extent of identity and their relative orientation. Homology in the 50 pseudogene and target is most essential and evidence for polarity in the gene conversion mechanism(s) has been presented (101). In addition to gene conversion, a number of mechanisms, including imprecise joining, somatic point mutation, single nonrandom nucleotide additions (102), and V gene replacement (99), further diversify light-chain genes. The diversification mechanisms found in avians are of particular interest from an evolutionary standpoint in terms of the stability and maintenance of V9 (donors). Comparisons of VL9 and functional alleles between three inbred lines of chicken identified multiple interstrain polymorphisms as well as

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

ANTIGEN BINDING RECEPTORS

125

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

polymorphisms that could arise by meiotic gene conversion. The incidence of mutations within the VL9 is lower than in the corresponding flanking regions, consistent with the hypothesis that the VL9 cluster behaves as a functional multigene family under selective pressure for its role in the diversification of the antibody repertoire (103). The similarities between the organization of the Ig heavy- and light-chain loci in both avian and cartilaginous fish models suggests parallel evolution of both form (organization) and function (mechanisms of diversification) in antigen binding receptors (37).

PHYLOGENETIC ORIGINS OF SOMATIC HYPERMUTATION Targeted somatic hypermutation has been demonstrated in all vertebrate Ig gene systems thus far examined (11) and is undoubtedly an ancient mechanism for the diversification of the antigen-binding sites of Igs, regardless of their genomic organization (8). The degree of somatic hypermutation varies by approximately an order of magnitude among vertebrate groups (11). Somatic hypermutation in mammals is associated with affinity maturation (10), which can be several orders of magnitude greater than in anuran amphibians and teleost fish. Temporal increases in antibody affinity have not been observed in cartilaginous fish (18, 104). The discrepancy between the presence of somatic hypermutation and the absence of affinity maturation in the cartilaginous fish is difficult to reconcile. One explanation is that the absence of germinal centers in cartilaginous fish may result in a lower efficiency of cellular selection of high-affinity mutants (10, 18, 105). However, this supposition raises the question of why targeted somatic mutation would have evolved prior to the development of a selection system. One possibility is that somatic mutation could diversify the preimmune repertoire prior to selection, as occurs in sheep Peyer’s patches (106). However, recent studies indicate that NAR mutations are accumulated after exposure to antigen and are not used to generate the preimmune repertoire (35), which argues against this theory. Although the basis for mutation of the shark NAR gene could be different from the other cluster-type genes, this observation is consistent with NAR mutations being a shared character with mammalian somatic mutation and, thus, ancestral. It is possible that rather than repeated rounds of mutation and expansion in response to antigen receptor signaling, as occurs in mammalian germinal centers (107), lymphocytes in lower vertebrate groups are biased toward terminal differentiation. It is also possible that highaffinity mutants reside in other (unknown) tissue(s) in these animals. Further study is required to relate differences in antigen binding receptor structures in various vertebrate groups to the differences observed in immune function assays.

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

126

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

B CELL MEMORY AND THE SECONDARY IMMUNE RESPONSE The evolution of antigen binding receptors involves organizational and mechanistic change and occurs within the context of highly specialized systems that potentiate function of effector molecules. B and T cell memory, defined in terms of the ability to respond more rapidly and effectively to secondary challenge with a specific antigen, are found in all jawed vertebrates, with the possible exception of the cartilaginous fish (66). Specifically, upon secondary antigenic challenge, horned sharks do not show a dramatic rise in levels of antibody. Rather, the increase requires repeated monthly immunizations and occurs over a much longer time interval than is seen in mammals (108, 109). Both teleost fish and amphibians demonstrate a secondary rise in antibody levels that is greater and more rapid than the primary response, but less vigorous than that seen in mammals. The secondary response is generally 10- to 20-fold higher than the primary response, without a measurable increase in affinity. It has been shown that a low priming dose is required to achieve optimal memory development in carp (Cyprinus carpio) (110). Whether IgM memory in mammals exists is debatable, and thus the role of isotype switching in triggering changes characteristic of the memory response is unclear. The modest secondary response consisting primarily of IgM antibody, which is seen in teleost fish, indicates that a limited form of memory exists in the absence of class switching. Anuran amphibians (such as Xenopus) exhibit a peak in the primary response at 4 weeks. The secondary response peak occurs at 2 weeks postimmunization and is accompanied by class switching from IgM primarily to IgY. The secondary response is associated with 10–40 times greater antibody production (111). However, in urodele amphibians [such as the newt (Notophthalmus viridescens)] class switching does not appear to occur. The peak level of antibody in response to T-dependent antigens does not increase in the secondary response, although the rise in antibody occurs more rapidly (112). Current views on mammalian memory emphasize the importance of the affinity of antigen receptor–antigen interactions in controlling which signaling pathways are activated and, thus, how the functional response is modulated (113). A strong-affinity interaction results in a full activation of all possible signaling pathways, leading to activation of the effector function gene expression program. However, a low-affinity interaction results in partial activation, which sensitizes the cell to further stimulation without inducing proliferation or effector function and allows entry into the memory compartment. Studies of memory B cells (114) support a model in which clonal expansion occurs primarily in cells that are undergoing the effector program response, not in cells that are

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

127

fated to become memory cells. The observation that a low priming is needed to achieve memory development in carp whereas high-priming doses result in poor memory responses suggests a similar mechanism in the development of memory in teleost fish (110). Two key differences between cartilaginous fish and higher vertebrate immunity are the low-affinity Ig receptors that do not mature, and the apparent lack of sensitization leading to a memory response. It is difficult to understand how the shark lymphocytes, which bear low-affinity antigen receptors, become activated to a full effector response unless there are major differences in the way their antigen receptors interact with the signaling apparatus, or unless there is a significant change in the signaling threshold.

THE IDENTIFICATION OF T CELL ANTIGEN RECEPTOR HOMOLOGS IN NONMAMMALIAN VERTEBRATES T cell–mediated immunity was thought to exist in all the phylogenetic groups in which rearranging Ig genes were known to exist. However, with the exception of chicken TCRβ (115), nonmammalian TCRs have proven refractory to isolation by DNA cross hybridization strategies that were used successfully to identify Ig genes in lower vertebrates. The primary basis for this technical disparity most likely relates to the higher rates of sequence divergence among TCRs. High rates of divergence are characteristic of immune genes relative to their nonimmune counterparts (116). The forces driving this rapid divergence are unclear but could stem from a generally lower level of constraint on immune proteins. Alternatively, this rate could result from sequence divergence in response to interactions with molecules—such as the TCRs themselves and MHC proteins—that are under heavy positive selection to vary (117). Paradoxically, as more data are acquired, it appears that although TCR genes are generally diverged in primary sequence, TCRs may be more conserved than Igs in terms of gene organization, diversification mechanisms, and subclass heterogeneity.

T CELL ANTIGEN RECEPTOR GENES IN CARTILAGINOUS FISH The particular interest in TCR genes in cartilaginous fish relates not only to their phylogenetic position but also to functionally unique aspects of immunity in these species, e.g. apparent absence of affinity maturation and chronic allograft rejection (118). A PCR strategy based on shared 3–4 amino acid motifs in FR2 and FR3 of higher vertebrate TCR and VL genes of higher vertebrates (119) was used to amplify candidate gene segments using horned shark cDNA as template.

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

128

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

Sequence analyses of cDNAs selected with the candidate segments identified clones with V, J, and C regions that were most TCRβ-like. Comparisons of cDNA clones reveal (a) a Dβ region core element and (b) that combinatorial joining plays a role TCRβ diversification in shark. Genomic Vβ and Jβ region sequences were shown to have typical RSS elements separated by 23-bp (V) and 12-bp (J) spacers, respectively (119). Southern blotting indicates that Vβ and Cβ gene elements are present in multiple copies yet are organized (unlike the shark Ig heavy-chain genes) in several tandem arrays consisting of multiple Vβ and Jβ elements, similar to mammalian TCR genes. The differences in restriction enzyme digestion patterns of genomic lambda clones containing Cβ Ig domain exons and cDNA analyses are consistent with multiple Cβ region genes (120). However, this multiplicity may not be the rule for TCR genes in cartilaginous fish, as the four classes of TCR genes that have been identified thus far from the skate (see below) are encoded in one or at most two copies (121). Additional information about Vβ diversity in horned shark has been obtained by sequencing large numbers of spleen TCRβ cDNAs. Seven Vβ families, 18 different Jβ sequences, and a putative Dβ core sequence (GGGACAAC) were identified (120). Phylogenetic analyses comparing the shark Vβ sequences to those of other vertebrates suggest that at least some Vβ family divergence occurred prior to the divergence of the cartilaginous fish and the lineage leading to mammals. Two TCRδ-like genes have been identified in horned shark (122). Initially it was unclear whether these genes represented a TCRδ ortholog; however, the later identification of homologs of TCRα and TCRδ in the skate (121) and the relationship of these genes to the horned shark genes suggest that the α/δ divergence occurred prior to or early in jawed vertebrate evolution. The two related but divergent TCRδ homologs as well as the multiple TCRβ loci may relate to genomic polyploidy in horned shark (122). Determination of orthology among mammalian and cartilaginous fish TCR genes is problematic given the divergent nature of TCR genes. A more comprehensive, though by no means exhaustive, approach based on short primer PCR was used to amplify TCR genes from the clearnose skate (121), a welldefined developmental model (123). cDNAs were identified that are similar to the four mammalian TCR gene types in terms of V and C region sequence, absence or presence (inferred) of D regions, and other junctional characteristics. TM regions associated with each of the four TCR types conform to conserved antigen receptor TM motifs described in higher vertebrates (124). Sequencing of representatives of the different V families for each putative gene type is completely consistent with TCR α, β, γ , and δ orthology. Specifically, four skate Vα and six distinct Jα region families were identified. TCRα

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

129

junctional sequences are short, consistent with direct V-J joining. Six skate Vβ region families, two potential Dβ core elements, and four Jβ region families are associated with the Cβ homolog. Although each of the Dβs is G-rich, neither is otherwise similar to the putative Dβ region identified from horned shark TCRβ. Five Vγ and two Jγ families were identified and direct Vγ Jγ joinings were observed, although some cDNAs possess more complex junctions. Five Vδ and two Jδ families have been identified. Three potential Dδ core elements are present in TCRδ VJ junctions. The long junctional lengths are characteristic of mammalian TCRδ genes. Pulsed field gel analyses suggest that Cα and Cδ are linked, although the distance between the elements is probably considerably greater than in mammals.

T CELL ANTIGEN RECEPTOR GENES IN RAY-FINNED FISH TCR genes have been described from four representative species of ray-finned fish. A TCRβ gene from the rainbow trout was cloned using a degenerate PCR amplification strategy with primers that targeted conserved V sequence motifs (125). The sequences of amplified products were used to generate specific primers, and an anchor PCR strategy employing thymocyte cDNA as template was used to clone a corresponding Cβ region. Three trout Vβ families, exhibiting between 30% and 38% pairwise amino acid identity, and 10 different Jβ sequences were described (126). The Jβ segments possess canonical 50 7-mer elements, and although a number of these lack typical 9-mers, they are utilized frequently in TCR rearrangements (127). A core Dβ-like sequence (GGACAGGG) in trout is identical to a corresponding sequence in axolotol, chicken, and mouse Dβ regions. Genomic analyses show that the Dβ segment in trout is flanked by typical RSS elements. Although Southern blotting analyses potentially indicate multiple Cβ-hybridizing elements, only a single Cβ sequence was identified. Junctional lengths of the in-frame trout TCRβ CDR3 regions are somewhat shorter than those observed in mice and chickens and can be accounted for by the relatively short (eight nucleotide) Dβ. Approximately 40% of trout Dβ-Jβ junctions lack N-region nucleotide additions compared with 26% in the mouse. The TCRβ VDJ junctions, along with a portion of Cβ, have been recovered from Atlantic salmon (Salmo salar) using reverse transcription-PCR and leukocyte cDNA as template (128). Only one Cβ region, closely related to trout Cβ, has been fully characterized. Minor sequence differences in salmon Cβ were attributed to allelic variation or Taq error; interpretation of these data is complicated by a recent ancestral tetraploidization event, which is expected to result in two TCRβ loci. Eight Jβ region sequences were isolated, of which

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

130

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

five can be paired readily with trout Jβ. A single Dβ core sequence, which is identical to that of trout, has been inferred, along with junctional diversity. A rainbow trout TCRα homolog has been isolated using anchored PCR and a thymocyte RNA template (129). Cα-specific primers were used to amplify Vα-Jα regions in a 50 -RACE PCR analysis. Of 40 cDNA sequences analyzed, 6 different Vα families and 32 different Jα segments were identified. The high level of J diversity is characteristic of TCRα. Sequence comparisons of the trout Cα reveal a close match with a Southern pufferfish putative TCRα sequence (122). The catfish, in which lymphocytes and immunocyte cell lines are well characterized, has proven to be an excellent model system for studying lower vertebrate immune function (130). TCRα and β have been identified by V region-directed PCR amplification with degenerate primers from peripheral blood leukocyte cDNA. Three Vα families and a single Cα gene as well as five Vβ and two Cβ families were identified. Southern blotting using Vα and Vβ family-specific probes are consistent with additional genes. Seven different Jα and seven Jβ regions have been identified, as have two potential Dβ regions, one of which matches corresponding sequences found in a variety of vertebrate TCRβ genes. Analyses of the transcription patterns of these genes in phenotypically characterized catfish T and B cell lines is consistent with their assignments.

T CELL ANTIGEN RECEPTOR GENES IN AMPHIBIANS The TCRβ genes of Xenopus were isolated by an anchored PCR strategy employing a DNA template from a spleen/thymus cDNA library (131). The single TCRβ-like C region that was identified shares TM region features (124) as well as predicted peptide identity (31%) with axolotol TCRβ. Ten different, highly divergent Vβ gene families and 10 different Jβ segments have been identified. Two putative Dβ contributions, one of which matches the trout, axolotol, chicken, and human Dβ sequences, have been inferred along with a possible third Dβ sequence, similar to that identified in horned shark. N-region diversification most likely occurs in the rearranged junctional regions. RSS sequences for Vβ, 50 -Dβ, and Jβ were identified in partially rearranged transcripts and are similar to those found in other TCR genes. Notably, the X. laevis TCRβ probes do not cross hybridize with DNA from members of the Xenopus tropicalis species group (131), which diverged ∼120 MYA from a common ancestor with the X. laevis group. This lack of sequence similarity is unusual among genes of the Xenopus immune system (Ig, MHC, etc) which typically cross hybridize between these species. The finding is an example of the higher rate of sequence divergence for TCR than Ig genes. Although the genomic

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

131

sequence of a putative TCRα V region from Xenopus has been reported (122), a corresponding cDNA has not been described. The first TCR gene from a nonamniote was identified in the axolotol using an anchored PCR strategy employing degenerate primers complementing a conserved region in TCRβ and Ig light-chain C regions (132). Two closely related Cβ and two additional Cβ sequences were identified. Nine different Vβ families were described, of which the Vβ7.1 gene is utilized most frequently (133). The junctional regions of 189 Vβ7.1 cDNAs from animals at 2.5, 10, and 24 months were compared and equivalent numbers of Cβ1 and Cβ2 positive transcripts were identified along with nine different Jβ elements. Three Jβ1 elements were associated with Cβ1 and six Jβ2 elements were associated with Cβ2. Two different Dβ elements were identified; Dβ1, Jβ1, and Cβ1 elements associate preferentially, as do Dβ2, Jβ2, and Cβ2. Dβ2-Jβ1-Cβ1 and Dβ1-Jβ2-Cβ2 transcripts that could arise by trans- or inversional rearrangements, as well as by rearrangements that skip Jβ and Cβ elements, respectively, were also identified. Several Vβ-Jβ rearrangements were identified that appear to arise either from direct Vβ-Jβ joining or from extensive exonucleolytic trimming. N region addition increases from 40% of junctions in the 2.5-month-old animals, to 73% in the 10- and 24-month-old animals. A PCR strategy employing 30 Vα- and Cα-TM specific primers was used to identify a TCRα homolog in the axolotol. Five different Vα families and 14 different Jα elements have been identified. As with TCRβ from this species (132), a high fraction of rearrangements are out-of-frame (134).

T CELL ANTIGEN RECEPTOR GENES IN AVIANS As indicated above, chicken TCRβ genes were the first nonmammalian TCR genes to be described (115). Two Vβ gene families, which have mammalian Vβ counterparts, were identified that appear to define the chicken TCR2 and TCR3 cell lineages (135, 136). A total of six Vβ1 and four Vβ2 genes encode the chicken Vβ repertoire. All Vβ genes rearrange to a single 14 nucleotide Dβ element. Four Jβ segments and a single Cβ element were identified. The extent of N-region addition increases during development and is consistent with ontogenetic up-regulation of terminal deoxynucleotide transferase (TdT) in the thymus (137). The chicken Dβ region encodes a glycine residue in all three reading frames that presumably participates in the formation of a CDR3encoded loop structure (137). The chicken TCRα genes were isolated by a coprecipitation-peptide sequencing strategy employing antibodies to chicken CD3 (138). Multiple Vα and Jα elements are linked to a single Cα region. Two Vα gene families, which exhibit ∼24% amino acid identity, account for all Vα expression by T cell lines

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

132

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

(136). Southern blotting is consistent with more than 25 Vα elements as well as multiple Jα elements. Typical RSS configurations are associated with chicken genomic Vα and Jα elements. In mammalian TCRα/δ rearrangements, Vα elements associate with Dδ-JδCδ at a low but significant frequency. cDNA library screening with Vα and Cα probes identified Vα +/Cα − clones. Sequencing revealed a putative Cδ gene exhibiting 33% amino acid identity with human and mouse Cδ. Northern blot analyses showed that this gene is transcribed in thymus and spleen but is absent in chicken αβ T cell lines. Two Vδ families have been identified, and Cδ associates with both chicken Vα families (136). TCRγ genes in chicken were isolated by a short primer degenerate PCR strategy using chicken genomic DNA as template, followed by screening of a chicken spleen cDNA library with a candidate TCRγ -like amplified product (122). Two different Vγ families and a putative Cγ region were identified. The extracellular Ig domain of putative chicken Cγ exhibits ∼31% identity with mouse Cγ . A third Vγ family was identified and each of the three families consists of eight to ten members (139). Three Jγ segments and a single Cγ region gene have been described. Transcripts that hybridize with a Cγ probe are prevalent in a γ δ cell line and are absent in αβ T cell and B cell lines. No evidence was found for the early waves of invariant rearrangement of TCRγ δ T cells seen for mouse and human (139). Notably, chicken—like sheep, cattle, and pigs—possess a high frequency of γ δ T cells and also express a more complex repertoire of γ and δ rearrangements. Thus, the TCRγ δ characteristic described for the mouse and human may not be typical of vertebrates as a whole. In the initial stages of the comparison of complex genetic systems, such as TCRs, similarities are likely to be recognized before differences are defined. With this bias in mind, the emerging consensus is that the broad-scale features of mammalian TCR genes can be found throughout the living representatives of the jawed vertebrates. This putative conservation of structure may result from an increased level of functional constraint that possibly represents an outcome of the more central role these receptors play in adaptive immunity. However, it is clear that this constraint does not extend to the TCR primary sequence, which exhibits extraordinary divergence relative to the Igs. It is likely that additional interpretations will emerge as TCR gene structures from an even wider variety of vertebrate taxa are elucidated and more extensive genomic studies are reported.

T CELL MEMORY IN LOWER VERTEBRATES The recent description of TCR α, β, γ , and δ genes in diverse vertebrate groups raises questions about conservation of T cell function throughout vertebrate phylogeny. T cell function typically has not been examined directly in

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

133

nonmammalian species, but accelerated alloimmune reactions to secondary skin grafts have been used as a measure of T cell memory in many lower vertebrate species. Although alloantigens are probably poor indicators of the T cell response (140), these studies generally showed mildly accelerated responses to second-set grafts (141); however, the specificity of these results is unclear because of a lack of a third-party, unrelated donor control group. Given these caveats, two major types of allograft rejection patterns are observed. Mammals, avians, anuran amphibians, and teleost fish exhibit acute primary allograft rejection followed by accelerated secondary rejection, whereas urodele amphibians, reptiles, cartilaginous fish, and lamprey respond more slowly to primary allograft and show a gradual stepwise increase in reactivity to repeated allografts (141–143). The basis for the differences in allograft rejection patterns in these animal groups remains unknown, but it is clear that there is no simple phylogenetic pattern that implies stepwise acquisition of this cell-mediated function. Strong evidence for cell-mediated immune memory in teleost fish is suggested by a study that used recombinant protein fragments from viral haemorrhagic septicemia rhabdovirus (VHSV) to stimulate trout leukocytes from survivors of VHSV infection or leukocytes from uninfected trout (144). Cellmediated immunity was detected by lymphocyte proliferation assays to detect increased sensitivity to activation. The persistence of sensitized, antigenspecific cells that proliferated in response to subsequent exposure to the primary stimulus is indicative of memory. Furthermore, the recombinant proteins were able to provide partial protection to trout after secondary challenge with the virus. The findings that chondrichthyan TCR and MHC genes are as diverse as their mammalian homologs essentially negates early speculation that restricted diversity in these molecules could relate to differences seen in allograft rejection in cartilaginous fish (4, 66, 121). The recent definition of molecular markers for different lymphocyte populations (22, 30, 42, 121, 145, 146), along with progress in elasmobranch culture systems (147) (C Luer, C Walsh, personal communication), and a better understanding of the molecular mechanisms of mammalian memory should facilitate efforts to measure similar parameters in the cartilaginous fish.

JAWLESS VERTEBRATE IMMUNITY The jawless vertebrates (Agnatha) are comprised of two extant groups (hagfishes and lampreys), which are the surviving remnants of a major evolutionary radiation and represent the most divergent extant vertebrate groups relative to the species that have been described in this review (Figure 1a). This divergence is reflected in significant anatomic, biochemical, and physiological differences relative to jawed vertebrates. The current consensus view places

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

134

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

each of these two agnathan groups in lineages derived from separate common ancestors (polyphyly) with the jawed vertebrates (148), although evidence also exists for a single common ancestor (monophyly) (149). Although the mode of divergence of the hagfish and lamprey from common ancestor(s) with the higher vertebrates has important implications as to how to interpret immunity in these species, there is no doubt that study of them will offer a unique phylogenetic view into the immune system of the jawed vertebrates. The agnathans lie across a phylogenetic boundary with regard to our current knowledge of vertebrate antigen binding receptors. Agnathans lack both a spleen and a thymus, which are present in all jawed vertebrate groups. Furthermore, the inducible immune response in lamprey is associated with a highly specific molecule(s) of a lectin-like character (24, 26). Although no fully systematic descriptions have been reported, a variety of approaches for identifying homologs of Ig, TCR, MHC I and II, and RAG genes have been unsuccessful with hagfish (Eptatretus stoutii) and lamprey (Petromyzon marinus), underscoring the major phylogenetic break in adaptive immune function between the jawed and jawless vertebrates (Figure 2). This failure could relate to the absence of such a receptor or to either qualitative or quantitative complications, e.g. low levels of expression, transient expression at different stages of development or expression in “atypical” tissues. Alternatively, antigen binding receptor gene homologs in these species may be present but have diverged appreciably. The successes in recovering Ig, MHC, and TCR in jawed vertebrates using PCRbased techniques has to some degree depended on the statistical advantages afforded by multiplicity of target motifs as well as inferred knowledge of their anatomic sites of expression. If the primary evolutionary event(s), presumably a transposition (5) (see below), which led to the rearrangement of segmental elements, was followed (under radically new selective pressure) by an expansion of gene number, then nonrearranging homologs in agnathans are possibly present in single- or low-copy number. Recent observations of extreme interspecies sequence divergence among orthologous novel immune-type receptor genes underscores how limited variation can impede the identification of homologous structures, even among different lineages of teleost fish (J Yoder, unpublished observation). A universal genetic strategy directed at the identification of the homologs of Ig and TCR in jawless vertebrates may not be feasible. Although it is plausible that homologs of relatively unconserved cell surface proteins may be difficult to identify in lamprey and hagfish, it should be possible to isolate highly conserved genes such as RAG (see below) from these species if homologs are present, i.e. failure to identify them is not likely to represent a technical artifact. It should be noted that the proportional difference in divergence time between the chondrichthyans and other jawed vertebrates (∼500 MYA) is not greatly different from that of the Agnathans and jawed vertebrates (<600 MYA),

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

135

which suggests that more than mere clock-like divergence is responsible for the failure to isolate homologs in these species. Although presently based on entirely negative data, the findings from agnathan representatives are entirely consistent with the emergence of the jawed vertebrate adaptive immune system by a transposition event that took place after their divergence. An important consideration of this real or perceived difference as regards antigen binding receptors in agnathans relates to the developmental biology of these species. Specifically, lampreys undergo metamorphosis from ammocoete larvae to the adult form. Characterization of gut-associated tissue (typhlosole) and the opisthonephros in the larval form of lampreys indicates the presence of lymphocytic/plasmocytic lineages. The typhlosole involutes prior to the first internal signs of metamorphosis and the opisthonephros involutes during metamorphosis. Metamorphosis is associated with a gain in lymphohemopoietic character of the protovertebral arch (supraneural body) (150; A Miracle, unpublished observation). It is possible that the adult (parasitic phase) of the lamprey life cycle is associated with a markedly different immune system than that seen in the larval form. Even if jawless vertebrates lack readily recognizable homologs of MHC I, MHC II, TCR, and Ig, viewing the emergence of these molecules in the jawed vertebrates as a sudden event relative to the origins of jawed vertebrates needs to be reconsidered, given the significant time period separating these vertebrate radiations, a period of time that clearly was of sufficient length to allow for drastic morphological change. Finally, it remains possible that species that diversified prior to the jawless vertebrates, e.g. protochordates, could retain homologs of antigen binding receptor genes that are more related to their higher vertebrate counterparts.

GENES AFFECTING DIFFERENTIATION OF LYMPHOCYTES AND THE GENERATION OF IMMUNOLOGICAL DIVERSITY The commitment of specific hematopoietic lineages to the expression of specific antigen binding receptor genes is significant both in terms of defining the anatomical sites of synthesis of immune receptors and in the potential for identifying related lineages in species that diverged prior to the jawed vertebrates. T, B, and NK cell lineage differentiation is dependent on the expression of the genes related to the Ikaros transcription factor. The Ikaros gene undergoes differential processing to produce a variety of different zinc-finger DNA binding transcription factors (151, 152). Ikaros is expressed in chickens from day 2 onward, prior to the transfer of the first precursor cells to the lymphoid primordia, thymus and bursa (153), as well as in Xenopus (154). In trout, Ikaros,

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

136

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

a single copy gene, is expressed in six different spliced isoforms also found in mammals and in two novel forms (154). The expression of Ikaros in trout and Xenopus is primarily lymphoid and occurs early in development (154). Unique differentially processed forms of Ikaros as well as related family members [Helios, Aiolos (152)] have been identified in skate (R Haire, A Miracle, unpublished observation). Regardless of the function of these genes, they are potentially valuable markers of immune cells in divergent species. Identification and characterization of various other transcription factors, which in general are more conserved than cell surface markers, in different vertebrate species could prove informative in understanding similarities and differences in B cell development and function. Specifically, studies in higher vertebrates have identified transcription factors that are critical to both the initial formation of germinal centers and the differentiation pathways toward either the plasma cell or the memory cell phenotype, e.g. Blimp-1 and BSAP (Pax-5) promote plasma cell or germinal center formation, respectively. These factors also influence germinal center B cells, promoting either plasma cell or memory cell formation following affinity maturation (155). The evolutionary stability of RSSs (5, 32, 156, 157) and the requisite involvement of RAG1 and 2 in Ig and TCR gene rearrangement represent other potential approaches for examining the immune systems of species found below the phylogenetic level of the jawed vertebrates. Specifically, RAG1 has been identified in mammals, avians, amphibians, ray-finned fish, and cartilaginous fish (5). Shared regions of RAG1 in sandbar shark and zebrafish (Danio rerio) as well as in other higher vertebrates reveal high degrees of amino acid identity (158, 159). RAG2 is somewhat less conserved phylogenetically than is RAG1 (160), but RAG1 and RAG2 are found in the same relative close proximity in lower vertebrates as in mammals (161). In mammals, avians, and amphibians, RAG1 and 2 are encoded by single exons, in opposite transcriptional orientation. However, in zebrafish the coding region of RAG1 is interrupted by two introns (162), and in trout, the coding region is interrupted by a single intron (163). These intronic relationships need to be considered in the design of strategies to identify RAG homologs in other species. The conserved close physical linkage of RAG1 and 2 in trout, zebrafish, and Xenopus (160–162) and higher vertebrates suggests that the identification of a homolog of one of these genes would lead to the isolation of the second. The observations of RAG2 gene expression in oocytes in Xenopus (160) and early in development in trout (161) are of strategic significance in terms of potential PCR template sources in other species. Although identification of RAG in a jawless vertebrate species would be of great significance, there is a distinct possibility that its presence in jawed vertebrates is the result of a horizontal gene transfer event and thus a homolog would not necessarily be present in jawless vertebrates or protochordates.

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

137

TdT functions in generating junctional diversity, which has been documented in species as phylogenetically distant from mammals as the cartilaginous fish (18, 121). TdT expression has been characterized in species as phylogenetically distant from man as amphibians (164) and ray-finned fish (165). Bruton’s tyrosine kinase (BTK), a nonreceptor protein tyrosine kinase, is essential for normal B lymphocyte development but is not expressed in T cells; a Btk homolog has been identified in skate (145). CD45, a TM tyrosine phosphatase that is required for antigen-induced signal transduction in mammalian lymphocytes, is expressed in horned shark (146). Investigations into the concordance of expression of antigen receptors, lymphocyte-specific enzymes such as RAG and TdT, transcription factors, and other genes implicated in lymphoid development have the potential for providing additional insight into the sites of hematopoiesis and the molecular events governing lymphocyte development and activation in diverse vertebrate groups. However, it also must be recognized that the transcription factors are often utilized in a variety of unrelated systems.

ORIGINS OF ANTIGEN BINDING RECEPTORS The phylogenetic approach that has been at the core of this review has provided a wealth of information regarding the evolution of rearranging gene families; however, it offers little in the way of defining the early evolutionary origins of antigen binding receptors, other than to constrain the time of origin and diversification of all six rearranging gene classes (TCRα, β, γ , δ, and Ig lightchain and Ig heavy-chain genes) to a period prior to the divergence of the cartilaginous and ray-finned fishes, ∼460–550 MYA. Inferring characteristics of primitive receptor forms through phylogenetic analysis of existing rearranging genes is complicated by the highly derived nature of the antigen binding receptors. Nevertheless, some relationships are evident, e.g. TCRβ is more related to TCRγ , and TCRα is more related to TCRδ at the sequence level than either is between these two groups. The D regions appear to have evolved or were lost more than once during the evolution of TCRs. Additional sources of information may prove useful in cases where the limited data from primary sequence is inadequate, e.g. the syntenic organization of genes encoded in the chromosomal vicinity of the rearranging receptors could be useful in relating them to one another. Also, as more information regarding the coreceptors and transcriptional regulation of antigen binding receptor genes becomes available, a more complete picture of the early evolutionary history of these proteins may emerge from affinities in regulation and function. Nonrearranging genes that descended from the prerearranging ancestral antigen binding receptor may still exist in invertebrates and protochordates as well

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

138

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

LITMAN, ANDERSON & RAST

as in jawed vertebrates. A number of potential nonrearranging ancestral-like candidate genes are known that contain TCR/Ig-like V regions, including CD8β and VpreB. Notably, both of these genes are encoded very near rearranging genes in mice and have functions that are consistent with an ancient association with these antigen recognition systems. A particularly interesting set of nonrearranging genes termed novel immune-type receptors (NITR) was isolated initially from the Southern pufferfish using a V region degenerate PCR strategy that was employed to isolate TCR genes (122). The prototypic NITR gene (Sn193) consists of a V domain and a C2 domain, as well as TM and cytoplasmic (Cyt) regions. The V region domain exhibits remarkable similarity to TCR and Ig V regions beyond that which is displayed by other nonrearranging V region–containing Ig domain proteins; a J-like region is located 30 of V. P1 artificial chromosomes were isolated using Sn193 as well as several related genes as probes, and a large number of related genes were identified in tandem linkage over a >100-kb region. The NITR genes exhibit extensive TCR-like V region diversification, and the majority characterized thus far contain ITIM motifs in their cytoplasmic regions (S Strong, G Mueller, R Litman, N Hawke, R Haire, A Miracle, J Rast, C Amemiya, G Litman, unpublished observation). NITR genes may represent a link between conventional V-type recognition, innate function, and signaling pathways known to function in recognition molecules such as certain NK receptors that are members of a newly defined superfamily of ITIM-containing genes, the inhibitory receptor superfamily (166). A number of additional V region domain–containing cell surface molecules, including CD8β, OX-2, CTX (167), and ChT1 (168, 169), have varying degrees of Ig/TCR-like character. The two Xenopus CTX genes contain a V and C2 Ig domain, both of which are split by introns. A J-like region is present, including a glycine bulge motif C terminal to the V region (167, 170). One of the Xenopus genes is encoded within the MHC region. The chicken ChT1 protein has a domain and intron/exon structure similar to that of CTX but is not linked to the chicken MHC (168; K Katevuo, personal communication). Both these genes are expressed in thymic T cells and possibly derive from an early TCR ancestral molecule. Recently, mammalian homologs of CTX and ChT1 were identified; however, these genes are not expressed in thymus (169). It is of great interest to researchers to determine if CTX/ChT1 genes can be detected in ray-finned and/or cartilaginous fish. All vertebrate rearranging receptors, along with class I and II MHC proteins, contain Ig C1-type domains. If the MHC genes are primitively nonrearranging, then the initial hypothesized transposition event that led to rearranging receptors is likely to have interrupted a V-C1-TM-Cyt-type molecule. The V-C2-TM-Cyt general structure, found in a variety of vertebrate genes along with the pufferfish NITR, may reflect the antigen receptor predecessor of IgC1-containing, nonrearranging receptor.

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

Annual Reviews

T1: ARS

AR078-05

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN BINDING RECEPTORS

139

Further clarification of the origins and interrelationships of these putative receptors will come through phylogenetic comparisons between rearranging and nonrearranging antigen binding receptors found in jawed vertebrates. Such analyses will also be significant in terms of designing strategies for identification of related systems in agnathans, lower chordates, hemichordates, and echinoderms. The recent discovery that RAG1 and RAG2 proteins together constitute a transposase, capable of excising a piece of DNA containing recombination signals from a donor site and inserting into a target DNA molecule, has enormous bearing on mechanisms whereby nonrearranging receptors diverged into the rearranging antigen binding receptor genes (171).

CONCLUSIONS The origins of Ig and TCR, as inferred from studies of contemporary species, can be traced to a distant period in the phylogenetic development of the jawed vertebrates. Available information suggests that the character of TCR genes has been conserved throughout vertebrates, whereas Ig genes have undergone a series of changes, of which some segregate to major phylogenetic radiations and others are specific to less inclusive taxonomic groups. A series of genetic divergences can be recognized throughout the jawed vertebrates that reflect a process of continued evolutionary variation of potential mechanisms of diversification. One such step separates species in which heavy- and light-chain genes are encoded at a single locus from those in which large numbers of individual loci are present. In the latter case, which is found in all cartilaginous fish (e.g. sharks), certain gene clusters are joined in the germline, whereas others hypermutate extensively. It is likely that the regulation of expression of single and multicluster Ig gene loci differs markedly. Other variations in Ig structure and regulation are seen throughout the radiations of ray- and fleshy-finned fish as well as the tetrapods, including the following: the use of alternative processing mechanisms, partial precommitment of rearranging segment elements, nonassociation of light chains with heavy chains, and the use of both gene conversion and antigen-independent somatic hyperconversion as a primary basis for varying receptor structure. Some of the exceptional phenomena typically are associated with specialized lymphocyte compartments and corresponding unique microenvironments. The basis for the remarkable plasticity in form and function of both Ig and TCR is not understood. Studies to date in jawless vertebrates have failed to provide evidence for antigen binding receptors resembling those found in jawed vertebrates, nor have homologous structures been identified in more phylogenetically distant forms. It remains open to question whether other heretofore unrecognized genetic systems related to Ig and TCR also function in antigen binding. Several nonrearranging genes that have been characterized

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

140

11:2

QC: ARS/Uks

T1: ARS

Annual Reviews

AR078-05

LITMAN, ANDERSON & RAST

in jawed vertebrates exhibit varying levels of sequence identity with the rearranging antigen binding receptors and could represent either modern forms of progenitor genes or derived products of Ig and TCR loci. Such gene products, which would function in an innate capacity, could offer new information as to how the highly complex diversified process of antigen recognition has evolved.

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ACKNOWLEDGMENTS We would like to thank Chris Amemiya, Louis Du Pasquier, Martin Flajnik, Robert Haire, Noel Hawke, Kasia Katevuo, Carl Luer, Ann Miracle, Tatsuya Ota, David Schatz, Scott Strong, Gregory Warr, and Jeffrey Yoder for sharing their unpublished findings and their valuable comments. We would also like to thank Michael Sexton for his assistance with the figures and Barbara Pryor for editorial assistance. This work was supported by Grant R37 AI23338 to GWL from the National Institutes of Health. MKA is supported by the Stowers Institute for Medical Research. JPR is supported by an NRSA Grant GM 18478. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Williams AF, Barclay AN. 1988. The immunoglobulin superfamily-domains for cell surface recognition. Annu. Rev. Immunol. 6:381–405 2. Medzhitov R, Janeway CA. 1997. Innate immunity: impact on the adaptive immune response. Curr. Opin. Immunol. 9:4–9 3. Medzhitov R, Preston-Hurlburt P, Janeway CA. 1997. A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 388:394–97 4. Bartl S, Baltimore D, Weissman IL. 1994. Molecular evolution of the vertebrate immune system. Proc. Natl. Acad. Sci. USA 91:10769–70 5. Thompson CB. 1995. New insights into V(D)J recombination and its role in the evolution of the immune system. Immunity 3:531–39 6. Litman GW, Rast JP. 1996. The organization and structure of immunoglobulin and T-cell receptor genes in the most phylogenetically distant jawed vertebrates: evolutionary implications. Res. Immunol. 147:226–33 7. Kokubu F, Litman R, Shamblott MJ, Hinds K, Litman GW. 1988. Diverse organization of immunoglobulin VH gene

8.

9.

10. 11.

12.

13.

14.

loci in a primitive vertebrate. EMBO J. 7:3413–22 Anderson MK, Shamblott MJ, Litman RT, Litman GW. 1995. The generation of immunoglobulin light chain gene diversity in Raja erinacea is not associated with somatic rearrangement, an exception to a central paradigm of B cell immunity. J. Exp. Med. 181:109–19 Zheng B, Xue W, Kelsoe G. 1994. Locusspecific somatic hypermutation in germinal centre T cells. Nature 372:556– 59 Wagner SD, Neuberger MS. 1996. Somatic hypermutation of immunoglobulin genes. Annu. Rev. Immunol. 14:441–57 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 Hinds KR, Litman GW. 1986. Major reorganization of immunoglobulin VH segmental elements during vertebrate evolution. Nature 320:546–49 Harding FA, Cohen N, Litman GW. 1990. Immunoglobulin heavy chain gene organization and complexity in the skate, Raja erinacea. Nucleic Acids Res. 18:1015–20 Rast JP, Amemiya CT, Litman RT, Strong

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

T1: ARS

Annual Reviews

AR078-05

ANTIGEN BINDING RECEPTORS

15. 16.

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

17.

18.

19.

20.

21.

22.

23.

24.

25.

SJ, Litman GW. 1998. Distinct patterns of IgH structure and organization in a divergent lineage of chondrichthyan fishes. Immunogenetics 47:234–45 Carroll RL. 1988. Vertebrate Paleontology and Evolution. New York: Freeman Kumar S, Hedges SB. 1998. A molecular timescale for vertebrate evolution. Nature 392:917–20 Litman GW, Berger L, Murphy K, Litman R, Hinds KR, Erickson BW. 1985. Immunoglobulin VH gene structure and diversity in Heterodontus, a phylogenetically primitive shark. Proc. Natl. Acad. Sci. USA 82:2082–86 Hinds-Frey KR, Nishikata H, Litman RT, Litman GW. 1993. Somatic variation precedes extensive diversification of germline sequences and combinatorial joining in the evolution of immunoglobulin heavy chain diversity. J. Exp. Med. 178:825–34 Kokubu F, Hinds K, Litman R, Shamblott MJ, Litman GW. 1988. Complete structure and organization of immunoglobulin heavy chain constant region genes in a phylogenetically primitive vertebrate. EMBO J. 7:1979–88 Kokubu F, Hinds K, Litman R, Shamblott MJ, Litman GW. 1987. Extensive families of constant region genes in a phylogenetically primitive vertebrate indicate an additional level of immunoglobulin complexity. Proc. Natl. Acad. Sci. USA 84:5868–72 Anderson M, Amemiya C, Luer C, Litman R, Rast J, Niimura Y, Litman G. 1994. Complete genomic sequence and patterns of transcription of a member of an unusual family of closely related, chromosomally dispersed immunoglobulin gene clusters in Raja. Int. Immunol. 6:1661–70 Greenberg AS, Avila D, Hughes M, Hughes A, McKinney EC, Flajnik MF. 1995. A new antigen receptor gene family that undergoes rearrangement and extensive somatic diversification in sharks. Nature 374:168–73 Anderson MK, Strong SJ, Litman RT, Luer CA, Amemiya CT, Rast JP, Litman GW. 1998. A long form of the IgX gene in the skate exhibits a striking resemblance to the novel IgW and IgNARC genes in the shark. Immunogenetics 47: In press Rast JP, Anderson MK, Litman GW. 1995. The structure and organization of immunoglobulin genes in lower vertebrates. See Ref. ?, pp. 315–41 Ota T, Nei M. 1994. Divergent evolution and evolution by the birth-and-death

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

141

process in the immunoglobulin VH gene family. Mol. Biol. Evol. 11:469–82 Litman GW, Rast JP, Shamblott MJ, Haire RN, Hulst M, Roess W, Litman RT, Hinds-Frey KR, Zilch A, Amemiya CT. 1993. Phylogenetic diversification of immunoglobulin genes and the antibody repertoire. Mol. Biol. Evol. 10:60–72 Roux KH, Greenberg AS, Greene L, Strelets L, Avila D, McKinney EC, Flajnik MF. 1998. Molecular convergence of the nurse shark (new) antigen receptor (NAR) and unusual mammalian immunoglobulins. PNAC 95:11,804–9 Hamers-Casterman C, Atarhouch T, Muyldermans S, Robinson G, Hamers C, Songa EB, Bendahman N, Hamers R. 1993. Naturally occurring antibodies devoid of light chains. Nature 363:446–48 Desmyter A, Transue TR, Ghahroudi MA, Thi M-HD, Poortmans F, Hamers R, Muyldermans S, Wyns L. 1996. Crystal structure of a camel single-domain VH antibody fragment in complex with lysozyme. Nat. Struct. Biol. 3:803–11 Harding FA, Amemiya CT, Litman RT, Cohen N, Litman GW. 1990. Two distinct immunoglobulin heavy chain isotypes in a primitive, cartilaginous fish, Raja erinacea. Nucleic Acids Res. 18:6369–76 Chess A, Simon I, Cedar H, Axel R. 1994. Allelic inactivation regulates olfactory receptor gene expression. Cell 78:823– 34 Litman GW, Anderson MK, Rast JP, Amemiya CT. 1996. Organization and mechanism of rearrangement of immunoglobulin genes in lower vertebrates. In Wier’s Handbook of Experimental Immunology, Immunochemistry and Molecular Biology, ed. LA Herzenberg, DM Weir, L Herzenberg, C Blackwell, 1:1– 14. Boston: Blackwell Sci. 5th ed. Berstein RM, Schluter SF, Shen S, Marchalonis JJ. 1996. A new high molecular weight immunoglobulin class from the carcharhine shark: implications for the properties of the primordial immunoglobulin. Proc. Natl. Acad. Sci. USA 93:3289– 93 Greenberg AS, Hughes AL, Guo J, Avila D, McKinney EC, Flajnik MF. 1996. A novel “chimeric” antibody class in cartilaginous fish: IgM may not be the primordial immunoglobulin. Eur. J. Immunol. 26:1123–29 Diaz M, Greenberg AS, Flajnik MF. 1998. Somatic hypermutation of the new antigen receptor gene (NAR) in the nurse shark does not generate the repertoire: possible role in antigen-driven reactions

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

142

36.

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

11:2

QC: ARS/Uks

T1: ARS

Annual Reviews

AR078-05

LITMAN, ANDERSON & RAST in the absence of germinal centers. PNAS. In press Shamblott MJ, Litman GW. 1989. Complete nucleotide sequence of primitive vertebrate immunoglobulin light chain genes. Proc. Natl. Acad. Sci. USA 86: 4684–88 Shamblott MJ, Litman GW. 1989. Genomic organization and sequences of immunoglobulin light chain genes in a primitive vertebrate suggest coevolution of immunoglobulin gene organization. EMBO J. 8:3733–39 Litman GW, Haire RN, Hinds KR, Amemiya CT, Rast JP, Hulst MA. 1992. Evolutionary development of the B-cell repertoire. Ann. NY Acad. Sci. 651:360– 68 Hohman VS, Schluter SF, Marchalonis JJ. 1992. Complete sequence of a cDNA clone specifying sandbar shark immunoglobulin light chain: gene organization and implications for the evolution of light chains. Proc. Natl. Acad. Sci. USA 89:276–80 Hohman VS, Schuchman DB, Schluter SF, Marchalonis JJ. 1993. Genomic clone for the sandbar shark lambda light chain: generation of diversity in the absence of gene rearrangement. Proc. Natl. Acad. Sci. USA 90:9882–86 Hohman VS, Schluter SF, Marchalonis JJ. 1995. Diversity of Ig light chain clusters in the sandbar shark (Carcharhinus plumbeus). J. Immunol. 155:3922–28 Rast JP, Anderson MK, Ota T, Litman RT, Margittai M, Shamblott MJ, Litman GW. 1994. Immunoglobulin light chain class multiplicity and alternative organizational forms in early vertebrate phylogeny. Immunogenetics 40:83–99 Greenberg AS, Steiner L, Kasahara M, Flajnik MF. 1993. Isolation of a shark immunoglobulin light chain cDNA clone encoding a protein resembling mammalian type kappa light chains: implications for the evolution of light chains. Proc. Natl. Acad. Sci. USA 90:10603–7 Sitnikova T, Nei M. 1998. Evolution of immunoglobulin kappa chain variable region genes in vertebrates. Mol. Biol. Evol. 15:50–60 Amemiya CT, Litman GW. 1990. The complete nucleotide sequence of an immunoglobulin heavy chain gene and analysis of immunoglobulin gene organization in a primitive teleost species. Proc. Natl. Acad. Sci. USA 87:811–15 Ventura-Holman T, Jones JC, Ghaffari SH, Lobb CJ. 1994. Structure and genomic organization of VH gene segments

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

in the channel catfish: members of different VH gene families are interspersed and closely linked. Mol. Immunol. 31:823– 32 Ghaffari SH, Lobb CJ. 1989. Cloning and sequence analysis of channel catfish heavy chain cDNA indicate phylogenetic diversity within the IgM immunoglobulin family. J. Immunol. 142:1356–65 Wilson MR, Marcuz A, van Ginkel F, Miller NW, Clem LW, Middleton D, Warr GW. 1990. The immunoglobulin M heavy chain constant region gene of the channel catfish, Ictalurus punctatus: an unusual mRNA splice pattern produces the membrane form of the molecule. Nucleic Acids Res. 18:5227–33 Lee MA, Bengten E, Daggfeldt A, Rytting A-S, Pilstr¨om L. 1993. Characterisation of rainbow trout cDNAs encoding a secreted and membrane-bound Ig heavy chain and the genomic intron upstream of the first constant region. Mol. Immunol. 30:641–48 Warr GW, Miller NW, Clem LW, Wilson MR. 1992. Alternative splicing pathways of the immunoglobulin heavy chain transcript of a teleost fish, Ictalurus punctatus. Immunogenetics 35:253–56 Wilson MR, Ross DA, Miller NW, Clem LW, Middleton DL, Warr GW. 1995. Alternative pre-mRNA processing pathways in the production of membrane IgM heavy chains in holostean fish. Dev. Comp. Immunol. 19:65–77 Rycyzyn MA, Wilson MR, Warr GW, Clem LW, Miller NW. 1996. Membrane immunoglobulin-associated molecules on channel catfish B lymphocytes. Dev. Comp. Immunol. 20:341–51 Magor BG, Wilson MR, Miller NW, Clem LW, Middleton DL, Warr GW. 1994. An Ig heavy chain enhancer of the channel catfish Ictalurus punctatus: evolutionary conservation of function but not structure. J. Immunol. 153:5556–63 Hayman JR, Ghaffari SH, Lobb CJ. 1993. Heavy chain joining region segments of the channel catfish. Genomic organization and phylogenetic implications. J. Immunol. 151:3587–96 Roman T, Charlemagne J. 1994. The immunoglobulin repertoire of the rainbow trout (Oncorhynchus mykiss): definition of nine Igh-V families. Immunogenetics 40:210–16 Roman T, Andersson E, Bengten E, Hansen J, Kaattari S, Pilstrom L, Charlemagne J, Matsunaga T. 1996. Unified nomenclature of Ig VH genes in rainbow trout (Oncorhynchus mykiss): definition

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

T1: ARS

Annual Reviews

AR078-05

ANTIGEN BINDING RECEPTORS

57.

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

of eleven VH families. Immunogenetics 43:325–26 Roman T, de Guerra A, Charlemagne J. 1995. Evolution of specific antigen recognition: size reduction and restricted length distribution of the CDRH3 regions in the rainbow trout. Eur. J. Immunol. 25:269–73 Wilson M, Bengt´en E, Miller NW, Clem LW, Du Pasquier L, Warr GW. 1997. A novel chimeric Ig heavy chain from a teleost fish shares similarities to IgD. Proc. Natl. Acad. Sci. USA 94:4593– 97 Lundqvist M, Bengt´en E, Str¨omberg S, Pilstr¨om L. 1996. Ig light chain gene in the Siberian sturgeon (Acipenser baeri): implications for the evolution of the immune system. J. Immunol. 157:2031– 38 Daggfeldt A, Bengten E, Pilstr¨om L. 1993. A cluster type organization of the loci of the immunoglobulin light chain in Atlantic cod (Gadus morhua L.) and rainbow trout (Oncorhynchus mykiss Walbaum) indicated by nucleotide sequences of cDNAs and hybridization analysis. Immunogenetics 38:199–209 Ghaffari SH, Lobb CJ. 1993. Structure and genomic organization of immunoglobulin light chain in the channel catfish. An unusual genomic organizational pattern of segmental genes. J. Immunol. 151:6900–12 Partula S, Schwager J, Timmusk S, Pilstrom L, Charlemagne J. 1996. A second immunoglobulin light chain isotype in the rainbow trout. Immunogenetics 45:44–51 Ghaffari SH, Lobb CJ. 1997. Structure and genomic organization of a second class of immunoglobulin light chain genes in the channel catfish. J. Immunol. 159:250–58 Amemiya CT, Ohta Y, Litman RT, Rast JP, Haire RN, Litman GW. 1993. VH gene organization in a relict species, the coelacanth Latimeria chalumnae: evolutionary implications. Proc. Natl. Acad. Sci. USA 90:6661–65 Ota T, Rast JP, Litman GW, Amemiya CT. 1997. Lungfish immunoglobulins and their evolutionary implications. Dev. Comp. Immunol. 21:159 (Abstr.) Du Pasquier L, Flajnik M. 1998. Origin and evolution of the vertebrate immune system. In Fundamental Immunology, ed. WE Paul. New York: Raven. 4th ed. In press Mussmann R, Wilson M, Marcuz A, Courtet M, Du Pasquier L. 1996. Membrane exon sequences of the three Xeno-

68. 69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

143

pus Ig classes explain the evolutionary origin of mammalian isotypes. Eur. J. Immunol. 26:409–14 Mussmann R, Du Pasquier L, Hsu E. 1996. Is Xenopus IgX an analog of IgA? Eur. J. Immunol. 26:2823–30 Mussmann R, Courtet M, Schwager J, Du Pasquier L. 1997. Microsites for immunoglobulin switch recombination breakpoints from Xenopus to mammals. Eur. J. Immunol. 27:2610–19 Schwager J, Burckert N, Courtet M, Du Pasquier L. 1989. Genetic basis of the antibody repertoire in Xenopus: analysis of the VH diversity. EMBO J. 8:2989–3001 Hsu E, Schwager J, Alt FW. 1989. Evolution of immunoglobulin genes: VH families in the amphibian Xenopus. Proc. Natl. Acad. Sci. USA 86:8010–14 Haire RN, Amemiya CT, Suzuki D, Litman GW. 1990. Eleven distinct VH gene families and additional patterns of sequence variation suggest a high degree of immunoglobulin gene complexity in a lower vertebrate, Xenopus laevis. J. Exp. Med. 171:1721–37 Haire RN, Ohta Y, Litman RT, Amemiya CT, Litman GW. 1991. The genomic organization of immunoglobulin VH genes in Xenopus laevis shows evidence for interspersion of families. Nucleic Acids Res. 19:3061–66 Schwager J, Burckert N, Courtet M, Du Pasquier L. 1991. The ontogeny of diversification at the immunoglobulin heavy chain locus in Xenopus. EMBO J. 10:2461–70 Fellah JS, Kerfourn F, Wiles MV, Schwager J, Charlemagne J. 1993. Phylogeny of immunoglobulin heavy chain isotypes: structure of the constant region of Ambystoma mexicanum upsilon chain deduced from cDNA sequence. Immunogenetics 38:311–17 Fellah JS, Jacques C, Charlemagne J. 1994. Characterization of immunoglobulin heavy chain variable regions in the Mexican axolotl. Immunogenetics 39:201–6 Golub R, Fellah JS, Charlemagne J. 1997. Structure and diversity of the heavy chain VDJ junctions in the developing Mexican axolotl. Immunogenetics 46:402–9 Patel HM, Hsu E. 1997. Abbreviated junctional sequences impoverish antibody diversity in urodele amphibians. J. Immunol. 159:3391–99 Schwager J, Burckert N, Schwager M, Wilson M. 1991. Evolution of immunoglobulin light chain genes: analysis of Xenopus IgL isotypes and their con-

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

144

80.

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

81.

82.

83. 84.

85. 86.

87.

88.

89.

90. 91.

11:2

QC: ARS/Uks

T1: ARS

Annual Reviews

AR078-05

LITMAN, ANDERSON & RAST tribution to antibody diversity. EMBO J. 10:505–11 Zezza DJ, Stewart SE, Steiner LA. 1992. Genes encoding Xenopus laevis Ig L chains: implications for the evolution of kappa and lambda chains. J. Immunol. 149:3968–77 Haire RN, Ota T, Rast JP, Litman RT, Chan FY, Zon LI, Litman GW. 1996. A third Ig light chain gene isotype in Xenopus laevis consists of six distinct VL families and is related to mammalian lambda genes. J. Immunol. 157:1544–50 Litman GW, Murphy K, Berger L, Litman RT, Hinds KR, Erickson BW. 1985. Complete nucleotide sequences of three VH genes in Caiman, a phylogenetically ancient reptile: evolutionary diversification in coding segments and variation in the structure and organization of recombination elements. Proc. Natl. Acad. Sci. USA 82:844–48 Turchin A, Hsu E. 1996. The generation of antibody diversity in the turtle. J. Immunol. 156:3797–805 Reynaud C-A, Bertocci B, Dahan A, Weill J-C. 1994. Formation of the chicken Bcell repertoire: ontogenesis, regulation of Ig gene rearrangement, and diversification by gene conversion. Adv. Immunol. 57:353–78 Vainio O, Imhof BA. 1995. The immunology and developmental biology of the chicken. Immunol. Today 16:365–70 Weill J-C, Reynaud C-A. 1995. Generation of diversity by post-rearrangement diversification mechanisms: the chicken and the sheep antibody repertoires. See Ref. ?, pp. 267–88 Magor KE, Warr GW, Bando Y, Middleton DL, Higgins DA. 1998. Secretory immune system of the duck (Anas platyrhynchos). Identification and expression of the genes encoding IgA and IgM heavy chains. Eur. J. Immunol. 28:1063–68 Magor KE, Warr GW, Middleton D, Wilson MR, Higgins DA. 1992. Structural relationship between the two IgY of the duck, Anas platyrhynchos: molecular genetic evidence. J. Immunol. 149:2627– 33 Magor KE, Higgins DA, Middleton DL, Warr GW. 1994. One gene encodes the heavy chains for three different forms of IgY in the duck. J. Immunol. 153:5549–55 Warr GW, Magor KE, Higgins DA. 1995. IgY: clues to the origins of modern antibodies. Immunol. Today 16:392–98 Reynaud C-A, Dahan A, Anquez V, Weill J-C. 1989. Somatic hyperconversion diversifies the single VH gene of the chicken

92.

93.

94.

95.

96. 97.

98.

99.

100.

101.

102.

103.

104.

with a high incidence in the D region. Cell 59:171–83 Reynaud C-A, Anquez V, Weill J-C. 1991. The chicken D locus and its contribution to the immunoglobulin heavy chain repertoire. Eur. J. Immunol. 21:2661–70 Knight KL. 1992. Restricted VH gene usage and generation of antibody diversity in rabbit. Annu. Rev. Immunol. 10:593– 616 Parng CL, Hansal S, Goldsby RA, Osborne BA. 1996. Gene conversion contributes to Ig light chain diversity in cattle. J. Immunol. 157:5478–86 Meyer A, Parng CL, Hansal SA, Osborne BA, Goldsby RA. 1997. Immunoglobulin gene diversification in cattle. Int. Rev. Immunol. 15:165–83 Sun J, Butler JE. 1996. Molecular characterization of VDJ transcripts from a newborn piglet. Immunology 88:331–39 Thompson CB, Nieman PE. 1987. Somatic diversification of the chicken immunoglobulin light chain gene is limited to the rearranged variable gene segment. Cell 48:369–78 Reynaud C-A, Anquez V, Grimal H, Weill J-C. 1987. A hyperconversion mechanism generates the chicken light chain preimmune repertoire. Cell 48:379–88 Kondo T, Arakawa H, Kitao H, Hirota Y, Yamagishi H. 1993. Signal joint of immunoglobulin Vλ1-Jλ and novel joints of chimeric V pseudogenes on extrachromosomal circular DNA from chicken bursa. Eur. J. Immunol. 23:245–49 McCormack WT, Carlson LM, Tjoelker LW, Thompson CB. 1989. Evolutionary comparison of the avian IgL locus: combinatorial diversity plays a role in the generation of the antibody repertoire in some avian species. Int. Immunol. 1:332–41 McCormack WT, Thompson CB. 1990. Chicken IgL variable region gene conversions display pseudogene donor preference and 50 to 30 polarity. Genes Dev. 4:548–58 McCormack WT, Tjoelker LW, Carlson LM, Petryniak B, Barth CF, Humphries EH, Thompson CB. 1989. Chicken IgL gene rearrangement involves deletion of a circular episome and addition of single nonrandom nucleotides to both coding segments. Cell 56:785–91 McCormack WT, Hurley EA, Thompson CB. 1993. Germ line maintenance of the pseudogene donor pool for somatic immunoglobulin gene conversion in chickens. Mol. Cell. Biol. 13:821–30 Nahm MH, Kroese FG, Hoffmann JW. 1992. The evolution of immune mem-

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

T1: ARS

Annual Reviews

AR078-05

ANTIGEN BINDING RECEPTORS

105.

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

106.

107.

108.

109.

110.

111. 112.

113. 114.

115.

116. 117.

118.

ory and germinal centers. Immunol. Today 13:438–41 Wilson M, Hsu E, Marcuz A, Courtet M, Du Pasquier L, Steinberg C. 1992. What limits affinity maturation of antibodies in Xenopus—the rate of somatic mutation or the ability to select mutants? EMBO J. 11:4337–47 Reynaud C-A, Garcia C, Hein WR, Weill J-C. 1995. Hypermutation generating the sheep immunoglobulin repertoire is an antigen-independent process. Cell 80:115–25 Jacob J, Kelsoe G, Rajewsky K, Weiss U. 1991. Intraclonal generation of antibody mutants in germinal centres. Nature 354:389–92 M¨akel¨a O, Litman GW. 1980. Lack of heterogeneity in anti-hapten antibodies of a phylogenetically primitive shark. Nature 287:639–40 Litman GW, Erickson BW, Lederman L, M¨akel¨a O. 1982. Antibody response in Heterodontus. Mol. Cell. Biochem. 45: 49–57 van Muiswinkel WB, Wiegertjes GF. 1997. Immune responses after injection vaccination of fish. Dev. Biol. Stand. 90: 55–57 Du Pasquier L. 1993. Phylogeny of B cell development. Curr. Opin. Immunol. 5:185–93 Ruben LN. 1983. IgM memory: long lived hapten-specific memory in the newt, Notophthalmus viridescens. Immunology 48:385–92 Farber DL. 1998. Differential TCR signaling and the generation of memory T cells. J. Immunol. 160:535–39 Decker DJ, Linton P-J, Zaharevitz S, Biery M, Gingeras TR, Klinman NR. 1995. Defining subsets of naive and memory B cells based on the ability of their progeny to somatically mutate in vitro. Immunity 2:195–203 Tjoelker LW, Carlson LM, Lee K, Lahti J, McCormack WT, Leiden JM, Chen CLH, Cooper MD, Thompson CB. 1990. Evolutionary conservation of antigen recognition: the chicken T-cell receptor β chain. Proc. Natl. Acad. Sci. USA 87: 7856–60 Murphy PM. 1993. Molecular mimicry and the generation of host defense protein diversity. Cell 72:823–26 Hughes AL. 1997. Rapid evolution of immunoglobulin superfamily C2 domains expressed in immune system cells. Mol. Biol. Evol. 14:1–5 Smith LC, Davidson EH. 1992. The echinoid immune system and the phylogenetic

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

145

occurrence of immune mechanisms in deuterostomes. Immunol. Today 13:356– 62 Rast JP, Litman GW. 1994. T cell receptor gene homologs are present in the most primitive jawed vertebrates. Proc. Natl. Acad. Sci. USA 91:9248–52 Hawke NA, Rast JP, Litman GW. 1996. Extensive diversity of transcribed TCR-β in a phylogenetically primitive vertebrate. J. Immunol. 156:2458–64 Rast JP, Anderson MK, Strong SJ, Luer C, Litman RT, Litman GW. 1997. α, β, γ , and δ T cell antigen receptor genes arose early in vertebrate phylogeny. Immunity 6:1–11 Rast JP, Haire RN, Litman RT, Pross S, Litman GW. 1995. Identification and characterization of T-cell antigen receptor related genes in phylogenetically diverse vertebrate species. Immunogenetics 42:204–12 Luer CA. 1989. Elasmobranchs (sharks, skates, and rays) as animal models for biomedical research. In Nonmammalian Animal Models for Biomedical Research, ed. AD Woodhead, pp. 121–47. Boca Raton, FL: CRC Campbell KS, Backstrom BT, Tiefenthaler G, Palmer E. 1994. CART: a conserved antigen receptor transmembrane motif. Semin. Immunol. 6:393–410 Partula S, Fellah JS, de Guerra A, Charlemagne J. 1994. Caract´erisation d’ADNc de la chaine β de r´ecepteur des lymphocytes T chez la truite arc-en-ciel. C. R. Acad. Sci. Ser. III 317:765–70 Partula S, de Guerra A, Fellah JS, Charlemagne J. 1995. Structure and diversity of the T cell antigen receptor β-chain in a teleost fish. J. Immunol. 155:699–706 de Guerra A, Charlemagne J. 1997. Genomic organization of the TCR β-chain diversity (Dβ) and joining (Jβ) segments in the rainbow trout: presence of many repeated sequences. Mol. Immunol. 34:653–62 Hordvik I, Jacob ALJ, Charlemagne J, Endresen C. 1996. Cloning of T-cell antigen receptor β chain cDNAs from Atlantic salmon (Salmo salar). Immunogenetics 45:9–14 Partula S, de Guerra A, Fellah JS, Charlemagne J. 1996. Structure and diversity of the TCR α-chain in a teleost fish. J. Immunol. 157:207–12 Wilson MR, Zhou H, Bengten E, Clem LW, Stuge TB, Warr GW, Miller NW. 1998. T-cell receptors in channel catfish: structure and expression of TCR α and β genes. Mol. Immunol. In press

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

146

11:2

QC: ARS/Uks

T1: ARS

Annual Reviews

AR078-05

LITMAN, ANDERSON & RAST

131. Chr´etien I, Marcuz A, Fellah J, Charlemagne J, Du Pasquier L. 1997. The T cell receptor β genes of Xenopus. Eur. J. Immunol. 27:763–71 132. Fellah JS, Kerfourn F, Guillet F, Charlemagne J. 1993. Conserved structure of amphibian T-cell antigen receptor β chain. Proc. Natl. Acad. Sci. USA 90: 6811–14 133. Fellah JS, Kerfourn F, Charlemagne J. 1994. Evolution of T cell receptor genes: extensive diversity of Vβ families in the Mexican axolotl. J. Immunol. 153:4539– 45 134. Fellah JS, Kerfourn F, Dumay AM, Aubet G, Charlemagne J. 1997. Structure and diversity of the T-cell receptor α chain in the Mexican axolotl. Immunogenetics 45:235–41 135. Lahti JM, Chen CL, Tjoelker LW, Pickel JM, Schat KA, Calnek BW, Thompson CB, Cooper MD. 1991. Two distinct αβ T-cell lineages can be distinguished by the differential usage of T-cell receptor Vβ gene segments. Proc. Natl. Acad. Sci. USA 88:10956–60 136. Chen CH, Six A, Kubota T, Tsuji S, Kong F-K, G¨obel TWF, Cooper MD. 1996. T cell receptors and T cell development. Curr. Top. Microbiol. Immunol. 212:36– 53 137. McCormack WT, Tjoelker LW, Stella G, Postema CE, Thompson CG. 1991. Chicken T-cell receptor β-chain diversity: an evolutionarily conserved Dβ encoded glycine turn within the hypervariable CDR3 domain. Proc. Natl. Acad. Sci. USA 88:7699–703 138. Gobel TWF, Chen C-LH, Lahti J, Kubota T, Kuo C-L, Aebersold R, Hood L, Cooper MD. 1994. Identification of T cell receptor α-chain genes in the chicken. Proc. Natl. Acad. Sci. USA 91:1094–98 139. Six A, Rast JP, McCormack WT, Dunon D, Courtois D, Li Y, Chen CH, Cooper MD. 1996. Characterization of avian Tcell receptor γ genes. Proc. Natl. Acad. Sci. USA 93:15329–34 140. Hall BM, Roser BJ, Dorsch SE. 1977. Magnitude of memory to the major histocompatibility complex. Nature 268:532– 34 141. Perey DYE, Finstad J, Pollara B, Good RA. 1968. Evolution of the immune response. VI. First and second set skin homograft rejections in primitive fishes. Lab. Invest. 19:91–97 142. Borysenko M, Hildemann WH. 1970. Reactions to skin allografts in the horn shark, Heterodontus francisci. Transplantation 10:545–51

143. Hildemann WH. 1970. Transplantation immunity in fishes: Agnatha, Chondrichthyes and Osteichthyes. Transpl. Proc. 2:253 144. Estepa A, Thiry M, Coll JM. 1994. Recombinant protein fragments from haemorrhagic septicaemia rhabdovirus stimulate trout leukocyte anamnestic responses in vitro. J. Gen. Virol. 75:1329–38 145. Haire RN, Strong SJ, Litman GW. 1997. Identification and characterization of a homologue of Bruton’s tyrosine kinase, a Tec kinase involved in B cell development, in a modern representative of a phylogenetically ancient vertebrate species. Immunogenetics 46:349–51 146. Okamura M, Matthews RJ, Robb B, Litman GW, Bork P, Thomas ML. 1996. Comparison of CD45 extracellular domain sequences from divergent vertebrate species suggests the conservation of three fibronectin type III domains. J. Immunol. 157:1569–75 147. Luer CA, Walsh CJ, Bodine AB, Wyffels JT, Scott TR. 1995. The elasmobranch thymus: anatomical, histological, and preliminary functional characterization. J. Exp. Zool. 273:342–54 148. Forey P, Janvier P. 1993. Agnathans and the origin of jawed vertebrates. Nature 361:129–34 149. Stock DW, Whitt GS. 1992. Evidence from 18S ribosomal RNA sequences that lampreys and hagfishes form a natural group. Science 257:787–89 150. Zapata AG, Cooper EL, eds. 1990. The Immune System: Comparative Histophysiology. Chichester, UK: Wiley. 335 pp. 151. Georgopoulos K, Bigby M, Wang JH, Molnar A, Wu P, Winandy S, Sharpe A. 1994. The Ikaros gene is required for the development of all lymphoid lineages. Cell 79:143–56 152. Georgopoulos K, Winandy S, Avitahl N. 1997. The role of the Ikaros gene in lymphocyte development and homeostasis. Annu. Rev. Immunol. 15:155–76 153. Liippo J, Lassila O. 1997. Avian Ikaros gene is expressed early in embryogenesis. Eur. J. Immunol. 27:1853–57 154. Hansen JD, Strassburger P, Du Pasquier L. 1997. Conservation of a master hematopoietic switch gene during vertebrate evolution: isolation and characterization of Ikaros from teleost and amphibian species. Eur. J. Immunol. 27:3049–58 155. Liu Y-J, Banchereau J. 1997. Regulation of B cell commitment to plasma cells or to memory B cells. Semin. Immunol. 9:235– 40 156. Ramsden DA, Baetz K, Wu GE. 1994.

P1: APR/spd/mbg

P2: SAT/ARS/ary

January 18, 1999

11:2

QC: ARS/Uks

T1: ARS

Annual Reviews

AR078-05

ANTIGEN BINDING RECEPTORS

157. 158.

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

159.

160.

161.

162.

163.

164.

Conservation of sequence in recombination signal sequence spacers. Nucleic Acids Res. 22:1785–96 Lewis SM, Wu GE. 1997. The origins of V(D)J recombination. Cell 88:159–62 Greehalgh P, Steiner LA. 1995. Recombination activating gene 1 (Rag1) in zebrafish and shark. Immunogenetics 41:54–55 Bernstein RM, Schluter SF, Bernstein H, Marchalonis JJ. 1996. Primordial emergence of the recombination activating gene 1 (RAG1): sequence of the complete shark gene indicates homology to microbial integrases. Proc. Natl. Acad. Sci. USA 93:9454–59 Greenhalgh P, Olesen CEM, Steiner LA. 1993. Characterization and expression of recombination activating genes (RAG-1 and RAG-2) in Xenopus laevis. J. Immunol. 151:3100–10 Hansen JD, Kaattari SL. 1996. The recombination activating gene 2 (RAG2) of the rainbow trout Oncorhynchus mykiss. Immunogenetics 44:203–11 Willett CE, Cherry JJ, Steiner LA. 1997. Characterization and expression of the recombination activating genes (rag1 and rag2) of zebrafish. Immunogenetics 45:394–404 Hansen JD, Kaattari SL. 1995. The recombination activating gene 1 (RAG1) of rainbow trout (Oncorhynchus mykiss): cloning, expression and phylogenetic analysis. Immunogenetics 42:188–95 Lee A, Hsu E. 1994. Isolation and characterization of the Xenopus terminal de-

165.

166. 167.

168.

169.

170.

171.

147

oxynucleotidyl transferase. J. Immunol. 152:4500–7 Hansen JD. 1997. Characterization of rainbow trout terminal deoxynucleotidyl transferase structure and expression. TdT and RAG1 co-expression define the trout primary lymphoid tissues. Immunogenetics 46:367–75 Lanier LL. 1998. NK cell receptors. Annu. Rev. Immunol. 16:359–93 Chr´etien I, Robert J, Marcuz A, GarciaSanz JA, Courtet M, Du Pasquier L. 1996. CTX, a novel molecule specifically expressed on the surface of cortical thymocytes in Xenopus. Eur. J. Immunol. 26:780–91 Katevuo K, Imhof BA, Boyd R, Chidgey A, Bean A, Dunon D, G¨obel TWF, Vainio O. 1998. ChT1, an IgSF molecule required for T cell differentiation. Submitted for publication Chr´etien I, Marcuz A, Courtet M, Katevuo K, Vainio O, Heath JK, White SJ, Du Pasquier L. 1998. CTX, a Xenopus thymocyte receptor, defines a molecular family conserved throughout vertebrates. Eur. J. Immunol. 28:1–11 Du Pasquier L, Chr´etien I. 1996. CTX, a new lymphocyte receptor in Xenopus, and the early evolution of Ig domains. Res. Immunol. 147:218–26 Agrawal A, Eastman QM, Schatz DG. 1998. Transposition mediated by the V(D)J recombination proteins RAG1 and RAG2: implications for the origins of the antigen-specific immune system. Nature 394:744–51

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:109-147. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:149–87 c 1999 by Annual Reviews. All rights reserved Copyright °

TRANSCRIPTIONAL REGULATION OF T LYMPHOCYTE DEVELOPMENT AND FUNCTION Chay T. Kuo and Jeffrey M. Leiden Departments of Medicine and Pathology, University of Chicago, Chicago, Illinois 60637; e-mail: [email protected] KEY WORDS:

transcription factors, lineage determination, thymocyte development, T cell activation, Th1/Th2 differentiation

ABSTRACT The development and function of T lymphocytes are regulated tightly by signal transduction pathways that include specific cell-surface receptors, intracellular signaling molecules, and nuclear transcription factors. Since 1988, several families of functionally important T cell transcription factors have been identified. These include the Ikaros, LKLF, and GATA3 zinc-finger proteins; the Ets, CREB/ATF, and NF-κB/Rel/NFAT transcription factors; the Stat proteins; and HMG box transcription factors such as LEF1, TCF1, and Sox4. In this review, we summarize our current understanding of the transcriptional regulation of T cell development and function with particular emphasis on the results of recent gene targeting and transgenic experiments. In addition to increasing our understanding of the molecular pathways that regulate T cell development and function, these results have suggested novel targets for genetic and pharmacological manipulation of T cell immunity.

INTRODUCTION T lymphocytes are critical regulators of mammalian immune responses to pathogens and tumor cells. They are also important effectors of allergies, transplant rejection, and autoimmunity. The development and function of the T lymphocyte lineage are regulated tightly by signaling pathways that involve lineage-restricted cell-surface receptors, intracellular signaling molecules, and nuclear transcription factors (reviewed in 1–5). Since 1988, the analysis of 149 0732-0582/99/0410-0149$08.00

P1: SAT/MBG/SPD

January 21, 1999

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

150

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

T cell–specific transcriptional regulatory elements has resulted in the identification of several families of transcription factors that appear to be important regulators of T cell development, T cell quiescence and activation, and T cell survival and death. These factors include the Ikaros, LKLF, and GATA zinc-finger proteins; the Ets, HMG box, and CREB/ATF transcription factors; members of the NF-κB/Rel/NFAT family; and the Stat proteins (reviewed in 4, 6). As is true of most biological systems, the more we learn about T cell transcription, the more complex the system appears. It is now clear, for example, that the expression of most T cell genes is controlled by the assembly on promoter and enhancer elements of large complexes of diverse transcription factors (reviewed in 7, 8). The assembly of such transcription factor complexes is regulated both by the geometry of their binding sites on the DNA and by direct protein-protein interactions among the transcription factors themselves and between transcription factors and coactivator proteins which lack DNA-binding activity. A further level of complexity derives from the fact that most T cell transcription factors belong to multigene families and that multiple members of a single family are often expressed in thymocytes and T cells (4). Despite this apparent degree of complexity, recent gene targeting studies in mice have identified unique roles for a number of transcription factors in T cell development and function. In addition to increasing our basic understanding of T cell biology, these studies have suggested novel targets for genetic and pharmacological manipulation of T cell immunity. In this review, we summarize our current understanding of the transcriptional regulation of T cell development and function, with particular emphasis on insights derived from recent gene targeting and transgenic experiments. The reader is referred to several recent reviews for additional information in this area (4–6, 9).

TRANSCRIPTIONAL REGULATION OF T AND NK CELL LINEAGE COMMITMENT The hematopoietic cell lineages, including B and T lymphocytes and natural killer (NK) cells, are derived from a mesodermally derived progenitor, the pluripotent hematopoieitic stem cell (PHSC), which resides predominantly in the fetal liver and the adult bone marrow (reviewed in 10, 11). Recent studies have identified and characterized a common (Lin−, IL-7R+, Thy1−, Sca-1low, c-kitlow) lymphoid progenitor that has the capacity to generate T, B, and NK cells but lacks progenitor activity for the other hematopoietic lineages (12). Additional studies of the mouse thymus have identified a common T/NK progenitor (13, 14). These results have established a lineage of early lymphoid development (see Figure 1). Genetic studies in mice have identified several transcription factors required for the development and survival of the PHSC. These factors include GATA2,

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

151

Figure 1 A lineage of early lymphoid development with known progenitor cells. The roles of the Ikaros, Ets1, and GATA3 transcription factors in the development of the NK and T cell lineages are depicted schematically.

c-myb, Tal-1/Scl, AML1, and PU.1. The reader is referred to several recent reviews of early hematopoietic development for additional information about these proteins (4, 15). Two transcription factors, Ikaros and GATA3, have been implicated in the earliest stages of T cell lineage commitment. A third transcription factor, Ets1, is not required for the development of mature T and B lymphocytes but is necessary for the development of the NK cell lineage. We discuss the roles of these transcription factors in lymphocyte development in more detail in the following sections.

Ikaros The Ikaros gene encodes a family of lymphoid-restricted zinc-finger transcription factors related to the Drosophila protein Hunchback (16, 17). The different isoforms of Ikaros, produced by alternative splicing, contain an identical C-terminal activation domain and two zinc-finger dimerization domains (18). However, each of the six known isoforms contains different combinations of the four N-terminal zinc-finger DNA-binding domains (18, 19). Because at least three of the four N-terminal zinc fingers are required for DNA binding, only Ikaros isoforms 1, 2, and 3 can bind to the consensus DNA core motif GGGA (18). Ikaros expression is detected first in the embryonic yolk sac (the site of early hematopoiesis) at embryonic day 8 (E8) in mice (17). The gene is also expressed in fetal and adult thymocytes, in mature T and B cells, and in NK cells (16, 17). This pattern of expression suggested that Ikaros might be an important regulator of lymphoid development. Ikaros is a member of a

P1: SAT/MBG/SPD

January 21, 1999

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

152

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

multigene family of related zinc-finger proteins. Other members of the family include Aiolos and LyF-1 (19, 20). Early genetic studies in mice to assess the role of Ikaros in lymphoid development involved the production of a targeted deletion of DNA-binding zinc fingers 1 through 3 (21). These Ikaros-mutant mice lacked all T, B, and NK cells and their earliest defined progenitors (21). In contrast, the development of the other hematopoietic lineages appeared unaffected. These results were consistent with a model in which Ikaros is required for the differentiation and/or survival of a common lymphoid progenitor and suggested that Ikaros defined the earliest transcriptional checkpoint in lymphoid lineage commitment. Subsequent studies have suggested a more complicated role for Ikaros in lymphocyte development. Further analysis of the original Ikaros-mutant mice demonstrated that the gene targeting event had resulted in the expression of a stable dominant negative form of Ikaros that can dimerize with Ikaros-related proteins (e.g. Aiolos) but cannot bind DNA (22, 23). Thus, the phenotype of these mice likely reflected the loss of function of multiple Ikaros-related proteins. To more accurately assess the necessary role(s) of Ikaros in lymphoid development, Wang et al (24) produced a null allele of the Ikaros gene by deletion of the last coding exon, which encodes the transactivation and dimerization domains of the protein. In the absence of Ikaros, B cell development was arrested before the immature pro-B/pre-B cell precursor stage in both embryonic liver and adult bone marrow. Consistent with the early block in fetal B cell development, the fetal thymi of the Ikaros-null mice were devoid of recognizable lymphoid cells, suggesting an early block in fetal thymocyte ontogeny. However, Ikaros-null thymi began to be repopulated with developing T cells within 3–5 days after birth. Within several weeks of birth, these mice developed nearly wild-type numbers of total thymocytes. Interestingly, however, the Ikaros-deficient T cells displayed a number of functional defects including (a) preferential differentiation and clonal expansion of the CD4+ single-positive (SP) thymocytes, (b) hyperproliferation of both thymocytes and splenic T cells in response to T cell receptor (TCR) cross-linking, (c) significant reductions in thymic dendritic APC and NK cell numbers, and (d ) selective defects in TCRγ /δ T cell populations. These results suggested that one or more Ikaros-related proteins (but not Ikaros itself) are required for the development or survival of the common lymphoid progenitor and that Ikaros itself is necessary for both early B cell development and for a variety of mature T cell functions. The identification and mutation of Ikaros-related proteins and dimerization partners, and the identification of Ikaros target genes will be helpful in further elucidating the function of this important family of lymphoid-restricted zinc-finger transcription factors.

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

153

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

GATA3 The GATA family of vertebrate transcription factors contains six members, GATA1–GATA6, each of which contains a highly related DNA-binding domain composed of two evolutionarily conserved zinc fingers (25). GATA proteins bind as monomers to both a consensus GATA sequence (A/T)GATA(A/G) and related nonconsensus GATA motifs such as CGATGG and AGATTA (26, 27). Mammalian GATA proteins can be divided into two subfamilies based on their structures and their patterns of expression. GATA1 (28, 29), GATA2 (30, 31), and GATA3 (32, 33) belong to one subfamily and are expressed in overlapping patterns in hematopoietic cells, whereas GATA4 (34, 35), GATA5 (36, 37), and GATA6 (36, 38) comprise a second subfamily of transcription factors that are expressed in the developing heart, gut, and smooth muscle cells. GATA proteins regulate early transcriptional checkpoints during mammalian development. GATA1 is required for erythroid differentiation (39, 40). GATA2 controls the determination and/or survival of the pluripotent hematopoietic stem cell (41). GATA4 is critical for embryonic heart tube formation and ventral morphogenesis (42, 43). GATA3 was first identified as a transcription factor that binds to the TCRα gene enhancer (32). Expression of GATA3 in the hematopoietic lineages is restricted to T cells and NK cells (32). Consensus GATA3-binding sites are required for the expression of multiple T cell genes including the TCRα, -β, and -δ genes; and the CD8α gene (32, 33, 44–46). During embryonic development, GATA3 is expressed at high levels in the central nervous system, kidney, and rudimentary thymus (47, 48). Targeted disruption of the GATA3 gene in mice resulted in embryonic lethality between E11 and E12 (49). The GATA3-deficient embryos displayed massive internal bleeding, marked growth retardation, and severe deformities of the brain and spinal cord; however, the primary defect responsible for the embryonic lethality seen in these mice is unclear. The early embryonic lethality observed in the GATA3−/− mice precluded an analysis of the role of GATA3 in T cell development. Accordingly, Ting et al (51) used the RAG2 complementation system (50) in conjunction with GATA3−/− embryonic stem (ES) cells to further evaluate the role of GATA3 in T cell development and function. Because mature B and T cells cannot develop in the absence of RAG2, all B and T cells present in chimeric mice produced by injection of RAG2−/− blastocysts with GATA3−/− ES cells were derived from the GATA3-deficient ES cells (50). GATA3−/−RAG2−/− chimeric mice contained normal B cell populations. However, there was an absence of doublepositive (DP) and single-positive (SP) cells in the thymus and a complete lack of mature T cells in the peripheral lymphoid organs of the chimeric animals (51). Molecular and biochemical analyses of the defect in T cell development

P1: SAT/MBG/SPD

January 21, 1999

154

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

demonstrated that GATA3 is required for the development and /or survival of the earliest committed double-negative (DN) thymocytes or their precursors (51). Thus, these studies defined GATA3 as the earliest known transcription factor required specifically for T cell lineage commitment. The mechanism by which GATA3 regulates T cell lineage commitment and the targets of GATA3 in early T cell progenitors are unknown and are the subject of ongoing investigation. As described below, GATA3 also plays a role in the differentiation of Th2 cells after T cell activation. Thus, this single zinc-finger transcription factor appears to regulate multiple stages of T cell differentiation.

Ets1 and IRF1 Ets1 is the prototype member of a large family of eukaryotic Ets transcription factors, some of which are also protooncogenes (52, 53). All members of this family share a related winged helix-loop-helix DNA-binding domain and recognize a conserved purine-rich sequence motif centered around a GGAA/T core (54). Ets proteins play important roles in regulating gene expression and cellular differentiation in a large number of species, from flies and worms to humans (55–57). Ets1 is a 56-kDa protein that binds to functionally important sites in the TCRα and TCRβ enhancers (reviewed in 54). Ets1 cooperates with the AP1 transcriptional complex to activate cellular growth factor responses (54, 58). In adult mice, Ets1 is expressed preferentially at high levels in B, T, and NK cells (54, 222). The activity of Ets1 is highly regulated at both the transcriptional and posttranslational levels by T cell activation (59, 60). In resting T cells, Ets1 is expressed in an unphosphorylated form that can bind to DNA and regulate transcription from Ets-dependent promoters. Upon T cell activation, Ets1 is phosphorylated on four Ser residues (61). This phosphorylation both inactivates the DNA-binding activity of the protein and dramatically accelerates its degradation (61). These findings suggested that Ets1 might play a critical role in regulating gene expression in resting T cells. To assess the role of Ets1 in lymphocyte development and function, two groups produced targeted mutations of the Ets1 gene (62, 63). Both mutations resulted in null alleles. As described below, Ets1 is not required for the development of mature B and T lymphocytes; however, mature Ets1-deficient T and B cells displayed several functional defects. In contrast, Ets1 is required for the development and/or survival of the NK cell lineage in mice. Ets1-deficient mice displayed markedly reduced or absent splenic CD3−DX5+ NK cells, and splenocytes from these animals failed to lyse a variety of NK cell targets in vitro or to produce IFN-γ after stimulation with poly (IC) (222). Moreover, the Ets1-deficient mice, unlike their wild-type littermates, consistently developed tumors after injection with NK-susceptible RMA-S tumor cells. These studies demonstrated that Ets1 is required specifically for the development of

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

155

the NK cell lineage, and they defined a novel NK-cell lineage developmental pathway. The targets of Ets1 in NK cells and the mechanism by which Ets1 deficiency resulted in the specific arrest of NK cell development are unknown. Gene targeting experiments have shown that a second transcription factor, IRF1 (interferon response factor 1), is also required for the development of functional NK cells in mice (64, 65). Recent bone marrow transplant experiments have shown that IRF1 promotes NK cell differentiation by inducing the expression of IL-15 from bone marrow stromal cells rather than via an intrinsic NK-cell mechanism (66).

TRANSCRIPTION FACTORS THAT REGULATE THYMOCYTE ONTOGENY Since 1988, we have learned a great deal about thymocyte development (see Figure 2). We now understand the lineage relationships among the different thymocyte subsets and have developed both cell-surface and molecular markers that facilitate the detailed analysis of thymocyte development in genetically manipulated mice (Figure 2). Murine T cell precursors initially migrate from the fetal liver to the fetal thymus on E13 (reviewed in 11). The earliest T cell precursors are so-called triple-negative (or double-negative, DN) cells that lack expression of the CD3/T cell antigen receptor (TCR) complex and the CD4 and CD8 coreceptors. These DN cells subsequently express the recombinaseactivating genes, RAG1 and RAG2; rearrange their TCRβ genes, and express mature TCRβ in conjunction with pre-TCRα on their cell surface (reviewed in 5, 67). Engagement of this pre-TCR is required to rescue these early progenitors from cell death, to stimulate their proliferation, and to allow their

Figure 2 Illustration of thymocyte ontogeny. The double-negative (DN), immature single-positive (ISP), double-positive (DP), and single-positive (SP) thymocyte subsets along with representative lineage-specific cell-surface markers are shown. The roles of the TCF1, LEF1, Sox4, and LKLF transcription factors are depicted schematically.

P1: SAT/MBG/SPD

January 21, 1999

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

156

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

differentiation into DP thymocytes that coexpress CD4 and CD8 (reviewed in 68). DP thymocytes, which comprise more than 70% of the lymphoid cells in the thymus, undergo two important selective events that together result in the elimination of as much as 95% of the DP population. During negative selection, cells expressing TCRs with high affinity for self-antigen plus self–major histocompatability complex (MHC) molecules are eliminated by a process of programmed cell death (reviewed in 69). In contrast, during positive selection, DP cells that express TCRs that are capable of lower-affinity interactions with self-MHC (plus peptide antigen) are rescued from apoptosis and allowed to differentiate into SP CD4+ helper or CD8+ cytotoxic T cells (reviewed in 69, 70). These two selective events efficiently eliminate self-reactive clones and promote the development of a T cell repertoire that is capable of recognizing a variety of foreign antigenic peptides in conjunction with self-MHC molecules on antigen-presenting cells. During the final stages of thymocyte ontogeny, the selected SP cortical thymocytes up-regulate their levels of TCR expression and migrate to the thymic medulla, from which they are exported to populate the peripheral lymphoid organs (reviewed in 71). A complete molecular understanding of thymocyte development would require the elucidation of the extracellular signals, cell-surface receptors, intracellular signaling molecules, transcription factors, and gene targets that regulate each of the developmental stages described above. Although much remains to be learned before we accomplish this goal, since 1993 we have made important strides in understanding the players involved in thymocyte ontogeny. Recent studies have identified three related HMG (high mobility group) box transcription factors—TCF1, LEF1, and Sox4—as important regulators of thymocyte development. These proteins appear to be important for the expansion of DN thymocytes and for their differentiation into DP cells. We describe the role of these transcription factor in T cell development in detail in the following sections.

TCF1/LEF1 TCF1 (T cell factor 1) and LEF1 (lymphoid enhancer-binding factor 1) are two closely related members of the HMG box family of DNA-binding proteins. The two proteins share nearly identical HMG DNA-binding domains and also display a high degree of sequence homology outside of their DNA-binding regions (72–74). TCF1 was first isolated by its ability to bind to the CD3-ε gene enhancer element (72). Subsequent studies showed that TCF1 can also bind with high affinity to the A/TA/TCAAAG consensus motif found in the TCRα, TCRβ, and TCRδ enhancers (75). Unlike TCF1, LEF1 was first isolated from pre-B cells by subtraction hybridization (74). Subsequently, it was shown that LEF1 can also bind to and transactivate the same A/TA/TCAAAG consensus motif of the TCRα enhancer bound by TCF1 (74). During murine embryonic

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

157

development, TCF1 and LEF1 are widely expressed in the developing limb buds, tooth buds, neural crest, urogenital ridge, lung, and thymus; however, their postnatal expression is restricted to lymphocytes (76). In addition to their structural relatedness and shared DNA-binding specificities, TCF1 and LEF1 also display largely overlapping patterns of expression during thymocyte development (76, 77). Both genes are expressed in all T cell subsets from early DN cells to mature peripheral SP T cells. However, TCF1 is expressed slightly earlier during DN thymocyte differentiation than LEF1, and the level of expression of TCF1 is higher than that of LEF1 in both thymocytes and mature T cells (77). Only LEF1 is expressed in the B cell lineage during pro- and pre-B cell maturation (76). Both LEF1 and TCF1 may function by bending the DNA elements to which they are bound (78). Such protein-mediated DNA bending may promote the assembly of multiple transcription factors on complex enhancers such as the TCRα and TCRβ enhancers and may also facilitate contacts between proteins bound to the enhancers and promoters of a single gene. To better understand the function of TCF1 and LEF1 in T cell development, gene targeting was used to produce TCF1- and LEF1-deficient mice (77, 79). Two different mutations of the TCF1 gene were generated: The first deleted exon, VII, which encodes an essential part of the HMG box DNA-binding domain, and the second deleted exon, V, which encodes a domain of the protein that is conserved between Drosophila, mice, and humans (77). The TCF1(VII)deficient mice were healthy and fertile; however, they exhibited a specific block in the proliferation of DN thymocytes and their maturation to the DP stage of thymocyte development. This block, which was incomplete, resulted in a 10- to 100-fold reduction in the numbers of DP and SP thymocytes. As a result, mature peripheral T cell numbers were reduced by 10-fold in the lymph nodes and 3-fold in the spleens of these animals. Further analysis of the TCF1(VII )-mutant mice demonstrated that the block in thymocyte development involved the immature single-positive (ISP) (CD8+) thymocytes, the subset that immediately precedes the DP stage of thymocyte development. In the TCF1(VII )-deficient mice these ISP cells failed to undergo the vigorous cell cycling normally seen in wild-type animals and subsequently failed to differentiate efficiently into DP thymocytes (77). The TCF1(V )-mutant mice displayed a similar but milder phenotype with less severe reductions in the numbers of DP and SP cells. Unlike the TCF1-mutant mice, LEF1-deficient animals had no detectable lymphoid defects at birth (79). However, the LEF1-deficient mice died postnatally with developmental defects of multiple organs, including the mammary glands, mesencephalic nucleus of the trigeminal nerve, teeth, whiskers, and hair. The lack of a lymphoid phenotype in the LEF1-deficient mice, when taken together with the finding that both TCF1 and LEF1 can bind to and activate the TCRα enhancer raised the possibility that TCF1 and LEF1 function

P1: SAT/MBG/SPD

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

158

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

in a partially redundant fashion in T cell development. To address this question, TCF1 and LEF1 double-mutant mice were generated (80). To circumvent the postnatal lethality caused by the lack of LEF1, fetal thymic organ cultures (FTOC) were used to study T cell development in these mice. FTOC from LEF1−/−TCF1(V)−/− animals exhibited a partial block at the CD44−CD25+ stage of DN thymocyte development and a complete block in the differentiation of more mature ISP to DP thymocytes (80). As a consequence of these developmental defects, these mice lacked DP and mature SP thymocytes and peripheral T cells (80). These results demonstrated that LEF1 and TCF1 play partially redundant roles in regulating the transition of DN to DP thymocytes. The mechanism and target genes that are responsible for the ability of these proteins to promote DP thymocyte expansion and maturation are unknown. Although the double-mutant mice displayed reduced expression of the TCRα gene, this cannot account for the phenotype of the animals as TCRα expression is not required for the DN to DP transition. The TCRβ gene contains a potential TCF1/LEF1-binding site, and TCRβ gene expression (in conjunction with pre-TCRα expression) is required for the proliferation and maturation of DN thymocytes. However, both pre-TCRα and TCRβ gene expression were normal in the TCF1/LEF1deficient thymocytes. Curiously, the defect in DP thymocyte development seen in the double-mutant mice could be rescued by treatment of the FTOC with an α-CD3 mAb, suggesting that the TCR signaling pathway responsible for the maturation of the DN thymocytes was intact in these animals. Further studies will be needed to elucidate more precisely the mechanism(s) by which TCF1 and LEF1 regulate early thymocyte maturation.

Sox4 Sox4 is the third member of the HMG box family of T cell transcription factors. It was first identified using a low-stringency PCR screen of mouse T cell RNA and is highly related to TCF1, LEF1, and the sex-determining gene, Sry (81). In vitro experiments showed that Sox4 can bind with high affinity to the LEF1- and TCF1-binding motif A/TA/TCAAAG found in many T cell enhancers and can activate transcription via its C-terminal serine-rich domain (81). In adult mice, Sox4 expression is restricted to immature B and T cells and the gonads (81). During embryonic development, Sox4 is also expressed at high levels in the endocardial cushions and ridges of the developing mouse heart (82). Mouse embryos containing targeted mutations of the Sox4 gene died at E14 from defects in heart development (82). These mice displayed severely impaired development of the semilunar valves and cardiac outflow tract, resulting in circulatory failure (82). To understand the function of Sox4 in lymphocyte

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

159

development, reconstitution experiments were performed in which Sox4−/− fetal liver cells were injected into sublethally irradiated mice. These experiments demonstrated that Sox4 is required for the expansion and differentiation of pro-B cells (82). In agreement with these experiments, the in vitro differentiation of B cell precursors from Sox4−/− fetal liver cells in response to IL-7 was reduced by approximately 10-fold (82). In contrast, apparently normal T lymphocyte reconstitution (at least as measured by numbers of TCRα/β cells in the lymph node) was observed after adoptive transfer of Sox4-deficient fetal liver cells to irradiated recipients (82). Further analysis of the role of Sox4 in T cell development was performed using FTOC from wild-type or Sox4-deficient animals (83). In these studies, immature thymocytes from E13 Sox4−/− embryos showed a 10- to 50-fold reduction in their ability to differentiate into DP and SP thymocytes as compared to age-matched wild-type cells (83). Consistent with these results, the Sox4−/− fetal liver progenitor cells also displayed a reduced ability to compete with wildtype cells in reconstituting thymic populations after injection into sublethally irradiated mice (83). Thus, it appears that Sox4, like TCF1 and LEF1, may play an important role in the proliferation and maturation of DN thymocytes. The finding of similar defects in thymocyte maturation in the LEF1/TCF1- and Sox4-deficient mice suggests either that the total level of expression of these three proteins is critical for a single pathway of DN thymocyte development or, alternatively, that the three proteins regulate distinct but parallel pathways that are each at least partially required for DN thymocyte proliferation and maturation. Finally, as discussed below, recent gene targeting studies have suggested that NFATc, a member of the NFAT/Rel/NF-κB family of transcription factors, may also play a role in regulating the maturation of DN to DP thymocytes. NFATc-deficient mice displayed a partial block in the maturation of CD25lowCD44high DN cells to the DP stage of thymocyte development.

TRANSCRIPTION FACTORS REGULATING T CELL QUIESCENCE AND ACTIVATION Naive SP (CD4+ or CD8+) peripheral T cells circulate through the blood and peripheral lymphoid organs in a proliferatively and transcriptionally quiescent state until they encounter an antigen-presenting cell bearing a cognate peptide bound to an appropriate MHC molecule (reviewed in 84). Engagement of the TCR/CD3 complex by peptide plus MHC results in T cell activation, a dynamic process characterized by the highly regulated expression of more than 100 activation-specific genes and concomitant cell cycle progression and proliferation (reviewed in 85–87). As the foreign antigen is eliminated and the immune response winds down, the majority of activated T cells are eliminated

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

160

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

by an apoptotic process termed activation-induced cell death (AICD) (reviewed in 88, 89). A small percentage of antigen-specific T cells escapes cell death and enters a pool of long-lived memory T cells that can be distinguished from naive T cells by their cell-surface phenotype (reviewed in 90). Such memory cells play important roles in generating secondary immune responses upon subsequent reexposure to antigen. Since 1993, a great deal has been learned about the signaling pathways and transcription factors that regulate T cell activation after TCR engagement. Current evidence suggests that most activation-specific genes are regulated by the simultaneous binding of multiple transcription factors including members of the NF-κB/Rel/NFAT, AP1, and CREB/ATF families. In addition, genetic studies in mice have demonstrated that Ets1 is required to generate SP T cells that are competent to receive TCR-mediated activation signals. Several recent studies have suggested that T cell quiescence, like T cell activation, is actively regulated. For example, the maintenance of the naive T cell pool (at least in the case of CD8+ cells) appears to require continuous stimulation or “tickling” by class I MHC-bearing cells (91). At the transcriptional level, the zinc-finger transcription factor LKLF appears to be required to program and maintain the quiescent phenotype in mature SP thymocytes and T cells. We describe the roles of these individual transcription factors in T cell quiescence and activation in detail in the following sections.

LKLF LKLF (lung Kruppel-like factor) belongs to a family of related zinc-finger transcription factors that includes the mammalian proteins EKLF, BKLF, and GKLF (92–96). All members of this family share related DNA-binding domains composed of three contiguous zinc fingers located at the C-termini of the proteins. These evolutionarily conserved DNA-binding domains have sequence similarities to that of the Drosophila protein Kruppel, a transcriptional repressor that determines body patterning (97, 98). They are also related to the zinc-finger DNA-binding domains of several other mammalian transcription factors, including Sp1, BETB2, and Wilms’ tumor 1 (99–101). The KLF family members bind with high affinity to CACCC sequence motifs (or CACCC boxes) that are present in many mammalian promoters and enhancers (92, 94). The prototype member of the KLF family, EKLF, was identified initially as an erythroid lineage-specific transcription factor (92). In vitro experiments showed that EKLF can bind specifically to a critical CACCC motif in the β-globin gene promoter to enhance transcriptional activation (92, 102). Subsequent gene targeting experiments revealed that EKLF is required for γ - to β-globin switching during the terminal stages of erythroid maturation (103, 104).

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

161

LKLF was isolated by two research groups using a low-stringency screen with probes derived from the zinc-finger region of EKLF (93, 105). LKLF is a 40-kDa transcription factor structurally related to the other KLF family members only in its zinc-finger DNA-binding domain. During mouse development, LKLF is expressed at high levels in the lung, vasculature, and lymphoid organs (93, 105). In the thymus, LKLF is expressed exclusively in lymphoid cells in the thymic medulla, a region that contains mature SP thymocytes (105). LKLF is also expressed at high levels in the white pulp of the spleen, a site of residence of splenic T cells (105). Consistent with these findings, Northern blot analyses demonstrated high-level LKLF expression in both CD4+ and CD8+ SP thymocytes and splenocytes, but undetectable expression in less mature DN and DP thymocytes (105). Interestingly, the expression of LKLF is rapidly extinguished at both the mRNA and protein levels after T cell activation (105). Thus, expression of the LKLF gene is developmentally activated during the transition to the mature SP stage of thymocyte development, remains elevated in resting SP T cells, and is rapidly extinguished after the activation of these cells by TCR cross-linking. To assess the role of LKLF in T cell development and function, LKLFdeficient mice were produced using a gene targeting approach that deleted the entire coding region of the gene (106). Consistent with the high level of LKLF expression in developing vascular endothelial cells, the LKLF-deficient embryos exhibited defects in blood vessel formation and died from intra-embryonic and intra-amniotic hemorrhages between E12.5 and E14.5 (106). This early embryonic lethality precluded studies of the role of LKLF in T cell development; therefore, Kuo et al (105) used the RAG2 complementation system (50), in conjunction with the LKLF−/− embryonic stem cells, to assess the role of LKLF in thymocyte ontogeny and T cell function. LKLF was not required for T cell development. LKLF−/−RAG2−/− chimeric mice developed all T cell subsets; however, thymi from the LKLF−/−RAG2−/− mice displayed both cellular and morphological defects. Histological analyses demonstrated significantly increased cellularity in the medulla with reciprocally decreased cellularity in the cortex as compared to control thymi (CT Kuo & JM Leiden, unpublished data). Consistent with these findings, thymi from LKLF−/−RAG2−/− chimeric mice contained decreased numbers of DP cells and an increased proportion of SP cells (105). In addition, the CD4+ and CD8+ SP thymocytes from the LKLF−/−RAG2−/− animals displayed a spontaneously activated cell-surface phenotype (CD44high, CD69high, L-selectinlow) and significantly increased rates of apoptosis in response to treatment with both dexamethasone and γ -irradiation in vitro (105; CT Kuo & JM Leiden, unpublished data). Although the LKLFdeficient SP thymocytes displayed this activated cell-surface phenotype, they

P1: SAT/MBG/SPD

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

162

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

did not express high levels of the high-affinity IL-2 receptor and were not proliferating actively (105). In addition to the defects in thymocyte maturation, the LKLF−/−RAG2−/− mice displayed several profound defects in the phenotype and function of peripheral T cells. First, SP LKLF−/− splenic and lymph node T cells, like their thymic counterparts, displayed an abnormal activated cell-surface phenotype (CD44high, CD69high and L-selectinlow) (105). In addition, total numbers of lymph node and splenic T cells were reduced by more than 90%. Finally, the LKLF−/−RAG2−/− animals uniformly lacked circulating CD4+ and CD8+ T cells. The severe reduction of mature T cells in the periphery of the LKLF−/−RAG2−/− mice was largely the result of increased rates of peripheral T cell apoptosis. TUNEL assays demonstrated large numbers of apoptotic CD3+ T cells in the spleens and lymph nodes of the LKLF−/−RAG2−/− mice (105). Consistent with these findings, purified LKLF−/− splenic T cells displayed a greater than fivefold increased rate of cell death during culture in vitro (105). This increased rate of T cell apoptosis both in vitro and in vivo was associated with increased levels of cell-surface expression of Fas ligand (FasL) on the SP LKLF−/− splenic and lymph node T cells; however, it was not clear whether the observed increase in FasL expression represented a primary cause of peripheral T cell death or was simply a marker of the partially activated phenotype of the LKLF-deficient T cells. These results suggested that the quiescent phenotype of SP T cells is regulated actively at the level of transcription and that LKLF is required to program and maintain this quiescent phenotype. They also raised important questions about the mechanisms by which LKLF regulates this terminal stage of T cell development. No LKLF target genes are known. In addition, the pathways that regulate LKLF expression during SP thymocyte maturation and LKLF degradation following T cell activation are unknown. It will be interesting to determine the role of LKLF in the generation and maintenance of T cell memory and in the class I MHC-dependent survival of peripheral CD8+ T cells.

Ets1 As described above, Ets1 is expressed in a transcriptionally active form in resting SP thymocytes and T cells and is inactivated and degraded after T cell activation (59–61). It is also expressed in B cells and NK cells in mice [107; (K Barton, N Huthusamy, C Fischer, CN Ting, TL Waluras, LL Lanier, JM Leiden), 222]. Gene targeting studies in mice have demonstrated at least three important roles for Ets1 in lymphocyte development and function. First, as described above, Ets1 is required for the development of the NK cell lineage mice (Barton et al, submitted for publication). Second, Ets1-deficient animals contained increased numbers of plasma cells and 5- to 10-fold elevated

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

163

serum IgM levels (63; Barton et al, 222). Thus, Ets1 appears to be required to prevent the spontaneous differentiation of B cells into IgM-secreting plasma cells. The mechanism by which Ets1 regulates B cell differentiation and potential Ets1 target genes in B cells is unknown. Third, although Ets1 is not required for the development of mature SP (CD4+ and CD8+) T cells (62, 63), thymocyte numbers in the Ets1-deficient animals were reduced by approximately 65% (Barton et al, 222). Moreover, although total numbers of peripheral SP T cells were normal in the Ets1−/− mice, CD8+ Ets1-deficient T cells reproducibly expressed low levels of CD4, suggesting a failure of terminal maturation in these cells. SP T cells from the Ets1-deficient mice also displayed a profound defect in activation after cross-linking of the TCR (62, 63; Barton et al, 222). These same Ets1-deficient T cells proliferated normally in response to stimulation with phorbol myristate acetate (PMA) plus ionomycin, suggesting that the activation defect in these cells was membrane proximal (Barton et al, 222). A logical model that could explain these results is that Ets1 normally regulates the expression of one or more proximal signal transduction molecules in T cells (and perhaps in the highly related NK cell lineage). The reduced expression of this molecule in the Ets1-deficient T cells would lead to a membrane-proximal activation defect that might result in both decreased thymocyte expansion (thereby accounting for the decreased numbers of thymocytes) and defective TCR-mediated proliferation. The absence of this same molecule in NK cells might be responsible for the observed developmental defect in this lineage. Candidates for such Ets1-regulated proximal signal transduction molecules include p56lck, Zap70, Vav, LAT, and SLP76. Ets1-binding sites have been identified in the proximal promoter of the p56lck gene (108). The patterns of expression of these molecules in the Ets1-deficient T cells is under investigation.

NF-κB The mammalian NF-κB transcription factors, which include NF-κB1 (p50/ p105), NF-κB2 (p52/p100), RelA (p65), c-Rel, and RelB, play important roles in the regulation of immune and inflammatory responses, cellular proliferation, and cell death (reviewed in 109–111). Transcriptional activation or repression of NF-κB target genes requires the binding of NF-κB dimers to κB DNA-binding sites, and evidence suggests that different NF-κB dimers regulate the expression of different genes (112, 113). In most cells, the transcriptional activity of NF-κB proteins is controlled at the posttranslational level by association with members of the IκB family of inhibitory proteins. NF-κB proteins are retained in the cytoplasm of unstimulated cells in an inactive form via interactions with one or more of the seven known IκB proteins (reviewed in 110, 114). In response to a variety of stimuli—including cell-surface antigen receptor cross-linking

P1: SAT/MBG/SPD

January 21, 1999

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

164

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

and exposure to cytokines, bacterial components, viruses, UV radiation, or oxidative stress—IκB-α and IκB-β proteins are phosphorylated and degraded and free cytoplasmic NF-κB dimers are translocated rapidly to the nucleus, where they regulate κB-dependent gene expression (115–117). NF-κB transcription factors are thought to play important roles in T cell activation and development (118, 119). Functionally important κB-binding sites have been identified in a large number of T cell transcriptional regulatory elements, including the IL-2, IL-2 receptor α, GM-CSF, and MIP-2 promoters (reviewed in 120). Preformed NF-κB proteins are present in the cytoplasm of thymocytes and resting peripheral T cells (121). TCR engagement results in the rapid inactivation of IκB-α and subsequent nuclear migration of active NF-κB dimers (119, 122, 123). Gene targeting experiments have demonstrated that RelB, c-Rel, and NF-κB1 play distinct roles in regulating the development and function of the mammalian immune system. NF-κB1-deficient mice displayed defects in B cell proliferation in response to the mitogen lipopolysaccharide (LPS) but not to IgM cross-linking (124, 125); however, T cell development appeared to be normal in these animals. Mice lacking RelA died around E15 from massive hepatocyte apoptosis (126, 127); however, progenitor cells derived from RelA−/− mice gave rise to normal T cells, suggesting that RelA is not required for T cell development (127, 128). RelB is required for the differentiation and survival of dendritic cells and thymic medullary epithelial cells, and RelB−/− mice displayed severe defects in cellular immune responses (129, 130). This immune dysfunction was worsened in NF-κB1/RelB doubleknockout mice, suggesting that the lack of RelB is compensated by other NFκB1–containing dimers (131). In contrast, c-Rel-deficient mice demonstrated defective B and T cell proliferation in response to mitogen stimulation and markedly decreased IL-2 production after TCR engagement (132). Despite these proliferative defects, T cell development appeared to be normal in the c-Rel−/− mice. Thus, these gene targeting experiments did not identify an essential role for NF-κB proteins in T cell development, although they did identify a lymphocyte-autonomous and essential role for c-Rel in IL-2 production and T cell proliferation after TCR engagement. Although gene targeting experiments have been useful for identifying the essential nonredundant roles of individual NF-κB transcription factors in mammalian development, the interpretation of these experiments can be obscured both by potential functional redundancies of related NF-κB proteins in the mutant animals and by early embryonic lethality, which precludes a complete analysis of T cell development and function. Moreover, because NF-κB proteins are expressed in many different cell types it can be difficult to determine whether observed defects in lymphocyte development and function are lymphocyte autonomous. To circumvent some of these problems, several groups have

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

165

produced transgenic mice expressing a superinhibitory form of the IκB-α protein under the control of T cell–specific promoters and enhancers (133, 134; T Hettmann, JD Donato, M Karin, JH Leiden, submitted for publication). This mutant form of IκB-α cannot be phosphorylated and degraded in response to TCR engagement and, therefore, constitutively inhibits all forms of inducible NF-κB activity after T cell activation (135–137). Studies of the mutant IκB-α (mIκB-α) transgenic mice revealed several important functions for NF-κB proteins in T cell development and function, some of which were not appreciated from analyses of the mutant NK-κB mice produced by gene targeting. First, although NF-κB is not required for normal thymocyte development, the mIκB-α transgenic animals displayed significantly decreased numbers of peripheral CD8+ T cells (133, 134; T Hettmann, JD Donato, M Karin, JH Leiden, submitted for publication). The magnitude of the reduction of the CD8+ T cell population was proportional to the level of transgene expression (133; T Hettmann, JD Donato, M Karin, JH Leiden, submitted for publication). This defect may have reflected an essential role for NF-κB in the export of CD8+ cells from the thymus (where they were produced normally) or a role of an NF-κB signal in the class I MHC-dependent survival signal for CD8+ peripheral T cells. Second, the IκB-α transgenic T cells displayed a severe proliferative defect in response to TCR cross-linking and to stimulation with concanavalin A (ConA) or PMA plus ionomycin. This proliferative defect, which was only partially rescued by α-CD28 costimulation, was associated with marked reductions in the production of several cytokines, including IL-2, IL-3, and GM-CSF. These experiments identified an essential role for NF-κB proteins in T cell proliferation and cytokine production after TCR engagement. Moreover, they demonstrated that the CD28 costimulatory signal is only partially mediated by an NF-κB-dependent signaling pathway. The third and perhaps most surprising role for NF-κB was its essential function as a pro-apoptotic signaling molecule in DP thymocytes. Wild-type DP thymocytes are exquisitively sensitive to a variety of apoptotic signals, including TCR cross-linking, glucocorticoids, and ionizing radiation. In contrast, the mIκB-α DP thymocytes were resistant to α-CD3 treatment in vivo (T Hettmann, JD Donato, M Karin, JH Leiden, submittted for publication); however, they remained fully sensitive to γ -irradiation-induced cell death. Apoptosis of wild-type DP thymocytes in response to α-CD3 treatment was preceded by the marked downregulation of the antiapoptotic gene, Bcl-xL. In contrast, the mIκB-α transgenic DP thymocytes maintained high-level expression of Bcl-xL after α-CD3 administration in vivo (T Hettmann, JD Donato, M Karin, JH Leiden, submitted for publication). Thus, these experiments showed that NF-κB proteins are required for α-CD3–induced apoptosis of DP thymocytes

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

166

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

via a pathway that involves the regulation of the antiapoptotic gene Bcl-xL. It will be interesting to determine whether NF-κB proteins regulate the Bcl-xL promoter directly and whether the mIκB-α–mutant mice exhibit defects in negative selection, a process that also involves a TCR-mediated apoptotic signal in DP thymocytes.

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

CREB CREB (cAMP response element–binding protein) is a 43-kDa basic/leucine zipper (b-ZIP) transcription factor that binds to the octanucleotide CRE element (TGANNTCA) both as a homodimer and as heterodimers in conjunction with other members of the CREB/ATF superfamily of transcription factors (138–140). The transcriptional activity of CREB is regulated by phosphorylation of a single serine residue (Ser133) (141). In resting cells (including resting T cells), CREB exists in an unphosphorylated state that can bind to DNA but is transcriptionally inactive (141). After cell activation, CREB is phosphorylated on Ser133, which activates its transcriptional activity at least in part by promoting its interaction with the 256-kDa coactivator protein, CBP (142–144). The CREB-CBP complex can interact with and activate the basal transcription complex. CREB phosphorylation and activation can be mediated by multiple signaling pathways in different cell lineages. These pathways include (a) a PKA-dependent pathway activated by increased intracellular concentrations of cAMP, (b) a calmodulin kinase–dependent pathway activated by increased intracytoplasmic Ca2+, and (c) a Ras-dependent pathway in which RSK2 can phosphorylate CREB on Ser133 (140, 141, 145–147). Recent studies have demonstrated that each of these pathways is functional in T cells; however, after TCR engagement, CREB is phosphorylated rapidly on Ser133 by a single pathway that involves activation of p56lck, protein kinase C, Ras, Raf-1, MEK, and RSK2 (148). Functionally important CRE elements are present in the promoters and enhancers of many T cell–specific genes, including the TCRα enhancer, TCR Vβ promoter, CD3δ enhancer, and the CD8α promoter (149–152). To better understand the function of CREB in T cell development and function, Barton et al (153) produced transgenic mice expressing a dominant negative form of CREB under the control of the T cell–specific CD2 promoter/enhancer. This dominant negative CREB mutant (Ser133 to Ala) retains DNA-binding activity but is rendered transcriptionally inactive and unresponsive to activation signals. The CREBA133 transgenic mice exhibited normal T cell development; however, SP thymocytes from these animals displayed a profound defect in T cell proliferation after stimulation with α-CD3, PMA plus ionomycin, or ConA (153). This reduced proliferative response was associated with markedly decreased IL-2 production; however, the proliferative defect of these cells could

P1: SAT/MBG/SPD

P2: NBL/plb

January 21, 1999

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

167

not be rescued by the addition of exogenous IL-2, suggesting that additional proliferative pathways were defective in the CREBA133 thymocytes. Because activation of T cells in the absence of IL-2 leads to programmed cell death, it was not surprising that the CREBA133 thymocytes displayed a G1 cell cycle arrest and subsequent apoptotic death in response to a number of different activation signals. Consistent with previous findings that CREB plays an important role in regulating the expression of AP1 proteins, Barton et al and other researchers (153–156) demonstrated that the T cell proliferative defects seen in the CREBA133 mice were associated with the decreased induction of c-jun, c-fos, Fra-2, and FosB after TCR cross-linking. These results were consistent with a model in which the rapid phosphorylation of CREB on Ser133 after TCR cross-linking is required for the induction of AP1 and IL-2 and subsequent cell cycle progression and proliferation. In the absence of functional CREB, T cells were arrested at the G1/S checkpoint and underwent programmed cell death in response to a variety of activation signals. Given the importance of CREB in regulating T cell activation and proliferation, it will be interesting to identify additional functionally important CREB target genes in activated T cells.

NFAT NFAT (nuclear factor of activated T cells) was first identified as an inducible nuclear protein complex that could bind to a T/AGGAAAATN TGTTTCA sequence motif present in the distal antigen receptor response element of the human IL-2 promoter (157). Subsequently, important NFAT-binding sites were also identified in the transcriptional regulatory regions of multiple activationspecific T cell genes, including the IL-3, IL-4, GM-CSF, and tumor necrosis factor (TNF ) α genes (reviewed in 158). The NFAT transcriptional complex that bound to these sites contains a cytoplasmic subunit that is expressed in resting T cells, and a nuclear component that is present in T cells after stimulation with phorbol esters (159). The nuclear component of NFAT is the AP1 transcription factor complex, which is composed of dimers of Fos and Jun proteins (160). Subsequently, two novel and related genes encoding the cytoplasmic component of NFAT were cloned. These genes, NFATp (also called NFAT1) and NFATc (also called NFAT2), encoded DNA-binding domains that were distantly related to those of the Rel/NF-κB proteins described above (161–164). Two additional NFAT genes (NFAT3 and NFAT4) were cloned by low-stringency screening approaches using the Rel DNA-binding domain of NFATp (165). Thus, the NFAT gene family has four known members. Cotransfection of NFATp or NFATc expression vectors (with or without AP1) transactivates an IL-2 promotor reporter construct, demonstrating directly the role of these factors in positively regulating IL-2 gene transcription (166, 167).

P1: SAT/MBG/SPD

January 21, 1999

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

168

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

More recent studies have demonstrated that the NFAT proteins are retained in the cytoplasm of resting T cells in a Ser phosphorylated form (168; reviewed in 158). After T cell activation, NFAT proteins are dephosphorylated by the protein phosphatase calcineurin (reviewed in 158, 169). This dephosphorylation unmasks a nuclear localization signal, facilitating the rapid translocation of NFAT proteins to the nucleus, where they pair with AP1 and bind to consensus NFAT sites (158, 169). The immunosuppressive drugs cyclosporin A and FK506 prevent the calcineurin-mediated dephosphorylation of NFAT, accounting for at least some of their immunosuppressive effects on T cells (170–172). The four NFAT genes display markedly different patterns of expression, suggesting that they serve distinct functions in mammalian organisms. NFATp and NFATc are expressed in T cells and thymocytes (reviewed in 158); however, NFATp is also expressed in skeletal myocytes, pancreas, and placenta, whereas NFATc is highly expressed in skeletal muscle and testis (165). NFAT3 is not expressed in T cells but is widely expressed in adult tissues; high-level expression is observed in the heart, lung, kidney, placenta, and testis (165). In contrast, NFAT4 expression is localized tightly to thymocytes and skeletal muscle (165, 173, 174). Recent gene targeting experiments have begun to elucidate the roles of NFAT proteins in T cell development and function. NFATp-deficient mice produced by targeted deletion were viable and fertile (175, 176). Thymocyte development was normal in these mice; however, they demonstrated a number of immune defects generally characterized by abnormally increased T cell activity. Splenic T cells from the NFATp−/− mice displayed increased proliferation in response to α-CD3 treatment, an increased primary immune response to Leishmania major, and an increased secondary immune response to ovalbumin in vitro (175, 176). In addition, the NFATp−/− mice accumulated abnormally large numbers of eosinophils and generated increased levels of serum IgE in an in vivo model of allergic inflammation (175, 176). NFATp-deficient B cells were also hyperproliferative after activation with α-IgM or α-CD40 (175). Despite these hyperproliferative responses, the induction of IL-3, IL-13, GM-CSF, TNFα, FasL, and CD40L were defective after α-CD3 stimulation of NFATp-deficient splenocytes (175). The induction of IL-2 after α-CD3 stimulation was essentially normal in these mice (175, 176). Finally, as described below, the NFATp-deficient animals displayed dysregulated IL-4 production, which resulted in a skewing toward the Th2 pathway of T cell differentiation both in vitro and in vivo. These results demonstrated a surprisingly complex role for NFATp in T cell function. On the one hand, NFATp appeared to play an important negative regulatory role both in T cell proliferation and in the maintenance of IL-4 production. On the other hand, NFATp was clearly required for the transcriptional induction of multiple genes after T cell activation, including cytokines such as IL-3, IL-13, GM-CSF, and TNFα and cell-surface molecules such as FasL and

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

169

CD40. NFATp did not appear to be required in the regulation of IL-2 gene expression. Targeted deletion of NFATc resulted in embryonic circulatory failure and subsequent embryonic lethality around E14.5 (177, 178). Heart failure was caused by specific defects in the formation of the cardiac valves and interventricular septum (177, 178). Thus, in line with its expression in endocardial cushion cells, NFATc appears to play a critical role in cardiac septation and valve formation. To study the function of NFATc in the immune system, researchers used the RAG2 complementation system in conjunction with NFATc−/− embryonic stem cells (179, 180). NFATc−/−RAG2−/− chimeric mice displayed reduced numbers of thymocytes and peripheral T cells because of a partial block in the maturation of CD25lowCD44− DN thymocytes to DP cells (179, 180). The NFATc−/− T cells exhibited mildly impaired proliferation and normal IL-2 production after activation with a variety of stimuli (179, 180). Thus, like NFATp, NFATc does not appear to be required for either IL-2 production or T cell proliferation. Perhaps the most striking defect observed in the NFATp-deficient T cells was a significant reduction in IL-4 production after T cell activation (179, 180). The NFATc-deficient animals displayed a marked decrease in serum IgG1 and IgE levels and decreased production of IgG1 after stimulation with lipopolysaccharide in vitro, both of which were consistent with decreased Th2 responses. These results suggested that NFATc, unlike NFATp, is important for early thymocyte development and also plays an important positive role in regulating IL-4 and Th2 responses both in vitro and in vivo. These studies also suggested that the NFAT family members each play quite distinct roles in the regulation of T cell development and function. This finding presumably reflects the distinct spatial and temporal patterns of expression of the proteins and possibly their ability to bind differentially to different NFAT sites. The NFATp and NFATc gene targeting experiments did not reveal a role for either NFATp or NFATc in regulating IL-2 gene expression or proliferation after T cell activation. It will be interesting to elucidate the phenotype of the NFAT4-deficient mice (the only other known NFAT family member expressed in T cells) and to produce mice deficient in multiple NFAT family genes by intercrossing.

TRANSCRIPTIONAL REGULATION OF THE Th1/Th2 RESPONSE CD4+ helper T cells can be divided into two subsets based on their patterns of cytokine expression and their subsequent roles in immune responses (reviewed in 181, 182). T helper 1 (Th1) cells secrete IL-2 and IFNγ and are important regulators of cell-mediated immune responses, whereas T helper 2 (Th2) cells secrete IL-4, IL-5, and IL-10 and mediate predominantly humoral and

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

170

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

Figure 3 Transcriptional regulation of Th1/Th2 T helper cell differentiation. The roles of Stat4, Stat6, GATA3, NFAT, and c-Maf transcription factors in regulating the differentiation of naive T helper cells into Th1 and Th2 subsets are depicted schematically.

eosinophilic responses (reviewed in 183, 184). Alterations of the ratio of Th1 to Th2 responses are important determinants of susceptibility to viral and parasitic infections, allergies, antitumor responses, and autoimmunity (reviewed in 185). Thus, an understanding of the molecular pathways that regulate the differential expression of cytokine genes and the differentiation of these two T helper subsets holds promise for the therapeutic manipulation of the mammalian immune system. Four families of transcription factors have been identified that appear to be important regulators of the Th1/Th2 switch (see Figure 3). Two members of the Stat (signal transducers and activators of transcription) family appear to play reciprocal roles in this response: Stat6 is required for IL-4-responsive Th2 differentiation, and Stat4 is necessary for IL-12-responsive Th1 development. The zinc-finger protein, GATA3, which, as described above, is required for the earliest stages of DN thymocyte development, also appears to be an important regulator of Th2 differentiation. NFATp and NFATc appear to regulate the balance of Th2 responses: NFATp down-regulates late IL-4 production, and NFATc is a positive regulator of IL-4 production. Finally, the b-ZIP protein c-Maf is a critical early regulator of IL-4 production and Th2 differentiation. We discuss the roles of these transcription factors in Th1/Th2 differentiation in detail in the following sections.

Stat4/6 Stat4 and Stat6 belong to a family of STAT transcription factors known to mediate an array of cytokine-induced responses (reviewed in 186, 187). Members

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

171

of this family share structurally related SH2 and SH3 domains (188). Before cytokine receptor engagement, Stat proteins are retained in the cytoplasm in a hypophosphorylated and inactive form. After cytokine receptor-mediated signaling, Stat proteins are phosphorylated by the Janus kinase (Jak) family of protein tyrosine kinases (reviewed in 186, 187). Phosphorylated Stat proteins dimerize (via their SH2 domains) and rapidly translocate to the nucleus, where they bind to and transactivate genes containing the gamma-activated sequence (GAS) TTCN4GAA (reviewed in 186, 187). Stat4 was first identified by a low-stringency PCR screening approach using primers derived from Stat1 and Stat2 cDNAs (189, 190). Stat4 is a 85-kDa protein that, unlike other Stat family members, is preferentially expressed in the spleen, lung, muscle, testis, and myloid cells (189). After IL-12 stimulation, which initiates the differentiation of naive CD4+ T cells into the Th1 phenotype, Stat4 is tyrosine phosphorylated and becomes transcriptionally active (191, 192). The finding that Stat4 is activated only after T cell stimulation with IL-12 raised the possibility that it was involved in the process of T helper cell differentiation. Stat4-deficient mutant mice produced by gene targeting were viable and fertile and had no apparent anatomical abnormalities; however, these mice demonstrated severe defects in all aspects of IL12-mediated immune responses (193, 194). Interferon-γ production, T cell proliferation, and NK cell cytolytic functions were all severely defective in the Stat4-deficient mice (193, 194). Furthermore, the ability of Stat4−/− CD4+ T cells to differentiate into the Th1 cells in response to Listeria monocytogenes or IL-12 was also defective (194). Thus, these results demonstrated a nonredundant and essential role for Stat4 in regulating IL-12-mediated Th1 immune responses. A related Stat protein, Stat6, is widely expressed in adult tissues and in hematopoietic cell lines and becomes activated in response to IL-4-mediated signaling (195–197). The engagement of the IL-4 receptor leads to tyrosine phosphorylation of Stat6 by Jak1 and Jak3 and results in the activation of IL-4regulated genes such as IgE, IL-4R, FcR, and class II MHC that contain GASbinding sites (198–202). The function of Stat6 in IL-4 signaling was analyzed in three independently derived Stat6-deficient mice (203–205). Although Stat6 is expressed in many adult tissues, Stat6-deficient animals showed no gross abnormalities; however, these mice displayed severe defects in IL-4-mediated responses (203–205). Stat6−/− lymphocytes failed to up-regulate class II MHC, CD23, and IL-4R in response to IL-4 stimulation (203–205). In addition, they failed to proliferate in response to exogenous IL-4 (203–205). Consistent with these defects in IL-4 responsiveness, Stat6−/− B cells did not undergo IgE class-switching after anti-IgD treatment (203–205). Perhaps most strikingly, the naive Stat6-deficient T cells failed to differentiate into Th2 cells after in

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

172

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

vitro stimulation in the presence of either IL-4 or IL-13 (203–205). Thus, these results demonstrated the importance of Stat6 in regulating both IL-4-mediated immune responses and the development of Th2 cells.

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

GATA3 In addition to serving as a critical regulator of early T cell development, the GATA3 transcription factor regulates both Th2 cytokine expression and the differentiation of naive CD4+ T helper cells to Th2 cells (206, 207). Using representational difference analysis (RDA), researchers found that GATA3 was up-regulated selectively during the differentiation of Th2 but not Th1 cells (206, 207). Because GATA3-deficient mice are embryonic lethal (49), and GATA3−/−RAG2−/− chimeric mice do not generate mature T cells (51), standard loss-of-function experiments could not be performed to analyze the role of GATA3 in Th2 development. Zheng & Flavell (206) stably transfected Th2 T cell clones with a GATA3 antisense cDNA expression vector. Reduction of GATA3 protein levels in the transfected Th2 cell clones was associated with the abrogation of Th2-restricted IL-4, IL-6, and IL-13 expression by these cells. Expression of other Th2-specific cytokines, such as IL-5 and IL-10, was also inhibited. In complementary gain-of-function experiments, GATA3 was expressed ectopically in the B cell lymphoma M12 and in transgenic mice (206). Transient transfection of M12 B cells, which do not normally express IL-4, with a GATA3 expression vector in conjunction with PMA plus ionomycin stimulation resulted in the transactivation of a cotransfected IL-4 promoter reporter construct (206). To study the role of GATA3 in Th2 differentiation in vivo, Zheng & Flavell (206) produced CD4-GATA3 transgenic mice, in which GATA3 was expressed ectopically in all CD4+ T cells by placing it under the transcriptional control of the CD4 promoter. T cells from these animals, which were stimulated to differentiate into Th1 cells in vitro by treatment with ConA, IL-2, IL-12, and anti-IL-4 mAb, displayed abnormal expression of the Th2 cytokines IL-4, IL-5, IL-6, and IL-10. Further support for the role of GATA3 in Th2 differentiation came from experiments in which GATA3 was shown to bind specifically to the IL-5 promoter and to activate the expression of an IL-5 promoter construct after cotransfection into both Th1 cells and nonlymphoid cells (207, 208). These experiments demonstrated a critical role for GATA3 in regulating both Th2 cytokine production and Th2 differentiation in vitro and in vivo. The fact that GATA3 plays important roles in both early T cell development and later Th2 differentiation suggests that its transcriptional effects in different T cell subsets may be regulated by the differential expression of additional transcription factors or coactivator proteins.

P1: SAT/MBG/SPD

P2: NBL/plb

January 21, 1999

13:7

QC: NBL

Annual Reviews

AR078-06

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

173

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NFAT Because NFAT proteins are expressed at equivalent levels in both Th1 and Th2 cells, and multiple Th1 and Th2 cytokine genes (including the IL-2 and IL-4 genes) contain functionally important NFAT-binding sites, it seemed unlikely that NFAT proteins played important roles in regulating the balance of Th1/Th2 responses. Nevertheless, analyses of the NFATp- and NFATc-deficient mice have suggested an interesting reciprocal relationship for these two proteins in the regulation of Th2 responses. The NFATp-deficient mice demonstrated markedly dysregulated production of IL-4 after activation in vitro and in vivo (175, 209). Early IL-4 production was deficient in these mice, demonstrating an essential role for NFATp in the initial induction of IL-4 transcription after T cell activation (175). In contrast, whereas the wild-type T cells rapidly downregulated IL-4 production, NFATp-deficient T cells displayed prolonged and elevated IL-4 production (175, 209). This prolonged expression of IL-4 correlated with evidence of increased Th2 and decreased Th1 differentiation both in vitro and in vivo (175, 209). Thus, NFATp appears to play a negative regulatory role in the generation of Th2 responses in vivo. In contrast to the results obtained in the NFATp−/− mice, NFATc-deficient T cells demonstrated severe reductions in IL-4 and IL-6 production and decreased development of Th2 cells after T cell activation (179, 180). Furthermore, there was a selective loss of IL-4-driven immunoglobulin isotypes in the NFATc−/−RAG2−/− chimeric mice and decreased production of Th2 cytokines in vitro after T cell stimulation (179, 180). Thus, although both NFATp and NFATc can bind to and transactivate the IL-4 promoter in transient transfection assays, these proteins play distinct roles in regulating IL-4 production and Th2 differentiation in normal T cells in vivo. Both proteins appear to be important for the early induction of the IL-4 gene after T cell activation; however, NFATp alone is necessary to down-regulate IL-4 gene expression during later phases of the immune response. How can NFATp (but not NFATc) serve as both an activator and repressor of IL-4 transcription? Possible mechanisms include specific posttranslational modifications of the protein during the later phases of the immune response or its ability to partner with coactivators and repressors whose expression are regulated temporally after T cell activation.

c-Maf c-Maf is a 42-kDa transcription factor that belongs to a superfamily of b-ZIP proteins that also includes the AP1 and CREB/ATF families (210). c-Maf is the cellular homologue of the v-maf oncogene, the transforming gene of the avian retrovirus AS42, which was isolated initially from a spontaneous musculoaponeurotic fibrosarcoma in the chicken (211). Like v-maf, the c-maf

P1: SAT/MBG/SPD

January 21, 1999

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

174

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

protooncogene can cause cellular transformation when overexpressed in chicken embryo fibroblasts (210). The Maf protein family can be subdivided into a subfamily of proteins containing N-terminal transactivation domains, such as c-Maf, MafB, and Nrl, and a second subfamily that lacks this domain, including MafK, MafF, MafG, and p18 (212–216). Like other b-ZIP family members, Maf proteins can form homo- and heterodimers with one another and with Fos and Jun (217, 218). c-Maf homodimers bind to distinct palindromic sequences, TGCTGACTCAGCA and TGCTGACGTCAGCA, with equal affinity (217). The fact that AP1-binding sites are present in both c-Maf consensus sequence motifs suggests that c-Maf and AP1 transcription factors may regulate similar or overlapping sets of target genes. The first evidence for the involvement of c-Maf in Th2 differentiation came from an analysis of its pattern of expression in T cell subsets. c-Maf mRNA was expressed preferentially in Th2 T cell clones and was specifically upregulated as naive splenic T cells were differentiated down the Th2 pathway (219). Transient transfection experiments demonstrated that c-Maf can bind to and transactivate the IL-4 promoter in both Th1 clones and in the B cell lymphoma M12 (219). More impressively, cotransfection of c-Maf and NFATp expression vectors induced endogenous IL-4 gene expression in M12 B cells (219). Recently, a newly cloned NFAT-interacting protein (NIP45) was also shown to participate in the synergistic activation of IL-4 expression (220). Cotransfection of c-Maf, NFATp, and NIP45 expression vectors resulted in a greater than 200-fold induction of an IL-4 promoter reporter construct in HepG2 cells and a similar induction of the endogenous IL-4 gene in M12 B cells (220). The mechanism by which NIP45 synergizes with c-Maf and NFAT is unknown as the protein lacks detectable DNA-binding activity and does not display structural similarities to other known proteins. These studies suggested a hierarchy of transcription factors that together control the differentiation of Th2 cells. In this model, c-Maf appears to play a central regulatory role because it is expressed preferentially in Th2 cells and in conjunction with NFAT can activate the endogenous IL-4 promoter in non–T cells. Expression of c-Maf, NFAT, and NIP45, perhaps along with GATA3, activates IL-4 expression, which, via binding to the IL-4 receptor, activates Stat6 to drive the differentiation program of Th2 cells. Despite the remarkable progress in understanding this pathway, a number of important questions remain unanswered. First, how are c-Maf and GATA3 induced during early Th2 differentiation? Second, what is the relationship between c-Maf and GATA3 gene expression? Third, which of these proteins is necessary for IL-4 expression and Th2 differentiation? In this regard, c-Maf knockouts and transgenics will be particularly informative. Fourth, do these transcription factors interact directly and do they

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

175

bind cooperatively to Th2-specific promoters? Finally, what are the equivalent transcription factors that drive Th1 differentiation?

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

SUMMARY AND FUTURE DIRECTIONS Recently, we have learned a great deal about the transcriptional programs that regulate T cell development and function. We have identified many of the nuclear proteins that regulate individual steps in thymocyte ontogeny, T cell quiescence and activation, and T cell effector functions. Genetic studies in mice have revealed unique and essential roles for some of these transcription factors in T cell development and function in vivo (see Table 1). Of equal importance, we have begun to understand how the activities of these T cell transcription factors are regulated in response to specific extracellular signals. From these studies, it has become clear that T cells, like most other cell types, have evolved complex highly regulated molecular pathways that allow the integration of multiple stimulatory and inhibitory signals via posttranslational modifications and combinatorial interactions of multiple transcription factors on their cognate promoters and enhancers. Despite these advances in our understanding of T cell transcription, much remains to be learned. We need to understand more about the relationships between the transcription factors that regulate early thymocyte development, and we need to place them into hierarchies or pathways similar to what has been accomplished in the elegant studies of the transcriptional regulation of early Drosophila development. For example, does Ikaros regulate the expression of GATA3, and if so, is this a direct or indirect effect of Ikaros? Similarly, what are the downstream targets of GATA3 in early T cell progenitors? We also need to know more about the transcription factors that regulate positive and negative selection of DP thymocytes and their subsequent differentiation into SP T cells. This will likely involve the identification of one or more previously unknown T cell transcription factors. Relatively little is known about the transcription factors that control Th1 differentiation or the establishment and maintenance of the memory phenotype in T cells. Of equal importance, we need to understand the combinatorial interactions between transcription factors and how these influence the patterns of T cell gene expression in response to different signals. For example, how is it that GATA3 functions both as a very early lineage-determining gene in DN thymocytes and as an important regulator of Th2 differentiation in mature SP T cells? Similarly, how do the different NFAT proteins positively and negatively regulate IL-4 gene expression and Th2 differentiation? The answers to these questions will provide us with exciting new insights into the function of the mammalian immune system. They will

Preferentially in T, B, and NK cells (54; Barton et al, submitted)

Endocardial cushions, gonads, T and B cells (81, 82)

HMG box WWCAAAG (81) Zinc finger CACCC core (93)

Sox4

LKLF

Vascular endothelial cells, lung, T and B cells (93, 105)

Limb buds, tooth buds, neural crest, lung, thymus urogenital ridge (76)

SP thymocytes, quiescent peripheral T cells (105)

Mature T cell lines (81)

Throughout T cell development (76, 77)

Throughout thymocyte development, highest in SP cells, downregulated upon T cell activation (60, 61)

Throughout thymocyte maturation, Th2 cells (32, 48, 206, 207)

Throughout T cell development (16, 17)

KO: embryonic lethal d12.5–14.5, intra-embryonic + intra-amniotic hemorrhage (106) RAG: spontaneously activated cell-surface phenotype, peripheral T cell apoptosis (105)

KO: embryonic lethal d14, circulatory failure due to defects in heart formation (82) Thymic organ culture: deficient transition of DN to DP thymocytes (83)

KO: TCF1: marked reduction in DP and SP thymocyte numbers (77); LEF1: neonatal lethal with defects in teeth, hair, whiskers, and mammary gland formation (79) KO/KO: complete block in ISP to DN thymocyte transition (80)

KO: severe defects in NK cell number and function; decreased thymocyte numbers; low-level expression of CD4 on CD8+ T cells and T cell activation defects listed below (222) RAG: decreased T cell proliferation after activation, increased T cell apoptosis (62, 63)

KO: embryonic lethal d11–12, severe internal bleeding, CNS developmental defects (49) RAG: normal B cells but complete lack of T cell generation (51) Transgene/cell line: induces Th2 cytokine gene expression in Th1 cells (206)

KO (null): early B cell developmental block, block in fetal thymocyte generation, adult thymcoyte maturation, and T cell proliferation defects (24) KO (dominant negative): complete lack of T, B, and NK cells and progenitors (21)

Phenotype of mutant mice

Annual Reviews

TCF1/ HMG box LEF1 WWCAAAG (72–74)

Ets GGAW core (54)

Ets1

Fetal yolk sac; T, B, and NK cells (16, 17)

T cell expression

13:7

CNS, kidney, thymus (47, 48)

Zinc finger GGGA core (18)

Ikaros

Expression pattern

P2: NBL/plb

GATA3 Zinc finger WGATAR (26, 27)

Family/motif

176

Factor

January 21, 1999

Table 1 Transcription factors in T cell developmenta

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

P1: SAT/MBG/SPD QC: NBL

AR078-06

KUO & LEIDEN

STAT TTCN4GAA (186, 187) b-ZIP CNS, connective, TGCTGACTCAGCA renal tissues (221) (217)

Stat4/6

c-Maf

Cell line: activates Th2-specific IL-4 gene expression in Th1 cells and non-T cells (219)

a KO, Knock-out mice; KO/KO, double knock-out mice containing targeted mutations of two related transcription factors; RAG, RAG-deficient chimeric mice produced by injection of RAG-deficient blastocysts with homozygous null ES cells; cell line, transfected cell line overexpressing a specific transcription factor; transgene, transgenic mice; thymic organ culture, thymic organ cultures from mice containing targeted deletions.

Upregulated during Th2 cell differentiation (219)

KO: Stat4: defects in IL-12-mediated immune responses and Th1 differentiation (193, 194); Stat6: defects in IL-4-mediated immune responses and Th2 differentiation (203–205)

NFATp, c, 4: throughout KO: NFATp: hyperproliferative T and B cells, enhanced Th2-mediated T cell development (158) immune responses (175, 176, 209); NFATc: embryonic lethal d14.5 from defects in cardiac valve and septum formation defects (177, 178) RAG: NFATc: defects in IL-4 and IL-6 production, Th2-mediated immune responses (179, 180)

Transgene (dominant negative): defects in T cell activation, AP-1 induction, IL-2 production, and block in cell cycle progression; increased apoptosis following activation (153)

Annual Reviews

Stat4: spleen, lung, muscle Not determined Stat6: widely expressed (189, 195–197)

widely expressed muscle, testis widely expressed thymus, muscle (158, 165)

Rel GGAAA core (158)

NFAT

Throughout T cell development (140)

KO: NF-κB1: B cell proliferation defects (124, 125); RelA: embryonic lethal d15, hepatocyte apoptosis (126, 127); RelB: dendritic and thymic medullary epithelium cell survival defects (129, 130); c-Rel: defects in T & B cell proliferation and IL-2 production (132) Transgene (dominant negative IκB-α): defects in CD8+ SP thymocyte maturation and peripheral CD8+ T cell numbers (133, 134; Hettmann et al, submitted)

P2: NBL/plb

NFATp: NFATc: NFAT3: NFAT4:

Widely expressed (140)

b-ZIP TGANNTCA (140)

CREB

Widely expressed during T cell development (119, 121)

13:7

Widely expressed (109–111)

January 21, 1999

NF-κB Rel GGGRNNYYCC (117)

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

P1: SAT/MBG/SPD QC: NBL

AR078-06

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

177

P1: SAT/MBG/SPD

January 21, 1999

178

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN

also present important new opportunities for therapeutic manipulations of T cell function in diseases such as cancer, allergies, and autoimmunity. ACKNOWLEDGMENTS We thank Ms P Lawrey for help with preparation of the manuscript. This work was supported in part by a grant from the NIAID to JML (AI29673-07).

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Weissman IL. 1994. Developmental switches in the immune system. Cell 76:207– 18 2. Weiss A, Littman DR. 1994. Signal transduction by lymphocyte antigen receptors. Cell 76:263–74 3. Janeway CA, Bottomly K. 1994. Signals and signs for lymphocyte responses. Cell 76:275–85 4. Clevers HC, Grosschedl R. 1996. Transcriptional control of lymphoid development: lessons from gene targeting. Immunol. Today 17:336–43 5. Fischer A, Malissen B. 1998. Natural and engineered disorders of lymphocyte development. Science 280:237–43 6. Clevers H, Ferrier P. 1998. Transcriptional control during T-cell development. Curr. Opin. Immunol. 10:166–71 7. Leiden JM. 1993. Transcriptional regulation of T cell receptor genes. Annu. Rev. Immunol. 11:539–70 8. Leiden JM, Thompson CB. 1994. Transcriptional regulation of T-cell genes during T-cell development. Curr. Opin. Immunol. 6:231–37 9. Georgopoulos K. 1997. Transcription factors required for lymphoid lineage commitment. Curr. Opin. Immunol. 9:222–27 10. Orkin SH. 1995. Hematopoiesis: How does it happen? Curr. Opin. Cell Biol. 7:870–77 11. Shortman K, Wu L. 1996. Early T lymphocyte progenitors. Annu. Rev. Immunol. 14:29–47 12. Kondo M, Weissman IL, Akashi K. 1997. Identification of clonogenic common lymphoid progenitors in mouse bone marrow. Cell 91:661–72 13. 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

14.

15. 16.

17.

18.

19.

20.

21.

22.

precursors of T lymphocytes and natural killer cells. Cell 69:139–50 Sanchez MJ, Muench MO, Roncarolo MG, Lanier LL, Phillips JH. 1994. Identification of a common T/natural killer cell progenitor in human fetal thymus. J Exp Med. 180:569–76 Shivdasani RA, Orkin SH. 1996. The transcriptional control of hematopoiesis. Blood 87:4025–39 Georgopoulos K, Moore DD, Derfler B. 1992. Ikaros, an early lymphoid-specific transcription factor and a putative mediator for T cell commitment. Science 258:808–12 Georgopoulos K, Winandy S, Avitahl N. 1997. The role of the Ikaros gene in lymphocyte development and homeostasis. Annu. Review. Immunol. 15:155–76 Molnar A, Georgopoulos K. 1994. The Ikaros gene encodes a family of functionally diverse zinc finger DNA-binding proteins. Mol. Cell. Biol. 14:8292–303 Hahm K, Ernst P, Lo K, Kim GS, Turck C, Smale ST. 1994. The lymphoid transcription factor LyF-1 is encoded by specific, alternative spliced mRNAs derived from the Ikaros gene. Mol. Cell. Biol. 14:7111– 23 Morgan B, Sun L, Avitahl N, Andrikopoulos K, Ikeda T, Gonzales E, Wu P, Neben S, Georgopoulos K. 1997. Aiolos, a lymphoid restricted transcription factor that interacts with Ikaros to regulate lymphocyte differentiation. EMBO J. 16:2004–13 Georgopoulos K, Bigby M, Wang JH, Molnar A, Wu P, Winandy S, Sharpe A. 1994. The Ikaros gene is required for the development of all lymphoid lineages. Cell 79:143–56 Sun L, Liu A, Georgopoulos K. 1996. Zinc finger-mediated protein interactions modulate Ikaros activity, a molecular

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

23.

24.

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

25.

26. 27. 28.

29. 30.

31.

32.

33.

34.

control of lymphocyte development. EMBO J. 15:5358–69 Winandy S, Wu P, Georgopoulos K. 1995. A dominant mutation in the Ikaros gene leads to rapid development of leukemia and lymphoma. Cell 83:289–99 Wang JH, Nichogiannopoulou A, Wu L, Sun L, Sharpe AH, Bigby M, Georgopoulos K. 1996. Selective defects in the development of the fetal and adult lymphoid system in mice with an Ikaros null mutation. Immunity 5:537–49 Pandolfi P, Roth M, Karis A, Leonard M, Dzierzak E, Grosveld F, Engel JD, Lindenbaum MH. 1995. Targeted disruption of the GATA-3 gene causes severe abnormalities in the nervous system and in the fetal liver haematopoiesis. Nat. Genet. 11:40–44 Ko LJ, Engel JD. 1993. DNA-binding specificities of GATA transcription factor family. Mol. Cell. Biol. 13:4011–22 Merika M, Orkin SH. 1993. DNA-binding specificity of GATA family transcription factors. Mol. Cell. Biol. 13:3999–4010 Tsai SF, Martin DI, Zon LI, D’Andrea AD, Wong GG, Orkin SH. 1989. Cloning of cDNA for the major DNA binding protein of the erythroid lineage through expression in mammalian cells. Nature 339:446–51 Evans T, Felsenfeld G. 1989. The erythroid-specific transcription factor eryf1: a new finger protein. Cell 58:877–85 Lee ME, Temizer DH, Clifford JA, Quertermous T. 1991. Cloning of the GATA-binding protein that regulates endothelin-1 gene expression in endothelial cells. J. Biol. Chem. 266:16188–92 Dorfman DM, Wilson DB, Bruns GAP, and Orkin SH. 1992. Human transcription factor GATA-2: evidence for regulation of preproendothelin-1 gene expression in endothelial cells. J. Biol. Chem. 267:1279–85 Ho IC, Vorhees PN, Marin BK, Oakley SF, Tsai SH, Orkin, Leiden JM. 1991. Human GATA-3: a lineage-restricted transcription factor that regulates the expression of the T cell receptor alpha-gene. EMBO J. 10:1187–92 Ko LJ, Yamamoto M, Leonard MW, George KM, Ting P, Engel JD. 1991. Murine and human T-lymphocyte GATA-3 factors mediate transcription through a cis-regulatory element within the human T-cell receptor delta gene enhancer. Mol. Cell. Biol. 11:2778–84 Arceci RJ, King AA, Simon MC, Orkin SH, Wilson DB. 1993. Mouse GATA-4: a retinoic acid-inducible GATA-binding

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

179

transcription factor expressed in endodermally derived tissues and heart. Mol. Cell. Biol. 13:2235–46 Kelley C, Blumberg H, Zon LI, Evans T. 1993. GATA-4 is a novel transcription factor expressed in endocardium of the developing heart. Development 118:817–27 Laverriere AC, MacNeill C, Mueller C, Poelmann RE, Burch JB, Evans T. 1994. GATA-4/5/6, a subfamily of three transcription factors transcribed in developing heart and gut. J. Biol. Chem. 269:23177– 84 Morrisey EE, Ip HS, Tang Z, Lu MM, Parmacek MS. 1997. GATA5: a transcriptional activator expressed in a novel temporally and spatially-restricted pattern during embryonic development. Dev. Biol. 183:21–36 Morrisey EE, Ip HS, Lu MM, Parmacek MS. 1996. GATA6: a zinc finger transcription factor that is expressed in multiple cell lineages derived from lateral mesoderm. Dev. Biol. 177:309–22 Pevny L, Simon MC, Robertson E, Klein WH, Tsai SF, Dagati V, Orkin SH, Constantini F. 1991. Erythroid differentiation in chimeric mice blocked by a targeted mutation in the gene for transcription factor GATA-1. Nature 349:257–60 Fujiwara Y, Browne CP, Cunniff K, Goff SC, Orkin SH. 1996. Arrested development of embryonic red cell precursors in mouse embryos lacking transcription factor GATA-1. Proc. Natl. Acad. Sci. USA 93:12355–58 Tsai FY, Keller G, Kuo FC, Weiss M, Chen J, Rosenblatt M, Alt FW, Orkin SH. 1994. An early haematopoietic defect in mice lacking the transcription factor GATA-2. Nature 371:221–26 Kuo CT, Morrisey EE, Anandappa R, Sigrist K, Lu MM, Parmacek MS, Soudais C, Leiden JM. 1997. GATA4 transcription factor is required for ventral morphogenesis and heart tube formation. Genes Dev. 11:1048–60 Molkentin JD, Lin Q, Duncan SA, Olson EN. 1997. Requirement of the transcription factor GATA4 for heart tube formation and ventral morphogenesis. Genes Dev. 11:1061–72 Henderson AJ, McDougall D, Leiden JM, Calame KL. 1994. GATA elements are necessary for the activity and tissue specificity of the T-cell beta chain transcriptional enhancer. Mol. Cell. Biol. 14:4286– 94 Joulin V, Bories D, Eleouet JF, Labastie MC, Chretien S, Mattei MG, Romeo PH. 1991. A T-cell-specific TCR delta DNA

P1: SAT/MBG/SPD

January 21, 1999

180

46.

47.

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN binding protein is a member of the human GATA family. EMBO J. 10:1809–16 Landry DB, Engel JD, Sen R. 1993. Functional GATA-3 binding sites within murine CD8 alpha upstream regulatory sequences. J. Exp. Med. 178:941–49 George KM, Leonard MW, Roth ME, Lieuw KH, Kioussis D, Grosveld F, Engel JD. 1994. Embryonic expression and cloning of the murine GATA-3. Genes Dev. 120:2673–86 Oosterwegel M, Timmerman J, Leiden J, Clevers H. 1992. Expression of GATA-3 during lymphocyte differentiation and mouse embryogenesis. Dev. Immunol. 3:1–11 Pandolfi PP, Roth ME, Karis A, Leonard MW, Dzierzak E, Grosveld FG, Engel JD, Lindenbaum MH. 1995. Targeted disruption of the GATA3 gene causes severe abnormalities in the nervous system and in fetal liver haematopoiesis. Nat. Genet. 11:40–44 Chen J, Landsford R, Steward V, Young F, Alt FW. 1993. RAG2-deficient blastocyst complementation: an assay of gene function in lymphocyte development. Proc. Natl. Acad. Sci. USA 90:4528–32 Ting CN, Olson MC, Barton KP, Leiden JM. 1996. Transcription factor GATA-3 is required for development of the T-cell lineage. Nature 384:474–78 Leprince D, Gegonne A, Coll J, de Taisne C, Schneeberger A, Lagrou C, Stehelin D. 1983. A putative second cell-derived oncogene of the avian leukaemia retrovirus E26. Nature 306:395–97 Nunn MF, Seeburg PH, Moscovici C, Duesberg PH. 1983. Tripartite structure of the avian erythroblastosis virus E26 transforming gene. Nature 306:391–95 Bassuk AG, Leiden JM. 1997. The role of Ets transcription factors in the development and function of the mammalian immune system. Adv. Immunol. 64:65–104 Lai ZC, Rubin GM. 1992. Negative control of photoreceptor development in Drosophila by the product of the yan gene, an ETS domain protein. Cell 70:609– 20 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 Su GH, Chen HM, Muthusamy N, Garrett-Sinha LA, Baunoch D, Tenen DG, Simon MC. 1997. Defective B cell receptor-mediated responses in mice lacking the Ets protein, Spi-B. EMBO J. 16:7118–29

58. Bassuk AG, Leiden JM. 1995. A direct physical association between ETS and AP-1 transcription factors in normal human T cells. Immunity 3:223–37 59. Pognonec P, Boulukos KE, Gesquiere JC, Stehelin D, Ghysdael J. 1988. Mitogenic stimulation of thymocytes results in the calcium-dependent phosphorylation of c-ets-1 proteins. EMBO J. 7:977– 83 60. Bhat NK, Komschlies KL, Fujiwara S, Fisher RJ, Mathieson BJ, Gregorio TA, Young HA, Kasik JW, Ozato K, Papas TS. 1989. Expression of ets genes in mouse thymocyte subsets and T cells. J. Immunol. 142:672–78 61. Rabault B, Ghysdael J. 1994. Calciuminduced phosphorylation of ETS1 inhibits its specific DNA binding activity. J. Biol. Chem. 269:28143–51 62. Muthusamy N, Barton K, Leiden JM. 1995. Defective activation and survival of T cells lacking the Ets-1 transcription factor. Nature 377:639–42 63. Bories JC, Willerford DM, Grevin D, Davidson L, Camus A, Martin P, Stehelin D, Alt FW. 1995. Increased T-cell apoptosis and terminal B-cell differentiation induced by inactivation of the Ets-1 protooncogene. Nature 377:635–38 64. Matsuyama T, Kimura T, Kitagawa M, Pfeffer K, Kawakami T, Watanabe N, Kundig TM, Amakawa R, Kishihara K, Wakeham A. 1993. Targeted disruption of IRF-1 or IRF-2 results in abnormal type I IFN gene induction and aberrant lymphocyte development. Cell 75:83–97 65. 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 66. Ogasawara K, Hida S, Azimi N, Tagaya Y, Sato T, Yokochi-Fukuda T, Waldmann TA, Taniguchi T, Taki S. 1998. Requirement for IRF-1 in the microenvironment supporting development of natural killer cells. Nature 391:700–3 67. Zuniga-Pflucker JC, Lenardo MJ. 1996. Regulation of thymocyte development from immature progenitors. Curr. Opin. Immunol. 8:215–24 68. von Boehmer H, Fehling HJ. 1997. Structure and function of the pre-T cell receptor. Annu. Rev. Immunol. 15:433–52 69. Robey E, Fowlkes BJ. 1994. Selective events in T cell development. Annu. Rev. Immunol. 12:675–705

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION 70. Jameson SC, Hogquist KA, Bevan MJ. 1995. Positive selection of thymocytes. Annu. Rev. Immunol. 13:93–126 71. Scollay R. 1991. T-cell subset relationships in thymocyte development. Curr. Opin. Immunol. 3:204–9 72. van de Wetering M, Oosterwegel M, Dooijes D, Clevers H. 1991. Identification and cloning of TCF-1, a T lymphocyte– specific transcription factor containing a sequence-specific HMG box. EMBO J. 10:123–32 73. Waterman ML, Fischer WH, Jones KA. 1991. A thymus-specific member of the HMG protein family regulates the human T cell receptor C alpha enhancer. Genes Dev. 5:656–69 74. Travis A, Amsterdam A, Belanger C, Grosschedl R. 1991. LEF-1, a gene encoding a lymphoid-specific protein with an HMG domain, regulates T-cell receptor alpha enhancer function. Genes Dev. 5:880–94 75. Oosterwegel M, van de Wetering M, Dooijes D, Klomp L, Winoto A, Georgopoulos K, Meijlink F, Clevers H. 1991. Cloning of murine TCF-1, a T cell– specific transcription factor interacting with functional motifs in the CD3-epsilon and T cell receptor alpha enhancers. J. Exp. Med. 173:1133–42 76. Oosterwegel M, van de Wetering M, Timmerman J, Kruisbeek A, Destree O, Meijlink F, Clevers H. 1993. Differential expression of the HMG box factors TCF-1 and LEF-1 during murine embryogenesis. Development 118:439–48 77. Verbeek S, Izon D, Hofhuis F, RobanusMaandag E, te Riele H, van de Wetering M, Oosterwegel M, Wilson A, MacDonald HR, Clevers H. 1995. An HMG-boxcontaining T-cell factor required for thymocyte differentiation. Nature 374:70–74 78. Love JJ, Li X, Case DA, Giese K, Grosschedl R, Wright PE. 1995. Structural basis for DNA bending by the architectural transcription factor LEF-1. Nature 376:791–95 79. van Genderen C, Okamura RM, Farinas I, Quo RG, Parslow TG, Bruhn L, Grosschedl R. 1994. Development of several organs that require inductive epithelialmesenchymal interactions is impaired in LEF-1-deficient mice. Genes Dev. 8:2691–703 80. 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

181

81. 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 82. Schilham MW, Oosterwegel MA, Moerer P, Ya J, de Boer P, van de Wetering M, Verbeek S, Lamers WH, Kruisbeek AM, Cumano A, Clevers H. 1996. Defects in cardiac outflow tract formation and proB-lymphocyte expansion in mice lacking Sox-4. Nature 380:711–14 83. Schilham MW, Moerer P, Cumano A, Clevers HC. 1997. Sox-4 facilitates thymocyte differentiation. Eur. J. Immunol. 27:1292–95 84. Crabtree GR. 1989. Contingent genetic regulatory events in T lymphocyte activation. Science 243:355–61 85. Chan AC, Desai DM, Weiss A. 1994. The role of protein tyrosine kinases and protein tyrosine phosphatases in T cell antigen receptor signal transduction. Annu. Rev. Immunol. 12:555–92 86. Cantrell D. 1996. T cell antigen receptor signal transduction pathways. Annu. Rev. Immunol. 14:259–74 87. Alberola-Ila J, Takaki S, Kerner JD, Perlmutter RM. 1997. Differential signaling by lymphocyte antigen receptors. Annu. Rev. Immunol. 15:125–54 88. Russell JH. 1995. Activation-induced death of mature T cells in the regulation of immune responses. Curr. Opin. Immunol. 7:382–88 89. Abbas AK. 1996. Die and let live: eliminating dangerous lymphocytes. Cell 84:655–57 90. Dutton RW, Bradley LM, Swain SL. 1998. T cell memory. Annu. Rev. Immunol. 16: 201–23 91. Tanchot C, Lemonnier FA, Perarnam B, Freitas AA, Rocha B. 1997. Differential requirements for survival and proliferation of CD8 naive or memory T cells. Science 276:2057–62 92. Miller IJ, Bieker JJ. 1993. A novel, erythroid cell-specific murine transcription factor that binds to the CACCC element and is related to the Kruppel family of nuclear proteins. Mol. Cell. Biol. 13:2776– 86 93. Anderson KP, Kern CB, Crable SC, Lingrel JB. 1995. Isolation of a gene encoding a functional zinc finger protein homologous to erythroid Kruppel-like factor: identification of a new multigene family. Mol. Cell. Biol. 15:5957–65 94. Crossley M, Whitelaw W, Perkins A, Williams G, Fujiwara Y, Orkin SH. 1996. Isolation and characterization of the

P1: SAT/MBG/SPD

January 21, 1999

182

95.

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

96.

97.

98.

99.

100.

101.

102.

103.

104.

105.

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN cDNA encoding BKLF/TEF-2, a major CACCC-box-binding protein in erythroid cells and selected other cells. Mol. Cell. Biol. 16:1695–705 Shields JM, Christy RJ, Yang VW. 1996. Identification and characterization of a gene encoding a gut-enriched Kruppellike factor expressed during growth arrest. J. Biol. Chem. 271:20009–17 Yet SF, McA’Nulty MM, Folta SC, Yen HW, Yoshizumi M, Hsieh CM, Layne MD, Chin MT, Wang H, Perrella MA, Jain MK, Lee ME. 1998. Human EZF, a Kruppel-like zinc finger protein, is expressed in vascular endothelial cells and contains transcriptional activation and repression domains. J. Biol. Chem. 273:1026–31 Preiss A, Rosenberg UB, Kienlin A, Seifert E, Jackle H. 1985. Molecular genetics of Kruppel, a gene required for segmentation of the Drosophila embryo. Nature 313:27–32 Rosenberg UB, Schroder C, Preiss A, Kienlin A, Cote S, Riede I, Jackle H. 1986. Structural homology of the product of the Drosophila Kruppel gene with Xenopus transcription factor IIIA. Nature 319:336–39 Nardelli J, Gibson TJ, Vesque C, Charnay P. 1991. Base sequence discrimination by zinc-finger DNA-binding domains. Nature 349:175–78 Sogawa K, Imataka H, Yamasaki Y, Kusume H, Abe H, Fujii-Kuriyama Y. 1993. cDNA cloning and transcriptional properties of a novel GC box-binding protein, BTEB2. Nucleic Acids Res. 21: 1527–32 Gessler M, Poustka A, Cavenee W, Neve RL, Orkin SH, Bruns GA. 1990. Homozygous deletion in Wilms tumours of a zinc-finger gene identified by chromosome jumping. Nature 343:774– 78 Bieker JJ, Southwood CM. 1995. The erythroid Kruppel-like factor transactivation domain is a critical component for cellspecific inducibility of a β-globin promoter. Mol. Cell. Biol. 15:852–60 Nuez B, Michalovich D, Bygrave A, Ploemacher R, Grosveld RF. 1995. Defective haematopoiesis in fetal liver resulting from inactivation of the EKLF gene. Nature 375:316–18 Perkins AC, Sharpe AH, Orkin SH. 1995. Lethal β-thalassaemia in mice lacking the erythroid CACCC-transcription factor EKLF. Nature 375:318–22 Kuo CT, Veselits ML, Leiden JM. 1997. LKLF: a transcriptional regulator of

106.

107. 108.

109. 110. 111.

112.

113.

114.

115. 116.

117.

118.

119.

single-positive T cell quiescence and survival. Science 277:1986–90 Kuo CT, Veselits ML, Barton KP, Lu MM, Clendenin C, Leiden JM. 1997. The LKLF transcription factor is required for normal tunica media formation and blood vessel stabilization during murine embryogenesis. Genes Dev. 11:2996–3006 Chen JH. 1985. The proto-oncogene c-ets is preferentially expressed in lymphoid cells. Mol. Cell. Biol. 5:2993–3000 Leung S, McCracken S, Ghysdael J, Miyamoto NG. 1993. Requirement of an ETS-binding element for transcription of the human lck type 1 promoter. Oncogene 8:989–97 Baeuerle PA, Henkel T. 1994. Function and activation of NF-κB in the immune system. Annu. Rev. Immunol. 12:141–79 Baldwin AS Jr. 1996. The NF-κB and IκB proteins: new discoveries and insights. Annu. Rev. Immunol. 14:649–83 Ghosh S, May MJ, Kopp EB. 1998. NFκB and Rel proteins: evolutionary conserved mediators of immune responses. Annu. Rev. Immunol 16:225–60 Baeuerle PA, Baltimore D. 1989. A 65kD subunit of active NF-κB is required for inhibition of NF-κB by IκB. Genes Dev. 3:1689–98 Siebenlist U, Franzoso G, Brown K. 1994. Structure, regulation and function of NF-κB. Annu. Rev. Cell Biol. 10:405– 55 Verma IM, Stevenson JK, Schwarz EM, Van Antwerp D, Miyamoto S. 1995. Rel/NF-κB/IκB family: intimate tales of association and dissociation. Genes Dev. 9:2723–35 Thanos D, Maniatis T. 1995. NF-κB: a lesson in family values. Cell 80:529–32 Thompson JE, Phillips RJ, ErdjumentBromage H, Tempst P, Ghosh S. 1995. IκB-beta regulates the persistent response in a biphasic activation of NF-κB. Cell 80:573–82 Kunsch C, Ruben SM, Rosen CA. 1992. Selection of optimal kappa B/Rel DNAbinding motifs: interaction of both subunits of NF-κB with DNA is required for transcriptional activation. Mol. Cell. Biol. 12:4412–21 Beg AA, Finco TS, Nantermet PV, Baldwin AS Jr. 1993. Tumor necrosis factor and interleukin-1 lead to phosphorylation and loss of IκB alpha: a mechanism for NF-κB activation. Mol. Cell. Biol. 13:3301–10 Sen J, Venkataraman L, Shinkai Y, Pierce JW, Alt FW, Burakoff SJ, Sen R. 1995. Expression and induction of nuclear

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

120. 121.

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

122.

123.

124.

125.

126.

127.

128.

129.

130.

131.

factor-κB-related proteins in thymocytes. J. Immunol. 154:3213–21 Kopp EB, Ghosh S. 1995. NF-κB and rel proteins in innate immunity. Adv. Immunol. 58:1–27 Feuillard J, Dargemont C, Ferreira V, Tarantino N, Debre P, Raphael M, Korner M. 1997. Nuclear Rel-A and c-Rel protein complexes are differentially distributed within human thymocytes. J. Immunol. 158:2585–91 Brown K, Park S, Kanno T, Franzoso G, Siebenlist U. 1993. Mutual regulation of the transcriptional activator NF-κB and its inhibitor, IκB-alpha. Proc. Natl. Acad. Sci. USA 90:2532–36 Henkel T, Machleidt T, Alkalay I, Kronke M, Ben-Neriah Y, Baeuerle PA. 1993. Rapid proteolysis of IκB-alpha is necessary for activation of transcription factor NF-κB. Nature 365:182–85 Sha WC, Liou HC, Tuomanen EI, Baltimore D. 1995. Targeted disruption of the p50 subunit of NF-κB leads to multifocal defects in immune responses. Cell 80:321–30 Snapper CM, Zelazowski P, Rosas FR, Kehry MR, Tian M, Baltimore D, Sha WC. 1996. B cells from p50/NF-κB knockout mice have selective defects in proliferation, differentiation, germ-line CH transcription, and Ig class switching. J. Immunol. 156:183–91 Beg AA, Sha WC, Bronson RT, Ghosh S, Baltimore D. 1995. Embryonic lethality and liver degeneration in mice lacking the RelA component of NF-κB. Nature 376:167–70 Doi TS, Takahashi T, Taguchi O, Azuma T, Obata Y. 1997. NF-κB RelA-deficient lymphocytes: normal development of T cells and B cells, impaired production of IgA and IgG1 and reduced proliferative responses. J. Exp. Med. 185:953–61 Horwitz BH, Scott ML, Cherry SR, Bronson RT, Baltimore D. 1997. Failure of lymphopoiesis after adoptive transfer of NF-κB-deficient fetal liver cells. Immunity 6:765–72 Burkly L, Hession C, Ogata L, Reilly C, Marconi LA, Olson D, Tizard R, Cate R, Lo D. 1995. Expression of relB is required for the development of thymic medulla and dendritic cells. Nature 373:531–36 Weih F, Carrasco D, Durham SK, Barton DS, Rizzo CA, Ryseck RP, Lira SA, Bravo R. 1995. Multiorgan inflammation and hematopoietic abnormalities in mice with a targeted disruption of RelB, a member of the NF-κB/Rel family. Cell 80:331–40 Weih F, Durham SK, Barton DS, Sha WC,

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

142.

183

Baltimore D, Bravo R. 1997. p50-NFκB complexes partially compensate for the absence of RelB: severely increased pathology in p50 (−/−) relB(−/−) doubleknockout mice. J. Exp. Med. 185:1359–70 Kontgen F, Grumont RJ, Strasser A, Metcalf D, Li R, Tarlinton D, Gerondakis S. 1995. Mice lacking the c-rel proto-oncogene exhibit defects in lymphocyte proliferation, humoral immunity, and interleukin-2 expression. Genes Dev. 9:1965–77 Boothby MR, Mora AL, Scherer DC, Brockman JA, Ballard DW. 1997. Perturbation of the T lymphocyte lineage in transgenic mice expressing a constitutive repressor of nuclear factor (NF)-κB. J. Exp. Med. 185:1897–907 Esslinger CW, Wilson A, Sordat B, Beermann F, Jongeneel CV. 1997. Abnormal T lymphocyte development induced by targeted overexpression of IκB alpha. J. Immunol. 158:5075–78 Traenckner EB, Pahl HL, Henkel T, Schmidt KN, Wilk S, Baeuerle PA. 1995. Phosphorylation of human IκB-alpha on serines 32 and 36 controls IκB-alpha proteolysis and NF-κB activation in response to diverse stimuli. EMBO J. 14:2876– 83 DiDonato J, Mercurio F, Rosette C, WuLi J, Suyang H, Ghosh S, Karin M. 1996. Mapping of the inducible IκB phosphorylation sites that signal its ubiquitination and degradation. Mol. Cell. Biol. 16: 1295–304 Brown K, Gerstberger S, Carlson L, Franzoso G, Siebenlist U. 1995. Control of IκB-alpha proteolysis by site-specific, signal-induced phosphorylation. Science 267:1485–88 Hoeffler JP, Meyer TE, Yun Y, Jameson JL, Habener JF. 1988. Cyclic AMPresponsive DNA-binding protein: structure based on a cloned placental cDNA. Science 242:1430–33 Gonzalez GA, Yamamoto KK, Fischer WH, Karr D, Menzel P, Biggs W, Vale WW, Montminy MR. 1989. A cluster of phosphorylation sites on the cyclic AMPregulated nuclear factor CREB predicted by its sequence. Nature 337:749–52 Habener JF. 1990. Cyclic AMP response element binding proteins: a cornucopia of transcription factors. Mol. Endocrinol. 4:1087–94 Gonzalez GA, Montminy MR. 1989. Cyclic AMP stimulates somatostatin gene transcription by phosphorylation of CREB at serine 133. Cell 59:675–80 Brindle P, Linke S, Montminy M.

P1: SAT/MBG/SPD

January 21, 1999

184

143.

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

144.

145.

146.

147.

148.

149.

150.

151.

152.

153.

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN 1993. Protein-kinase-A-dependent activator in transcription factor CREB reveals new role for CREM repressors. Nature 364:821–24 Chrivia JC, Kwok RP, Lamb N, Hagiwara M, Montminy MR, Goodman RH. 1993. Phosphorylated CREB binds specifically to the nuclear protein CBP. Nature 365: 855–59 Kwok RP, Lundblad Jr, Chrivia JC, Richards JP, Bachinger HP, Brennan RG, Roberts SG, Green MR, Goodman RH. 1994. Nuclear protein CBP is a coactivator for the transcription factor CREB. Nature 370:223–26 Ginty DD, Bonni A, Greenberg ME. 1994. Nerve growth factor activates a Rasdependent protein kinase that stimulates c-fos transcription via phosphorylation of CREB. Cell 77:713–25 Enslen H, Tokumitsu H, Soderling TR. 1995. Phosphorylation of CREB by CaMkinase IV activated by CaM-kinase IV kinase. Biochem. Biophys. Res. Commun. 207:1038–43 Brindle P, Nakajima T, Montminy M. 1995. Multiple protein kinase A-regulated events are required for transcriptional induction by cAMP. Proc. Natl. Acad. Sci. USA 92:10521–25 Muthusamy N, Leiden JM. 1998. A protein kinase C-, ras-, and RSK2-dependent signal transduction pathway activates the cAMP-responsive element-binding protein transcription factor following T cell receptor engagement. J. Biol. Chem. 273: 22,841–47 Mayall TP, Sheridan PL, Montminy MR, Jones KA. 1997. Distinct roles for PCREB and LEF-1 in TCR alpha enhancer assembly and activation on chromatin templates in vitro. Genes Dev. 11:887– 99 Anderson SJ, Miyake S, Loh DY. 1989. Transcription from a murine T-cell receptor V beta promoter depends on a conserved decamer motif similar to the cyclic AMP response element. Mol. Cell. Biol. 9:4835–45 Gao MH, Kavathas PB. 1993. Functional importance of the cyclic AMP response element–like decamer motif in the CD8 alpha promoter. J. Immunol. 150:4376– 85 Gupta A, Terhorst C. 1994. CD3 delta enhancer. CREB interferes with the function of a murine CD3-delta A binding factor (M delta AF). J. Immunol. 152:3895–903 Barton K, Muthusamy N, Chanyangam M, Fischer C, Clendenin C, Leiden JM. 1996. Defective thymocyte proliferation

154.

155.

156.

157.

158.

159.

160.

161.

162.

163.

164.

and IL-2 production in transgenic mice expressing a dominant-negative form of CREB. Nature 379:81–85 Berkowitz LA, Riabowol KT, Gilman MZ. 1989. Multiple sequence elements of a single functional class are required for cyclic AMP responsiveness of the mouse c-fos promoter. Mol. Cell. Biol. 9:4272– 81 Sheng M, Dougan ST, McFadden G, Greenberg ME. 1988. Calcium and growth factor pathways of c-fos transcriptional activation require distinct upstream regulatory sequences. Mol. Cell. Biol. 8:2787–96 Ofir R, Dwarki VJ, Rashid D, Verma IM. 1991. CREB represses transcription of fos promoter: role of phosphorylation. Gene Expression 1:55–60 Shaw JP, Utz PJ, Durand DB, Toole JJ, Emmel EA, Crabtree GR. 1988. Identification of a putative regulator of early T cell activation genes. Science 241:202–5 Rao A, Luo C, Hogan PG. 1997. Transcription factors of the NFAT family: regulation and function. Annu. Rev. Immunol. 15:707–47 Flanagan WM, Corthesy B, Bram RJ, Crabtree GR. 1991. Nuclear association of a T-cell transcription factor blocked by FK-506 and cyclosporin A. Nature 352:803–7 Northrop JP, Ullman KS, Crabtree GR. 1993. Characterization of the nuclear and cytoplasmic components of the lymphoid-specific nuclear factor of activated T cells (NF-AT) complex. J. Biol. Chem. 268:2917–23 Jain J, McCaffrey PG, Miner Z, Kerppola TK, Lambert JN, Verdine GL, Curran T, Rao A. 1993. The T-cell transcription factor NFATp is a substrate for calcineurin and interacts with Fos and Jun. Nature 365:352–55 McCaffrey PG, Luo C, Kerppola TK, Jain J, Badalian TM, Ho AM, Burgeon E, Lane WS, Lambert JN, Curran T. 1993. Isolation of the cyclosporin-sensitive T cell transcription factor NFATp. Science 262:750–54 Northrop JP, Ho SN, Chen L, Thomas DJ, Timmerman LA, Nolan GP, Admon A, Crabtree GR. 1994. NF-AT components define a family of transcription factors targeted in T-cell activation. Nature 369:497–502 Jain J, Burgeon E, Badalian TM, Hogan PG, Rao A. 1995. A similar DNA-binding motif in NFAT family proteins and the Rel homology region. J. Biol. Chem. 270: 4138–45

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T LYMPHOCYTE DEVELOPMENT AND FUNCTION 165. Hoey T, Sun YL, Williamson K, Xu X. 1995. Isolation of two new members of the NF-AT gene family and functional characterization of the NF-AT proteins. Immunity 2:461–72 166. Rooney JW, Sun YL, Glimcher LH, Hoey T. 1995. Novel NFAT sites that mediate activation of the interleukin-2 promoter in response to T-cell receptor stimulation. Mol. Cell. Biol. 15:6299–310 167. Randak C, Brabletz T, Hergenrother M, Sobotta I, Serfling E. 1990. Cyclosporin A suppresses the expression of the interleukin 2 gene by inhibiting the binding of lymphocyte-specific factors to the IL-2 enhancer. EMBO J. 9:2529–36 168. Ruff VA, Leach KL. 1995. Direct demonstration of NFATp dephosphorylation and nuclear localization in activated HT-2 cells using a specific NFATp polyclonal antibody. J. Biol. Chem. 270:22602–7 169. Crabtree GR, Clipstone NA. 1994. Signal transmission between the plasma membrane and nucleus of T lymphocytes. Annu. Rev. Biochem. 63:1045–83 170. O’Keefe SJ, Tamura J, Kincaid RL, Tocci MJ, O’Neill EA. 1992. FK506- and CsA-sensitive activation of the interleukin-2 promoter by calcineurin. Nature 357:692–94 171. Park J, Yaseen NR, Hogan PG, Rao A, Sharma S. 1995. Phosphorylation of the transcription factor NFATp inhibits its DNA binding activity in cyclosporin Atreated human B and T cells. J. Biol. Chem. 270:20653–59 172. Shaw KT, Ho AM, Raghavan A, Kim J, Jain J, Park J, Sharma S, Rao A, Hogan PG. 1995. Immunosuppressive drugs prevent a rapid dephosphorylation of transcription factor NFAT1 in stimulated immune cells. Proc. Natl. Adad. Sci. USA 92:11205–9 173. Masuda ES, Naito Y, Tokumitsu H, Campbell D, Saito F, Hannum C, Arai K, Arai N. 1995. NFATx, a novel member of the nuclear factor of activated T cells family that is expressed predominantly in the thymus. Mol. Cell. Biol. 15:2697– 706 174. Ho SN, Thomas DJ, Timmerman LA, Li X, Francke U, Crabtree GR. 1995. NFATc3, a lymphoid-specific NFATc family member that is calcium-regulated and exhibits distinct DNA binding specificity. J. Biol. Chem. 270:19898–907 175. Hodge MR, Ranger AM, Charles de la Brousse F, Hoey T, Grusby MJ, Glimcher LH. 1996. Hyperproliferation and dysregulation of IL-4 expression in NF-ATpdeficient mice. Immunity 4:397–405

185

176. Xanthoudakis S, Viola JP, Shaw KT, Luo C, Wallace JD, Bozza PT, Luk DC, Curran T, Rao A. 1996. An enhanced immune response in mice lacking the transcription factor NFAT1. Science 272:892–95 177. de la Pompa JL, Timmerman LA, Takimoto H, Yoshida H, Elia AJ, Samper E, Potter J, Wakeham A, Marengere L, Langille BL, Crabtree GR, Mak TW. 1998. Role of the NF-ATc transcription factor in morphogenesis of cardiac valves and septum. Nature 392:182–86 178. Ranger AM, Grusby MJ, Hodge MR, Gravallese EM, de la Brousse FC, Hoey T, Mickanin C, Baldwin HS, Glimcher LH. 1998. The transcription factor NF-ATc is essential for cardiac valve formation. Nature 392:186–90 179. Yoshida H, Nishina H, Takimoto H, Marengere LE, Wakeham AC, Bouchard D, Kong YY, Ohteki T, Shahinian A, Bachmann M, Ohashi PS, Penninger JM, Crabtree GR, Mak TW. 1998. The transcription factor NF-ATc1 regulates lymphocyte proliferation and Th2 cytokine production. Immunity 8:115–24 180. Ranger AM, Hodge MR, Gravallese EM, Oukka M, Davidson L, Alt FW, de la Brousse FC, Hoey T, Grusby M, Glimcher LH. 1998. Delayed lymphoid repopulation with defects in IL-4-driven responses produced by inactivation of NF-ATc. Immunity 8:125–34 181. Fitch FW, McKisic MD, Lancki DW, Gajewski TF. 1993. Differential regulation of murine T lymphocyte subsets. Annu. Rev. Immunol. 11:29–48 182. Abbas AK, Murphy KM, Sher A. 1996. Functional diversity of helper T lymphocytes. Nature 383:787–93 183. Seder RA, Paul WE. 1994. Acquisition of lymphokine-producing phenotype by CD4+ T cells. Annu Rev Immunol 12: 635–73 184. Murphy KM. 1998. T lymphocyte differentiation in the periphery. Curr. Opin. Immunol. 10:226–32 185. Constant SL, Bottomly K. 1997. Induction of Th1 and Th2 CD4+ T cell responses: the alternative approaches. Annu. Rev. Immunol. 15:297–322 186. Schindler C, Darnell JE Jr. 1995. Transcriptional responses to polypeptide ligands: the JAK-STAT pathway. Annu. Rev. Biochem. 64:621–51 187. Leonard WJ, O’Shea JJ. 1998. Jaks and STATs: biological implications. Annu. Rev. Immunol. 16:293–322 188. Darnell JE Jr, Kerr IM, Stark GR. 1994. Jak-STAT pathways and transcriptional activation in response to IFNs and other

P1: SAT/MBG/SPD

January 21, 1999

186

189.

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

190.

191.

192.

193.

194.

195.

196.

197.

198.

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

KUO & LEIDEN extracellular signaling proteins. Science 264:1415–21 Yamamoto K, Quelle FW, Thierfelder WE, Kreider BL, Gilbert DJ, Jenkins NA, Copeland NG, Silvennoinen O, Ihle JN. 1994. Stat4, a novel gamma interferon activation site-binding protein expressed in early myeloid differentiation. Mol. Cell. Biol. 14:4342–49 Zhong Z, Wen Z, Darnell JE Jr. 1994. Stat3 and Stat4: members of the family of signal transducers and activators of transcription. Proc. Natl. Acad. Sci. USA 91:4806–10 Bacon CM, Petricoin EF 3rd, Ortaldo Jr, Rees RC, Larner AC, Johnston JA, O’Shea JJ. 1995. Interleukin 12 induces tyrosine phosphorylation and activation of STAT4 in human lymphocytes. Proc. Natl. Acad. Sci. USA 92:7307–11 Jacobson NG, Szabo SJ, Weber-Nordt RM, Zhong Z, Schreiber RD, Darnell JE Jr, Murphy KM. 1995. Interleukin 12 signaling in T helper type 1 (Th1) cells involves tyrosine phosphorylation of signal transducer and activator of transcription (Stat)3 and Stat4. J. Exp. Med. 181:1755– 62 Thierfelder WE, van Deursen JM, Yamamoto K, Tripp RA, Sarawar SR, Carson RT, Sangster MY, Vignali DA, Doherty PC, Grosveld GC, Ihle JN. 1996. Requirement for Stat4 in interleukin-12mediated responses of natural killer and T cells. Nature 382:171–74 Kaplan MH, Sun YL, Hoey T, Grusby MJ. 1996. Impaired IL-12 responses and enhanced development of Th2 cells in Stat4deficient mice. Nature 382:174–77 Hou J, Schindler U, Henzel WJ, Ho TC, Brasseur M, McKnight SL. 1994. An interleukin-4-induced transcription factor: IL-4 Stat. Science 265:1701–6 Schindler C, Kashleva H, Pernis A, Pine R, Rothman P. 1994. STF-IL-4: a novel IL-4-induced signal transducing factor. EMBO J. 13:1350–56 Quelle FW, Shimoda K, Thierfelder W, Fischer C, Kim A, Ruben SM, Cleveland JL, Pierce JH, Keegan AD, Nelms K. 1995. Cloning of murine Stat6 and human Stat6, Stat proteins that are tyrosine phosphorylated in responses to IL-4 and IL-3 but are not required for mitogenesis. Mol. Cell. Biol. 15:3336–43 Chen XH, Patel BK, Wang LM, Frankel M, Ellmore N, Flavell RA, LaRochelle WJ, Pierce JH. 1997. Jak1 expression is required for mediating interleukin4-induced tyrosine phosphorylation of insulin receptor substrate and Stat6

199.

200.

201.

202.

203.

204.

205.

206.

207.

208.

209.

signaling molecules. J. Biol. Chem. 272: 6556–60 Witthuhn BA, Silvennoinen O, Miura O, Lai KS, Cwik C, Liu ET, Ihle JN. 1994. Involvement of the Jak-3 Janus kinase in signalling by interleukins 2 and 4 in lymphoid and myeloid cells. Nature 370:153– 57 Messner B, Stutz AM, Albrecht B, Peiritsch S, Woisetschlager M. 1997. Cooperation of binding sites for STAT6 and NF κB/rel in the IL-4-induced up-regulation of the human IgE germline. J. Immunol. 159:3330–37 Kotanides H, Reich NC. 1996. Interleukin-4-induced STAT6 recognizes and activates a target site in the promoter of the interleukin-4 receptor gene. J. Biol. Chem. 271:25555–61 Patel BKR, Pierce JH, LaRochelle WJ. 1998. Regulation of interleukin 4mediated signaling by naturally occurring dominant negative and attenuated forms of human Stat6. Proc. Natl. Acad. Sci. USA 95:172–77 Takeda K, Tanaka T, Shi W, Matsumoto M, Minami M, Kashiwamura S, Nakanishi K, Yoshida N, Kishimoto T, Akira S. 1996. Essential role of Stat6 in IL-4 signalling. Nature 380:627–30 Shimoda K, van Deursen J, Sangster MY, Sarawar SR, Carson RT, Tripp RA, Chu C, Quelle FW, Nosaka T, Vignali DA, Doherty PC, Grosveld G, Paul WE, Ihle JN. 1996. Lack of IL-4-induced Th2 response and IgE class switching in mice with disrupted Stat6 gene. Nature 380:630– 33 Kaplan MH, Schindler U, Smiley ST, Grusby MJ. 1996. Stat6 is required for mediating responses to IL-4 and for development of Th2 cells. Immunity 4:313– 19 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 Zhang DH, Cohn L, Ray P, Bottomly K, Ray A. 1997. Transcription factor GATA3 is differentially expressed in murine Th1 and Th2 cells and controls Th2-specific expression of the interleukin-5 gene. J. Biol. Chem. 272:21597–603 Lee HJ, O’Garra A, Arai K, Arai N. 1998. Characterization of cis-regulatory elements and nuclear factors conferring Th2-specific expression of the IL-5 gene: a role for a GATA-binding protein. J. Immunol. 160:2343–52 Kiani A, Viola JP, Lichtman AH, Rao A. 1997. Down-regulation of IL-4 gene tran-

P1: SAT/MBG/SPD

January 21, 1999

P2: NBL/plb

13:7

QC: NBL

Annual Reviews

AR078-06

T LYMPHOCYTE DEVELOPMENT AND FUNCTION

210.

211.

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

212.

213.

214.

215.

scription and control of Th2 cell differentiation by a mechanism involving NFAT1. Immunity 7:849–60 Kataoka K, Nishizawa M, Kawai S. 1993. Structure-function analysis of the maf oncogene product, a member of the b-Zip protein family. J. Virol. 67:2133–41 Nishizawa M, Kataoka K, Goto N, Fujiwara KT, Kawai S. 1989. v-maf, a viral oncogene that encodes a “leucine zipper” motif. Proc. Natl. Acad. Sci. USA 86:7711–15 Swaroop A, Xu JZ, Pawar H, Jackson A, Skolnick C, Agarwal NA. 1992. Conserved retina-specific gene encodes a basic motif/leucine zipper domain. Proc. Natl. Acad. Sci. USA 89:266–70 Fujiwara KT, Kataoka K, Nishizawa M. 1993. Two new members of the maf oncogene family, mafK and mafF, encode nuclear b-Zip proteins lacking putative trans-activator domain. Oncogene 8:2371–80 Igarashi K, Itoh K, Motohashi H, Hayashi N, Matuzaki Y, Nakauchi H, Nishizawa M, Yamamoto M. 1995. Activity and expression of murine small Maf family protein MafK. J. Biol. Chem. 270:7615– 24 Andrews NC, Kotkow KJ, Ney PA, Erdjument-Bromage H, Tempst P, Orkin SH. 1993. The ubiquitous subunit of erythroid transcription factor NF-E2 is a small basic-leucine zipper protein related

216.

217.

218.

219.

220.

221.

222.

187

to the v-maf oncogene. Proc. Natl. Acad. Sci. USA 90:11488–92 Kataoka K, Fujiwara KT, Noda M, Nishizawa M. 1994. MafB, a new Maf family transcription activator that can associate with Maf and Fos but not with Jun. Mol. Cell. Biol. 14:7581–91 Kataoka K, Noda M, Nishizawa M. 1994. Maf nuclear oncoprotein recognizes sequences related to an AP-1 site and forms heterodimers with both fos and jun. Mol. Cell. Biol. 14:700–12 Kerppola TK, Curran T. 1994. Maf and Nrl can bind to AP-1 sites and form heterodimers with Fos and Jun. Oncogene 9:675–84 Ho IC, Hodge MR, Rooney JW, Glimcher LH. 1996. The proto-oncogene c-maf is responsible for tissue-specific expression of interleukin-4. Cell 85:973–83 Hodge MR, Chun HJ, Rengarajan J, Alt A, Lieberson R, Glimcher LH. 1996. NFAT-Driven interleukin-4 transcription potentiated by NIP45. Science 274:1903–5 Nakayama H, Yamasaki H, Nishizawa M, Goto N. 1995. Tissue distribution of the DNA binding oncoprotein Maf during chicken development. Int. Dev. Biol. 39:957–64 Barton K, Muthusamy N, Fischer C, Ting C-N, Lanier L, Leiden JM. 1998. The Ets-1 transcription factor is required for the development of the natural killer cell lineage in mice. Immunity. In press.

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:149-187. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:189–220 c 1999 by Annual Reviews. All rights reserved Copyright °

NATURAL KILLER CELLS IN ANTIVIRAL DEFENSE: Function and Regulation by Innate Cytokines Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, and Thais P. Salazar-Mather Department of Molecular Microbiology and Immunology, Division of Biology and Medicine, Brown University, Providence, Rhode Island 02912; e-mail: Christine [email protected] KEY WORDS:

virus infections, interferons, interleukin 12, chemokines

ABSTRACT Natural killer (NK) cells are populations of lymphocytes that can be activated to mediate significant levels of cytotoxic activity and produce high levels of certain cytokines and chemokines. NK cells respond to and are important in defense against a number of different infectious agents. The first indications for this function came from the observations that virus-induced interferons α/β (IFN-α and -β) are potent inducers of NK cell-mediated cytotoxicity, and that NK cells are important contributors to innate defense against viral infections. In addition to IFN-α/β, a wide range of other innate cytokines can mediate biological functions regulating the NK cell responses of cytotoxicity, proliferation, and gamma interferon (IFN-γ ) production. Certain, but not all, viral infections induce interleukin 12 (IL-12) to elicit NK cell IFN-γ production and antiviral mechanisms. However, high levels of IFN-α/β appear to be unique and/or uniquely dominant in the context of viral infections and act to regulate other innate responses, including induction of NK cell proliferation in vivo and overall negative regulation of IL-12 production. A detailed picture is developing of particular innate cytokines activating NK cell responses and their consorted effects in providing unique endogenous milieus promoting downstream adaptive responses, most beneficial in defense against viral infections.

189 0732-0582/99/0410-0189$08.00

P1: SAT/ary

P2: NBL/plb

January 25, 1999

190

12:16

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

INTRODUCTION Natural killer (NK) cells are populations of lymphocytes that can contribute to protective responses against a variety of infections and cancers (1–5). Although they mediate certain functions overlapping with those delivered by classical T cells, they can be distinguished by their participation in innate immunity and early defense. NK cells were originally identified by their ability to spontaneously mediate lysis of certain susceptible tumor cell lines and their large granular lymphocyte morphology (1) and were characterized as non-T cells lacking expression of T cell antigen receptors and the CD3 complex (1, 2, 6–9). Their role in innate defense against infections was first indicated by identification of virus-induced type I interferon enhancement of NK cell–mediated cytotoxicity (10–12). A number of recent advances suggest more diverse responses and functions for NK-lineage cells. For example, it is now known that they (a) can produce subsets of soluble factors, including antimicrobial and immunoregulatory cytokines, under particular conditions of stimulation (1, 2, 13–22) and (b) have a related NK T cell population (23–25). Thus, in the context of early events after microbial challenges, innate cytokines, produced by cells of the innate immune system and/or nonimmune cells, have the potential to elicit particular and different responses mediated by NK non-T and/or NK T cell populations. Combinations of the innate responses have the potential to deliver immediate effects, as well as immunoregulatory functions promoting the most beneficial downstream adaptive responses, for defense against particular pathogens (5, 18, 26–28). Although it is now known that NK cells can respond to infections with a number of different classes of agents, the best evidence for their importance in defense is with viruses. The major functions for NK cells and the pathways regulating them during viral infections are reviewed here. Known and potential mechanisms are discussed. The focus is on unique or dominant characteristics potentially distinguishing infections with viruses from those with bacteria and parasites.

CELLS Although NK cells were first characterized based on biological functions and histological morphology, cell surface determinants have been identified. For obvious reasons, studies with human materials have primarily examined populations from peripheral blood. Studies with rodents have evaluated numerous compartments including bone marrow, spleen, and lymph nodes, and occasionally including blood, peritoneal exudates, liver, and lung. The studies in humans and mice identify the primary peripheral NK cell population in

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NK CELLS IN ANTIVIRAL DEFENSE

191

spontaneous states as being non-T lineage: dependent upon bone marrow but not on thymus or T and B cells for development, and present as mature cells with decreasing frequencies, respectively, in blood, spleen, and bone marrow (1). Experiments examining activated populations after systemic challenges with viruses, the type 1 interferons, i.e. alpha/beta interferons (IFN-α/β) or the chemical type 1 IFN inducers, i.e. tilerone or analogues of viral nucleic acid such as polyinosinic-polycytidylic acid (polyI:C), have identified responses of the blood, spleen, and bone marrow NK non-T cells as well as their accumulation in liver and lung (1, 2, 7–9, 29–33). Under the systemic induction conditions examined, this “classical” NK cell subset is not found at high frequencies in lymph nodes. However, the population can be activated and/or induced to accumulate at these sites after regional challenges with antigens and/or nonviral and viral microbial infections (19, 34–36). A population of NK T cells has been identified based on cell surface expression of certain classical NK and T cell determinants (23–25). These cells are present in thymus, bone marrow, and spleen. With the exception of thymus, they are generally at low frequencies in immune compartments. However, NK T cells constitute a high proportion of the leukocyte populations in unchallenged livers (25). Details concerning characterization of these cells as well as regulation pathways for them are given below.

Classical NK Cells Classical NK cells do not rearrange and express T cell antigen receptors (TCR), are CD3−, and develop despite deficiencies in machinery for rearranging antigen receptors and/or in T and B cell populations, i.e. SCID and RAG− (1, 2, 6–9). These cells can mediate perforin-dependent lysis, undergo proliferation, make the cytokines gamma interferon (IFN-γ ), tumor necrosis factor (TNF), and granulocyte/macrophage colony stimulation factor (GM-CSF) (1, 2), and produce the beta chemokines macrophage inflammatory protein 1α (MIP-1α), macrophage inflammatory protein 1β (MIP-1β), and the factor regulated on activation—normal T cell expressed and secreted (RANTES) (37, 38). NK cells from certain mouse strains express the marker NK1.1 (39), and those from a wider range of strains express DX5 (40). Human NK cells express CD56 (41). Asialo ganglio-N-tetraosylceramide (AGM1) is expressed at high levels on NK cells in all species (1, 2). However, none of these markers is exclusive to classical NK cells. Under certain culture conditions, NK cells can be driven to produce interleukin 5 (IL-5) (17). They are activated by IFN-α/β to mediate elevated lysis of sensitive cells and to lyse a broader range of cells including virus-infected cells (10–12). NK cell lytic activity is induced in response to early endogenously expressed IFN-α/β (see Table 1) (42–44). In addition to activating cytotoxicity, conditions of in vivo IFN-α/β exposure induce limited NK cell

P1: SAT/ary

P2: NBL/plb

January 25, 1999

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

192

QC: NBL

12:16

Annual Reviews

AR078-07

BIRON ET AL

blastogenesis and proliferation through interleukin 2 (IL-2)–independent pathways (2, 8, 45–49; LP Cousens, CA Biron, unpublished observations). Despite the in vivo effects, IFNs activate NK cell cytotoxicity but do not support NK cell growth in culture. Thus the NK cell proliferative response is dependent on IFNα/β–induced in vivo cytokines and/or conditions not yet identified. Although IL-2 can promote NK cell proliferation (1), the effect does not appear to be important during endogenous immunocompetent responses to viral infections because: (a) the factor is made subsequent to peak IFN-α/β and NK cell responses; and (b) in vivo activated T cells have a competitive advantage for IL-2 utilization as a result of being induced to express IL-2 receptors (R) (48–52) that are high affinity—in comparison to the NK cells’ intermediate affinity IL-2 receptors. However, endogenous IL-2 may contribute to NK cell proliferation in the absence of induced CD8 T cell responses under certain limited conditions (50). The cytokine interleukin 12 (IL-12) is a potent inducer of NK cell IFN-γ (13). This factor is generated in some, but not all, infections and is required for the induction of NK cell IFN-γ production under these conditions (32, 44, 53, 54). Classical NK cells also can be stimulated through their cell surface receptors for immunoglobulins, FcR, to mediate antibody-dependent cellular cytotoxicity (ADCC) and express IFN-γ (1), and NK1.1 can signal for IFN-γ production (15, 16). However, the role of these pathways in activating endogenous NK cell responses remains to be conclusively demonstrated in the host.

NK T Cells A population of T cells sharing characteristics with classical NK cells has been identified based on expression of NK cell markers, i.e. NK1.1 in mouse strains expressing the determinant (23–25, 55). These NK1+ T cells are found at high frequencies in unchallenged livers (25, 55). They express a limited T cell receptor (TCR) repertoire and are predominantly TCR α/β expressing Vα14/Jα281 in the mouse (56, 57). NK T cells depend on the nonclassical, major histocompatibility complex (MHC) class I-like CD1 molecules for their development (23, 55, 58–60). As a result, they are reduced in mice, especially young mice, lacking the β2 microglobulin molecule (β2M) used for expression of class I type MHC molecules (60, 61), and in CD1-deficient mice (62). Certain activated NK T cells can lyse cells sensitive to classical NK cell-mediated cytotoxicity (63). Their proliferation can be supported by IL-2, and they can be induced to release IL-4 by stimulation through the CD3 complex (15, 64, 65). Like classical NK cells, NK T cells can be induced to produce IFN-γ by stimulation through NK1.1 molecules or IL-12 exposure (15, 64). Much of the interest in NK T cells stems from their potential to make IL-4 during innate responses (64, 65) and their recognition of lipid structures, such as bacterial cell wall components, presented by CD1 molecules (66). These characteristics have the potential to be

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NK CELLS IN ANTIVIRAL DEFENSE

193

important in the context of nonviral infections but are unlikely to be critical during viral infections. Early expression of IL-4 has not yet been observed during challenges with viruses (67), and T cell antigens from these agents are likely to be predominantly or peptides. However, CD1 molecules can present peptides (68). Moreover, the NK T cell responsiveness to IL-12 and high frequencies in liver suggest that these cells could have a role in the context of infections with IL12-inducing viruses in this organ. Interestingly, a dramatic expansion of cells expressing the T cell marker Thy-1 and NK1.1 is induced in both spleen and lymph node during chronic retroviral infections in the model of murine acquired immunodeficiency syndrome (MAIDS) (69), but these populations have not been specifically identified as canonical NK T cells. Overall, little is known about endogenous NK T cell subset responses to, and functions during, viral infections.

ANTIVIRAL FUNCTIONS Classical NK cell responses and functions have been evaluated in the context of a wide range of viral infections (Table 1) (8, 11, 36, 38, 42–45, 47, 53, 54, 70– 106). Activation of elevated NK cell-mediated cytotoxicity has been observed during a number of different viral infections including those with representatives of the arenaviruses, e.g. lymphocytic choriomeningitis virus (LCMV) (8, 42), the herpesviruses, e.g. murine cytomegalovirus (MCMV) (43, 44, 53, 54) and herpes simplex virus (HSV) (77), the orthomyxoviruses, e.g. influenza virus (11), and the picornaviruses, e.g. Coxsackie virus (91). Infection-induced NK cell IFN-γ production responses have been demonstrated in a subset of these including MCMV (44, 53) and influenza virus (86). For the most part, peak NK cell responses of cytotoxicity and IFN-γ production occur within the first several hours to days after primary infections, whereas adaptive T and B cell responses take more than a week to develop. The importance of NK cells in early defense has been demonstrated by documentation of increased sensitivity to a number of mouse infections, including MCMV (53, 72, 80), HSV (76), influenza virus (85), and Coxsackie virus (91), as a result of NK cell depletions or deficiencies. Their presence in T cell-deficient mice has made it possible to conclusively assign cytotoxic and IFN-γ responses as well as antiviral functions to classical NK non-T cells in these systems. The contribution of NK cells to defense against human viral infections is supported by data from natural infections. Low NK cell cytotoxic activity is linked with increased human sensitivity to severe disseminating herpesgroup virus infections, including those with HSV (77, 78), Epstein-Barr virus (EBV) (83, 84), and human cytomegalovirus (HCMV) (78, 82); NK cell defects occur at late times after human immunodeficiency virus (HIV) infections (102, 103); and absence of NK cells has been associated with disease manifestations of

P1: SAT/ary

P2: NBL/plb

January 25, 1999

QC: NBL

12:16

194

Annual Reviews

AR078-07

BIRON ET AL

Table 1 Pr´ecis of published findings on natural killer cells and viral infections Virus Adenoviridae Adenoviruses

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Arenaviridae LCMV Pichinde virus Coronaviridae MHV

Herpesviridae HSV-1

VZV MCMV

HCMV

EBV

Orthomyxoviridae Influenza virus

Papovaviridae Papillomaviruses

Findings

Sensitivity of virus-transformed cells to NK cellmediated cytolysis in vitro correlates inversely with oncogenicity

70, 71

Induction of cytotoxicity but not IFN-γ production; viral titers do not increase upon depletion of NK cells Induction of cytotoxicity; depletion of NK cells increases replication of NK-susceptible virus strains

8, 42, 54, 72, 74, 75 73–75

Induction of cytotoxicity; contrasting reports on effects of 72, 75 NK cell depletion on susceptibility to infection that may be attributable to differences in strain variants Depletion of NK cells increases susceptibility; low cytotoxicity or NK cell deficiency in humans leads to disseminated infections NK cell deficiency in humans associates with disseminated infection Induction of both cytotoxicity and IFN-γ production; depletion or deficiency of NK cells increases susceptibility; resistance gene, Cmv-1, links with NK gene complex Low cytotoxicity associates with increased mortality following bone marrow transplantation; NK cell deficiency in humans leads to disseminated infection Humans with Chediak-Higashi Syndrome exhibit severe chronic infections; persistent defective cytotoxicity associates with chronic active infection; improvement correlates with increased cytotoxicity

76–78

78 36, 43, 44, 53, 54, 72, 74, 75, 79–81 78, 82

83, 84

Depletion of NK cells increases morbidity and mortality; induction of cytotoxicity and IFN-γ production

11, 85, 86

NK cell deficiency in humans correlates with recurrent cervical carcinoma in situ and condylomata

87

Paramyxoviridae Measles, mumps, Induction of cytotoxicity and Sendai viruses Picornaviridae Coxsackie virus

Reference(s)

Induction of cytotoxicity; depletion of NK cells increases viral replication and results in more severe myocarditis

88–90

91, 92

(Continued )

P1: SAT/ary

P2: NBL/plb

January 25, 1999

QC: NBL

12:16

Annual Reviews

AR078-07

NK CELLS IN ANTIVIRAL DEFENSE

195

Table 1 (Continued ) Virus EMCV

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

TMEV

Findings

Induction of cytotoxicity; correlation of resistance to 93–95 diabetogenic virus strain in female Swiss mice with induction of IFN-γ production; abrogation of resistance in C57BL/6J male mice with NK cell depletion Susceptibility of mouse strain correlates with lower NK 96 cell cytotoxicity; depletion of NK cells in resistant mouse strain increases severity of encephalitis

Poxviridae Vaccinia virus

Induction of cytotoxicity; depletion of NK cells increases viral replication Ectromelia virus Depletion of NK cells increases viral titers; resistance in different strains of mice associates with NK gene complex

Retroviridae HTLV-1 HIV

Rhabdoviridae VSV

Togaviridae Sindbis virus

Reference(s)

72, 97 98, 99

NK cells mediate cytotoxicity against virus-infected cells Immune-complex armed NK cells mediate lysis against CD4+ T cells; NK cells from infected patients produce antiviral CC chemokines in vitro; cytotoxicity, but not ADCC, decreases over the course of infection

100 38, 101–103

Induction of cytotoxicity; poor in vivo tumorigenicity of persistently infected cell lines correlates with susceptibility to NK cell cytotoxicity in vitro

97, 104

Induction of cytotoxicity; lack of correlation between NK cell cytotoxicity and resistance to infection

105, 106

papilloma viral infections in a female patient (87). The occurrence of chronic active EBV infections with low NK cell activity in both EBV-infected and EBVseronegative members of a family suggests that pre-existing NK cell deficiencies contribute to a lack of EBV control (84). Thus there are striking correlations between poor or no NK cell function and susceptibility to viruses. However, the most convincing data supporting a role for NK cells in defense against human viral infections come from a long-term longitudinal study of a female patient identified with a complete lack of NK phenotype cells as well as no spontaneous or IL-2-inducible NK cell cytotoxic function (78). This individual first presented as a 13-year-old with an overwhelming chicken pox infection and evidence of varicella pneumonia. After recovery, her NK cell deficiency was identified, other immune parameters were shown to be normal or near normal, and both antibody and memory T cell responses to varicella-zoster

P1: SAT/ary

P2: NBL/plb

January 25, 1999

196

QC: NBL

12:16

Annual Reviews

AR078-07

BIRON ET AL

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

virus (VZV) were demonstrated. She maintained her NK cell deficiency and developed primary life-threatening HCMV infection and a severe HSV infection during the next five-year period. Thus she presented with an abnormal sensitivity to the herpesgroup virus VZV, and she demonstrated abnormal sensitivities to two other herpesgroup viruses after identification of her NK cell deficiency. Taken together, the studies of mouse and human infections overwhelmingly indicate that NK cells promote defense against viruses. The most definitive documentation is for control of primary herpesvirus infections.

Cytotoxicity Although NK cells clearly mediate in vivo defense against viruses, mechanisms for their effects are not understood completely. Presumably, NK cell cytotoxicity could deliver antiviral activity by lysing virus-infected cells at times prior to replicating virion assembly and/or virus spreading through cell-to-cell contact. Such a defense mechanism would require direct contact between NK cells and virus-infected target cells, as well as the presence of positive (but absence of negative) signaling from the target to the effector cells for release of cytolytic molecules (see below). NK cell cytotoxicity is activated whenever the IFN-α/β cytokines are induced. Both of these occur at high and systemic levels during numerous viral infections (42–44, 107), and the endogenous NK cell cytotoxic response is dependent upon the IFN-α/β responses (43, 44). Moreover, as an apparent strategy to avoid recognition by T cells, certain viruses reduce expression of the MHC I molecules capable of stimulating negative receptors on NK cell surfaces (108–113). Nevertheless, a prominent role for NK cell-mediated lysis in antiviral defense has not been established. NK cells clearly are induced to mediate elevated cytotoxicity without contributing to viral resistance during LCMV infections (72). In the case of NK cell defense during MCMV infections, studies carried out with mice lacking the cytotoxic pathway (i.e. perforin-deficient), indicate that this mechanism is not responsible for the early NK cell-mediated effects in liver (44, 114) but may contribute to those in spleen (114). Thus, induction of NK cell cytotoxicity is common during many viral infections, but the function accounts for only a proportion, if any, of the antiviral effects mediated by NK cells. Given the reproducibility and magnitude of virus-elicited responses, NK cell cytotoxic activity is likely to contribute to some as-yet-undefined aspect of defense or immunoregulation.

Cytokine Production NK cells can also make cytokines with antiviral functions, including IFN-γ and TNF. In contrast to the limited evidence supporting a role for cytotoxicity, the importance of NK cell-produced IFN-γ in antiviral defense has been definitively established. This cytokine response is elicited at detectable protein levels during

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NK CELLS IN ANTIVIRAL DEFENSE

197

some but not all viral infections (44, 53, 54, 86). Induction is associated with a demonstrable role for NK cells in antiviral defense: NK cell IFN-γ production is observed during MCMV (44, 53, 54, 115) and influenza virus (86), but not LCMV, infections (54, 116). Those infections eliciting NK cell IFN-γ also induce production of biologically active IL-12 p70 heterodimer, and the NK cell response is dependent upon this endogenous IL-12 (44, 54, 86). Thus, NK cell IFN-γ production during viral infection is a consequence of virus-induced IL-12. During MCMV infection, the NK cell IFN-γ response is systemic, with serum levels reaching 10,000 pg/ml at peak times (115). NK cell-produced IFN-γ clearly contributes to protection in MCMV-infected livers (53). The effects may result from several mechanisms because the virus is sensitive to both IFN-γ –mediated inhibition of replication in infected cells (117) and pathways of antiviral defense dependent upon inducible nitric oxide synthase (iNOS) (114, 118) and/or delivered by IFN-γ -activated macrophages (119). Because wide ranges of viruses are known to be sensitive to these mechanisms (117– 121), such NK cell-activated pathways are likely to mediate protection during a number of other infections. Given the potential for systemic levels, NK cell– produced IFN-γ effects may act distally. However, even under the conditions of high systemic IFN-γ production during MCMV infection, peak antiviral functions in liver require proximity of NK cells (32). TNF also can be produced by NK cells. However, in contrast to IFN-γ , a wide range of cell types can make this factor at early times during infections, and NK cells are not required for TNF induction during MCMV infections (44, 53). Thus, IFN-γ clearly contributes, and TNF may contribute, to the antiviral effects mediated by NK cells, but NK cells are required for early IFN-γ and for responses dependent upon this cytokine.

Chemokine Production In addition to production of the aforementioned antiviral cytokines, NK cells can make certain of the low-molecular-weight family of cytokines, called chemokines (37, 38). These factors have important chemoattractant and proinflammatory functions (122). Moreover, under the specific circumstances of HIV infections, chemokines also can mediate direct antiviral effects (123, 124). This is because HIV uses chemokine receptors as co-receptors with CD4 molecules (123, 125, 126), and natural receptor ligands can block virus binding and infection (124). One of these ligands is MIP-1α. Although T cells were first shown to be capable of delivering this particular chemokine-mediated mechanism of protection, NK cells have recently been shown to mediate the same effect (38). Given that CCR5 (a chemokine receptor for MIP-1α) is the coreceptor predominantly used by the macrophage tropic HIV isolates associated with spread between individuals (126), this NK cell MIP-1α antiviral pathway

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

198

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL

is likely to be important in early protection against primary infections with the virus.

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

CYTOKINE-MEDIATED REGULATION OF FUNCTION A variety of cytokines have been demonstrated to activate particular NK cell responses and/or to be induced by cytokines activating NK cell functions (see Table 2) (1, 2, 10–14, 17, 18, 31, 32, 37, 43, 44, 54, 127–153). Pathways in the context of innate immune responses—the times at which NK cells are at peak activation and mediating crucial antiviral defense functions—are keys to regulation. In addition to NK cell-produced IFN-γ , other cytokines that may be part of innate responses include IFN-α/β, IL-12, TNF alpha (TNF-α), IL-1α, IL-1β, IL-6, IL-10, transforming growth factors beta (TGF-β), IL-15, and the IFN-γ – inducing factor (IGIF), sometimes called IL-18. Many of these can function to regulate the innate NK cell responses of cytotoxicity, cytokine production, and/or proliferation. Table 2 Innate cytokine and chemokine regulation of natural killer cell functions Cytokine Cytokines IFN-α/β

IL-12

TNF IL-1α,β IL-15

IGIF

IL-10 TGF-β Chemokines MIP-1α MIP-1β MCP-1,2,3 RANTES

Direct and indirect effects on NK cells

Induces cell trafficking and cytotoxic activity; stimulates proliferation in vivo; inhibits IL-12 production; elicits expression of IL-15 mRNA Stimulates IFN-γ production; induces cytoxicity and is critical for the response during certain non-viral, but not viral, infections; synergizes with IL-15 to induce expression of MIP-1α Synergizes with IL-12 for IFN-γ production Synergizes with IL-12 for IFN-γ production Promotes cell growth and maturation; synergizes with IL-12 for induction of IFN-γ production; synergizes with IL-12 to induce expression of MIP-1α Stimulates IFN-γ production; induces cytotoxic activity in culture and following in vivo administration; synergizes with IL-12 for IFN-γ production and cytotoxic activity Inhibits IL-12 production Inhibits IFN-γ production; inhibits IL-12 production; blocks proliferation and cytotoxicity Induces chemotaxis in culture and following infection in vivo Induces chemotaxis in culture Induces chemotaxis in culture Induces chemotaxis in culture

Reference(s)

2, 10, 11, 12, 31, 43, 44–50, 127, 156, 157 1, 2, 13, 35, 37, 44, 53, 54, 86, 128–131, 137 129–131 132 133–141

132, 142–144

145, 146 147–150

32, 151–153 152, 153 151–153 151–153

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NK CELLS IN ANTIVIRAL DEFENSE

199

However, in the context of infections, cytokine effects vary depending upon doses induced and/or interactions with other expressed factors. Accumulating evidence indicates that the host, as a result of identifying particular general pathogen characteristics, responds with different innate cytokine profiles to elicit defense mechanisms most effective against the agent (26). In addition to accessing mechanisms promoting beneficial responses, the elicited innate cytokines are likely to mediate a number of other effects, including inhibiting suboptimal immune responses and protecting against detrimental effects resulting from toxic levels and/or combinations of cytokines. Moreover, the low-molecular-weight chemokines also are likely to be important in facilitating delivery of innate NK cell functions as a result of promoting their accumulation in critical sites. Therefore, the overall picture of an innate immune response to any particular infectious agent may have (a) elements of key, important, and/or dominant responses, (b) absence of other potential responses because inducing factor is not stimulated and/or an inhibitory response is elicited to block the pathway, and (c) presence or absence of factors promoting localization of cells for delivery of innate immune functions. Finally, particularly in the case of viruses, certain innate cytokines can mediate effects directly inhibiting microbial replication in infected cells. Characterization of the regulation and function of NK cell responses to infections must be considered within these larger contexts.

Interferons alpha/beta The virus-induced type 1 interferons, IFN-α/β, are perhaps the best understood innate cytokines in regard to defense functions. IFN-α/β clearly interfere with virus replication (154, 155). In addition, these cytokines have the potential to mediate a broad range of immunoregulatory functions affecting both innate and adaptive responses. At extremely high concentrations in culture, IFN-α/β can be inhibitors of lymphocyte proliferation. However, as stated above, NK cell cytotoxic activity is induced by IFN-α/β, and the cytokines elicit NK cell proliferation in vivo (2, 8, 45–50). At high but physiologically relevant concentrations, IFN-α/β can negatively regulate IL-12 expression in vitro and in vivo (156, 157), and this function contributes to the lack of detectable IL-12 production during LCMV infections (156). The cytokines also elicit IL-15 mRNA expression in culture (127). Because IL-12 is a potent stimulator of NK cell IFN-γ production and IL-15 can be an NK cell proliferation factor (134–141), virus induction of IFN-α/β can control expression of other innate cytokines with effects on NK cell responses. In addition to their known and potential immunoregulatory functions controlling NK cell responses, IFN-α/β can induce effects and/or conditions that may promote CD8 T cell responses in vivo, including enhanced antigen processing and presentation through the MHC class I pathway (158), modified cell trafficking patterns to localize effector cells to sites

P1: SAT/ary

P2: NBL/plb

January 25, 1999

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

200

QC: NBL

12:16

Annual Reviews

AR078-07

BIRON ET AL

of infections and/or to concentrate T and B cells to sites of antigen presentation (31, 159–162), limited proliferation of memory T cells (163), and enhanced expression of functional heterodimeric IL-12 receptor (IL-12R) (164). Thus, these cytokines have the potential to be the dominant regulatory factors shaping innate and adaptive immune responses to viral infections. In humans and mice, the multigene members of this cytokine family include those coding for a single IFN-β and multiple IFN-α proteins. Many viruses are potent inducers of IFN-α/β. Double-stranded RNA (dsRNA), not detectable in normal cells but produced in infected cells, is a known potent inducer of IFN-α/β (154, 162). However, other pathways of induction apparently exist, including interactions through cell surface receptors (165–167). A common heterodimeric receptor is used by IFN-α and IFN-β. Virtually all nucleated cells express IFN-α/βR, and can be induced to make some form of the cytokines. The cytokines clearly promote antiviral states in cells, but the final mediators of IFN-α/β-induced defense functions are understood incompletely. The best-characterized pathways are those initiated by the IFN-α/β–induced enzymes (20 -50 )-oligoadenylate synthetase and the dsRNA-dependent protein kinase PKR to inhibit protein synthesis (154). IFN-α/β–induced antiviral pathways can block replication of a broad range of viruses, including vesicular stomatitis virus (VSV), influenza virus, vaccinia virus, HSV, picornaviruses, and reoviruses. Extensive precise information is available concerning the receptor and signal transduction molecules required for induction of IFN-α/β–mediated effects (154, 168–170). The postligand binding intracellular pathways activate the tyrosine kinases Tyk2 and JAK1 (171, 172), with consequential induction of the downstream signal transducers and activators of transcription STAT1, STAT2, STAT3, and STAT4 (173–179). Activated STAT1 and STAT2 induce the interferon regulatory factor-1 (IRF-1) transcription factor (180). In addition, the RAF-1 and MAP (181, 182) and the VAV and Rac-1 kinase systems (183, 184) are activated. Thus, multiple independent and divergent intracellular signaling pathways are activated by IFN-α/β. Although the pathways by which IFN-α/β activate NK cell-mediated lysis are poorly understood, peak signaling for this response appears to be dependent upon IRF-1 (185).

Interleukin 12 In addition to inducing NK cell IFN-γ production, the innate cytokine IL-12 can contribute to preferential development of T helper type (Th) 1 cells producing IFN-γ over Th2 cells producing IL-4 and IL-5 (27, 28). The biologically active form of this cytokine is a p70 heterodimer comprised of p35 and p40 molecules. IL-12 mRNA is elevated during a range of viral infections (186–188). However, production of biologically active factor has been reported in only a few infections, including those with MCMV and influenza virus (54, 86, 116). The

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NK CELLS IN ANTIVIRAL DEFENSE

201

factor is required for NK cell IFN-γ production during these viral infections (54, 86). Interestingly, LCMV infections fail to induce IL-12 expression but are strong inducers of IL-12-independent T cell IFN-γ production (54, 67, 116, 189; LP Cousens, CA Biron, unpublished results). Influenza virus (86) and mouse hepatitis virus (MHV) (190) also have IL-12-independent pathways for development of T cell IFN-γ responses. Moreover, the conditions of endogenous T cell responses to LCMV are associated with dramatically increased sensitivity to detrimental responses induced by IL-12 exposure (191, 192). Thus, in the context of viral infections, there must be other pathways promoting T cell IFN-γ production, and regulation of IL-12 expression appears to be particularly important. A limited range of cell types are known major producers of biologically active IL-12. These include macrophage lineage cells, neutrophils and dendritic cells (193, 194). The IL-12R is comprised of β1 and β2 chains (195, 196). Studies with IFN-γ in mouse (197) and IFN-α and IFN-γ in human (164, 198) have shown that these factors promote T cell responsiveness to IL-12 by facilitating preservation of IL-12R β2 chain expression. However, in the context of high IFN-α/β levels during viral infections, potential promotion of IL-12-mediated effects as a result of enhanced IL-12R expression would be negated by IFN-α/β-mediated inhibition of the cytokine (156, 157). Characterization of postligand binding intracellular pathways induced by the IL-12R indicate that the Tyk2 and JAK2 kinases are activated with downstream recruitment of STAT3 and STAT4 (199–201), and that STAT4 is an essential element for all the known biological responses induced by IL-12, including NK cell IFN-γ production (202, 203).

Other Innate Cytokines NK cell IFN-γ is an innate cytokine. Although activation of NK cell cytotoxicity during infections with intracellular protozoan parasites is dependent upon the IL-12 to NK cell IFN-γ pathway (128, 131), this is not the case during infections with viruses inducing both IFN-α/β and IL-12 (44, 53). Under the viral conditions, a clear dichotomy of functions with IL-12 required for NK cell IFN-γ production but not for enhanced cytotoxicity, and IFN-α/β (but not IL-12 or NK cell IFN-γ ) production dominant for the enhanced lytic activity. Other innate cytokines with known or potential positive immunoregulatory functions are IL-1α, IL-1β, and TNF-α, and the two newly identified innate cytokines, IL-15 and IGIF. IL-15, can act as an NK cell growth and maturation factor (134–141). It uses the common γ and β, but not the α, chains of the IL-2R (134, 136, 137, 204). An alternative α chain is utilized for the IL-15R (205, 206). IL-15 can drive proliferation of NK cells and memory T cells (127, 135, 139). IGIF has structural similarity to IL-1β (207), is processed to an active form by caspases including the IL-1β converting enzyme (ICE)

P1: SAT/ary

P2: NBL/plb

January 25, 1999

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

202

12:16

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL

(208, 209), uses the IL-1R-related protein (IL-1Rrp) for its receptor (210), and shares certain intracellular signaling pathways with IL-1 (211). IL-15–deficient mice lack NK cells (140), and IGIF-deficient mice have reduced NK cell activity upon challenge with bacterial products (143). TNF, IL-1, IL-15, and IGIF all have been shown to synergize with IL-12 for induction of NK cell IFN-γ production, and to promote NK cell IFN-γ production following parasitic or bacterial infections or challenges with products from these organisms (13, 129–132, 137, 143). However, information on the induction and function of these during viral infections is only now becoming available. Although certain bacterial products induce high levels of IL-1α and IL-1β, these proteins are either undetectable or induced to more modest levels during CMV infections (115, 212). A number of viral infections induce significant levels of TNF-α and IL-6 (44, 53, 115, 212–216). Similar to infections with other agents, endogenous TNF function promotes NK cell IFN-γ production during MCMV infection (44). However, in this context, the effect is relatively modest. Moreover, endogenous expression of the factor also results in conditions modestly interfering with the concurrent IFN-α/β induction of cytotoxicity and proliferation (44). IL-15 mRNA expression is induced in vitro by human herpesvirus-6 (HSV-6) and human herpesvirus-7 (HSV-7) (217, 218), by IFN-α/β (127), and in vivo during LCMV and MCMV infections (GC Pien, LP Cousens, CA Biron, unpublished observations). Given the linkage of this factor’s induction to IFNα/β, it is likely to be an elicited and functional cytokine during viral infections. IGIF has been shown to be induced by influenza virus A infections of human monocytes in culture, and synergize with IFN-α/β for induction of IFN-γ by human T cells (219). Taken together, these observations provide good evidence supporting the potential importance of IL-15 and IGIF—but less the importance for IL-1α, IL-1β, and TNF—in regulation of NK cell responses under the conditions of viral infections and IFN-α/β induction. Cytokines with the potential to be produced and function for the negative regulation of NK cells during innate responses are IL-10 and TGF-β. IL-10 can inhibit induction of IL-12 (37, 145, 146), and TGF-β can block NK cell proliferation and cytotoxicity as well as inhibit induction of IL-12 and NK cell IFN-γ production (147–150). Pathways exist for eliciting these cytokines during both innate and adaptive immune responses. IL-10 is induced under certain conditions of innate immunity to negatively regulate other pro-inflammatory cytokines (146). It can be produced in response to viral infections (220) but it does not appear to be induced or to function at early times during infections with either LCMV or MCMV (LP Cousens, CA Biron, unpublished results). Adaptive immune responses to LCMV promote production of biologically active forms of TGF-β1 (221), and this appears to contribute to the negative regulation of NK cell responses at later times after viral infections (149).

P1: SAT/ary

P2: NBL/plb

January 25, 1999

QC: NBL

12:16

Annual Reviews

AR078-07

NK CELLS IN ANTIVIRAL DEFENSE

203

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Chemokines NK cells can be induced in vivo to traffic with specific localization patterns following treatments with polyI:C or polyI:C-related chemicals and infections with LCMV or MCMV (31, 32). In the case of the chemical treatments, TNF promotes accumulation of NK cells within livers by inducing expression of the cellular adhesion molecule VCAM-1 on vascular endothelial cells, and as a result promotes VCAM-1/VLA-4 interactions between vasculature and NK cells (222, 223). In contrast, NK cell accumulation is induced in MCMV-infected livers (32), but TNF is not required for this response (224). Thus, alternative pathways must exist to promote NK cell migration during viral infections. Other cytokines (IL-12 and IL-15 in particular) can promote the adherence of human NK cells to cultured vascular endothelial cells and subsequent in vitro chemotaxis (225, 226). In addition, chemokines with the potential to promote NK cell chemoattraction and migration can be induced by cytokines. In particular, culture studies with freshly isolated human cells or cell lines show MIP-1α to induce NK cell chemotaxis (151–153), and during infections of mice with MCMV, MIP-1α is required for a focal NK cell migration into livers (32). IFN-γ , IL-12, and IL-15 can independently or interactively stimulate MIP-1α production (37, 227). Other chemokines with clear requirements for cytokine induction are the monokine induced by interferon gamma (mig) (228), and the human interferon-inducible protein 10 (IP-10) (229) along with its mouse counterpart the chemokine responsive to gamma-2 (crg-2) (230). Thus, these chemokines have the potential to regulate NK cell migration during infections that induce IFN-γ expression. All of the chemokines are induced at least at the mRNA level during certain viral infections (38, 230–232), and MIP-1α is induced at the protein level during MCMV and HIV infections (32, 38). These factors may be jointly required and/or function in different specific anatomical compartments. In this regard, it is interesting to note that although MIP-1α is required for the MCMV-induced NK cell migration into liver (32), it may not be necessary for another unique NK cell–dependent trafficking pattern induced in the spleen (31; TP Salazar-Mather, CA Biron, unpublished results).

CELL SURFACE RECEPTORS In addition to cytokine receptors, NK cells have a variety of well-defined cell surface molecules that can deliver positive or negative stimulatory signals. These have been recently reviewed by others (233–238). Although there are large numbers of such candidate receptors, the evidence linking any of these to NK cell-mediated antiviral defense is virtually nonexistent. However, there are persuasive rationales for the potential contribution of certain of these to the delivery of NK cell–mediated defense, and for others in the regulation of potentially

P1: SAT/ary

P2: NBL/plb

January 25, 1999

204

QC: NBL

12:16

Annual Reviews

AR078-07

BIRON ET AL

detrimental effects mediated by activated NK cells, during viral infections. In particular, receptors for immunoglobulins may play roles in ADCC against, and stimulation to deliver positive signals is required for direct NK cell-mediated lysis of, virus-infected cells. Furthermore, negative regulation of activated NK cell cytotoxicity must be important in limiting potential detrimental effects delivered to normal host cells.

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Fc Receptors The first characterized cell surface molecules for activation of NK cells were the Fc receptors of the type III, identified as CD16. Upon cross-linking through bound immunoglobulins, these molecules can deliver positive signals to mediate ADCC and/or produce IFN-γ (1). The signal transduction pathway (reviewed in 233) occurs via activation of src-family tyrosine kinases and phosphorylation of immunoreceptor tyrosine-based activating motifs (ITAM) within the noncovalently associated intermediates, the γ subunit of the high-affinity IgE receptor, i.e. FcεRI-γ , in mouse NK cells and either this molecule or the ζ subunit of the TCR complex in human NK cells. Phosphorylation of ZAP70, activation of phospholipases, stimulation of phosphatidylinosital 3-kinase and MAP kinase induction, p21 ras activation, and translocation of NFATp and NFATc also occur. Because the FcR-activated defense pathways would require virus-specific antibody attachment to infected target cells, they should come into play only after antibody production at late times after primary, or during secondary, infections. To date, the evidence for effectiveness of antibody-activated NK cell responses, as in vivo antiviral defense mechanisms, is minimal (77, 239). New studies suggest that in contrast to direct cytotoxicity, NK cell–mediated ADCC is maintained later into HIV infection (103) and may contribute to CD4 T cell decline during HIV disease (101). These observations support a role for NK cell-mediated ADCC in pathogenesis during, and possible defense against, viral infection.

Other Receptors Activating NK Cells A variety of other cell surface molecules have been reported to contribute to NK cell activation for cytotoxicity. However, receptors critical to antiviral defense have not been identified. As stated above, NK1.1, a member of the NK receptor protein 1 (NKR-P1) family, can induce NK cell IFN-γ production (15, 16), but the natural ligand for this receptor has not been defined. If, as the induction of high levels of certain cytokines suggests, NK cell IFN-γ responses can be exclusively or primarily a result of soluble cytokine stimulation and production during viral infections, interactions with such positive receptors may not be necessary or may be only marginally enhancing for NK cell IFN-γ production and function under these conditions. However, NK cell–mediated cytotoxicity

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NK CELLS IN ANTIVIRAL DEFENSE

205

requires stimulation through receptors inducing positive signals for release of the chemical mediators of the function. Measles (88), mumps (89), and Sendai (90) virus glycoproteins have been reported to promote cytotoxicity as a result of stimulating NK cells and/or facilitating binding of NK to target cells, but the NK cell receptor counterparts interacting with these await characterization. An interesting group of molecules with activating functions are isoforms of the MHC class I receptors delivering negative signals to NK cells described below (233–238). The positive induction receptors express shorter cytoplasmic tails lacking critical elements for negative signaling. Recent evidence indicates that certain of these noncovalently associate in human and mouse with a disulfidebonded homodimer, DAP12, containing an ITAM motif in its cytoplasmic tail, that this molecule is phosphorylated upon cross-linking of the complex, and that there are downstream interactions with ZAP 70 and Syk proteins (240, 241). However, a perplexing issue is how receptors having similar external binding domains but different intracellular pathways to alternatively mediate positive or negative signals would be regulated and appropriately accessed to induce antiviral defense mechanisms as needed.

MHC Class I Receptors Delivering Negative Signals to NK Cells A flood of information has recently occurred concerning multiple types of negative receptors on NK cells (233–238). Many of these receptors respond to stimulation by MHC class I expressed on surfaces of target cells. Such receptors were first predicted to exist as a mechanism to let NK cells know if neighboring cells were normal and as a result, protect them from inappropriate and detrimental NK cell attack (242). In mice, the best-defined are the Ly49 receptors, type II membrane glycoproteins of the C-type lectin superfamily. In humans, the best-characterized are the killing inhibitory receptors (KIRs), type I glycoproteins of the Ig superfamily. Both the human and mouse receptors are products of gene clusters with several homologous genes. The individual receptors have ligand specificity for products of different groups of classical MHC class I alleles. Recently, an additional negative receptor complex with a type II membrane family component, CD94/NKG2A, has been identified on human NK cells, and shown to recognize MHC leader sequence peptide presented by the nonclassical MHC molecule, HLA-E (243). These molecules are able to deliver negative signals to NK cells through the immunoreceptor tyrosinebased inhibitory motifs (ITIM) in their cytoplasmic domains to recruit tyrosine phosphatases and cut off proximal positive signaling pathways (reviewed in 233–238). They have the potential to inhibit a number of cellular responses in addition to cytotoxicity, including induction of cytokine production, provided that the negative signals are delivered in the vicinity of positive signals.

P1: SAT/ary

P2: NBL/plb

January 25, 1999

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

206

12:16

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL

Negative signaling through NK cell receptors would require both MHC class I expression and cell-to-cell contact for delivery. An intriguing correlation is that the viruses most clearly associated with NK cell–mediated defense, i.e. those in the herpesgroup, have mechanisms for decreasing expression of MHC class I molecules on infected cell surfaces (108–113). The effects are thought to be selected to protect infected cells from CD8 T cell recognition. As a result of concurrent removal of class I negative signals, however, these viruses may render themselves more sensitive to NK cell–mediated cytotoxicity and/or provide environments more supportive for induction of NK cell IFN-γ production. NK cells may be particularly important for defense under precisely such conditions of suboptimal T cell effectiveness. However, correlation of the ability to sabotage class I expression and a demonstrable role for NK cells in defense do not prove cause and effect. Interestingly, cytomegaloviruses have genes coding for class I homologues, e.g. UL18 in HCMV and m144 in MCMV, which may serve as decoys to deliver negative signals and to protect virus-infected cells with blocked class I expression from NK cell–mediated lysis (110, 244– 246). Further work is needed to establish a definitive role for decoy molecules in avoidance of NK cell–mediated defenses. However, certain host viral resistance genes have been mapped to a genetic region with clustered NK cell surface receptors for class I molecules in the mouse (81, 99, 247). Because the site has genes encoding for receptors of both positive and negative signals, the location suggests that the viral resistance gene products are receptors either receiving positive signals for activation or substituting for allelic variants capable of receiving negative signals from certain virus-infected cells. In summary, much remains to be learned about the significance of NK cell negative receptor signaling, but the evidence accumulating indicates that these pathways are involved in regulating NK cell–mediated effects in the host.

SHAPING OF ADAPTIVE IMMUNE RESPONSES In addition to immediate effects, innate responses can have immunoregulatory roles to provide conditions promoting and/or shaping downstream adaptive responses. Although understanding of these is advancing in the context of bacterial and parasitic infections (5, 18, 26–28), their characterization in the context of viral infections is in its infancy. Immune responses to viruses include an assortment of those also observed during infections with other organisms, but they often have components unique to, and/or uniquely dominant during, viral infections (reviewed in 248). As stated above, these include high circulating levels of IFN-α/β and numerous IFN-α/β–mediated effects. Moreover, because many viral protein products are readily available for processing and presentation by MHC class I molecules, CD8 T cells can have dramatic responses and

P1: SAT/ary

P2: NBL/plb

January 25, 1999

QC: NBL

12:16

Annual Reviews

AR078-07

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NK CELLS IN ANTIVIRAL DEFENSE

207

play prominent roles during viral infections, and T cell cytotoxic functions can sometimes be induced at high endogenous levels. Finally, T cell IFN-γ responses are generally induced and detected at high levels, but mixtures of other Th1 and Th2 cytokines, including IL-4, also can be expressed. Limited information is available concerning shaping of adaptive immune responses by innate immunity during viral infections (249). It is clear that IL12-independent pathways must exist for the induction of T cell IFN-γ responses (54, 86, 190). One report suggests that NK cells can play a role in the induction of specific CTL during influenza virus infection (250). However, this is not a universal finding in all viral infections (249, 251; CA Biron, KB Nguyen, unpublished data), and CTL development is independent of IFN-γ in certain viral infections (252, 253). Because the IFN-α/β cytokines have the ability to mediate a broad range of immunoregulatory functions with the potential of promoting T cell responses (158 –164), they are good candidates for dominating innate cytokine responses and shaping adaptive immune responses during viral infections.

SUMMARY We now have extensive knowledge concerning NK cells, regulation of their responses, and the significance of their biological functions in the host. This is particularly true in the context of viral infections (represented schematically in Figure 1), where NK cells have been shown to be induced to mediate high levels of cytotoxicity in response to endogenously expressed IFN-α/β and, under certain conditions, produce impressive amounts of IL-12-induced IFN-γ and mediate antiviral defense. In addition, NK cells have been demonstrated to respond to chemokines for localization and delivery of antiviral functions in particular tissues and to produce certain of these factors for antiviral defense against HIV. The picture developing suggests certain apparently dominant cytokines (i.e. IFN-α/β) modulating other cytokines as well as NK cell responses. It also provides insights into the environment produced by the host to fight off viral infections and promote protective immunity. Although much progress has been made, many important questions on the regulation and function of NK cells remain unanswered: Is there a role for the NK T cell population in defense and immunoregulation? Are there other players in the innate cytokine milieu contributing to NK cell regulation in particular settings? Are cell-to-cell contact events required for certain of the NK cell–mediated effects? If so, how do the NK cell receptors recognize and respond to determinants specifically identifying infected cells? Do NK cells play immunoregulatory roles in vivo? How do the innate and NK cell responses act to shape downstream adaptive immune responses and promote those most beneficial for defense? Given the

P1: SAT/ary

P2: NBL/plb

January 25, 1999

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

208

12:16

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL

Figure 1 Characterized and potential innate cytokine-mediated pathways accessing early NK cell, and promoting downstream adaptive immune, responses in defense against viral infections. The profiles of early cytokines are likely to be modified in the context of particular agents. However, there are certain unique or uniquely dominant features to those elicited by viruses. In particular, many viral infections induce systemic and high levels of IFN-α/β. These cytokines are good inducers of NK cell cytotoxicity, and elicit the proliferation of NK cell and memory T cell populations in vivo. A candidate intermediary factor for proliferation is IL-15. IL-12 is induced in some, but not all, viral infections. When present, it is tightly regulated and expressed within a narrow window of time. It is required for NK cell IFN-γ responses to viruses, and these responses contribute significantly to NK cell–mediated antiviral functions. IGIF is an example of a cytokine having known synergistic effects with IL-12 for the induction of NK cell IFN-γ . Others with synergistic functions are TNF, IL-1, and IL-15, but IL-1 may not be expressed at high levels in response to all viruses. The IFN-α/β responses negatively regulate IL-12 production during viral infections. Chemokine induction for NK cell migration, and NK cell production of chemokines for defense, are both likely to play roles in certain specific conditions of viral infections. Clearly, there are IL-12-independent pathways for promoting T cell IFN-γ responses during viral infections. Dark solid arrows represent known in vivo pathways or responses. Broken arrows represent proposed pathways. The X represents inhibition.

demonstrated importance of NK cells in defense against viruses, addressing these questions in the context of viral infections promises to further reveal both specific and global mechanisms for the regulation and function of innate immune responses. ACKNOWLEDGMENTS We thank our many colleagues past and present for the contributions to the study of NK cells and endogenous immune responses to viral infections. Work

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

NK CELLS IN ANTIVIRAL DEFENSE

209

in the authors’ laboratory is supported by Public Health Service Grants RO1 CA41268, KOI CA79076 and RO1 MH47674. GC Pien is a predoctoral fellow of the Howard Hughes Medical Institute. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Literature Cited 1. Trinchieri G. 1989. Biology of natural killer cells. Adv. Immunol. 47:187–376 2. Biron CA. 1997. Activation and function of natural killer cell responses during viral infections. Curr. Opin. Immunol. 9:24–34 3. Unanue ER. 1997. Inter-relationship among macrophages, natural killer cells and neutrophils in early stages of Listeria resistance. Curr. Opin. Immunol. 9:35–43 4. Scharton-Kersten TM, Sher A. 1997. Role of natural killer cells in innate resistance to protozoan infections. Curr. Opin. Immunol. 9:44–51 5. Scott P, Trinchieri G. 1995. The role of natural killer cells in host-parasite interactions. Curr. Opin. Immunol. 7:34–40 6. Lanier LL, Spits H, Phillips JH. 1992. The developmental relationship between NK cells and T cells. Immunol. Today 13:392– 95 7. Biron CA, van den Elsen P, Tutt MM, Medveczky P, Kumar V, Terhorst C. 1987. Murine natural killer cells stimulated in vivo do not express the T cell receptor α, β, T3δ, or T3ε genes. J. Immunol. 139:1704–10 8. Biron CA, Su HC, Orange JS. 1996. Function and regulation of natural killer (NK) cells during viral infections: characterization of responses in vivo. Methods: Companion Methods Enzymol. 9:379–93 9. Wang B, Biron C, She J, Higgins K, Sunshine M-J, Lacy E, Lonberg N, Terhorst C. 1994. A block in both early T lymphocyte and natural killer cell development in transgenic mice with high-copy numbers of the human CD3εgene. Proc. Natl. Acad. Sci. USA 91:9402–6 10. Trinchieri G, Santoli D, Koprowski H. 1978. Spontanteous cell-mediated cytotoxicity in humans: role of interferon and immunoglobulins. J. Immunol. 120: 1849–55 11. Santoli D, Trinchieri G, Koprowski H. 1978. Cell-mediated cytotoxicity against virus-infected target cells in humans. II. Interferon induction and activation of natural killer cells. J. Immunol. 121:532–38

¨ A, Wigzell H, Senik A, 12. Gidlund M, Orn Gresser I. 1978. Enhanced NK cell activity in mice injected with interferon and interferon inducers. Nature 273:759–61 13. Chan SH, Perussia B, Gupta JW, Kobayashi M, Pospisil M, Young HA, Wolf SF, Young D, Clark SC, Trinchieri G. 1991. Induction of interferon γ production by natural killer cell stimulatory factor: characterization of the responder cells and synergy with other inducers. J. Exp. Med. 173:869–79 14. Bancroft GJ, Schreiber RD, Unanue ER. 1991. Natural immunity: a T-cellindependent pathway of macrophage activation, defined in scid mouse. Immunol. Rev. 124:5–24 15. Arase H, Arase N, Saito T. 1996. Interferon γ production by natural killer (NK) cells and NK1.1+ T cells upon NKRP1 cross-linking. J. Exp. Med. 183:2391– 96 16. Kim S, Yokoyama WM. 1998. NK cell granule exocytosis and cytokine production inhibited by Ly49A engagement. Cell. Immunol. 183:106–12 17. Warren HS, Kinnear BF, Phillips JH, Lanier LL. 1995. Production of IL-5 by human NK cells and regulation of IL-5 secretion by IL-4, IL-10, and IL-12. J. Immunol. 154:5144–52 18. Biron CA, Gazzinelli RT. 1995. Effects of IL-12 on immune responses to microbial infections: a key mediator in regulating disease outcome. Curr. Opin. Immunol. 7:485–96 19. Scharton TM, Scott P. 1993. Natural killer cells are a source of interferon γ that drives differentiation of CD4+ T cell subsets and induces early resistance to Leishmania major in mice. J. Exp. Med. 178:567–77 20. Wysocka M, Kubin M, Vieira LQ, Ozmen L, Garotta G, Scott P, Trinchieri G. 1995. Interleukin-12 is required for interferon-γ production and lethality in lipopolysaccharide-induced shock in mice. Eur. J. Immunol. 25:672–76

P1: SAT/ary

P2: NBL/plb

January 25, 1999

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

210

12:16

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL

21. Ozmen L, Pericin M, Hakimi J, Chizzonite RA, Wysocka M, Trinchieri G, Gately M, Garotta G. 1994. Interleukin 12, interferon γ , and tumor necrosis factor α are key cytokines of the generalized Schwartzman reaction. J. Exp. Med. 180:907–15 22. Heremans H, Dillen C, van Damme J, Billiau 1994. Essential role for natural killer cells in the lethal lipopolysaccharideinduced Shwartzman-like reaction in mice. Eur. J. Immunol. 24:1155–60 23. Bendelac A, Rivera MN, Park S-H, Roark JH. 1997. Mouse CD1-specific NK1 T cells: development, specificity, and function. Annu. Rev. Immunol. 15:535– 62 24. Vicari AP, Zlotnik A. 1996. Mouse NK1.1+ T cells: a new family of T cells. Immunol. Today 17:71–76 25. MacDonald HR. 1995. NK1.1+ T cell receptor-α/β+ cells: new clues to their origin, specificity, and function. J. Exp. Med. 182:633–38 26. Medzhitov R, Janeway CA. 1997. Innate immunity: impact on the adaptive immune response. Curr. Opin. Immunol. 9:4–9 27. Hsieh CS, Macatonia SE, Tripp CS, Wolf SF, O’Garra A, Murphy KM. 1993. Development of TH1 CD4+ T cells through IL-12 produced by Listeriainduced macrophages. Science 260:547– 49 28. Manetti R, Parronchi P, Giudizi MG, Piccini MP, Maggi E, Trinchieri G, Romagnani S. 1993. Natural killer cell stimulatory factor (interleukin 12 [IL-12]) induces T helper type 1 (Th1)-specific immune responses and inhibits the development of IL-4-producing Th cells. J. Exp. Med. 177:1199–204 29. Wiltrout RH, Mathieson BJ, Talmadge JE, Reynolds CW, Zhang SR, Herberman RB, Ortaldo JR. 1984. Augmentation of organ-associated natural killer activity by biological response modifiers: isolation and characterization of large granular lymphocytes from the liver. J. Exp. Med. 160:1431–49 30. Wiltrout RH, Pilaro AM, Gruys ME, Talmadge JE, Longo DL, Ortaldo JR, Reynolds CW. 1989. Augmentation of mouse liver-associated natural killer activity by biologic response modifiers occurs largely via rapid recruitment of large granular lymphocytes from the bone marrow. J. Immunol. 143:372–78 31. Salazar-Mather TP, Ishikawa R, Biron CA. 1996. Natural killer (NK) cell trafficking and cytokine expression in

32.

33.

34.

35.

36.

37.

38.

39. 40.

41.

42.

splenic compartments after interferoninduction and viral infection. J. Immunol. 157:3054–64 Salazar-Mather TP, Orange JS, Biron CA. 1998. Early murine cytomegalovirus (MCMV) infection induces liver natural killer (NK) cell inflammation and protection through macrophage inflammatory protein 1α (MIP-1α)-dependent pathways. J. Exp. Med. 187:1–14 Biron CA, Habu S, Okumura K, Welsh RM. 1984. Lysis of uninfected and virusinfected cells in vivo: a rejection mechanism in addition to that mediated by natural killer cells. J. Virol. 50:698–707 Bogen SA, Fogelman I, Abbas AK. 1983. Analysis of IL-2, IL-4, and IFN-γ producing cells in situ during immune responses to protein antigens. J. Immunol. 150:4197–205 Scharton-Kersten T, Scott P. 1995. The role of the innate immune response in Th1 cell development following Leishmania major infection. J. Leukocyte Biol. 57:515–22 Bigger JE, Thomas CA, Atherton SS. 1998. NK cell modulation of murine cytomegalovirus retinitis. J. Immunol. 160: 5826–31 Bluman EM, Bartynski KJ, Avalos BR, Caligiuri MA. 1996. Human natural killer cells produce abundant macrophage inflammatory protein-1α in response to monocyte-derived cytokines. J. Clin. Invest. 97:2722–27 Oliva A, Kinter AL, Vaccarezza M, Rubbert A, Catanzaro A, Moir S, Monaco J, Ehler L, Mizell S, Jackson R, Li Y, Romano JW, Fauci AS. 1998. Natural killer cells from Human Immunodeficiency Virus (HIV)-infected individuals are an important source of CC-chemokines and suppress HIV-1 entry and replication in vitro. J. Clin. Invest. 102:223–31 Koo GC, Peppard JR. 1984. Establishment of monoclonal anti-NK1.1 antibody. Hybridoma 3:301–3 Ortaldo JR, Winkler-Pickett R, Mason AT, Mason LH. 1998. The Ly-49 family: regulation of cytotoxicity and cytokine production in murine CD3+ cells. J. Immunol. 160:1158–65 Robertson JM, Caligiuri MA, Manley TJ, Levine H, Ritz J. 1990. Human natural killer cell adhesion molecules. Differential expression after activation and participation in cytolysis. J. Immunol. 145:3194–201 Welsh RM. 1978. Cytotoxic cells induced during lymphocytic choriomeningitis virus infection of mice. I. Character-

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

NK CELLS IN ANTIVIRAL DEFENSE

43.

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

44.

45. 46.

47.

48.

49.

50.

51.

52.

53.

54.

ization of natural killer cell induction. J. Exp. Med. 148:163–81 Grundy (Chalmer) JE, Trapman J, Allan JE, Shellam GR, Melief CJM. 1982. Evidence for a protective role of interferon in resistance to murine cytomegalovirus and its control by non-H-2-linked genes. Infect. Immun. 37:143–50 Orange JS, Biron CA. 1996. Characterization of early IL-12, IFN-α/β, and TNF effects on antiviral state and NK cell responses during murine cytomegalovirus infection. J. Immunol. 156:4746–56 Biron CA, Welsh RM. 1982. Blastogenesis of natural killer cells during viral infection in vivo. J. Immunol. 129:2788–95 Biron CA, Sonnenfeld G, Welsh RM. 1984. Interferon induces natural killer cell blastogenesis in vivo. J. Leukocyte Biol. 35:31–37 Biron CA, Turgiss LR, Welsh RM. 1983. Increase in NK cell number and turnover rate during acute viral infection. J. Immunol. 131:1539–45 Biron CA, Young HA, Kasaian MT. 1990. Interleukin 2-induced proliferation of murine natural killer cells in vivo. J. Exp. Med. 171:173–88 Kasaian MT, Biron CA. 1990. Cyclosporin A inhibition of interleukin 2 gene expression, but not natural killer cell proliferation, after interferon induction in vivo. J. Exp. Med. 171:745–62 Su HC, Orange JS, Fast LD, Chan AT, Simpson SJ, Terhorst C, Biron CA. 1994. IL-2-dependent NK cell responses discovered in virus-infected β 2-microglobulin-deficient mice. J. Immunol. 153:5674–81 Kasaian MT, Biron CA. 1989. The activation of IL-2 transcription in L3T4+ and Lyt-2+ lymphocytes during virus infection in vivo. J. Immunol. 142:1287–92 Kasaian MT, Biron CA. 1990. Effects of cyclosporin A on IL-2 production and lymphocyte proliferation during infection of mice with lymphocytic choriomeningitis virus. J. Immunol. 144:299–306 Orange JS, Wang B, Terhorst C, Biron CA. 1995. Requirement for natural killer cell-produced interferon γ in defense against murine cytomegalovirus infection and enhancement of this defense pathway by interleukin 12 administration. J. Exp. Med. 182:1045–56 Orange JS, Biron CA. 1996. An absolute and restricted requirement for IL-12 in natural killer cell interferon-γ production and antiviral defense: studies of natural killer and T cell responses in contrasting viral infections. J. Immunol. 156:1138–42

211

55. Ohteki T, MacDonald HR. 1994. Major histocompatibility complex class I related molecules control the development of CD4+ and CD4− subsets of natural killer 1.1+ T cell receptor-α/β+ cells in the liver of mice. J. Exp. Med. 180:699– 704 56. Lantz O, Bendelac A. 1994. An invariant T cell receptor α chain is used by a unique subset of major histocompatibility complex class I-specific CD4+ and CD4−CD8− T cells in mice and humans. J. Exp. Med. 180:1097–106 57. Taniguchi M, Koseko H, Tokuhisa T, Masuda K, Sato H, Kondo E, Kawano T, Cui J, Perkes A, Koyasu S, Makino Y. 1996. Essential requirement of an invariant Vα14 T cell antigen receptor expression in the development of natural killer T cells. Proc. Natl. Acad. Sci. USA 93:11025–28 58. Bendelac A. 1995. Positive selection of mouse NK1+ T cells by CD1expressing cortical thymocytes. J. Exp. Med. 182:2091–96 59. Bendelac A, Lantz O, Quimby ME, Yewdell JW, Bennink JR, Brutkiewicz RR. 1995. CD1 recognition by mouse NK1+ T lymphocytes. Science 268:863– 65 60. Bendelac A, Killeen N, Littman, DR, Schwartz RH. 1994. A subset of CD4+ thymocytes selected by MHC class I molecules. Science 263:1774–78 61. Murakami M, Paul WE. 1998. Agedependent appearance of NK1.1+ T cells in the livers of β2-microglobulin knockout and SJL mice. J. Immunol. 160:2649– 54 62. 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 63. Koyasu S. 1994. CD3+CD16+NK1.1 + B220+ large granular lymphocytes arise from both αβTCR+ CD4−CD8− and γ δ TCR+ CD4−CD8− cells. J. Exp. Med. 179:1957–72 64. Yoshimoto T, Paul WE. 1994. CD4pos NK1.1pos T cells promptly produce IL4 in response to in vivo challenge with anti-CD3. J. Exp. Med. 179:1285–95 65. Chen H, Paul WE. 1997. Cultured NK1.1+ CD4+ T cells produce large amounts of IL-4 and IFN-γ upon activation by anti-CD3 or CD1. J. Immunol. 159:2240–44 66. Maher JK, Kronenberg M. 1997. The role of CD1 molecules in immune responses to infection. Curr. Opin. Immunol. 9:456–61

P1: SAT/ary

P2: NBL/plb

January 25, 1999

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

212

12:16

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL

67. Su HC, Cousens LP, Fast LD, Slifka MK, Bungiro RD, Ahmed R, Biron CA. 1998. CD4+ and CD8+ T cell interactions in IFN-γ and IL-4 responses to viral infections: requirements for IL-2. J. Immunol. 160:5007–17 68. Casta˜no AR, Tangri S, Miller JE, Holcombe HR, Jackson MR, Huse WD, Kronenberg M, Peterson PA. 1995. Peptide binding and presentation by mouse CD1. Science 269:223–26 69. Makino M, Winkler DF, Wunderlich J, Hartley JW, Morse HC, Holmes KL. 1993. High expression of NK-1.1 antigen is induced by infection with murine AIDS virus. Immunology 80:319–25 70. Raska K Jr, Gallimore PH. 1982. An inverse relation of the oncogenic potential of adenovirus-transformed cells and their sensitivity to killing by syngeneic natural killer cells. Virology 123:8–18 71. Sheil JM, Gallimore PH, Zimmer SG, Sopori ML. 1984. Susceptibility of adenovirus-2-transformed rat cell lines to natural killer (NK) cells: direct correlation between NK resistance and in vivo tumorigenesis. J. Immunol. 132:1578–82 72. 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 73. Walker CM, Rawls WE, Rosenthal KL. 1984. Generation of memory cellmediated immune responses after secondary infection of mice with pichinde virus. J. Immunol. 132:469–74 74. Welsh RM, Brubaker JO, Vargas-Cortes M, O’Donnell CL. 1991. Natural killer (NK) cell response to virus infections in mice with severe combined immunodeficiency. The stimulation of NK cells and the NK cell-dependent control of virus infections occur independently of T and B cell function. J. Exp. Med. 173:1053– 63 75. Welsh RM, Dundon PL, Eynon EE, Brubaker JO, Koo GC, O’Donnell CL. 1990. Demonstration of the antiviral role of natural killer cells in vivo with a natural killer cell-specific monoclonal antibody (NK1.1). Nat. Immun. Cell Growth Regul. 9:112–20 76. Habu S, Akamatsu K, Tamaoki N, Okumura K. 1984. In vivo significance of NK cell on resistance against virus (HSV-1) infections in mice. J. Immunol. 133:2743– 47 77. Ching C, Lopez C. 1979. Natural killing of herpes simplex virus type 1-infected target cells: normal human responses and

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

influence of antiviral antibody. Infect. Immun. 26:49–56 Biron CA, Byron KS, Sullivan JL. 1989. Severe herpesvirus infections in an adolescent without natural killer cells. N. Engl. J. Med. 320:1731–35 Bancroft GJ, Shellam GR, Chalmer JE. 1981. Genetic influences on the augmentation of natural killer (NK) cells during murine cytomegalovirus infection: correlation with patterns of resistance. J. Immunol. 126:988–94 Bukowski JF, Woda BA, Welsh RM. 1984. Pathogenesis of murine cytomegalovirus infection in natural killer cell-depleted mice. J. Virol. 52:119–28 Scalzo AA, Fitzgerald NA, Wallace CR, Gibbons AE, Smart YC, Burton RC, Shellam GR. 1992. The effect of the Cmv1 resistance gene, which is linked to the natural killer cell gene complex, is mediated by natural killer cells. J. Immunol. 149:581–89 Quinnan GV Jr, Kirmani N, Rook AH, Manischewitz JF, Jackson L, Moreschi G, Santos GW, Saral R, Burns WH. 1982. Cytotoxic T cells in cytomegalovirus infection: HLA-restricted T-lymphocyte and non-T-lymphocyte cytotoxic responses correlate with recovery from cytomegalovirus infection in bone-marrowtransplant recipients. N. Engl. J. Med. 307:7–13 Merino F, Henle W, Ramirez-Duque P. 1986. Chronic active Epstein-Barr virus infection in patients with ChediakHigashi syndrome. J. Clin. Immunol. 6: 299–305 Joncas J, Monczak Y, Ghibu F, Alfieri C, Bonin A, Ahronheim G, Rivard G. 1989. Brief report: killer cell defect and persistent immunological abnormalities in two patients with chronic active Epstein-Barr virus infection. J. Med. Virol. 28:110–17 Stein-Streilein J, Guffee J. 1986. In vivo treatment of mice and hamsters with antibodies to asialo GM1 increases morbidity and mortality to pulmonary influenza infection. J. Immunol. 136:1435–41 Monteiro JM, Harvey C, Trinchieri G. 1998. Role of interleukin-12 in primary influenza virus infection. J. Virol. 72:4825–31 Ballas ZK, Turner JM, Turner DA, Goetzman EA, Kemp JD. 1990. A patient with simultaneous absence of “classical” natural killer cells (CD3−, CD16+, and NKH1+) and expansion of CD3+, CD4−, CD8−, NKH1+ subset. J. Allergy Clin. Immunol. 85:453–59 Casali P, Oldstone MB. 1982. Mecha-

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

NK CELLS IN ANTIVIRAL DEFENSE

89.

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

90.

91.

92. 93.

94.

95.

96.

97.

98.

99.

nisms of killing of measles virus-infected cells by human lymphocytes: interferonassociated and unassociated cell-mediated cytotoxicity. Cell. Immunol. 70:330– 44 Harfast B, Orvell C, Alsheikhly A, Andersson T, Perlmann P, Norrby E. 1980. The role of viral glycoproteins in mumps-virus-dependent lymphocytemediated cytotoxicity in vitro. Scand. J. Immunol. 11:391–400 Alsheikhly A, Orvell C, Harfast B, Andersson T, Perlmann P, Norrby E. 1983. Sendai virus induced cell mediated cytotoxicity in vitro. The role of viral glycoproteins in cell mediated cytotoxicity. Scand. J. Immunol. 17:129–38 Godeny EK, Gauntt CJ. 1986. Involvement of natural killer cells in coxsackievirus B3-induced murine myocarditis. J. Immunol. 137:1695–702 Godeny EK, Gauntt CJ. 1987. Murine natural killer cells limit coxsackievirus B3 replication. J. Immunol. 139:913–18 Hassin D, Fixler R, Bank H, Klein AS, Hasin Y. 1985. Cytotoxic T lymphocytes and natural killer cell activity in the course of mengo virus infection of mice. Immunology 56:701–5 McFarland HI, Bigley NJ. 1989. Sexdependent, early cytokine production by NK-like spleen cells following infection with the D variant of encephalomyocarditis virus (EMCV-D). Viral Immunol. 2: 205–14 White LL, Smith RA. 1990. D variant of encephalomyocarditis virus (EMCV-D)induced diabetes following natural killer cell depletion in diabetes-resistant male C57BL/6J mice. Viral Immunol. 3:67– 76 Paya CV, Patick AK, Leibson PJ, Rodriguez M. 1989. Role of natural killer cells as immune effectors in encephalitis and demyelination induced by Theiler’s virus. J. Immunol. 143:95–102 Stitz L, Althage A, Hengartner H, Zinkernagel R. 1985. Natural killer cells vs cytotoxic T cells in the peripheral blood of virus-infected mice. J. Immunol. 134:598–602 Jacoby RO, Bhatt PN, Brownstein DG. 1989. Evidence that NK cells and interferon are required for genetic resistance to lethal infection with ectromelia virus. Arch. Virol. 108:49–58 Delano ML, Brownstein DG. 1995. Innate resistance to lethal mousepox is genetically linked to the NK gene complex on chromosome 6 and correlates with early restriction of virus replication by cells

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

213

with an NK phenotype. J. Virol. 69:5875– 77 Ruscetti FW, Mikovits JA, Kalyanaraman VS, Overton R, Stevenson H, Stromberg K, Herberman RB, Farrar WL, Ortaldo JR. 1986. Analysis of effector mechanisms against HTLV-I- and HTLVIII/LAV-infected lymphoid cells. J. Immunol. 136:3619–24 Skowron G, Cole BF, Zheng D, Accetta G, Yen-Lieberman B. 1997. gp120-directed antibody-dependent cellular cytotoxicity as a major determinant of the rate of decline in CD4 percentage in HIV-1 disease. AIDS 11:1807–14 Bonavida B, Katz J, Gottlieb M. 1986. Mechanism of defective NK cell activity in patients with acquired immunodeficiency syndrome (AIDS) and AIDSrelated complex. I. Defective trigger on NK cells for NKCF production by target cells, and partial restoration by IL 2. J. Immunol. 137:1157–63 Katz JD, Mitsuyasu R, Gottlieb MS, Lebow LT, Bonavida B. 1987. Mechanism of defective NK cell activity in patients with acquired immunodeficiency syndrome (AIDS) and AIDS-related complex. II. Normal antibody-dependent cellular cytotoxicity (ADCC) mediated by effector cells defective in natural killer (NK) cytotoxicity. J. Immunol. 139:55– 60 Minato N, Bloom BR, Jones C, Holland J, Reid LM. 1979. Mechanism of rejection of virus persistently infected tumor cells by athymic nude mice. J. Exp. Med. 149:1117–33 Griffin DE, Hess JL. 1986. Cells with natural killer activity in the cerebrospinal fluid of normal mice and athymic nude mice with acute Sindbis virus encephalitis. J. Immunol. 136:1841–45 Hirsch RL. 1981. Natural killer cells appear to play no role in the recovery of mice from Sindbis virus infection. Immunology 43:81–89 Trgovcich J, Aronson JF, Johnston RE. 1996. Fatal Sindbis virus infection of neonatal mice in the absence of encephalitis. Virology 224:73–83 Campbell AE, Slater JS. 1994. Downregulation of major histocompatibility complex class I synthesis by murine cytomegalovirus early gene expression. J. Virol. 68:1805–11 Del Val M, Hengel H, H¨acker H, Hartlaub U, Ruppert T, Lucin P, Koszinowski UH. 1992. Cytomegalovirus prevents antigen presentation by blocking the transport of peptide-loaded major histocompatibil-

P1: SAT/ary

P2: NBL/plb

January 25, 1999

214

110. 111.

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

112.

113.

114.

115.

116. 117.

118. 119.

120.

121. 122.

12:16

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL ity complex class I molecules into the medial-golgi compartment. J. Exp. Med. 176:729–38 Hengel H, Koszinowski UH. 1997. Interference with antigen processing by viruses. Curr. Opin. Immunol. 9:470–77 Hill A, Jugovic P, York I, Russ G, Bennink J, Yewdell J, Ploegh H, Johnson D. 1995. Herpes simplex virus turns off the TAP to evade host immunity. Nature 375:411–15 Jones TR, Hanson LK, Sun L, Slater JS, Stenberg RM, Campbell AE. 1995. Multiple independent loci within the human cytomegalovirus unique short region down-regulate expression of major histocompabitility complex class I heavy chains. J. Virol. 69:4830–41 Wiertz EJ, Jones TR, Sun L, Bogyo M, Geuze HJ, Ploegh HL. 1996. The human cytomegalovirus US11 gene product dislocates MHC class I heavy chains from the endoplasmic reticulum to the cytosol. Cell 84:769–79 Tay CH, Welsh RM. 1997. Distinct organdependent mechanisms for the control of murine cytomegalovirus infection by natural killer cells. J. Virol. 71:267–75 Ruzek MC, Miller AH, Opal SM, Pearce BD, Biron CA. 1997. Characterization of early cytokine responses and an IL6-dependent pathway of endogenous glucocorticoid induction during murine cytomegalovirus infection. J. Exp. Med. 185:1185–92 Biron CA, Orange JS. 1995. IL-12 in acute viral infectious disease. Res. Immunol. 146:590–600 Lucin P, Jonjic S, Messerle M, Polic B, Hengel H, Koszinowski UH. 1994. Late phase inhibition of murine cytomegalovirus replication by synergistic action of interferon-γ and tumour necrosis factor. J. Gen. Virol. 75:101–10 MacMicking J, Xie QW, Nathan C. 1997. Nitric oxide and macrophage function. Annu. Rev. Immunol. 15:323–50 Heise MT, Virgin HW IV. 1995. The T cell independent role of γ interferon and tumor necrosis factor α in macrophage activation during murine cytomegalovirus and herpes simplex virus infections. J. Virol. 69:904–9 Karupiah G, Xie QW, Buller RM, Nathan C, Duarte C, MacMicking JD. 1993. Inhibition of viral replication by interferonγ -induced nitric oxide synthase. Science 261:1445–48 Reiss CS, Komatsu T. 1998. Does nitric oxide play a critical role in viral infections? J. Virol. 72:4547–51 Oppenheim JJ, Zachariae COC, Mukaida

123. 124.

125.

126. 127.

128.

129.

130.

131.

132.

N, Matsushima K. 1991. Properties of the novel proinflammatory supergene “intercrine” cytokine family. Annu. Rev. Immunol. 9:617–48 Berger EA. 1997. HIV entry and tropism: the chemokine receptor connection. AIDS 11:S3–S16 Cocchi F, DeVico AL, Garzino-Demo A, Arya SK, Gallo RC, Lusso P. 1995. Identification of RANTES, MIP-1α, and MIP1β as the major HIV-suppressive factors produced by CD8+ T cells. Science 270:1811–15 Alkhatib G, Combadiere C, Broder CC, Feng Y, Kennedy PE, Murphy PM, Berger EA. 1996. CC-CKR5: a RANTES, MIP1α, MIP-1β receptor as a fusion cofactor for macrophage-tropic HIV-1. Science 272:1955–58 Moore JP, Trkola A, Dragic T. 1997. Coreceptors for HIV-1 entry. Curr. Opin. Immunol. 9:551–62 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 Scharton-Kersten T, Afonso LCC, Wysocka M, Trinchieri G, Scott P. 1995. IL-12 is required for natural killer cell activation and subsequent T helper 1 cell development in experimental leishmaniasis. J. Immunol. 154:5320–30 Tripp CS, Wolf SF, Unanue ER. 1993. Interleukin 12 and tumor necrosis factor α are costimulators of interferon γ 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 Gazzinelli RT, Hieny S, Wynn T, Wolf S, Sher A. 1993. IL-12 is required for the T-lymphocyte-independent induction of IFN-γ by an intracellular parasite and induces resistance in T-cell-deficient hosts. Proc. Natl. Acad. Sci. USA 90:6115–19 Hunter CA, Subauste CS, Van Cleave VH, Remington JS. 1994. Production of γ interferon by natural killer cells from Toxoplasma gondii-infected SCID mice: regulation by interleukin-10, interleukin-12, and tumor necrosis factor α. Infect. Immun. 62:2818–24 Hunter CA, Timans J, Pisacane P, Menon S, Cai G, Walker W, Aste-Amezaga M, Chizzonite R, Bazan JF, Kastelein RA. 1997. Comparison of the effects of interleukin-1α, interleukin-1β, and interferon-γ -inducing factor on the production of interferon-γ by natural killer. Eur. J. Immunol. 27:2787–92

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NK CELLS IN ANTIVIRAL DEFENSE 133. Doherty TM, Seder RA, Sher A. 1996. Induction and regulation of IL-15 expression in murine macrophages. J. Immunol. 156:735–41 134. Tagaya Y, Bamford RN, DeFilippis AP, Waldmann TA. 1996. IL-15: a pleiotropic cytokine with diverse receptor/signaling pathways whose expression is controlled at multiple levels. Immunity 4:329– 36 135. Warren HS, Kinnear BF, Kastelein RL, Lanier LL. 1996. Analysis of the costimulatory role of IL-2 and IL-15 in initiating proliferation of resting (CD56dim) human NK cells. J. Immunol. 156:3254– 59 136. Carson WE, Giri JG, Lindemann MJ, Linett ML, Ahdieh M, Paxton R, Anderson D, Eisenmann J, Grabstein K, Caligiuri MA. 1994. Interleukin (IL) 15 is a novel cytokine that activates human natural killer cells via components of the IL-2 receptor. J. Exp. Med. 180:1395– 403 137. Carson WE, Ross ME, Baiocchi RA, Marien MJ, Boiani N, Grabstein K, Caligiuri MA. 1995. Endogenous production of interleukin 15 by activated human monocytes is critical for optimal production of interferon-γ by natural killer cells in vitro. J. Clin. Invest. 96:2578–82 138. Puzanov IJ, Bennett M, Kumar V. 1996. IL-15 can substitute for the marrow microenvironment in the differentiation of natural killer cells. J. Immunol. 157:4282–85 139. Williams NS, Moore TA, Schatzle JD, Puzanov IJ, Sivakumar PV, Zlotnik A, Bennett M, Kumar V. 1997. Generation of lytic natural killer 1.1+, Ly-49− cells from multipotential murine bone marrow progenitors in a stroma-free culture: definition of cytokine requirements and developmental intermediates. J. Exp. Med. 186:1609–14 140. Williams NS, Klem J, Puzanov IJ, Sivakumar PV, Schatzle JD, Bennett M, Kumar V. 1998. Natural killer cell differentiation: insights from knockout and transgenic mouse models and in vitro systems. Immunol. Rev. In press 141. Warren HS, Kinnear BF, Kastelein RL, Lanier LL. 1996. Analysis of the costimulatory role of IL-2 and IL-15 in initiating proliferation of resting (CD56dim) human NK cells. J. Immunol. 156:3254–59 142. Okamura H, Nagata K, Komatsu T, Tanimoto T, Nukata Y, Tanabe F, Akita K, Torigoe K, Okura T, Fukuda S, Kurimoto M. 1995. A novel costimulatory factor for γ interferon induction found in the livers

143.

144.

145.

146.

147.

148.

149.

150.

151.

152.

215

of mice causes endotoxic shock. Infect. Immun. 63:3966–72 Takeda K, Tsutsui H, Yoshimoto T, Adachi O, Yoshida N, Kishimoto T, Okamura H, Nakanishi K, Akira S. 1998. Defective NK cell activity and Th1 response in IL-18-deficient mice. Immunity 8:383– 90 Nakamura K, Okamura H, Nagata K, Komatsu T, Tamura T. 1993. Purification of a factor which provides a costimulatory signal for γ interferon production. Infect. Immun. 61:64–70 D’Andrea A, Aste-Amezaga M, Valiante NM, Ma X, Kubin M, Trinchieri G. 1993. Interleukin 10 (IL-10) inhibits human lymphocyte interferon γ -production by suppressing natural killer cell stimulatory factor/IL-12 synthesis in accessory cells. J. Exp. Med. 178:1041–48 Snijders A, Hilkens CM, van der Pouw Kraan TC, Engel M, Aarden LA, Kapsenberg ML. 1996. Regulation of bioactive IL-12 production in lipopolysaccharidestimulated human monocytes is determined by the expression of the p35 subunit. J. Immunol. 156:1207–12 Rook AH, Kehrl JH, Wakefield LM, Roberts AB, Sporn MB, Burlington DB, Lane HC, Fauci AS. 1986. Effects of transforming growth factor β on the functions of natural killer cells: depressed cytolytic activity and blunting of interferon responsiveness. J. Immunol. 136:3916–20 Ortaldo Jr, Mason LH, O’Shea JJ, Smyth LA, Falk MJ, Kennedy ICS, Longo DL, Ruscetti RW. 1991. Mechanistic studies of transforming growth factor-β inhibition of IL-2-dependent activation of CD3− large granular lymphocyte functions: regulation of IL-2Rβ (p75) signal transduction. J. Immunol. 146:3791–98 Su HC, Leite-Morris KA, Braun L, Biron CA. 1991. A role for transforming growth factor-β1 in regulating natural killer cell and T lymphocyte proliferative responses during acute infection with lymphocytic choriomeningitis virus. J. Immunol. 147:2717–27 Bellone G, Aste-Amezaga M, Trinchieri G, Rodeck U. 1995. Regulation of NK cell functions by TGF-β1. J. Immunol. 155:1066–73 Maghazachi AA, Al-Aoukaty A, Schall TJ. 1994. C-C chemokines induce the chemotaxis of NK and IL-2-activated NK cells. J. Immunol. 153:4969–77 Taub DD, Sayers TJ, Carter CRD, Ortaldo JR. 1995. α and β chemokines induce NK cell migration and enhance NK-mediated cytolysis. J. Immunol. 155:3877–88

P1: SAT/ary

P2: NBL/plb

January 25, 1999

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

216

12:16

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL

153. Loetscher P, Seitz M, Clark-Lewis I, Baggiolini M, Moser B. 1996. Activation of NK cells by CC chemokines: chemotaxis, Ca+ mobilization, and enzyme release. J. Immunol. 156:322–27 154. Vilcek J, Sen GC. 1996. Interferons and other cytokines. In Fundamental Virology, ed. BN Fields, DM Knipe, PM Howley. 11:341–65. Philadelphia/New York: Lippincott-Raven. 1340 pp. 3rd ed. 155. Pfeffer LM, Dinarello CA, Herberman RB, Williams BR, Borden EC, Bordens R, Walter MR, Nagabhushan TL, Trotta PP, Pestka S. 1998. Biological properties of recombinant α-interferons: 40th anniversary of the discovery of interferons. Cancer Res. 58:2489–99 156. Cousens LP, Orange JS, Su HC, Biron CA. 1997. Interferon-α/β inhibition of interleukin 12 and interferon-γ production in vitro and endogenously during viral infection. Proc. Natl. Acad. Sci. USA 94:634–39 157. McRae BL, Semnani RT, Hayes MP, van Seventer GA. 1998. Type I IFNs inhibit human dendritic cell IL-12 production and Th1 cell development. J. Immunol. 160:4298–304 158. Lindahl P, Gresser I, Leary P, Tovey M. 1976. Interferon treatment of mice. Enhanced expression of histocompatibility antigens on lymphoid cells. Proc. Natl. Acad. Sci. USA 73:1284–87 159. Dunnick JK, Galasso GJ. 1980. Update on clinical trials with exogenous interferon. J. Infect. Dis. 142:293–99 160. Gresser I, Guy-Grand D, Maury C, Maunoury MT. 1981. Interferon induces peripheral lymphadenopathy in mice. J. Immunol. 127:1569–75 161. Korngold R, Blank KJ, Murasko DM. 1983. Effect of interferon on thoracic duct lymphocyte output: induction with either polyI:C or vaccinia virus. J. Immunol. 130:2236–40 162. Ishikawa R, Biron CA. 1993. IFN induction and associated changes in splenic leukocyte distribution. J. Immunol. 150: 3713–27 163. Tough DF, Borrow P, Sprent J. 1996. Induction of bystander T cell proliferation by viruses and type I interferon in vivo. Science 272:1947–50 164. Rogge L, Barberis-Maino, Biffi M, Passini N, Presky DH, Gubler U, Sinigaglia F. 1997. Selective expression of an interleukin-12 receptor component by human T helper 1 cells. J. Exp. Med. 185:825–31 165. Feldman SB, Ferraro M, Zheng H-M, Patel N, Gould-Fogerite S, Fitzgerald-

166.

167.

168. 169.

170.

171.

172.

173.

174.

175.

176.

177.

Bocarsly P. 1994. Viral induction of low frequency interferon-α producing cells. Virology 201:1–7 Fitzgerald-Bocarsly P, Howell DM, Pettera L, Tehrani S, Lopez C. 1991. Immediate-early gene expression is sufficient for induction of natural killer cell-mediated lysis of herpes simplex virus type 1-infected fibroblasts. J. Virol. 65:3151–60 Lebon P. 1985. Inhibition of herpes simplex virus type-1 induced interferon synthesis by monoclonal antibodies against viral glycoprotein D and by lysosomotropic drugs. J. Gen. Virol. 66:2781– 86 Darnell JE Jr. 1997. STATS and gene regulation. Science 277:1630–35 Schindler C, Darnell JE Jr. 1995. Transcriptional responses to polypeptide ligands: the JAK-STAT pathway. Annu. Rev. Biochem. 64:621–51 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 Silvennoinen O, Ihle JN, Schlessinger J, Levy DE. 1993. Interferon-induced nuclear signaling by Jak protein tyrosine kinases. Nature 366:583–85 Shual K, Ziemiecki A, Wilks AF, Harpur AG, Sadowski HB, Gilman MZ, Darnell JE Jr. 1993. Polypeptide signaling to the nucleus through tyrosine phosphorylation of Jak and Stat proteins. Nature 366:580– 83 Velazquez L, Fellous M, Stark GR, Pellegrini S. 1992. A protein tyrosine kinase in the interferon α/β signaling pathway. Cell 70:313–22 Schindler C, Shuai K, Prezioso VR, Darnell JE Jr. 1992. Interferon-dependent tyrosine phosphorylation of a latent cytoplasmic transcription factor. Science 257:809–13 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 Meraz MA, White JM, Sheehan KC, Bach EA, Rodig SJ, Dighe AS, Kaplan DH, Riley JK, Greenlund AC, Campbell D, Carver-Moore K, DuBois RN, Clark R, Aguet M, Schreiber RD. 1996. Targeted disruption of the Stat1 gene in mice reveals unexpected physiologic specificity in the JAK-STAT signaling pathway. Cell 84:431–42 Durbin JE, Hackenmiller R, Simon MC,

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

NK CELLS IN ANTIVIRAL DEFENSE

178.

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

179.

180.

181.

182.

183.

184.

185.

186.

187.

Levy DE. 1996. Targeted disruption of the mouse Stat1 gene results in compromised innate immunity to viral disease. Cell 84:443–50 Beadling C, Guschin D, Witthuhn BA, Ziemiecki A, Ihle JN, Kerr IM, Cantrell DA. 1994. Activation of JAK kinases and STAT proteins by interleukin-2 and interferon α, but not the T cell antigen receptor, in human T lymphocytes. EMBO J. 13:5605–15 Cho SS, Bacon CM, Sudarshan C, Rees RC, Finbloom D, Pine R, O’Shea JJ. 1996. Activation of STAT4 by IL-12 and IFN-α: evidence for the involvement of ligandinduced tyrosine and serine phosphorylation. J. Immunol. 157:4781–89 Li X, Leung S, Qureshi S, Darnell JE Jr, Stark GR. 1996. Formation of STAT1STAT2 heterodimers and their role in the activation of IRF-1 gene transcription by interferon-α. J. Biol. Chem. 271:5790–94 Stancato LF, Sakatsume M, David M, Dent P, Dong F, Petricoin EF, Krolewski JJ, Silvennoinen O, Saharinen P, Pierce J, Marshall CJ, Sturgill T, Finbloom DS, Larner AC. 1997. Beta interferon and oncostatin M activate Raf-1 and mitogen-activated protein kinase through a JAK1-dependent pathway. Mol. Cell Biol. 17:3833–40 David M, Petricoin E 3rd, Benjamin C, Pine R, Weber MJ, Larner AC. 1995. Requirement for MAP kinase (ERK2) activity in interferon α- and interferon βstimulated gene expression through STAT proteins. Science 269:1721–23 Platanias LC, Sweet ME. 1994. Interferon α induces rapid tyrosine phosphorylation of the vav proto-oncogene product in hematopoietic cells. J. Biol. Chem. 269:3143–46 Crespo P, Schuebel KE, Ostrom AA, Gutkind JS, Bustelo XR. 1997. Phosphotyrosine-dependent activation of Rac1 GDP/GTP exchange by the vav protooncogene product. Nature 385:169–72 Duncan GS, Mittrucker H-W, Kagi D, Matsuyama T, Mak TW. 1996. The transcription factor interferon regulatory factor-1 is essential for natural killer cell function in vivo. J. Exp. Med. 184:2043– 48 Coutelier JP, Van Broeck J, Wolf SF. 1995. Interleukin-12 gene expression after viral infection in the mouse. J. Virol. 69:1955– 58 Kanangat S, Thomas J, Gangappa S, Babu JS, Rouse BT. 1996. Herpes simplex virus type 1-mediated up-regulation of IL-12 (p40) mRNA expression. Implications in

188.

189.

190.

191.

192.

193.

194.

195.

196.

197.

198.

217

immunopathogenesis and protection. J. Immunol. 156:1110–16 Schijns VECJ, Wierda CMH, van Hoeij M, Horzinek MC. 1996. Exacerbated viral hepatitis in IFN-γ receptor-deficient mice is not suppressed by IL-12. J. Immunol. 157:815–21 Cousens LP, Orange JS, Biron CA. 1995. Endogenous IL-2 contributes to T cell expansion and IFN-γ production during lymphocytic choriomeningitis virus infection. J. Immunol. 155:5690–99 Schijns VECJ, Haagmans BL, Wierda CMH, Kruithof B, Heijnen IAFM, Alber G, Horzinek MC. 1998. Mice lacking IL12 develop polarized Th1 cells during viral infection. J. Immunol. 160:3958–64 Orange JS, Wolf SF, Biron CA. 1994. Effects of IL-12 on the response and susceptibility to experimental viral infections. J. Immunol. 152:1253–64 Orange JS, Salazar-Mather TP, Opal SM, Spencer RL, Miller AH, McEwen BS, Biron CA. 1995. Mechanism of interleukin 12-mediated toxicities during experimental viral infections: role of tumor necrosis factor and glucocorticoids. J. Exp. Med. 181:901–14 Trinchieri G. 1995. Interleukin-12: a proinflammatory cytokine with immunoregulatory functions that bridge innate resistance and antigen-specific adaptive immunity. Annu. Rev. Immunol. 13:251– 76 Sousa CR, Hieny S, Scharton-Kersten T, Jankovic D, Charest H, Germain RN, Sher A. 1997. In vivo microbial stimulation induces rapid CD40 ligand-independent production of interleukin 12 by dendritic cells and their redistribution to T cell areas. J. Exp. Med. 186:1819–29 Chua AO, Chizzonite R, Desai BB, Truitt TP, Nunes P, Minetti LJ, Warrier RR, Presky DH, Levine JF, Gately MK, Gubler U. 1994. Expression cloning of a human IL-12 receptor component. A new member of the cytokine receptor superfamily with strong homology to gp130. J. Immunol. 153:128–36 Presky DH, Yang H, Minetti LJ, Chua AO, Nabavi N, Wu CY, Gately MK, Gubler U. 1996. A functional interleukin 12 receptor complex is composed of two β-type cytokine receptor subunits. Proc. Natl. Acad. Sci. USA 93:14002–7 Szabo SJ, Dighe AS, Gubler U, Murphy KM. 1997. Regulation of the interleukin (IL)-12R β 2 subunit expression in developing T helper 1 (Th1) and Th2 cells. J. Exp. Med. 185:817–24 Gollob JA, Kawasaki H, Ritz J. 1997.

P1: SAT/ary

P2: NBL/plb

January 25, 1999

218

199.

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

200.

201.

202.

203.

204.

205.

206.

207.

12:16

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL Interferon-γ and interleukin-4 regulate T cell interleukin-12 responsiveness through the differential modulation of high-affinity interleukin-12 receptor expression. Eur. J. Immunol. 27:647–52 Bacon CM, McVicar DW, Ortaldo Jr, Rees RC, O’Shea JJ, Johnston JA. 1995. Interleukin 12 (IL-12) induces tyrosine phosphorylation of JAK2 and TYK2: differential use of Janus family tyrosine kinases by IL-2 and IL-12. J. Exp. Med. 181:399–404 Bacon CM, Petricoin EF 3rd, Ortaldo Jr, Rees RC, Larner AC, Johnston JA, O’Shea JJ. 1995. Interleukin 12 induces tyrosine phosphorylation and activation of STAT4 in human lymphocytes. Proc. Natl. Acad. Sci. USA 92:7307–11 Jacobson NG, Szabo SJ, Weber-Nordt RM, Zhong Z, Schreiber RD, Darnell JE Jr, Murphy KM. 1995. Interleukin 12 signaling in T helper type 1 (Th1) cells involves tyrosine phosphorylation of signal transducer and activator of transcription (Stat)3 and Stat4. J. Exp. Med. 181:1755– 62 Thierfelder WE, van Deursen JM, Yamamoto K, Tripp RA, Sarawar SR, Carson RT, Sangster MY, Vignali DA, Doherty PC, Grosveld GC, Ihle JN. 1996. Requirement for Stat4 in interleukin-12mediated responses of natural killer and T cells. Nature 382:171–74 Kaplan MH, Sun YL, Hoey T, Grusby MJ. 1996. Impaired IL-12 responses and enhanced development of Th2 cells in Stat4deficient mice. Nature 382:174–77 Grabstein KH, Eisenman J, Shanebeck K, Rauch C, Srinivasan S, Fung V, Beers C, Richardson J, Schoenborn MA, Ahdieh M, Johnson L, Alderson MR, Watson JD, Anderson D, Giri J. 1994. Cloning of a T cell growth factor that interacts with the β chain of the interleukin-2 receptor. Science 264:965–68 Giri JG, Kumaki M, Ahdieh M, Friend DJ, Loomis K, Shanebeck K, DuBose R, Cosman D, Park LS, Anderson DM. 1995. Identification and cloning of a novel IL-15 binding protein that is structurally related to the α chain of the IL-2 receptor. EMBO J. 14:3654–63 Chae D-W, Nosaka Y, Strom TB, Maslinski W. 1996. Distribution of IL-15 receptor α-chains on human peripheral blood mononuclear cells and effect of immunosuppressive drugs on receptor expression. J. Immunol. 157:2813–19 Bazan JF, Timans JC, Kastelein RA. 1996. A newly defined interleukin-1? Nature 379:591

208. Gu Y, Kuida K, Tsutsui H, Ku G, Hsiao K, Fleming MA, Hayashi N, Higashino K, Okamura H, Nakanishi K, Kurimoto M, Tanimoto T, Flavell RA, Sato V, Harding MW, Livingston DJ, Su MS-S. 1997. Activation of interferon-γ inducing factor mediated by interleukin-1β converting enzyme. Science 275:206–9 209. Ghayur T, Banerjee S, Hugunin M, Butler D, Herzog L, Carter A, Quintal L, Sekut L, Talanian R, Paskind M, Wong W, Kamen R, Tracey D, Allen H. 1997. Caspase-1 processes IFN-γ -inducing factor and regulates LPS-induced IFN-γ production. Nature 386:619–23 210. Torigoe K, Ushio S, Okura T, Kobayashi S, Taniai M, Kunikata T, Murakami T, Osamu S, Kojima H, Fujii M, Ohta T, Ikeda M, Ikegami H, Kurimoto M. 1997. Purification and characterization of the human interleukin-18 receptor. J. Biol. Chem. 272:25737–42 211. Robinson D, Shibuya K, Mui A, Zonin F, Murphy E, Sana T, Hartley SB, Menon S, Kastelein R, Bazan F, O’Garra A. 1997. IGIF does not drive Th1 development but synergizes with IL-12 for interferonγ production and activates IRAK and NFκB. Immunity 7:571–81 212. Pulliam L, Moore D, West DC. 1995. Human cytomegalovirus induces IL-6 and TNF α from macrophages and microglial cells: possible role in neurotoxicity. J. Neurovirol. 1:219–27 213. Gilles PN, Fey G, Chisari FV. 1992. Tumor necrosis factor α negatively regulates hepatitis B virus gene expression in transgenic mice. J. Virol. 66:3955–60 214. Trgovcich J, Ryman K, Extrom P, Eldridge JC, Aronson JF, Johnston RE. 1997. Sindbis virus infection of neonatal mice results in a severe stress response. Virology 227:234–38 215. Moskophidis D, Frei K, Lohler J, Fontana A, Zinkernagel RM. 1991. Production of random classes of immunoglobulins in brain tissue during persistent viral infection paralleled by secretion of interleukin6 (IL-6) but not IL-4, IL-5, and γ interferon. J. Virol. 65:1364–69 216. Kanangat S, Babu JS, Knipe DM, Rouse BT. 1996. HSV-1-mediated modulation of cytokine gene expression in a permissive cell line: selective upregulation of IL-6 gene expression. Virology 219:295– 300 217. Atedzoe BN, Ahmad A, Menezes J. 1997. Enhancement of natural killer cell cytotoxicity by the human herpesvirus-7 via IL-15 induction. J. Immunol. 159:4966– 72

P1: SAT/ary

P2: NBL/plb

January 25, 1999

12:16

QC: NBL

Annual Reviews

AR078-07

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NK CELLS IN ANTIVIRAL DEFENSE 218. Flamand L, Stefanescu I, Menezes J. 1996. Human herpesvirus-6 enhances natural killer cell cytotoxicity via IL-15. J. Clin. Invest. 97:1373–78 219. Sareneva T, Matikainen S, Kurimoto M, Julkunen I. 1998. Influenza A virusinduced IFN-α/β and IL-18 synergistically enhance IFN-γ gene expression in human T cells. J. Immunol. 160:6032–38 220. Mo XY, Sarawar SR, Doherty PC. 1995. Induction of cytokines in mice with parainfluenza pneumonia. J. Virol. 69:1288–91 221. Su HC, Ishikawa R, Biron CA. 1993. Transforming growth factor-β expression and natural killer cell responses during virus infection of normal, nude, and SCID mice. J. Immunol. 151:4874–90 222. Pilaro AM, Taub DD, McCormick KL, Williams HM, Sayers TJ, Fogler WE, Wiltrout RH. 1994. TNF-α is a principal cytokine involved in the recruitment of NK cells to liver parenchyma. J. Immunol. 153:333–42 223. Fogler WE, Volker K, McCormick KL, Watanabe M, Ortaldo Jr, Reynolds CW. 1996. NK cell infiltration into lung, liver, and subcutaneous B16 melanoma is mediated by VCAM-1/VLA-4 interaction. J. Immunol. 156:4707–14 224. Orange JS, Salazar-Mather TP, Opal SM, Biron CA. 1997. Mechanisms for virusinduced liver disease: a TNF-mediated pathology independent of NK and T cells during murine cytomegalovirus infection. J. Virol. 71:9248–58 225. Allavena P, Paganin C, Zhou D, Bianchi G, Sozzani S, Mantovani A. 1994. Interleukin-12 is chemotactic for natural killer cells and stimulates their interaction with vascular endothelium. Blood 84:2261–68 226. Allavena P, Giardina G, Bianchi G, Mantovani A. 1997. IL-15 is chemotactic for natural killer cells and stimulates their adhesion to vascular endothelium. J. Leukocyte Biol. 61:729–35 227. Lukacs NW, Strieter RM, Elner VM, Evanoff HL, Burdick M, Kunkel SL. 1994. Intercellular adhesion molecule-1 mediates the expression of monocytederived MIP-1α during monocyte-endothelial cell interactions. Blood 83:1174– 78 228. Farber JM. 1990. A macrophage mRNA selectively induced by γ -interferon encodes a member of the platelet factor 4 family of cytokines. Proc. Natl. Acad. Sci. USA 87:5238–42 229. Luster AD, Unkeless JC, Ravetch JV. 1985. γ -interferon transcriptionally reg-

230.

231.

232.

233. 234.

235. 236. 237. 238. 239.

240.

241.

242.

219

ulates an early-response gene containing homology to platelet proteins. Nature 315:672–76 Amichay D, Gazzinelli RT, Karupiah G, Moench TR, Sher A, Farber JM. 1996. Genes for chemokines mumig and crg2 are induced in protozoan and viral infections in response in response to IFNγ with patterns of tissue expression that suggest nonredundant roles in vivo. J. Immunol. 157:4511–57 Lane TL, Asensio VC, Yu N, Paoletti AD, Campbell IL, Buchmeier MJ. 1998. Dynamic regulation of a- and b-chemokine expression in the central nervous system during mouse hepatitis virus-induced demyelinating disease. J. Immunol. 160:970–78 Asensio VC, Campbell IL. 1997. Chemokine gene expression in the brains of mice with lymphocytic choriomeningitis. J. Virol. 71:7832–40 Lanier LL. 1998. NK cell receptors. Annu. Rev. Immunol. 16:359–93 Moretta A, Bottino C, Vitale M, Pende D, Biassoni R, Mingari MC, Moretta L. 1996. Receptors for HLA class-I molecules in human natural killer cells. Annu. Rev. Immunol. 14:619–48 Yokoyama WM. 1998. Natural killer cell receptors. Curr. Opin. Immunol. 10:298– 305 Long EO, Wagtmann N. 1997. Natural killer cell receptors. Curr. Opin. Immunol. 9:344–50 Vivier E, Da¨eron M. 1997. Immunoreceptor tyrosine-based inhibition motifs. Immunol. Today 18:286–91 Parham P, ed. 1997. NK cells, MHC class I antigens and missing self. Immunol. Rev. 155:1–221 Allison AC. 1974. Interactions of antibodies, complement components and various cell types in immunity against viruses and pyogenic bacteria. Transplant. Rev. 19:3– 55 Lanier LL, Corliss BC, Wu J, Leong C, Philips JH. 1998. Immunoreceptor DAP12 bearing a tyrosine-based activation motif is involved in activating NK cells. Nature 391:703–7 Smith KM, Wu J, Bakker ABH, Phillips JH, Lanier LL. 1998. Ly-49D and Ly-49H associate with mouse DAP12 and form activating receptors. J. Immunol. 161:7– 10 K¨arre K, Ljunggren HG, Piontek G, Kiessling R. 1986. Selective rejection of H-2 deficient lymphoma variants suggest alternative immune defense strategy. Nature 319:675–78

P1: SAT/ary

P2: NBL/plb

January 25, 1999

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

220

12:16

QC: NBL

Annual Reviews

AR078-07

BIRON ET AL

243. Braud VM, Allan SJ, O’Callaghan CA, S¨oderstr¨om K, D’Andrea A, Ogg GS, Lazetic S, Young NT, Bell JI, Phillips JH, Lanier LL, McMichael AJ. 1998. HLA-E binds to natural killer cell receptors CD94/NKG2A, B, and C. Nature 391:795–99 244. Cosman D, Fanger N, Borges L, Kubin M, Chin W, Peterson L, Hsu ML. 1997. A novel immunoglobulin superfamily receptor for cellular and viral MHC class I molecules. Immunity 7:273–82 245. Farrell HE, Hally H, Lynch DM, Fleming P, Shellam GR, Scalzo AA, DavisPoynter NJ. 1997. Inhibition of natural killer cells by a cytomegalovirus MHC class I homologue in vivo. Nature 386:510–14 246. Reyburn HT, Mandelboim O, Val`esGomez M, Davis DM, Pazmani L, Strominger JL. 1997. The class I MHC homologue of human cytomegalovirus inhibits attack by natural killer cells. Nature 386:514–16 247. Scalzo AA, Lyons PA, Fitzgerald NA, Forbes CA, Yokoyama WM, Shellam GR. 1995. Genetic mapping of Cmv1 in the region of mouse chromosome 6 encoding the NK gene complex-associated loci Ly49 and musNKR-P1. Genomics 27:435–41

248. Ahmed R, Biron CA. 1999. Immunity to viruses. In Fundamental Immunology. ed. WE Paul. New York: Raven. In press. 4th ed. 249. Biron CA. 1998. Interfaces between innate and adaptive immunity: role of interferons in shaping immune responses to viral infections. Semin. Immunol. 10:In press 250. Kos FJ, Engleman EG. 1996. Role of natural killer cells in the generation of influenza virus-specific cytotoxic T cells. Cell. Immunol. 173:1–6 251. Wang B, Nguyen K, Zhang X, Stamm L, David J, Satoskar A, Biron C. 1998. Use of a novel natural killer cell deficiency mouse model to evaluate role for NK cells in shaping adoptive immune responses during parasitic and viral infections. FASEB J. 12:A904 (Abstr.) 252. Huang S, Hendriks W, Althage A, Hemmi S, Bluethmann H, Kamijo R, Vilcek J, Zinkernagel RM, Aguet M. 1993. Immune response in mice that lack the interferon-γ receptor. Science 259:1742– 45 253. Graham MB, Dalton DK, Giltinan D, Braciale VL, Stewart TA, Braciale TJ. 1993. Response to influenza infection in mice with a targeted disruption in the interferon γ gene. J. Exp. Med. 178:1725–32

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:189-220. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:221–53

MATURE T LYMPHOCYTE APOPTOSIS—Immune Regulation in a Dynamic and Unpredictable Antigenic Environment1 Michael Lenardo, Francis Ka-Ming Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, and Lixin Zheng Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland 20892; e-mail: [email protected] KEY WORDS:

death, cytokine, lymphokine, Fas/APO-1/CD95, Bcl-2, propriocidal, feedback, caspase, tumor necrosis factor, receptor, Autoimmune Lymphoproliferative Syndrome

ABSTRACT Apoptosis of mature T lymphocytes preserves peripheral homeostasis and tolerance by countering the profound changes in the number and types of T cells stimulated by diverse antigens. T cell apoptosis occurs in at least two major forms: antigen-driven and lymphokine withdrawal. These forms of death are controlled in response to local levels of IL-2 and antigen in a feedback mechanism termed propriocidal regulation. Active antigen-driven death is mediated by the expression of death cytokines such as FasL and TNF. These death cytokines engage specific receptors that assemble caspase-activating protein complexes. These signaling complexes tightly regulate cell death but are vulnerable to inherited defects. Passive lymphokine withdrawal death may result from the cytoplasmic activation of caspases that is regulated by mitochondria and the Bcl-2 protein. The human disease, Autoimmune Lymphoproliferative Syndrome (ALPS) is due to dominant-interfering mutations in the Fas/APO-1/CD95 receptor and other components of the death pathway. The study of ALPS patients reveals the necessity of apoptosis for preventing autoimmunity and allows the genetic investigation of apoptosis in humans. Immunological, cellular, and molecular evidence indicates that throughout the life of a T cell, apoptosis may be evoked in 1 The

US government has right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

221

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

222

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL

excessive, harmful, or useless clonotypes to preserve a healthy and balanced immune system.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

For effective action...it is not only essential that we possess good effectors, but that the performance of these effectors be properly monitored back and that the readings of these monitors be properly combined with other information coming in...to produce a properly proportioned output to the effectors. N Wiener (1961)

INTRODUCTION Immune responses often involve the dramatic expansion of specifically reactive T cells with potent and potentially toxic effector functions (1–3). The doubling time of a T cell after antigen stimulation can be as little as 4.5 hours; consequently, in a week’s time, a single T cell could multiply to almost 1 × 1012 cells and thereby double the total number of T cells in a human being (4). However, an average protein containing 2 to 10 epitopes could be recognized by as many as 10 to 1000 naive cells depending on the major histocompatibility alleles. During certain viral infections, the total number of virus-specific CD8 cells in a mouse can increase as much as tenfold (5). Thymic output can average 1% of the peripheral pool each day early in life (6). Countering these influxes of T cells is programmed cell death, or apoptosis. A balanced economy of cell production and destruction is important because the lymphocyte compartment has a limited capacity (7). Stretching that capacity has deleterious effects. For example, hypersplenism produces blood cell destruction, anemia, and thrombocytopenia. Moreover, it is generally held that maximal response capability requires that the lymphoid compartment should not be filled with a predominance of one or a few clonotypes. In addition, expanded clonotypes that have substantial autoreactivity must be downsized or eliminated. Therefore, T cell number is regulated to meet the triple requirements of containment, diversity and tolerance.

IMMUNE PHYSIOLOGY OF MATURE T CELL APOPTOSIS: PROPRIOCIDAL REGULATION It is uncertain whether T lymphocytes, like other formed blood elements, have a predetermined life span. Nevertheless, IL-2 and antigen can evoke apoptosis under specific conditions. Mature post-thymic T cells initially resist apoptosis but can become highly susceptible to apoptosis when cell cycling is initiated in response to antigen (8–10). To understand T cell apoptosis, it is important to recognize that the T cell response to antigen occurs in two contingent phases

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

MATURE T LYMPHOCYTE APOPTOSIS

223

comprising distinct molecular events: an activation phase and a subsequent proliferation phase (11). Key-activation-phase events are the induction of the genes for IL-2, and its high affinity receptor. Contrary to the oft-used moniker, “activation-induced cell death,” there is little or no apoptosis caused by TCR engagement during the initial activation phase of a resting T cell (9, 10, 12–15). During initial activation, apoptosis is avoided to allow protective immune responses to develop. The proliferative phase of the T cell response takes place when IL-2 engages its receptor and initiates cell cycle progression. Once T cells have gone through one, and perhaps several, cell cycles and enter late G1 or S phase, they become exquisitely susceptible to apoptosis (8, 16–20) (Figure 1). This susceptibility explains why T hybridomas, perpetually cycling cells, die in response to TCR cross-linking. Although T hybridomas have been thought to mimic thymocyte deletion, their death resembles that of peripheral T cells at the molecular level (21–25). Cell cycle progression and apoptosis susceptibility both involve the c-myc proto-oncogene, but not p53 (26–28). Although certain T cell clones show marked differences in susceptibility, the ability of cycling lymph node T cells to antigen-induced apoptosis is generally shared by all the major T cell subsets including: αβ, γ δ, CD4, CD8, TH1, and TH2, as well as by their relatives, natural killer cells (13, 29–33). The susceptibility of proliferating T cells to apoptosis positions IL-2 as the key regulator of T cell apoptosis (Figure 1). This new regulatory effect of IL-2 emerged from studies showing antigen-induced apoptosis in vitro and superantigen-induced death in vivo was strongly promoted by IL-2 (9). Furthermore, IL-2 and IL-2 receptor knockout mice that show that deficiencies in IL-2 signaling were paradoxically found to have the abnormal accumulation of activated T cells associated with defective TCR-induced apoptosis (34–39). The concept of propriocidal regulation, or feedback control by apoptosis, was borne out of the need to explain this new role of IL-2 that conflicted with its known proliferative effects (8, 13, 40, 41). According to this theory, IL-2 provides feedback susceptibility to apoptosis and the level of antigen stimulation determines whether apoptosis actually occurs. The fate of cycling T cells is thus linked to the prevailing conditions of the immune response. If there is no further antigen stimulation, the expression of IL-2 and its receptor falls and passive or “lymphokine withdrawal” apoptosis ensues. Passive apoptosis thereby decreases the expanded population of T cells at the conclusion of an immune response. In contrast, if cycling T cells are strongly stimulated by antigen, active or “antigen-induced” apoptosis occurs. Death caused by TCR re-engagement restrains T cell expansion when the antigen is continuously or repeatedly encountered. Even though the cycling T cells are destined to die after strong antigen re-engagement, effector functions, such as lymphokine production or

P2: NBL/SPD

February 11, 1999

224

11:32

QC: PSA/UKS

Annual Reviews

Figure 1 The propriocidal or feedback-response paradigm of mature T lymphocyte apoptosis (9, 40, 258, 344). The T cell response to antigen occurs in two molecularly distinct phases: activation, which leads to the production of IL-2; and proliferation, which is due to cell cycle progression caused by IL-2 (11). Cycling T cells become highly susceptible to apoptosis and whether death ensues depends on the environmental conditions. Active apoptosis occurs if strong secondary TCR engagement is encountered. Passive apoptosis occurs after cessation of antigen and IL-2 stimulation. A small number of cells escape both death pathways and these are believed to become the “memory” population.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

P1: KKK/PLB T1: PSA

AR078-08

LENARDO ET AL

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

MATURE T LYMPHOCYTE APOPTOSIS

225

cytotoxicity, are still potently expressed (37, 42, 43). Antigen-stimulated death differs from activation in that it is generated solely by TCR engagement; costimulatory molecules such as CD28, that powerfully promote the activation of resting cells, have no effect on antigen-induced death (14, 44). The active and passive forms of apoptosis have molecular differences. Passive apoptosis requires new protein synthesis, is strongly inhibited by Bcl-2 and related molecules, and may involve mitochondrial apoptosis mechanisms rather than death cytokines. Active apoptosis requires TCR stimulation, and involves death cytokines such as FasL and TNF, is independent of new protein synthesis after death, and is inefficiently inhibited by Bcl-2. Passive apoptosis can be prevented by several T cell growth cytokines that all use the common gamma chain IL-2, IL-4, IL-7, and IL-15; the protection can occur even without cell cycle induction and is unaffected by the lpr mutation (45–48) (L Zheng & M Lenardo, unpublished results). Active apoptosis can occur at concentrations of these cytokines that cause proliferation, but IL-2 causes the greatest proliferation and susceptibility to active apoptosis (8, 47). The accumulation of activated T cells in IL-2- and IL-2R-deficient mice may be due to the stimulatory effects of other T cell growth cytokines unopposed by the propriocidal effect of IL-2 (47, 49–51). As a result of these two forms of apoptosis, the feedback response eliminates T cells if there is too much or too little Ag and IL-2. Balanced between the extremes of IL-2 and antigen, active immune responses take place with varying amounts of T cell proliferation and death. A fraction of cycling T cells may escape both active and passive apoptosis and become a long-lived pool of “memory” T cells (5). The need for the propriocidal mechanism is perhaps dictated by the stochastic nature of the immune response. A nearly random set of T cell clonotypes will encounter antigen in unpredictable ways. Cybernetic analysis of complex systems (systems for which the final outcome cannot be predicted from the initial conditions) has determined that feedback control is essential for regulated responses (52, 53). Feedback control consists of two main components: the ability to “sense” present conditions and the capacity to respond. In the immune system, “sensing” of the environment occurs via the receptors for IL-2 and antigen. The “response” is either survival or death. An apoptosis response constitutes negative feedback, which has an intrinsic tendency to stabilize a complex system (52). Propriocidal regulation contrasts with previous concepts of immune autoregulation because the servomechanism is “built in” to each T cell and is governed by molecular feedback rather than regulatory lymphocyte interactions (54, 55). This new regulatory paradigm potentially explains the observation made in many different systems that repeated or chronic antigenic stimulation, which is now known to be a potent stimulus for apoptosis, causes “suppression” or “exhaustion” rather than a better anamnestic response

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

226

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL

(13, 46, 56–59). Also, passive IL-2 withdrawal could account for the striking T cell deletion that re-equilibrates the T cell repertoire after a strong immune response (58, 60–67). Feedback control allows the delivery of potent but homeostatically regulated responses in the face of numerous antigens in a constantly changing and unpredictable environment.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

TWO PATHWAYS OF ACTIVATED T CELL APOPTOSIS Antigen-Induced (Active) Apoptosis is Due to Fas Ligand and TNF Active T cell apoptosis takes place indirectly by the antigen-induced expression of death cytokines, chiefly FasL/APO-1L and tumor necrosis factor (TNF) (24, 50, 68–74). Additional homologues of TNF have been detected in activated T cells, and the role of these homologues in T cell death requires further examination (75, 76). Marked T cell apoptosis defects, lymphoproliferation, and autoimmunity occur in certain strains of mice homozygous for the gld and lpr alleles, which are genetic defects in FasL and Fas, respectively (68, 69, 72, 77–79). These mouse strains were originally employed as models of systemic lupus erythematosus, but manifested much greater lymphoproliferation than is typical of this disease in humans (80). The dramatic lymphoid expansion in lpr and gld mice illustrates the importance of apoptosis for T cell regulation (50, 68, 69, 81, 82). Fas-deficient T cells exhibit reduced but clearly evident TCR-induced death, and the residual apoptosis is blocked by inhibiting TNF (50, 83). Although no obvious T cell phenotype has been uncovered in TNF or TNFR knockout mice, TCR-induced T cell deletion in vitro and in vivo can be inhibited by blocking TNF (42, 84–87). In resting cells, the genes for FasL and TNF are weakly induced by TCR stimulation, but in IL-2 stimulated T cells these death cytokines are powerfully induced (37). This difference is part of the reason why TCR engagement kills cycling but not resting T cells and thereby restricts the feedback death effect to when it is needed to control T cell expansion. Both FasL and TNF are found in cell surface-anchored forms, raising the issue of whether a two-cell interaction is needed for death. However, both molecules can be readily cleaved from the membrane by metalloproteinases and evidence supports the idea that a single T cell can kill itself (23, 88–90). In contrast, B cell killing through Fas is executed only by T cells expressing FasL because B cells do not express FasL (91, 92). This observation may provide the raison d’ˆetre for the indirect mechanism of self-killing in T lymphocytes. Expression of the death ligands by T cells allows them to exert control over the fate of other cells involved in the immune response and to carry out cytotoxic effector functions (93, 94).

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

MATURE T LYMPHOCYTE APOPTOSIS

227

FasL and TNF wield their effects by interacting with cognate cell surface receptors whose expression is induced by TCR signals: Fas/APO-1/CD95 and TNFRs, type 1 and type 2 (86, 95–100). These receptors are members of a growing TNFR superfamily; members of this superfamily share cysteine-rich extracellular domains that are important for ligand binding (97, 99, 101–103). FasL induction depends on the TCR signaling apparatus and a series of regulatory proteins including nur77, TDAG51, and NF-IL6, among others (25, 104–106). However, Fas is constitutively present on circulating memory T cells that increase in number during aging (107). The type 2 TNFR (p75) is expressed after activation on mature T cells, whereas the surface expression of the type 1 TNFR (p55) is difficult to detect and may not be obligatory for T cell apoptosis (50). The crystal structure of TNF bound to its receptor reveals a characteristic homotrimeric assembly with a threefold axis of symmetry perpendicular to the membrane (108). This quaternary structure is apparently conserved for Fas and other members of the TNFR superfamily and directs signal transmission (97, 99, 109–111). Several apoptosis-inducing members of the TNFR superfamily—Fas, p55 TNFR, DR3, DR4, and DR5—share a conserved region in their cytoplasmic tail called the “death domain” (112–120). This domain is necessary and sufficient for apoptosis and recruits cytoplasmic signaling proteins after receptor aggregation (112, 113, 121, 122). Interestingly, the p75 TNFR contains in its cytoplasmic tail a signaling structure that binds a group of proteins containing “TRAF domains;” the presence of this differing structure may betoken a distinct means of influencing apoptosis (123–129) (Figure 2). Apoptosis is more rapidly induced with FasL than with TNF (50, 130). In vitro studies suggest that Fas and TNF preferentially control the death of CD4+ and CD8+ T cells, respectively, but there are exceptions to this distinction (85, 131). In the next section we describe only the main apoptosis signal pathway for Fas. We expect signal transduction from the other death domain-containing receptors to follow similar principles (132–136) (Figure 2). Engagement by FasL initiates a death signal by causing the aggregation of specific cytoplasmic signaling proteins on the death domain of Fas (137–142). The key components of the death signaling complex are FADD/Mort1, an adaptor protein, and the proenzyme form of Caspase-8/Mach/caspase-8, a protease that induces apoptosis when activated (138, 143–146). FADD and pro-caspase8 pre-exist in the cytoplasm and only come together on the cytoplasmic tail of Fas after FasL engagement. This tripartite complex is required for death and depends on a homotypic interaction between death domains in Fas and FADD as well as between “death effector domains” in FADD and caspase-8 (147–150). The death domain and the DED share a similar overall tertiary structure comprising a characteristic nest of 6 alpha helices, but there is no evidence that they can interact in a heterotypic fashion (151, 152). Death signaling is independent

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

P1: KKK/PLB P2: NBL/SPD

February 11, 1999

228

11:32

QC: PSA/UKS

LENARDO ET AL

T1: PSA

Annual Reviews AR078-08

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

MATURE T LYMPHOCYTE APOPTOSIS

229

of new protein synthesis; in fact, inhibitors of transcription or translation enhance apoptosis perhaps by removing labile inhibitors (153, 154). The fact that it can be difficult to measure the tripartite complex in certain cell types may indicate technical limitations or the presence of an alternative Fas signal mechanism (137, L Zheng & M Lenardo, unpublished observations). Formation of this complex seals the fate the T cell within an hour or two and the mortally damaged T cell disintegrates in the next few hours (155–158). Hence, Fas renders a rapid feedback response and reverses T cell expansion by directly eliminating T cells. The mechanism of death after the aggregation of FADD and caspase-8 involves caspase-8 processing and proteolytic release of the active enzyme. Active caspase-8 initiates a chain of lethal proteolytic events including the activation of other caspases (139). Caspases are cystinyl-aspartate-requiring proteinases that cleave specific cellular substrates to generate the cytoskeletal and chromatin changes of apoptosis (159–162). Caspase activation is necessary and sufficient for apoptosis because specific caspase inhibitors block Fasinduced T cell apoptosis, and the expression of active caspase-8 alone is lethal (84, 163–167). Although a cascade of caspase-1 (Interleukin-1ß-converting enzyme or ICE) and then caspase 3 was proposed for Fas death, recent evidence suggests that neither is required and that a caspase-3-related enzyme is ←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 2 Signaling by Fas/APO-1/CD95, TNFR1 and TNFR2. For active apoptosis, trimeric death cytokines, FasL, and TNF interact with trimers formed by individual receptor molecules. This interaction promotes the recruitment of downstream signaling molecules that are distinct for each receptor. Fas induces apoptosis via recruitment of FADD and caspase-8 as described in the text. Shown are the signaling impediments caused by the ALPS mutations in the death domain and by FLIPS in FADD:caspase-8 DED interactions. TNFR1 can signal either for apoptosis through TRADD and FADD, or deliver anti-apoptotic signals via recruitment of alternate signaling complexes consisting of TRADD, RIP, and TRAF-2 (124). The anti-apoptotic signaling complex probably activates the NF-κB pathway through the NIK kinase that binds TRAF-2 (345). NF-κB presumably induces the expression of as yet unidentified survival molecules. Signaling through TNFR2 proceeds through the recruitment of TRAF proteins, most likely a complex of TRAF1/2 as indicated, although other TRAF proteins can bind TNFR2 in vitro. The TRAF1/2 complex can also activate NF-κB, but experiments in TRAF-2-deficient mice have revealed other NF-κB-independent anti-apoptotic pathways (264, 346). Paradoxically, TNFR2 signaling may also sensitize cells for apoptosis through TNFR1 (dashed line), although the mechanism is not well understood (FKM Chan & M Lenardo, unpublished observations). The cytoplasmic mechanism of caspase activation is apparently initiated by a mitochondrial permeability transition that allows the release of cytochrome c. Cytochrome c causes the association of APAF1 and caspase 9, leading to the activation of caspase 9 and triggering a caspase cascade. Bcl-2 family proteins can block apoptosis at the point of mitochondrial permeability transition and possibly by directly inhibiting the APAF-1/Caspase-9 complex, which activates downstream caspases (308).

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

230

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL

a downstream participant in the pathway (168–173). Certain evidence indicates that spingomyelinases, kinases, or other signaling enzymes could play a role in Fas death, but, no posttranslational modifications of caspase-8, Fas, or FADD, have been documented to be necessary for caspase-8 activation (174–197). Rather, caspase-8 undergoes spontaneous autoprocessing if two or more precursors are juxtaposed (167, 198). In an unknown way, death also involves proteins related to familial Alzheimer’s disease genes (199, 200). The principal regulatory step in the Fas death pathway, however, is the signaling event in which a specific complex comprising at least three proteins causes the processing and release of an active apoptosis-inducing caspase. The death mechanism entrained to Fas has been implicated in various disease processes. Two diseases clearly arise from Fas abnormalities: ALPS, which is due to deficient Fas function, as discussed below (201), and hepatitis, in which apoptosis results from inappropriate triggering of Fas on hepatocytes (202–208). Expression of FasL on non-lymphoid cells has been proposed as a means of “immune privilege” in which tolerance is established by killing any reactive T cells in the vicinity of the FasL-bearing cells, but these data have been controversial (209–216). In fact, evidence has shown that the overexpression of FasL on pancreatic islet cells fails to create tolerance and leads to more rapid rejection of transplanted islet cells by the recruitment of neutrophils (217, 218). Evidence also suggests that TCR- and Fas-mediated apoptosis participate in the depletion of CD4+ T cells in AIDS (219–225). This effect may involve “bystander” killing of uninfected cells secondary to generalized activation of the immune system (226). Infection by the human immunodeficiency virus (HIV) is highly lethal to Fas-defective peripheral blood T cells from ALPS patients, suggesting that direct viral killing is Fas independent (227). Death of T cells in AIDS may also be due to defective Bcl-X expression (228). Finally, the presence of Fas on non-lymphoid cells may contribute to organ-specific autoimmune diseases in which activated FasL-expressing T cells are generated and infiltrate tissue parenchyma (229–231). TNF may play similar pathogenic roles in disease (232–234).

Regulation of the Sensitivity of T Cells to FasL and TNF The sensitivity of T cells to FasL and TNF killing is subject to many influences that are biologically important. These include the general state of the cell as well as highly specific inhibitory molecules. The sensitization of resting T cells to antigen-induced death by IL-2 appears to involve a greater response of cycling T cells to death cytokines in addition to the greater amount of these cytokines that they produce. Sensitization requires, in part, progressing past early G1 phase of the cell cycle and may require multiple cell cycles (8, 18). Human T cells may become more responsive to Fas-induced death after prolonged

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

MATURE T LYMPHOCYTE APOPTOSIS

231

cultivation in IL-2 because of a more efficient recruitment of caspase-8 and FADD (235, 236). Specific cellular and viral inhibitors can also cause resistance to Fas-induced death. The Molluscum Contagiosum viral protein MC159L and the E8 protein from the equine Herpesvirus each contain two DED domains that will bind the FADD DED and thereby block Fas-induced apoptosis (237–239). A cellular gene product expressed in lymphocytes—and is called variously Usurpin, I-FLICE, FLAME, MRIT, or Casper—blocks Fas death by preventing DED interactions (239–243). While this cellular protein is structurally related to caspase-8, it has amino acid substitutions that incapacitate its enzymatic function (242). These DED-containing inhibitory proteins have been called FLICE (caspase-8) Inhibitory Proteins (FLIPs): the former v-FLIP (for viral) and the latter c-FLIP (for cellular). The level of expression of c-FLIP decreases after IL-2 stimulation and has been proposed to regulate the sensitivity of T cells to Fas-induced death in this context (241, 244). In experimental settings, caspase inhibitory proteins such as crmA, p35, and IAP proteins can also block Fas-induced apoptosis; of these, IAP proteins are expressed in T cells and may play a physiologically important role (164, 245–251). In cell culture systems, there is evidence both for and against the ability of the Bcl-2 and its homologues to potently block Fas-induced T cell death (252–256). These differences may be attributable to the degree to which different cell types rely on mitochondrial changes during Fas- or TNFR-induced death (see below) (137, 257). Biochemical and genetic evidence in mice suggests that Fas and Bcl-2 affect distinct death pathways in nontransformed T cells (257–260). The p53 protein is not required for TCRinduced death (27). The NF-κB family of gene transactivators are induced by TNF and strongly inhibit apoptosis, potentially enabling a death response to TNF to be shifted to an activation response (261–263) (Figure 2). TNF can also effect a shift from death to activation independently of NF-κB (264). A large number of factors expressed in T cells associate with the cytoplasmic region of Fas. These factors include Daxx, FAP, and various components of the ubiquitin-proteasome complex (135, 265–268). These molecules have striking effects on Fas death in certain transfection assays, but their physiological roles in T lymphocyte death are still the subject of investigation. Other important modulatory effects on T cell death may result from bacterial lipopolysaccharide and cytokines in the milieu surrounding an activated T cell during an immune response (269–273). Finally, one of the most immunologically important means of facilitating T cell death by FasL or TNF is through TCR occupancy. Even in the first observations of antigen-induced T cell apoptosis, it was clear that only T cells that were specifically stimulated through the TCR, and not bystander cells, were killed (9). Subsequent studies explicitly demonstrated that in heterogeneous

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

232

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL

pools of cycling, susceptible T cells, only those that receive TCR engagement at the time of Fas stimulation will undergo death (274, 275). Thus, TCR stimulation is needed twice to kill a T cell: once to stimulate the production of FasL and/or TNF and again, at the time that Fas or TNFR are stimulated by these cytokines, to permit death. Recent experiments with altered peptide ligands have revealed that there is a “competency to die” TCR signal distinct from the signal that induces lymphokine production (276) (Combadiere & M Lenardo, unpublished results). The biochemical mechanism of this “competency” function of TCR is not known but preliminary evidence indicates that FADD and caspase-8 recruitment are not affected (M Siegel, R Lenardo, unpublished results). The requirement for TCR signaling during Fas engagement could prevent bystander T cell killing in heterogeneous immune responses.

Lymphokine Withdrawal (Passive) Apoptosis A striking effect that accompanies antigen clearance at the end of an immune response is the dramatic deletion of specifically activated T cells. Immune responses involving superantigens, peptide antigens, and virus infection show this effect, suggesting that it plays a fundamental role in T cell homeostasis by reducing the specifically expanded T cell clonotypes when they are no longer needed (58, 60–66) (Figure 1). The coincidence of declining antigen and lymphocyte attenuation suggests that passive apoptosis results from the loss of a trophic substance such as IL-2. In vitro, cycling T cells rapidly undergo apoptosis after being deprived of IL-2 (277). This process requires new gene expression because it is blocked by actinomycin D and cycloheximide. In contrast, death induced by cross-linking Fas or related receptors is independent of new protein synthesis. IL-2 lymphokine withdrawal apoptosis is therefore a likely cause of the drop in T cells at the end of an antigen response. The continuous infusion of IL-2 has been shown to reverse the decline of T cells after superantigen stimulation in vivo (278). Lymphokine withdrawal apoptosis involves activation of a caspase other than ICE and is potently inhibited by Bcl-2 and Bcl-X, which are cytoplasmic membrane-bound apoptosis-inhibitory proteins (84, 252, 260, 279–285). Genetic crosses between Bcl-2 overexpressing transgenic mice and lpr mice result in a synergistic increase in lymphadenopathy compared to lpr only (257, 259). These data indicate that Fas and Bcl-2 are likely to affect independent death pathways whereas Bcl-2 and Bcl-X appear to affect the same apoptosis pathway (286). Bcl-2 represents a new class of oncogenes; members of this new class enhance cell survival rather than stimulate proliferation (283, 287, 288). Translocations involving the Bcl-2 gene and the Ig heavy chain locus are found in follicular and diffuse B cell lymphomas and Bcl-2 transgenes expressed in T lymphocytes lead

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

MATURE T LYMPHOCYTE APOPTOSIS

233

to T lymphomas (289). Full transformation may require abnormalities of other oncogenes, such as c-myc, in addition to dysregulation of Bcl-2. Although it is not clear that Bcl-2 is involved, lymphoid malignancy has been associated with an inherited Fas mutation in one family and somatic Fas mutations or defective Fas function in other human tumors (290–295). Thus, dysregulation of T cell apoptosis appears to foster the development of lymphoid malignancy. Lymphokine withdrawal apoptosis has no known requirement for death cytokines or their receptors. Rather, it seems to involve the direct cytoplasmic activation of caspases, possibly as a result of mitochondrial damage (284, 296). Disturbances in the electrochemical gradient on the inner mitochondrial membrane, and the subsequent release of cytochrome c and other mitochondrial proteins, induce caspase activation and apoptosis (297–302). This condition is thought to arise from a variation in the mitochondrial permeability transition pore, which dissipates the inner mitochondrial membrane potential, 9 (184). After its release, cytochrome c, and the CED4 homologue APAF-1 associate with caspase-9 and cause its proteolytic activation (303, 304). Active caspase-9 can then process caspase-3/CPP32 and other effector caspases and possibly initiate a positive feedback loop because caspase-9 is a substrate for caspase-3. (305). Bcl-2/Bcl-X may inhibit this form of apoptosis by blocking the release of cytochrome c via binding to the mitochondrial membrane, to APAF-1, or to both, as well as by having inhibitory effects on the active caspase complex (306–311). Similar mechanisms may also underlie T cell death due to genotoxic agents and glucocorticoids (C Zacharchuk, personal communication). How lymphokine withdrawal induces mitochondrial changes and apoptosis is currently under study. One possible mechanism was found in IL-3 dependent cell lines. In these cells IL-3 maintains phosphorylation of the pro-apoptotic bcl-2 family member BAD through the PI-3K/Akt pathway. Phosphorylated BAD is sequestered by 14-3-3 proteins, preventing interaction with other bcl-2 family members on intracellular membranes. After IL-3 withdrawal, BAD becomes dephosphorylated, is released from the 14-3-3 protein, and can form inhibitory dimers with other bcl-2 family members, blocking the anti-apoptotic function of these proteins (347). Whether this mechanism applies to other types of lymphokine withdrawal is not known.

MOLECULES CONTROLLING QUIESCENT T CELL SURVIVAL OR DEATH A great majority of lymphocytes live in a quiescent state, residing in G0 of the cell cycle and performing none of the effector functions of activated cells. This is true of naive lymphocytes and possibly also of some memory cells

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

234

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL

that have entered a long-term pool after antigen stimulation (312). Bcl-2 is essential for the survival of quiescent or “resting” cells: Bcl-2-deficient mice lose all lymphocytes as they grow older (313). Bcl-X deficient lymphocytes are also prone to apoptosis (314). In addition, resting lymphocyte survival is subtly regulated by the antigen receptor (315–318). Naive T cells require their restricting MHC molecule to survive, and memory cells need MHC but not necessarily their restricting element. Resting B cells also require antigen receptor expression, and perhaps engagement, to survive (318, 319). These peripheral survival signals may be limited in a way that fosters competition between resident cells and recent arrivals to create a peripheral lymphoid pool that is constant in size (7, 320). Thus antigen receptors and perhaps a ligand causing partial activation may ensure a normal lymphocyte life span in a resting state. The molecular control of resting cell survival is incompletely understood but appears to involve gene transcription events. One molecule involved may be the “lung kruppel-like factor” (LKLF), a transcription factor that is expressed in single positive thymocytes and resting (but not activated) T cells (321). Ablation of this factor causes apoptosis in resting T cells. Genetic deficiencies of the NF-κB transcription factor and related molecules also increase spontaneous and TNF-induced lymphocyte apoptosis (261, 262). How these transcription factors promote survival is unclear, but presumably it involves the induction of longevity genes or the inhibition of death genes. Naive B and T cell death has been associated in certain studies with Fas and FasL gene expression, respectively, but no functional role has been demonstrated (319, 321). Cytokines are likely to play an important role because IL-4, IL-6, and IL-7 can promote the survival of resting T cells (322, 323). The molecules governing resting cell survival warrant further study. This process may ensure a diverse repertoire as well as eliminate useless, potentially harmful, or malignant lymphocytes.

AUTOIMMUNE LYMPHOPROLIFERATIVE DISEASE (ALPS)—A GENETIC APPROACH TO STUDYING CELL DEATH IN HUMANS Genetic analysis in simple organisms, such as the nematode Caenorhabditis elegans, has powerfully advanced our understanding of programmed cell death (324). Death in Caenorhabditis elegans is not ligand directed or modulated as a response to physiological need; rather, it is hard-wired in a cell-autonomous fashion into the developmental program. Three major components of nematode cell death—CED3, CED4, and CED9—have functional homologues in the mammalian molecules: caspases, APAF-1, and Bcl-2, respectively (159, 303,

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

MATURE T LYMPHOCYTE APOPTOSIS

235

304, 325). In T lymphocytes, the principal forms of regulated death involve receptor signaling (either agonist or antagonist) and thus are a response to the external environment rather than hard-wired cell fates. Genetic analysis in humans may identify important molecules that have evolved for specialized roles in regulating mammalian apoptosis. An opportunity to undertake a genetic approach to studying T cell apoptosis came in the definition of the Autoimmune Lymphoproliferative Syndrome (ALPS) and its relationship to lpr mice (293, 326). ALPS is an autosomal dominant disorder with incomplete penetrance in which a defect in TCR-induced apoptosis is associated with chronic nonmalignant lymphoproliferation, autoimmunity, and an increase in a normally rare population of αβ T cells that are CD4− CD8−. ALPS patients also manifest increases in single positive αβ T cells, γ δ T cells, and B cells along with characteristic histopathologic changes in the lymph node and spleen that in rare instances can progress to lymphoma. In ALPS patients, antigen-induced apoptosis is defective in T cells, and Fasinduced apoptosis is defective in T and B cells (154, 293, 327). The unchecked lymphoproliferation may result from stimulation by self-antigens as well as foreign antigens. Autoimmune diseases such as anemia and thrombocytopenia in ALPS are due to a disturbance in humoral immunity that involves increased TH2 cells and the production of autoantibodies (272). Most of these features are observed in lpr mice, which led to the hypothesis that ALPS patients have the same pathogenic defect as lpr mice (328). We tested this hypothesis and determined that T cells from ALPS children exhibited clear defects in TCR-induced apoptosis. Further studies showed that defective apoptosis in most patients was due to inactivating mutations in the Fas gene that clearly demonstrated the molecular equivalence between lpr mice and ALPS (154, 327). Related findings have been made in many human subpopulations (185, 293, 329–333). The study of apoptosis defects in ALPS has revealed that of the 50 cases seen at the NIH, approximately 85% are associated with mutations in the Fas gene (326). Most ALPS cases are due to the inheritance of a mutant gene rather than a de novo mutation. A FasL gene mutation has been observed in only one patient who manifested symptoms of both ALPS and systemic lupus erythematosus (334). The remaining ALPS cases are of great interest because they do not have mutations in the Fas, FasL, FADD or caspase-8 genes (J Wang & M Lenardo, unpublished observations). These unusual cases should allow a genetic exploration of novel regulatory aspects of antigen-induced apoptosis in human lymphocytes. Even in regard to individuals harboring Fas mutations, critical genetic and environmental determinants of disease remain to be discovered. We have repeatedly observed individuals with identical Fas mutations in the same kindred exhibiting widely different clinical phenotypes, sometimes

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

236

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL

manifesting no disease at all. Studies from inbred strains of mice suggest that this variation in disease stems from genetic background differences (335, 336). The lpr gene on an MRL background is associated with a shortened life span due to vasculitis and glomerulonephritis (77). When the lpr gene is crossed onto C57BL/6 or other mouse backgrounds, the disease is indolent, suggesting that “background” genes play a decisive role in the disease. The Fas mutations in ALPS patients are genotypically different than in lpr mice. In ALPS, Fas mutations are almost always heterozygous dominantinterfering mutations rather homozygous loss-of-function mutations like lpr (154, 293). So far, only one ALPS case has been attributed to a loss-of-function mutation that is homozygous by consanguinity (327). The Fas mutations in the NIH cohort most frequently encode abnormal Fas proteins with amino acid substitutions in the intracytoplasmic “death domain.” These typically prevent the binding of FADD (D Martin & M Lenardo, unpublished observations). The dominant-interfering effect is caused by the physical association of wild-type and mutant Fas chains in receptor trimers that prevents the formation of a specific hexameric signaling complex involving Fas and FADD (R Siegel, J Song & M Lenardo, unpublished observations). Thus while the formation of a specific signaling complex may tightly control the transmission of death signals, it creates a vulnerability to dominant-interfering mutations in the Fas gene. Because only heterozygosity is required to cause a functional defect, even outbred populations such as humans will manifest deleterious effects caused by these types of alleles.

ANTIGEN-INDUCED APOPTOSIS AS A THERAPEUTIC APPROACH FOR T CELL-MEDIATED DISEASES When it was first recognized that antigens could specifically induce the death of activated T cells, therapeutic uses of this effect were envisioned (13, 46, 258, 337, 338). T cell-mediated diseases such as autoimmune diseases, allograft rejection, and allergies could be ameliorated if the antigen-specific T cell component could be eliminated. Just as vaccines serve to stimulate and increase reactive T cells, antigen-induced apoptosis therapy could potentially exploit a natural immunoregulatory mechanism to reduce or eliminate pathogenic T cells. An antigen-specific approach to T cell-mediated diseases could augment or supplant current therapies in the future. Steroids and cyclosporin A are generally immunosuppressive and have other serious side effects. Therapeutic antigen-induced apoptosis has been validated in principle in experimental allergic encephalomyelitis in mice (46). Nonetheless, the direct application of this approach to human disease is challenging because of the difficulties in defining pathogenic antigens against widely differing MHC

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

MATURE T LYMPHOCYTE APOPTOSIS

237

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

backgrounds and in preparing such proteins for pharmacological use. In addition, “determinant spreading” could theoretically expand the number of pathogenic epitopes during the disease process (339). Notwithstanding these challenges, this is a promising area for future endeavors. A large number of serious and widespread diseases such as diabetes, rheumatoid arthritis, multiple sclerosis, and others appear to have a T cell component. Also eliminating pathogenic T cells may greatly facilitate the success of bone marrow and whole organ transplantation.

CONCLUSION The population of mature T lymphocytes is tightly controlled for cell number, repertoire diversity, and self-tolerance by programmed death or apoptosis. Apoptosis can occur throughout the life of a T cell, both in its resting and activated state. However, apoptosis plays an especially important role after antigen activation in governing immune homeostasis and tolerance. Propriocidal or feedback-response regulation of T cell apoptosis is guided by IL-2 and antigen levels in the local immune environment (9, 340). The level of cell cycling induced by IL-2 is a quantitative determinant of the degree of death induced by T cell receptor re-engagement (8, 9). Active antigen-induced T cell death constrains the immune response under conditions of high IL-2 and antigen, whereas passive lymphokine withdrawal T cell death removes excessive T cells and occurs under conditions of low IL-2 and antigen stimulation such as at the end of an immune response. Antigen-induced death is mediated by the death cytokines FasL and TNF and their respective receptors that recruit and activate caspases. The lpr mouse and the human disease ALPS reveal that lymphocyte accumulation and autoimmunity are the consequences of defects in antigen-induced apoptosis. The study of ALPS patients offers a genetic approach to uncovering new components of the apoptosis pathway and related mechanisms of immune homeostasis in man. The rapid progress in understanding mature T cell apoptosis has provided a platform from which we can explore a number of additional questions: (i) the molecular basis of thymocyte death induced by positive, negative, and betachain selection during development (341–343); (ii) the mechanism underlying the requirement of TCR:MHC interactions for resting T cell survival (316); (iii) how cell cycle progression caused by IL-2 contributes to T cell death by TCR engagement or lymphokine withdrawal; (iv) whether T cell death regulation is the same or different in distinct lymphoid compartments such as the spleen, lymph node, and the mucosa; (v) how CD4 and CD8 T cells escape death to become memory cells; (vi) the nature of antigens involved in common autoimmune diseases that would allow antigen-induced apoptosis

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

238

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

therapy, (vii) the mechanism of viral killing of CD4+ T cells by HIV; and (viii) in ALPS, the derivation of the CD4−, CD8− T cells, the genetic factors that control the penetrance of the disease, and the specific antigens that trigger the organ-specific autoimmunity (293, 344). In summary, many fruitful lines of investigation could provide a fuller understanding of lymphocyte longevity in cellular and molecular terms, and facilitate the development of novel therapies for T cell-mediated disease as well as vaccines against formidable pathogens such as HIV. ACKNOWLEDGMENTS The explosive growth of investigation in this area made it impossible to discuss each of the hundreds of related studies. The papers cited in this review should be viewed only as representative. We apologize to many investigators whose important work was not explicitly cited. We thank Jacques F.A.P. Miller for helpful ideas on T cell homeostasis and the manuscript; we also thank Galen Fisher for reading the manuscript. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Moskophidis D, Laine E, Zinkernagel RM. 1993. Peripheral clonal deletion of antiviral memory CD8+ T cells. Eur. J. Immunol. 23:3306–11 2. Marrack P, Blackman M, Kushnir E, Kappler J. 1990. The toxicity of staphylococcal enterotoxin B in mice is mediated by T cells. J. Exp. Med. 171:455–64 3. Pantaleo G, Demarest JF, Soudeyns H, Graziosi C, Denis F, Adelsberger JW, Borrow P, Saag MS, Shaw GM, Sekaly RP, Fauci AS. 1994. Major expansion of CD8+ T cells with a predominant V-beta usage during the primary immune response to HIV. Nature 370:463–67 4. 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 5. Murali-Krishna K, Altman JD, Suresh M, Sourdive DJ, Zajac AJ, Miller JD, Slansky J, Ahmed R. 1998. Counting antigenspecific CD8 T cells: a reevaluation of bystander activation during viral infection. Immunity 8:177–87 6. Scollay RG, Butcher EC, Weissman IL.

7.

8.

9. 10.

11. 12.

1980. Thymus cell migration. Quantitative aspects of cellular traffic from the thymus to the periphery in mice. Eur. J. Immunol. 10:210–18 Berzins SP, Boyd RL, Miller JF. 1998. The role of the thymus and recent thymic migrants in the maintenance of the adult peripheral lymphocyte pool. J. Exp. Med. 187:1839–48 Boehme SA, Lenardo MJ. 1993. Propriocidal apoptosis of mature T lymphocytes occurs at S phase of the cell cycle. Eur. J. Immunol. 23:1552–60 Lenardo MJ. 1991. Interleukin-2 programs mouse αβ T lymphocytes for apoptosis. Nature 353:858–61 Russell JH, White CL, Loh DY, MeleedyRey P. 1991. Receptor-stimulated death pathway is opened by antigen in mature T cells. Proc. Natl. Acad. Sci. USA 88:2151–55 Crabtree GR. 1989. Contingent genetic regulatory events in T lymphocyte activation. Science 243:355–61 Green DR, Bissonnette RP, Glynn JM, Shi Y. 1992. Activation-induced apoptosis in lymphoid systems. Semin. Immunol. 4:379–88

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

MATURE T LYMPHOCYTE APOPTOSIS 13. Critchfield JM, Boehme SA, Lenardo MJ. 1995. The regulation of antigen-induced apoptosis in mature T lymphocytes. In Apoptosis and the Immune Response, ed. CD Gregory, pp. 55–114. New York: Wiley-Liss 14. Noel PJ, Boise LH, Green JM, Thompson CB. 1996. CD28 costimulation prevents cell death during primary T cell activation. J. Immunol. 157:636–42 15. Radvanyi LG, Shi Y, Vaziri H, Sharma A, Dhala R, Mills GB, Miller RG. 1996. CD28 costimulation inhibits TCRinduced apoptosis during a primary T cell response. J. Immunol. 156:1788–98 16. Radvanyi LG, Shi Y, Mills GB, Miller RG. 1996. Cell cycle progression out of G1 sensitizes primary-cultured nontransformed T cells to TCR-mediated apoptosis. Cell. Immunol. 170:260–73 17. Zhu L, Anasetti C. 1995. Cell cycle control of apoptosis in human leukemic T cells. J. Immunol. 154:192–200 18. Lissy NA, Van Dyk LF, Becker-Hapak M, Vocero-Akbani A, Mendler JH, Dowdy SF. 1998. TCR antigen-induced cell death occurs from a late G1 phase cell cycle check point. Immunity 8:57–65 19. Fournel S, Genestier L, Robinet E, Flacher M, Revillard JP. 1996. Human T cells require IL-2 but not G1/S transition to acquire susceptibility to Fas-mediated apoptosis. J. Immunol. 157:4309–15 20. Gonzalo JA, Baixeras E, Gonzalez-Garcia A, George-Chandy A, Van Rooijen N, Martinez C, Kroemer G. 1994. Differential in vivo effects of a superantigen and an antibody targeted to the same T cell receptor. Activation-induced cell death vs passive macrophage-dependent deletion. J. Immunol. 152:1597–608 21. Ashwell JD, Cunningham RE, Noguchi PD, Hernandez D. 1987. Cell growth cycle block of T cell hybridomas upon activation with antigen. J. Exp. Med. 165:173–94 22. Mercep M, Noguchi PD, Ashwell JD. 1989. The cell cycle block and lysis of an activated T cell hybridoma are distinct processes with different Ca2+ requirements and sensitivity to cyclosporine A. J. Immunol. 142:4085–92 23. Brunner T, Mogil RJ, LaFace D, Yoo NJ, Mahboubi A, Echeverri F, Martin SJ, Force WR, Lynch DH, Ware CF, Green DR. 1995. Cell-autonomous Fas (CD95)/Fas-ligand interaction mediates activation-induced apoptosis in T-cell hybridomas. Nature 373:441–44 24. Dhein J, Walczak H, Baumler C, Debatin KM, Krammer PH. 1995. Au-

25.

26.

27.

28. 29.

30.

31.

32.

33.

34.

35.

36.

239

tocrine T-cell suicide mediated by APO1/(Fas/CD95). Nature 373:438–41 Fruman DA, Mather PE, Burakoff SJ, Bierer BE. 1992. Correlation of calcineurin phosphatase activity and programmed cell death in murine T cell hybridomas. Eur. J. Immunol. 22:2513–17 Shi YF, Glynn JM, Guilbert LJ, Cotter TG, Bissonnette RP, Green DR. 1992. Role for c-myc in activation-induced apoptotic cell death in T-cell hybridomas. Science 257:212–14 Boehme SA, Lenardo MJ. 1996. TCRmediated death of mature T lymphocytes occurs in the absence of p53. J. Immunol. 156:4075–78 Evan GI, Brown L, Whyte M, Harrington E. 1995. Apoptosis and the cell cycle. Curr. Opin. Cell Biol. 7:825–34 Janssen O, Wesselborg S, Heckl-Ostreicher B, Pechhold K, Bender A, Schondelmaier S, Moldenhauer G, Kabelitz D. 1991. T cell receptor/CD3-signaling induces death by apoptosis in human T cell receptor gamma delta + T cells. J. Immunol. 146:35–39 Boehme SA, Lenardo MJ. 1993. Ligandinduced apoptosis of mature T lymphocytes (propriocidal regulation) occurs at distinct stages of the cell cycle. Leukemia 7(Suppl. 2):S45–49 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 Ramsdell F, Seaman MS, Miller RE, Picha KS, Kennedy MK, Lynch DH. 1994. Differential ability of Th1 and Th2 T cells to express Fas ligand and to undergo activation-induced cell death. Int. Immunol. 6:1545–53 Ortaldo JR, Mason AT, O’Shea JJ. 1995. Receptor-induced death in human natural killer cells: involvement of CD16. J. Exp. Med. 181:339–44 Willerford DM, Chen JZ, 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, Kundig TM, Furlonger C, Wakeham A, Timms E, Matsuyama T, Schmits R, Simard JJ, Ohashi PS, Griesser H, et al. 1995. Deregulated T cell activation and autoimmunity in mice lacking interleukin-2 receptor beta. Science 268:1472–76 Kneitz B, Herrmann T, Yonehara S, Schimpl A. 1995. Normal clonal expan-

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

240

37.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

38.

39. 40.

41.

42.

43.

44.

45.

46.

47.

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL sion but impaired Fas-mediated cell death and anergy induction in interleukin-2deficient mice. Eur. J. Immunol. 25:2572– 77 Zheng L, Trageser CL, Willerford DM, Lenardo MJ. 1998. T cell growth cytokines cause the superinduction of molecules mediating antigen-induced T lymphocyte death. J. Immunol. 160:763– 69 Hunig T, Schimpl A. 1997. Systemic autoimmune disease as a consequence of defective lymphocyte death. Curr. Opin. Immunol. 9:826–30 Abbas AK. 1996. Die and let live: eliminating dangerous lymphocytes. Cell 84:655–57 Lenardo MJ, Boehme S, Chen L, Combadiere B, Fisher G, Freedman M, McFarland H, Pelfrey C, Zheng L. 1995. Autocrine feedback death and the regulation of mature T lymphocyte antigen responses. Int. Rev. Immunol. 13:115–34 Morgan DA, Ruscetti FW, Gallo R. 1976. Selective in vitro growth of T lymphocytes from normal human bone marrows. Science 193:1007–8 Alexander-Miller MA, Leggatt GR, Sarin A, Berzofsky JA. 1996. Role of antigen, CD8, and cytotoxic T lymphocyte (CTL) avidity in high dose antigen induction of apoptosis of effector CTL. J. Exp. Med. 184:485–92 Pelfrey CM, Tranquill LR, Boehme SA, McFarland HF, Lenardo MJ. 1995. Two mechanisms of antigen-specific apoptosis of myelin basic protein (MBP)-specific T lymphocytes derived from multiple sclerosis patients and normal individuals. J. Immunol. 154:6191–202 Boehme SA, Zheng L, Lenardo MJ. 1995. Analysis of the CD4 coreceptor and activation-induced costimulatory molecules in antigen-mediated mature T lymphocyte death. J. Immunol. 155:1703–12 Boise LH, Minn AJ, June CH, Lindsten T, Thompson CB. 1995. Growth factors can enhance lymphocyte survival without committing the cell to undergo cell division. Proc. Natl. Acad. Sci. USA 92:5491– 95 Critchfield JM, Racke MK, ZunigaPflucker JC, Cannella B, Raine CS, Goverman J, Lenardo MJ. 1994. T cell deletion in high antigen dose therapy of autoimmune encephalomyelitis. Science 263:1139–43 Kung JT, Beller D, Ju ST. 1998. Lymphokine regulation of activation-induced apoptosis in T cells of IL-2 and IL-2R beta

48.

49.

50.

51.

52. 53. 54. 55. 56.

57.

58.

59.

60.

61.

knockout mice. Cell. Immunol. 185:158– 63 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 Van Parijs L, Biuckians A, Ibragimov A, Alt FW, Willerford DM, Abbas AK. 1997. Functional responses and apoptosis of CD25 (IL-2R alpha)-deficient T cells expressing a transgenic antigen receptor. J. Immunol. 158:3738–45 Zheng L, Fisher G, Miller RE, Peschon J, Lynch DH, Lenardo MJ. 1995. Induction of apoptosis in mature T cells by tumour necrosis factor. Nature 377:348–51 Wang R, Ciardelli TL, Russell JH. 1997. Partial signaling by cytokines: cytokine regulation of cell cycle and Fasdependent, activation-induced death in CD4+ subsets. Cell. Immunol. 182:152– 60 de Rosnay J. 1997. Cybernetics/Feedback Principia Cybernetica web. http:// pespmc1.vub.ac.be/feedback.html Weiner NL. 1961. Cybernetics, Control and Communication in the Animal and Machine. New York: MIT Press Benacerraf B, Greene MI, Sy MS, Dorf ME. 1982. Suppressor T cell circuits. Ann. NY Acad. Sci. 392:300–8 Dorf ME, Benacerraf B. 1984. Suppressor cells and immunoregulation. Annu. Rev. Immunol. 2:127–57 Mitchison NA. 1964. Induction of immunological paralysis in two zones of dosage. Proc. R. Soc. London Ser. B 161: 275–92 Critchfield JM, Zuniga-Pflucker JC, Lenardo MJ. 1995. Parameters controlling the programmed death of mature mouse T lymphocytes in high-dose suppression. Cell. Immunol. 160:71–78 Mamalaki C, Tanaka Y, Corbella P, Chandler P, Simpson E, Kioussis D. 1993. T cell deletion follows chronic antigen specific T cell activation in vivo. Int. Immunol. 5:1285–92 Bertolino P, Heath WR, Hardy CL, Morahan G, Miller JF. 1995. Peripheral deletion of autoreactive CD8+ T cells in transgenic mice expressing H-2Kb in the liver. Eur. J. Immunol. 25:1932–42 Jones LA, Chin LT, Longo DL, Kruisbeek AM. 1990. Peripheral clonal elimination of functional T cells. Science 250:1726– 29 Rellahan BL, Jones LA, Kruisbeek AM, Fry AM, Matis LA. 1990. In vivo induction of anergy in peripheral Vb8+ T cells

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

MATURE T LYMPHOCYTE APOPTOSIS

62.

63.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

64.

65. 66.

67.

68.

69.

70.

71.

72.

73.

by staphylococcal enterotoxin B. J. Exp. Med. 172:1091–1100 Webb S, Morris C, Sprent J. 1990. Extrathymic tolerance of mature T cells: clonal elimination as a consequence of immunity. Cell 63:1249–56 Kawabe Y, Ochi A. 1991. Programmed cell death and extrathymic reduction of V-beta-8+ CD4+ T cells in mice tolerant to Staphylococcus aureus enterotoxin B. Nature 349:245–48 MacDonald HR, Baschieri S, Lees RK. 1991. Clonal expansion precedes anergy and death of V-beta-8+ peripheral T cells responding to staphylococcal enterotoxin B in vivo. Eur. J. Immunol. 21:1963–66 Rocha B, von Boehmer H. 1991. Peripheral selection of the T cell repertoire. Science 251:1225–28 Kyburz D, Aichele P, Speiser DE, Hengartner H, Zinkernagel RM, Pircher H. 1993. T cell immunity after a viral infection versus T cell tolerance induced by soluble viral peptides. Eur. J. Immunol. 23:1956–62 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 Russell JH, Wang R. 1993. Autoimmune gld mutation uncouples suicide and cytokine/proliferation pathways in activated, mature T cells. Eur. J. Immunol. 23:2379–82 Russell JH, Rush B, Weaver C, Wang R. 1993. Mature T cells of autoimmune lpr/lpr mice have a defect in antigenstimulated suicide. Proc. Natl. Acad. Sci. USA 90:4409–13 Alderson MR, Tough TW, Davis-Smith T, Braddy S, Falk B, Schooley KA, Goodwin RG, Smith CA, Ramsdell F, Lynch DH. 1995. Fas ligand mediates activationinduced cell death in human T lymphocytes. J. Exp. Med. 181:71–77 Lynch DH, Watson ML, Alderson MR, Baum PR, Miller RE, Tough T, Gibson M, Davis-Smith T, Smith CA, Hunter K, et al. 1994. The mouse Fas-ligand gene is mutated in gld mice and is part of a TNF family gene cluster. Immunity 1:131–36 Takahashi T, Tanaka M, Brannan CI, Jenkins NA, Copeland NG, Suda T, Nagata S. 1994. Generalized lymphoproliferative disease in mice, caused by a point mutation in the Fas ligand. Cell 76:969– 76 Ju ST, Panka DJ, Cui H, Ettinger R, el-Khatib M, Sherr DH, Stanger BZ, Marshak-Rothstein A. 1995. Fas(CD95)/

74.

75.

76.

77.

78.

79.

80. 81.

82.

83.

84.

85.

241

FasL interactions required for programmed cell death after T-cell activation. Nature 373:444–48 Cui H, Sherr DH, el-Khatib M, Matsui K, Panka DJ, Marshak-Rothstein A, Ju ST. 1996. Regulation of T-cell death genes: selective inhibition of FasL- but not Fas-mediated function. Cell. Immunol. 167:276–84 Siegel RM, Martin DA, Hornung F, Dudley EC, Zheng L, Lenardo MJ. 1998. Signaling through Fas (CD95, APO-1) and related death receptors. In Signal Transduction and Cell Cycle Inhibitors, ed. S Gutkind. New York: Human Press Gruss HJ, Dower SK. 1995. Tumor necrosis factor ligand superfamily: involvement in the pathology of malignant lymphomas. Blood 85:3378–404 Cohen PL, Eisenberg RA. 1991. Lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu. Rev. Immunol. 9:243–69 Watanabe-Fukunaga R, Brannan CI, Copeland NG, Jenkins NA, Nagata S. 1992. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356:314– 47 Ramsdell F, Seaman MS, Miller RE, Tough TW, Alderson MR, Lynch DH. 1994. gld/gld mice are unable to express a functional ligand for Fas. Eur. J. Immunol. 24:928–33 Theofilopoulos AN, Dixon FJ. 1995. Murine models of systemic lupus erythematosus. Adv. Immunol. 37:269 Gillette-Ferguson I, Sidman CL. 1994. A specific intercellular pathway of apoptotic cell death is defective in the mature peripheral T cells of autoimmune lpr and gld mice. Eur. J. Immunol. 24:1181–85 Singer GG, Abbas AK. 1994. The fas antigen is involved in peripheral but not thymic deletion of T lymphocytes in T cell receptor transgenic mice. Immunity 1:365–71 Tucek-Szabo CL, Andjelic S, Lacy E, Elkon KB, Nikolic-Zugic J. 1996. Surface T cell Fas receptor/CD95 regulation, in vivo activation, and apoptosis. Activation-induced death can occur without Fas receptor. J. Immunol. 156:192– 200 Sarin A, Conan-Cibotti M, Henkart PA. 1995. Cytotoxic effect of TNF and lymphotoxin on T lymphoblasts. J. Immunol. 155:3716–68 Sytwu HK, Liblau RS, McDevitt HO. 1996. The roles of Fas/APO-1 (CD95) and TNF in antigen-induced programmed cell

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

242

86.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL death in T cell receptor transgenic mice. Immunity 5:17–30 Speiser DE, Sebzda E, Ohteki T, Bachmann MF, Pfeffer K, Mak TW, Ohashi PS. 1996. Tumor necrosis factor receptor p55 mediates deletion of peripheral cytotoxic T lymphocytes in vivo. Eur. J. Immunol. 26:3055–60 Marino MW, Dunn A, Grail D, Inglese M, Noguchi Y, Richards E, Jungbluth A, Wada H, Moore M, Williamson B, Basu S, Old LJ. 1997. Characterization of tumor necrosis factor-deficient mice. Proc. Natl. Acad. Sci. USA 94:8093–98 Kayagaki N, Kawasaki A, Ebata T, Ohmoto H, Ikeda S, Inoue S, Yoshino K, Okumura K, Yagita H. 1995. Metalloproteinase-mediated release of human Fas ligand. J. Exp. Med. 182:1777–83 Mariani SM, Matiba B, Armandola EA, Krammer PH. 1997. Interleukin 1 betaconverting enzyme related proteases/ caspases are involved in TRAIL-induced apoptosis of myeloma and leukemia cells. J. Cell Biol. 137:221–29 Tanaka M, Suda T, Takahashi T, Nagata S. 1995. Expression of the functional soluble form of human fas ligand in activated lymphocytes. EMBO J. 14:1129–35 Rothstein TL, Wang JK, Panka DJ, Foote LC, Wang Z, Stanger B, Cui H, Ju ST, Marshak-Rothstein A. 1995. Protection against Fas-dependent Th1-mediated apoptosis by antigen receptor engagement in B cells. Nature 374:163–65 Wang J, Taniuchi I, Maekawa Y, Howard M, Cooper MD, Watanabe T. 1996. Expression and function of Fas antigen on activated murine B cells. Eur. J. Immunol. 26:92–96 Kagi D, Vignaux F, Ledermann B, Burki K, Depraetere V, Nagata S, Hengartner H, Golstein P. 1994. Fas and perforin pathways as major mechanisms of T cellmediated cytotoxicity. Science 265:528– 30 Sieg S, Huang Y, Kaplan D. 1997. Viral regulation of CD95 expression and apoptosis in T lymphocytes. J. Immunol. 159:1192–99 Yonehara S, Ishii A, Yonehara M. 1989. A cell-killing monoclonal antibody (antiFas) to a cell surface antigen codownregulated with the receptor of tumor necrosis factor. J. Exp. Med. 169:1747– 56 Trauth BC, Klas C, Peters AM, Matzku S, Moller P, Falk W, Debatin KM, Krammer PH. 1989. Monoclonal antibodymediated tumor regression by induction of apoptosis. Science 245:301–5

97. Nagata S, Golstein P. 1995. The Fas death factor. Science 267:1449–56 98. Matiba B, Mariani SM, Krammer PH. 1997. The CD95 system and the death of a lymphocyte. Semin. Immunol. 9:59–68 99. Smith CA, Farrah T, Goodwin RG. 1994. The TNF receptor superfamily of cellular and viral proteins: activation, costimulation, and death. Cell 76:959–66 100. Oehm A, Behrmann I, Falk W, Pawlita M, Maier G, Klas C, Li-Weber M, Richards S, Dhein J, Trauth BC, et al. 1992. Purification and molecular cloning of the APO-1 cell surface antigen, a member of the tumor necrosis factor/nerve growth factor receptor superfamily. Sequence identity with the Fas antigen. J. Biol. Chem. 267:10709–15 101. Orlinick JR, Vaishnaw A, Elkon KB, Chao MV. 1997. Requirement of cysteinerich repeats of the Fas receptor for binding by the Fas ligand. J. Biol. Chem. 272:28889–94 102. Armitage RJ. 1994. Tumor necrosis factor receptor superfamily members and their ligands. Curr. Opin. Immunol. 6:407–13 103. Wallach D, Boldin M, Varfolomeev E, Beyaert R, Vandenabeele P, Fiers W. 1997. Cell death induction by receptors of the TNF family: towards a molecular understanding. FEBS Lett. 410:96–106 104. Park CG, Lee SY, Kandala G, Choi Y. 1996. A novel gene product that couples TCR signaling to Fas(CD95) expression in activation-induced cell death. Immunity 4:583–91 105. Winoto A. 1997. Genes involved in T-cell receptor-mediated apoptosis of thymocytes and T-cell hybridomas. Semin. Immunol. 9:51–58 106. Wada N, Matsumura M, Ohba Y, Kobayashi N, Takizawa T, Nakanishi Y. 1995. Transcription stimulation of the Fasencoding gene by nuclear factor for interleukin-6 expression upon influenza virus infection. J. Biol. Chem. 270: 18007–12 107. Miyawaki T, Uehara T, Nibu R, Tsuji T, Yachie A, Yonehara S, Taniguchi N. 1992. Differential expression of apoptosisrelated Fas antigen on lymphocyte subpopulations in human peripheral blood. J. Immunol. 149:3753–58 108. Banner DW, D’Arcy A, Janes W, Gentz R, Schoenfeld HJ, Broger C, Loetscher H, Lesslauer W. 1993. Crystal structure of the soluble human 55 kd TNF receptorhuman TNF-beta complex: implications for TNF receptor activation. Cell 73:431– 45 109. Bajorath J, Aruffo A. 1997. Prediction of

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

MATURE T LYMPHOCYTE APOPTOSIS

110.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

111.

112.

113.

114.

115. 116.

117.

118.

119.

120.

121.

the three-dimensional structure of the human Fas receptor by comparative molecular modeling. J. Comput. Aided Mol. Des. 11:3–8 Fadeel B, Lindberg J, Achour A, Chiodi F. 1998. A three-dimensional model of the Fas/APO-1 molecule: cross-reactivity of anti-Fas antibodies explained by structural mimicry of antigenic sites. Int. Immunol. 10:131–40 Peitsch MC, Tschopp J. 1995. Comparative molecular modeling of the Fas-ligand and other members of the TNF family. Mol. Immunol. 32:761–72 Tartaglia LA, Ayres TM, Wong GH, Goeddel DV. 1993. A novel domain within the 55 kd TNF receptor signals cell death. Cell 74:845–53 Itoh N, Nagata S. 1993. A novel protein domain required for apoptosis. Mutational analysis of human Fas antigen. J. Biol. Chem. 268:10932–37 Jeremias I, Herr I, Boehler T, Debatin KM. 1998. TRAIL/Apo-2-ligandinduced apoptosis in human T cells. Eur. J. Immunol. 28:143–52 Golstein P. 1997. Cell death: TRAIL and its receptors. Curr. Biol. 7:R750–53 Marsters SA, Sheridan JP, Donahue CJ, Pitti RM, Gray CL, Goddard AD, Bauer KD, Ashkenazi A. 1996. Apo-3, a new member of the tumor necrosis factor receptor family, contains a death domain and activates apoptosis and NF-kappa-B. Curr. Biol. 6:1669–76 Pan G, Ni J, Wei YF, Yu G, Gentz R, Dixit VM. 1997. An antagonist decoy receptor and a death domain-containing receptor for TRAIL. Science 277:815–18 Pan G, O’Rourke K, Chinnaiyan AM, Gentz R, Ebner R, Ni J, Dixit VM. 1997. The receptor for the cytotoxic ligand TRAIL. Science 276:111–13 Schneider P, Thome M, Burns K, Bodmer JL, Hofmann K, Kataoka T, Holler N, Tschopp J. 1997. TRAIL receptors 1 (DR4) and 2 (DR5) signal FADDdependent apoptosis and activate NFkappa-B. Immunity 7:831–36 Sheridan JP, Marsters SA, Pitti RM, Gurney A, Skubatch M, Baldwin D, Ramakrishnan L, Gray CL, Baker K, Wood WI, Goddard AD, Godowski P, Ashkenazi A. 1997. Control of TRAIL-induced apoptosis by a family of signaling and decoy receptors. Science 277:818–21 Vandevoorde V, Haegeman G, Fiers W. 1997. Induced expression of trimerized intracellular domains of the human tumor necrosis factor (TNF) p55 receptor elicits TNF effects. J. Cell Biol. 137:1627–38

243

122. Varfolomeev EE, Boldin MP, Goncharov TM, Wallach D. 1996. A potential mechanism of “cross-talk” between the p55 tumor necrosis factor receptor and Fas/APO1: proteins binding to the death domains of the two receptors also bind to each other. J. Exp. Med. 183:1271–75 123. Grell M, Krammer PH, Scheurich P. 1994. Segregation of APO-1/Fas antigen- and tumor necrosis factor receptor-mediated apoptosis. Eur. J. Immunol. 24:2563–66 124. Hsu H, Shu HB, Pan MG, Goeddel DV. 1996. TRADD-TRAF2 and TRADDFADD interactions define two distinct TNF receptor 1 signal transduction pathways. Cell 84:299–308 125. Rothe M, Wong SC, Henzel WJ, Goeddel DV. 1994. A novel family of putative signal transducers associated with the cytoplasmic domain of the 75 kDa tumor necrosis factor receptor. Cell 78:681–92 126. Rothe M, Sarma V, Dixit VM, Goeddel DV. 1995. TRAF2-mediated activation of NF-kappa-B by TNF receptor 2 and CD40. Science 269:1424–27 127. Rothe M, Pan MG, Henzel WJ, Ayres TM, Goeddel DV. 1995. The TNFR2-TRAF signaling complex contains two novel proteins related to baculoviral inhibitor of apoptosis proteins. Cell 83:1243–52 128. Shu HB, Takeuchi M, Goeddel DV. 1996. The tumor necrosis factor receptor 2 signal transducers TRAF2 and c- IAP1 are components of the tumor necrosis factor receptor 1 signaling complex. Proc. Natl. Acad. Sci. USA 93:13973–78 129. Zheng L, Fisher G, Combadiere B, Horning F, Martin D, Pelfrey C, Wang J, Lenardo MJ. 1996. Mature T lymphocyte apoptosis in the healthy and diseased immune system. Adv. Exp. Med. Biol. 406:229–239 130. Clement MV, Stamenkovic I. 1994. Fas and tumor necrosis factor receptor-mediated cell death: similarities and distinctions. J. Exp. Med. 180:557–67 131. Heath WR, Kurts C, Miller JF, Carbone FR. 1998. Cross-tolerance: a pathway for inducing tolerance to peripheral tissue antigens. J. Exp. Med. 187:1549–53 132. Chinnaiyan AM, Dixit VM. 1997. Portrait of an executioner: the molecular mechanism of FAS/APO-1-induced apoptosis. Semin. Immunol. 9:69–76 133. Chinnaiyan AM, Tepper CG, Seldin MF, O’Rourke K, Kischkel FC, Hellbardt S, Krammer PH, Peter ME, Dixit VM. 1996. FADD/MORT1 is a common mediator of CD95 (Fas/APO-1) and tumor necrosis factor receptor-induced apoptosis. J. Biol. Chem. 271:4961–65

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

244

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL

134. Wong GH, Goeddel DV. 1994. Fas antigen and p55 TNF receptor signal apoptosis through distinct pathways. J. Immunol. 152:1751–55 135. Wang J, Lenardo MJ. 1997. Molecules involved in cell death and peripheral tolerance. Curr. Opin. Immunol. 9:818–25 136. Speiser DE, Lee SY, Wong B, Arron J, Santana A, Kong YY, Ohashi PS, Choi Y. 1997. A regulatory role for TRAF1 in antigen-induced apoptosis of T cells. J. Exp. Med. 185:1777–83 137. Scaffidi C, Fulda S, Srinivasan A, Friesen C, Li F, Tomaselli KJ, Debatin KM, Krammer PH, Peter ME. 1998. Two CD95 (APO-1/Fas) signaling pathways. EMBO J. 17:1675–87 138. Muzio M, Chinnaiyan AM, Kischkel FC, O’Rourke K, Shevchenko A, Ni J, Scaffidi C, Bretz JD, Zhang M, Gentz R, Mann M, Krammer PH, Peter ME, Dixit VM. 1996. FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-inducing signaling complex. Cell 85:817–27 139. Medema JP, Scaffidi C, Kischkel FC, Shevchenko A, Mann M, Krammer PH, Peter ME. 1997. FLICE is activated by association with the CD95 death-inducing signaling complex (DISC). EMBO J. 16:2794–804 140. Kischkel FC, Hellbardt S, Behrmann I, Germer M, Pawlita M, Krammer PH, Peter ME. 1995. Cytotoxicity-dependent APO-1 (Fas/CD95)-associated proteins form a death-inducing signaling complex (DISC) with the receptor. EMBO J. 14:5579–88 141. Bazzoni F, Beutler B. 1995. How do tumor necrosis factor receptors work? J. Inflamm. 45:221–38 142. Wong B, Choi Y. 1997. Pathways leading to cell death in T cells. Curr. Opin. Immunol. 9:358–64 143. Boldin MP, Varfolomeev EE, Pancer Z, Mett IL, Camonis JH, Wallach D. 1995. A novel protein that interacts with the death domain of Fas-Apo-1 contains a sequence motif related to the death doman. J. Biol. Chem. 270:7795–98 144. Boldin MP, Mett IL, Varfolomeev EE, Chumakov I, Shemer-Avni Y, Camonis JH, Wallach D. 1995. Self-association of the “death domains” of the p55 tumor necrosis factor (TNF) receptor and Fas/APO1 prompts signaling for TNF and Fas/APO1 effects. J. Biol. Chem. 270:387–91 145. Boldin MP, Goncharov TM, Goltsev YV, Wallach D. 1996. Involvement of MACH, a novel MORT1/FADD-interacting pro-

146.

147.

148.

149.

150.

151.

152.

153.

154.

155.

156.

tease, in Fas/APO-1- and TNF receptorinduced cell death. Cell 85:803–15 Chinnaiyan AM, O’Rourke K, Tewari M, Dixit VM. 1995. FADD, a novel death domain-containing protein, interacts with the death domain of Fas and initiates apoptosis. Cell 81:505–12 Yeh WC, Pompa JL, McCurrach ME, Shu HB, Elia AJ, Shahinian A, Ng M, Wakeham A, Khoo W, Mitchell K, El-Deiry WS, Lowe SW, Goeddel DV, Mak TW. 1998. FADD: essential for embryo development and signaling from some, but not all, inducers of apoptosis. Science 279:1954–58 Zhang J, Cado D, Chen A, Kabra NH, Winoto A. 1998. Fas-mediated apoptosis and activation-induced T-cell proliferation are defective in mice lacking FADD/Mort1. Nature 392:296–300 Newton K, Harris AW, Bath ML, Smith KGC, Strasser A. 1998. A dominant interfering mutant of FADD/MORT1 enhances deletion of autoreactive thymocytes and inhibits proliferation of mature T lymphocytes. EMBO J. 17:706–18 Walsh CM, Wen BG, Chinnaiyan AM, O’Rourke K, Dixit VM, Hedrick SM. 1998. A role for FADD in T cell activation and development. Immunity 8:439–49 Huang B, Eberstadt M, Olejniczak ET, Meadows RP, Fesik SW. 1996. NMR structure and mutagenesis of the Fas (APO-1/CD95) death domain. Nature 384:638–41 Eberstadt M, Huang B, Chen Z, Meadows RP, Ng SC, Zheng L, Lenardo MJ, Fesik SW. 1998. NMR structure and mutagenesis of the FADD (Mort1) death-effector domain. Nature 392:941–45 Itoh N, Yonehara S, Ishii A, Yonehara M, Mizushima S, Sameshima M, Hase A, Seto Y, Nagata S. 1991. The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis. Cell 66:233–43 Fisher GH, Rosenberg FJ, Straus SE, Dale JK, Middleton LA, Lin AY, Strober W, Lenardo MJ, Puck JM. 1995. Dominant interfering Fas gene mutations impair apoptosis in a human autoimmune lymphoproliferative syndrome. Cell 81:935– 46 Weis M, Schlegel J, Kass GE, Holmstrom TH, Peters I, Eriksson J, Orrenius S, Chow SC. 1995. Cellular events in Fas/APO-1mediated apoptosis in JURKAT T lymphocytes. Exp. Cell. Res. 219:699–708 Cifone MG, Roncaioli P, De Maria R, Camarda G, Santoni A, Ruberti G, Testi R. 1995. Multiple pathways originate at

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

MATURE T LYMPHOCYTE APOPTOSIS

157.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

158.

159. 160.

161.

162.

163.

164.

165.

166.

167.

168.

the Fas/APO-1 (CD95) receptor: sequential involvement of phosphatidylcholinespecific phospholipase C and acidic sphingomyelinase in the propagation of the apoptotic signal. EMBO J. 14:5859–68 Chow SC, Weis M, Kass GE, Holmstrom TH, Eriksson JE, Orrenius S. 1995. Involvement of multiple proteases during Fas-mediated apoptosis in T lymphocytes. FEBS Lett. 364:134–38 Scaffidi C, Medema JP, Krammer PH, Peter ME. 1997. FLICE is predominantly expressed as two functionally active isoforms, caspase-8/a and caspase-8/b. J. Biol. Chem. 272:26953–58 Nicholson D, Thornberry N. 1997. Caspases: killer proteases. Trends Biochem. Sci. 22:299–306 Miller DK, Myerson J, Becker JW. 1997. The interleukin-1 beta converting enzyme family of cysteine proteases. J. Cell. Biochem. 64:2–10 Casiano CA, Martin SJ, Green DR, Tan EM. 1996. Selective cleavage of nuclear autoantigens during CD95 (Fas/APO-1)mediated T cell apoptosis. J. Exp. Med. 184:765–70 Alnemri ES, Livingston DJ, Nicholson DW, Salvesen G, Thornberry NA, Wong WW, Yuan J. 1996. Human ICE/CED-3 protease nomenclature. Cell 87:171 Longthorne VL, Williams GT. 1997. Caspase activity is required for commitment to Fas-mediated apoptosis. EMBO J. 16:3805–12 Tewari M, Dixit VM. 1995. Fas- and tumor necrosis factor-induced apoptosis is inhibited by the poxvirus crmA gene product. J. Biol. Chem. 270:3255–60 Los M, Van de Craen M, Penning LC, Schenk H, Westendorp M, Baeuerle PA, Droge W, Krammer PH, Fiers W, Schulze-Osthoff K. 1995. Requirement of an ICE/CED-3 protease for Fas/APO-1mediated apoptosis. Nature 375:81–83 Memon SA, Hou J, Moreno MB, Zacharchuk CM. 1998. Apoptosis induced by a chimeric Fas/FLICE receptor: lack of requirement for Fas- or FADDbinding proteins. J. Immunol. 160:2046– 49 Martin DA, Siegel RM, Zheng L, Lenardo MJ. 1998. Membrane oligomerization and release activates the caspase 8 (FLICE/MACHα1) death signal. J. Biol. Chem. 273:4345–49 Schlegel J, Peters I, Orrenius S, Miller DK, Thornberry NA, Yamin TT, Nicholson DW. 1996. CPP32/apopain is a key interleukin 1 beta converting enzyme-like protease involved in Fas-mediated apop-

245

tosis. J. Biol. Chem. 271:1841–44 169. Enari M, Talanian RV, Wong WW, Nagata S. 1996. Sequential activation of ICElike and CPP32-like proteases during Fasmediated apoptosis. Nature 380:723–36 170. Kuida K, Zheng TS, Na S, Kuan C, Yang D, Karasuyama H, Rakic P, Flavell RA. 1996. Decreased apoptosis in the brain and premature lethality in CPP32deficient mice. Nature 384:368–72 171. Smith DJ, McGuire MJ, Tocci MJ, Thiele DL. 1997. IL-1 beta convertase (ICE) does not play a requisite role in apoptosis induced in T lymphoblasts by Fasdependent or Fas-independent CTL effector mechanisms. J. Immunol. 158:163–70 172. Fernandes-Alnemri T, Takahashi A, Armstrong R, Krebs J, Fritz L, Tomaselli KJ, Wang L, Yu Z, Croce CM, Salveson G, et al. 1995. Mch3, a novel human apoptotic cysteine protease highly related to CPP32. Cancer Res. 55:6045–52 173. Fernandes-Alnemri T, Litwack G, Alnemri ES. 1995. Mch2, a new member of the apoptotic Ced-3/Ice cysteine protease gene family. Cancer Res. 55:2737–42 174. Eischen CM, Dick CJ, Leibson PJ. 1994. Tyrosine kinase activation provides an early and requisite signal for Fas-induced apoptosis. J. Immunol. 153:1947–54 175. Huang S, Jiang Y, Li Z, Nishida E, Mathias P, Lin S, Ulevitch RJ, Nemerow GR, Han J. 1997. Apoptosis signaling pathway in T cells is composed of ICE/Ced-3 family proteases and MAP kinase kinase 6b. Immunity 6:739–49 176. Goillot E, Raingeaud J, Ranger A, Tepper RI, Davis RJ, Harlow E, Sanchez I. 1997. Mitogen-activated protein kinasemediated Fas apoptotic signaling pathway. Proc. Natl. Acad. Sci. USA 94:3302– 7 177. Atkinson EA, Ostergaard H, Kane K, Pinkoski MJ, Caputo A, Olszowy MW, Bleackley RC. 1996. A physical interaction between the cell death protein Fas and the tyrosine kinase p59fynT. J. Biol. Chem. 271:5968–71 178. Kolesnick RN, Haimovitz-Friedman A, Fuks Z. 1994. The sphingomyelin signal transduction pathway mediates apoptosis for tumor necrosis factor, Fas, and ionizing radiation. Biochem. Cell Biol. 72:471– 74 179. Cifone MG, De Maria R, Roncaioli P, Rippo MR, Azuma M, Lanier LL, Santoni A, Testi R. 1994. Apoptotic signaling through CD95 (Fas/Apo-1) activates an acidic sphingomyelinase. J. Exp. Med. 180:1547–52 180. Watts JD, Gu M, Polverino AJ, Patter-

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

246

181.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

182.

183.

184.

185.

186.

187.

188.

189.

190.

191.

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL son SD, Aebersold R. 1997. Fas-induced apoptosis of T cells occurs independently of ceramide generation. Proc. Natl. Acad. Sci. USA 94:7292–96 De Maria R, Rippo MR, Schuchman EH, Testi R. 1998. Acidic sphingomyelinase (ASM) is necessary for fas-induced GD3 ganglioside accumulation and efficient apoptosis of lymphoid cells. J. Exp. Med. 187:897–902 Cock JG, Tepper AD, de Vries E, van Blitterswijk WJ, Borst J. 1998. CD95 (Fas/APO-1) induces ceramide formation and apoptosis in the absence of a functional acid sphingomyelinase. J. Biol. Chem. 273:7560–65 Kennedy NJ, Budd RC. 1998. Phosphorylation of FADD/MORT1 and Fas by kinases that associate with the membraneproximal cytoplasmic domain of Fas. J. Immunol. 160:4881–88 Penninger JM, Kroemer G. 1998. Molecular and cellular mechanisms of T lymphocyte apoptosis. Adv. Immunol. 68:51– 144 Dianzani U, Bragardo M, DiFranco D, Alliaudi C, Scagni P, Buonfiglio D, Redoglia V, Bonissoni S, Correra A, Dianzani I, Ramenghi U. 1997. Deficiency of the Fas apoptosis pathway without Fas gene mutations in pediatric patients with autoimmunity/lymphoproliferation. Blood 89:2871–79 Cahill MA, Peter ME, Kischkel FC, Chinnaiyan AM, Dixit VM, Krammer PH, Nordheim A. 1996. CD95 (APO-1/Fas) induces activation of SAP kinases downstream of ICE-like proteases. Oncogene 13:2087–96 Lenczowski JM, Dominguez L, Eder AM, King LB, Zacharchuk CM, Ashwell JD. 1997. Lack of a role for Jun kinase and AP-1 in Fas-induced apoptosis. Mol. Cell Biol. 17:170–81 Toyoshima F, Moriguchi T, Nishida E. 1997. Fas induces cytoplasmic apoptotic responses and activation of the MKK7JNK/SAPK and MKK6-p38 pathways independent of CPP32-like proteases. J. Cell Biol. 139:1005–15 Sillence DJ, Allan D. 1997. Evidence against an early signalling role for ceramide in Fas-mediated apoptosis. Biochem. J. 324:29–32 Holmstrom TH, Chow SC, Elo I, Coffey ET, Orrenius S, Sistonen L, Eriksson JE. 1998. Suppression of Fas/APO-1mediated apoptosis by mitogen-activated kinase signaling. J. Immunol. 160:2626– 36 Inohara N, del Peso L, Koseki T, Chen

192.

193.

194.

195.

196.

197.

198.

199.

200.

201.

202.

203.

S, Nunez G. 1998. RICK, a novel protein kinase containing a caspase recruitment domain, interacts with CLARP and regulates CD95-mediated apoptosis. J. Biol. Chem. 273:12296–300 Juo P, Kuo CJ, Reynolds SE, Konz RF, Raingeaud J, Davis RJ, Biemann HP, Blenis J. 1997. Fas activation of the p38 mitogen-activated protein kinase signalling pathway requires ICE/CED-3 family proteases. Mol. Cell Biol. 17:24– 35 Kennedy NJ, Budd RC. 1998. Phosphorylation of FADD/MORT1 and Fas by kinases that associate with the membraneproximal cytoplasmic domain of Fas. J. Immunol. 160:4881–88 Lahti JM, Xiang J, Kidd VJ. 1995. Cell cycle-related protein kinases and T cell death. Adv. Exp. Med. Biol. 376:247– 58 Latinis KM, Koretzky GA. 1996. Fas ligation induces apoptosis and Jun kinase activation independently of CD45 and Lck in human T cells. Blood 87:871–75 Schraven B, Peter ME. 1995. APO-1 (CD95)-mediated apoptosis in Jurkat cells does not involve src kinases or CD45. FEBS Lett. 368:491–94 Tian Q, Taupin J, Elledge S, Robertson M, Anderson P. 1995. Fas-activated serine/threonine kinase (FAST) phosphorylates TIA-1 during Fas-mediated apoptosis. J. Exp. Med. 182:865–74 Yang X, Chang HY, Baltimore D. 1998. Autoproteolytic activation of procaspases by oligomerization. Mol. Cell. 1:319–25 Vito P, Lacana E, D’Adamio L. 1996. Interfering with apoptosis: Ca(2+)-binding protein ALG-2 and Alzheimer’s disease gene ALG-3. Science 271:521–25 D’Adamio L, Lacana E, Vito P. 1997. Functional cloning of genes involved in T-cell receptor-induced programmed cell death. Semin. Immunol. 9:17–23 Puck JM, Sneller MC. 1997. ALPS: an autoimmune human lymphoproliferative syndrome associated with abnormal lymphocyte apoptosis. Semin. Immunol. 9:77–84 Kondo T, Suda T, Fukuyama H, Adachi M, Nagata S. 1997. Essential roles of the Fas ligand in the development of hepatitis. Nat. Med. 3:409–13 Ogasawara J, Watanabe-Fukunaga R, Adachi M, Matsuzawa A, Kasugai T, Kitamura Y, Itoh N, Suda T, Nagata S. 1993. Lethal effect of the anti-Fas antibody in mice. Nature 364:806–9 Erratum. 1993. Nature 365:568

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

MATURE T LYMPHOCYTE APOPTOSIS 204. Lacronique V, Mignon A, Fabre M, Viollet B, Rouquet N, Molina T, Porteu A, Henrion A, Bouscary D, Varlet P, Joulin V, Kahn A. 1996. Bcl-2 protects from lethal hepatic apoptosis induced by an anti-Fas antibody in mice. Nat. Med. 2:80–86 205. Ruggieri A, Harada T, Matsuura Y, Miyamura T. 1997. Sensitization to Fasmediated apoptosis by hepatitis C virus core protein. Virology 229:68–76 206. Seino K, Kayagaki N, Takeda K, Fukao K, Okumura K, Yagita H. 1997. Contribution of Fas ligand to T cell-mediated hepatic injury in mice. Gastroenterology 113:1315–22 207. Galle PR, Hofmann WJ, Walczak H, Schaller H, Otto G, Stremmel W, Krammer PH, Runkel L. 1995. Involvement of the CD95 (APO-1/Fas) receptor and ligand in liver damage. J. Exp. Med. 182:1223–30 208. Strand S, Hofmann WJ, Grambihler A, Hug H, Volkmann M, Otto G, Wesch H, Mariani SM, Hack V, Stremmel W, Krammer PH, Galle PR. 1998. Hepatic failure and liver cell damage in acute Wilson’s disease involve CD95 (APO-1/Fas) mediated apoptosis. Nat. Med. 4:588–93 209. Kang SM, Hoffmann A, Le D, Springer ML, Stock PG, Blau HM. 1997. Immune response and myoblasts that express Fas ligand. Science 278:1322–24 210. Chervonsky AV, Wang Y, Wong FS, Visintin I, Flavell RA, Janeway CA Jr, Matis LA. 1997. The role of Fas in autoimmune diabetes. Cell 89:17–24 211. Bellgrau D, Gold D, Selawry H, Moore J, Franzusoff A, Duke RC. 1995. A role for CD95 ligand in preventing graft rejection. Nature 377:630–32 212. Allison J, Seino K, Yagita H. 1998. Can expression of CD95 (Fas/APO-1) ligand on grafts or tumor cells prevent their rejection? Springer Semin. Immunopathol. 19:311–22 213. Allison J, Georgiou HM, Strasser A, Vaux DL. 1997. Transgenic expression of CD95 ligand on islet beta cells induces a granulocytic infiltration but does not confer immune privilege upon islet allografts. Proc. Natl. Acad. Sci. USA 94:3943– 47 214. Griffith TS, Brunner T, Fletcher SM, Green DR, Ferguson TA. 1995. Fas ligand-induced apoptosis as a mechanism of immune privilege. Science 270:1189– 92 215. Lau HT, Yu M, Fontana A, Stoeckert CJ Jr. 1996. Prevention of islet allograft rejection with engineered myoblasts expressing FasL in mice. Science 273:109–12

247

216. Sabelko KA, Kelly KA, Nahm MH, Cross AH, Russell JH. 1997. Fas and Fas ligand enhance the pathogenesis of experimental allergic encephalomyelitis, but are not essential for immune privilege in the central nervous system. J. Immunol. 159:3096– 99 217. Kang SM, Lin Z, Ascher NL, Stock PG. 1998. Fas ligand expression on islets as well as multiple cell lines results in accelerated neutrophilic rejection. Transpl. Proc. 30:538 218. Kang SM, Schneider DB, Lin Z, Hanahan D, Dichek DA, Stock PG, Baekkeskov S. 1997. Fas ligand expression in islets of Langerhans does not confer immune privilege and instead targets them for rapid destruction. Nat. Med. 3:738–43 219. Groux H, Torpier G, Monte D, Mouton Y, Capron A, Ameisen JC. 1992. Activation-induced death by apoptosis in CD4+ T cells from human immunodeficiency virus-infected asymptomatic individuals. J. Exp. Med. 175:331–40 220. Katsikis PD, Garcia-Ojeda ME, TorresRoca JF, Tijoe IM, Smith CA, Herzenberg LA. 1997. Interleukin-1 beta converting enzyme-like protease involvement in Fasinduced and activation-induced peripheral blood T cell apoptosis in HIV infection. TNF-related apoptosis-inducing ligand can mediate activation-induced T cell death in HIV infection. J. Exp. Med. 186:1365–72 221. Westendorp MO, Frank R, Ochsenbauer C, Stricker K, Dhein J, Walczak H, Debatin KM, Krammer PH. 1995. Sensitization of T cells to CD95-mediated apoptosis by HIV-1 Tat and gp120. Nature 375:497–500 222. Sloand EM, Young NS, Kumar P, Weichold FF, Sato T, Maciejewski JP. 1997. Role of Fas ligand and receptor in the mechanism of T-cell depletion in acquired immunodeficiency syndrome: effect on CD4+ lymphocyte depletion and human immunodeficiency virus replication. Blood 89:1357–63 223. Silvestris F, Cafforio P, Frassanito MA, Tucci M, Romito A, Nagata S, Dammacco F. 1996. Overexpression of Fas antigen on T cells in advanced HIV-1 infection: differential ligation constantly induces apoptosis. AIDS 10:131–41 224. Gougeon ML, Lecoeur H, Dulioust A, Enouf MG, Crouvoiser M, Goujard C, Debord T, Montagnier L. 1996. Programmed cell death in peripheral lymphocytes from HIV-infected persons: increased susceptibility to apoptosis of CD4 and CD8 T cells correlates with lymphocyte activa-

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

248

225.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

226.

227.

228.

229.

230.

231.

232.

233. 234.

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL tion and with disease progression. J. Immunol. 156:3509–20 Badley AD, McElhinny JA, Leibson PJ, Lynch DH, Alderson MR, Paya CV. 1996. Upregulation of Fas ligand expression by human immunodeficiency virus in human macrophages mediates apoptosis of uninfected T lymphocytes. J. Virol. 70:199– 206 Muro-Cacho CA, Pantaleo G, Fauci AS. 1995. Analysis of apoptosis in lymph nodes of HIV-infected persons. Intensity of apoptosis correlates with the general state of activation of the lymphoid tissue and not with stage of disease or viral burden. J. Immunol. 154:5555–66 Gandhi RT, Chen BK, Straus SE, Dale JK, Lenardo MJ, Baltimore D. 1998. HIV-1 directly kills CD4+ T cells by a Fasindependent mechanism. J. Exp. Med. 187:1113–22 Blair PJ, Boise LH, Perfetto SP, Levine BL, McCrary G, Wagner KF, St. Louis DC, Thompson CB, Siegel JN, June CH. 1997. Impaired induction of the apoptosis-protective protein Bcl-xL in activated PBMC from asymptomatic HIV-infected individuals. J. Clin. Immunol. 17:234–46 De Maria R, Boirivant M, Cifone MG, Roncaioli P, Hahne M, Tschopp J, Pallone F, Santoni A, Testi R. 1996. Functional expression of Fas and Fas ligand on human gut lamina propria T lymphocytes. A potential role for the acidic sphingomyelinase pathway in normal immunoregulation. J. Clin. Invest. 97:316–22 Waldner H, Sobel RA, Howard E, Kuchroo VK. 1997. Fas- and FasL-deficient mice are resistant to induction of autoimmune encephalomyelitis. J. Immunol. 159:3100–3 White CA, McCombe PA, Pender MP. 1998. The roles of Fas, Fas ligand and Bcl-2 in T cell apoptosis in the central nervous system in experimental autoimmune encephalomyelitis. J. Neuroimmunol. 82:47–55 Jacob CO, McDevitt HO. 1988. Tumour necrosis factor-alpha in murine autoimmune “lupus” nephritis. Nature 331:356– 58 Beutler B, Bazzoni F. 1998. TNF, apoptosis and autoimmunity: a common thread? Blood Cells Mol. Dis. 24:216–30 Heumann D, Le Roy D, Zanetti G, Eugster HP, Ryffel B, Hahne M, Tschopp J, Glauser MP. 1995. Contribution of TNF/ TNF receptor and of Fas ligand to toxicity in murine models of endotoxemia and bacterial peritonitis. J. Inflamm. 47:173– 79

235. Owen-Schaub LB, Yonehara S, Crump WL, Grimm EA. 1992. DNA fragmentation and cell death is selectively triggered in activated human lymphocytes by Fas antigen engagement. Cell. Immunol. 140:197–205 236. Peter ME, Kischkel FC, Scheuerpflug CG, Medema JP, Debatin KM, Krammer PH. 1997. Resistance of cultured peripheral T cells towards activation-induced cell death involves a lack of recruitment of FLICE (MACH/caspase 8) to the CD95 death-inducing signaling complex. Eur. J. Immunol. 27:1207–12 237. Bertin J, Armstrong RC, Ottilie S, Martin DA, Wang Y, Banks S, Wang GH, Senkevich TG, Alnemri ES, Moss B, Lenardo MJ, Tomaselli KJ, Cohen JI. 1997. Death effector domain-containing herpesvirus and poxvirus proteins inhibit both Fas- and TNFR1-induced apoptosis. Proc. Natl. Acad. Sci. USA 94:1172– 76 238. Thome M, Schneider P, Hofmann K, Fickenscher H, Meinl E, Neipel F, Mattmann C, Burns K, Bodmer JL, Schroter M, Scaffidi C, Krammer PH, Peter ME, Tschopp J. 1997. Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature 386:517–21 239. Hu S, Vincenz C, Ni J, Gentz R, Dixit VM. 1997. I-FLICE, a novel inhibitor of tumor necrosis factor receptor-1– and CD-95–induced apoptosis. J. Biol. Chem. 272:17255–57 240. Han D, Chaudhary PM, Wright ME, Friedman C, Trask BJ, Riedel R, Baskin D, Schwartz SM, Hood L. 1997. MRIT, a novel death-effector domain-containing protein, interacts with caspases abd BclXl and initiates cell death. Proc. Natl. Acad. Sci. USA 94:11333–38 241. Irmler M, Thome M, Hahne M, Schneider P, Hofmann K, Steiner V, Bodmer JL, Schroter M, Burns K, Mattmann C, Rimoldi D, French LE, Tschopp J. 1997. Inhibition of death receptor signals by cellular FLIP. Nature 388:190–95 242. Rasper DM, Vaillancourt JP, Hadano S, Houtzager VM, Seiden I, et al. 1998. Cell death attenuation by “Usurpin,” a mammalian DED-caspase homologue that precludes caspase-8 recruitment and activation by the CD-95 (Fas, APO-1) receptor complex. Cell Death Differ. 5:271– 88 243. Shu HB, Halpin DR, Goeddel DV. 1997. Casper is a FADD- and caspase-related inducer of apoptosis. Immunity 6:751–63 244. Van Parijs L, Peterson DA, Abbas AK. 1998. The Fas/Fas ligand pathway

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

MATURE T LYMPHOCYTE APOPTOSIS

245.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

246.

247.

248.

249.

250.

251.

252.

253.

254.

255. 256.

and Bcl-2 regulate T cell responses to model self and foreign antigens. Immunity 8:265–74 Roy N, Deveraux QL, Takahashi R, Salvesen GS, Reed JC. 1997. The c-IAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J. 16:6914– 25 Chu ZL, McKinsey TA, Liu L, Gentry JJ, Malim MH, Ballard DW. 1997. Suppression of tumor necrosis factor-induced cell death by inhibitor of apoptosis c-IAP2 is under NF-kappa-B control. Proc. Natl. Acad. Sci. USA 94:10057–62 Deveraux QL, Takahashi R, Salvesen GS, Reed JC. 1997. X-linked IAP is a direct inhibitor of cell-death proteases. Nature 388:300–4 Digby MR, Kimpton WG, York JJ, Connick TE, Lowenthal JW. 1996. ITA, a vertebrate homologue of IAP that is expressed in T lymphocytes. DNA Cell Biol. 15:981–88 Deveraux QL, Roy N, Stennicke HR, Van Arsdale T, Zhou Q, Srinivasula SM, Alnemri ES, Salvesen GS, Reed JC. 1998. IAPs block apoptotic events induced by caspase-8 and cytochrome c by direct inhibition of distinct caspases. EMBO J. 17:2215–23 Liston P, Roy N, Tamai K, Lefebvre C, Baird S, Cherton-Horvat G, Farahani R, McLean M, Ikeda JE, MacKenzie A, Korneluk RG. 1996. Suppression of apoptosis in mammalian cells by NAIP and a related family of IAP genes. Nature 379:349–53 Uren AG, Pakusch M, Hawkins CJ, Puls KL, Vaux DL. 1996. Cloning and expression of apoptosis inhibitory protein homologs that function to inhibit apoptosis and/or bind tumor necrosis factor receptor-associated factors. Proc. Natl. Acad. Sci. USA 93:4974–78 Boise LH, Minn AJ, Noel PJ, June CH, Accavitti MA, Lindsten T, Thompson CB. 1995. CD28 costimulation can promote T cell survival by enhancing the expression of Bcl-XL. Immunity 3:87–98 Boise LH, Thompson CB. 1997. Bcl-x(L) can inhibit apoptosis in cells that have undergone Fas-induced protease activation. Proc. Natl. Acad. Sci. USA 94:3759–64 Memon SA, Moreno MB, Petrak D, Zacharchuk CM. 1995. Bcl-2 blocks glucocorticoid- but not Fas- or activationinduced apoptosis in a T cell hybridoma. J. Immunol. 155:4644–52 Ashkenas J, Werb Z. 1996. Proteolysis and the biochemistry of life-or-death decisions. J. Exp. Med. 183:1947–51 Armstrong RC, Aja T, Xiang J, Gaur S,

257.

258.

259.

260.

261.

262.

263.

264.

265.

266.

267.

268.

249

Krebs JF, Hoang K, Bai X, Korsmeyer SJ, Karanewsky DS, Fritz LC, Tomaselli KJ. 1996. Fas-induced activation of the cell death-related protease CPP32 is inhibited by Bcl-2 and by ICE family protease inhibitors. J. Biol. Chem. 271:16850–55 Strasser A, Harris AW, Huang DC, Krammer PH, Cory S. 1995. Bcl-2 and Fas/APO-1 regulate distinct pathways to lymphocyte apoptosis. EMBO J. 14:6136–47 Critchfield JM, Lenardo MJ. 1995. Antigen-induced programmed T cell death as a new approach to immune therapy. Clin. Immunol. Immunopathol. 75: 13–19 Reap EA, Felix NJ, Wolthusen PA, Kotzin BL, Cohen PL, Eisenberg RA. 1995. bcl-2 transgenic Lpr mice show profound enhancement of lymphadenopathy. J. Immunol. 155:5455–62 Tamura A, Katsumata M, Greene MI, Yui K. 1996. Inhibition of apoptosis and augmentation of lymphoproliferation in bcl-2 transgenic Fas/Fas ligand-defective mice. Cell Immunol. 168:220–28 Beg AA, Baltimore D. 1996. An essential role for NF-kappa-B in preventing TNF-alpha-induced cell death. Science 274:782–84 Boothby MR, Mora AL, Scherer DC, Brockman JA, Ballard DW. 1997. Perturbation of the T lymphocyte lineage in transgenic mice expressing a constitutive repressor of nuclear factor (NF)-kappa-B. J. Exp. Med. 185:1897–907 Kelliher MA, Grimm S, Ishida Y, Kuo F, Stanger BZ, Leder P. 1998. The death domain kinase RIP mediates the TNFinduced NF–B signal. Immunity 8:297– 303 Lee SY, Reichlin A, Santana A, Sokol KA, Nussenzweig MC, Choi Y. 1997. TRAF2 is essential for JNK but not NF-kappa-B activation and regulates lymphocyte proliferation and survival. Immunity 7:703– 13 Sato T, Irie S, Kitada S, Reed JC. 1995. FAP-1: a protein tyrosine phosphatase that associates with Fas. Science 268:411–15 Yang X, Korshravi-Far, Chang HY, Baltimore D. 1997. Daxx, a novel Fas-binding protein activates Jnk and Apoptosis. Cell 89:1067–76 Chu K, Niu X, Williams LT. 1995. A Fasassociated protein factor, FAF1, potentiates Fas-mediated apoptosis. Proc. Acad. Natl. Sci. USA 92:11894–98 Ting AT, Pimentel-Muinos FX, Seed B. 1996. RIP mediates tumor necrosis factor

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

250

269.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

270.

271.

272.

273.

274.

275.

276.

277.

278.

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL receptor 1 activation of NF-κB but not Fas/APO-1-initiated apoptosis. EMBO J. 15:6189–96 Belizario JE, Dinarello CA. 1991. Interleukin 1, interleukin 6, tumor necrosis factor, and transforming growth factor beta increase cell resistance to tumor necrosis factor cytotoxicity by growth arrest in the G1 phase of the cell cycle. Cancer Res. 51:2379–85 Choe J, Kim HS, Zhang X, Armitage RJ, Choi YS. 1996. Cellular and molecular factors that regulate the differentiation and apoptosis of germinal center B cells. Anti-Ig down-regulates Fas expression of CD40 ligand-stimulated germinal center B cells and inhibits Fas-mediated apoptosis. J. Immunol. 157:1006–16 Estaquier J, Tanaka M, Suda T, Nagata S, Golstein P, Ameisen JC. 1996. Fasmediated apoptosis of CD4+ and CD8+ T cells from human immunodeficiency virus-infected persons: differential in vitro preventive effect of cytokines and protease antagonists. Blood 87:4959–66 Fuss IJ, Strober W, Dale JK, Fritz S, Pearlstein GR, Puck JM, Lenardo MJ, Straus SE. 1997. Characteristic T helper 2 T cell cytokine abnormalities in autoimmune lymphoproliferative syndrome, a syndrome marked by defective apoptosis and humoral autoimmunity. J. Immunol. 158:1912–18 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 Hornung F, Zheng L, Lenardo MJ. 1997. Maintenance of clonotype specificity in CD95/Apo-1/Fas-mediated apoptosis of mature T lymphocytes. J. Immunol. 159:3816–22 Wong B, Arron J, Choi Y. 1997. T cell receptor signals enhance susceptibility to Fas-mediated apoptosis. J. Exp. Med. 186:1939–44 Combadiere B, e Sousa CR, Germain RN, Lenardo MJ. 1998. Selective induction of apoptosis in mature T lymphocytes by variant T cell receptor ligands. J. Exp. Med. 187:349–55 Duke RC, Cohen JJ. 1986. IL-2 addiction: withdrawal of growth factor activates a suicide program in dependent T cells. Lymphokine Res. 5:289–99 Kuroda K, Yagi J, Imanishi K, Yan XJ, Li XY, Fujimaki W, Kato H, Miyoshi-Akiyama T, Kumazawa Y, Abe H, Uchiyama T. 1996. Implantation of IL-2-containing osmotic pump prolongs the survival of

279.

280.

281.

282.

283. 284.

285.

286.

287.

288.

289.

290.

superantigen-reactive T cells expanded in mice injected with bacterial superantigen. J. Immunol. 157:1422–31 Deng G, Podack ER. 1993. Suppression of apoptosis in a cytotoxic T-cell line by interleukin 2-mediated gene transcription and deregulated expression of the protooncogene bcl-2. Proc. Natl. Acad. Sci. USA 90:2189–93 Vasilakos JP, Ghayur T, Carroll RT, Giegel DA, Saunders JM, Quintal L, Keane KM, Shivers BD. 1995. IL-1 beta converting enzyme (ICE) is not required for apoptosis induced by lymphokine deprivation in an IL-2-dependent T cell line. J. Immunol. 155:3433–42 Boise LH, Gonzalez-Garcia M, Postema CE, Ding L, Lindsten T, Turka LA, Mao X, Nunez G, Thompson CB. 1993. bcl-x, a bcl-2-related gene that functions as a dominant regulator of apoptotic cell death. Cell 74:597–608 Strasser A, Harris AW, Cory S. 1991. bcl-2 transgene inhibits T cell death and perturbs thymic self-censorship. Cell 67:889–99 Yang E, Korsmeyer SJ. 1996. Molecular thanatopsis: a discourse on the BCL2 family and cell death. Blood 88:386–401 Ohta T, Kinoshita T, Naito M, Nozaki T, Masutani M, Tsuruo T, Miyajima A. 1997. Requirement of the caspase3/CPP32 protease cascade for apoptotic death following cytokine deprivation in hematopoietic cells. J. Biol. Chem. 272:23111–16 Broome HE, Dargan CM, Krajewski S, Reed JC. 1995. Expression of Bcl-2, Bcl-x, and Bax after T cell activation and IL-2 withdrawal. J. Immunol. 155:2311– 17 Chao DT, Linette GP, Boise LH, White LS, Thompson CB, Korsmeyer SJ. 1995. Bcl-XL and Bcl-2 repress a common pathway of cell death. J. Exp. Med. 182:821– 28 Vaux DL, Cory S, Adams JM. 1988. Bcl-2 gene promotes haemopoietic cell survival and cooperates with c-myc to immortalize pre-B cells. Nature 335:440–42 Linette GP, Hess JL, Sentman CL, Korsmeyer SJ. 1995. Peripheral T-cell lymphoma in lckpr-bcl-2 transgenic mice. Blood 86:1255–60 Strasser A, Harris AW, Bath ML, Cory S. 1990. Novel primitive lymphoid tumours induced in transgenic mice by cooperation between myc and bcl-2. Nature 348:331–33 Beltinger C, Kurz E, Bohler T, Schrappe M, Ludwig WD, Debatin KM. 1998.

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

MATURE T LYMPHOCYTE APOPTOSIS

291.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

292.

293.

294.

295.

296.

297.

298.

299.

300.

301.

CD95 (APO-1/Fas) mutations in childhood T-lineage acute lymphoblastic leukemia. Blood 91:3943–51 Natoli G, Costanzo A, Ianni A, Templeton DJ, Woodgett JR, Balsano C, Levrero M. 1997. Activation of SAPK/JNK by TNF receptor 1 through a noncytotoxic TRAF2- dependent pathway. Science 275:200–3 Tamiya S, Etoh K, Suzushima H, Takatsuki K, Matsuoka M. 1998. Mutation of CD95 (Fas/Apo-1) gene in adult T-cell leukemia cells. Blood 91:3935–42 Sneller MC, Wang J, Dale JK, Strober W, Middelton LA, Choi Y, Fleisher TA, Lim MS, Jaffe ES, Puck JM, Lenardo MJ, Straus SE. 1997. Clinical, immunologic, and genetic features of an autoimmune lymphoproliferative syndrome associated with abnormal lymphocyte apoptosis. Blood 89:1341–48 Karawajew L, Wuchter C, Ruppert V, Drexler H, Gruss HJ, Dorken B, Ludwig WD. 1997. Differential CD95 expression and function in T and B lineage acute lymphoblastic leukemia cells. Leukemia 11:1245–52 Friesen C, Fulda S, Debatin KM. 1997. Deficient activation of the CD95 (APO1/Fas) system in drug-resistant cells. Leukemia 11:1833–41 Kluck RM, Bossy-Wetzel E, Green DR, Newmeyer DD. 1997. The release of cytochrome c from mitochondria: a primary site for Bcl-2 regulation of apoptosis . Science 275:1132–36 Yang J, Liu XS, Bhalla K, Kim CN, Ibrado AM, Cai J, Peng TI, Jones DP, Wang X. 1997. Prevention of apoptosis by Bcl-2: release of cytochrome-c from mitochondria blocked. Science 275:1129–32 Liu XS, Kim CN, Yang J, Jemmerson R, Wang XD. 1996. Induction of apoptotic program in cell-free extracts: requirement for dATP and cytochrome c. Cell 86:147–57 Marchetti P, Castedo M, Susin SA, Zamzami N, Hirsch T, Macho A, Haeffner A, Hirsch F, Geuskens M, Kroemer G. 1996. Mitochondrial permeability transition is a central coordinating event of apoptosis. J. Exp. Med. 184:1155–60 Susin SA, Zamzami N, Castedo M, Hirsch T, Marchetti P, Macho A, Daugas E, Geuskens M, Kroemer G. 1996. Bcl-2 inhibits the mitochondrial release of an apoptogenic protease. J. Exp. Med. 184: 1331–41 Marchetti P, Hirsch T, Zamzami N, Castedo M, Decaudin D, Susin SA, Masse B, Kroemer G. 1996. Mitochondrial

302.

303.

304.

305.

306. 307.

308. 309.

310.

311.

312.

251

permeability transition triggers lymphocyte apoptosis. J. Immunol. 157:4830– 36 Susin SA, Zamzami N, Castedo M, Daugas E, Wang HG, Geley S, Fassy F, Reed JC, Kroemer G. 1997. The central executioner of apoptosis: multiple connections between protease activation and mitochondria in Fas/APO-1/CD95- and ceramide- induced apoptosis. J. Exp. Med. 186:25–37 Zou H, Henzel WJ, Liu X, Lutschg A, Wang X. 1997. Apaf-1, a human protein homologous to C. elegans CED-4, participates in cytochrome c-dependent activation of caspase-3. Cell 90:405–13 Li P, Nijhawan D, Buduhardjo I, Srinivasula S, Ahmad M, Alnemri E, Wang X. 1997. Cytochrome c and dATP-dependent formation of Apaf-1/Caspase-9 complex initiates an apoptotic protease cascade. Cell 91:479–89 Srinivasula SM, Ahmad M, FernandesAlnemri T, Litwack G, Alnemri ES. 1996. Molecular ordering of the Fas-apoptotic pathway: the Fas/APO-1 protease Mch5 is a CrmA-inhibitable protease that activates multiple Ced-3/ICE-like cysteine proteases. Proc. Acad. Natl. Sci. USA 93:14486–91 Reed JC. 1997. Cytochrome c: Can’t live with it–can’t live without it. Cell 91:559– 62 Vander Heiden MG, Chandel NS, Williamson EK, Schumacker PT, Thompson CB. 1997. Bcl-xL regulates the membrane potential and volume homeostasis of mitochondria [see comments]. Cell 91:627– 37 Hengartner MO. 1998. Apoptosis. Death cycle and Swiss army knives. Nature 391:441–42 Hengartner MO, Horvitz HR. 1994. The ins and outs of programmed cell death during C. elegans development. Philos. Trans. R. Soc. London Ser. B 345:243– 46 Rosse T, Olivier R, Monney L, Rager M, Conus S, Fellay I, Jansen B, Borner C. 1998. Bcl-2 prolongs cell survival after Bax-induced release of cytochrome c. Nature 391:496–99 Li F, Srinivasan A, Wang Y, Armstrong RC, Tomaselli KJ, Fritz LC. 1997. Cellspecific induction of apoptosis by microinjection of cytochrome c. Bcl-xL has activity independent of cytochrome c release. J. Biol. Chem. 272:30299–305 Croft M, Bradley LM, Swain SL. 1994. Naive versus memory CD4 T cell response to antigen. Memory cells are less

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

252

313.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

314.

315.

316.

317. 318. 319.

320.

321.

322.

323.

324.

325.

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

LENARDO ET AL dependent on accessory cell costimulation and can respond to many antigenpresenting cell types including resting B cells. J. Immunol. 152:2675–85 Veis DJ, Sorenson CM, Shutter JR, Korsmeyer SJ. 1993. Bcl-2-deficient mice demonstrate fulminant lymphoid apoptosis, polycystic kidneys, and hypopigmented hair. Cell 75:229–40 Motoyama N, Wang F, Roth KA, Sawa H, Nakayama K, Negishi I, Senju S, Zhang Q, Fujii S, et al. 1995. Massive cell death of immature hematopoieticcells and neurons in Bcl-x-deficient mice. Science 267:1506–10 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 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 Freitas AA, Rocha B. 1997. Lymphocyte survival: a red queen hypothesis. Science 277:1950 Neuberger MS. 1997. Antigen receptor signaling gives lymphocytes a long life. Cell 90:971–73 Lam KP, Kuhn R, Rajewsky K. 1997. In vivo ablation of surface immunoglobulin on mature B cells by inducible gene targeting results in rapid cell death. Cell 90:1073–83 Tanchot C, Rocha B. 1997. Peripheral selection of T cell repertoires: the role of continuous thymus output. J. Exp. Med. 186:1099–106 Kuo CT, Veselits ML, Leiden JM. 1997. LKLF: a transcriptional regulator of single-positive T cell quiescence and survival. Science 277:1986–90 Teague TK, Marrack P, Kappler JW, Vella AT. 1997. IL-6 rescues resting mouse T cells from apoptosis. J. Immunol. 158:5791–96 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 Horvitz HR, Shaham S, Hengartner MO. 1994. The genetics of programmed cell death in the nematode Caenorhabditis elegans. Cold Spring Harbor Symp. Quant. Biol. 59:377–85 Vaux DL, Weissman IL, Kim SK. 1992. Prevention of programmed cell death in

326.

327.

328.

329.

330.

331.

332.

333.

334.

335. 336.

Caenorhabditis elegans by human bcl-2. Science 258:1955–57 Straus SE, Lenardo MJ, Puck JM, Strober W, Sneller MC. 1998. Autoimmune Lymphoproliferative Syndrome: an inherited disorder of lymphocyte apoptosis. Ann. Int. Med. In press Rieux-Laucat F, Diest FL, Roberts IA, Debatin KM, Fisher A, Villartay JP. 1995. Mutations in Fas-associated with human lymphoproliferative syndrome and autoimmunity. Science 268:1347–51 Sneller MC, Straus SE, Jaffe ES, Jaffe JS, Fleisher TA, Stetler-Stevenson M, Strober W. 1992. A novel lymphoproliferative/autoimmune syndrome resembling murine lpr/gld disease. J. Clin. Invest. 90:334–41 Bettinardi A, Brugnoni D, Quiros-Roldan E, Malagoli A, La Grutta S, Correra A, Notarangelo LD. 1997. Missense mutations in the Fas gene resulting in autoimmune lymphoproliferative syndrome: a molecular and immunological analysis. Blood 89:902–9 Drappa J, Vaishnaw AK, Sullivan KE, Chu JL, Elkon KB. 1996. Fas gene mutations in the Canale-Smith syndrome, an inherited lymphoproliferative disorder associated with autoimmunity. N. Engl J. Med. 335:1643–49 Kasahara Y, Wada T, Niida Y, Yachie A, Seki H, Ishida Y, Sakai T, Koizumi F, Koizumi S, Miyawaki T, Taniguchi N. 1998. Novel Fas (CD95/APO-1) mutations in infants with a lymphoproliferative disorder. Int. Immunol. 10:195–202 Le Deist F, Emile JF, Rieux-Laucat F, Benkerrou M, Roberts I, Brousse N, Fischer A. 1996. Clinical, immunological, and pathological consequences of Fasdeficient conditions. Lancet 348:719– 23 Pensati L, Costanzo A, Ianni A, Accapezzato D, Iorio R, Natoli G, Nisini R, Almerighi C, Balsano C, Vajro P, Vegnente A, Levrero M. 1997. Fas/Apo1 mutations and autoimmune lymphoproliferative syndrome in a patient with type 2 autoimmune hepatitis. Gastroenterology 113:1384–89 Wu J, Wilson J, He J, Xiang L, Schur PH, Mountz JD. 1996. Fas ligand mutation in a patient with systemic lupus erythematosus and lymphoproliferative disease. J. Clin. Invest. 98:1107–13 Kimura M, Matsuzawa A. 1994. Autoimmunity in mice bearing lprcg: a novel mutant gene. Int. Rev Immunol. 11:193–210 Vidal S, Kono DH, Theofilopoulos AN. 1998. Loci predisposing to autoimmu-

P1: KKK/PLB

P2: NBL/SPD

February 11, 1999

11:32

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-08

MATURE T LYMPHOCYTE APOPTOSIS

337.

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

338.

339.

340. 341.

nity in MRL-Fas lpr and C57BL/6-Faslpr mice. J. Clin. Invest. 101:696–702 Ishigami T, White CA, Pender MP. 1998. Soluble antigen therapy induces apoptosis of autoreactive T cells preferentially in the target organ rather than in the peripheral lymphoid organs. Eur. J. Immunol. 28:1626–35 McFarland HI, Critchfield JM, Racke MK, Mueller JP, Nye SH, Boehme SA, Lenardo MJ. 1995. Amelioration of autoimmune reactions by antigen-induced apoptosis of T cells. Adv. Exp. Med. Biol. 383:157–66 Lehmann PV, Sercarz EE, Forsthuber T, Dayan CM, Gammon G. 1993. Determinant spreading and the dynamics of the autoimmune T-cell repertoire. Immunol. Today 14:203–8 Lenardo MJ. 1997. The molecular regulation of lymphocyte apoptosis. Semin. Immunol. 9:1–5 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

253

342. Ashton-Rickardt PG, Tonegawa S. 1994. A differential-avidity model for T-cell selection. Immunol. Today 15:362–66 343. Zuniga-Pflucker JC, Lenardo MJ. 1996. Regulation of thymocyte development from immature progenitors. Curr. Opin. Immunol. 8:215–24 344. Lenardo MJ. 1996. Fas and the art of lymphocyte maintenance. J. Exp. Med. 183:721–24 345. Malinin NL, Boldin MP, Kovalenko AV, Wallach D. 1997. MAP3K-related kinase involved in NF–B induction by TNF, CD95 and IL-1. Nature 385:540–44 346. Lee SY, Lee SY, Choi Y. 1997. TRAFinteracting protein (TRIP): a novel component of the tumor necrosis factor receptor (TNFR)- and CD30-TRAF signaling complexes that inhibits TRAF2mediated NF-κB activation. J. Exp. Med. 185:1275–85 347. del Peso L, Gonzalez-Garcia M, Page C, Herrera R, Nunez G. 1997. Interleukin-3induced phosphorylation of BAD through the protein kinase Akt. Science 278:687– 89

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:221-253. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:255–81 c 1999 by Annual Reviews. All rights reserved Copyright °

IMMUNOLOGIC BASIS OF ANTIGEN-INDUCED AIRWAY HYPERRESPONSIVENESS Marsha Wills-Karp Department of Environmental Health Sciences, Johns Hopkins School of Hygiene and Public Health, Baltimore, Maryland 21205; e-mail: [email protected] KEY WORDS:

allergy, asthma, IL-4, IL-13, IL-5, T lymphocytes, cytokines

ABSTRACT The incidence, morbidity, and mortality of asthma has increased worldwide over the last two decades. Asthma is a complex inflammatory disease of the lung characterized by variable airflow obstruction, airway hyperresponsiveness (AHR), and airway inflammation. The inflammatory response in the asthmatic lung is characterized by infiltration of the airway wall with mast cells, lymphocytes, and eosinophils. Although asthma is multifactorial in origin, the inflammatory process in the most common form of the disease (extrinsic asthma) is believed to be a result of inappropriate immune responses to common aero-allergens in genetically susceptible individuals. As such, it has been hypothesized that CD4+ T cells that produce a Th2 pattern of cytokines play a pivotal role in the pathogenesis of this disease. Through the release of cytokines such as IL-4, IL-13, and IL-5, these cells orchestrate the recruitment and activation of the primary effector cells of the allergic response, the mast cell and the eosinophil. Activation of these cells results in the release of a plethora of inflammatory mediators that individually or in concert induce changes in airway wall geometry and produce the symptoms of the disease. The aim of this review is to discuss our current understanding of the pathophysiologic mechanisms by which Th2 cytokines induce airway disease, and the factors that predispose to the generation of these pathogenic cells in response to inhalation of ubiquitous aero-allergens. Elucidation of the exact immunological basis for allergic asthma may yield immunotherapeutic strategies to reverse the development of pathogenic Th2-mediated immune responses and reduce the morbidity and mortality associated with this disease.

255 0732-0582/99/0410-0255$08.00

P1: APR/SPD

P2: NBL/plb

January 21, 1999

256

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

INTRODUCTION Asthma is a chronic inflammatory disease of the lung which has been increasing in prevalence, morbidity, and mortality over the last two decades. Despite this increase in the disease, considerable progress has been made in our understanding of the pathophysiology underlying the disease process. Although it was once thought that asthma was a disease of airway smooth muscle, over the last decade it has become widely accepted that asthma is primarily an inflammatory disease. Structurally, the airways of asthmatics are characterized by the presence of chronic inflammation with intense infiltration of the bronchial mucosa by lymphocytes, eosinophils, and mast cells, along with epithelial desquamation, goblet cell hyperplasia, and thickening of the submucosa (1, 2). These cellular findings have consistently been associated with the main physiologic abnormalities of the disease, including variable airflow obstruction and AHR (3–9). Although asthma is multifactorial in origin, atopy, the genetic predisposition for the development of an IgE-mediated response to common aero-allergens, is the strongest identifiable predisposing factor for the development of asthma. Most childhood asthma is allergic in nature and referred to as extrinsic asthma (10, 11). In this form of the disease, the inflammatory process is thought to arise as a result of inappropriate immune responses to commonly inhaled allergens. After repeated low-dose exposure to allergens, atopic individuals develop specific IgE antibodies to the allergens. Subsequent exposure to allergens initiates a secondary humoral response. When the sequence of events following an allergen provocation are examined in allergic volunteers, the allergic reaction can be divided into both early (within minutes) bronchospastic responses and late (hours after exposure) inflammatory responses (Figure 1). The early (immediate) response is characterized by rapid onset of mucosal edema, increases in airway smooth muscle tone, and airway narrowing, associated with mast cell degranulation (12). Some allergic asthmatics also develop late-phase responses that begin three to six hours after antigen challenge and may persist for several days in the absence of therapy. In these responses, airway narrowing is associated with the migration of neutrophils, eosinophils, and lymphocytes from the blood into lung parenchyma and airway epithelium (12–15). This has led to the concept that the immediate response after allergen challenge is mediated by mast cells, whereas eosinophils are the predominant cells in the late asthmatic reaction. It has recently become appreciated that as orchestrators of the inflammatory response, T lymphocytes, in particular CD4 T cells, play a pivotal role in the pathogenesis of asthma. This article reviews our current understanding of the pathogenesis of this disease with particular emphasis placed on the pathogenic

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T CELLS IN ASTHMA

257

Figure 1 The early- and late-phase reactions to inhaled antigen and the consequence of these reactions on pulmonary physiology. Reprinted from Nadel and Busse, Am. J. Respir. Crit. Care Med. 157:S130–S138.

role of Th2 cytokine-producing cells. Because our fullest understanding of the mechanisms involved in the inflammatory response in the airways is of the allergen-driven asthmatic reaction, the primary focus is placed on extrinsic, or atopic asthma. Results from experiments on humans will be cited when available, although much of our current understanding of the disease has been derived from study of animal models, particularly murine models in which manipulation of immunological processes is possible.

CD4 T CELLS IN ASTHMA As the primary orchestrator of the specific immune response, the T lymphocyte has been implicated in the pathogenesis of allergic airway disease (16–19). In support of a role for T lymphocytes in the pathogenesis of asthma, consistently elevated numbers of T lymphocytes have been found in the bronchoalveolar lavage (BAL) fluids and bronchial biopsies from asthmatics (16–19). These T cells are predominantly of the CD4 population, based on the fact that the numbers of CD8 cells are equivalent to those in normal individuals. Cell recruitment to the airways is suggested by the finding that increased numbers of CD4+ T cells in BAL are concurrent with decreased numbers of CD4+ T cells in peripheral blood following allergen challenge (20). Experimental data support a

P1: APR/SPD

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

258

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP

generalized increase in T cell activation in asthmatics, with increased expression of interleukin-2 receptor (IL-2R), class II histocompatibility antigens (HLADR), and very late activation antigen-1 (VLA-1) in blood, BAL, and bronchial biopsies (19, 21). Functional subsets of CD4+ T cells have been distinguished at both clonal and population levels by the unique profiles of cytokines that they produce (22, 23). The differential presence of these cytokine phenotypes in a variety of allergic and infectious diseases both in mice and in humans has provided descriptive power and theoretical insight into disease pathogenesis (23–27). Th1 cells produce interleukin-2, TNF-β, and interferon gamma (IFN-γ ), and are critical in the development of cell-mediated immunity (22, 23). On the other hand, Th2 cells produce IL-4, IL-13, IL-5, IL-9, IL-6, and IL-10 and are important in the stimulation of IgE production, mucosal mastocytosis, and eosinophilia (22, 23). A possible immunopathogenic role for Th2 cells in asthma has been postulated on the role that these cytokines play in IgE synthesis and eosinophil regulation. Several lines of evidence from experiments provide support for the involvement of these cytokines in the pathogenesis of asthma. Firstly, the T cells from asthmatic patients express a unique pattern of cytokines consistent with a Th2 pattern (21, 28, 29). Specifically, T cells from both the BAL fluids and bronchial biopsies of allergic asthmatics express elevated levels of mRNA for IL-4, IL-13, GM-CSF, and IL-5 (21, 28, 29). Secondly, it has been shown that successful steroid treatment increases IFN-γ levels in the BAL of asthmatics patients, and simultaneously decreases IL-4 and IL-5 levels (30). Thirdly, the increased numbers of activated T cells observed in asthmatics correlate with the numbers of activated eosinophils, the magnitude of the decrement in peak expiratory flow rates, and the severity of the disease (31, 32). The production of this pattern of cytokines appears to be genetically controlled and established early during childhood (33). In an elegant longitudinal study of children, Martinez and colleagues (33) show that the propensity to develop asthma is associated with low stimulated levels of IFN-γ in children at 9 months of age, which suggests that a type-1 response is a protective factor. Furthermore, low stimulated levels of IL-2 and IFN-γ at 9 months of age were positively correlated with parental immediate skin test reactivity. In this regard, genetic linkage studies by multiple groups have shown, although not conclusively, that asthma is linked to a region of human chromosome 5q in which the genes for IL-4, IL-5, and IL-13 are localized (34, 35). Additional support for this hypothesis as it relates to humans has been provided in a population-based study described by Shirakawa (36). The study found that among Japanese schoolchildren there was a strong inverse association between delayed-type hypersensitivity to Bacillus Calmette-Guerin (BCG) and atopy. Positive tuberculin responses predicted a lower incidence of asthma,

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T CELLS IN ASTHMA

259

lower serum IgE levels, and cytokine profiles biased toward a type-1 profile. These results suggest that exposure and response to BCG may, by modification of immune cytokine profiles, inhibit atopic disorders. Although considerable descriptive evidence suggests that CD4+ T lymphocytes and Th2 cytokines are important in the pathogenesis of airway hyperresponsiveness in asthmatic humans, definitive proof is difficult to obtain in human subjects. Therefore, experimental animal models have been extremely useful in contributing to a delineation of the role of CD4+ T cells and T-cell-derived cytokines in the pathogenesis of asthma. Direct evidence of a causal role for CD4+ T cells in the development of AHR was first provided in a murine model of allergen-induced AHR (37). In this model, ovalbumin sensitization and local lung challenge resulted in increases in airway responsiveness to acetylcholine challenge, and in an eosinophilic inflammatory response in the lung. Depletion of CD4+ T cells in sensitized mice prior to local lung antigen challenge with specific monoclonal antibodies prevented the development of allergen-induced allergic airway responses. Conversely, depletion of CD8 cells has been shown not to affect airway responses to allergen challenge in mice (38). Definitive evidence of a pathogenic role for Th2 cytokines in allergen-driven pathophysiologic processes has been provided by studies in which IL-4 and IL-5 have been manipulated through either antibody blockade (39, 40) or gene targeting (41–43). Furthermore, AHR can be induced in naive mice by adoptive transfer of Th2 clones into their lungs (44). On the other hand, studies have shown that the administration of agents such as IL-12 and IFN-γ that inhibit Th2 cytokine production and stimulate Th1 pathways prevent the development of antigen-induced AHR and inflammation in murine models (44, 45). Along these lines, prior infection of sensitized mice with BCG (46) or administration of CpG oligodeoxynucleotides (47) have resulted in suppression of eosinophilic airway inflammation concomitant with a shift in cytokine production to a protective type-1 profile.

INDUCTION OF TH2 CELL DIFFERENTIATION IN ALLERGIC ASTHMA Although considerable evidence supports a pathogenic role for Th2 cytokines in asthma, very little is known about the underlying cause(s) of the aberrant expansion of Th2 cells. The data from studies using T cell receptor transgenic mice (48, 49) suggest that naive T helper cells have the ability to differentiate into either Th1 or Th2 cells. Naive T cells generate IL-2 when activated by antigen presentation and costimulatory signals (i.e. B7 molecules, CD58, and CD40) and progress through a multipotential cell that generates a mixed spectrum of cytokines including IL-2, IL-4, and IFN-γ (50). Although many

P1: APR/SPD

P2: NBL/plb

January 21, 1999

260

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

immune responses are likely to retain the (Th0) phenotype, responses may be polarized to either the Th1 or Th2 type, a process markedly influenced by the nature and dose of the antigen and the cytokine milieu during initial priming by specific antigens (51). The elegant use of mice transgenic for a specific alpha/beta T cell receptor has provided details of this critical involvement of the cytokine environment on subset differentiation during naive lymphocyte priming (49, 50). In these models, IL-12 directly primes CD4+ T cells for Th1 differentiation (50). Th2 differentiation, in turn, is critically dependent upon IL-4 (49).

Role of Specific Allergens Recent studies indicate that a number of allergens from diverse sources have enzymatic activity that may subvert the immune response toward the Th2 phenotype. For example, Der p1 is a 25 kDa cysteine protease which has been shown to cleave CD25, the 55 kDa alpha subunit of the IL-2R (52, 53). As a result of cleavage of CD25, peripheral blood T cells show markedly diminished proliferation and IFN-γ secretion in response to potent stimulation by anti-CD3 Ab. These findings indicate that Der p1 could upset the balance of Th1/Th2 subset distribution by decreasing growth and expansion of the Th1 subset and as a consequence augmenting expansion of the Th2 subset that favors a pro-allergic response. Der p1 may also contribute to the allergic phenotype by cleaving CD23 on B cells that would normally serve to inhibit IgE synthesis and thereby disrupt the IgE reulatory mechanism (54). Through its ability to disrupt epithelial architecture, Der p1 may also facilitate its own passage across the epithelium, thus enhancing its (own) access to immune cells. Although allergens such as Der p1 can potentially create a microenvironment conducive to Th2 cell expansion, normal individuals do not mount Th2 responses to these allergens, which suggests that despite the nature of these antigens other factors also contribute to the allergic outcome following inhalation of aero-allergens in susceptible individuals.

Antigen Presentation in the Asthmatic Lung A pivotal step in induction of a T cell immune response is the uptake, processing, and presentation of Ag by professional antigen-presenting cells (APCs). In the lung, several potential APCs exist, including alveolar macrophages (AMs), epithelial cells, and dendritic cells DCs. Early studies focused on the accessory cell capacities of AMs, but numerous studies have shown that AMs are poor accessory cells (55). In fact, AMs have been shown to inhibit mitogen-stimulated T cell proliferation. Airway epithelial cells have also been thought to have antigen-presenting capabilities. Because epithelial cells line the respiratory tract, they are ideally

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T CELLS IN ASTHMA

261

located to encounter antigens and present them to T cells. A role for airway epithelium as APCs has been suggested by the observation that APCs express many of the costimulatory molecules important in antigen presentation such as MHCII, CD40, B7 molecules, and ICAM-1 (55a). However, to date there is no definite evidence of their role in antigen presentation in the lung. The most important professional APC in the lung is probably DC. It has been shown that DCs form an extensive network above the basement membrane of the airway epithelium that ensures accessibility to inhaled antigens (56). It is hypothesized that upon encountering inhaled antigen, airway DCs migrate to the draining lymph nodes of the lung, upregulate expression of costimulatory ligands, and interact with naive T lymphocytes, initiating a primary immune response. Lambrecht et al (57) have demonstrated the important role of DC in the lung for development of allergic airways responses in an elegant study of mice. In the study, the herpes simplex virus type-1 thymidine kinase suicide gene was overexpressed in cells of DC myeloid lineage. Activation of this gene was accomplished by treating mice with the nucleotide analogue ganciclovir prior to lung challenge with allergen. In sensitized thymidine kinase transgenic mice, the infiltration of lymphocytes and eosinophils into the airways following local antigen challenge was almost completely abolished, which illustrated the reliance of antigen presentation in the lung on DCs. Several lines of evidence suggest that DC function in the asthmatic lung may be altered. For example, the number of DCs recovered in the BAL is greater in asthmatic patients than in normal controls (58). Moreover, van den Heuvel (59) demonstrated that when stimulated in vitro under the same conditions, circulating monocytes (MOs) from atopic subjects develop into more potent accessory cells than those from normal control subjects. Although the mechanism(s) behind the increased accessory function are currently unknown, it is possible that there are intrinsic differences in monocyte derived dendritic cell populations between atopic and normal individuals, or that perhaps they respond differently to the cytokines IL-4 and GM-SCF used to expand them in vitro. Further evidence for altered DC function in asthmatics is provided by the demonstration that the proportion of DCs expressing the alpha subunit of FC epsilon RI (FcεRI) is significantly increased in asthmatics as compared to non-atopic controls (60). This increase in FcεRI on DCs may also be the result of higher endogenous levels of IL-4, a known modulator of immunoglobulin Fc receptors. Although the role of this receptor on DCs is not known, it is assumed that FcεRI-alpha expression by DCs may facilitate the capture and internalization of allergens, which can then be processed and presented in the context of MHCII to CD4 T lymphocytes. If this mechanism is used in an asthmatic individual already sensitized to a particular allergen, FcεRI+ DCs may exacerbate the chronic allergic inflammatory response by ensuring efficient accumulation

P1: APR/SPD

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

262

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP

of allergen-specific T cells. Another observation that suggests DC may contribute to the allergic phenotype is that systemic steroid treatment dramatically inhibits the recruitment of DC into the respiratory tract epithelium during acute inflammation (61). Although collectively these studies suggest a role for DC in augmented presentation of allergens to T cells, they do not explain the preferential expansion of CD4 T cells into Th2-producing cells in the asthmatic lung. In this regard, murine DCs have been shown to produce IL-12 in response to CD40/CD40L interactions, with a resulting differentiation of naive T cells toward a Th1 phenotype (62). However, to my knowledge murine DCs do not produce IL-4. DC cells are, however, capable of driving differentiation of CD4 T cells toward a Th2 phenotype when inhibitors of IL-12 such as IL-4, IL-10, and PGE2 have been exogenously added to cultures, which suggests that alterations in the production of these mediators in the microenvironment are more likely than an intrinsic defect in the DC to be the primary defect in asthma. However, since the study of DC function is in its infancy, it remains to be seen whether DCs exert a primary influence over expansion of Th2 cells in the asthmatic lung.

Cytokine Regulation of Th2 Differentiation in the Asthmatic Lung Because IL-4 and IL-12 are known regulators of T cell differentiation, alterations in either production of or responsiveness to these cytokines could result in the polarization of T cell responses to allergens observed in the asthmatic lung. The differentiation of uncommitted T cell precursors into Th2 cells is largely driven by IL-4. Although IL-4’s role in the development of acute Th2 responses and maintenance of the response has been recently disputed (63), it is still likely that IL-4 plays an important role in the initial priming of T cells. Indeed, the results of several studies suggest that perhaps genes regulating IL-4 production may be altered by asthma. Firstly, a specific polymorphism in the IL-4 gene has been shown to correlate with high serum IgE levels and enhanced IL-4 gene expression (64). Secondly, Hershey et al (65) have demonstrated that expression of a mutant form of the IL-4 receptor α chain in allergic patients is associated with increased IL-4 signalling. Lastly, studies in mice have shown that alterations in a number of the family members of the NFAT protein family may result in altered IL-4 gene expression and polarization of T cell responses toward the type-2 pattern. NFAT proteins are expressed in T cells, B cells, and mast cells, and control the transcription of a number of genes relevant to asthma, including IL-4. Rao and colleagues (66) recently demonstrated that NFAT-p (NFAT-1)-deficient mice have an increased number of eosinophils in the bone marrow and blood concomitant with increased production of Th2

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T CELLS IN ASTHMA

263

cytokines. These experiments suggest that deficiencies in this transcription factor may lead to the development of the allergic phenotype. It has also been shown that other NFAT family members, including NF-ATC1 (NF-ATc), are important in differentiation of T cells. Several groups have shown that loss of NF-ATc activity results in impaired T lymphocyte activity and secretion of IL-4 in mice (67, 68). Taken together, these studies suggest that altered gene expression of the NFAT proteins may result in an imbalance in the regulation of genes important for Th2 differentiation. Collectively, these results suggest that the known genetic predisposition of allergic asthma in humans may involve dysregulation of the genes controlling IL-4 production or the IL-4 receptor. Although Th2 cell polarization in asthma can clearly arise as a result of aberrant expression of the genes important in Th2 differentiation, alterations in factors controlling expansion of the opposing Th1 pathways may also play an important role. In this regard, interleukin-12, a product of MAs and other APCs, is the primary determinant of T cell differentiation to a Th1 pattern (69). Studies of IL-12 production in the lungs of human asthmatics do indeed suggest that impaired IL-12 production occurs, but it is not known at the present time whether this is a primary or secondary event. Naseer and colleagues (70) demonstrated that the number of cells positive for IL-12 p40 mRNA is significantly lower in asthmatic patients than in normal controls. Furthermore, successful steroid treatment was characterized by a significant increase in the numbers of cells expressing IL-12 mRNA, whereas steroid therapy in steroid-resistant patients did not result in an increase in IL-12-expressing cells. Additionally, another group has shown that Staphylococcus aureus (SAC)-induced production of IL-12 p70 in whole blood cultures from asthmatic patients was significantly less in comparison with non-atopic control subjects (71). Further support for the importance of IL-12 in prevention of antigen-induced allergic airway responses has been provided by the observation that blockade of endogenous production of IL-12 in naturally resistant murine strains (C3H/HeJ) renders them susceptible to the development of allergen-induced AHR and eosinophilic inflammation (72). Taken together, these studies suggest that dysregulation of endogenous IL-12 levels may be an important mechanism governing the pathogenesis of allergic airway disorders. The mechanisms which give rise to alterations in IL-12 production are unclear at the present time; however, several potential mechanisms exist, such as altered expression of the genes encoding either one or both of the individual subunits of the functional cytokine, or alterations in receptor signalling pathways. Although specific polymorphisms in the genes encoding p40 and p70 have not been reported, the gene encoding the p40 subunit of the functional heterodimer is located in the region of human chromosome 5q that has previously been linked to human asthma. In addition, recent studies in mice have implied

P1: APR/SPD

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

264

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP

that loss of IL-12 responsiveness and subsequent inability of mice to mount Th1 responses is due to altered expression of the IL-12 receptor B2 subunit (73). IL-12-dependent signalling in human Th1 cells was shown to correlate with the selective expression of the transcripts encoding the signalling component of the IL-12 receptor B2 and with the presence of high-affinity IL-12 binding sites selectively located on Th1 cells (74). Thus it is proposed that not only is IL-12RB2 a marker of Th1 cells, but that lack of expression of this receptor subunit may lead to a Th2-polarized immune response. To my knowledge, IL-12 receptor expression has not been examined in T cells from human asthmatics. As IL-12 production is deficient in the lungs of human asthmatics, we might propose that selective expression and regulation of the IL-12RB2 subunit may be a potential mechanism by which aberrant Th2 cell differentiation occurs in allergic diseases. Alternatively, the deficient production of IL-12 in asthma may occur as a result of altered regulation of IL-12 production by mediators and cytokines, which either positively (IFN-γ ) (75), or negatively (IL-4, PGE2) (76, 77) regulate its production. Clearly, IL-4 and PGE2 levels are elevated in allergic diseases, whereas IFN-γ levels are reduced. Thus a better understanding of the mechanisms underlying dysregulation of IL-12 production and responsiveness awaits studies designed to examine IL-12 receptor expression and modulation of IL-12 production in asthma.

MECHANISMS OF TH2 CYTOKINE-INDUCED EOSINOPHILIC INFLAMMATION AND AHR Although an immunopathogenic role for Th2 cells is suggested by the role of IL-4, IL-13, and IL-5 in IgE synthesis and eosinophil differentiation and activation, the exact mechanisms by which Th2 cytokines mediate eosinophilic inflammation and subsequent AHR are still not entirely clear. Th2 cytokines may potentially induce AHR through direct or indirect effects on B cells, mast cells, and eosinophils (Figure 2). To my knowledge, no one has examined the direct effects of IL-4, IL-13, or IL-5 on airway reactivity in in vitro systems that are relatively devoid of cellular, neural, and humoral components. Of the uniquely Th2 cytokines, IL-4 likely plays a particularly important role in the allergic diathesis. In addition to its pivotal role in Th2 differentiation of naive T cells into Th2-producing cells, IL-4 also induces a plethora of cellular responses that are potentially important in the development of allergic airway diseases including: (a) its role in B cell-class switching to IgE production (78); (b) in conjunction with IL-3, it is a growth factor for mast cells (79); and (c) its ability to upregulate vascular cell adhesion molecule-1 (VCAM-1) expression which leads to preferential migration of eosinophils into tissues (80).

P2: NBL/plb

January 21, 1999 13:49

QC: KKK/uks

Annual Reviews

Figure 2 Proposed mechanism of the role of CD4+ T cells in the pathogenesis of asthma.

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

P1: APR/SPD T1: NBL

AR078-09

T CELLS IN ASTHMA

265

P1: APR/SPD

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

266

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP

Initial studies in IL-4 mice supported a pivotal role for IL-4 in both the early- and late-phase allergic response. However, recent studies have shown that blockade of IL-4 prior to antigen challenge in sensitized mice is not effective in attenuating AHR or eosinophilic inflammation (40). Although these results were initially puzzling, the interpretation at the time was that perhaps IL-4 was essential for the initial expansion of Th2 cytokine-producing cells, but was not required for maintenance of this response. In contrast, studies conducted by Gavett et al (81) demonstrated that blockade of the IL-4 receptor prior to antigen provocation in sensitized mice effectively inhibited both AHR and eosinophilic accumulation. Although these discrepant results were initially difficult to resolve, recent studies in Stat6-deficient animals have provided a potential explanation. Kuperman et al (82) showed that a deficiency in the Stat6 molecule, which mediates most of the cellular actions of both IL-4 and IL-13, abolished antigen-induced eosinophilic inflammation and AHR. These results indicated that perhaps the effectiveness of IL-4 receptor blockade was due to inhibition of IL-13-mediated processes and not to those mediated by IL-4. Although these results suggest that IL-13 rather than IL-4 plays the central role in driving allergic responses, direct proof of this hypothesis is not available at this time. Because IL-4 and IL-13 have many overlapping actions, further studies will be necessary to determine the exact contributions of each of these mediators to the allergic response. Therefore, in the following discussion regarding the role of IL-4 in allergic responses, I point out the instances of IL-13 implication. One of the potential effector mechanisms by which IL-4 may induce allergic inflammation and airway function changes is in its role as the primary inducer of immunoglobulin-class switching in B cells that leads to the synthesis and secretion of IgE (78). The combination of IL-4’s effects on IgE synthesis and mast cell growth suggests that its role in AHR may be mediated primarily through mast cell activation. As will be discussed in more detail, IgE activation of mast cells leads to the synthesis and release of a number of inflammatory mediators that may contribute to the bronchoconstriction, vascular changes, and mucus changes observed in the early-phase response to allergen challenge. Excessive production of airway mucus glycoproteins is found consistently in the lungs of asthmatics, and in particular in those patients who die in status asthmatics (83). Evidence of a role for IL-4 in mucus production is provided by studies conducted in IL-4 transgenic mice that exhibit profound increases in mucus-containing cells in the airway epithelium (84). This fact is supported by the finding that antigen-induced increases in mucus-containing cells is ablated by blockade of the IL-4 receptor (81). Also consistent with this hypothesis is the demonstration that IL-4 induces expression of the MUC5 gene in the airways of IL-4 transgenic mice (84). The effects of IL-4 on goblet-cell hyperplasia may be mediated through IL-4’s involvement in mast cell activation and eosinophil

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T CELLS IN ASTHMA

267

recruitment since both of these effector cells release the potent mucus secretogues LT, 15-HETES, and PAF. Interestingly, Cohn et al (85) demonstrated that although mucus changes could be induced via transfer of Th2 clones into the airways, clones derived from IL-4-deficient mice could still confer mucus production, which suggests that Th2 cells are important for mucus production but that IL-4 is not essential. Taken together with the finding that the airways of Stat6-deficient animals do not show evidence of mucus production following allergen provocation, we may conclude that IL-13 is the type-2 cytokine mediating mucus changes in the airways. This speculation is supported by studies in N. brasiliensis-infected IL-13−/− mice in which the authors demonstrate that IL-13 mediates parasite-induced goblet-cell hyperplasia (86). The relative contribution of IL-4 and IL-13 to allergen-driven mucus hypersecretion in asthma remains to be determined. IL-4 may induce AHR by way of another mechanism—its potential role in eosinophil recruitment. Numerous studies have shown that IL-4 blockade eliminates antigen-induced increases in eosinophils (42, 43, 81). Although the exact mechanisms involved in IL-4’s ability to induce tissue eosinophilia are unknown, there are several potential mechanisms. IL-4 may mediate pulmonary eosinophilia through its role in Th2 cell differentiation and the subsequent production of IL-5. In addition, IL-4 may regulate eosinophil influx by regulating VCAM-1 expression on the endothelium and/or stimulating release of specific chemokines from resident airway cells (87). In support of a role for IL-4/IL-13mediated VCAM-1 in pulmonary eosinophilia, numerous studies have shown that VCAM-1 is necessary for eosinophil recruitment into the lung in response to antigen provocation in mice (88). Because interleukin-5 has been shown to be the primary determinant of eosinophil differentiation, activation, and survival, it is a likely candidate in the development of AHR (89). In bone marrow, IL-5 is important for stimulation of eosinophilopoiesis and promotion of the terminal differentiation of myeloid precursors into eosinophils. IL-5 also increases eosinophil adhesion to vascular endothelial cells, promotes the migration of eosinophils from the blood into tissues, prolongs eosinophil survival in tissues, and augments the cytotoxic activity of eosinophils. Furthermore, IL-5 may activate pulmonary eosinophils and cause them to release cytotoxic products. As stated above, although IL-5 is derived from CD4 T cells, more recent studies show that it is released by mast cells and by eosinophils themselves (90). The importance of IL-5 in antigeninduced eosinophilia has been well established in several species (39, 41, 91). For example, blockade of endogenous IL-5 levels in antigen-sensitized guinea pigs (91) and in mice (39) has resulted in significant suppression of both BAL and tissue eosinophilia. Consistent with these observations is the demonstration that intratracheal administration of IL-5 induces eosinophil accumulation in the

P1: APR/SPD

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

268

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP

guinea pig lung in vivo (91). More definitely, mice in which the IL-5 gene has been disrupted do not develop aero-allergen-induced pulmonary eosinophilia (41). Reconstitution of these mice with IL-5 by administration of recombinant vaccinia viruses engineered to express IL-5 completely restored aero-allergeninduced eosinophilia. If taken at face value, this result would suggest that IL-5 alone is sufficient for restoration of tissue eosinophilia. However, it has been shown that overexpression of IL-5 in mice results in a marked elevation of circulating eosinophils, but that the tissue levels of eosinophils remain similar to their wildtype controls (92). These results suggest that other processes also contribute in the recruitment of eosinophils to sites of inflammation in the lung. Chemokines, or chemotactic cytokines, attract and activate lymphocytes, granulocytes, and monocytes. They are 7–16 kDa proteins with four conserved cysteines. Based on the position of the first two cysteines, these chemotactic cytokines are divided into two subfamilies which differ in terms of what cells they attract. The C-C chemokines (including eotaxin, RANTES, MCP-3, and MCP-1) are particularly relevant to allergic inflammation. Elevations in a number of chemokines including MCP-1, RANTES, and eotaxin have been shown in the lungs of asthmatic patients (93). One chemokine of particular importance in airway disease is eotaxin because it has been shown to be exquisitely selective for eosinophil recruitment (94). Studies show that increases in eotaxin mRNA-positive cells are inversely correlated to the dose of histamine, and induce a 20% drop in forced expiratory volume in one second in asthmatic patients (93). Furthermore, eotaxin and IL-5 have been shown to cooperate in the orchestration of eosinophil accumulation in tissues because IL-5 increases responsiveness to eotaxin (95). Although eotaxin probably plays an important role in lung eosinophilia, studies have shown that neutralization of this cytokine results only in partial suppression of antigen-driven eosinophil migration (96). In turn, these results suggest that other chemokines may also be involved in controlling eosinophil chemotaxis in the lung. In this regard, mMCP-5, a chemokine produced by lung MAs and smooth muscle cells, has also been implicated in eosinophil recruitment into the murine lung (97). Several lines of evidence suggest that Th2 cells regulate chemokine expression in the lung following antigen challenge. Specifically, MacLean and colleagues (98) have demonstrated that anti-CD3 mAb administration prior to antigen challenge in sensitized mice was associated with a reduction in eotaxin mRNA expression. The demonstration that IL-4 stimulates eotaxin production in fibroblasts provides support for this conclusion (87). Similarly, mMCP-5 expression in mast cells has been shown to be lymphocyte-dependent. Thus we may postulate that eotaxin, mMCP-5, and other chemokines yet to be defined may act in a sequential manner to regulate migration of eosinophils to the lung during an asthmatic response.

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T CELLS IN ASTHMA

269

Based on the independent roles of IL-4 and IL-5 discussed above, we would envision the following paradigm for regulation of allergen-induced pulmonary eosinophilia by CD4 Th2 cytokines: Following antigen-specific induction of Th2 cytokine production, IL-5 would rapidly induce differentiation of eosinophils from myeloid precursors in the bone marrow and stimulate their release into the bloodstream. IL-4 and/or IL-13 would promote eosinophil egress from the vascular compartment by upregulating VCAM-1 expression on vascular endothelial cells and by stimulating chemokine production (eotaxin and MCP-5) by endothelial cells, epithelial cells, and smooth muscle cells in the lung. Once the cells accumulate in the tissues, locally produced IL-5 would promote their actions by prolonging their survival in tissues. Within this paradigm, allergendriven eosinophil recruitment into the lung is coordinately regulated by the Th2 cytokines, IL-4, IL-13, and IL-5.

ROLE OF EFFECTOR CELLS IN AHR It is considered that the principal effector cells of the allergic airway response are mast cells and eosinophils. Although these two cells have unique functions and release a unique profile of mediators, they also produce an overlapping array of mediators which are known to contribute to the allergic diathesis. It has long been thought that mast cells contribute to the early-phase response and that eosinophils mediate the last-phase response and the structural changes, regardless of the fact that the recent discovery of mast-cell production of a number of chemokines and cytokines may dispel this notion. Next we explore the individual role of each of these effector cells in the pathophysiology of allergic asthma.

Mast Cells Although an increase in mast cell numbers in the lungs of asthmatics has not been demonstrated, numerous studies have shown evidence of mucosal mastcell activation in the lungs of asthmatics (99). Mast cells are thought to contribute to the pathogenesis of allergic airway responses through IgE-dependent mechanisms. It is hypothesized that IgE produced by allergen-reactive B cells binds to FcεR receptors present on the surface of mast cells and basophils, and that when challenged with allergen, these cells release vasoactive mediators as well as chemotactic factors and cytokines that promote leukocyte infiltration and exacerbate the inflammatory response. Through the production and release of these proinflammatory molecules, mast cells set in motion a series of events that result in the immediate response to inhaled antigens and may also contribute to the late-phase allergic response (LPR). Mast cell release of pre-formed mediators such as histamine, PGD2, LTC4, and PAF within minutes after inhalation of antigens together produce the

P1: APR/SPD

P2: NBL/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

270

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP

symptoms of the early response to allergen challenge. Each of these mediators is found in increased levels in the lungs of asthmatics, and can enhance activation of inflammatory cells, cause microvascular leakage, increase mucus production, and induce bronchoconstriction (99). Although it has long been thought that mast cells only contribute to the early acute response (EAR), recent evidence demonstrates their potential contribution to the late-phase response. Through the elaboration of the cytokines TNF-α, IL-1, IL-3, GM-CSF, IFN-γ , IL-4, IL-5, IL-6, and IL-8, IL-16 and the chemokines MCP-1, MIP-1B, MIP-1a, and RANTES, mast cells may contribute to the cellular component of the late-phase response. They may also perpetuate or amplify the eosinophilic inflammatory process by preferentially expressing IL-4. Indeed, immunohistochemical studies of bronchial biopsies of allergic patients suggest that mast cells are the primary source of IL-4 protein (100). Thus, mast cells may favor the acquisition of the Th2 phenotype by providing a continuously high local concentration of IL-4. Mast cell mediators may also contribute to the airway wall remodeling observed in the asthmatic lung since many of the mediators they release influence connective tissue turnover. Specifically, histamine and tryptase have been shown to stimulate fibroblast growth and collagen synthesis in vitro and in vivo (101). Support of a role for mast cells in the LPR comes from studies in which mast cell stabilizers such as sodium nedocromil effectively inhibit both the EAR and the late-phase response to allergen exposure. Further support of this tenet is provided by the finding that anti-IgE pre-treatment does not elicit AHR in mast-cell deficient mice (102). Recent clinical trials utilizing a monoclonal antibody against IgE resulted in modest improvements in the lung function of asthmatics (103). However, IgE-deficient mice can develop antigen-induced AHR, which indicates that although mast cells are capable of inducing AHR, mechanisms other than IgE-mediated mast cell degranulation are also important in the development of AHR (104).

Eosinophils Eosinophilic inflammation is clearly a hallmark of both allergic and non-allergic asthma, and considerable evidence suggests that there is an association between pulmonary eosinophil infiltration and AHR in human asthma (105). Eosinophils release a myriad of mediators and cytokines that are potentially important in the allergic response including: (a) the eosinophil-specific proteins, eosinophilic cationic protein (ECP), and major basic protein (MBP), and eosinophil-derived neurotoxin (EDN); (b) a number of cytokines (IL-2, IL-3, IL-4, IL-5, GM-CSF, IL-6, IL-10, IFN-γ , IL-12, TGF-α, and TGF-β); and (c) lipid mediators such as LTs, PGE2, and PAF (90).

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

T CELLS IN ASTHMA

271

Eosinophils are postulated to induce AHR through the actions of these basic proteins on the airway wall. In fact, ECP and MBP have been detected in the BAL of patients with asthma (106). These proteins are cytotoxic to the airway epithelium. Damage to the airway epithelium may lead to AHR by removing enzymes important in the degradation of neuropeptides and/or in the loss of epithelial-derived relaxing factor. In addition, these proteins may have direct effects since MBP has been shown to potentiate contraction of airway tissues in vitro without showing evidence of epithelial damage (107). Further evidence for this hypothesis is provided by the fact that administration of MBP to primates induces AHR (108). MBP may induce AHR through its demonstrated ability to competitively inhibit binding of M2 receptors to acetylcholine autoreceptors on parasympathetic nerves that may result in increased release of acetylcholine (109). The importance of eosinophils in M2-receptor dysfunction has been shown in a guinea pig model of asthma. For example, blockade of inflammatory cells, especially eosinophils, prevents M2-receptor dysfunction in antigen-challenged guinea pigs (110). Further support for this hypothesis is provided by the fact that blockade of IL-5 in antigen-sensitized animals also protects the M2-muscarinic receptor function (111). Thus the demonstrated presence of eosinophil-derived proteins in the asthmatic lung, coupled with their potent actions on airway epithelial cells and/or neural receptors, strongly suggests a role for these mediators in the development of allergen-induced AHR and airway obstruction. Although there is a substantial body of evidence supporting the role of eosinophils in asthma, a number of studies on humans have not confirmed a correlation between eosinophils and disease severity (112). A similar conflict also exists in mouse studies of allergic asthma. Studies in IL-5 gene knockout mice show that both eosinophilia and AHR are IL-5-dependent (41). Results from a study in which IL-5 transgenic mice over-express IL-5 in the lung also support the role of IL-5 and eosinophils in the development of AHR (113). In contrast, Corry et al (114) demonstrated that anti-IL-4 treatment of mice prevented AHR in spite of the presence of eosinophils in the lung. Also, the development of AHR in that model was not affected by an anti-IL-5 treatment that abrogated eosinophilia. Conflicting results in murine models may be explained by the recent finding that eosinophils in the lungs of certain strains of mice do not degranulate following antigen challenge. In addition, since the role of eosinophils in AHR has mostly been based on IL-5 depletion, it is conceivable that IL-5 has other actions that induce AHR in addition to the recruitment of eosinophils. In support of this hypothesis, Kraneveld et al (115) have shown that IL-5 induces AHR through release of tachykinins in the guinea pig airway, an action which is independent of its effect on eosinophils.

P1: APR/SPD

P2: NBL/plb

January 21, 1999

272

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

MECHANISMS OF INFLAMMATION-INDUCED AHR Although the studies to date have not elucidated the precise mechanisms by which Th2 cytokines mediate AHR, they have suggested that either eosinophils, mast cells, or both are likely to play a role in this process. These cells are thought to induce the physiologic changes of asthma by means of the secretion of an array of inflammation mediators including histamine, leukotrienes, PAF, eosinophil-derived basic proteins, and various proteases. Because mast cells and eosinophils both produce many of these mediators, the importance of each cell type has been difficult to assess. This uncertainty is highlighted by the fact that the murine studies discussed in this review show that AHR can be induced in the absence of either one or the other cell type. Since these cell types release an overlapping array of mediators, it may be that development of AHR is achieved through separate pathways that result in the release of a common mediator such as leukotrienes. Alternately, AHR can be induced by two or more independent processes that both result in AHR by means of different mechanisms. Perhaps the varying severity of the disease is due to involvement of multiple independent processes working in concert and resulting in a more severe phenotype. The existence of two separable clinical phenotypes would support this conclusion. For example, extrinsic asthma, possibly in concert with IL-5-mediated eosinophil responses, is most certainly an IgE-dependent process. On the other hand, although the inciting agent in intrinsic patients is unknown, this pathway is probably not IgE-dependent, but rather IL-5- and eosinophil-dependent. Genetic linkage studies conducted to date would also support the hypothesis that multiple independent or additive pathways contribute to the asthmatic phenotype because asthma has clearly been shown to be a multigenic disease. Thus, in the next sections, I explore the potential mechanisms by which many of the above-mentioned mediators induce AHR without explicit regard for the origin of the mediators.

Bronchoconstrictor Effects of Inflammatory Mediators A variety of inflammatory mediators released by mast cells and eosinophils directly constrict airways. Among these are histamine, leukotrienes, PAF, endothelin, and eosinophil granule proteins. The contractile effects of these agents are mediated via direct stimulation of specific receptors on smooth muscle and through mechanisms including induction of the release of other contractile agonists. Because each of these of mediators is increased in the airways of asthmatics, one could envision airway smooth muscle bathed in a virtual soup of bronchoconstrictors, resulting in hyperresponsiveness. However, blockade of each mediator independently has not convincingly shown that they contribute to AHR, with one exception: leukotrienes. A more likely explanation

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

T CELLS IN ASTHMA

273

for these observations is that inflammation induces alterations in the airway wall geometry that convey stimuli to an already thickened airway wall, conceivably resulting in an even greater narrowing of the airway for a given stimulus. These airway changes include: (a) altered neural regulation of airway tone; (b) increases in muscle content; (c) increased mucus secretion; (d ) airway wall edema; and/or (e) airway epithelial desquamation. Evidence exists to support each of these possibilities.

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Altered Neural Regulation of Airway Tone The neuronal pathways controlling airway tone include the sympathetic, parasympathetic, and peripheral sensory nerves. A number of inflammatory mediators are known to alter function of these neural pathways. First, the eosinophil-derived MBP has been shown to alter release of the neurotransmitter acetylcholine from the parasympathetic nerve terminal by binding the M2 autoreceptor on the nerve ending (109). This leads to an increase in the release of acetylcholine and an enhanced narrowing of the airway wall for any given stimulus. In addition, responsiveness to sympathetic input regulating relaxation of the airway via ß-adrenergic receptors on airway smooth muscle has also been shown to be impaired in asthmatics (116). One hypothesis regarding the mechanism of this defect is that the cytokines IL-1 and TNF-α alter β-adrenoceptor signalling mechanisms and lead to impaired ability of the smooth muscle to relax (117). In this regard, polymorphisms in the β-adrenergic receptor have been shown to correlate with the severity of the disease (118). Inflammation is also associated with enhanced release of a number of sensory neuropeptides including substance P and neurokinins that possibly augment reflex constriction of airway smooth muscle.

Airway Smooth Muscle Hypertrophy Another alteration in airway smooth muscle noted in the lungs of asthmatics and experimental animals is an increase in airway smooth muscle mass (ASM). Recent evidence suggests a role for several inflammatory mediators in the induction of muscle growth through hyperplasia. In Brown-Norway rats for example, ovalbumin sensitization and repeated challenge induced an increase in the quantity of ASM in the airways (119). This increase was inhibited by administration of nedocromil sodium prior to antigen challenge. Nedocromil sodium may inhibit airway smooth muscle hyperplasia by inhibiting a variety of mediators released by mast cells including histamine, leukotrienes, 15-HETES, and endothelin that have each been shown to increase growth of ASM. In addition, IL-1 and IL-6, both present in the asthmatic airway, can potentially increase the proliferation of ASM cells (120). Thus, numerous mediators of the asthmatic response may contribute to the increased muscle mass observed in patients

P1: APR/SPD

P2: NBL/plb

January 21, 1999

274

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP

with asthma. This increase in muscle mass may theoretically contribute to an exaggerated narrowing of the airways to a variety of environmental stimuli.

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Mucus Hypersecretion Excessive production of airway mucus glycoproteins is a consistent finding in the lungs of asthmatics, in particular in those patients who die in status asthmatics. There is evidence for hyperplasia of submucosal glands and increased numbers of epithelial goblet cells in the lungs of asthmatics. The increased secretory response is most likely due to a combination of effects of inflammatory mediators (i.e. IL-4, IL-13, LT, PAF) on submucosal glands, neural elements controlling submucosal glands and/or on goblet cells (121). Although the exact contribution of mucus plugging of the airways to AHR has not been determined experimentally, the fact that mucus plugging is a consistent feature of fatal asthma suggest that it plays an important role in the pathogenesis of asthma.

Airway Wall Edema Airway edema is thought to be a result of inflammatory-induced increases in airway microvascular permeability. Inflammatory mediators released in the airway wall during the asthmatic response including histamine, PAF, and leukotrienes are potent inducers of increased bronchial permeability and are thus promoters of bronchial edema and airway wall swelling. It is hypothesized that thicker airway walls due to edema would theoretically result in greater airway narrowing for the same degree of airway smooth muscle contraction. However, studies designed to examine the direct effects of airway wall edema in the absence of inflammation have shown that edema in and of itself does not induce AHR (122). Thus the correlation of airway reactivity in asthmatics with airway wall edema may more accurately reflect the increase in inflammatory cells and mediators which accompany inflammation-induced microvascular leakage rather than a physical effect on airway smooth shortening.

Airway Epithelial Desquamation Airway epithelial damage correlates with airway hyperreactivity and appears to be mediated by inflammation of the airways, particularly due to the presence of eosinophils. In atopic asthmatics, positive correlations were observed between the concentrations of MBP, the numbers of desquamated epithelial cells in BAL fluid, and the degree of bronchial hyperresponsiveness (123). Loss of airway epithelium may contribute to allergen-induced obstruction of airways by exposing irritant receptors of nerves, which may increase the response of the airways to various stimuli, or by inducing the release of a variety of inflammatory mediators PGs, 15-lipoxygenase products, ET-1, nitric oxide, and several cytokines (IL-6, GM-SCF, TNF-α and RANTES) which either directly induce changes in the airway or elicit the recruitment of inflammatory cells. Furthermore, loss of

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

T CELLS IN ASTHMA

275

the epithelial barrier may provide greater access of allergens to dendritic cells and specific immune cells such as T and B cells, serving to amplify the inflammatory process. Thus the disruption or loss of the airway epithelium through a variety of mechanisms may play a pivotal role in the development of AHR, airway inflammation, and airflow obstruction.

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

CONCLUSION In summary, asthma is a complex inflammatory lung disease which is multifactorial in origin. However, it likely originates from an abnormal specific immune response to inhaled antigens. The presentation of inhaled allergens to CD4 T cells in the lungs of susceptible individuals results in the production of the Th2 cytokines, IL-4, IL-13, and IL-5, which orchestrate the differentiation, recruitment, and activation of mast cells and eosinophils in the airway mucosa. Such effector cells release a plethora of inflammatory mediators (e.g. histamine, LTs, PAF, eosinophil-derived basic proteins, and proteases) which have overlapping effects on the airway wall. Individually or in concert, these mediators cause acute bronchoconstriction, disruption of the airway epithelial layer, alterations in neural control of airway tone, increased mucus production, and increased smooth muscle mass. Each of these consequences of the inflammatory process could conceivably induce AHR, but it is more likely that they occur in combination. Although the mechanisms underlying the aberrant production are unclear, current studies suggest that perhaps altered regulation of genes controlling either IL-4/IL-13 or IL-12 production may lead to the expansion of deleterious Th2 cells in response to otherwise harmless inhaled antigens. It is hoped that as we gain a better understanding of the underlying pathophysiology of the disease, this knowledge will lead to the development of immunotherapeutics strategies to reduce or prevent the morbidity and mortality associated with this disease. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. McFadden ER, Gilbert IA. 1992. Asthma. N. Engl. J. Med. 327:1928–37 2. Beasley R, Roche WR, Roberts JA, Holgate ST. 1989. Cellular events in the bronchi in mild asthma and after bronchial provocation. Am. Rev. Respir. Dis. 139:806–17 3. Holgate ST. 1997. Asthma: a dynamic disease of inflammation and repair. Ciba Found. Symp. 206:5–28

4. Cockcroft D, Ruffin RE, Dolovich J, Hargreave FE. 1977. Allergen-induced bronchial reactivity. Clin. Allergy 7:503– 13 5. O’Byrne PM. 1998. Allergen-induced airway hyperresponsiveness. J. Allergy Clin. Immunol. 81:119–27 6. Casale TB, Wood D, Richerson HB. 1987. Elevated bronchoalveolar lavage fluid histamine levels in allergic asth-

P1: APR/SPD

P2: NBL/plb

January 21, 1999

276

7.

8.

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

9.

10.

11.

12.

13.

14.

15. 16. 17.

18.

19.

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP matics are associated with methacholine bronchial hyperresponsiveness. J. Clin. Invest. 79:1197–1203 Metgzer WJ, Richerson HB, Worden K, Monick M, Hunninghake GW. 1986. Bronchoalveolar lavage of allergic asthmatic patients following allergen bronchoprovocation. J. Clin. Invest. Chest 89: 477–83 Snapper JR. 1990. Inflammation and airway function: the asthma syndrome. Am. Rev. Respir. Dis. 141:531–33 Alexander AG, Barnes NC, Kay AB. 1992. Trial of cyclosporin in corticosteroid-dependent chronic severe asthma. Lancet 339:324–28 Cockcroft DW, Killian DN, Mellon JJ, Hargreave FE. 1977. Bronchial reactivity to inhaled histamine: A method and clinical survey. Clin. Allergy 7:235–43 Sears MR, Burrows B, Flannery EM, Herbison G, Hewitt CJ, Hodaway MD. 1991. Relation between airway responsiveness and serum IgE in children with asthma and in apparently normal children. N. Engl. J. Med. 325:1067–71 Pauwels R. 1989. The relationship between airway inflammation and bronchial hyperresponsiveness. Clin. Exp. Allergy 19:395–98 Bousquet J, Chanez P, Lacoste JY, Barneon G, Ghavanian N, Enander I, Verge P, Ahlstedt S, Simony-La fontaine J, Godard P, Michel FB. 1990. Eosinophilic inflammation in asthma. N. Engl. J. Med. 323:1033–39 De Monchy JGR, Kauffman HF, Venge P, Koeter GH, Jansen HM, Sluiter HJ, De Vries K. 1985. Bronchoalveolar eosinophilia during allergen-induced late asthmatic reactions. Am. Rev. Respir. Dis. 131:373–6 Cockcroft DW. 1989. Airway hyperresponsiveness and late asthmatic responses. Chest 94:178–80 Lemanske RF Jr, Busse WW. 1997. Asthma. J. Am. Med. Assoc. 278:1855– 73 Gerblich AA, Salik H, Schuyler MR. 1991. Dynamic T-cell changes in peripheral blood and bronchoalveolar lavage after antigen bronchoprovocation in asthmatics. Am. Rev. Respir. Dis. 143:533–37 Azzawi M, Bradley B, Jeffery PK, Frew AJ, Assoufi B, Collins JV, Durham S, Kay AB. 1990. Identification of activated T lymphocytes and eosinophils in bronchial biopsies in stable atopic asthmatics. Am. Rev. Respir. Dis. 142:1407–13 Corrigan CJ, Kay AB. 1990. CD4+ Tlymphocyte activation in acute severe

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

asthma: relationship to disease severity and atopic status. Am. Rev. Respir. Dis. 141:970–77 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 Walker C, Bode E, Boer L, Hansel TT, Blaser K, Virchow JC Jr. 1992. Allergic and nonallergic asthmatics have distinct patterns of T-cell activation and cytokine production in peripheral blood and bronchoalveolar lavage. Am. Rev. Respir. Dis. 146:109–15 Mosmann TR, Cherwinski H, Bond MW, Gieldin 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 Street NE, Mosmann TR. 1991. Functional diversity of T-lymphocytes due to secretion of different cytokine patterns. FASEB J. 5:171–77 Peltz G. 1991. A role for CD4+ T-cell subsets producing a selective pattern of lymphokines in the pathogenesis of human chronic inflammatory and allergic diseases. Immunol. Rev. 123:23–35 Kapsenberg ML, Wierenga EA, Bos JD, Jansen HM. 1991. Functional subsets of allergen-reactive human CD4+ T cells. Immunol. Today 12:392–95 Del Prete GF, de Carli M, Mastromauro C, Biagiotti R, Macchia D, Falagiani P, Ricci M, Romagnani S. 1991. Purified protein derivative of Mycobacterium tuberculosis and excretory-secretory antigen(s) of Toxocara-canis expand in vitro human T-cells with stable and opposite (type-1 T-helper or type 2-T-helper) profile of cytokine production. J. Clin. Invest. 88: 346–50 Sher A, Fiorentino D, Caspar P, Pearce E, Mosmann TR. 1991. Production of IL-10 by CD4+ lymphocytes correlates with down-regulation of Th1 cytokine synthesis in helminth infections. J. Immunol. 147:2713–16 Del Prete GF, De Carli M, D’Elios MM, Maestrelli P, Ricci M, Fabbri L, Romagnini S. 1993. Allergen exposure induces the activation of allergen-specific Th2 cells in the airway mucosa of patients with allergic respiratory disorders. Eur. J. Immunol. 23:1445–49 Robinson DS, Hamid Q, Ying S, Tsicopoulos A, Barkans J, Bentley AM, Corrigan C, Durham SR, Kay AB. 1992.

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

T CELLS IN ASTHMA

30.

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

31.

32.

33.

34.

35.

36.

37.

38.

39.

Predominant TH2-like bronchoalveolar T-lymphocyte population in atopic asthma. N. Engl. J. Med. 326:298–304 Wilkinson JR, Lane SJ, Lee TH. 1991. The effects of corticosteroids on cytokine generation and expression of activation antigens by monocytes in bronchial asthma. Int. Arch. Allergy Clin. Immunol. 94:220–21 Umetsu DT, DeKruyff RH. 1997. Th1 and Th2 CD4+ cells in the pathogenesis of allergic diseases. Proc. Soc. Exp. Biol. Med. 215:11–20 Robinson DS, Hamid Q, Bentley A, Ying S, Kay AB, Durham SR. 1993. Activation of CD4+ T cells, increased TH2-type cytokine mRNA expression, and eosinophil recruitment in bronchoalveolar lavage after allergen inhalation challenge in patients with atopic asthma. J. Allergy Clin. Immunol. 92:313–24 Martinez FD, Stern DA, Wright AL, Holberg CJ, Taussig LM, Halonen M. 1995. Association of interleukin-2 and interferon-gamma production by blood mononuclear cells in infancy with parental allergy skin tests and with subsequent development of atopy. J. Allergy Clin. Immunol. 96:652–60 Marsh DG, Neely JD, Breazeale DR, Ghosh B, Freidhoff LR, Ehrlich-Kautzky E, Schou C, Krishnaswany G, Beaty TH. 1994. Linkage analysis of IL4 and other chromosome 5q31.1 markers and total serum immunoglobulin E concentrations. Science 264:1152–56 Postma DS, Bleecker ER, Amelung PJ, Holroyd KJ, Xu J, Panhuysen CI, Meyers DA, Levitt RC. 1995. Genetic susceptibility to asthma: bronchial hyperresponsiveness coinherited with a major gene for atopy. N. Engl. J. Med. 333:894–900 Shirakawa T, Enomoto T, Shimazu S, Hopkin JM. 1997. The inverse association between tuberculin responses and atopic disorder. Science 275:77–79 Gavett SH, Chen X, Finkelman F, WillsKarp M. 1994. Depletion of murine CD4+ T lymphocytes prevents antigeninduced airway hyperreactivity and pulmonary eosinophilia. Am. J. Respir. Cell Mol. Biol. 10:587–93 Gonzalo J-A, Lloyd CM, Kremer L, Finger E, Martinez C, Siegelman MH, Cybulsky Guiterrez-Ramos J-A. 1996. Eosinophil recruitment to the lung in a murine model of allergic inflammation. The role of T cells, chemokines, and adhesion receptors. J. Clin. Invest. 98:2332–45 Kung TT, Stelts DM, Zurcher JA, Adams GK, Egan RW, Kreutner W, Watnick AS,

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

277

Jones H, Chapman RW. 1995. Involvement of IL-5 in a murine model of allergic pulmonary inflammation: prophylactic and therapeutic effect of an anti-IL-5 antibody. Am. J. Respir. Cell. Mol. Biol. 13:360–65 Coyle AJ, Le Gros G, Bertrand C, Tsuyuki S, Heusser CH, Kopf M, Anderson GP. 1995. Interleukin-4 is required for the induction of lung Th2 mucosal immunity. Am. J. Respir. Cell. Mol. Biol. 13:54–59 Foster PS, Hogan SP, Ramsay AJ, Matthaei KI, Young IG. 1996. Interleukin 5 deficiency abolishes eosinophilia, airways hyperreactivity and lung damage in a mouse asthma model. J. Exp. Med. 183:195–210 Brusselle G, Kips J, Joos G, Bluethmann H, Pauwels R. 1995. Allergen-induced airway inflammation and bronchial responsiveness in wild-type and interleukin4-deficient mice. Am. J. Respir. Cell. Mol. Biol. 12:254–59 Lukacs NW, Strieter RM, Chensue SW, Kunkel SL. 1994. Interleukin-4dependent pulmonary eosinophil infiltration in a murine model of asthma. Am. J. Respir. Cell. Mol. Biol. 10:526–32 Li X, Chopra RK, Chou T, Schofield BH, Wills-Karp M, Huang S. 1996. Mucosal IFN-γ gene transfer inhibits pulmonary allergic responses in mice. J. Immunol. 157:3216–19 Lack G, Bradley KL, Hamelmann E, Renz Lader HJ, Leung DYM, Larsen G, Gelfand EW. 1996. Nebulized IFN-γ inhibits the development of secondary allergic responses in mice. J. Immunol. 157:1432–39 Gavett SH, O’Hearn D, Li X, Huang S, Finkelman FD, Wills-Karp M. 1995. Interleukin 12 inhibits antigen-induced airway hyperresponsiveness, inflammation, and Th2 cytokine expression in mice. J. Exp. Med. 182:1–10 Erb KJ, Holloway JW, Sobeck A, Moll H, Le Gros G. 1998. Infection of mice with Mycobacterium bovis-Bacillus Calmette-Guerin (BCG) suppresses allergeninduced airway eosinophilia. J. Exp. Med. 187:561–69 Kline JN, Waldschmidt TJ, Businga TR, Lemish JE, Weinstock JV, Thorne PS, Krieg AM. 1998. Modulation of airway inflammation by CpG oligodeoxynucleotides in a murine model of asthma. J Immunol. 160:2555–59 Seder RA, Paul WE, Davis MM, Fazekas de St Groth B. 1992. The presence of interleukin 4 in vitro priming determines the lymphokine-producing potential of

P1: APR/SPD

P2: NBL/plb

January 21, 1999

278

50.

51.

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

52. 53.

54.

55.

55a.

56.

57.

58.

59.

60.

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP CD4+ T cells from T cell receptor transgenic mice. J. Exp. Med. 176:1091–98 Hsieh CS, Macatonia SE, Tripp CS, Wolf SF, O’Garra A, Murphy KM. 1993. Development of Th1 CD4+ T cells through IL-12 produced by Listeriainduced macrophages. Science 260:547– 49 Locksley RM. 1994. Th2 cells: Help for helminths. J. Exp. Med. 179:1405–7 Thomas WR. 1993. Mite allergens groups I-VII. A catalogue of enzymes. Clin. Exp. Allergy 23:350–53 Schulz O, Sewell HF, Shakib F. 1998. Proteolytic cleavage of CD25, the alpha subunit of the human T cell interleukin 2 receptor, by Der p 1, a major mite allergen with cysteine protease activity. J. Exp. Med. 187:271–75 Shakib F, Schulz O, Sewell H. 1998. A mite subversive: cleavage of CD23 and CD25 by Der p 1 enhances allergenicity. Immunol. Today 19:313–16 Lee TH, Lane SJ. 1992. The role of macrophages in the mechanisms of airway inflammation in asthma. Am. Rev. Respir. Dis. 145:S27–S30 Nakajima J, Ono M, Takeda M, Kawauchi M, Furuse A, Takizawa H. 1997. Role of costimulatory molecules on airway epithelial cells acting as alloantigen presenting cells. Transplant. Proc. 29:2297– 300 Holt PG, Schon-Hegrad MA, Oliver J, Holt BJ, McMenamin PG. 1990. A contiguous network of dendritic antigenpresenting cells within the respiratory epithelium. Int. Arch. Allergy Appl. Immunol. 91:155–59 Lambrecht BN, Salomon B, Klatzman D, Pauwels RA. 1998. Dendritic cells are required for the development of chronic eosinophilic airway inflammation in response to inhaled antigen in sensitized mice. J. Immunol. 160:4090–97 Bellini A, Vittori E, Martin M, Ackerman V, Mattoli S. 1993. Intraepithelial dendritic cells and selective activation of Th2like lymphocytes in patients with atopic asthma. Chest 103:997–1005 Van den Heuvel MM, Vanhee DDC, Postmus PE, Hoefsmit ECM, Beelen RHJ. 1998. Functional and phenotypic differences of monocyte-derived dendritic cells from allergic and nonallergic patients. J. Allergy Clin. Immunol. 101:90–95 Tunon-de-lara JM, Redington AE, Bradding P, Church MK, Hartley JA, Semper AE, Holgate ST. 1996. Dendritic cells in normal and asthmatic airways: expression of the alpha subunit of the high affin-

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

ity immunoglobulin E receptor (FcεRIalpha). Clin. Exp. Allergy 26:648–55 Holt PG, Thomas JA. 1997. Steroids inhibit uptake and/or processing but not presentation of antigen by airway dendritic cells. Immunology 91:145–50 Macatonia SE, Hosken NA, Litton M, Viera Physical Chemistry, Hsieh CS, Culpepper JA, Wysocka M, Trinchieri G, Murphy KM, O’Garra A. 1995. Dendritic cells produced IL-12 and direct the development of Th1 cells from naive CD4+ T cells. J. Immunol. 154:5071–79 Huang H, Hu-Li J, Chen H, Ben-Sasson SZ, Paul WE. 1997. IL-4 and IL-13 production in differentiated T helper 2 cells is not IL-4 dependent. J. Immunol. 15:3731– 38 Rosenwasser LJ, Klemm DJ, Dresback JK, Inamura H, Mascali JJ, Klinnert M, Borish L. 1995. Promoter polymorphisms in the chromosome 5 gene cluster in asthma and atopy. Clin. Exp. Allergy 25(Suppl.2):74–78 Hershey GKK, Friedrich MF, Esswein LA, Thomas ML, Chatila TA. 1997. The association of atopy with a gain-offunction mutation in the subunit of the interleukin-4 receptor. N. Engl. J. Med. 337:1720–25 Viola JPB, Kiani A, Bozza PT, Rao A. 1998. Regulation of allergic inflammation and eosinophilic recruitment in mice lacking the transcription factor NFAT1: Role of Interleukin-4 (IL-4) and IL-5. Blood 91:2223–30 Yoshida H, Nishina H, Takimoto H, Marengere LEM, Wakeham AC, Bouchard D, Kong Y-Y, Ohteki T, Shahinian A, Bachmann M, Ohashi PS, Penninger JM, Crabtree GR, Mak TW. 1998. The transcription factor NF-ATc1 regulates lymphocyte proliferation and Th2 cytokine production. Immunity 8:115–24 Ranger AM, Hodge MR, Gravallese EM, Oukka M, Davidson L, Alt FW, de la Brousse FC, Hoey T, Grusby M, Glimcher LH. 1998. Delayed lymphoid repopulation with defects in IL-4-driven responses produced by inactivation of NF-AT-c. Immunity 8:125–34 Trinchieri G. 1994. Interleukin-12: A cytokine produced by antigen-presenting cells with immunoregulatory functions in the generation of T-helper cells type 1 and cytotoxic lymphocytes. Blood 84:4008– 27 Naseer T, Minshall EM, Leung DYM, Laberge S, Ernst P, Martin RJ, Hamid Q. 1997. Expression of IL-12 and IL-13 mRNA in asthma and their modulation in

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

T CELLS IN ASTHMA

71.

72.

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

73.

74.

75.

76.

77.

78.

79.

80.

81.

response to steroid therapy. Am. J. Respir. Crit. Care Med. 155:845–51 Van der Pouw Kraan TCTM, Boeije LCM, De Groot ER. 1997. Reduced production of IL-12 and IL-12-dependent IFN-γ release in patients with allergic asthma. J. Immunol. 158:5560–65 Keane-Myers M, Wysocka M, Trinchieri G, Wills-Karp M. 1998. Resistance to antigen-induced airway hyperresponsiveness requires endogenous production of interleukin-12. J. Immunol. 162:919–26 Szabo SJ, Jacobson NG, Dighe AS, Gubler U, Murphy KM. 1995. Developmental commitment to the Th2 lineage by extinction of IL-12 signalling. Immunity 2:665–75 Rogge L, Barneris-Maino L, Biffi M, Oassini N, Presky DH, Gubler U, Sinigaglia F. 1997. Selective expression of an interleukin-12 receptor component by human T helper 1 cells. J. Exp. Med. 185:825–31 Ma X, Chow JM, Gri G, Carra G, Gerosa F, Wolf SF, Dzialo R, Trinchieri G. 1996. The interleukin-12 gene promoter is primed by interferon-gamma in monocytic cells. J. Exp. Med. 183:147–57 D’Andrea A, Ma X, Aste-amezaga M, Paganin C, Trinchieri G. 1995. Stimulatory and inhibitory effects of interleukin (IL)-4 and IL-13 on the production of cytokines by human peripheral blood mononuclear cells: Priming for IL12 and tumor necrosis factor-α production. J. Exp. Med. 181:537–46 Van der Pouw Kraan TCTM, Boeije LCM, Smeenk RJT, Wijdenes J, Aarden LA. 1995. Prostaglandin-E2 is a potent inhibitor of human interleukin 12 production. J. Exp. Med. 181:775–79 Finkelman FD, Katona IM, Urban JF Jr, Holmes J, Ohara J, Tung AS, Sample JG, Paul WE. 1988. Interleukin-4 is required to generate and sustain in vivo IgE responses. J. Immunol. 141:2335–41 Madden KB, Urban JF Jr, Ziltener HJ, Schrader JW, Finkelman FD, Katona IM. 1991. Antibodies to IL-3 and IL-4 suppress helminth-induced intestinal mastocytosis. J. Immunol. 147:1387–91 Schleimer RP, Sterbinsky S, Kaiser S. 1992. IL-4 induced adherence of human eosinophils and basophils but not neutrophils to endothelium. Association with expression of VCAM-1. J. Immunol. 148:1086–92 Gavett SH, O’Hearn DJ, Karp CL, Patel EA, Schofield BH, Finkelman FD, Wills-Karp M. 1997. Interleukin-4 receptor blockade prevents airway responses

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

279

induced by antigen challenge in mice. Am. J. Physiol. 272:L253–61 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 Aikawa T, Shimura S, Hidetada S, Ebina M, Takishima T. 1992. Marked goblet cell hyperplasia with mucus accumulation in the airways of patients who died of severe acute asthma attack. Chest 101:916–21 Temann UA, Prasad B, Gallup MW, Basbaum C, Ho SB, Flavell RA, Rankin JA. 1997. A novel role for murine IL-4 in vivo: induction of MUC5AC gene expression and mucin hypersecretion. Am. J. Respir. Cell. Mol. Biol. 16:471–78 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 McKenzie GJ, Bancroft A, Grencis RK, McKenzie AN. 1998. A distinct role for interleukin-13 in Th2 cell mediated immune responses. Curr. Biol. 8:339–42 Mochizuki M, Bartels J, Mallet AI, Christophers E, Schroder JM. 1998. IL-4 induces eotaxin: a possible mechanism of selective eosinophil recruitment in helminth infection and atopy. J. Immunol. 160:60–68 Nakajima H, Sano H, Nishimura T, Yoshida S, Iwamoto I. 1994. Role of vascular cell adhesion molecule 1/very late activation antigen 4 and intercellular adhesion molecule 1/lymphocyte function-associated antigen 1 interactions in antigen-induced eosinophil and T cell recruitment into the tissue. J. Exp. Med. 179:1145–54 Campbell HD, Tucker WQ, Hort Y, Martinson ME, Mayo G, Clutterbuck EJ, Sanderson CJ, Young IG. 1987. Molecular cloning, nucleotide sequence, and expression of the gene encoding human eosinophil differentiation factor (interleukin 5). Proc. Natl. Acad. Sci. USA 84:6629–33 Weller PF, Lim K, Wan HC, Dvorak AM, Wong DT, Cruikshank WW, Kornfeld H, Center DW. 1996. Role of the eosinophil in allergic reactions. Eur. Respir. J. 22:S109–15 Van Oosterhout AJ, Ladenius AR, Savelkoul HF, Van Ark I, Delsman KC, Nijkamp FP. 1993. Effect of anti-IL-5 and IL-5 on

P1: APR/SPD

P2: NBL/plb

January 21, 1999

280

92.

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

93.

94.

95.

96.

97.

98.

99. 100.

101.

102.

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

WILLS-KARP airway hyperreactivity and eosinophils in guinea pigs. Am. Rev. Respir. Dis. 147: 548–52 Luster AD, Rothenberg ME. 1997. Role of the monocyte chemoattractant protein and eotaxin subfamily of chemokines in allergic inflammation. J. Leukocyte Biol. 62:620–33 Holgate ST, Bodey KS, Janezic A, Frew AJ, Kaplan AP, Teran LM. 1997. Release of RANTES, MIP-1 alpha, and MCP-1 into asthmatic airways following endobronchial allergen challenge. Am. J. Respir. Crit. Care Med. 156:1377–83 Ying S, Robinson DS, Meng Q, Rottman J, Kennedy R, Ringler DJ, Mackay CR, Daugherty BL, Springer MS, Durham SR, Williams TJ, Kay AB. 1997. Enhanced expression of eotaxin and CCR3 mRNA and protein in atopic asthma. Association with airway hyperresponsiveness and predominant co-localization of eotaxin mRNA to bronchial epithelial and endothelial cells. Eur. J. Immunol. 27:3507– 16 Mould AW, Mattaei KI, Young IG, Foster PS. 1997. Relationship between interleukin-5 and eotaxin in regulating blood and tissue eosinophilia in mice. J. Clin. Invest. 99:1064–71 Rothenberg ME, MacLean JA, Pearlman E, Luster AD, Leder P. 1997. Targeted disruption of the chemokine eotaxin only partially reduces antigen-induced tissue eosinophilia. J. Exp. Med. 185:785–90 Jia G-Q, Gonzalo JA, Lloyd C, Kremer L, Lu L, Martinez-C, Wershil BK, Gutierrez-Ramos JC. 1996. Distinct expression and function of the novel mouse chemokine monocyte chemotactic protein-5 in lung allergic inflammation. J. Exp. Med. 184:1939–51 MacLean JA, Ownbey, Luster AD. 1996. T cell-dependent regulation of eotaxin in antigen-induced pulmonary eosinophilia. J. Exp. Med. 184:1461–69 Metcalfe DD, Baram Y, Mekori YA. 1997. Mast cells. Physiol. Rev. 77:1033–79 Bradding P, Feather IH, Howarth PH, Mueller R, Roberts JA, Britten K, Bews JP, Hunt TC, Okayama Y, Heusser CH. 1992. Interleukin 4 is localized to and released by human mast cells. J. Exp. Med. 176:1381–86 Costa JJ, Weller PF, Galli SJ. 1997. The cells of the allergic response. Mast cells, basophils and eosinophils. J. Am. Med. Assoc. 278:1815–22 Martin TR, Takeishi T, Katz HR, Austen KF, Drazen JM, Galli SJ. 1993. Mast cell activation enhances airway respon-

103.

104.

105.

106. 107.

108.

109.

110.

111.

112. 113.

siveness to methacholine in the mouse. J. Clin. Invest. 91:1176–82 Fahy JV, Fleming HE, Wong HH, Leu JT, Su JQ, Reimann J, Fick RB Jr, Boushey HA. 1997. The effect of an anti-IgE monoclonal antibody on the early- and latephase responses to allergen-inhalation in asthmatic subjects. Am. J. Respir. Crit. Care Med. 155:1828–34 Melhop PD, Van de Rijn M, Goldberg AB, Brewer JP, Kurup VP, Martin TR, Oettgen HC. 1997. Allergen-induced bronchial hyperreactivity and eosinophilic inflammation ocurr in the absence of IgE in mouse model of asthma. Proc. Natl. Acad. Sci. USA 94:1344–49 Wardlaw AJ, Dunnette S, Gleich GJ, Collins JV, Kay AB. 1988. Eosinophils and mast cells in bronchoalveolar lavage in subjects with asthma: relationship to bronchial hyperreactivity. Am. Rev. Respir. Dis. 137:62–69 Gleich GJ. 1990. The eosinophil and bronchial asthma: current understanding. J. Allergy Clin. Immunol. 85:422–36 Flavahan NA, Slitman NR, Gleich GJ, Vanhoutte PM. 1988. Human eosinophil major basic protein causes hyperreactivity of respiratory smooth muscle: role of the epithelium. Am. Rev. Respir. Dis. 138:685–88 Gundel RH, Letts G, Gleich GJ. 1991. Human eosinophil major basic protein induces airway constriction and airway hyperresponsiveness in primates. J. Clin. Invest. 87:1470–73 Jacoby DB, Gleich GJ, Fryer AD. 1993. Human eosinophil major basic protein is an endogenous allosteric antagonist at the inhibitory muscarinic M2 receptor. J. Clin. Invest. 91:1314–18 Gambone LM, Elbon CL, Fryer AD. 1994. Ozone-induced loss of neuronal M2 muscarinic receptor function is prevented by cyclosphosphamide. J. Appl. Physiol. 77:1492–99 Elbon CL, Jacoby DB, Fryer AD. 1995. Pretreatment with an antibody to interleukin-5 prevents the loss of pulmonary M2 muscarinic receptor function in antigen-challenged guinea pigs. Am. J. Respir. Cell. Mol. Biol. 12:320–28 McFadden ER. 1994. Asthma: Morphologic-physiologic interactions. Am. J. Respir. Crit. Care Med. 150:S23–S26 Lee JJ, McGarry MP, Farmer SC, Denzler KL, Larson KA, Carrigan PE, Brenneise IE, Horton MA, Haczku A, Gelfand EW, Leikauf GD, Lee NA. 1997. Interleukin-5 expression in the lung epithelium of transgenic mice leads to pulmonary changes

P1: APR/SPD

P2: NBL/plb

January 21, 1999

13:49

QC: KKK/uks

T1: NBL

Annual Reviews

AR078-09

T CELLS IN ASTHMA

114.

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

115.

116.

117.

118.

pathognomonic of asthma. J. Exp. Med. 185:2143–56 Corry DB, Folkesson HG, Warnock ML, Erle DJ, Matthay MA, Wiener-Kronish JP, Locksley RM. 1996. Interleukin 4, but not interleukin 5 or eosinophils, is required in a murine model of acute airway hyperreactivity. J. Exp. Med. 183:109–17 Kraneveld AD, Nijkamp FP, Van Oosterhout AJM. 1997. Role of neurokinin-2 receptor in interleukin-5-induced airway hyperresponsiveness but not eosinophilia in guinea pigs. Am. J. Respir. Crit. Care Med. 156:367–74 Bai TR, Mak JCW, Barnes PJ. 1992. A comparison of ß-adrenergic receptors and in vitro relaxant responses to isoproterenol in asthmatic airway smooth muscle. Am. J. Respir. Cell. Mol. Biol. 6:647– 51 Wills-Karp M, Uchida Y, Lee J, Jinot J, Hirata A, Hirata F. 1993. Organ culture with proinflammatory cytokines reproduces impairment of the β-adrenoceptormediated relaxation in tracheas of a guinea pig model. Am. J. Respir. Cell. Mol. Biol. 8:153–59 Liggett SB. 1997. Polymorphisms of the beta2-adrenergic receptor and asthma.

119.

120.

121.

122.

123.

281

Am. J. Respir. Crit. Care Med. 156:S156– 62 Du T, Sapienza S, Wang CG, Renzi PM, Pantano R, Rossi P, Martin JG. 1996. Effect of nedocromil sodium on allergeninduced airway responses and changes in the quantity of airway smooth muscle in rats. J. Allergy Clin. Immunol. 98:400–7 De S, Zelazny ET, Souhrada JF, Souhrada M. 1995. IL-1 beta and IL-6 induced hyperplasia and hypertrophy of cultured guinea pig smooth muscle cells. J. Appl. Physiol. 78:1555–63 Maron Z, Shelhamer JH, Bach MK, Morton DR, Kaliner M. 1982. Slow-reacting substances, leukotrienes C4 and D4, increase the release of mucous from human airways in vitro. Am. Rev. Respir. Dis. 126:449–51 Blosser S, Mitzner W, Wagner EM. 1994. Effects of increased bronchial blood flow on airway morphometry, resistance and reactivity. J. Appl. Physiol. 76:1624–29 Wardlaw AJ, Dunnette S, Gleich GJ, Collins JV, Kay AB. 1988. Eosinophils and mast cells in bronchoalveolar lavage in subjects with mild asthma. Relationship to bronchial hyperreactivity. Am. Rev. Respir. Dis. 137:62–69

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:255-281. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SAT/PKS/mbg

P2: NBL/KKK/spd

February 11, 1999

12:0

QC: NBL

Annual Reviews

AR078-10

Annu. Rev. Immunol. 1999. 17:283–95 c 1999 by Annual Reviews. All rights reserved Copyright °

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

REGULATION OF T CELL FATE BY NOTCH Ellen Robey Department of Molecular and Cell Biology, 47l Life Sciences Addition, University of California, Berkeley, California 94720; email: [email protected] KEY WORDS:

thymus, T cell development, lineage committment, positive selection, Notch ligands

ABSTRACT The transmembrane receptor Notch participates in diverse cell fate decisions throughout embryonic development. Notch receptors and their ligands are expressed in the mammalian thymus, raising the possibility that Notch could regulate T cell fate decisions. Expression of a constitutively activated form of Notch in developing thymocytes causes thymocytes normally destined for the CD4 lineage to adopt the CD8 lineage instead. This suggests that Notch activity normally acts to direct CD4+CD8+ precursors to the CD8 lineage. The choice between CD4 and CD8 T cell fates is also controlled by MHC recognition during positive selection, implying that recognition of class I or II MHC might regulate Notch signaling. Possible models for the regulation of Notch by MHC recognition during CD4 versus CD8 lineage determination are discussed.

INTRODUCTION During development, precursor cells often receive signals from their environment that direct them to the appropriate lineage. One such signal is mediated by Notch, a transmembrane receptor that functions in diverse developmental scenarios to control the choice between alternative cell fates (reviewed in (1–4)). The Notch family of receptors includes the C. elegans homologs LIN-12 and GLP-1 as well as Drosophila Notch, and Notch1-4 in mammals. These receptors are broadly expressed throughout embryonic development and control cell fate decisions in many different tissues. 283 0732-0582/99/0410-0283$08.00

P1: SAT/PKS/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

284

P2: NBL/KKK/spd

12:0

QC: NBL

Annual Reviews

AR078-10

ROBEY

Figure 1 A schematic diagram of the Notch signaling pathway. Notch is activated by binding to its ligand on a neighboring cell. A CSL transcription factor (CBF1, Su(H), LAG-1) associates with the intracellular domain of Notch. Ligand binding leads to the activation of CSL (14–19), perhaps by inducing the cleavage and nuclear translocation of the intracellular domain of Notch (20, 21). Activated CSL proteins then bind to the promoters of a variety of target genes, some of which are themselves transcription factors (22–26). This cascade of transcriptional regulation ultimately leads the precursor to adopt a particular developmental fate. The major structural motifs of Notch and its ligands are indicated (not to scale). All members of the Notch family of receptors contain multiple EGF-like and LNG (LIN-12, Notch, GLP-1) repeats in their extracellular domains, and ankryn-like repeats in their intracellular domains. The ligands for Notch comprise a family of transmembrane proteins containing EGF-like repeats and a characteristic DSL domain (Delta, Serrate, LAG-2) in their extracellular domains. Small open rectangles represent EGF-like repeats, open square represents DSL domain, shaded rectangles represent LNG repeats, ovals represent ankryn-like repeats.

The Notch signaling pathway has been highly conserved throughout evolution (Figure 1). Notch activity is generally regulated by ligand binding, and ligands for Notch comprise a family of related transmembrane proteins that includes Delta and Serrate in Drosophila and Delta-like (Delta1-3) and Serrate-like ligands (Jagged1/Serrate1 and Jagged2/Serrate2) in vertebrates (5–13). Notch signaling leads to the activation of an associated transcription factor, termed CSL (mammalian CBF1, Drosophila Su(H), C. elegans LAG-1) (14–19), by a poorly understood process that may involve the proteolytic cleavage and nuclear translocation of the intracellular domain of Notch (20, 21). Activated CSL proteins then turn on the transcription of a variety of target genes, some of which encode other transcription factors including enhancer of split (ESR) and the HES genes (22–26). This cascade of transcription factor activation triggered by Notch leads the precursor to adopt a

P1: SAT/PKS/mbg

February 11, 1999

P2: NBL/KKK/spd

12:0

QC: NBL

Annual Reviews

AR078-10

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NOTCH IN T CELL DEVELOPMENT

285

particular developmental fate, presumably by regulating cell type–specific gene expression. There are strong indications that Notch signaling controls the choice of cell fate in the mammalian thymus. Mouse thymocytes express at least two Notch family members: Notch1 (27–30) and Notch3 (31; and G Weinmaster, unpublished). In addition, at least one Notch ligand, Jagged2, is known to be expressed in the thymus, both by thymocytes and by thymic stomal cells (9, 32; A Itano, E Robey, unpublished). Recent studies from our lab using transgenic mice expressing an activated form of Notch in the thymus, as well as studies with mice that are heterozygous for a gene disruption of Notch1, indicate that Notch activity controls both the CD4 versus CD8 lineage choice (30) and the αβ versus γ δ lineage choice (33, 34). In this review I briefly discuss our published work on the effects of Notch activity on CD4 versus CD8 lineage determination. I then turn to a speculative discussion of how Notch signaling in thymocytes might be regulated normally, and how signals through the T cell antigen receptor complex and Notch may work together to direct thymocyte precursors to the appropriate mature T cell lineage.

THE INFLUENCE OF AN ACTIVATED FORM OF NOTCH ON THYMIC DEVELOPMENT One way to probe the role of Notch in thymic development is to examine the effects of a constitutively activated form of Notch in the thymus. Studies in Drosophila and C. elegans have demonstrated that truncations of Notch that remove most or all of the extracellular domain lead to ligand independent Notch activity and result in cell fate transformations that are the opposite of those seen with loss-of-function Notch mutations: extra secondary fate cells at the expense of primary fate cells (35–38). Expression of a similarly truncated version of the mouse Notch1 gene (NotchIC) in thymocytes also leads to a striking cell fate transformation: extra CD8 lineage cells and fewer CD4 lineage cells (30). While the effect of a loss-of-function Notch mutation on the CD4 versus CD8 lineage decision has not been determined, the ample precedent for the effects of activated Notch in invertebrate systems strongly suggests that endogenous Notch activity in the thymus acts to favor the development of CD8 lineage cells and inhibit the development of CD4 lineage cells. The interpretation of the effects of activated Notch on the CD4 versus CD8 lineage decision is based not only on the precedent from invertebrate systems but also on the highly selective effect of the NotchIC transgene on thymic development. This is shown most clearly by the analysis of thymocyte population dynamics in these mice using a continuous BrdU labeling protocol (30) (summarized in Figure 2). In normal mice, the CD4+CD8+ precursors represent a

P1: SAT/PKS/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

286

P2: NBL/KKK/spd

12:0

QC: NBL

Annual Reviews

AR078-10

ROBEY

Figure 2 Thymocyte population dynamics in wild-type and NotchIC transgenic mice. Proliferation, maturation, and apoptosis in a thymocyte clone is diagrammed for wild-type mice and for mice expressing an activated form of Notch under the Lck proximal promoter (NotchIC transgenic mice). The diagrams are based on the analysis of thymocyte population dynamics as measured by a continuous BrdU labeling protocol (30). Activated Notch does not grossly alter the lifespan of the CD4+CD8+ population, nor the fraction of CD4+CD8+ cells that develop into mature thymocytes. The major effect of the mutation is to increase the proportion of thymocytes that develop as CD8 lineage thymocytes (shaded circles) and decrease the proportion of thymocytes that develop as CD4 lineage cells (open circles). See text for discussion.

P1: SAT/PKS/mbg

February 11, 1999

P2: NBL/KKK/spd

12:0

QC: NBL

Annual Reviews

AR078-10

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NOTCH IN T CELL DEVELOPMENT

287

short-lived population as reflected by the rapid appearance of BrdU label in this population and the short turnover time of 3–4 days. Expression of activated Notch in thymocytes does not alter the labeling kinetics of the CD4+CD8+ precursor population, indicating that Notch activity does not grossly alter the proliferation or life-span of this population. In contrast, inhibition of apoptosis in CD4+CD8+ thymocytes by overexpression of a BCL-2 transgene (39) leads to a dramatic increase in the turnover time of the CD4+CD8+ population (A Itano and E Robey, unpublished observations). In normal mice the mature αβTCR+CD4+CD8− (CD4 lineage) and αβTCR+CD4−CD8+ (CD8 lineage) thymocytes are not dividing and are derived from nonproliferating precursors, as indicated by the 0.5–2-day lag in the labeling kinetics of these mature thymocytes (40, 41). In NotchIC transgenic mice, mature CD4 and CD8 lineage cells still show a pronounced lag in labeling kinetics, indicating that activated Notch does not lead to the inappropriate proliferation of mature thymocytes. Because mature CD4 and CD8 lineage thymocytes are not proliferating and do not acquire the BrdU label directly, the accumulation of label in these mature populations reflects the conversion of labeled precursors into mature thymocytes. In wild-type mice, BrdU label accumulates more rapidly in mature CD4 lineage thymocytes than in CD8 lineage thymocytes, indicating that a greater proportion of precursors normally choose the CD4 lineage than choose the CD8 lineage. In NotchIC transgenic mice the opposite is true, indicating that a greater proportion of precursors develop as CD8 lineage cells than as CD4 lineage cells. This analysis shows that activated Notch does not alter the overall population dynamics of thymic development but, instead, alters the proportion of CD4+CD8+ precursors that develop as CD8 versus CD4 lineage cells. Although Notch activity increases the proportion of precursors hat choose the CD8 over the CD4 lineage, it does not alter the overall efficiency of thymocyte maturation. In normal mice, as in NotchIC transgenic mice, the vast majority of CD4+CD8+ precursors do not develop into mature thymocytes but instead undergo programmed cell death. This suggests that activated Notch does not override the requirement for positive selection: the recognition by thymocytes of MHC proteins on thymic epithelial cells that rescue thymocytes from programmed cell death and allow them to mature into long-lived CD4 or CD8 lineage cells (reviewed in 42, 43). Together, these data suggest that activated Notch causes thymocytes bearing TCRs specific for class II MHC proteins, which would normally develop as CD4 lineage cells, to develop as CD8 lineage cells instead. Analysis of the effect of activated Notch in the absence of class I or class II MHC proteins confirms this suggestion (30; summarized in Table 1). Activated Notch allows CD8 lineage thymocytes to develop in the absence of class I or class II MHC proteins, but not in the absence of both class I

P1: SAT/PKS/mbg

P2: NBL/KKK/spd

February 11, 1999

12:0

288

QC: NBL

Annual Reviews

AR078-10

ROBEY

Table 1 The effect of MHC mutations and an activated form of Notch (NotchIC transgene) on the development of CD4 and CD8 lineage thymocytes. Plus and minus signs indicate the relative steady state size of the mature CD8 and CD4 lineage thymocyte populations. Data summarized from (30). Class I MHC− mice are β2 microglobulin mutant (65) and Class II MHC− mice are IAb mutant (66)

Genotype

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Wild type

”

Class I MHC−

”

Class II MHC−

”

Class I and II MHC−

”

NotchIC transgene

CD8 lineage (αβTCR+CD4−CD8+)

No Yes No Yes No Yes No Yes

+ +++ − ++ + ++ − −

CD4 lineage (αβTCR+CD4+CD8−) +++ + +++ + − − − −

and class II MHC. Thus, the effect of activated Notch in MHC mutant mice, together with the kinetic analysis of thymic development, supports the interpretation that Notch activity selectively alters the choice of CD4+CD8+ precursors between the CD4 and CD8 lineage such that precursors that recognize class II MHC develop as CD8 lineage cells rather than as CD4 lineage cells.

NOTCH AND THE T CELL ANTIGEN RECEPTOR: HOW MIGHT THEY WORK TOGETHER? How Notch activity is normally regulated in the thymus is an open question that is an important area for future investigation. In invertebrates, where the role of the Notch family of receptors in cell fate decisions has been extensively studied, a number of diverse scenarios for Notch regulation have been described. I discuss three of these scenarios to illustrate the range of possibilities. I then present two possible scenarios for Notch regulation in the thymus that, while speculative, may provide a framework for future investigation.

Scenarios from Invertebrates In many cases both a Notch-like receptor and its ligand are expressed by a group of equivalent precursor cells, and Notch signaling between neighboring precursors determines which cell will adopt a primary or secondary fate (44–46). This scenario, termed lateral signaling, operates in the decision between the anchor cell fate and the ventral uterine precursor cell fate (AC versus VU) during vulval development in C. elegans (Figure 3a) (46). In this case two precursor cells of equivalent developmental potential initially express similar

P1: SAT/PKS/mbg

February 11, 1999

P2: NBL/KKK/spd

12:0

QC: NBL

Annual Reviews

AR078-10

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NOTCH IN T CELL DEVELOPMENT

289

Figure 3 Three different scenarios from invertebrates for regulation of the Notch family of receptors. (a) Lateral signaling, stochastic choice of cell fate. In the AC versus VU decision in C. elegans, an initial fluctuation in the level of receptor (LIN-12) or ligand (LAG-2) is amplified over time by a self-reinforcing feedback mechanism such that eventually one cell becomes the sending cell and adopts the primary fate (open circle) while its neighbor becomes the receiving cell and adopts the secondary fate (shaded circle). (Adapted from 44, 46) The arrows denote ligand and the asterisks denote activated receptor. (b) Lateral signaling, regulated choice of cell fate. In vulval development in C. elegans, lateral signaling between vulval precursor cells through the Notch-like receptor LIN-12 and a secreted signal from the nearby anchor cell act together to specify primary, secondary, and tertiary fates (48). (c) In MP2 neuroblast development, sibling precursors are both in contact with ligand (Delta) on neighboring cells (55). Notch signaling is inhibited by Numb, a protein that is asymmetrically distributed between the precursor cells (55, 57). See text for discussion.

levels of receptor (LIN-12) and ligand (LAG-2). Over time an asymmetry develops: One precursor begins to express more ligand and less receptor, while its neighbor expresses less ligand and more receptor. This asymmetry is amplified over time until the ligand expressing cell adopts the primary fate (AC) and the receptor expressing cell adopts the secondary fate (VU). These and other data (44–46) lead to a model in which an initial random fluctuation in the levels of receptor or ligand is subsequently amplified by a self-reinforcing feedback loop, such that the cell that receives more Notch signal responds by turning up the receptor and turning down the ligand, whereas its neighbor that receives less Notch signal does the opposite. In the AC versus VU decision, the choice of cell fate appears to be stochastic. There are other lateral signaling scenarios, however, in which precursor cells

P1: SAT/PKS/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

290

P2: NBL/KKK/spd

12:0

QC: NBL

Annual Reviews

AR078-10

ROBEY

are not equivalent and other cues act together with Notch to influence the choice between primary and secondary cell fate (discussed in 3, 47). For example, at a later stage in vulval development in C. elegans, LIN-12 signaling between vulval precursor cells (VPC) regulates the choice between primary and secondary fates (Figure 3b) (48, 49). In this case, however, the choice between primary and secondary fates is not random but is controlled by a secreted signal, produced by the nearby anchor cell, binding to a receptor tyrosine kinase on the VPC. The VPC that is the closest to the source of the secreted signal adopts the primary fate, while its immediate neighbors adopt the secondary fate. While it is not yet understood precisely how this information is integrated, it is likely that the secreted signal from the anchor cell influences lateral signaling between the VPCs, causing the VPC that is closest to the anchor cell to signal its immediate neighbors through LIN-12 to adopt the secondary fate (50). While lateral signaling is a common feature of Notch regulation, there are also situations in which the Notch ligand is not expressed by the precursor cells, but rather is present on another cell type (51–55). This type of signaling appears to operate in the development of MP2 neurons in Drosophila (Figure 3c) (55, 56). In this case, Notch signaling is regulated, not by the presence or absence of the ligand, which appears to be available to both precursor cells, but by the presence or absence of Numb, a protein that is asymmetrically distributed between precursors and that inhibits Notch signaling (57). These three examples from invertebrates provide some indication of the wide variety of mechanisms by which Notch signaling can be regulated.

Models for CD4 versus CD8 Lineage Determination Any model for Notch signaling in the thymus must incorporate the wellestablished role of MHC recognition in guiding the CD4 versus CD8 lineage decision (reviewed in 42, 43, 58). During positive selection, MHC proteins on thymic epithelial cells are recognized via the T cell antigen receptor (TCR) and the CD4 or CD8 co-receptors on developing thymocytes such that recognition of class I MHC proteins by a class I–specific TCR and the CD8 co-receptor leads to CD8 cell development, and recognition of class II MHC by a class II–specific TCR and the CD4 co-receptor leads to CD4 cell development. Much attention has focused on the role of the co-receptors in CD4 versus CD8 lineage determination. Although the precise role of CD4 and CD8 in this process has not yet been resolved, one possibility is that quantitative differences in signaling might influence the CD4 and CD8 lineage decision, so that engagement of CD8 (which weakly activates the tyrosine kinase, Lck) would lead to the CD8 fate, whereas engagement of CD4 (which activates Lck more strongly) would lead to the CD4 cell fate (59–61). The notion that binding of class I or class II MHC would lead to differential signaling through CD4

P1: SAT/PKS/mbg

February 11, 1999

P2: NBL/KKK/spd

12:0

QC: NBL

Annual Reviews

AR078-10

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NOTCH IN T CELL DEVELOPMENT

291

Figure 4 Two possible scenarios for Notch regulation in the CD4 versus CD8 lineage decision. (a) Lateral signaling model. In this model, both Notch and its ligand are expressed by thymocytes. An initial difference in Notch signaling between a thymocyte undergoing positive selection and its neighbors is imposed by MHC recognition such that recognition of class I MHC enhances Notch signaling, and recognition of class II MHC inhibits Notch signaling. The initial difference in Notch signaling between thymocytes could then be amplified by feedback regulation. (b) In this model, the Notch ligand is expressed on thymic epithelial cells and would be available to all cells undergoing positive selection. Notch signaling would be regulated by MHC recognition so that recognition of class I MHC would enhance Notch signaling and recognition of class II MHC would inhibit Notch signaling. See text for discussion.

and CD8, which would in turn modulate Notch signaling, is incorporated in both models described below. In the model depicted in Figure 4a, lateral Notch signaling between thymocytes is modulated by the recognition of MHC proteins on thymic epithelial cells. According to this model, Notch signaling between thymocytes would operate by a self-reinforcing feedback mechanism such that an initial difference in Notch signaling between thymocytes could be amplified over time. This initial difference between thymocytes would not be due to a random fluctuation, as in the AC versus VU decision (Figure 3a), but would rather be imposed by

P1: SAT/PKS/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

292

P2: NBL/KKK/spd

12:0

QC: NBL

Annual Reviews

AR078-10

ROBEY

differential signaling generated upon recognition of MHC class I or II. Class I recognition would enhance Notch signaling (perhaps by engaging CD8) and class II recognition would inhibit Notch signaling (perhaps by engaging CD4). Feedback regulation would then amplify the difference in Notch signaling between thymocytes, with the result that thymocytes that initially receive more Notch signal would become CD8 cells, and those that receive less Notch signal would become CD4 cells. This model is analogous to the development of VPCs (Figure 3b) in which lateral signaling between precursors is modulated by a distinct signal from another cell type. One possible objection that could be raised to this model is that a thymocyte that happens to be next to a thymocyte undergoing positive selection might have a cell fate imposed upon it that is not compatible with the specificity of its TCR for class I or class II MHC. This, however, may not be a significant concern given that the vast majority of thymocytes are destined to die in the thymus. Thus, the rare thymocyte that undergoes positive selection will be surrounded by “bystander” thymocytes that could serve as compliant partners in a lateral signaling scenario. In the model depicted in Figure 4b the ligand for Notch is expressed, not by thymocytes, but by thymic epithelial cells. In this model, thymocytes that recognize class I or class II MHC would be in contact with Notch ligands and Notch signaling would be modulated by differential signals emanating from the TCR and co-receptors upon class I or class I MHC recognition. This model bears some similarity to the development of the MP2 lineage neuroblasts (Figure 3c) in which precursors destined for the primary and secondary fate are both in contact with ligand and Notch signaling is regulated downstream of ligand binding. Does the existing information concerning Notch ligands argue for or against either of these models? The expression pattern of the Serrate-like ligand, Jagged2, is compatible with either model since it is expressed on both thymocytes and thymic stromal cells (9, 32; and A Itano and E Robey, unpublished). A gene disruption of Jagged2 leads to a reduction in fetal γ δ lineage thymocytes, but does not affect the adult type γ δ lineage cells, CD4, or CD8 cells (62; and A Itano, unpublished). Given the ever growing number of vertebrate Notch ligands (5–12), it is likely that multiple Notch ligands will be involved in thymic development and may serve overlapping, redundant functions that could complicate interpretation of their mutant phenotypes. What about the regulation of Notch expression during thymic development? A lateral signaling model might predict autoregulation of Notch expression, and there are indications that Notch1 levels are increased in response to Notch activity in thymocytes (30, 63). In addition, a report of Notch2 upregulation as thymocytes differentiate into CD8 cells in culture (64) is also compatible with such a model. In the absence of evidence for feedback regulation of ligand expression in thymocytes, however, the lateral signaling model remains speculative.

P1: SAT/PKS/mbg

P2: NBL/KKK/spd

February 11, 1999

12:0

QC: NBL

Annual Reviews

AR078-10

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

NOTCH IN T CELL DEVELOPMENT

293

Clearly there is not enough data at the moment to support or rule out either of these models. Nor is this an exhaustive list of the possibilities. For example, Notch signaling could be autocrine, or access to Notch ligands could be limited by anatomical or temporal factors. In addition, while this discussion has focused on the CD4 versus CD8 lineage decision, γ δ versus αβ lineage determination also appears to be controlled by both Notch and TCR signaling (through the preTCR and the γ δTCR, reviewed in 34), and the same kinds of possibilities and questions exist for this earlier T cell fate decision. The resolution of these questions awaits more information about the expression and regulation of Notch and its ligands in the thymus and how Notch and TCR signals are integrated to direct thymocytes to their appropriate mature T cell fate. ACKNOWLEDGMENTS Work in my laboratory is supported by National Institutes of Health and the American Cancer Society. I thank BJ Fowlkes, Iva Greenwald, Paul Sternberg, and members of my laboratory for comments on the manuscript. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Greenwald I. 1994. Structure/function studies of lin-12/Notch proteins. Curr. Opin. Genet. Dev. 1994:556–62 2. Artavanis-Tsakonas S, Matsuno K, Fortini M. 1995. Notch signaling. Science 268:225–32 3. Kimble J, Simpson P. 1997. The Lin12/ Notch signalling pathway and its regulation. Annu. Rev. Cell Dev. Biol. 13:333–61 4. Robey E. 1997. Notch in vertebrates. Curr. Opin. Genet. Dev. 7:551–57 5. Chitnis A, Henrique D, Lewis J, IshHorowicz D, Kintner C. 1995. Primary neurogenesis in Xenopus embryos regulated by a homologue of the Drosophila neurogenic gene Delta. Nature 375:761–66 6. Henrique D, Adam J, Myat A, Chitnis A, Lewis J, Ish-Horowicz D. 1995. Expression of a Delta homologue in prospective neurons in the chick. Nature 375:787– 90 7. Lindsell C, Shawber C, Boulter J, Weinmaster G. 1995. Jagged: a mammalian ligand that activates Notch1. Cell 80:909–17 8. Bettenhausen B, Hrabe de Angelis M, Simon D, Guenet J-L, Gossler A. 1995. Transient and restricted expression during mouse embryogenesis of Dll1, a murine

9.

10.

11.

12.

13.

gene closely related to Drosophila Delta. Development 121:2407–18 Shawber C, Boulter J, Lindsell C, Weinmaster G. 1996. Jagged2: a serrate-like gene expressed during rat embryogenesis. Dev. Biol. 180:370–76 Myat A, Henrique D, Ish-Horowicz D, Lewis J. 1996. A chick homologue of Serrate and its relationship with Notch and Delta homologues during central neurogenesis. Dev. Biol. 174:233–47 Jen W-C, Wettstein D, Turner D, Chitnis A, Kintner C. 1997. The Notch ligand, X-Delta-2, mediates segmentation of the paraxial mesoderm in Xenopus embryos. Development 124:1169–78 Dunwoodie S, Henrique D, Harrison S, Beddington R. 1997. Mouse Dll3: a novel divergent Delta gene which may complement the function of other Delta homologues during early pattern formation in the mouse embryo. Development 124:3065– 76 Li L, Milner LA, Yu D, Iwata M, Banta A, et al. 1998. The human homolog of rat Jagged1 expressed by marrow stroma inhibits differentiation of 32D cells through interaction with Notch1. Immunity 8:43–55

P1: SAT/PKS/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

294

P2: NBL/KKK/spd

12:0

QC: NBL

Annual Reviews

AR078-10

ROBEY

14. Schweisguth F, Posakony J. 1992. Suppressor of Hairless, the Drosophila homolog of the mouse recombination signalbinding protein gene, controls sensory organ cell fates. Cell 69:1199–1212 15. Fortini ME, Artavanis-Tsakonas S. 1994. The suppressor of Hairless protein participates in Notch receptor signaling. Cell 79:273–82 16. Christensen S, Kodoyianni V, Bosenberg M, Friedman L, Kimble J. 1996. lag-1, a gene required for lin-12 and glp-1 signaling in C. elegans, is homologous to human CBF1 and Drosophila Su(H). Development 122:1373–83 17. Jarriault S, Brou C, Logeat F, Schroeter E, Kopan R, Israel A. 1995. Signalling downstream of activated mammalian Notch. Nature 377:355–58 18. Tamura K, Taniguchi Y, Minoguchi S, Sakai T, Tun T, et al. 1995. Physical interaction between a novel domain of the receptor Notch and the transcription factor RBP-Jk/Su(H). Curr. Biol. 5:1416–23 19. Hsieh JJD, Henkel T, Salmon P, Robey E, Peterson MG, Hayward D. 1996. Truncated mammalian Notch1 activates CBF1/RBPJk repressed genes by a mechanism resembling that of Epstein-Barr virus EBNA2. Mol. Cell. Biol. 16:952– 59 20. Struhl G, Adachi A. 1998. Nuclear access and action of Notch in vivo. Cell 93:649– 60 21. Schroeter E, Kisslinger J, Kopan R. 1998. Notch-1 signalling requires ligand-induced proteolytic release of intracellular domain. Nature 393:382–86 22. Lecourtois M, Schweisguth F. 1995. The neurogenic Suppressor of Hairless DNAbinding protein mediates the transcriptional activation of Enhancer of split complex genes triggered by Notch signaling. Genes Dev. 9:2598–2608 23. Bailey A, Posakony J. 1995. Suppressor of Hairless directly activates transcription of Enhancer of split complex genes in response to Notch receptor activity. Genes Dev. 9:2609–22 24. de Celis JF, de Celis J, Ligoxygakis P, Priess A, Delidakis C, Bray S. 1996. Functional relationships between Notch, Su(H) and the bHLH genes of the E(spl) complex: the E(spl) genes mediate only a subset of Notch activities during imaginal development. Development 122:2719–28 25. de la Poma J, Wakeman A, Correla K, Samper E, Brown S, et al. 1997. Conservation of the Notch signaling pathway in mammalian neurogenesis. Development 124:1139–48

26. Wettstein D, Turner D, Kintner C. 1997. The Xenopus homolog of Drosophila Suppressor of Hairless mediates Notch signaling during primary neurogenesis. Development 124:693–702 27. Ellisen LW, Bird J, West DC, Soreng AL, Reynolds TC, et al. 1991. TAN-1, the human homolog of the Drosophila Notch gene, is broken by chromosomal translocations in T lymphoblastic neoplasms. Cell 66:649–61 28. Weinmaster G, Roberts VJ, Lemke G. 1991. A homologue of Drosophila Notch expressed during mammalian development. Development 113:199–205 29. Hasserjian R, Aster J, Davi D, Weinberg D, Sklar J. 1996. Modulated expression of Notch1 during thymocyte development. Blood 88:970–76 30. 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 31. Lardelli M, Dahlstrand J, Lendahl U. 1994. The novel Notch homologue mouse Notch 3 lacks specific epidermal growth factorrepeats and is expressed in proliferating neuroepithelium. Mech. Dev. 46:123–36 32. Luo B, Astar J, Hasserjian R, Kuo F, Sklar J. 1997. Isolation and functional analysis of a cDNA for human Jagged2, a gene encoding a ligand for the Notch1 receptor. Mol. Cell. Biol. 17:6057–67 33. Washburn T, Schweighoffer E, Gridley T, Chang D, Fowlkes B, et al. 1997. Notch activity influences the αβ vs. γ δ T cell lineage decision. Cell 88:833–43 34. Robey E, Fowlkes B. 1998. The αβ versus γ δ T-cell lineage choice. Curr. Opin. Immunol. 10:181–87 35. Struhl G, Fitzgerald K, Greenwald I. 1993. Intrinsic activity of the Lin-12 and Notch intracellular domains in vivo. Cell 74:331– 45 36. Rebay I, Fehon R, Artavanis-Tsakonas S. 1993. Specific truncations of Drosophila Notch define dominant activated and dominant negative forms of the receptor. Cell 74:319–29 37. Lieber T, Kidd S, Alcamo E, Corbin V, Young M. 1993. Antineurogenic phenotypes induced by truncated Notch proteins indicate a role in signal transduction and may point to a novel function for Notch in nuclei. Genes Dev. 7:1949–65 38. Roehl H, Kimble J. 1993. Control of cell fate in C. elegans by a GLP-1 peptide consisting primarily of ankyrin repeats. Nature 364:632–35 39. Sentman CL, Shutter JR, Hockenbery D,

P1: SAT/PKS/mbg

February 11, 1999

P2: NBL/KKK/spd

12:0

QC: NBL

Annual Reviews

AR078-10

NOTCH IN T CELL DEVELOPMENT

40.

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

41.

42. 43.

44. 45. 46.

47.

48.

49. 50.

51. 52.

53.

Kanagawa O, Korsmeyer S. 1991. bcl-2 inhibits multiple forms of apoptosis but not negative selection in thymocytes. Cell 67:879–88 Egerton M, Scollay R, Shortman K. 1990. Kinetics of mature T-cell development in the thymus. Proc. Natl. Acad. Sci. USA 87: 2579–82 Huesmann M, Scott B, Kisielow P, von Boehmer H. 1991. Kinetic and efficacy of positive selection in the thymus of normal and T cell receptor transgenic mice. Cell 66:533 Jameson SC, Hogquist K, Bevan M. 1995. Positive selection of thymocytes. Annu. Rev. Immunol. 13:93–126 Marrack P, Kappler J. 1997. Positive selection of thymocytes bearing alpha-beta T cell receptors. Curr. Opin. Immunol. 9: 250–55 Seydoux G, Greenwald I. 1989. Cell autonomy of lin-12 function in a cell fate decision in C. elegans. Cell 57:1237–45 Heitzler P, Simpson P. 1991. The choice of cell fate in the epidermis of Drosophila. Cell 64:1083–92 Wilkinson H, Fitzgerald K, Greenwald I. 1994. Reciprocal changes in expression of the receptor lin-12 and its ligand prior to commitment in a C. elegans cell fate decision. Cell 79:1187–98 Simpson P. 1997. Notch signalling in development: on equivalence groups and assymetric developmental potential. Curr. Opin. Genet. Dev. 7:537–42 Horvitz H, Sternberg P. 1991. Multiple intercellular signalling systems control the development of the Caenorhabditis elegans vulva. Nature 351:535–41 Simske J, Kim S. 1995. Sequential signalling during Caenorhabditis elegans vulval development. Nature 375:142–46 Levitan D, Greenwald I. 1998. LIN-12 protein expression and localization during vulval development in C. elegans. Development 125:3101–9 Kimble J, White J. 1981. On the control of germ cell development in Caenorhabditis elegans. Dev. Biol. 81:208–19 Henderson S, Gao E, Lambie E, Kimble J. 1994. lag-2 may encode a signaling ligand for the GLP-1 and LIN-12 receptors of C. elegans. Development 120:2913–24 Tax F, Yeargers J, Thomas J. 1994. Sequence of C. elegans lag-2 reveals a cellsignalling domain shared with Delta and Serrate of Drosphila. Nature 368:150– 3

295

54. Newman AP, White JG, Sternberg PW. 1995. The C. elegans lin-12 gene mediates induction of ventral uterine specialization by the anchor cell. Development 121:263– 71 55. Spana E, Doe C. 1996. Numb antagonizes Notch signaling to specify sibling neuron cell fates. Neuron 17:21–26 56. Campos-Ortega J. 1996. Numb diverts Notch pathway of the tramtrack. Neuron 17:1–4 57. Jan Y, Jan L. 1995. Maggot’s hair and bug’s eye: role of cell interactions and intrinsic factors in cell fate specification. Neuron 14:1–5 58. Robey EA, Fowlkes BJ. 1994. Selective events in T cell development. Annu. Rev. Immunol. 12:675–705 59. Itano A, Salmon P, Kioussis D, Tolaini M, Corbella P, Robey E. 1996. The cytoplasmic domain of CD4 promotes the development of CD4 lineage T cells. J. Exp. Med. 183:731–41 60. Matechak E, Killeen N, Hedrick S, Fowlkes B. 1996. MHC class II-specific T cells can develop in the CD8 lineage when CD4 is absent. Immunity 4:337–47 61. Sharp LL, Schwarz DA, Bott CM, Marshall CJ, Hedrick SM. 1997. The influence of the MAPK pathway on T cell lineage commitment. Immunity 7:609–18 62. Jiang R, Lan Y, Chapman H, Shawber C, Norton C, Serreze D, Weinmaster G, Gridley T. 1998. Defects in limb, craniofacial, and thymic development in Jagged2 mutant mice. Genes Dev. 12:1046–57 63. Girard L, Hanna Z, Beaulieu N, Hoemann N, Hoemann C, Simard C, Kozak C, Jolicoeur P. 1996. Frequent provirus insertional mutagenesis of Notch1 in thymomas of MMTVD/myc transgenic mice suggests a collaboration of c-myc and Notch1 for oncogenesis. Genes Dev. 10: 1930–44 64. Takahama Y, Tokoro Y, Sugawara T, Negishi I, Nakauchi H. 1997. Pertussis toxin can replace T cell receptor signals that induce positive selection of CD8 T cells. Eur. J. Immunol. 27:3318–31 65. Zijlstra M, Bix M, Simister NE, Loring JM, Raulet DH, Jaenisch R. 1990. Beta 2microglobulin deficient mice lack CD4−8+ cytotoxic T cells. Nature 344:742–46 66. Grusby MJ, Johnson RS, Papaioannou VE, Glimcher LH. 1991. Depletion of CD4+ T cells in major histocompatibility complex class II-deficient mice. Science 253:1417– 20

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:283-295. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:297–329 c 1999 by Annual Reviews. All rights reserved Copyright °

THE CD1 SYSTEM: Antigen-Presenting Molecules for T Cell Recognition of Lipids and Glycolipids Steven A. Porcelli Division of Rheumatology, Immunology, and Allergy, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts 02115; e-mail: [email protected]

Robert L. Modlin Division of Dermatology, Department of Microbiology and Immunology, and the Molecular Biology Institute, University of California School of Medicine, Los Angeles, California 90095; e-mail: [email protected] KEY WORDS:

cell-mediated immunity, T cells, microbial immunity, nonpeptide antigens, glycolipid antigens

ABSTRACT Recent studies have identified the CD1 family of proteins as novel antigenpresenting molecules encoded by genes located outside of the major histocompatibility complex. CD1 proteins are conserved in all mammalian species so far examined and are prominently expressed on cells involved in antigen presentation, which suggests a role in activation of cell-mediated immunity. This has now been confirmed by functional studies demonstrating the ability of CD1 proteins to restrict the antigen-specific responses of T cells in humans and mice. Identification of naturally occurring antigens presented by CD1 has revealed the surprising finding that these are predominantly a variety of foreign lipids and glycolipids, including several found prominently in the cell walls and membranes of pathogenic mycobacteria. Structural, biochemical, and biophysical studies support the view that CD1 proteins bind the hydrophobic alkyl portions of these antigens directly and position the polar or hydrophilic head groups of bound lipids and glycolipids for highly specific interactions with T cell antigen receptors. Presentation of antigens by CD1 proteins requires uptake and intracellular processing by antigen presenting cells, and evidence exists for cellular pathways leading to the presentation of both exogenous and endogenous lipid antigens. T cells recognizing antigens presented by CD1 have a range of functional activities that suggest they

297 0732-0582/99/0410-0297$08.00

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

298

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

are likely to mediate an important component of antimicrobial immunity and may also contribute to autoimmunity and host responses against neoplastic cells.

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

INTRODUCTION T cells occupy a central position in the generation of cell-mediated immune responses and are of key importance to the maintenance of immunological memory. The appreciation that T cells recognize peptide fragments of protein antigens bound to class I or class II antigen-presenting molecules encoded by the major histocompatibility complex (MHC) has been one of the key discoveries leading to the development of our current understanding of the cell-mediated immune response (1, 2). In fact, this paradigm has become so dominant that it is now difficult to conceive that T cell recognition of anything other than MHC/peptide complexes could play a significant role in specific cell-mediated immune responses. Nevertheless, it is now clear that other mechanisms exist by which specific T cell recognition of antigens that are chemically distinct from peptides may occur. This had long been suggested by a variety of incompletely characterized models of T cell responses to infectious agents (3–11) and has, in recent years, acquired a more solid foundation from detailed studies of human TCR γ δ + T cells that recognize small nonpeptide molecules of various microbes (12, 13). Perhaps the most clearly established paradigm of nonpeptide antigen recognition by T cells has come from studies of the CD1 system of MHC class Ilike proteins. These evolutionarily conserved proteins are now recognized as comprising a family of antigen-presenting molecules with unusually hydrophobic ligand binding grooves that are capable of presenting nonpeptide lipid and glycolipid antigens to T cells. These findings have enlarged the paradigm for generation of specific cell-mediated responses through T cell recognition of a broad and ubiquitous class of molecules that were not previously known to be T cell antigens. Here we review (a) the recent developments in this emerging field that have begun to clarify the molecular and cellular mechanisms enabling the presentation of lipid antigens by CD1 proteins and (b) their subsequent effects on the T cell response.

THE CD1 SYSTEM OF MHC-RELATED PROTEINS Historical Perspective: Discovery of CD1 Proteins and Genes The discovery of CD1 and its designation as the first cluster of differentiation (CD) has a special significance in the history of immunology research. The CD nomenclature, introduced at the First International Workshop on Human

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN PRESENTATION BY CD1

299

Leukocyte Differentiation Antigens, groups or clusters monoclonal antibodies with similar reactivity, as judged by immunofluorescence and other techniques (14). One of the first monoclonal antibodies made against a human cell-surface antigen, shortly after the introduction of the hybridoma technique by Milstein and colleagues, was specific for the protein that is now known as CD1a (15). This antibody initially defined the first cluster of differentiation markers, thus marking the start of the continuing effort to systematically define and classify the differentiation antigens of human leukocytes. Subsequently, isolation of additional monoclonal antibodies subdivided the CD1 cluster into a group of distinct serologic and biochemical entities, indicating that human CD1 was actually a family of related proteins (16–18). With the pioneering molecular studies also from the laboratory of Milstein that defined the genes for these proteins, it was definitively shown that CD1 was in fact a family of related proteins encoded by separate closely linked genes (19–21). These studies, along with early biochemical analyses, revealed the MHC-related structure and tissue distribution of CD1 proteins and pointed the way to cellular studies that eventually revealed their function as a novel class of antigen-presenting molecules for T cell responses.

Genomic Organization and Evolutionary Diversification of CD1 The human CD1 family is encoded by five nonpolymorphic and closely linked genes located on chromosome 1 and are, therefore, unlinked to the MHC on chromosome 6 (22, 23). These show an intron/exon structure similar to MHC class I genes (21) and encode polypeptides with significant homology to both MHC class I and II proteins (19, 20, 24). Of the five CD1 genes in the human genome, four (the CD1A, -B, -C, and -D genes) are known to be expressed as proteins (21, 25, 26). These proteins represent distinct CD1 isoforms, which in humans are designated CD1a, -b, -c, and -d. The fifth human CD1 gene, designated CD1E, lacks obvious pseudogene features and has been shown to be transcribed (27), but no protein product (i.e. a CD1e protein) has yet been identified. CD1 genes and proteins have been studied in several other mammalian species besides humans, including ungulates [cows and sheep (28–35)], lagomorphs [rabbits and guinea pigs (36, 37; K LeClair, personal communication)], and rodents [mice and rats (38–43)]. In all mammals so far examined, CD1 genes and proteins have been found, although the differences in the size and complexity of the CD1 families of different mammals are striking (Table 1). A separation of the known CD1 genes and proteins into two groups, now generally referred to as group I and group II CD1, was first proposed by Calabi and colleagues based mainly on homology of nucleotide and amino acid sequences (44). As described in subsequent sections of this review, this division of CD1 molecules into groups I and II now also appears to be supported by

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

300

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

PORCELLI & MODLIN Table 1 Size and complexity of CD1 gene families in various mammals

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Species Human Mouse Rat Guinea pig Rabbit Sheep

Number of genes for indicated isoform

Total CD1 genesa

CD1A

CD1B

CD1C

CD1D

CD1E

5 2 1 ∼10 ∼8 ∼7

1 0 0 ?b ? ?

1 0 0 ≥5 ≥1 ≥3

1 0 0 ≥3 ? ?

1 2 1 ? ≥1 ?

1 0 0 ≥1 ? ?

a The number of genes per haploid genome is shown. This has been estimated based on genomic Southern blotting for rabbit (36), sheep (35), and guinea pig (K LeClair, personal communication). b ?, Indicates that there is insufficient data to determine the presence or absence of the indicated isoform.

a variety of studies on the expression and function of the two groups of CD1 proteins. Group I CD1 includes the products of the human CD1A, -B, and -C genes and their homologues in other mammals. These are the classic CD1 antigens first identified as differentiation antigens that are expressed on immature cortical thymocytes and subsequently shut off during the process of T cell maturation (45–49). Group II CD1 is currently defined as the products of the human CD1D gene and its close homologues in other species. These include all the CD1 proteins expressed in mice and rats and also one of the two forms of rabbit CD1 identified by cDNA cloning. A comparison of the CD1 loci in humans and mice reveals the marked difference in the CD1 families of these species (Figure 1), which is an important point given the central role that the mouse currently plays as an animal model in immunology research. The available data indicate that the genomes of all strains of mice examined lack group I CD1 genes, although they have maintained at least two group II CD1 genes (38, 50, 51). The two CD1 genes in the mouse genome are extremely similar to each other (approximately 90–95% sequence identity in all domains) and clearly represent a relatively recent duplication event. The gene referred to as CD1D1 (or MCD1.1) is now established to be expressed and functional (52, 53). In contrast, current data suggest that in at least one strain of mouse (C57BL/6), a frame shift mutation extinguishes expression of CD1D2 (54), and it remains unclear whether this gene gives rise to a functional protein in other strains. The finding that group I CD1 proteins are absent from mice and rats, whereas group II is present in these and probably most or all other mammals, raises questions about the different functions and relative importance of the two groups. It is currently unknown whether group I CD1 genes were once present and then

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN PRESENTATION BY CD1

301

Figure 1 Genomic maps of the human and mouse CD1 loci. (Open boxes) Genes for which the protein product is known to be expressed in vivo; (shaded boxes) genes that are known to be transcriptionally active but not yet established to give rise to a protein product in vivo; (arrows) the direction of transcription. Previous names for genes refer to the original nomenclature of Milstein and colleagues (see text for references). (Reprinted with permission from Reference 57.)

deleted from the forerunners of modern mice and rats, or whether group II CD1 is the precursor of the group I genes and failed to diversify in rodents. However, the presence of both groups in rabbits, which are believed to be closely related to rodents in terms of evolutionary origin (55), strongly suggests that the former explanation is more likely. Furthermore, the finding that the murine CD1 genes are located near the boundary point of an area of chromosomal synteny between mice and humans, presumably the breakpoint of an ancient translocation event, provides circumstantial evidence that previously present group I CD1 genes may have been deleted in rodents (56). In either case, at least two explanations could account for the finding that the absence of group I CD1 genes is tolerated by mice and rats. One possiblilty is that the environmental factors (presumably specific pathogens) that are responsible for selecting the maintenance or development of group I CD1 are not relevant to mice and rats. Alternatively, the function of group I CD1 may have been replaced or compensated for by other mechanisms that have evolved in rodents. It is important to note that the absence of group I CD1 in mice means that in many cases it may be difficult to make direct extrapolations from this animal to humans. This ultimately may mandate the use of other animals (e.g. guinea pigs or nonhuman primates) in

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

302

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

order to acquire a broader and more accurate understanding of the role of the CD1 system in human immune responses.

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Evolutionary Relation of CD1 to the MHC The presence of at least one CD1 gene in all mammalian species studied to date indicates that this family is evolutionarily ancient and must predate the extensive radiation of mammalian species that is predicted to have occurred approximately 60–80 million years ago (Figure 2). Also supporting the view that CD1 is an ancient lineage of antigen-presenting molecules is the observation that CD1 proteins show comparable levels of similarity and divergence at the amino acid and nucleic acid sequence level to both MHC class I and class II proteins (20, 57). This similar level of homology to both families of MHC proteins implies that CD1 may have diverged from a primordial ancestral antigen-presenting molecule at a point close to the divergence of the precursors of modern MHC class I and class II molecules. Because both MHC class I and class II genes are known to exist in species as ancient as cartilaginous fish (58–60), one might expect to find CD1 genes also present in most vertebrate species. This possibility remains untested, as no studies establishing either the

Figure 2 Hypothetical evolutionary tree for CD1 and major histocompatibility complex (MHC) class I and class II genes, based on nucleotide and amino acid sequence homologies. CD1 is proposed to have diverged from the common ancestral gene at a distant point in time close to the point at which the separate MHC class I and II lineages diverged. The presence of the same CD1 isoforms in a variety of mammals indicates that the subsequent diversification of CD1 into distinct isoforms (i.e. CD1a, -b, -c, -d, and -e) must have occurred prior to the extensive radiation of mammalian species that occurred between 60 and 80 million years (Myr) ago. (Adapted from Reference 149.)

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN PRESENTATION BY CD1

303

presence or absence of CD1 genes or proteins in vertebrate species more ancient than mammals have been reported. Another feature that bears emphasis is the striking lack of polymorphism among CD1 heavy chains in different outbred individuals (61, 62), which suggests that CD1 proteins are not subject to the same evolutionary forces responsible for the extensive allelic polymorphism of classical MHC class I and II loci. The homology of CD1 proteins to MHC class I heavy chains, although significant, is not extremely high (57, 63). Both CD1 and MHC class I proteins have three extracellular domains of similar size, designated α1, α2, and α3. In the most membrane distal α1 domains, only very minimal if any homology can be detected between CD1 and MHC class I at the amino acid or nucleic acid sequence level. In the α2 and α3 domains, the homology with the corresponding MHC class I domains rises to as much as approximately 35% at the amino acid sequence level, depending on the particular CD1 and MHC class I molecules that are compared. This is far less than the homology between the products of different MHC class I loci (e.g. HLA-A, B, or C and nonclassical MHC class I molecules such as HLA-E, F, and G), which are typically 70% homologous or more (64, 65). In contrast, this is much closer to the level of similarity observed between MHC class I and the intestinal immunoglobulin receptor FcRn, a protein encoded outside of the MHC that is known to have an MHC class I–like three-dimensional structure (66). Another important point to consider is the relationship between the different CD1 isoforms (i.e. CD1a, -b,-c, -d, and -e). It has been consistently found that in comparing the sequences of CD1 genes and proteins of nonhuman species with those of human CD1 family members, it is possible to identify each nonhuman sequence as a clear homologue of one of the human isoforms (57). For example, CD1 sequences have been identified in several species (sheep, guinea pig, rabbit) that are clearly direct homologues of human CD1b (35, 36). This is evident from the observation that these sequences of nonhuman species are more closely related to human CD1b than the latter is to other human CD1 isoforms (i.e. CD1a, -c, -d, or -e). This evolutionary preservation of distinct isoforms of CD1 is very different from what is generally observed for MHC-encoded molecules, for which distinct interspecies homologues of individual MHC class I and II loci can not readily be identified. This feature of CD1 suggests that the different CD1 isoforms may have evolved specialized functions early in the course of mammalian evolution, thus resulting in strong selection for the preservation of their structure during the subsequent divergence of different species.

CD1 Protein Structure CD1 genes encode polypeptides with a predicted molecular mass of approximately 33,000, although the presence of three or more N-linked glycans

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

304

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

generally raises their observed mass into a range between 41,000 and 55,000 (16, 26, 62, 67–69). All CD1 proteins studied to date are expressed on the surface of cells as type I transmembrane proteins that associate noncovalently with β2-microglobulin (β2-m). In most cases, this association appears to be necessary for efficient folding and surface expression of the CD1 protein (70, 71), although some possible exceptions to this rule have been noted (72–74). By analogy with MHC class I, the polypeptides encoded by CD1 genes are typically referred to as CD1 heavy chains. Given the limited but significant homology between CD1 and MHC class I heavy chains, as well as the similar overall domain organization and β2-m association, it was predicted that CD1 proteins could adopt a folded structure similar to that of MHC class I proteins (44, 57, 75). This prediction was recently shown to be correct with the solution of the three-dimensional structure of mouse CD1d1 by X-ray crystallography (76), which revealed a remarkable similarity in overall shape to MHC class I proteins (Figure 3—See color section, p. C-0, at back of book). The membrane distal α1 and α2 domains of CD1d1 were found to adopt the typical antigen-binding superdomain structure found in all MHC class I and II molecules, consisting of two antiparallel stretches of α-helices overlying a floor of β-pleated sheet. As in all MHC class I structures, the two membrane distal domains of the heavy chain are supported by an immunoglobulin-like α3 domain and its associated β2-m subunit. The most unique and potentially revealing aspect of the CD1d1 structure relates to its putative ligand binding groove. Similar to both classes of MHCencoded antigen-presenting molecules, this consists of an opening between the two rows of α-helices that descends into a cavity within the core of this portion of the protein. However, in CD1d1 this cavity is much deeper than the peptide binding grooves of MHC class I or II. Instead of having a series of six to nine small pockets to accommodate individual amino acid side chains, as in MHC molecules, the CD1 groove is more accurately described as having just two large pockets, designated A0 and F0 . Most notably, the interior of the CD1 groove is formed mainly by hydrophobic amino acid residues. This creates a surface that is almost entirely of neutral electrostatic potential and has little or no potential for hydrogen bonding or other polar interactions. The groove in CD1d1 appears to be closed at either end and is covered over much of its length, such that it may be accessible only through a narrow entrance extending from the center of the groove to the center of the F0 pocket. These features strongly suggest that the groove of CD1d1 would not be likely to interact with its ligands in the same way that peptide binding to MHC molecules is known to occur, and they also indicate that CD1d1 would most likely bind very hydrophobic ligands. Structures of other CD1 proteins are not yet available, but molecular modeling of some of these (e.g. human CD1b and -c) suggest features similar to

P2: NBL/plb

February 27, 1999 16:50

QC: NBL/tkj

Annual Reviews

Figure 3 Comparison of the crystal structures of CD1 and MHC class I. (Left) Backbone ribbon diagram of mouse CD1d1: (red) α-helices; (blue) β-strands; and (brown) loops. (Center) Ribbon diagram of the mouse MHC class I molecule H-2Kb: (cyan) α-helices; (green) β-strands; and (brown) loops. (Right) Superposition using alignment of β2m-domains highlights some of the differences between CD1d1 and H-2Kb. Note in particular the shifting of the α-helices in the α1 and α2 domains. This produces a deeper and more voluminous groove in CD1d1, which is narrower at its entrance compared with H-2Kb. The color scheme is the same as in the left and center panels. (Reprinted with permission from Reference ?.)

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

P1: SAT/VKS T1: NBL

AR078-01

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

ANTIGEN PRESENTATION BY CD1

305

those observed for murine CD1d1 (M Degano, B Segelke, IA Wilson, personal communication).

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Cellular Expression and Tissue Distribution of CD1 Proteins The expression of CD1 proteins has been extensively studied in humans and rodents. The group I CD1 proteins in humans were first identified as differentiation markers expressed on immature cortical thymocytes. It is now well known that these proteins are also expressed on a variety of specialized antigen-presenting cells, especially dendritic cells present in both lymphoid and nonlymphoid tissues (57, 77). Group I CD1 proteins are also inducible in vitro on circulating human monocytes by exposure to granulocyte macrophage–colony stimulating factor (GM-CSF) (78, 79), which suggests that they might be up-regulated on tissue macrophages in many inflammatory lesions. A subset of human B lymphocytes also expresses the CD1c protein (80). This appears to be developmentally regulated and is much more prominent on circulating B cells of infants than adults (81, 82). In addition, CD1c expression has been shown to be up-regulated on marginal zone B cells of lymphoid follicles (83). Information concerning the regulation and tissue expression of group II CD1 proteins (i.e. human CD1d and its homologues) indicates both similarities and differences compared with the human group I CD1 proteins. Several studies of human CD1d have found expression of these proteins by normal human gastrointestinal epithelia (25, 72, 84). In contrast, group I CD1 proteins appear not to be expressed at this site. The putative CD1d protein expressed by gut epithelia appears unusual in that a large fraction of it lacks glycosylation and appears to be expressed on the cell surface mainly without associated β2-m (72). One study in mice (85) and one in rats (86) also reported group II CD1 expression by the intestinal epithelium, but this has not been consistently found (73). The reason for the discrepancies remains unclear but could relate to the use of different monoclonal antibody reagents for detection. It has also been shown that group II CD1 proteins are widely expressed on hematopoietic cells in both humans and mice (73, 87; M Exley, SA Porcelli, SP Balk, unpublished data). In both species, expression has been detected at high levels on a majority of thymocytes. The expression of group II CD1 (CD1d1) appears to be down-regulated during the thymic maturation process, but in mice this is not complete and substantial residual expression can still be detected on most mature mouse T cells (73, 87). Mouse CD1d1 is also expressed constitutively by most B cells and is particularly up-regulated on a population of splenic marginal zone B cells (87). Human CD1d has also been detected on a subset of circulating T and B lymphocytes and resting monocytes, although generally at rather low levels (M Exley, SA Porcelli, SP Balk, unpublished data).

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

306

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

Although studies in this area are ongoing, the overall impression at this point is that expression of group II CD1 is more strongly constitutive in mice than in humans, in which it may be regulated by inducible factors. Interestingly, CD1d is not up-regulated by GM-CSF treatment of human monocytes in vitro under the same conditions that strongly up-regulate group I CD1 expression (SA Porcelli, unpublished data), and cytokines or other factors that regulate the expression of group II CD1 proteins have not yet been discovered. These differences in cellular expression and regulation of groups I and II CD1 further support the classification of the CD1 proteins into these two categories. Expression of group II CD1 by dendritic cells has been less extensively studied than group I CD1 expression on this cell type. However, mouse splenic dendritic cells constitutively express CD1d1 (87), and preliminary studies indicate that isolated human epidermal dendritic cells (i.e. Langerhans cells) are uniformly positive at a low to moderate level for CD1d (M Sugita, SA Porcelli, unpublished data).

CD1 AND T CELL RECOGNITION The discovery that human T cells recognize CD1 and mediate typical cellmediated immune functions constitutes the first solid evidence that CD1 gene products play an immunological role. The MHC class I–like structure of CD1 proteins and their prominent expression on antigen-presenting cells provided the initial stimulus to investigate the role of these molecules in T cell activation and antigen presentation. The first direct evidence implicating CD1 in T cell function was provided by human circulating CD4−CD8− T cell clones expressing either αβ or γ δ TCRs that lysed tumor cells expressing specific isoforms of human CD1 (i.e. CD1a or CD1c) (88). This finding was subsequently confirmed for circulating γ δ T cells (89), and similar findings were also reported for human intestinal intraepithelial lymphocyte lines (90). Several years after these initial findings in the human system were reported, studies of mice also demonstrated CD1-reactive T cells, both in the residual CD4+ population of MHC class II–deficient mice (74) and in the NK1+ T cell fraction of normal mice (52). Interestingly, all the CD1-reactive T cells demonstrated in these initial studies were responsive in the absence of any deliberately added foreign antigen, which suggests that such responses represent a form of T cell autoreactivity that is inherent in the normal lymphocyte pool. A second significant step was made with the derivation of human T cell lines that responded to Mycobacterium tuberculosis antigens in a CD1b-restricted fashion (78). The antigen-specific responses of these T cells were absolutely dependent on CD1b expression and could be demonstrated to be independent of MHC class I and II expression by the antigen-presenting cells. The existence of CD1b-restricted T cells specific for mycobacteria was subsequently

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN PRESENTATION BY CD1

307

confirmed in other studies (91–93) and extended to show similar T cell responses restricted by CD1c (94) and CD1a (JP Rosat, MB Brenner, personal communication). Although the first examples of CD1-restricted T cells were phenotypically CD4−8− or double negative (DN) T cells, it is now apparent that CD1 recognition or restriction is more broadly distributed among T cell subsets. In particular, CD1-restricted T cells specific for mycobacterial antigens within the CD8+ subset have been clearly demonstrated (95; JP Rosat, MB Brenner, personal communication). These T cells express the CD8 αβ heterodimers classically found on MHC class I–restricted cytolytic T lymphocytes (CTL). It is currently unclear what role if any the CD8 molecule plays in human CD1-restricted T cell recognition of antigen. However, several studies indicate that mouse CD1 interacts with CD8 molecules (115, 152, 153). Recent studies suggest the existence of human CD4+ T cells responsive to mycobacterial antigens presented by CD1 proteins (P Sieling, RL Modlin, unpublished data), and many mouse T cells reactive with CD1 are also CD4+ (74, 96). Thus, the CD1 family may be involved in the function of T cells found within all the major phenotypic subsets as currently defined. The recognition of CD1 or CD1-restricted microbial antigens by T cells shows all of the hallmarks of immune recognition mediated by antigen-specific, clonotypic TCRs, and this has now been definitively established by TCR gene transfer studies. Thus, the transfection of cloned TCRα and TCRβ cDNAs isolated from CD1-restricted, mycobacteria-specific T cell lines into Jurkat cells conferred both CD1 restriction and antigen specificity on the resulting transfectants (E Grant, MB Brenner, personal communication). In limited studies carried out to date, the TCRs of human mycobacteria-specific, CD1-restricted T cells have been found to be formed from a variety of different germline V and J segments, and to encode substantial junctional diversity (E Grant, MB Brenner, personal communication). Thus, antigen recognition through this pathway, as for the MHC-dependent pathways, involves a range of clonally diverse receptors and may, therefore, mediate recognition of a wide array of potential foreign antigens.

CD1-Restricted NK T Cells Most knowledge about the function of CD1 in mice has come from studies of a unique subset of T cells often referred to as NK T cells, so called because of their expression of cell-surface proteins previously associated mainly with the natural killer (NK) cell lineage (96, 97). Several lines of investigation over the past 10 years have revealed a range of unusual properties for this specialized population of T cells in mice. Perhaps foremost among these is their expression of an invariant TCRα chain (Vα14-Jα281 with no N region additions or deletions) and limited TCRβ chain repertoire (98), thus endowing them with an antigen

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

308

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

receptor repertoire of limited diversity. In addition, these T cells are unusual in their ability to secrete large amounts of interleukin-4 (IL-4) and other cytokines rapidly upon TCR engagement (99, 100). The development of mouse NK T cells has been shown to be dependent on expression of CD1d1 because these cells fail to develop in mice that have had either the CD1D1 gene (53) or both mouse CD1 genes (101, 102) inactivated by targeted gene disruption. Consistent with this finding, isolated mouse NK T cells and NK T cell hybridomas appear to be universally reactive to mouse CD1 proteins in studies carried out in vitro (52). The functions of NK T cells in the immune response remain unclear, but many intriguing observations have been made that suggest they are an important component of the immune system. Numerically, they represent a major fraction of the T cell compartment, accounting for 20–30% of the T cells in the liver and bone marrow and up to 1% of splenocytes, amounting to a total number of about one million T cells in each of these tissues in mice (96). Given that these cells have limited TCR variability and therefore are likely to be responding to a narrow spectrum of ligands, a population of this size is likely to give rise to substantial responses. The rapid production of IL-4 by NK T cells has implicated them in the early programming of immune responses, in some cases leading to outcomes associated with Th2 responses that are important for antibody production and immunity to extracellular parasites (103). However, in other cases their activation has been associated with outcomes more typical of a Th1-type inflammatory response (104). Other lines of investigation have recently shown a profound role of mouse NK T cells in IL-12–induced tumor rejection (105) and also a potential role in regulating autoimmunity (106, 154). Also consistent with the hypothesis that NK T cells perform a critical role in immunity is the marked evolutionary conservation of this subset between mice and humans. Thus, recent studies have identified human T cells expressing an invariant Vα24-JαQ TCRα chain highly homologous to that expressed by murine NK T cells and coexpressing a variety of NK cell-associated markers (107–109). These human NK T cells also produce high levels of both IL-4 and interferon-γ upon activation and are almost universally reactive with the human CD1d protein expressed in various cell types by transfection (107). As in mice, their function remains unknown, although recent studies have linked reductions in their numbers and alterations in their cytokine secretion patterns to progression of human autoimmune disorders (110, 111). The issue of how the NK T cell populations in humans and mice relate to other CD1-restricted T cell subsets, such as those reactive with mycobacterial antigens, remains to be resolved. The relatively fixed TCR structure of NK T cells stands in marked distinction to the diverse TCR repertoire observed for CD1-restricted mycobacteria-specific T cells in humans. Moreover, emerging

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

ANTIGEN PRESENTATION BY CD1

309

data suggest that even in the murine system, CD1 restriction extends to populations of T cells that do not express the canonical Vα14-Jα281 invariant TCRα chain used by most or all NK T cells (54, 74; SM Behar, MB Brenner, personal communication). This finding, along with the extensive data now available from studies of human CD1-restricted T cells, strongly suggests that development and activation of the NK T cell system is likely to be only one aspect of the function of CD1 in the murine model.

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Identification of CD1-Presented Antigens Perhaps the most striking feature of the CD1 system is the chemical identity of the antigens it presents to T cells (Figure 4). Direct purification of the antigen recognized by the prototype CD1b-restricted M. tuberculosis–specific human T cell line DN1 revealed this to be a subset of mycolic acids, a heterogeneous group of branched- and long-chain fatty acids unique to mycobacteria and a restricted group of related bacterial species (112). This finding suggested the remarkable conclusion that CD1 proteins could be antigen-presenting molecules that have evolved the ability to present nonpeptide lipid antigens to T cells. Subsequent studies confirmed and extended this finding, showing CD1b-restricted T cell recognition of structurally defined glycolipids (91, 113). It has also now been established that both human CD1c (94; DB Moody, SA Porcelli, unpublished data) and CD1a (JP Rosat, MB Brenner, personal communication) can similarly present lipid antigens of mycobacteria to T cells. Comparison of the structures of the known and proposed CD1-presented lipid antigens suggests a structural motif that is common to all these compounds (Figure 4). This consists of a hydrophobic portion composed of branched or dual acyl chain function, which is covalently coupled to a hydrophilic cap formed by the polar or charged groups of the lipid and its associated carbohydrates. Several studies have provided insight into how the structural features of CD1-presented lipid and glycolipid antigens correlate with their ability to be presented to and recognized by T cells. Initial studies of the M. tuberculosis mycolic acids presented by human CD1b demonstrated that recognition of this lipid was completely blocked when the carboxylate was derivatized with a bulky bromophenacylbromide group (112). This suggested that the hydrophilic end of this lipid participated in the specific interaction of this antigen with either the CD1b protein or with the TCR of the responding T cells. Subsequently, these studies were extended by an analysis of two Mycobacterium leprae–reactive, CD1b-restricted T cell lines recognizing a major structurally defined mycobacterial cell wall constituent known as lipoarabinomannan (LAM) (91). This molecule belongs to the family of glycosylphosphatidyl inositols and is composed of a hydrophobic lipid-containing phosphatidyl inositol group attached to a large and complex hydrophilic heteropolysaccharide (114).

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

310

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

Figure 4 Structures of CD1b-presented antigens. (Left) Representative members of three different classes of CD1b-presented antigens. Each of these antigens is a naturally occurring glycolipid component of mycobacterial cell walls. (Right) The CD1d antigen is a synthetic glycolipid with a ceramide-like lipid structure that is recognized by most or all mouse and human CD1d-restricted NK T cells. Note that all these antigens have a common general structure composed of a hydrophilic head group and two aliphatic tails, thus defining one proposed structural motif for a class of CD1presented antigens.

Studies of these T cell lines showed that the acyl chains of LAM were absolutely necessary for presentation of LAM to T cells because removal of these by alkaline hydrolysis resulted in loss of activity. A requirement for some but not the entire carbohydrate portion of LAM was also demonstrated for recognition of this compound by CD1b-restricted T cells. In this case, one T cell line required the presence of virtually the entire carbohydrate region for optimal recognition, whereas the second T cell line recognized a subunit of LAM with

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN PRESENTATION BY CD1

311

a smaller carbohydrate backbone composed of the inositol group with two to six covalently linked mannose groups (PIM2–6). These results provided further support for the hypothesis that the hydrophilic (in this case carbohydrate) portion of the antigen was involved in specific TCR-mediated recognition, while also confirming the necessity of the hydrophobic lipid moiety for presention by CD1. A more complete and detailed demonstration of the relation between the chemical structure of the antigen and its recognition by CD1-restricted T cells has come from studies using the CD1b-presented glycolipid antigen glucose monomycolate (GMM) (113). T cells reactive with a naturally occurring form of GMM containing a C80 wax-ester mycolate could also recognize a fully synthetic GMM containing a simple C32 mycolate, indicating that the fine structure of the lipid moiety was not critical to the presentation and recognition of this glycolipid. In contrast, elimination of the branched structure of the lipid moiety led to complete loss of recognition, which suggests that the dual alkyl chain motif was required. Most notably, the T cell response to GMM could be shown to be extremely specific for the precise stereochemistry of the polar head group, as substituting mannose or galactose for glucose led to a complete loss of T cell recognition. These findings, together with recent information on the structure of CD1 proteins, suggest a straightforward mechanism for lipid and glycolipid presentation by CD1 proteins, which is discussed in detail below. Studies of antigen presentation by mouse CD1d molecules have also identified a number of specific ligands that bind to CD1 and are presented to murine T cells. In contrast to the lipids and glycolipids consistently found in studies seeking natural microbial antigens produced by human CD1 proteins, initial studies in the murine system sought and identified synthetic peptide ligands for recombinant mouse CD1d1 by using a powerful random peptide library screening technique (115). This yielded a collection of peptides with high CD1d1 binding affinity (KD ∼ 10−7 M) that were quite hydrophobic and appeared to contain a motif with three anchor positions occupied by aromatic or bulky hydrophobic amino acids. CD1-restricted T cells specific for these peptides could be demonstrated, which suggests that these findings were of immunological significance. Also raising the possibility of peptide antigen presentation by mouse CD1, one report has recently appeared in which a DNA vaccination approach appeared to give rise to mouse CD1d1-restricted T cells specific for the protein antigen ovalbumin (116). Together, these results raised the interesting possibility that mouse CD1d1, and possibly other CD1 proteins, might maintain at least some potential for interacting with peptides in a manner that leads to their presentation to T cells. However, the possibility that mouse CD1d1 plays a major role as a peptidepresenting molecule now seems less appealing in light of several other findings favoring the hypothesis that the antigens bound and presented by mouse CD1

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

312

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

proteins, as for their human counterparts, are predominantly lipids and glycolipids. Thus, mouse CD1-restricted NK1.1+ T cells have recently been shown to recognize synthetic ceramide-containing glycolipid antigens (117, 154a, 158). Reactivity to these glycolipids is also consistently seen for human CD1drestricted NK T cell clones (155, 156). These synthetic compounds have an unusual structure, consisting of an α-anomeric hexose sugar (glucose or galactose) linked to an acylphytosphingosine moiety. Lipids with this structure are currently known to occur naturally only in marine sponges, but it is possible that they represent structural analogues of lipid antigens that occur in relevant pathogens or in abnormal tissues. It has also been demonstrated in one study that immunopurified mouse CD1d1 contains associated lipids, whereas no associated peptides could be detected (118). In fact, structural studies of acid-eluted ligands from mouse CD1d1 have identified cellular glycosylphosphatidyl inositols (GPI) as possibly the major bound endogenous ligands of this protein. The possible significance of this result is emphasized by the fact that the structure of mammalian GPI is extremely similar to that of the mycobacterial antigens LAM and PIM, GPIs identified in earlier studies of human CD1b-presented antigens (91). An interesting possibility to consider is that structural variations in the endogenous GPIs or other lipids presented by CD1d1 on different cell types could account for the patterns of tissue-specific recognition of CD1d1 observed for mouse NK T cell hybridomas (54, 119).

UPTAKE AND PROCESSING PATHWAYS OF CD1-PRESENTED ANTIGENS The identification of CD1 as a system for the presentation of lipid antigens raises numerous fundamental questions about how such antigens are taken up and processed by antigen-presenting cells. In all cases so far studied, CD1-restricted T cell recognition of nonpeptide antigens could be shown to require uptake and delivery to an intracellular compartment in APCs (78, 91). Furthermore, like MHC class II presentation, the presentation of lipid antigens by human CD1b (78, 91) and by murine CD1d1 (117) is inhibited by agents that prevent endosomal acidification (e.g. chloroquine and concanamycin A), indicating a crucial endosomal step in the pathway. However, the peptide transporter complex TAP-1/2, which is required for assembly and stable expression of MHC class I proteins, is not required either for expression of CD1 proteins (120, 121) or for their antigen-presenting function (78, 94). Likewise, HLADM complexes, which are required for efficient antigen presentation by MHC class II, are not necessary for normal expression and function of human CD1b or CD1c proteins (78, 94). These findings indicate that the intracellular pathways

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN PRESENTATION BY CD1

313

involved in protein assembly and antigen processing are in some ways distinct between CD1 and MHC class I and II, as might be anticipated from the marked differences in the properties of the antigens presented by these systems. Current evidence suggests that either exogenous (i.e. taken up by phagocytosis or endocytosis) or endogenous (i.e. produced by pathogens living within an APC) lipid antigens can enter the CD1 antigen processing route and be presented to CD1-restricted T cells. The mechanism by which exogenous lipid antigens are taken up by APCs has been studied in detail only for LAM, which represents a large and heavily glycosylated CD1b-presented glycolipid antigen (122). It has been clearly demonstrated that the macrophage mannose receptor (MR) is involved in the uptake of LAM, probably through its ability to bind the mannose core of this mycobacterial glycolipid. The presentation of LAM by monocyte-derived CD1+ APCs has been shown to be dependent on MR uptake of the antigen because presentation is completely blocked when this process is competitively inhibited by soluble mannan and by antibodies to the MR. Furthermore, CD1b+ transfectant cell lines that lack expression of MR are unable to present LAM (SA Porcelli, unpublished data), although they are capable of taking up and presenting other lipid and glycolipid antigens (78, 94, 113). Overall, these results suggest that the uptake of relatively small and predominantly hydrophobic antigens (such as mycolic acids and GMM) may not involve interaction with specific receptors on APCs, whereas the much larger and more hydrophilic glycolipids such as LAM may require specific receptor-mediated uptake for presentation. Uptake of CD1-presented lipid antigens by APCs leads to their delivery to endosomes, and the requirement for endosomal acidification in CD1b-restricted presentation suggests that association of lipid antigens with CD1 most likely occurs in an acidic endosomal compartment. This hypothesis is strongly supported by the finding that the human CD1b protein localizes prominently in a variety of acidic endosomal compartments in APCs, including those in which MHC class II molecules are known to be loaded with peptides (122, 123). These MHC class II–containing compartments (MIICs) are lipid-rich late endosomes with either a multilammellar or multivesicular membrane arrangement. Localization of CD1b to MIICs and other endocytic compartments is mediated by a targeting motif (YXXZ, where Y is tyrosine, X is any amino acid, and Z is a bulky hydrophobic residue) in the short cytoplasmic tail of this protein (123). Similar results have been reported for mouse CD1d1 (119). Sequences corresponding to the targeting motif interact with at least two different adapter protein complexes (AP-1 and AP-2), which direct proteins bearing the motif into clathrin-coated pits and vesicles (124). In the case of human CD1b, cytoplasmic tail-mediated endosomal targeting has been shown to be required for efficient presentation of exogenous lipid antigens to CD1b-restricted T cells

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

314

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

(125). A YXXZ motif is also present in most other CD1 proteins, with one clear exception being human CD1a, which appears not to traffic to MIICs and may be loaded with antigens at a different site (M Sugita, MB Brenner, personal communication). The lipid-rich composition of the MIICs may provide an ideal site in which to concentrate foreign lipid antigens, such as those presented by CD1b. In fact, one CD1b-presented antigen, LAM, has been shown to colocalize with CD1b at this site (122). The acidic pH (∼4.0) of MIICs also may be important, as this promotes a conformational change in CD1b that increases accessibility to the hydrophobic interior of the protein and facilitates binding of lipid ligands by purified CD1b in vitro (126). In addition, MIICs contain a wide array of degradative enzymes, which may be involved in the trimming of large glycan components of some CD1-presented antigens, such as those found in LAM. Currently, little is known about the potential of antigen-presenting cells to modify the covalent structure of bacterial lipid antigens. However, one recent study has shown that trehalose-6,6’-dimycolate, a relatively small mycobacterial cell wall glycolipid that contains both mycolic acid and GMM as components of its structure, is not processed by APCs into components that can be recognized by mycolic acid or GMM-specific CD1b-restricted T cells (127). This suggests that for some bacterial lipid antigens, APCs may not possess enzymes capable of cleaving their covalent bonds, and that intracellular antigen processing may reflect only the requirements for loading intact lipid antigens or their precursors onto CD1. In contrast, for LAM, which has an extremely bulky polysaccharide component, it has been proposed that enzymatic processing to reduce the antigen to a smaller core structure most likely is a component of the intracellular processing (91, 122). In addition to the exogenous pathway of antigen loading for CD1b, there is an endogenous pathway as well. This has been demonstrated by infecting APCs in vitro with virulent M. tuberculosis (95, 125). Both DN and CD8+ T cell lines were able to recognize and lyse M. tuberculosis–infected targets in a CD1b restricted manner. Effective antigen presentation in this system also depended on the trafficking of CD1b to endosomes because deletion of the cytoplasmic tail YXXZ targeting sequence markedly diminished the recognition of infected cells by all CD1b-restricted T cell lines tested. Although the details of this pathway for endogenous lipids are not yet known, it seems unlikely that antigen is loaded into CD1b molecules in the same endocytic vesicles in which the live mycobacteria reside because these compartments do not acidify normally (128, 129). It is more likely that secreted or shed lipid antigens of M. tuberculosis are able to translocate to other subcellular compartments that do not harbor bacilli and are therefore able to acidify. In support of this hypothesis, immunoelectron microscopy of M. tuberculosis–infected macrophages indicates that

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

ANTIGEN PRESENTATION BY CD1

315

LAM is present in endosomal structures that do not contain M. tuberculosis organisms (129).

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Mechanism of Lipid Antigen Interaction with CD1 Proteins Together with the demonstration that naturally occurring CD1-presented antigens are lipids, the structure of the putative CD1 ligand-binding grooves points strongly to a molecular model for how antigens interact with these proteins in a way that leads to specific recognition by the antigen receptors of T cells (113, 126, 130). Although unlikely to bind peptides in the manner of MHC class I and II molecules, the putative antigen-binding groove demonstrated in mouse CD1d1, and predicted to be present in other CD1 proteins, is well suited to bind the twin acyl tails of the known CD1b-presented antigens (76). This mechanism of binding would mimic that which has been shown for nonspecific lipid transport proteins, which bury the hydrophobic tail of their ligands in an electrostatically neutral binding pocket lined by hydrophobic residues (131). In this way, a relatively nonspecific interaction of the lipid with the hydrophobic surface of the CD1 groove could provide most of the binding energy needed to generate a stable interaction between the antigen and the CD1 protein. The anchoring of lipid antigens into the CD1 groove through hydrophobic interactions would also be predicted to orient amphipathic lipid and glycolipid antigens such that their polar head groups would be positioned outside the ligand-binding groove or near the groove entrance where several hydrophilic residues are present (76). Such positioning of the antigen would in theory allow the hydrophilic head groups and specific residues on the α-helical face of the CD1 protein to interact directly with TCRs, leading to the highly specific recognition that has been observed for CD1-restricted T cells. This proposed mechanism of antigen interaction is strongly supported by findings on the structural basis for CD1-restricted T cell recognition of LAM and PIM (91), and especially by studies of the glycolipid GMM that showed exquisite fine specificity for the carbohydrate but not the lipid component of this antigen (113). Also arguing strongly for this mechanism is the finding that recently obtained crystals of recombinant mouse CD1d1 protein contain a distinct electron density buried in the hydrophobic groove. Although the origin of this material is uncertain, its structure appears most consistent with that of a bound lipid (BW Segelke, AR Casta˜no, EA Stura, PA Peterson, IA Wilson, personal communication). It has also proven possible to directly observe the interaction of human CD1b with three of its known ligands, the mycobacterial glycolipid antigens LAM, PIM, and GMM, using evanescent wave sensor and surface plasmon resonance measurements (126). Binding of intact LAM and GMM was shown to be detectable only at acidic pH, and for LAM the optimal pH was determined

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

316

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

to be pH 4.0. In contrast, an interaction between deacylated LAM and CD1b could not be detected, consistent with the proposal that CD1b directly binds the acyl functions of this antigen. At acid pH, the CD1b-LAM equilibrium binding constant (KD) was determined to be 3.2 × 10−8 M, which indicates an affinity similar to those determined for interactions between immunogenic peptides and MHC class I molecules (132, 133). Binding of PIM and GMM could be shown to have affinities approximately one order of magnitude less than that measured for LAM, and all three of these glycolipids could be shown to bind to the same or closely adjacent sites on CD1b by competition studies. Another study using different techniques has examined the binding affinity between a soluble form of mouse CD1d1 and phosphatidylinositol, the proposed lipid anchor of the major cellular ligand of CD1d1 (118). This revealed a KD at neutral pH of approximately 4 × 10−7 M, which is extremely similar to that measured at acid pH for PIM and GMM binding to CD1b. Whether this reflects a true difference in the pH requirement for lipid ligand binding by different CD1 proteins will require further detailed investigation. Other biophysical studies carried out in vitro have provided further insight into the mechanism by which lipid antigens are loaded onto CD1b, demonstrating that the properties of the CD1b protein are dramatically altered by acidic pH to facilitate its direct interaction with hydrophobic ligands (126). Circular dichroism analysis of CD1b indicates reversible unfolding of the α-helical portions of the molecule at acid pH. This may lead to marked changes in the accessibility of the hydrophobic ligand-binding groove in CD1b because the α-helical portions of the protein form the walls and outlet of the groove. Indeed, for human CD1b it appears that the unfolding of the α-helices in the presence of low pH exposes a hydrophobic binding site, as detected by the enhanced binding and emission of the fluorescent probe 1-anilo-naphthalene-8-sulfonic acid (126). Collectively, these data are consistent with a model in which the hydrophobic ligand-binding groove of CD1b becomes exposed in the acid milieu of the endosome, thus allowing the direct binding of the hydrophobic portions of lipid and glycolipid antigens present at this intracellular site. This mode of antigen binding would bury the hydrophobic alkyl chain component within the core of the CD1 protein and leave the hydrophilic or charged cap of the antigen exposed at the opening of the groove where it may make direct contacts with the TCRs of specific T cells. This model thus accounts for the requirement for a hydrophobic alkyl component in all of the CD1-presented antigens so far studied and explains the relative lack of specificity of T cells for this portion of the antigen. In addition, it also makes clear the reason for the exquisite specificity for structural features of the polar and hydrophilic ends of these antigens (113).

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

ANTIGEN PRESENTATION BY CD1

317

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

THE ROLE OF CD1-RESTRICTED T CELLS IN MICROBIAL IMMUNITY Investigation of human leprosy has provided strong evidence for involvement of the CD1-restricted T cells in host immune response to infection. Leprosy provides an ideal model to study the role of human T cell subsets in host defense against microbial pathogens because infection by the causative bacillus, M. leprae, results in disease manifestations that encompass an immunological spectrum (134). At one pole of the spectrum are patients with tuberculoid leprosy, who are able to restrict the growth of the pathogen. These individuals mount strong cell-mediated immune responses to M. leprae, resulting in a generally benign clinical state characterized by relatively few skin lesions containing low numbers of bacilli. In contrast, at the opposite pole are patients with lepromatous leprosy who are unable to contain the infection. These individuals have poor cell-mediated immunity against the pathogen and, consequently, have widespread lesions with an enormous bacterial burden. In a study of skin biopsy specimens from human leprosy patients, CD1a, -b, and -c expression was found to be up-regulated on mature CD83+ dendritic cells infiltrating dermal granulomas (159). The frequency of CD1+ cells correlated with the level of cell-mediated immunity to M. leprae, being tenfold more abundant in the granulomas of patients with the immunologically responsive tuberculoid form of the disease compared with the unresponsive lepromatous form. The prominence of CD1+ cells in tuberculoid lesions is likely influenced by the local cytokine environment and may directly reflect the high expression of GM-CSF, a key differentiation factor for dendritic cells, in these lesions (135, 136). In contrast, the low frequency of CD1+ cells in lepromatous lesions correlates with low levels of GM-CSF in these lesions and may also be due to directly inhibitory factors in the local cytokine milieu. For example, IL-10 is strongly expressed in lepromatous lesions (135, 136) and is likely to be a key inhibitor of the CD1 system. This cytokine inhibits GM-CSF secretion by antigen-stimulated peripheral blood mononuclear cells (137) and also inhibits the CD1 expression normally induced by GM-CSF on monocytes (138). Because administration of GM-CSF to lepromatous leprosy patients results in the infiltration of CD1a+ cells into skin lesions (139), it may be possible to develop immunotherapeutic strategies to up-regulate this antigen presentation pathway. The bacilli themselves may also influence CD1 expression, as infection of CD1+ antigen-presenting cells with virulent mycobacteria causes a down-regulation of CD1 expression, but not of MHC class I or II expression (160). This effect required infection of the cells with live mycobacteria because heat-killing of the bacteria completely abrogated the effect and was associated with decreased

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

318

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

the steady state CD1 mRNA levels. The down-regulation of CD1 expression by mycobacteria may represent a novel immune evasion mechanism for this pathogen. The finding that mycobacteria have apparently targeted this pathway as part of their strategy in evading the host immune response suggests that the CD1-dependent T cell response is a significant component of host resistance to these pathogens. Consistent with the prominent expression of CD1 in leprosy lesions, it has also proven possible to derive CD1-restricted T cells from the tissue lesions and blood of patients with mycobacterial infections. Initially, a CD1b-restricted, M. leprae–specific T cell line was derived from a cutaneous lesion of a subject infected with M. leprae (91). In addition, M. tuberculosis–reactive T cells that recognize lipid antigens of M. tuberculosis in the context of human CD1 molecules have been derived form normal individuals (78, 94), patients with tuberculosis (95), and patients coinfected with M. tuberculosis and HIV (93). There is currently little information about the precise frequency of CD1-restricted T cells in the T cell repertoire, and what the details are of how these T cells may expand and whether they persist following antigenic challenges are major questions that have not yet been addressed. Several different mechanisms are apparent by which CD1-restricted T cells may contribute to protective immunity to microbial pathogens. Host defense against intracellular pathogens, such as mycobacteria, is thought to involve two major effector T cell pathways. First, mycobacteria-specific group I CD1restricted T cells release high levels of interferon γ and low levels of IL-4 (93), typical of the Th1 pattern of cytokines required for activation of macrophagemediated killing of intracellular pathogens and development of effective cellmediated immunity against such organisms (91). Second, mycobacteria-reactive CD1-restricted T cells typically show a high degree of cytolytic activity in vitro against antigen-pulsed CD1+ mononuclear phagocytes (95, 127), and they also recognize and lyse CD1+ targets infected with live virulent M. tuberculosis bacilli (95, 125). Lysis of chronically or productively infected macrophages would be expected to contribute to host defense either by directly killing the bacteria or by disbursing the pathogen and thereby allowing freshly recruited macrophages to take up and destroy it (140, 141). The apoptotic death of mononuclear phagocytes harboring bacteria could limit the reservoir of host cells for the pathogen and also increase the ratio between Th1 cytokine producing T cells and infected cells. Infected macrophages that have undergone apoptosis can also be rapidly ingested by dendritic cells, which may facilitate the generation of additional CTL to combat the infection (142). Studies of the mechanisms by which CD1-restricted M. tuberculosis–specific T cells lyse mycobacteria-infected target cells have revealed an interesting dichotomy in the lytic pathways used by different phenotypic subsets of CTLs

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN PRESENTATION BY CD1

319

(95). The cytotoxicity of CD4−CD8− CD1-restricted T cells was mediated by Fas/FasL interaction because anti-Fas and anti-FasL antibodies could block cytotoxicity. In contrast, target cell lysis by CD8+ T cell clones (either CD1restricted or MHC class I-restricted) was not inhibited by anti-Fas or anti-FasL antibodies but could be blocked by depletion of cytotoxic granules by strontium pretreatment. Lysis of infected target cells by CD4−CD8− T cells had no effect on the viability of intracellular M. tuberculosis bacilli, whereas CD1-restricted CD8+ T cells killed approximately 50% of the intracellular bacteria during an 18-h coincubation with infected cells. Recent data link this direct antimicrobial effect of CD8+ CTL (including those that are either CD1 or MHC class II restricted) to their expression of granulysin (157), a protein component of CTL granules (143, 144). The reason for the existence of two distinct subsets of CTL that use different mechanisms to kill infected cells is not yet clear, but it is likely that these contribute in different ways to host defense against intracellular infection.

The Potential Range of CD1-Presented Lipid and Glycolipid Antigens At present, nearly all of the known naturally occurring CD1-presented foreign antigens have been isolated from mycobacteria. However, it seems likely that T cell recognition through the CD1 system will extend to a much wider range of microbial pathogens and may also include antigens of host cell origin. It is in fact already known that human CD1b-restricted T cells can recognize glycosylated mycolates produced by organisms that are phylogenetically related to the mycobacteria, including rhodococci and nocardia (DB Moody, SA Porcelli, unpublished data). Based on the putative motif of dual alkyl chains linked to a polar cap that has been proposed for CD1b-presented antigens, a variety of other more widespread potential targets for CD1-restricted T cells can be postulated (Figure 5). These include the lipoteichoic acids of gram-positive organisms and also components or precursors of the ubiquitous lipopolysaccharides of gram-negative bacteria (145). The basic dual alkyl chain motif is also found in the abundant capsular polysaccharides of virulent gram-negative bacilli such as Haemophilus influenzae and Neisseria meningitidis, which are major targets of protective antibody responses against these organisms. Although such responses have generally been thought to be T cell independent, it has recently been suggested that CD1restricted T cells could be involved (146). This possibility is supported by the expression of CD1 proteins by B cells, as shown first in humans (80) and more recently in mice (73, 87). Other complex pathogens, such as protozoal or multicellular parasites, are also known to harbor a range of unique lipids that could hypothetically function as CD1-presented antigens. Although a few

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

320

12:3

QC: NBL/apm

Annual Reviews

T1: PSA

AR078-11

PORCELLI & MODLIN

Figure 5 Some examples of the potential range of CD1-presented antigens are depicted, based on the proposed CD1b antigen motif illustrated schematically in the center of the diagram. The glycolipids illustrated for mycobacteria and actinomycetes are known to be recognized by CD1brestricted T cells (91, 112; DB Moody, SA Porcelli, unpublished data). The glycolipids illustrated below have not yet been shown to be recognized by CD1-restricted T cells. These are examples of potential self antigens found in normal human tissues that conform to the CD1b antigen motif and common glycolipids of gram-negative (lipopolysaccharide) and gram-positive (lipoteichoic acid) bacteria that may also contain the motif within their structures. GMM, Glucose monomycolate.

examples, such as the lipophosphoglycans of Leishmania species (147), have been partially explored in terms of their structure and antigenicity, this remains for the most part a largely uncharted territory in the search for molecular targets of immune recognition. We anticipate future studies delineating involvement of CD1-restricted T cells in these and other infectious diseases. The possibility that lipids produced by self tissues could also act as CD1presented antigens has implications for autoimmunity and tumor immunity. Multiple examples of T cells that recognize CD1 proteins in the absence of deliberately added foreign antigens, and therefore appear to be autoreactive T cells, have been isolated from both humans (78, 89, 90, 94) and mice (52, 54, 74, 119).

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN PRESENTATION BY CD1

321

The recent finding that cellular glycolipids can associate with CD1 proteins has strengthened the view that such T cells may in fact be responsive to self-lipid ligands bound to CD1 (118). This implies that mechanisms must exist to maintain tolerance for CD1-restricted T cells, presumably at least in part through positive and negative selection steps that are analogous to those that take place in the thymus for conventional MHC-restricted T cells. Ceramide-containing self glycolipids such as glucocerebroside and other gangliosides are similar in their overall structure to some of the known bacterial and synthetic lipid antigens recognized by human and mouse CD1-restricted T cells. The possibility that these could become altered or dysregulated in inflammatory diseases affecting lipid-rich tissues, such as multiple sclerosis, remains an attractive mechanism for activating CD1-restricted T cells in such lesions. In addition, many known tumor antigens are glycolipid-associated carbohydrate epitopes including gangliosides and other ceramide-containing structures (148), and this could represent another situation in which the cell-mediated immune response might be targeted to nonpeptide antigen recognition by CD1.

CONCLUDING REMARKS The discovery of nonpeptide lipid and glycolipid antigen recognition by CD1restricted T cells defines a new paradigm for immune recognition and provides a novel mechanism for host responses to infection. The existence of this MHC-independent pathway for T cell activation may substantially expand the immune repertoire and could have important implications for many aspects of cell-mediated immunity. It seems likely that CD1 has evolved as part of the unique adaptation of the immune system to its task of combating a myriad of microbial pathogens. The implications of lipid antigen presentation to T cells are potentially broad and may extend to antimicrobial and anti-tumor immunity as well as to immunoregulation and autoimmunity. One immediate practical consideration for this new insight into antigen presentation resides in the area of vaccine development. Much effort is being devoted to the development of protein subunit vaccines for bacterial and parasitic diseases, which may vary in effectiveness according to the MHC haplotype of the individual and the ability of the microbe to modulate the particular epitopes targeted. Because CD1 molecules are nonpolymorphic, the nonpeptide antigens they present may offer particular advantages as vaccine subunits. In addition, CD1-presented lipid antigens appear to be critical components of microbial organisms that can not be readily altered by random single-step mutations, and thus they represent relatively fixed targets. A next critical step will be to determine whether CD1restricted T cell responses to nonpeptide antigens can contribute significantly to protective immune responses to microbial pathogens. Studies employing

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

322

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

PORCELLI & MODLIN

CD1-restricted antigens as vaccines offer new promise in the fight against infectious disease and should also provide an integrated understanding of how CD1 participates in the immune response.

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ACKNOWLEDGMENTS This work was supported by grants from the NIH (AI 22553, AI 36069, AR 40312, AR 01854, and AI 40135), the Arthritis Foundation (SAP, Investigator Award), the American Cancer Society (SAP), and the UNDP/World Bank/World Health Organization Special Programme for Research and Training in Tropical Diseases (IMMLEP) (RLM). We wish to note our special appreciation for the insight and encouragement we have derived from long-term interactions with Drs. Barry Bloom and Michael Brenner. In addition, we thank Peter Sieling, Bill Ernst, Ethan Grant, Sam Behar, Branch Moody, Ken LeClair, Jenny Gumperz, Brent Segelke, Massimo Degano, Ian Wilson, Mitchell Kronenberg, and numerous other colleagues for helpful discussions and for sharing unpublished data. Special thanks to Branch Moody and Robin Jackman for help in preparation of the figures for this article. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Pamer EG, Cresswell P. 1998. Mechanisms of MHC class I-restricted antigen processing. Annu. Rev. Immunol. 16:323–58 2. Cresswell P. 1994. Assembly, transport, and function of MHC class II molecules. Annu. Rev. Immunol. 12:259–93 3. Shapiro ME, Onderdonk AB, Kasper DL, Finberg RW. 1982. Cellular immunity to Bacteroides fragilis capsular polysaccharide. J. Exp. Med. 155:1188– 97 4. Jackson S, Folks TM, Wetterskog DL, Kindt TJ. 1984. A rabbit helper T cell clone reactive against group-specific streptococcal carbohydrate. J. Immunol. 133:1553–57 5. Mitchell GF, Handman E. 1985. Tlymphocytes recognise Leishmania glycoconjugates. Parasitol. Today 1:61–63 6. Katsuki M, Kakimoto K, Kawata S, Kotani S, Koga T. 1987. Induction of delayed-type hypersensitivity by the T cell line specific to bacterial peptidoglycans. J. Immunol. 139:3570–72 7. Domer JE, Garner RE, Befidi-Mengue RN. 1989. Mannan as an antigen in cell-

8.

9.

10.

11.

12.

mediated immunity (CMI) assays and as a modulator of mannan-specific CMI. Infect. Immun. 57:693–700 Moll H, Mitchell GF, McConville MJ, Handman E. 1989. Evidence of Tcell recognition in mice of a purified lipophosphoglycan from Leishmania major. Infect. Immun. 57:3349–56 Baer H, Chaparas SD. 1964. Tuberculin reactivity of a carbohydrate component of unheated BCG culture filtrate. Science 146:245–47 Mehra V, Brennan PJ, Rada E, Convit J, Bloom BR. 1984. Lymphocyte suppression in leprosy induced by unique M. leprae glycolipid. Nature 308:194– 96 Nataraj C, Brown ML, Poston RM, Shawar SM, Rich RR, Fischer Lindau K, Kurlander R. 1996. H2-M3wt-restricted, Listeria monocytogenes-specific CD8 T cells recognize a novel, hydrophobic, protease-resistant, periodatesensitive antigen. Int. Immunol. 8:367– 78 Tanaka Y, Morita CT, Nieves E, Brenner MB, Bloom BR. 1995. Natural and

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

ANTIGEN PRESENTATION BY CD1

13.

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

synthetic nonpeptide antigens recognized by human gamma/delta T cells. Nature 375:155–58 Constant P, Davodeau F, Peyrat M, Poquet Y, Puzo G, Bonneville M, Fournie J. 1994. Stimulation of human gamma delta T cells by nonpeptidic mycobacterial ligands. Science 264:267–70 Bernard A, Boumsell L, Hill C. 1984. Joint report of the First International Workshop on Human Leucocyte Antigens by the investigators of the participating laboratories. In Leucocyte Typing—Human Leucocyte Differentiation Antigens Detected by Monoclonal Antibodies, ed. A Bernard, L Boumsell, J Dausset, C Milstein, SF Schlossman, pp. 9–135. Berlin: Springer-Verlag McMichael A, 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 Knowles RW, Bodmer WF. 1982. A monoclonal antibody recognising a human thymus leukemia-like antigen associated with β2-microglobulin. Eur. J. Immunol. 12:676–81 Favaloro EJ, Bradstock KF, Kamath S, Dowden G, Gillis D, George V. 1986. Characterization of a p43 human thymocyte antigen. Dis. Markers 4:261–70 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 Calabi F, Milstein C. 1986. A novel family of human major histocompatibility complex-related genes not mapping to chromosome 6. Nature 323:540–43 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 Martin LH, Calabi F, Lefebvre F-A, Bilsland CAG, Milstein C. 1987. Structure and expression of the human thymocyte antigens CD1a, CD1b, and CD1c. Proc. Natl. Acad. Sci. USA 84:9189–93 Yu CY, Milstein C. 1989. A physical map linking the five CD1 human thymocyte differentiation antigen genes. EMBO J. 8:3727–32 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

24.

25.

26.

27.

28.

29.

30.

31. 32. 33.

34.

35.

36.

323

human thymocyte CD1 antigen genes to chromosome 1. EMBO J. 7:2801–5 Balk SP, Bleicher PA, Terhorst C. 1989. Isolation and characterization of a cDNA and gene coding for a fourth CD1 molecule. Proc. Natl. Acad. Sci. USA 86:252–56 Blumberg RS, Terhorst C, Bleicher P, McDermott FV, Allan CH, Landau SB, Trier JS, Balk SP. 1991. Expression of a nonpolymorphic MHC class I-like molecule, CD1D, by human intestinal epithelial cells. J. Immunol. 147:2518– 24 Bilsland CAG, Milstein C. 1991. The identification of the β2-microglobulin binding antigen encoded by the human CD1D gene. Eur. J. Immunol. 21:71–78 Calabi F, Bilsland CAG, Yu CY, Bradbury A, Belt KT, Martin LH, Milstein C. 1989. Recent progress in the molecular study of CD1. See Ref. 150, pp. 254–58 MacHugh ND, Bensaid A, Davis WC, Howard CJ, Parsons KR, Jones B, Kaushal A. 1988. Characterization of a bovine thymic differentiation antigen analogous to CD1 in the human. Scand. J. Immunol. 27:541–47 Parsons KR, MacHugh ND. 1991. Individual antigens of cattle. Bovine CD1 (BoCD1). Vet. Immunol. Immunopathol. 27:37–41 Mackay CR, Maddox JF, GogolinEwens KJ, Brandon MR. 1985. Characterization of two sheep lymphocyte differentiation antigens, SBU-T1 and SBU-T6. Immunology 55:729–37 Dutia BM, Hopkins J. 1991. Analysis of the CD1 cluster in sheep. Vet. Immunol. Immunopathol. 27:189–94 Hopkins J, Dutia BM. 1991. Workshop studies on the ovine CD1 homologue. Vet. Immunol. Immunopathol. 27:97–99 Hopkins J, Dutia BM, Bujdoso R, McConnell I. 1989. In vivo modulation of CD1 and MHC class II expression by sheep afferent lymph dendritic cells. Comparison of primary and secondary immune responses. J. Exp. Med. 170:1303–18 Rhind SM, Dutia BM, Howard CJ, Hopkins J. 1996. Discrimination of two subsets 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 Calabi F, Belt KT, Yu CY, Bradbury A, Mandy WJ, Milstein C. 1989. The rabbit CD1 and the evolutionary conservation

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

324

37.

38.

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

39.

40.

41.

42.

43.

44. 45.

46.

47.

48.

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

PORCELLI & MODLIN of the CD1 gene family. Immunogenetics 30:370–77 Wang CR, Chen G-H, Mandy WJ. 1987. Identification of a rabbit class I-like thymocyte-specific antigen. J. Immunol. 138:3352–59 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. Isolation and expresion 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 Ichimiya S, Kikuchi K, Matsuura A. 1994. Structural analysis of the rat homologue of CD1. Evidence for evolutionary conservation of the CD1D class and widespread transcription by rat cells. J. Immunol. 153:1112–23 Kasai K, Matsuura A, Kikuchi K, Hashimoto Y, Ichimiya S. 1997. Localization of rat CD1 transcripts and protein in rat tissues: an analysis of rat CD1 expression by in situ hybridization and immunohistochemistry. Clin. Exp. Immunol. 109:317–22 Matsuura A, Hashimoto Y, Kinebuchi M, Kasai K, Ichimiya S, Katabami S, Chen H, Shimizu T, Kikuchi K. 1997. Rat CD1 antigen: structure, expression and function. Transplant. Proc. 29:1705–6 Calabi F, Jarvis JM, Martin L, Milstein C. 1989. Two classes of CD1 genes. Eur. J. Immunol. 19:285–92 Lanier L, Allison JP, Phillips JH. 1986. Correlation of cell surface antigen expression on human thymocytes by multicolor flow cytometric analysis: implications for differentiation. J. Immunol. 137:2501–7 Reinherz EL, Kung PC, Goldstein G, Levey RH, Schlossman SF. 1980. Discrete stages of human intrathymic differentiation: analysis of normal thymocytes and leukemic lymphoblasts of T-cell lineage. Proc. Natl. Acad. Sci. USA 77:1588–92 Lopez-Botet M, Moretta L. 1985. Functional characterization of human thymocytes: a limiting dilution analysis of precursors with proliferative and cytolytic activites. J. Immunol. 134:2299–304 Bhan AK, Reinherz EL, Sibrand P, McCluskey RT, Schlossman SF. 1980. Location of T cell and major histocompat-

49.

50. 51.

52.

53.

54.

55. 56.

57. 58.

59.

60.

61.

ibility complex antigens in the human thymus. J Exp. Med. 152:771–82 Blue ML, Levine H, Daley JF, Branton KRJ, Schlossman SF. 1989. Expression of CD1 and class I MHC antigens by human thymocytes. J. Immunol. 142:2714– 20 Bradbury A, Calabi F, Milstein C. 1990. Expression of CD1 in the mouse thymus. Eur. J. Immunol. 20:1831–36 Balk SP, Bleicher PA, Terhorst C. 1991. Isolation and expression of cDNA encoding the murine homologues of CD1. J. Immunol. 146:768–74 Bendelac A, Lantz O, Quimby ME, Yewdell JW, Bennink JR, Brutkiewicz RR. 1995. CD1 recognition by mouse NK1+ T lymphocytes. Science 268: 863–65 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 Park SH, Roark JH, Bendelac A. 1998. Tissue-specific recognition of mouse CD1 molecules. J. Immunol. 160:3128– 34 Novacek MJ. 1997. Mammalian evolution: an early record bristling with evidence. Curr. Biol. 7:R489–91 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 Porcelli S. 1995. The CD1 family: a third lineage of antigen presenting molecules. Adv. Immunol. 59:1–98 Kasahara M, Mckinney EC, Flajnik MF, Ishibashi T. 1993. The evolutionary origin of the major histocompatibility complex: polymorphism of class II α chain genes in the cartilaginous fish. Eur. J. Immunol. 23:2160–65 Bartl S, Weissman IL. 1994. Isolation and characterization of major histocompatibility complex class IIB genes from the nurse shark. Proc. Natl. Acad. Sci. USA 91:262–66 Okamura K, Ototake M, Nakanishi T, Kurosawa Y, Hashimoto K. 1997. The most primitive vertebrates with jaws possess highly polymorphic MHC class I genes comparable to those of humans. Immunity 7:777–90 van Agthoven A, Terhorst C. 1982. Further biochemical characterization of the human thymocyte differentiation antigen T6. J. Immunol. 128:426–32

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

ANTIGEN PRESENTATION BY CD1 62. Amiot M, Dastot H, Fabbi M, Degos L, Bernard A, Boumsell L. 1988. Intermolecular complexes between three human CD1 molecules on normal thymus cells. Immunogenetics 27:187–95 63. Hughes AL. 1991. Evolutionary origin and diversification of the mammalian CD1 antigen genes. Mol. Biol. Evol. 8:185–201 64. Srivastava R, Lambert ME. 1991. Molecular organization and expression of class I HLA gene family. See Ref. 151, pp. 100–54 65. Parham P, Lawlor DA, Lomen CE, Ennis PD. 1989. Diversity and diversification of HLA-A,B,C alleles. Proc. Natl. Acad. Sci. USA 142:3937–50 66. Burmeister WP, Gastinel LN, Simister NE, Blum ML, Bjorkman PJ. 1994. ˚ resolution of Crystal structure at 2.2 A the MHC-related neonatal Fc receptor. Nature 372:336–43 67. 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 68. 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 69. Mosser DD, Duchaine J, Martin LH. 1991. Biochemical and developmental characterization of the murine cluster of differentiation 1 antigen. Immunology 73:298–303 70. Sugita M, Porcelli SA, Brenner MB. 1997. Assembly and retention of CD1b heavy chains in the endoplasmic reticulum. J. Immunol. 159:2358–65 71. Bauer A, Huttinger R, Staffler G, Hansmann C, Schmidt W, Majdic O, Knapp W, Stockinger H. 1997. Analysis of the requirement for beta 2-microglobulin for expression and formation of human CD1 antigens. Eur. J. Immunol. 27:1366–73 72. Balk SP, Burke S, Polischuk JE, Frantz ME, Yang L, Porcelli S, Colgan SP, Blumberg RS. 1994. β2-Microglobulinindependent MHC class Ib molecule expressed by human intestinal epithelium. Science 265:259–62 73. Brossay L, Jullien D, Cardell S, Sydora BC, Burdin N, Modlin RL, Kronenberg M. 1997. Mouse CD1 is mainly expressed on hemopoietic-derived cells. J. Immunol. 159:1216–24 74. Cardell S, Tangri S, Chan S, Kronenberg M, Benoist C, Mathis D. 1995. CD1restricted CD4+ T cells in major histo-

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

325

compatibility complex class II-deficient mice. J. Exp. Med. 182:993–1004 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 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 Cattoretti G, Berti E, Parravicini C, Buscaglia M, Cappio F, Caputo R, Cerri A, Crosti L, Delia D, Gaiera G, Polli N. 1989. Expression of CD1 molecules on dendritic cells: ontogeny, epitope analysis on normal and malignant cells, and tissue distribution. See Ref. 150, pp. 262–64 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 Kasinrerk W, Baumruker T, Majdic O, Knapp W, Stockinger H. 1993. CD1 molecule expression on human monocytes induced by granulocyte-macrophage colony-stimulating factor. J. Immunol. 150:579–84 Small TN, Knowles RW, Keever C, Kernan NA, Collins N, O’Reilly RJ, Dupont B, Flomenberg N. 1987. M241 (CD1) expression on B lymphocytes. J. Immunol. 138:2864–68 Small TN, Keever C, Collins N, Dupont B, O’Reilly RJ, Flomenberg N. 1989. Characterization of B cells in severe combined immunodeficiency disease. Hum. Immunol. 25:181–93 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 Cattoretti G, Berti E, Mancuso A, D’Amato L, Schiro’ R, Soligo D, Delia D. 1987. CD1: a MHC class I related family of antigens with widespread distribution on resting and activated cells. In Leukocyte Typing III, ed. A McMichael, PC Beverley, SP Cobbold, MJ Crumpton, W Gilks, FM Golch, N Hogg, M Horton, N Ling, ICM MacLennon, DY Mason, C Milstein, pp. 89–91. Oxford, UK: Oxford Univ. Press Canchis PW, Bhan AK, Landau SB, Yang L, Balk SP, Blumberg RS. 1993. Tissue distribution of the nonpolymorphic major histocompatibility

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

326

85.

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

86.

87.

88.

89.

90.

91.

92.

93.

94.

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

PORCELLI & MODLIN complex class I-like molecule, CD1d. Immunology 80:561–65 Bleicher PA, Balk SP, Hagen SJ, Blumberg RS, Flotte TJ, Terhorst C. 1990. Expression of murine CD1 on gastrointestinal epithelium. Science 250:679– 82 Burke S, Landau S, Green R, Tseng CC, Nattakom T, Canchis W, Yang L, Kaiserlian D, Gespach C, Balk S, Blumberg RS. 1994. Rat cluster of differentiation 1 molecule: expression on the surface of intestinal epithelial cells and hepatocytes. Gastroenterology 106:1143–49 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 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 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 Balk SP, Ebert EC, Blumenthal RL, McDermott FV, Wucherpfennig KW, Landau SB, Blumberg RS. 1991. Oligoclonal expansion and CD1 recognition by human intestinal intraepithelial lymphocytes. Science 253:1411–15 Sieling PA, Chatterjee D, Porcelli SA, Prigozy TI, Mazzaccaro RJ, Soriano T, Bloom BR, Brenner MB, Kronenberg M, Brennan PJ. 1995. CD1-restricted T cell recognition of microbial lipoglycan antigens. Science 269:227–30 Thomssen H, Ivanyi J, Espitia C, Arya A, Londei M. 1995. Human CD4−CD8− αβ+ T-cell receptor T cells recognize different mycobacteria strains in the context of CD1b. Immunology 85:33–40 Gong J, Stenger S, Zack JA, Jones BE, Bristol GC, Modlin RL, Morrissey PJ, Barnes PF. 1998. Isolation of mycobacterium-reactive CD1-restricted T cells from patients with human immunodeficiency virus infection. J. Clin. Invest. 101:383–89 Beckman EM, Melian A, Behar SM, Sieling PA, Chatterjee D, Furlong ST, Matsumoto R, Rosat JP, Modlin RL, Porcelli SA. 1996. CD1c restricts responses of mycobacteria-specific T cells. Evidence for antigen presentation by a

95.

96.

97. 98.

99.

100.

101.

102.

103.

104.

105.

106.

second member of the human CD1 family. J. Immunol. 157:2795–803 Stenger S, Mazzaccaro RJ, Uyemura K, Cho S, Barnes PF, Rosat JP, Sette A, Brenner MB, Porcelli SA, Bloom BR, Modlin RL. 1997. Differential effects of cytolytic T cell subsets on intracellular infection. Science 276:1684–87 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 Bendelac A. 1998. Mouse NK1+ T cells. Curr. Opin. Immunol. 7:367–74 Lantz O, Bendelac A. 1994. An invariant T cell receptor α 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 Yoshimoto T, Paul WE. 1994. CD4pos NK1. 1pos T cells promptly produced IL4 in response to in vivo challenge with anti-CD3. J. Exp. Med. 179:1285–95 Hayakawa K, Lin BT, Hardy RR. 1992. Murine thymic CD4+ T cell subsets: a subset (Thy0) that secretes diverse cytokines and overexpresses the Vβ8 T cell receptor gene family. J. Exp. Med. 176:269–74 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 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 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 Cui J, Shin T, Kawano T, Sato H, Kondo E, Toura I, Kaneko Y, Koseki H, Kanno M, Taniguchi M. 1997. Requirement for Vα14 NKT cells in IL-12-mediated rejection of tumors. Science 278:1623–26 Mieza MA, Itoh T, Cui JQ, Makino Y, Kawano T, Tsuchida K, Koike T,

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

ANTIGEN PRESENTATION BY CD1

107.

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

108.

109.

110.

111.

112.

113.

114.

115.

116.

Shirai T, Yagita H, Matsuzawa A, Koseki H, Taniguchi M. 1996. Selective reduction of V alpha 14+ NK T cells associated with disease development in autoimmune-prone mice. J. Immunol. 156:4035–40 Exley M, Garcia J, Balk SP, Porcelli S. 1997. Requirements for CD1d recognition by human invariant Vα24+ CD4−CD8− T cells. J. Exp. Med. 186: 109–20 Davodeau F, Peyrat M-A, Necker A, Dominici R, Blanchard F, Leget C, Gaschet J, Costa P, Jacques Y, Godard A, Vi´e H, Poggi A, Romagn´e F, Bonneville M. 1997. Close phenotypic and functional similarities between human and murine αβ T cells expressing invariant TCR αchains. J. Immunol. 158:5603–11 Prussin C, Foster B. 1997. TCR Vα24 and Vβ11 coexpression defines a human NK1 T cell analog containing a unique Th0 subpopulation. J. Immunol. 159:5862–70 Sumida T, Sakamoto A, Murata H, Makino Y, Takahashi H, Yoshida S, Nishioka K, Iwamoto I, Taniguchi M. 1995. Selective reduction of T cells bearing invariant V alpha 24J alpha Q antigen receptor in patients with systemic sclerosis. J. Exp. Med. 182:1163–68 Wilson SB, Kent SC, Patton KT, Orban T, Jackson RA, Exley M, Porcelli S, Schatz DA, Atkinson MA, Balk SP, Strominger JL, Hafler DA. 1998. Extreme Th1 bias of invariant Vα24JαQ T cells in type 1 diabetes. Nature 391:177–81 Beckman EM, Porcelli SA, Morita CT, Behar SM, Furlong ST, Brenner MB. 1994. Recognition of a lipid antigen by CD1-restricted αβ+ T cells. Nature 372:691–94 Moody DB, Reinhold BB, Guy MR, Beckman EM, Frederique DE, Furlong ST, Ye S, Reinhold VN, Sieling PA, Modlin RL, Besra GS, Porcelli SA. 1997. Structural requirements for glycolipid antigen recognition by CD1brestricted T cells. Science 278:283–86 Prinzis S, Chatterjee D, Brennan PJ. 1993. Structure and antigenicity of lipoarabinomannan from Mycobacterium bovis BCG. J. Gen. Microbiol. 139:2649–58 Casta˜no AR, Tangri S, Miller JE, Holcombe HR, Jackson MR, Huse WD, Kronenberg M, Peterson PA. 1995. Peptide binding and presentation by mouse CD1. Science 269:223–26 Lee DJ, Abeyratne A, Carson DA, Corr M. 1998. Induction of an antigen-

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

327

specific, CD1-restricted cytotoxic T lymphocyte response in vivo. J Exp. Med. 187:433–38 Kawano T, Cui J, Koezuka Y, Toura I, Kaneko Y, Motoki K, Ueno H, Nakagawa R, Sato H, Kondo E, Koseki H, Taniguchi M. 1997. CD1d-restricted and TCR-mediated activation of Valpha14 NKT cells by glycosylceramides. Science 278:1626–29 Joyce S, Woods AS, Yewdell JW, Bennink JR, De Silva AD, Boesteanu A, Balk SP, Cotter RJ, Brutkiewicz RR. 1998. Natural ligand of mouse CD1d1: cellular glycosylphosphatidylinositol. Science 279:1541–44 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 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 Hanau D, Fricker D, Bieber T, EspositoFarese ME, Bausinger H, Cazenave JP, Donato L, Tongio MM, de la Salle H. 1994. CD1 expression is not affected by human peptide transporter deficiency. Hum. Immunol. 41:61–68 Prigozy TI, Sieling PA, Clemens D, Stewart PL, Behar SM, Porcelli SA, Brenner MB, Modlin RL, Kronenberg M. 1997. The mannose receptor delivers lipoglycan antigens to endosomes for presentation to T cells by CD1b molecules. Immunity 6:187–97 Sugita M, Jackman RM, van Donselaar E, Behar SM, Rogers RA, Peters PJ, Brenner MB, Porcelli SA. 1996. Cytoplasmic tail-dependent localization of CD1b antigen-presenting molecules to MIICs. Science 273:349–52 Marks M, Ohno H, Kirchhausen T, Bonifacino J. 1997. Protein sorting by tyrosine-based signals: adapting to the Ys and wherefores. Trends Cell Biol. 7:124–28 Jackman RM, Stenger S, Lee A, Moody DB, Rogers RA, Niazi KR, Sugita M, Modlin RL, Peters PJ, Porcelli SA. 1998. The tyrosine-containing cytoplasmic tail of CD1b is essential for its efficient presentation of bacterial lipid antigens. Immunity 8:341–51 Ernst WA, Maher J, Cho S, Niazi KR, Chatterjee D, Moody DB, Besra GS, Watanabe Y, Jensen PE, Porcelli SA, Kronenberg M, Modlin RL. 1998.

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

328

127.

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

128.

129.

130.

131.

132.

133.

134.

135.

136.

137.

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

PORCELLI & MODLIN Molecular interaction of CD1b with lipoglycan antigens. Immunity 8:331–40 Moody DB, Reinhold BB, Reinhold VN, Besra GS, Porcelli SA. 1998. Uptake and processing of glycosylated mycolates for presentation to CD1b-restricted T cells. Immunol. Lett. In press Sturgill-Koszycki S, Schlesinger PH, Chakraborty P, Haddix PL, Collins HL, Fok AK, Allen RD, Gluck SL, Heuser J, Russell DG. 1994. Lack of acidification in Mycobacterium phagosomes produced by exclusion of the vesicular proton-ATPase. Science 263:678–81 Xu S, Cooper A, Sturgill-Koszycki S, van Heyningen T, Chatterjee D, Orme I, Allen P, Russell DG. 1994. Intracellular trafficking in Mycobacterium tuberculosis and Mycobacterium avium-infected macrophages. J. Immunol. 153:2568– 78 Porcelli SA, Segelke BW, Sugita M, Wilson IA, Brenner MB. 1998. The CD1 family of lipid antigen presenting molecules. Immunol. Today 19:362–68 Shin DH, Lee JY, Hwang KY, Kim KK, Suh SW. 1995. High-resolution crystal structure of the non-specific lipidtransfer protein from maize seedlings. Structure 3:189–99 Sette A, Sidney J, del Guercio MF, Southwood S, Ruppert J, Dahlberg C, Grey HM, Kubo RT. 1994. Peptide binding to the most frequent HLA-A class I alleles measured by quantitative molecular binding assays. Mol. Immunol. 31:813–22 Guercio M, Sidney J, Hermanson G, Perez C, Grey H, Kubo R, Sette A. 1995. Binding of a peptide antigen to multiple HLA alleles allows definition of an A2like supertype. J. Immunol. 154:685–93 Ridley DS, Jopling WH. 1966. Classification of leprosy according to immunity. A five-group system. Int. J. Lepr. 34:255–73 Yamamura M, Uyemura K, Deans RJ, Weinberg K, Rea TH, Bloom BR, Modlin RL. 1991. Defining protective responses to pathogens: cytokine profiles in leprosy lesions. Science 254:277– 79 Yamamura M, Wang X-H, Ohmen JD, Uyemura K, Rea TH, Bloom BR, Modlin RL. 1992. Cytokine patterns of immunologically mediated tissue damage. J. Immunol. 149:1470–75 Sieling PA, Abrams JS, Yamamura M, Salgame P, Bloom BR, Rea TH, Modlin RL. 1993. Immunosuppressive roles for interleukin-10 and interleukin-4 in

138.

139.

140. 141. 142.

143.

144.

145. 146.

147. 148.

149. 150.

human infection: in vitro modulation of T cell responses in leprosy. J. Immunol. 150:5501–10 Thomssen H, Kahan M, Londei M. 1995. Differential effects of interleukin10 on the expression of HLA class II and CD1 molecules induced by granulocyte/macrophage colony- stimulating factor/interleukin-4. Eur. J. Immunol. 25:2465–70 Kaplan G, Walsh G, Guido LS, Meyn P, Burkhardt RA, Abalos RM, Barker J, Frindt PA, Fajardo TT, Celona R, Cohn ZA. 1992. Novel responses of human skin to intradermal recombinant granulocyte/macrophage-colonystimulating factor: Langerhans cell recruitment, keratinocyte growth, and enhanced wound healing. J. Exp. Med. 175:1717–28 Kaufmann SHE. 1988. CD8+ T lymphocytes in intracellular microbial infections. Immunol. Today 9:168–74 Kaufmann SHE. 1993. Immunity to intracellular bacteria. Annu. Rev. Immunol. 11:129–63 Albert ML, Sauter B, Bhardwaj N. 1998. Dendritic cells acquire antigen from apoptotic cells and induce class Irestricted CTLs. Nature 392:86–89 Pena SV, Hanson DA, Carr BA, Goralski TJ, Krensky AM. 1997. Processing, subcellular localization, and function of 519 (granulysin), a human late T cell activation molecule with homology to small, lytic, granule proteins. J. Immunol. 158:2680–88 Pena SV, Krensky AM. 1997. Granulysin, a new human cytolytic granuleassociated protein with possible involvement in cell-mediated cytotoxicity. Semin. Immunol. 9:117–25 Ratledge C, Wilkinson SG. 1988. Microbial Lipids, Vol. 1. San Diego, CA: Academic. 963 pp. Fairhurst RM, Wang CX, Sieling PA, Modlin RL, Braun J. 1998. CD1restricted T cells and resistance to polysaccharide-encapsulated bacteria. Immunol Today 19:257–59 Turco SJ, Descoteaux A. 1992. The lipophosphoglycan of Leishmania parasites. Annu. Rev. Microbiol. 46:65–94 Hakomori S. 1985. Aberrant glycosylation in cancer cell membranes as focused on glycolipids: overview and perspectives. Cancer Res. 45:2405–14 Calabi F, Yu CY, Bilsland CAG, Milstein C. 1991. CD1: From structure to function. See Ref. 151, pp. 215–43 Knapp W, Dorkin B, Guilks WR, Reiber

P1: PKS/ARY

P2: PKS/NBL/mbg

February 11, 1999

12:3

QC: NBL/apm

T1: PSA

Annual Reviews

AR078-11

ANTIGEN PRESENTATION BY CD1

151.

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

152.

153.

154.

154a.

EP, Schmitt RE, Stein H, von dem Borneed AEG, eds. 1989. Leucocyte Typing IV. Oxford: Oxford Univ. Press Srivastava R, Ram BP, Tyle P, eds. 1991. Immunogenetics of the Major Histocompatibility Complex. New York: VCH Publ. Teitell M, Holcombe HR, Brossay L, Hagenbaugh A, Jackson MJ, Pond L, Balk SP, Terhorst C, Peterson PA, Kronenberg M. 1997. Nonclassical behavior of the mouse CD1 class I-like molecule. J. Immunol. 158:2143–49 Bendelac A, Killeen N, Littman DR, Schwartz RH. 1994. A subset of CD4+ thymocytes selected by MHC class I molecules. Science 263:1774–78 Hammond KL, 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 Burdin N, Brossay L, Koezuka Y, Smiley ST, Grusby MJ, Gui M, Taniguchi M, Hayakawa K, Kronenberg M. 1998. Selective ability of mouse CD1 to present glycolipids: alpha-galactosylceramide specifically stimulates Valpha 14+ NK T lymphocytes. J. Immunol. 161:3271– 81

329

155. Spada F, Koezuka Y, Porcelli SA. 1998. CD1d-restricted recognition of synthetic glycolipid antigens by human NK T cells. J. Exp. Med. 188:1529–34 156. Brossay L, Chioda MC, Burdin N, Koezuka Y, Casorati G, Dellabona P, Kronenberg M. 1998. CD1d mediated recognition of a alpha-galactosylceramide by Nk T cells is highly conserved through mammalian evolution. J. Exp. Med. 188: 1521–28 157. Stenger S, Hanson DA, Teitlebaum R, Dewan P, Niazi KR, Froelich CJ, Ganz T, Thoma-Uzynski S, Melian A, Bogdan C, Porcelli SA, Bloom BR, Krensky AM, Modlin RL. 1988. An antimicrobial activity of cytolytic T cells mediated by granulysin. Science. 282:121–25 158. Brossay L, Naidenko O, Burdin N, Sakai T, Kronenberg M. Structural requirements for galactosylceramide recognition by CD1 restricted NK T cells. J. Immunol. In press 159. Sieling PA, Jullien D, Dahlem M, Tedder TF, Rea TH, Modlin RL, Porcelli SA. 1999. CD1 expression by dendritic cells in human leprosy lesions: correlation with effective host immunity. J. Immunol. In press 160. Stenger S, Niazi KR, Modlin RL. 1998. Down-regulation of CD1 on antigenpresenting cells by infection with Mycobacterium tuberculosis. J. Immunol. 161:3582–88

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:297-329. Downloaded from arjournals.annualreviews.org by HINARI on 08/31/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:331–67 c 1999 by Annual Reviews. All rights reserved Copyright °

TUMOR NECROSIS FACTOR RECEPTOR AND Fas SIGNALING MECHANISMS D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, and M. P. Boldin Department of Biological Chemistry, Weizmann Institute, Rehovot, 76100, Israel; e-mail: [email protected] KEY WORDS:

apoptosis, caspase, MAP kinase, NF-κB, signaling

ABSTRACT Four members of the tumor necrosis factor (TNF) ligand family, TNF-α, LT-α, LT-β, and LIGHT, interact with four receptors of the TNF/nerve growth factor family, the p55 TNF receptor (CD120a), the p75 TNF receptor (CD120b), the lymphotoxin beta receptor (LTβR), and herpes virus entry mediator (HVEM) to control a wide range of innate and adaptive immune response functions. Of these, the most thoroughly studied are cell death induction and regulation of the inflammatory process. Fas/Apo1 (CD95), a receptor of the TNF receptor family activated by a distinct ligand, induces death in cells through mechanisms shared with CD120a. The last four years have seen a proliferation in knowledge of the proteins participating in the signaling by the TNF system and CD95. The downstream signaling molecules identified so far—caspases, phospholipases, the three known mitogen activated protein (MAP) kinase pathways, and the NF-κB activation cascade—mediate the effects of other inducers as well. However, the molecules that initiate these signaling events, including the death domain- and TNF receptor associated factor (TRAF) domain-containing adapter proteins and the signaling enzymes associated with them, are largely unique to the TNF/nerve growth factor receptor family.

INTRODUCTION The study of cell response to ligands of the tumor necrosis factor (TNF) family is one of the most dynamic research areas in the signaling field today. Vast 331 0732-0582/99/0410-0331$08.00

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

332

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

amounts of phenomenological information have accumulated. Most of the available knowledge concerns the cell-killing activity of some of these ligands, a subject that has gained increased attention with the recent surge of interest in cell death mechanisms. The isolation and cloning of TNF-α and lymphotoxin (LT)-α in 1985 (196–199), and later of their receptors, the p55 and p75 TNF receptors (CD120a and CD120b) (200–204), confirmed that cell killing by these two ligands—a phenomenon then already known for almost 20 years (205–208)—is a receptor-induced effect. However, it was only after the discovery of Fas/Apo1 (CD95), a death-inducing receptor functionally related to CD120a (208–211), and its ligand (Fas-L) (212) that confidence in the biological significance of this death-inducing function prompted intense studies of the signaling mechanisms involved. The cloning of TNF-α and LT-α also confirmed indications that these cytokines have many other activities in addition to cell killing, mainly related to inflammation. The known range of activities has expanded with the recent discovery of two additional ligand molecules, LT-β (213) and LIGHT (11), which together with LT interact with two additional receptors, the lymphotoxin β receptor (LTβR) (214) and herpes virus entry mediator (HVEM) (10). The pattern of activities mediated by this group of ligands and receptors is outstanding in its complexity as well as in its diametric consequences, which range from destruction of tissues to orchestration of immune organogenesis. The nature of the mechanisms that control this complex response has gained wide interest even beyond the realm of basic science. Particular attention to TNF and its function has come from the fields of biomedicine and biotechnology because increasing evidence implicates dysregulation of the function of this cytokine in the pathology of many diseases. Molecular understanding in this field has lagged significantly behind the phenomenological knowledge. Today, however, the use of affinity purification and two-hybrid screening techniques, and the availability of data banks to identify proteins on the basis of sequence homology, allow rapid progress in elucidation of the signaling mechanisms of the TNF/Fas systems. Although still fragmentary, this knowledge has already yielded important lessons. Alongside principles of signaling shared with other pathways, features are emerging that are unique to the signaling for death of cells and to the functioning of the TNF receptor family in general. Space restrictions necessitate limitation of this review to aspects of the signaling mechanisms that currently seem most worthy of highlighting. The references cited are mostly the more recent and lesser-known studies, and readers are referred to other recent reviews for more detailed information and references. We apologize to the authors of many important relevant studies that could not be cited here.

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

TNF RECEPTOR AND Fas SIGNALING

333

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

RECEPTORS AND LIGANDS OF THE TNF AND Fas/APO1 SYSTEMS1 The ligands and receptors whose signaling is the subject of this review belong to the large TNF-related ligand and TNF/nerve growth factor (NGF) receptor families. In addition to the receptor-ligand interaction motifs that define these families—a β-sheet receptor-binding structure arranged in β-jellyroll topology and a cysteine-rich repeat ligand-binding module (reviewed in 8, 215)—they share several other common structural and functional features. With the exception of LT-α, which is secreted by cells, all members of the ligand family are formed as type II transmembrane proteins and can therefore act in a juxtacrine manner. Some of them are subject to proteolytic processing, allowing them to act in a soluble form, either as ligands (as in the case of TNF-α) or as inhibitors of signaling (as in the case of Fas-L). It seems that all members of the TNF family act in the form of trimers, and most of them act as homotrimers. The only known exception is LT-β, which functions after forming heterotrimers with LT-α, apparently because of its inability to assemble properly on its own (9). The molecular structures recognized by the receptors reside in the groove between neighboring ligand monomers. Each of the receptors of the TNF system can bind to either one of two different structures. CD120a and CD120b bind both to TNF-α and to LT-α. The LTβR binds to LT-β and to LIGHT, and HVEM [initially identified by virtue of its binding to a Herpes simplex protein (10)], binds to LIGHT and (with low affinity) to LT-α (11). CD95, however, is known to bind only to Fas-L (Figure 1).

PHYSIOLOGICAL FUNCTIONS AND CELLULAR EFFECTS OF THE TNF AND Fas SYSTEMS The TNF and Fas (TNF/Fas) systems regulate immune defense. Their effects are characterized by a remarkable duality—induction of damage on the one hand accompanied by induction of repair and expansion on the other. Cell death is induced alongside cell growth and resistance to death; hematopoiesis is suppressed simultaneously with induction of hematopoietic growth factors; inflammation is promoted and then suppressed. Together, these contrasting effects result in an intense yet brief response, allowing adjustment to insult. Our knowledge of the relative contributions of the different components of these systems to their overall functioning is incomplete. Most of the current knowledge has to do with the functions of CD120a and CD95; least is known about the function of the recently identified HVEM and LIGHT. 1 For

other recent reviews, see (1 –7).

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

334

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

Figure 1 Ligands and receptors of the TNF and Fas systems, their docking proteins, and their known structural modules.

In vivo assessment of the consequences of obliteration of individual receptor or ligand functions indicates that these proteins serve different yet overlapping physiological roles. The most salient in vivo consequences of obliteration of CD120a function are deficient defense against certain intracellular pathogens and restrained inflammatory response (12, 13). Restriction of the inflammatory response, although in a different manner, is also the most obvious change accompanying obliteration of CD120b function (14). Inhibition of LTβR function

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TNF RECEPTOR AND Fas SIGNALING

335

is most clearly reflected in deficient lymphoid organ development (15 and references therein). Inhibition of CD95 function is manifested mainly in excessive expansion of lymphoid organs, pointing to the critical role of CD95 in regulation of lymphocyte survival (16). However, more thorough examination revealed that the TNF receptors could also contribute to lymphoid organ development (15 and references therein) and to lymphocyte death (17), the LTβR to inflammatory disorders (JL Browning, submitted to Gastroenterology) and CD95 to inflammation (18). The individual cellular responses to CD120a appear to concern various aspects of innate immunity and also—although to a lesser extent—of adaptive immunity. This receptor controls defense on the level of the individual cell, for example, by inducing death of pathogen-afflicted cells. On the level of multicellular organs, CD120a controls defense by coordinating the inflammatory process, and on the level of the whole organism, by inducing changes such as fever, loss of appetite, or elevation of acute-phase serum proteins. CD120b, which is activated preferentially by the cell-bound form of TNF (19), is so far known to induce only a few effects. In most cases, the observed effect can also be induced by triggering CD120a. The known in vitro effects of the LTβR are growth stimulation of fibroblasts and a cytocidal effect on some tumor cell lines (20, 21), whereas HVEM induces enhancement of growth (and probably also of some other functions) of T lymphocytes (22). CD95 has been found to induce death of cells but may also stimulate cell growth and induce synthesis of the cytokines interleukin (IL)-6 and IL-8 (e.g. 23).

THE SIGNALING MECHANISMS General Considerations: Reliability of Interpretation The flood of new molecular data concerning the functions of the TNF/Fas systems raises a need to define measures of quality control to evaluate their significance. Two issues are of particular concern in this connection: primary versus secondary targets of signaling and the exact role of the signaling molecules themselves. TARGETS FOR SIGNALING: WHICH ARE DIRECT AND WHICH ARE SECONDARY?

Sorting out the primary targets of signaling among the profusion of molecular events has proved particularly difficult in the case of the TNF/Fas systems. The degree of uncertainty can be appreciated from the fact that, as opposed to the common practice in the signaling field, in most cases targets have been identified by elucidation of their corresponding signaling mechanisms, rather than the other way around. Several features unique to the TNF/Fas systems have contributed to this high degree of uncertainty.

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

336

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

1. Cell death induction. The cell death process is notorious for its multiplicity of molecular events and the difficulties involved in distinguishing the primary ones. Over the years of study of death induction by the TNF/Fas systems, several different mechanisms have been proposed as the initiators of these processes. It was only through sequential analysis of the death-inducing signaling cascades, however, that a direct target for their action, activation of specific caspases, was discovered. 2. Lipid-derived mediators. The identity of the initial targets in the induction of lipid-derived mediators by the TNF/Fas systems is particularly uncertain, for several reasons: (a) In cells where these systems trigger cytocidal effects, lipases are activated at a late stage of the response as part of the death process itself; (b) the inflammatory function of the TNF system also involves induction of lipid-derived mediators as part of the cellular response, rather than as the signaling mechanism itself; (c) lack of familiarity with the techniques involved has prevented all but a few laboratories from contributing materially to the knowledge of the formation of lipid mediators. The lipase activation pathways described below are therefore supported only by a limited amount of data, and only parts of them are clear enough to allow full identification of the lipase involved. 3. Transcription factors. As with the mechanism of death induction by the TNF/Fas systems, the most solid information about the targets of the signaling mechanisms that control gene expression has come from the study of the mechanisms themselves. Analysis of these signaling cascades has confirmed that NF-κB and AP1, two groups of transcription factors with central roles in inflammation and immune response regulation, are direct targets of these signaling systems (see below). Several other transcription factors, including IRF1, NF-IL6, and others, are known to be affected. However, owing to the complexity of the gene-activation effects of the TNF system and the multiplicity of interactions among different transcription factors, there is still only partial knowledge of the identity of the factors affected directly by these systems. No information is available on the molecular targets in posttranscriptional modulation of gene expression by these systems, such as that thought to be mediated by the JNK and p38 MAPK cascades (see below). SIGNALING MOLECULES: THE TEST SYSTEMS APPLIED AND THEIR PHYSIOLOGICAL RELEVANCE In many cases (several of which are cited in this review),

initial notions about particular events activated by the TNF/Fas systems have later turned out to be erroneous. As in any other experimental endeavor, all techniques applied in this field involve varying degrees of distortion of the natural situation. Knowledge of the interactions of signaling proteins gained

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

TNF RECEPTOR AND Fas SIGNALING

337

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

by their artificial expression either in mammalian cells or in transfected yeast (two-hybrid tests) has proved less reliable than that gained by monitoring the interaction of the proteins expressed endogenously in cells. Similarly, knowledge of protein function gained by assessing the effects of high-level expression of the proteins or their mutants in transfected cells has often proved to be less reliable that the knowledge gained by elimination of the endogenous expression of the proteins in cells through mutation or targeted disruption of their genes. Much of our current knowledge of the molecular interactions in the TNF/Fas systems is still based on the less reliable method of enforced expression testing.

INITIATION OF SIGNALING Triggering As in many other receptor-induced processes, the signaling activity of the TNF family of receptors is triggered upon juxtaposition of the intracellular domains of the receptor molecules following binding of the ligand molecules to their extracellular domains. The trimeric structure of the ligands probably contributes to this juxtaposition, although formation of dimeric receptor molecules also seems to suffice for triggering. Signaling by the receptors can be triggered merely by imposing aggregation of the receptor molecules, for example, with antireceptor antibodies (e.g. 216, 217). It seems likely, however, that this process involves not only translocation of the receptor molecules on the cell surface but also an induced conformational change that may account for the reported ability of certain fragments of TNF-α to initiate signaling (e.g. 24, 25). That the receptors play an active role in their induced triggering is indicated by the finding that both the death domain (DD) motif in the intracellular domain of CD120a (see below) and the receptor’s extracellular domain undergo self-association. It was suggested that the self-association of the DD might contribute to initiation and amplification of signaling, whereas dimerization of the extracellular domain may prevent spontaneous initiation of signaling by keeping their intercellular domains apart (26, 27). As with other receptors that have intrinsic protein kinase activity, in the initiation of signaling by the receptors of the TNF/Fas systems the activation step seems to occur by cross-modification of enzymes found in the signaling complexes (see the description below of the way in which caspase-8 is activated). In this case, however, the enzymatic activity is mediated not by the receptor itself but by other, recruited proteins.

Role of Docking Proteins The recruitment of signaling molecules to the receptors of the TNF/Fas systems occurs through an intermediate phase of adapter proteins, most of which

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

338

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

have no enzymatic activity of their own. These adapter (docking) proteins, like the ligands and receptors, are modular in structure. Several of the docking proteins can bind to each other through specific modules distinct from those involved in their binding to signaling molecules. The result is the formation of a network of proteins that dictate signaling for different effects through binding to different enzymes. This network is composed of two main parts, each involving a distinct major protein-binding motif that prompts homophilic protein interactions. One part involves several docking proteins, including MORT1/FADD, TRADD, RIP, and RAIDD/CRADD, which bind to each other as well as to CD120a and CD95 through a DD motif found both in the docking proteins and in the receptors. The other part is centered around a group of adapter molecules that share a protein-binding motif called the TRAF domain (Figure 2). The two parts of the network are linked through association of the TRAF domain in the adapter protein TRAF2 with the regions upstream of the DD in the adapter proteins TRADD and RIP (28–30). The functions of these two parts of the docking protein network are not entirely distinct. As described below, the DD-containing adapter proteins are involved mainly in death induction, yet the DD-containing protein MORT1/FADD seems to be involved in induction of lymphocyte growth as well (31). Conversely, the TRAF domain complex, which is involved mainly in gene regulation, also seems to affect the induction of death (32–34).

STRUCTURES INVOLVED Motifs in the Extracellular Domains of the Receptors and the Ligands Apart from the receptor-ligand interacting motifs shared by the ligands and receptors of the TNF-related families, the extracellular domains of TNF-α and its receptors (and probably of other ligands and receptors of these families) contain membrane-proximal regions whose structural features, which have yet to be identified, render these molecules susceptible to shedding. The enzyme responsible for the shedding of TNF-α (and perhaps also of its receptors), TACE, is a transmembrane protease of the adamylisin family. The crystal structure of its catalytic domain was recently defined (35).

Motifs Identified in the Intracellular Domains of the Receptors and the Proteins That Bind to Them 1. The death domains. A shared sequence motif in the intracellular domains of CD120a and CD95 was dubbed the death domain in view of its critical

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

339

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TNF RECEPTOR AND Fas SIGNALING

Figure 2 Regulation of the direct caspase activation and the NF-κB activating cascades by the TNF receptors and CD95.

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

340

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

role in the cytocidal effect of these receptors. Its presence was later detected in many other proteins, only some of which participate in death induction [reviewed in (36, 37)]. Structural modeling of the DDs in various proteins, as well as NMR studies of the DD in CD95 (38, 39) and the low-affinity NGF (40), indicated that this motif is composed of six amphipatic α-helical regions arranged antiparallel to one another. The DD serves as a proteindocking site and apparently also as a transducer of conformational changes. It participates mainly in homotypic interactions. Four adapter proteins that take part in the signaling by CD120a and CD95—MORT1/FADD (41, 42), TRADD (43), RIP (44), and RAIDD (45, 46)—contain DDs. The proteins bind to the receptors, to each other, or both and, with the exception of MORT1/FADD, also self-associate through homotypic DD interactions. An exception is the adapter protein DAXX, which binds to the DD of CD95, but is itself devoid of a DD (47). A protein called MADD/Rab3-GAP, containing a region with a low degree of homology to the DD, was shown in one study to bind through this region to the DD of CD120a (48). Enforced expression of this protein or of its putative DD homology region was found to affect the activation of JNK, ERK, and cPLA2 by TNF-α. The same protein also serves as a GDP/GTP exchange protein for certain members of the Rab family of small G proteins that associate with synaptic vesicles and regulate neurotransmitter release. A short splice variant of the protein, called DENN, is phosphorylated by the brain-specific JNK isoform (JNK3) and translocated with it in neurons to the nucleoli in response to hypoxia (49, 50). 2. Regions upstream of the DDs. The membrane-proximal region in CD120a, upstream of the DD, participates in signaling both independently and in cooperation with the DD (51–53). Three proteins are known to bind to it. A protein called FAN, which is required for neutral sphingomyelinase activation by this receptor, binds through a WD-repeat region to a stretch of nine residues (residues 309–319) upstream of the DD (52). A regulatory component of the 26 proteasomes, called 55.11, p97 or TRAP2, binds to a region just upstream of the FAN-binding site (residues 234–308), possibly allowing direct regulation of proteasomal function by the receptor (54, 55). A protein of unknown function, called TRAP1, closely related to hsp90, binds to recognition sites diffusely spread upstream of the DD (56). 3. Regions downstream of the DDs. As in some other DD-containing proteins (36), the regions downstream of the DDs in CD120a and CD95 have a high content of serine, threonine, and proline residues, whose functional significance is still unknown. In the human CD95, though apparently not

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

TNF RECEPTOR AND Fas SIGNALING

341

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

in the mouse receptor, the three most C-terminal residues (SVL) serve as a binding motif for a protein tyrosine phosphatase, FAP1, which can downregulate death induction by the receptor (57, 58). 4. The TRAFs and their binding regions in the receptors. The TRAFs are adapter proteins that share a sequence homology C-terminal motif (the TRAF domain), N-terminal ring finger and zinc finger motifs, and a central coiledcoil region. They participate in the signaling activity of several receptors of the TNF/NGF family, including the three receptors of the TNF system that do not contain DDs; they bind to these latter receptors through their TRAF domains. CD120b binds to TRAF 2 and (indirectly, through binding to TRAF2) also to TRAF1 (59, 60). LTβR binds to TRAF2, 3 and 5 (61), and HVEM binds to TRAF1, 2, 3, and 5 (62, 63). With other receptors, the TRAFs could be shown to bind to distinct intracellular domain sequence motifs (e.g. 64). Both the TRAF domain region and the N-terminal part of the TRAF protein are required for the signaling function. Yet, while several signaling and regulatory proteins are known to bind to the TRAF homology region (TRAF-C) and to the region immediately upstream of it (TRAF-N) (see 65 for review), no protein that binds to the N-terminal part of the TRAFs has yet been identified. Preliminary evidence suggests that this region can act independently as a transcription regulator, after being transported by an unknown mechanism to the nucleus (66). 5. Conserved phosphorylation sites. A conserved tyrosine in the CD120a DD (Tyr 331 in the human receptor) can be phosphorylated by pp60src, apparently affecting the function of a serine/threonine kinase associated with this domain (67). A conserved tyrosine is present at a similar site in CD95. A cluster of serines at the TRAF2-binding site in CD120b contains a casein kinase I phosphorylation motif, whose phosphorylation was suggested to down-regulate signaling by the receptor (68–70).

Motifs in the Intracellular Domains of the Ligands and Reversed Signaling The intracellular domain of TNF is phosphorylated in cells (71). A substrate site for phosphorylation by casein kinase I is found in this domain, as well as in the intracellular domains of most other members of the TNF ligand family, including Fas-L, and apparently participates in signaling by these domains upon ligand binding to their receptors (reversed signaling; A Watts, submitted to EMBO J.). It was suggested that an additional conserved site in the Fas-L intracellular domain, an SH3 binding motif, allows anchoring of the ligand molecules to the cytoskeleton (72).

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

342

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SPECIFIC SIGNALING PATHWAYS The known enzymes through which the receptors of the TNF/Fas systems initiate signaling include proteases of the caspase family, phospholipases, and protein kinases. Some of the multiple functional changes regulated by these enzymes reflect altered gene activity, resulting from direct activation of transcription factors (e.g. NF-κB), increased synthesis of such factors (e.g. IRF1), or modulation of translation rate or message stability. Other effects, like the induction of cell death, occur independently of gene activation. Although this review is concerned with intracellular signaling, it should be noted that the TNF/Fas systems also control the formation of molecules that transmit signals among cells. The formation of these extracellular mediators and of the intracellular signaling mediators are closely linked. The activation of caspases within cells results in processing of the precursors for extracellular polypeptide mediators such as IL-1β and IL-18, whereas the activation of phospholipases yields compounds that can be metabolized to lipid extracellular mediators such as prostaglandins and PAF. These polypeptide and lipid extracellular mediators, together with mediators like IL-8 whose genes are activated by the kinase cascades stimulated by the TNF/Fas systems, act outside of their producing cells to perpetuate the signals initiated within them, and coordinate the multicellular inflammatory processes that these systems induce.

Direct Caspase Activation Cascades The caspases, a family of evolutionarily conserved cysteine proteases that cleave proteins at specific substrate sites downstream of aspartate residues, play crucial roles in apoptotic processes and in the formation of several proinflammatory mediators (reviewed in 73). These proteases normally exist in cells as inactive precursors, yet upon death induction become activated by processing at internal caspase substrate sites, allowing a cascade-like caspase activation process. The precursors of some of the caspases bind through the region upstream of their protease moiety to regulatory proteins that control their processing. Three caspases have been found to associate through motifs in these prodomains to homologous motifs found in adapter proteins of the CD120a and CD95 signaling complexes. Caspase-8 (MACH/FLICE/Mch5) (74–76) and caspase-10 (FLICE2/Mch4) (76, 77) bind through duplicated N-terminal death effector domains (DEDs) to an N-terminal DED in MORT1/FADD, and caspase-2 binds through an N-terminal motif called CARD to a CARD domain in RAIDD (45, 46). Duplicated N-terminal DED also occurs in a protein with multiple names, including FLIP, Casper, and CASH (reviewed in 78), which displays sequence homology to the caspases, yet lacks several residues critical for

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TNF RECEPTOR AND Fas SIGNALING

343

protease activity. The DED and CARD motifs display some sequence and structural similarities to the DD, raising the possibility that the conformational changes underlying their homotypic associations are similar to those in the DD (79, 80). Apparently, all three caspases mentioned above participate in the induction of death, while CASH serves as a regulator (an inhibitor, or—according to several studies—a stimulator) of the death process. At the moment, however, the only one of these proteins for which there is direct evidence of involvement in death induction beyond that gained in enforced expression studies is caspase-8. There is conclusive evidence for recruitment of this caspase to the Fas signaling complex (75). Moreover, targeted disruption of the caspase-8 gene was found to ablate death induction by TNF or by CD95 ligation (81). The processing of caspase-8 upon ligation of CD95 or CD120a seems to result from juxtaposition of the caspase-8 molecules recruited to the receptors, apparently through the mild proteolytic activity of the unprocessed caspase-8 molecules themselves (82, 83). Knowledge of the events after caspase-8 activation is fragmentary. In vitro, caspase-8 is capable of processing and activating almost all other caspases (84). Within cells, however, it seems to act in a much more restricted manner, resulting in the sequential activation, first of caspase-9 (85), then of caspase-3 and caspase-7, and later of caspase-6 (86). Even the processing of caspase-9 may not be directly mediated by caspase-8 but rather may occur as a consequence of the cleavage of other proteins. Recent findings indicate that BID, a mammalian homolog of the nematode death inhibitory CED9 protein, and plectin, a major cytoskeletal protein, serve as direct substrates of caspase-8 (87, 88, 218). The COOH-terminal fragment of BID formed in its processing by caspase-8 translocates from the cytosol to the mitochondria, causes their clustering, and initiates mitochondrial changes characteristic of apoptotic events, such as permeability transition and leakage of cytochrome c. This last event may well trigger self-processing of caspase-9, catalyzed by Apaf1, a mammalian homolog of the nematode major regulatory protein CED4, which is subject to allosteric activation by cytochrome c (89). 2 In cells in which CD95 ligation causes relatively weak caspase-8 processing, death induction by this receptor indeed involves a crucial amplification role of mitochondrial functions (90, 91). In such cells, Bcl-2 or Bcl-xL, two death-inhibitory proteins related to CED9, can block CD95-mediated death induction by inhibiting death-related events occurring consequently to caspase-8 activation. In most types of cells, 2 BID is apparently not the only cytosolic protein that mediates caspase-8-induced mitochondrial apoptotic changes. BID depleted cytosol preparations can also induce, after their treatment with caspase-8, cytochrome c release from mitochondria and loss of mitochondrial membrane potential [M Steemans et al 1998. Caspase-8 induced mitochondrial permeability transition through a nonprotease intermediate. J. Interferon Cytokine Res. 18:A-79 (Abstr.), and M Steemans, personal communication].

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

344

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

however, caspase-8 signals for death in a way that appears to be independent of mitochondrial involvement. In these cells, Bcl-2 and Bcl-xL, which block the death-related mitochondrial events as well as the caspase-8-induced cleavage of plectin after CD95 ligation (though not the processing of caspase-8 itself ), are incapable of providing effective protection from death induction by CD95 or TNF (91–94). The mechanism of caspase-9 activation in these cells remains to be clarified. There is no evidence for any mechanistic link between initiating events in death induction by the TNF/Fas systems and by other apoptotic triggers such as growth factor withdrawal. However, the TNF/Fas systems seem to use the same downstream mechanisms, subsequent to caspase-9 activation, as those activated by other apoptotic processes.

Death Induction Independent of Direct Caspase Activation The available knowledge of the molecular changes underlying death processes induced by the TNF/Fas systems suggests that these processes depend on the cooperative functioning of several different mechanisms. The relative contribution of these mechanisms may vary from one cell type to another. Among these contributing factors are the mitochondria-associated death processes. As described above, inhibition of these processes can cause arrest of the TNF/Fasinduced death in some cells, but not in others. This is also the case with some enzymatic activities that appear to be critical for inducing the death processes; yet they exert their effects in a cell-specific manner. Such enzymes include chymotrypsin-like proteases (95), cellular phospholipase A2 (96 and references therein), lysosomal enzymes such as cathepsin D and the acid sphingomyelinase, and protein kinases like JNK [e.g. (97, 98) and references therein]. As discussed below, several of these death-related activities are stimulated by CD120a or CD95 through signaling pathways independent of the direct activation of caspase-8. CD120a may have the ability directly to affect even the mitochondria in a way that might contribute to death induction (kinesinmediated translocation to a perinuclear site) via a region distinct from that involved in caspase-8 activation [upstream of the DD and the FAN-binding motif (53)]. Whether the contribution of these activities to death induction can allow initiation of the death process independently of the direct caspase activation process is a matter of debate. Two very recent findings are of relevance here: 1. Targeted disruption of either MORT1/FADD or caspase-8 in mice was found to result in complete unresponsiveness of fibroblasts derived from the mutated mice (and in the case of the MORT1/FADD null mice, also of lymphocytes) to death induction by TNF or by CD95 ligation. Thus, at least in

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

TNF RECEPTOR AND Fas SIGNALING

345

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

fibroblasts, the direct caspase-8 activation pathway plays an indispensable role in apoptosis induction (31, 81, 99). 2. Although induction of apoptosis by the TNF/Fas systems has not been observed in the absence of caspase action, very recent studies clearly show that both CD120a and CD95 have the ability to induce in some cells, though not in others, a caspase-independent necrotic process. Which, if any, of the cell line-specific, death-related mechanisms listed above contribute to this effect is not yet known. There are some indications that the necrotic process involves a critical role of mitochondria-produced oxygen radicals. Notably, in cells exhibiting this effect, caspase blockage actually results in augmentation of death, suggesting that the caspases act to suppress mechanisms of necrosis (100–102).

Phospholipase Activation Cascades Effects of the TNF system on the formation of lipid mediators reflect, to some extent, enhanced expression of the enzymes involved. The signaling by both CD120a and CD95, however, also seems to have a direct effect on certain enzymes that produce such mediators. Knowledge of these signaling mechanisms is still rather limited (see above, section on Reliability of Interpretation). SPHINGOMYELINASE ACTIVATION3

Increased sphingomyelinase (SMase) activity in response to TNF-α application or CD95 ligation has been observed in various cells, either shortly after stimulation or as a late event that occurs secondarily to other signaling events or as part of the apoptotic process. Ceramide, the product of SMase action on membrane sphingomyelin, is thought to act as a secondary mediator that, through modulation of the activity of certain enzymes, enhances the response to stress. Initial speculation that ceramide also affects NF-κB activation by TNFα could not be confirmed (105). There is some evidence that, however, ceramide effects contribute to JNK activation and death induction by the TNF/Fas systems. It was suggested that the latter activity is mediated only by the lysosome-bound acid SMase (see below) and involves activation of the lysosomal protease cathepsin D, to which ceramide seems to bind specifically (106). Although some mammalian proteins with SMase activity have been cloned, it is not clear whether these or other enzymes are involved in the TNF/Fas effects. Nor is there any information on the nature of the structural changes in SMases that underlie their activation by CD120a or CD95. At least two distinct target enzymes appear to be involved. Neutral sphingomyelinase (nSMase) Both CD120a and CD95 activate this cell membrane-associated enzymatic function (107, 108). Although the 3 (See

98, 103–104 for reviews.)

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

346

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

mechanism of the CD95 effect is still unknown, CD120a activation has been shown to involve the receptor-associated adapter protein FAN (52). Cells from mice with targeted disruption of FAN fail to show nSMase activation by TNF-α. The mice also lack natural killer cell development (109). Acid sphingomyelinase (aSMase) and phosphatyidylcholine–specific phospolipase C (PC-PLC) As with the nSMase, the activity of aSMases, which reside in the lysosomes, is enhanced by both CD120a and CD95, but the receptor region involved in the aSMase effect is the DD (107, 108). Both receptors also activate PC-PLC, an effect involving TRADD and the DD regions in the receptors. D609, an inhibitor of PC-PLC, blocks aSMase activation by TNF, suggesting that the PC-PLC products play a role in this activation (110). The TNF-induced activation of phospholipase A2 (PLA2) has attracted considerable attention in view of the role of secondary mediators produced by the enzyme in the proinflammatory and pyrogenic activity of this cytokine. PLA2 provides arachidonic acid, the precursor for the eicosanoids (cytoplasmic PLA2 preferentially acts on sn-2-arachidonoyl phospholipids), and the precursor for PAF, when the sn-1 position of the phospholipid is an alkyl ether linkage. As in SMase activity, PLA2 can be activated by the TNF/Fas systems in different ways and at different times after receptor ligation. At least four different mechanisms are known to participate in these effects: (a) induced expression of the gene encoding the secreted PLA2, which acts remotely from its producing cells (see, e.g. 111); (b) induced expression of the cytoplasmic PLA2 (cPLA2) gene, observed several hours after TNF-α application (112); (c) early activation of cPLA2 (within minutes of ligand application) and its translocation from the cytoplasm to the plasma membrane, which seems to involve cPLA2 phosphorylation by any of the three MAP kinase cascades (113–115). Though this activation is modest, it strongly synergizes with that of Ca2+ ions (112); (d) Late increase in activity of a cellular PLA2 as part of the apoptotic program. This increase appears to be secondary to caspase activation (116). It is unlikely to involve the cPLA2 that is inactivated by the caspases. Rather, it seems to reflect activation of a type VI Ca2+-independent PLA2 (iPLA2) (96, 116). Apart from PLA2, phospholipase D (117) and—as mentioned above at least in some cells—sphingomyelinases (118) are activated at a late stage of the process of death induction. PHOSPHOLIPASE A2 (PLA2)

Protein Kinase and Protein Phosphatase Activation Changes in cellular protein phosphorylation patterns, reflecting alterations in the activity of a variety of protein kinases and phosphatases, are observed in cells shortly after TNF treatment (e.g. 119, 120). These phosphorylation and

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TNF RECEPTOR AND Fas SIGNALING

347

dephosphorylation events seem to be the exclusive mode of gene modulation by the TNF/Fas systems. They also appear to contribute to the regulation of all of their other effects. Most of the current knowledge on the identity of the enzymes involved in these phosphorylation events concerns the MAP kinases, an evolutionarily conserved group of protein kinase cascades whose basic module comprises three consecutively active enzymes: a proline-directed serine/threonine kinase (MAPK), a dual-specificity kinase (MAP2K) that activates the MAPK by phosphorylating both a serine and a tyrosine residue, and a serine/threonine kinase (MAP3K), which activates the MAP2K. The way in which the MAP3Ks become activated is not well understood. In several systems, the activation involves small G proteins. These proteins apparently prompt phosphorylation of the MAP3Ks by other, heterogeneous kinases (MAP4Ks). The substrates affected by these cascades are highly heterogeneous, some themselves being protein kinases (dubbed MAPKAPKs). Mammalian cells contain three known MAPK cascades (reviewed in 121–124); all are activated by both the TNF and the Fas systems, though the components of the activated cascades, as well as their targets, may vary from one cell type to another, in keeping with the cell type-specific patterns of responses to the TNF/Fas systems. The three cascades have different functions but cross-react on several levels. A related cascade, which is involved in NF-κB activation by these systems, was recently elucidated (reviewed 125, 126). STRESS-ACTIVATED PROTEIN KINASE 1 (SAPK1)/C-JUN N-TERMINAL KINASE (JNK) MAP KINASE CASCADE Contribution to the function of the TNF/Fas systems

As implied by their name, the SAPK cascades induce adaptive responses to a variety of stress signals. They do so mainly through induced changes in gene expression. The SAPK1/JNK pathway participates in the regulation of gene expression by the TNF/Fas systems both by enhancing the function of transcription factors, of which the most thoroughly studied is AP1 (123), and by affecting the stability of certain messages (127). It affects transcription through the phosphorylation of various transcription factors, including c-Jun, ATF2, Elk-1, and CREB. Among the genes it affects are those of collagenase IL-1α and c-Jun. Prolonged JNK activation results in death of some cells by an unknown mechanism, suggesting that this enzyme might be involved in the TNF/Fas-induced signaling for death. As with various other activities of the TNF and Fas systems, their activating effect on the SAPK cascades is restricted by antagonizing mechanisms and is therefore mostly transient, followed by a rather long period of lethargy. TNF-induced activation of phosphatase(s) may contribute to the transient character of SAPK1/JNK activation (128).

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

348

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

Mode of activation and the kinases involved All five receptors of the TNF/Fas systems can activate the SAPK1/JNK cascade (62, 129, 130). The effect of the TNF receptors involves TRAF2 (131), and according to limited evidence also MADD (48), as well as effects of the ceramide formed upon aSMase stimulation (reviewed in 98, yet see 132). The effect of CD95 involves the DD-associated adaptor protein DAXX (47, 219). In addition, the SAPK1/JNK pathway can be activated late in the process of death induction through caspase-mediated processing and activation of kinases that act in this pathway, for example PAK2 (133), PAK65 (134) and MEKK1 (135, 136). At least in macrophages, TNF affects the p46 isoform of JNK1 (MAPK) much more strongly than the p54 isoform (137). The main MAP2K mediating JNK activation by the TNF/Fas systems is MKK7 (138, 139). Recent studies point to two caspase-independent pathways through which this activation can occur. One of these pathways involves ASK1, a MAP3K whose carboxy-terminal kinase-flanking region has the ability to bind to TRAF2, 5 and 6, and its aminoterminal region to DAXX. Binding of ASK1 to either of the two adapter proteins results in displacement of an inhibitory intermolecular interaction between the two kinase-flanking regions, allowing activation of the kinase by both the TNF receptors and CD95 (141, 219, 220). The other pathway involves the MAP4K GCK. Similarly to ASK1, GCK binds to TRAF2. It also binds to the MAP3K MEKK1 and apparently activates it in a stimulus-dependent manner (221). GCKR/GLK, a kinase related to GCK, also seems to contribute in a similar way, to JNK activation by TNF (140, 222). Limited evidence suggests that JNK activation by TNF may also be mediated by the MAP3Ks TAK1 (perhaps through an effect of ceramide) (223) and MLK2 (142). SAPK2/P38 CASCADE Contribution to the function of the TNF/Fas systems Among the known target proteins of the SAPK2/p38 cascade are several transcription factors, some of which (such as ATF2) are identical to those affected by the JNKs, and also cytosolic proteins such as cPLA2 and hsp27. In many cells, the phosphorylation of hsp27—which is mediated by MAPKAP2, a kinase phosphorylated by p38 (the MAPK in this cascade)—is the most prominent TNF-α-induced serine/threonine phosphorylation event. Most of the information on the physiological significance of activation of this cascade is based on assessment of the effects of certain bicyclic imidazole inhibitors, which at least at low concentrations, appear to affect the function of the p38 kinase in a specific manner (143). These data suggest involvement of the kinase in the up-regulation of various inflammation-related genes, like TNFα itself, prostaglandin H synthase 2, collagenase-1, IL-6, and IL-8,

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

TNF RECEPTOR AND Fas SIGNALING

349

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

through effects on transcription of the genes, translation of the transcripts or their stability (144, and references therein). The functional significance of hsp27 phosphorylation in response to TNF is not known. Mode of activation and the kinases involved In all in vivo studies reported so far, induction of the p38 and JNK pathways has occurred simultaneously. In certain in vivo situations, however, it is possible to observe differential responses of the two pathways to some stimuli, suggesting that they share both common and distinct activation mechanisms. The MAP2Ks activating the p38 kinases were reported to be MKK2 and MKK3 (in response to TNF; 147) and MKK6 (in response to CD95 ligation) (139). As in the case of the JNK pathway, these kinases are activated by the TNF/Fas systems in both a caspase-dependent and a caspase-independent manner (139, 145). Of the two caspase-independent JNK-activation pathways (described above), the one involving GCK (or GCKR) and MEKK1 seems to lead rather specifically to JNK activation (221 and references within). The other, mediated by ASK1, can activate p38 as well (146). Occurrence of a signaling pathway that can mediate specific activation of p38 by TNF was suggested in a recent report describing association of a “p38 specific” MAP3K activity with the intermediate region of RIP (the region linking its death domain to the kinase domain in RIP). The identity of the RIP-associated enzyme mediating this p38-specific function is still unknown (221). Contribution to the function of the TNF and Fas systems Phosphorylation of the p42 and p44 ERKs (the MAPKs in this cascade) is the most prominent tyrosine phosphorylation event observed in certain cells in response to TNF (148, 149). There are also cells in which these kinases become activated upon Fas ligation (108). In many cells, however, activation of this cascade by the TNF/Fas systems is milder than that of SAPK cascades. The ERKs seem to contribute to the growth-stimulatory effects of TNF and Fas as well as to their effects on cell differentiation and inflammation. Their substrates include transcription factors such as Elk-1 and cytosolic proteins such as cPLA2.

ERK/MAP KINASE CASCADE

Mode of activation and the kinases involved Both the p42 and the p44 ERKs are activated by TNF. In murine macrophages, CD120a ligation was found preferentially to trigger activation of the p42 isoform through phosphorylation of the MAP2K MEK1, which in turn is activated by the MAP3K MEK kinase (150). In some other cells the MAP3K seems to be cRaf1 (151, 152). In studies of the involvement of cRaf1 in the activation of this cascade by the TNF/Fas systems, it was suggested that this kinase is activated through phosphorylation, either by a ceramide-regulated kinase activated by the FANnSMase pathway (151) or by protein kinase C zeta (153). The mechanism of the

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

350

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

reported stimulatory effect of the CD120a-associated adapter protein MADD on the ERK/MAP kinase pathway is unknown (48). NF-κB ACTIVATION KINASE CASCADE Contribution to the function of the TNF/ Fas systems The group of transcription factors collectively called NF-κB contributes to the control of expression of many of the genes that participate in inflammation and immune response. In most cells, these factors normally occur in a latent form imposed by their association with inhibitory proteins collectively termed I-κB, which dictate the cytoplasmic location of the proteins. They can, however, become activated in response to a wide range of inducers, including all receptors of the TNF system, and in some cells also CD95. The proteins controlled by NF-κB include many that contribute to the proinflammatory functions of TNF-α. Cells deficient in NF-κB function display increased sensitivity to the cytocidal effect of TNF and much lower dependence on treatment with protein-synthesis inhibitors for exhibiting such an effect, suggesting that some proteins regulated by NF-κB serve to protect cells against undue killing by TNF (reviewed in 154).

Kinases involved As with various other inducing agents, activation of NFκB by the two TNF receptors, and probably also the activation mediated by the two other receptors of the TNF system and by CD95, occur by triggering the phosphorylation of I-κBα at serines 32 and 36. Such phosphorylation targets this inhibitor for proteasomal degradation. Two structurally homologous serine/threonine kinases, IKK1 and IKK2, which mediate the phosphorylation of I-κB in response to these and various other inducers, were recently identified (155–159). These two enzymes associate within a macromolecular complex of ∼700,000 Daltons (the signalosome) that apparently contains several other regulatory enzymes and structural proteins. Two protein kinases homologous to MAP3Ks, NIK and MEKK1 (the latter also functions as a MAP3K in the JNK cascade), can activate the signalosome. NIK activates it mainly through phosphorylation of IKK1, and MEKK1 mainly through phosphorylation of IKK2 (160–163). Of these two MAP3K homologs, NIK is differentially involved in mediating the NF-κB-stimulating effects of receptors of the TNF family and IL1, while MEKK1 plays a central role in NF-κB activation by the HTLV-I Tax protein (161, 164). Although NIK does not activate JNK or the p38 kinase (146, 165), its enforced expression leads by unknown mechanisms to AP1 activation (166). Mode of activation Studies of enforced expression in cultured cells indicated involvement of the adapter proteins TRADD, RIP, and TRAF2 in NF-κB activation by CD120a, as well as involvement of TRAF2 in NF-κB activation by CD120b and by CD95. Specific binding of NIK to TRAF2 also indicates

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TNF RECEPTOR AND Fas SIGNALING

351

involvement of this adapter protein in NF-κB activation (161). Yet, although targeted disruption or mutation of RIP indeed results in ablation of TNF-induced NF-kB activation (167, 168), fibroblasts deficient in TRAF2 can still display such activation, suggesting that the role of this adapter protein in the process is dispensable (131). A number of other protein kinases, for example, certain protein kinase C species (153), are reportedly involved in the TNF-induced NF-κB activation. Whether these kinases operate by activating the NIK-IKK pathway or in other ways remains to be clarified. Several studies suggest that TNF also controls NFκB function at some post-I-κB phosphorylation step(s), through involvement of additional signaling pathways such as the SAPK2/p38 cascade (169). OTHER KINASES INVOLVED IN THE FUNCTION OF THE TNF/Fas SYSTEMS Fragmentary evidence points to the involvement of a number of additional protein kinases in the signaling activities of the TNF/Fas systems. Protein kinase C, most prominently the epsilon isoform but also other isoforms like zeta, display increased activity shortly after TNF application, probably partly in response to the diacyl glycerol formed by the PC–PLC activated by CD120a. These kinase isoforms might contribute to the activation of the NF-κB and the ERK/MAP kinase cascades as well as to the TNF-mediated induction of resistance to its own cytotoxicity. They may also contribute to the TNF-induced down-regulation and shedding of its own receptors (153, 170, 171). A poorly defined enzyme with beta casein kinase activity and apparently tyrosine kinase activity is stimulated rather specifically by TNF, IL-1, and IL-18. Its activation mechanism is unknown, except that it appears not to involve phosphorylation of the enzyme (172). Limited evidence indicates that Jak1, Jak2, and Tyk2, tyrosine kinases that are centrally involved in signaling for the effects of the interferons and of various receptors of the hematopoietin family, can also bind to the intracellular domain of CD120a and mediate activation of the transcriptional factors STAT1, 3, 5, and 6 (173). There is some evidence for the association of undefined serine/threonine kinase(s) with the intracellular domains of CD120a, CD120b (reviewed in 4; also see 68), LTβR (174), and CD95 (175) and phosphorylation of these receptors by them. Phosphorylation of CD120b by its associated kinase(s) (which displays casein kinase I activity) results in decreased signaling activity (68). The serine/threonine kinase(s) associated with the membrane-proximal region in CD95 are also capable of phosphorylating the CD95-associated adapter protein MORT1/FADD (175). RIP, a DD-containing adapter protein that participates in the activation of JNK and NF-κB, has serine/threonine kinase activity and can self-phosphorylate,

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

352

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

but its substrate proteins and the functional significance of this kinase activity are not yet known (29, 44).

REGULATION OF THE RESPONSE

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

General Considerations The multiple and contrasting activities of the receptors of the TNF/Fas systems could not result in any meaningful functional consequence were they not adjusted and coordinated by regulatory mechanisms. The ability of these systems to elicit antagonistic effects that can counterbalance each other probably contributes to this adjustment. Molecules that contribute to the restriction of the response can be identified in these signaling systems on almost all mechanistic levels—the availability of ligands and receptors, interaction of the receptors with docking proteins and of docking proteins with signaling enzymes, the function of the signaling enzymes and of amplification processes, and the function of proteins participating in the eventual phenotypic changes (reviewed in 176). One important feature of the process of decision taking in this regulation is that, once taken, the decision is perpetuated by the suppression of alternative options. In cells where death is induced, NF-κB action is prevented by inhibition of TRADD recruitment (177), and activation of the ERKs—although not of JNK or p38 kinase—is prevented through the action of some caspases (178). Conversely, once NF-κB is activated, it elicits the transcription of proteins with anti-apoptotic function (reviewed in 154). Another major mode of regulation affecting both the quality and the quantity of the response is the control of expression of the various receptors and ligands. This regulation occurs on the levels of transcription, translation, intracellular transport, and shedding. There are marked differences in the cellular expression patterns of the individual receptors and ligands, and for most (with the exception of CD120a, which is constitutively expressed by almost all types of cells) the expression is largely dependent on extracellular stimuli (reviewed in 7).

Localization of the Signaling Events As discussed above, (see Reliability of Interpretation), enforced expression studies in transfected cells could lead to erroneous impression of the function of signaling proteins. This might be the result of abnormal placing of transfected proteins in the cell, in a way that is inconsistent with the compartmentalization necessary for maintaining specificity in their function. Evidence has indeed been presented for restricted localization of various components of the TNF/Fas signaling cascades, as well as for strictly defined induced changes in these localizations as part of the signaling process. Some examples are given below.

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TNF RECEPTOR AND Fas SIGNALING

353

SIGNALING INITIATION AND OTHER PROTEIN TRANSLOCATION EVENTS None of the signaling proteins known to participate in signaling induction by CD120a, CD120b, or CD95 was found to associate with the receptors before triggering. For some of these molecules, receptor recruitment upon triggering may simply reflect their increased affinity for the receptors, to which they may well have bound loosely a priori. This was clearly shown, however, not to be the case with TRADD. At least in endothelial cells, this protein is largely located in the Golgi region before stimulation (179). Translocation of signaling molecules from the cell interior to the cell membrane is followed by the translocation of signal-mediating proteins in the reverse orientation, to target sites within the cell. The TNF/Fas systems use all three known modes of targeting of gene-activating signals to the nucleus: phosphorylation of nucleus-resident factors by activated kinases following their translocation to the nucleus (for example, the phosphorylation of cJun by JNK), direct phosphorylation of transcription factors by receptor-associated kinases (for example, phosphorylation of the STATs by the JAKs), and phosphorylation of a protein that holds back a transcription factor in the cytoplasm, thus allowing its translocation to the nucleus (for example, phosphorylation of I-κB by the IKKs). The fragmentary information of the death processes induced by these systems indicates that targeting of the death signals is defined just as carefully. As with the control of the transcription factor NF-κB, activation of CAD (an enzyme participating in the cleavage of DNA as part of the apoptotic process) occurs by inactivation of an inhibitory protein that holds it back in the cytoplasm, thus allowing its transport into the nucleus. In this case, however, the enzyme responsible for inactivation of the inhibitor is not a kinase, but a caspase (180). The location of the caspases themselves within the cell also seems to be restricted by specific molecular interactions. In some cells caspase-8, before being recruited by the activated receptors, is mainly associated with the mitochondria [as are caspase-7 (181) and to some extent caspase-3 (182)]. After processing, however, it is associated specifically with the cytoskeleton (87). Caspase-8 can also associate with a specific endoplasmic reticulum protein (183). LOCALIZATION OF THE EVENTS RELATED TO SIGNALING THROUGH THE MAP KINASE AND NIK-IKK CASCADES In addition to interacting through the actual

substrate recognition sites, kinases of the MAP kinase cascades can interact with each other, as well as with their substrates, through distinct docking sites. These interactions may not, however, fully account for the specificity of action of these kinases within the cell, which greatly exceeds their specificity observed in vitro. In yeast cells, colocalization of the different components of the pheromone-responsive MAPK cascade is dictated by their binding to a common scaffolding protein, STE5. Putative MAPK scaffolding proteins, one specific to the ERK and the other to the JNK MAP kinase cascades, have recently been

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

354

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

identified (224, 225). TRAF2 also seems to act as scaffolding protein to which numerous proteins participating in the signaling for MAPK activation can bind. In this case, however, the majority of these proteins are not the kinases themselves but other adapters such as RIP and TRADD and regulatory molecules such as cIAP2, TRAF1, TRIP, A20, and I-TRAF/TANK (reviewed in 65). Two additional macromolecular complexes specifically involved in the pathway that leads to NF-κB activation are the signalosome (157, 160) and the proteasome (184). The exact locations of these complexes in the cell, the extent to which they are distinct, and the way in which proteins such as NIK and the NF-κB complex are translocated between them remain to be further clarified.

Interactions of Different Signaling Systems In addition to the ability of each of the individual receptors of the TNF/Fas systems to induce multiple effects, the pleiotropicity of these systems reflects cross talk between the receptors. It also involves various kinds of interactions with other receptors, most notably with the receptors for cytokines such as the interferons, IL-1, various growth factors, and insulin, whose physiological roles are intimately related to those of these systems. INTERACTIONS OF THE DIFFERENT RECEPTORS OF THE TNF/Fas SYSTEMS Both similarities and differences between the death processes induced by CD120a and CD95 have been noted. It now seems that the similarities reflect activation of a shared death-inducing signaling pathway (31, 81, 99). The differences may well be a function of the different ways by which the receptors activate this shared caspase-8 pathway (for example, the inability of MORT1/FADD to bind directly to CD120a). They may also reflect superimposed effects of other signaling pathways, activated differentially by the two receptors. Shared signaling pathways may well account for the close functional relationship between CD120a and CD120b. The following three mechanisms have been suggested to contribute to these similarities: (a) cross talk between signaling molecules. The two receptors indeed use the same proximal molecules both in activating NF-κB [NIK (161)] and in inducing cell death (MORT1/FADD; W Declercq, personal communication). Several points of evidence indicate involvement of TRAF2 in the pro-apoptotic CD120b effect, perhaps by mediating the deviation of anti-apoptotic molecules such as cIAP1 and 2 or CASH (which bind to TRAF2) (185, 186) from CD120a (33, 34); (b) induced synthesis of cell-bound TNF-α and autocrine activation of CD120a by it (M Grell, personal communication); (c) enhancement of TNF-α binding to the CD120a by the CD120b at low TNF-α concentration (a ligand passing mechanism) (187). As in CD120b, the functions shared between the LTβR and CD120a seem to be mediated by TRAF molecules; however, the TRAF species involved in

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

TNF RECEPTOR AND Fas SIGNALING

355

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

death induction by this receptor (TRAF3) is distinct from those responsible for NF-αB activation (TRAF2 and TRAF5) (32). With the exception of the LTβ2/LTα1 heterotrimer, each of the ligands in the TNF system can bind more than one kind of receptor species. It thus seems possible that signaling initiation by these ligands involves not only imposed juxtaposition of the same receptor molecules but also heteroassociation of different receptors that bind to the same ligand. Some evidence indicates that CD120a and CD120b can indeed bind simultaneously to the same TNF-α molecule (188). INTERACTIONS WITH THE IL-1S There is almost no known effect of TNF that cannot be induced in some cells by IL-1, and for many of these effects, there is pronounced synergism in the function of the two cytokines. This relationship at least partly reflects shared early postreceptor events activated by these ligands. Like the TNF receptors, the IL-1 receptor uses a member of the TRAF family, TRAF6, in its signaling activity, and TRAF6 shares various functional features with the TNF-activated TRAF2. It binds common signaling molecules like NIK (161, 165). It also hetero-associates with TRAF2, a mechanism that may contribute to the synergism of the signaling pathways activated by IL-1 and TNF. INTERACTIONS WITH THE IFNS TNF-α has various antiviral effects, like those of the IFNs. Conversely, the IFNs, like TNF-α, have cytotoxic effects that can preferentially affect virus-infected cells (although, unlike with TNF-α, this effect—like all others of the IFNs—is dependent on gene activation). The ability of CD120a to activate the Jak kinases and Tyk, which play a principal role in signaling by the IFN receptors (173), might contribute to this close similarity of function. INTERACTIONS WITH THE EGF RECEPTOR FAMILY The cytocidal effects of both TNF-α and CD95 ligation are antagonized by certain growth-inducing ligands. There is particular interest in the mediation of such effects by receptors of the EGF family, as it appears that these receptors interact with those of TNF-α on several mechanistic levels. TNF-α induces increased expression both of the EGF receptor (EGF-R) and of its ligand TGF-α, although it suppresses the synthesis of the EGF-R homolog HER2/ERBB2 (189). Triggering or overexpression of HER2/ERBB2 or other members of the EGF-R family, with consequent activation of their tyrosine kinase function, endows cells with resistance to TNF-α cytotoxicity although they maintain expression of the TNF receptors (e.g. 190). Moreover, triggering of the TNF receptors induces serine or threonine phosphorylation of the EGF-R, apparently mediated by the ERK and JNK MAPKs (191), and in some cells, it induces tyrosine phosphorylation of the EGF-R. The latter, which results in increased in vitro kinase activity of

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

356

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

the receptor and signaling for c-fos gene expression (192), probably contributes to the resistance of the cells to TNF-α cytotoxicity (193). INTERACTIONS WITH THE INSULIN SIGNALING PATHWAY TNF-α inhibits signaling by the insulin receptor, an effect that—because TNF-α is constitutively formed by adipocytes—is believed to contribute to obesity-linked insulin resistance. The inhibitory effect, which is triggered by CD120a, is manifested in decreased association of the adapter protein IRS1 with the insulin receptor. It is also manifested by an inhibitory effect of IRS1 on the tyrosine proteinkinase activity of the insulin receptor, which is required for signaling by this receptor. These effects reflect increased serine phosphorylation of IRS1. It was suggested that the kinase mediating this phosphorylation is activated by ceramide, formed as a consequence of sphingomyelinase activation by TNF-α. Its identity is not known, however, and there is no knowledge of its substrate site(s) in IRS1 (194, 195).

CONCLUDING REMARKS “The world is embodied in a drop of dew,” said Goethe. To paraphrase this poetic insight, tracking the chains of interactions of even a single molecule may reward us with a view of the whole world. The vista obtained, however, depends on the starting point. The odyssey begun by tracking the molecules linked to the receptors of the TNF and Fas systems has already given us a unique outlook on the world of biological regulation. It has made a significant contribution to knowledge of the regulation of cell death, an aspect of biology that until recently was largely neglected, and is likely to contribute further to our understanding of how death of the individual cell and damage on the level of the whole tissue occur. There are additional functions that are unique to the TNF and Fas systems, for example, the function of lymph node organogenesis. Their exploration may well also make a novel contribution to the study of signaling. Perhaps the most intriguing next frontier in this odyssey, however, is one that concerns not so much the features unique to one particular signaling system as much as the features shared between the various members of the TNF/NGF family. The last few years have seen rapid growth in the numbers of known members of this family and of the family of TNF-related ligands. The ability to induce both death and growth of cells is rather common in these families, but, at the same time, they are also in charge of regulating almost any other biological function that comes to mind. Despite their large sizes and the wide range of activities they mediate, these families display a rather conserved pattern of molecular structures and mechanisms. Further studies of the TNF and Fas systems and of the various other receptors of the TNF/NGF family should provide

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

TNF RECEPTOR AND Fas SIGNALING

357

a better understanding of the unique mechanistic features of the combination of molecular structures that characterize this family. They will also increasingly reveal the advantages of this particular combination, used so abundantly and in so many ways by nature for the control of immune defense.

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ACKNOWLEDGMENTS The authors thank Jeffery Browning, Wim Declercq, Marja Jaattela, Stefan Leu, Peter Krammer, Martin Kroenke, Marcus Peter, Jordan Pober, and Peter Vandenabeele for advice and for providing unpublished results for inclusion in the manuscript and Shirley Smith for editorial assistance. Work cited from the authors’ laboratory was supported by grants from Inter-Lab Ltd., Ness Ziona, Israel, from Ares Trading SA, Switzerland, and from the Israeli Ministry of Arts and Sciences. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Vandenabeele P, Declercq W, Beyaert R, Fiers W. 1995. Two tumor necrosis factor receptors: structure and function. Trends Cell Biol. 5:392–400 2. Ware CF, VanArsdale TL, Crowe PD, Browning JL. 1995. The ligands and receptors of the lymphotoxin system. Curr. Top. Microbiol. Immunol. 198:175– 218 3. Wallach D. 1996. A decade of accumulated knowledge and emerging answers: the 6th International Congress on TNF Rhodes, Greece, May 1996. Eur. Cytokine Netw. 7:713–24 4. Darnay BG, Aggarwal BB. 1997. Early events in TNF signaling: a story of associations and dissociations. J. Leukocyte Biol. 61:559–66 5. Wallach D, Boldin M, Varfolomeev E, Beyaert R, Vandenabeele P, Fiers W. 1997. Cell death induction by receptors of the TNF family: towards a molecular understanding. FEBS Lett. 410:96–106 6. Nagata S. 1997. Apoptosis by death factor. Cell 88:355–65 7. Wallach D, Bigda J, Engelmann H. 1998. The TNF family and related molecules. In The Cytokine Network and Immune Functions, ed. J Th´eze, London/New York: Oxford Univ. Press. In press 8. Van Ostade X, Tavernier J, Fiers W. 1994. Structure-activity studies of human tumor necrosis factors. Protein Eng. 7:5–22

9. Williams Abbott L, Walter BN, Cheung TC, Goh CR, Porter AG, Ware CF. 1997. The lymphotoxin-alpha (LTalpha) subunit is essential for the assembly, but not for the receptor specificity, of the membraneanchored LTalpha1beta2 heterotrimeric ligand. J. Biol. Chem. 272:19451–56 10. Montgomery RI, Warner MS, Lum BJ, Spear PG. 1996. Herpes simplex virus-1 entry into cells mediated by a novel member of the TNF/NGF receptor family. Cell 87:427–36 11. Mauri DN, Ebner R, Montgomery RI, Kochel KD, Cheung TC, Yu G-L, Ruben S, Murphy M, Eisenberg RJ, Cohen GH, Spear PG, Ware CF. 1998. LIGHT, a new member of the TNF superfamily, and lymphotoxin are ligands for herpesvirus entry mediator. Immunity 8:21–30 12. Rothe J, Lesslauer W, Lotscher H, Lang Y, Koebel P, Kontgen F, Althage A, Zinkernagel R, Steinmetz M, Bluethmann H. 1993. Mice lacking the tumour necrosis factor receptor 1 are resistant to TNFmediated toxicity but highly susceptible to infection by Listeria monocytogenes. Nature 364:798–802 13. Pfeffer K, Matsuyama T, Kundig TM, Wakeham A, Kishihara K, Shahinian A, Wiegmann K, Ohashi PS, Kronke M, Mak TW. 1993. Mice deficient for the 55 kd tumor necrosis factor receptor are resistant to endotoxic shock, yet succumb to L.

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

358

14.

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

15.

16. 17.

18. 19.

20.

21.

22.

23.

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL monocytogenes infection. Cell 73:457– 67 Erickson SL, de Sauvage FJ, Kikly K, Carver Moore K, Pitts Meek S, Gillett N, Sheehan KC, Schreiber RD, Goeddel DV, Moore MW. 1994. Decreased sensitivity to tumour-necrosis factor but normal T-cell development in TNF receptor2-deficient mice. Nature 372:560–63 Rennert PD, James D, Mackay F, Browning JL, Hochmann PS. 1998. Lymph node genesis is induced by signaling through the lymphotoxin beta receptor. Immunity 9:71–79 Nagata S, Suda T. 1995. Fas and Fas ligand: lpr and gld mutations. Immunol. Today 16:39–43 Hernandez Caselles T, Stutman O. 1993. Immune functions of tumor necrosis factor. I. Tumor necrosis factor induces apoptosis of mouse thymocytes and can also stimulate or inhibit IL-6-induced proliferation depending on the concentration of mitogenic costimulation. J. Immunol. 151:3999–4012 Seino K, Kayagaki N, Okumura K, Yagita H. 1997. Antitumor effect of locally produced CD95 ligand. Nat. Med. 3:165–70 Grell M, Douni E, Wajant H, L¨ohden M, Clauss M, Baxeiner B, Georgopoulos S, Lesslauer W, Kollias G, Pfizenmaier K, Scheurich P. 1995. The transmembrane form of tumor necrosis factor is the prime activating ligand of the 80 kDa tumor necrosis factor receptor. Cell 83:793–802 Mackay F, Majeau GR, Hochman PS, Browning JL. 1996. Lymphotoxin beta receptor triggering induces activation of the nuclear factor kappa B transcription factor in some cell types. J. Biol. Chem. 271:24934–38 Browning JL, Miatkowski K, Sizing I, Griffiths D, Zafari M, Benjamin CD, Meier W, Mackay F. 1996. Signaling through the lymphotoxin beta receptor induces the death of some adenocarcinoma tumor lines. J. Exp. Med. 183:867–78 Kwon BS, Tan KB, Ni J, Lee KO, Kim KK, Kim YJ, Wang S, Gentz R, Yu GL, Harrop J, Lyn SD, Silverman C, Porter TG, Truneh A, Young PR. 1997. A newly identified member of the tumor necrosis factor receptor superfamily with a wide tissue distribution and involvement in lymphocyte activation. J. Biol. Chem. 272:14272–76 Rensing Ehl A, Hess S, Ziegler Heitbrock HW, Riethmuller G, Engelmann H. 1995. Fas/Apo-1 activates nuclear factor kappa B and induces interleukin-6 production. J. Inflamm. 45:161–74

24. Kapas L, Hong L, Cady AB, Opp MR, Postlethwaite AE, Seyer JM, Krueger JM. 1992. Somnogenic, pyrogenic, and anorectic activities of tumor necrosis factor-alpha and TNF-alpha fragments. Am. J. Physiol. 263:R708–15 25. Rathjen DA, Ferrante A, Aston R. 1993. Differential effects of small tumour necrosis factor-alpha peptides on tumour cell cytotoxicity, neutrophil activation and endothelial cell procoagulant activity. Immunology 80:293–99 26. Boldin MP, Mett IL, Varfolomeev EE, Chumakov I, Shemer AY, Camonis JH, Wallach D. 1995. Self-association of the “death domains” of the p55 tumor necrosis factor (TNF) receptor and Fas/APO1 prompts signaling for TNF and Fas/APO1 effects. J. Biol. Chem. 270:387–91 27. Naismith JH, Devine TQ, Brandhuber BJ, Sprang SR. 1995. Crystallographic evidence for dimerization of unliganded tumor necrosis factor receptor. J. Biol. Chem. 270:13303–7 28. Hsu H, Shu H-B, Pan M-G, Goeddel DV. 1996. TRADD-TRAF2 and TRADDFADD interactions define two distinct TNF receptor 1 signal transduction pathways. Cell 84:299–308 29. Hsu H, Huang J, Shu H-B, Baichwal V, Goeddel DV. 1996. TNF-dependent recruitment of the protein kinase RIP to the TNF receptor-1 signaling complex. Immunity 4:387–96 30. Takeuchi M, Rothe M, Goeddel DV. 1996. Anatomy of TRAF2. Distinct domains for nuclear factor-κB activation and association with tumor necrosis factor signaling proteins. J. Biol. Chem. 271:19935– 42 31. Zhang J, Cado D, Chen A, Kabra NH, Winoto A. 1998. Fas-mediated apoptosis and activation-induced T-cell proliferation are defective in mice lacking FADD/Mort1. Nature 392:296–300 32. Force WR, Cheung TC, Ware CF. 1997. Dominant negative mutants of TRAF3 reveal an important role for the coiled coil domains in cell death signaling by the lymphotoxin-beta receptor. J. Biol. Chem. 272:30835–40 33. Declercq W, Denecker G, Fiers W, Vabdenabeele P. 1998. Cooperation of both TNF receptors in inducing apoptosis: involvement of the TNF receptor-associated factor binding domain of the TNF receptor 75. J. Immunol. 161:390–99 34. Weiss T, Grell M, Siemienski K, M¨uhlenbeck F, D¨urkop H, Pfizenmaier K, Scheurich P, Wajant H. 1998. TNFR80dependent enhancement of TNFR60-

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

TNF RECEPTOR AND Fas SIGNALING

Annu. Rev. Immunol. 1999.17:331-367. 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. 46.

induced cell death is mediated by TRAF2 and is specific for TNFR60. J. Immunol. 161:3136–42 Maskos K, Fernandez-Catalan C, Huber R, Bourenkov GP, Bartunik H, Ellestad GA, Reddy P, Wolfson MF, Rauch CT, Castner BJ, Davis R, Clarke HR, Petersen M, Fitzner JN, Cerretti DP, March CJ, Paxton RJ, A. BR, Bode W. 1998. Crystal structure of the catalytic domain of human tumor necrosis factor-alphaconverting enzyme. Proc. Natl. Acad. Sci. USA 95:3408–12 Feinstein E, Wallach D, Boldin M, Varfolomeev E, Kimchi A. 1995. The death domain: a module shared by proteins with diverse cellular functions. Trends Biochem. Sci. 29:342–44 Hofmann K, Tschopp J. 1995. The death domain motif found in Fas (Apo-1) and TNF receptor is present in proteins involved in apoptosis and axonal guidance. FEBS Lett. 371:321–23 Huang B, Eberstadt M, Olejniczak ET, Meadows RP, Fesik SW. 1996. NMR structure and mutagenesis of the Fas (APO-1/CD95) death domain. Nature 384:638–41 Eberstadt M, Huang B, Olejniczak ET, Fesik SW. 1997. The lymphoproliferation mutation in Fas locally unfolds the Fas death domain [letter]. Nat. Struct. Biol. 4:983–85 Liepinsh E, Ilag LL, Otting G, Ibanez CF. 1997. NMR structure of the death domain of the p75 neurotrophin receptor. EMBO J. 16:4999–5005 Boldin MP, Varfolomeev EE, Pancer Z, Mett IL, Camonis JH, Wallach D. 1995. A novel protein that interacts with the death domain of Fas/APO1 contains a sequence motif related to the death domain. J. Biol. Chem. 270:7795–98 Chinnalyan AM, O’Rourke K, Tewari M, Dixit VM. 1995. FADD, a novel death domain-containing protein, interacts with the death domain of Fas and initiates apoptosis. Cell 81:505–12 Hsu H, Xiong J, Goeddel DV. 1995. The TNF receptor 1-associated protein TRADD signals cell death and NF-κB activation. Cell 81:495–504 Stanger BZ, Leder P, Lee T-H, Kim E, Seed B. 1995. RIP: a novel protein containing a death domain that interacts with Fas/APO-1 (CD95) in yeast and causes cell death. Cell 81:513–23 Duan H, Dixit VM. 1997. RAIDD is a new ‘death’ adapter molecule. Nature 385:86– 89 Ahmad M, Srinivasula SM, Wang L,

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

359

Talanian RV, Litwack G, FernandesAlnemri T, Alnemri ES. 1997. CRADD, a novel human apoptotic adaptor molecule for caspase-2, and FasL/tumor necrosis factor-interacting protein RIP. Cancer Res. 57:615–19 Yang X, Khosravi Far R, Chang HY, Baltimore D. 1997. Daxx, a novel Fas-binding protein that activates JNK and apoptosis. Cell 89:1067–76 Schievella AR, Chen JH, Graham JR, Lin LL. 1997. MADD, a novel death domain protein that interacts with the type 1 tumor necrosis factor receptor and activates mitogen-activated protein kinase. J. Biol. Chem. 272:12069–75 Brown TL, Howe PH. 1998. MADD is highly homologous to a Rab3 guaninenucleotide exchange protein (Rab3GEP). Curr. Biol. 8:R191 Zhang Y, Zhou L, Miller CA. 1998. A splicing variant of a death domain protein that is regulated by a mitogen-activated kinase is a substrate for c-Jun N-terminal kinase in the human central nervous system. Proc. Natl. Acad. Sci. USA 95:2586– 91 Tartaglia LA, Ayres TM, Wong GH, Goeddel DV. 1993. A novel domain within the 55 kd TNF receptor signals cell death. Cell 74:845–53 Adam Klages S, Adam D, Wiegmann K, Struve S, Kolanus W, Schneider Mergener J, Kronke M. 1996. FAN, a novel WD-repeat protein, couples the p55 TNFreceptor to neutral sphingomyelinase. Cell 86:937–47 De Vos K, Goossens V, Boone E, Grooten J, Vercammen D, Vancompernolle K, Vandenabeele P, Haegeman G, Fiers W, Grooten J. 1998. The 55-kDa tumor necrosis factor receptor induced clustering of mitochondria through its membrane-proximal region. J. Biol. Chem. 273:9673–80 Boldin MP, Mett IL, Wallach D. 1995. A protein related to a proteasomal subunit binds to the intracellular domain of the p55 TNF receptor upstream to its ‘death domain’. FEBS Lett. 367:39–44 Tsurumi C, Shimizu Y, Saeki M, Kato S, Demartino GN, Slaughter CA, Fujimuro M, Yokosawa H, Yamasaki M, Hendil KB, Toh e A, Tanahashi N, Tanaka K. 1996. cDNA cloning and functional analysis of the p97 subunit of the 26S proteasome, a polypeptide identical to the type-1 tumor-necrosis-factorreceptor-associated protein-2/55.11. Eur. J. Biochem. 239:912–21 Song HY, Dunbar JD, Zhang YX, Guo D,

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

360

57.

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

58.

59.

60.

61.

62.

63.

64.

65.

66.

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL Donner DB. 1995. Identification of a protein with homology to hsp90 that binds the type 1 tumor necrosis factor receptor. J. Biol. Chem. 270:3574–81 Sato T, Irie S, Kitada S, Reed JC. 1995. FAP-1: a protein tyrosine phosphatase that associates with Fas. Science 268:411–15 Cuppen E, Nagata S, Wieringa B, Hendriks W. 1997. No evidence for involvement of mouse protein-tyrosine phosphatase-BAS-like Fas-associated phosphatase-1 in Fas-mediated apoptosis. J. Biol. Chem. 272:30215–20 Rothe M, Wong SC, Henzel WJ, Goeddel DV. 1994. A novel family of putative signal transducers associated with the cytoplasmic domain of the 75 kDa tumor necrosis factor receptor. Cell 78:681–92 Rothe M, Sarma V, Dixit VM, Goeddel DV. 1995. TRAF2-mediated activation of NF-kappa B by TNF receptor 2 and CD40. Science 269:1424–27 Nakano H, Oshima H, Chung W, Williams Abbott L, Ware CF, Yagita H, Okumura K. 1996. TRAF5, an activator of NF-kappaB and putative signal transducer for the lymphotoxin-beta receptor. J. Biol. Chem. 271:14661–64 Marsters SA, Ayres TM, Skubatch M, Gray CL, Rothe M, Ashkenazi A. 1997. Herpesvirus entry mediator, a member of the tumor necrosis factor receptor (TNFR) family, interacts with members of the TNFR-associated factor family and activates the transcription factors NFkappaB and AP-1. J. Biol. Chem. 272: 14029–32 Hsu H, Solovyev I, Colombero A, Elliot R, Kelley M, Boyle WJ. 1997. ATAR, a novel tumor necrosis factor receptor family member, signals through TRAF2 and TRAF5. J. Biol. Chem. 272:13471– 74 Devergne O, Hatzivassiliou E, Izumi KM, Kaye KM, Kleijnen MF, Kieff E, Mosialos G. 1996. Association of TRAF1, TRAF2, and TRAF3 with an EpsteinBarr virus LMP1 domain important for B-lymphocyte transformation: role in NF-kappaB activation. Mol. Cell. Biol. 16:7098–108 Wallach D, Boldin MP, Kovalenko AV, Malinin NL, Mett IL, Camonis JH. 1998. The yeast two-hybrid screening technique and its use in the study of protein-protein interactions in apoptosis. Curr. Opin. Immunol. 10:131–36 Min W, Bradley JR, Galbraith JJ, Jones SJ, Ledgerwood EC, Pober JS. 1998. The N-terminal domains target TNF receptor-

67.

68.

69.

70.

71.

72.

73. 74.

75.

76.

associated factor-2 to the nucleus and display transcriptional regulatory activity. J. Immunol. 161:319–24 Darnay BG, Aggarwal BB. 1997. Inhibition of protein tyrosine phosphatases causes phosphorylation of tyrosine-331 in the p60 TNF receptor and inactivates the receptor-associated kinase. FEBS Lett. 410:361–67 Beyaert R, Vanhaesebroeck B, Declercq W, Van Lint J, Vandenabele P, Agostinis P, Vandenheede JR, Fiers W. 1995. Casein kinase-1 phosphorylates the p75 tumor necrosis factor receptor and negatively regulates tumor necrosis factor signaling for apoptosis. J. Biol. Chem. 270:23293– 99 Darnay BG, Singh S, Aggarwal BB. 1997. The p80 TNF receptor-associated kinase (p80TRAK) associates with residues 354–397 of the p80 cytoplasmic domain: similarity to casein kinase. FEBS Lett. 406:101–5 Ng PW, Janicke RU, Porter AG. 1998. Mutations which abolish phosphorylation of the TRAF-binding domain of TNF receptor 2 enhance receptor-mediated NF-kappaB activation. Biochem. Biophys. Res. Commun. 244:756–62 ´ Duda E, Wallach D. 1995. PhosP´ocsik E, phorylation of the 26 kDa TNF precursor in monocytic cells and in transfected HeLa cells. J. Inflamm. 45:152–60 Takahashi T, Tanaka M, Inazawa J, Abe T, Suda T, Nagata S. 1994. Human Fas ligand: gene structure, chromosomal location and species specificity. Int. Immunol. 6:1567–74 Nicholson DW, Thornberry NA. 1997. Caspases: killer proteases. Trends Biochem. Sci. 22:299–306 Boldin MP, Goncharov TM, Goltsev YV, Wallach D. 1996. Involvement of MACH, a novel MORT1/FADD-interacting protease, in Fas/APO1- and TNF receptorinduced cell death. Cell 85:803–15 Muzio M, Chinnaiyan AM, Kischkel FC, O’Rourke K, Shevchenko A, Ni J, Scaffidi C, Bretz JD, Zhang M, Gentz R, Mann M, Krammer PH, Peter ME, Dixit VM. 1996. FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-inducing signaling complex. Cell 85:817–27 Fernandes-Alnemri T, Armstrong RC, Krebs J, Srinivasula SM, Wang L, Bullrich F, Fritz LC, Trapani JA, Tomaselli KJ, Litwack G, Alnemri ES. 1996. In vitro activation of CPP32 and Mch3 by Mch4, a novel human apoptotic cysteine protease containing two FADD-like domains.

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TNF RECEPTOR AND Fas SIGNALING Proc. Natl. Acad. Sci. USA 93:7464–69 77. Vincenz C, Dixit VM. 1997. Fas-associated death domain protein interleukin1beta-converting enzyme 2 (FLICE2), an ICE/Ced-3 homologue, is proximally involved in CD95- and p55-mediated death signaling. J. Biol. Chem. 272:6578–83 78. Wallach D. 1997. Apoptosis. Placing death under control [news; comment]. Nature 388:123,125–26 79. Hofmann K, Bucher P, Tschopp J. 1997. The CARD domain: a new apoptotic signalling motif. Trends Biochem. Sci. 22: 155–56 80. Eberstadt M, Huang B, Chen Z, Meadows RP, Ng SC, Zheng L, Lenardo MJ, Fesik SW. 1998. NMR structure and mutagenesis of the FADD (Mort1) death-effector domain. Nature 392:941–45 81. Varfolomeev EE, Schuchmann M, Luria V, Chiannilkulchai N, Beckmann JS, Mett IL, Rebrikov D, Brodianski VM, Kemper OC, Kollet O, Lapidot T, Soffer D, Sobe T, Avraham KB, Goncharov T, Holtmann H, Lonai P, Wallach D. 1998. Targeted disruption of the mouse caspase-8 gene ablates cell-death induction by the TNF receptors, Fas/Apo1 and DR3 and is lethal prenatally. Immunity 9:267–76 82. Medema JP, Scaffidi C, Kischkel FC, Shevchenko A, Mann M, Krammer PH, Peter ME. 1997. FLICE is activated by association with the CD95 death-inducing signaling complex (DISC). EMBO J. 16:2794–804 83. Muzio M, Stockwell BR, Stennicke HR, Salvesen GS, Dixit VM. 1998. An induced proximity model for caspase-8 activation. J. Biol. Chem. 273:2926–30 84. Srinivasula SM, Ahmad M, FernandesAlnemri T, Litwack G, Alnemri ES. 1996. Molecular ordering of the Fas-apoptotic pathway: the Fas/APO-1 protease Mch5 is a CrmA-inhibitable protease that activates multiple Ced-3/ICE-like cysteine proteases. Proc. Natl. Acad. Sci. USA 93:14486–91 85. Pan G, O’Rourke K, Dixit VM. 1998. Caspase-9, Bcl-XL, and Apaf-1 form a ternary complex. J. Biol. Chem. 273: 5841–45 86. Hirata H, Takahashi A, Kobayashi S, Yonehara S, Sawai H, Okazaki T, Yamamoto K, Sasada M. 1998. Caspases are activated in a branched protease cascade and control distinct downstream processes in Fas-induced apoptosis. J. Exp. Med. 187:587–600 87. Stegh AH, Herrmann H, Weisenberger D, Lichter P, Krammer PH, Peter ME. 1998. Identification of the cytoskeletal

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

361

crosslinker protein plectin as a major in vivo substrate for capsase-8 during CD95-mediated apoptosis. J. Interferon Cytokine Res. 18:A–80 (Abstr.) Luo X, Budihardjo I, Zou Hm Slaughter C, Wang X. 1998. Bid, a Bcl2 interacting protein, mediates cytochrome c release from mitochondria in response to activation of cell surface death receptors. Cell 94:481–90 Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X. 1997. Cytochrome c and dATP-dependent formation of Apaf-1/Caspase-9 complex initiates and apoptotic protease cascade. Cell 91:479–89 Kuwana T, J SJ, Muzio M, Dixit V, Newmeyer DD, Kornbluth S. 1998. Apoptosis induction by caspase-8 is amplified through the mitochondrial release of cytochrome c. J. Biol. Chem. 273:16589–94 Scaffidi C, Fulda S, Srinivasan A, Friesen C, Li F, Tomaselli KJ, Debatin K-M, Krammer PH, Peter ME. 1998. Two CD95 (APO-1/Fas) signaling pathways. EMBO J. 17:1675–87 Boise LH, Thompson CB. 1997. Bcl-x(L) can inhibit apoptosis in cells that have undergone Fas-induced protease activation. Proc. Natl. Acad. Sci. USA 94:3759–64 Medema JP, Scaffidi C, Krammer PH, Peter ME. 1998. Bcl-xL acts downstream of caspase-8 activation by the CD95 death-inducing signaling complex. J. Biol. Chem. 273:3388–93 Srinivasan A, Li F, Wong A, Kodandapani L, Smidt RJ, Krebs JF, Fritz LC, Wu JC, Tomaselli KJ. 1998. Bcl-xL functions downstream of caspase-8 to inhibit Fas- and tumor necrosis factor receptor 1-induced apoptosis of MCF7 breast carcinoma cells. J. Biol. Chem. 273:4523–29 Weitzen M, Granger GA. 1980. The human LT system. VIII. A target celldependent enzymatic activation step required for the expression of the cytotoxic activity of human lymphotoxin. J. Immunol. 125:719–24 Atsumi G-i, Tajima M, Hadano A, Nakatani Y, Murakami M, Kudo I. 1998. Fas-induced arachidonic acid release is mediated by Ca2+-independent phospholipase A2, which undergoes proteolytic inactivation. J. Biol. Chem. 1998:13870–77 Deiss LP, Galinka H, Berissi H, Cohen O, Kimchi A. 1996. Cathepsin D protease mediates programmed cell death induced by interferon-gamma, Fas/APO-1 and TNF-alpha. EMBO J. 15:3861–70 Pena LA, Fuks Z, Kolesnick R. 1997.

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

362

99.

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

100.

101.

102.

103. 104. 105.

106.

107.

108.

109.

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL Stress-induced apoptosis and the sphingomyelin pathway. Biochem. Pharmacol. 53:615–21 Yeh W, de la Pompa JL, McCurrach ME, Shu H, Elia AJ, Shahinian A, Ng M, Wakeham A, Khoo W, Mitchell K, ElDeiry WS, Lowe SW, Goeddel DV, Mak TW. 1998. FADD: essential for embryo development and signaling from some, but not all, inducers of apoptosis. Science 279:1954–58 Vercammen D, Beyaert R, Denecker G, Goossens V, Van Loo G, Declercq W, Grooten J, Fiers W, Vandenabeele P. 1998. Inhibition of caspases increases the sensitivity of L929 cells to necrosis mediated by tumor necrosis factor. J. Exp. Med. 187:1477–85 Lemaire C, Andreau K, Souvannavong V, Adam A. 1998. Inhibition of caspase activity induced a switch from apoptosis to necrosis. FEBS Lett. 425:266–270 Vercammen D, Brouckaert G, Denecker G, Van de Craen M, Declercq W, Fiers W, Vandenabeele P. 1998. Dual signaling of the Fas receptor: initiation of both apoptotic and necrotic cell death pathways. J. Exp. Med. 188:919–30 Heller RA, Kronke M. 1994. Tumor necrosis factor receptor-mediated signaling pathways. J. Cell. Biol. 126:5–9 Hannun YA. 1996. Functions of ceramide in coordinating cellular responses to stress. Science 274:1855–59 Zumbansen M, Stoffel W. 1997. Tumor necrosis factor alpha activates NF-kappaB in acid sphingomyelinasedeficient mouse embryonic fibroblasts. J. Biol. Chem. 272:10904–09 Wickel M, Heinrich M, Rosenbaum C, Gahr J, Schwandner R, Kreder D, Brunner J, Kr¨onke M, Sch¨utze S. 1998. Identification of cathepsin D as a novel ceramidetarget involved in TNF signaling. J. Interferon Cytokine Res. 18:A–77 (Abstr.) Wiegmann K, Schutze S, Machleidt T, Witte D, Kronke M. 1994. Functional dichotomy of neutral and acidic sphingomyelinases in tumor necrosis factor signaling. Cell 78:1005–15 Cifone MG, Roncaioli P, De Maria R, Camarada G, Santoni A, Ruberti G, Testi R. 1995. Multiple pathways originate at the Fas/APO1 (CD95) receptor: sequential involvement of phosphatidylcholinespecific phospholipase C and acidic sphingomyelinase in the propagation of the apoptotic signal. EMBO J. 14:5859– 68 Kreder D, Schwandner R, Wiegmann K, Adam-Klage S, Schere G, Bernardo K,

110.

111.

112.

113.

114.

115.

116.

117.

118.

Pfeffer K, Kr¨onke M. 1998. FAN mediates TNF-dependent activation of neutral sphingomyelinase in vivo. J. Interferon Cytokine Res. 18:A–54 (Abstr.) Schutze S, Potthoff K, Machleidt T, Berkovic D, Wiegmann K, Kronke M. 1992. TNF activates NF-kappaB by phosphatidylcholine-specific phospholipase C-induced “acidic” sphingomyelin breakdown. Cell 71:765–76 van den Bosch H, Schalkwijk C, Pfeilschifter J, Marki F. 1992. The induction of cellular group II phospholipase A2 by cytokines and its prevention by dexamethasone. Adv. Exp. Med. Biol. 318:1– 10 Hoeck WG, Ramesha CS, Chang DJ, Fan N, Heller RA. 1993. Cytoplasmic phospholipase A2 activity and gene expression are stimulated by tumor necrosis factor: dexamethasone blocks the induced synthesis. Proc. Natl. Acad. Sci. USA 90: 4475–79 Lin LL, Wartmann M, Lin AY, Knopf JL, Seth A, Davis RJ. 1993. cPLA2 is phosphorylated and activated by MAP kinase. Cell 72:269–78 Kramer RM, Roberts EF, Um SL, Borsch Haubold AG, Watson SP, Fisher MJ, Jakubowski JA. 1996. p38 mitogenactivated protein kinase phosphorylates cytosolic phospholipase A2 (cPLA2) in thrombin-stimulated platelets. Evidence that proline-directed phosphorylation is not required for mobilization of arachidonic acid by cPLA2. J. Biol. Chem. 271: 27723–29 Hern´andez M, Bay´on Y, S´anchez Crespo M, Nieto ML. 1997. Thrombin produces phosphorylation of cytosolic phospholipase A2 by a mitogen-activated protein kinase kinase-dependent mechanism in the human astrocytoma cell line 1321N1. Biochem. J. 328:263–69 Wissing D, Mouritzen H, Egeblad M, Poirier GG, Jaattela M. 1997. Involvement of caspase-dependent activation of cytosolic phospholipase A2 in tumor necrosis factor-induced apoptosis. Proc. Natl. Acad. Sci. USA 94:5073–77 De Valck D, Beyaert R, Van Roy F, Fiers W. 1993. Tumor necrosis factor cytotoxicity is associated with phospholipase D activation. Eur. J. Biochem. 212:491– 97 Tepper AD, Cock JG, de Vries E, Borst J, van Blitterswijk WJ. 1997. CD95/Fasinduced ceramide formation proceeds with slow kinetics and is not blocked by caspase-3/CPP32 inhibition. J. Biol. Chem. 272:24308–12

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TNF RECEPTOR AND Fas SIGNALING 119. Guy GR, Chua SP, Wong NS, Ng SB, Tan YH. 1991. Interleukin 1 and tumor necrosis factor activate common multiple protein kinases in human fibroblasts. J. Biol. Chem. 266:14343–52 120. Wang XY, Kafka M, Dvilansky A, Nathan I. 1996. The roles of protein phosphorylation/dephosphorylation in tumor necrosis factor antitumor effects. J. Interferon Cytokine Res. 16:1021–25 121. Seger R, Krebs EG. 1995. The MAPK signaling cascade. FASEB J. 9:726–35 122. Paul A, Wilson S, Belham CM, Robinson CJ, Scott PH, Gould GW, Plevin R. 1997. Stress-activated protein kinases: activation, regulation and function. Cell Signal. 9:403–10 123. Karin M. 1996. The regulation of AP-1 activity by mitogen-activated protein kinases. Philos. Trans. R. Soc. London B Biol. Sci. 351:127–34 124. Woodgett JR, Kyriakis JM, Avruch J, Zon LI, Zanke B, Templeton DJ. 1996. Reconstitution of novel signalling cascades responding to cellular stresses. Philos. Trans. R. Soc. London B Biol. Sci. 351:135–41 125. Stancovski I, Baltimore D. 1997. NF-kB activation: The IkB kinase revealed? Cell 91:299–302 126. Verma IM, Stevenson J. 1997. IkappaB kinase: beginning, not the end. Proc. Natl. Acad. Sci. USA 94:11758–60 127. Chen C-Y, Del Gatto-Konczak F, Wu Z, Karin M. 1998. Stabilization of interleukin-2 mRNA by the c-Jun NH2terminal kinase pathway. Science 280: 1945–49 128. Guo YL, Baysal K, Kang B, Yang LJ, Williamson JR. 1998. Correlation between sustained c-Jun N-terminal protein kinase activation and apoptosis induced by tumor necrosis factor-alpha in rat mesangial cells. J. Biol. Chem. 273:4027– 34 129. Reinhard C, Shamoon B, Shyamala V, Williams LT. 1997. Tumor necrosis factor alpha-induced activation of c-jun N-terminal kinase is mediated by TRAF2. EMBO J. 16:1080–92 130. Chen C-M, Zimmerman G, Wu Lee Y-H, Grell M. 1998. Functional analysis of the cytoplasmic region in the lymphotoxin-β receptor (LT-βR). J. Interferon Cytokine Res. 18:A–18 (Abstr.) 131. Yeh W, Shahinian A, Speiser D, Kraunus J, Billia F, Wakeham A, de la Pompa J, Ferrick D, Hum B, Iscove N, Ohashi P, Rothe M, Goeddel DV, Mak T. 1997. Early lethality, functional NF-kB activation, and increased sensitivity to TNF-

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

142.

363

induced cell death in TRAF2-deficient mice. Immunity 7:715–25 Adam D, Ruff A, Strelow A, Wiegmann K, Kr¨onke M. 1998. Induction of stress-activated protein kinase/C-Jun aminoterminal kinases by the p55 tumor nectrosis factor receptor does not require sphingomyelinases. J. Interferon Cytokine Res. 18:A–60 (Abstr.) Rudel T, Bokoch GM. 1997. Membrane and morphological changes in apoptotic cells regulated by caspase-mediated activation of PAK2. Science 276:1571–74 Lee N, MacDonald H, Reinhard C, Halenbeck R, Roulston A, Shi T, Williams LT. 1997. Activation of hPAK65 by caspase cleavage induces some of the morphological and biochemical changes of apoptosis. Proc. Natl. Acad. Sci. USA 94:13642–47 Cardone MH, Salvesen GS, Widmann C, Johnson G, Frisch SM. 1997. The regulation of anoikis: MEKK-1 activation requires cleavage by caspases. Cell 90:315– 23 Deak JC, Cross JV, Lewis M, Qian Y, Parrott LA, Distelhorst CW, Templeton DJ. 1998. Fas-induced proteolytic activation and intracellular redistibution of the stress-signaling kinase MEKK1. Proc. Natl. Acad. Sci. USA 95:5595–600 Chan ED, Winston BW, Jarpe MB, Wynes MW, Riches DW. 1997. Preferential activation of the p46 isoform of JNK/SAPK in mouse macrophages by TNF alpha. Proc. Natl. Acad. Sci. USA 94:13169–74 Moriguchi T, Toyoshima F, Masuyama N, Hanafusa H, Gotoh Y, Nishida E. 1997. A novel SAPK/JNK kinase, MKK7, stimulated by TNF alpha and cellular stresses. EMBO J. 16:7045–53 Toyoshima F, Moriguchi T, Nishida E. 1997. Fas induces cytoplasmic apoptotic responses and activation of the MKK7JNK/SAPK and MKK6-p38 pathways independent of CPP32-like proteases. J. Cell Biol. 139:1005–15 Diener K, Wang XS, Chen C, Meyer CF, Keesler G, Zukowski M, Tan TH, Yao Z. 1997. Activation of the c-Jun N-terminal kinase pathway by a novel protein kinase related to human germinal center kinase. Proc. Natl. Acad. Sci. USA 94:9687–92 Ichijo H, Nishida E, Irie K, ten Dijke P, Saitoh M, Moriguchi T, Takagi M, Matsumoto K, Miyazono K, Gotoh Y. 1997. Induction of apoptosis by ASK1, a mammalian MAPKKK that activates SAPK/JNK and p38 signaling pathways. Science 275:90–94 Cuenda A, Dorow DS. 1998. Differential activation of stress-activated protein

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

364

143.

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

144.

145.

146.

147.

148.

149.

150.

151.

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL kinase SKK4/MKK7 and SKK1/MKK4 by the mixed-lineage kinase-2 and mitogen-activated protein kinase (MKK) kinase-1. Biochem. J. 333:11–15 Cuenda A, Rouse J, Doza YN, Meier R, Cohen P, Gallagher TF, Young PR, Lee JC. 1995. SB 203580 is a specific inhibitor of a MAP kinase homologue which is stimulated by cellular stresses and interleukin-1. FEBS Lett. 364:229–33 Ridley SH, Sarsfield SJ, Lee JC, Bigg HF, Cawston TE, Taylor DJ, DeWitt DL, Saklatvala J. 1997. Actions of IL-1 are selectively controlled by p38 mitogenactivated protein kinase: regulation of prostaglandin H synthase-2, metalloproteinases, and IL-6 at different levels. J. Immunol. 158:3165–73 Juo P, Kuo CJ, Reynolds SE, Konz RF, Raingeaud J, Davis RJ, Biemann HP, Blenis J. 1997. Fas activation of the p38 mitogen-activated protein kinase signalling pathway requires ICE/CED3 family proteases. Mol. Cell. Biol. 17:24–35. Erratum. 1997. Mol. Cell. Biol. 17:1757 Carpentier I, Declercq W, Malinin NL, Wallach D, Fiers W, Beyaert R. 1998. TRAF2 plays a dual role in NF-kappaBdependent gene activation by mediating the TNF-induced activation of p38 MAPK and IkappaB kinase pathways. FEBS Lett. 425:195–98 Winston BW, Chan ED, Johnson GL, Riches DW. 1997. Activation of p38mapk, MKK3, and MKK4 by TNFalpha in mouse bone marrow-derived macrophages. J. Immunol. 159:4491–97 Vietor I, Schwenger P, Li W, Schlessinger J, Vilcek J. 1993. Tumor necrosis factor-induced activation and increased tyrosine phosphorylation of mitogenactivated protein (MAP) kinase in human fibroblasts. J. Biol. Chem. 268: 18994–99 Van Lint J, Agostinis P, Vandevoorde V, Haegeman G, Fiers W, Merlevede W, Vandenheede JR. 1992. Tumor necrosis factor stimulates multiple serine/threonine protein kinases in Swiss 3T3 and L929 cells. Implication of casein kinase-2 and extracellular signal-regulated kinases in the tumor necrosis factor signal transduction pathway. J. Biol. Chem. 267:25916–21 Winston BW, Remigio LK, Riches DW. 1995. Preferential involvement of MEK1 in the tumor necrosis factor-alphainduced activation of p42mapk/erk2 in mouse macrophages. J. Biol. Chem. 270: 27391–4 Yao B, Zhang Y, Delikat S, Mathias S, Basu S, Kolesnick R. 1995. Phosphoryla-

152.

153.

154. 155.

156.

157.

158.

159.

160.

161.

162.

163.

tion of Raf by ceramide-activated protein kinase. Nature 378:307–10 Kalb A, Bluethmann H, Moore MW, Lesslauer W. 1996. Tumor necrosis factor receptors (Tnfr) in mouse fibroblasts deficient in Tnfr1 or Tnfr2 are signaling competent and activate the mitogen-activated protein kinase pathway with differential kinetics. J. Biol. Chem. 271:28097–104 Berra E, Diaz Meco MT, Lozano J, Frutos S, Municio MM, Sanchez P, Sanz L, Moscat J. 1995. Evidence for a role of MEK and MAPK during signal transduction by protein kinase C zeta. EMBO J. 14:6157–63 Wallach D. 1997. Cell death induction by TNF: a matter of self control. Trends Biochem. Sci. 22:107–46 Regnier CH, Song HY, Gao X, Goeddel DV, Cao Z, Rothe M. 1997. Identification and characterization of an IkappaB kinase. Cell 90:373–83 DiDonato JA, Hayakawa M, Rothwarf DM, Zandi E, Karin M. 1997. A cytokineresponsive IkappaB kinase that activates the transcription factor NF-kappaB. Nature 386:548–55 Mercurio F, Zhu H, Murray BW, Shevchenko A, Bennett BL, Li J, Young DB, Barbosa M, Mann M, Manning A, Rao A. 1997. IKK-1 and IKK-2: cytokine-activated IkappaB kinases essential for NF-kappaB activation [see comments]. Science 278:860–66 Zandi E, Rothwarf DM, Delhase M, Hayakawa M, Karin M. 1997. The IkappaB kinase complex (IKK) contains two kinase subunits, IKKalpha and IKKbeta, necessary for IkappaB phosphorylation and NF-kappaB activation. Cell 91:243–52 Woronicz JD, Gao X, Cao Z, Rothe M, Goeddel DV. 1997. IkappaB kinase-beta: NF-kappaB activation and complex formation with IkappaB kinase-alpha and NIK. Science 278:866–69 Lee FS, Hagler J, Chen ZJ, Maniatis T. 1997. Activation of the IkappaB alpha kinase complex by MEKK1, a kinase of the JNK pathway. Cell 88:213–22 Malinin NL, Boldin MP, Kovalenko AV, Wallach D. 1997. MAP3K-related kinase involved in NF-κB induction by TNF, CD95 and IL-1. Nature 385:540–44 Ling L, Cao Z, Goeddel DV. 1998. NFκB-induced kinase activates IKK-α by phosphorylation of Ser-176. Proc. Natl. Acad. Sci. USA 95:3792–97 Nakano H, Shindo M, Sakon S, Nishinaka S, Mihara M, Yagita H, Okumura K. 1998. Differential regulation of IkappaB

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

TNF RECEPTOR AND Fas SIGNALING

164.

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

165.

166.

167.

168.

169.

170.

171.

172.

kinase alpha and beta by two upstream kinases, NF-kappaB-inducing kinase and mitogen-activated protein kinase/ERK kinase kinase-1. Proc. Natl. Acad. Sci. USA 95:3537–42 Yin MJ, Christerson LB, Yamamoto Y, Kwak YT, Xu S, Mercurio F, Barbosa M, Cobb MH, Gaynor RB. 1998. HTLV-I Tax protein binds to MEKK1 to stimulate IkappaB kinase activity and NF-kappaB activation. Cell 93:875–84 Song HY, Regnier CH, Kirschning CJ, Goeddel DV, Rothe M. 1997. Tumor necrosis factor (TNF)-mediated kinase cascades: bifurcation of nuclear factorkappaB and c-jun N-terminal kinase (JNK/SAPK) pathways at TNF receptorassociated factor 2. Proc. Natl. Acad. Sci. USA 94:9792–96 Natoli G, Costanzo A, Moretti F, Fulco M, Balsano C, Levrero M. 1997. Tumor necrosis factor (TNF) receptor 1 signaling downstream of TNF receptor-associated factor 2. Nuclear factor kappaB (NFkappaB)-inducing kinase requirement for activation of activating protein 1 and NFkappaB but not of c-Jun N-terminal kinase/stress-activated protein kinase. J. Biol. Chem. 272:26079–82 Ting AT, Pimentel Muinos FX, Seed B. 1996. RIP mediates tumor necrosis factor receptor 1 activation of NF-kappaB but not Fas/APO-1-initiated apoptosis. EMBO J. 15:6189–96 Kelliher MA, Grimm S, Ishida Y, Kuo F, Stanger BZ, Leder P. 1998. The death domain kinase RIP mediates the TNFinduced NF-kB signal. Immunity 8:297– 303 Vanden Berghe W, Plaisance S, Boone E, De Bosscher K, Schmitz ML, Fiers W, Haegeman G. 1998. p38 and extracellular signal-regulated kinase mitogenactivated protein kinase pathways are required for nuclear factor-kappaB p65 transactivation mediated by tumor necrosis factor. J. Biol. Chem. 273:3285–90 Holtmann H, Wallach D. 1987. Down regulation of the receptors for tumor necrosis factor by interleukin 1 and 4 beta-phorbol-12-myristate-13-acetate. J. Immunol. 139:1161–67 Mire Sluis A, Meager A. 1994. Tumor necrosis factor (TNF)-induced protein phosphorylation in a human rhabdomyosarcoma cell line is mediated by 60-kD TNF receptors (TR60). Blood 83:2211–20 Guesdon F, Knight CG, Rawlinson LM, Saklatvala J. 1997. Dual specificity of the interleukin 1- and tumor necrosis factor-

173.

174.

175.

176.

177.

178.

179.

180.

181.

182.

183.

365

activated beta casein kinase. J. Biol. Chem. 272:30017–24 Guo D, Dunbar JD, Yang CH, Pfeffer LM, Donner DB. 1998. Induction of Jak/STAT signaling by activation of the type 1 TNF receptor. J. Immunol. 160:2742–50 Wu MY, Hsu TL, Lin WW, Campbell RD, Hsieh SL. 1997. Serine/threonine kinase activity associated with the cytoplasmic domain of the lymphotoxin-beta receptor in HepG2 cells. J. Biol. Chem. 272:17154–59 Kennedy NJ, Budd RC. 1998. Phosphorylation of FADD/MORT1 and Fas by kinases that associate with the membraneproximal cytoplasmic domain of Fas. J. Immunol. 160:4881–88 Wallach D, Kovalenko AV, Varfolomeev EE, Boldin MP. 1998. Death-inducing functions of ligands of the TNF family: a Sanhedrin verdict. Curr. Opin. Immunol. 10:279–88 Madge LA, Sierra-Honigman MR, Pober JS. 1998. Suppression of TNFα- but not IL-1-induced NFκB activation in human endothelial cells by initiation of cell death. J. Interferon Cytokine Res. 18:A71 (Abstr.) Widmann C, Gibson S, Johnson GL. 1998. Caspase-dependent cleavage of signaling proteins during apoptosis: a turnoff mechanism for anti-apoptotic signals. J. Biol. Chem. 273:7141–47 Jones S, Prins JB, Savidge J, Ledgerwood EC, Johnson D, Pober JS, Bradley JR. 1998. TNF recruits TRADD to the plasma membrane but not to trans-golgi network, the principal subcellular localization of TNFR1. J. Interferon Cytokine Res. 18:A51 (Abstr.) Enari M, Sakahira H, Yokoyama H, Okawa K, Iwamatsu A, Nagata S. 1998. A caspase-activated DNase that degrades DNA during apoptosis and its inhibitor ICAD. Nature 391:43–50 Chandler JM, Cohen GM, MacFarlane M. 1998. Different subcellular distribution of caspase-3 and caspase-7 following Fasinduced apoptosis in mouse liver. J. Biol. Chem. 273:10815–18 Mancini M, Nicholson DW, Roy S, Thornberry NA, Peterson EP, Casciola Rosen LA, Rosen A. 1998. The caspase-3 precursor has a cytosolic and mitochondrial distribution: implications for apoptotic signaling. J. Cell Biol. 140:1485– 95 Ng FW, Nguyen M, Kwan T, Branton PE, Nicholson DW, Cromlish JA, Shore GC. 1997. p28 Bap31, a Bcl-2/Bcl-XLand procaspase-8-associated protein in

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

366

184. 185.

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

186. 187.

188.

189.

190.

191.

192.

193.

194.

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

WALLACH ET AL the endoplasmic reticulum. J. Cell. Biol. 139:327–38 Hilt W, Wolf DH. 1996. Proteasomes: destruction as a programme. Trends Biochem. Sci. 21:96–102 Rothe M, Pan M-G, Henzel WJ, Ayres TM, Goeddel DV. 1995. The TNFR2TRAF signaling complex contains two novel proteins related to baculoviral inhibitor of apoptosis proteins. Cell 83: 1243–52 Shu HB, Halpin DR, Goeddel DV. 1997. Casper is a FADD- and caspase-related inducer of apoptosis. Immunity 6:751–63 Tartaglia LA, Pennica D, Goeddel DV. 1993. Ligand passing: the 75-kDa tumor necrosis factor (TNF) receptor recruits TNF for signaling by the 55-kDa TNF receptor. J. Biol. Chem. 268:18542– 48 Pinckard JK, Sheehan KC, Schreiber RD. 1997. Ligand-induced formation of p55 and p75 tumor necrosis factor receptor heterocomplexes on intact cells. J. Biol. Chem. 272:10784–89 Kalthoff H, Roeder C, Gieseking J, Humburg I, Schmiegel W. 1993. Inverse regulation of human ERBB2 and epidermal growth factor receptors by tumor necrosis factor alpha. Proc. Natl. Acad. Sci. USA 90:8972–76 Hudziak RM, Lewis GD, Shalaby MR, Eessalu TE, Aggarwal BB, Ullrich A, Shepard HM. 1988. Amplified expression of the HER2/ERBB2 oncogene induces resistance to tumor necrosis factor alpha in NIH 3T3 cells. Proc. Natl. Acad. Sci. USA 85:5102–6 Bird TA, Saklatvala J. 1990. Downmodulation of epidermal growth factor receptor affinity in fibroblasts treated with interleukin 1 or tumor necrosis factor is associated with phosphorylation at a site other than threonine 654. J. Biol. Chem. 265:235–40 Perez M, Donato NJ. 1996. Activation of epidermal growth factor receptor tyrosine phosphorylation by tumor necrosis factor correlates with loss of cytotoxic activity. J. Interferon Cytokine Res. 16:307–14 Izumi H, Ono M, Ushiro S, Kohno K, Kung HF, Kuwano M. 1994. Cross talk of tumor necrosis factor-alpha and epidermal growth factor in human microvascular endothelial cells. Exp. Cell. Res. 214:654–62 Kanety H, Hemi R, Papa MZ, Karasik A. 1996. Sphingomyelinase and ceramide suppress insulin-induced tyrosine phosphorylation of the insulin receptor substrate-1. J. Biol. Chem. 271:9895–97

195. Peraldi P, Hotamisligil GS, Buurman WA, White MF, Spiegelman BM. 1996. Tumor necrosis factor (TNF)-alpha inhibits insulin signaling through stimulation of the p55 TNF receptor and activation of sphingomyelinase. J. Biol. Chem. 271:13018– 22 196. Aggarwal BB, Moffat B, Harkins RN. 1984. Human lymphotoxin. Production by a lymphoblastoid cell line, purification, and initial characterization. J. Biol. Chem. 259:686–91 197. Gray PW, Aggarwal BB, Benton CV, Bringman TS, Henzel WJ, Jarrett JA, Leung DW, Moffat BNP, Svedersky LP, Palladino MA, Nedwin GE. 1984. Cloning and expression of cDNA for human lymphotoxin, a lymphokine with tumor necrosis activity. Nature 312:721–24 198. Aggarwal BB, Kohr WJ, Hass PE, Moffat B, Spencer SA, Henzel WJ, Bringman TS, Nedwin GE, Goeddel DV, Harkins RN. 1985. Human tumor necrosis factor. Production, purification, and characterization. J. Biol. Chem. 260:2345–54 199. Pennica D, Nedwin GE, Hayflick JS, Seeburg PH, Derynck R, Paladino MA, Kohr WJ, Aggarwal BB, Goeddel DV. 1984. Human tumor necrosis factor: precursor structure, cDNA cloning, expression, and homology to lymphotoxin. Nature 312:724–29 200. Engelmann H, Novick D, Wallach D. 1990. Two tumor necrosis factor binding proteins purified from human urine. Evidence for immunological cross reactivity with cell surface tumor-necrosis-factor receptors. J. Biol. Chem. 265:1531–36 201. Loetscher H, Schlaeger EJ, Lahm HW, Pan YC, Lesslauer W, Brockhaus M. 1990. Purification and partial amino acid sequence analysis of two distinct tumor necrosis factor receptors from HL60 cells. J. Biol. Chem. 265:20131–38 202. Loetscher H, Pan Y-CE, Lahm H-W, Gentz R, Brockhaus M, Tabuchi H, Lesslauer W. 1990. Molecular cloning and expression of the human 55 kd tumor necrosis factor receptor. Cell 61:351–59 203. Schall TJ, Lewis M, Koller KJ, Lee A, Rice GC, Wong GHW, Gatanaga T, Granger GA, Leutz R, Raab H, Kohr WJ, Goeddel DV. 1990. Molecular cloning and expression of a receptor for human tumor necrosis factor. Cell 61:361–70 204. Smith CA, Davis T, Anderson D, Solam L, Beckmann MP, Jerzy R, Dower SK, Cosman D, Goodwin RG. 1990. A receptor for tumor necrosis factor defines an unusual family of cellular and viral proteins. Science 248:1019–23

P1: PSA/MBG

P2: PKS/nbl/vks

February 12, 1999

10:20

QC: ARS/ABE

T1: ARS

Annual Reviews

AR078-12

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TNF RECEPTOR AND Fas SIGNALING 205. Granger GA, Kolb WP. 1968. Lymphocyte in vitro cytotoxicity: mechanisms of immune and non-immune small lymphocyte mediated target L cell destruction. J. Immunol. 101:111–20 206. Ruddle NH, Waksman BH. 1968. Cytotoxicity mediated by soluble antigen and lymphocytes in delayed hypersensitivity. III. Analysis of mechanisms. J. Exp. Med. 128:1267–79 207. Carswell EA, Old LJ, Kassel S, Green S, Fiore N, Williamson B. 1975. An endotoxin-induced serum factor that causes necrosis of tumors. Proc. Natl. Acad. Sci. USA 72:3666–70 208. Yonehara S, Ishii A, Yonehara M. 1989. A cell-killing monoclonal antibody (antiFas) to a cell surface antigen codownregulated with the receptor of tumor necrosis factor. J. Exp. Med. 169:1747–56 209. Trauth BC, Klas C, Peters AM, Matzku S, Moller P, Falk W, Debatin KM, Krammer PH. 1989. Monoclonal antibodymediated tumor regression by induction of apoptosis. Science 245:301–5 210. Itoh N, Yonehara S, Ishii A, Yonehara M, Mizushima S, Sameshima M, Hase A, Seto Y, Nagata S. 1991. The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis. Cell 66:233–43 211. Oehm A, Behrmann I, Falk W, Pawlita M, Maier G, Klas C, Li WM, Richards S, Dhein J, Trauth BC, Ponstingl H, Krammer PH. 1992. Purification and molecular cloning of the APO-1 cell surface antigen, a member of the tumor necrosis factor/nerve growth factor receptor superfamily. Sequence identity with the Fas antigen. J. Biol. Chem. 267:10709–15 212. Suda T, Takahashi T, Goldstein P, Nagata S. 1993. Molecular cloning and expression of the Fas ligand, a novel member of the tumor necrosis factor family. Cell 75:1169–78 213. Browning JL, Androlewicz MJ, Ware CF. 1991. Lymphotoxin and an associated 33kDa glycoprotein are expressed on the surface of an activated human T cell hybridoma. J. Immunol. 147:1230–37 214. Baens M, Chaffanet M, Cassiman JJ, van den Berghe H, Marynen P. 1993. Construction and evaluation of a hncDNA library of human 12p transcribed sequences derived from a somatic cell hybrid. Genomics 16:214–18 215. Bazan JF. 1993. Emerging families of cytokines and receptors. Curr. Biol. 3:603–6 216. Engelmann H, Holtmann H, Brakebusch

217.

218. 219.

220.

221.

222.

223.

224.

225.

367

C, Shemer Avni Y, Sarov I, Nophar Y, Hadas E, Leitner O, Wallach D. 1990. Antibodies to a soluble form of a tumor necrosis factor receptor have TNF-like activity. J. Biol. Chem. 265:14497–504 Dhein J, Daniel PT, Trauth BC, Oehm A, Moller P, Krammer PH. 1992. Induction of apoptosis by monoclonal antibody antiAPO-1 class switch variants is dependent on cross-linking of APO-1 cell surface antigens. J. Immunol. 149:3166–73 Li H, Zhu H, Xu C-j, Yuan J. 1998. Cleavage of BID by caspase 8 mediates the mitochondrial damage. Cell 94:491–501 Chang HY, Nishitoh H, Yang X, Ichijo H, Baltimore D. 1998. Activation of Apoptosis Signal-regulating Kinase 1 (ASK1) by the death adaptor Daxx. Science 281:1860–63 Nishitoh H, Saitoh M, Mochida Y, Takeda K, Nakano H, Rothe M, Miyazono K, Ichijo H. 1998. ASK1 is essential for JNK/SAPK activation by TRAF2. Mol. Cell 2:1–20 Yuasa T, Ohno S, Kehrl JH, Kyriakis JM. 1998. Tumor necrosis factor signaling to stress-activated protein kinase (SAPK)/Jun NH2-terminal kinase (JNK) and p38. Germinal center kinase couples traf2 to mitogen-activated protein kinase/erk kinase kinase 1 and sapk while receptor interacting protein associates with a mitogen-activated protein kinase kinase kinase upstream of mkk6 and p38. J. Biol. Chem. 273:22681–92 Shi CS, Kehrl JH. 1997. Activation of stress-activated protein kinase/c-Jun N-terminal kinase, but not NF-kappaB, by the tumor necrosis factor (TNF) receptor 1 through a TNF receptor-associated factor 2- and germinal center kinase related-dependent pathway. J. Biol. Chem. 272:32102–7 Shirakabe K, Yamaguchi K, Shibuya H, Irie K, Matsuda S, Moriguchi T, Gotoh Y, Matsumoto K, Nishida E. 1997. TAK1 mediates the ceramide signaling to stressactivated protein kinase/c-Jun N-terminal kinase. J. Biol. Chem. 272:8141–44 Schaeffer HJ, Catling AD, Eblen ST, Collier LS, Krauss A, Weber MJ. 1998. MP1: A MEK binding partner that enhances enzymatic activation of the MAP kinase cascade. Science 281:1668–71 Whitmarsh AJ, Cavanagh J, Tournier C, Yasuda J, Davis RJ. 1998. A mammalian scaffold complex that selectively mediates MAP kinase activation. Science 281:1671–74

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:331-367. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

QC: ARS/uks

12:15

T1: KKK

Annual Reviews

AR078-13

Annu. Rev. Immunol. 1999. 17:369–97 c 1999 by Annual Reviews. All rights reserved Copyright °

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

STRUCTURAL BASIS OF T CELL RECOGNITION K. Christopher Garcia The Scripps Research Institute, Department of Molecular Biology, 10550 North Torrey Pines Road, La Jolla, California 92037; e-mail: [email protected]

Luc Teyton The Scripps Research Institute, Department of Immunology, 10550 North Torrey Pines Road, La Jolla, California 92037

Ian A. Wilson The Scripps Research Institute, Department of Molecular Biology, and The Skaggs Institute for Chemical Biology, 10550 North Torrey Pines Road, La Jolla, California 92037 KEY WORDS:

three-dimensional structure, T cell receptor, TCR, protein-protein recognition, protein crystallography, MHC-peptide complexes, TCR-MHC complexes

ABSTRACT Exciting breakthroughs in the last two years have begun to elucidate the structural basis of cellular immune recognition. Crystal structures have been determined for full-length and truncated forms of αβ T cell receptor (TCR) heterodimers, both alone and in complex with their peptide-MHC (pMHC) ligands or with antiTCR antibodies. In addition, a truncated CD8 coreceptor has been visualized with a pMHC. Aided in large part by the substantial body of knowledge accumulated over the last 25 years on antibody structure, a number of general conclusions about TCR structure and its recognition of antigen can already be derived from the relatively few TCR structures that have been determined. Small, but important, variations between TCR and antibody structures bear on their functional differences as well as on their specific antigen recognition requirements. As observed in antibodies, canonical CDR loop structures are already emerging for some of the TCR CDR loops. Highly similar docking orientations of the TCR Vα domains in the TCR/pMHC complex appear to play a primary role in dictating orientation, but the Vβ positions diverge widely. Similar TCR contact positions, but whose exact amino acid content can vary, coupled with relatively poor interface shape complementarity, may explain the flexibility and short half-lives of many TCR

369 0732-0582/99/0410-0369$08.00

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

370

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

interactions with pMHC. Here we summarize the current state of this field, and suggest that the knowledge gap between the three-dimensional structure and the signaling function of the TCR can be bridged through a synthesis of molecular biological and biophysical techniques.

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTRODUCTION The formation of a complex between a T cell receptor (TCR) and a peptideMHC ligand (pMHC) represents the molecular solution to the recognition of an antigen in the cellular immune response (1). Although TCR recognition of pMHC is functionally analogous to antibody–antigen interaction in the humoral system (2), T cell recognition is a more complex process from a genetic, structural, and biological standpoint (3–5). Specificity in cellular responses arises from a TCR repertoire as extensive as that of antibodies, but is additionally coupled to polymorphism in the MHC that controls the size and diversity of the peptide repertoire presented. Furthermore, the TCR does not bind to pMHC in isolation, but as part of a multicomponent signaling complex that includes the CD3 γ , δ, ε, and ζ chains, and coreceptors CD8 or CD4 (6–9). Thus, the TCR functions not only to bind pMHC, but also to trigger the signaling components of the complex when an antigenic pMHC is encountered (5). Although TCRs and antibodies are both assembled by genetic recombination of V, D, J, and C segments, only antibodies undergo somatic mutation to increase their affinity for antigen (i.e. affinity maturation) (10, 11). Consistent with their function of clearing infective agents from the circulation, affinity-matured antibodies can recognize a broad spectrum of different antigens including proteins, peptides, carbohydrates, small molecules, and DNA (12). TCRs, on the other hand, recognize processed antigens, such as peptides, but only when bound to an MHC molecule (13). The TCR/pMHC affinities are substantially weaker than those of antibody/antigen complexes, and are usually in the low micromolar range (14, 15). The TCR, then, has the problem of achieving sufficient antigen specificity from the small structural differences that result from peptide binding to the highly conserved surface of the MHC. The structural polymorphism of this TCR-antigen complex is centered at the heart of the interface between the TCR and pMHC. The postengagement signaling decisions, however, are distributed among the numerous monomorphic components of the TCR signaling complex in a kinetic balance of weak macromolecular interactions (7, 16) that permits a fine control of the various antagonist, agonist, weak agonist, or null signaling outcomes (17, 18). A structural understanding of this subtle balance between binding and kinetics has been greatly illuminated in the last few years, initially from crystal

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL RECOGNITION

371

structures of TCR fragments (19–21), and then from the first crystal structures of intact αβ T cell receptors (22–24) and their complexes with pMHC (22, 25–28). This structural information has begun to clarify how the TCR is adapted for pMHC recognition (3, 29). A generalized orientation for how the TCR docks onto the surface of the pMHC has emerged from these structural studies (22, 25–28). This understanding of how the structure of the TCR and its complexes are related to its function has only been possible through a combination of biophysical and molecular biological approaches. In this review, we cover the recent exciting progress in structural immunology that has stemmed from the x-ray crystallographic analyses of TCRs, TCR/pMHC, and coreceptor complexes.

T CELL RECEPTOR VERSUS ANTIBODY A wealth of structural information has accumulated in the last 25 years about immunoglobulin structure and function (12, 30–32). Crystal structures of antibodies and their complexes with a variety of antigens (>100 structures) have appeared well in advance of TCRs because of the early availability of large amounts of antibodies from myelomas and, subsequently, from hybridoma and in vitro technologies (33). These many crystal structures and years of analysis have been necessary to obtain a general understanding of the structure and function of antibody molecules. This plethora of knowledge about antibodies can now provide the necessary framework to be able to derive a more rapid understanding of TCR structure/function relationships from a much smaller number of structures, since many of the same principles would presumably apply. In addition, this extensive literature on antibody structure should also prevent us from reaching premature conclusions about TCRs based on an insufficient TCR structural database. For example, some of the prevailing assertions that had to be modified after the determination of a larger body of antibody structures have included signaling through to elbow angle of the Fab, lock-and-key recognition of antibody–antigen complexes, and the role of conformational changes and water molecules in the antibody–antigen interface (12, 32, 34).

Expression and Crystallization of the TCR TCR structural studies have lagged far behind those of antibodies because of the enormous problems encountered in producing sufficient quantities of soluble TCR material (4). An extensive literature is now available on the production of recombinant soluble TCR and TCR domains (4). Soluble TCRs have a number of features that may contribute to the expression problems, such as inefficient αβ chain-pairing, when compared with the expression of light (L) and heavy (H) chains of antibodies. Attempts to solve this problem have included the

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

372

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

addition of Fab constant domains to the ends of the TCR chains (35, 36) or, more recently, by the addition of leucine zippers to the carboxy-terminal ends of the soluble α and β chains (37, 38). Alternatively, the pairing problem can be overcome by engineering a single-chain Fv TCR with a covalent linker that connects the Vα and Vβ chains (39–41). Another feature of the TCR that distinguishes it from an antibody Fab is its extensive glycosylation, with up to seven N-linked sites distributed among the α and β chains. On the one hand, glycosylation heterogeneity from eukaryotic expression systems can pose a problem for crystallization (42), but on the other, carbohydrates may help aid correct and stable folding in vivo. Expression from Drosophila melanogaster cells results in generally smaller and simpler highmannose glycosylation on glycoproteins (43) and has been used for the 2C TCR (44), as well as numerous other immunological proteins (45–50). An ingenious solution to the problem of complex glycosylation in mammalian cells has been used for the N15 TCR by expression in a CHO-derived cell line that does not attach complex carbohydrates (51). Many other efforts to produce TCR from Escherichia coli as single-chain Fv TCR constructs have resulted in partially folded and aggregated material (52, 53). Recently, full-length TCR α and β chains have been secreted from E. coli (38), as well as refolded from inclusion bodies and crystallized as complexes with pMHC (25, 28, 54). Certainly, E. coli expression would still be the preferred method from the standpoint of cost, but it has not yet worked for all TCRs. Thus, TCR expression is still an empirical exercise, and numerous strategies should be explored in parallel to increase the chance of success. The expression of crystallizable αβ heterodimer, both alone (44, 51) and in complex with pMHC (26, 28, 44, 54), has now finally been achieved by a few groups (Table 1). The difficulties in expressing the αβ heterodimer were evidenced from the initial expression and crystallization of an individual TCR β chain and Vα fragment (Table 1) (19, 20, 55, 56). The crystallization of the 2C TCR, compared with numerous other αβ pairs, was aided by its relatively high expression levels and efficient αβ chain pairing in Drosophila melanogaster cells. Future directions in TCR expression will probably see an increased number of single-chain Fv TCRs, such as was successful for one structure (Table 1, Figure 2d) (23, 40). The question remains, however, as to the role of the constant domains in transducing the signal of pMHC engagement, and hence their removal eliminates the possibility of providing that sort of information.

THE STRUCTURE OF THE αβ TCR The crystal structures of αβ TCR heterodimers, both alone, as full-length or FvTCR complexes with Fabs, and in complex with pMHC, have now been

14.3.d

1934.4

(24) 1nfd

(27) 2ckb

(25) 1ao7

1460 — 14.0◦ ˚ (2.7 A) 150◦ (28) 1bd2

1390 2200 4.7◦ ˚ (1.0 A) 150◦

(23) 1kb5

1660 — 5.8◦ ˚ (3.0 A) —

(19) 1bec

155◦

— — —

(20)

1500 — 6.0◦ ˚ (1.0 A) —

(75) 1tvd

—

— — —

(21) 1jck

155◦

Vβ8.2 Jβ2.1, Cβ1 (II) I-Ed HA murine myeloma ˚ 3.5 A SEC3,2 SuperAgs

14.3.d/SAg

b

“ref” and “secr” denote refolded and secreted expression systems, respectively. ˚ probe (138), and VαVβ pairing (deviation from pseudo two-fold) and elbow angles calculated as described (22). Buried surface areas calculated with MS using a 1.7 A Value in parentheses denotes the translational component of the superpositioning.

1390 2400 14.5◦ ˚ (1.9 A) 140◦

1300 2380 3.5◦ ˚ (0.8 A) 149◦

HLA-A2 — human a E. colisecr ˚ 1.9 A unliganded Vδ

Vδ3, Dδ2 Jδ1

ES204

Annual Reviews

c

KB5-C20

12:15

a

B7

Vα8, Jα19 Vα2.3, Jα24 Vα17.2, Jα54 Vα2.3, Jα10 Vα4.2, Jα4 (I) (II) (II) (II) (I) — Vβ5.2, Dβ2 Vβ12.3, Dβ2.1 Vβ12.3 Vβ2, Dβ2 Vβ8.2 Jβ2.4, Cβ2 Jβ2.7, Cβ2 Jβ2.7, Cβ2 Jβ2.3 Jβ2.1, Cβ1 (I) (II) (II) (III) (II) H-2Kb H-2Kb HLA-A2 HLA-A2 H-2Kb I-Ed I-Au dEV8 VSV Tax Tax — HA Acl-9 murine murine human human murine murine murine a a a E. coli ref myeloma myeloma E. coli secr Drosophila CHO-lec E. coli ref ˚ ˚ ˚ ˚ ˚ ˚ ˚ 3.0 A 2.8 A 2.6 A 2.5 A 2.6 A 1.8 A 2.2 A H-2Kb-dEV8 Fab-H57 HLA-A2-Tax HLA-A2-Tax Fab-Desir´e1 unliganded unliganded β chain homodimer

—

A6

QC: ARS/uks

˚ 2):b Buried Surface Area (A VαVβ 1300 CαCβ 2380 VαVβ — pairingc elbow 149◦ angle ref. (22) (PDB) 1tcr

Vα3, Jα58 (I) Vβ8.2, Dβ2 Jβ2.4, Cβ2 (I) H-2Kb, Ld dEV8, p2Ca murine Drosophila ˚ 2.5 A unliganded heterodimer

N15

P2: NBL/mbg/plb

(Kabat) MHC peptide species source resolut. complex

α chain (Kabat) β chain

2C bound

February 11, 1999

2C free

Table 1 Descriptions of TCR molecules whose crystal structures have been determined

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SAT/ary T1: KKK

AR078-13

T CELL RECOGNITION

373

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

374

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

Figure 1 Comparison of crystal structures of T cell receptors and an antiprotein Fab. The three TCR structures shown are (a) 2C, (c) N15 (from the N15-Fab H57 complex), and (d ) KB5-C20 single-chain Fv TCR. Carbohydrates are shown as ball-and-sticks. For purposes of comparison we show (b) the antitissue factor Fab 5G9 (140). Figure drawn with MOLSCRIPT (141) and Raster3D (142).

reported by several groups (Table 1). The first TCR αβ heterodimer crystal structure published was that of the murine TCR 2C (Figure 1a), which also simultaneously reported a low-resolution crystal structure of 2C in complex with its pMHC, H-2Kb-dEV8 (22). 2C, one of the first TCRs to be cloned, is one of the most extensively studied TCRs because of its alloreactivity (57–64). The recent Fab complexes of the full-length mouse N15 (Table 1, Figure 1c) (24), and KB5-C20 scFv TCR (Table 1, Figure 1d ) (23) have now given us the opportunity to compare different TCRs directed as the same MHC, H-2Kb, but

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL RECOGNITION

375

Figure 2 Comparison of TCR structures after superpositioning of their Vβ domains (a–c), and Cα domains (d ). (a) The elbow angles of the full-length TCR structures are relatively constant, as seen from this sideview along the axis of rotation between the V and C domains. (b) The quaternary arrangements of the Vα, Vβ, Cα, and Cβ domains vary somewhat, although the individual domains superimpose closely. (c) A view onto the top of the TCR V domains, superimposed on their Vβ domains, clearly shows the wide range of Vα/Vβ pairing angles seen so far. (d ) The three Cα domain structures determined (2C, N15, and B7) are superimposed on the back β-sheet, showing the noncanonical structures of the top strands in all three structures. Figure prepared with InsightII (MSI, San Diego, Calif).

with different peptide antigens. The first human full-length TCRs, A6 and B7, were determined as TCR/pMHC complexes with the same pMHC, HLA-A2Tax (Table 1, Figure 4b,c, see color plates) (25, 28). When considering the overall architecture of the heterodimer, the TCR structure (Figure 1a,c) appears to be generally similar to that of a Fab fragment (Figure 1b). The two variable (V) domains and two (C) domains are similarly situated with the association of the TCR α and β chains being reminiscent

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

376

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

of the L and H chain pairing of Fabs, consistent with earlier predictions (65– 67). However, many differences are apparent upon closer inspection of the ˚ vs ∼46 A) ˚ structures. The TCR is wider across the middle than an Fab (∼56 A because of the protrusion of a loop in the Cβ domain that appears to be a general feature of all β chains (19) (Figure 2a,b). The TCR is also more asymmetric and squat than an Fab because of the more parallel crossing angle of the β-sheets ˚ shift off-center in the position into the Cα/Cβ interface, and the roughly 5 A of the pseudo-2-fold relating Cα to Cβ. This asymmetry is accentuated by the smaller size of the Cα domain as compared with the Cβ (Figure 1a,c).

The Constant Domains The structures of the individual Cα and Cβ domains (Figures 1 & 2) and their orientations with respect to each other have been remarkably consistent for ˚ rmsd for all the full-length TCRs determined so far (approximately 0.6–0.8 A ˚ the Cβ’s, and 0.8–1.1 A rmsd for Cα). Another distinguishing feature of the TCR constant domains as compared with Fabs is the large buried surface area ˚ 2) and highly polar nature of the Cα/Cβ interface, which differs from (∼2300 A ˚ 2) and generally more hydrophobic CL/CH1 interface. The the smaller (∼1700 A distribution of charges in the Cα/Cβ interface is markedly skewed toward being more acidic for Cα and more basic for Cβ. This large polar interface is reminiscent of the CH3–CH3 homodimer interaction in an Ig module. The Cβ domains ˚ rmsd for themselves are more structurally similar to a CH3 domain (∼1.6 A ˚ 89 residues) than to antibody CH1 constant domains (∼2.0–2.2 A rmsd). Perhaps the most interesting feature of the Cβ domains is the large 14-residue FG loop, which extends out to the side of both the mouse and human Cβ do˚ rmsd) in all of the mains and adopts roughly similar conformations (0.9–1.4 A structures, although its actual position varies. This loop appears relatively rigid, owing to a small hydrophobic core, but its exact function has yet to be defined, although it has been proposed to facilitate TCR interaction with coreceptors and the CD3 chains (68). Indeed, mutagenesis and modeling studies, in conjunction with the N15 TCR structure determination, have proposed that a cavity underneath the Cβ FG loop acts as a CD3ε docking site (68). The Cβ/Vβ interaction surface is very extensive and likely contributes to the lack of variation in the V-to-C disposition of the TCR structures seen so far (19). Even though structurally analogous residues to the Fab ball-and-socket joint (67) exist in the TCR (22), the “elbow angle” (69) of the TCR has been fairly constant at around 140–150◦ , compared with the analogous value in antibodies, which spans around 120–225◦ (Figure 2a) (32). The Cα domain structure was the most unusual feature of the first fulllength heterodimer (2C) determined (Figure 1a). Since then, two other Cα domains (N15 and B7) have clearly confirmed its deviation from an Ig fold.

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL RECOGNITION

377

˚ rmsd The superpositions of Cα from these different TCRs range from 0.6 A ˚ for the top strandlike elements G, F, for the ABED back β-sheet, to 1.2–1.8 A and C (Figure 2d). From a crystallographic standpoint, the Cα domains have presented the biggest technical challenges, as they are somewhat flexible, and their electron densities have not been as well defined as the other more rigid domains; all of the TCRs show higher thermal factors (B values) for the Cα domain. It was anticipated that the Cα domains would have some structural differences from canonical Ig folds because of low sequence identity (∼15%) compared with Ig C domains, a characteristic shared by γ δ TCR Cγ domains (1). Although the Cα does contain the two conserved cysteine residues that covalently link the top strands (G, F, and C) with the bottom β-sheet (A, B, E, and D), only 50 amino acids connect the two cysteines, opposed to the usual 65 residues in Ig “C type” folds (Figure 2d ). The structural results of this deletion are that the Cα does not have a C0 and a C00 strand, and also has a very short FG loop which is missing a conserved Trp residue normally found in other Ig folds (see Figure 3 of Ref. 22). Additionally, the F strand forms an unusual one-turn minihelix. The net result is that the top strands do not hydrogenbond (H-bond) into a continuous sheet and may, therefore, be somewhat pliable (Figure 2d), possibly to facilitate an interaction with the CD3 signaling components. Two of the TCR structures (N15 and 2C) contain carbohydrate moieties (Figure 1a,c). In N15, the sugars have been trimmed by Endo-H to a single GlcNAc moiety, but clear electron density is seen for all seven potential N-linked glycosylation sites, with five of them on the C domains (24, 51). In 2C, the sugars are untrimmed, and four sites (three on the C domains, and one on Vα) had interpretable electron density for several sugar residues (22). In particular, an N-linked sugar, attached to Asn185 of the Cα, forms H-bonds with the neighboring Cβ domain and may play a role in stabilizing the Cα/Cβ interface (especially in soluble TCR), as seen, for example, as a role for carbohydrate in the Fc structure (70). The human B7 TCR does not contain any carbohydrates, owing to its expression in E. coli, but nevertheless, in this case, has a structured Cα domain in the crystal, although again with higher B values (28).

The Variable Domains The V domains of the TCRs are also highly similar to antibody V domains (Figure 1b), but with numerous small, but important, differences. The quaternary arrangement observed in these relatively few examples of Vα/Vβ interfaces surprisingly spans a wide range, similar to those in the extensive database of Fab VL/VH pairings (Figure 1b,c) (12, 71, 72). So far, the relative Vα/Vβ rotational dispositions between TCRs vary from 3 to 14◦ between TCRs (Table 1) (compared with up to 16◦ for Fabs) (Figure 2c), and the buried surface area

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

378

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

˚ 2 (2C) to 1650 A ˚ 2 (KB5-C20), which between Vα and Vβ ranges from 1300 A 2 ˚ –1700 A ˚ 2). On the other correlates well with similar values for Fabs (1000 A hand, the differences in relative pairing between the V and C modules in TCR structures are insufficient to indicate a signaling function through the elbow region, as already concluded for antibodies (73). However, the surprising range in the chain pairing angle of the Vα and Vβ domains clearly has an impact on its specificity as differences in the relative orientation of the Vα and Vβ chains will result in interactions with different parts of the pMHC surface. The Vα/Vβ interface is composed of a core of conserved residues (22) that are similar to those found in VL/VH interfaces (74). In the TCR, the hydrophobic core is surrounded at the periphery and base of the β-strands by conserved H-bonds, especially between Gln37α and Gln37β (22). The Vα/Vβ interface appears more 2-fold symmetric than its VL/VH counterpart owing to the use of identical side chains at similar positions in each chain. This 2-fold symmetry could explain the tendency of TCR α and β chains to homodimerize readily in solution. The V domains themselves contain approximately 24 residues in the core of the β-sheet sandwich that are conserved in mouse and human TCRs, as well as in Ig folds (66, 67). Hence, the TCR V domains superimpose closely on Ig V domains, with the conserved residues performing similar structural roles in the TCR and antibody. The Vα and Vβ domains do not segregate clearly as to which is structurally more similar to VH or VL. Superpositions of Vα onto VH ˚ rmsd for and onto VL indicate a generally better fit for Vα onto VL (∼1.6 A ˚ rmsd for up to up to 70 residues), but for Vβ, the fit with VL or VH (∼2.0 A 70 residues) is roughly equivalent. Strand switches involving the C00 strand in Vα (20) for all TCR structures (Figure 3a), and for one Vβ (23) have not been seen in antibodies. The result of the strand switches, particularly for C00 in Vα, is that the path of CDR2 shifts by ∼90◦ and creates a more narrow, compact TCR-combining site (Figure 3a), which would be favorable for docking onto the long, narrow groove of the pMHC. The strand switch could be another mechanism of altering specificity of the TCR for an MHC antigen. The recent structure of a Vδ domain, the only example of a γ δ TCR fragment, more closely resembles an antibody VH domain than either a Vα or Vβ, and does not contain a C00 strand switch, instead having the antibody-like C0 –C00 interaction (75).

The TCR Combining Site The “combining site” of TCRs is made up of hypervariable loops, or complementarity-determining regions (CDRs) (30) 1, 2, and 3 from the α and β chains, and also another loop on the β chain, termed HV4, which exhibits some hypervariability and is within the binding site for some superantigens (Figure 1, 3a) (21, 76). Some variability exists in the corresponding Vα α4 loop, but is

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL RECOGNITION

379

Figure 3 Conformations and positions of the CDR loops of the TCR structures. (a) The individual Vα and Vβ domains of each TCR crystal structure are individually aligned on the 2C Vα and Vβ structures, allowing the relative positions of the CDRs (in black) to be clearly seen. The range of pairing between Vα and Vβ is not indicated in this figure (see Figure 2c). (b) The individual CDRs are superimposed: CDR1α (25-31, the two canonical subclasses are shaded differently), CDR2α (48-55), CDR1β (25-31, the conserved His29 is drawn), CDR2β (48-55), and HV4 (69-75). Figure prepared with InsightII (MSI, San Diego, Calif).

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

380

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

less pronounced (1). The combining sites of the TCRs determined so far are relatively flat (4), similar to antiprotein antibodies (77), and consistent with the TCR’s function of binding to the generally flat, undulating surface of the pMHC (78). The A6, B7, and KB5 TCRs are somewhat different from 2C in that their CDR3β’s are much longer and protrude from the center of the combining site (Figure 3a); most of the TCRs have a cleft between the two CDR3s, presumably to accommodate a central upfacing peptide side chain from the pMHC. It was speculated, after the cloning of the first few TCR chains, that the much higher sequence diversity seen for the CDR3s implied a concentration of the peptide discrimination function to the CDR3s, with CDRs 1 and 2 interacting with more conserved structural elements of the MHC (1, 79). In fact, the TCR genes contain many more J regions than antibodies for the purpose of increasing V-J(α) and V-D-J(β) junctional diversity in the CDR3s (1).

The Complementarity-Determining Region (CDR) Loops The positions of the CDRs within the TCR binding site (Figure 3a) are all similar to antibodies (4) except for CDR2, which undergoes the strand switch in Vα (20) and occasionally in Vβ (23). The limited sequence diversity of TCR CDRs 1 and 2, as well as fewer V genes as compared with antibodies (1), suggests that enough representative structures may be available to begin an assessment of whether TCRs have “canonical” CDR structures, as seen for antibodies (80). The large number of Fab crystal structures determined so far has resulted in the clear delineation of structural subfamilies for CDRs 1 to 3 from the L chain and CDRs 1 and 2 of the H chain; the conformation of the individual CDRs, except H3, can be reasonably well predicted based on their primary sequence (81). CDR1β is highly similar in four out of the six Vβ structures in Kabat sub˚ rmsd) (Figure 3b). The outliers N15 (1.7 A ˚ rmsd) and KB5 group II (0.2–0.5 A ˚ (1.8 A rmsd) (Kabat subgroups I and III, respectively) contain a proline instead of the relatively conserved, loop-stabilizing glutamine at position 25. Gln and His at positions 25 and 29 (position 30 in KB5 due to an insertion) are relatively conserved in many mouse and human Vβ CDR1s, and serve the same structural role of stabilizing the center of the turn. Hence, these CDR1β structures probably represent canonical structures for most Vβ chains (82). CDR2β is constrained to be a β turn in all of the structures so far (similar to CDR-L2 in antibodies), and the superposition for all members of Kabat subgroup II is ˚ rmsd for backbone atoms (Figure 3b). Most of very close with a 0.2–0.7 A the CDR2β’s are stabilized by an H-bond between Ser49 and Arg69 of HV4. ˚ rmsd, respectively), which Again, the outliers are N15 and KB5 (1.2 and 1.6 A represent different Kabat subgroups; also, KB5 undergoes a C00 strand switch, where it deviates substantially from the others. The four closely superposing

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL RECOGNITION

381

structures all contain a stretch of Gly-small side chain-Gly residues from positions 51-53 at the extreme tip of the turn, a feature likely to be common across many Vβ family members (82). Five of the six HV4 structures superimpose ˚ rmsd (Figure 3b), indicating close strucwith rmsd ranging from 0.2 to 0.7 A tural similarity, which is aided by the intraloop-stabilizing effect of Arg69β. ˚ which has a generally similar structure but The outlier is N15 (rmsd of 1.7 A), lacks Arg69; in this case, the extreme tip of the loop folds away from the center of the combining site. All of the TCR CDR1α conformations are stabilized by a conserved hydrophobic interaction between framework positions 24 and 32 (highly conserved as Tyr24 and Leu or Phe32). However, the conformation of CDR1α clearly segregates into subfamilies according to Kabat subgroups (Figure 3b), with the subgroup I members (2C, N15, and 1934.4) showing one conformation and subgroup II (A6, B7, and KB5) another. The predominant driving force for the subgroup II conformation appears to be a conserved hydrophobic core in the center of the turn composed of positions 24, 29, and 32, which cause the center of the loop to be pulled in. In subgroup I, the hydrophobic core contains smaller side chains at position 30. In most CDR1α’s of mouse and human, position 27 is a Ser or Thr, and positions 24 and 32 have hydrophobic character, strongly implying that this loop-stabilizing core is a general theme, and that all CDR1α structures may have similar sets of canonical conformations. CDR2α is highly similar to the corresponding CDR2β in that its short sequence requires it to make a tight turn (Figure 3b). The subgroup I TCRs (Table 1) superimpose very closely for residues in the first half of the turn ˚ rmsd), while B7 and KB5 deviate. Substantial vari(residues 48–52, ∼0.5 A ation in the backbone for the latter half of CDR2α is presumably due to the different degrees of strand switch in each structure (Figure 3b), and the high solvent exposure of this strand. The α4 loop of TCR Vα domains has some sequence hypervariability, although to a lesser extent than the other CDRs, and so is not classified as a bona fide CDR loop. However, its proximity to the TCR combining site suggests the possibility that it could interact with a pMHC ligand, as seen in A6/HLA-A2/Tax (25). All of the α4 loops superimpose very ˚ rmsd) in both their conformations as tight turns and positions closely (∼1.0 A within the binding site (Figure 3a), indicating these conformations are likely to be conserved in other TCRs. The CDR3α and CDR3β have substantially different conformations in each TCR structure (Figure 3a), a reflection of their varied lengths (6 to 12 residues) and sequence. This observation is interesting in light of the fact that canonical structures of CDR-L3 have been found in immunoglobulins (81). The lack of a canonical CDR3α would be consistent with the increased genetic diversity of the CDR3α relative to CDR-L3 (83). In the TCR, in addition to many

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

382

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

more J regions, an increased diversity in CDR3α results from an increased frequency of N and P additions at the Vα–Jα junctions, as well in the Vβ–Dβ–Jβ junctions (1). Therefore, in contrast to antibodies, where the major diversity is concentrated in the CDR-H3 owing to substantial length and sequence variation and noncoded additions, the TCR distributes this diversity over both CDR3s, as reflected in their varied structures. The CDR loops of the ES204 Vδ domain structure display features of both antibody and TCR CDRs. The conformation and position of CDR2δ is highly ˚ rmsd), partly a consequence of similar to the CDR-H2 of antibodies (∼1.0 A 0 00 the antibody-like C –C interaction. CDR1δ, however, most closely resem˚ rmsd) and is stabilized by similar intraloop bles the TCR CDR1α (∼1.0–1.9 A hydrophobic interactions between residues 24 and 33 analogous to the hydrophobic core of CDR1α. The conservation of the loop-stabilizing residues in CDR1δ suggests that this conformation will be canonical. However, more Vδ structures, as well as Vγ and γ δ heterodimers, are needed before firm generalizations can be made about γ δ TCRs. The overall relative similarities, albeit to different extents, of the CDR1 and 2 loops from both α and β chains, combined with their conserved short lengths, have ramifications for the recognition of the pMHC ligands. The tips of these loops, which are mostly contact residues for the pMHC, as seen in the TCR/pMHC complex crystal structures (Figure 4, see color plate), correspond to the same sequence positions, although different amino acids can occupy these positions in each TCR. Thus, the TCR residues that contact the pMHC will almost always include the residues at the apices of the CDRs, and hence define a rather small subset of the total CDR residues that should greatly simplify mutational and modeling studies. The tips of the CDRs in all of the structures analyzed to date comprise the following residues: CDR1α, residues 27–30; CDR2α, residues 50–52; CDR1β, residues 27–30; CDR2β, residues 52 and 53; and HV4, residues 72 and 73.

TCR COMPLEXES WITH PEPTIDE-MHC The holy grail of the TCR structural studies has been to elucidate the mechanism of MHC restriction. The main question is how the TCR binds to, and discriminates among, the tremendous diversity of highly similar peptide-MHC surfaces. One key to the answer has come from the myriad of crystal structures of peptide-MHC complexes that have now been elucidated to a very detailed atomic level (46, 84–93). We now know, as a result of structural, biochemical, and genetic studies, how MHC molecules of both classes bind antigenic peptides and display them to the T cell receptor (78, 94–98). However, a key advance has been the crystallization and structure determination of a complex

March 4, 1999

P2: NBL/plb

12:42

QC: NBL/tkj

Annual Reviews

Figure 4 Crystal structures of three TCR/pMHC complexes. The bound peptide is drawn in red with side chains shown as ball-and-stick. The CDR loops are colored: α1, blue; α2, purple; α3, green; β1, pink; β2, cyan; β3, yellow; HV4, orange. Figure drawn using MIDAS 2.0 (143). The PDB codes for the coordinates used are: 2ckb (2C/H-2Kb-dEV8, (27), 1ao7 (A6/HLA-A2-Tax, (25)), and for B7-HLA-A2/Tax (DC Wiley, personal communication (28)).

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SAT/VKS T1: NBL

AR078-01

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL RECOGNITION

383

between the TCR and pMHC. Definition of the functional topology of this complex has been the subject of numerous elegant biological experiments over the last 10 years (99–106). The earliest standard was set forth by the classic paper of Jorgensen et al (100) in which variant peptide immunization verified the paradigm of the CDR3s’ reading out the central region of the bound peptide, while the less diverse CDRs 1 and 2 interacted with more conserved helical elements of the MHC molecule (100). Other, more recent experiments have further examined this general topology. For example, Sun et al derived a diagonal orientation from an extensive mutational analysis of H-2Kb (102), and Sant’ Angelo et al (104) predicted the Vα chain would lie over the N terminus of the peptide and the Vβ would lie over the C-terminal half of the peptide (104). Reaching the goal of a TCR/pMHC complex crystal has required advances in multiple fields such as recombinant protein expression, x-ray crystallography, and macromolecular characterization (i.e. BIAcore) (16). The crystal structure determinations of the TCR/pMHC complexes have been challenging because of the moderate resolution of the x-ray diffraction data and the ensuing crystallographic problems associated with such a large and elongated macromolecular complex. So far, the only high-resolution TCR/pMHC complex crystal structures are for MHC class I complexes (Figure 4a–c, Table 2): 2C/H-2Kb-dEV8 ˚ (27), A6-HLA-A2-Tax (2.6 A) ˚ (25), and B7-HLA-A2-Tax (2.5 A) ˚ (28), (3.0 A) ˚ with one low-resolution (∼6.0 A) molecular replacement model of N15/H-2KbVSV also being reported recently (26).

2C/H-2Kb-dEV8 The first murine complex structure reported was that of the 2C TCR with H-2Kb and a mouse self-peptide dEV8 (22), which has subsequently been refined to ˚ resolution (Figure 4a) and compared with the unbound TCR (see Figure 4 3A of Ref. 27). 2C is an alloreactive TCR in which self- and foreign pMHC ligands have been defined (62, 64). 2C is positively selected in the presence of H-2Kb and negatively selected in the presence of H-2Kbm3 and H-2Ld (27, 59–61, 107). The dEV8 peptide is probably one of a degenerate set of similar peptides that are able to stimulate differentiation of 2C T cells, and is, therefore, an excellent peptide for an example of a positively selecting self-peptide complex (108). Alloligands are also known for H-2Ld (QL9 or p2Ca peptides) and the mutant H-2Kbm3 (dEV8), which converts the positively selecting H-2Kb to an allo-MHC (62, 107). The structure of the 2C/H-2Kb-dEV8 complex revealed that the TCR was oriented in an approximate diagonal orientation over the pMHC composite surface, with the α chain over the N-terminal half of the peptide and the β chain over the C-terminal half (Figures 4a & 5a). The CDR3α and β both lie over the central positions (P4 to P6) of the peptide, but the CDR3β, which has

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

384

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Table 2 Descriptions of TCR/pMHC complexes whose crystal structures have been determined 2C/H-2Kb-dEV8

A6/HLA-A2-Tax

B7/HLA-A2-Tax

Buried surface area:a ˚ 2) TCR/pMHC (A ˚ 2) pMHC (A Heavy chain (%) Peptide (%) ˚ 2) TCR (A ˚ 2) Vα (A CDR1α (%) CDR2α (%) CDR3α (%) ˚ 2) Vβ (A CDR1β (%) CDR2β (%) CDR3β (%) HV4 (%) SC

1876 966 75 25 910 480 24 13 15 430 18 16 10 1 0.46

1810 910 66 33 900 576 24 10 24 324 2 1 33 0 0.63

1710 900 66 33 810 551 28 13 22 260 2 10 21 0 0.64

Total contacts: Vα CDR1α CDR2α CDR3α α4 Vβ CDR1β CDR2β CDR3β HV4 MHC Peptide References PDB code

41 23 10 5 8 0 18 10 6 2 0 27 14 (27) 2ckb

46 30 10 4 13 2 16 1 0 15 0 27 19 (25) 1a07

63 49 15 10 24 0 14 0 3 11 0 34 29 (28) 1bd2

a

˚ probe (138). Buried surface areas calculated with MS using a 1.7 A

an unusually high number of glycines, has surprisingly minimal contact with the peptide, perhaps explaining the alloreactive nature of this TCR (Table 2). The CDR1s of both chains straddle the center of the N- and C-terminal ends of the peptide binding groove and contact both peptide and MHC, while the CDR2s have almost exclusive contact with the α-helices of the MHC heavy ˚ 2), the chain. Although the buried surface in the interface is large (∼1800 A surface complementarity (SC) between the TCR and MHC is very poor (Table 2) (27, 109), with large gaps present in the interface, which is consistent with the weak affinity of 2C for H-2Kb-dEV8 (KD ∼ 10−5 M). Most of the buried

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL RECOGNITION

385

surface (75%) and TCR contacts are with the helices of the MHC, with the P1, P2, P4, and P6 and P7 positions of the peptide extending up from the groove to form direct or indirect H-bonds and van der Waal interactions with the TCR (Table 2). The peptide contacts are relatively precarious in that the side chains must be completely extended to reach the TCR, in some cases, as for P1, requiring bridging waters (see Figure 3 of Ref. 27). The chemical nature of the interface is relatively uncharged, with substantial van der Waal contact between the backbone atoms of the TCR and MHC. Two positions on the TCRα chain (27α and 51α) that have been shown to play a role in class I vs II restriction (110) are in contact with conserved MHC residues (Figure 5c, see color plate). The structure of 2C/H-2Kb-dEV8 was extremely informative about various aspects of T cell recognition. The most obvious is the dominance of the MHC helices in the recognition of pMHC, which has been supported by a recent alanine scan of 2C against its alloligand, H-2Ld-QL9 (111). From this alanine scan, the interaction surface was found to be relatively flat in energetics in that most of the TCR contact residues contribute roughly equal energies, with the total interaction energy being distributed over the entire surface area (111). These data support a “scanning” model in which there exists enough binding energy between the MHC heavy chain and the TCR to enable the TCR to dock on the pMHC and read out the peptide contents (14, 112). Those peptides that contribute a sufficient number of energetically favorable contacts with the TCR can then provide the slight amount of additional kinetic stabilization required for signaling to occur (112). Another important result from the 2C complex structure is that specificity can be achieved even in the absence of extensive CDR3β contact, and that the other CDRs can contribute much of the peptide specificity. This feature may not be general because of the small glycine-rich CDR3β found in 2C, as compared with longer CDR3β loops in most other TCRs. The 2C TCR complex has also allowed the first comparison of a bound and unbound TCR. Large conformational changes are induced upon binding of the TCR to the pMHC in CDRs 1α and 3α, and a minor change in CDR3β, all peptide-contacting CDRs; these changes are highly significant compared with those seen to date for binding of antiprotein antibodies to their antigens (71), but are similar to the large conformational changes observed for antihapten antibodies. These structural accommodations are important when one considers that TCRs must be able to respond to numerous different peptide ligands bound to an MHC, including degenerate sets of positively selecting peptides (113– 115) for which dEV8 may be representative for 2C. Hence, a “plasticity” exists in the TCR’s biological response to and recognition of pMHC ligands, and so a certain flexibility of the combining site to adapt to multiple ligands would be a mechanism of expanding the recognition repertoire (116, 117).

P1: SAT/VKS

12:42

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

March 4, 1999

P2: NBL/plb

Figure 5 Overlaps and footprints of the interface regions of the three TCR/pMHC complexes. The α 1 and α 2 domains of H-2Kb-dEV8 and HLA-A2-Tax have been superimposed for the three complexes, in order to see where their respective TCRs interact. The CDR loops are colored: 2C, blue; A6, brown; B7, green. For the purposes of clarity, only the CDR loops of the TCRs are drawn. a) View looking down in the pMHC surface. b) Looking down the pMHC helices with the Va CDR loops in front. c)Side view clearly showing the difference in b chain CDR contacts between the complexes. [Figure prepared with InsightII (MSI, San Diego, CA).]

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

386

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

In the 2C system, a model of 2C in complex with its much more reactive and higher affinity alloligand H-2Ld-p2Ca has been constructed based on the high conservation of the helical TCR-contact residues on the MHC between H-2Kb and H-2Ld (93). In the 2C/H-2Ld-peptide complex, a pronounced bulge in the C-terminus of the peptide, not present in H-2Kb-bound peptides, presents an acidic residue (Asp-P8) to the electropositive 2C β-chain, close to the HV4 residue Arg69, as well as substantially increasing the number of TCR-pMHC contacts. In the 2C/H-2Kb-dEV8 complex, the negative charge in this area arises from the MHC helix (Asp77) and is much farther away from Arg69, possibly explaining the reduced reactivity and affinity for the H-2Kb-dEV8 complex. Thus, an explanation for the alloreactivity in this system seems to reside in a form of molecular mimicry in which the negative charge on the Kb helix (Asp77), whose mutation renders Kb into an alloligand, is mimicked in Ld by a different structural element (i.e. the peptide).

A6/HLA-A2-Tax The A6 TCR in complex with HLA-A2 and the HTLV-derived Tax peptide was the first human TCR complex reported (Figure 4b) (25), and although the overall orientation is similar, it is strikingly different from the 2C complex in many important ways (Table 2). The A6 TCR is derived from a patient with HTLV-1 in which an extraordinarily high proportion of circulating CTLs are directed against the Tax peptide, thus providing a highly antigenic pMHC complex to contrast with the partial agonist self-peptide dEV8 with 2C. In the A6 complex, the TCR also has a diagonal orientation over the pMHC (Figure 5), and the authors speculate that this is due to an intercalation of the TCR between the two “peaks” of the MHC α1 and α2 helices (25, 102). The TCR contacts with the pMHC are from both similar and different positions on the TCR and MHC. The α chain contacts the MHC helices in much the same manner as 2C, with the tips of CDRs 1 and 2 in essentially identical locations (Figure 4b), further buttressing the argument that the Vα is primarily driving the orientation. The β chain, however, tilts off the pMHC surface (Figure 4b & 5c) such that only the long CDR3β has intimate and extensive contact with peptide and MHC, while CDRs 1β, 2β, and HV4 are elevated above the surface. The Vα/Vβ pairing of A6 is somewhat unusual when compared with the other TCRs (and antibodies) in the manner in which Vβ is twisted away from Vα. This pairing difference (Table 1) results in the footprint of the A6 Vβ chain being further rotated toward and tilted away from the α1 helix of the MHC compared to 2C. Extensive contact between A6 and the Tax peptide (Table 2) is in large part due to the much more exposed, elevated central region of the peptide in the HLA-A2 structure, which is accentuated by A6 pulling the peptide even farther out of the groove compared with its unliganded structure (86). The central P5-Tyr sits in a

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL RECOGNITION

387

pocket between the two CDR3s, with significant contact between the side chains and backbone of each upfacing peptide residue (P1, P2, P4, P6, P7, and P8). ˚ 2) and side-chain As for the 2C complex, the interface surface area (∼1800 A contacts are dominated by the α helices of the pMHC (Table 2). The shape complementarity of A6/HLA-A2-Tax is significantly better than for 2C/H-2KbdEV8, consistent with the greater antigenicity of the human complex; however, the shape complementarity of both complexes is still somewhat poor compared with antibody-protein interfaces (109). It is interesting to speculate that the larger number of TCR-peptide contacts contributes to the greater antigenicity of the Tax peptide.

B7/HLA-A2-Tax The structural work in the HLA-A2-Tax system has now been extended by the recent determination of the structure of another human TCR, B7, in complex with the same pMHC (Figure 4c) (28). B7 retains the same Vβ, but has a different Vα (Table 1). Again, this complex was refolded from E. coli inclusion bodies, and although it does not contain the interchain disulfide, all of the domains are ordered in the electron density. B7 has a footprint similar to A6 on the pMHC (Figure 5), although the position of the Vβ is slightly different owing to a Vα/Vβ pairing difference from A6; the B7 Vα/Vβ pairing is more similar to the 2C TCR (Table 1). The most interesting result from the B7 complex is that although 13 TCR and 18 MHC contact positions are identical between the two complexes, the amino acids used by the TCRs are completely different, providing a clear explanation of how one pMHC can be recognized by multiple TCRs (28). Seven identical atomic contacts are maintained between the two complexes, out of approximately 17 (B7) to 20 (A6) of the total TCR contacts. The electrostatic potentials of the two different TCR-binding surfaces are quite different, with B7 providing a negatively charged P5 binding pocket and A6 a positively charged one. The differential effects of partial agonist peptides for A6 and B7 are rationalized by probable loss of contacts or steric clashes caused by various peptide amino acid substitutions in the two interfaces. The TCR footprint is again dominated by Vα (Table 2), which has a substantially different sequence in B7 and A6, and the β chain CDR1 and HV4 are raised off the pMHC surface, presumably pushed off by the long CDR3β, as seen in A6. The tips of the CDRs 1α and 2α are in very similar locations, as in both the 2C and A6 complexes, reinforcing the notion that these CDR positions play a fundamental role in driving the TCR orientation on the pMHC. Indeed, the only absolutely conserved contact in the B7 complex is between Ser51α and the HLA residue Ala158, which is also a conserved contact in 2C/H-2Kb. ˚ molecular replacement solution of the murine The crystallization and 6 A N15 TCR in complex with H-2Kb-VSV has recently been reported (26) and

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

388

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

is consistent with extensive mutagenesis data on the interaction of this TCR with the pMHC (105). The low resolution allows us to discuss this complex only in generalities, but it appears similar in orientation and footprint to the 2C/H-2Kb-dEV8 complex, although variations in such parameters as “twist,” “tilt angle,” and “shift” may result in subtly different TCR/pMHC interfaces (26). The β chain of N15 appears to contact the pMHC in a way similar to 2C in that the Vβ sits relatively flat on the pMHC.

GENERAL PRINCIPLES FOR TCR RECOGNITION OF PEPTIDE-MHC We briefly encapsulate in this section some of the general conclusions that we can draw from the TCR/pMHC crystal structures determined to date. 1. The overall diagonal footprint of the TCR over the pMHC appears to be general and conserved, but the actual positions of the TCR domains and distribution of MHC vs peptide contacts vary widely (Figure 5-see color plate, Table 2). While the footprints of the Vα domains are in almost identical positions, the Vβ domain footprints differ substantially (Figure 5a–c). For both Vα and Vβ, the positions and incidence angle of the TCR with respect of the pMHC are unique in each of the structures. Nevertheless, the roughly diagonal overall footprint will probably be found in all class I TCR complexes, as well as in class II TCRs. This diagonal orientation observed in the crystal structures is consistent with mutational and immunological analyses from many different class I and II systems. From the structures themselves, the interdigitation of the long axis of the TCR-combining site between the high points of the grooves of the pMHC does appear to maximize the interaction between the two surfaces. Arguments for a unique orientation can also be made from inferences regarding coreceptor interaction. The CD8 and CD4 coreceptors are nonpolymorphic, and, therefore, almost certainly interact with the MHC and TCR in a conserved fashion. Consequently, the topology of the TCR/pMHC docking must also be conserved to accommodate the identical positioning of the coreceptor. 2. The MHC helices dominate the TCR/pMHC interface (Table 2), facilitating scanning of the bound peptide by the TCR. This idea had its foundation in early experiments in positive selection, which suggested that the TCRbinding energy must be primarily directed against the most conserved features of the MHC (118, 119). The conserved helical framework enhances the TCR’s ability to sample many different bound peptides by providing a conserved structural scaffold.

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

T CELL RECOGNITION

389

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

3. Even though the bound peptide conformations vary considerably between class I and class II pMHC complexes, the upfacing side chains are likely to engage in similar interactions with their respective TCRs. Indeed, class II peptides have a highly conserved polyproline type II conformation that severely restricts their conformation in the MHC-binding groove (78, 96, 98). However, because of their increased length over class I peptides, peptide residues outside the central 9-mer, such as P10 and P11, are also in a position to interact with the TCR. 4. Poor shape complementarity between the peptide and the TCR facilitates the TCR’s ability to adapt to different bound ligands with different biological outcomes, as seen with agonist, antagonists, and altered peptide ligands (17). Additionally, the poor shape complementarity can explain the short TCR/pMHC half-lives, which are a critical variable in TCR signaling (120). Perhaps more highly antigenic peptides have greater contact with the TCR, increasing their half-lives, as for A6 and B7, which appear to have a higher affinity (28). 5. The Vα appears to be the predominant driving force for steering the TCR/ pMHC orientation. The crystal structures of all the complexes so far have almost identical locations for the Vα on the pMHC, with extraordinarily close superpositions of the tips of CDR1α and CDR2α loops (Figure 5b), which have been implicated as critical for the class I vs II restriction of a TCR (110). 6. Conformational changes in the CDR loops are an important mechanism of expanding TCR specificity. The inherent ability of the TCR to recognize multiple peptide ligands, in the face of its restricted genetic diversity relative to antibodies, may require it to harness conformational changes as a means of expanding its repertoire (116, 117, 121). 7. No large-scale conformational changes are obvious in the complex structures that might have an impact on signal transduction. In all of the complexes, the rather large relative movements of the α3 and β 2 m domains of the MHC are also seen in unliganded MHC structures; thus, the biological relevance of these movements in the complexes is uncertain at present. However, in 2C-H-2Kb-dEV8, there is not only a segmental movement of the α3/β 2 m, but also a large change in the α3/β 2 m domain pairing between the free and the bound forms (27). Such conformational changes could certainly play a critical role in signaling, but at present there is insufficient evidence to reach any such conclusion.

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

390

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

8. A surprising range of Vα/Vβ chain pairings has already been seen in the handful of TCR structures determined to date (Figure 2c, Table 1). The range of chain pairings could be an additional mechanism of expanding the recognition repertoire of the TCR, especially considering that the highly variable junctional residues at V-J and V-D-J are involved in the Vα/Vβ interface and can strongly influence the pairing of these domains. 9. No evidence of higher order oligomerization is present in TCR/pMHC crystals of the complexes. From biological experiments, multimerization and clustering appear to be integral aspects of TCR signaling (122, 123); however, the complex structures are all monomeric. Earlier models for dimerization of MHC and TCR have not been supported by crystal structures of the complexes (20, 87, 124). This result is somewhat surprising in light of recent data that higher order TCR/MHC assemblies can form, and are stable, at solution concentrations and conditions approaching those used for crystallization (125).

CORECEPTORS IN THE TCR COMPLEX The TCR does not bind to the pMHC in isolation but as part of a complex with coreceptors CD8 (αα or αβ) or CD4 and CD3. These coreceptors are monomorphic and do not influence the specificity of the TCR recognition of pMHC, but they do dramatically affect the qualitative nature of that interaction. CD8 and CD4 augment the TCR/pMHC interaction, both in vitro and in vivo (18, 49, 126, 127). How this is accomplished has been a difficult structural question because it is not firmly established whether CD8 or CD4 bind to the TCR, pMHC, or both in the signaling complex. While it has been generally suspected that CD4 simultaneously binds TCR and MHC (128–131), mutagenesis experiments have shown more clearly that CD8 is able to bind the MHC independent of the TCR (132–134). Clarification of the CD8/pMHC interaction, originally defined by mutagenesis, has now been achieved from a crystal structure of human CD8αα bound to HLA-A2 (135, 136). In this complex, the globular head of CD8αα, whose structure was previously shown to resemble an antibody Fv fragment (137), binds to the side of the MHC containing the “acidic loop” of the α3 domain in a manner similar to the way an antibody binds its antigen (i.e. through the CDR-like loops). The interface is very large and comprises not only the CDR loops of CD8 but also the side of one of the CD8α domains, which contacts the underside of the β-sheet floor of the MHC α1 and α2 domains. The crystal structure is a monomer, and the CD8αα is oriented at approximately 90◦ relative to the long axis of the MHC. The CD8αα construct used in the crystal did not contain the long stalk region that connects the globular head to the membrane.

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

T CELL RECOGNITION

391

Even so, the perpendicular orientation of the CD8αα to the long axis of the TCR/MHC complex makes it difficult to visualize how the stalk region would contact the TCR. Further studies will need to be directed at obtaining a structure of CD8αβ and clarifying the role of the stalk region in the affinity-enhancement effect observed for CD8 both in vitro and on cells (49, 126).

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUTURE PERSPECTIVES Many of the larger issue questions about the structural basis of T cell recognition have been answered by the recent structural results. However, as for the understanding of antibody-antigen interactions, true clarification will come in the details when a large enough database of structures is available so that generalizations can be made that will stand the test of time. No structures exist yet for a class II TCR/pMHC complex or any of the CD3 components, although these will no doubt appear soon. The obvious direction for this field of structural immunology is toward the reconstitution of the entire TCR signaling complex in order to obtain a molecular understanding of the signaling process. To approach these formidable structural problems, a combination of molecular biology, protein engineering, crystallography, and other lower resolution biophysical methods such as cryoelectron microscopy will probably be required to obtain a complete picture. Additionally, we need to begin to bridge the gap between the somewhat static results of x-ray crystallography and the dynamic nature of the TCR signaling complex. ACKNOWLEDGMENTS The authors wish to thank Massimo Degano, Jeff Speir, David Kranz, Larry Pease, Don Wiley, Frank Carbone, Ellis Reinherz, Jia-huai Wang, Alex Smolyar, Mark Davis, Randy Stefanko, Anders Brunmark, Michael Jackson, and Per Peterson for helpful discussions, support, reagents, and coordinates. The authors are supported by NIH RO1 CA58896 (IAW), AI42266 (IAW), and AI42267 (LT), and the RW Johnson Pharmaceutical Research Institute. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Davis MM, Bjorkman PJ. 1988. T-cell antigen receptor genes and T-cell recognition. Nature 334:395–402 2. Davies DR, Metzger H. 1983. Structural basis of antibody function. Annu. Rev. Immunol. 1:87–117

3. Bjorkman PJ. 1997. MHC restriction in three dimensions: a view of T cell receptor/ligand interactions. Cell 89:167–70 4. Wilson IA, Garcia KC. 1997. T-cell receptor structure and TCR complexes. Curr. Opin. Struct. Biol. 7:839–48

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

392

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

5. Garcia KC, Teyton L. 1998. TCR-peptide-MHC interactions: biological lessons from structural studies. Curr. Opin. Biotech. 9:338–43 6. Meuer SC, Acuto O, Hercend T, Schlossman SF, Reinherz EL. 1984. The human T-cell receptor. Annu. Rev. Immunol. 2:23–50 7. Clevers H, Alarcon B, Wileman T, Terhorst C. 1988. The T cell receptor/CD3 complex: a dynamic protein ensemble. Annu. Rev. Immunol. 6:629–62 8. Janeway CA Jr. 1992. The T cell receptor as a multicomponent signalling machine: CD4/CD8 coreceptors and CD45 in T cell activation. Annu. Rev. Immunol. 10:645– 74 9. Jorgensen J, Reay P, Ehrich E, Davis M. 1992. Molecular components of Tcell recognition. Annu. Rev. Immunol. 10: 835–73 10. Tonegawa S. 1983. Somatic generation of antibody diversity. Nature 302:575–81 11. Kronenberg M, Siu G, Hood LE, Shastri N. 1986. The molecular genetics of the T-cell antigen receptor and T-cell antigen recognition. Annu. Rev. Immunol. 4:529– 91 12. Wilson IA, Stanfield RL. 1994. Antibodyantigen interactions: new structures and new conformational changes. Curr. Opin. Struct. Biol. 4:857–67 13. Germain R. 1994. MHC-dependent antigen processing and peptide presentation: providing ligands for T lymphocyte. Cell 76:287–99 14. Matsui K, Boniface JJ, Reay PA, Schild H, Fazekas de St Groth B, Davis MM. 1991. Low affinity interaction of peptide-MHC complexes with T cell receptors. Science 254:1788–91 15. Karjalainen K. 1994. High sensitivity, low affinity—paradox of T-cell receptor recognition. Curr. Opin. Immunol. 6:9–12 16. Davis MM, Boniface JJ, Reich Z, Lyons D, Hampl J, Arden B, Chien Y. 1998. Ligand recognition αβ T cell receptors. Annu. Rev. Immunol. 16:523–44 17. Kersh GJ, Allen PM. 1996. Essential flexibility in the T-cell recognition of antigen. Nature 380:495–8 18. Madrenas J, Chau LA, Smith J, Bluestone JA, Germain RN. 1997. The efficiency of CD4 recruitment to ligand-engaged TCR controls the agonist/partial agonist properties of peptide-MHC molecule ligands. J. Exp. Med. 185:219–29 19. Bentley G, Boulot G, Karjalainen K, Mariuzza R. 1995. Crystal structure of the β chain of a T cell antigen receptor. Science 267:1984–7

20. Fields BA, Ober B, Malchiodi EL, Lebedeva MI, Braden BC, Ysern X, Kim JK, Shao X, Ward ES, Mariuzza RA. 1995. Crystal structure of the Vα domain of a T cell antigen receptor. Science 270:1821–4 21. Fields BA, Malchiodi EL, Li H, Ysern X, Stauffacher CV, Schlievert PM, Karjalainen K, Mariuzza RA. 1996. Crystal structure of a T-cell receptor β-chain complexed with a superantigen. Nature 384:188–92 22. Garcia KC, Degano M, Stanfield RL, Brunmark A, Jackson MR, Peterson PA, Teyton L, Wilson IA. 1996. An αβ T cell receptor structure at 2.5 A and its orientation in the TCR-MHC complex. Science 274:209–19 23. Housset D, Mazza G, Gregoire C, Piras C, Malissen B, Fontecilla-Camps JC. 1997. The three-dimensional structure of a Tcell antigen receptor VαVβ heterodimer reveals a novel arrangement of the Vβ domain. EMBO J. 16:4205–16 24. Wang J, Lim K, Smolyar A, Teng M, Liu J, Tse AG, Liu J, Hussey RE, Chishti Y, Thomson CT, Sweet RM, Nathenson SG, Chang HC, Sacchettini JC, Reinherz EL. 1998. Atomic structure of an αβ T cell receptor (TCR) heterodimer in complex with an anti-TCR Fab fragment derived from a mitogenic antibody. EMBO J. 17:10–26 25. 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 26. Teng MK, Smolyar A, Tse AGD, Liu JH, Liu J, Hussey RE, Nathenson SG, Chang HC, Reinherz EL, Wang JH. 1998. Identification of a common docking topology with substantial variation among different TCR-peptide-MHC complexes. Curr. Biol. 8:409–12 27. Garcia KC, Degano M, Pease LR, Huang M, Peterson PA, Teyton L, Wilson IA. 1998. Structural basis of plasticity in T cell receptor recognition of a self peptideMHC antigen. Science 279:1166–72 28. Ding YH, Smith KJ, Garboczi DN, Utz U, Biddison WE, Wiley DC. 1998. Two human T cell receptors bind in a similar diagonal mode to the HLA-A2/Tax peptide complex using different TCR amino acids. Immunity 8:403–11 29. Padlan EA, Margulies DH. 1997. Feeling out the complex. Curr. Biol. 7:17–20 30. Amzel LM, Poljak RJ. 1979. Threedimensional structure of immunoglobulins. Annu. Rev. Biochem. 48:961–97 31. Davies DR, Padlan EA, Sheriff S. 1990.

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

T CELL RECOGNITION

32. 33. 34.

Annu. Rev. Immunol. 1999.17:369-397. 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.

Antibody-antigen complexes. Annu. Rev. Biochem. 59:439–73 Padlan E. 1994. Anatomy of the antibody molecule. Mol. Immunol. 31:169–217 Morrison SL. 1992. In vitro antibodies: strategies for production and application. Annu. Rev. Immunol. 10:239–65 Stanfield RL, Wilson IA. 1995. Proteinpeptide interactions. Curr. Opin. Struct. Biol. 5:103–13 Mariuzza RA, Winter G. 1989. Secretion of a homodimeric Vα Cκ T-cell receptorimmunoglobulin chimeric protein. J. Biol. Chem. 264:7310–6 Weber S, Traunecker A, Oliveri F. Gerhard W, Karjalainen K. 1992. Specific low-affinity recognition of major histocompatibility complex plus peptide by soluble T-cell receptor. Nature 356:793–6 Chang HC, Bao Z, Yao Y, Tse AG, Goyarts EC, Madsen M, Kawasaki E, Brauer PP, Sacchettini JC, Nathenson SG, et al. 1994. A general method for facilitating heterodimeric pairing between two proteins: application to expression of alpha and beta T-cell receptor extracellular segments. Proc. Natl. Acad. Sci. USA 91:11408–12 Golden A, Khandekar SS, Osburne MS, Kawasaki E, Reinherz EL, Grossman TH. 1997. High-level production of a secreted, heterodimeric αβ murine T-cell receptor in Escherichia coli. J. Immunol. Methods 206:163–9 Schlueter CJ, Schodin BA, Tetin SY, Kranz DM. 1996. Specificity and binding properties of a single-chain T cell receptor. J. Mol. Biol. 256:859–69 Gregoire C, Malissen B, Mazza G. 1996. Characterization of T cell receptor singlechain Fv fragments secreted by myeloma cells. Eur. J. Immunol. 26:2410–6 Plaksin D, Polakova K, McPhie P, Margulies DH. 1997. A three-domain T cell receptor is biologically active and specifically stains cell surface MHC/peptide complexes. J. Immunol. 158:2218–27 Strong RK, Penny DM, Feldman RM, Weiner LP, Boniface JJ, Davis MM, Bjorkman PJ. 1994. Engineering and expression of a secreted murine TCR with reduced N-linked glycosylation. J. Immunol. 153:4111–21 Roberts DB, Mulvany WJ, Dwek RA, Rudd PM. 1998. Mutant analysis reveals an alternative pathway for N-linked glycosylation in Drosophila melanogaster. Eur. J. Biochem. 253:494–8 Garcia KC, Tallquist MD, Pease LR, Brunmark A, Scott CA, Degano M, Stura EA, Peterson PA, Wilson IA, Teyton L.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

393

1997. αβ T cell receptor interactions with syngeneic and allogeneic ligands: affinity measurements and crystallization. Proc. Natl. Acad. Sci. USA 94:13838–43 Jackson MR, Song ES, Yang Y, Peterson PA. 1992. Empty and peptide-containing conformers of class I major histocompatibility complex molecules expressed in Drosophila melanogaster cells. Proc. Natl. Acad. Sci. USA 89:12117–21 Fremont DH, Matsumura M, Stura EA, Peterson PA, Wilson IA. 1992. Crystal structures of two viral peptides in complex with murine MHC class I H-2Kb. Science 257:919–27 Zeng Z, Castano AR, Segelke BW, Stura EA, Peterson PA, Wilson IA. 1997. Crystal structure of mouse CD1: an MHClike fold with a large hydrophobic binding groove. Science 277:339–45 Dessen A, Lawrence CM, Cupo S, Zaller DM, Wiley DC. 1997. X-ray crystal structure of HLA-DR4 (DRA∗ 0101, DRB1∗ 0401) complexed with a peptide from human collagen II. Immunity 7:473– 81 Garcia K, Scott C, Brunmark A, Carbone F, Peterson P, Wilson I, Teyton L. 1996. CD8 enhances formation of stable T-cell receptor/MHC class I molecule complexes. Nature 384:577–81 Kwong PD, Wyatt R, Robinson J, Sweet RW, Sodroski J, Hendrickson WA. 1998. Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 393:648–59 Liu J, Tse AGD, Chang HC, Liu J, Wang J, Hussey RE, Chishti Y, Rheinhold B, Spoerl R, Nathenson SG, Sacchettini JC, Reinherz EL. 1996. Crystallization of a deglycosylated T cell receptor (TCR) complexed with an anti-TCR Fab fragment. J. Biol. Chem. 271:33639–46 Novotny J, Ganju RK, Smiley ST, Hussey RE, Luther MA, Recny MA, Siliciano RF, Reinherz EL. 1991. A soluble, singlechain T-cell receptor fragment endowed with antigen-combining properties. Proc. Natl. Acad. Sci. USA 88:8646–50 Hilyard KL, Reyburn H, Chung S, Bell JI, Strominger JL. 1994. Binding of soluble natural ligands to a soluble human T-cell receptor fragment produced in Escherichia coli. Proc. Natl. Acad. Sci. USA 91:9057–61 Garboczi DN, Utz U, Ghosh P, Seth A, Kim J, VanTienhoven EA, Biddison WE, Wiley DC. 1996. Assembly, specific binding, and crystallization of a human TCR-αβ with an antigenic Tax peptide

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

394

55.

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON from human T lymphotropic virus type 1 and the class I MHC molecule HLA-A2. J. Immunol. 157:5403–10 Boulot G, Bentley G, Karjalainen K, Mariuzza R. 1994. Crystallisation and preliminary X-ray diffraction analysis of the β chain of a T-cell antigen receptor. J. Mol. Biol. 235:795–7 Fields BA, Ysern X, Poljak RJ, Shao X, Ward ES, Mariuzza RA. 1994. Crystallization and preliminary X-ray diffraction study of a bacterially produced T-cell antigen receptor Vα domain. J. Mol. Biol. 239:339–41 Saito H, Kranz DM, Takagaki Y, Hayday AC, Eisen HN, Tonegawa S. 1984. A third rearranged and expressed gene in a clone of cytotoxic T lymphocytes. Nature 312:36–40 Saito H, Kranz DM, Takagaki Y, Hayday AC, Eisen HN, Tonegawa S. 1984. Complete primary structure of a heterodimeric T-cell receptor deduced from cDNA sequences. Nature 309:757–62 Sha WC, Nelson CA, Newberry RD, Kranz DM, Russell JH, Loh DY. 1988. Selective expression of an antigen receptor on CD8-bearing T lymphocytes in transgenic mice. Nature 335:271–4 Sha WC, Nelson CA, Newberry RD, Kranz DM, Russell JH, Loh DY. 1988. Positive and negative selection of an antigen receptor on T cells in transgenic mice. Nature 336:73–6 Sha WC, Nelson CA, Newberry RD, Pullen JK, Pease LR, Russell JH, Loh DY. 1990. Positive selection of transgenic receptor-bearing thymocytes by Kb antigen is altered by Kb mutations that involve peptide binding. Proc. Natl. Acad. Sci. USA 87:6186–90 Udaka K, Tsomides TJ, Eisen HN. 1992. A naturally occurring peptide recognized by alloreactive CD8+ cytotoxic T lymphocytes in association with a class I MHC protein. Cell 69:989–98 Udaka K, Wiesmuller K-H, Kienle S, Jung G, Walden P. 1996. Self-MHCrestricted peptides recognized by an alloreactive T lymphocyte clone. Immunology 157:670–8 Tallquist MD, Pease LR. 1995. Alloreactive 2C T cells recognize a self peptide in the context of the mutant Kbm3 molecule. J. Immunol. 155:2419–26 Hedrick SM, Nielsen EA, Kavaler J, Cohen DI, Davis MM. 1984. Sequence relationships between putative T-cell receptor polypeptides and immunoglobulins. Nature 308:153–8 Novotny J, Tonegawa S, Saito H, Kranz

67. 68.

69.

70.

71.

72.

73.

74.

75.

76.

77. 78. 79.

80.

D, Eisen H. 1986. Secondary, tertiary and quaternary structure of T-cell-specific immunoglobulin-like polypeptide chains. Proc. Natl. Acad. Sci. USA 83:742–6 Chothia C, Boswell D, Lesk A. 1988. The outline structure of the T-cell αβ receptor. EMBO J. 7:3745–55 Ghendler Y, Smolyar A, Chang HC, Reinherz EL. 1998. One of the CD3ε subunits within a T cell receptor complex lies in close proximity to the Cβ FG loop. J. Exp. Med. 187:1529–36 Lesk AM, Chothia C. 1988. Elbow motion in the immunoglobulins involves a molecular ball-and-socket joint. Nature 335:188–90 Deisenhofer J. 1981. Crystallographic refinement and atomic models of a human Fc fragment and its complex with fragment B of protein A from Staphylococcus aureus at 2.9- and 2.8-A resolution. Biochemistry 20:2361–70 Stanfield RL, Takimoto-Kamimura M, Rini JM, Profy AT, Wilson IA. 1993. Major antigen-induced domain rearrangements in an antibody. Structure 1:83–93 Li H, Lebedeva MI, Ward ES, Mariuzza RA. 1997. Dual conformations of a T cell receptor Vα homodimer: implications for variability in VαVβ domain association. J. Mol. Biol. 269:385–94 Colman PM. 1988. Structure of antibodyantigen complexes: implications for immune recognition. Adv. Immunol. 43:99– 132 Chothia C, Novotny J, Bruccoleri R, Karplus M. 1985. Domain association in immunoglobulin molecules. The packing of variable domains. J. Mol. Biol. 186:651–63 Li H, Lebedeva MI, Llera AS, Fields BA, Brenner MB, Mariuzza RA. 1998. Structure of the Vδ domain of a human γ δ T-cell antigen receptor. Nature 391:502– 6 Choi Y, Herman A, DiGiusto D, Wade T, Marrack P, Kappler J. 1990. Residues of the variable region of the T-cell-receptor β chain that interact with S. aureus toxin superantigens. Nature 346:471–3 Davies DR, Cohen GH. 1996. Interactions of protein antigens with antibodies. Proc. Natl. Acad. Sci. USA 93:7–12 Madden D. 1995. The three-dimensional structure of peptide-MHC complexes. Annu. Rev. Immunol. 13:587–622 Claverie J, Prochnicka-Chalufour A, Bougueleret L. 1989. Implications of a fab-like structure for the T-cell receptor. Immunol. Today 10:10–14 Chothia C, Lesk A. 1987. Canonical

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

T CELL RECOGNITION

81.

82.

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

structures for the hypervariable regions of immunoglobulins. J. Mol. Biol. 196:901– 17 Al-Lazikani B, Lesk AM, Chothia C. 1997. Standard conformations for the canonical structures of immunoglobulins. J. Mol. Biol. 273:927–48 Bentley G, Mariuzza R. 1996. The structure of the T cell antigen receptor. Annu. Rev. Immunol. 14:563–90 Rock EP, Sibbald PR, Davis MM, Chien YH. 1994. CDR3 length in antigenspecific immune receptors. J. Exp. Med. 179:323–8 Bjorkman PJ, Sapar MA, Samraoui B, Bennett WS, Strominger JL, Wiley DC. 1987. Structure of the human class I histocompatibility antigen, HLA-A2. Nature 329:506–12 Zhang W, Young AC, Imarai M, Nathenson SG, Sacchettini JC. 1992. Crystal structure of the major histocompatibility complex class I H-2Kb molecule containing a single viral peptide: implications for peptide binding and T-cell receptor recognition. Proc. Natl. Acad. Sci. USA 89:8403–7 Madden DR, Garboczi DN, Wiley DC. 1993. The antigenic identity of peptideMHC complexes: a comparison of the conformations of five viral peptides presented by HLA-A2. Cell 75:693–708 Stern LJ, Brown JH, Jardetzky TS, Gorga JC, Urban RG, Strominger JL, Wiley DC. 1994. Crystal structure of the human class II MHC protein HLA-DR1 complexed with an influenza virus peptide. Nature 368:215–21 Jardetzky T, Brown J, Gorga J, Stern L, Urban R, Chi Y, Stauffacher CJ, Wiley D. 1994. Three-dimensional structure of a human class II histocompatibility molecule complexed with superantigen. Nature 368:711–8 Young AC, Zhang W, Sacchettini JC, Nathenson SG. 1994. The three-dimensional ˚ resolution: structure of H-2Db at 2.4 A implications for antigen-determinant selection. Cell 76:39–50 Smith KJ, Reid SW, Stuart DI, McMichael AJ, Jones EY, Bell JI. 1996. An altered position of the α2 helix of MHC class I is revealed by the crystal structure of HLA-B∗ 3501. Immunity 4:203–13 Fremont DH, Monnaie D, Nelson CA, Hendrickson WA, Unanue ER. 1998. Crystal structure of I-Ak in complex with a dominant epitope of lysozyme. Immunity 8:305–17 Scott CA, Peterson PA, Teyton L, Wilson

93.

94.

95.

96.

97.

98. 99.

100.

101.

102.

103.

104.

395

IA. 1998. Crystal structures of two I-Adpeptide complexes reveal that high affinity can be achieved without large anchor residues. Immunity 8:319–29 Speir JA, Garcia KC, Brunmark A, Degano M, Peterson PA, Teyton L, Wilson IA. 1998. Structural basis of 2C TCR allorecognition of H-2Ld peptide complexes. Immunity 8:553–62 Bjorkman PJ, Saper MA, Samraoui B, Bennett WS, Strominger JL, Wiley DC. 1987. The foreign antigen binding site and T cell recognition regions of class I histocompatibility antigens. Nature 329:512–8 Matsumura M, Fremont DH, Peterson PA, Wilson IA. 1992. Emerging principles for the recognition of peptide antigens by MHC class I molecules. Science 257:927– 34 Stern LJ, Wiley DC. 1994. Antigenic peptide binding by class I and class II histocompatibility proteins. Structure 2:245– 51 Wilson IA, Fremont DH. 1993. Structural analysis of MHC class I molecules with bound peptide antigens. Semin. Immunol. 5:75–80 Jones EY. 1997. MHC class I and class II structures. Curr. Opin. Immunol. 9:75–9 Engel I, Hedrick S. 1988. Site-directed mutations in the VDJ junctional region of a T cell receptor β chain cause changes in antigenic peptide recognition. Cell 54:473–84 Jorgensen JL, Esser U, de St Groth BF, Reay PA, Davis MM. 1992. Mapping T-cell receptor-peptide contacts by variant peptide immunization of single-chain transgenics. Nature 355:224–30 Wucherpfennig KW, Strominger JL. 1995. Molecular mimicry in T cellmediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 80:695–705 Sun R, Shepherd SE, Geier SS, Thomson CT, Sheil JM, Nathenson SG. 1995. Evidence that the antigen receptors of cytotoxic T lymphocytes interact with a common recognition pattern on the H-2Kb molecule. Immunity 3:573–82 Hong SC, Chelouche A, Lin RH, Shaywitz D, Braunstein NS, Glimcher L, Janeway CA Jr. 1995. An MHC interaction site maps to the amino-terminal half of the T cell receptor α chain variable domain. J. Immunol. 154:5153–64 Sant’Angelo D, Waterbury G, PrestonHurlburt P, Yoon S, Medzhitov R, Hong S, Janeway C Jr. 1996. The specificity and orientation of a TCR to its peptide-MHC class II ligands. Immunity 4:367–76

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

396

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

GARCIA, TEYTON & WILSON

105. Chang HC, Smolyar A, Spoerl R, Witte T, Yao Y, Goyarts EC, Nathenson SG, Reinherz EL. 1997. Topology of T cell receptor-peptide/class I MHC interaction defined by charge reversal complementation and functional analysis. J. Mol. Biol. 271:278–93 106. Smith KD, Lutz CT. 1997. Alloreactive T cell recognition of MHC class I molecules: the T cell receptor interacts with limited regions of the MHC class I long α helices. J. Immunol. 158:2805–12 107. Pullen JK, Hunt HD, Horton RM, Pease LR. 1989. The functional significance of two amino acid polymorphisms in the antigen-presenting domain of class I MHC molecules. Molecular dissection of Kbm3. J. Immunol. 143:1674–9 108. Tallquist MD, Weaver AJ, Pease LR. 1998. Degenerative recognition of alloantigenic peptides on a positivelyselecting class I molecule. J. Immunol. 160:802–9 109. Lawrence MC, Colman PM. 1993. Shape complementarity at protein/protein interfaces. J. Mol. Biol. 234:946–50 110. Sim BC, Zerva L, Greene MI, Gascoigne NRJ. 1996. Control of MHC restriction of TCR Vα CDR1 and CDR2. Science 273:963–6 111. Manning TC, Schlueter CJ, Brodnicki TC, Parke EA, Speir JA, Garcia KC, Teyton L, Wilson IA, Kranz DM. 1998. Alanine scanning mutagenesis of an αβ T cell receptor: mapping the energy of antigen recognition. Immunity 8:413–25 112. Valitutti S, Muller S, Cella M, Padovan E, Lanzavecchia A. 1995. Serial triggering of many T-cell receptors by a few peptideMHC complexes. Nature 375:148–51 113. Ignatowicz L, Kappler J, Marrack P. 1996. The repertoire of T cells shaped by a single MHC/peptide ligand. Cell 84:521–9 114. Pawlowski TJ, Singleton MD, Loh DY, Berg R, Staerz UD. 1996. Permissive recognition during positive selection. Eur. J. Immunol. 26:851–7 115. Bevan MJ. 1997. In thymic selection, peptide diversity gives and takes away. Immunity 7:175–8 116. Kuzushima K, Sun R, van Bleek GM, Vegh Z, Nathenson SG. 1995. The role of self peptides in the allogeneic crossreactivity of CTLs. J. Immunol. 155:594– 601 117. Ausubel LJ, Kwan CK, Sette A, Kuchroo V, Hafler DA. 1996. Complementary mutations in an antigenic peptide allow for crossreactivity of autoreactive Tcell clones. Proc. Natl. Acad. Sci. USA 93:15317–22

118. Germain RN. 1990. Immunology. Making a molecular match. Nature 344:19–22 119. Nikolic-Zugic J, Bevan MJ. 1990. Role of self-peptides in positively selecting the T-cell repertoire. Nature 344:65–7 120. Ysern X, Li H, Mariuzza RA. 1998. Imperfect interfaces. Nat. Struct. Biol. 5:412–4 121. Nanda NK, Arzoo KK, Geysen HM, Sette A, Sercarz EE. 1995. Recognition of multiple peptide cores by a single T cell receptor. J. Exp. Med. 182:531–9 122. Germain R. 1993. Seeing double. Curr. Biol. 3:586–9 123. Weiss A, Littman D. 1994. Signal transduction by lymphocyte antigen receptors. Cell 76:263–74 124. Fremont DH, Hendrickson WA, Marrack P, Kappler J. 1996. Structures of an MHC class II molecule with covalently bound single peptides. Science 272:1001–4 125. Reich Z, Boniface JJ, Lyons DS, Borochov N, Wachtel EJ, Davis MM. 1997. Ligand-specific oligomerization of T-cell receptor molecules. Nature 387:617–20 126. Luescher IF, Vivier E, Layer A, Mahiou J, Godeau F, Malissen B, Romero P. 1995. CD8 modulation of T-cell antigen receptor-ligand interactions on living cytotoxic T lymphocytes. Nature 373:353–6 127. Hampl J, Chein YH, Davis MM. 1997. CD4 augments the response of a T cell to agonist but not to antagonist ligands. Immunity 7:379–85 128. Doyle C, Strominger JL. 1987. Interaction between CD4 and class II MHC molecules mediates cell adhesion. Nature 330:256–9 129. Rojo JM, Saizawa K, Janeway CA Jr. 1989. Physical association of CD4 and the T-cell receptor can be induced by anti-Tcell receptor antibodies. Proc. Natl. Acad. Sci. USA 86:3311–5 130. Dianzani U, Shaw A, al-Ramadi BK, Kubo RT, Janeway CA Jr. 1992. Physical association of CD4 with the T cell receptor. J. Immunol. 148:678–88 131. Vignali DA, Carson RT, Chang B, Mittler RS, Strominger JL. 1996. The two membrane proximal domains of CD4 interact with the T cell receptor. J. Exp. Med. 183:2097–107 132. Potter TA, Bluestone JA, Rajan TV. 1987. A single amino acid substitution in the α3 domain of an H-2 class I molecule abrogates reactivity with CTL. J. Exp. Med. 166:956–66 133. Salter RD, Benjamin RJ, Wesley PK, Buxton SE, Garrett TP, Clayberger C, Krensky AM, Norment AM, Littman DR, Parham P. 1990. A binding site for the

P1: SAT/ary

P2: NBL/mbg/plb

February 11, 1999

12:15

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-13

T CELL RECOGNITION

134.

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

135.

136.

137.

T-cell co-receptor CD8 on the α3 domain of HLA-A2. Nature 345:41–6 Shen L, Potter TA, Kane KP. 1996. Glu227 → Lys substitution in the acidic loop of major histocompatibility complex class I α3 domain distinguishes low avidity CD8 coreceptor and avidity-enhanced CD8 accessory functions. J. Exp. Med. 184:1671–83 Gao GF, Tormo J, Gerth UC, Wyer JR, McMichael AJ, Stuart DI, Bell JI, Jones EY, Jakobsen BK. 1997. Crystal structure of the complex between human CD8αα and HLA-A2. Nature 387:630–4 Gao GF, Gerth UC, Wyer JR, Willcox BE, O’Callaghan CA, Zhang Z, Jones EY, Bell JI, Jakobsen BK. 1998. Assembly and crystallization of the complex between the human T cell coreceptor CD8α homodimer and HLA-A2. Protein Sci. 7:1245–9 Leahy DJ, Axel R, Hendrickson WA. 1992. Crystal structure of a soluble form of the human T cell coreceptor CD8 at 2.6 A resolution. Cell 68:1145–62

397

138. Connolly ML. 1983. Analytical molecular surface calculation. J. Appl. Crystallog. 16:548–58 139. Kabat EA, Wu TT, Perry HM, Gottesman KS, Foeller C. 1991. Sequences of Proteins of Immunological Interest. 5th ed. Bethesda, Md: National Institutes of Health 140. Huang M, Syed R, Stura EA, Stone MJ, Stefanko RS, Ruf W, Edgington TS, Wilson IA. 1998. The mechanisms of an inhibitory antibody on TF-initiated blood coagulation revealed by the crystal structures of human tissue factor, Fab 5G9 and TF.G9 complex. J. Mol. Biol. 275:873–94 141. Kraulis PJ. 1991. MOLSCRIPT. J. Appl. Crystallog. 24:946–50 142. Merritt EA, Murphy MEP. 1994. Raster3D Version 2.0: A Program for Photorealistic Molecular Graphics. Acta Cryst. D50:869–73 143. Ferrin TE, Huang CC, Jarvis LE, Langridge R. 1988. The MIDAS display system. J. Mol. Graph. 6:13–27

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:369-397. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:399–433 c 1999 by Annual Reviews. All rights reserved Copyright

DEVELOPMENT AND MATURATION OF SECONDARY LYMPHOID TISSUES Yang-Xin Fu1 The Center for Immunology, Department of Pathology, Washington University School of Medicine, Saint Louis, Missouri 63110; e-mail: [email protected]

David D. Chaplin The Center for Immunology, Howard Hughes Medical Institute and Department of Internal Medicine, Washington University School of Medicine, Saint Louis, Missouri 63110; e-mail: [email protected] KEY WORDS:

follicular dendritic cells, lymphotoxin, lymph node, Peyer’s patch, tumor necrosis factor

ABSTRACT The secondary lymphoid tissues are located at strategic sites where foreign antigens can be efficiently brought together with immune system regulatory and effector cells. The organized structure of the secondary lymphoid tissues is thought to enhance the sensitivity of antigen recognition and to support proper regulation of the activation and maturation of the antigen-responsive lymphoid cells. Although a substantial amount is known about the cellular elements that compose the lymphoid and nonlymphoid components of the secondary lymphoid tissues, information concerning the signals that control the development of the tissues and that maintain the organized tissue microenvironment remain undefined. Studies over the past few years have identified lymphotoxin as a critical signaling molecule not only for the organogenesis of secondary lymphoid tissues but for the maintenance of aspects of their microarchitecture as well. Additional signaling molecules that contribute to the formation of normal lymphoid tissue structure are being identified at an accelerating pace. Analyses of mouse strains with congenital defects in different aspects of secondary lymphoid tissue development are beginning to clarify the role of these tissues in immune responses and 1 Present address: Department of Pathology, University of Chicago, 5814 Maryland, Chicago, IL 60637.

399 0732-0582/99/0410-0399$08.00

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

400

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

host defense. This review focuses on studies defining recently identified crucial signals for the biogenesis of secondary lymphoid organs and for the maintenance of their proper microarchitecture. It also discusses new insights into how the structure of these tissues supports effective immune responses.

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTRODUCTION The immune system is functionally compartmentalized into primary lymphoid organs and secondary lymphoid tissues. The role of primary lymphoid tissues, in which lymphocyte precursors develop into immunocompetent naive lymphocytes, has been extensively studied and is not discussed here. Secondary lymphoid tissues are the spleen, lymph nodes (LN), and organized lymphoid tissues associated with mucosal surfaces, including the tonsils, bronchial-associated lymphoid tissues, gut-associated lymphoid tissues, Peyer’s patches (PP), and other less-prominent organized clusters of lymphoid cells associated with the gastrointestinal, genitourinary, and respiratory tracts. These lymphoid tissues are located at strategic sites where foreign antigens entering the body from either the skin or a mucosal surface can be trapped and concentrated. Lymphocytes, antigen transporting and presenting cells, and other regulatory cells are also located in these anatomically defined tissues and are thought to be organized into structures that optimize cellular interactions that support the efficient removal of unwanted pathogens (1–3). The structures of LN, PP, and spleen have been thoroughly reviewed (4–6). Each lymphoid tissue has a unique architecture, but they share some common features. Generally, T and B lymphocytes are segregated into distinct areas. The area of T cell predominance contains small numbers of B cells and substantial numbers of dendritic cells, which are thought to present antigens for the initial activation of the T and B cells. The B cell areas contain primary follicles that represent sites where antigen-activated B cells that have received T cell help can expand, mature, and undergo the germinal center reaction prior to becoming antibody-producing cells and memory B cells. Packaging immune cells into secondary lymphoid tissues is thought to enhance the efficiency of immune responses by (a) arranging B and T cells in geographic locations that favor their interactions with antigen-presenting cells and perhaps other regulatory cells and (b) providing a framework to permit rapid circulation of naive cells through a space where antigens are concentrated. The efficiency generated by these organized tissue structures is shown by experiments, such as those by Kundig and coworkers (7), in which immunization with 106 fibrosarcoma cells was required to prime T cells when the tumor cells were injected subcutaneously, whereas only 500 cells were required when they were injected directly into the spleen. Zinkernagel et al (8) recently proposed that

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

SECONDARY LYMPHOID TISSUES

401

both geographical localization of antigens to secondary lymphoid tissues and the dose and time of antigen exposure are key variables determining whether a regulated, productive immune response occurs (8). The ability of secondary lymphoid tissues to concentrate and retain antigens in proximity to the initially rare antigen-specific cells may be critical in order for this to occur.

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

STRUCTURAL FEATURES OF NORMAL SECONDARY LYMPHOID TISSUES The microarchitecture of the LN has been the most fully characterized of all of the secondary lymphoid tissues. The infrastructure of the LN has been demonstrated at the ultrastructural level by removing cells and then analyzing fixed material by scanning electron microscopy. LN are fed by two vascular systems: the lymphatic vasculature, which delivers antigens and antigen-transporting cells from peripheral tissues to the node and returns fluid and cells to the circulation, and the blood vasculature, which brings circulating lymphocytes into the node. The afferent lymphatics empty into the marginal sinus immediately beneath the LN capsule and drain ultimately via the medullary sinus into the efferent lymphatic. Under the marginal sinus lies the LN cortex, which is separated into an outer cortex, consisting mostly of B lymphocytes, and an inner paracortex, consisting mostly of T lymphocytes. Within the outer cortex, the B cells are organized into primary follicles that support the formation of germinal centers following immunization with a T cell–dependent antigen. An important nonlymphocytic component of the primary follicles is the clusters of follicular dendritic cells (FDC). FDC with their abundant complement receptors and immunoglobulin Fc receptors are thought to focus immune complexes within the B cell follicle in a fashion that is crucial for the development of effective isotypeswitched and memory B cell responses (9–12). Internal to the paracortex are the medullary cords, populated prominently by macrophages and plasma cells and which lead to the medullary sinus. The functional unit of the paracortex is the paracortical cord, which stretches from the base of a B cell follicle to an underlying medullary cord. The paracortical cord is approximately 100–1000 µm in diameter and is thought to provide a space in which antigen-presenting cells (primarily dendritic cells) can encounter rare antigen-specific T lymphocytes and favor their activation and subsequent maturation (13). Each segment of paracortex appears to be composed of hundreds of paracortical cords. In contrast, in the area of the primary B cell follicles, the tissue is penetrated by only a few fibers. This suggests that the requirements for cell trafficking change dramatically when cells leave the paracortex and enter the follicle structure. B and T lymphocytes from the blood enter the LN by crossing the specialized high endothelial venules, which are located in the paracortex near the junction with

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

402

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

the outer cortex (14). If they traffic through the paracortical cord and medullary cord without being activated by antigen on an antigen-presenting cell, they return to the circulation via the efferent lymphatic vessels and the thoracic duct. For a naive lymphocyte that does not encounter its cognate antigen, such a circuit through the node is thought to take fewer than 24 h. The mucosa-associated lymphoid tissues, including the PP, bronchial-associated lymphoid tissues, gut-associated lymphoid tissues, and others, have a general structure similar to the LN, with distinct T and B cell areas and B cell follicles (5, 15, 16), but they differ prominently in the pathway by which antigens enter the lymphoid compartment. Rather than traffic through an afferent lymphatic, antigens enter the mucosa-associated lymphoid tissues across the mucosal epithelium. Often the mucosal epithelial cells immediately overlying the lymphoid tissue are specialized for uptake of antigen from the lumen of the mucosa. The M cells overlying PP in the gastrointestinal tract have been shown to transport particles as large as intact microorganisms (17). On the abluminal surface of the mucosa-associated lymphoid tissues, lymphocytes return to the blood circulation via efferent lymphatic vessels that ultimately join the thoracic duct like efferent lymphatic vessels from LN. The spleen, the largest single lymphoid organ in mammals, contains up to 25% of the body’s mature lymphocytes (18). It is separated into two major components, the red pulp and the white pulp (Figure 1). The red pulp has been thought of primarily as a filter in which aged or damaged erythrocytes are removed from the circulation. It consists of a reticular network containing stromal cells and a large population of macrophages. It also contains a variably large population of plasma cells and can be a site of substantial immunoglobulin production. The white pulp represents the organized lymphoid compartment in which regulated activation and maturation of antigen-dependent B and T cells occur. Unlike the LN, the spleen has a single vascular supply, with immune cells and antigen entering the tissue with the blood via the splenic artery. In humans, approximately 5% of the total cardiac output is directed through the spleen (19). The blood enters via the splenic artery, which branches into trabecular arteries and, ultimately, into central arterioles that penetrate the white pulp nodules. Surrounding the central arterioles is a T cell–rich compartment designated the periarteriolar lymphoid sheath (PALS). The PALS also contains abundant interdigitating dendritic cells that are thought to serve as important antigen-presenting cells early in the immune response in this tissue. The central arteriole, after it penetrates the PALS, forms a marginal sinus that is lined with a mucosal addressin cell adhesion molecule-1 (MAdCAM-1)–expressing endothelium (20). Also associated with this marginal sinus endothelium is a specialized layer of metallophilic macrophages that are thought to regulate the entry of antigen into the white pulp tissue (21). Additional venous sinuses

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SECONDARY LYMPHOID TISSUES

403

Figure 1 Structure of the spleen white pulp nodule. The white pulp nodule is separated into a central T cell–rich zone [periarteriolar lymphoid sheath (PALS)] surrounded by B cell–rich primary follicles. Within each primary follicle is a cluster of follicular dendritic cells (FDC). The white pulp nodule is separated from the red pulp (RP) by the MAdCAM-1+ marginal sinus (MS). The MS is embedded in a layer of marginal zone (MZ) lymphocytes. Also adjacent to the MS is a layer of metallophilic macrophages that are thought to be important to regulate antigen trafficking into the red and white pulp spaces. The bridging channels (BC) are thought to represent areas by which lymphocytes enter and leave the white pulp. CA, central arteriole.

are present within the red pulp, and these are thought to coalesce into larger segmental venules that ultimately exit the spleen as the major splenic vein. In the white pulp, the B cells are organized into two compartments (2, 22). The first consists of naive B cells and at least some memory B cells and includes the marginal zone cells that are adjacent to the MAdCAM-1–expressing marginal sinus (23, 24). The second is composed of follicle-associated cells, which in the resting state are organized into primary follicles. These primary follicle B cells surround clusters of FDC, similar to the arrangement of primary follicles in LN. Altogether, the components of the white pulp form a highly ordered structure that is thought to be critical for proper regulation of immune responsiveness in this tissue (3, 25).

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

404

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

Although the anatomical features of the secondary lymphoid tissues are well defined, little is known about the mechanisms that establish the normal cellular compartments within these structures during ontogeny, or that regulate trafficking of cells through these compartments during normal function of the mature tissue (1–3, 13). The general conservation of structure between the various tissues has been taken as evidence that the organized structure is critical for the function of the tissues; however, in fact, the relationship between organized tissue structure and effective immune responsiveness has not been extensively investigated (8, 22, 26). Historically, analyses of mice with inherited lack of the thymus have been highly informative for definition of the function of that structure (27); however, until recently, similar model systems for studying the consequences of congenital lack of secondary lymphoid tissues have not existed. In the last several years, studies analyzing the independent roles of lymphotoxin (LT) and tumor necrosis factor (TNF) have provided valuable models in which various aspects of secondary lymphoid tissue structure are disturbed, and the analyzing of these models is beginning to establish the requirements for normal secondary lymphoid structure in the host immune response (22, 28). This article reviews studies that identify some of the signals required for organogenesis and maturation of secondary lymphoid tissues and discusses new insights into the cellular interactions between lymphocytes and stromal cells in these lymphoid tissues.

MUTANT MOUSE STRAINS WITH ALTERED LYMPHOID TISSUE DEVELOPMENT Three spontaneously arising mutant mouse strains have been identified that manifest disturbances of the structures of primary and/or secondary lymphoid tissues. Examination of these strains is leading to important insights into the normal mechanisms controlling development of these tissues and the biological and immunological processes that normally occur within them. The focus of this article is the secondary lymphoid tissues, but a historically important example affecting the development of lymphoid organs, the nude mutation, is discussed first. The autosomal recessive nude mutation is phenotypically invisible in heterozygotes but leads to congenital absence of the thymus and lifelong hairlessness in homozygotes (27). Recent studies by Boehm and coworkers (29–31) have demonstrated that mutations in the whn gene encoding a transcription factor of the forkhead/winged-helix class causes the nude phenotype. Whn is expressed in thymic epithelial precursors and cells within the hair follicle. Although whn expression is not required for the formation of the thymic epithelial primordium prior to the entry of lymphocyte progenitors, the whn

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SECONDARY LYMPHOID TISSUES

405

gene product is required for the subsequent differentiation of this epithelial primordium into the specialized epithelia of the subcapsular, cortical, and medullary compartments (29). Analysis of spontaneous nude mice and mice targeted in vitro for mutations in the whn locus shows that normal development of αβ T cell-receptor-bearing T lymphocytes is dependent on the development of functional thymus tissue. Less thoroughly studied is the mouse strain carrying the semidominant Dh (dominant hemimelia) mutation (32). In homozygous form, this mutation is embryonic lethal. Heterozygous Dh /+ mice show skeletal and visceral abnormalities, including prominent asplenia. Studies using chimeric embryos in which Dh /Dh or Dh /+ embryos were aggregated with C3H/He embryos have demonstrated that when a spleen develops (presumably under the influence of the normal C3H/He cells), its structure is grossly normal and the Dh cells can contribute to at least several components of the splenic tissue (33). This suggests that the Dh mutation affects an early step in spleen organogenesis but not the ability of cells to migrate to the spleen. Although further analysis of this strain may yield important insights into the cellular interactions that commit to spleen organogenesis, because the Dh mutation affects many other somatic tissues, it is unlikely that analysis of this strain will provide straightforward insights into the role of the spleen in normal immune responsiveness. Perhaps more tractable to immunological analysis will be the recently discovered autosomal recessive alymphoplasia (aly) mutation (34). This mutation, which maps to mouse chromosome 11, determines in its homozygous form broad alterations in immune tissue structure and function without gross defects in other somatic tissues. aly/aly mice have complete absence of LN and PP. The thymus is present, but with disturbed structure. Although phenotypically mature T cells are found in blood and spleen, their numbers are reduced and they show depressed function manifested in all measures of T cell activation. The structure of the spleen is diffusely altered, with atrophic white pulp nodules and absence of detectable B cell follicle structure. Total serum immunoglobulins are low, and antigen-specific antibody responses are substantially reduced (34, 35). Comparison of the ontogeny of gut-associated lymphoid tissues in aly/aly, wildtype (wt), and severe combined immunodeficiency (scid ) mice has yielded insights into the development of mouse PP (36). Adachi and colleagues (36) found that clusters of vascular cell adhesion molecule-1 (VCAM-1)–expressing cells form in the wall of the gut by day 15 after conception. Over the next 2 days, these VCAM-1+ clusters of apparently stromal cells become infiltrated with a homogeneous-appearing population of major histocompatibility class II+, IL-7R+, CD4+, and CD3− cells. These cells are then gradually replaced by T and B lymphocytes, as well as by the other cellular elements of the normal, mature PP.

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

406

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

In addition to the natural mouse mutations discussed above, targeted mutagenesis of mice using embryonic stem cells has yielded several examples in which formation of certain lymphoid tissue structures is altered or ablated. Studies by Roberts et al (37) demonstrated the absolute requirement for expression of hox11 for the formation of the spleen. Although hox11 is expressed in other tissues, absence of its gene leads to loss of only the spleen, with the only other detected phenotypic alterations being a modest increase in size of the stomach and pancreas (38). Disturbances in the development of secondary lymphoid tissues have also been observed in mice deficient in LT or TNF, as well as in certain intracellular signaling molecules. Defects due to absence of these molecules are discussed below. Recent interesting studies have examined immune host defense in both the asplenic hox11-deficient and the aly/aly mice (26). Hox11-deficient mice showed modestly delayed antibody responses to vesicular stomatitis virus (VSV) but were otherwise resistant to infection. In contrast, aly/aly mice, with their profound disturbance of lymphoid tissue structure and abnormal B and T lymphocytes responsiveness, were highly susceptible to VSV. They generated a delayed and reduced immunoglobulin (Ig) M anti-VSV response and failed to generate a protective IgG response. This failure to produce an IgG response was not the manifestation of an intrinsic defect in isotype switching in the aly/aly lymphocytes. If aly/aly spleen cells were adoptively transferred into an irradiated wt mouse (with morphologically normal secondary lymphoid tissues), these cells were able to generate a protective, isotype-switched IgG response. Thus, when the structure of the secondary lymphoid tissues is disturbed, maturation of the antibody response is importantly impaired.

AN ESSENTIAL ROLE FOR LYMPHOTOXIN IN THE DEVELOPMENT OF SECONDARY LYMPHOID TISSUES Until recently, there had been little progress in defining essential signals that supported the development of normal secondary lymphoid tissues. As described above, analyses of aly/aly and scid mice have provided clues regarding the initial steps of PP development (36). It is likely that secondary lymphoid tissue organogenesis occurs in several discrete steps and depends on the interaction of several different types of cells using several different types of cell-cell signaling processes. A full understanding of the processes that lead to the formation of normal secondary lymphoid tissue structure will require the identification of the primordial cell types that interact during the initial biogenesis of the tissues and the signals they use to communicate with each other during this organogenesis. In addition, it appears that ongoing signals are required to maintain the normal

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

SECONDARY LYMPHOID TISSUES

407

tissue structures, even in mature animals. Recent studies using gene targeting and transgenic methods are beginning to define some of the important signals that govern these complex processes. The following section focuses on the role of LT and TNF, as recent studies of these molecules are rapidly increasing our understanding of their biological functions and are demonstrating their pivotal roles in the formation of normal peripheral lymphoid tissue structures.

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The TNF Family of Cytokines and Receptors TNF (also designated TNFα) and lymphotoxin-α (LTα) were first identified based on their cytotoxic activities against various cell lines and tumor cells (reviewed in 39). In supernatants of activated cultured cells, TNF and LTα are structurally related homotrimeric proteins. These homotrimeric ligands can interact with and activate each of the two defined TNF receptors, TNFR-I/p55 and TNFR-II/p75 (Figure 2). Because of the similar binding of these two ligands to the defined TNF receptors, they were generally accepted to be functionally redundant, differing only in that LTα is a conventionally secreted glycoprotein, whereas TNF is synthesized with a long N-terminal peptide that anchors it at the cell surface as a type II membrane protein (40). Secretion of TNF is the result of cleavage of the membrane protein from the cell surface by the action of the TNFα converting enzyme (TACE), a metalloproteinase produced by the same cell that synthesizes the TNF (41, 42). Studies in the early 1990s demonstrated that LT also exists in a membrane-associated form, with a single LTα chain noncovalently associated with two copies of a structurally related type II transmembrane protein designated lymphotoxin β (LTβ) (43–45). The genes encoding TNF, LTα, and LTβ are genetically linked, encoded within a 25-kb portion of the class III region of the major histocompatibility complex (46). The membrane LT heterotrimer (LTα 1β 2, or mLT) is not a substrate for TACE and appears to exist only as a membrane-associated protein. It shows no detectable affinity for TNFR-I or TNFR-II but binds and signals through another receptor of the TNFR family, designated the lymphotoxin β receptor (LTβR) (47). The LTβR appears to be specific for mLT and shows no measurable affinity for the homotrimeric TNF or LTα 3 ligands. Membrane LT has also been detected in vitro as an LTα 2β 1 trimer. The LTα 2β 1 heterotrimer can interact with TNFR-I and TNFR-II, but its ability to interact with and activate the LTβR is not fully defined. Also unclear is whether the LTα 2β 1 heterotrimer is present in biologically meaningful quantities in vivo. Unlike TNFR-I and TNFR-II, which are expressed very broadly, the LTβR is not expressed on lymphoid cells. Rather, it is expressed on stromal cells in various lymphoid tissues (28). Because the LTα 1β 2 ligand is membrane associated, it is likely that LTα 1β 2-mediated responses involve physical contact between the mLT-expressing cell and its LTβR-bearing target. Thus, TNF and

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

408

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

Figure 2 Ligands and receptors (R) of the lymphotoxin (LT)/tumor necrosis factor (TNF) family. The four major ligands of the LT/TNF family are shown as homo- and heterotrimeric proteins produced by the effector cell. TNF and LTβ are synthesized as type II membrane proteins, anchoring the primary translation products to the surface of the producing cell. The three major receptors for these ligands show considerable homology in their extracellular domains. Their unique intracellular domains provide mechanisms by which they can each transmit independent signals following ligand binding. The black bar in the intracellular domain of TNFR-I designates the “death domain” that is required to effect the apoptotic response to TNF or LTα 3 in sensitive target cells. Membrane TNF can bind and activate both TNFR-I and TNFR-II similarly to soluble TNF. TACE, TNFα converting enzyme.

LT encode two sets of ligands: the membrane and secreted forms of TNF and the LTα 3, which interact with TNFR-I and TNFR-II, forming one set; and the membrane LTα 1β 2 heterotrimer, which interacts with the LTβR forming the other set. Acting through the different intracellular domains of their receptors, these two sets of ligands and receptors would be expected to mediate independent sets of cellular and tissue responses.

Critical Role of mLT in the Formation of LN and PP The first experiments that linked LTα with the biogenesis of secondary lymphoid tissues used gene targeting to generate a mouse strain with homozygous

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SECONDARY LYMPHOID TISSUES

409

LTα deficiency (48). LTα −/− mice showed a profound defect in formation of LN and complete absence of PP. The spleen was present, but its microarchitecture was grossly disturbed (discussed below). These observations established that LTα was a critical factor, required for the formation of these secondary lymphoid tissue structures. These results were confirmed by Banks et al (49), who independently produced an LTα −/− mouse strain. Of interest, the congenital absence of LN and PP in the LTα −/− mice was observed in the context of apparently normal lymphatic vasculature, with retained efficient transport of India ink injected into the footpad to the spleen (48). Thus, the development of the lymphatic vasculature is not obligately linked to the formation of intact LN. The signals required for the development of lymphatic vessels have not yet been identified, but the recent observation that mice carrying an overexpressing vascular endothelial growth factor-C transgene have hyperplasia of many lymphatic vessels suggests that vascular endothelial growth factor-C may deliver one of the important developmental signals (50). Of interest, defective LN biogenesis was not absolute in the LTα −/− mice. In more than 95% of the mice, LN were completely absent. In the remaining 2–4%, a mesenteric LN was detected (51). In wt mice, mesenteric LN are present as a short chain consisting of several discrete nodes. In the small fraction of LTα −/− mice with mesenteric LN, the normal mesenteric chain was reduced to a single node. The presence of a mesenteric LN in some LTα −/− mice indicates that some LTα-independent signal can substitute for LTα in the development of this particular node. The nature of this LTα-independent signal remains to be characterized. That LTα was essential for the formation of LN and PP was unanticipated because mouse strains deficient in either of the two defined TNF receptors (TNFR-I or TNFR-II) had not been recognized to manifest any defects in lymphoid tissue structure (52, 53). In fact, LN appear to form with normal distribution even in mice with targeted ablation of both TNFR-I and TNFR-II (Y-X Fu, DD Chaplin, unpublished data). This suggested that disturbed LN and PP formation in LTα −/− mice was not mediated by ablation of signals through the homotrimeric ligand/TNFR-I/TNFR-II arm of the TNF ligand/receptor family, but rather that it might represent failure of signaling through the mLT/LTβR arm of the family. The LN and PP defect observed in LTα −/− mice is developmentally fixed. If wt bone marrow (BM) is infused into lethally irradiated LTα −/− mice, although the wt BM–derived cells home to and repopulate the spleen, they are unable to induce the formation of LN or PP (54). In contrast, when LTα −/− BM was infused into lethally irradiated wt mice, the LTα −/− cells showed apparently full potential to repopulate the wt LN, PP, and spleen. Thus, the failure to develop detectable LN or PP in LTα −/− mice was not due to an inability of LTα −/−

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

410

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

BM–derived cells to traffic to the lymphoid tissues. Rather, the absence of LN and PP in the LTα −/− mice represented a true failure of biogenesis of these organs, with some LTα-dependent developmental signal being required before adulthood in order for LN and PP structures to form.

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Membrane Form of LT Supports the Formation of LN and PP Recognizing that LTα probably acted during ontogeny to signal the formation of LN and PP, Rennert et al (55) adopted a novel strategy to investigate the molecular form of LT that was active in these processes. They prepared TNFR-Iand LTβR-immunoglobulin (LTβR-Ig) fusion proteins to achieve in vivo neutralization of the LTα 3 homotrimer and the membrane LTα 1β 2 heterotrimer respectively. Receptor-Ig fusion proteins have been used productively before, both to render the ligand binding domain of the receptor divalent and to enhance stability in vivo. These investigators recognized that if the Ig domain was derived from an IgG, then the receptor-Ig fusion protein could cross the placenta of a pregnant mouse and neutralize the ligand in developing embryos. When an LTβR-Ig fusion was administered to pregnant wt mice at gestational day 18, the offspring were born with unaltered LN structure, but with a total absence of PP. If the fusion protein was administered on gestational day 16, then PP and popliteal LN were ablated, although mesenteric, axillary, and inguinal nodes were retained. If the fusion protein was delivered on gestational day 12 or earlier, then PP, popliteal, axillary, and inguinal LN were all ablated, with only mesenteric LN and certain cervical LN being retained. Administration of a TNFR-I–Ig fusion protein to pregnant wt mice at any time during gestation did not interfere with the development of either LN or PP. Given that the LTβR binds to the membrane form of LT and not the homotrimer, whereas TNFR-I binds the homotrimer but not the membrane LTα 1β 2 form, this study demonstrated clearly that the membrane form of LT was the active signaling molecule in LN and PP biogenesis. Furthermore, this study defined specific time windows during which mLT must act to support the development of different anatomically defined sets of LN and PP. The critical period of mLT expression varied for different sets of LN. Interestingly, the mesenteric LN were resistant to ablation by the LTβR-Ig, reminiscent of the fraction of LTα −/− mice with retained development of the mesenteric node. This confirms that there is a fundamental difference between the mesenteric LN and other LN and suggests that there is not an absolute requirement for an mLT signal for the development of this tissue. Subsequent studies showed that several cervical, sacral, and lumbar LN, in addition to the mesenteric LN, were also resistant to ablation by the soluble LTβR-Ig fusion protein (56). This defined a subset of enteric-associated LN that have no absolute requirement for mLT to induce their development. The

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SECONDARY LYMPHOID TISSUES

411

selective role of mLT in LN and PP biogenesis was underscored by the observation that treatment of mice with a TNFR-I–Ig fusion protein did not interfere with formation of LN or PP. In related studies, an alternative approach was used to neutralize the function of membrane LT (57). A mouse strain was created in which a transgene encoding a soluble LTβR-Ig fusion protein was expressed under the control of the CMV promoter. This promoter yields only very low levels of expression during embryonic development, with high-level expression beginning approximately 3 days after birth. Mice expressing the transgene showed variable loss of PP formation but no defect in LN biogenesis. This confirmed that the ligand neutralized by the soluble LTβR, mLT, gives essential signals for the formation of PP, and that these signals must be present near the time of birth. That the organogenic mLT signal must be delivered during a discrete window of time is supported by the observation that when PP did form in the LTβR-Ig transgenic animals, their overall morphology appeared normal. This was at a time when high levels of receptor-Ig fusion protein could be detected in the circulation. Thus, once commitment to development of a PP occurred, then further mLT signals appeared not to be required to sustain the tissue structure. This is consistent with earlier experiments in which BM from an LTα −/− donor was used to reconstitute hematopoiesis in a lethally irradiated wt mouse (54). After BM reconstitution, only mLT-deficient cells were present in the circulation. Nevertheless, LN structures were retained for the remainder of the life of these animals. These experiments support the concept that during ontogeny, decisions regarding biogenesis of LN and PP are made during a specific window of time and that once a decision for or against formation of the secondary lymphoid tissue has been made, the phenotype is fixed. Expression or lack of expression of mLT outside that window of time does not alter the LN/PP phenotype. Additional data supporting the role of mLT as a key signaling molecule in the biogenesis of LN and PP came from later studies in which the LTβ gene was inactivated by gene targeting (58). LTβ −/− mice were unable to express the membrane LTα 1β 2 heterotrimer but were presumed to retain expression of the LTα 3 homotrimer. The LTβ −/− mice manifested absence of peripheral LN and PP but retained formation of the mesenteric LN and certain cervical LN. Because mesenteric and cervical LN were retained in the LTβ −/− mice and in mice treated with the LTβR-Ig fusion protein (55, 56) and were generally absent in LTα −/− mice (48, 49, 51), it was suggested that the development of mesenteric and cervical nodes was controlled uniquely by LTα. A subsequent study showed that crossing the LTβ −/− genotype onto the TNFR-I−/− genotype resulted in loss of the mesenteric LN, in addition to the loss of other peripheral LN and PP (59). Similar results were obtained using

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

412

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

the receptor-fusion protein technology when wt pregnant mice were treated with a combination of LTβR-Ig and TNFR-I–Ig fusion proteins. In this case, in addition to the peripheral LN and PP, development of the mesenteric LN was extinguished (60). Given that the mesenteric LN is present in mice with isolated deficiency of TNFR-I, this suggests that although mLT plays an essential role in the formation of most peripheral LN and PP and a dominant role in the formation of mesenteric and enteric-associated LN, LTα 3 signaling through TNFR-I might rescue formation of the mesenteric and enteric-associated LN when LTβ is absent. Arguing against a role for LTα 3 is the observation that treatment of wt pregnant mice with a combination of LTβR-Ig fusion protein and neutralizing anti-TNF antibody blocked mesenteric LN formation similarly to treatment with the combination of LTβR-Ig and TNFR-I–Ig fusion proteins (60). This suggests synergy between mLT and TNF for the formation of mesenteric LN, rather than between mLT and LTα 3. Nevertheless, a dominant role for signals from mLT through the LTβR is generally accepted, and the role of LTβR-dependent signaling is underscored by the recent observation that LTβR−/− mice manifest total ablation of all LN, including the mesenteric node (61). From a different perspective, it was recently shown that treatment of pregnant LTα −/− mice with an agonist anti-LTβR monoclonal antibody beginning on day 12 of gestation specifically induced the genesis of both peripheral and enteric-associated LN in the LTα-deficient progeny (60). Thus, signaling through the LTβR appears to be the only essential signal delivered by LTα-containing ligands. In addition, specificity for the location of developing LN appears to be independent of the mLT-expressing cell but dependent on the location of the LTβR-bearing cell. Although several studies have demonstrated an unequivocal and dominant role for mLT in the induction of PP formation, controversy remains regarding a potential role for signaling via TNFR-I for the development of these structures. Neumann et al (62) identified extensive disorganization of gut-associated lymphoid tissue in TNFR-I−/− mice, with absence of morphologically defined PP. In apparent conflict are studies by Rennert et al (55, 56), in which treatment of pregnant mice with soluble TNFR-I–Ig fusion protein failed to block PP formation, regardless of the time or duration of treatment. In related studies, it was found that PP formation was preserved in mice lacking either TNF or TNFR-I, although there were reduced numbers of PP and those that formed were smaller than in wt mice (63). These data together suggest that mLT provides the dominant signal for PP organogenesis but that signaling by TNF through TNFR-I may be required for full development of normal PP structure. It is clear that for LN and PP to develop, the mLT signal must be present during mid- to late-gestation and, for PP, during the few days after birth.

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

SECONDARY LYMPHOID TISSUES

413

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

OTHER SIGNALS ESSENTIAL FOR THE ORGANOGENESIS OF SECONDARY LYMPHOID TISSUES The studies described above have established that mLT interacting with the LTβR provides key signals that support the development of LN and PP, but they also have demonstrated that different secondary lymphoid tissue elements require different signals at specific times and that additional signals are important for the normal development of specific tissues. The careful attention being paid to the structure of experimental mouse strains with altered expression of a range of cellular signaling proteins is aiding in the identification of additional signals that contribute to the development of the normal repertoire of secondary lymphoid tissues. Integration of these signals into models of how the development and maintenance of secondary lymphoid tissue structure are controlled should occur in the context of identifying the nature of the cells that deliver and receive these signals.

BLR1 Burkitt’s lymphoma receptor-1 (BLR1), recently renamed CXCR5, was identified as an orphan chemokine receptor expressed in Burkitt’s lymphoma B cells. Mice with a targeted null mutation in BLR1 manifest loss of inguinal LN and severe reduction in PP biogenesis (64). Examination of BLR1−/− mice demonstrated that this receptor was required for trafficking of B cells to certain lymphoid tissues and to selected compartments within the spleen. Analyses of scid mice have shown that LN, PP, and spleen form in the absence of B and T cells. Consequently, it must be assumed that other cells besides B or T cells can express BLR1 or the BLR1 ligand that specifies BLR1-dependent induction of inguinal LN and PP development. A ligand for BLR1 designated BLC (B lymphocyte chemoattractant)/BCA-1 (B cell-attracting chemokine-1) has been recently identified (65, 66) and found to be produced constitutively in secondary lymphoid tissues by cells located within B cell follicles, perhaps the FDC. Concerning the potential roles of BLC/BCA-1 and BLR1/CXCR5 in the organogenesis of inguinal LN and PP, it is unlikely that FDC or B cells are the cells that act during ontogeny. Future studies should address the nature of cells that express BLC/BCA-1 and BLR1/CXCR5 at sites of secondary lymphoid tissue organogenesis during embryonic development.

Ikaros Ikaros, a member of the kruppel family of zinc finger DNA-binding proteins, is recognized as a key regulator for the specification and development of all lymphoid lineages (67). A mutation that deletes the N-terminal zinc finger

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

414

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DNA-binding domain from the Ikaros proteins blocks lymphocyte development at its earliest recognizable stage (68). Mice carrying this mutation lack mature T and B lymphocytes and also natural killer (NK) cells, as well as their earliest described precursors. The production of subsets of dendritic cells is also defective (69, 70). Interestingly, these Ikaros mutant mice show defective formation of LN and PP (67). When a different mutation of the Ikaros gene was targeted to the C terminus of the protein, thymocyte development was considerably restored, with the numbers of αβ T cells substantially recovered; however, development of secondary lymphoid tissues was still grossly disturbed (69).

Common Cytokine Receptor γ Chain and JAK-3 The common γ chain (γ c) of the interleukin (IL)-2, IL-4, IL-7, IL-9, and IL-15 receptors is defective in humans with X-linked severe combined immunodeficiency. Mice lacking γ c expression manifested a profound defect in the development of T and B cells and had no detectable NK cells. These mice lack peripheral LN and have mesenteric LN that are small (71, 72). The Janus family tyrosine kinase JAK3 is the key signaling molecule known to be associated with γ c, so it was hypothesized that defects in JAK3 might cause an XSCID-like phenotype in mice (73). JAK-3−/− mice have profoundly reduced numbers of T cells and lack B and NK cells (74–76). Similar to mice lacking γ c, JAK-3−/− mice have a broad defect in the formation of peripheral LN and PP (76) and show no recovery of these lymphoid tissues after reconstitution with BM from wt mice (JW Verbsky, Y-X Fu, & DD Chaplin, unpublished data). Both γ c and JAK-3 targeted mice have defects in the development of several hematopoietic cell lineages, in addition to their defective peripheral LN and PP development. Experiments have not yet been performed to determine whether it is the lack of certain BM-derived cell lineages or the lack of specific signaling via γ c or JAK-3 that leads to the impaired secondary lymphoid tissue development in these mutant mouse strains.

LYMPHOTOXIN IS REQUIRED FOR DEVELOPMENT OF NORMAL SPLENIC WHITE PULP STRUCTURE Membrane LT Establishes White Pulp B Cell / T Cell Segregation and B Cell Follicle Structure Not all elements of the secondary lymphoid system are lost in LTα −/− and LTβ −/− mice. Although LN and PP are dramatically deficient in mice lacking mLT, the spleen is retained. Together with the observation that Hox11-deficient mice have isolated absence of the spleen without impairment of LN or PP development, this observation confirms that the organogenesis of the LN and PP is controlled by signals different from those that control the organogenesis of

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SECONDARY LYMPHOID TISSUES

415

the spleen. In mLT-deficient mice, although the spleen forms, the organization of its white pulp compartment is grossly disturbed (48, 51, 58, 77–79). The overall size of the spleen is retained, but the white pulp nodules are generally reduced in size and their margins are blurred. The organized MAdCAM-1+ marginal sinus that normally separates the white pulp from the red pulp cannot be detected and appears to be absent. Similarly undetectable is the population of MOMA-1+ metallophilic macrophages that normally associates with the marginal sinus endothelium and that is thought to participate in the trafficking of antigen and cells from the blood circulation into the lymphoid compartment. The altered white pulp structure affects essentially all aspects of the normal organization, including disturbance of the normal segregation of the distinct T and B cell compartments. In LTα −/− mice, there is nearly complete loss of segregated B and T cell zones and apparent dissociation of the central arteriole from the T cell zone (48, 80). The B cells and T cells appear equally scattered throughout the white pulp nodule and B cell follicle structure is lost, including disappearance of the network of FDC that is normally located near the center of each follicle. Although the marginal sinus structure seems equally disturbed in the LTα −/− and LTβ −/− mice, the extent of ablation of normal T and B cell segregation appears more extreme in the LTα −/− mice. Recent studies by Alexopoulou et al (81) have shown that the grossly disturbed T cell/B cell segregation in LTα −/− mice can be repaired by breeding into the strain a TNF-expressing transgene. The ability of the TNF transgene to restore segregated B and T cell zones correlated with reduced TNF expression in the LTα −/− mice (81; A-S Johannson, DD Chaplin, unpublished data). Although the mechanism underlying this reduced TNF expression in the LTα −/− strain has not been definitively established, it is manifested at the level of TNF mRNA expression and may be a consequence of the retention in the LTα locus of the neor expression cassette used for gene targeting. Other investigators have observed effects of transgenes used for insertional gene targeting on loci upstream or downstream of the targeted locus (82). In the Alexopoulou et al study, the expression of the TNF transgene did not restore B cell follicle structure, but rather its action appeared limited to the formation of independent B and T cell zones. Although there is no loss of T and B zones in mice singly deficient in TNF (79, 83), this finding demonstrates that there can be interactions between LT and TNF in the formation of lymphoid tissue structure, and it underscores the general problem that gene targeting can alter the expression of loci adjacent to the intentionally targeted locus. Support for the coordinated action mLT and TNF also comes from studies showing that treatment of developing mouse embryos with soluble LTβR-Ig together with soluble TNFR-I-Ig had effects on the formation of lymphoid tissue structure that were not seen with treatment with either soluble receptor alone (60).

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

416

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

Disturbance of white pulp T and B cell segregation similar to that seen in LTα −/− and LTβ −/− mice was observed in LTβR−/− mice, which suggests that the LT-dependent signal for this segregation was delivered by the mLT heterotrimer (61). Consistent with this interpretation was the earlier finding that transgenic expression of a soluble LTβR-Ig fusion protein beginning shortly after birth resulted in similar loss of T cell/B cell segregation and follicle structure (57).

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Lymphotoxin-Dependent Establishment of Splenic T Cell and B Cell Zones Experiments in which either mixtures of lymphoid cells, purified cell populations, or BM are transferred from one strain of mice to another have been highly informative for defining rules that govern the compartmentalization of cells within the splenic white pulp. For example, mice carrying the severe combined immunodeficiency mutation (scid ), a result of mutation of the DNA-dependent protein kinase that functions during T and B cell antigen receptor rearrangement, have spleen white pulp nodules that are small, that are bordered by a marginal sinus, and that contain primarily NK and dendritic cells. It is interesting that when wt splenocytes were transferred into a scid recipient, within 1 day after infusion the transferred cells had partitioned into periarteriolar T cell and marginal B cell zones (84). In contrast, when wt splenocytes were transferred to a LTα −/− mouse, B cell/T cell segregation did not occur (80). These data suggest that in a B cell– and T cell–independent but LTα-dependent fashion, the underlying addressing mechanism that specifies segregated B and T cell zones could be laid down. In the scid mouse system, this maintenance of the addressing system appeared to require ongoing mLT expression, as pretreatment of the scid recipients for 1 week prior to transfer with soluble LTβR-Ig fusion protein ablated the ability of the transferred cells to segregate (84). This was not due to neutralization of mLT on the transferred splenocytes, because splenocytes from an LTα −/− donor could segregate effectively in an untreated scid recipient. Other investigators have found less plasticity in the ability of T and B cells to segregate into discrete zones in the white pulp. In earlier studies using BM from either a wt or a LTα −/− donor to reconstitute lethally irradiated wt recipients, it was found that similar to wt cells, the LTα-deficient cells segregated into discrete T and B cell zones (80). When BM from an LTα −/− donor was used, the crisp distinction between the T and B cell zones in the normal white pulp was partially lost, which suggests that LTα expression is required to retain complete T cell/B cell segregation; however, even 6 months after BM transfer, considerable segregation of B and T cells was still retained. This indicated that it was not necessary for B or T lymphocytes to express LTα in order to target existing white pulp B and T cell zones. In contrast, when BM from either wt or LTα −/− donors was used to reconstitute lethally irradiated LTα −/− recipients,

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

SECONDARY LYMPHOID TISSUES

417

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

segregation into discrete T and B cell areas did not occur. This suggested that the ability to form discrete white pulp B cell and T cell zones is a fixed feature of the microenvironment, imprinted by the time mice reach maturity. In contrast, if mice underwent blockage of mLT either during prenatal development by transplacental exposure to an LTβR-Ig fusion protein (55) or immediately after birth by transgenic expression of an LTβR-Ig fusion protein under the control of the cytomegalovirus promoter (57), then B cell/T cell segregation was ablated.

Partial Plasticity of B Cell Follicle Structure The studies described above showed that once the splenic white pulp has become populated with lymphocytes, the T cell/B cell segregation phenotype becomes relatively fixed. If segregated T cell and B cell zones have formed under the influence of mLT, then removal of mLT-expressing lymphocytes does not ablate segregation. If B and T cells are not segregated, because of congenital absence of LTα, then introduction of LTα-expressing cells cannot induce segregation. In dramatic contrast, cell transfer studies show that B cell follicle structure is highly plastic and dependent on the ongoing presence of LTα-expressing cells (80). When LTα −/− BM was used to reconstitute irradiated wt mice, B cell follicle structure, including the FDC network, was lost. In contrast, when irradiated LTα −/− mice were reconstituted with wt BM, the transferred cells were able to induce the formation of a robust FDC network. These FDC networks were competent to support the formation of functional germinal centers (GC) with isotype switching. Thus, FDC networks and B cell follicular structure are plastic, requiring ongoing expression of LTα for their maintenance, and responsive to the introduction of LTα-expressing cells into an LTα −/− environment. Similar results were obtained using mice doubly deficient in TNF and LTα (85).

Immune Dysfunction in LTα −/− Mice With their disturbed B cell and T cell compartmentalization and loss of normal B cell follicle structure, LTα −/− mice provide a model to examine the role of normal secondary lymphoid tissue structure on immune responsiveness. When LTα −/− mice were immunized with sheep red blood cells (SRBC), they produced high levels of antigen-specific IgM but low or no antigen-specific IgG in either primary or secondary responses (80). To test whether this failure to produce class-switched serum IgG antibody was due to the altered splenic microarchitecture in these mice, or perhaps to a requirement for LTα expression by lymphocytes cooperating in the antibody response, reciprocal splenocyte and BM transfers were performed. When irradiated LTα −/− mice were reconstituted with wt splenocytes and immunized immediately with SRBC, splenic

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

418

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

FU & CHAPLIN

Table 1 Correlation between the presence of LTα-expressing cells and the appearance of FDC, GC, and IgG responsesa Altered spleen

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Donorb

Cells transferredc

Recipient

Microarchitecture

Short-term reconstitution +/+ Spleen +/+ Spleen −/− Spleen −/− Spleen

+/+ −/− +/+ −/−

No Yes No Yes

Long-term reconstitution +/+ BM +/+ BM −/− BM −/− BM

+/+ −/− +/+ −/−

No Yes No Yes

T cell/B cell zones

FDCd

GC

IgG

No Yes No Yes

++ — + —

++ +/− ++ —

++ — ++ —

No Yes Slight Yes

++ ++ — —

++ ++ — —

++ ++ — —

a LT, Lymphotoxin; FDC, follicular dendritic cells; GC, germinal centers; Ig, immunoglobulin; BM, bone marrow. b +/+ and −/−, Wild-type and LTα −/− mice, respectively. c For transfer of spleen cells, a suspension of viable cells was prepared from a single donor spleen and infused intravenously with 108 sheep red blood cells (SRBC) 3 h after the recipient had been irradiated with 750 rad. For BM transfer, the recipients were lethally irradiated (1050 rad) and reconstituted with BM, then immunized with SRBC 6–8 weeks later. Tissue and sera were collected 10 days after immunization. Anti-SRBC IgG was determined by ELISA. Anti-Thy1.2 and anti-B220 were the markers for T cells and B cells, anti-CR1/2 or FDC-M1 for FDC, and peanut agglutinin for GC. d ++, Similar to unirradiated wild-type mice; +, weak staining or IgG response; —, no response.

microarchitecture remained disturbed and there was no IgG response. This suggests that merely providing mature immune cells, including T cells, B cells, NK cells, and macrophages, is not sufficient to generate a productive IgG response. In contrast, when irradiated wt animals received splenocytes from LTα −/− mice, follicle structure and a strong IgG response were retained (Table 1). Thus, LTα-deficient B cells and T cells have no intrinsic defect in their ability to generate an IgG response. Rather, the altered microenvironment characteristic of LTα −/− mice appears to impair the ability to switch to a productive IgG response. Further studies using BM transfer, as discussed above, showed that when wt BM was transferred to LTα −/− recipient mice, B cell follicles were, over time, restored, and the ability to sustain an isotype-switched IgG response was also, over time, restored (80).

MULTIPLE SIGNALS ARE REQUIRED FOR THE FORMATION OF MATURE B CELL FOLLICLES The studies reviewed above indicate that LTα expression is needed inductively for the formation of functional B cell follicles, and also tonically for the

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SECONDARY LYMPHOID TISSUES

419

maintenance of mature follicles with FDC clusters that can support the formation of a high-affinity, isotype-switched Ig response. Absence of FDC networks and B cell follicles is also seen in LTβ −/− mice (58, 79). The similar primary follicle phenotype of LTα −/− and LTβ −/− mice suggests that the form of LT required to develop and maintain the FDC network is the membrane LTα 1β 2 heterotrimer. Support for this comes from experiments in which specific neutralization of membrane LT in mice expressing transgenic LTβR-Ig fusion protein (57) or treated with soluble LTβR-Ig fusion protein (55, 86) shows loss of B cell follicles. Final confirmation that membrane LT provides essential signals for the development of the FDC network has come from analyses of mice deficient in LTβR that also show absence of morphologically defined FDC networks, GC, and IgG responses (61). In addition to signals through the LTβR, signals via TNFR-I are also required for the formation of FDC networks and GC (77, 87). The signals that are delivered by TNFR-I are likely delivered by TNF as the ligand, because mice deficient in TNF also show absence of FDC clusters (83, 88). Thus, signaling via both the TNF/TNFR-I axis and the mLT/LTβR axis appears to be required for development of FDC networks. Consistent with this, both LTβR and TNFR-I are expressed on FDC (89, 90). Although LTα −/−, LTβ −/−, and TNF−/− mice all show failure to form splenic FDC networks, this does not mean that mLT and TNF deliver identical tissue morphogenic signals. LTα −/−, LTβ −/−, and LTβR−/− mice manifest profound disturbances of splenic T cell/B cell segregation (58, 61, 80), whereas TNF−/− and TNFR-I−/− mice retain segregated T cell/B cell zones (83, 87, 91). Thus, in the spleen, the dominant role of TNF appears to concern the formation of primary B cell follicles. Both TNF−/− and TNFR-I−/− mice have impaired development of primary B cell follicles, whereas the distribution of B cells in the marginal zone appears intact. We have observed that the formation of morphologically defined and functionally defined primary B cell follicles is linked to the formation of FDC clusters (Table 1). It remains an interesting question whether a lack of organized FDC prevents the formation of primary B cell follicles or whether some other developmental failure leading to a lack of primary B cell follicles prevents the formation of clusters of FDC. The studies outlined above indicate clearly that signaling through both the LTβR and TNFR-I is required for the formation of FDC networks within splenic primary follicles. These findings must be reconciled with observations comparing the ability of soluble LTβR-Ig and TNFR-I–Ig fusion proteins to disrupt spleen primary follicle structure (86). When LTβR-Ig was administered to wt mice in an intraperitoneal dose of 100 µg once a week, after 2–4 weeks extensive disruption of the white pulp structures was observed, with loss of T cell/B cell organization, loss of expression of several markers of marginal sinus

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

420

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

structure, and failure to form morphologically detectable GC. These changes were accompanied by an inability to generate isotype-switched Ig responses following immunization with SRBC. Although the distribution of FDC was not specifically studied, it can be assumed that the FDC network was ablated following treatment with the soluble receptor because, in other studies, withdrawal of membrane LT signaling resulted in loss of the FDC reticulum (80). In contrast, when wt mice were injected twice a week for up to 3 weeks with up to 300 µg of TNFR-I–Ig fusion protein, there was no detectable interference with either the GC reaction or the generation of an isotype-switched antibody response (86). An explanation for the apparent discrepancy between these data and those obtained using genetically deficient mice may relate to the time courses of the different experiments. It may be that in order to affect FDC function and the maturation of the immunoglobulin response, TNF blockade must be more prolonged than blockade of membrane LT. Perhaps TNF signaling is required for the induction but not the maintenance of an FDC network, whereas mLT signaling appears to be required for both induction and maintenance. In this case, if FDC have a relatively long half-life, then blocking TNF signaling would have a much less dramatic impact compared with blocking mLT signaling. Or, more trivially, it may be impossible to block the presumed cell-cell interactions signaled by TNF/TNFR-I using this kind of soluble receptor reagent. This may be particularly true because biological potency of the soluble TNFR-I–Ig fusion proteins has generally been verified for the neutralization of soluble TNF. The ability of the soluble reagent to block the biological actions of membrane TNF is less well defined. Of relevance here are studies that have demonstrated that membrane TNF signals with selectivity through TNFR-II (92). This may be important because other studies have shown that the disturbed spleen follicle phenotype of TNF−/− mice can be restored by selective transgenic expression of the membrane form of TNF (83). Thus, treatment of mice with soluble TNFR-I–Ig might fail to block signals mediated by membrane TNF, and administration of soluble TNFR-II–Ig might produce a different result compared with administration of soluble TNFR-I–Ig.

THE LT-DEPENDENT SIGNALS THAT SUPPORT FORMATION OF THE FDC NETWORK ARE DELIVERED BY B CELLS The studies described above indicate clearly that the LTα expressing cells that are required for the formation and maintenance of the FDC network are BM derived. Although the full repertoire of LTα-expressing cells has not been rigorously defined, the major cell lineages known to produce LTα are all BM derived and include T, B, and NK cells (28, 93). Recent studies have begun to dissect

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SECONDARY LYMPHOID TISSUES

421

the nature of the cell lineage that delivers the LTα-dependent signal leading to the apparently de novo induction of splenic FDC networks (94). Immunohistochemical analyses of the spleens of T cell–deficient mice (with targeted null mutations of the genes encoding both the T cell receptor β and δ chains) show white pulp nodules consisting primarily of B lymphocytes, NK cells, and dendritic cells, with the majority of nodules containing clusters of FDC. This established that neither αβ nor γ δ T cells are required for either the induction or maintenance of FDC clusters. In contrast, the spleens of B cell–deficient mice (with targeted null mutation of the Ig heavy-chain locus) contained white pulp nodules consisting primarily of T lymphocytes, NK cells, and dendritic cells and devoid of detectable FDC. These results indicated that B cells were absolutely required for the development of FDC clusters (Table 2). Within 3 weeks of intravenous infusion of purified wt splenic B cells into recombination activating gene-1 (RAG-1)–deficient mice, FDC clusters were induced in both spleen and LN. Treatment of the recipient mice with LTβR-Ig fusion protein at the time of B cell transfer blocked the development of the FDC clusters, establishing a requirement for mLT in the reconstitution process. To exclude the possibility that NK cells (present in the T cell–deficient and RAG-1−/− mice) might play an essential role in the induction or maintenance of FDC networks, CD3ε transgenic mice lacking both mature T and mature NK cells (95) were investigated. Robust FDC clusters were found in the splenic white pulp of the T cell– and NK cell–deficient mice (Table 2). Results supporting a key role for LTα-expressing B cells have been obtained elsewhere (84). Together these studies show that B cells, in a LTα-dependent fashion, provide a signal that induces Table 2 B cells, but not T cells or NK cells, are required for the development of the splenic white pulp FDC networka Miceb

Lymphocytesc

B cell folliclesd

FDC

MAdCAM-1

wt RAG-1−/− BCR−/− TCR−/− CD3ε trangene JAK3−/−

T, B, NK NK NK, T B, NK B Few

+ — — + + —

+ — — + + —

+ +/− +/− + + —

a NK, Natural killer; FDC, follicular dendritic cells; MAdCAM-1, mucosal address in cell adhesion molecule-1. b Mouse strains tested: wt, wild type; RAG-1−/−, recombination activating gene 1 deficient; BCR−/−, B cell receptor deficient (immunoglobulin H targeted); TCR−/−, T cell receptor deficient (TCRβ and δ targeted); CD3ε transgene, overexpressing CD3ε (95); JAK3−/−, Janus kinase 3 deficient (75). c Lymphocyte populations present in the spleen of the indicated mouse strain. d +, Detectable at a level similar to wt mice; +/−, weakly positive staining; —, not detectable.

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

422

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

and maintains clusters of FDC in spleen and LN. Mature T cells and NK cells are not required for this to occur. It is important, however, to keep in mind that transgenic and gene targeted mice that manifest congenital absence of selective cell lineages may continue to be populated by immature cells from the affected lineage. So, for example, the CD3ε transgenic mice that have no detectable NK cell activity and no mature circulating or tissue NK cells may continue to contain NK precursors. Although no biological functions of such precursors have yet been defined, it remains possible that they may, during their own maturation, contribute to the maturation of other lineages and tissue environments. Identifying the B cell as the cell responsible for delivering an mLT signal that induces the formation of FDC clusters expands the repertoire of important B cell functions. The role of the B cell is not limited solely to the presentation of antigen to T cells and the production of antibodies to neutralize and eliminate these antigens. B cells also act by delivering key signals that support the development of the lymphoid tissue structure that is itself required for the B cell to express its mature functions. Whether B cells express mLT or not, and whether they gain access to sites appropriate for the induction of FDC clusters, will have a profound impact on the quality of the immune response. The target cell with which the B cell interacts to induce mature FDC structure has not been defined. Given that LTβR and TNFR-I are both expressed on FDC themselves (89, 90), it is reasonable to assume that the B cell delivers its LT signal directly to the FDC precursor. The differentiation pathway that leads to the formation of mature FDC is poorly defined. The elusive FDC precursor has been thought not to be derived from hematopoietic stem cells but rather to belong to a fibroblast-like cell (96–101). Recently, however, it was demonstrated that FDC can be transferred to newborn scid recipients using either BM or fetal liver (10). Whether the FDC developed from a hematopoietic precursor or a stromal precursor remains undefined. With the understanding that B cells provide key signals supporting the development of primary lymphoid follicle structure, we can begin to make predictions about the types of molecules whose function will be required for normal primary follicles to form. Any molecule that is required for normal B cell development or to render B cells competent to deliver an mLT signal will also be required for the development of primary follicle structures. Although many studies have identified molecules that are required for normal B cell development and activation, most of them have used as their functional readout the ability to form GC and serum isotype-switched antibodies. The requirements for the formation of a GC can be anticipated to be different from the requirements to form a mature FDC network. GC are thought to depend for their formation on productive interactions between B cells, T cells, FDC, and perhaps other antigen-presenting cells (25, 102). Loss of any molecule essential for these interactions is expected

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SECONDARY LYMPHOID TISSUES

423

to lead to loss of secondary B cell follicles and GC, without a priori disrupting the FDC network. Molecules that are now known to support the development of GC structures include the intercellular signaling proteins CD40 and CD40L (103), CD19 (104), CD28 (105), and B7-2 (106), among others. These molecules have the potential to act directly at the level of the B cell/T cell interactions that are required for maturation of B cells into GC cells. That additional intracellular signaling molecules are required for the development of the GC reaction has also been recognized but without a clear understanding of which cell lineage(s) they affect. These additional molecules include members of the NF-kB/rel family, with p52 (107, 108) and Bcl-3 (109, 110) both required for the formation of GC. Mice deficient in the transcriptional repressor Bcl-6 also show failure to form GC (111, 112). Alterations in the formation of GC are not always absolute. For example, mice deficient in the complement receptors CR1 and CR2 (encoded in mice as alternatively spliced products of a single gene) show underdeveloped GC and manifest reduced but not absent primary and secondary IgG responses (113). Because the complement receptors encoded by this gene are expressed on both FDC and B cells, their absence could affect GC formation at multiple levels. For many of these factors, it remains unclear whether they are required to support cellular interactions necessary for the activation of GC B cells or rather to support the development of follicular structure (such as the FDC network) that is required for GC to develop.

WHAT CELLS DELIVER AND RECEIVE LT SIGNALS DURING SECONDARY LYMPHOID TISSUE ORGANOGENESIS? We know that B lymphocytes provide a crucial membrane LT signal that is required for the formation of the FDC network in primary and secondary B cell follicles. We know little about the nature of the cell type(s) that deliver mLT signals that specify the initial organogenesis process itself. Similarly, we have little information about the nature of the mLT-expressing cells that support the formation of segregated T and B cell zones in the splenic white pulp. For these processes, we do not know whether the mLT-dependent signals are required for the earliest stages of lymphoid organ biogenesis, or whether they are required to nurture nascent lymphoid organs. Regardless of when the mLT-dependent signal leads to commitment for LN, PP, and organized splenic white pulp structure, it is clear that two of the major known mLT-expressing cell lineages, mature B and T lymphocytes, cannot be essential for delivering this mLT-dependent signal. This is based on the observation that both scid and RAG-1−/− mice have intact development of LN and are able to segregate

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

424

17:35

QC: ARS/uks

Annual Reviews

T1: KKK

AR078-14

FU & CHAPLIN

transferred B and T cells within their splenic white pulp, in spite of their absence of mature B and T cell lineages. In addition, the initial steps of the organogenesis of secondary lymphoid tissues appear to precede the maturation of T and B lymphocytes during ontogeny. This suggests again that mature T and B cells may not be required for the formation of these tissues. NK cells, one of the non-T and non-B cell lineages that are retained in scid and RAG-1−/− mice, are also able to express LT, and cells of the NK lineage should certainly be considered as cells that might use mLT as one of the signals for the initiation of development of normal secondary lymphoid tissue structure. The possibility that an immature cell of one of the hematopoietic lineages (B cell, T cell, NK cell, or uncommitted) might provide a critical mLT-dependent signal should be kept in mind. Precursor B and T lineage cells are thought to be present in normal or greater than normal numbers in scid and RAG-1−/− mice. A potential role for NK cells in LN and PP biogenesis was suggested not only by the recognition that NK cells represent an LT-expressing lineage that is retained in scid and RAG-1−/− mice, but also by the finding that LN and PP are absent or present in reduced numbers in mice carrying mutations that lead to loss of mature NK cells [γ c−/− (71, 72), JAK-3−/− (74–76), and CD3ε transgenic (95); Y-X Fu, JW Verbsky, G Huang, DD Chaplin, unpublished data]; however, we have been unable to block the development of LN or PP by depletion of NK cells in the developing fetus by administering various anti-NK cell antibodies to pregnant mice. Because these antibodies only bound to mature NK cells, it remains possible that immature NK cell precursors might contribute to delivering the inductive signals for LN and PP formation. In this regard, the recent observation of CD4+CD3−LTαβ + cells in the LN of developing mouse embryos may be of particular significance (114). These cells express α4β7 integrin and appear to differentiate into NK cells, dendritic antigen-presenting cells, and follicular cells but not into T or B lymphocytes. Although they have only been identified within already developing LN structures, they have many of the characteristics of the cell type that might deliver the mLT-dependent signal for the earliest stages of secondary lymphoid tissue commitment. It is important to keep in mind that the commitment to the formation of different lymphoid tissues may occur at different times, may require the interactions of different types of cells, and may involve different intercellular and intracellular signaling molecules. The temporal distribution of commitment events has been shown in the development of PP and different sets of LN, which was blocked depending on the time of administration of the LTβR-Ig fusion protein (55). Additional studies in which LTα was expressed in LTα −/− mice under the control of the rat insulin promoter showed that the development of some but not all LN was restored (115, 116). The molecular basis for partial restoration of LN structures in this model remains undefined, but it underscores the hypothesis

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SECONDARY LYMPHOID TISSUES

425

that selective spacial or temporal expression of LT determines which secondary lymphoid tissues are formed. In addition, different secondary lymphoid tissues may require a different complex of signals for their development. For example, mice deficient in BLR1 have selective loss of inguinal LN (64), and absence of mature NK cells correlates with a more extensive but still partial failure to form LN (71, 72, 74–76, 95; Y-X Fu, JW Verbsky, G Huang, DD Chaplin, unpublished data). Finally, lack of LTα leads to an essentially complete failure to form LN and PP (48, 49). Further definition of the cellular interactions required during the early steps of secondary lymphoid tissues biogenesis will be difficult because of the small numbers of cells that are likely to be involved, and the general inaccessibility of embryonic mammalian tissues to provocative manipulation. Nevertheless, identification of candidate cell lineages such at the CD4+CD3−LTαβ + cells seen in embryonic LN (114) may provide new leverage to track the formation of the tissues.

INDEPENDENT SIGNALS REGULATE DEVELOPMENT OF PRIMARY AND SECONDARY FOLLICLE STRUCTURE IN SPLEEN AND LN Additional data demonstrating the complexity of signals controlling secondary lymphoid tissue structure come from studies comparing the structures of mesenteric LN and spleen in both LTα −/− and TNFR-I−/− mice (51). Consistent with data described above, the splenic white pulp of the LTα −/− mice showed loss of discrete T and B cell zones, loss of the MAdCAM-1–staining marginal sinus, loss of discrete B cell follicles and FDC networks, and loss of the ability to generate peanut agglutinin+ (PNA+) GC B cell clusters after immunization with T cell–dependent antigens. The spleens of TNFR-I−/− mice (and TNF−/− mice) (Y-X Fu, MW Marino, DD Chaplin, unpublished data) showed a similar absence of marginal sinus MAdCAM-1 staining, lack of discrete B cell follicles and FDC networks, and lack of splenic GC, but they retained substantial segregation of T and B cell zones. A small fraction of the LTα −/− mice have a single mesenteric LN. It is interesting that although the spleens in these mice show grossly disturbed B cell/T cell segregation, the mesenteric LN, when it develops, shows segregation of B cell and T cell zones similar to that seen in the mesenteric LN of wt mice. In stark contrast, in TNFR-I−/− mice, although the spleens show preserved B cell/T cell segregation, the mesenteric LN is totally disordered, with no distinct B and T cell areas (Table 3). Of additional interest, although both LTα −/− and TNFR-I−/− mice fail to form GC in the spleen following intraperitoneal or intravenous immunization with SRBC, they both manifested prominent GC-like clusters of PNA+ cells in their mesenteric LN. To study further the role of LTα in the formation of these PNA+

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

426

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

FU & CHAPLIN Table 3 LTα and TNF independently regulate the development of primary and secondary follicle structure in spleen and lymph nodesa

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T/B zone

FDC clusters

GC

Mice

Spleen

MLN

Spleen

MLN

Spleen

MLN

wt LTα −/− wt to RAG-1−/− LTα −/− to RAG-1−/− TNF−/− TNFR-I−/−

+ — + +/− + +

+ + + + — —

+ — + — — —

+ — + — — —

+ + + — — —

+ + + + + +

a LT, Lymphotoxin; TNF, tumor necrosis factor; FDC, follicular dendritic cells; GC, germinal centers MLN, mesenteric lymph node; +, similar to wt; +/−, modestly disorganized; —, disorganized or undetectable.

clusters of GC-like cells in the mesenteric LN, RAG-1α −/− mice were treated with an infusion of spleen cells from either LTα −/− or wt mice and immediately immunized with SRBC. Ten days later, the RAG-1−/− mice reconstituted with wt splenocytes showed typical PNA+ GC in their reconstituted spleens, whereas RAG-1−/− mice reconstituted with LTα −/− splenocytes did not. RAG-1−/− mice reconstituted with either wt or LTα −/− splenocytes showed robust PNA+ clusters in their reconstituted mesenteric LN (Table 3). Comparison of the LTα −/− and the TNFR-I−/− mice shows that the signals that regulate the development of T and B cell zones are different in the spleen and the mesenteric LN. They show further that the signals that regulate B cell activation to produce clusters of PNA+ cells differ between the spleen and mesenteric LN. The nature of the PNA+ clusters that are induced in the mesenteric LN of these targeted mice following immunization with SRBC remains unclear. They form in the absence of detectable clusters of FDC and are not associated with the production of antigen-specific serum IgG. They, therefore, probably do not represent functionally intact GC but rather are likely to represent partially activated clusters of proliferating cells. Further analysis of these structures may provide insight into the activation steps required for the generation of the normal GC response. Comparison of the LTα −/− and the TNFR-I−/− mice provides strong evidence that the rules governing the development of organized structure in the spleen and mesenteric LN are different, and that the signals required for activation of B cells in these two compartments are also distinct. This may be one of the morphological correlates of the prior observations that different lymphoid tissues express distinct cytokine profiles following antigen stimulation (117), and that the quality of the immune response may be quite different depending on whether the responding lymphocytes are activated in the LN or in the spleen (118).

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

SECONDARY LYMPHOID TISSUES

427

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONCLUDING REMARKS The membrane form of LT is now recognized as providing crucial signals required for the biogenesis of secondary lymphoid tissues and primary B cell follicles within these structures. TNF mediates independent signals for primary B cell follicle structure and can also cooperate with mLT contributing to other aspects of secondary lymphoid tissue development. Additional molecules participating in the development of normal secondary lymphoid tissue structure are being identified (summarized in Table 4). Some of the actions of mLT lead to the development of fixed characteristics of the lymphoid tissues, but some other structural features are plastic, indicating a need for ongoing expression of mLT to sustain normal tissue elements. This is particularly true for the formation of FDC clusters throughout the secondary lymphoid tissues. Withdrawal of mLT-expressing cells results over a period of days to weeks in loss of recognizable FDC and loss of primary B cell follicle structure. Of special interest, B cells have been identified as the lineage that delivers the mLT signal for formation and maintenance of the FDC network. Thus, B cells use mLT as an inductive signal for the formation of the lymphoid tissue structure that is Table 4 Phenotypic effects of LTα, LTβ, and TNF on lymphoid tissue developmenta Mutant strain or treatment with soluble receptor

Lymph nodes

Peyer’s patches

Segregated splenic B and T zones Disorganized

LTα −/−

Absentb

Absent

LTβ −/− TNF−/− TNFR-I−/−

Absentc Present Present

TNFR-II−/− LTβR−/− LTβR-Ig

Present Absent Partially absente Present

Present

Absent

Absent

TNFR-I–Ig LTβR-Ig+ TNFR-I–Ig aly/aly BLR1−/−

Splenic marginal zones

References

Absent

Absent

Absent Disorganized Reduced Retained Reduced Retained

Disorganized Absent Enlarged Absent ND Absent

Absent Absent Absent

Present Absent Absentd

ND Present Disorganized Absent Disorganized Absent

Present Absent Absent

48, 49, 51, 77, 80, 88, 94, 101 58 83, 88 62, 63, 77, 87, 101 53, 77 61 55–57, 86

Absent Absent Partially Present absentg

Disorganized

Primary B cell FDC follicles networks

Retained Disorganized Mildly disorganized Mildly disorganized ND

Disorganized

Absentf

Absentf

57, 86

ND

Absent

Absent

60

Disorganized ND

Disorganized ND ND ND

ND ND

34 64

a LT, Lymphotoxin; TNF, tumor necrosis factor; FDC, follicle dendritic cells; R, receptor; ND, not determined; Ig, immunoglobulin; BLR1, Burkitt’s lymphoma receptor-1. b Mesenteric lymph nodes (LN) retained in a small fraction of LTα −/− mice. c Mesenteric and cervical LN retained in all LTβ −/− mice. d Peyer’s patches ablated if LTβR-Ig Fc administered at birth or earlier. e Mesenteric, cervical, lumbar, and sacral LN resistant to ablation. f FDC networks and primary follicles sensitive during embryonic development. g Absence of inguinal LN.

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

428

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

FU & CHAPLIN

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

required for the B cell to express its fully differentiated functions. The actions of mLT and TNF in signaling the formation of secondary lymphoid tissues are mediated primarily through the LTβR and TNFR-I, but the participation of additional, recently defined receptors in this family has not yet been excluded (119). Analyses of mice with selective abnormalities of components of secondary lymphoid tissue structures are providing a unique opportunity to relate lymphoid tissue structure and immune responsiveness. Detailed definition of the signals that sustain and modulate lymphoid macro- and microarchitecture may identify fruitful new targets for immunomodulating drug therapy. ACKNOWLEDGMENTS David Chaplin is an investigator of the Howard Hughes Medical Institute. DD Chaplin and Y-X Fu are supported by grants from the National Institutes of Health. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Goodnow CC. 1997. Chance encounters and organized rendezvous. Immunol. Rev. 156:5–10 2. MacLennan IC, Gulbranson-Judge A, Toellner KM, Casamayor-Palleja M, Chan E, Sze DM, Luther SA, Orbea HA. 1997. The changing preference of T and B cells for partners as T-dependent antibody responses develop. Immunol. Rev. 156:53–66 3. Steinman RM, Pack M, Inaba K. 1997. Dendritic cells in the T-cell areas of lymphoid organs. Immunol. Rev. 156:25–37 4. Castenholz A. 1990. Architecture of the lymph node with regard to its function. Curr. Top. Pathol. 84:1–32 5. Witmer MD, Steinman RM. 1984. The anatomy of peripheral lymphoid organs with emphasis on accessory cells: lightmicroscopic immunocytochemical studies of mouse spleen, lymph node, and Peyer’s patch. Am. J. Anat. 170:465–81 6. Timens W. 1991. The human spleen and the immune system: not just another lymphoid organ. Res. Immunol. 142:316– 20 7. Kundig TM, Bachmann MF, DiPaolo C, Simard JJ, Battegay M, Lother H, Gessner A, Kuhlcke K, Ohashi PS, Hengartner H, Zinkernagel RM. 1995. Fibroblasts as efficient antigen-presenting cells in lymphoid organs. Science 268:1343–47

8. Zinkernagel RM, Ehl S, Aichele P, Oehen S, Kundig T, Hengartner H. 1997. Antigen localisation regulates immune responses in a dose- and time-dependent fashion— a geographical view of immune reactivity. Immunol. Rev. 156:199–209 9. Burton GF, Conrad DH, Szakal AK, Tew JG. 1993. Follicular dendritic cells and B cell costimulation. J. Immunol. 150:31– 38 10. Kapasi ZF, Qin D, Kerr WG, KoscoVilbois MH, Shultz LD, Tew JG, Szakal AK. 1998. Follicular dendritic cell (FDC) precursors in primary lymphoid tissues. J. Immunol. 160:1078–84 11. Liu Y-J, Xu J, de Bouteiller O, Parham CL, Grouard G, Djossou O, de SaintVis B, Lebecque S, Banchereau J, Moore KW. 1997. Follicular dendritic cells specifically express the long CR2/ CD21 isoform. J. Exp. Med. 185:165– 70 12. Tew JG, Wu JH, Qin DH, Helm S, Burton GF, Szakal AK. 1997. Follicular dendritic cells and presentation of antigen and costimulatory signals to B cells. Immunol. Rev. 156:39–52 13. Gretz JE, Anderson AO, Shaw S. 1997. Cords, channels, corridors and conduits: critical architectural elements facilitating cell interactions in the lymph node cortex. Immunol. Rev. 156:11–24

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SECONDARY LYMPHOID TISSUES 14. Butcher EC, Picker LJ. 1996. Lymphocyte homing and homeostasis. Science 272:60–66 15. Kuper CF, Koornstra PJ, Hameleers DM, Biewenga J, Spit BJ, Duijvestijn AM, van Breda Vriesman PJ, Sminia T. 1992. The role of nasopharyngeal lymphoid tissue. Immunol. Today 13:219–24 16. Sminia T, van der Brugge-Gamelkoorn GJ, Jeurissen SH. 1989. Structure and function of bronchus-associated lymphoid tissue (BALT). Crit. Rev. Immunol. 9:119–50 17. Neutra MR, Kraehenbuhl JP. 1992. M cell-mediated antigen transport and monoclonal IgA antibodies for mucosal immune protection. Adv. Exp. Med. Biol. 327:143–50 18. Brown AR. 1992. Immunological functions of splenic B lymphocytes. Crit. Rev. Immunol. 11:395–417 19. Yamauchi H. 1968. Estimation of blood flow in the Banti spleen on anatomical basis. Tohoku. J. Exp. Med. 95:63–77 20. Kraal G, Schornagel K, Streeter PR, Holzmann B, Butcher EC. 1995. Expression of the mucosal vascular addressin, MAdCAM-1, on sinus-lining cells in the spleen. Am. J. Pathol. 147:763–71 21. Kraal G, Janse M. 1986. Marginal metallophilic cells of the mouse spleen identified by a monoclonal antibody. Immunology 58:665–69 22. Chaplin DD, Fu Y-X. 1998. Cytokine regulation of secondary lymphoid organ development. Curr. Opin. Immunol. 10:289–97 23. van Krieken JH, von Schilling C, Kluin PM, Lennert K. 1989. Splenic marginal zone lymphocytes and related cells in the lymph node: a morphologic and immunohistochemical study. Hum. Pathol. 20:320–25 24. Oldfield S, Liu YJ, Beaman M, MacLennan CM. 1988. Memory B cells generated in T cell-dependent antibody responses colonise the splenic marginal zone. Adv. Exp. Med. Biol. 237:93–98 25. MacLennan IC. 1994. Germinal centers. Annu. Rev. Immunol. 12:117–39 26. Karrer Y, Althage A, Odermatt B, Roberts CW, Korsmeyer SJ, Miyawaki S, Hengartner H, Zinkernagel RM. 1997. On the key role of secondary lymphoid organs in antiviral immune responses studied in alymphoplastic (aly/aly) and spleenless (hox11(−/−)) mutant mice. J. Exp. Med. 185:2157–70 27. Flanagan SP. 1966. ‘Nude,’ a new hairless gene with pleoptropic effects in the mouse. Genet. Res. 8:295–309

429

28. Ware CF, VanArsdale TL, Crowe PD, Browning JL. 1995. The ligands and receptors of the lymphotoxin system. Curr. Top. Microbiol. Immunol. 198:175–218 29. Nehls M, Kyewski B, Messerle M, Waldschutz R, Schuddekopf K, Smith AJH, Boehm T. 1996. Two genetically separable steps in the differentiation of thymic epithelium. Science 272:886–89 30. Schuddekopf K, Schorpp M, Boehm T. 1996. The whn transcription factor encoded by the nude locus contains an evolutionarily conserved and functionally indispensable activation domain. Proc. Natl. Acad. Sci. USA 93:9661–64 31. Schlake T, Schorpp M, Nehls M, Boehm T. 1997. The nude gene encodes a sequence-specific DNA binding protein with homologs in organisms that lack an anticipatory immune system. Proc. Natl. Acad. Sci. USA 94:3842–47 32. Searle AG. 1964. The genetics and morphology of two “luxoid” mutants in the house mouse. Genet. Res. 5:171–97 33. Suto J, Wakayama T, Imamura K, Goto S, Fukuta K. 1995. Incomplete development of the spleen and the deformity in the chimeras between asplenic mutant (Dominant hemimelia) and normal mice. Teratology 52:71–77 34. Miyawaki S, Nakamura Y, Suzuka H, Koba M, Yasumizu R, Ikehara S, Shibata Y. 1994. A new mutation, aly, that induces a generalized lack of lymph nodes accompanied by immunodeficiency in mice. Eur. J. Immunol. 24:429–34 35. Shinkura R, Matsuda F, Sakiyama T, Tsubata T, Hiai H, Paumen M, Miyawaki S, Honjo T. 1996. Defects of somatic hypermutation and class switching in alymphoplasia (aly) mutant mice. Int. Immunol. 8:1067–75 36. Adachi S, Yoshida H, Kataoka H, Nishikawa S. 1997. Three distinctive steps in Peyer’s patch formation of murine embryo. Int. Immunol. 9:507–14 37. Roberts CW, Shutter JR, Korsmeyer SJ. 1994. Hox11 controls the genesis of the spleen. Nature 368:747–49 38. Roberts CW, Sonder AM, Lumsden A, Korsmeyer SJ. 1995. Development expression of Hox11 and specification of splenic cell fate. Am. J. Pathol. 146:1089– 101 39. Ruddle NH. 1992. Tumor necrosis factor (TNF-α) and lymphotoxin (TNFβ). Curr. Opin. Immunol. 4:327–32 40. Kriegler M, Perez C, DeFay K, Albert I, Lu SD. 1988. A novel form of TNF/cachectin is a cell surface cytotoxic transmembrane protein: ramifications for

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

430

41.

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

42.

43.

44.

45.

46.

47.

48.

49.

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

FU & CHAPLIN the complex physiology of TNF. Cell 53:45–53 Black RA, Rauch CT, Kozlosky CJ, Peschon JJ, Slack JL, Wolfson MF, Castner BJ, Stocking KL, Reddy P, Srinivasan S, Nelson N, Boiani N, Schooley KA, Gerhart M, Davis R, Fitzner JN, Johnson RS, Parton RJ, March CJ, Cerretti DP. 1997. A metalloproteinase disintegrin that releases tumour-necrosis factorα from cells. Nature 385:729–33 Moss ML, Jin S-LC, Milla ME, Burkhart W, Carter HL, Chen W-J, Clay WC, Didsbury JR, Hassler D, Hoffman CR, Kost TA, Lambert MH, Leesnitzer MA, McCauley P, McGeehan G, Mitchell J, Moyer M, Pahel G, Rocque W, Overton LK, Schoenen F, Seaton T, Su J-L, Warner J, Willard D, Becherer JD. 1997. Cloning of a disintegrin metalloproteinase that processes precursor tumour-necrosis factorα. Nature 385:733–36 Browning JL, Androlewicz MJ, Ware CF. 1991. Lymphotoxin and an associated 33kDa glycoprotein are expressed on the surface of an activated human T cell hybridoma. J. Immunol. 147:1230–37 Browning JL, Ngam-ek A, Lawton P, DeMarinis J, Tizard R, Chow EP, Hession C, O’Brine-Greco B, Foley SF, Ware CF. 1993. Lymphotoxin β, a novel member of the TNF family that forms a heteromeric complex with lymphotoxin on the cell surface. Cell 72:847–56 Browning JL, Dougas I, Ngam-ek A, Bourdon PR, Ehrenfels BN, Miatkowski K, Zafari M, Yampaglia AM, Lawton P, Meier W, Benjamin CP, Hession C. 1995. Characterization of surface lymphotoxin forms. Use of specific monoclonal antibodies and soluble receptors. J. Immunol. 154:33–46 Lawton P, Nelson J, Tizard R, Browing JL. 1995. Characterization of the mouse lymphotoxin-β gene. J. Immunol. 154:239–46 Crowe PD, VanArsdale TL, Walter BN, Ware CF, Hession C, Ehrenfels B, Browning JL, Din WS, Goodwin RG, Smith CA. 1994. A lymphotoxin-β-specific receptor. Science 264:707–10 De Togni P, Goellner J, Ruddle NH, Streeter PR, Fick A, Mariathasan S, Smith SC, Carlson R, Shornick LP, StraussSchoenberger J, Russell JH, Karr RW, Chaplin DD. 1994. Abnormal development of peripheral lymphoid organs in mice deficient in lymphotoxin. Science 264:703–7 Banks TA, Rouse BT, Kerley MK, Blair PJ, Godfrey VL, Kuklin NA, Bouley DM,

50.

51.

52.

53.

54.

55.

56.

57.

58.

Thomas J, Kanangat S, Mucenski ML. 1995. Lymphotoxin-α-deficient mice. Effects on secondary lymphoid organ development and humoral immune responsiveness. J. Immunol. 155:1685–93 Jeltsch M, Kaipainen A, Joukov V, Meng X, Lakso M, Rauvala H, Swartz M, Fukumura D, Jain RK, Alitalo K. 1997. Hyperplasia of lymphatic vessels in VEGF-C transgenic mice. Science 276:1423–25 Fu Y-X, Huang G, Matsumoto M, Molina H, Chaplin DD. 1997. Independent signals regulate development of primary and secondary follicle structure in spleen and mesenteric lymph node. Proc. Natl. Acad. Sci. USA 94:5739–43 Pfeffer K, Matsuyama T, Kundig TM, Wakeham A, Kishihara K, Shahinian A, Wiegmann K, Ohashi PS, Kronke M, Mak TW. 1993. Mice deficient for the 55 kd tumor necrosis factor receptor are resistant to endotoxic shock, yet succumb to L. monocytogenes infection. Cell 73:457– 67 Erickson SL, de Sauvage FJ, Kikly K, Carver-Moore K, Pitts-Meek S, Gillett N, Sheehan KCF, Schreiber RD, Goeddel DV, Moore MW. 1994. Decreased sensitivity to tumour necrosis factor but normal T-cell development in TNF receptor-2-deficient mice. Nature 372:560– 63 Mariathasan S, Matsumoto M, Baranyay F, Nahm MH, Kanagawa O, Chaplin DD. 1995. Absence of lymph nodes in lymphotoxin-α (LTα)-deficient mice is due to abnormal organ development, not defective lymphocyte migration. J. Inflamm. 45:72–78 Rennert PD, Browning JL, Mebius R, Mackay F, Hochman PS. 1996. Surface lymphotoxin α/β complex is required for the development of peripheral lymphoid organs. J. Exp. Med. 184:1999–2006 Rennert PD, Browning JL, Hochman PS. 1997. Selective disruption of lymphotoxin ligands reveals a novel set of mucosal lymph nodes and unique effects on lymph node cellular organization. Int. Immunol. 9:1627–39 Ettinger R, Browning JL, Michie SA, van Ewijk W, McDevitt HO. 1996. Disrupted splenic architecture, but normal lymph node development in mice expressing a soluble lymphotoxin-β receptor-IgG1 fusion protein. Proc. Natl. Acad. Sci. USA 93:13102–7 Koni PA, Sacca R, Lawton P, Browning JL, Ruddle NH, Flavell RA. 1997. Distinct roles in lymphoid organogenesis for lymphotoxins α and β revealed in

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

SECONDARY LYMPHOID TISSUES

59.

Annu. Rev. Immunol. 1999.17:399-433. 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.

lymphotoxin β-deficient mice. Immunity 6:491–500 Koni PA, Flavell RA. 1998. A role for tumor necrosis factor receptor type 1 in gut-associated lymphoid tissue development—genetic evidence of synergism with lymphotoxin β. J. Exp. Med. 187: 1977–83 Rennert PD, James D, Mackay F, Browning JL, Hochman PS. 1998. Lymph node genesis is induced by signaling through the lymphotoxin β receptor. Immunity 9:71–79 Futterer A, Mink K, Luz A, Kosco-Vilbois MH, Pfeffer K. 1998. The lymphotoxin β receptor controls organogenesis and affinity maturation in peripheral lymhoid tissues. Immunity 9:59–70 Neumann B, Luz A, Pfeffer K, Holzmann B. 1996. Defective Peyer’s patch organogenesis in mice lacking the 55-kD receptor for tumor necrosis factor. J. Exp. Med. 184:259–64 Pasparakis M, Alexopoulou L, Grell M, Pfizenmaier K, Bluethmann H, Kollias G. 1997. Peyer’s patch organogenesis is intact yet formation of B lymphocyte follicles is defective in peripheral lymphoid organs of mice deficient for tumor necrosis factor and its 55-kDa receptor. Proc. Natl. Acad. Sci. USA 94:6319–23 Forster R, Mattis AE, Kremmer E, Wolf E, Brem G, Lipp M. 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 Burkittslymphoma 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 Georgopoulos K, Winandy S, Avitahl N. 1997. The role of the Ikaros gene in lymphocyte development and homeostasis. Annu. Rev. Immunol. 15:155–76 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 Wang JH, Nichogiannopoulou A, Wu L, Sun L, Sharpe AH, Bigby M,

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

431

Georgopoulos K. 1996. Selective defects in the development of the fetal and adult lymphoid system in mice with an Ikaros null mutation. Immunity 5:537–49 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 Leonard WJ, Shores EW, Love PE. 1995. Role of the common cytokine receptor gamma chain in cytokine signaling and lymphoid development. Immunol. Rev. 148:97–114 Cao X, Shores EW, Hu-Li J, Anver MR, Kelsall BL, Russell SM, Drago J, Noguchi M, Grinberg A, Bloom ET. 1995. Defective lymphoid development in mice lacking expression of the common cytokine receptor gamma chain. Immunity 2:223– 38 Russell SM, Tayebi N, Nakajima H, Riedy MC, Roberts JL, Aman MJ, Migone T-S, Noguchi M, Markert ML, Buckley RH, O’Shea JJ, Leonard WJ. 1995. Mutation of Jak3 in a patient with SCID: essential role of Jak3 in lymphoid development. Science 270:797–800 Nosaka T, van Deursen JM, Tripp RA, Thierfelder WE, Witthuhn BA, McMickle AP, Doherty PC, Grosveld GC, Ihle JN. 1995. Defective lymphoid development in mice lacking Jak3. Science 270:800–2 Thomis DC, Gurniak CB, Tivol E, Sharpe AH, Berg LJ. 1995. Defects in B lymphocyte maturation and T lymphocyte activation in mice lacking Jak3. Science 270:794–97 Park SY, Saijo K, Takahashi T, Osawa M, Arase H, Hirayama N, Miyake K, Nakauchi H, Shirasawa T, Saito T. 1995. Developmental defects of lymphoid cells in Jak3 kinase-deficient mice. Immunity 3:771–82 Matsumoto M, Mariathasan S, Nahm MH, Baranyay F, Peschon JJ, Chaplin DD. 1996. Role of lymphotoxin and the type I TNF receptor in the formation of germinal centers. Science 271:1289–91 Matsumoto M, Lo SF, Carruthers CJL, Min J, Mariathasan S, Huang G, Plas DR, Martin SM, Geha RS, Nahm MH, Chaplin DD. 1996. Affinity maturation without germinal centres in lymphotoxin-αdeficient mice. Nature 382:462–66 Alimzhanov MB, Kuprash DV, KoscoVilbois MH, Luz A, Turetskaya RL, Tarakhovsky A, Rajewsky K, Nedospasov SA, Pfeffer K. 1997. Abnormal development of secondary lymphoid tissues

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

432

80.

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

81.

82.

83.

84.

85.

86.

87.

88.

89.

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

FU & CHAPLIN in lymphotoxin-β-deficient mice. Proc. Natl. Acad. Sci. USA 94:9302–7 Fu Y-X, Molina H, Matsumoto M, Huang G, Min J, Chaplin DD. 1997. Lymphotoxin-α supports development of splenic follicular structure that is required for IgG responses. J. Exp. Med. 185:2111–20 Alexopoulou L, Pasparakis M, Kollias G. 1998. Complementation of lymphotoxinα knockout mice with tumor necrosis factor-expressing transgenes rectifies defective splenic structure and function. J. Exp. Med. 188:745–54 Pham CT, MacIvor DM, Hug BA, Heusel JW, Ley TJ. 1996. Long-range disruption of gene expression by a selectable marker cassette. Proc. Natl. Acad. Sci. USA 93:13090–95 Pasparakis M, Alexopoulou L, Episkopou V, Kollias G. 1996. Immune and inflammatory responses in TNFα-deficient mice: a critical requirement for TNFα in the formation of primary B cell follicles, follicular dendritic cell networks, and germinal centers, and in the maturation of the humoral immune response. J. Exp. Med. 184:1397–411 Gonzalez M, Mackay F, Browning JL, Kosco-Vilbois MH, Noelle RJ. 1998. The sequential role of lymphotoxin and B cells in the development of splenic follicles. J. Exp. Med. 187:997–1007 Muller M, Eugster H-P, Le Hir M, Shakhov A, di Padova F, Maurer C, Quesniaux VFJ, Ryffel B. 1996. Correction or transfer of immunodeficiency due to TNF-LTα deletion by bone marrow transplantation. Mol. Med. 2:247–55 Mackay F, Majeau GR, Lawton P, Hochman PS, Browning JL. 1997. Lymphotoxin but not tumor necrosis factor functions to maintain splenic architecture and humoral responsiveness in adult mice. Eur. J. Immunol. 27:2033–42 Le Hir M, Bleuthmann H, Kosco-Vilbois MH, Muller M, di Padova F, Moore M, Ryffel B, Eugster H-P. 1996. Tumor necrosis factor receptor-1 signaling is required for differentiation of follicular dendritic cells, germinal center formation, and full antibody responses. J. Inflamm. 47:76–80 Korner H, Cook M, Riminton DS, Lemckert FA, Hoek RM, Ledermann B, Kontgen F, Fazekas de St Groth B, Sedgewick JD. 1997. Distinct roles for lymphotoxinα and tumor necrosis factor in organogenesis and spatial organization of lymphoid tissue. Eur. J. Immunol. 27:2600–9 Murphy M, Pike-Nobile L, Browning JL,

90. 91.

92.

93. 94.

95.

96.

97.

98.

99.

100. 101.

Ware CF, Epstein LB. 1995. Expression of lymphotoxin-β receptor in human thymus and spleen. Proc. 9th Int. Congr. Immunol., p. 770 (Abstr.) Ryffel B, Mihatsch MJ. 1993. TNF receptor distribution in human tissues. Int. Rev. Exp. Pathol. 34B:149–56 Marino MW, Dunn A, Grail D, Inglese M, Noguchi Y, Richards E, Jungbluth A, Wada H, Moore M, Williamson B, Basu S, Old LJ. 1997. Characterization of tumor necrosis factor-deficient mice. Proc. Natl. Acad. Sci. USA 94:8093–98 Grell M, Douni E, Wajant H, Lohden M, Clauss M, Maxeiner B, Georgopoulos S, Lesslauer W, Kollias G, Pfizenmaier K, Scheurich P. 1995. The transmembrane form of tumor necrosis factor is the prime activating ligand of the 80 kDa tumor necrosis factor receptor. Cell 83:793–802 Paul NL, Ruddle NH. 1988. Lymphotoxin. Annu. Rev. Immunol. 6:407–38 Fu Y-X, Huang GM, Wang Y, Chaplin DD. 1998. B lymphocytes induce the formation of follicular dendritic cell clusters in a lymphotoxin-α-dependent fashion. J. Exp. Med. 187:1009–18 Wang B, Hollander GA, Nichogiannopoulou A, Simpson SJ, Orange JS, Gutierrez-Ramos JC, Burakoff SJ, Biron CA, Terhorst C. 1996. Natural killer cell development is blocked in the context of aberrant T lymphocyte ontogeny. Int. Immunol. 8:939–49 Mandel TE, Phipps RP, Abbot A, Tew JG. 1980. The follicular dendritic cell: long term antigen retention during immunity. Immunol. Rev. 53:29–59 Cerny A, Zinkernagel RM, Groscurth P. 1988. Development of follicular dendritic cells in lymph nodes of B-cell-depleted mice. Cell Tissue Res. 254:449–54 Kapasi ZF, Burton GF, Shultz LD, Tew JG, Szakal AK. 1993. Induction of functional follicular dendritic cell development in severe combined immunodeficiency mice. Influence of B and T cells. J. Immunol. 150:2648–58 Yoshida K, van den Berg TK, Dijkstra CD. 1994. The functional state of follicular dendritic cells in severe combined immunodeficient (SCID) mice: role of the lymphocytes. Eur. J. Immunol. 24:464–68 Imai Y, Yamakawa M. 1996. Morphology, function and pathology of follicular dendritic cells. Pathol. Int. 46:807–33 Matsumoto M, Fu Y-X, Molina H, Huang G, Kim J, Thomas D, Nahm MH, Chaplin DD. 1997. Distinct roles of lymphotoxinα and the type I tumor necrosis factor (TNF) receptor in the establishment

P1: KKK/SAT/ary

P2: KKK/mbg

January 18, 1999

17:35

QC: ARS/uks

T1: KKK

Annual Reviews

AR078-14

SECONDARY LYMPHOID TISSUES

102. 103.

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

104.

105.

106.

107.

108.

109.

110.

of follicular dendritic cells from nonbone marrow-derived cells. J. Exp. Med. 186:1997–2004 Liu YJ, Arpin C. 1997. Germinal center development. Immunol. Rev. 156:111–26 Foy TM, Laman JD, Ledbetter JA, Aruffo A, Claassen E, Noelle RJ. 1994. gp39CD40 interactions are essential for germinal center formation and the development of B cell memory. J. Exp. Med. 180:157– 63 Sato S, Steeber DA, Jansen PJ, Tedder TF. 1997. CD19 expression levels regulate B lymphocyte development. Human CD19 restores normal function in mice lacking endogenous CD19. J. Immunol. 158:4662–69 Ferguson SE, Han S, Kelsoe G, Thompson CB. 1996. CD28 is required for germinal center formation. J. Immunol. 156:4576–81 Borriello F, Sethna MP, Boyd SD, Schweitzer AN, Tivol EA, Jacoby D, Strom TB, Simpson EM, Freeman GJ, Sharpe AH. 1997. B7-1 and B7-2 have overlapping, critical roles in immunoglobulin class switching and germinal center formation. Immunity 6:303–13 Weih F, Warr G, Yang H, Bravo R. 1997. Multifocal defects in immune responses in RelB-deficient mice. J. Immunol. 158:5211–18 Franzoso G, Carlson L, Poljack L, Shores EW, Epstein S, Leonardi A, Grinberg A, Tran T, Schartonkersten T, Anver M, Love P, Brown K, Siebenlist U. 1998. Mice deficient in nuclear factor (NF)-kappaB/p52 present with defects in humoral responses, germinal center reactions, and splenic microarchitecture. J. Exp. Med. 187:147–59 Schwarz EM, Krimpenfort P, Berns A, Verma IM. 1997. Immunological defects in mice with a targeted disruption of Bcl3. Genes Dev. 11:187–97 Franzoso G, Carlson L, Schartonkersten T, Shores EW, Epstein S, Grinberg A, Tran T, Shacter E, Leonardi A, Anver M, Love P, Sher A, Siebenlist U. 1997. Critical roles for the Bcl-3 oncoprotein in T cell-mediated immunity, splenic microar-

111.

112.

113.

114.

115.

116.

117.

118.

119.

433

chitecture, and germinal center reactions. Immunity 6:479–90 Dent AL, Shaffer AL, Yu X, Allman D, Staudt LM. 1997. Control of inflammation, cytokine expression, and germinal center formation by BCL-6. Science 276:589–92 Fukuda T, Yoshida T, Okada S, Hatano M, Miki T, Ishibashi K, Okabe S, Koseki H, Hirosawa S, Taniguchi M, Miyasaka N, Tokuhisa T. 1997. Disruption of the Bcl6 gene results in an impaired germinal center formation. J. Exp. Med. 186:439–48 Molina H, Holers VM, Li B, Fang YF, Mariathasan S, Goellner J, StraussSchoenberger J, Karr RW, Chaplin DD. 1996. Markedly impaired humoral immune response in mice deficient in complement receptors 1 and 2. Proc. Natl. Acad. Sci. USA 93:3357–61 Mebius RE, Rennert P, Weissman IL. 1997. Developing lymph nodes collect CD4+CD3− LTβ + cells that can differentiate to APC, NK cells, and follicular cells but not T or B cells. Immunity 7:493–504 Kratz A, Campos-Neto A, Hanson MS, Ruddle NH. 1996. Chronic inflammation caused by lymphotoxin is lymphoid neogenesis. J. Exp. Med. 183:1461–72 Sacca R, Turley S, Soong L, Mellman I, Ruddle NH. 1997. Transgenic expression of lymphotoxin restores lymph nodes to lymphotoxin-α-deficient mice. J. Immunol. 159:4252–60 Villavedra M, Carol H, Hjulstrom M, Holmgren J, Czerkinsky C. 1997. “PERFEXT”: a direct method for quantitative assessment of cytokine production in vivo at the local level. Res. Immunol. 148:257– 66 Forsthuber T, Yip HC, Lehmann PV. 1996. Induction of TH1 and TH2 immunity in neonatal mice. Science 271:1728– 30 Mauri DN, Ebner R, Montgomery RI, Kochel KD, Cheung G-L, Yu TC, Ruben S, Murphy M, Eisenberg RJ, Cohen GH, Spear PG, Ware CF. 1998. LIGHT, a new member of the TNF superfamily, and lymphotoxin α are ligands for herpesvirus entry mediator. Immunity 8:21–30

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:399-433. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:435–66 c 1999 by Annual Reviews. All rights reserved Copyright °

THE STRUCTURAL BASIS OF T CELL ACTIVATION BY SUPERANTIGENS Hongmin Li,# Andrea Llera,# Emilio L. Malchiodi,#,∗ and Roy A. Mariuzza# #Center for Advanced Research in Biotechnology, University of Maryland Biotechnology Institute, Rockville, Maryland 20850, and ∗ Instituto de Estudios de la Inmunidad Humoral, CONICET, Catedra de Inmunolgia, FFyB, UBA, 1113 Buenos Aires, Argentina; e-mail: [email protected]

KEY WORDS:

staphylococcal enterotoxins, T cell receptor, T cell stimulation, three-dimensional structure

ABSTRACT Superantigens (SAGs) are a class of immunostimulatory and disease-causing proteins of bacterial or viral origin with the ability to activate large fractions (5–20%) of the T cell population. Activation requires simultaneous interaction of the SAG with the Vβ domain of the T cell receptor (TCR) and with major histocompatibility complex (MHC) class II molecules on the surface of an antigenpresenting cell. Recent advances in knowledge of the three-dimensional structure of bacterial SAGs, and of their complexes with MHC class II molecules and the TCR β chain, provide a framework for understanding the molecular basis of T cell activation by these potent mitogens. These structures along with those of TCRpeptide/MHC complexes reveal how SAGs circumvent the normal mechanism for T cell activation by peptide/MHC and how they stimulate T cells expressing TCR β chains from a number of different families, resulting in polyclonal T cell activation. The crystal structures also provide insights into the basis for the specificity of different SAGs for particular TCR β chains, and for the observed influence of the TCR α chain on SAG reactivity. These studies open the way to the design of SAG variants with altered binding properties for TCR and MHC for use as tools in dissecting structure-activity relationships in this system.

435 0732-0582/99/0410-0435$08.00

P1: SAT/spd

P2: KKK/plb

January 21, 1999

436

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTRODUCTION T lymphocytes recognize a wide variety of antigens through highly diverse cellsurface glycoproteins known as T cell receptors (TCRs). These disulfide-linked heterodimers are composed of α and β, or γ and δ, chains that have variable (V) and constant (C) regions that are structurally homologous to those of antibodies (1, 2). Unlike antibodies, however, which recognize antigen alone, αβ TCRs recognize antigen only in the form of peptides bound to major histocompatibility complex (MHC) molecules. In addition, TCRs interact with a class of viral or bacterial proteins known as superantigens (SAGs), which stimulate T cells bearing particular Vβ elements, resulting in the massive release of T cell– derived cytokines such as interleukin (IL)-2 and tumor necrosis factor (TNF) β, generally followed by the eventual disappearance or inactivation of responding T cells (3–4). Activation of the T cell requires simultaneous interaction of the SAG with the TCR and with MHC class II molecules on an antigen-presenting cell (APC). The best-characterized group of SAGs belongs to the so-called pyrogenic toxin SAG family, which includes the staphylococcal enterotoxins (SE)A through I (except F), staphylococcal toxic shock syndrome toxin-1 (TSST-1), streptococcal superantigen (SSA), and streptococcal pyrogenic exotoxins (SPE)A-C and -F (6–8). These bacterial SAGs have in common the following characteristics: (a) they are among the most potent pyrogens known, (b) they are all capable of inducing a highly lethal toxic shock syndrome, and (c) they share a typical three-dimensional structure consisting of two domains, termed large and small. The small domain is a β-barrel made up of two β-sheets, whereas the large domain contains a β-grasp motif, an α-helix packed against a mixed β-sheet that connects the peripheral strands (9–16). Nevertheless, each of these molecules has unique biological properties and stimulates the proliferation of T cells with different Vβ regions. Among their biological effects, the staphylococcal enterotoxins are characterized by their ability to induce emesis and diarrhea, whereas TSST-1 lacks emetic activity. The streptococcal toxins do not cause enteric problems but they are associated with cardiotoxicity. The level of sequence homology between the pyrogenic toxins varies widely, and they can be divided into groups based on sequence similarities. The highest degree of homology is achieved by SEA, SED, and SEE (between 53–81%), followed by the group of SEB, the SECs, SPEA, and SSA, with 50–66% of sequence homology. All the rest, including SPEB, SPEC, SPEF, and TSST-1, have poor or no homology to any other toxin, or to each other. Bacterial SAGs that do not belong to the pyrogenic toxin family include the staphylococcal exfoliative toxins (ET) A and B (17, 18), Mycoplasma arthritidis mitogen (MAM) (19), and Yersinia pseudotuberculosis mitogen (20, 21).

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL ACTIVATION BY SUPERANTIGENS

437

Among superantigenic proteins of viral origin, only mouse mammary tumor virus (MMTV)-encoded SAGs have been defined in detail (22). It has been demonstrated that mouse Mls endogenous SAGs are encoded by MMTV proviral DNA that has been integrated into the germline, demonstrating a link between endogenous SAGs and infectious agents. Other reports have shown superantigenic activity by the rabies virus nucleocapsid protein (23) and by two human tumor viruses, cytomegalovirus (24) and Epstein-Barr virus (25). Recently, the envelope gene of an endogenous human retrovirus isolated from pancreatic islets was shown to encode an MHC class II–dependent SAG specific for Vβ7 (26). The past four years have witnessed a remarkable series of advances in knowledge of the three-dimensional structure of TCRs (27–32) and of their complexes with peptide/MHC (33–35) and SAGs (36, 37). In this review, we focus on TCRSAG interactions and describe current understanding of the structural basis of T cell activation by SAGs. After giving an overview of the biological effects of bacterial and viral SAGs, we discuss the affinity and kinetics of TCR and MHC binding to these molecules. We then describe the three-dimensional structures of MHC-SAG and TCR β chain-SAG complexes. These structures, along with those of TCR-peptide/MHC complexes, reveal how SAGs circumvent the normal mechanism for T cell activation by peptide/MHC and how they stimulate T cells expressing TCR β chains from a number of different families, resulting in polyclonal T cell activation. Finally, we discuss the structural basis for the specificity of different SAGs for particular TCR β chains and for the observed influence of the TCR α chain on SAG reactivity.

BIOLOGICAL EFFECTS OF SUPERANTIGENS T Cell Anergy and Deletion The specificity of interaction of SAGs with the Vβ domain of the TCR has provided a unique opportunity to examine the fate of reactive T cells in vivo independently of functional assays. Such studies have revealed that responding T cells can proliferate, become nonresponsive (anergy), or even die (deletion) (38, 39). In the in vivo recognition of endogenous SAGs, intrathymic deletion in Vβ-specific subsets occurs at the double-positive (CD4+, CD8+) stage of development, and deletion is correspondingly apparent in both the mature CD4+ and CD8+ subsets (40–42). In the case of exogenous SAGs, an early report showed that mice injected from birth with SEB virtually lack Vβ3+ and Vβ8+ mature thymocytes, giving the first formal demonstration that clonal deletion can accompany induced tolerance to a foreign antigen (43). Subsequent studies confirmed this report and showed that in adult mice, SEB-specific mature T cells can, after an initial expansion, be rendered anergic in both in vivo and in vitro

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

438

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

models (44, 45). Moreover, SEB-induced death of Vβ8+ cells is independent of an intact thymus, because it also occurs in adult thymectomized animals (46). The presence of a SAG in the MMTV genome can guarantee the existence of actively dividing populations of T and B cells through the ability of the SAG to stimulate T cells, and, thus, let the virus complete its replication cycle (22). This was confirmed using mice transgenic for the MMTV (C3H) sag gene (47). The SAG-mediated deletion of Vβ14+ T cells during early life conferred resistance to infection to these mice. Similar results were observed for a different exogenous MMTV, called SW, where the corresponding SAG stimulates Vβ6+ cells (48). Because MMTV infection occurs during the shaping of the immune repertoire in neonatal life and because infection is persistent, there is a gradual deletion of SAG-reactive T cells (49). Finally, there is no evidence for a SAG-independent pathway of MMTV transmission, and only MMTV with functional sag genes can be transmitted through milk (50, 51). Once the SAG is stably integrated into the mouse genome, it can be inherited by successive generations. When expressed endogenously, it causes deletion of cognate T cells and prevents a reinfection with the same strain of virus that produces the SAG (47, 48).

Toxic Shock Syndrome and Food Poisoning Toxic shock syndrome (TSS) is an acute, life-threatening intoxication characterized by high fever, hypotension, rash, multiorgan dysfunction, and cutaneous desquamation that is caused by staphylococcal or streptococcal pyrogenic toxins (6, 52–54). The interaction of the pyrogenic toxins with TCR and MHC activates both the T cell, for secretion of TNFβ, IL-2, and γ interferon, and the APC, for secretion of TNFα and IL-1. The resulting massive cytokine release is believed to be responsible for capillary leak and hypotension, and it is also likely to cause the erythematous rash in TSS patients (52, 54). Staphylococcal enterotoxins are among the most common causes of food poisoning in humans. It has been suggested that the enterotoxic effects are directly related to their superantigenic activity, i.e. dependent on T cell stimulation and probably caused by massive cytokine release (3, 6). However, some evidence suggests that the emetic and T cell proliferative activities of the toxins may be distinct (3, 55–57). In fact, the induction of emesis has been attributed to leukotriene or histamine release (58). It has been shown that SEB and SEA can rapidly cross an epithelial membrane in intact, fully functional form, thus gaining access to T cells. On the contrary, TSST-1, which lacks emetic activity, although able to transcytose epithelium, may be more easily destroyed by digestive enzymes in the stomach and intestine (59). Thus, the ability to cause enterotoxicity may be related to the resistance to digestion of the enterotoxins.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

T CELL ACTIVATION BY SUPERANTIGENS

439

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Autoimmune Diseases In autoimmune diseases, a breakdown of self-tolerance leads to the generation of an immune response against a specific target antigen or antigens. A large body of clinical and epidemiological evidence indicates that infections are important in the induction of autoimmune disorders such as autoimmune myocarditis (60) and insulin-dependent diabetes mellitus (61). One mechanism by which this has long been thought to occur is through the activation of autoreactive T cells by epitopes on microbial antigens that are cross-reactive with antigens on target organs (62). For example, a number of viral and bacterial peptides have been identified that efficiently activate myelin basic protein (MBP)-specific T cell clones from multiple sclerosis (MS) patients (63, 64). More recently, it has been proposed that SAGs derived from bacteria, mycoplasma, or viruses may initiate autoimmune disease by activating specific anti-self T cell clones (3, 65). Indeed, microbial SAGs have been shown to trigger clinical relapses of autoimmune disease in several animal models, as discussed below. The expansion of selected Vβ families in the affected organs or peripheral blood of certain individuals with autoimmune disease has also been documented (3, 65). However, it is a common observation that different TCR repertoire studies of the same disease can provide different results (66). Even when it is unlikely that SAGs by themselves initiate an autoimmune disease (65), they may modulate disease pathogenesis. In susceptible individuals, the activation of autoreactive T cells is a necessary, but not sufficient, condition for the development of an autoimmune disease. A sufficient degree of clonal expansion of autoreactive T cells may be a major limiting factor, and SAGs may induce such an expansion. Alternatively, the activation of B lymphocytes and other APC through the SAG bridge may lead to the secretion of autoantibodies and interleukins that contribute to inflammation. Evidence for an autoimmune origin of MS comes from (a) the presence of CD4+ T cells and cells expressing MHC class II molecules in inflamed tissues (67), (b) the finding that MS is associated with certain MHC class II alleles (68), and (c) the demonstration that MBP-specific T cells are clonally expanded in MS patients (69–72). In experimental autoimmune encephalomyelitis (EAE), a model for MS, administration of SEB to PL/J mice following immunization with a peptide derived from MBP (Ac1-11) was found to induce paralysis in mice with subclinical disease and to trigger relapses in mice that are in remission following an initial episode of paralysis (73–75). It was shown that these effects are the direct result of stimulation by SEB of Vβ8-expressing encephalitogenic T cells specific for MBP Ac1-11. An analysis of the TCR β chain repertoire of synovial T cells from rheumatoid arthritis (RA) patients revealed a selective expansion of Vβ14-bearing T cells compared with the levels in the peripheral blood of the same

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

440

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

individuals (76). A mechanism for the pathogenesis of RA was proposed in which a microbial SAG activates disease-mediating Vβ14+ T cells and allows these activated cells to enter the synovial tissue, where they persist because of reactivation by autoantigens. In collagen II–induced arthritis (CIA), a model for RA, mice are immunized with native porcine type II collagen and develop joint swelling. It has been demonstrated that T cells expressing Vβ8 are important in the development of CIA (77–80) and that administration of SEB 10 days prior to a collagen II challenge protects mice from CIA (81). In both EAE and CIA, the response to self-antigens is controlled by a potent regulatory T cell circuitry based on recognition of different determinants derived from the TCR Vβ8.2 chain (81–83). The SAG MAM, which derives from a naturally occurring mouse arthritogenic mycoplasma, activates Vβ8+ T cells. Administration of MAM has been shown to markedly exacerbate arthritis in mice that were convalescent from CIA, or to trigger arthritis in animals previously immunized with collagen II but that had failed to develop clinical disease (19). Insulin-dependent diabetes mellitus (IDDM) is an autoimmune disease affecting pancreatic β cells that secrete insulin. A relationship between viral infections and the development of IDDM has been long suspected (84). An analysis of pancreatic islet–infiltrating T cells from patients with IDDM revealed preferential expression of the Vβ7 gene segment, but no selection for particular Vα segments or Vβ-D-Jβ junctional sequences (85, 86). This led to the proposal that a SAG associated with pancreatic islets may be involved in the pathogenesis of IDDM. This putative SAG was recently identified by Conrad et al (26), who isolated a novel human endogenous retrovirus from supernatants of IDDM islets and showed that the envelope gene encodes an MHC class II–dependent SAG specific for Vβ7. Kawasaki disease (KD) is an acute febrile illness with symptoms similar to toxic shock syndrome. Several studies have revealed a significantly elevated level of circulating Vβ2+ and, to a lesser extent, Vβ8.1+ T cells in patients with acute KD, compared with control populations (87–90). Sequencing of these β chains revealed extensive junctional region diversity, which suggests activation by SAG and not a specific disease-associated antigen. Bacteriaproducing toxins that activate Vβ2+ T cells (TSST-1 and SPEB/SPEC) were isolated from 13 out of 16 KD patients but only 1 out of 15 in the control group (91). Nevertheless, other groups were not able to document the expansion of any Vβ family during the acute phase of KD (92, 93). Differences in population studied, method and time of collecting samples, and techniques used could potentially explain the differences in results. Alternatively, the expansion of selected Vβ families may not be related to the pathogenesis of the disease (94).

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/uks

15:3

T1: KKK

Annual Reviews

AR078-15

T CELL ACTIVATION BY SUPERANTIGENS

441

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Skin Diseases The staphylococcal toxins ETA and ETB induce the symptoms associated with staphylococcal scalded skin syndrome, characterized by a specific intraepidermal separation of layers of the skin (95). X-ray crystallographic studies of ETA have shown that its overall structure is similar to that of the chymotrypsin-like serine protease family of enzymes (17, 18). The catalytic triad includes the residue Ser195, which when mutated to cysteine abolishes the characteristic separation of epidermal layers, although the ability to induce T cell proliferation is not altered (17, 96). These findings suggest that skin separation is the result of a specific proteolysis by ETA, and not related to its superantigenic activity; the latter is probably involved in the edema or redness associated with scalded skin syndrome. Psoriasis is a disease characterized by increased proliferation of epidermal cells associated with an inflammatory component. Patients with acute guttate psoriasis often have flares of psoriasis following streptococcal infections (97). Histological examination of early skin lesions shows that infiltration of lymphocytes and macrophages into the skin precedes the characteristic epidermal proliferation of psoriasis. The predominant distribution of Vβ2-, Vβ3-, and Vβ5-bearing T cells in lesional skin of acute guttate psoriasis has been described (98–100). However, there are conflicting reports about the restricted T cell receptor repertoire in chronic psoriasis (98, 101), and no increase in SAG-producing Staphylococcus aureus has been seen in chronic psoriatic patients (102). It is unlikely, then, that SAGs are essential to the continuance of psoriasis, although they may be exacerbating factors or triggers for the disease. Atopic dermatitis is a chronic pruritic inflammation of the skin characterized by local infiltration of monocytes and lymphocytes, mast cell degranulation, and immediate and delayed hypersensitivity (103). There are numerous reports that S. aureus can exacerbate this disease, and S. aureus was isolated from the affected skin of more than 90% of patients (104). More than half of the patients had S. aureus that secreted SEA, SEB, and TSST-1 (105). Sera from 57% of atopic dermatitis patients contained immunoglobulin E specific for one or more of these SAGs. Thus, epicutaneous superantigenic toxins might induce specific immunoglobulin E in atopic dermatitis patients, as well as mast cell degranulation.

AFFINITY AND KINETICS OF TCR BINDING TO SUPERANTIGENS Gascoigne & Ames first demonstrated direct binding of a soluble TCR β chain (mouse Vβ3) to SEA presented by MHC class II molecules on cells (106).

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

442

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

However, the affinity was too low to be measured in their cell-binding assay. More recently, the development of surface plasmon resonance techniques for detecting macromolecular interactions (107) has permitted the precise measurement of kinetic and affinity constants for TCR binding to SAGs, as well as to peptide/MHC complexes (108). By this method, a soluble human TCR (Vβ3.1) was found to bind immobilized SEB in the absence of MHC class II molecules with a dissociation constant (KD) of 0.8 µM; the on-rate (kon) of the interaction was 1.3 × 104 M−1 s−1 and the off-rate (koff) 1.1 × 10−2 s−1 (109). Specific binding of soluble 14.3.d TCR β chain (mouse Vβ8.2) was demonstrated to SEB, SEC1, SEC2, SEC3, and SPEA, consistent with the known proliferative effects of these SAGs on T cells expressing Vβ8.2 (110). In contrast, SEA, which does not stimulate Vβ8.2-bearing cells, did not bind the recombinant β chain. Affinities ranged from 3 µM for SEC3 to 140 µM for SEB; kon and koff were too fast to be accurately measured, but were estimated at >105 M−1 s−1 and >0.1 s−1, respectively (110, 111). The unpaired β chain was shown to fully retain the SAG-binding activity of the assembled 14.3.d αβ TCR heterodimer (110). A KD of 1.1 µM was measured for the binding of SEC2 to the mouse D10 TCR (Vβ8.2), with a kon of 1.7 × 104 M−1 s−1 and a koff of 1.9 × 10−2 s−1 (112). These values closely resemble those for the interaction of D10 TCR with its cognate peptide/MHC class II ligand, which has a KD of 2.1 µM and a kon and koff of 1.0 × 104 M−1 s−1 and 2.2 × 10−2 s−1 , respectively (112). In each of the above examples, TCR-SAG binding is characterized by low affinities (>10−6 M) and very fast kon and koff (>104 M−1 s−1 and >10−2 s−1 , respectively). It is noteworthy that low affinities and rapid dissociation kinetics have also been reported for the interaction of other T cell surface glycoproteins with their ligands, such as the adhesion molecule CD2 with CD48 (113, 114). In particular, the affinities of TCR-SAG interactions (10−4–10−6 M) are remarkably similar to those reported for the binding of TCRs to specific peptide/MHC complexes (10−4–10−7 M) (115–117) and are much weaker than those of antigen-antibody reactions (typically 10−8–10−11 M). In the case of adhesion molecules, fast dissociation rates may facilitate deadhesion, a requirement for cell motility (113). In the case of TCRs, rapid off-rates may permit a single peptide/MHC complex to sequentially bind and trigger a large number of TCRs (up to 200), as proposed in the serial triggering (118, 119) and kinetic proofreading (120) models of T cell activation, until a certain activation threshold is reached. The finding that the binding of bacterial SAGs to the TCR is characterized by low affinities and fast dissociation kinetics suggests that SAGs mimic the interaction of peptide/MHC complexes with the TCR in terms of affinity and kinetics and that some form of serial engagement may also operate in T cell activation by SAGs (111, 119, 121). The relationship between the affinity of SAGs for TCR and MHC and their ability to activate T cells has been investigated using site-directed mutants of

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL ACTIVATION BY SUPERANTIGENS

443

SEC3 and SEB (111). In order to mimic normal physiological conditions as closely as possible, resting lymph node T cells bearing the 14.3.d TCR from RAG-2−/− TCR transgenic mice were used to measure the stimulatory effects of mutant SAGs on BALB/c spleen cells expressing I-Ed, or on MHC class II–negative mouse fibroblasts transfected with a gene encoding human leukocyte antigen (HLA)-DR1. A clear and simple relationship was observed between the affinity of SAGs for the TCR and their mitogenic potency: the tighter the binding of a particular mutant of SEC3 or SEB to the TCR β chain, the greater its ability to stimulate T cells. The affinities of the SAGs tested ranged from 3.5 µM to >250 µM. However, an apparent exception to this simple affinity-activity rule was the finding that SEB stimulated transgenic T cells about 10-fold better than did SEC3, even though the affinity of SEB for the TCR β chain (140 µM) is much lower than that of SEC3 (111). To determine whether the surprisingly strong mitogenic potency of SEB relative to SEC3 could be attributed to tighter binding to MHC class II on APC, the binding of SEB and SEC3 to soluble recombinant HLA-DR1 was measured by sedimentation equilibrium: Whereas SEB bound to DR1 with a KD of 14 µM, the corresponding value for SEC3 was 48 µM. Therefore, the unexpectedly high mitogenic potency of SEB relative to SEC3 can be explained by the stronger binding of SEB to MHC class II. This indicates that mitogenic potency is the result of an interplay between TCR-SAG and SAG-MHC interactions, such that a relatively small (threefold) increase in the affinity of a SAG for MHC can overcome a large (35-fold) decrease in the affinity of a SAG for the TCR. With the apparent affinities of SAGs for both TCR and MHC class II molecules in the micromolar range, nearly all SAG molecules will be unbound at physiological SAG concentrations (10−12–10−15 M) (121). Under these conditions, it is difficult to understand how a SAG can effectively cross-link the T cell and APC. The problem is seemingly less severe for peptide/MHC because the peptide is, in effect, irreversibly bound to MHC. One possible explanation for the ability of SAGs to trigger T cells at concentrations orders of magnitude less than their KDs is that accessory molecules such as CD4 help stabilize the TCR-SAG-MHC complex sufficiently for activation to occur. Another is that the overall stability of the TCR-SAG-MHC complex is considerably greater than would be expected from considering the TCR-SAG and SAG-MHC interactions independently. That is, the binding of SAGs may be a cooperative process in which the SAG-MHC complex binds the TCR with greater affinity than does the SAG alone. This hypothesis is supported by the finding that the affinity of SEB for a soluble human TCR was significantly enhanced by the addition of soluble HLA-DR1 (109). The potential role of the TCR α chain in stabilizing the TCR-SAG-MHC complex is discussed in a later section.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

444

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

STRUCTURE OF SUPERANTIGEN–MHC CLASS II COMPLEXES The three-dimensional structures of three SAG-MHC class II complexes have been determined by Wiley and colleagues: (a) the complex between SEB and ˚ resolution (122), (b) the complex between SEB and HLAHLA-DR1 to 2.7-A ˚ resolution (123), and (c) the complex between TSST-1 and HLADR4 to 2.5-A ˚ resolution (124). In the SEB–HLA-DR1 complex [Figure 1A DR1 to 3.5-A (see color plates)], SEB binds to the α1 domain of DR1, contacting residues from the first and third turns of the β-sheet and from the N-terminal portion of the α-helix (122). The binding of SEB to DR4 is similar (123). The ability of SEB to bind many different DR alleles can therefore be explained by its exclusive interaction with the DR1 α chain, which is conserved in all DR molecules. Residues of SEB in contact with DR1 derive mainly from the small domain of the SAG, although several residues from the large domain also contact the DR α chain. SEB binds away from the peptide-binding groove of DR1 and does not contact the bound peptide. The affinity of SEB for DR1 was reported as approximately 0.5 µM in a cell-binding assay (125) and as 14 µM using soluble DR1 (111). Although the TSST-1 binding site on HLA-DR1 overlaps that of SEB, the two SAGs bind differently (124). Whereas SEB binds primarily off one edge of the peptide-binding groove (Figure 1A), TSST-1 extends over nearly half the binding groove and contacts the α-helix of the α1 domain of DR1, the bound peptide, and part of the α-helix of the β1 domain of DR1 (Figure 1B). This binding mode suggests that the interaction of TSST-1 with MHC class II molecules may be partially peptide dependent. In agreement with the crystal structure, certain peptides were found to promote the presentation of TSST-1 by I-Ab up to 5000-fold (126). In contast, the binding of SEB to I-Ab and I-Edk is peptide independent (126, 127). Although no crystal structures have been reported for SEA complexed with MHC class II molecules, mutagenesis and binding studies have demonstrated that SEA possesses two distinct, yet cooperative, binding sites for class II molecules: (a) a low-affinity site (KD = 10−5 M) to the DR1 α chain analogous to the DR1-binding site of SEB, and (b) a Zn2+-dependent, high-affinity site (KD = 10−7 M) to the polymorphic DR1 β chain (127–129). Binding of one SEA molecule to the DR1 β chain enhances the binding of a second SEA molecule to the DR1 α chain (128, 129). Surprisingly, mutations in the Zn2+dependent site completely abolish SEA activity, even though it can still bind the DR1 α chain through its low-affinity SEB-like site. This suggests that MHC cross-linking on the surface of APC may be an essential feature of SEA function. This conclusion is supported by the demonstration that SEA2-DR1

P1: SAT/VKS

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

March 4, 1999

P2: NBL/plb

12:42

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Figure 1 Three-dimensional structures of the complexes between HLA-DR1 and SEB and between HLA-DR1 and TSST-1. (A) Ribbons diagram of the SEB-HLA-Dr1 complex (122). (B) Ribbons diagram of the TSST-HLA-DR-1 complex (124). Colors are as follows: SEB (blue), TSST-1 (pink), DR1 α1 domain (green), DR1 β domain (yellow), and peptide (red).

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL ACTIVATION BY SUPERANTIGENS

445

trimers exist in solution (130), as well as by the finding that SEA mediates signaling through the APC by direct cross-linking of DR1 molecules (131, 132). The high sequence identity between SEE and SEA (see Figure 3A) suggests that they may bind class II molecules similarly. Two additional SAGs have been described that cross-link class II molecules, but by different mechanisms than SEA. The crystal structure of SED shows that this SAG forms dimers in the presence of Zn2+ by coordinating two Zn2+ ions between the large domains of two SED molecules (15). Binding to MHC class II molecules is believed to occur through the small domain to the class II α chain in a manner similar to SEB, resulting in a tetrameric class II α–SED-SED–class II α complex on the APC. The three-dimensional structure of SPEC reveals that the class II α chainbinding site on the small domain has been replaced by SPEC dimer interface (16). Instead, SPEC binds only to the class II β chain. This could potentially lead to the formation of class II β–SPEC-SPEC–class II β tetramers. Dimeric SAGs like SED and SPEC may facilitate TCR dimerization and subsequent T cell triggering. Endogenous SAGs encoded by MMTV can be efficiently presented to T cells only by B cells, through interaction with MHC class II molecules (22). Although direct binding has been demonstrated between recombinant forms of MMTV SAGs and MHC class II molecules (133, 134), the interaction remains poorly understood. I-E molecules are the best presenters for all the described MMTV SAGs (135). In addition, C57BL mice that lack I-E molecule, and thus are not able to present SAG to T cells, are resistant to milk-borne MMTV (C3H) (136). Analysis of class II mutants that lost the ability to present bacterial SAGs revealed that bacterial SAGs have different binding requirements than do MMTV SAGs (137). Another study showed, however, an overlap in at least one binding site for MMTV and SEA on the MHC molecule (138). Recently it was shown that N-linked glycosylation is required for effective B cell presentation of MMTV SAGs to T cells (139).

STRUCTURE OF TCR β CHAIN–SUPERANTIGEN COMPLEXES The three-dimensional structures of several TCR β chain–SAG complexes have been determined to date, each involving the VβCβ chain of the mouse 14.3.d TCR specific for a hemagglutinin peptide of influenza virus bound by the I-Ed class II molecule: (a) the complex between the β chain and SEC2 to ˚ resolution (36), (b) the complex with SEC3 to 3.5-A ˚ resolution (36), 3.5-A ˚ (c) the complex with SEB to 2.4-A resolution (37), and (d ) the complex with a mutant of SEB in which valine at position 26 is replaced by tyrosine (SEB ˚ resolution (37). The SEB V26Y mutant was designed on the V26Y) to 2.6-A

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

446

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

basis of the structure of the TCR β–SEC3 complex to bind the β chain more tightly than wild-type SEB: Its KD is 12 µM, approximately 12 times lower than that of SEB (KD = 140 µM), but still four times higher than that of SEC3 (KD = 3 µM) (111). The mutant is fourfold more active in T cell proliferation assays than is wild-type SEB, consistent with its enhanced affinity. The crystal structures of the TCR β–SEC2/3 complexes identified the regions of the β chain recognized by SEC and showed how SAGs circumvent the normal mechanism for T cell activation by specific peptide/MHC complexes. However, ˚ precluded a detailed analythe moderate resolution of these structures (3.5 A) sis of the interface between the two proteins in terms of hydrogen bonds, van der Waals interactions, and solvent structure. To achieve a more complete description of a β-SAG interface, as well as to assess whether conformational changes occur in either or both proteins upon complex formation, the structures of the complexes between the 14.3.d β chain and SEB and SEB V26Y were determined to high resolution (37). These structures, along with those of αβ TCR heterodimers (29–31) and TCR-peptide/MHC complexes (33–35), can account for the specificity of different SAGs for particular β chains and for the influence of the TCR β chain on SAG reactivity (5, 140–142).

Overall Structure of the TCR β–SEB and TCR β–SEC Complexes The overall structure of the β-SEB complex is shown in Figure 2A (see color plates). The complex is formed through contacts between the Vβ domain and the small and large domains of SEB. The complementarity-determining region (CDR)2 of the TCR β chain and, to lesser extents, hypervariable region (HV)4 and framework regions (FR)2 and -3 bind in the cleft between the two domains of the SAG (37). This binding mode is similar to that observed in the β-SEC2 and β-SEC3 complexes (36), but with several differences, as discussed below. SEC2 and SEC3 bind identically to the 14.3.d β chain, and none of the four amino acid differences between SEC2 and SEC3 is located in the complex interface. This is reflected in the KDs of the two SAGs, which are both approximately 3 µM. The TCR-binding sites of SEC3 and SEB, in contrast, differ at positions 20, 26, and 91; these differences presumably account for the 45-fold weaker affinity of SEB for the 14.3.d β chain (111). ˚2 ˚ 2 (685 A The solvent-excluded surface area for the β-SEB complex is 1343 A ˚ 2 from SEB); the buried surface area for the β-SEC3 from Vβ and 658 A ˚ 2). These values are within the range observed complex is similar (1300 A for antigen-antibody complexes (143) but somewhat smaller than the approx˚ 2 of buried surface in TCR-peptide/MHC complexes (33–35). imately 1800 A As shown in Figure 2A, the TCR-binding site of SEB is adjacent to, but distinct from, the MHC-binding site of this SAG (122, 123). This spatial proximity

P1: SAT/VKS

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

March 4, 1999

P2: NBL/plb

12:42

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Figure 2 TCR β-SEB complex. (A) Ribbons diagram of the VβCβ-SEB complex (37). Colors are as follows: Vβ (yellow), CDR1 (pink), CDR2 (red), CDR3 (gray), HV4 (blue), Cβ (brown), SEB large domain (green), and SEB small domain (blue). Residues of Vβ and SEB involved in interactions in the TCR-SAG interface are red. Residues of SEB in contact with MHC in the structure of the SEB-HLA-DR1 complex (122) are yellow. The SEB disulfide loop (light gray), which is not visible in the electron density map of the β-SEB complex, was modeled according to the uncomplexed SEB crystal structure (9). (B) Interactions in the β-SEB interface. View is the same as in panel A. Vβ atoms are colored accordingly to atom type: carbon, nitrogen and oxygen atoms are yellow, blue, and red, respectively. SEB atoms are colored green (large domain) and blue (small domain). SEB residues are indicated with asterisks. Hydrogen bonds are dotted brown lines.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

T CELL ACTIVATION BY SUPERANTIGENS

447

suggests that the two binding sites may not be completely independent; that is, the affinity of the TCR for SEB alone may be lower (or higher) than its affinity for SEB bound to MHC class II molecules.

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Structure of the β-SEB and β-SEC Interfaces The Vβ residues in contact with SEB are as follows: His47 of FR2; Tyr50, Ala52, Gly53, Ser54, and Thr55 of CDR2; Glu56, Lys57, Tyr65, Lys66, and Ala67 of FR3; and Pro70 and Ser71 of HV4 (Table 1). The FR2, CDR2, FR3, and HV4 regions contribute 7%, 50%, 34%, and 9%, respectively, of the total contacts to SEB. The crystal structure therefore readily accounts for mutational and genetic evidence implicating Vβ CDR2 and HV4 in SAG recognition (3). In the β-SEC3 complex (36), the Vβ residues in contact with the SAG are as follows: Tyr50, Ala52, Gly53, Ser54, and Thr55 of CDR2; Glu56, Lys57, and Lys66 of FR3; and Pro70 and Ser71 of HV4 (Table 1). The CDR2, FR3, and HV4 regions contribute 63%, 32%, and 7%, respectively, of the total contacts to SEC3. Thus, although there are several differences in contacting residues in the two complexes (e.g. Vβ FR2 His47, which contacts SEB but not SEC3, and Vβ CDR2 Gly51, which contacts SEC3 but not SEB), CDR2 and FR3 account for the majority of interactions with the SAG in both complexes, with HV4 playing only a secondary role. The binding sites on the TCR for SAG and peptide/MHC class I molecules only partially overlap. As shown in Table 1, only Vβ residues Tyr50, Ala52, Thr55, and Glu56 contact both SEB and peptide/MHC in the 2C TCR-dEV8/H-2Kb complex (34). The SAG residues in contact with Vβ are as follows: Asn60, Tyr90, and Tyr91 (Val91 in SEC3) of the small domain; and Thr18, Gly19, Leu20 (Thr20 in SEC3), Glu22 (in β-SEB only), Asn23, Tyr26 (in β-SEC3 and β-SEB V26Y only), Phe177 (Phe176 in SEB), and Glu210 of the large domain (Table 1). Residues Asn23, Asn60, and Tyr90 are strictly conserved among bacterial SAGs reactive with mouse Vβ8.2, including SEC1–3 and SPEA, and have been shown to constitute energetic hot spots for binding the 14.3.d β chain (111) (Figure 3A, see color). The structures of the β-SEB and β-SEC complexes enable us to understand how SEB and SEC, which have nearly identical Vβ specificities, can each stimulate T cells expressing Vβ domains from a number of different families (3). As shown in Figure 2B, all the hydrogen bonds between SEB and Vβ are formed between SEB side-chain atoms and Vβ main-chain atoms, except for a hydrogen bond between the main-chain oxygen of SEB Thr18 and the side chain of Vβ Lys57 (SEB Thr18 O-Nζ Lys57 Vβ). Four of the mainchain–side-chain hydrogen bonds in the β-SEB complex are also present in the β-SEC3 complex: Vβ Gly53 O-Nε2 Gln210 SEB, Vβ Thr55 N-Oδ1 Asn23 SEB, Vβ Thr55 O-Nδ2 Asn23 SEB, and Vβ Pro70 O-Nδ2 Asn60 SEB (Table 1).

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SAT/VKS

March 4, 1999

P2: NBL/plb

12:42

QC: NBL/tkj T1: NBL

Annual Reviews AR078-01

P1: SAT/VKS

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

March 4, 1999

P2: NBL/plb

12:42

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Figure 3 Residues defining the interaction of bacterial SAGs with TCR β chains. (A) Sequence alignment of bacterial SAGs (SEB, SEC1-3, SPEA, SEA, SED, SEE, SPEC, TSST-1) based on structural information. The α-carbon skeletons were first optimally superposed. Sequences were then manually adjusted to minimize the number of gaps while respecting the structural similarity. Residues 116-173 and 232-239 are not shown. SEC3 residues in contact with the TCR β chain are boxed in colors according to the loss of binding free energy (11G) upon alanine substitution: (red) >2.5 kcal/mol; (yellow) 1.5-2.5 kcal.mol; (green) 0.5-1.5 kcal/mol; (blue) <0.5 kcal/mol (111). The homologous residues in the other SAGs are only boxed in color if they are identical with those in SEC3. SEB residues contacting MHC in the crystal structure of the SEB-HLA-DR1 complex (122), and the corresponding residues of SEC3, are boxed in cyan if identical in SEB and SEC3 and in magenta if different. (B) Sequence alignment of selected mouse (m) and human (h) TCR β chains reactive with SEB or SEC(3). Only SAG-containing residues are shown.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

448

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL Table 1 Contacts between the 14.3.d TCR β chain and staphylococcal SAGsa Hydrogen bonds β

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

G53 T55c

K57 P70

SEBb O N O O Nζ O

SEC3b

Q210 N23

Nε2 Oδ1 Nδ2

T18 N60

O Nδ2

Q210 N23 T20

Nε2 Oδ1 Nδ2 Oγ 1

N60

Nδ2

Van der Waals contactsd β

SEB

No. of contacts

H47

L20 F177 Y91

1 4 10

Y90 N23

4 1

S54

N23

2

T55c

L20 N23 E22 F177 L20 N23 G19 L20 F177 F177 F177 N60 N60

4 2 2 1 1 1 3 2 1 6 4 2 2

Y50c G51 A52c G53

E56c K57 Y65 K66 A67 P70 S71

SEC3

No. of contacts

V91 V91 Y90 N23 Y26 Q210 N23 V91 T20 N23

1 4 5 1 7 4 5 1 2 4

T20

1

G19 T20

4 4

F176 F176 L58 N60

7 3 1 1

a TCR, T cell receptor; SAG, superantigen; SE, staphylococcal enterotoxin; MHC, major histocompatibility complex. b Data for the β-SEB and β-SEC3 complexes are from References 37 and 36, respectively. c Vβ residues in contact with peptide/MHC in the 2C TCR-dEV8/H2Kb complex (34). d ˚ Van der Waals contacts <4.0 A.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL ACTIVATION BY SUPERANTIGENS

449

The importance of these conserved interactions to complex stabilization is demonstrated by the finding that SEC3 mutants Asn23 → Ala and Gln210 → Ala bind the TCR β chain 70-fold less tightly than does the wild-type SAG, whereas SEC3 Asn60 → Ala binds 16-fold less tightly (111). A recognition mechanism involving a major role for main-chain hydrogen bonds can be highly sequence independent, enabling SEB or SEC to recognize virtually any Vβ domain in which the positions of the relevant main-chain atoms are close to those of mouse Vβ8.2; a similar binding mode has been described for peptide/MHC complexes (144, 145). A sequence alignment of mouse and human Vβ families reactive with SEB or SEC illustrates the diversity of amino acids that can be accommodated at Vβ-contacting positions (Figure 3B). Four water molecules were found to form hydrogen bonds bridging Vβ and SEB: Vβ Ala67 O-H2O-Oε2 Glu22 SEB, Vβ Tyr50 O-H2O-O Tyr91 SEB, Vβ Tyr65 O-H2O-Nδ2 Asn178 SEB, and Vβ Lys66 Nζ -H2O-N Phe177 SEB. Bound water molecules have also been observed in the combining site of antibodies, where they act to increase complementarity in the interface with antigen (146–148). There are no direct contacts between SEB or SEC and Vβ CDR3, which folds away from the SAG (Figure 2A, Table 1); this is consistent with the finding that bacterial and viral SAGs stimulate T cells expressing particular Vβ elements without obvious selection for Vβ CDR3 length or sequence (3–5). However, this does not rule out the possibility that, depending on its conformation, Vβ CDR3 may in certain cases modulate SAG reactivity. For example, when the Vβ domain of TCR A6 (33) is superposed onto the 14.3.d Vβ domain in the VβCβ-SEB structure, SEB Tyr94 is predicted to contact Leu98 of A6 Vβ CDR3, located at the tip of this long protruding loop (not shown). Similar interactions may explain the observed influence of Vβ CDR3 residues on T cell reactivity toward MAM (149) and mouse retroviral Mtv-9 SAG (150). Alternatively, these SAGs may bind the TCR in different orientations than SEB. The latter possibility is supported by the finding that reactivity to the mouse retroviral SAG Mls-1 is affected by mutations at Vβ positions 19, 20, and 24, which are not part of the interface with SEB or SEC (3, 4, 151).

Conformational Changes in the TCR β–SEB Interface The availability of high-resolution crystal structures for uncomplexed 14.3.d β chain (27) and SEB (9) permits an assessment of whether any conformational changes occur in Vβ or the SAG upon complex formation. The free and complexed Vβ domains superpose with a root-mean-square (RMS) difference ˚ for all α-carbon atoms. Likewise, the unbound and bound SEB of 0.33 A ˚ Thus, there are no mamolecules superpose with a RMS difference of 0.54 A. jor rearrangements in the polypeptide backbones of Vβ or SEB associated with

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

450

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

complex formation, as also noted for the β-SEC3 complex (36). However, a number of adjustments in Vβ and SAG side-chain positions are evident when comparing the free and bound structures of the 14.3.d β chain and SEB. Certain of these changes are necessary to avoid steric clashes between the TCR and SAG, whereas others probably serve to maximize productive interactions between the two proteins. For example, the side chain of SEB residue Tyr91 undergoes a 120◦ rotation away from Vβ in order to avoid a collision with CDR2 (37). The structural rearrangements in the β-SEB interface are of similar magnitude to those observed in antigen-antibody complexes in which the antigen is a protein (146, 147, 152, 153). They imply a limited “induced fit” mechanism for TCR-SAG recognition analogous to that described for antigen-antibody interactions (152, 154). A further indication of flexibility in TCR-SAG association comes from the finding that the two β-SEB molecules in the asymmetric unit of the crystal, although similar, are not identical. A rotation of 6◦ is required to optimize the overlap between SEB molecules in the two complexes following superposition of their Vβ domains (37). The changes in the conformation of interface residues in the 14.3.d β-SEB complex are not as large as those in the 2C TCR-dEV8/H-2Kb complex, in ˚ relawhich the CDR1 and CDR3 loops of the Vα domain are displaced 4–6 A tive to their positions in the unliganded 2C TCR structure (34). This probably indicates that the SAG has evolved to optimize its fit to the TCR. Indeed, calculations of shape complementarity (155) reveal that the β-SEB interface is about as tightly packed as antigen-antibody interfaces but significantly more tightly packed than TCR–peptide/MHC class I interfaces (33, 34, 156). It is important to emphasize, however, that, for both 2C TCR-dEV8/H-2Kb and 14.3.d TCR β–SEB complexes, the observed conformational changes are localized to the interfaces between the proteins and are not transmitted to the constant regions of the TCR. Thus, the possibility that changes in TCR conformation upon ligand binding are responsible for initiating T cell signaling can probably be ruled out. Rather, mechanisms based on ligand-induced TCR oligomerization (157) are more likely to account for T cell activation by peptide/MHC or SAGs.

Structural Basis for the Vβ-binding Specificity of SEB and SEC The structure of the 14.3.d β chain–SEB complex explains why SEB recognizes certain Vβ families but not others. As discussed above, all the hydrogen bonds between SEB and mouse Vβ8.2 are formed between SEB side chains and Vβ main-chain atoms (Figure 2B, Table 1), such that the positions of these main-chain atoms should be similar in Vβ domains reactive with SEB but significantly different in Vβs that do not bind this SAG. A comparison of Vβ domains

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL ACTIVATION BY SUPERANTIGENS

451

of known three-dimensional structure confirms this expectation. Thus, SEB activates T cells bearing mouse Vβ8 and human Vβ12, but not mouse Vβ2 or Vβ5 (3). When human Vβ12.3 (33) is superposed onto mouse Vβ8.2 (Figure 4A, see color plates), the RMS difference in α-carbon positions for 14 ˚ However, when mouse Vβ2.3 residues in the SEB-binding site is only 0.9 A. ˚ (30) is superposed onto mouse Vβ8.2 (Figure 4B), the RMS difference is 3.0 A. This difference is largely attributable to a strand switch in Vβ2.3 relative to other Vβ domains of known structure: In Vβ2.3, the c00 strand is hydrogen bonded to the d strand of the adjacent (outer) β sheet, whereas in other Vβs the c00 strand is associated with the c0 strand in the same (inner) sheet (27, 29, 31, 33). A consequence of the c00 strand switch is a repositioning of CDR2 and FR3, which contribute 50% and 34%, respectively, of the total contacts to SEB. It is interesting that no bacterial or viral SAGs have been described with reactivity toward members of the mouse Vβ2 family (3), consistent with the unique folding topology of Vβ2.3. Similarly, when mouse Vβ5.2 (31) is superposed onto mouse Vβ8.2 (Figure 4C ), the RMS difference in α-carbon positions for residues in the ˚ This difference is mainly attributable to a displaceSEB-binding site is 2.1 A. 00 ment of the c strand in a direction opposite from that of the c00 strand in the mouse Vβ2.3 domain, which again results in a repositioning of CDR2 and FR3. Except for MAM, which reacts with mouse Vβ5.1, no SAGs specific for members of the mouse Vβ5 family have been reported (3). These results indicate that the relative position of the c00 strand in Vβ domains is critical in determining their reactivity toward different microbial SAGs and suggest that Vβs reactive with SEB or SEC (mouse Vβ3, 7, 10, and 17; human Vβ3, 5, 12, 13, 14, 15, 17, and 20) probably have a β-strand topology in their SAG-binding sites similar to that of mouse Vβ8.2. The structures of the β-SEB and β-SEC complexes also explain why T cells expressing mouse Vβ8.2 are stimulated by SEB, SEC1–3, and SPEA, but not by SEA, SED, SEE, TSST-1, or SPEC (3, 158). When SEC3 (12) is superposed onto SEB (9), the RMS difference in α-carbon positions for 11 residues in the ˚ (Figure 5A, see color plates). Although the TCR-binding site is only 0.51 A three-dimensional structure of SPEA is not known, a sequence alignment with SEB and SEC reveals that it retains several key Vβ-contacting residues, in particular Asn60, Tyr90, and Gln210 (Figure 3A). Alanine-scanning mutagenesis has shown that these three residues are hot spots for the binding of SEC3 to the 14.3.d β chain (111), in agreement with the fact that SAGs having other residues at these positions display different Vβ-binding specificities. Thus, SEA, SED, SEE, SPEC, and TSST-1, which do not activate Vβ8.2-bearing T cells, differ from SEB at nearly all Vβ-contacting positions, in particular 90 and 210 (Figure 3A). Furthermore, when SEA (14) is superposed onto SEB, the RMS difference in β-carbon positions for residues in the TCR-binding site

P1: SAT/VKS

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

March 4, 1999

P2: NBL/plb

12:42

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Figure 4 Comparison of Vβ structures in the region of the SAG-binding site. (A) Mouse Vβ8.2 (red) (27) superposed onto human Vβ12.3 (blue) (33). The CDR loops are numbered 1, 2 and 3; HV4 is labeled 4. (B) Mouse Vβ8.2 (red) superposed onto mouse Vβ2.3 (yellow) (30). (C) Mouse Vβ8.2 (red) superposed onto mouse Vβ5.2 (green) (31). The SEB-binding site of mouse Vβ8.2 is circled in each panel. The c00 and d strands are labeled.

P1: SAT/VKS

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

March 4, 1999

P2: NBL/plb

12:42

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Figure 5 Structural comparison of SEB with other bacterial SAGs. SEB is oriented with its TCRbinding site directly facing the reader. (A) α-carbon diagram of SEB (yellow) (9) superposed onto SEC3 (dark blue) (12). (B) SEB (yellow) superposed onto SEA (green) (14). (C) SEB (yellow) superposed onto TSST-1 (pink) (1911). (D) SEB (yellow) superposed onto SPEC (light blue) (16). Regions of SEB in contact with Vβ in the 14.3.d β-SEB complex are red; key contact residues are labeled.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

452

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

˚ (Figure 5B). As shown in Figures 5C and D, the putative TCR-binding is 2.8 A sites of TSST-1 (10, 11) and SPEC (16) are markedly different from that of SEB; this can account for the finding that the Vβ specificities of TSST-1 and SPEC do not overlap with those of SEB or SEC (3).

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

COMPARISON OF TCR-SAG-PEPTIDE/MHC AND TCR-PEPTIDE/MHC COMPLEXES Although the three-dimensional structure of a TCR-SAG-peptide/MHC complex has not been determined, a model of this complex may be readily constructed by least-squares superposition of (a) the 14.3.d VβCβ-SEB complex, (b) the SEB–peptide/HLA-DR1 complex (122), and (c) the 2C TCR αβ heterodimer (29), which uses the same Vβ element (mouse Vβ8.2) as does 14.3.d TCR (Figure 6A, see color). The accuracy of this model, in which the SAG is seen to bridge the APC and the T cell, depends on the assumption that there are no major conformational changes in any of the individual components upon complex formation; such changes are unlikely, given that none are observed in the TCR β-SEB or SEB-peptide/DR1 complexes (37, 122). This model may be compared with the structure of the 2C TCR complexed with peptide/MHC class I (34) (Figures 6B and C ). Assuming that TCRs bind MHC class I and class II molecules in similar orientations, as recently argued on the basis of structural considerations (33), it is apparent that the binding of peptide/MHC to TCR in the TCR-peptide/MHC complex is different from that in the TCR-SEB-peptide/MHC complex and that there is only partial overlap between the binding sites on the TCR for SEB and for peptide/MHC. In the TCR-peptide/MHC complex (Figure 6B), the peptide antigen, as well as both the α1 and α2 helices of the class I molecule, simultaneously engage the TCR combining site. By contrast, in the model of the TCR-SEB-peptide/MHC complex (Figure 6A), the peptide is effectively removed from the TCR combining site, and there are no direct contacts between the TCR β chain and the MHC class II α1 or β1 (which corresponds to α2 in class I) helices. However, as discussed below, the MHC β1 helix is predicted to interact with the TCR Vα domain. In addition, the rotational orientation of TCR and MHC molecules in the TCR-SEB-peptide/MHC complex is different from that in the TCR-peptide/MHC complex (compare Figures 6A and C, respectively, in which the TCRs are shown in the same orientation): The MHC class II molecule in Figure 6A must be rotated approximately 40◦ counterclockwise around a vertical axis to align it with the MHC class I molecule in Figure 6C. Therefore, even though the TCR engages peptide/MHC differently in the two types of complexes, the end result—highly efficient T cell activation—is similar. This implies that the specific geometry of TCR engagement by peptide/MHC

P1: SAT/VKS

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

March 4, 1999

P2: NBL/plb

12:42

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Figure 6 Comparison of TCR-SAG-peptide/MHC and TCR-peptide/MHC complexes. (A) Model of the TCR-SEB-peptide/MHC class II complex constructed by least-squares superposition of: 1) the 14.3.d VβCβ-SEB complex (37), 2) the SEB-peptide/HLA-DR1 complex (122), and 3) the 2C TCR αβ heterodimer (29). (B) Structure of the 2C TCR-peptide/MHC class I complex (34). The complex is oriented such that the MHC molecule is approximately aligned with that in panel A. (C) Another view of the 2C TCR-peptide/MHC class I complex. The complex is oriented such that the TCRs in panels A and C are aligned.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL ACTIVATION BY SUPERANTIGENS

453

may be less critical than are other factors, such as the affinity and kinetics of the binding reaction, in triggering T cells. By acting as a wedge between the TCR β chain and the MHC class II α chain, the SAG is able to circumvent the normal mechanism for T cell triggering by specific peptide/MHC complexes. The result is polyclonal activation of whole populations of T cells expressing particular Vβ elements, largely irrespective of the peptide/MHC specificity of the corresponding TCRs. The absence of direct contacts between peptide and TCR in the model of the TCR-SEB–peptide/MHC class II complex in Figure 6A can explain the finding that formation of a TCR-SEB-DR1 complex was not affected by the presence of several different DR1-bound peptides (109). However, depending on the particular peptide bound by the MHC class II molecule, and on the conformations of the Vα and Vβ CDR3 loops, the peptide may in certain cases make a small number of contacts with the TCR and so modulate SAG activity. It must be emphasized that other SAGs may bind differently to TCR and/or MHC class II than does SEB or SEC, thereby affecting the geometry of the TCR-SAG-peptide/MHC complex. In the case of TSST-1 bound to HLA-DR1 (Figure 1B), the SAG reaches across the antigen-binding groove, such that peptide-SAG and SAG-MHC β chain interactions may also contribute to complex stabilization (124). Furthermore, in contrast to SEB and SEC, the mode of binding of TSST-1 to DR1 would probably also preclude direct TCR-MHC interactions. This may represent an extreme example of the model in Figure 6A and illustrates how variations in the structure and positioning of the SAG wedge provide a means for different SAGs to modulate the degree of TCR-MHC interactions in the TCR-SAG-peptide/MHC complex.

Role of the TCR α Chain in Stabilization of the TCR-SAG-Peptide/MHC Complex There is increasing evidence that the TCR α chain may, in certain cases, play a role in stabilizing the TCR-SAG-peptide/MHC complex and thereby influence T cell reactivity to bacterial or viral SAGs (5, 142, 159). For example, it was shown that Vα4 is expressed by Vβ6+ T cell hybridomas that react with SEB but not by Vβ6+ hybridomas that do not respond to this SAG (160). Transfection experiments demonstrated that the Vα4 α chain transferred SEB responsiveness regardless of whether the Vβ6 β chain was derived from a responsive or nonresponsive hybridoma (161). These effects of the TCR α chain on T cell activation by SAGs may be mediated through an interaction between Vα and the MHC class II β chain (5, 142, 161, 162). Thus, mutations at position 77 of the I-Ek β chain (141) and at positions 77 and 81 of HLA-DR1 β chain (140) were found to greatly reduce the T cell response to SEB without affecting binding of the SAG to MHC class II, which suggests contacts to the TCR. In addition,

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

454

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

the affinity of SEB for a soluble human TCR was observed to be significantly enhanced by the addition of soluble HLA-DR1 (109). These results may be understood in terms of the model of the TCR-SEB-peptide/MHC complex in Figure 6A, in which the Vα domain of the 2C TCR is predicted to interact with ˚ between Vα CDR2 the MHC β1 helix. A close-up of putative contacts (<4 A) residues Ser51, Gly52, and Asp53 and the class II β chain residues Asp76 and Thr77 is shown in Figure 7A (see color plates). Thus, the overall stability of the TCR-SAG-peptide/MHC complex is probably determined by the combined strengths of three separate sets of interactions: TCR β chain–SAG, SAG–MHC α chain, and MHC β chain–TCR α chain. The preferential expression of certain Vα regions among SAG-reactive T cells has been interpreted as evidence that these particular Vαs interact with MHC more favorably than do other Vαs during SAG-mediated T cell activation (5, 142, 159–161). The availability of crystal structures for several αβ TCR heterodimers (29–31, 33) allows an examination of the possible effects of different TCR α chains on Vα-MHC interactions in the TCR-SAG-peptide/MHC complex. By superposing the Vβ domain of the 14.3.d β-SEB complex onto the Vβ domain of TCR A6 (33), N15 (31), or KB5-C20 (30), as described above for the 2C TCR, it is apparent that the extent of interaction between the Vα and MHC β1 domains in the TCR-SEB-peptide/MHC complex depends mostly on the relative orientation of Vα and Vβ domains in each TCR heterodimer. Because the variability in the geometry of Vα/Vβ association among these TCRs is considerable (2, 31), large differences are observed in the extent of Vα-MHC β1 interactions. These are illustrated in Figure 7. For the 2C TCR, as discussed above, Vα CDR2 Ser51, Gly52, and Asp53 are predicted to contact MHC β1 Asp76 and Thr77 (Figure 7A). For the A6 TCR, Vα CDR1 Gln30, and CDR2 Tyr50, Ser51 and Asn52 contact MHC β1 Glu69, Ala73, Asp76, Thr77, and His81 (Figure 7B). For the N15 TCR, Vα CDR1 Leu29 and CDR2 Thr51 contact MHC β1 Ala73, Thr77, and His81 (Figure 7C ). For TCR KB-C50, no contacts are predicted because of the particular geometry of Vα/Vβ association of this TCR (not shown). In all cases where contacts between Vα and the MHC molecule are expected to occur, however, these involve Vα CDR2 and residues on the MHC β1 helix pointing away from the peptide-binding groove. Thus, depending on the geometry of Vα/Vβ association and on the structure of the Vα CDR2 loop, Vα-MHC class II interactions may (a) contribute to stabilizing the TCR-SEB-peptide/MHC complex and thus increase reactivity toward the SAG, (b) have no net effect on complex stability and not affect reactivity, or (c) destabilize the complex through unfavorable contacts and thereby decrease reactivity. In this way, the TCR α chain may modulate the level of activation by SEB of T cells expressing the same Vβ but different Vαs.

P1: SAT/VKS

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

March 4, 1999

P2: NBL/plb

12:42

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Figure 7 Differences in Vα-MHC β chain contacts as a result of differences in Vα/Vβ orientation and Vα CDR sequences. (A) Close-up of putative contacts between the MHC class II β chain (yellow) and Vα of the 2C TCR (blue) in the model of the TCR-SEB-peptide/MHC complex in Figure 6A. The bound peptide is red. Only those MHC and Vα residues predicted to form direct contacts are labeled. In the upper right hand corner is a portion of the MHC class II α chain in green. (B) Contacts between the MHC class II β chain and Vα of the A6 TCR (orange). The model of the TCR-SEB-peptide/MHC complex was constructed in the same way as that in Figure 7A, except using TCR A6 (33) instead of TCR 2C (29). The Vβ domain of TCR A6 was superposed onto 14.3.d Vβ by overlapping structurally equivalent FR residues. (C) Contacts between the MHC class II β chain and Vα of the N15 TCR (pink). The Vβ domain of TCR N15 (31) was superposed onto 14.3.d Vβ by overlapping structurally equivalent FR residues.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

T CELL ACTIVATION BY SUPERANTIGENS

455

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POTENTIAL THERAPEUTIC APPLICATIONS OF SUPERANTIGENS Because bacterial SAGs are such extremely potent activators of the immune system, efforts are currently underway in a number of laboratories to engineer them for therapeutic applications (163). Knowledge of the three-dimesional structure of SAGs, and of their binding sites for TCR and MHC, may be used to design variants with altered binding properties toward these ligands, resulting in desired biological effects. The potential applications of SAG derivatives include cancer immunotherapy and the treatment of infectious and autoimmune diseases. The recruitment of antigen-specific cytotoxic T lymphocytes (CTLs) is a major goal for the immunotherapy of malignant tumors. However, the frequency of tumor-specific CTLs is generally too low to interfere with progressive tumor growth. An attractive approach for immunotherapy is to use antibodies specific for tumor-associated antigens to target large numbers of T cells to the tumor. Taking advantage of the ability of SAGs to activate large populations of T cells, chemical conjugates of SEA and the colon carcinoma– reacting monoclonal antibodies (mAbs) C215 or C242 were shown to mediate T cell–dependent destruction of colon carcinoma cells lacking MHC class II molecules (164). The SEA-mAb–mediated cytotoxicity was MHC class II independent and did not require antigen-specific effector CTLs. In subsequent work, a recombinant fusion protein of SEA and the Fab region of the C215 mAb was found to efficiently target T cells to lyse C215+ MHC class II–negative human colon carcinoma cells (165). Treatment of mice carrying B16 melanoma cells expressing transfected C215 antigen resulted in 85–99% inhibition of tumor growth and allowed long-term survival. In similar experiments, SEA bound to specific anti-carcinoma cell or anti-ganglioside GD2 mAbs displayed T cell–mediated cytotoxicity toward MHC class II–negative lymphatic leukemia cell lines or neuroblastoma cells, respectively (166, 167). The demonstration of a Zn2+-dependent MHC class II binding site with high affinity in the large domain of SEA (14, 128, 129), which is distinct from the low-affinity SEB-like binding site in the small domain, prompted the introduction of a point mutation (Asp227 → Ala) in the high-affinity site in order to lower the systemic toxicity of Fab-SEA conjugates (168). Thus, after treatment with Fab-SEA Asp227 → Ala, a 100- to 1000-fold reduction in serum levels of IL-6 and TNF was observed in mice compared with the wild-type conjugate, without affecting anti-tumor activity. These results suggest that bacterial SAGs can be converted into tolerable immunotoxins for cancer therapy (169).

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

456

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

Inactivated forms of SAGs might be useful as vaccines to protect against staphylococcal or streptococcal toxic shock. Mutants of SEB that do not bind TCR do not induce T cell proliferation and therefore do not cause toxic shock (170). Animals immunized with mutants of SEA with attenuated binding to TCR or MHC class II developed high titers of anti-SEA antibodies and were fully protected against challenge with the wild-type toxin (171). Formalinized SEB toxoid–containing microspheres have been tested for efficacy in rhesus monkeys as a vaccine candidate for respiratory toxicosis and toxic shock (172). Protective immunity correlated with antibody levels in both the circulation and the respiratory tract. Similar results were obtained with intranasal or intramuscular immunization by using meningococcal outer-membrane protein proteasome-SEB toxoid. The proteosome-SEB toxoid vaccine was efficacious in protecting 100% of monkeys against severe symptomatology and death from aerosolized-SEB intoxication (173). T cell activation by SAGs is generally followed by the disappearance or inactivation of the responding T cells, resulting in clonal deletion of cells bearing specific Vβ elements (3, 4). Chronic exposure to low concentrations of SAGs permits clonal deletion to occur directly, without the cells first passing through a state of hyperreactivity (174, 175). In mice with EAE or lupus nephritis, both of which serve as models for human autoimmune diseases, treatment with SEB resulted in a reduction in symptoms or in a cure of the disease (176–179). In both models, autoimmunity is known to be mediated by self-reactive T cells expressing a single Vβ element (mouse Vβ8), such that the effect of the SAG most likely results from the specific elimination of pathogenic T cells bearing that particular Vβ. These experiments suggest that SAGs could potentially be used for the prevention of autoimmune disease by selectively eliminating specific T cell populations. In cases where T cells from several Vβ families might be involved in the disease process, SAGs engineered to recognize these Vβs could be used for therapy. Knowledge of the three-dimensional structure of TCR-SAG complexes should facilitate the design of SAGs with predefined Vβ specificities. A number of concerns must be addressed, however, before SAGs can be used as therapeutic agents (180). Administration of SAGs may lead to the release of dangerous levels of cytokines such as TNF, resulting in toxic shock. In addition, inadvertent stimulation of autoreactive T cells could trigger autoimmune disease. It appears likely, however, that the systemic toxicity of SAGs can be dissociated from the superantigenic effects of these molecules through structure-based genetic engineering, as described above for SEA-Fab conjugates (168) and SEA vaccines (171). Similarly, mutations of TSST-1 have been described that alter either lethality or superantigenicity, without significantly affecting the other property (181–183).

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

T CELL ACTIVATION BY SUPERANTIGENS

457

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

FUTURE DIRECTIONS Although the X-ray crystallographic studies described above represent important advances in our understanding of SAG interactions with TCR and MHC, the apparent diversity of these interactions clearly illustrates the need for further work in this area. For example, TSST-1 binds differently to MHC class II molecules than does SEB (Figure 1) (122–124), and its interaction with the TCR presumably differs as well. Unlike SEB and TSST-1, which have only one class II–binding site, mutagenesis and biochemical experiments indicate that SEA has two such sites (127–129). However, because the crystal structure of an SEA–MHC class II complex has not been determined, the precise locations of these sites are unknown. SED and SPEC crystallize as homodimers (15–16), which may facilitate TCR oligomerization and T cell triggering. However, X-ray crystallographic studies are required to elucidate the geometry of the putative SED2TCR2, SED2MHC2, SPEC2TCR2, and SPEC2MHC2 tetramers. Our knowledge of SAG structure is currently limited to the staphylococcal and streptococcal pyrogenic toxins (9–16) and to a staphylococcal exfoliative toxin (17, 18). No structural information is available for SAGs produced by mycoplasma, such as MAM (19), or by viruses (22–26), including the muchstudied MMTV SAGs. Human endogenous retroviral SAGs that are associated with autoimmune diseases (26) may become a focus of future attention. Finally, the structures of entire TCR-SAG-MHC complexes must be determined in order to define experimentally the putative Vα-MHC interactions discussed in this review. X-ray crystallographic studies of TCR-SAG, SAG-MHC, and TCR-SAGMHC complexes will open the way for the design of SAG variants with altered binding properties for TCR and MHC for use as tools in dissecting structureactivity relationships in this system. The relative contributions of TCR-SAG and SAG-MHC interactions to T cell stimulation can be defined by engineering panels of mutant SAGs with both higher and lower affinities for TCR and MHC than the wild-type toxins have. It has been shown that SAGs mimic the interaction of peptide/MHC complexes with the TCR in terms of affinities and kinetics (109–112). It remains to be established, however, whether there is an optimum affinity for T cell activation by SAGs (or by peptide/MHC), as predicted by the serial triggering (118, 119) and kinetic proofreading (120) models of T cell activation, such that SAGs with either higher or lower affinities than this optimum value exhibit decreased ability to stimulate T cells. Alternatively, SAGs with progressively higher affinities for the TCR relative to the wild type may stimulate T cells increasingly well, until some plateau of maximum stimulation is attained. Mutants with altered affinities for the TCR can be used to distinguish between these possibilities and thereby define the affinity and kinetic

P1: SAT/spd

P2: KKK/plb

January 21, 1999

458

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL

parameters governing T cell activation by SAGs. The role, if any, of cooperative interactions involving the TCR α chain in stabilizing the TCR-SAG-MHC complex can be similarly addressed. By thus combining X-ray crystallography with mutagenesis and binding studies, a comprehensive understanding of the physical basis of T cell activation by microbial SAGs should emerge.

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ACKNOWLEDGMENTS This research was supported by NIH grants AI36900 and AI42937 and National Multiple Sclerosis Society grant RG2747 to RAM. ELM is supported by grants from CONICET, UBA, and Fundacin Antorchas. We thank IA Wilson for coordinates of the 2C TCR-dEV8/H-2Kb complex, CV Stauffacher for coordinates of SEC3, EN Baker and J Fraser for coordinates of SPEC, and DC Wiley for coordinates of the TSST-1/HLA-DR1 complex. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Bentley GA, Mariuzza RA. 1995. The structure of the T cell antigen receptor. Annu. Rev. Immunol. 14:563–90 2. Wilson IA, Garcia KC. 1997. T-cell receptor structure and TCR complexes. Curr. Opin. Struct. Biol. 7:839–48 3. Kotzin BL, Leung DYM, Kappler J, Marrack P. 1993. Superantigens and their potential role in human disease. Adv. Immunol. 54:99–166 4. Scherer MT, Ignatowicz L, Winslow GM, Kappler JW, Marrack P. 1993. Superantigens: bacterial and viral proteins that manipulate the immune system. Annu. Rev. Cell. Biol. 9:101–128 5. Webb SR, Gascoigne NRJ. 1994. T-cell activation by superantigens. Curr. Opin. Immunol. 6:467–475 6. Bohach GA, Fast DJ, Nelson RD, Schlievert PM. 1990. Staphylococcal and streptococcal pyrogenic toxins involved in toxic shock syndrome and related illnesses. Crit. Rev. Microbiol. 17:251–272 7. Betley MJ, Borst DW, Regassa LB. 1992. Staphylococcal enterotoxins, toxic shock syndrome toxin and streptococcal pyrogenic exotoxins: a comparative study of their molecular biology. Chem. Immunol. 55:1–35 8. Bohach GA. 1997. Staphylococcal enterotoxins B and C. In Superantigens: Molecular Biology, Immunology and Relevance to Human Disease, ed. DYM

9.

10.

11.

12.

13.

14.

Leung, BT Huber, PM Schlievert, Marcel Dekker Inc. NY, pp. 167–198 Swaminathan S, Furey W, Pletcher J, Sax M. 1992. Crystal structure of staphylococcal enterotoxin B, a superantigen. Nature 359:801–806 Prasad GS, Earhart CA, Murray DL, Novick RP, Schlievert PM, Ohlendorf DH. 1993. Structure of toxic shock syndrome toxin–1. Biochemistry 32:13761– 13766. Acharya KR, Passalacqua EF, Jones EY, Harlos K, Stuart DI, Brehm RD, Tranter HS. 1994. Structural basis of superantigen action inferred from crystal structure of toxic–shock syndrome toxin–1. Nature 367:94–97 Hoffmann ML, Jablonski LM, Crum KK, Hackett SP, Chi Y–I, Stauffacher CV, Stevens DL, Bohach GA. 1994. Predictions of T cell receptor and major histocompatibility complex binding sites on staphylococcal enterotoxin C3. Infect. Immun. 62:3396–3407 Papageorgiou A, Acharya KR, Shapiro R, Passalacqua EF, Brehm RD, Tranter HS. 1995. Crystal structure of the superantigen enterotoxin C2 from Stapylococcus aureus reveals a zinc–binding site. Structure 3:769–779 Schad EM, Zaitseva I, Zaitsev VN, Dohlsten M, Kalland T, Schlievert PM, Ohlendorf DH, Svensson LA. 1995.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

T CELL ACTIVATION BY SUPERANTIGENS

15.

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

16.

17.

18.

19.

20.

21.

22. 23. 24.

Crystal structure of the superantigen staphylococcal enterotoxin type A. EMBO J. 14:3292–3301 Sundstrom M, Abrahmsen L, Antonsson P, Mehindate K, Mourad W, Dohlsten M. 1996. The crystal structure of staphylococcal enterotoxin type D reveals Zn2+– mediated homodimerization. EMBO J. 15:6832–6840 Roussell A, Anderson BF, Baker HM, Fraser JD, Baker EN. 1997. Crystal structure of the streptococcal superantigen SPE–C: dimerization and zinc–binding suggest a novel mode of interaction with MHC class II molecules. Nature Struct. Biol. 4:635–643 Vath GM, Earhart CA, Rago JV, Kim MH, Bohach GA, Schlievert PM, Ohlendorf DH. 1997. The structure of the superantigen exfoliative toxin A suggests a novel regulation as a serine protease. Biochemistry 36:1559–1566 Cavarelli J, Prevost G, Bourguet W, Moulinier L, Chevrier B, Delagoutte B, Bilwes A, Mourey L, Rifau S, Piemont Y, Moras D. 1997. The structure of Staphylococcus aureus epidermolytic toxin A, an ˚ resolution. atypic serine protease, at 1.7 A Structure 5:813–824 Cole BC, Griffiths MM. 1993. Triggering and exacerbation of autoimmune arthritis by the Mycoplasma arthritidis superantigen MAM. Arthritis Rheumat. 36:994– 1002 Ito Y, Abe J, Yoshino K, Takeda T, Koshaka T. 1995. Sequence analysis of the gene for a novel superantigen produced by Yersinia pseudotuberculosis and expression of the recombinant protein. J. Immunol. 154:5896–5906 Miyoshi–Akiyama T, Abe A, Kato H, Kawahara K, Narimatsu H, Uchiyama T. 1995. DNA sequencing of the gene encoding a bacterial superantigen, Yersinia pseudotuberculosis–derived mitogen (YPM), and characterization of the gene product, cloned YPM. J. Immunol. 154:5228–5234 Acha–Orbea H, MacDonald HR. 1995. Superantigens of mouse mammary tumor virus. Annu. Rev. Immunol. 13:459–486 Astoul E, Lafage M, Lafon M. 1996. Rabies superantigen as a Vβ T–dependent adjuvant. J. Exp. Med. 183:1623–1631 Dobrescu D, Ursea B, Pope M, Asch AS, Posnett DM. 1995. Enhanced HIV–1 replication in Vβ12 T cells due to human cytomegalovirus in monocytes: evidence for a putative herpesvirus superantigen. Cell 82:753–763

459

25. Sutkowski N, Palkama T, Ciurli C, Sekal R–P, Thorley–Lawson DA, Huber B. 1996. An Epstein–Barr virus–associated superantigen. J. Exp. Med. 184:971–980 26. Conrad B, Weissmahr RN, Boni J, Arcari R, Schupbach J, Mach B. 1997. A human endogenous retroviral superantigen as candidate autoimmune gene in type I diabetes. Cell 90:303–313 27. Bentley GA, Boulot G, Karjalainen K, Mariuzza RA. 1995. Crystal structure of the β chain of a T cell antigen receptor. Science 267:1984–1987 28. Fields BA, Ober B, Malchiodi EL, Lebedeva MI, Braden B, Ysern X, Kim J–K, Shao X, Ward ES, Mariuzza RA. 1995. Crystal structure of the Vα domain of a T cell antigen receptor. Science 270: 1821–1824 29. Garcia KC, Degano M, Stanfield RL, Brunmark A, Jackson MR, Peterson PA, Teyton L, Wilson IA. 1996. An αβ T cell ˚ and its orientareceptor structure at 2.5 A tion in the TCR–MHC complex. Science 274:209–219 30. Housset D, Mazza G, Gregoire C, Piras C, Malissen B, Fontecilla–Camps JC. 1997. The three–dimensional structure of a T-cell antigen receptor VαVβ heterodimer reveals a novel arrangement of the Vβ domain. EMBO J. 16:4205–4216 31. Wang J, Lim K, Smolyar A, Teng M, Liu J, Tse AGD, Liu J, Hussey RE, Chishti Y, Thomson CTT, Sweet RM, Nathenson SG, Chang H, Sacchettini JC, Reinherz EL. 1998. Atomic structure of an αβ T cell receptor (TCR) heterodimer in complex with an anti–TCR Fab fragment derived from a mitogenic antibody. EMBO J. 17:10–26 32. Li H, Lebedeva MI, Llera AS, Fields BA, Brenner MA, Mariuzza RA. 1998. Structure of the Vδ domain of a human γ δ T-cell antigen receptor. Nature 391:502– 506 33. 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 384:134–141 34. Garcia KC, Degano M, Pease LR, Huang M, Peterson PA, Teyton L, Wilson IA. 1998. Structural basis of plasticity in T cell receptor recognition of a self peptide–MHC antigen. Science 279: 1166–1172 35. Ding Y–H, Smith KJ, Garboczi DN, Utz U, Biddison WE, Wiley DC. 1998. Two human T cell receptors bind in a similar diagonal mode to the HLA–A2/Tax

P1: SAT/spd

P2: KKK/plb

January 21, 1999

460

36.

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

37.

38.

39. 40. 41.

42.

43.

44.

45.

46.

47.

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL peptide complex using different TCR amino acids. Immunity 8:403–411 Fields BA, Malchiodi EL, Li H–M, Ysern X, Stauffacher CV, Schlievert PM, Karjalainen K, Mariuzza RA. 1996. Crystal structure of the β chain of a T-cell receptor complexed with a superantigen. Nature 384:188–192 Li H, Llera A, Tsuchiya D, Leder L, Ysern X, Schlievert PM, Karjalainen K, Mariuzza RA. 1998. Three–dimensional structure of the complex between a T cell receptor β chain and the superantigen staphylococcal enterotoxin B. Immunity MacDonald HR, Lees RK, Baschieri S, Herrmann T, Lussow AR. 1993. Peripheral T-cell reactivity to bacterial superantigens in vivo: the response/anergy paradox. Immunol. Rev. 133:105–117 Woodland DL, Wen R, Blackman MA. 1997. Why do superantigens care about peptides? Immunol. Today 18:18–22 Kappler JW, Roehm N, Marrack P. 1987. T cell tolerance by clonal elimination in the thymus. Cell 49:273–280 MacDonald HR, Schneider R, Lees RK, Howe RC, Acha-Orbea H, Festenstein H, Zinkernagel RM, Hengartner H. 1988. T-cell receptor Vβ use predicts reactivity and tolerance to Mlsa–encoded antigens. Nature 332:40–45 Pullen AM, Bill J, Kubo RT, Marrack P, Kappler JW. 1991. Analysis of the interaction site for the self superantigen Mls– 1a on T cell receptor Vβ. J. Exp. Med. 173:1183–1192 White J, Herman A, Pullen AM, Kubo R, Kappler JW, Marrack P. 1989. The Vβ– specific superantigen staphylococcal enterotoxin B: stimulation of mature T cells and clonal deletion in neonatal mice. Cell 56:27–35 Rellahan BL, Jones LA, Kruisbeek AM, Fry AM, Matis LA. 1990. In vivo induction of anergy in peripheral Vβ8+ T cells by staphylococcal enterotoxin B. J. Exp. Med. 172:1091–1100 O’Hehir RE, Lamb JR. 1990. Induction of specific clonal anergy in human T lymphocytes by Staphylococcus aureus enterotoxins. Proc. Natl. Acad. Sci. USA 87:8884–8888 Kawabe Y, Ochi A. 1991. Programmed cell death and extrathymic reduction of Vβ8+ CD4+ T cells in mice tolerant to Staphylococcus aureus enterotoxin B. Nature 349:245–248 Golovkina TV, Chervonsky A, Dudley JP, Ross SR. 1992. Transgenic mouse mammary tumor virus superantigen expression prevents viral infection. Cell 69:637–645

48. Held W, Waanders G, Shakov AN, Scarpellino L, Acha–Orbea H, MacDonald HR. 1993. Superantigen–induced immune stimulation amplifies mouse mammary tumor virus infection and allows virus transmission. Cell 74:529–540 49. Ignatowicz L, Kappler J, Marrack P. 1992. The effects of chronic infection with a superantigen–producing virus. J. Exp. Med. 175:917–923 50. Golovkina TV, Dudley JP, Jaffe A, Ross SR. 1995. Mouse mammary tumor viruses with functional superantigen genes are selected during in vivo infection. Proc. Natl. Acad. Sci. USA 92:4828– 4832 51. Wrona TJ, Lozano M, Binhazim AA, Dudley JP. 1998. Mutational and functional analysis of the C–terminal region of the C3H mouse mammary tumor virus superantigen. J. Virol. 72:4746–4755 52. Deresiewicz RL. 1997. Staphylococcal toxic shock syndrome. In Superantigens: Molecular Biology, Immunology and Relevance to Human Disease, ed. DYM Leung, BT Huber, PM Schlievert, Marcel Dekker Inc. NY, pp. 435–479 53. Stevens DL. 1997. Streptococcal toxic shock syndrome. In Superantigens: Molecular Biology, Immunology and Relevance to Human Disease, ed. DYM Leung, BT Huber, PM Schlievert, Marcel Dekker Inc. NY, pp. 481–501 54. Rago JV, Schlievert PM. 1998. Mechanisms of pathogenesis of staphylococcal and streptococcal superantigens. Curr. Top. Microbiol. Immunol. 225:81–97 55. Alber G, Hammer DK, Fleischer B. 1990. Relationship between enterotoxic– and T lymphocyte–stimulating activity of staphylococcal enterotoxin. Br. J. Immunol. 144:4501–4506 56. Harris TO, Grossman D, Kappler JW, Marrack P, Rich RR, Betley MJ. 1993. Lack of complete correlation between emetic and T cell stimulatory activities of staphylococcal enterotoxins. Infect. Immun. 61:3175–3183 57. Hovde CJ, Marr JC, Hoffmann ML, Hackett SP, Chi YI Crum KK, Stevens DL, Stauffacher CV, Bohach GA. 1994. Investigation of the role of the disulphide bond in the activity and structure of staphylococcal enterotoxin C1. Mol. Microbiol. 13:897–909 58. Sears CL, Kaper JB. 1996. Enteric bacterial toxins: mechanisms of action and linkage to intestinal secretion. Microbiol. Rev. 60:167–215 59. Hamad ARA, Marrack P, Kappler JW. 1997. Transcytosis of staphylococcal

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

T CELL ACTIVATION BY SUPERANTIGENS

60.

61.

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

62. 63.

64.

65.

66.

67.

68.

69.

70.

71.

superantigen toxins. J. Exp. Med. 185: 1447–1454 Rose NR, Wolfgram LJ, Herskowitz A, Beisel KW. 1986. Postinfectious autoimmunity: two distinct phases of coxsackie B3–induced myocarditis. Ann. N.Y. Acad. Sci. 475:146–156 Ray CG, Palmer JP, Crossley JR, Williams RH. 1980. Coxsackie B virus antibody responses in juvenile–onset diabetes mellitus. Clin. Endocrinol. 12:375– 378 Oldstone MBA. 1990. Molecular mimicry and autoimmune disease. Cell 50: 819–820 Wucherpfenning KW, Strominger JL. 1995. Molecular mimicry in T-cell mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 80:695–705 Hemmer B, Fleckenstein BT, Vergelli M, Jung G, McFarland H, Martin R. 1997. Identification of high potency microbial and self ligands for a human autoreactive class II–restricted T cell clone. J. Exp. Med. 185:1651–1659 Renno T, Acha–Orbea H. 1996. Superantigens in autoimmune disease: still more shades of gray. Immunol. Rev. 154: 175–191 Bach JF. 1995. T cell receptor use in organ–specific human autoimmune diseases other than rheumatoid arthritis and multiple sclerosis. Ann. N.Y. Acad. Sci. 756:453–459 Traugott U. 1987. Multiple sclerosis: relevance of class I and class II MHC– expressing cells to lesion development. J. Neuroimmunol. 16:283–302 Olerup O, Hillert J. 1990. HLA class II associated genetic susceptibility in multiple sclerosis: a critical evaluation. Tissue Antigens 38:1–5 Allegretta M, Nicklas JA, Sriram S, Albertini RJ. 1990. T cells responsive to myelin basic protein in patients with multiple sclerosis. Science 247:718–721 Wucherpfennig KW, Zhang J, Witek C, Matsui M, Modabber Y, Ota K, Hafler DA. 1994. Clonal expansion and persistence of human T cells specific for an immunodominant myelin basic protein peptide. J. Immunol. 150:5581–5592 Zhang J, Markovic S, Lacet B, Raus J, Weiner HL, Hafler DA. 1994. Increased frequency of interleukin 2–responsive T cells specific for myelin basic protein and proteolipid protein in peripheral blood and cerebrospinal fluid of patients with multiple sclerosis. J. Exp. Med. 179: 973–984

461

72. Stinissen P, Raus J, Zhang J. 1997. Autoimmune pathogenesis of multiple sclerosis: role of autoreactive T lymphocytes and new immunotherapeutic strategies. Crit. Rev. Immunol. 17:33–75 73. Brocke S, Gaur A, Piercy C, Gautam K, Gijbels C, Fathman CG, Steinman L. 1993. Induction of relapsing paralysis in experimental autoimmune encephalomyelitis by bacterial superantigen. Nature 365:642–644 74. Racke MK, Quigley L, Cannella B, Raine C, McFarlin DE, Scott DE. 1994. Superantigen modulation of experimental allergic encephalomyelitis: activation or anergy determines outcome. J. Immunol. 152:2051–2059 75. Brocke S, Hausmann S, Steinman L, Wucherpfennig KW. 1998. Microbial peptides and superantigens in the pathogenesis of autoimmune diseases of the central nervous system. Sem. Immunol. 10:57–67 76. Paliard X, West SG, Lafferty JA, Clements JR, Kappler JW, Marrack P, Kotzin BL. 1991. Evidence for the effects of a superantigen in rheumatoid arthritis. Science 253:325–329 77. Haqqi TM, Anderson GD, Banerjee S, David CS. 1992. Restricted heterogeneity in T-cell antigen receptor Vβ gene usage in the lymph nodes and arthritic joints of mice. Proc. Natl. Acad. Sci. USA 89:1253–1255 78. Osman GE, Toda M, Kanagawa O, Hood LE. 1993. Characterization of the T cell receptor repertoire causing collagen arthritis in mice. J. Exp. Med. 177:387– 395 79. Chiocchia G, Boissier MC, Fournier C. 1991. Therapy against murine collagen– induced arthritis with T-cell receptor Vβ–specific antibodies. Eur. J. Immunol. 21:2899–2905 80. Moder KG, Luthra HS, Griffiths M, David CS. 1993. Prevention of collagen– induced arthritis in mice by depletion of T cell receptor Vβ8 bearing T cells with monoclonal antibodies. Br. J. Rheumat. 32:26–30 81. Kumar V, Sercarz E. 1993. The involvement of T cell receptor peptide–specific regulatory CD4+ T cells in recovery from antigen–induced autoimmune disease. J. Exp. Med. 178:909–916 82. Kumar V, Stellrecht K, Sercarz E. 1996. Inactivation of CD4 regulatory T cells results in chronic autoimmunity. J. Exp. Med. 184:1609–1617 83. Kumar V, Aziz F, Sercarz E, Miller A. 1997. Regulatory T cells specific for the

P1: SAT/spd

P2: KKK/plb

January 21, 1999

462

84.

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL same framework 3 region of the Vβ8.2 chain are involved in the control of collagen II–induced arthritis and experimental autoimmune encephalomyelitis. J. Exp. Med. 185:1725–1733 Singh B, Prange S, Jevnikar AM. 1998. Protective and destructive effects of microbial infection in insulin–dependent diabetes mellitus. Sem. Immunol. 10:79– 86 Conrad B, Weldmann E, Trucco G, Rudert WA, Behboo R, Ricordi C, RodriquezRilo H, Finegold D, Trucco M. 1994. Evidence for superantigen involvement in insulin–dependent diabetes mellitus aetiology. Nature 371:351–355 Luppi P, Trucco M. 1996. Superantigens in insulin–dependent diabetes mellitus. Springer Semin. Immunopathol. 17:333– 362 Abe J, Kotzin BL, Jujo K, Melish ME, Glode MP, Kohsaka T, Leung DY. 1992. Selective expansion of T cells expressing T-cell receptor variable region Vβ2 and Vβ8 in Kawasaki disease. Proc. Natl. Acad. Sci. USA 89:4066–4070 Abe J, Kotzin BL, Meissner C, Melish ME, Takahashi M, Fulton D, Romagne F, Malissen B, Leung DYM. 1993. Characterization of T cell receptor changes in acute Kawasaki disease. J. Exp. Med. 177:791–796 Curtis N, Zheng R, Lamb JR, Levin M. 1995. Evidence for a superantigen– mediated process in Kawasaki disease. Arch. Dis. Child. 72:308–311 Leung DY, Giorno RC, Kazemi LV, Flynn PA, Busse JB. 1995. Evidence for superantigen involvement in cardiovascular injury due to Kawasaki syndrome. J. Immunol. 155:5018–5021 Leung DY, Meissner HC, Fulton DR, Murray DL, Kotzin BL, Schlievert PM. 1993. Toxic shock syndrome toxin– secreting Staphylococcus aureus in Kawasaki syndrome. Lancet 342:1385– 1388 Pietra BA, De Inocencio J, Giannini EH, Hirsch R. 1994. TCR V beta family repertoire and T cell activation markers in Kawasaki disease. J. Immunol. 153:1881– 1888 Sakaguchi M, Kato H, Nishiyori A, Sagawa K, Itoh K. 1995. Characterization of CD4+ T helper cells in patients with Kawasaki disease (KD): preferential production of tumor necrosis factor– alpha (TNF–alpha) by Vbeta2 or Vbeta8 CD4+ T helper cells. Clin. Exp. Immunol. 99:276–282 Jason J, Montana E, Donald JF, Seidman

95.

96.

97.

98.

99.

100.

101.

102.

103.

104.

M, Inge KL, Campbell R. 1998. Kawasaki disease and the T-cell antigen receptor. Hum. Immunol. 59:29–38 Iandolo JJ, Chapes SK. 1997. The exfoliative toxins of Staphylococcus aureus. In Superantigens: Molecular Biology, Immunology and Relevance to Human Disease, ed. DYM Leung, BT Huber, PM Schlievert, Marcel Dekker Inc. NY, pp. 231–255 Prevost G, Rifai S, Chaix ML, Piemont Y. 1991. Functional evidence that the Ser–195 residue of staphylococcal exfoliative toxin A is essential for biological activity. Infect. Immun. 59:3337–3339 Henderson CA, Highet AS. 1988. Acute psoriasis associated with Lancefield group C and group G cutaneous streptococcal infections. Br. J. Dermatol. 118: 559–562 Lewis HM, Baker BS, Bokth S, Powles AV, Garioch JJ, Valdimarsson H, Fry L. 1993. Restricted T cell receptor Vβ gene usage in the skin of patients with guttate and chronic plaque psoriasis. Br. J. Dermatol. 129:514–520 Chang JCC, Smith LR, Froning KJ, Schwabe BJ, Laxer JA, Caralli LL, Kurland HH, Karasek MA, Wilkinson DI, Carlo DJ, Brostoff SW. 1994. CD8+ T cells in psoriatic lesions preferentially use T-cell receptor Vbeta 3 and/or Vbeta 13.1 genes. Proc. Natl. Acad. Sci USA 91:9282–9286 Leung DYM, Travers JB, Giorno R, Norris DA, Skinner R, Aelion J, Kazemi LV, Kim MH, Trumble AE, Kotb M, Schlievert PM. 1995. Evidence for streptococcal superantigen–driven process in acute guttate psoriasis. J. Clin. Invest. 96:2106– 2112 Boehncke WH, Dressel D, Manfras B, Zollner TM, Wettstein A, Bohm BO, Sterry W. 1995. T-cell repertoire in chronic plaque–stage psoriasis is restricted and lacks enrichment of superantigen–associated Vβ regions. J. Invest. Dermatol. 104:725–728 Sayama K, Midorikawa K, Hanakawa Y, Sugai M, Hashimoto K. 1998. Superantigen production by Staphylococcus aureus in psoriasis. Dermatology 196:194– 198 Leung DY. 1995. Atopic dermatitis: the skin as a window into the pathogenesis of chronic allergic diseases. J. Allergy Clin. Immunol. 96:302–318 Leyden JJ, Marples RR, Klingman AM. 1974. Staphylococcus aureus in the lesions of atopic dermatitis. Br. J. Dermatol. 90:525–530

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL ACTIVATION BY SUPERANTIGENS 105. Leung DY, Harbeck R, Bina P, Reiser RF, Yang E, Norris DA, Hanifin JM, Sampson HA. 1993. Presence of IgE antibodies to staphylococcal exotoxins on the skin of patients with atopic dermatitis: evidence for a new group of allergens. J. Clin. Invest. 92:1374–1380 106. Gascoigne NRJ, Ames KT. 1991. Direct binding of secreted T-cell receptor β chain to superantigen associated with class II major histocompatibility complex protein. Proc. Natl. Acad. Sci. USA 88:613– 616 107. Malmqvist M. 1993. Surface plasmon resonance for detection and measurement of antigen–antibody affinity and kinetics. Curr. Biol. 5:282–286 108. Margulies DH, Plaskin D, Khilko SN, Jelonek MT. 1996. Studying interactions involving the T-cell antigen receptor by surface plasmon resonance. Curr. Opin. Immunol. 8:262–270 109. Seth A, Stern LS, Ottenhoff THM, Engel I, Owen MJ, Lamb JR, Klausner RD, Wiley DC. 1994. Binary and ternary complexes between T-cell receptor, class II MHC and superantigen in vitro. Nature 369:324–327 110. Malchiodi EL, Eisenstein E, Fields BA, Ohlendorf DH, Schlievert PM, Karjalainen K, Mariuzza RA. 1995. Superantigen binding to a T cell receptor β chain of known three–dimensional structure. J. Exp. Med. 182:1833–1845 111. Leder L, Llera A, Lavoie PM, Lebedeva MI, Li H, Sekaly R-P, Bohach GA, Gahr PJ, Schlievert PM, Karjalainen K, Mariuzza RA. 1998. A mutational analysis of the binding of staphylococcal enterotoxins B and C3 to the T cell receptor β chain and major histocompatibility complex class II. J. Exp. Med. 187:823–833 112. Khandekar SS, Brauer PP, Naylor JW, Chang H–C, Kern P, Newcomb JR, LeClair KP, Stump HS, Bettencourt BM, Kawasaki E, Banerji J, Profy AT, Jones B. 1997. Affinity and kinetics of the interactions between an αβ T-cell receptor and its superantigen and class II–MHC/peptide ligands. Mol. Immunol. 34:493–503 113. van der Merwe PA, Brown MH, Davis SJ, Barclay AN. 1993. Affinity and kinetics of the interaction of the cell adhesion molecules rat CD2 and CD48. EMBO J. 12:4945–4954 114. van der Merwe PA, Barclay AN, Mason DW, Davis EA, Morgan BP, Tone M, Krishnam AKC, Ianelli C, Davis SJ. 1994. Human cell–adhesion molecule CD2 binds CD58 (LFA–3) with a very low affinity and an extremely fast disso-

115.

116.

117.

118.

119.

120.

121. 122.

123.

124.

125.

126.

463

ciation rate but does not bind CD48 or CD59. Biochemistry 33:10149–10160 Corr M, Slanetz AE, Boyd LF, Jelonek MT, Khilko S, Al-Ramadi BK, Kim YS, Maher SE, Bothwell AL, Margulies DH. 1994. T cell receptor–MHC class I peptide interactions: affinity, kinetics, and specificity. Science 265:946–948 Matsui K, Boniface JJ, Steffner P, Reay PA, Davis MM. 1994. Kinetics of T cell receptor binding to peptide/I–Ek complexes: correlation of the dissociation rate with T-cell responsiveness. Proc. Natl. Acad. Sci. USA 91:12862–12866 Alam SM, Travers PJ, Wung JL, Nasholds W, Redpath S, Jameson SC, Gascoigne NRJ. 1996. T-cell receptor affinity and thymocyte positive selection. Nature 381:616–620 Valitutti S, Muller S, Cella M, Padovan E, Lanzavecchia A. 1995. Serial triggering of many T-cell receptors by a few peptide– MHC complexes. Nature 375:148– 151 Viola A, Lanzavecchia A. 1996. T cell activation determined by T cell receptor number and tunable thresholds. Science 273:104–106 Rabinowitz JD, Beeson C, Lyons DS, Davis MM, McConnell HM. 1996. Kinetic discrimination in T-cell activation. Proc. Natl. Acad. Sci. USA 93:1401–1405 Proft T, Fraser J. 1998. Superantigens: just like peptides only different. J. Exp. Med. 187:819–821 Jardetzky TS, Brown JH, Gorga JC, Urban RG, Chi YI, Stauffacher C, Strominger JL, Wiley DC. 1994. Three– dimensional structure of a human class II histocompatibility molecule complexed with superantigen. Nature 368:711–718 Dessen A, Lawrence CM, Cupo S, Zaller DM, Wiley DC. 1997. X–ray crystal structure of HLA–DR4 (DRA∗ 0101, DRB1∗ 0401) complexed with a peptide from human collagen II. Immunity 7:473– 481 Kim J, Urban RG, Strominger JL, Wiley DC. 1994. Toxic shock syndrome toxin–1 complexed with a class II major histocompatibility molecule HLA–DR1. Science 266:1870–1874 Mollick JA, Chintagumpala M, Cook RG, Rich RR. 1991. Staphylococcal exotoxin activation of T-cells. Role of exotoxin– MHC class II binding affinity and class II isotype. J. Immunol. 146:463–468 Wen R, Cole GA, Surman S, Blackman MA, Woodland DL. 1996. Major histocompatibility complex class II–associated peptides control the presentation of

P1: SAT/spd

P2: KKK/plb

January 21, 1999

464

127.

128.

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

129.

130.

131.

132.

133.

134.

135.

136.

137.

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL bacterial superantigens to T cells. J. Exp. Med. 183:1083–1092 Kozono H, Parker D, White J, Marrack P, Kappler J. 1995. Multiple binding sites for bacterial superantigens on soluble class II molecules. Immunity 3:187–196 Hudson KR, Tiedemann RE, Urban RG, Lowe SC, Strominger JL, Fraser JD. 1995. Staphylococcal enterotoxin A has two cooperative binding sites on major histocompatibility complex class II. J. Exp. Med. 182:711–720 Abrahmsen L, Dohlsten M, Segren S, Bjork P, Jonsson E, Kalland T. 1995. Characterization of two distinct MHC class II binding sites in the superantigen staphylococcal enterotoxin A. EMBO J. 14:2978–2986 Tiedemann RE, Urban RJ, Strominger JL, Fraser JD. 1995. Isolation of HLA– DR1.(staphylococcal enterotoxin A)2 trimers in solution. Proc. Natl. Acad. Sci. USA 92:12156–12159 Mehindate K, Thibodeau J, Dohlsten M, Kalland T, Sekaly R–P, Mourad W. 1995. Cross–linking of major histocompatibility complex class II molecules by staphylococcal enterotoxin A is a requirement for inflammatory cytokine gene expression. J. Exp. Med. 182:1573–1577 Tiedemann RE, Fraser JD. 1996. Cross– linking of MHC class II molecules by staphylococcal enterotoxin A is essential for antigen–presenting cell and T cell activation. J. Immunol. 157:3958–3966 Winslow GM, Marrack P, Kappler JW. 1994. Processing and major histocompatibility complex binding of the MTV7 superantigen. Immunity 1:23–33 Mottershead DG, Hsu P–N, Urban RG, Strominger JL, Huber BT. 1995. Direct binding of the Mtv7 superantigen (Mls–1) to soluble MHC class II molecules. Immunity 2:149–154 Held W, Waanders GA, MacDonald HR, Acha–Orbea H. 1994. MHC class II hierarchy of superantigen presentation predicts efficiency of infection with mouse mammary tumor virus. Int. Immunol. 6:1403–1407 Pucillo C, Cepeda R, Hodes RJ. 1993. Expression of a MHC class II transgene determines superantigenicity and susceptibility to mouse mammary tumor virus infection. J. Exp. Med. 178:1441–1445 Thibodeau J, Labrecque N, Denis F, Huber BT, Sekaly R–P. 1994. Binding sites for bacterial and endogenous retroviral superantigens can be dissociated on major histocompatibility complex class II molecules. J. Exp. Med. 179:1029–1034

138. Torres BA, Griggs ND, Johnson HM. 1993. Bacterial and retroviral superantigens share a common binding region on class II MHC antigens. Nature 364:152– 154 139. McMahon CW, Bogatzki LY, Pullen AM. 1997. Mouse mammary tumor virus superantigens require N–linked glycosylation for effective presentation to T cells. Virology 228:161–170 140. Labrecque N, Thibodeau J, Mourad W, Sekaly R–P. 1994. T cell receptor–major histocombatibility complex class II interaction is required for the T cell response to bacterial superantigens. J. Exp. Med. 180:1921–1929 141. Deckhut AM, Chien Y, Blackman MA, Woodland DL. 1994. Evidence for a functional interaction between the β chain of major histocompatibility complex class II and the T cell receptor α chain during recognition of a bacterial superantigen. J. Exp. Med. 180:1931–1935 142. Blackman MA, Woodland DL. 1996. Role of the T cell receptor α–chain in superantigen recognition. Immunol. Res. 15:98–113 143. Padlan EA. 1994. Anatomy of the antibody molecule. Mol. Immunol. 31:169– 217 144. Fremont DH, Matsumura M, Stura EA, Peterson PA, Wilson IA. 1992. Crystal structures of two viral peptides in complex with MHC class I H–2Kb. Science 257:919–927 145. Madden D. 1995. The three–dimensional structure of peptide–MHC complexes. Annu. Rev. Immunol. 13:587–622 146. Bhat TN, Bentley GA, Boulot G, Greene MI, Tello D, Dall’Acqua W, Souchon H, Schwarz FP, Mariuzza RA, Poljak RJ. 1994. Bound water molecules and conformational stabilization help mediate an antigen–antibody association. Proc. Natl. Acad. Sci. USA 91:1089–1093 147. Fields BA, Goldbaum FA, Ysern X, Poljak RJ, Mariuzza RA. 1995. Molecular basis of antigen mimicry by an anti– idiotope. Nature 374:739–742 148. Dall’Acqua W, Goldman ER, Lin W, Teng C, Tsuchiya D, Li H, Ysern X, Braden BC, Li Y, Smith-Gill SJ, Mariuzza RA. 1998. A mutational analysis of binding interactions in an antigen–antibody protein– protein complex. Biochemistry 37:7981– 7991 149. Hodtsev AS, Choi Y, Spanopoulou E, Posnett DN. 1998. Mycoplasma superantigen is a CDR3–dependent ligand for the T cell antigen receptor. J. Exp. Med. 187:319– 327

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

T CELL ACTIVATION BY SUPERANTIGENS 150. Ciurli C, Posnett DN, Sekaly R–P, Denis F. 1998. Highly biased CDR3 usage in restricted sets of β chain variable regions during viral superantigen 9 response. J. Exp. Med. 187:253–258 151. Hong S–C, Waterbury G, Janeway CA Jr. 1996. Different superantigens interact with distinct sites in the Vβ domain of a single T cell receptor. J. Exp. Med. 183:1437–1446 152. Davies DR, Padlan EE. 1992. Twisting into shape. Curr. Biol. 2:254–256 153. Tulip WR, Varghese JN, Laver WG, Webster RG, Coleman PM. 1992. Refined crystal structure of the influenza virus N9 neuraminidase–NC41 Fab complex. J. Mol. Biol. 227:122–148 154. Wilson IA, Stanfield RL. 1993. Antigen– antibody interactions. Curr. Opin. Struct. Biol. 3:113–118 155. Lawrence MC, Coleman PM. 1993. Shape complementarity at protein/protein interfaces. J. Mol. Biol. 234:946–950 156. Ysern X, Li H, Mariuzza RA. 1998. Imperfect interfaces. Nature Struct. Biol. 5:412–414 157. Reich Z, Boniface JJ, Lyons DS, Borochov N, Wachtel EJ, Davis MM. 1997. Ligand–specific oligomerization of Tcell receptor molecules. Nature 387:617– 620 158. Imanishi KH, Igarashi H, Uchiyama T. 1990. Activation of murine T cells by streptococcal pyrogenic exotoxin A. J. Immunol. 145:3170–3176 159. Woodland DL, Blackman MA. 1993. How do T cell receptors, MHC molecules and superantigens get together? Immunol. Today 14:208–212 160. Borrero H, Donson D, Cervera C, Rexer C, Macphail S. 1995. T cell receptor Vα4 is expressed by a subpopulation of Vβ6 T cells that respond to the bacterial superantigen staphylococcal enterotoxin B. J. Immunol. 154:4247–4260 161. Donson D, Borrero H, Rutman M, Pergolizzi R, Malhado N, Macphail S. 1997. Gene transfer directly demonstrates a role for TCR Vα elements in superantigen recognition. J. Immunol. 158:5229–5236 162. Daly K, Nguyen P, Hankley D, Zhang WJ, Woodland DL, Blackman MA. 1995. Contribution of the TCR α–chain to the differential recognition of bacterial and retroviral superantigens. J. Immunol. 155:27–34 163. Abrahms´en L. 1995. Superantigen engineering. Curr. Opin. Struct. Biol. 5:464– 470 ˚ 164. Dohlsten M, Hedlund G, Akerblom E, Lando PA, Kalland T. 1991. Monoclonal

165.

166.

167.

168.

169.

170.

171.

172.

173.

465

antibody–targeted superantigens: A different class of anti–tumor agents. Proc. Natl. Acad. Sci. USA 88:9287–9291 Dohlsten M, Abrahms´en L, Bj¨ork P, Lando PA, Hedlund G, Forsberg G, Brodin T, Gascoigne NRJ, F¨orberg C, Lind P, Kalland T. 1994. Monoclonal antibody–superantigen fusion proteins: Tumor–specific agents for T-cell–based tumor therapy. Proc. Natl. Acad. Sci. USA 91:8945–8949 Ihle J, Holzer U, Krull F, Dohlsten M, Kalland T, Niethammer D, Dannecker GE. 1995. Antibody–targeted superantigens induce lysis of major histocompatibility complex class II–negative T-cell leukemia lines. Cancer Res. 55:623–628 Holzer U, Bethge W, Krull F, Ihle J, Handgretinger R, Reisfeld RA, Dohlsten M, Kalland T, Niethammer D, Dannecker GE. 1995. Superantigen–staphylococcal– enterotoxin–A–dependent and antibody– targeted lysis of GD2–positive neuroblastoma cells. Cancer Immunol. Immunother. 41:129–136 Hansson J, Ohlsson L, Persson R, Andersson G, Ilb¨ack N–G, Litton MJ, Kalland T, Dohlsten M. 1997. Genetically engineered superantigens as tolerable antitumor agents. Proc. Natl. Acad. Sci. USA 94:2489–2494 Giantonio BJ, Alpaugh RK, Schultz J, McAleer C, Newton DW, Shannon B, Guedez Y, Kotb M, Vitek L, Persson R, Gunnarsson PO, Kalland T, Dohlsten M, Persson B, Weiner LM. 1997. Superantigen–based immunotherapy: a phase I trial of PNU–214565, a monoclonal antibody–staphylococcal enterotoxin A recombinant fusion protein, in advanced pancreatic and colorectal cancer. J. Clin. Oncol. 15:1994–2007 Kappler JW, Herman A, Clements J, Marrack P. 1992. Mutations defining functional regions of the superantigen staphylococcal enterotoxin B. J. Exp. Med. 175:387–396 Bavari S, Dyas B, Ulrich RG. 1996. Superantigen vaccines: a comparative study of genetically attenuated receptor– binding mutants of staphylococcal enterotoxin A. J. Infect. Dis. 174:338–345 Tseng J, Komisar JL, Trout RN, Hunt RE, Chen JY, Johnson AJ, Pitt L, Ruble DL. 1995. Humoral immunity to aerosolized staphylococcal enterotoxin B (SEB), a superantigen, in monkeys vaccinated with SEB toxoid–containing microspheres. Infect. Immun. 63:2880–2885 Lowell GH, Colleton C, Frost D, Kaminski RW, Hughes M, Hatch J,

P1: SAT/spd

P2: KKK/plb

January 21, 1999

466

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

174.

175.

176.

177.

178.

15:3

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-15

LI ET AL Hooper C, Estep J, Pitt L, Topper M, Hunt RE, Baker W, Baze WB. 1996. Immunogenicity and efficacy against lethal aerosol staphylococcal enterotoxin B challenge in monkeys by intramuscular and respiratory delivery of proteosome– toxoid vaccines. Infect. Immun. 64:4686– 4693 McCormack JE, Callahan JE, Kappler J, Marrack PC. 1993. Profound deletion of mature T cells in vivo by chronic exposure to exogenous superantigen. J. Immunol. 150:3785–3792 Miethke T, Wahl C, Heeg K, Wagner H. 1993. Acquired resistance to superantigen–induced T cell shock. Vβ selective T cell unresponsiveness unfolds directly from a transient state of hyperreactivity. J. Immunol. 150:3776–3784 Kim C, Siminovitch KA, Ochi A. 1991. Reduction of lupus nephritis in MRL/lpr mice by a bacterial superantigen treatment. J. Exp. Med. 174:1431–1437 Rott O, Wekerle H, Fleischer B. 1992. Protection from experimental allergic encephalomyelitis by application of a bacterial superantigen. Int. Immunol. 4:347– 353 Soos JM, Schiffenbauer J, Johnson HM. 1993. Treatment of PL/J mice with superantigen staphylococcal enterotoxin B, prevents development of experimental

179.

180.

181.

182.

183.

allergic encephalomyelitis. J. Neuroimmunol. 43:39–43 Racke MK, Quigley L, Cannella B, Raine C, McFarlin DE, Scott DE. 1994. Superantigen modulation of experimental allergic encephalomyelitis: activation or anergy determines outcome. J. Immunol. 152:2051–2059 Leung DYM, Schlievert PM. 1997. Superantigens in human disease. In Superantigens: Molecular Biology, Immunology and Relevance to Human Disease, ed. DYM Leung, BT Huber, PM Schlievert, Marcel Dekker Inc. NY, pp. 581–601 Hurley JM, Shimonkevitz R, Hanagan A, Enney K, Boen E, Malmstron S, Kotzin BL, Matsumura M. 1995. Identification of class II major histocompatibility complex and T cell receptor binding sites in the superantigen toxic shock syndrome toxin–1. J. Exp. Med. 181:2229–2235 Murray DL, Prasad GS, Earhart CA, Leonard BAB, Kreiswirth BN, Novick RP, Ohlendorf DH, Schlievert PM. 1994. Immunobiologic and biochemical properties of mutants of toxic shock syndrome toxin–1. J. Immunol. 152:87–95 Deresiewicz RL, Wood J, Chan M, Finberg RW, Kasper DL. 1994. Mutations affecting the activity of toxic shock syndrome toxin–1. Biochemistry 33:12844– 12851

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:435-466. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:467–522

THE DYNAMICS OF T CELL RECEPTOR SIGNALING: Complex Orchestration and the Key Roles of Tempo and Cooperation1 ˇ Ronald N. Germain and Irena Stefanov´ a Lymphocyte Biology Section, Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland 20892; e-mail: ronald [email protected], [email protected] KEY WORDS:

kinetic proofreading, antagonists, feedback regulation

ABSTRACT T cells constantly sample their environment using receptors (TCR) that possess both a germline-encoded low affinity for major histocompatibility complex (MHC) molecules and a highly diverse set of CDR3 regions contributing to a range of affinities for specific peptides bound to these MHC molecules. The decision of a T cell “to sense and to respond” with proliferation and effector activity rather than “to sense, live on, but not respond” is dependent on TCR interaction with a low number of specific foreign peptide:MHC molecule complexes recognized simultaneously with abundant self peptide-containing complexes. Interaction with self-complexes alone, on the other hand, generates a signal for survival without a full activation response. Current models for how this distinction is achieved are largely based on translating differences in receptor affinity for foreign versus self ligands into intracellular signals that differ in quality, intensity, and/or duration. A variety of rate-dependent mechanisms involving assembly of molecular oligomers and enzymatic modification of proteins underlie this differential signaling. Recent advances have been made in measuring TCR:ligand interactions, in understanding the biochemical origin of distinct proximal and distal signaling events resulting from TCR binding to various ligands, and in appreciating the role of feedback pathways. This new information can be synthesized into a model of 1 The US government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

467

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

468

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

how self and foreign ligand recognition each evoke the proper responses from T cells, how these two classes of signaling events interact, and how pathologic responses may arise as a result of the underlying properties of the system. The principles of signal spreading and stochastic resonance incorporated into this model reveal a striking similarity in mechanisms of decision-making among T cells, neurons, and bacteria.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PROLOGUE A lover of classical music, I have purchased tickets to a rehearsal session of a major symphony orchestra. Entering the music hall, I see the players already seated and quickly pick out the various sections and then the individual performers. Soon, I am aware of each component of the orchestra, their seating arrangement with respect to each other and to the hall. I know from past experience the shape, construction, and timber of each instrument. Yet I have no idea of the music I will hear, lacking knowledge of either the score chosen or the interpretation it will be given. The conductor enters, takes the podium, raps for the musicians’ attention, and signals by a downbeat for the playing to begin. The conductor is trying a new interpretation of the composition, the score of which is before him. He signals for the strings to begin the opening theme, then for the woodwinds to join in, but waits too long to bring in the brass. Cacophony rather than melodious intersection ensues, and the playing stops before the full power of the composition develops. The conductor indicates to the orchestra to begin again at the first measure—the strings start but this time the indicated tempo is too fast, the woodwinds enter late, and again the music comes to a halt. A third try, and this time the pacing is correct, the conductor able to bring in each player and instrumental section at the proper time and in harmonius cooperation to create a melodius whole. The beauty of the piece is heard in full over its entire length, reaching a peak and then finally ceasing at the proper moment.

INTRODUCTION The biochemistry of signaling by integral membrane receptors is an area of intense investigation. An increasingly sophisticated array of tools has permitted the identification of new pathways and of new components of known signaling pathways. Yet when one looks at this large body of data, it is easy to feel like the music lover described above. One can recognize the instruments, even see the organization of the orchestra into sections, and know the identity of the conductor (the receptor whose interaction with ligand will start the signaling cascade) but not be able to anticipate the actual symphony of events that will connect the known players or the ultimate biologic outcome of the initial interaction.

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

469

In particular, such knowledge fails to integrate the crucial role of tempo in the outcome of receptor-ligand association, with the kinetics of events too often studied in isolation. The dramatic effects of changing not the pace of events within one part of the system, but rather the kinetic relationship of activity in the distinct subsystems to one another, are given insufficient attention. The primary aim of this review is to lay out the evidence for a key role of dynamic processes in regulating T cell receptor (TCR) signaling events and the downstream outcomes of these events. We first discuss why the characteristics of TCR structure and ligand diversity, and of the interaction between these receptors and ligands, together demand that the signaling system behave in particular ways. We then describe what is known about the relevant details of ligand-receptor interaction, the resultant signal generation process, and the downstream effects of these signaling events. Finally, we make the argument that additional kinetic complexity and cross-talk between and among various receptors are likely to be central to the proper functioning of the T cell adaptive immune response. A major goal of this review is to increase recognition of the importance of relative rates and subsystem cooperativity in immune activation. By doing so, we hope that the vast amount of data being generated on individual proteins or protein interactions can be placed in a biologically relevant context with respect to the physiologic behavior of the cells in which these biochemical reactions are taking place.

TCRs AND THEIR LIGANDS—FACTS AND IMPLICATIONS The T cells of higher vertebrates have two well-defined clonally distributed receptor structures (TCR) termed αβ and γ δ (1). Because there is limited information available on the ligands of γ δ TCR (2–4) or concerning the signaling events in γ δ T cells, we focus in this review on αβ TCR-bearing cells and the consequences of αβ TCR binding to the known partners of this receptor. These ligands are comprised of peptides bound to class I and class II major histocompatibility complex (MHC) molecules (5) and glycolipids bound to the MHC-like molecule CD1 (6–8). Although many of the principles described here are likely to apply to both categories, little is known about TCR interaction with lipid:CD1 complexes or signaling in the responding cells, so again, we do not discuss this subject in any depth.

The Structure of Peptide:MHC Molecule Ligands and the Biochemical Nature of Their Interaction with the TCR MHC class I and class II molecules are integral membrane glycoproteins whose structures have been well studied at the crystallographic level (9). They have

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

470

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

nearly identical protein backbones in the homologous membrane-distal regions (α1 and α2 of class I and α1 and β1 of class II) that make up the single integrated protein domain that serves as the peptide binding segment and target of interaction with the TCR (10). MHC proteins are extremely polymorphic, but the bulk of the variation is confined to regions within the binding groove that affect association with peptide, not interaction with other protein molecules (10, 11). Each allele of class I or class II has a variety of pockets within the peptide binding domain whose shape, chemical composition, and position within the groove are dictated by a combination of conserved and polymorphic residues. The result is a differing set of peptide binding sites that each interacts strongly with a specific array of amino acid side chains possessing a particular spacing (9, 12, 13). As a consequence, each allele of MHC molecule binds best to a certain subset of peptide sequences derived from self and foreign proteins. The preferred/optimal amino acid residues within a peptide and their proper spatial distribution for high-affinity association with a particular MHC molecule are together referred to as the binding motif for that MHC allele (14–17). The majority of peptides binding stably to that MHC allele will have this motif, but other peptides can interact with faster off-rates (18). Which peptides actually wind up bound to expressed MHC molecules depends on the proteins present in the cytosol or endosomes of a cell, the activity of various proteases, and the function of several accessory molecules involved in regulation of MHC molecule binding site availability, MHC molecule localization, and peptide transport (19). Although there are some exceptions (e.g. see 20), in general MHC molecules cannot discriminate between peptides derived from self-proteins and those derived from pathogen proteins, and hence, even on infected cells, the majority of MHC molecules will be occupied with peptides derived from the much more abundant self-protein pool and whose sequences match closely the binding motifs of the available MHC molecules (21–24). Most nucleated cells express between 104–106 surface MHC class I molecules, and those cells that express class II molecules either constitutively or as a result of cytokine induction display similar total numbers. Analysis by mass spectroscopy of peptides eluted from immunoisolated MHC molecules indicates that the expressed product of a single MHC allele contains as many as 1000–2000 different (self-)peptides represented at greater than 10 copies per cell, with some species being present in several hundred or more copies (23, 24). During active processing of acquired pathogen proteins, the estimated number of copies accumulated on the surface membrane of an individual cell ranges from as few as 102 to more than 104 (25–28). At least for memory or effector T cells, only 10–100 complexes per cell are necessary to evoke effector responses (29–33). Thus, it is typical for a T cell to respond with activation to presentation of specific ligand at between 0.01% and 0.1% of the

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

471

total peptide:MHC molecule display on the presenting cell surface. The other 99.9–99.99% of the same allelic product is occupied with self-peptides. The detailed structures of many peptide:MHC class I and several peptide: MHC class II molecules have been solved at atomic resolution (for reviews see 9, 13; see also 34–36). In almost all cases, only a few side chains of the peptide are not buried in pockets mediating association with the MHC molecule and these protrude into solvent above the surrounding MHC binding region. No other features of a foreign ligand besides these few protruding side chains distinguish it from the numerically much more prevalent self-complexes, except perhaps in some cases for a (subtle) change in the shape of the MHC molecule (37). The location of the exposed parts of the bound peptide makes it difficult to imagine recognition by the immunoglobulin-sized TCR without some interaction of the receptor with the surrounding MHC molecule (1). In agreement with this expectation, the recently solved structures of several TCR:peptide:MHC molecule cocrystals reveal the TCR to make many bonds to the surrounding MHC molecule as well as to the exposed side chains of the peptide. The ligand for αβ TCR can thus be considered a single surface, comprised of the helices that constitute the upper portion of the MHC binding groove together with the exposed portion of the peptide within (38–43). The affinity of interaction of this ligand with the TCR thus arises from the sum of binding interactions to the several parts of this conjoint ligand surface. The structure of the MHC helical regions is quite conserved and, to the best of current knowledge, essentially the same among various peptide-occupied versions of the same allelic MHC product. As a consequence, only those specific bonds to the small number of exposed side chains that vary between self and foreign peptides can give rise to differences in either affinity or allosteric effects that permit distinctions between self-complexes and foreign complexes in the activation of mature T cells through the TCR.

Relating TCR Structure to Specificity αβ TCRs are generated by somatic rearrangement of germline V, D, and J gene segments, the nontemplated insertion and/or deletion of junctional residues, and combinatorial association of completed protein chains (1). The outcome of these several events is a primary receptor repertoire containing a highly diverse set of combining sites distributed clonally among immature CD4+CD8+ thymocytes (44). Certain portions of the combining site (germline V, D, and J regions) can be selected over evolutionary time, whereas others (CDR3 residues affected by N-nucleotide addition or nuclease trimming) cannot. αβ T cells are focused on the recognition of pathogens or their products through MHC molecule peptide presentation, and the exposed regions of the peptide binding domain of MHC molecules have a highly conserved structure. It therefore makes sense that the germline elements of the TCR would coevolve with the MHC molecules

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

472

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

to provide a surface that promotes effective interactions in support of more precise binding events dictated by the specific peptide and the nongermline features of the receptor (1, 45, 46). This prediction has also been borne out by recent crystallographic analysis (38–43). The CDR1 and CDR2 regions of germline encoded Vβ and Vα segments of the TCR make contact in a largely stereotypical manner with the exposed helical segments of the class I or class II peptide binding domain. The CDR3 region of the TCR, formed by contributions of the D, J, and N regions of the rearranged receptor, make the primary contacts with the exposed (epitopic) residues of the bound peptide within the MHC molecule. Additional evidence for a generic fit between TCR germline segments and MHC molecules comes from studies of both individual TCR isolated from the developing T cells of MHC-deficient mice (47), so that no selection by MHC molecules can bias the TCR analyzed, and from studies of the responses of entire populations of newly generated immature thymocytes possessing unselected repertoires to variations in MHC display by thymic stromal cells (48; MS Vacchio, J Lee, JD Ashwell, submitted for publication). In the first case, a substantial fraction of the individual TCR isolated in the form of immortalized hybridoma cells show functional reactivity to a random panel of MHC-bearing presenting cells. In the second case, measurement of CD5 and CD69 levels on immature thymocytes shows a homogenous increase in these inducible gene products in concert with rising total MHC molecule expression on the thymic stroma, independent of MHC allele or class. Together, these data confirm early predictions of a strong bias of germline TCR segments toward broad reactivity with MHC molecules (1, 45, 49), and they further show that the interaction is not merely permissive but of sufficient strength to lead to intracellular signals. At the same time, the level of CD5 and CD69 up-regulation induced among the total pool of TCR-bearing thymocytes is less than that associated with those few cells progressing through thymocyte-positive selection and maturation. Thus, the strength of the conserved interaction of all germline αβ segments with MHC molecules is such that even exposure of the TCR to 104-fold more total MHC molecules than the number involved in specific foreign antigen activation of mature T cells does not lead to signaling exceeding the selection threshold. The corollary of this observation is that the additional contribution of the CDR3 region that distinguishes the TCR on individual cells must make a key contribution to TCR-ligand interaction leading to effective signaling for biological responses. Because it does not appear that the molecular mechanisms that generate diversity in the CDR3 region can produce TCR with an innate capacity to bind better or in an efficacious manner only foreign peptide:MHC ligands, there must be one or more mechanisms that adapt the primary repertoire to the ambient self-ligand pool.

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

DYNAMICS OF TCR SIGNALING

473

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Problems in Ligand Discrimination by TCR The preceding two sections outlined the enormous complexity of ligands on antigen presenting cell membranes, the vastly greater abundance of self-peptide– containing MHC molecules over foreign peptide-containing molecules when the latter are present at all, the lack of linkage among self-protein sequences, self-MHC molecules, and TCR genes, the generic binding of MHC molecules by germline-encoded elements of the TCR, the small number of specific bonds made by CDR3 regions of the TCR to unique features of a particular peptide:MHC molecule combination, and the resulting inability of the primary repertoire in the thymus to distinguish between self and foreign peptide:MHC molecule complexes. Yet despite this, mature T cells can be activated to effector function by very low numbers of foreign ligands in a sea of such self-complexes, while remaining in a quiescent state when exposed to the vastly greater number of self-ligands found on the same presenting cells in the absence of foreign antigen. The issues raised by this distinction in the stimulatory properties of self and foreign ligands are made even more cogent by the importance of self-recognition in thymic positive selection (44) and, even more remarkably, peripheral T cell survival (50–54). Self-ligands clearly bind well enough to the TCR of the selected thymocytes to generate physiologically relevant intracellular signals within immature thymocytes and also mature T cells. Thus the question arises of how the immune system is organized at a cellular and molecular level to give rise to sensitive and, most important, selective T cell activation to effector function in response to foreign ligands, given their near identity to self, their unfavorably skewed relative density on presenting cell surfaces, and this evidence for the potential of the more abundant self-ligands to engage and signal through the positively selected TCR. The situation with TCRs and MHC molecules differs markedly from that of other receptor:ligand systems. Receptors for growth factors, cytokines, hormones, neurotransmitters, chemokines, and other endogenous ligands coevolve with these mediators. Except for rare mutations, each receptor:ligand pair has a defined affinity and a three-dimensional relationship in the bound state. As a consequence, whether signaling by the receptor requires an allosteric change, homodimerization, heterodimerization, or alteration in ion channel behavior, the duration of the binding event involving these partners is predictable, as is any structural change in the receptor or ligand that is required for proper functioning. Such predictability permits a given ligand to trigger only the correct receptor and limits its capacity to directly stimulate or inhibit other receptors present in the same environment. In stark contrast, the specific on- and offrates for interaction of any given TCR with any specific peptide:MHC ligand are dictated by the nonprogrammed fine structure of the CDR3 region in the context of the particular pair of germline V segments that form the receptor

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

474

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

dimer, together with the variation in epitopic peptide side chains and local protein structure of the ligand. Thus, any mechanism for regulation of signaling that involves the affinity of the receptor:ligand pair, for example, must deal with the diversity in this parameter among all TCR produced in the thymus for the extensive set of self as well as for the unknown set of foreign peptide:MHC ligands ultimately present in the host environment. Furthermore, it must do so when the structural basis for any such difference can be as limited as a single amino acid side chain.

Receptor:Ligand Interactions—Generalities About Binding Constants and Pharmacology What are the implications of these properties of the TCR and its universe of ligands for our understanding of how recognition and response of T cells are coupled and controlled? The study of many examples of drug:receptor interactions has shown that variation in the affinity of a molecule for a receptor can dictate whether it has the properties of agonist, partial agonist, antagonist, or null compound (see 55). A definition of these terms is useful. In principle, an agonist evokes all signaling events of which a receptor is capable. Stronger or weaker agonists do this with greater or lesser potency, but all signaling and functional results of ligand:receptor interaction are the same in each case. Partial agonists (a) evoke some but not other signaling consequences, (b) if studied functionally stimulate some but not other responses, (c) change the relative dose-response relationships among several measurable outcomes, or (d ) change the plateau level of one but not another response. Antagonists inhibit responses normally stimulated upon exposure to a known agonist. Lastly, a null compound does not interact with the receptor sufficiently to signal or functionally perturb the cell at all and does not bind strongly enough to competitively interfere with agonist function. How can variations in binding affinity give rise to the apparently nonproportional changes in function that characterize partial agonists or antagonists of a receptor? The most straightforward model postulates that there is a time delay between ligand binding and the molecular events that characterize the full complement of receptor signaling activities (55–61). By binding to the receptor for a time less than this, a partial agonist or antagonist fails to evoke one or more downstream signaling events typically evoked by the full agonist, hence the selective changes in those responses coupled to the different downstream events. Strong versus weak agonists, on the other hand, occupy the receptor long enough to accomplish all signaling events, though the latter bind less avidly and hence require higher concentrations to produce equivalent absolute occupancy and evoke the same response. Antagonists arise due to lower affinity than partial agonists, such that downstream events are further attenuated, but

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

475

occupancy at high concentrations can compete with offered agonist for binding. This explanation of ligand pharmacology does not take account of any feedback events that might modify receptor or cellular behavior, nor of the situations in which tight binding can still fail to be efficacious (allosteric signaling) (62, 63). If one considers carefully the properties of the TCR repertoire and the peptide:MHC molecule displays on cells in the body, it becomes clear that variations in affinity are an intrinsic property of the T cell immune system. The junctional diversity of TCR must result in a substantial range of binding constants among the pool of TCR for individual peptide:MHC molecule ligands on a presenting cell surface. In turn, it seems inescapable that such differences in affinity would segregate particular TCR:ligand pairs into agonist, partial agonist, antagonist, and null ligand sets. One can look at this from either of two perspectives: (a) For any given TCR in the unselected primary repertoire, the diverse ligands on a presenting cell should constitute a mix of these various pharmacologic categories, at differing concentrations, depending on the peptide:MHC molecule combinations present and the particular structure of the TCR in question; (b) for any given peptide:MHC ligand, the diverse set of TCR in the total repertoire created by rearrangement, nucleotide trimming and addition, and chain pairing is likely to include receptors whose interaction with that ligand can be placed in the different pharmacologic categories. Thus, ligand diversity and receptor diversity combine to create a situation such that in individual T cells, as well as among the entire pool of T cells, signaling events and response patterns characteristic of agonists, partial agonists, and antagonists should in principle coexist. Some data in evidence of this concept have been reported (64, 65). Although this analysis suggests it should have been evident to immunologists many years ago that peptide:MHC molecule ligands would exist for any given TCR with the properties of partial agonists or antagonists, this was not at all the case. Serendipitous observations, rather than experiments based on prediction, first revealed the existence of such ligands (66–69), and the field was surprised (and to a great extent skeptical) when the initial papers on this subject appeared (70). However, the past few years have seen a substantial expansion of work in this area, with strong confirmation of the early findings in a wide array of murine and human experimental systems.

TCR:LIGAND INTERACTIONS—AFFINITY, SIGNALING, AND RESPONSE The preceding sections have outlined the essential genetic and structural characteristics of the TCR recognition system, discussed the type of ligand discrimination and sensitivity it must achieve for physiological activity, and introduced the evidence for distinct pharmacologic properties of TCR ligands seen in

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

476

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

functional assays (see Appendix 1). In building a suitable model of how this recognition system is organized to account for these properties, the next step is to review what is known about the kinetics/affinity of TCR:ligand interaction, the relationship of these parameters to ligand pharmacology, and the consequences of these binding events in terms of intracellular signal generation.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Measurements of TCR and of TCR/Coreceptor Binding to Peptide:MHC Molecule Ligands The recognition that certain combinations of peptides and MHC molecules created ligands with the functional properties of partial agonists and antagonists, combined with the understanding from other systems that affinity can determine such alternative pharmacologic properties, led several investigators to undertake direct measurement of TCR affinity for a series of related ligands. Various methods have been used, including direct analysis by surface plasmon resonance of the binding and dissociation of soluble αβ TCR and soluble peptide-associated class I or class II molecules (87–93), competitive inhibition by soluble peptide:MHC molecule complexes of binding to the TCR of monoclonal antibodies whose affinities had been independently measured (94), Scatchard analysis using labeled soluble TCR analogs and surface displayed peptide:MHC molecule complexes (95) or the converse (96, 97), and measurement of covalently cross-linked proteins during association or dissociation reactions using cell-bound TCR and soluble peptide:MHC molecule ligands (98, 99). Although the absolute affinity or association/dissociation constants derived from these various studies differ, the range of values are nearly all indicative of relatively weak binding characterized by moderately fast on- and off-rates (100, 101). Affinity constants tend to cluster between 0.1 and 50 µM, with dissociation t1/2 in the 10- to 50-s range. In comparison to antibodies in general, this strength of binding is three to six orders of magnitude less avid. Even in comparison to the few monoclonal antibodies specific for peptide:MHC complexes (27, 82, 102), TCR tend to be two to three orders of magnitude less strong. There are several limitations to these studies. For those involving soluble αβ TCR, we do not know if removal of the TCR from the membrane environment affects its binding to ligand. Many integral membrane proteins that exist as oligomers have important subunit interactions mediated by the transmembrane regions (103–106). The elimination of these segments in preparing soluble TCR may alter the relative orientation or stability of the remaining domains and affect affinity. Even more important, there is normally significant contact between the αβ TCR and the immunoglobulin-like regions of CD3 components (107–109); the extent to which this influences how the αβ dimer interacts with ligand is unknown. These concerns do not apply to those cases in which soluble ligand is

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

477

used to measure binding to TCR expressed on the T cell plasma membrane, but these latter studies do not have the same level of kinetic resolution as the plasmon resonance measurements made with soluble TCR-ligand pairs. In both of these situations, the measurements are made at room temperature (25◦ C) rather than 37◦ C, so relating the absolute values obtained by these methods to the actual rates of binding and dissociation during physiologic recognition reactions is not possible. Nor do any of the published results take into account that under physiologic circumstances the TCR and the ligand occupy what have been termed two-dimensional volumes, rather than the three-dimensional volumes involved in these measurement assays. There are methods of correcting for this difference that have been applied to other ligand-receptor pairs involving membrane proteins (110), but these have not yet been used with the TCR. Perhaps most important, only a few of these analyses have included an assessment of the effect of CD4 or CD8 coreceptor binding on the apparent affinities or rate constants for TCR-ligand interaction. Given the role of these other MHC-specific, peptide-unspecific binding proteins in T cell development and peripheral activation, understanding how they affect TCR occupancy and the resulting signals is critical. Garcia et al analyzed binding of a soluble form of the peptide:MHC class I–specific 2C TCR to soluble peptide:MHC ligand with and without the addition of soluble CD8αα dimers to the reaction (91). Remarkably, these investigators observed a sevenfold decrease in the dissociation-rate of the TCR:ligand interaction, with a substantial effect that persisted even after removal of the CD8. This was proposed to arise from the ability of CD8 to stabilize TCR:ligand association during a process of structural adaption. This surprising result has not yet been confirmed by independent studies, and some data suggest an artifactual origin to the phenomenon (111). In another series of experiments, soluble ligand binding to membrane-associated TCR was assessed using photoactivated cross-linking of ligand and receptor (98). By employing T cells with and without CD8 coexpression, the authors could estimate the contribution of coreceptor to the ligand on- and off-rates, observing a lengthening of the t1/2 of dissociation from 7 s to 90 s in the presence of CD8. At present, CD4 has not been shown directly to mediate a similar prolongation of TCR:ligand association, although this is predicted by analogy from the CD8 data and from the effect of CD4 in functional experiments (112).

Relating Binding Events to Ligand Pharmacology Despite these limitations, useful information has been gathered for a small number of TCRs on their binding to a series of related ligands with properties ranging from agonist to partial agonist to antagonist to null complexes. In most cases, the trend is for the less potent ligands to have a lower affinity and, in particular, a faster dissociation-rate (88, 92, 100, 113). This agrees with the

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

478

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

general concept that when allosteric effects are not involved, the pharmacologic properties of a ligand are dictated by its strength of binding, more specifically by the duration of its occupancy of the receptor (55, 114). McKeithan was the first to formalize this relationship for the TCR, suggesting that the phenomenon of kinetic proofreading in other biochemical pathways could be used to explain how a cascade of signal transduction reactions could improve discrimination among a related set of ligands binding to a defined receptor (56). The basis of kinetic proofreading involves the use of a second process to “time” the duration of a binding event. Only if association of two molecules persists until the timing event is completed can the binding event in question contribute at all to the next step in a signaling cascade or biochemical process. Any interaction lasting a shorter time makes no contribution to downstream events, providing a sharp all-or-none effect rather than an outcome proportional to the average time of occupancy or the overall number of engagements. The prototype of this system is the coupling of GTP hydrolysis to incorporation of the amino acid carried by an amino acyl tRNA into a growing polypeptide chain (115). Only when codon:anticodon interaction results in association long enough for the GTP hydrolytic event to occur is peptide bond formation permitted; dissociation of the amino acyl tRNA and mRNA before this GTP-based timing event transpires results in no bond formation, providing an absolute barrier to misincorporation of amino acids into a protein as long as the dissociation rates of all codon:anticodon mismatches are faster than the rate of GTP hydrolysis. Thus, inappropriate binding interactions can occur, but they do not lead to proportional errors in protein synthesis. For the TCR, McKeithan proposed that a series of biochemical events, most likely phosphorylation based, would be initiated upon TCR:ligand engagement. If a functional response required a cascade of such events, the signal transmitted through each step would be limited to only those engagement events lasting longer than some defined minimum time (57, 58, 60), and the signal transmitted through several steps would be reduced by these kinetic attenuation factors. Faster dissociating ligands would show much greater attenuation factors that would multiply through the several steps of the cascade, leading to increasingly large differences in downstream signals for gene activation for two ligands that differ only modestly in binding to the TCR. Because of the branching nature of downstream signaling events within a T cell, some signals will be more affected than others, giving rise to different ratios of activated signaling molecules and, hence, to the potential to alter the relative response of two genes, as seen with partial agonists. Other investigators have proposed variations on this kinetic proofreading theme (59, 61, 116), but it remains the basic model for translating differences in TCR:ligand affinity into changes in ligand potency and functional effect.

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

DYNAMICS OF TCR SIGNALING

479

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Intracellular Signaling upon TCR Engagement A prediction of this proofreading model is that if the ordered arrangement of intracellular signaling events occurring upon receptor engagement was known, one would see greater attenuation of more distal than proximal events when using partial agonists versus agonists; that is, the ratio of measured signals would be different for the pharmacologically distinct ligands. This progressive attenuation should not be observed using lower concentrations of the agonist, because for individual TCR signaling in an autonomous fashion, a lower density of ligand does not change the length of individual TCR occupancy and, hence, should not differentially change the downstream events. The overall intensity of all signals should decrease in parallel as ligand density is lowered, with the ratio of signals remaining constant. The complex biochemical events arising from TCR engagement have been reviewed elsewhere (117–119) and cannot be discussed in full detail here. However, a summary of the most relevant aspects of TCR signaling is important for interpretation of experiments seeking to test the basic tenets of this widely accepted model and for understanding how the signal transduction apparatus is organized to allow physiologically relevant discrimination among closely related ligands. The first biochemical changes known to occur upon TCR engagement involve the activation of src family kinases (120, 121), most notably Lck (122) and possibly Fyn (123). This has been proposed to result from the transphosphorylation of kinase molecules either in two coaggregated TCR complexes or in a single TCR associated with an Lck-coupled CD4 or CD8 coreceptor. Very recent data using soluble monovalent peptide:MHC molecule ligands suggest that the latter are the more efficient events (J Delon, C Gr´egoire, B Malissen, S Darche, F Lemaˆıtre, P Kourilsky, J-P Abastrado, A Troutmann, Immunity 9:467–73). The enzymatically active (Lck) kinase begins to add phosphates to tyrosines present in the TCR-associated ζ (124) as well as CD3 ε, γ , and δ chains (125). The relevant tyrosines reside within stretches of amino acids with the motif [YxxI/L (7–8 amino acids) YxxI/L] (126). This pattern has been termed the immunoreceptor tyrosine-based activation motif (ITAM), and it is found not only in TCR-associated chains but also in signaling proteins associated with FcR, B cell, and natural killer–activating receptors (127). CD3 δ, γ , and ε each contain a single ITAM with a pair of tyrosines, whereas ζ has three ITAM motifs in each chain, for a total of six in the disulfide-linked form of ζ typically found associated with the TCR. The relative rate at which activated Lck phosphorylates the ITAMs in these different TCR-associated proteins is not known. Using antisera specific for each of the six phosphotyrosine-containing regions of a single mouse ζ chain, Kersh et al provided some evidence for a nonrandom order of phosphorylation events involving this protein (128), although the origin of this sequential process has not been determined.

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

480

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

The connection between ITAM phosphorylation and other biochemical changes in the cell is complex. Although some investigators have reported the ability of singly phosphorylated sites in the ITAMs to bind adapter proteins such as Shc (129–131), it is generally believed that biphosphorylated ITAMs play the primary role in downstream signaling, through their recruitment of the syk-family kinase ZAP-70 (130–132). This kinase has tandem SH2 domains that have been shown by mutagenesis (132), binding studies (132–135), and crystallography (136) to interact with high avidity with the two phosphorylated tyrosines in a single ζ ITAM. In contrast to Syk (137), this binding per se does not activate the kinase function of ZAP-70 (138), which depends on tyrosine phosphorylation by Lck (138–140). Kinase-active ZAP-70 in turn tyrosine phosphorylates a major substrate called LAT (141), which serves as an important adapter for interaction with Grb-2, phosphatidylinositol 3-kinase, phospholipase C (PLC) and other downstream signaling molecules or adapters that mediate more distal steps in the signal transduction process. Other adapter proteins such as TRIM may be directly tyrosine phosphorylated by src kinases rather than by activated ZAP-70 and contribute to downstream signaling by recruiting such molecules as phosphatidylinositol 3-kinase to the membrane (142). An intriguing possibility is that complex linkers like LAT and possibly TRIM serve the role of scaffold proteins, whose function has been well-documented in several eukaryotic systems (143). Such scaffold proteins can act as a site of integration of signals arising from several biochemical pathways and provide an opportunity for hierarchical control of signal throughput to downstream effectors. In accord with this concept, LAT associates with SLP-76 (144–146), possibly via Grb2, and SLP-76 in turn appears to be directly linked to phosphorylation and activation of PLC-γ , as well as Ras-pathway stimulation (147). Once activated, PLC-γ cleaves phosphatidylinositol 4,5-bisphosphate and generates the second messengers inositol-1,4,5-triphosphate (IP3) and diacylglycerol. IP3 binds to receptors in the endoplasmic reticulum to initiate release of intracellular Ca2+ stores. The rise in Ca2+ in combination with diacylglycerol activates protein kinase C (148). Activation of the IP3 receptors and release of stored Ca2+ are coordinated with opening of CRAC channels that allow influx of extracellular Ca2+ into the cytoplasm (149). Release from intracellular stores by IP3 accounts for the early rise in intracellular free Ca2+ that accompanies TCR engagement, whereas CRAC channel opening is essential to sustained rises in the level of intracellular free Ca2+. Increased intracellular Ca2+ activates the phosphatase calcineurin, which dephosphorylates the cytoplasmic component of the transcription factor NF-AT (150), resulting in its accumulation in the nucleus in association with members of the AP-1 family (151, 152). These and other modifications of the activity and localization of preformed positive- and

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

481

negative-acting transcription factors link proximal TCR second messenger generation to the changes in gene activity we typically measure as cellular responses (153). Given this information about the biochemistry of TCR signal transduction and gene activation, two different approaches are available for experimentally testing the prediction of progressive attenuation made by the kinetic proofreading model and for determining where in the signaling cascade the influence of affinity makes itself most apparent. One method is to begin at some distal step of the signaling cascade, show a difference when using ligands of distinct functional quality, and then work back to the more proximal events to see at what step and to what extent attenuation affects the preceding steps in the pathway. The alternative is to begin as proximally as possible and to identify the extent of attenuation as one moves more distally. Both approaches have been taken, by different investigators. Some have looked at parameters such as changes in ˇ a, WE intracellular pH (154, 155), Ca2+ (156, 157), or MAPK (158; I Stefanov´ Biddison, R Germain, unpublished data), with the latter being removed from the first events known to follow TCR engagement by 3–10 biochemical steps. ˇ a, WE Biddison, Results of analyses of Ca2+ (157) and MAPK (158; I Stefanov´ R Germain, unpublished data) in particular have indicated that changes in ligand quality affect signal duration, and not just intensity, implying that a simple proofreading model with stepwise attenuation cannot fully explain the properties of the signaling system. Others have examined the earliest known changes, namely activation of TCR-associated src family kinases such as Lck, and the tyrosine phosphorylation sites in protein chains associated directly with the αβ TCR, such as ζ . These studies involving more proximal phosphorylation events have been revealing because they allow evaluation of qualitative rather than just quantitative aspects of the signaling events.

Relating Ligand Pharmacology to Early TCR Phosphorylation Events The first direct evidence for altered intracellular signaling in response to TCR ligands with partial agonist properties came from two groups studying proximal tyrosine phosphorylation events. Using nontransformed mouse Th1 clones, both reported a change in the pattern of ζ phosphorylation and a defect in the phosphorylation of ZAP-70 (159, 160). One of the studies also documented that although ZAP-70 was not activated, this kinase did bind to the incompletely phosphorylated ζ chains (160). In both cases, agonist induced the appearance of two forms of phosphorylated ζ monomers, now referred to as p23 and p21, in approximately a 1:1 ratio. In contrast, partial agonist led to little if any p23 accumulation but did induce the formation of p21 phospho-ζ . Agonist also induced substantial phosphorylation of CD3ε, whereas partial agonists did so

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

482

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

weakly, if at all, and clearly to a disproportionate extent compared with p21 phospho-ζ formation. Decreasing concentrations of agonist, in contrast, led to a lower intensity of signal for p21 ζ , p23 ζ , phospho-CD3ε, and phosphorylated ZAP-70, but the ratio of these products remained relatively constant as the overall signal declined with ligand concentration. These data made clear that the signaling seen with partial agonists did not represent just a weak form of agonist signaling due to lower overall TCR occupancy resulting from use of lower affinity ligands. Rather, the data were fully consistent with the standard definition of partial agonist in that there was a nonparallel change in signals associated with receptor occupancy. These specific alterations in early downstream signaling from the TCR have been reproduced with many different T cells, clones, lines, and even fresh ˇ a& cells, using mouse CD4+ and CD8+ T-lymphocytes (161, 162; I Stefanov´ RN Germain, unpublished data). Similar data on the patterns of TCR-associated phosphorylation in response to ligands of varying pharmacologic quality have ˇ been obtained with human cells (76; I Stefanov´ a, B Hemmer, M Vargelli, R Martin, WE Biddison, RN Germain, unpublished data), although here the analysis involves nonreduced ζ dimers migrating at apparent molecular masses of 32,000 (mouse p21) and 38,000 (mouse p23). Formation of p21 phospho-ζ has also been reported in response to antagonists showing no agonist function at any tested concentration, providing clear evidence for TCR signaling without a corresponding typical activation response. The only exception to this overall pattern of TCR signaling so far is with mouse Th2 cells, which do not seem to produce the expected amount of phosphorylated ζ or ZAP-70 in response ˇ to agonist as compared with Th1 cells bearing identical TCR (I Stefanov´ a& RN Germain, unpublished data). The significance of this difference is unclear, but it emphasizes that the patterns of signaling observed are not an inviolate representation of TCR:ligand affinity but are regulated by the nature, concentration, and state of other signaling molecules in the cell that interpret this binding event. The relationship between different mobility forms of phosphorylated ζ and specific tyrosine phosphorylation events has been partially clarified by immunoblotting studies using antisera to each tyrosine in each ITAM and T cells stimulated with agonist or partial agonist ligands (128). Several distinct patterns of tyrosine phosphorylation appear to give rise to the p21 and to the p23 mobility forms of phospho-ζ seen in blotting experiments. Some results suggested that p21 ζ isoforms containing only monophosphorylated ITAMs could bind ZAP-70 stably, but further study is needed to verify this interpretation in light of other data (134, 135). p23 ζ also appears to consist of several distinct phosphorylated species, but only those having all tyrosines phosphorylated seem associated with the presence of phosphorylated ZAP-70 in the cell. Different ligands appear to induce production of various ratios of the p21 mobility forms,

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

483

and partial agonists were reported to give rise to p23 forms lacking the critical phosphotyrosines needed for ZAP-70 activation. Perhaps the most intriguing data in this study were those revealing an inviolate order for phosphorylation of specific sites in ζ . If each site were independently available for modification by Lck, the mutation of a more easily phosphorylated site should not prevent a strong agonist capable of prolonged Lck activation from inducing the phosphorylation of the less easily modified sites. The observation of such a limitation effect suggests that the modification of the second location depends on an event such as a change in the structure of the ζ chain, rather than just the length of time the TCR is occupied and Lck is active and able to cause phosphorylation. At first blush, this is seemingly inconsistent with a simple kinetic explanation for the progressive phosphorylation of the ζ sites. However, one could imagine that this necessary change is part of the proofreading mechanism, akin to the absolute inhibition of peptide bond formation for amino acyl tRNAs bound incorrectly to the mRNA via a codon:anticodon mismatch. If the later phosphorylated tyrosines play specific roles in downstream signaling events, then placing an absolute barrier to such modifications unless an earlier phosphorylation takes place ensures that the signaling event proceeds for a minimum length of time.

The Role of CD4 and CD8 Coreceptors in Determining Ligand Quality and Signaling Phenotype Ligand pharmacology seems to generally track affinity for the TCR, and ligands with distinct functional properties also give rise to alternative early phosphorylation/ZAP-70 activation events. This raises the question of how variations in TCR affinity can be translated into distinct patterns of receptor-associated phosphorylation. More specifically, in the context of a kinetic proofreading model, what are the timing events that set barriers to further signaling in a discrete fashion in relationship to receptor occupancy? One possible “clock” is the ζ chain itself, which may go through a series of alterations in folding that permit eventual tyrosine phosphorylation of key sites for downstream signal propagation only if (Lck) kinase activity is maintained long enough to drive completely all initial steps in the series. At the moment, it is not apparent whether this potential mechanism has any flexibility, i.e. that it can be modified by the cell to deal with environmental changes. Another possible early timing mechanism in the kinetic proofreading system for which more evidence exists involves the Lck-coupled CD4 and CD8 coreceptors. If the function of these MHC molecule-binding proteins were simply to enhance the avidity of ligand binding to the TCR, as is often stated, one would imagine that they should have evolved to become a permanent part of the receptor complex (see Appendix 2). In this way, every TCR would have this

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

484

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

benefit upon any interaction with a MHC molecule. Instead, their existence as separate molecules in the same membrane as the TCR allows the time it takes a CD4 or CD8 molecule to coengage a peptide:MHC molecule complex already bound to a TCR to provide the cell with a measure of the affinity of the initial TCR:ligand association (61, 163–165). This may be an especially important clock, as such recruitment not only can provide a specific biochemical boost to the signaling cascade by bringing additional Lck to the TCR complex, it also affects the duration of receptor occupancy itself, allowing more time for the system to pass other checkpoints in the proofreading mechanism. Several observations support this concept that the rate of CD4/CD8 recruitment to occupied TCR is a key step in regulation of signaling efficacy and effective kinetic proofreading. On the functional side, a number of reports have shown that changes in the density of CD4 or CD8 are associated with changes in the pharmacologic properties of specific ligand:TCR pairs (164–168). In addition, interference with CD4 function leads to a change in early receptor associated phosphorylation from that typical of agonist (high p23 phospho-ζ and phosphorylated ZAP-70) to that typical of partial agonist/antagonist (low p23, high p21 phospho-ζ and bound but not phosphorylated ZAP-70) (164). It is thus apparent that variations in the available density of CD4 or CD8 alter both the pharmacology and biochemistry of responses to defined TCR:ligand pairs, consistent with a key role for the rate of (Lck-coupled) coreceptor recruitment in defining the efficacy of TCR ligands. The use of nonassociated proteins capable of binding to the same ligand as the TCR in setting affinity thresholds for effective intracellular signaling has many advantages. First, such a system should be largely independent of the density of the total pool of TCR but highly dependent on the density of the coreceptor.2 This is because as long as the system treats each receptor as a separate unit, it only evaluates the length of time a ligand is bound to a particular TCR. Hence, changing the density of surface TCR display within broad bounds should not affect the affinity required for effective signal generation. In contrast, reducing the density of coreceptors immediately increases the average time to association with a ligand-engaged TCR and imposes a requirement for higher ligand affinity for useful signaling. This could permit a thymocyte with a particular receptor whose affinity for environmental self-ligands is close to that needed for activation to “tune” its proofreading clock by lowering coreceptor density, thus eliminating effective signaling by such ligands (166). Although such coreceptor density tuning has been reported, a key question remains—is this the consequence of cell selection or of active regulation of expression, and 2 This is probably true only above a certain threshold. When the density of TCR falls too low, changes in the rate of overall signal generation rather than the nature of signaling per TCR come into play with respect to effective T cell activation measured functionally (33, 169, 170).

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

DYNAMICS OF TCR SIGNALING

485

is this a stable trait of the cell or is it a change requiring constant feedback control? Can the variation in coreceptor density be autonomously maintained by the naive cell in the periphery and can it be maintained after several cell divisions in response to foreign ligand activation of the cell? These are all crucial issues for future study.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE FEEDBACK REGULATION OF TCR SIGNALING In the preceding section, much of the description of TCR signal transduction was presented in linear terms, with step A leading to step B in a strictly forward direction, and with the rate of TCR-ligand disengagement controlling the number of such steps completed for any occupancy event. Although a substantial amount of data argues in favor of this relatively simple forward rate concept, it is widely appreciated that other receptor systems are regulated by feedback processes that are themselves activated by the initial signaling events. A primary role of such feedback systems is to limit the extent and duration of receptor activity, especially in the continued presence of ligand. Several proteins that limit effective TCR signaling have been described and, in a few cases, their function linked biochemically and kinetically to the upstream events stimulated by TCR interaction with ligand. More recently, evidence has emerged that positive feedback regulation also helps control TCR signal transduction, and evidence has been obtained that the interplay of this positive control element with a negative regulatory mechanism does more than just modulate signal strength or duration. Kinetic competition between these feedback pathways appears to be an integral part of the mechanisms underlying functional discrimination among related ligands of the TCR.

Negative Regulation of TCR Signaling Given the central role of Lck and of phosphorylated CD3/ζ chains in effective intracellular signaling by the TCR, it is not surprising that a number of distinct inhibitory events target these molecules. Csk is a tyrosine kinase whose preferred substrates are the inhibitory tyrosines of src-family kinases such as Lck residue Y505 (171, 172). Csk binding to Lck and enzymatic targeting of tyrosine 505 are enhanced by Lck phosphorylation at the activating tyrosine 394 (173), which is typically the site of Lck autophosphorylation during effective TCR signaling. Thus, TCR engagement leads to activation of Lck, and Lck activation recruits Csk, which in turn inhibits Lck, producing a classical feedback inhibition loop. Some information is available concerning the relative rates of Lck activation, Csk activation, and Csk-mediated Lck inactivation (174), but variant ligands have not been examined. A second mechanism of inhibition

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

486

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

of TCR signaling involves CD45. Although initially believed to function as a positive regulator of TCR function through removal of the Csk-added phosphates at position 505 of Lck (171), more recent studies have indicated that CD45 can also serve to inhibit TCR signaling (175) or reduce the duration of such signaling by removing phosphates from ζ chain ITAMs (176). Another molecule recently shown to inhibit TCR signaling is c-Cbl. T cells from mice with null mutations in the c-cbl locus show increased levels of tyrosine phosphorylated ZAP-70 upon TCR cross-linking (177). As with the above regulatory proteins, little information on the relationship of its activation to the normal kinetic progression of TCR signaling or to ligand quality is available. A fourth system for interfering with TCR signaling involves the membrane protein CTLA-4 (178). This molecule is a homolog of CD28 and binds the same CD80 and CD86 ligands as the costimulatory receptor (179). However, the consequences of this engagement are quite different. If the CD80/86 ligands are on the same presenting cell surface as the TCR ligands, allowing colocalization within the zone of cell-cell contact (180), or if CTLA-4 is co–cross-linked to engaged TCR using antibodies (180, 181), TCR-dependent cell activation is reduced. CTLA-4 has been reported to bind the cytosolic tyrosine phosphatase SHP-2 (182), and very recent data indicate that coclustering of CTLA-4:SHP-2 complexes with the TCR leads to dephosphorylation of many of the substrate molecules typically tyrosine phosphorylated in response to TCR engagement (J Bluestone, personal communication). Because CTLA-4 tends to appear at high levels on the cell surface only hours after initiation of TCR signaling (183), this regulatory system may not primarily operate on the early phase of TCR signaling but rather during a later sustained phase that has been shown to be critical for cell cycle progression. Evidence consistent with a major role for CTLA-4:SHP-2 in controlling clonal expansion comes from the phenotype of mice deficient in CTLA-4 expression, in which disregulated CD4+ and CD8+ T cell proliferation leads to massive lymphocyte accumulation (184–186). Although all of these mechanisms of signal attenuation are of evident importance, none has been shown to play a key role in discrimination among ligands with different TCR binding properties rather than simply limiting signaling on interaction of any ligand with the TCR. Furthermore, we also know little about the way in which the activation of these feedback pathways is linked temporally to the initiation of TCR signaling and how they are controlled so that effective downstream transmission of messages from agonist-engaged TCR takes place but does not proceed too long or to too great an extent. A different feedback regulatory system, however, has now been characterized with respect to many of these particular issues and found to contribute to ligand discrimination as well as to down-modulation and recovery from normal agonist signaling. The protein in question is SHP-1, another cytosolic tyrosine

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

487

phosphatase (187). Mutations in the gene encoding this molecule give rise to the moth-eaten phenotype (188–190). Analysis of antibody-induced signaling by the TCR of mature thymocytes from such mutant mice revealed prolonged activation of downstream molecules such as MAPK (191). In vitro, SHP-1 was reported to dephosphorylate ZAP-70 (192). Furthermore, expression of a dominant-negative form of SHP-1 in a T-hybridoma increased the response of the transfected cell to TCR signaling by peptide:MHC molecule ligand. Most recently, SHP-1 has been shown to be present in complexes with the TCR of ˇ cells exposed to various ligands (I Stefanov´ a, B Hemmer, M Vergelli, R Martin, WE Biddison, RN Germain, unpublished data). These data all indicate a role for SHP-1 in termination of effective TCR signaling, possibly through interference with the activity of ZAP-70 and/or Lck in the receptor complex itself. Studies have been conducted that look at the relationship between SHP-1 recruitment to the TCR, the inhibition of signaling by such receptors, and the nature of the ligand engaging the TCR. These experiments have revealed a striking effect of ligand quality on the rate of SHP-1 appearance in engaged TCR complexes, as well as the role of SHP-1 in terminating normal signaling by agonist, and in desensitizing T cells to ligand reexposure. Consistent data have been obtained to date with human CD4+ T cells, human CD8+ T cells, and mouse CD4+ Th1 clones. In each case studied, the appearance of SHP-1 in TCR complexes as evaluated by coimmunoprecipitation is rapid if the ligand has antagonist activity, intermediate for partial agonists, and slowest for agonists ˇ (I Stefanov´ a, B Hemmer, M Vergelli, R Martin, WE Biddison, RN Germain, unpublished data). Thus, a negative feedback system is more rapidly engaged using ligands that are of lower functional potency and that have an apparent intrinsic limitation in their capacity for sustained signaling owing to more rapid dissociation from the TCR. This finding implies that whatever kinetic limitations exist in such “forward” signaling through the TCR due to abbreviated occupancy by partial agonists or antagonists, the engagement is of sufficient duration and efficacy to yield the signals required for SHP-1 association with the receptor complex. The relationship between poor signaling and rapid recruitment of SHP-1 suggests that the inappropriately early interaction of TCR occupied by partial agonists or antagonist with SHP-1 also contributes to the low functional potency of these ligands. Finally, the rapid recruitment of SHP-1 to TCR that engage antagonists may contribute to the ability of such ligands to inhibit responses to simultaneously available agonist, by creating a pool of TCR held in an incompetent state by the docked phosphatase. Direct evidence that antagonists can desensitize TCR by promoting the docking of SHP-1 has come from studies with human CD4+ T cells. Exposure of such cells to presenting cells bearing an antagonist for several hours, followed by the removal of these presenting cells,

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

488

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

results in the creation of a large pool of TCR complexes containing SHP-1. If these cells are then exposed to agonist, the pattern of early TCR-associated phosphorylation resembles that of partial agonist signaling, and the response ˇ is substantially blunted, even though antagonist is not present (I Stefanov´ a, B Hemmer, M Vergelli, R Martin, RN Germain, unpublished data). A similar desensitization of TCR occurs using agonist after 1–2 h of effective signaling, and the T cell regains sensitivity to stimulation only in proportion to the decay ˇ in this phosphatase:receptor association (I Stefanov´ a, B Hemmer, M Vergelli, R Martin, RN Germain, unpublished data).

ERK-1 Positive Feedback Regulation of SHP-1 Negative Feedback Inhibition There is a logical disconnect between (a) the notion of SHP-1 as a feedback regulatory molecule whose activity should be recruited in proportion to the activation of the TCR and (b) the observation that partial agonists and antagonists can evoke even more rapid and extensive recruitment of SHP-1 to the TCR complex than agonists that more potently trigger TCR signaling. Two different models could be offered to explain this apparent discrepancy. One possibility is that partial agonists and antagonists could be more effective than agonists in triggering a particular biochemical change in the TCR that is relevant to SHP-1 recruitment, for example the specific tyrosine phosphorylations leading to p21 phospho-ζ . The problem with this explanation is that it does not account for why agonist should be highly effective at promoting SHP-1 binding at a later time after initiation of signaling—the off-rate of TCR binding should not change and in a simple proofreading model, neither should the drive to full ζ phosphorylation or any other qualitative difference between partial agonist/antagonist versus agonist signaling. An alternative explanation more in line with the evidence that agonist can trigger events that the partial agonist/antagonists fail to elicit or elicit poorly is that at least one such agonist-related signaling event serves to protect the TCR from SHP-1 recruitment. In the absence of this factor, the binding of SHP-1 is unopposed when using partial agonists/antagonists, hence the faster and more extensive binding of the phosphatase. Some other limitation to continued full efficacy of agonists in producing this regulator of SHP-1 recruitment would then need to be invoked to explain the late binding of SHP-1. We have recently identified a product of agonist signaling that interferes with SHP-1 recruitment to the TCR, as the latter hypothesis predicts. What is surprising is that it is active ERK-1, a serine/threonine kinase activated by the Ras pathway and traditionally considered a downstream effector of TCR signaling that connects cytoplasmic to nuclear events (193). Even more remarkable, the action of ERK-1 as a positive feedback regulator involves the binding of ERK-1

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

489

directly to Lck in the TCR complex, where its enzymatic modification of this tyrosine kinase interferes with SHP-1 docking. These observations tie together a number of isolated results already in the literature concerning Lck serine phosphorylation (194), Lck:MAPK binding (195), and SHP-1 tyrosine phosphorylation (196), none of whose physiological significance was previously understood. An important lead in the discovery of this role of ERK-1 in regulation of TCR signal transduction came from analyzing the magnitude and timing of MAPK activation in human CD8+ T cells exposed to agonist or partial agonist ligand ˇ (I Stefanov´ a, WE Biddison, RN Germain, submitted for publication). In contrast to the most obvious expectation of a kinetic proofreading model in which the activation of increasingly more distal signaling molecules should be highly attenuated using partial versus complete agonists, the peak height of the MAPK response was not reproducibly lower for partial agonist compared with agonist. What was always true was that the peak response was delayed with partial compared with full agonist, and the decline from the peak of MAPK activity was also more rapid for the less-effective ligand. This pattern had the properties one might expect from a molecule that would help the agonist-engaged TCR to resist inactivation and that would be defective/deficient with partial agonist. The delay in the rise of MAPK activity would not provide effective protection to the initial pool of partial agonist-engaged TCR, and the unopposed function of a negative regulatory molecule (which we now understand to be SHP-1) would lead to a rapid decline in signaling that would also affect the activity of MAPK itself. Even though MAPK at the time was considered strictly a far-downstream effector molecule with no direct relationship to proximal TCR signaling events, these kinetic features of the MAPK response led us to examine whether it may play a role in early TCR signaling events in relationship to SHP-1 recruitment. A variety of experimental approaches was used to examine this hypothesis, which together provided strong evidence for a direct action of the ERK-1 isoform of MAPK in limiting receptor association with SHP-1. We now understand some aspects of the mechanism involved in ERK-1 inhibition of SHP-1 binding to the TCR. MAPK can serine phosphorylate Lck at residue 59 (194), leading to a change in electrophoretic mobility from an apparent molecular mass of 56,000 to one of 59,000. This change from p56 to p59 occurs in concert with the appearance of ERK-1 in TCR immunoprecipitates from agonist exposed cells, and reversion to p56 coincides with SHP-1 ˇ appearance in the TCR complex (I Stefanov´ a, WE Biddison, RN Germain, submitted for publication). Serine 59 modification may alter the binding properties of the Lck SH2 domain, which when unmodified may bind SHP-1 through a C-terminal sequence containing a phosphorylated tyrosine. Together these observations suggest that SHP-1 recruitment to the TCR may be regulated by

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

490

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

the tyrosine phosphorylation of SHP-1, which transfection data indicate can be mediated by Lck itself, consistent with this being a feedback loop. The concept that Lck associates via its SH2 domain with SHP-1 tyrosine phosphorylated by Lck fits well with the findings of Songyang et al, who showed that Src family kinases have a preferred substrate specificity for sequences that matched the SH2 consensus of the same enzyme (197). In vitro studies show that SHP-1 binding to active Lck lacking the Ser-59 modification leads to its dephosphorylation at position 394, inhibiting kinase function. Activation of SHP-1 phosphatase function requires binding of its N- and C-terminal SH2 domains with tyrosine phosphorylated targets (198, 199). Therefore, the putative Lck SH2-mediated binding of tyrosine phosphorylated SHP-1 is likely to be a rate limiting step in recruitment, but interactions of the phosphatase with other tyrosine phosphorylated proteins in the Lck-TCR complex would appear to be key to the activity of the inhibitory protein once docked. This sequence of events helps explain how opposing biochemical events focused on Lck can be used to augment both the sensitivity and selectivity of TCR signaling. TCR binding to agonist promotes effective signaling not limited in the early steps of the pathway by any proofreading events, in particular coreceptor:Lck recruitment. This results in a rapid raise in active ERK-1 that binds to TCR complexes via Lck and modifies this key kinase, possibly also in unengaged receptors entering the zone of presenting cell contact. By protecting these incoming TCR from SHP-1 association, this helps enforce effective signaling over a sustained time interval, as these protected TCR interact with available agonist. In contrast, with partial agonist/antagonist, other proofreading mechanisms acting early in signaling slow the rate of increase in active MAPK, whereas the proximal event of Lck activation may contribute to a buildup of phosphorylated SHP-1. This pool can associate with the Lck before a level of functional ERK-1 is reached that is adequate to protect most of the TCR in the zone of cell contact. The SHP-1 associated with Lck in these TCR has the potential to inhibit kinase function through dephosphorylation of Y394, which in turn can further depress effective signaling beyond that due to just the early proofreading events linked to dissociation rate, limiting the total signal produced and especially its overall duration. Thus, in complete contrast to agonist where dominance of the ERK-1 positive feedback pathway helps amplify effective signaling, the inadequacy of the positive pathway and dominance of the SHP-1 negative pathway for kinetic reasons now suppress signaling. In the case of some self-ligands lacking the quality of full agonists, this may limit their capacity to exceed the threshold for functional cell activation. Grossman & Paul have clearly pointed out the importance of a rapid increase in positive signaling, resulting in effective cell activation (200). Although it did not deal explicitly with positive feedback of the kind described here, the emphasis we

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

491

place on the importance of a rapid rise in induced second messenger levels for effective signaling is fully consistent with their proposals. A characteristic of systems with opposing positive and negative feedback regulation is the tendency to “overshoot” in each direction, resulting in oscillations in signal intensity over time (201). Such oscillations have been observed in T cells exposed to presenting cells bearing agonist ligand, and they are especially striking at low ligand densities where the positive regulatory events ˇ cannot fully dominate quickly after initiation of ligand exposure (I Stefanov´ a & RN Germain, unpublished data). Given the evidence that oscillations in such second messengers as intracellular Ca2+ can dictate the nature of gene activation events (202, 203), these findings are clearly of significance in understanding the role not only of ligand quality but also of amount in dictating the functional responses of T cells (204, 205).

TCR Activation, Down-modulation, and Serial Engagement In addition to biochemical interference with signal propagation, another mechanism of inhibitory feedback regulation is characteristic of many receptor systems. At some point after initiation of intracellular signaling due to ligand binding, many surface receptors undergo internalization, which is followed either by degradation or by removal of ligand and recycling of receptors back to the plasma membrane in inactive form. The TCR is no exception, and it is well documented that either antibody cross-linking or exposure to stimulatory ligands leads to loss of TCR from the cell surface beginning a few minutes after TCR signaling begins, and which is followed by receptor degradation (206). Any simple picture of the link between TCR engagement and receptor internalization has become complicated as ligands other than anti-TCR antibodies or strong agonists have been analyzed. Ligands of low potency, but nonetheless stimulatory of various T cell responses, can lead to little measurable TCR loss from the cell surface (71, 73, 76, 207, 208). Some have concluded from these data that there is no relationship between TCR down-modulation and effective signaling for responses. However, on closer inspection, these data can be reinterpreted in light of the studies of other groups (33) and the concept of hierarchical response thresholds. Analysis of TCR down-modulation using human CD4+ T cell clones and ligands of varying potency provides strong evidence that the removal of TCR complexes from the cell surface depends on “efficacious” signaling by that TCR, as indicated by p38 phospho-ζ and phosphorylated ZAP-70 generation (76). The clear implication of these results is that TCR are only lost from the surface if they complete the signaling process; TCR arrested in an incompletely active state do not undergo internalization. Therefore, for responses that can be elicited without extensive TCR loss, if

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

492

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

these are the responses with the lowest 50% effective dose (that is, those most sensitive to signaling), then very few fully activated TCR may be needed to reach the threshold for this response, consistent with the low level of TCR down-modulation observed (73). Use of much more potent ligands will lead to much greater TCR loss, but the signaling here will be in substantial excess of that needed to reach the threshold for the sensitive response. One of the most important concepts to emerge from these analyses of TCR down-modulation is the idea of heterogeneity in TCR signaling by individual TCR on a single cell (76). Immunoblotting results with a wide variety of cells always show both p21/p32 and p23/p38 ζ formation, even using strong agonists. Because a population of cells is used for each analysis, this could be due to variations in signaling among different cells, with some incapable of having any TCR reaching the fully active state and others with TCR all reaching this state if occupied. Alternatively, some TCR on a cell might complete the signaling process whereas others on the same cell might arrest the process before completion. The linkage between TCR down-modulation and full signaling provides evidence against the former model, because the loss of surface TCR is homogeneous among the treated cells, rather than bimodal as this hypothesis predicts. What changes with the quality/affinity of the ligand for the TCR is the fraction of engaged receptors that enter one or another state, resulting in a wide array of ratios of the different signals that accompany complete and incomplete downstream transmission of signals by the various engaged TCR complexes. This evidence for the coexistence of TCR that signal differentially within a single cell has crucial implications for the issue of cross-regulation of TCR function, as is discussed below. Another significant issue connected to TCR down-modulation is that of serial engagement of receptors by individual peptide:MHC molecule ligands. Using TCR loss from the cell surface as an assay, Valitutti et al first reported that the loss of TCR exceeds the number of ligands on the presenting cell membrane, sometimes by a ratio of as much as 200:1 at very low ligand densities (209). These investigators proposed that this is an amplification mechanism of central importance to ensuring responses by T cells to very low ligand levels, as would be the case early in an infectious process. Although some investigators have interpreted their results in the context of this model (99), only recently have the quantitative aspects of the initial observation been confirmed for human cells, although the same study has failed to observe as extensive an effect with mouse T cells (Y Itoh, B Hemmer, R Martin, RN Germain, submitted for publication). Clearance from the cell surface of fully activated TCR complexes once they disengage from ligand might contribute to efficient use of small ligand numbers by preventing rebinding of the ligand to a TCR that has already contributed its quantum of signal to the T cell.

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

DYNAMICS OF TCR SIGNALING

493

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

TCR INTERACTIONS: OLIGOMERIZATION, CROSS TALK, AND POSITIVE AND NEGATIVE COOPERATIVITY Although it is possible to rationalize much of the existing data on TCR signaling in terms of small units of one TCR and one or two CD4 or CD8 coreceptors, a variety of findings point to an important role of higher-order oligomers in physiologic signaling events. In addition, the new information of ERK-1 and SHP-1 regulation of TCR signaling provides evidence for biochemical cross talk that may involve TCR complexes that are not directly linked to the receptor initially generating the activated enzyme. This evidence for more complex interactions among populations of TCR and for cross regulation of signal transduction needs to be integrated with the basic model of kinetic proofreading to provide a more complete understanding of how TCR occupancy relates to cellular response.

Higher-Order Protein-Protein Associations Involving the TCR and CD4/CD8 Coreceptors Many experiments using antibodies as ligands have shown that coaggregation of TCR and especially of TCR and CD4 or CD8 provide the most effective signal to T cells (164, 210). Microscopic studies demonstrate that TCR and coreceptors accumulate at the site of T cell contact, with the presenting cell bearing specific ligand (211, 212). At first glance, these data might be taken to indicate a role for direct association of multiple TCR and coreceptors during effective signal generation. But because the antibody studies do not necessarily reflect the situation when physiologic (operationally monovalent) receptor ligands are employed and because the resolution of the light microscopic images is too low to distinguish between mere concentration of TCR and coreceptors in a region of the membrane versus physical interaction of these proteins, it is not possible to use these findings to make strong arguments in either direction about the occurrence and importance of higher-order associations. Other observations make a more convincing case for specific formation of complex oligomers during physiologic ligand recognition. Various reports suggest a 2:1 stoichiometry of interaction between CD4 and the TCR during active signaling (213, 214), and crystals of soluble CD4 show a well-ordered dimer (215). Mapping of the sites on MHC class II molecules controlling functional interaction with CD4 unexpectedly revealed two discrete regions whose alterations diminished or eliminated CD4-dependent responses (216). In the crystallographic structure of MHC class II molecules (10), these two sites are on opposite faces of an individual class II heterodimer (one in α2 and one in β2), a distribution incompatible with simultaneous binding to a single CD4 molecule, but one that fits well with asymmetric cross-linking via the arms of the CD4

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

494

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

dimer. Alternatively, the two class II sites lie together in the crystallographic “dimer of dimers” observed with various different HLA-DR molecules, raising the possibility that they form a conjoint surface for binding to one arm of the CD4 dimer, with a second class II “dimer of dimers” binding symmetrically to the other CD4 subunit. Although not all class II molecule crystals show the same dimer orientation (35), there is limited evidence from microscopic studies of the distribution of class II molecules on cell membranes (217) and from mutagenesis studies (S Fleury, J Brown, L Stern, D Wiley, RN Germain, unpublished data) that the HLA-DR dimer-of-dimers model has relevance to the natural state of the protein and to the generation of effective signals for the T cell. Further support for some sort of oligomerization in the function of CD4 comes from the identification of a mutation at position Phe43 of human CD4 that acts as a dominant negative when coexpressed with wild-type CD4 (218), a finding that is not easily explained unless some higher-order state of CD4 is involved in its function (219). The most direct evidence for ligand-induced clustering of TCR comes from the work of Reich et al (220). Using light scattering methods, these investigators observed the formation of complexes with six or more subunits when soluble TCR and soluble peptide:MHC molecule ligands were admixed. Neither component alone showed self-association at the concentrations used, and only peptide:MHC molecule combinations known to bind well to the TCR in question showed this effect, which they did at concentrations matching the known Kd of the pair. The mechanism for such clustering was not addressed in the light scattering experiments. It could reflect an allosteric change in the bound TCR that generates a site for self-association, or it could depended on the capacity of the bound ligand:TCR combination to enter into a multipoint association that yields a higher avidity of binding than either component would have for itself (57, 164, 221). What advantage would oligomerization offer the T cell? The possibilities include both extracellular and intracellular effects (221). On the outside of the cell, the tight clustering of TCR and coreceptors can help achieve an effective occupancy by ligand that is greater than individual complexes alone. This occurs because when the density of receptor binding sites is sufficiently high, the likelihood of ligand rebinding to the same or a neighboring receptor on dissociation is greatly increased in comparison to diffusion away from the receptor (222). If the process resulting in a loss of signaling activity on ligand dissociation (for example, the diffusion apart of a TCR and an Lck-associated coreceptor) is a slower process than is the rate of ligand rebinding, then such reassociation will result in the receptor behaving as if occupied by a higheraffinity ligand with a slower off-rate. Thus, ligand-imposed TCR:coreceptor clustering could amplify the efficacy of the ligand. This effect would not apply

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

495

equally to all ligands, because as the off-rate increases, the average time receptors have available to diffuse out of any clusters that have been created increases, reducing the rebinding effect. Therefore, this mechanism can contribute to discrimination among ligands of diverse binding affinity, such that there is a reinforcement effect for ligands with affinities above a certain threshold but not for those below this threshold. To the extent that longer effective occupancy allows the engaged receptor complex to pass through additional kinetic proofreading checkpoints, this phenomenon can amplify the differences among a series of peptide:MHC molecule combinations. At the interface between the outside and inside of the cell, tight clustering can contribute to formation of a zone of exclusion for other proteins, whether on the presenting cell membrane, on the T cell membrane, or associated with the inner leaflet of the T cell membrane (110, 212, 223). Very close packing of physically linked TCR and coreceptors will tend to exclude other proteins from the inside of the growing cluster. If such excluded molecules include such negative regulatory proteins as CD45 or CTLA-4, this will contribute to the maintenance of signaling by those TCR sequestered away from the outer rim of the cluster. This clustering can also contribute to the trapping of regulatory molecules in the same zone of the cell, either passively within the lattice formed by the TCR-coreceptor subunits or actively through direct contact with these molecules.

Cross-Talk Among TCR The site at which close physical approximation of engaged TCR has the best documented effect involves the inside of the T cell. Here there is intriguing evidence for communication between different TCR complexes in the same cell. The role of SHP-1 in mediating the negative effect of TCR antagonists has been mentioned above, but primarily in the context of association of this phosphatase with the TCR directly engaging the antagonist. One characteristic of antagonist-engaged TCR containing SHP-1 is the lack of enzymatic activity of the associated Lck. Analysis of the proteins associated with the TCR of cloned T cells bearing two distinct receptors with different peptide and MHC molecule specificities has revealed that exposure to both receptor ligands results in the rapid accumulation of Lck protein lacking kinase activity in association not only with the antagonist-engaged TCR but also with the agonist-engaged ˇ receptors (BN Dittel, I Stefanov´ a, RN Germain, CA Janeway Jr., unpublished data). Thus, the effect of antagonist occupancy of one set of TCR may be to promote SHP-1 binding to another within the same cell, even though the latter are engaging only agonist. Such “spreading” of the SHP-1 inhibitory signal from antagonist-engaged TCR to agonist-engaged TCR would obviously be favored if the two sets of receptors are in close proximity, as the clustering model

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

496

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

would predict. These findings provide a possible alternative explanation for the inhibition of high-affinity FcRε engagement by low-affinity ligands (224). Rather than reflecting Lyn sequestration, as the authors propose, enhanced recruitment of a phosphatase to the FcRε bound to high-affinity ligand due to low-affinity ligation of many FcRε in the same cell could be the explanation. Perhaps the most interesting manner in which clustering can affect T cell responses comes from the conjoint effect of occupancy prolongation by ligand rebinding and the spreading not of negative mediators like SHP-1 but of positive regulators such as ERK-1. As already discussed, ligands with too low an intrinsic affinity for the TCR are unlikely to be able to take advantage of the rebinding effect to generate fully effective signals on their own. If there are also agonist ligands of higher affinity on the same presenting cells surface, however, the physical commingling of the two types of ligand could promote a greater average occupancy of those receptors initially recruited into the cluster by selfligands [a variant of this suggestion has been proposed (225)]. At the same time, the local elevation of active ERK-1 induced by the subset of agonist-engaged TCR could affect the other TCR brought into close contact, protecting them from SHP-1. The combination of these two effects would be to transform at least a fraction of the TCR that normally would engage only ineffective ligands and fail to produce useful downstream signals into an additional cohort of effective signaling complexes that augment the output of that subset of TCR directly engaged by a limiting density of agonist. The ability of a MAPK positive feedback system to augment the sensitivity of a signaling response has already been documented in differentiating in Xenopus oocytes (226), consistent with the concept that biological systems utilize common subroutines for specific tasks in particular circumstances. This hypothetical scenario can be cast in very relevant terms by considering the situation with a dendritic cell bearing both self-peptide–containing MHC molecules matching those involved in thymocyte positive selection and a small number of complexes containing foreign peptide serving as agonist for a given TCR. Because the TCR has been selected to show at least some level of effective engagement with the self-peptide ligands, though not to perceive them alone as agonists, the clustering of TCR engaged by self-ligands would take place together with the TCR engaging the foreign (agonist) ligand. Through the occupancy enhancement mediated by ligand rebinding, in combination with the ERK-1 protective mechanism, a fraction of this set of self-ligand–engaged TCR would now switch from one incapable for stimulating the T cell to one adding to the magnitude of signal generation by foreign ligand alone. Evidence for this cooperativity comes from preliminary experiments examining the functional responses and the biochemical signals generated by T cells in the presence or absence of synthetic peptides that resemble known agonists for a

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

497

TCR but that lack agonist or antagonist function. The addition of such peptides to the surface display of a presenting cell bearing a fixed level of agonist seems ˇ to have a striking effect on response at the biochemical level (I Stefanov´ a, RN Germain, unpublished data). The standard mass action or MichaelisMenton equations predict that an 81-fold increase in ligand will be required for changing responses from 10% to 90% of maximum (227, 228). In actuality, in the absence of added nonstimulatory peptide and using nonhematopoietic presenting cells that may be deficient in presenting the self-peptides involved in selection, this rise requires an even greater change in agonist concentration, a condition termed subsensitivity. The copresentation of ligands containing the otherwise inactive altered peptide or certain self-ligands results in a phenomenon termed ultrasensitivity, in which the rise from 10% to 90% of maximum response requires less than the 81-fold change in agonist concentration. If borne out in additional experiments, this dramatic effect raises the possibility that thymocyte selection on self-ligands may not be used merely to ensure that available self-MHC molecules are suitable for presentation of foreign peptides that can be seen by the TCR. Rather, a major driving force for such selection may be to ensure the capacity of the T cell to take advantage of this amplification mechanism that permits very sensitive responses to low densities of foreign antigen in the periphery. Testing whether the T cell is capable of this mode of signaling synergy may also be the reason for using self-ligands to maintain naive mature T cell viability, because T cells unable to show such signal amplification may be less useful for responses to an infectious agent.

DIFFERENTIAL REGULATION OF SIGNALING BY THYMOCYTES AND MATURE T CELLS: SURVIVAL VERSUS RESPONSE Changing Signaling During T Cell Maturation Much of the preceding discussion has treated T cells in a generic manner, but it is apparent that the relationship between ligand structure and biological effect, if not signaling, varies during T cell maturation. Many comparative studies of the responses to a set of related ligands of thymocytes versus mature cells expressing the same TCR have documented either enhanced positive (229) or negative (230) selection using peptide:MHC molecule combinations incapable of activating the mature cell. From this and related data, the idea that TCR signaling is tuned to a lower sensitivity during thymocyte maturation has arisen (231, 232). However, these models invoking a general loss of TCR sensitivity do not fully consider the flexibility or sophisticated discriminatory control with respect to ligand structure that we are beginning to appreciate is a characteristic of the

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

498

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

TCR signaling machinery. Recently, we directly examined the sensitivity of immature CD4+CD8+ versus mature CD4+ T cells to a series of related ligands with the properties of agonist, partial agonist, or antagonist for the mature ˇ cells bearing a defined TCR (B Lucas, I Stefanov´ a, N Dautigny, RN Germain, submitted for publication). The results revealed a remarkable and unanticipated feature of T cell development. Ligands with agonist versus partial agonist properties for the mature T cell, and showing the characteristic differences in ζ and ZAP-70 phosphorylation, acted as strong and weak agonists for the immature T cell on a functional level. They showed the same signaling pattern, with high p23 phospho-ζ and phosphorylated ZAP-70 generation in both cases, although the partial agonist for mature cells was somewhat less potent (i.e. fit the definition of a weak rather than partial agonist). Similar biochemical data for just immature thymocytes has been reported in another TCR transgenic model (233). These results help explain a major paradox—how ligands incapable of producing active ZAP-70 in mature cells promote positive selection of immature thymocytes whose development to CD4+ or CD8+ mature cells appears to require the function of this same kinase. Because many ligands that are only antagonists for mature cells act as weak agonists for immature cells and produce activated ZAP-70, one can understand how this enzyme can be used to promote early differentiation and yet not be activated to a level consonant with mature cell effector function by the same ligand in the periphery. The functional studies showed that sensitivity to the stronger ligand is maintained, or in some circumstances increased, whereas the response to the weaker ligand decreases by several hundredfold. In short, rather than seeing a uniform loss of TCR sensitivity to all ligands, as predicted by prevailing models, we saw an opposing or divergent change in response to two closely related ligands. The result was to maintain the capacity for activation by one ligand, while losing all capacity to see activation with another. Consider the stronger ligand as representative of a foreign antigen to which the developing T cell is not exposed in the thymus (and hence is not negatively selected), and the weaker stimulus as a positively selecting self-ligand. In such a case, it is obvious that this capacity for divergent changes in potency allows maintenance of high sensitivity to foreign antigen while providing a barrier to activation by self-ligands, i.e. enhanced self-nonself discrimination. The details of how a selective change in ligand discrimination occurs during T cell maturation are not fully worked out. SHP-1 is lower in immature thymocytes (192; K Yasutomo, B Lucas, RN Germain, unpublished data) and rises during development. Thus, the negative feedback regulation that we have argued helps limit mature T cell responses to self-ligands involved in positive selection may not be fully operative in the immature cells. This could allow useful signaling by TCR engaged with lower-affinity ligands than would be

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

499

effective in mature cells with a fully functional negative feedback loop. In addition, the proportion of coreceptors (especially CD8) that is Lck coupled in immature versus mature cells may differ (234). Based on the rise in surface coreceptor display from the immature to mature state (235) and the steady level of Lck (236), it seems likely that the fraction of Lck-coupled coreceptors declines during the maturation process. In the context of coreceptor-based kinetic proofreading, as described above, such a presumptive change in Lck-coreceptor association would imply that ligands with affinities falling below a rising threshold would lose efficacy as the cell matured, whereas those sufficiently above this timing limit would show little effect, consistent with our findings. Direct analysis of wild-type and genetically manipulated immature and mature T cells will be needed to determine whether this hypothetical mechanism actually contributes to developmental regulation of ligand sensitivity. Finally, several studies have indicated that MAPK may play a central role in positive selection (237–239). Although this is traditionally viewed as due to contributions to nuclear signaling, the recent evidence for the contribution of ERK-1 to effective TCR signal generation may indicate that an important role of this kinase is in ensuring adequate signaling during positive selection through protection of TCR against inactivation by the rising levels of SHP-1 seen during maturation, a period during which persistent TCR signaling is required for survival (240, 241; K Yasutomo, B Lucas, RN Germain, submitted data).

Partial Signaling in Response to Self-Ligands One benefit of a mechanism that does not prevent all signaling of mature cells by self-ligands of the quality involved in positive selection, but that prevents overt activation by these complexes, is the possible use of incomplete signals for other cellular processes. This is a point well made by others looking at regulation of B cell receptor signaling by the Fcγ RIIB in B cells (242) and appears to apply also to T cells. A number of groups have provided evidence that naive T cells require the presence of self-MHC for survival in vivo (50–54). Duke reported an allele-specific association of syngeneic cells with alloreactive cytotoxic T cells that was inadequate for cytolysis on its own but that supported “bystander” killing when the alloantigen agonist was present (243). This was taken as evidence for self-recognition by the TCR as a peripheral manifestation of thymic positive selection. Others have shown that freshly isolated ˇ thymocytes (244) and mature T cells (128, 245; I Stefanov´ a, RN Germain, unpublished data) have partially phosphorylated ζ chains and nonphosphorylated ZAP-70 associated with these chains, a biochemical phenotype identical to that seen on stimulation of T cells with suboptimal ligands. Although the analysis of thymocytes from MHC-deficient mice by van Oers et al implies that at least some of this phosphorylation is not dependent on MHC recognition (244), we

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

500

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

recently observed a relationship among the presence of self-MHC in an animal, the maintenance of this state of receptor partial activation, and the survival of ˇ T cells (JR Dorfman, I Stefanov´ a, RN Germain, unpublished data). This is concordant with in vitro data on enhanced T cell survival on stimulation with partial agonists (246). It is interesting that exposure of T cells to an MHC-deficient environment in vivo results in loss of protein tyrosine phosphorylation involving only a few selected molecules, especially ζ , whereas in vitro culture leads ˇ to the rapid loss of tyrosine phosphates from multiple proteins (I Stefanov´ a, JR Dorfman, RN Germain, unpublished data), differences that correlate with more rapid death in culture. These observations raise the possibility that the incomplete signaling induced by TCR engagement of self-ligands may be a necessary signal, but not a sufficient one, for prolonged survival of mature cells in the naive state. As a whole, these findings suggest that the biochemical mechanisms used by the TCR to discriminate among self and foreign ligands may be designed to prevent the cell from reaching thresholds necessary for effector molecule production, without eliminating all signaling on contact with self-complexes.

BIOCHEMICAL FINGERPRINTING OF TCR LIGANDS AND THE PARTIAL AGONIST QUALITY OF SOME SELF-ANTIGENS The findings that ligands of differing functional properties give rise to distinct patterns of TCR-associated tyrosine phosphorylation and that some stimulatory peptide:MHC molecule complexes are pharmacologic partial agonists have had a major impact on our understanding of the relationship between an immunogen and the T cells it elicits. It is often assumed that the T cells that can be isolated from a primed host are those whose receptors respond to processed antigen as a strong agonist. Yet one of the primary points made above is that the diversity of binding structures in a normal repertoire is likely to result in any given peptide:MHC molecule ligand having the characteristics of a null ligand for some receptors, a partial agonist for others, and an agonist for still others. Whether only the latter relationship leads to clonal expansion and effector function has until recently not been considered an issue. The occasional demonstrations of heteroclitic (more potent) stimulation of a T cell by a ligand slightly different from the initial immunogen have usually been ascribed to better MHC molecule binding by the antigenic peptide and an increased density of available ligand, rather than to better interaction with the TCR (247–249). The introduction of methods for the extensive screening of peptide sequences as T cell stimuli and the application of the new data on TCR phosphorylation patterns as a functionindependent means of evaluating ligand quality have begun to dramatically change this widely held view of the immunogen as the most effective ligand.

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

501

They also provide new insights into the relationship between the TCR repertoire and either self-antigens or pathogen antigens subject to immune selection. It is now possible to produce and functionally evaluate either large pooled libraries of peptides with individual or multiple changes at each position in a specified starting sequence (86, 250, 251) or many individual peptides with slight variations in this starting sequence (16). By these methods, a number of peptide:MHC molecule complexes have been found that are functionally much more potent in stimulating a cloned T cell than the sequence corresponding to the original immunogen (75, 250). In many of these instances, the improvement in stimulation occurs without a gain in binding to the MHC molecule; it also is clone-specific in effect. Together, this argues that the changes in response are due to alteration of TCR binding rather than peptide presentation. It thus appears that many T cells that are isolated following immunization and expansion with a given antigen do not possess TCR capable of optimal responses to the determinant used for their stimulation. Insight into why this should be the case came from perceiving a pattern in the results of such ligand scans and from the nature of the TCR-associated signaling accompanying exposure to the starting ligand and to the more potent analogs. Such heteroclitic ligands have been found most readily when the T cells in question are reactive with self-antigens, for example myelin basic protein (76, 250; ˇ B Hemmer, I Stefanov´ a, M Vergelli, R Martin, RN Germain, unpublished data) or proteolipid protein from the central nervous system (75). Little improvement over the natural ligand is seen for clones specific for true foreign antigens, ˇ such as determinants in influenza virus (B Hemmer, I Stefanov´ a, M Vergelli, R Martin, RN Germain, unpublished data). In the context of both intrathymic and peripheral repertoire selection, this can be imagined to reflect deletion or inactivation of those clones with TCR capable of binding self-ligands with high affinity, leaving the repertoire depleted of such T cells (252). Only TCR recognizing self-ligands as weak partial agonists may survive, presumably doing so because the normal level of self-presentation is too low to signal the cell for either thymic elimination or peripheral unresponsiveness (anergy) (253, 254). When high levels of ligand in adjuvant are used with experimental animals, or high levels of peptide with human T cells, the threshold for activation even with partial agonist signaling can be exceeded and these cells isolated as growing clones. Clear evidence of the partial agonist nature of the analyzed self-ligands comes from examination of TCR ζ and ZAP-70 phosphorylation. In almost ˇ every case tested with human T cells (76; B Hemmer, I Stefanov´ a, M Vergelli, R Martin, RN Germain, unpublished data), the pattern of phosphorylation is that characteristic of partial agonists. A predominance of the p32 phospho-ζ form is observed, as is ZAP-70 binding with little or no measurable phosphorylation. The very best of the heteroclitic ligands (“superagonists” in the

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

502

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

literature) give a pattern of signaling characteristic of true agonist, with high levels of p38 phospho-ζ and of phosphorylated ZAP-70. Thus, the modified ligands are actually the “real” agonists for the TCR of these cells, and the putative “agonist” self-ligands prove to be partial agonists based on function and signal generation. The ability to “biochemically fingerprint” the interaction of the priming antigen with the TCR using ζ and ZAP-70 phosphorylation (at least for CD8+ cells and Th1/Th0 CD4+ clones) provides a direct method for evaluating how close the immunogen is to the optimal agonist for a particular TCR. The more the pattern resembles that of a partial agonist, the more likely it is that a better ligand can be found. This is of immense practical importance for tumor vaccines. Tumor-specific T cells reacting with unmodified self-antigens are likely to show the same properties as the T cells reactive with central nervous system proteins, i.e. partial agonist signaling and responsiveness. Use of the biochemical fingerprinting approach may allow one to predict the existence of heteroclitic ligands that once identified functionally can be verified as optimal by biochemical analysis. Such optimized ligands can be incorporated into vaccines and should be much more potent that the natural self-ligand. Although the target antigen on the tumor cell will remain a partial agonist for the effectors that are elicited, one should gain tremendously in dose efficacy for priming and potentially also in the magnitude of the ensuing response, making available more effectors that should still be triggered by the self-partial agonist on the tumor targets. These data on the partial agonist nature of many self-ligand reactive T cells also provides new insight into the pruning of the peripheral T cell pool in support of functional self-tolerance. It has long been assumed that the high-affinity T cells are lost (252, 255); these biochemical data now support that argument in a direct fashion. This applies as well to the “mimicry” model of autoimmunity (256, 257). Rather than the pathogen determinant being able to cross-reactively activate T cells specific for self-antigens, where the latter are considered the “true” ligands of the TCR, one can clearly view the ligands created by pathogen peptides as the real agonists and the T cells they activate as “cross-reacting” with self-partial agonists during subsequent autoaggressive reactions (B Hemmer, ˇ I Stefanov´ a, M Vergelli, R Martin, RN Germain, unpublished data). This is akin to the scenario proposed above for immunization with synthetic variants of self-tumor antigens. Primed T cells appear to be more sensitive to ligand than do naive cells (258); thus, these foreign agonist-activated cells can now respond to the ambient level of self-ligand acting as a partial agonist, when as resting cells they could not. Once this occurs, tissue damage can increase the level of self-antigen presentation, sustaining responses even though the self-ligands are less potent partial agonists for the autoreactive cells.

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

503

Finally, there is emerging evidence that clones derived by “primary” in vitro stimulation using cells from donors naive to a particular infectious agent show a partial agonist signaling pattern when confronted with the priming peptide ˇ (B Hemmer, I Stefanov´ a, M Vergelli, R Martin, RN Germain, unpublished data). Such results suggest the possibility that these cells are actually derived from memory T cells specific for another antigen that can be reactivated by high concentrations of a cross-reactive ligand, namely the antigen used in the in vitro stimulation. This pattern of hierarchical cross-reaction is fully consistent with data recently presented in relationship to the phenomenon of original antigenic sin, as applied to T cells (259). Priming with wild-type lymphocytic choriomeningitis virus followed by immunization with a mutant lymphocytic choriomeningitis virus led to CTL responses more specific for the priming virus than for the second virus. It is remarkable that the pattern was not reciprocal— immunization first with the mutant virus followed by the wild-type virus did not lead to a response dominated by anti-mutant reactive T cells. This asymmetry and the relative dose-response curves for killing of targets bearing the wildtype or mutant glycoprotein peptide imply that the immunoselected variations in the viral protein result in determinants with the just-described characteristics of self-ligands, namely a partial agonist quality for those T cells still available in the repertoire. This is fully consistent with the selection of variants whose altered determinants act as potent antagonists for available anti-viral TCR, as seen with HIV (260) and hepatitis virus (261), which may represent a more extensive form of this type of viral antigen evolution under pressure from the immune system. These findings suggest that a re-evaluation of the effective breadth of the TCR repertoire in the face of microbial genetic variation may be warranted and may lead to a different view of potential pathogen exploitation of this limitation of adaptive immune responses.

CODA Cogito, ergo sum. I think, therefore I am. Ren´e Descartes Sentio, ergo sum. I sense, therefore I am. T cell

Stochastic Resonance and the Role of Self-Recognition in T cell Activation This review began by pointing out the unique tasks in ligand discrimination that need to be accomplished by clonally distributed αβ TCR and the lymphocytes

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

504

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

on which they are expressed. We described the multiple layers of kinetically controlled events that subserve this function, beginning with TCR:ligand binding affinity and involving the rate of CD4/CD8 coreceptor recruitment, ζ chain phosphorylation, binding of negative regulators such as SHP-1, and positive feedback control by ERK-1. Together these mechanisms (and surely others) parse the very complex display of peptide:MHC molecule combinations on antigen-presenting cells to deliver signals that result in life or death choices in the thymus, and life, activation, or death choices in the periphery. Our description of the details of TCR signaling has been cast largely in terms of conventional thinking about ligand:receptor interactions and downstream transduction processes. It seems fitting to end by attempting to bridge between this information and insights from nonimmunological systems that may provide alternative ways of understanding the subtle, but critical, underpinnings of TCR-dependent control of T cell behavior. One major issue is the physiologic significance of what appears to be persistent “weak” signaling by TCR engaging self-peptide:MHC molecule ligands in the peripheral immune system. Although primarily considered to be involved in transmitting signals necessary for maintenance of viability, such constant low-level signaling could play another role. Stochastic resonance is a process involving the summation of random noise in a system with specific incoming signals (262). This generates a total signal that allows a subthreshold level of ligand-dependent receptor input to exceed the triggering limit and produce a measurable response, and it has been proposed to enhance sensitivity in a number of natural systems, including sensory neurons (263). The incomplete biochemical signature associated with the TCR that presumably arises from self-recognition could be considered just such a low tonic level of “noise” that permits T cell responses to foreign ligand densities that would otherwise be nonstimulatory, without limiting the capacity of the same cell to sense suprathreshold inputs from agonist ligand alone. We discussed above the notion of ultrasensitivity that could arise from the spread of positive (ERK-1 mediated) feedback signals from agonist-engaged TCR to other TCR on the same cell. Such an amplification process would constitute a “rain-drop effect” (lateral activity spread) and is in general accord with the model proposed by Bray et al (264). However, in contrast to the bacterial chemotactic receptors that were used to illustrate this model and that are supposed to switch randomly between a high- and a low-affinity conformation without the assistance of ligand, TCR may need priming by self-ligand to become competent for such positive cooperativity. If this is the case, then effective stimulation of T cells by low ligand densities early in infectious processes requires a specific type of cooperation among TCR. Occupancy by agonist (which overcomes the activation threshold in one TCR unit) can be assisted in cellular triggering by

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

505

neighboring TCRs only if the latter are appropriately primed by self-ligand; in this sense, foreign antigen cannot autonomously trigger a response under some circumstances, but requires the “approval” of the self-environment. Furthermore, T cells with a higher intrinsic activation threshold (e.g. those recently exposed to antagonist or in postactivation states that result in docking of negative regulators in the receptor complexes) will be excluded from such positive cooperation. Even if a few receptors receive a strong agonist stimulus, this signal will not be heard by the genetic machinery of the T cell without further amplification, which will be denied in these latter situations. This process is akin to that of adaptation in bacteria, which reduces noise while maintaining responsiveness by decreasing the degree of receptor aggregation and, hence, limiting lateral spread (264). Cooperativity can also occur in a negative mode. We already mentioned infectious spreading of the negative regulator SHP-1 to nonengaged receptors. It is remarkable that SHP-1 spread does not seem to require priming of TCR by self-ligand. Thus, it seems to be easier to silence TCR by negative cooperation than to engage them in signal amplification, where the assistance of self-ligand seems to be necessary for activity spread. This may be so because it is less dangerous to allow TCR negative regulation without additional environmental control than to permit positive cooperativity without such limitation.

Possible “Small World” Character of the TCR Signal Transduction Apparatus One of the most remarkable features of signal transduction is the capacity of individual molecules to be involved in signaling pathways whose activity is controlled by different receptors. Presumably, the disadvantages of such use of the same molecule for different purposes is outweighed by the evolutionarily conservative nature of the strategy and the power of combinatorial control and signal integration in regulating cellular function. At the moment, we are unable to predict the biochemical consequences of simultaneous engagement of two different receptors that employ overlapping sets of signaling molecules. One possibility would be independent, parallel signaling, a result that assumes an unlimited supply of the critical molecules. It seems more likely, however, that there will be cross-talk between the systems, with the signaling “orchestrated” between simultaneously engaged receptors from the onset of ligand occupancy. This may involve both positive and negative effects of modified substrates and regulatory molecules recruited or activated by one or the other receptor system (e.g. see 265). In this regard it is interesting to consider studies demonstrating association of ZAP-70, Lck, and CD45 with the α-interferon receptor (266) or possible involvement of signaling components of cytokine receptors in TCRmediated activation (267). It is especially intriguing that both the TCR and some

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

506

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

cytokine receptors are controlled by the negative regulator SHP-1 (268–270). SHP-1 spreading (mentioned above in the context of TCR antagonism and negative cooperativity) may thus affect not only receptors of the same class (i.e. other TCR), but different receptor complexes competent or primed for SHP-1 binding (e.g. the interleukin 2 receptor). The same principle of “receptor unspecific infectious activity” may also apply to positive cooperation. Thus, the connection topology among surface receptors communicating in clusters may resemble “small-world” networks (271) [a phenomenon also known as six degrees of separation, in which a few random interactions among many individuals (nodes) in an extended network markedly increases the rate of communication and connectivity within the system as a whole]. This phenomenon has been suggested to operate in neuronal and other systems, where it enhances the speed of signal propagation and promotes synchronization of cellular networks.

CONCLUSION Although many of these last considerations regarding “computational” signaling in T cells are speculative, they integrate known facts concerning a T cell’s sense of its environment and provide a conceptually intriguing model amenable to further study. This is especially true with regard to (a) the principle of “effective decision making” in a “noisy” environment, (b) cooperation to complete the task that an individual unit (receptor) is unable to carry out alone with respect to signal amplification, and (c) adaptation and short-term memory (200, 232). These several phenomena connect T cells in a striking way with emerging concepts involving neurons and bacteria. Given the sophisticated simplicity of these principles, it should be more our anticipation than surprise that they may be widely used by many biological systems, including the topic of this review.

APPENDIX 1 To achieve a more uniform use of terms related to TCR ligands and to clarify the process by which these ligands are properly categorized, it is helpful to consider several aspects of T cell responses along with certain features of the peptide antigen presentation system. Simultaneous analysis of multiple distinct responses of cloned T cells following receptor engagement has made clear that the 50% effective dose (ED50) (the apparent ligand concentration needed for half-maximal response) is different for each measured parameter. This property of T cells has been termed hierarchical organization of response activation thresholds (71–76). The differences in the thresholds for some parameters are quite large, ranging over several orders of magnitude, and it is not unusual for those responses with the highest response threshold not to reach plateau within

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

507

the experimentally attainable range of ligand concentrations. With respect to ligand, peptide or protein antigen alone is not the proximate ligand of the TCR. Yet the dose/concentration axis in virtually all experiments denotes how much of these were offered, not the actual number/density of specific peptide:MHC molecule complexes present on the surface of the presenting cells in the culture. Finally, a T cell does not respond to the average concentration of free ligand in the culture but to the specific density of peptide:MHC molecule ligand on the individual presenting cell membrane with which it interacts. Response plateaus in the face of increases in offered antigen can artifactually arise because of the saturation of available MHC molecule binding sites on the presenting cells. These properties of T cells and their ligands have consequences for analyzing and characterizing the properties of specific peptide:MHC molecule combinations. The distinction between “weak” and “partial” agonist is often improperly made because of these factors, yet the difference is great with respect to mechanistic issues. A weak agonist should evoke the same set of intracellular signals in the same proportion as a strong agonist but require more ligand to reach the same absolute signal strength. A partial agonist should be unable to evoke this full set of signals. The consequences of this difference in signaling are that the weak agonist can give rise to all functional responses in the same proportion, the same hierarchical relationship, and with the same plateaus as the strong agonist, provided enough ligand is offered. The altered signaling with true partial agonist will lead to a relative inability to evoke one or more functional responses, or a change in the ratio of two responses or in the plateau response (57). Because most studies begin with function, rather than signaling, misclassification of ligands into these two categories based on biological readouts can make it difficult to perceive consistent patterns of change in intracellular biochemical changes associated with weak versus partial agonists, because ligands in both categories are inappropriately lumped together. To correctly identify a partial agonist at the functional level, the analysis needs to be conducted so that one or more of the criteria of altered response ratios or altered plateaus are met. Because of hierarchial thresholds and the nature of peptide:MHC molecule ligands just discussed, this is not always easily accomplished. First, the specific responses are best compared by normalizing them to the same scale, most often by expressing the results for two different parameters in terms of percent of maximum for each response. How does one know when 100%/maximum is reached? Typically, this is considered to be the response at plateau, i.e. when further ligand addition no longer leads to an increased response. However, unless one independently measures the number of peptide:MHC molecules on the presenting cell population by any of several available techniques [biotinylated peptides (77), monoclonal antibodies specific for peptide:MHC molecule complexes (27, 78–82)], the only way to tell if

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

508

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

further peptide or protein addition still leads to additional useful ligand formation is to see at least one functional response continue to rise. Thus, even the use of normalized responses can be problematic. Ideally, two or more distinct responses will reach plateau while another response continues to rise upon addition of more antigen, indicating that for the former parameters, the plateau is not an artifact of limited ligand generation. One can then examine whether the ratio of the ED50s for these two parameters changes as other ligands for the TCR are tested, indicating the partial agonist nature of a ligand. Otherwise, maximal ligand density may have been reached, arresting the rise in all responses at that relative point in their individual dose-response curves and creating artifactual plateaus at different points. A related point of confusion involves the loss of detectable generation of one response with retention of measurable levels of another as one switches between ligands (66, 75, 83, 84). This presumably qualitative change in T cell behavior is often ascribed to the partial agonist nature of a ligand. However, such a result does not necessarily indicate true differential stimulation of the two responses, if the one that can no longer be detected was less sensitive (had a higher threshold) to begin with (57). Consider a modified ligand that is a weak rather than partial agonist. If each of two assayed responses actually has an equal shift in its ED50, at certain ligand densities the more easily induced parameter will remain above the absolute detection threshold whereas the less sensitive one will go to background. What matters for properly classifying the ligand is whether the relative sensitivities of the two responses have changed. This means that it is essential to ensure that the response that remains measurable reaches a level that would predict detection of the less-sensitive response, based on the ratio of response sensitivities seen with the original, more potent ligand. Failure to see the less-sensitive response under these conditions indicates that these relative thresholds have changed, consistent with partial rather than weak agonist function. If the measurable response cannot be driven to an adequate level to meet this standard, then the distinction between weak versus partial agonist cannot be properly made. Antagonists pose less of a problem. Provided that careful controls have been performed to eliminate (a) nonspecific toxicity of the peptide or protein in use, (b) activation-induced cell death (85) resulting from higher rather than lower ligand potency, and (c) the possibility that the second antigen causes a decrease in stimulatory ligand levels on the presenting cells (67–69), inhibition of an expected response to one ligand by copresentation of a second suffices to classify the latter as a TCR antagonist. It is worth noting that such inhibitory ligands may nevertheless also possess some agonist activity, usually at a much higher concentration (69). Thus the terms partial agonist and antagonist may both be correctly applied to the same ligand, depending on the context.

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING

509

A last problem arises from a reticence of investigators to change their historical characterization of a ligand upon acquiring additional data. The term agonist has been used almost universally to describe any peptide:MHC molecule capable of evoking a functional response from a T cell. With no other information available, this is a correct use of the term, but it should be obvious that further examination of the system may reveal that this ligand is actually a partial agonist or even an antagonist when other parameters of response are measured or other ligands are employed with the same T cell. Some investigators have tested synthetic variants of the peptide involved in forming the proband “agonist” and found ligands of greater potency. The term superagonist has been used to describe these latter ligands (75, 86), and in comparison to the initially analyzed peptide:MHC molecule complex, this seems to make sense. However, this leads to ligand description based on the luck of the draw, i.e. based on which peptide:MHC molecule combination the experimenter happened to use first, rather than on intrinsic properties of the receptor:ligand pair. If the ligand of greater potency and with the ability to evoke a different pattern of functional responses had been analyzed initially, it would have been called the agonist and the less-potent peptide:MHC molecule combination termed a weak agonist or partial agonist. If this latter ligand had originally been used not to directly stimulate the T cell by itself but found in a copresentation assay to inhibit responses to the more potent ligand, it would have been called an antagonist. These scenarios draw attention to the contextual nature of these terms when defined by functional assays selected by the investigator. They are useful only to the extent that the experimental basis for the definition is made clear and the relative nature of the designation is appreciated. In general, it would seem useful to revise one’s classification as more information about a specific TCR’s ligand preference arises, rather than to invent new terms to deal with mislabeling of a ligand because of limited information at early times of analysis. Thus, the discovery of a more potent ligand, and recognition of the limited functional capacity of a ligand previously considered to be an agonist, should lead to the more potent ligand being given this name. The initial, less-potent or incompletely activating ligand should be renamed as a weak or partial agonist in accord with its characteristics, rather than the more potent ligand being called a superagonist. Ultimately, as detailed in the section on Biochemical Fingerprinting, it may prove best to classify ligands based on the signals they evoke from a particular TCR rather than on function, or by a combination of both assessments.

APPENDIX 2 Regarding this issue of why the CD4 and CD8 coreceptors are not stably associated with the TCR, it is valuable to think about whether high- or low-affinity

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

510

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

interactions better support discrimination between two closely related ligands for a given receptor (163). Consider two chemical structures, dinitrophenyl (DNP) and trinitrophenyl (TNP). If our task is to create an ELISA assay that allows clear discrimination between proteins conjugated to either of the two structures, then antibodies with very high affinity for one or the other ligand are unlikely to be satisfactory. The reason for this is that if the absolute affinity is high for the anti-TNP reagent, say 10−11 liters/M, then the affinity for the structurally very similar DNP might still be 10−9 liters/M. This is a big difference in energetic terms, but for an ELISA assay with the usual times of binding and washing before substrate addition, both the TNP- and DNP-containing ligands would be captured and remain largely bound, giving rise to a measurable signal. It would thus be difficult to distinguish two solutions containing TNP-proteins at different concentrations from one containing a TNP-protein and the another a DNP conjugate. If, however, the assay antibodies had an affinity of 10−9 liters/M for TNP and 10−7 liters/M for DNP (the same energetic difference as in the first case), now the absolute off-rate for DNP-protein would be such that the binding and washing reactions would lead to little or no residual signal, while still permitting detection of TNP-proteins present at reasonable concentrations. If one reflects on this scenario, it is clear that it is a form of kinetic proofreading, with the experimentally determined washing time of the ELISA acting as the clock and setting a threshold for signal generation. This example clearly shows that the nature, and especially the rates, of the reactions that can be used as “clocks” determine the absolute receptor affinities that can be employed for effective discrimination among closely related ligands. The slower the off-rate for a reaction (the higher the affinity), the slower any proofreading clock must run to have a chance of discriminating between two different states of receptor occupancy. Were the TCR to have an affinity for either MHC itself, or for many peptide:MHC molecule complexes, approaching those of somatically mutated immunoglobulins for their antigens, the off-rates would be measured in tens of minutes, much too great to be reasonably timed by typical individual biochemical events. Thus, it would be impossible for a TCR to distinguish among peptide-loaded MHC molecules of very similar structure, because even the worst binder would be stable for too long to limit signaling through normal cellular reactions. As affinity decreases (off-rate increases), the times that are relevant to the system begin to approach those more characteristic of enzymatic reactions or protein association kinetics within a cell. Under these conditions, small differences in off-rates begin to overlap with these biochemical events of signaling, and, hence, become subject to kinetic limitation in signal progression related to occupancy times. For this reason, useful TCR bind to their ligands with off-rates in the range of fractions of

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

DYNAMICS OF TCR SIGNALING

511

a second to at most a few seconds at 37◦ C, compatible with biological reaction rates. Much more avid TCR (or stable TCR:coreceptor complexes) would either be constantly driven to signal by the high ambient level of self peptidecontaining MHC molecules on cells or require substantial desensitization of the signaling apparatus that would make the T cell unable to respond effectively to low foreign ligand densities.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ACKNOWLEDGMENTS The authors wish to thank their colleagues and collaborators whose data and intellectual contributions have been indispensable to the evolution of the ideas presented in this review. We especially wish to recognize W Biddison, B Hemmer, M Vergelli, R Martin, B Dittel, and CA Janeway Jr; our ongoing studies with these investigators have played a key role in the development of the concepts of feedback control and signal spreading as applied to TCR recognition that we emphasize here. We recognize that not all contributions have been fully documented, and for any such omissions, we apologize and hope our scientific colleagues will understand the constraints of the format. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Davis MM, Bjorkman PJ. 1988. T-cell antigen receptor genes and T-cell recognition. Nature 334:395–402 2. Ito K, Van KL, Bonneville M, Hsu S, Murphy DB, Tonegawa S. 1990. Recognition of the product of a novel MHC TL region gene (27b) by a mouse γ δ T cell receptor. Cell 62:549–61 3. O’Brien RL, Fu YX, Cranfill R, Dallas A, Ellis C, Reardon C, Lang J, Carding SR, Kubo R, Born W. 1992. Heat shock protein Hsp60-reactive γ δ-cells— A large, diversified T lymphocyte subset with highly focused specificity. Proc. Natl. Acad. Sci. USA 89:4348–52 4. Schild H, Mavaddat N, Litzenberger C, Ehrich EW, Davis MM, Bluestone JA, Matis L, Draper RK, Chien YH. 1994. The nature of major histocompatibility complex recognition by γ δ T cells. Cell 76:29–37 5. Germain RN. 1994. MHC-dependent antigen processing and peptide presentation: providing ligands for T lymphocyte activation. Cell 76:287–99 6. Beckman EM, Porcelli SA, Morita CT, Behar SM, Furlong ST, Brenner MB.

7.

8.

9.

10.

11.

1994. Recognition of a lipid antigen by CD1-restricted αβ+ T cells. Nature 372:691–94 Sieling PA, Chatterjee D, Porcelli SA, Prigozy TI, Mazzaccaro RJ, Soriano T, Bloom BR, Brenner MB, Kronenberg M, Brennan PJ, Modlin RL. 1995. CD1restricted T cell recognition of microbial lipoglycan antigens. Science 269:227–30 Prigozy TI, Sieling PA, Clemens D, Stewart PL, Behar SM, Porcelli SA, Brenner MB, Modlin RL, Kronenberg M. 1997. The mannose receptor delivers lipoglycan antigens to endosomes for presentation to T cells by CD1b molecules. Immunity 6:187–97 Stern LJ, Wiley DC. 1994. Antigenic peptide binding by class I and class II histocompatibility proteins. Structure 2:245– 51 Brown JH, Jardetzky TS, Gorga JC, Stern LJ, Urban RG, Strominger JL, Wiley DC. 1993. Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1. Nature 364:33–39 Bjorkman PJ, Saper MA, Samraoui B, Bennett WS, Strominger JL, Wiley DC.

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

512

12.

Annu. Rev. Immunol. 1999.17:467-522. 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.

23.

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A 1987. The foreign antigen binding site and T cell recognition regions of class I histocompatibility antigens. Nature 329:512– 18 Garrett TP, Saper MA, Bjorkman PJ, Strominger JL, Wiley DC. 1989. Specificity pockets for the side chains of peptide antigens in HLA-Aw68. Nature 342:692–96 Matsumura M, Fremont DH, Peterson PA, Wilson IA. 1992. Emerging principles for the recognition of peptide antigens by MHC class I molecules. Science 257:927– 34 Falk K, R¨otzschke O, Stevanovic S, Jung G, Rammensee HG. 1991. Allele-specific motifs revealed by sequencing of selfpeptides eluted from MHC molecules. Nature 351:290–96 Falk K, R¨otzschke O, Stevanovic S, Jung G, Rammensee HG. 1994. Pool sequencing of natural HLA-DR, DQ, and DP ligands reveals detailed peptide motifs, constraints of processing, and general rules. Immunogenetics 39:230–42 Hammer J, Bono E, Gallazzi F, Belunis C, Nagy Z, Sinigaglia F. 1994. Precise prediction of major histocompatibility complex class II-peptide interaction based on peptide side chain scanning. J. Exp. Med. 180:2353–58 Sinigaglia F, Hammer J. 1995. Motifs and supermotifs for MHC class II binding peptides. J. Exp. Med. 181:449–51 Cerundolo V, Elliott T, Elvin J, Bastin J, Rammensee HG, Townsend A. 1991. The binding affinity and dissociation rates of peptides for class I major histocompatibility complex molecules. Eur. J. Immunol. 21:2069–75 Yewdell JW, Bennink JR. 1999. Immunodominance in MHC class I-restricted T lymphocyte responses. Annu. Rev. Immunol. 17: In press Wang CR, Fischer-Lindahl K, Deisenhofer J. 1996. Crystal structure of the MHC class Ib molecule H2-M3. Res. Immunol. 147:313–21 Rudensky A, Preston HP, Hong SC, Barlow A, Janeway CA Jr. 1991. Sequence analysis of peptides bound to MHC class II molecules. Nature 353:622–27 Chicz RM, Urban RG, Lane WS, Gorga JC, Stern LJ, Vignali DA, Strominger JL. 1992. Predominant naturally processed peptides bound to HLA-DR1 are derived from MHC-related molecules and are heterogeneous in size. Nature 358:764–68 Hunt DF, Michel H, Dickinson TA, Shabanowitz J, Cox AL, Sakaguchi K, Appella E, Grey HM, Sette A. 1992. Peptides presented to the immune system by

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

the murine class II major histocompatibility complex molecule I-Ad. Science 256:1817–20 Hunt DF, Henderson RA, Shabanowitz J, Sakaguchi K, Michel H, Sevilir N, Cox AL, Appella E, Engelhard VH. 1992. Characterization of peptides bound to the class I MHC molecule HLA-A2.1 by mass spectrometry. Science 255:1261–63 R¨otzschke O, Falk K, Deres K, Schild H, Norda M, Metzger J, Jung G, Rammensee HG. 1990. Isolation and analysis of naturally processed viral peptides as recognized by cytotoxic T cells. Nature 348:252–54 Van Bleek GM, Nathenson SG. 1990. Isolation of an endogenously processed immunodominant viral peptide from the class I H-2Kb molecule. Nature 348:213– 16 Porgador A, Yewdell JW, Deng Y, Bennink JR, Germain RN. 1997. Localization, quantitation, and in situ detection of specific peptide-MHC class I complexes using a monoclonal antibody. Immunity 6:715–26 Anton LC, Yewdell JW, Bennink JR. 1997. MHC class I-associated peptides produced from endogenous gene products with vastly different efficiencies. J. Immunol. 158:2535–42 Demotz S, Grey HM, Sette A. 1990. The minimal number of class II MHC-antigen complexes needed for T cell activation. Science 249:1028–30 Harding CV, Unanue ER. 1990. Quantitation of antigen-presenting cell MHC class II/peptide complexes necessary for T-cell stimulation. Nature 346:574–76 Christinck ER, Luscher MA, Barber BH, Williams DB. 1991. Peptide binding to class I MHC on living cells and quantitation of complexes required for CTL lysis. Nature 352:67–70 Sykulev Y, Joo M, Vturina I, Tsomides TJ, Eisen HN. 1996. Evidence that a single peptide-MHC complex on a target cell can elicit a cytolytic T cell response. Immunity 4:565–71 Schodin BA, Tsomides TJ, Kranz DM. 1996. Correlation between the number of T cell receptors required for T cell activation and TCR-ligand affinity. Immunity 5:137–46 Madden DR, Garboczi DN, Wiley DC. 1993. The antigenic identity of peptideMHC complexes: a comparison of the conformations of five viral peptides presented by HLA-A2. Cell 75:693–708 Fremont DH, Hendrickson WA, Marrack P, Kappler J. 1996. Structures of an MHC

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

DYNAMICS OF TCR SIGNALING

36.

37.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

38.

39.

40.

41.

42.

43.

44. 45. 46.

47.

48.

class II molecule with covalently bound single peptides. Science 272:1001–4 Fremont DH, Monnaie D, Nelson CA, Hendrickson WA, Unanue ER. 1998. Crystal structure of I-Ak in complex with a dominant epitope of lysozyme. Immunity 8:305–17 Fremont DH, Matsumura M, Stura EA, Peterson PA, Wilson IA. 1992. Crystal structures of two viral peptides in complex with murine MHC class I H-2Kb. Science 257:919–27 Garcia KC, Degano M, Stanfield RL, Brunmark A, Jackson MR, Peterson PA, Teyton L, Wilson IA. 1996. An αβ T cell receptor structure at 2.5 A and its orientation in the TCR-MHC complex. Science 274:209–19 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 Speir JA, Garcia KC, Brunmark A, Degano M, Peterson PA, Teyton L, Wilson IA. 1998. Structural basis of 2C TCR allorecognition of H-2Ld peptide complexes. Immunity 8:553–62 Garcia KC, Degano M, Pease LR, Huang M, Peterson PA, Teyton L, Wilson IA. 1998. Structural basis of plasticity in T cell receptor recognition of a self peptideMHC antigen. Science 279:1166–72 Ding YH, Smith KJ, Garboczi DN, Utz U, Biddison WE, Wiley DC. 1998. Two human T cell receptors bind in a similar diagonal mode to the HLA-A2/Tax peptide complex using different TCR amino acids. Immunity 8:403–11 Teng MK, Smolyar A, Tse AG, Liu JH, Liu J, Hussey RE, Nathenson SG, Chang HC, Reinherz EL, Wang JH. 1998. Identification of a common docking topology with substantial variation among different TCR-peptide-MHC complexes. Curr. Biol. 8:409–12 Robey E, Fowlkes BJ. 1994. Selective events in T cell development. Annu. Rev. Immunol. 12:675–705 Germain RN. 1990. Making a molecular match. Nature 344:19–22 Germain RN. 1993. Antigen processing and presentation. In Fundamental Immunology, ed. WE Paul, pp. 629–76. New York: Raven. 3rd ed. Zerrahn J, Held W, Raulet DH. 1997. The MHC reactivity of the T cell repertoire prior to positive and negative selection. Cell 88:627–36 Merkenschlager M, Graf D, Lovatt M, Bommhardt U, Zamoyska R, Fisher AG.

49. 50.

51.

52.

53.

54.

55.

56.

57.

58. 59.

60.

61.

513

1997. How many thymocytes audition for selection? J. Exp. Med. 186:1149–58 Jerne NK. 1971. The somatic generation of immune recognition. Eur. J. Immunol. 1:1–9 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 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 Brocker T. 1997. Survival of mature CD4 T lymphocytes is dependent on major histocompatibility complex class II-expressing dendritic cells. J. Exp. Med. 186:1223–32 Rooke R, Waltzinger C, Benoist C, Mathis D. 1997. Targeted complementation of MHC class II deficiency by intrathymic delivery of recombinant adenoviruses. Immunity 7:123–34 Markiewicz MA, Girao C, Opferman JT, Sun J, Hu Q, Agulnik AA, Bishop CE, Thompson CB, Ashton-Rickardt PG. 1998. Long-term T cell memory requires the surface expression of selfpeptide/major histocompatibility complex molecules. Proc. Natl. Acad. Sci. USA 95:3065–70 Bergman RN, Hechter O. 1978. Neurohypophyseal hormone-responsive renal adenylate cyclase. IV. A random-hit matrix model for coupling in a hormonesensitive adenylate cyclase system. J. Biol. Chem. 253:3238–50 McKeithan TW. 1995. Kinetic proofreading in T-cell receptor signal transduction. Proc. Natl. Acad. Sci. USA 92:5042– 46 Germain RN, Levine EH, Madrenas J. 1995. The T-cell receptor as a diverse signal transduction machine. Immunologist 3/4:113–21 Jameson SC, Bevan MJ. 1995. T cell receptor antagonists and partial agonists. Immunity 2:1–11 Sloan-Lancaster J, Allen PM. 1996. Altered peptide ligand-induced partial T cell activation: molecular mechanisms and role in T cell biology. Annu. Rev. Immunol. 14:1–27 Madrenas J, Germain RN. 1996. Variant TCR ligands: New insights into the molecular basis of antigen-dependent signal transduction and T cell activation. Semin. Immunol. 8:83–101 Kersh GJ, Allen PM. 1996. Essential

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

514

62.

63.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

64.

65.

66.

67.

68.

69.

70. 71.

72.

73.

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A flexibility in the T-cell recognition of antigen. Nature 380:495–98 Mannie MD. 1991. A unified model for T cell antigen recognition and thymic selection of the T cell repertoire. J. Theor. Biol. 151:169–92 Janeway CA Jr. 1995. Ligands for the T-cell receptor: hard times for avidity models. Immunol. Today 16:223–25 Evavold BD, Sloan-Lancaster J, Wilson KJ, Rothbard JB, Allen PM. 1995. Specific T cell recognition of minimally homologous peptides: evidence for multiple endogenous ligands. Immunity 2:655–63 Williams O, Tarazona R, Wack A, Harker N, Roderick K, Kioussis D. 1998. Interactions with multiple peptide ligands determine the fate of developing thymocytes. Proc. Natl. Acad. Sci. USA 95:5706–11 Evavold BD, Allen PM. 1991. Separation of IL-4 production from Th cell proliferation by an altered T cell receptor ligand. Science 252:1308–10 De Magistris MT, Alexander J, Coggeshall M, Altman A, Gaeta FC, Grey HM, Sette A. 1992. Antigen analog-major histocompatibility complexes act as antagonists of the T cell receptor. Cell 68:625–34 Racioppi L, Ronchese F, Matis LA, Germain RN. 1993. Peptide-major histocompatibility complex class II complexes with mixed agonist/antagonist properties provide evidence for ligand-related differences in T cell receptor-dependent intracellular signaling. J. Exp. Med. 177:1047– 60 Jameson SC, Carbone FR, Bevan MJ. 1993. Clone-specific T cell receptor antagonists of major histocompatibility complex class I-restricted cytotoxic T cells. J. Exp. Med. 177:1541–50 Marx J. 1995. The T cell receptor begins to reveal its many facets. Science 267:459–60 Valitutti S, M¨uller S, Dessing M, Lanzavecchia A. 1996. Different responses are elicited in cytotoxic T lymphocytes by different levels of T cell receptor occupancy. J. Exp. Med. 183:1917–21 Itoh Y, Germain RN. 1997. Single cell analysis reveals regulated hierarchical T cell antigen receptor signaling thresholds and intraclonal heterogeneity for individual cytokine responses of CD4+ T cells. J. Exp. Med. 186:757–66 Bachmann MF, Oxenius A, Speiser DE, Mariathasan S, Hengartner H, Zinkernagel RM, Ohashi PS. 1997. Peptideinduced T cell receptor down-regulation on naive T cells predicts agonist/partial agonist properties and strictly correlates

74.

75.

76.

77.

78. 79.

80.

81.

82.

83.

with T cell activation. Eur. J. Immunol. 27:2195–203 Carballido JM, Faith A, Carballido-Perrig N, Blaser K. 1997. The intensity of T cell receptor engagement determines the cytokine pattern of human allergen-specific T helper cells. Eur. J. Immunol. 27:515– 21 Nicholson LB, Waldner H, Carrizosa AM, Sette A, Collins M, Kuchroo VK. 1998. Heteroclitic proliferative responses and changes in cytokine profile induced by altered peptides: implications for autoimmunity. Proc. Natl. Acad. Sci. USA 95:264–69 Hemmer B, Stefanova I, Vergelli M, Germain RN, Martin R. 1998. Relationships among TCR ligand potency, thresholds for effector function elicitation, and the quality of early signaling events in human T cells. J. Immunol. 160:5807–14 Busch R, Strang G, Howland K, Rothbard JB. 1990. Degenerate binding of immunogenic peptides to HLA-DR proteins on B cell surfaces. Int. Immunol. 2:443–51 Rudensky A, Rath S, Preston HP, Murphy DB, Janeway CA Jr. 1991. On the complexity of self. Nature 353:660–62 Murphy DB, Rath S, Pizzo E, Rudensky AY, George A, Larson JK, Janeway CA Jr. 1992. Monoclonal antibody detection of a major self peptide—MHC class-II complex. J. Immunol. 148:3483–91 Zhong G, Reis e Sousa C, Germain RN. 1997. Antigen-unspecific B cells and lymphoid dendritic cells both show extensive surface expression of processed antigenmajor histocompatibility complex class II complexes after soluble protein exposure in vivo or in vitro. J. Exp. Med. 186:673– 82 Zhong G, Reis e Sousa C, Germain RN. 1997. Production, specificity, and functionality of monoclonal antibodies to specific peptide:MHC class II complexes formed by active processing of exogenous protein. Proc. Natl. Acad. Sci. USA 94:13856–61 Dadaglio G, Nelson CA, Deck MB, Petzold SJ, Unanue ER. 1997. Characterization and quantitation of peptideMHC complexes produced from hen egg lysozyme using a monoclonal antibody. Immunity 6:727–38 Ikagawa S, Matsushita S, Chen YZ, Ishikawa T, Nishimura Y. 1996. Single amino acid substitutions on a Japanese cedar pollen allergen (Cry j 1)-derived peptide induced alterations in human T cell responses and T cell receptor antagonism. J. Allergy Clin. Immunol. 97:53–64

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DYNAMICS OF TCR SIGNALING 84. Combadiere B, Reis e Sousa C, Germain RN, Lenardo MJ. 1998. Selective induction of apoptosis in mature T lymphocytes by variant T cell receptor ligands. J. Exp. Med. 187:349–55 85. Lenardo MJ. 1991. Interleukin-2 programs mouse αβ T lymphocytes for apoptosis. Nature 353:858–61 86. Vergelli M, Hemmer B, Kalbus M, Vogt AB, Ling N, Conlon P, Coligan JE, McFarland H, Martin R. 1997. Modifications of peptide ligands enhancing T cell responsiveness imply large numbers of stimulatory ligands for autoreactive T cells. J. Immunol. 158:3746–52 87. Corr M, Slanetz AE, Boyd LF, Jelonek MT, Khilko S, al-Ramadi BK, Kim YS, Maher SE, Bothwell AL, Margulies DH. 1994. T cell receptor-MHC class I peptide interactions: affinity, kinetics, and specificity. Science 265:946–49 88. Matsui K, Boniface JJ, Steffner P, Reay PA, Davis MM. 1994. Kinetics of T-cell receptor binding to peptide/I-Ek complexes: correlation of the dissociation rate with T-cell responsiveness. Proc. Natl. Acad. Sci. USA 91:12862–66 89. Seth A, Stern LJ, Ottenhoff TH, Engel I, Owen MJ, Lamb JR, Klausner RD, Wiley DC. 1994. Binary and ternary complexes between T-cell receptor, class II MHC and superantigen in vitro. Nature 369:324– 27 90. al-Ramadi BK, Jelonek MT, Boyd LF, Margulies DH, Bothwell AL. 1995. Lack of strict correlation of functional sensitization with the apparent affinity of MHC/peptide complexes for the TCR. J. Immunol. 155:662–73 91. Garcia KC, Scott CA, Brunmark A, Carbone FR, Peterson PA, Wilson IA, Teyton L. 1996. CD8 enhances formation of stable T-cell receptor/MHC class I molecule complexes. Nature 384:577–81 92. Alam SM, Travers PJ, Wung JL, Nasholds W, Redpath S, Jameson SC, Gascoigne NR. 1996. T-cell-receptor affinity and thymocyte positive selection. Nature 381:616–20 93. Seibel JL, Wilson N, Kozono H, Marrack P, Kappler JW. 1997. Influence of the NH2-terminal amino acid of the T cell receptor alpha chain on major histocompatibility complex (MHC) class II + peptide recognition. J. Exp. Med. 185:1919– 27 94. Matsui K, Boniface JJ, Reay PA, Schild H, Fazekas de St. Groth B, Davis MM. 1991. Low affinity interaction of peptide-MHC complexes with T cell receptors. Science 254:1788–91

515

95. Weber S, Traunecker A, Oliveri F, Gerhard W, Karjalainen K. 1992. Specific low-affinity recognition of major histocompatibility complex plus peptide by soluble T-cell receptor. Nature 356:793– 96 96. Sykulev Y, Brunmark A, Jackson M, Cohen RJ, Peterson PA, Eisen HN. 1994. Kinetics and affinity of reactions between an antigen-specific T cell receptor and peptide-MHC complexes. Immunity 1:15–22 97. Sykulev Y, Brunmark A, Tsomides TJ, Kageyama S, Jackson M, Peterson PA, Eisen HN. 1994. High-affinity reactions between antigen-specific T-cell receptors and peptides associated with allogeneic and syngeneic major histocompatibility complex class I proteins. Proc. Natl. Acad. Sci. USA 91:11487–91 98. Luescher IF, Vivier E, Layer A, Mahiou J, Godeau F, Malissen B, Romero P. 1995. CD8 modulation of T-cell antigen receptor-ligand interactions on living cytotoxic T lymphocytes. Nature 373:353– 56 99. Kessler BM, Bassanini P, Cerottini JC, Luescher IF. 1997. Effects of epitope modification on T cell receptor-ligand binding and antigen recognition by seven H-2Kd-restricted cytotoxic T lymphocyte clones specific for a photoreactive peptide derivative. J. Exp. Med. 185:629–40 100. Margulies DH. 1997. Interactions of TCRs with MHC-peptide complexes: a quantitative basis for mechanistic models. Curr. Opin. Immunol. 9:390–95 101. Davis MM, Boniface JJ, Reich Z, Lyons D, Hampl J, Arden B, Chien Y. 1998. Ligand recognition by αβ T cell receptors. Annu. Rev. Immunol. 16:523–44 102. Andersen PS, Stryhn A, Hansen BE, Fugger L, Engberg J, Buus S. 1996. A recombinant antibody with the antigen-specific, major histocompatibility complex-restricted specificity of T cells. Proc. Natl. Acad. Sci. USA 93:1820–24 103. Tan L, Turner J, Weiss A. 1991. Regions of the T cell receptor α and β chains that are responsible for interactions with CD3. J. Exp. Med. 173:1247–56 104. Cosson P, Bonifacino JS. 1992. Role of transmembrane domain interactions in the assembly of class II MHC molecules. Science 258:659–62 105. Rutledge T, Cosson P, Manolios N, Bonifacino JS, Klausner RD. 1992. Transmembrane helical interactions—ζ -chain dimerization and functional association with the T-cell antigen receptor. EMBO J. 11:3245–54

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

516

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

106. Campbell KS, Backstrom BT, Tiefenthaler G, Palmer E. 1994. CART: a conserved antigen receptor transmembrane motif. Semin. Immunol. 6:393–410 107. Dietrich J, Neisig A, Hou X, Wegener AM, Gajhede M, Geisler C. 1996. Role of CD3γ in T cell receptor assembly. J. Cell Biol. 132:299–310 108. Backstrom BT, Milia E, Peter A, Jaureguiberry B, Baldari CT, Palmer E. 1996. A motif within the T cell receptor α chain constant region. Immunity 5:437–47 109. Ghendler Y, Smolyar A, Chang HC, Reinherz EL. 1998. One of the CD3ε subunits within a T cell receptor complex lies in close proximity to the Cβ FG loop. J. Exp. Med. 187:1529–36 110. Shaw AS, Dustin ML. 1997. Making the T cell receptor go the distance: a topological view of T cell activation. Immunity 6:361–69 111. Jelonek MT, Classon BJ, Hudson PJ, Margulies DH. 1998. Direct binding of the MHC class I molecule H-2Ld to CD8: interaction with the amino terminus of a mature cell surface protein. J. Immunol. 160:2809–14 112. Janeway CA Jr. 1991. The co-receptor function of CD4. Semin. Immunol. 3:153– 60 113. Lyons DS, Lieberman SA, Hampl J, Boniface JJ, Chien Y, Berg LJ, Davis MM. 1996. A TCR binds to antagonist ligands with lower affinities and faster dissociation rates than to agonists. Immunity 5:53– 61 114. Kent UM, Mao SY, Wofsy C, Goldstein B, Ross S, Metzger H. 1994. Dynamics of signal transduction after aggregation of cell-surface receptors: studies on the type I receptor for IgE. Proc. Natl. Acad. Sci. USA 91:3087–91 115. Alberts B, Bray D, Lewis J, Raff M, Roberts K, Watson JD. 1994. The fidelity of protein synthesis is improved by two proofreading mechanisms. In Molecular Biology of the Cell, ed. B Alberts, D Bray, J Lewis, et al, pp. 239–41. New York/London: Garland 116. Rabinowitz JD, Beeson C, Lyons DS, Davis MM, McConnell HM. 1996. Kinetic discrimination in T-cell activation. Proc. Natl. Acad. Sci. USA 93:1401– 5 117. Chan AC, Desai DM, Weiss A. 1994. The role of protein tyrosine kinases and protein tyrosine phosphatases in T cell antigen receptor signal transduction. Annu. Rev. Immunol. 12:555–92 118. Weiss A, Iwashima M, Irving B, van Oers NS, Kadlecek TA, Straus D, Chan A.

119.

120.

121.

122.

123.

124.

125. 126. 127.

128.

129.

130.

131.

1994. Molecular and genetic insights into T cell antigen receptor signal transduction. Adv. Exp. Med. Biol. 365:53–62 Zhao H, Koretzky GA. 1997. Regulation of signal transduction through the T cell antigen receptor. J. Lab. Clin. Med. 130:126–31 Samelson LE, Patel MD, Weissman AM, Harford JB, Klausner RD. 1986. Antigen activation of murine T cells induces tyrosine phosphorylation of a polypeptide associated with the T cell antigen receptor. Cell 46:1083–90 June CH, Fletcher MC, Ledbetter JA, Samelson LE. 1990. Increases in tyrosine phosphorylation are detectable before phospholipase C activation after T cell receptor stimulation. J. Immunol. 144:1591–99 Straus DB, Weiss A. 1992. Genetic evidence for the involvement of the lck tyrosine kinase in signal transduction through the T cell antigen receptor. Cell 70:585– 93 Samelson LE, Phillips AF, Luong ET, Klausner RD. 1990. Association of the fyn protein-tyrosine kinase with the T-cell antigen receptor. Proc. Natl. Acad. Sci. USA 87:4358–62 Iwashima M, Irving BA, van Oers NS, Chan AC, Weiss A. 1994. Sequential interactions of the TCR with two distinct cytoplasmic tyrosine kinases. Science 263:1136–39 Wange RL, Samelson LE. 1996. Complex complexes: signaling at the TCR. Immunity 5:197–205 Reth M. 1992. Antigen receptors on B lymphocytes. Annu. Rev. Immunol. 10:97–121 Isakov N. 1997. ITIMs and ITAMs. The Yin and Yang of antigen and Fc receptorlinked signaling machinery. Immunol. Res. 16:85–100 Kersh EN, Shaw AS, Allen PM. 1998. Fidelity of T cell activation through multistep T cell receptor ζ phosphorylation. Science 281:572–75 Ravichandran KS, Lee KK, Songyang Z, Cantley LC, Burn P, Burakoff SJ. 1993. Interaction of Shc with the ζ chain of the T cell receptor upon T cell activation. Science 262:902–5 Osman N, Lucas SC, Turner H, Cantrell D. 1995. A comparison of the interaction of Shc and the tyrosine kinase ZAP-70 with the T cell antigen receptor zeta chain tyrosine-based activation motif. J. Biol. Chem. 270:13981–86 Zenner G, Vorherr T, Mustelin T, Burn P. 1996. Differential and multiple binding

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

DYNAMICS OF TCR SIGNALING

132.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

133.

134.

135.

136.

137.

138.

139.

140.

141.

of signal transducing molecules to the ITAMs of the TCR-ζ chain. J. Cell Biochem. 63:94–103 Koyasu S, Tse AG, Moingeon P, Hussey RE, Mildonian A, Hannisian J, Clayton LK, Reinherz EL. 1994. Delineation of a T-cell activation motif required for binding of protein tyrosine kinases containing tandem SH2 domains. Proc. Natl. Acad. Sci. USA 91:6693–97 Gauen LK, Zhu Y, Letourneur F, Hu Q, Bolen JB, Matis LA, Klausner RD, Shaw AS. 1994. Interactions of p59fyn and ZAP-70 with T-cell receptor activation motifs: defining the nature of a signalling motif. Mol. Cell Biol. 14:3729–41 Isakov N, Wange RL, Burgess WH, Watts JD, Aebersold R, Samelson LE. 1995. ZAP-70 binding specificity to T cell receptor tyrosine-based activation motifs: the tandem SH2 domains of ZAP-70 bind distinct tyrosine-based activation motifs with varying affinity. J. Exp. Med. 181:375–80 Bu JY, Shaw AS, Chan AC. 1995. Analysis of the interaction of ZAP-70 and syk protein-tyrosine kinases with the T-cell antigen receptor by plasmon resonance. Proc. Natl. Acad. Sci. USA 92:5106–10 Hatada MH, Lu X, Laird ER, Green J, Morgenstern JP, Lou M, Marr CS, Phillips TB, Ram MK, Theriault K, et al. 1995. Molecular basis for interaction of the protein tyrosine kinase ZAP-70 with the Tcell receptor. Nature 377:32–38 Couture C, Baier G, Altman A, Mustelin T. 1994. p56lck-independent activation and tyrosine phosphorylation of p72syk by T-cell antigen receptor/CD3 stimulation. Proc. Natl. Acad. Sci. USA 91:5301– 5 Chan AC, Dalton M, Johnson R, Kong GH, Wang T, Thoma R, Kurosaki T. 1995. Activation of ZAP-70 kinase activity by phosphorylation of tyrosine 493 is required for lymphocyte antigen receptor function. EMBO J. 14:2499–508 Wange RL, Guitian R, Isakov N, Watts JD, Aebersold R, Samelson LE. 1995. Activating and inhibitory mutations in adjacent tyrosines in the kinase domain of ZAP-70. J. Biol. Chem. 270:18730–33 Kong G, Dalton M, Wardenburg JB, Straus D, Kurosaki T, Chan AC. 1996. Distinct tyrosine phosphorylation sites in ZAP-70 mediate activation and negative regulation of antigen receptor function. Mol. Cell Biol. 16:5026–35 Zhang W, Sloan-Lancaster J, Kitchen J, Trible RP, Samelson LE. 1998. LAT: the ZAP-70 tyrosine kinase substrate that

142.

143. 144.

145.

146.

147.

148.

149. 150.

151. 152. 153.

154.

517

links T cell receptor to cellular activation. Cell 92:83–92 Bruyns E, Marie-Cardine A, Kirchgessner H, Sagolla K, Shevchenko A, Mann M, Autschbach F, Bensussan A, Meuer S, Schraven B. 1998. T cell receptor (TCR) interacting molecule (TRIM), a novel disulfide-linked dimer associated with the TCR-CD3-ζ complex, recruits intracellular signaling proteins to the plasma membrane. J. Exp. Med. 188:561–75 Pawson T, Scott JD. 1997. Signaling through scaffold, anchoring, and adaptor proteins. Science 278:2075–80 Jackman JK, Motto DG, Sun Q, Tanemoto M, Turck CW, Peltz GA, Koretzky GA, Findell PR. 1995. Molecular cloning of SLP-76, a 76-kDa tyrosine phosphoprotein associated with Grb2 in T cells. J. Biol. Chem. 270:7029–32 Wu J, Motto DG, Koretzky GA, Weiss A. 1996. Vav and SLP-76 interact and functionally cooperate in IL-2 gene activation. Immunity 4:593–602 Raab M, da Silva AJ, Findell PR, Rudd CE. 1997. Regulation of Vav-SLP-76 binding by ZAP-70 and its relevance to TCR. Immunity 6:155–64 Yablonski D, Kuhne MR, Kadlecek T, Weiss A. 1998. Uncoupling of nonreceptor tyrosine kinases from PLC-γ 1 in an SLP-76-deficient T cell. Science 281:413–16 Weiss A, Imboden JB. 1987. Cell surface molecules and early events involved in human T lymphocyte activation. Adv. Immunol. 41:1–38 Lewis RS, Cahalan MD. 1995. Potassium and calcium channels in lymphocytes. Annu Rev. Immunol. 13:623–53 Crabtree GR, Clipstone NA. 1994. Signal transmission between the plasma membrane and nucleus of T lymphocytes. Annu. Rev. Biochem. 63:1045–83 Jain J, Loh C, Rao A. 1995. Transcriptional regulation of the IL-2 gene. Curr. Opin. Immunol. 7:333–42 Karin M, Liu Z, Zandi E. 1997. AP-1 function and regulation. Curr. Opin. Cell Biol. 9:240–46 Ullman KS, Northrop JP, Verweij CL, Crabtree GR. 1990. Transmission of signals from the T lymphocyte antigen receptor to the genes responsible for cell proliferation and immune function: the missing link. Annu. Rev. Immunol. 8:421–52 Rabinowitz JD, Beeson C, Wulfing C, Tate K, Allen PM, Davis MM, McConnell HM. 1996. Altered T cell receptor ligands trigger a subset of early T cell signals. Immunity 5:125–35

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

518

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A

155. Beeson C, Rabinowitz J, Tate K, Gutgemann I, Chien YH, Jones PP, Davis MM, McConnell HM. 1996. Early biochemical signals arise from low affinity TCRligand reactions at the cell-cell interface. J. Exp. Med. 184:777–82 156. Sloan-Lancaster J, Steinberg TH, Allen PM. 1996. Selective activation of the calcium signaling pathway by altered peptide ligands. J. Exp. Med. 184:1525–30 157. Wulfing C, Rabinowitz JD, Beeson C, Sjaastad MD, McConnell HM, Davis MM. 1997. Kinetics and extent of T cell activation as measured with the calcium signal. J. Exp. Med. 185:1815–25 158. Chau LA, Bluestone JA, Madrenas J. 1998. Dissociation of intracellular signaling pathways in response to partial agonist ligands of the T cell receptor. J. Exp. Med. 187:1699–709 159. Sloan-Lancaster J, Shaw AS, Rothbard JB, Allen PM. 1994. Partial T cell signaling: altered phospho-ζ and lack of ZAP70 recruitment in APL-induced T cell anergy. Cell 79:913–22 160. Madrenas J, Wange RL, Wang JL, Isakov N, Samelson LE, Germain RN. 1995. ζ phosphorylation without ZAP-70 activation induced by TCR antagonists or partial agonists. Science 267:515–18 161. Reis e Sousa C, Levine EH, Germain RN. 1996. Partial signaling by CD8+ T cells in response to antagonist ligands. J. Exp. Med. 184:149–57 162. La Face DM, Couture C, Anderson K, Shih G, Alexander J, Sette A, Mustelin T, Altman A, Grey HM. 1997. Differential T cell signaling induced by antagonist peptide-MHC complexes and the associated phenotypic responses. J. Immunol. 158:2057–64 163. K¨onig R, Fleury S, Germain RN. 1996. The structural basis of CD4-MHC class II interactions: coreceptor contributions to T cell receptor antigen recognition and oligomerization-dependent signal transduction. Curr. Top. Microbiol. Immunol. 205:19–46 164. Madrenas J, Chau LA, Smith J, Bluestone JA, Germain RN. 1997. The efficiency of CD4 recruitment to ligand-engaged TCR controls the agonist/partial agonist properties of peptide-MHC molecule ligands. J. Exp. Med. 185:219–30 165. Hampl J, Chien Y-H, Davis MM. 1997. CD4 augments the response of a T cell to agonist but not to antagonist ligands. Immunity 7:379–85 166. Jameson SC, Hogquist KA, Bevan MJ. 1994. Specificity and flexibility in thymic selection. Nature 369:750–52

167. Mannie MD, Rosser JM, White GA. 1995. Autologous rat myelin basic protein is a partial agonist that is converted into a full antagonist upon blockade of CD4. Evidence for the integration of efficacious and nonefficacious signals during T cell antigen recognition. J. Immunol. 154:2642–54 168. Vidal K, Hsu BL, Williams CB, Allen PM. 1996. Endogenous altered peptide ligands can affect peripheral T cell responses. J. Exp. Med. 183:1311–21 169. Viola A, Lanzavecchia A. 1996. T cell activation determined by T cell receptor number and tunable thresholds. Science 273:104–6 170. Kim DT, Rothbard JB, Bloom DD, Fathman CG. 1996. Quantitative analysis of T cell activation: role of TCR/ligand density and TCR affinity. J. Immunol. 156:2737–42 171. Bergman M, Mustelin T, Oetken C, Partanen J, Flint NA, Amrein KE, Autero M, Burn P, Alitalo K. 1992. The human p50csk tyrosine kinase phosphorylates p56lck at Tyr-505 and down regulates its catalytic activity. EMBO J. 11:2919–24 172. Sondhi D, Xu W, Songyang Z, Eck MJ, Cole PA. 1998. Peptide and protein phosphorylation by protein tyrosine kinase Csk: insights into specificity and mechanism. Biochemistry 37:165–72 173. Bougeret C, Delaunay T, Romero F, Jullien P, Sabe H, Hanafusa H, Benarous R, Fischer S. 1996. Detection of a physical and functional interaction between Csk and Lck which involves the SH2 domain of Csk and is mediated by autophosphorylation of Lck on tyrosine 394. J. Biol. Chem. 271:7465–72 174. Chow LM, Fournel M, Davidson D, Veillette A. 1993. Negative regulation of T-cell receptor signalling by tyrosine protein kinase p50csk. Nature 365:156– 60 175. Mustelin T, Williams S, Tailor P, Couture C, Zenner G, Burn P, Ashwell JD, Altman A. 1995. Regulation of the p70zap tyrosine protein kinase in T cells by the CD45 phosphotyrosine phosphatase. Eur. J. Immunol. 25:942–46 176. Furukawa T, Itoh M, Krueger NX, Streuli M, Saito H. 1994. Specific interaction of the CD45 protein-tyrosine phosphatase with tyrosine-phosphorylated CD3ζ chain. Proc. Natl. Acad. Sci. USA 91:10928–32 177. Murphy MA, Schnall RG, Venter DJ, Barnett L, Bertoncello I, Thien CB, Langdon WY, Bowtell DD. 1998. Tissue hyperplasia and enhanced T-cell signalling

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

DYNAMICS OF TCR SIGNALING

178. 179.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

180.

181.

182.

183.

184.

185.

186.

187. 188.

189.

via ZAP-70 in c-Cbl-deficient mice. Mol. Cell Biol. 18:4872– 82 Thompson CB, Allison JP. 1997. The emerging role of CTLA-4 as an immune attenuator. Immunity 7:445–50 Linsley PS, Brady W, Urnes M, Grosmaire LS, Damle NK, Ledbetter JA. 1991. CTLA-4 is a second receptor for the B cell activation antigen B7. J. Exp. Med. 174:561–69 Krummel MF, Allison JP. 1995. CD28 and CTLA-4 have opposing effects on the response of T cells to stimulation. J. Exp. Med. 182:459–65 Walunas TL, Lenschow DJ, Bakker CY, Linsley PS, Freeman GJ, Green JM, Thompson CB, Bluestone JA. 1994. CTLA-4 can function as a negative regulator of T cell activation. Immunity 1:405– 13 Marengere LE, Waterhouse P, Duncan GS, Mittrucker HW, Feng GS, Mak TW. 1996. Regulation of T cell receptor signaling by tyrosine phosphatase SYP association with CTLA-4. Science 272:1170– 73 Linsley PS, Greene JL, Tan P, Bradshaw J, Ledbetter JA, Anasetti C, Damle NK. 1992. Coexpression and functional cooperation of CTLA-4 and CD28 on activated lymphocytes-T. J. Exp. Med. 176:1595– 604 Waterhouse P, Penninger JM, Timms E, Wakeham A, Shahinian A, Lee KP, Thompson CB, Griesser H, Mak TW. 1995. Lymphoproliferative disorders with early lethality in mice deficient in CTLA-4. Science 270:985–88 Tivol EA, Borriello F, Schweitzer AN, Lynch WP, Bluestone JA, Sharpe AH. 1995. Loss of CTLA-4 leads to massive lymphoproliferation and fatal multiorgan tissue destruction, revealing a critical negative regulatory role of CTLA-4. Immunity 3:541–47 Chambers CA, Sullivan TJ, Allison JP. 1997. Lymphoproliferation in CTLA-4deficient mice is mediated by costimulation-dependent activation of CD4+ T cells. Immunity 7:885–95 Tonks NK, Neel BG. 1996. From form to function: signaling by protein tyrosine phosphatases. Cell 87:365–68 Kozlowski M, Mlinaric-Rascan I, Feng GS, Shen R, Pawson T, Siminovitch KA. 1993. Expression and catalytic activity of the tyrosine phosphatase PTP1C is severely impaired in motheaten and viable motheaten mice. J. Exp. Med. 178:2157– 63 Shultz LD, Schweitzer PA, Rajan TV,

190.

191.

192.

193.

194.

195.

196.

197.

198.

199.

519

Yi T, Ihle JN, Matthews RJ, Thomas ML, Beier DR. 1993. Mutations at the murine motheaten locus are within the hematopoietic cell protein-tyrosine phosphatase (Hcph) gene. Cell 73:1445–54 Tsui HW, Siminovitch KA, de Souza L, Tsui FW. 1993. Motheaten and viable motheaten mice have mutations in the haematopoietic cell phosphatase gene. Nat. Genet. 4:124–29 Pani G, Fischer KD, Mlinaric-Rascan I, Siminovitch KA. 1996. Signaling capacity of the T cell antigen receptor is negatively regulated by the PTP1C tyrosine phosphatase. J. Exp. Med. 184:839–52 Plas DR, Johnson R, Pingel JT, Matthews RJ, Dalton M, Roy G, Chan AC, Thomas ML. 1996. Direct regulation of ZAP-70 by SHP-1 in T cell antigen receptor signaling. Science 272:1173–76 Robbins DJ, Zhen E, Cheng M, Xu S, Ebert D, Cobb MH. 1994. MAP kinases ERK1 and ERK2: pleiotropic enzymes in a ubiquitous signaling network. Adv. Cancer Res. 63:93–116 Winkler DG, Park I, Kim T, Payne NS, Walsh CT, Strominger JL, Shin J. 1993. Phosphorylation of Ser-42 and Ser-59 in the N-terminal region of the tyrosine kinase p56lck. Proc. Natl. Acad. Sci. USA 90:5176–80 August A, Dupont B. 1996. Association between mitogen-activated protein kinase and the ζ chain of the T cell receptor (TcR) with the SH2,3 domain of p56lck. Differential regulation by TcR cross-linking. J. Biol. Chem. 271:10054– 59 Lorenz U, Ravichandran KS, Pei D, Walsh CT, Burakoff SJ, Neel BG. 1994. Lck-dependent tyrosyl phosphorylation of the phosphotyrosine phosphatase SH-PTP1 in murine T cells. Mol. Cell Biol. 14:1824–34 Songyang Z, Carraway III KL, Eck MJ, Harrison SC, Feldman RA, Mohammadi M, Schlessinger J, Hubbard SR, Smith DP, Eng C, Lorenzo ML, Ponder BAJ, Mayer BJ, Cantley LC. 1995. Catalytic specificity of protein-tyrosine kinases is critical for selective signalling. Nature 373:536–39 Pei D, Lorenz U, Klingmuller U, Neel BG, Walsh CT. 1994. Intramolecular regulation of protein tyrosine phosphatase SH-PTP1: a new function for Src homology 2 domains. Biochemistry 33:15483– 93 Pei D, Wang J, Walsh CT. 1996. Differential functions of the two Src homology 2 domains in protein tyrosine phosphatase

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

520

200.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

201.

202.

203.

204.

205.

206.

207.

208.

209.

210.

211.

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A SH-PTP1. Proc. Natl. Acad. Sci. USA 93:1141–45 Grossman Z, Paul WE. 1992. Adaptive cellular interactions in the immune system: the tunable activation threshold and the significance of subthreshold responses. Proc. Natl. Acad. Sci. USA 89:10365–69 Palsson BO, Lightfoot EN. 1985. Mathematical modelling of dynamics and control in metabolic networks. IV. Local stability analysis of single biochemical control loops. J. Theor. Biol. 113:261– 77 Dolmetsch RE, Xu K, Lewis RS. 1998. Calcium oscillations increase the efficiency and specificity of gene expression. Nature 392:933–36 Li W, Llopis J, Whitney M, Zlokarnik G, Tsien RY. 1998. Cell-permeant caged InsP3 ester shows that Ca2+ spike frequency can optimize gene expression. Nature 392:936–41 Constant S, Pfeiffer C, Woodard A, Pasqualini T, Bottomly K. 1995. Extent of T cell receptor ligation can determine the functional differentiation of naive CD4+ T cells. J. Exp. Med. 182:1591–96 Hosken NA, Shibuya K, Heath AW, Murphy KM, O’Garra A. 1995. The effect of antigen dose on CD4+ T helper cell phenotype development in a T cell receptor-αβ-transgenic model. J. Exp. Med. 182:1579–84 Valitutti S, Muller S, Salio M, Lanzavecchia A. 1997. Degradation of T cell receptor (TCR)-CD3-ζ complexes after antigenic stimulation. J. Exp. Med. 185: 1859–64 Cai Z, Kishimoto H, Brunmark A, Jackson MR, Peterson PA, Sprent J. 1997. Requirements for peptide-induced T cell receptor downregulation on naive CD8+ T cells. J. Exp. Med. 185:641–51 Salio M, Valitutti S, Lanzavecchia A. 1997. Agonist-induced T cell receptor down-regulation: molecular requirements and dissociation from T cell activation. Eur. J. Immunol. 27:1769–73 Valitutti S, M¨uller S, Cella E, Padovan E, Lanzavecchia A. 1995. Serial triggering of many T-cell receptors by a few peptideMHC complexes. Nature 375:148–51 Ledbetter JA, Deans JP, Aruffo A, Grosmaire LS, Kanner SB, Bolen JB, Schieven GL. 1993. CD4, CD8 and the role of CD45 in T-cell activation. Curr. Opin. Immunol. 5:334–40 Kupfer A, Singer SJ. 1989. Cell biology of cytotoxic and helper T cell functions: immunofluorescence microscopic studies

212.

213.

214.

215.

216.

217.

218.

219.

220.

221. 222.

223.

of single cells and cell couples. Annu. Rev. Immunol. 7:309–37 Monks CR, Freiberg BA, Kupfer H, Sciaky N, Kupfer A. 1998. Threedimensional segregation of supramolecular activation clusters in T cells. Nature 395:82–86 Saizawa K, Rojo J, Janeway CA Jr. 1987. Evidence for a physical association of CD4 and the CD3:αβ T-cell receptor. Nature 328:260–63 Anderson P, Blue ML, Schlossman SF. 1988. Comodulation of CD3 and CD4. Evidence for a specific association between CD4 and approximately 5% of the CD3:T cell receptor complexes on helper T lymphocytes. J. Immunol. 140:1732–37 Wu H, Kwong PD, Hendrickson WA. 1997. Dimeric association and segmental variability in the structure of human CD4. Nature 387:527–30 K¨onig R, Shen X, Germain RN. 1995. Involvement of both major histocompatibility complex class II α and β chains in CD4 function indicates a role for ordered oligomerization in T cell activation. J. Exp. Med. 182:779–87 Cherry RJ, Wilson KM, Triantafilou K, O’Toole P, Morrison IE, Smith PR, Fernandez N. 1998. Detection of dimers of human leukocyte antigen (HLA)-DR on the surface of living cells by singleparticle fluorescence imaging. J. Cell Biol. 140:71–79 Sakihama T, Smolyar A, Reinherz EL. 1995. Oligomerization of CD4 is required for stable binding to class II major histocompatibility complex proteins but not for interaction with human immunodeficiency virus gp120. Proc. Natl. Acad. Sci. USA 92:6444–48 Sakihama T, Smolyar A, Reinherz EL. 1995. Molecular recognition of antigen involves lattice formation between CD4, MHC class II and TCR molecules. Immunol. Today 16:581–87 Reich Z, Boniface JJ, Lyons DS, Borochov N, Wachtel EJ, Davis MM. 1997. Ligand-specific oligomerization of T-cell receptor molecules. Nature 387:617–20 Germain RN. 1997. T-cell signaling: the importance of receptor clustering. Curr. Biol. 7: R640–44 Posner RG, Lee B, Conrad DH, Holowka D, Baird B, Goldstein B. 1992. Aggregation of IgE receptor complexes on rat basophilic leukemia cells does not change the intrinsic affinity but can alter the kinetics of the ligand-IgE interaction. Biochemistry 31:5350–56 Davis SJ, van der Merwe PA. 1996. The

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

18:5

QC: APR

Annual Reviews

AR078-16

DYNAMICS OF TCR SIGNALING

224. 225.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

226.

227.

228.

229.

230.

231.

232.

233.

234.

235.

structure and ligand interactions of CD2: implications for T-cell function. Immunol. Today 17:177–87 Torigoe C, Inman JK, Metzger H. 1998. An unusual mechanism for ligand antagonism. Science 281:568–72 Janeway CA Jr, Kupfer A, Viret C, Boursalian T, Goverman J, Bottomly K, Sant’Angelo D. 1998. T-cell development, survival, and signalling. A new concept of the role of self-peptide:self-MHC complexes. Immunologist 6:5–12 Ferrell JE Jr, Machleder EM. 1998. The biochemical basis of an all-or-none cell fate switch in Xenopus oocytes. Science 280:895–98 Koshland DE Jr, Goldbeter A, Stock JB. 1982. Amplification and adaptation in regulatory and sensory systems. Science 217:220–25 Goldbeter A, Koshland DE Jr. 1981. An amplified sensitivity arising from covalent modification in biological systems. Proc. Natl. Acad. Sci. USA 78:6840–44 Hogquist KA, Jameson SC, Heath WR, Howard JL, Bevan MJ, Carbone FR. 1994. T cell receptor antagonist peptides induce positive selection. Cell 76:17–27 Pircher H, Rohrer UH, Moskophidis D, Zinkernagel RM, Hengartner H. 1991. Lower receptor avidity required for thymic clonal deletion than for effector T-cell function. Nature 351:482–85 Sebzda E, Kundig TM, Thomson CT, Aoki K, Mak SY, Mayer JP, Zamborelli T, Nathenson SG, Ohashi PS. 1996. Mature T cell reactivity altered by peptide agonist that induces positive selection. J. Exp. Med. 183:1093–104 Grossman Z, Singer A. 1996. Tuning of activation thresholds explains flexibility in the selection and development of T cells in the thymus. Proc. Natl. Acad. Sci. USA 93:14747–52 Smyth LA, Williams O, Huby RDJ, Norton T, Acuto O, Ley SC, Kioussis D. 1998. Altered peptide ligands induce quantitatively but not qualitatively different intracellular signals in primary thymocytes. Proc. Natl. Acad. Sci. USA 95:8193–98 Wiest DL, Ashe JM, Abe R, Bolen JB, Singer A. 1996. TCR activation of ZAP70 is impaired in CD4+CD8+ thymocytes as a consequence of intrathymic interactions that diminish available p56lck. Immunity 4:495–504 Lucas B, Germain RN. 1996. Unexpectedly complex regulation of CD4/CD8 coreceptor expression supports a revised model for CD4+CD8+ thymocyte differentiation. Immunity 5:461–77

521

236. Olszowy MW, Leuchtmann PL, Veillette A, Shaw AS. 1995. Comparison of p56lck and p59fyn protein expression in thymocyte subsets, peripheral T cells, NK cells, and lymphoid cell lines. J. Immunol. 155:4236–40 237. Alberola-Ila J, Hogquist KA, Swan KA, Bevan MJ, Perlmutter RM. 1996. Positive and negative selection invoke distinct signaling pathways. J. Exp. Med. 184:9–18 238. Alberola-Ila J, Forbush KA, Seger R, Krebs EG, Perlmutter RM. 1995. Selective requirement for MAP kinase activation in thymocyte differentiation. Nature 373:620–23 239. Sharp LL, Schwarz DA, Bott CM, Marshall CJ, Hedrick SM. 1997. The influence of the MAPK pathway on T cell lineage commitment. Immunity 7:609–18 240. Kisielow P, Miazek A. 1995. Positive selection of T cells: rescue from programmed cell death and differentiation require continual engagement of the T cell receptor. J. Exp. Med. 181:1975–84 241. Schmitt S, Muller KP, Kyewski BA. 1997. Two separable T cell receptor signals reconstitute positive selection of CD4 lineage T cells in vivo. Eur. J. Immunol. 27:2139–44 242. Ono M, Okada H, Bolland S, Yanagi S, Kurosaki T, Ravetch JV. 1997. Deletion of SHIP or SHP-1 reveals two distinct pathways for inhibitory signaling. Cell 90:293–301 243. Duke RC. 1989. Self recognition by T cells. I. Bystander killing of target cells bearing syngeneic MHC antigens. J. Exp. Med. 170:59–71 244. van Oers NSC, Killeen N, Weiss A. 1994. ZAP-70 is constitutively associated with tyrosine-phosphorylated TCR ζ in murine thymocytes and lymph node T cells. Immunity 1:675–85 245. van Oers NS, Tao W, Watts JD, Johnson P, Aebersold R, Teh HS. 1993. Constitutive tyrosine phosphorylation of the T-cell receptor (TCR) ζ subunit: regulation of TCR-associated protein tyrosine kinase activity by TCR ζ . Mol. Cell Biol. 13:5771–80 246. Matsushita S, Kohsaka H, Nishimura Y. 1997. Evidence for self and nonself peptide partial agonists that prolong clonal survival of mature T cells in vitro. J. Immunol. 158:5685–91 247. Schwartz RH. 1985. T-lymphocyte recognition of antigen in association with gene products of the major histocompatibility complex. Annu. Rev. Immunol. 3:237–61 248. Ronchese F, Schwartz RH, Germain RN. 1987. Functionally distinct subsites on a

P1: ARS/mbg

P2: APR/KKK/SPD

January 19, 1999

522

249.

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

250.

251.

252. 253.

254.

255.

256. 257.

258.

259.

260.

18:5

QC: APR

Annual Reviews

AR078-16

ˇ ´ GERMAIN & STEFANOV A class II major histocompatibility complex molecule. Nature 329:254–56 Parkhurst MR, Salgaller ML, Southwood S, Robbins PF, Sette A, Rosenberg SA, Kawakami Y. 1996. Improved induction of melanoma-reactive CTL with peptides from the melanoma antigen gp100 modified at HLA-A∗ 0201-binding residues. J. Immunol. 157:2539–48 Hemmer B, Vergelli M, Gran B, Ling N, Conlon P, Pinilla C, Houghten R, McFarland HF, Martin R. 1998. Predictable TCR antigen recognition based on peptide scans leads to the identification of agonist ligands with no sequence homology. J. Immunol. 160:3631–36 Hemmer B, Vergelli M, Pinilla C, Houghten R, Martin R. 1998. Probing degeneracy in T-cell recognition using peptide combinatorial libraries. Immunol. Today 19:163–68 Sprent J, Lo D, Gao EK, Ron Y. 1988. T cell selection in the thymus. Immunol. Rev. 101:173–90 Schild H, Rotzschke O, Kalbacher H, Rammensee HG. 1990. Limit of T cell tolerance to self proteins by peptide presentation. Science 247:1587–89 Salemi S, Caporossi AP, Boffa L, Longobardi MG, Barnaba V. 1995. HIVgp120 activates autoreactive CD4-specific T cell responses by unveiling of hidden CD4 peptides during processing. J. Exp. Med. 181:2253–57 Liu GY, Fairchild PJ, Smith RM, Prowle JR, Kioussis D, Wraith DC. 1995. Low avidity recognition of self-antigen by T cells permits escape from central tolerance. Immunity 3:407–15 Oldstone MB. 1987. Molecular mimicry and autoimmune disease. Cell 50:819– 20 Wucherpfennig KW, Strominger JL. 1995. Molecular mimicry in T cellmediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 80:695–705 Alexander MA, Damico CA, Wieties KM, Hansen TH, Connolly JM. 1991. Correlation between CD8 dependency and determinant density using peptide-induced, Ld-restricted cytotoxic T lymphocytes. J. Exp. Med. 173:849–58 Klenerman P, Zinkernagel RM. 1998. Original antigenic sin impairs cytotoxic T lymphocyte responses to viruses bearing variant epitopes. Nature 394:482–85 Klenerman P, Rowland JS, McAdam S, Edwards J, Daenke S, Lalloo D, et al.

261.

262.

263. 264.

265.

266.

267. 268.

269.

270.

271.

1994. Cytotoxic T-cell activity antagonized by naturally occurring HIV-1 Gag variants. Nature 369:403–7 Bertoletti A, Sette A, Chisari FV, Penna A, Levrero M, De Carli M, Fiaccadori F, Ferrari C. 1994. Natural variants of cytotoxic epitopes are T-cell receptor antagonists for antiviral cytotoxic T cells. Nature 369:407–10 Wiesenfeld K, Moss F. 1995. Stochastic resonance and the benefits of noise: from ice ages to crayfish and SQUIDs. Nature 373:33–36 Collins JJ, Chow CC, Imhoff TT. 1995. Stochastic resonance without tuning. Nature 376:236–38 Bray D, Levin MD, Morton-Firth CJ. 1998. Receptor clustering as a cellular mechanism to control sensitivity. Nature 393:85–88 Haughn L, Gratton S, Caron L, Sekaly RP, Veillette A, Julius M. 1992. Association of tyrosine kinase p56(lck) with CD4 inhibits the induction of growth through the αβ T-Cell receptor. Nature 358:328–31 Petricoin EF, 3rd, Ito S, Williams BL, Audet S, Stancato LF, Gamero A, Clouse K, Grimley P, Weiss A, Beeler J, Finbloom DS, Shores EW, Abraham R, Larner AC. 1997. Antiproliferative action of interferon-α requires components of T-cell-receptor signalling. Nature 390:629–32 Saito T. 1998. Negative regulation of T cell activation. Curr. Opin. Immunol. 10:313–21 Yi T, Mui AL, Krystal G, Ihle JN. 1993. Hematopoietic cell phosphatase associates with the interleukin-3 (IL-3) receptor β chain and down-regulates IL-3induced tyrosine phosphorylation and mitogenesis. Mol. Cell Biol. 13:7577–86 Imani F, Rager KJ, Catipovic B, Marsh DG. 1997. Interleukin-4 (IL-4) induces phosphatidylinositol 3-kinase (p85) dephosphorylation. Implications for the role of SHP-1 in the IL-4 -induced signal in human B cells. J. Biol. Chem. 272:7927–31 Migone TS, Cacalano NA, Taylor N, Yi T, Waldmann TA, Johnston JA. 1998. Recruitment of SH2-containing protein tyrosine phosphatase SHP-1 to the interleukin 2 receptor; loss of SHP-1 expression in human T-lymphotropic virus type I-transformed T cells. Proc. Natl. Acad. Sci. USA 95:3845–50 Watts DJ, Strogatz SH. 1998. Collective dynamics of ‘small-world’ networks. Nature 393:440–42

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:467-522. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

QC: KKK

12:20

Annual Reviews

AR078-17

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:523–54 c 1999 by Annual Reviews. All rights reserved Copyright °

THE REGULATION OF CD4 AND CD8 CORECEPTOR GENE EXPRESSION DURING T CELL DEVELOPMENT Wilfried Ellmeier Molecular Pathogenesis Program, Skirball Institute of Biomolecular Medicine, and Howard Hughes Medical Institute, New York University Medical Center, New York, NY 10016

Shinichiro Sawada Department of Neuro-Psychiatry, Saitama Medical School, 38 Morohongo, Morovama, Iruma-gun, Saitama, 350-04, Japan

Dan R. Littman Molecular Pathogenesis Program, Skirball Institute of Biomolecular Medicine, and Howard Hughes Medical Institute, New York University Medical Center, New York, NY 10016 KEY WORDS:

promoter, enhancer, gene silencing, lineage commitment, thymocytes

ABSTRACT The two major subsets of T lymphocytes in the peripheral immune system, the helper and cytotoxic T cells, are defined by their expression of either the CD4 or the CD8 glycoproteins, respectively. Expression of these molecules, which serve as coreceptors by interacting specifically with either MHC class II or class I molecules, also defines discrete stages of T cell development within the thymus. Thus, CD4+ and CD8+ single-positive (SP) thymocytes arise from common progenitor double positive (DP) cells that express both CD4 and CD8, during a process known as positive selection. The molecular mechanisms underlying the developmental choice toward the helper or cytotoxic lineage remain poorly understood. Because regulation of coreceptor gene expression appears to be coupled to the phenotypic choice of the differentiating T cell, it is likely that shared signaling pathways direct CD4 and CD8 transcription and the development of an uncommited DP thymocyte toward either the helper or cytotoxic lineage.

523 0732-0582/99/0410-0523$08.00

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

524

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

Therefore, an understanding of how CD4 and CD8 expression is regulated will not only provide insights into transcriptional control mechanisms in T cells, but may also result in the identification of molecular factors that are involved in lineage choices during T cell development. In this review, we summarize recent progress that has been made toward an understanding of how CD4 and CD8 gene expression is regulated.

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTRODUCTION Development of T lymphocytes bearing αβ T cell receptors (TCRs) specific for classic major histocompatibility complex (MHC) molecules is arguably the best defined ontogenetic system in vertebrates. Progression of thymocytes through distinct developmental stages is marked by induced or repressed expression of numerous regulatory components and cell surface molecules. Among these, expression of the CD4 and the CD8 coreceptor molecules most clearly defines the different stages of thymocyte ontogeny. Coordinate upregulation of CD4 and CD8 on thymocytes marks the successful antigen-independent selection of double negative (DN) cells with functionally rearranged TCR-β chain genes and their transit to the double positive (DP, CD4+CD8+) stage during which they proceed to rearrange the TCR-α chain gene and undergo antigen-dependent selection. The subsequent downregulation of either CD4 or CD8 is a hallmark of positive selection of T cells with functionally useful TCRs; this regulation of coreceptor expression is coupled to the commitment of the single positive (SP, CD4+CD8−or CD4−CD8+) T cells to become helper or cytotoxic effector cells. There is a strong likelihood that factors regulating the expression of CD4 or CD8 are also involved in directing DP thymocytes toward the helper or cytotoxic lineage, respectively. Therefore, it is important to understand how the CD4 and CD8 genes are transcriptionally regulated during T cell development and to identify cis- and trans-acting elements involved in their regulation. This review focuses on our current understanding of the regulation of expression of the coreceptors.

A BRIEF SUMMARY OF T CELL DEVELOPMENT The majority of thymus-derived TCRαβ-positive T cells express either the CD4 or the CD8 coreceptor molecules. CD4-expressing cells almost always have a helper phenotype, and express TCRs specific for MHC class II, whereas CD8+ T cells have a cytotoxic phenotype and are MHC class I restricted. These cells develop in the thymus from common DP progenitor cells through positive selection of those cells that have TCRs with appropriate avidity for MHC/peptide complexes. DP thymocytes develop from progenitors that express neither CD4

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

CD4 AND CD8 GENE REGULATION

525

nor CD8, hence are designated double-negative (DN, CD4−CD8−) thymocytes. For a detailed and comprehensive discussion of general aspects of T cell development, we refer the reader to some excellent recently published reviews (1–4).

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

From Double-Negative to Double-Positive to Single-Positive Thymocytes Immature CD4−CD8− cells, which constitute approximately 1–5% of total thymocytes, are further subdivided into several developmental stages according to the surface expression of the CD117 (c-kit), CD44 (Pgp-1), and CD25 (IL-2 receptor α-chain) molecules. (For detailed reviews on early aspects of T cell development and also for TCRαβ versus γ δ lineage decisions, see 2, 5–9.) Development within this population progresses from the CD117+44+25− cell (also called the thymic lymphoid progenitor) to the CD117−44+25+ cell (pro-T cell), to the CD117−44−25+ cell (pre-T cell), and finally to the CD117−44−25− stage. Functional rearrangement and subsequent expression of the TCRβ-chain at the pre-T cell stage leads to the formation, together with pTα, of a functional pre-TCR complex whose signals result in proliferation and expansion of these thymocytes (a process termed β-selection). The inability to assemble a full pre-TCR signaling complex, as observed in mice with genetic deficiencies in RAG-1, RAG-2, TCRβ, pTα, CD3ε, p56lck/p59 fyn, Syk/ZAP70 (recently reviewed in 9), CD3γ (10) or SLP-76 (11, 12), results in developmental arrest or a major block at the pre-T cell stage and the lack of expansion of cells in this compartment. Signals transmitted through the pre-TCR complex on DN thymocytes not only cause proliferation and expansion, but also induce TCRα gene rearrangement and the expression of the CD4 and CD8 coreceptors, and therefore progression to the DP stage, which constitutes about 80–90% of thymocytes. Functional rearrangement of the TCRα locus and downmodulation of pTα expression result in the surface expression of the mature TCRαβ receptor complex on DP thymocytes. The development of SP thymocytes from DP cells depends on the expression of a functional TCRαβ complex and its productive interaction with MHC/peptide complexes expressed on thymic stromal cells (13). Positive (3) and negative selection (14) ensures that only the rare DP cells that express a TCR complex able to recognize self-MHC/peptide complexes with an appropriate avidity will develop into mature SP thymocytes. Studies during the last decade indicate that the CD4 and CD8 molecules contribute to this selection process by influencing both the avidity of the TCR-MHC interaction and the signaling function of the ligated TCR complex (15–17). These molecules function as coreceptors, interacting with membrane-proximal domains of the MHC molecules (18, 19), thus permitting simultaneous interaction of the TCR and

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

526

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

either CD4 or CD8 with MHC and also enhancing signaling owing to the increased proximity of the coreceptor-bound cytoplasmic protein tyrosine kinase p56lck (20, 21) to the TCR complex. The bridging of TCR and coreceptor in their interaction with MHC/self-peptide complexes is generally required for appropriate development of DP cells into mature SP thymocytes (and subsequently also in the activation of peripheral T cells upon recognition of MHC/antigen complexes). Thus, continued expression of a coreceptor is thought to be essential to maintain a requisite avidity that permits full maturation of the SP thymocyte, which is accompanied by changes in the expression of several surface molecules, including the upregulation of TCRαβ levels, the transient expression of CD69, and the downmodulation of HSA expression (22, 23). Finally, the selected and mature thymocytes seed the peripheral immune system, in which expression of either CD4 or CD8 marks the functional phenotype of the mature T cells.

Lineage Commitment of Double-Positive Thymocytes and the Role of Coreceptor Molecules The molecular mechanisms involved in lineage choice during the development of DP thymocytes into CD4 or CD8 SP cells have been a matter of considerable debate during the last several years (24–27). One of the key questions has been whether DP thymocytes are able to discriminate between TCR-MHC class I and TCR-MHC class II interactions at the onset of positive selection. The ability to distinguish between these interactions could lead to an instructive signal that would direct DP thymocytes toward either the CD8 or CD4 lineage. Because CD4 and CD8 bind to p56lck with different affinities by way of their cytoplasmic domains (21, 28, 29), models have proposed that either the strength of signals transmitted through Lck (30, 31) or the balance between TCR- and Lck-mediated signals determines the outcome of the lineage choice of DP thymocytes (32). However, coreceptor-Lck interactions are not absolutely essential for lineage development (33, 34), suggesting that even if instructive signals are required, they need not involve binding of Lck to the coreceptor. Alternative models have proposed that the initial TCR-MHC interaction results in a stochastic lineage choice by the DP thymocytes (i.e. downregulation of either the CD4 or the CD8 coreceptor) regardless of the MHC specificity of the receptor. Subsequently, only those thymocytes with a coreceptor whose MHC specificity matches that of the TCR would be positively selected (35–37). More recent studies suggest a more complex mechanism of lineage choice and indicate an asymmetric requirement of TCR-MHC interactions for commitment to the CD4 or CD8 lineage (38–42). Examination of highly purified thymocyte subpopulations showed that commitment to the CD4 lineage can occur upon

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD4 AND CD8 GENE REGULATION

527

TCR recognition of either MHC class I or II, whereas CD8 lineage commitment appears to be dependent on TCR interaction with MHC class I, suggesting that in the absence of an instructive CD8 lineage signal (requiring TCR and presumably CD8), DP thymocytes develop by default to the CD4 lineage (40, 41). Recent experiments indicate that overexpression of an activated form of Notch in thymocytes of transgenic mice favors the development of CD8 lineage cells, suggesting that the Notch signaling pathway is involved in instructing the CD8 lineage choice (43). However, a requirement for this signaling pathway in T cell lineage commitment has yet to be demonstrated. While the molecular mechanisms of lineage commitment remain largely uncharacterized, it is clear that the transcriptional regulation of the CD4 and CD8 genes is tightly linked to the functional program of the developing T cell (44–47). For example, class I–specific thymocytes that are obtained by expression of CD8 transgenes in CD4 lineage cells retain the helper phenotype of the CD4 SP cells, despite reactivity with the “wrong” class of MHC molecules. Such findings suggest that factors that positively and negatively regulate CD4 and CD8 expression in developing SP thymocytes are also involved in directing DP thymocytes toward the helper or cytotoxic lineage.

REGULATION OF CD4 GENE EXPRESSION Early studies suggested that regulation of mRNA stability contributes significantly to determining levels of CD4 (48, 49). For example Takahama & Singer showed that although CD4 and CD8 mRNA levels were attenuated upon in vitro stimulation of DP thymocytes, there was no effect on the rate of CD4 transcription (49). Subsequent studies, using both transfection assays and analyses of transgenic mice, have provided strong evidence that CD4 expression in both immature thymocytes and mature T cells is regulated primarily at the transcriptional level (50–52). These experiments have identified elements, described below, that are involved in both positive and negative regulation of CD4 gene transcription.

Cis-Regulatory Elements and Trans-Acting Factors Involved in CD4 Gene Regulation CD4 PROMOTER The minimal murine CD4 promoter has been mapped within a 101-bp fragment immediately 50 from the transcription initiation site by using transient transfection assays. Sequence analysis revealed the presence of two Myb binding sites within the 101-bp fragment, and mutational analysis demonstrated that both sites are necessary for full activity of the TATA box-less CD4 promoter (53, 54). Further analysis of the CD4 promoter by linker-scanning mutations led to the identification of three additional cis-acting regions: an

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

528

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

initiator-like sequence at the cap site, a Myc-associated zinc finger protein (MAZ) binding site, and an Ets protein consensus binding sequence (55). Although binding of Ets proteins to this site was not demonstrated in this study, mutation of the Ets binding site decreased promoter activity in the CD4+CD8− T cell clone D10 (55). For the human CD4 promoter, Ets family members can transactivate the CD4 promoter in transiently transfected HeLa cells, but the importance of Ets proteins in CD4 promoter function in T cells has not been determined (56). Some studies have suggested that the CD4 promoter has preferential activity in CD4+CD8− T cell lines compared with CD4−CD8+ cell lines or other T cell lines representative of early thymocyte lineages (53, 56). However, in combination with either the SV40 enhancer (50, 56) or the CD4 enhancer (50), expression of a reporter gene was obtained either in non–T cells or in CD4 SP and CD8 SP T cell lines, respectively. Furthermore, expression of a reporter gene was obtained both in CD4 and CD8 SP T cells in transgenic mice when the CD4 promoter was combined with the CD4 enhancer (51, 52). The discrepancy of these results may be explained by assuming that the CD4 promoter is preferentially active in the CD4 subset, but that its specificity is masked by the presence of the CD4 enhancer, which is active in all T cells. Alternatively, it is possible that the conflicting results are caused by performing in vitro transactivation assays in cell lines that might not properly reflect the transcriptional regulation and specificity observed in primary T cells in vivo. Further examination of the CD4 promoter in transgenic mice is required to resolve this issue. DNase I hypersensitive (DH) site analysis within 80 kb of the murine CD4 locus led to the identification of a cis-acting sequence that displayed enhancer activity in transiently transfected T cell lines, but not in B cell lines or fibroblasts (50). This enhancer, which is located approximately 13 kb upstream of the transcriptional start site of the CD4 gene, was active both in CD4 and CD8 SP T cell lines, indicating that subset-specificity of CD4 expression is not mediated by this enhancer. By using transfection assays, the enhancer was narrowed down to a 339-bp core fragment (50), and its function was validated by the ability to direct expression of various reporter genes in T cells and thymocytes of transgenic mice (51, 52, 57–59). The core enhancer was found to contain three (CD4-1, CD4-2, and CD4-3) nuclear protein binding sites (60). One of these (CD4-2) represented a consensus binding sequence for the HMG (high mobility group) family members TCR1/LEF1 (61). Binding of LEF1 to this site could be demonstrated since a polyclonal antiserum against LEF1 recognized a T cell- and thymus-specific nucleoprotein complex formed with this sequence in gel mobility shift assays

CD4 ENHANCER

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD4 AND CD8 GENE REGULATION

529

(S Sawada, R Grosschedl, DR Littman, unpublished observation). However, binding of TCF1/LEF1 was largely dispensable for activity in transient transfection assays (60). Both of the other nuclear protein binding sites (CD4-1 and CD4-3) contained E-box motifs, which are recognized by bHLH (basic helix–loop–helix) factors (62). One of these (CD4-3) contained two E-boxes (designated as 50 or 30 E-box) that were separated by a single nucleotide and formed a lymphocyte-specific complex with a nuclear protein that interacted with the 50 E-box in gel shift assays. Use of specific antibodies demonstrated that this complex contained the bHLH factors HEB and E2A (60), suggesting that a heterodimer (or hetero-oligomer) of HEB and E2A interacts with this 50 E-box. The T cell specificity of this nucleoprotein complex, which contains the ubiquitously expressed E2A (63), is most likely explained by the preferential expression of HEB in thymocytes (64). The functional importance of this 50 E-box was further demonstrated by mutational analysis. A mutation that disrupted binding of the E2A/HEB heterodimer abolished CD4 enhancer activity (60). This suggests that HEB plays an important role in the regulation of CD4 expression. Consistent with this result, mice with a targeted disruption of a functional domain of HEB had significantly reduced levels of CD4 expression on thymocytes, but not on mature splenic T cells (65). In addition, this study also revealed that expression of CD4 is dependent on the dosage of E2A and HEB. Since the level of expression of the CD4 gene is indistinguishable between immature and mature T cells, these results suggest that an interaction of the enhancer and the bHLH factor(s) is required for CD4 transcription in the immature stages of T lymphopoiesis, but not in mature thymocytes and T cells. In mature CD4+ T cells, CD4 transcription may be maintained by different mechanisms in which HEB is no longer required. Both the murine CD4 enhancer and a homologous human CD4 (hCD4) enhancer, located 6.5 kb upstream of the human CD4 gene (66), were shown to function in a T cell-specific manner in vivo. Transgenes in which either enhancer was linked to a hCD4 minigene (in which most of the large third intron had been deleted) directed expression of hCD4 on both peripheral T lymphocytes and thymocytes in mice (57–59, 66). Significantly, expression was restricted to cells that expressed endogenous CD4 owing to the presence of the CD4 silencer in the first intron (see below). One of the studies showed that in the absence of the enhancer, low level subset-specific expression of hCD4 could be detected in mature CD4 SP T cells, but not in DP thymocytes. The inclusion of the murine enhancer increased CD4 levels on mature thymocytes and resulted in expression in DP thymocytes (59). The in vivo function of the enhancer was also confirmed using a murine transgene that contained a human CD2 (hCD2) cDNA as a reporter gene driven by the murine CD4 promoter. The 339 bp murine enhancer directed hCD2 expression in both peripheral T

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

530

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

lymphocytes and thymocytes in the majority of transgenic mice analyzed. When the enhancer was omitted, hCD2 expression was detected, at a very low level, only in the peripheral T lymphocytes of one out of eleven founders, thus suggesting a position effect at the integration site in this animal (51). Together, these results indicate that the CD4 enhancer is required both for high level expression and for tissue-specificity of the transgenes in vivo. In addition, they suggest that CD4 promoter regulation may differ between mature T cells and DP thymocytes, although the results may also be explained by a greater sensitivity for detection of chromatin position effects in mature T cells of transgenic mice. Studies suggesting differences in positive regulation of CD4 in mature versus immature thymocytes are discussed in greater detail below. A second enhancer, corresponding to a DNase I hypersensitivity site, was identified approximately 24 kb upstream of the murine CD4 gene (67). In transient transfection assays, its activity appeared restricted to mature T cells. An Ets family member, Elf-1, which is expressed at a high level in T cells, was implicated in regulation of this enhancer (67). However, the relevance and celltype specificity of this regulatory region remains unclear, since in transgenic mice a combination of the proximal CD4 enhancer and this distal enhancer with other DNase I hypersensitivity sites 30 to the CD4 gene resulted in expression of a reporter gene not only in T cells, but also in B cells and macrophages (52). Whether the distal enhancer can specifically augment transcription in mature T cells in the absence of the proximal CD4 enhancer has not been shown in a transgenic mouse system. The CD4 gene resides in a gene-rich region of human chromosome 12 and mouse chromosome 6 (68, 69). It is flanked by the LAG-3 gene upstream of the T cell–specific enhancer and by a cluster of three other genes, including the heterotrimeric G-protein β3 subunit gene, within 20 kb of the polyadenylation site. Because the distal enhancer lies immediately 30 to the LAG-3 gene, which encodes a CD4-related molecule that is expressed in activated CD4 and CD8 lineage cells, as well as natural killer cells (70), it is possible that the enhancer is more relevant for the regulation of LAG-3. In addition, this gene-dense genomic region must contain multiple enhancer elements with variable tissue specificities. Therefore, validation of the role of any enhancer relevant for CD4 expression will require in vivo mutagenesis of the cis-acting sequences by gene targetting. In addition to its expression in T cells, the CD4 glycoprotein is expressed in macrophages and dendritic cells in humans (71, 72), but not in mice (73). In humans, it is expressed on cells of other hematopoietic lineages, such as eosinophils (74). The species difference in monocyte and macrophage expression is dictated by differences within cis-acting sequences of CD4 genes in these species. Thus, a transgene containing only 2.6 kb upstream of the human CD4 transcription start site plus exons 1–3 and the intervening introns was

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

CD4 AND CD8 GENE REGULATION

531

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

expressed in murine macrophages, indicating that the requisite trans-acting factors are present in mouse (59). Another human CD4 transgene was expressed in macrophages independent of the presence or absence of the murine CD4 enhancer (N Killeen, S Sawada, DR Littman, unpublished results). In contrast, murine CD4 transgenes were not expressed in monocytes or macrophages. Construction of chimeric mouse-human transgenes has permitted mapping of a macrophage-specific enhancer within the first intron of the human CD4 gene (JD Scarborough, N Killeen, DR Littman, unpublished results). However, further transgenic reporter analysis is required to define the minimal enhancer region. CD4 SILENCER Several early studies suggested that T cell subset-specific CD4 expression would be subject to negative regulation. For example, transfection studies showed that the CD4 enhancer and promoter directed expression in T cells regardless of the CD4 expression pattern in such cells (50); and whereas human CD4 transgenes with intact exon/intron organization were specifically expressed in CD4, but not CD8, SP cells (57–59, 66), another transgene lacking introns 1–4 was expressed in both lineages (51). The presence of DNase I hypersensitivity sites in the first intron of the murine gene additionally suggested that this region would be involved in subset-specific expression (50, 75). Because subset-specific expression could not initially be demonstrated in transfected T cells, transgenic analyses were employed to map elements required for CD4 SP-specific expression. The human CD2 and HLA-B7 cDNAs were used as transgenic reporters to show that a region within the first intron of the murine CD4 gene conferred subset-specific expression when combined with the CD4 enhancer and promoter (51, 52). Thus, a murine minigene containing the 339 bp CD4 enhancer and 0.5 kb promoter, followed by non-coding exon 1, intron 1, and exon 2 (in which the hCD2 reporter gene was introduced), displayed subset-specific expression in transgenic mice. It was further shown that sequences within intron 1, including the DH sites at about 2 kb downstream from the transcription initiation site (50, 75), were required to shut off expression of the reporter gene in CD8 SP cells (51). When the intronic element was deleted, the CD4 enhancer and promoter were active in both the CD4+CD8− and CD4−CD8+ subsets and, unexpectedly, also in DN thymocytes. This indicated that the CD4 enhancer described above (or a yet uncharacterized enhancer within the minigene) is active early in thymocyte development, but is normally silenced. Because mouse CD4 is expressed on precursor cells in the bone marrow (76) and on very immature pro-thymocytes (77), these results suggest that CD4 is shut off on thymic lymphoid progenitors (CD117+44+25−) owing to silencer activation in DN cells. The silencer activity must then be inactivated upon progression to the DP stage, and is reestablished during positive selection to the CD8 SP lineage.

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

532

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

Subset-specific expression was also observed when the CD4 proximal enhancer and the distal enhancer element were combined with the intronic element to direct transgenic expression of HLA-B7 (52). Interestingly, in the absence of the intronic sequence, inclusion of DH sites 30 of the CD4 polyadenylation site directed expression not only in T cells, but also in B cells and myeloid cells (52). Because expression in the presence of the silencer was limited to CD4 SP T lymphocytes, the results suggest that the silencer can also function in non-T lineage cells. Whether this is relevant in vivo remains to be determined. The murine CD4 intronic element was initially narrowed down to a 434-bp fragment that, when placed outside the coding region, could repress reporter transgene expression in CD8 lineage cells (51). This activity was independent of orientation or location relative to the transcribed region (51, 52) and also functioned in combination with the heterologous enhancer and promoter from the CD3δ gene (51). The last observation indicates also that the CD4 promoter is not absolutely required for CD4 subset-specific transcription in vivo, even though it has preferential activity for CD4 SP T cell lines on its own in transfection systems (53, 56). Taken together, the properties of this intronic sequence indicate that it is a bona fide silencer. A lineage-specific silencer was also identified in the first intron of the human CD4 gene, by using a human CD4 transgene in which exons 2–10 were replaced with cDNA sequence (78). Its activity was mapped to a 484 bp RsaI fragment the sequence of which was 77% identical to that of the murine silencer. Remarkably, the 50 190 bp of this fragment, which overlaps the 50 sequence 1–173 of the murine silencer, was sufficient to direct silencing in transgenic mice (see below). Together, these findings show, in three different systems, that combinations of the minimal enhancer, promoter, and silencer can direct developmental expression of reporter genes in T cells in a manner that is indistinguishable from that observed for endogenous CD4. While these results do not rule out the possibility that additional sequences contribute to CD4 gene regulation (as discussed further below), they indicate that it is highly unlikely that posttranscriptional mechanisms have a significant role in CD4 expression. DNase I footprinting and transient transfection assays have been coupled with transgenic studies to identify functional sites within the CD4 silencer. In one study, three sites were identified within the 434-bp silencer by footprinting with T cell nuclear extracts (Figure 1; 79). Deletion of any single site had no effect on silencer activity in transgenic mice. However, multiple deletions of these sites indicated that either site II alone or a combination of sites I and III is required for silencer activity, suggesting that there is an asymmetric redundancy in the mechanism of CD4 silencer function (79). In the human silencer, two sequences in the active 50 190-bp fragment were footprinted using nuclear extracts of human CD4 or CD8 SP cell lines (78). One of these corresponded

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD4 AND CD8 GENE REGULATION

533

Figure 1 Transcriptional regulatory elements within the murine CD4 locus. Organization of the murine CD4 locus and localization of DNase I hypersensitivity sites (50, 52, 75, 142). DNase I hypersensitivity sites are indicated by the vertical arrows. The cluster of hypersensitivity sites far downstream of the CD4 gene is approximately 20 kb 30 to the polyA site (75). All BamHI (B) sites are shown. The transcriptional orientation of the CD4 gene is shown by the horizontal arrow. Closed and open bars around the CD4 gene represent CD4 coding and noncoding exons, respectively (142). Proximal (E4P) and distal (E4D) enhancers (located approximately 13 kb and 24 kb upstream of the transcriptional start site, respectively (50, 67); the promoter (P4) and the silencer (S4) (51, 52) are shown as filled circles. The expanded section above the map shows putative factor binding sites on murine and human silencer subelements defined by DNase I footprinting and transfection studies. The top corresponds to the minimal human silencer sequence shown to function in transgenic mice, and depicts the two footprinted sites (78); the middle shows the three footprint sites detected in the murine silencer by Duncan et al (79); the lower shows a combination of sites (boxes A to H) defined by footprint analysis and by functional analysis of core sequence (base pairs 132–274) deletions in transient tranfection assays (S Sawada, S Mahanta, I Taniuchi, DR Littman, manuscript in preparation).

precisely to footprint I described by Duncan et al (Figure 1; 79), whereas the other did not correspond to any footprints in the murine silencer, but overlapped with site C that was detected by deletional analysis of the murine silencer (see below). Comparison of the results of the different studies suggests that there are different requirements for human versus murine silencer function in transgenic mice. Thus, the murine silencer is inactive if sequences corresponding to footprints II and III are both deleted, even though sequence 1–173 is fully

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

534

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

intact (79). In contrast, the human sequence corresponding to murine silencer sequence 1–173 is sufficient for in vivo function. Although this result may indicate that a key nuclear protein-binding motif is present in the human 50 silencer fragment and not in the murine equivalent, it remains possible that the conflicting results are due to differences in the overall design of the transgenic constructs. It will therefore be important to compare the functions of murine and human silencer fragments in the context of the same transgenic reporter construct. Redundancy of function was also observed in other studies in which an analysis of silencer function in transient transfection assays was used to guide the choice of mutations in transgenic constructs. When three copies of the murine silencer were linked to a CAT reporter vector regulated by the CD4 enhancer and promoter, expression of CAT was specifically repressed in the CD4−CD8+ 1200 M thymoma cell line, but not in CD4+CD8+ T cells [e.g. 1010-5 (51)]. Using this assay system, it was possible to further narrow down the relevant sequence involved in repression to a core consisting of residues 132–265 within the 434-bp silencer (Figure 1). Inclusion of three copies of this 134-bp fragment in the CD4-CAT vector resulted in 10–20-fold repression of transcriptional activity (S Sawada, S Mahanta, I Taniuchi, DR Littman, manuscript in preparation). Although the core silencer fragment functioned even better than the 434-bp silencer in transfection studies, it had no silencing activity in transgenic mice. However, inclusion of either the 50 or 30 flanking sequence restored full silencer activity in the transgenic constructs (S Sawada, JD Scarborough, DR Littman, unpublished results). The functional redundancy of the flanking sequences in animals, and their dispensability in transient transfection studies, suggest that they contain elements needed for organizing the chromatin structure in a manner that permits trans-acting silencing factors to be active. Deletion of residues 206–225 (motif F in Figure 1) abrogated silencer activity in multiple transgenic founders, suggesting that this sequence encompasses a motif that is essential for silencer function (Table 1). However, individual deletions of other segments, including motifs, C, D, and G, resulted in derepression of reporter expression in only a fraction of CD8 SP T cells (S Sawada, J Scarborough, DR Littman, unpublished results). Thus, reporter expression was either on or off in individual CD8 lineage cells, with variable ratios in different transgenic mouse lines. This phenotype resembles the phenomenon of position effect variegation in Drosophila (80, 81) and suggests that a mutant silencer becomes subject to the influence of the chromatin structure surrounding the integration site. However, loss of a single nuclear factor binding site appears not to completely inactivate the silencer but, rather, to decrease the probability of establishment of silencing. This suggests that multiple nuclear factors must bind to the silencer to ensure that it functions optimally, but that a

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

CD4 AND CD8 GENE REGULATION

535

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Table 1 Summary of expression phenotypes of a hCD2 reporter gene in CD8+ T cells. Expression of hCD2 was under the control of the CD4 enhancer and promoter plus the wild-type silencer (base pairs 132–434) or mutant silencer containing various deletions within the core sequence (S Sawada, J. Scarborough, DR Littman, unpublished). Transgene

Deletion

Phenotype

38 44 45 46 51 52

None (132–434) 1C (132–145) 1E (186–205) 1C/1E 1F (206–225) 1G (246–265)

Silencing Variegation Variegation No silencing No silencing Variegation

level of redundancy exists (Figure 2). The ratio of derepressed versus silenced reporter in CD8 SP cells was found to be stable and heritable in individual transgenic lines. Silencing may thus be established during a window in development, in which case it is stably maintained. Variegation may hence reflect the probability of establishing a stable silencing complex in a mutant locus that becomes susceptible to position effects. Studies on the mechanism of action of the CD4 silencer are in their preliminary stages. The silencer may change the structure of adjacent chromatin, possibly by influencing DNA methylation and/or the activity of histone deacetylases, or it may function by directly influencing components involved in initiating transcription of the CD4 gene (82–86). The observation of variegated expression of mutant transgenic reporter constructs is consistent with a heritable mechanism of CD4 silencing, but it remains possible that there is a requirement for sustained inhibitory interaction of silencing factors with components of the transcriptional machinery. It is also possible that the observed variegation reflects an artifact of the transgenic system used. Definitive studies of the mechanism of silencing will require modification of silencer sequences in the murine germ line. Preliminary studies indicate that the targeted deletion of a 1.6-kb fragment containing the silencer results in derepression of the CD4 gene in CD8 SP lineage cells (Y Zou, N Killeen, DR Littman, unpublished results). To determine whether CD4 silencing involves a heritable epigenetic mechanism, it will be important to examine whether, once established, silencing can be reversed by deletion of the relevant cis-acting sequences in conditional knockout mice. Biochemical characterization of trans-acting factors will also provide important insight into the mechanism of silencing. However, at the time of this writing, only HES-1, the mammalian homologue of the Drosophila Notch signaling pathway mediators Enhancer of split (for review, see 87–89) has been

P2: APR/spd/ary

February 11, 1999

536

12:20 Annual Reviews

Figure 2 Model for variegation of silencing in transgenic mice upon deletion of subelements within the silencer region. Several proteins binding to silencer subelements may be involved in recruitment of a transcriptional silencing complex that may consist of either histone deacetylases (HD), methylases, or other factors that are able to modify chromatin and to suppress gene transcription. Deletion of individual boxes (and the consequent loss of binding putative recruitment factors) may result in inefficient recruitment of the silencing complex and in a decreased probability of establishment of silencing during a critical developmental stage. Subsequent establishment or maintenance of silencing may then be subject to epigenetic regulation.

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SAT/mbg QC: KKK

AR078-17

ELLMEIER, SAWADA & LITTMAN

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD4 AND CD8 GENE REGULATION

537

shown to participate in the silencing process (90). Siu and colleagues demonstrated that HES-1 binds to the CD4 silencer region. Transient overexpression of HES-1 in D10 CD4 SP TH cells resulted in an approximately 60% repression of CD4 promoter and enhancer activity (measured by luciferase activity of a reporter construct) that was dependent on the presence of a functional HES-1 binding site within the silencer region. In addition, endogenous CD4 surface levels of the transfected T cell line were downmodulated. These data are in agreement with the proposed role of the Notch signaling pathway in directing DP thymocytes toward the CD8 SP T cell lineage (43) and may indicate that Notch signaling induces transcriptional downmodulation of CD4 in the CD8 lineage via the HES-1 protein. Further studies are required to confirm the proposed role of HES-1 in the regulation of CD4 gene expression in vivo. ADDITIONAL CD4 TRANSCRIPTIONAL CONTROL ELEMENTS? Although the expression of reporter genes regulated by a combination of CD4 enhancer, promoter, and silencer appears to recapitulate the pattern of expression of CD4 in transgenic mice, several reports have suggested that additional elements may be involved (91–94). An intronless CD4 minigene, in which the hCD4 cDNA was under the control of three tandem copies of the 339-bp murine CD4 enhancer upstream of a 1.1 kb human CD4 promoter fragment, directed expression of hCD4 in peripheral T cells (both TCRαβ and TCRγ δ lineages) and some NK cells (construct 1 in Figure 3). Strikingly, in the thymus, the transgene was expressed only in HSAlo SP thymocytes, but not in DP thymocytes or CD3− DN thymocytes (91). This study was the first to suggest that, even when the CD4 enhancer was included, CD4 expression may be differentially regulated in double-positive and single-positive thymocytes. Two subsequent studies provide additional evidence that specific cis-acting elements may be required for expression in DP, but not SP, thymocytes. When murine sequences upstream from the CD4 transcriptional start site, including the proximal and distal enhancers, were placed in front of the HLA-B7 reporter (construct 2), expression in transgenic mice was limited to mature thymocytes and T cells (92). Inclusion of additional genomic DNA containing DH-sites 50 to the murine CD4 gene restored weak expression in DP cells (construct 3), while the addition of sequences (DH sites 11–17) 30 of the CD4 gene (construct 4), which directed expression of the reporter in several hematopoietic lineages, rescued expression in the DP thymocytes (92). In another study, deletion of most of the sequence of intron 1 in a human CD4 minigene (in which exons 2–10 were replaced with cDNA sequence; construct 5) resulted in reduction of transgene expression in DP cells relative to that in mature T cells (94). The results of these studies need to be interpreted in the context of the observation that a murine minigene (construct 6), consisting of the murine proximal

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

538

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

Figure 3 Transgenic constructs used to dissect CD4 enhancer activity in immature versus mature thymocytes. Schematic map of transgenic constructs (after 51, 52, 78, 91, 92, 94; YR Zou & DRL, unpublished). Note that the maps are not to scale. Human CD4 (91), HLA-B7 (52, 92) and hCD2 (51) were used as reporter genes for monitoring enhancer activity. Thin lines represent genomic DNA containing additional DNase I hypersensitivity sites localized upstream of the CD4 gene (constructs 3 and 4) or sequences within the first intron of CD4 (constructs 5 and 6). The activity of the different transgenic constructs in different T cell populations is shown on the right and the reference reporting the observed expression pattern is indicated: +, +/−, or − indicate strong, weak or no expression of the reporter gene, respectively. Abbreviations: mE: murine proximal CD4 enhancer; DE: murine distal CD4 enhancer; hP4: human CD4 promoter; mP4, murine CD4 promoter.

enhancer, the promoter and part of intron 1, directed expression of a hCD2 reporter equally well in DP and SP thymocytes and in mature T cells (51). One possible explanation for the discrepancy is that an enhancer, located in the first intron, is required for expression in DP cells and was spared by the 6-kb deletion in the murine minigene, but was deleted in the human minigene (construct 5) described by Uematsu and colleagues (94). However, when exons 1 and 2 were fused in the murine minigene (construct 7), thus completely eliminating sequences within the first intron, expression of the hCD2 transgene in DP thymocytes was unaffected (YR Zou, DR Littman, unpublished results). Taken together, these findings suggest that differences in upstream sequences between the various transgenic constructs explain the disparate results. The most parsimonious explanation for these apparently conflicting results is that expression in DP cells is repressed by an element upstream of the 0.5-kb minimal promoter used in constructs 6 and 7. If such a negative regulatory element is present, it would have to be antagonized by a positive element that is distinct from either the proximal CD4 enhancer or the distal enhancer. Such an element may be present in the first intron of the CD4 gene but could be replaced by regulatory sequences in the gene-rich region 30 of the CD4 gene (68, 92). Further

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD4 AND CD8 GENE REGULATION

539

transgenic and gene targeting studies will be required to resolve this complex problem. The mechanism that regulates expression of CD4 in early thymic immigrants may also differ from the regulatory mechanism in mature thymocytes. Transgenes in which the silencer was deleted (Figure 3, construct 6) derepressed expression of the reporter gene in triple negative thymocytes (CD3−CD4−CD8−), indicating that silencer and enhancer functions are both normally intact in these cells (51, 94). The importance of the silencer was confirmed by the demonstration that CD4 was derepressed in these cells when the silencer region was deleted by gene targeting in mice (YR Zou, N Killeen, DR Littman, unpublished results). However, when the first intron was completely removed by fusing exons 1 and 2 (Figure 3, construct 7), expression of a reporter transgene was observed in DP and SP thymocytes but not in the DN cells (YR Zou, DR Littman, unpublished results). Thus, sequences in intron 1 may be involved in directing CD4 expression in the early thymic precursors, and other sequences, within the silencer region, shut this off as DN cells progress through development.

REGULATION OF CD8 GENE EXPRESSION Thymus-derived T cells usually express CD8α and CD8β heterodimers, whereas extrathymically derived intraepithelial lymphocytes (IEL) from the gut and a subset of human natural killer (NK) cells express CD8αα homodimers (95–97). This partially overlapping but distinct expression pattern of the CD8α and β genes, which are linked at a distance of about 36 kb on mouse chromosome 6 (98), indicates that their expression must be both coordinately and independently regulated. Evidence for independent regulation of CD8α and β gene expression has also been obtained in fusion experiments of CD8+ MHC class I– restricted T cells with the BW5147 (CD4−CD8−) thymoma cell line (99, 100). CD8 surface expression was repressed in hybridomas generated by fusion of these two cell lines. CD8α mRNA has not been detected in these hybridomas, suggesting negative regulation of CD8α expression at the transcriptional level (101). However, posttranscriptional regulatory mechanisms have not been ruled out. Interestingly, CD8β transcription remained unaffected in these hybridomas (102, 103). This indicates that negatively acting factors might be involved in the regulation of CD8α gene expression, at least in CD4−CD8− cells. Whether negative factors (or putative silencing factors) are also involved in the regulation of CD8α in mature peripheral CD4+ T cells is not known (see below). As mentioned above, a study by Takahama & Singer suggested that expression of CD4 and CD8 might not be exclusively regulated at the transcriptional level in DP thymocytes in contrast to mature thymocytes (49). Further support for solely transcriptional regulation of CD8α expression in mature murine T

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

540

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

cells was obtained by the generation of transgenic mice with a genomic clone containing the CD8α promoter and coding region linked to the human CD2 locus control region. Those mice expressed transgenic CD8α in both CD4+ and CD8+ T cells at similar levels (104). However, a recent study reported that posttranscriptional control mechanisms contribute to the regulation of human CD8α in mature CD4+ T cells (105). The authors showed by nuclear run-on assays that CD4+ T cells actively transcribed CD8α mRNA. This might indicate that, in contrast to murine T cells, mature human CD4+ T cells may utilize posttranscriptional regulatory mechanisms to suppress the expression of CD8α protein. Because CD8α expression can be induced on human CD4+ T cells by IL-4 (106), it is possible that the posttranscriptional block of CD8α expression can be overcome by IL-4.

Cis-Regulatory Elements and Trans-Acting Factors Involved in CD8α and CD8β Regulation CD8α PROMOTER AND PROMOTER-PROXIMAL REGIONS Several cis-regulatory elements involved in the regulation of both mouse and human CD8α gene expression have been identified by studying reporter gene expression in transfected cells or in transgenic mice. Initial studies were focused on the promoter and promoter-proximal regions of the CD8α gene. A cyclic AMP responsive element-like site present in the TATA box-less human CD8α promoter (nucleotides −143 to −133) is necessary for basal promoter activity (107). Mutation of this site led to a loss of transcriptional activation of a luciferase reporter construct. However, tissue and lineage specificity of CD8α expression does not reside within the promoter region, since basal activity was observed also in nonlymphoid cell lines and in CD4+ T cells. DNase I hypersensitivity studies performed by the same group led to the identification of a T cell–specific enhancer located in the last intron of the human CD8α gene (108, 109). A minimal 653-bp enhancer region containing a cluster of consensus binding sites for GATA-3, LyF-1, TCF-1, Ets-1, and bHLH proteins displayed T cell-specific activation of a luciferase construct (20- to 30-fold activation) in vitro, suggesting a prominent role for this enhancer in regulation of CD8α gene expression. In transgenic mice, however, this enhancer directed expression of a human CD8α reporter construct in NK cells, but not in T cells (110). Since CD8αα homodimers are expressed in a subset of human NK cells (97), it is likely that this intronic enhancer is involved in directing expression of CD8α in this cell type but does not mediate expression in the CD8 T cell lineage. This is further supported by the finding that a homologous intronic enhancer has not been detected within the body of the murine CD8α gene, which is not expressed in NK cells (110). Several binding sites for GATA-3 have been described in a region that coincides with a CD8-specific DH site about 4 kb upstream of the murine CD8α

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD4 AND CD8 GENE REGULATION

541

transcriptional initiation site (111). Coexpression of GATA-3 with a CAT reporter construct containing dimers of the GATA-3 binding sites and the minimal c-fos promoter in HeLa cells resulted in a 15- to 20-fold increase in CAT activity, suggesting that GATA-3 is involved in the regulation of CD8α gene expression. However, since GATA-3 is also expressed in CD4+ T cells (112), it is unlikely that it is involved in mediating lineage-specific expression of CD8α. Interestingly, a region approximately 300 bp upstream of the GATA-3 binding site has been implicated in the negative regulation of CD8α expression upon fusion of a CD8+ T cell line with the BW5147 thymoma cell line (101). A recent study indicated a role of this region in regulating the association of the CD8 locus with the nuclear matrix (113). Electromobility shift assays revealed two retarded bands, one of which was specific for T cells. Biochemical purification of the binding proteins identified SATB1, a matrix-attachment region binding protein (114) in the T cell–specific band, while CDP/Cux, another MAR-binding protein that negatively regulates expression of several genes (115–119) was present in the ubiquitously observed gel shift band. MARs are AT-rich stretches of DNA that have high affinity for the nuclear matrix, and they are involved in the transcriptional regulation of gene expression (120–122). Based on protein/DNA interaction studies including missing nucleotide approach, DNase I footprinting and EMSA, a model was suggested by Gottlieb and colleagues in which displacement of CDP/Cux by SATB1 favors transcription of CD8α (113), although recent reports indicate the SATB1 might also be involved in the negative regulation of gene expression (123, 124). The in vivo role of the putative MARs has not been addressed, although recent transgenic reporter expression experiments indicate an involvement of this region in the in vivo regulation of CD8 gene expression as well (125). CD8β PROMOTER The murine CD8β promoter has been analyzed for the presence of cis-regulatory elements (126). By combining the CD8β promoter with the SV40 enhancer in CAT reporter constructs, it has been demonstrated that its activity is restricted to T cells, although whether the activity is specific for the CD8 lineage has not been determined. Deletional analysis of the TATA box-less CD8β promoter revealed that a Sp l binding site (located at position −45 to −40 relative to the transcriptional start site) is necessary for promoter activity. Gel retardation experiments identified multiple complexes of which one appeared to be specific for T cells. Whereas some of the complexes most likely contain the ubiquitously expressed transcription factor Spl (127), the molecular nature of this T cell–specific complex was not determined. CD8 LINEAGE-SPECIFIC ENHANCER ELEMENTS Since analysis of promoterproximal regions failed to identify CD8 lineage-specific cis-regulatory elements, long-range DH site analyses over the whole CD8 locus have been

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

542

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

performed (104). Four clusters (I, II, III, and IV) of DH-sites have been described within a 80-kb murine genomic fragment spanning the whole CD8 locus, including 2 kb upstream of the CD8β gene and 25 kb downstream of the CD8α gene (104). Three of the clusters (clusters II, III, and IV) are specific for T cells (Figure 4), while cluster I (located approximately 20 kb downstream of the CD8α gene) was also observed in DNase I–treated liver nuclei. Transgenic mice generated with this 80-kb genomic fragment displayed appropriate developmental stage- and lineage-specific expression of the transgenic CD8α and β genes, demonstrating that the major cis-regulatory elements are localized within this large genomic fragment (104). A similar result was also reported for the human CD8β gene in transgenic mice generated with a 95-kb genomic fragment encompassing the human CD8β locus (128). In both studies, mosaic expression of the transgene was observed, most likely due to position effect variegation (129). Interference of transgene expression by surrounding genomic regions is usually observed in the absence of locus control regions LCR: cis-acting elements that mediate position-independent and copy number-dependent expression of a transgene (130–132). This might indicate that LCRs are not involved in facilitating expression of the endogenous CD8α and β genes. To characterize the role of individual DH sites within the 80-kb murine CD8 locus, a combination of DH sites or clusters was used to prepare reporter constructs that were analyzed for enhancer activity in transgenic mice (125, 133–136). These studies resulted in the identification of at least four enhancers designated E8I (corresponding to DH sites CIII-1,2), E8II (CIV-4,5), E8III (CIV-3), and E8IV (CIV-1,2) in the CD8 locus that display distinct developmental stage- and lineage-specific patterns of activity (Figure 4). E8I (CIII-1,2): an enhancer specific for mature CD8+ T cells and CD8αα + IEL The enhancer E8I (CIII-1,2), located approximately 16 kb upstream of the murine CD8α gene, was the first cis-acting element shown to have CD8 lineage-restricted enhancer activity. In transgenic mice, E8I (CIII-1,2) is sufficient to direct expression of a reporter gene in mature CD8 SP thymocytes and in CD4−CD8+ T cells, but not in DP thymocytes (133, 134). The tight correlation between onset of enhancer activity and the final steps of positive selection of CD8 SP thymocytes suggests an important role for E8I (CIII-1,2) in directing CD8 expression after completion of positive selection. E8I (CIII-1,2) also directs expression in extrathymically derived CD8αα + IEL of both TCRγ δ and TCRαβ lineages (133). It may therefore be specifically involved in the regulation of CD8α expression, both in thymus-derived and thymus-independent T cells, although a role for E8I (CIII-1,2) in regulating CD8β expression cannot be ruled out.

P2: APR/spd/ary

February 11, 1999 12:20 Annual Reviews

Figure 4 Transcriptional regulatory elements within the murine CD8 locus. Schematic map of the murine CD8 locus (98). The arrangement of DNase I hypersensitivity (DH) site clusters I to IV is according to Hostert and colleagues and the vertical arrows indicate individual DH sites (104). All BamHI (B), but only relevant EcoRI (E), restriction sites are shown as vertical bars. The position and the transcriptional orientation of the CD8α and CD8β genes are indicated by horizontal arrows, and the closed and open bars within the CD8α and CD8β genes indicate coding and noncoding exons, respectively (143, 144). The filled vertical bars below the CD8 locus indicate the genomic fragments used for the generation of transgenic constructs (104, 125, 133–136). The exact location of E8II, E8III, or E8IV (indicated as filled circles) within the BamHI/BamHI, BamHI/EcoRI, or EcoRI/EcoRI fragments (135), respectively, is not known. However, it is likely that they overlap with the DH sites identified within the subfragments (CIV-4,5 for E8II, CIV-3 for E8III, and CIV-1,2, for E8IV). The mature CD8+ T cell enhancer E8I (CIII-1,2) can be divided into two enhancer subelements designated E8IA (CIII-1) and E8IB (CIII-2) based on transgenic reporter expression studies (133–135). The developmental stage, subset and lineage-specificity of the individual enhancers or DH-sites localized on transgenic fragments in different T cell populations is shown on the left and the reference reporting the observed expression pattern is indicated: +, +/−, or − indicate strong, weak, or no expression of the reporter gene, respectively. For CD8+ IEL, both CD8αα and CD8αβ sublineages are shown. Abbreviations: LN: lymph nodes; IEL-CD8+: CD8-expressing intraepithelial lymphocytes of the gut; DP: double-positive thymocytes; nd: not determined.

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SAT/mbg QC: KKK

AR078-17

CD4 AND CD8 GENE REGULATION

543

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

544

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

The deletion of E8I (CIII-1,2) by homologous recombination confirmed a major role for this enhancer in regulating CD8α expression in thymus-independent IEL of both the TCRγ δ and TCRαβ lineages (125, 135). In the absence of E8I (CIII-1,2), a large fraction of TCRγ δ + IEL lost expression of CD8α, and the remaining CD8αα +TCRγ δ + IEL population displayed at least 3- to 5-fold lower surface levels of CD8α. A similar reduction (2- to 3-fold) in surface CD8α was also observed on TCRαβ + IEL. Based on these findings, the role of E8I (CIII-1,2) in regulating CD8 expression in thymus-derived T cells remains unclear. Because deletion of E8I (CIII-1,2) did not affect expression of CD8α and CD8β in thymus-derived T cells (125, 135), it may be argued that E8I (CIII-1,2) is not involved in the regulation of CD8α (and/or CD8β) expression in this lineage and that the enhancer activity observed in transgenic mice does not properly reflect the role of E8I (CIII-1,2) in its endogenous chromosomal location. This possibility cannot be ruled out, although it appears to be unlikely considering that a defined developmental stage-specific induction (at the HSAhiTCRint/hi to HSAloTCRhi transition) of enhancer activity and CD8 subset-specificity of E8I (CIII-1,2) were observed in numerous transgenic founders or lines (133, 134). In addition, CD8 SP thymocytes from mutant mice displayed a decrease of about 20% in surface CD8 levels (135), as compared with wild-type littermate controls, suggesting that this enhancer does function in the thymus (although this small difference in expression was not observed in the periphery). It is clear, however, that other cis-acting elements must be involved in the regulation of CD8 gene expression. Therefore, the results from the E8I (CIII-1,2) knockout suggest that other enhancers are able to fully compensate for loss of E8I (CIII-1,2) activity in thymus-dependent T cell lineages (see below). E8I (CIII-1,2) enhancer activity was initially mapped to a 7.6-kb genomic BamHI fragment that contains two CD8+ T cell line-specific DH sites from DH-cluster III (sites 1 and 2). Sequences corresponding to either site can individually direct CD8 subset-specific expression, indicating that two independent enhancer activities, designated E8IA and E8IB (CIII-1 and CIII-2, respectively), reside on this 7.6-kb genomic fragment (133, 134). Since transactivation assays have failed to detect enhancer activity in vitro (W Ellmeier, DR Littman, unpublished observation), further transgenic reporter expression assays have been used to narrow down to region containing enhancer activity. This has led to further narrowing of the E8IB (CIII-2) activity to a 1.6 kb HindIII fragment that mediates expression in CD8+ T cells and in CD8αα + IEL (135). E8II (CIV-4,5), E8III (CIV-3) and E8IV (CIV-1,2): developmental stage-specific enhancers at the CD8β locus The generation of transgenic mice with genomic

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD4 AND CD8 GENE REGULATION

545

fragments containing all DH sites localized around the murine CD8β gene (cluster IV) demonstrated the presence of other CD8 locus enhancers that direct expression in DP thymocytes and mature CD8-lineage T cells (135). This was independently confirmed by generating transgenic mice with constructs in which DH sites from cluster IV (CIV-1,2,3) were tested in combination with either cluster III (125), cluster II or both (136). By testing combinations of DH sites from cluster IV, it has been possible to dissect the cluster into three distinct enhancer activities (Figure 4). E8II (localized within a 4.3-kb genomic BamHI fragment containing DH sites CIV-4,5) directs expression of a reporter gene both in DP and in CD8 SP thymocytes and in the mature CD8+ T cell subset. Further transgenic reporter analysis is required to determine whether the DP and SP enhancer activities within E8II (CIV-4,5) are physically separable. E8III (CIV-3), which maps to a 4.1-kb BamHI fragment, displays an enhancer activity that is restricted to DP thymocytes and is shut off during positive selection at the TCRint/hi to TCRhi transition (135). An additional level of complexity of CD8 gene regulation is suggested by the finding that the combination of DH site clusters II and III directs expression of a reporter gene not only in CD8+ T cells (as one would expect, since cluster III, containing E8I, directs expression in the mature CD8+ T cell lineage), but also in DP thymocytes (125). Cluster II, which alone cannot direct expression of a reporter gene (104, 136), therefore gains new functionality in conjunction with cluster III, directing expression in DP thymocytes. As suggested from studies of Gottlieb and coworkers (113), the region upstream of the CD8α gene that overlaps with DH site CII-1 may represent a matrix-attachment region (MAR) (113). Thus, it is tempting to speculate that the combined activity of an enhancer (presumably E8I (CIII-1,2) within cluster III) and a MAR (localized within cluster II) facilitates the expression of a reporter gene in immature DP thymocytes. Further transgenic reporter expression studies and deletion of DHsites by homologous recombination in ES cells are necessary to understand the interactions between different cis-regions within clusters II and III and to determine whether cluster II indeed constitutes a MAR. Another example of the complexity of CD8 gene regulation is provided by the observation that an enhancer activity designated E8IV (localized on a 3-kb EcoRI fragment that contains DH sites CIV-1,2) directs expression not only in CD8+ T cells but also in a subset of CD4+ T cells (135). Although this enhancer activity was tested only in combination with either E8III (CIV-3) or E8IV (CIV1,2), this finding may explain the recent observation of low-level expression of CD8β in a subset of CD4+CD8− T cells (104). Expression of a CD8α transgene in the CD4 subset resulted in the appearance of about 25% mature CD4+ T cells that expressed CD8αβ heterodimers, while the majority of other transgenic CD4+ T cells expressed CD8αα homodimers. However, expression

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

546

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

of a transgenic reporter construct containing E8IV (CIV-1,2) in CD4+CD8− T cells has not been observed in other studies (125, 136). Since larger genomic fragments containing additional 50 sequences were used in those studies (see Figure 4), it remains possible that negative cis-acting sequences are present at the CD8β locus, and these suppress E8IV (CIV-1,2) enhancer activity in CD4+ T cells (see below). None of the enhancers identified around the murine CD8β locus was able to direct expression in CD8αα-expressing IEL, indicating that their activity is restricted to the thymus-derived lineage (135). These enhancers are hence unlikely to account for the residual low-level expression of CD8α in IEL of E8I (CIII-1,2)-deficient mice. Other CD8 locus enhancers, yet to be identified, must partially compensate for the loss of E8I (CIII-1,2) in CD8αα-positive IEL. Interestingly, transgenic mice generated with a 95-kb genomic fragment containing the human CD8β locus expressed the transgene in CD8αα +TCRαβ + IEL (128). This suggests that negative regulatory elements, absent from the 95-kb transgene, restrict expression of the human CD8β gene in CD8αα + IEL of the TCRαβ lineage. However, it is possible that regulatory elements for human CD8α expression are localized on this genomic fragment and direct ectopic expression of the human CD8β transgene in CD8αα +TCRαβ + IEL.

Enhancer Specificity in CD8α versus CD8β Gene Expression An issue that has not been addressed in the studies described above relates to the question of how specificity of CD8α and β gene expression in different T cell lineages is achieved. Because the four different enhancers are in close proximity to each other and since CD8α and CD8β display T cell lineage– dependent differences in their expression pattern, a tight regulatory interaction between the CD8α and β genes and the different enhancers must exist. E8I (CIII-1,2), which is necessary and sufficient to direct expression in CD8αα + IEL (133), is likely to function in conjuction with the CD8α promoter and may therefore be specific for regulating CD8α expression. In contrast, E8II (CIV4,5), E8III (CIV-3), and E8IV (CIV-1,2) direct expression in thymus-derived CD8αβ + T cells and may therefore regulate expression of CD8α or CD8β or both. Studies on the regulation of the murine HoxB cluster have recently demonstrated that some promoters can either share or compete for the same enhancer (137). Analyses of promoter and enhancer interactions in Drosophila have also shown that compatibility of these cis-regulatory elements, regulated by properties of the promoter region, is one way to achieve selectivity of gene expression in “gene-dense” areas (138–140). Since the transgenic studies used to identify and characterize enhancers have thus far utilized only the CD8α

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CD4 AND CD8 GENE REGULATION

547

promoter (133–135), it is possible that some of these enhancers may not have displayed their appropriate function if they interact in vivo specifically with the CD8β promoter. Incompatibility between enhancers and promoters could therefore explain why E8I (CIII-1,2), which is active in IEL, fails to direct CD8β gene expression in these T cells. Alternatively, insulator or boundary elements (120–122, 141) may be localized between E8I (CIII-1,2) and CD8β, thus preventing E8I (CIII-1,2) enhancer and CD8β promoter interactions. The generation of a transgenic construct in which E8I (CIII-1,2) is combined with a reporter gene driven by the CD8β promoter will help to distinguish between these two models.

Is There a CD8 Silencer? As summarized above, CD4 subset-specific expression is achieved by the combined activity of a T cell–specific enhancer (50) and a transcriptional silencer that shuts off CD4 expression in developing CD8 SP thymocytes and CD8+ T cells (51, 52). Therefore, one might expect that a similar regulatory mechanism is involved in the transcriptional regulation of CD8α and β gene expression. Indeed, the discovery of a negative cis-regulatory sequence mediating extinction of CD8α expression in a CD4−CD8− thymoma cell line supported this assumption (101). However, this putative negatively acting element, located approximately 4 kb upstream of the CD8α promoter, did not suppress expression of an hCD2 reporter gene in CD4+ T cells when combined with the murine CD4 enhancer and CD4 promoter in a transgenic construct (S Sawada, DR Littman, unpublished result). This result suggested that the negative cis-element does not work in CD4+CD8− T cells, although it remains possible that it is incompatible with heterologous regulatory sequences. In a similar transgenic construct, the 7.6-kb genomic fragment containing the E8I (CIII-1,2) enhancer did not silence the expression in CD4+ T cells of an hCD2 reporter under the control of the CD4 enhancer and CD4 promoter (133) or of the human CD2 locus control region (134). However, it is possible that CD8 locus-specific silencer elements may not be able to override the transcriptional activity of the CD4 enhancer or the hCD2 locus control region. The E8II (CIV-4,5) enhancer directs expression of a reporter gene in DP and CD8 SP thymocytes. This is reminiscent of the expression pattern (in DP and CD4 SP thymocytes) of a reporter gene generated by the combined activity of the CD4 enhancer and CD4 silencer in transgenic mice (51, 52). Therefore, before a final conclusion concerning the presence of a transcriptional silencer at the CD8 locus can be reached, additional transgenic constructs will be required to functionally dissect E8II (CIV-4,5), and also E8I (CIII-1,2) or E8IV (CIV-1,2), and determine whether they contain a potental silencing activity.

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

548

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONCLUSIONS The regulation of CD4 and CD8 gene expression in T cells of the TCRαβ lineage appears to have coevolved with the regulation of the functions of the two major subsets of cells in this lineage. Identification of the trans-acting factors involved in transcription of these coreceptor genes will likely provide important insight into helper versus cytotoxic lineage commitment and into the mechanism of positive selection of thymocytes. Although several basic elements that can fully or partially recapitulate the patterns of CD4 and CD8 expression in thymocyte development have been identified, no relevant trans-acting factors that contribute to subset specificity has yet been described. It is expected that such factors would bind to sequences within the CD4 silencer and CD8 enhancer [E8I (CIII-1,2)] and would be regulated in a manner that is specific for the respective sublineage. Recent results suggest that additional factors differentially regulate expression of the coreceptor genes in double-positive versus single-positive thymocytes. This additional layer of complexity in coreceptor gene regulation may reveal an intimate connection with the positive selection process. To understand the roles of the relevant cis-acting sequences in stage- and lineage-specific regulation of the coreceptor genes, it will be necessary to delete these individually in the murine germ line. Such genetic manipulations will help resolve conflicting data from multiple transgenic studies with CD4 regulatory sequences and will also deepen our understanding of differential regulation of CD8α and β gene expression in thymus-dependent T cells and IEL. The CD4 silencer additionally provides a unique opportunity to study negative transcriptional regulation in a vertebrate system. Conditional manipulation of this locus in mice is likely to result in a greater understanding of general mechanisms of gene repression and of epigenetic regulation. ACKNOWLEDGMENTS The authors would like to thank Derya Unutmaz, Sanjeev Mahanta, Ichiro Taniuchi, and Yong-Rui Zou for comments on the manuscript. We would also like to apologize to those scientists whose work was only indirectly cited in the context of a referenced published review because of space limitations. W Ellmeier was supported by an Erwin-Schr¨odinger postdoctoral fellowship from the Fonds zur F¨orderung der wissenschaftlichen Forschung (Vienna, Austria) and is currently an associate of the Howard Hughes Medical Institute. DR Littman is an investigator of the Howard Hughes Medical Institute. Visit the Annual Reviews home page at http://www.AnnualReviews.org

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

CD4 AND CD8 GENE REGULATION

549

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Literature Cited 1. Kisielow P, von Boehmer H. 1995. Development and selection of T cells: facts and puzzles. Adv. Immunol. 58:87–209 2. Zuniga-Pflucker JC, Lenardo MJ. 1996. Regulation of thymocyte development from immature progenitors. Curr. Opin. Immunol. 8:215–24 3. Marrack P, Kappler J. 1997. Positive selection of thymocytes bearing αβ T cell receptors. Curr. Opin. Immunol. 9:250–55 4. Killeen N, Irving BA, Pippig S, Zingler K. 1998. Signaling checkpoints during the development of T lymphocytes. Curr. Opin. Immunol. 10:360–67 5. Shortman K, Wu L. 1996. Early T lymphocyte progenitors. Annu. Rev. Immunol. 14:29–47 6. Fehling HJ, von Boehmer H. 1997. Early αβ T cell development in the thymus of normal and genetically altered mice. Curr. Opin. Immunol. 9:263–75 7. von Boehmer H, Fehling HJ. 1997. Structure and function of the pre-T cell receptor. Annu. Rev. Immunol. 15:433–52 8. Robey E, Fowlkes BJ. 1998. The αβ versus γ δ T-cell lineage choice. Curr. Opin. Immunol. 10:181–87 9. Rodewald HR, Fehling HJ. 1998. Molecular and cellular events in early thymocyte development. Adv. Immunol. 69:1– 112 10. Haks MC, Krimpenfort P, Borst J, Kruisbeek AM. 1998. The CD3γ chain is essential for development of both the TCRαβ and TCRγ δ lineages. EMBO J. 17:1871–82 11. Clements JL, Yang B, Ross-Barta SE, Eliason SL, Hrstka RF, Williamson RA, Koretzky GA. 1998. Requirement for the leukocyte-specific adapter protein SLP76 for normal T cell development. Science 281:416–19 12. Pivniouk V, Tsitsikov E, Swinton P, Rathbun G, Alt FW, Geha RS. 1998. Impaired viability and profound block in thymocyte development in mice lacking the adaptor protein SLP-76. Cell 94:229–38 13. Anderson G, Moore NC, Owen JJ, Jenkinson EJ. 1996. Cellular interactions in thymocyte development. Annu. Rev. Immunol. 14:73–99 14. Page DM, Kane LP, Onami TM, Hedrick SM. 1996. Cellular and biochemical requirements for thymocyte negative selection. Semin. Immunol. 8:69–82 15. Miceli MC, Parnes JR. 1993. Role of CD4 and CD8 in T cell activation and differentiation. Adv. Immunol. 53:59–122

16. Killeen N, Littman DR. 1996. The regulation and function of the CD4 coreceptor during T lymphocyte development. Curr. Top. Microbiol. Immunol. 205:89–106 17. Zamoyska R. 1998. CD4 and CD8: modulators of T cell receptor recognition of antigen and of immune responses? Curr. Opin. Immunol. 10:82–87 18. Garcia KC, Scott CA, Brunmark A, Carbone FR, Peterson PA, Wilson IA, Teyton L. 1996. CD8 enhances formation of stable T-cell receptor/MHC class I molecule complexes. Nature 384:577–81 19. Gao GF, Tormo J, Gerth UC, Wyer JR, McMichael AJ, Stuart DI, Bell JI, Jones EY, Jakobsen BK. 1997. Crystal structure of the complex between human CD8αα and HLA-A2. Nature 387:630–34 20. Rudd CE, Trevillyan JM, Dasgupta JD, Wong LL, Schlossman SF. 1988. The CD4 receptor is complexed in detergent lysates to a protein-tyrosine kinase (pp58) from human T lymphocytes. Proc. Natl. Acad. Sci. USA 85:5190–94 21. Veillette A, Bookman MA, Horak EM, Bolen JB. 1988. The CD4 and CD8 T cell surface antigens are associated with the internal membrane tyrosine-protein kinase p56lck. Cell 55:301–8 22. Bendelac A, Matzinger P, Seder RA, Paul WE, Schwartz RH. 1992. Activation events during thymic selection. J. Exp. Med. 175:731–42 23. Swat W, Dessing M, von Boehmer H, Kisielow P. 1993. CD69 expression during selection and maturation of CD4+8+ thymocytes. Eur. J. Immunol. 23:739– 46 24. von Boehmer H, Kisielow P. 1993. Lymphocyte lineage commitment: instruction versus selection. Cell 73:207–8 25. Chan CH, Benoist C, Mathis D. 1994. In favor of the selective model of positive selection. Semin. Immunol. 6:241–48 26. Davis CB, Littman DR. 1994. Thymocyte lineage commitment: is it instructed or stochastic? Curr. Opin. Immunol. 6:266– 72 27. von Boehmer H. 1996. CD4/CD8 lineage commitment: back to instruction? J. Exp. Med. 183:713–15 28. Wiest DL, Yuan L, Jefferson J, Benveniste P, Tsokos M, Klausner RD, Glimcher LH, Samelson LE, Singer A. 1993. Regulation of T cell receptor expression in immature CD4+CD8+ thymocytes by p56lck tyrosine kinase: basis for differential signaling by CD4 and CD8 in immature

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

550

29.

30.

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

31.

32.

33. 34.

35.

36.

37.

38.

39.

40.

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN thymocytes expressing both coreceptor molecules. J. Exp. Med. 178:1701–12 Ravichandran KS, Burakoff SJ. 1994. Evidence for differential intracellular signaling via CD4 and CD8 molecules. J. Exp. Med. 179:727–32 Itano A, Salmon P, Kioussis D, Tolaini M, Corbella P, Robey E. 1996. The cytoplasmic domain of CD4 promotes the development of CD4 lineage T cells. J. Exp. Med. 183:731–41 Matechak EO, Killeen N, Hedrick SM, Fowlkes BJ. 1996. MHC class II-specific T cells can develop in the CD8 lineage when CD4 is absent. Immunity 4:337– 47 Basson AM, Bommhardt U, Cole MS, Tso JY, Zamoyska R. 1998. CD3 ligation on immature thymocytes generates antagonist-like signals appropriate for CD8 lineage commitment, independently of T cell receptor specificity. J. Exp. Med. 187:1249–60 Killeen N, Littman DR. 1993. Helper T cell development in the absence of CD4− p56lck association. Nature 364:729–32 Chan IT, Limmer A, Louie MC, Bullock ED, Fung-Leung WP, Mak TW, Loh DY. 1993. Thymic selection of cytotoxic T cells independent of CD8α-Lck association. Science 261:1581–84 Chan SH, Cosgrove D, Waltzinger C, Benoist C, Mathis D. 1993. Another view of the selective model of thymocyte selection. Cell 73:225–36 Davis CB, Killeen N, Crooks ME, Raulet D, Littman DR. 1993. Evidence for a stochastic mechanism in the differentiation of mature subsets of T lymphocytes. Cell 73:237–47 van Meerwijk JP, Germain RN. 1993. Development of mature CD8+ thymocytes: selection rather than instruction? Science 261:911–15 Lucas B, Vasseur F, Penit C. 1995. Stochastic coreceptor shut-off is restricted to the CD4 lineage maturation pathway. J. Exp. Med. 181:1623– 33 Lundberg K, Heath W, Kontgen F, Carbone FR, Shortman K. 1995. Intermediate steps in positive selection: differentiation of CD4+8int TCRint thymocytes into CD4−8+TCRhi thymocytes. J. Exp. Med. 181:1643–51 Suzuki H, Punt JA, Granger LG, Singer A. 1995. Asymmetric signaling requirements for thymocyte commitment to the CD4+ versus CD8+ T cell lineages: a new perspective on thymic commitment and selection. Immunity 2:413–25

41. Benveniste P, Knowles G, Cohen A. 1996. CD8/CD4 lineage commitment occurs by an instructional/default process followed by positive selection. Eur. J. Immunol. 26:461–71 42. Lucas B, Germain RN. 1996. Unexpectedly complex regulation of CD4/CD8 coreceptor expression supports a revised model for CD4+CD8+ thymocyte differentiation. Immunity 5:461–77 43. Robey E, Chang D, Itano A, Cado D, Alexander H, Lans D, Weinmaster G, Salmon P. 1996. An activated form of Notch influences the choice between CD4 and CD8 T cell lineages. Cell 87:483–92 44. Baron A, Hafen K, von Boehmer H. 1994. A human CD4 transgene rescues CD4− CD8+ cells in β2-microglobulin-deficient mice. Eur. J. Immunol. 24:1933–36 45. Chan SH, Waltzinger C, Baron A, Benoist C, Mathis D. 1994. Role of coreceptors in positive selection and lineage commitment. EMBO J. 13:4482–89 46. Corbella P, Moskophidis D, Spanopoulou E, Mamalaki C, Tolaini M, Itano A, Lans D, Baltimore D, Robey E, Kioussis D. 1994. Functional commitment to helper T cell lineage precedes positive selection and is independent of T cell receptor MHC specificity. Immunity 1:269–76 47. Robey E, Itano A, Fanslow WC, Fowlkes BJ. 1994. Constitutive CD8 expression allows inefficient maturation of CD4+ helper T cells in class II major histocompatibility complex mutant mice. J. Exp. Med. 179:1997–2004 48. Paillard F, Sterkers G, Vaquero C. 1990. Transcriptional and post-transcriptional regulation of TCR, CD4 and CD8 gene expression during activation of normal human T lymphocytes. EMBO J. 9:1867– 72 49. Takahama Y, Singer A. 1992. Posttranscriptional regulation of early T cell development by T cell receptor signals. Science 258:1456–62 50. Sawada S, Littman DR. 1991. Identification and characterization of a T cellspecific enhancer adjacent to the murine CD4 gene. Mol. Cell. Biol. 11:5506–15 51. Sawada S, Scarborough JD, Killeen N, Littman DR. 1994. A lineage-specific transcriptional silencer regulates CD4 gene expression during T lymphocyte development. Cell 77:917–29 52. Siu G, Wurster AL, Duncan DD, Soliman TM, Hedrick SM. 1994. A transcriptonal silencer controls the developmental expression of the CD4 gene. EMBO J. 13:3570–79 53. Siu G, Wurster AL, Lipsick JS, Hedrick

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

CD4 AND CD8 GENE REGULATION

54.

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

55.

56.

57.

58.

59.

60.

61. 62.

63.

64.

SM. 1992. Expression of the CD4 gene requires a Myb transcription factor. Mol. Cell. Biol. 12:1592–604 Nakayama K, Yamamoto R, Ishii S, Nakauchi H. 1993. Binding of c-Myb to the core sequence of the CD4 promoter. Int. Immunol. 5:817–24 Duncan DD, Stupakoff A, Hedrick SM, Marcu KB, Siu G. 1995. A Myc-associated zinc finger protein binding site is one of four important functional regions in the CD4 promoter. Mol. Cell. Biol. 15:3179– 86 Salmon P, Giovane A, Wasylyk B, Klatzmann D. 1993. Characterization of the human CD4 gene promoter: transcription from the CD4 gene core promoter is tissue-specific and is activated by Ets proteins. Proc. Natl. Acad. Sci. USA 90: 7739–43 Gillespie FP, Doros L, Vitale J, Blackwell C, Gosselin J, Snyder BW, Wadsworth SC. 1993. Tissue-specific expression of human CD4 in transgenic mice. Mol. Cell. Biol. 13:2952–58 Killeen N, Sawada S, Littman DR. 1993. Regulated expression of human CD4 rescues helper T cell development in mice lacking expression of endogenous CD4. EMBO J. 12:1547–53 Hanna Z, Simard C, Laperriere A, Jolicoeur P. 1994. Specific expression of the human CD4 gene in mature CD4+CD8− and immature CD8+CD8+ T cells and in macrophages of transgenic mice. Mol. Cell. Biol. 14:1084–94 Sawada S, Littman DR. 1993. A heterodimer of HEB and an E12- related protein interacts with the CD4 enhancer and regulates its activity in T-cell lines. Mol. Cell. Biol. 13:5620–28 Clevers H, van de Wetering M. 1997. TCF/LEF factor earn their wings. Trends Genet. 13:485–9 Murre C, McCaw PS, Vaessin H, Caudy M, Jan LY, Jan YN, Cabrera CV, Buskin JN, Hauschka SD, Lassar AB, et al. 1989. Interactions between heterologous helixloop-helix proteins generate complexes that bind specifically to a common DNA sequence. Cell 58:537–44 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 Hu JS, Olson EN, Kingston RE. 1992. HEB, a helix-loop-helix protein related to E2A and ITF2 than can modulate the DNA-binding ability of myogenic regulatory factors. Mol. Cell. Biol. 12:1031–42

551

65. 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 66. Blum MD, Wong GT, Higgins KM, Sunshine MJ, Lacy E. 1993. Reconstitution of the subclass-specific expression of CD4 in thymocytes and peripheral T cells of transgenic mice: identification of a human CD4 enhancer. J. Exp. Med. 177:1343–58 67. Wurster AL, Siu G, Leiden JM, Hedrick SM. 1994. Elf-1 binds to a critical element in a second CD4 enhancer. Mol. Cell. Biol. 14:6452–63 68. Ansari-Lari MA, Muzny DM, Lu J, Lu F, Lilley CE, Spanos S, Malley T, Gibbs RA. 1996. A gene-rich cluster between the CD4 and triosephosphate isomerase genes at human chromosome 12p13. Genome Res. 6:314–26 69. Bruniquel D, Borie N, Triebel F. 1997. Genomic organization of the human LAG-3/CD4 locus. Immunogenetics 47:96–8 70. Triebel F, Jitsukawa S, Baixeras E, Roman-Roman S, Genevee C, ViegasPequignot E, Hercend T. 1990. LAG-3, a novel lymphocyte activation gene closely related to CD4. J. Exp. Med. 171:1393– 405 71. Kazazi F, Mathijs JM, Foley P, Cunningham AL. 1989. Variations in CD4 expression by human monocytes and macrophages and their relationships to infection with the human immunodeficiency virus. J. Gen. Virol. 70:2661–72 72. Wood GS, Warner NL, Warnke RA. 1983. Anti-Leu-3/T4 antibodies react with cells of monocyte/macrophage and Langerhans lineage. J. Immunol. 131:212–16 73. Crocker PR, Jefferies WA, Clark SJ, Chung LP, Gordon S. 1987. Species heterogeneity in macrophage expression of the CD4 antigen. J. Exp. Med. 166:613– 18 74. Lucey DR, Dorsky DI, Nicholson-Weller A, Weller PF. 1989. Human eosinophils express CD4 protein and bind human immunodeficiency virus 1 gp120. J. Exp. Med. 169:327–32 75. Sands JF, Nikolic-Zugic J. 1992. T cellspecific protein-DNA interactions occurring at the CD4 locus: identification of possible transcriptional control elements of the murine CD4 gene. Int. Immunol. 4:1183–94 76. Onishi M, Nagayoshi K, Kitamura K, Hirai H, Takaku F, Nakauchi H. 1993. CD4dull/+ hematopoietic progenitor cells

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

552

77.

78.

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

79. 80. 81.

82. 83. 84. 85. 86. 87.

88. 89. 90. 91.

92.

93.

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN in murine bone marrow. Blood 81:3217– 25 Wu L, Scollay R, Egerton M, Pearse M, Spangrude GJ, Shortman K. 1991. CD4 expressed on earliest T-lineage precursor cells in the adult murine thymus. Nature 349:71–74 Donda A, Schulz M, Burki K, De Libero G, Uematsu Y. 1996. Identification and characterization of a human CD4 silencer. Eur. J. Immunol. 26:493–500 Duncan DD, Adlam M, Sui G. 1996. Asymmetric redundancy in CD4 silencer function. Immunity 4:301–11 Karpen GH. 1994. Position-effect variegation and the new biology of heterochromatin. Curr. Opin. Genet. Dev. 4:281–91 Weiler KS, Wakimoto BT. 1995. Heterochromatin and gene expression in Drosophila. Annu. Rev. Genet. 29:577– 605 Kass SU, Pruss D, Wolffe AP. 1997. How does DNA methylation repress transcription? Trends Genet. 13:444–49 Pazin MJ, Kadonaga JT. 1997. What’s up and down with histone deacetylation and transcription? Cell 89:325–28 Siegfried Z, Cedar H. 1997. DNA methylation: a molecular lock. Curr. Biol. 7:305–7 Nan X, Cross S, Bird A. 1998. Gene silencing by methyl-CpG-binding proteins. Novartis Found. Symp. 214:6–16 Struhl K. 1998. Histone acetylation and transcriptional regulatory mechanisms. Genes. Dev. 12:599–606 Kimble J, Simpson P. 1997. The LIN12/Notch signaling pathway and its regulation. Annu. Rev. Cell Dev. Biol. 13:333– 61 Kopan R, Cagan R. 1997. Notch on the cutting edge. Trends Genet. 13:465–67 Greenwald I. 1998. LIN-12/Notch signaling: lessons from worms and flies. Genes Dev. 12:1751–62 Kim H, Siu G. 1998. The notch pathway intermediate HES-1 silences CD4 gene expression. Mol. Cell. Biol. 18:7166–75 Salmon P, Boyer O, Lores P, Jami J, Klatzmann D. 1996. Characterization of an intronless CD4 minigene expressed in mature CD4 and CD8 T cells, but not expressed in immature thymocytes. J. Immunol. 156:1873–79 Adlam M, Duncan DD, Ng DK, Siu G. 1997. Positive selection induces CD4 promoter and enhancer function. Int. Immunol. 9:877–87 Boyer O, Zhao JC, Cohen JL, Depetris D, Yagello M, Lejeune L, Bruel S, Mattei MG, Klatzmann D. 1997. Position-

94.

95.

96. 97.

98.

99.

100.

101.

102.

103.

104.

105.

dependent variegation of a CD4 minigene with targeted expression to mature CD4+ T cells. J. Immunol. 159:3383–90 Uematsu Y, Donda A, De Libero G. 1997. Thymocytes control the CD4 gene differently from mature T lymphocytes. Int. Immunol. 9:179–87 Jarry A, Cerf-Bensussan N, Brousse N, Selz F, Guy-Grand D. 1990. Subsets of CD3+ (T cell receptor αβ or γ δ) and CD3− lymphocytes isolated from normal human gut epithelium display phenotypical features different from their counterparts in peripheral blood. Eur. J. Immunol. 20:1097–103 Lefrancois L. 1991. Phenotypic complexity of intraepithelial lymphocytes of the small intestine. J. Immunol. 147:1746–51 Moebius U, Kober G, Griscelli AL, Hercend T, Meuer SC. 1991. Expression of different CD8 isoforms on distinct human lymphocyte subpopulations. Eur. J. Immunol. 21:1793–800 Gorman SD, Sun YH, Zamoyska R, Parnes JR. 1988. Molecular linkage of the Ly-3 and Ly-2 genes. Requirement of Ly2 for Ly-3 surface expression. J. Immunol. 140:3646–53 Carbone AM, Marrack P, Kappler JW. 1988. Remethylation at sites 50 of the murine Lyt-2 gene in association with shutdown of Lyt-2 expression. J. Immunol. 141:1369–75 Wilkinson MF, Doskow J, von Borstel RD, Fong AM, MacLeod CL. 1991. The expression of several T cell-specific and novel genes is repressed by trans-acting factors in immature T lymphoma clones. J. Exp. Med. 174:269–80 Lee WH, Banan M, Harriss JV, Hwang I, Woodward E, Youn HJ, Gottlieb PD. 1994. Cis-acting DNA elements and cell type-specific nuclear proteins which may play a role in regulation of mouse CD8α (Lyt-2) gene transcription. Int. Immunol. 6:1307–21 Gu JJ, Gottlieb PD. 1992. Inducible functions in hybrids of a Lyt-2+ BW5147 transfectant and the 2C CTL line. Immunogenetics 36:283–93 Hwang I, Gu JJ, Gottlieb PD. 1993. Differential susceptibility of mouse Lyt-2 and Lyt-3 genes to negative regulation. Immunogenetics 37:129–34 Hostert A, Tolaini M, Festenstein R, McNeill L, Malissen B, Williams O, Zamoyska R, Kioussis D. 1997. A CD8 genomic fragment that directs subsetspecific expression of CD8 in transgenic mice. J. Immunol. 158:4270–81 Gao MH, Walz M, Kavathas PB. 1996.

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

12:20

QC: KKK

Annual Reviews

AR078-17

CD4 AND CD8 GENE REGULATION

106.

107.

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

108.

109.

110.

111.

112.

113.

114.

115.

116.

Post-transcriptional regulation associated with control of human CD8α expression of CD4+ T cells. Immunogenetics 45:130–35 Paliard X, Malefijt RW, de Vries JE, Spits H. 1988. Interleukin-4 mediates CD8 induction on human CD4+ T-cell clones. Nature 335:642–44 Gao MH, Kavathas PB. 1993. Functional importance of the cyclic AMP response element-like decamer motif in the CD8α promoter. J. Immunol. 150:4376–85 Hambor JE, Mennone J, Coon ME, Hanke JH, Kavathas P. 1993. Identification and characterization of an Alu-containing, Tcell-specific enhancer located in the last intron of the human CD8α gene. Mol. Cell. Biol. 13:7056–70 Hanke JH, Hambor JE, Kavathas P. 1995. Repetitive Alu elements form a cruciform structure that regulates the function of the human CD8α T cell-specific enhancer. J. Mol. Biol. 246:63–73 Kieffer LJ, Bennett JA, Cunningham AC, Gladue RP, McNeish J, Kavathas PB, Hanke JH. 1996. Human CD8α expression in NK cells but not cytotoxic T cells of transgenic mice. Int. Immunol. 8:1617– 26 Landry DB, Engel JD, Sen R. 1993. Functional GATA-3 binding sites within murine CD8 alpha upstream regulatory sequences. J. Exp. Med. 178:941–49 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 Banan M, Rojas IC, Lee WH, King HL, Harriss JV, Kobayashi R, Webb CF, Gottlieb PD. 1997. Interaction of the nuclear matrix-associated region (MAR)-binding proteins, SATB1 and CDP/Cux, with a MAR element (L2a) in an upstream regulatory region of the mouse CD8α gene. J. Biol. Chem. 272:18440–52 Dickinson LA, Joh T, Kohwi Y, KohwiShigematsu T. 1992. A tissue-specific MAR/SAR DNA-binding protein with unusual binding site recognition. Cell 70:631–45 Superti-Furga G, Barberis A, Schaffner G, Busslinger M. 1988. The −117 mutation in Greek HPFH affects the binding of three nuclear factors to the CCAAT region of the gamma-globin gene. EMBO J. 7:3099–107 Skalnik DG, Strauss EC, Orkin SH. 1991. CCAAT displacement protein as a repressor of the myelomonocytic-specific gp91-phox gene promoter. J. Biol. Chem. 266:16736–44

553

117. Lievens PM, Donady JJ, Tufarelli C, Neufeld EJ. 1995. Repressor activity of CCAAT displacement protein in HL-60 myeloid leukemia cells. J. Biol. Chem. 270:12745–50 118. Mailly F, Berube G, Harada R, Mao PL, Phillips S, Nepveu A. 1996. The human cut homeodomain protein can repress gene expression by two distinct mechanisms: active repression and competition for binding site occupancy. Mol. Cell. Biol. 16:5346–57 119. Higgy NA, Tarnasky HA, Valarche I, Nepveu A, van der Hoorn FA. 1997. Cux/CDP homeodomain protein binds to an enhancer in the rat c-mos locus and represses its activity. Biochim. Biophys. Acta 1351:313–24 120. Ernst P, Smale ST. 1995. Combinatorial regulation of transcription. I: General aspects of transcriptional control. Immunity 2:311–19 121. Gerasimova TI, Corces VG. 1996. Boundary and insulator elements in chromosomes. Curr. Opin. Genet. Dev. 6:185–92 122. Geyer PK. 1997. The role of insulator elements in defining domains of gene expression. Curr. Opin. Genet. Dev. 7:242–48 123. Kohwi-Shigematsu T, Maass K, Bode J. 1997. A thymocyte factor SATB1 suppresses transcription of stably integrated matrix-attachment region-linked reporter genes. Biochemistry 36:12005–10 124. Liu J, Bramblett D, Zhu Q, Lozano M, Kobayashi R, Ross SR, Dudley JP. 1997. The matrix attachment region-binding protein SATB1 participates in negative regulation of tissue-specific gene expression. Mol. Cell. Biol. 17:5275–87 125. Hostert A, Garefalaki A, Mavria G, Tolaini M, Roderick K, Norton T, Mee PJ, Tybulewicz VLJ, Coles M, Kioussis D. 1998. Hierarchical interactions of control elements determine CD8a gene expression in subsets of thymocytes and peripheral T cells. Immunity 9:497–508 126. Kawachi Y, Otsuka F, Nakauchi H. 1996. Characterization of the mouse CD8β chain-encoding gene promoter region. Immunogenetics 44:358–65 127. Kadonaga JT, Carner KR, Masiarz FR, Tjian R. 1987. Isolation of cDNA encoding transcription factor Sp1 and functional analysis of the DNA binding domain. Cell 51:1079–90 128. Kieffer LJ, Yan L, Hanke JH, Kavathas PB. 1997. Appropriate developmental expression of human CD8β in transgenic mice. J. Immunol. 159:4907–12 129. Festenstein R, Tolaini M, Corbella P, Mamalaki C, Parrington J, Fox M, Miliou A,

P1: SAT/mbg

P2: APR/spd/ary

February 11, 1999

554

130.

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

131.

132.

133.

134.

135.

136.

12:20

QC: KKK

Annual Reviews

AR078-17

ELLMEIER, SAWADA & LITTMAN Jones M, Kioussis D. 1996. Locus control region function and heterochromatininduced position effect variegation. Science 271:1123–25 Grosveld F, van Assendelft GM, Greaves DR, Kollias G. 1987. Position-independent, high-level expression of the human β-globin gene in transgenic mice. Cell 51:975–85 Greaves DR, Wilson FD, Lang G, Kioussis D. 1989. Human CD2 30 -flanking sequences confer high-level, T cell-specific, position-independent gene expression in transgenic mice. Cell 56:979–86 Martin DI, Fiering S, Groudine M. 1996. Regulation of β-globin gene expression: straightening out the locus. Curr. Opin. Genet. Dev. 6:488–95 Ellmeier W, Sunshine MJ, Losos K, Hatam F, Littman DR. 1997. An enhancer that directs lineage-specific expression of CD8 in positively selected thymocytes and mature T cells. Immunity. 7:537–47 Hostert A, Tolaini M, Roderick K, Harker N, Norton T, Kioussis D. 1997. A region in the CD8 gene locus that directs expression to the mature CD8 T cell subset in transgenic mice. Immunity 7:525–36 Ellmeier W, Sunshine MJ, Losos K, Littman DR. 1998. Multiple developmental stage-specific enhancers regulate CD8 expression in developing thymocytes and in thymus-independent T cells. Immunity 9:485–96 Zhang XL, Seong R, Piracha R, Larijani M, Heeney M, Parnes JR, Chamberlain JW. 1998. Distinct stage-specific cisactive transcriptional mechanisms control expression of T cell coreceptor CD8 alpha at double- and single-positive stages

137.

138.

139.

140.

141.

142.

143.

144.

of thymic development. J. Immunol. 161:2254–66 Sharpe J, Nonchev S, Gould A, Whiting J, Krumlauf R. 1998. Selectivity, sharing and competitive interactions in the regulation of Hoxb genes. EMBO J. 17:1788–98 Li X, Noll M. 1994. Compatibility between enhancers and promoters determines the transcriptional specificity of gooseberry and gooseberry neuro in the Drosophila embryo. EMBO J. 13:400– 6 Merli C, Bergstrom DE, Cygan JA, Blackman RK. 1996. Promoter specificity mediates the independent regulation of neighboring genes. Genes Dev. 10:1260– 70 Ohtsuki S, Levine M, Cai HN. 1998. Different core promoters possess distinct regulatory activities in the Drosophila embryo. Genes Dev. 12:547–56 Felsenfeld G, Boyes J, Chung J, Clark D. Studitsky V. 1996. Chromatin structure and gene expression. Proc. Natl. Acad. Sci. USA 93:9384–88 Gorman SD, Tourvieille B, Parnes JR. 1987. Structure of the mouse gene encoding CD4 and an unusual transcript in brain. Proc. Natl. Acad. Sci. USA 84:7644–48 Liaw CW, Zamoyska R, Parnes JR. 1986. Structure, sequence, and polymorphism of the Lyt-2 T cell differentiation antigen gene. J. Immunol. 137:1037–43 Nakayama K, Shinkai YI, Okumura K, Nakauchi H. 1989. Isolation and characterization of the mouse CD8β chain (Ly-3) genes. Absence of an intervening sequence between V- and J-like gene segments. J. Immunol. 142:2540–46

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:523-554. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SAT/spd

P2: KKK/plb

January 21, 1999

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999. 17:555–92 c 1999 by Annual Reviews. All rights reserved Copyright °

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETIC ANALYSIS OF B CELL ANTIGEN RECEPTOR SIGNALING Tomohiro Kurosaki Department of Molecular Genetics, Institute for Liver Research, Kansai Medical University, 10-15 Fumizono-cho, Moriguchi 570-8506, Japan; e-mail: [email protected] KEY WORDS:

signal transduction, BCR, Syk, Btk, BLNK, PLC-γ 2, PI-3K, calcium signal

ABSTRACT In B lymphocytes, a signaling complex that contributes to cell fate decisions is the B cell antigen receptor (BCR). Data from knockout experiments in cell lines and mice have revealed distinct functions for the intracellular protein tyrosine kinases (Lyn, Syk, Btk) in BCR signaling and B cell development. Combinations of intracellular signaling pathways downstream of these PTKs determine the quality and quantity of BCR signaling. For example, concerted actions of the PLC-γ 2 and PI3-K pathways are required for proper calcium responses. Similarly, the regulation of ERK and JNK responses involves both PLC-γ 2 and GTPases pathways. Since the immune response in vivo is regulated by alteration of these signaling outcomes, achieving a precise understanding of intracellular molecular events leading to B lymphocyte proliferation, deletion, anergy, receptor editing, and survival still remains a challenge for the future.

INTRODUCTION The B cell antigen receptor (BCR) mediates the response of B lymphocytes to foreign antigen. The existence of the BCR was a central prediction of the clonal selection hypothesis of Burnet (1) and Talmage (2) that the diversity in the immune response is due to the selection and expansion of genetically committed antibody-producing cells (B lymphocytes). Although originally invoked to account for the proposed antigen-specific activation of B cell clones, it has become clear that the BCR plays a central role in determining the fate of B cells even before it encounters antigen (3). For instance, progression through 555 0732-0582/99/0410-0555$08.00

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

556

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

the pre-B cell stage of differentiation (during which antigen receptors are being assembled) depends on the presence of the pre-BCR, which is composed of functional heavy chains and surrogate light chains with their signaling subunits, Igα and Igβ. The outcomes of these developmental responses as well as antigen-specific responses are mediated by signal transduction through the BCR. Transmembrane signaling is further regulated or fine-tuned by an array of cytoplasmic signal transduction molecules that amplify or dampen signals during B cell development. Key cytoplasmic regulators include kinases, phosphatases, lipid metabolizing enzymes, and adaptor molecules that greatly influence the strength and quality of transmembrane signaling events. Cell surface regulators such as CD19, CD22, PIR-B, and Fcγ RIIB also inform B cells of their extracellular microenvironment, thereby defining signaling thresholds within B cells. The purpose of this review is to summarize the major advances made in defining the functions of cytoplasmic and cell surface regulatory proteins and to discuss the mechanisms by which the BCR induces such heterogenous cellular responses as cell survival, death, and receptor editing.

FUNCTION OF THE BCR COMPONENT The BCR is characterized by a complex hetero-oligomeric structure in which ligand binding and signal transduction are compartmentalized into distinct receptor subunits. The ligand-binding portion of this receptor is membrane immunoglobulin (mIg), which is a tetrameric complex of Ig heavy (H) and light (L) chains; the signal transduction component comprises a disulfide-bonded heterodimer of the Igα (CD79α) and Igβ (CD79β) molecules. In developing B cells, rearrangement of the Ig heavy chain (H) locus variable (V), diversity (D), and joining (J) segment occurs before light chain rearrangement. In this transitional stage, membrane-bound Igµ (mIgµ) associates with surrogate light chains (λ5 and VpreB) and with the Igα-Igβ to form a preBCR complex, the presence of which allows cell lineage progression to the pre-B cell stage as well as allelic exclusion (Figure 1). Indeed, targeted disruption of the mIgµ transmembrane domain (mIgµT−/−) results in loss of H chain allelic exclusion as well as a specific developmental block at the pro-B cell stage (4, 5), whereas λ5 mutant (λ5−/−) mice showed impaired, but not completely blocked, B cell development (6, 7) (Figure 1). Perhaps the most critical insights of the past decade in the study of BCR signal transduction have been the discovery of the immunoreceptor tyrosine-based activation motif (ITAM) within the Igα and Igβ subunits of the BCR (8), and recognition of the activation of protein tyrosine kinases (PTKs). The ITAMs in

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34 Annual Reviews

Figure 1 Function of BCR components and signaling molecules in B cell development.

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SAT/spd T1: KKK

AR078-18

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

557

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

558

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

the cytoplasmic regions of Igα and Igβ are characterized by six conserved amino acids (D/Ex7D/Ex2Yx2L/Ix7Yx2L/I). The function of the ITAMs was revealed in experiments in which chimeric receptors containing ITAMs were constructed and expressed in cultured lymphocytes. Clustering of these chimeric receptors using anti-receptor antibodies induced signaling events including PTK activation and calcium mobilization. This signaling function was abolished by mutations that changed either of the two conserved tyrosine residues of the ITAM to phenylalanine, indicating the importance of phosphorylation on these tyrosines (9, 10). Introducing similar chimeric receptors in recombination activating (RAG)−/− (11) or mIgµT−/− (12) mice provided the compelling evidence that signaling through a preBCR complex mediates the activation of B cell development. A mIgµ mutant (mIgµm) that is unable to associate with endogenous Igα/Igβ fails to signal for allelic exclusion and developmental progression to the pre-B cell stage. Chimeric molecules composed of mIgµm and the cytoplasmic domain of Igβ or Igα are sufficient to overcome this developmental block; more importantly, functional reconstitution is dependent on tyrosine residues within the ITAM (11). The effects of Igα and Igβ on B cell development were more directly examined by in vivo targeting experiments. Although Igβ −/− mice still have endogenous Igα, because of the inability of Igα to form a homodimeric signaling complex, this Igβ defect eventually results in the functional knockout of both Igβ and Igα. Mice deficient in Igβ exhibit a complete impairment in B cell development. Moreover, this block occurs before variable (VH) to diversity joining (DJH) recombination (13). There are at least two potential explanations for the Igβ −/− phenotype. The first is that Igα and Igβ may also perform signaling functions as part of a surrogate BCR (as proposed for Igα/Igβcalnexin) and that signaling through this surrogate BCR would be required to make cells competent to undergo V-DJ rearrangement. Supporting this concept, cross-linking Igβ in vivo in Rag-2-deficient mice induced progression from the pro-B to pre-B cell stages (14). The second explanation is that, although V-DJ rearrangement occurs in Igβ −/− mice, albeit at low levels, selection and amplification of productive rearrangements require Igα/Igβ. In addition to a requirement for Igβ in the efficient completion of V-DJ recombination, Igβ is also essential for negative selection of B cells that carry antibody genes encoding truncated Ig heavy chains (Dµ). Dµ is the truncated form of mIgµ, which is encoded by DJH rearrangements in reading frame 2 (RF2). In wild-type mice, this Dµ complex produces allelic exclusion, yet such cells cannot develop to any further differentiated stage because they are selected against. The readout of this negative selection is an under-representation of Igs with variable regions encoded in RF2 (Figure 1). Based on the observation of no bias against RF2 in either DJ or VDJ

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

559

sequences from sorted Igβ −/− pro-B cells (15), Nussenzweig and co-workers proposed that signals through Igβ are required for deleting the Dµ complex. The signaling capacity of Igα was examined by deleting its cytoplasmic domain. In the cytoplasmic Igα −/− mice, the BCR was properly assembled and was composed of mIgµ associated with complete Igβ and truncated Igα. These mice showed apparently normal pre-B cell development, whereas the number of peripheral B cells was decreased tenfold (16). This may be due to decreased persistent signals delivered by BCR. The requirement of maintained BCR expression in order for a mature B cell to persist has been demonstrated by using the technique of inducible gene targeting to delete the rearranged VH gene in mature B cells (17). This persistent signal differs from the activation signal in that it does not lead to entry into the cell cycle or to an upregulation of the costimulatory molecules-changes that occur when the BCR is cross-linked with high-affinity antigen or anti-immunoglobulin. Hence, two distinct signals are thought to be delivered by the BCR: a persistence signal and an activation signal. Presumably this distinction lies in the nature and extent of BCR cross-linking and/or in the developmental stages of B cells. Thus, the simplest explanation for the decreased number of peripheral B cells in cytoplasmic Igα −/− mice is that the persistence signal through BCR requires both Igα and Igβ while Igβ alone is sufficient for pre-B cell development. Whether this differential requirement for B cell development reflects quantitative or qualitative differences of signaling activity between Igα/Igβ and Igβ alone remains to be determined. Collectively, these series of elegant experiments show that the strength and quality of signals transmitted by the surrogate BCR, pre-BCR, and BCR affect the fate of B cells in many different ways, depending on the maturation stages of B cells.

FUNCTION OF PTKs AND PTPases As the BCR has no intrinsic PTK activity, this receptor utilizes several distinct families of cytoplasmic PTKs and PTPases. Indeed, three distinct types of PTKs have been found to be activated upon BCR engagement: the Src-PTKs (Lyn, Blk, and Fyn), Syk, and Btk. Recent gene-targeting experiments have dissected the functions of these PTKs and PTPases in BCR signaling as well as in B cell development.

Lyn Lyn−/− mice exhibit apparently normal B cell development in the bone marrow, whereas the number of peripheral B cells is decreased with an increased proportion of immature B cells (18, 19). Enhanced ERK activation and a subsequent hyperproliferative response upon BCR engagement was observed in these mice (20). Furthermore, Lyn−/− mice carrying transgenic BCRs (anti-HEL)

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

560

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

exhibited exaggerated negative selection responses to self-antigen, as expected from the enhanced signaling properties of Lyn−/− mice (21). Thus, the production of auto-antibodies in Lyn−/− mice is somewhat surprising. Since the antigen-independent formation of plasma cells is greatly exaggerated in the absence of Lyn, this may reflect an elevation of spontaneous plasma cell differentiation. An explanation for enhanced activation of B cells in vivo in the absence of Lyn has been envisaged by an analysis of CD22 knockout mice. CD22 is known to inhibit BCR-induced responses; indeed, CD22 knockout mice exhibit hyperresponsiveness to BCR stimulation (22–25). The inhibitory signal is mediated through the phosphorylation of the immunoreceptor tyrosine-based inhibitory motif (ITIM) located in the cytoplasmic domains of CD22 (26). Assuming that Lyn is the PTK responsible for the phosphorylation of CD22 ITIMs, loss of Lyn decreases both its phosphorylation and subsequent SHP-1 recruitment, resulting in a dampening of the inhibitory signals of CD22 and a subsequent hyperresponsiveness. In fact, tyrosine phosphorylation of CD22 and its subsequent association with SHP-1 were inhibited in Lyn−/− mice (27, 28). Moreover, using BCR transgenic system (anti-HEL), genetic interactions among Lyn, CD22, and SHP-1 have been examined by intercrossing heterozygotes for the mutation of these genes, demonstrating that negative selection in the presence of sHEL is exaggerated in a cumulative manner by 50% loss of these three gene products (21). Therefore, these data indicate that Lyn, CD22, and SHP-1 operate in a single pathway and that each of three elements is a limiting component. Collectively, these experiments indicate that BCR-activated Lyn subserves both a negative regulatory role (by phosphorylating inhibitory receptors such as CD22, which dampens BCR signaling) and positive regulatory roles.

Syk In contrast to Lyn−/− mice, in Syk−/− mice bone marrow B cells accumulate at the late pro-B stage, suggesting a block in B cell development at the pro-B to pre-B cell transition (29, 30). Although Rag2−/− mice show a similar B cell developmental arrest, Syk−/− mice exhibit the presence of rearrangements in the Ig H chain. These data indicate that Syk is required for signaling through the pre-BCR and thus for the clonal expansion of pre-B cells that have productively rearranged the Ig H chain genes. Moreover, like the Igβ −/− mice discussed above, Syk−/− mice show no bias against RF2 in DJ rearrangements (30). Thus, Syk participates in negative selection of the Dµ complex as well as positive selection for clonal expansion of pre-B cells expressing successfully rearranged Ig H chains (Figure 1). Hence, Syk appears to be more critically involved in B cell development, independent of Lyn. However, the argument has not been formally excluded that a complete knockout of Src-family PTKs (for example,

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

561

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

the triple knockout of Lyn, Fyn, and Blk) leads to a phenotype similar to that of the Syk knockout. In addition to the role of Syk in antigen-independent processes, Syk is essential for progression of immature B cells into the recirculating pool of B cells (31) (Figure 1). This is not due to the short life span of immature B cells by loss of Syk. Instead, Tybulewicz et al proposed the hypothesis that Syk might be involved in the expression of the chemokine receptor BLR1, based on the fact that BLR1−/− B cells, like Syk−/− B cells (32), are also unable to enter splenic follicles.

Btk Btk was isolated as the gene responsible for X-linked agammaglobulinemia (XLA) (33, 34). Peripheral B cells from XLA patients are rare and exhibit an immature phenotype, but the number of pre-B cells in the bone marrow is not significantly reduced, suggesting impaired cellular proliferation or increased cell death at the pre-B to B cell transition. In contrast, a point mutation in the Btk pleckstrin homology (PH) domain in mice causes a mild X-linked immunodeficiency (xid) phenotype (35). Peripheral B cells in xid mice are present, albeit somewhat reduced in number, and are skewed toward an immature phenotype, with an over-representation of IgMhighIgDlow cells and a deficiency in the IgMlowIgDhigh class (Figure 1). Strikingly, the peritoneal B-1 B cell population is absent in xid mice. IgMhighIgDlow peripheral xid B cells are refractory to BCRmediated activation in vitro and fail to respond to thymus-independent type II (TI-II) antigens, indicating the involvement of Btk in BCR signaling (36, 37). The difference in phenotype between XLA and xid initially suggested the possibility that the murine xid mutation may not be a complete loss-of-function allele. Analyses of null mutations in the mouse Btk (38, 39), however, proved that lack of Btk function indeed results in the xid phenotype, further confirming a selective disadvantage for Btk-deficient mice B cells during antigen-dependent development, but not during the pro-B to pre-B transition (40).

CD45 Protein tyrosine phosphorylation is controlled by the balance between activity of PTKs and that of protein tyrosine phosphatases. Particularly important in BCR signaling is the transmembrane tyrosine phosphatase CD45 and the SH2containing protein SHP-1. CD45−/− mice generate mature B cells but have perturbed B cell development, exhibiting a marked increase in IgMhighIgDlow (most normal mature B cells are IgMlowIgDhigh) (41, 42). These findings led to the prevailing view that CD45 plays a positive role in B cell development. Supporting this view, the calcium influx from extracellular sources as well as BCR-induced proliferation is decreased in CD45−/− mice (43). Mice carrying

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

562

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

transgenic BCRs (anti-HEL) were bred with CD45−/− mice, and B cell fates in the presence or absence of soluble antigen (HEL) were evaluated. Circulating HEL autoantigen mediated negative selection of mature CD45+/+ HELbinding B cells, whereas, in striking contrast, the autoantigen positively selected CD45−/− HEL-binding B cells, promoting their accumulation as long-lived IgMlowIgDhigh cells (44). The simple explanation for these results is that, even in the absence of CD45, a weak signal is induced by antigen, thereby promoting the maturation and/or survival of B cells. According to this model, mutations or genetic changes that reduce BCR signal strength, such as CD45−/−, result in autoantibody production because they inhibit negative selection. It is also possible that the BCR couples to a qualitatively different signaling pathway in the absence of CD45, leading to increased maturation and survival of B cells.

SHP In contrast to CD45−/− mice, B cell development and function are markedly abnormal in me/me mice (SHP-1 null), resulting in hyper gammaglobulinemia and auto-antibody production. Motheaten viable (mev/mev) mice express a mutant SHP-1 that exhibits markedly decreased PTPase activity. Both strains exhibit the same spectrum of abnomalities, although me/me mice have a more severe phenotype. In me/me mice, B cell progenitors are depleted, and activated B cell and plasma cell populations are increased. In addition, there is a large increase in B1 cell numbers. Since SHP-1 is expressed in almost all hematopoietic cells, the abnormality of B cells in SHP-1−/− mice is presumably due to a mixture of both intrinsic defects within B cells and extrinsic effects of other hematopoietic cells on B cells (45–47). To circumvent these trans-effects on B cells, Cyster and Goodnow reconstituted irradiated mice by using a mixture of bone marrow cells (20% from mev/mev cells and 80% from wild-type cells). This approach reduces trans effects from the overproduced mev/mev mutant myeloid cells in irradiated mice (48). While SHP-1-deficient HEL-specific B cells developed normally in the bone marrow in the absence of HEL, the peripheral mev/mev B cells were unusual, mimicking those of anergic wild-type B cells. However, in contrast to a profound desensitization of BCR signaling in anergic B cells, HEL antigen elicited a larger elevation of cytosolic calcium in SHP-1-deficient B cells, indicating that SHP-1 is a negative regulator of BCR signaling. More efficient elimination of self-reactive B cells (augumentation of negative selection) is due to this exaggerated signaling in SHP-1-deficient B cells. Thus, the production of auto-antibodies in mev/mev mice is somewhat paradoxical. Given the evidence that the predominant source of auto-antibodies in me/me mice is the pool of B1-type B cells (45–47), this can be explained by a favored production of B1 cells through loss of SHP-1. The concept that SHP-1 and CD45 play a negative or positive role, respectively, for BCR signaling has been

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

563

further verified by analysis of CD45−/SHP-1− mice. In these mice, defects of BCR signaling and development due to loss of CD45 were largely, but not completely, restored by the motheaten mutations (49).

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MODES OF PROTEIN-PROTEIN AND PROTEIN-LIPID INTERACTIONS As mentioned above, one outcome of BCR signaling is to alter the phosphorylation state of tyrosine, serine, or threonine residues of target proteins. However, many protein kinases and protein phosphatases have relatively broad substrate specificities. Thus, mechanisms for ensuring the speed and precision of BCR signal transduction must exist to organize the correct repertoires of enzymes into the correct sites of action. This function can be achieved either by recruitment of active signaling molecules into signaling networks or activation of dormant enzymes already positioned close to their substrates (Figure 2). The recent discovery of protein modules that participate in protein-protein interactions has provided critical insights into these aspects of receptor-mediated signal transduction (50). The SH2 domain and PTB domain recognize phosphotyrosine (pTyr) followed by three to five C-terminal residues and pTyr preceded by

Figure 2 Mechanisms for signaling complex formation. (A) Phosphorylation-dependent assembly of protein modules. SH2 domain and PTB domain are recruited to phosphotyrosine (p-Y), whereas PH domain recognizes the specific polyphosphoinositides. (B) Four signaling molecules (for instance enzymes) are co-localized by virtue of a scaffold protein.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

564

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

residues that form a β turn, respectively (Figure 2). The importance of the SH2 domain is typified by the association of autophosphorylated receptor tyrosine kinases with cytoplasmic proteins that contain the SH2 domains. Once the receptor tyrosine kinase is autophosphorylated by ligand binding, proteins containing an SH2 domain are recruited to the phosphorylated receptor, leading to the assembly of complexes of signaling proteins around the activated receptor. Similarly, subcellular organization of serine-threonine kinases and phosphatases occurs through interactions with the targeting subunits or anchoring proteins that localize these enzymes. In addition, proteins such as AKAP79 (51) serve as signaling scaffolds of several kinases and phosphatases (Figure 2). The SH3 domain contributes to protein-protein interactions by binding to proline-rich peptide sequences with the consensus PXXP, whereas the PH (pleckstrin homology) domain is involved in protein-lipid interactions. The PH domains bind the charged headgroups of specific polyphosphoinositides and thereby regulate the subcellular targeting of signaling proteins to specific regions of the plasma membrane (Figure 2). In this way, PH domains can couple the actions of phosphatidyl inositol (PI) kinases, inositol phosphatases, and phospholipases to the regulation of intracellular signaling. Examples of these modular proteins involved in B cell function are depicted in Figure 3. Recent

Figure 3 Signaling molecules found in B cells depicted to highlight their modular structures. Sites of tyrosine phosphorylation are indicated with P and proline-rich sites are indicated Pro. Domains with enzymatic functions are in open rectangles.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

565

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

advances in our understanding of BCR signal transduction (discussed below) suggest how these protein–protein and protein–lipid interactions integrate transmembrane events into B cell biological responses. In addition to these signaling modules, specialized membrane fractions, called the DIG fraction or detergent-resistant membrane rafts, appear to be important for the integrity of signaling molecules around antigen receptors and their subsequent activity. The functional importance of this specialized membrane compartment in TCR signaling has been recently shown by using filipin and nystatin, compounds that disrupt the DIG fraction (52).

ACTIVATION MECHANISMS OF PTKs Kinetic experiments in B cells have shown that the increased activity of the SrcPTKs Blk, Lyn, and Fyn following BCR activation occurs before the increased activity of Btk and Syk (53), suggesting that it is the Src-PTKs that are initially activated through BCR signaling. In vitro binding studies showed that Lyn and Fyn, not Src, interact with the resting BCR (54). This differential association between Lyn/Fyn and Src might reflect binding to a membrane fraction through palmitoylation rather than protein–protein interactions, since Lyn and Fyn, but not Src, are palmitoylated at one or two Cys residues (position 3 or 5 from N-terminal) in addition to being myristoylated at their N termini. Indeed, the functional importance of palmitoylation of Lck has been demonstrated in the context of TCR signaling (55). Targeted gene disruption of Lyn in the chicken B cell line DT40 results in a profound decrease in tyrosine phosphorylation of cellular proteins upon BCR engagement. Since this particular cell line predominantly expresses Lyn among the Src family PTKs, this residual receptor-induced phosphorylation of cellular proteins is presumably attributable to Syk. In fact, the complementary pattern of the receptor-induced phosphorylation observed with the loss of Syk supports this contention (56). Lyn/Syk doubly deficient DT40 cells exhibit no tyrosine phosphorylation induced by BCR cross-linking (57). Hence, two important conclusions can be drawn from these data; Lyn and Syk are the PTKs initially activated by BCR signaling, and these PTKs can be activated independently of each other, at least to some extent. That Syk is activated independently of Lyn is in marked contrast to TCR signaling in which the Syk homologue Zap-70 is activated downstream of Src-PTKs such as Lck, implicating distinct activating modes by Syk and Zap-70 (discussed below). The activities of Src-PTKs are regulated by tyrosine phosphorylation; autophosphorylation of Tyr416 within the Src catalytic domain is stimulatory, and carboxy-terminal phosphorylation of Tyr527 is inhibitory (58). Phosphorylation of Tyr527 mediates an intramolecular association with the kinase’s own SH2 domain, leading to a conformation that represses its kinase activity.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

566

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

The phosphorylation of the carboxy-terminal tyrosine is regulated by Csk and CD45. This concept is verified by the observations that the C-terminal tyrosine of Lyn is hyperphosphorylated in CD45-deficient DT40 cells (59) and CD45− J558 plasmacytoma (60), while this tyrosine is hypophosphorylated in Csk-deficient DT40 cells (61). As the extent of receptor-induced tyrosine phosphorylation of cellular substrates is significantly increased by loss of Csk (61) and decreased by loss of CD45 (59, 62, 63), it appears that the strength of BCR signaling is regulated through the balance between Csk and CD45 activities by determining the phosphorylation state of the carboxy-terminal tyrosine of the Src-PTKs. However, even in the absence of CD45, a weak signal such as a low level of calcium mobilization can be observed (43, 59), suggesting that Syk is still activated despite the downregulation of Src-PTKs. Although many studies indicate that CD45 plays a positive role in BCR signaling, negative effects of CD45 were also observed. CD45-negative variants of WEHI-231 cells show increased tyrosine phosphorylation of cellular proteins at the resting level and an augumented calcium response upon BCR stimulation (64). These discrepancies suggest that CD45 has both positive and negative roles in BCR signaling and that the balance of these opposite effects varies depending upon the distinct developmental stage of B cells and/or distinct CD45 isoforms and/or level of expression of CD45. Crystal structure analyses of Src-family PTKs reveal that the SH3 domain recognizes the linker region between the SH2 domain and the kinase domain (65, 66). Thus, two interactions, between the SH2 domain and phosphorylated Tyr527, and between the SH3 domain and the SH2-kinase linker region, are thought to be involved in keeping Src kinase in the inactive state. Activation of the kinase would presumably lead to the sequential dissociation of the SH2 and SH3 domains from their intramolecular ligands, removing these constraints. Thus, in addition to dephosphorylation of Tyr527, which is likely to be the key regulatory switch, more potent exogenous ligands for the SH2 or SH3 domain may compete for endogenous binding, thereby disrupting the inactive structure. Thus, in the case of BCR signaling, although the initial activation mechanism of Lyn is still unclear, the initially phosphorylated substrate would then bind the SH2 domain of Lyn, which could stimulate enzyme activity, thereby promoting processive phosphorylation. The mechanism of Syk activation has been extensively examined. Weiss and Littman initially proposed a sequential activation model, based on the cooperativity between Lck and Zap-70 in T cells (67). Most of the observations to date in B cells support that this mechanism also operates in B cells. According to this model, BCR engagement leads to phosphorylation of the Igα/Igβ ITAMs by Src-PTKs. Then Syk is recruited to the doubly-phosphorylated ITAM and is activated by Src-PTK phosphorylation (Figure 4). Detailed biochemical (68)

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

567

and genetic studies (69, 70), however, have underscored the existence of an additional activating mode of Syk that is independent of Src-PTKs. What are the underlying mechanisms that explain Src-PTK independent activation of Syk? Since association of Syk with the BCR complex is observed before receptor stimulation (71), receptor aggregation might directly stimulate the activity of the pre-associated Syk. An important mechanism of Syk activation is phosphorylation of the tyrosine located in the kinase activation loop, and the failure of Syk mutant (Tyr519 to Phe) to activate BCR signaling (72). This phosphorylation is probably mediated by autophosphorylation as well as Src-PTKs. Consistent with this notion, the kinase activity of Syk, but not Zap-70, is reported to be activated when bound to doubly-phosphorylated (dp)-ITAMs of Igα-Igβ (73). Thus, binding of Syk to the dp-ITAM may bring Syk into a state that is more susceptible to autophosphorylation, resulting in an increase in Syk specific activity. The existence of the coiled-coiled loop between the two SH2 domains, revealed by crystallographic analysis (74), might provide the structural basis for this concept. Syk binding to dp-ITAM may induce a conformational change of this coiled-coiled region, allowing its interaction with the kinase domain to alter enzymatic activity. The binding affinity of two SH2 domains to dp-ITAMs of Igα/Igβ, partially mediated by the region between the C-terminal SH2 domain and the kinase domain (75), also appears to be involved in the enzymatic activation of Syk. Although Syk is activated in the absence of Src-PTKs, the positive effects of Src-PTKs on Syk are clear. In fact, BCR-induced activation of Syk is dramatically inhibited by loss of Lyn in DT40 B cells (76). Hence, under physiological conditions, BCR ligation activates Src-PTKs as well as a pre-associated Syk, leading to phosphorylation of ITAMs in Igα-Igβ, and the subsequent recruitment of Syk. Then recruited Syk is activated by Src-PTK-dependent transphosphorylation and by autophosphorylation. The Tyr130 residue of Syk is autophosphorylated by activated Syk, and this phosphorylation may, in turn, lead to termination of Syk activation in BCR signaling (77). Tyr130 is located between the N-terminal and C-terminal Syk SH2 domains, the phosphorylation of which is important in mediating release of Syk from the BCR complex. Phosphorylation of Tyr130 may induce conformational changes of the Syk SH2 domains, leading to a low-affinity state of Syk with (dp)-ITAMs of Igα-Igβ. Another possibility is that a protein containing SH2 domain binds to the phosphorylated tyrosine, sequestering Syk from the receptor complex. In the mast cell system, overexpression of Cbl blocks Syk assembly from the activated FcεRI and blocks the enzymatic activity of Syk (78). Recent analysis of Cbl−/− mice also supports the negative role of Cbl in Syk/Zap-70 function (79). There seem to exist two additional mechanisms that allow Syk to be turned off. One potential mechanism is that tyrosine

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SAT/spd P2: KKK/plb

January 21, 1999

568

15:34

QC: KKK/mbg/plb

KUROSAKI

T1: KKK

Annual Reviews AR078-18

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

569

phosphatases such as SHP-1 mediate dephosphorylation of the Tyr519 residue. Another mechanism may be analogous to the JAK kinase inhibitor CIS (80– 82). SH2 containing proteins like CIS may bind to phosphorylated Syk after BCR stimulation, leading to inhibition of Syk enzymatic activity by inducing conformational changes of Syk. As increased activity of Btk follows the increased activity of Src-PTKs upon receptor cross-linking, it has been proposed that Btk is activated by Src-PTKmediated phosphorylation (53). In COS cells and fibroblasts, Btk is phosphorylated and activated by cotransfection with Src-PTKs (83, 84). The requirement for Src-PTKs in BTK activation was reexamined in the context of BCR signaling. In Lyn-deficient DT40 B cells, the BCR-induced tyrosine phosphorylation of Btk still occurred despite delayed time kinetics. In contrast, sustained Btk phosphorylation was clearly inhibited in Syk-deficient DT40 cells (85). A simple explanation of these results is the existence of two phases of tyrosine phosphorylation of Btk in BCR signaling: the initial phase and the sustained phase, mediated by Lyn and Syk, respectively. Supporting this conclusion, BCR-mediated phosphorylation of Btk in Blk/Lyn double-deficient splenic B cells was comparable to that in wild-type cells (86). In fibroblasts, Btk is activated through Src-PTK-dependent transphosphorylation of Tyr551, which is in the activation loop of the catalytic domain of Btk (83, 84). This results in a five- to tenfold increase in Btk enzymatic activity, subsequently leading to autophosphorylation at Tyr223 in the SH3 domain (87) (Figure 4). In B cells, Tyr551 and Tyr223 are also phosphorylated in response to BCR stimulation. Although it still remains unresolved which PTK(s) phosphorylates Tyr551 in B cells, transphosphorylation of Tyr551 is essential for participation of Btk in BCR signal transduction (85). Recent crystallographic analysis of Itk (a Btk/Tec family PTK expressed in T cells) shows the existence of an intramolecular interaction between the proline-rich region and the SH3 domain of Itk (88). Based on this structure, Schreiber et al proposed the existence of an equilibrium between an intramolecular, “closed” ←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 4 Model for PLC-γ 2 activation upon BCR engagement. (A) BCR ligation activates Lyn, leading to tyrosine phosphorylation of ITAMs within the Igα-Igβ subunits of the BCR, Then, Syk is recruited to phosphorylated ITAMs. PI-3,4,5-P3, a product of activated PI3-K, recruits Btk to membrane fraction by its binding to Btk PH domain. (B) Activated Lyn phosphorylates the tyrosine in the activation loop of Syk and Btk (Tyr519 of Syk and Tyr551 of Btk), so leading to their activation. (C ) Activated Syk phosphorylates BLNK and thereby brings BLNK to the membrane fraction, eventually leading to co-localization of Syk, BLNK, and Btk. The membrane-associated Btk might be further recruited to phosphorylated BLNK by using the Btk SH2 domain. (D) Then phosphorylated BLNK brings PLC-γ 2 into close proximity of a Syk-BLNK-Btk complex and thereby facilitates the tyrosine phosphorylation and subsequent activation of PLC-γ 2.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

570

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

complex and intermolecular, “open” complex in which Itk SH3 and SH2 domains are occupied by bidentate cellular ligands. Indeed, a novel protein, Sab, which preferentially associates with the Btk SH3 domain, has been recently cloned (89), and this molecule is able to inhibit Btk enzymatic activity upon its binding to Btk. This suggests that Sab functions as a trans-inhibitor for Btk (T Yamadori, Y Baba, T Kurosaki, M Matsushita, S Hashimoto, T Kishimoto, S Tsukada, submitted). Interestingly, Tyr223 of Btk is located within the interface of the interaction between the SH3 domain and the proline-rich region. Thus, an intriguing possibility is that phosphorylation of Tyr223 may disrupt this intramolecular interaction, thereby exposing binding sites for association with other signaling molecules. In addition to regulation by phosphorylation, other mechanisms of Btk regulation have been identified. Btk interacts with phosphatidylinositol-3,4,5trisphosphate through its PH domain, an interaction that is required for recruitment of Btk to membrane fractions (90–92). Since phosphatidylinositol3,4,5-trisphosphate (PI-3,4,5-P3) is generated by PI3-K, Btk is targeted to the plasma membrane after PI3-K activation, thereby rendering Btk susceptible to transphosphorylation at Tyr551 by Lyn and Syk (Figure 4).

EFFECTOR SYSTEMS Lipid Metabolizing Enzymes One of the critical downstream events following PTK activation is activation of lipid-metabolizing enzymes such as PLC-γ and PI3-K. In the case of the receptor PTKs, such as the PDGF receptor, receptor autophosphorylation sites direct the binding of PLC-γ 1 to the activated receptor, whereupon PLC-γ 1 becomes tyrosine phosphorylated and activated. Similar to this mechanism, PLC-γ 1 is tyrosine phosphorylated after recruitment through its SH2 domains to phosphorylated Tyr341 and Tyr345 of Syk in COS cells (93). However, the Syk mutant (Tyr341/345 to Phe) was still able to activate PLC-γ 2 in Sykdeficient DT40 cells (H Adachi, M Kurosaki, T Kurosaki, unpublished data). Since the SH2 domains of PLC-γ 2 are essential for its activation (94), these data raise a question about the phosphorylated target molecule of the PLC-γ 2 SH2 domains in B cells. Candidate molecules include BLNK (alternatively named SLP-65) (95, 96) and/or a B cell homologue of LAT (97, 98), as PLC-γ 2 and PLC-γ 1 are recruited to phosphorylated BLNK in B cells and to phosphorylated LAT in T cells, respectively. BCR-induced BLNK phosphorylation is apparently mediated by Syk, since BLNK phosphorylation is lost in Syk-deficient DT40 cells (95). BLNK has a structural domain similar to SLP-76 (99), in that it contains an N-terminal basic region with several tyrosine phosphorylation sites and a C-terminal SH2

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

571

domain (Figure 3). SLP-76 is expressed in T cells, NK cells, and mast cells, whereas BLNK is expressed only in B cells. BLNK associates with PLC-γ 1/2, Grb2, Vav, and Nck by SH2-phophotyrosine interactions, suggesting the possibility that BLNK is a central linker molecule in bridging Syk with downstream effector functions including PLC-γ 2 activation (95). In fact, BCR-induced PLC-γ 2 activation was abrogated in BLNK-deficient DT40 cells, and this defect was bypassed by expression of PLC-γ 2 as a membrane-associated form (M Ishiai, M Kurosaki, R Pappu, K Ohkawa, I Ronko, C Fu, A Iwamatsu, AC Chan, T Kurosaki, Immunity, in press) (Figure 4). LAT, a recently cloned adaptor molecule in T cells, consists of 223 amino acids with a putative transmembrane domain at the N-terminus. Although LAT contains no regions homologous to SH2, SH3, and PH domains, it contains several tyrosines predicted to mediate interactions with the SH2 domains of Grb2 and PLC-γ 1 (97, 98). Northern analysis demonstrates that LAT is expressed predominantly in T cells, NK cells, and mast cells but not in B cells, which correlates well with the expression pattern of SLP-76. It is possible that a homologue of LAT may exist in B cells. Thus, a complex of this LAT homologue and BLNK may participate in targeting PLC-γ 2 to the plasma membrane. Genetic analysis reveals a more complicated story for PLC-γ 2 activation upon BCR stimulation. Deletion of Btk in DT40 cells eliminates PLC-γ 2 activation through inhibition of phosphorylation of PLC-γ 2 (57), indicating that Btk, as well as Syk, is required for tyrosine phosphorylation and activation of PLC-γ 2. Supporting the importance of the Btk/Tec family in PLC-γ activation, B cells from XLA patients (100) and Itk−/− T cells (101) showed a profound reduction in IP3 production upon BCR and TCR engagement, respectively. Although the precise mechanism for the requirement for Syk and Btk is still unclear, these two PTKs could be required to phosphorylate PLC-γ 2 on distinct sites that are collectively required for activation. Alternatively, one kinase may be responsible for the phosphorylation that activates PLC-γ 2 and the other may be responsible for directing PLC-γ 2 to the plasma membrane. The former possibility appears more likely because both Syk- and Btk-deficient DT40 cells show inhibition of PLC-γ 2 tyrosine phosphorylation upon BCR engagement (Figure 4). An effect of the PI3-K pathway on PLC-γ 2 activation via two distinct mechanisms has been envisaged by recent reports. One mechanism is that PI-3,4,5-P3 is indirectly involved in PLC-γ 2 activation through its binding to the Btk PH domain and subsequent Btk activation (Figure 4). Indeed, cotransfection of the PI3-K p110 subunit and Btk increased PLC-γ 2 tyrosine phosphorylation and its activation (100). The second possibility is that the PLC-γ 2 PH domain binds to PI-3,4,5-P3 thereby being recruited to the plasma membrane, since the PLC-γ 1 PH domain binds preferentially to PI-3,4,5-P3 (102).

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

572

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

Not only PI3-K, but also phosphatidyliositol 4-phosphate 5-kinase (PIP5K) has been proposed to contribute to PLC-γ 2 activation because PIP-5K synthesizes the available substrate PI-4,5-P2 for PLC-γ 2. Since Vav1-activated Rac has been shown to interact with PIP-5K in other cell systems, Vav1 may participate in BCR signaling by generating PI-4,5-P2 through PIP-5K activation (103). PLC-γ 2 activation leads to hydrolysis of phospholipid, yielding IP3 and DAG. IP3 binds IP3 receptors (IP3Rs) located in the endoplasmic reticulum (ER), leading to calcium release from internal stores (94). Triple knockout of three IP3R isoforms abolishes BCR-induced calcium mobilization both from internal stores and from extracellular sources, whereas this overall calcium mobilization is still observed by single knockouts of these receptors (104). Thus, three IP3R isoforms are essential and functionally redundant mediators for overall BCR-induced calcium mobilization. These data support the capacitative model for calcium entry in which calcium influx across the plasma membrane is coupled to depletion of intracellular calcium stores (Figure 5). Although expression of only one isoform does not abolish calcium mobilization upon BCR ligation, detailed calcium signaling patterns differ significantly among these three IP3R isoforms. For instance, DT40 B cells expressing only type 2 IP3R showed regular and robust Ca2+ oscillations upon receptor ligation, whereas monophasic Ca2+ transient or rapidly damped Ca2+ oscillations were observed in mutant cells expressing either type 3 or type 1 alone, respectively (T Miyakawa, A Maeda, T Yamazawa, K Hirose, T Kurosaki, M Iino, submitted). Thus, differential and combinatorial expression of three IP3R isoforms appears to be one of the critical determinants for spatiotemporal Ca2+ signals, which in turn regulate the selectivity of transcription factors in B cells (105–107). IP3R channel activity may be modulated by phosphorylation. In fact, IP3R has been suggested to be a substrate of Src-PTKs in antigen receptor signaling, since TCR-induced tyrosine phosphorylation of the IP3R is reduced in Fyn−/− mice. Moreover, IP3R channel activity is increased by addition of Src-PTKs in vitro (108). BCR engagement generates the PI3,4,5-P3 product (109), indicating that the activation of PI3-K occurs in BCR signaling. The PI-3K inhibitor wortmannin blocks anti-Ig-induced growth inhibition of a human B cell line (110). As this growth arrest and/or BCR-induced apoptosis are thought to mimic one mechanism of BCR-induced tolerance of B cells, these data suggest that the PI-3K pathway modulates B cell tolerance. BCR stimulation induces tyrosine phosphorylation of the BCR-associated CD19 molecule. This phosphorylation occurs on sites that interact with the p85 regulatory subunit of PI-3K via its SH2 domains (111). Phosphorylation

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

573

of CD19 is greatly enhanced when the antigen is decorated with C3d fragments resulting from activation of the complement cascade. The resulting coligation of the BCR with CD19 promotes the ability of CD19 to recruit PI3-K, thereby leading to activation of PI3-K kinase activity. In addition to this mechanism, a proline-rich region of the p85 subunit binds to the SH3 domains of Lyn and Fyn, leading to an increase in PI3-K activity. Indeed, the BCR-induced activation of PI3-K is blocked by a peptide containing this proline-rich region (112). Thus, one hypothesis is that multimeric protein complexes are formed upon BCR ligation in which PI3-K is simultaneously bound to CD19 and to Fyn or Lyn, resulting in maximal PI3-K activation. Various PH domain-containing molecules including Vav, Btk, and PLC-γ 1 selectively bind PI3,4,5-P3, a product of PI3-K activity. This binding appears to contribute to conformational modification of these signaling molecules as well as their recruitment to the plasma membrane. After PI3,4,5-P3 generation by activated PI3-K, a rapid and prolonged elevation of PI3,4-P2 occurs following BCR cross-linking (109). This PI3,4-P2 could be the result of phosphorylation of PI4-P by PI3-K, or alternatively it could result from the removal of the 50 phosphate of PI3,4,5-P3. The importance of the latter mechanism is suggested by the fact that SHIP, an enzyme capable of catalyzing this reaction, is a prominent substrate of BCR-induced tyrosine phosphorylation. This SH2 domain-containing inositol polyphosphate 5-phosphatase (SHIP) was initially identified by virtue of its association with Shc upon stimulation of antigen receptors and cytokine receptors (113). SHIP is thought to exert two effects on the PI-3K pathway; it acts to decrease the amount of PI3,4,5-P3 and to increase its conversion to PI3,4-P2. This prediction was directly examined by studying SHIP-deficient DT40 B cells. These cells demonstrated enhanced PI3,4,5-P3 production upon BCR ligation. Furthermore, SHIP-deficient B cells exhibited sustained Ca2+ increase or long lasting Ca2+ oscillations upon receptor stimulation (181). Hyperactivation of the ITAM-bearing receptor FcεRI was also observed in SHIP−/− mice (114). These mice exhibit increased numbers of granulocyte-macrophage progenitors, due to hyper-responsiveness to stimulation via cytokine receptors such as macrophage-colony stimulating factor and IL3 receptors. In contrast, the number of B220 positive B cells in the bone marrow was significantly decreased, presumably due to trans-effects from overproduced SHIP−/− myeloid cells. This phenotype is similar to that of mev/mev mice, raising the intriguing question of whether SHP-1 and SHIP function independently within the same pathway or in an overlapping manner. In contrast to hyper-calcium responses in SHIP-deficient B cells, wortmannin treatment blocks BCR-induced calcium influx (115). Thus, following calcium mobilization from the ER by PLC-γ 2-evoked IP3, a capacitative process is regulated by the PI3-K pathway, leading to calcium influx from extracellular

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

574

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

Figure 5 Two models for BCR-induced calcium influx. (A) Level of IP3 generation, regulated by Btk and Syk, influences the state of emptiness of intracellular calcium stores such as endoplasmic reticulum (ER) stores. Sustained elevation of IP3 results in progressive depletion of ER stores, efficiently coupling ER store to opening of calcium channels on the plasma membrane (CRAC channels). This model was proposed by Scharenberg and Kinet (180), based on the observations that Btk-deficient B cells generate less IP3 than wild-type cells and thereby not allowing ER stores to be fully depleted, leading to inefficient coupling to CRAC channels. (B) In addition to the effect of Btk on sustained IP3 generation, Btk participates in the process to opening of CRAC channels.

sources. Moreover, based on the evidence that wortmannin-mediated inhibition of calcium influx can be restored by active Btk (116), it has been proposed that PI-3,4,5-P3-activated Btk participates in calcium influx by regulating depletion of intracellular calcium stores and/or calcium influx channels (Figure 5B). Uncovering the molecular mechanisms and identification of calcium channels in the plasma membrane will be important, since the amplitude and duration of BCR-induced Ca2+ control the differential activation of transcription factors (105).

GTP-Binding Proteins Ras is maximally stimulated within one to two minutes after receptor ligation, as indicated by its transition from the RasGDP to the RasGTP state. A dominant-negative transgene (Ras N17) expressed in B cells under the control of the Eµ enhancer arrests B cell development before the pre-BCR dependent stage (117). This phenotype is similar to that of mice lacking the high affinity IL-7 receptor α chain, suggesting that Ras is required for IL-7 to drive B cell development. Nevertheless, some leaky mature B cells emerge in transgenic

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

575

mice expressing Ras N17, and these cells exhibit inhibition of BCR-induced proliferation. Thus, one explanation is that a Ras-dependent signal is required at various B cell developmental stages, presumably reflecting its requirement in various receptor systems including the IL-7 receptor and the BCR. Sos, a nucleotide exchange factor for Ras, is present in the cytosol in the resting state, where it does not efficiently activate the membrane-bound Ras molecule. Since forced localization of Sos to the plasma membrane is sufficient to activate Ras (118), it is thought that Ras is activated by inducing the translocation of Sos to the membrane fraction. Several coupling molecules, including Shc, Grb2, and Cbl, have been implicated in Sos translocation. Shcdeficient DT40 B cells exhibited normal ERK activation, whereas this ERK activation was inhibited by loss of Grb2 or expression of dominant-negative Ras, suggesting the dispensability of Shc for Ras activation (119). Grb2 consists of one SH2 domain and two SH3 domains (Figure 3). The former domain can bind to tyrosine phosphorylated molecules such as the B cell LAT homologue, leading to Grb2 recruitment to the membrane fraction, while the SH3 domains can interact with the proline-rich regions in Sos. Since ERK activation was completely abrogated in Syk-deficient B cells (120), phosphorylation of the Grb2-associated molecule is presumably mediated by Syk. In contrast to the positive role of Grb2 in the Ras pathway, a negative role of Cbl through two distinct mechanisms has been proposed. Cbl is a 120-kDa protein with a variety of protein motifs, including multiple prolinerich SH3-binding sites and multiple tyrosines (Figure 3), the phosphorylation of which could promote binding to several SH2-containing proteins including CrkL. The first mechanism for involvement of Cbl in the Ras pathway is via Rap1. CrkL constitutively associates with the Rap1-specific exchange factor C3G by an interaction mediated by the SH3 domain of C3G. This CrkL-C3G complex is linked to BCR signaling by binding of CrkL via an SH2 domain to tyrosine phosphorylated Cbl (121, 122). Thus, BCR stimulation could promote formation of the Cbl-CrkL-C3G complex, thereby allowing Rap1 activation. Rap1 appears to negatively regulate Ras signaling pathways by competing for binding to Ras effectors such as Raf, thereby preventing their activation by GTP-bound Ras (123). The second mechanism is via competition for binding to Sos. The binding of Sos and Cbl to Grb2 is mutually exclusive, as the proline-rich regions of these proteins compete for the same binding site in the Grb2 SH3 domain (124). Thus, Cbl may negatively regulate the Ras pathway by inhibition of Sos recruitment to the membrane fraction. BCRinduced tyrosine phosphorylation of Cbl is abrogated in Lyn-deficient DT40 cells (125), suggesting that Cbl is a substrate of Lyn. In contrast to this report, Syk was shown to mediate tyrosine phosphorylation of Cbl upon FcεRI crosslinking (126), indicating that Cbl may be a substrate for both Src-PTK and

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

576

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

Syk with the relative contribution of each kinase varying in different cellular systems. Ras is downregulated following interaction with the Ras-GTPase activating protein, RasGAP. BCR stimulation results in tyrosine phosphorylation of rasGAP, and the GTPase activating activity of RasGAP appears to be decreased by tyrosine phosphorylation (127, 128). In addition, BCR stimulation leads to tyrosine phosphorylation of two proteins, of 190 kDa and 62 kDa, which then associate with RasGAP. The p190 protein turns out to be a GAP for the Rho family of small GTP-binding proteins, and thus its association with RasGAP may reflect a coordinated regulation of Ras and Rho GTP-binding families (129). The p62dok protein was recently cloned, and this protein contains a PH domain (130, 131). Current models for the involvement of p62dok in the regulation of Rho family GTPases are based on the assumption that tyrosine phosphorylation of p62dok will modulate the cellular distribution of associated RasGAP and RhoGAP and sequester these negative regulators away from active Ras and Rac, thereby prolonging the activation of Rac and Rho effector pathways. Vav1 is a 95-kDa guanine nucleotide exchange protein with selectivity for Rac. Vav1−/− mice display normal numbers and differentiation of B cells in the bone marrow and normal B cell populations in peripheral lymph nodes, but reduced number of B1 B cells in the peritoneal cavity. BCR-induced proliferation and calcium mobilization of peripheral Vav1-deficient B cells are compromised (132–136). Thus, even in the absence of Vav1, a sufficient signal for conventional (B2) B cell growth is provided, but the signal strength is below the threshold necessary for proliferation of B1 B cells. Since several reports have shown that Rac regulates JNK activation (137), the apparently normal JNK activation upon BCR engagement in Vav1−/− mice was somewhat surprising. The simple explanation is that the physiological function of Vav1 in lymphocytes is related not to JNK activation but rather to cytoskeletal regulation and calcium mobilization. Supporting this view, treatment of T cells with cytochalasin D, which blocks actin polymerization, produced defects in TCR signaling nearly identical to those observed in Vav1−/− mice (136). Another possibility is that the Vav1−/− phenotype could be accounted for by a functional redundancy of Vav1 and Vav2. Similar to recruitment of Vav1 to Zap-70 in T cells (138), Vav1 is recruited to the phosphorylated Tyr341 and Tyr345 of Syk through its SH2 domain and is subsequently phosphorylated in B cells (139). Bustelo and co-workers have presented evidence that tyrosine phosphorylated Vav1, not the nonphosphorylated form, catalyzes GDP/GTP exchange on Rac-1 (137). In addition to regulation of Vav1 by tyrosine phosphorylation, PI-3,4,5,-P3 a product of PI3-K activity, might be involved in Vav1 recruitment to the membrane fraction, since the PH domain of Vav1 selectively binds PI-3,4,5-P3. Interestingly, this lipid product

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

577

contributes to allosteric modification of GDP/GTP exchange activity of Vav1, thereby potentiating tyrosine-dependent enhancement of GDP/GTP exchange activity (140). Apart from these mechanisms, coligation of BCR and CD19 recruits Vav1 to the phosphorylated Tyr391 of CD19, leading eventually to membrane localization of Vav1. In fibroblasts, Rho GTPases including Rho, Rac, and Cdc42 coordinate the dynamic organization of the actin cytoskelton and the assembly of focal adhesions. Cdc42 can regulate T cell polarization toward APCs, a process critical for efficient APC-T cell contact and subsequent cytokine release. The importance of Cdc42 has been highlighted by the observation that the genetic defect in patients with Wiskott-Aldrich syndrome (WAS) maps to an effector protein for Cdc42, termed WASP. WAS is an X-linked hereditary immunodeficiency, characterized by thrombocytopenia and abnormal humoral- and cell-mediated immunity; defects include cytoskeletal and cell-activation abnormalities in lymphocytes (141). WASP can interact not only with Cdc42 (142), but also with PI-4,5-P2 and several SH3-containing proteins through its N-terminal PH domain and central proline-rich domain, respectively (141). One of these SH3-containing proteins is Btk, which is able to phosphorylate WASP in vitro, suggesting that WASP may be a potential substrate of Btk in B cells (Y Baba, S Nonoyama, M Matsushita, T Yamadori, S Hashimoto, K Imai, S Arai, T Kunikata, M Kurimoto, T Kurosaki, H Ochs, J Yata, T Kishimoto, S Tsukada, submitted).

Serine-Threonine Kinases and Phosphatases One event downstream of the activation of lipid metabolizing enzymes and GTP binding proteins is activation of serine-threonine kinases and phosphatases. IP3 and DAG, generated by PLC-γ 2 activation, evoke calcium mobilization and PKC activation, respectively. Calcium elevation leads to the activation of both calmodulin-dependent protein kinase II and the calmodulin-activated serine/threonine phosphatase calcineurin, the latter of which is the target of cyclosporin A and FK506 (143). One of the targets regulated by calcium elevation is the transcription factor NF-AT whose nuclear translocation is facilitated through its dephosphorylation by calcineurin (144, 145). The BCR-induced apoptosis of WEHI-231 cells and proliferation of splenic B cells are inhibited by cyclosporin A, indicating the importance of calcineurin for these responses (146, 147). Both phosphatases, and kinases that regulate cytoplasmic-nuclear localization of NF-AT have been identified. Cytoplasmic NF-AT is phosphorylated on the serine proline (SP) repeats and serine-rich region (SRR) by the coordinated activity of casein kinase Iα and MEKK1-mediated kinase (148), thereby masking the activity of nuclear localization sequences (NLSs). Dephosphorylation of NF-AT by calcineurin leads to an alteration

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

578

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

in an intramolecular interaction, exposing NLSs to the nuclear import machinery (149). Moreover, glycogen synthase-3 (GSK-3) has been shown to phosphorylate serine residues necessary for nuclear export, promoting nuclear exit of NF-AT (150). Tyrosine phosphorylation as well as serine phosphorylation may affect cytoplasmic-nuclear localization of transcription factors. HS1 possesses features that are characteristic of a transcription factor, namely, a putative nuclear localization signal and a motif composed of basic amino acids that resembles a DNA-binding protein. A variant WEHI B lymphoma line that is resistant to BCR-induced apoptosis exhibits a low level of HS1 expression, and transfection of HS1 into this variant B cell line restores apoptosis (151). BCR stimulation evokes HS1 phosphorylation on several Tyr residues, among which Tyr378 and Tyr397 are critical for inducing apoptosis as well as for HS1 translocation to the nucleus (152). Thus, it is speculated that phosphorylation at these tyrosines allows HS1 to be transported to the nucleus, where HS1 participates in the transcriptional regulation of target genes involved in apoptosis. Protein kinase C (PKC) isoforms expressed in B cells include α, β, δ, ζ , η, and θ, of which α, β, δ, ε and θ are regulated by DAG, a product of PLC-γ 2 activity. Similar to xid mice, PKCβI/II-deficient mice have a reduced number of mature B cells in the spleen and lymph nodes, barely detectable B-1 cells, and they are unable to mount a humoral immune response to thymus-independent type II (TI-II) antigens. Furthermore, BCR-mediated proliferation of PKCβ −/− B cell is significantly reduced, again similar to xid mice, placing PKCβ and Btk along the same signaling pathway (153). No significant change in BCRinduced Btk activation between PKCβ −/− and wild-type mice was noted. This finding, together with the evidence that PLC-γ 2 activation is inhibited by loss of Btk (57), suggests that Btk is upstream of PKCβ in BCR signaling. Another PKC isoform, PKCµ, is reported to associate with the BCR. Furthermore, BCR stimulation leads to markedly increased PKCµ activity, which is a downstream event of PLC-γ 2 activation (154). Phosphorylation of Syk by PKCµ decreases the ability of Syk to phosphorylate PLC-γ 1 in vitro, implying that PKCµ may play a negative feedback role in the PLC-γ pathway. A link between the PI3-K pathway and Akt activation is emerging. PI3,4-P2 activates the serine/threonine protein kinase Akt, through binding to the PH domain of Akt (PKB) (155, 156), so SHIP may stimulate Akt activation by generating this lipid product. However, recent identification of the serine/ threonine kinase PDK1 that acts upstream of Akt has made this scenario more complicated. PDK1 is stimulated by PI3,4,5-P3, not by PI3,4,-P2, and this activated PDK1 phosphorylates Akt on Thr308, leading to Akt activation (157). Thus, it appears that both binding to PI3,4-P2 and phosphorylation by PI3,4,5-P3-triggered PDK1 are required for Akt activation. One of the targets

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

579

of Akt is glycogen synthase kinase 3 (GSK3); activated Akt phosphorylates GSK3, leading to inactivation of GSK3 kinase activity (158). NF-AT nuclear translocation is regulated by inhibiting its export through inactivation of GSK3. Thus, Akt activation could eventually lead to enhancing NF-AT nuclear localization by inactivation of GSK3 activity (150). In more recent studies, Bad has been identified as a potential target of Akt, linking the PI3K pathway directly to the apoptotic machinery. The Bcl-2 family of related proteins contains protein–protein interaction domains that facilitate homo- and heterodimerization. Overexpression of some members (Bcl-2, Bcl-XL) in some systems promotes cell survival, whereas others (Bax and Bad) promote cell death. Bad dimerizes with Bcl-XL and Bcl-2 through interaction of the BH3 domain of Bad. Phosphorylation of Bad, mediated by Akt, promotes association with the 14-3-3 family proteins, which may promote survival by allowing heterodimerization of Bcl-XL with Bax, thereby preventing the pro-apoptotic function of Bax (159, 160). According to this model, PI-3K pathway transmits the anti-apoptotic signal through phosphorylation of Bad, and this signaling model could account for the fact that wortmannin blocks anti-Ig-induced growth inhibition, as discussed above. The MAPK cascades constitute a group of signal transduction pathways characterized by successive phosphorylation of coupled serine/threonine or dual specificity kinases. The conserved signaling modules consist of a mitogenactivated protein kinase (MAPK), a MAPK kinase (MAPKK), normally a dual-specificity kinase that phosphorylates MAPK on threonine and tyrosine residues, and a MAPKK kinase (MAPKKK), which phosphorylates MAPKK on serine and activates it. The best characterized MAPK pathway is the ERK pathway. This pathway receives a primary signal from Ras-GTP, which binds directly to Raf-1, the MAPKKK in this cascade. Activated Raf phosphorylates and activates MEK-1 and MEK-2 (the MAPKK), which both in turn phosphorylate ERK-1 and ERK-2. Phosphorylated ERKs form dimers, a step required for nuclear translocation and the subsequent phosphorylation of transcriptional regulatory proteins, including Fos, Jun, and members of the Ets family (161). Supporting the notion that Raf acts downstream of Ras, transgenic mice harboring a B cell–specific dominant-negative Ras were rescued by introducing constitutively active Raf. Although Ras appears to be critical for ERK activation in BCR signaling, stimulation of PKC with phorbor esters can also promote ERK activation. This pharmacological evidence was further strengthened by the genetic data that BCR-induced ERK activation was significantly inhibited in PLC-γ 2-deficient DT40 B cells (119) (Figure 6). Unlike the ERK pathway, the JNK (SAPK) pathway is regulated by the small GTP-binding proteins Rac1 and Cdc42. The JNK cascade elements

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

580

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

Figure 6 Model of BCR-evoked activation of MAP kinase family members.

are positioned in a conventional signaling cascade involving MKK1 (the MAPKKK), SEK1 (the MAPKK) and JNK. Among the targets of this cascade are the transcriptional factors c-Jun and ATF-2. JNK activation is inhibited by cyclosporine and loss of IP3 receptors, suggesting that this cascade requires calcium-sensitive calcineurin activity (119, 162). Furthermore, in addition to calcium, a particular PKC isoform, PKCθ (163), has been recently shown to participate in JNK activation (Figure 6). Regulation of the p38 kinase is also achieved through a serine kinase cascade and the small GTP-binding proteins Rac1 and Cdc42. Unlike the requirement for both PKC and calcium in JNK activation, maximum p38 activation appears to require only PKC, not calcium (119) (Figure 6). Although the precise physiological substrates of this cascade are unknown, p38 can phosphorylate ATF-2 and another kinase, MAPK-activated protein kinase 2, and its activation has been correlated with BCR-induced apoptosis.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

581

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MODULATION OF BCR SIGNALING BY CELL-SURFACE CO-RECEPTORS An area of intense investigation in the last three years has been the involvement of multiple transmembrane proteins in regulating the strength and quality of the BCR signal. B cell activation by the BCR is modified by co-receptors such as CD19, CD22, PIR-B, and Fcγ RII. Since phosphorylation of these coreceptors by ITAM-associated PTKs, such as Lyn, is critical to modulation of BCR signaling pathways, these co-receptors can be considered to act like adaptor proteins. Coligation of co-receptors and BCR brings Lyn into the close proximity with the cytoplasmic domains of co-receptors, thereby leading to their phosphorylation. Thus, information about the nature of each antigen as well as the microenvironment is recognized by co-receptors, is translated into the strength of phosphorylation and phosphorylation site(s) in the cytoplasmic domains of these co-receptors. For instance, tyrosine residues within the ITIM of the cytoplasmic domain of Fcγ RII is phosphorylated by coligation of Fcγ RII and BCR through antigen-antibody complexes, leading to recruitment of SHIP by phosphotyrosine-SH2 interaction (164). Recruitment of SHIP to the phosphorylated ITIM in Fcγ RII brings this enzyme into close proximity to its substrate, PI3,4,5-P3, inducing its breakdown (165, 166). As discussed above, since this lipid product, generated by PI-3K activation, is thought to increase calcium influx, the breakdown of PI3,4,5-P3 by SHIP results in a decrease in calcium influx. Because SHIP inhibits the BCR signal alone (181), SHIP conveys signals from the BCR as well as from Fcγ RII. Thus, SHIP is required for inhibiting calcium mobilization, and the Fcγ RII cytoplasmic domain acts to amplify this inhibition. In contrast to a requirement for SHIP in Fcγ RII-mediated inhibitory signal, protein tyrosine phosphatases SHP-1 and/or SHP-2 participate in CD22- (167) and PIR-B-mediated inhibition (168, 169). SHP-1 and/or SHP-2 are recruited to the phosphorylated ITIMs in CD22 and PIR-B, whereupon they become activated, resulting in the inhibition of tyrosine phosphorylation of PLC-γ 2 by dephosphorylating Syk and Btk (182).

DIFFERENTIAL SIGNALING BY THE BCR Signaling from the BCR provokes several different cellular responses, depending on the stage of differentiation of the B cell and on extracellular inputs. The best examples that opposite responses to antigen occur at different B cell stages are immature versus mature B cells. Cross-linking BCR on immature B cells aborts B cell development in vivo and transmits a signal that preferentially inactivates immature B cells in vitro, whereas stimulation of mature B cells

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

582

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

with anti-IgM antibodies triggers a mitogenic response. What is the underlying mechanism leading to distinct biological responses? Although BCR-induced calcium mobilization was comparable in mature versus immature B cells, induction of the Ras/ERK responsive egr-1 and c-fos genes was only detectable in mature B cells (170). Failure to activate egr-1 in immature B cells may reflect developmentally regulated methylation of this gene (171). Thus, one possibility is that developmental changes in transcription factor expression, chromatin structure, or covalent modification of genes could allow the same BCR-induced cytosolic signals to regulate different sets of genes in immature versus mature B cells. Arrested development of immature B cells followed by deletion or receptor editing (secondary light chain gene rearrangements) is triggered by a variety of high avidity self-antigens such as membrane bound H-2Kb (172), membrane bound HEL (173), erythrocyte surface antigens (174), and double-stranded DNA (175). It had been previously thought that a single cell type (IgM+IgD−) is capable of either antigen-induced receptor editing or apoptosis depending on the nature of the antigen. However, Nemazee and co-workers proposed an alternative model: that antigen-induced apoptosis arises relatively late in the immature B cell stage (IgMhighIgD− corresponding to transitional stage) and is preceded by a functionally distinct developmental stage capable of receptor editing (IgMlowIgD−) (Figure 1). Indeed, upon BCR ligation, IgMhighIgD− cells were more sensitive to apoptosis, and conversely IgMlowIgD− B cells were more competent to undergo upregulation of Rag2 mRNA than IgMhighIgD− cells. Moreover, both apoptosis and upregulation of Rag2 were induced by treatment with ionomycin on IgMhighIgD− and IgMlowIgD− B cells, respectively, demonstrating that calcium mobilization is likely responsible for both receptor editing and apoptosis, depending on the developmental stage (176). Identification of downstream targets of calcium mobilization that could explain these distinct responses certainly deserves future study. Transgenic mice expressing a well-defined HEL-specific receptor on most B cells (IgHEL transgenic mice) in combination with mice expressing HEL as a self-antigen (HEL-transgenic mice) provide a well-controlled model to analyze positive versus negative signaling at the same maturation stage of B cells. In IgHEL transgenic mice, the transgenic BCR has not been exposed to HEL in the bone marrow. Acute ligation of the BCR on these naive cells with foreign HEL rapidly induces a large biphasic calcium response. NF-AT and NF-κB are translocated to the nucleus, the c-myc gene is induced within 30 min, and JNK and ERK are activated. By contrast to single IgHEL transgenic mice, in HEL/IgHEL double-transgenic mice, which express soluble HEL as a circulating self-antigen, the concentration and avidity of antigen and the maturation stage of the cells are the same, but the timing of B cell encounter

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

583

with HEL and costimulation are different. In these double-transgenic mice, binding of soluble HEL begins as soon as BCRs are expressed on immature B cells in the bone marrow. During the initial encounter in the bone marrow and during chronic exposure in the periphery, no positive responses occur in the double-transgenic mice. These tolerant B cells exhibit chronic calcium oscillations, continuous shuttling of NF-AT to the nucleus and activation of ERK, whereas neither NFκB nor JNK is activated, nor is c-myc, B7.2, or CD69 induced (177). Thus, the negative response to chronic selfantigen triggers only a subset of the signals that are active in the acute positive response made by naive B cells. These distinct signals are likely to reflect differential defects in the activity of proximal signaling molecules in tolerant B cells. HEL-induced tyrosine phosphorylation of Syk in tolerant cells is dramatically inhibited compared with that in naive cells. HEL also induces very little increase in tyrosine phosphorylation of Igα and Igβ in tolerant cells. In contrast to downregulation of Syk in tolerant cells, the Src-family PTKs appear to be still active in tolerant cells (178). Thus, one potential explanation is that expression of Syk kinase inhibitors might be induced during chronic exposure to HEL. Apart from activation signals through the BCR, what is the molecular basis for the persistent signal by the BCR? Although little information is currently available regarding molecules involved in this persistent signal, some hints might be gleaned from experiments using pervanadate/H2O2 to stimulate J558L myeloma cell lines. Treatment of cells with this chemical inhibits PTPases, activates PTKs, and increases phosphorylation of substrates. Most PTK substrates are phosphorylated only in BCR-positive transfectants of J558L, but not in the BCR-negative parental cells (179). These data suggest that, once BCR is expressed on the cell surface, the BCR organizes PTKs and substrate proteins. Thus, assuming that components such as scaffold proteins are required for this organization of PTKs and their substrates, this scaffold protein molecule might be involved in generating the persistent signal.

FUTURE DIRECTIONS Just over a decade has passed since the first description of the ITAM within the Igα/Igβ subunits of the BCR. Remarkable progress has occurred in enumerating the signal transduction elements that link the BCR to the cell interior. However, as the signaling components become more and more detailed, we are in danger of losing an overview of the organization scheme that determines the function of different signaling pathways. Proper dissection of the signal transduction circuitry controlling B cell responsiveness will require the development of novel analytical techniques to measure biochemical changes in single cells.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

584

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

Deletion or activation of genes under regulated conditions, as well as the use of fluorescent proteins whose location and expression can be followed inside living cells, will be critical for these efforts. Deciphering the regulatory pathways that select anergy, apoptosis, editing, or activation from a menu of alternative signaling pathways will provide a means to develop novel immunomodulatory drugs.

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ACKNOWLEDGMENTS I wish to acknowledge Kazumi Noda for the artwork and members of my laboratory who shared their unpublished observations. I am grateful to Drs. Steven Greenberg, Satoshi Tsukada, and Msamitsu Iino for critical discussion. Work from my laboratory was supported by grants from the Ministry of Education, Science, Sports, and Culture of Japan. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Bumet FM. 1959. The Clonal Selection Theory of Acquired Immunity. Cambridge, UK: Cambridge Univ. Press 2. Talmage DW. 1959. Clonal selection theory. Science 129:1643–48 3. Rajewsky K. 1996. Clonal selection and learning in the antibody system. Nature 381:751–58 4. Kitamura D, Roes J, Kuhn R, Rajewsky K. 1991. A B cell deficient mouse by targeted disruption of the membrane exons of the immunoglobulin µ chain gene. Nature 350:423–26 5. Kitamura D, Rajewsky K. 1992. Targeted disruption of mu chain membrane exon causes loss of heavy-chain allelic exclusion. Nature 356:154–56 6. Kitamura D, Kudo A, Schaal S, Muller W, Melchers F, Rajewsky K. 1992. A critical role of lambda 5 protein in B cell development. Cell 69:823–31 7. Rolink A, Karasuyama H, Grawunder U, Haasner D, Kudo A, Melchers F. 1993. B cell development in mice with a defective lambda 5 gene. Eur. J. Immunol. 23:1284– 88 8. Reth M. 1989. Antigen receptor tail clue. Nature 338:383–84 9. Law DA, Chan VWF, Datta SK, DeFranco AL. 1993. B-cell antigen receptor motifs have redundant signalling capabilities and bind the tyrosine kinases PTK72, Lyn, and Fyn. Curr. Biol. 3:645–57

10. Sanchez M, Misulovin Z, Burkhardt AL, Mahajan S, Costa T, Franke R, Bolen JB, Nussenzweig M. 1993. Signal transduction by immunoglobulin is mediated through Igα and Igβ. J. Exp. Med. 178:1049–55 11. Papavasiliou F, Misulovin Z, Suh H, Nussenzweig MC. 1995. The role of Igβ in precursor B cell transition and allelic excludion. Science 268:408–11 12. Teh Y-M, Neuberger MS. 1997. The immunoglobulin (Ig)α and Igβ cytoplasmic domains are independently sufficient to signal B cell maturation and activation in transgenic mice. J. Exp. Med. 185:1753– 58 13. Gong S, Nussenzweig MC. 1996. Regulation of an early developmental checkpoint in the B cell pathway by Igβ. Science 272:411–14 14. Nagata K, Nakamura T, Kitamura F, Kuramochi S, Taki S, Campbell KS, Karasuyama H. 1997. The Igα/Igβ heterodimer on µ-negative pro B cells in competent for transducing signals to induce early B cell differentiation. Immunity 7:559–70 15. Gong S, Sanchez M, Nussenzweig MC. 1996. Counterselection against Dµ is mediated through Igα-Igβ. J. Exp. Med. 184:2079–84 16. Torres RM, Flaswinkel H, Reth M, Rajewsky K. 1996. Aberrant B cell

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

17.

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

development and immune response in mice with a compromised BCR complex. Science 272:1804–8 Lam K-P, K¨uhn R, Rajewsky K. 1997. In vivo ablation of surface immunogloubulin on mature B cells by inducible gene targeting results in rapid cell death. Cell 90:1073–83 Hibbs ML, Tarlinton DM, Armes J, Grail D, Hodgson G, Maglitto R, Stacker SA, Dunn AR. 1995. Multiple defects in the immune system of Lyn-deficient mice, culminating in autoimmune disease. Cell 83:301–11 Nishizumi H, Taniuchi I, Yamanashi Y, Kitamura D, Ilic D, Mori S, Watanabe T, Yamamoto T. 1995. Impaired proliferation of peripheral B cells and indication of autoimmune disease in Lyn-deficient mice. Immunity 3:549–60 Chan VWF, Meng F, Soriano P, DeFranco AL, Lowell CA. 1997. Characterization of the B lymphocyte populations in Lyndeficient mice and the role of Lyn in signal intiation and down-regulation. Immunity 7:69–81 Cornall RJ, Cyster JG, Hibbs ML, Dunn AR, Otipoby KL, Clark EA, Goodnow CC. 1998. Polygenic autoimmune traits: Lyn, CD22 and SHP-1 are limiting elements of a biochemical pathway regulating BCR signaling and selection. Immunity 8:497–508 O’Keefe TL, Williams GT, Davies SL, Nueberger MS. 1996. Hyperresponsive B cells in CD22-deficient mice. Science 274:798–801 Otipoby KL, Andersson KB, Draves KE, Klaus SJ, Farr AG, Kerner JD, Perlmutter RM, Law C-L, Clark EA. 1996. CD22 regulates thymus-independent responses and the lifespan of B cells. Nature 384:634–37 Nitschke L, Carsetti R, Ocker B, Kohler G, Lamers MC. 1997. CD22 is a negative regulator of B cell receptor signaling. Curr. Biol. 7:133–43 Sato S, Miller AS, Inaoki M, Bock CB, Jansen PJ, Tang MLK, Tedder TF. 1996. CD22 is both a positive and negative regulator of B lymphocyte antigen receptor signal transduction: altered signaling in CD22-deficient mice. Immunity 5:551–62 Doody GM, Justement LB, Delibrias CC, Matthews RJ, Lin J, Thomas ML, Fearon DT. 1995. A role in B cell activation for CD22 and the protein tyrosine phosphatase SHP. Science 269:242–44 Nishizumi H, Horikawa K, MlinaricRascan I, Yamamoto T. 1998. A doubleedged kinase Lyn: a positive and negative

28.

29.

30.

31.

32.

33.

34.

35.

36. 37.

38.

585

regulator for antigen receptor-mediated signals. J. Exp. Med. 187:1343–48 Smith KGC, Tarlinton DM, Doody GM, Hibbs ML, Fearon DT. 1998. Inhibition of the B cell by CD22: A requirement for Lyn. J. Exp. Med. 187:807–11 Turner M, Mee PJ, Costello PS, Williams O, Price AA, Duddy LP, Furlong MT, Geahlen RL, Tybulewicz VLJ. 1995. Perinatal lethality and blocked B-cell development in mice lacking the tyrosine kinase Syk. Nature 378:298–302 Cheng AM, Rowley B, Pao W, Hayday A, Bolen JB, Pawson T. 1995. Syk tyrosine kinase required for mouse viability and B-cell development. Nature 378:303–6 Turner M, Gulbranson-Judge A, Quinn ME, Walters AE, MacLennan ICM, Tybulewicz VLJ. 1997. Syk tyrosine kinases is required for the positive selection of immature B cells into the recirculating B cell pool. J. Exp. Med. 186:2013–21 Forster R, Mattis AE, Kremmer E, Wolf E, Brem G, Lipp M. 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 Tsukada S, Saffran DC, Rawlings DJ, Parolini O, Allen RC, Klisak I, Sparkes RS, Kubagawa H, Mohandas T, Quan S, Belmont JW, Cooper MD, Conley ME, Witte ON. 1993. Deficient expression of a B cell cytoplasmic tyrosine kinase in human X-linked agammaglobulinemia. Cell 72:279–90 Vetrie D, Vorechovsky I, Sideras P, Holland J, Davies A, Flinter F, Hammarstrom L, Kinnon C, Levinsky R, Bobrow M, Smith CIE, Bentley DR. 1993. The gene involved in X-linked agammaglobulinaemia is a member of the src family of protein-tyrosine kinases. Nature 361:226–33 Rawlings DJ, Saffran DC, Tsukada S, Largaespada DA, Grimaldi JC, Cohen RN, Mohr JF, Bazan M, Howard M, Copeland NG, Jenkins NA, Witte ON. 1993. Mutation of unique region of Bruton’s tyrosine kinase in immunodeficient XID mice. Science 261:358–61 Rawlings DJ, Witte ON. 1994. Bruton’s tyrosine kinase is a key regulator in B-cell development. Immunol. Rev. 138:105–19 Smith CI, Islam KB, Vorechovsky I, Olerup O, Wallin E, Rabbani H, Baskin B, Hammarstrom L. 1994. X-linked agammaglobulinemia and other immunoglobulin deficiencies. Immunol. Rev. 138:159– 83 Kerner JD, Appleby MW, Mohr S, Chien

P1: SAT/spd

P2: KKK/plb

January 21, 1999

586

39.

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

40.

41.

42.

43.

44.

45.

46. 47. 48.

49.

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI DJ, Rawlings DJ, Maliszewski CR, Witte ON, Perlmutter RM. 1995. Impaired expansion of mouse B cell progenitors lacking Btk. Immunity 3:301–12 Khan WN, Alt FW, Gerstein RM, Malynn BA, Larsson I, Rathbun G, Davidson L, M¨uller S, Kantor AB, Herzenberg LA, Rosen FS, Sideras P. 1995. Defective B cell development and function in Btkdeficient mice. Immunity 3:283–99 Hendriks RW, de Bruijn MFTR, Mass A, Dingjan GM, Karis A, Grosveld F. 1996. Inactivation of BTK by insertion of lacZ reveals defects in B cell development only past the pre-B cell stage. EMBO J. 15:4862–72 Byth KF, Conroy LA, Howlett S, Smith AJH, May J, Alexander DR, Holmes N. 1996. CD45-null transgenic mice reveal a positive regulatory role for CD45 in early thymocyte development, in the selection of CD4+ CD8+ thymocytes, and B cell maturation. J. Exp. Med. 183:1707–18 Kishihara K, Penninger J, Wallace VA, Kundig TM, Kawai K, Ohashi PS, Thomas ML, Furlonger C, Paige CJ, Mak TW. 1993. Normal B lymphocyte development but impaired T cell maturation in CD45-exon6 protein tyrosine phosphatase-deficient mice. Cell 74:143– 56 Benatar T, Carsetti R, Furlonger C, Kamalia N, Mak T, Paige CJ. 1996. Immunoglobulin-mediated signal transduction in B cells from CD45-deficient mice. J. Exp. Med. 183:329–34 Cyster JG, Healy JI, Kishihara K, Mak TW, Thomas ML, Goodnow CC. 1996. Regulation of B-lymphocyte negative and positive selection by tyrosine phosphatase CD45. Nature 381:325–28 Bignon JS, Siminovitch KA. 1994. Identification of PTP1C mutation as a genetic defect in motheaten and viable motheaten mice: a step toward defining the roles of protein tyrosine phosphtatases in the regulation of hemopoietic cell differentiation and function. Clin. Immunol. Immunopathol. 73:168–79 Shults LD, Sidman CL. 1987. Genetically determined murine models of immunodeficiency. Annu. Rev. Immunol. 5:367–403 Tsui FWL, Tsui HW. 1994. Molecular basis of the motheaten phenotype. Immunol. Rev. 136:185–206 Cyster JG, Goodnow CC. 1995. Protein tyrosine phosphatase 1C negatively regulates antigen receptor signaling in B lymphocytes and determines thresholds for negative selection. Immunity 2:13–24 Pani G, Siminovitch KA, Paige CJ. 1997.

50. 51.

52.

53.

54.

55.

56.

57.

58. 59.

60.

61.

The motheaten mutation rescues B cell signaling and development in CD45– deficient mice. J. Exp. Med. 186:581–88 Pawson T, Scott JD. 1997. Signaling through scaffold, anchoring, and adaptor proteins. Science 278:2075–80 Dell’Acqua ML, Faux MC, Thorburn J, Thoburn A, Scott JD. 1998. Membrane-targeting sequences on AKAP79 bind phosphatidylinositol-4,5-bisphosphate. EMBO J. 17:2246–60 Xavier R, Brennan T, Li Q, McCormack C, Seed B. 1998. Membrane compartmentation is required for efficient T cell activation. Immunity 8:723–32 Saouaf SJ, Mahajan S, Rowley RB, Kut SA, Fargnoli J, Burkhardt AL, Tsukada S, Witte ON, Bolen JB. 1994. Temporal differences in the activation of three classes of non-transmembrane protein tyrosine kinases following B-cell antigen receptor surface engagement. Proc. Natl. Acad. Sci. USA 91:9524–28 Pleiman CM, Abrams C, Gauen LT, Bedzyk W, Jongstra J, Shaw AS, Cambier JC. 1994. Distinct p53/56lyn and p59fyn domains associate with nonphosphorylated and phosphorylated Ig-α. Proc. Natl. Acad. Sci. USA 91:4268–72 Kabouridis PS, Magee AI, Ley SC. 1997. S-acylation of LCK protein tyrosine kinase is essential for its signalling function in T lymphocytes. EMBO J. 16:4983– 98 Takata M, Sabe H, Hata A, Inazu T, Homma Y, Nukada T, Yamamura H, Kurosaki T. 1994. Tyrosine kinases Lyn and Syk regulate B cell receptor-coupled Ca2+ mobilization through distinct pathways. EMBO J. 13:1341–49 Takata M, Kurosaki T. 1996. A role for Bruton’s tyrosine kinase in BCRmediated activation of phospholipase C-γ 2. J. Exp. Med. 184:31–40 Cooper JA, Howell B. 1993. The when and how of Src regulation. Cell. 73:1051– 54 Yanagi S, Sugawara H, Kurosaki M, Sabe H, Yamamura H, Kurosaki T. 1996. CD45 modulates phosphorylation of both autophosphorylation and negative regulatory tyrosines of Lyn in B cells. J. Biol. Chem. 271:30487–92 Pao LI, Cambier JC. 1997. Syk, but not Lyn, recruitment to B cell antigen receptor and activation following stimulation of CD45− B cells. J. Immunol. 158:2663– 69 Hata A, Sabe H, Kurosaki T, Takata M, Hanafusa H. 1994. Functional analysis of Csk in signal transduction through the

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

62.

63.

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

64.

65. 66. 67. 68. 69.

70.

71.

72.

73.

74.

B-cell antigen receptor. Mol. Cell. Biol. 14:7306–13 Pao LI, Bedzyk WD, Persin C, Cambier JC. 1997. Molecular targets of CD45 in B cell antigen receptor signal transduction. J. Immunol. 158:1116–24 Justement LB, Campbell KS, Chien NC, Cambier JC. 1991. Regulation of B cell antigen receptor signal transduction and phosphorylation by CD45. Science 252:1839–42 Ogimoto M, Katagiri T, Mashima K, Hasegawa K, Mizuno K, Yakura H. 1994. Negative regulation of apoptotic death in immature B cells by CD45. Int. Immunol. 6:647–54 Xu W, Harrison SC, Eck M. 1997. Threedimensional structure of the tyrosine kinase c-Src. Nature 385:595–602 Sicheri F, Moarefi I, Kuriyan J. 1997. Crystal structure of the Src family tyrosine kinase Hck. Nature 385:602–9 Weiss A, Littman DR. 1994. Signal transduction by lymphocyte antigen receptors. Cell 76:263–74 Kolanus W, Romeo C, Seed B. 1993. T cell activation by clustered tyrosine kinases. Cell 74:171–83 Chu DH, Spits H, Peyron J-F, Rowley RB, Bolen JB, Weiss A. 1996. The syk protein tyrosine kinase can function independently of CD45 or Lck in T cell antigen receptor signaling. EMBO J. 15:6251–61 Latour S, Fournel M, Veillette A. 1997. Regulation of T-cell antigen receptor signalling by Syk tyrosine protein kinase. Mol. Cell. Biol. 17:4434–41 Hutchcroft JE, Harrison ML, Geahlen RL. 1992. Association of the 72-kDa protein-tyrosine kinase PTK72 with the B cell antigen receptor. J. Biol. Chem. 267:8613–19 Kurosaki T, Johnson SA, Pao L, Sada K, Yamamura H, Cambier JC. 1995. Role of the Syk autophosphorylation site and SH2 domains in B cell receptor signaling. J. Exp. Med. 182:1815–23 Rowley RB, Burkhardt AL, Chao HG, Matsueda GR, Bolen JB. 1995. Syk protein-tyrosine kinase is regulated by tyrosine-phosphorylated Igα/Igβ immunoreceptor tyrosine activation motif binding and autophosphorylation. J. Biol. Chem. 270:11590–94 Hatada MH, Lu X, Laird ER, Green J, Morgenstern JP, Lou M, Marr CS, Phillips TB, Ram MK, Theriault K, Zoller MJ, Karas JL. 1995. Molecular basis for interaction of the protein tyrosine kinase ZAP-70 with the T-cell receptor. Nature 377:32–38

587

75. Latour S, Zhang J, Siraganian RP, Veillette A. 1998. A unique insert in the linker domain of Syk is necessary for its function in immunoreceptor signalling. EMBO J. 17:2584–95 76. Kurosaki T, Takata M, Yamanashi Y, Inazu T, Taniguchi T, Yamamoto T, Yamamura H. 1994. Syk activation by the Src-family tyrosine kinase in the B cell receptor signaling. J. Exp. Med. 179:1725– 29 77. Keshvara LM, Isaacson C, Harrison L, Geahlen RL. 1997. Syk activation and dissociation from the B-cell antigen receptor is mediated by phosphorylation of tyrosine 130. J. Biol. Chem. 272:10377–81 78. Ota Y, Samelson LE. 1997. The product of the proto-oncogene c-cbl: a negative regulator of the Syk tyrosine kinase. Science 276:418–20 79. Murphy MA, Schnall RG, Venter DJ, Barnett L, Bertoncello I, Thien CBF, Langdon WY, Bowtell DDL. 1998. Tissue hyperplasia and enhanced T-cell signalling via ZAP-70 in c-Cbl-deficient mice. Mol. Cell. Biol. 18:4872–82 80. Starr R, Willson TA, Viney EM, Murray LJL, Rayner JR, Jenkins BJ, Gonda TJ, Alexander WS, Metcalf D, Nicola NA, Hilton DJ. 1997. A family of cytokineinducible inhibitors of signalling. Nature 387:917–21 81. Endo TA, Masuhara M, Yokouchi M, Suzuki R, Sakamoto H, Mitsui K, Maysumoto A, Tanimura S, Ohtsubo M, Miaswa H, Miyazaki T, Leonor N, Taniguchi T, Fujita T, Kanakura Y, Komiya S, Yoshimura A. 1997. A new protein containing an SH2 domain that inhibits JAK kinases. Nature 387:921–24 82. Naka T, Narazaki M, Hirata M, Matsumoto T, Minamoto S, Aono A, Nishimoto N, Kajita T, Taga T, Yoshizaki K, Akira S, Kishimoto T. 1997. Structure and function of a new STAT-induced STAT inhibitor. Nature 387:924–29 83. Mahajan S, Fargnoli J, Burkhardt AL, Kut SA, Saouaf SJ, Bolen JB. 1995. Src family protein tyrosine kinases induce autoactivation of Bruton’s tyrosine kinase. Mol. Cell. Biol. 15:5304–11 84. Rawlings DJ, Scharenberg AM, Park H, Wahl MI, Lin S, Kato RM, Fluckiger A-C, Witte ON, Kinet J-P. 1996. Activation of BTK by a phosphorylation mechanism initiated by SRC family kinases. Science 271:822–25 85. Kurosaki T, Kurosaki M. 1997. Transphosphorylation of Btk on tyrosine 551 is critical for B cell antigen receptor function. J. Biol. Chem. 272:15595–98

P1: SAT/spd

P2: KKK/plb

January 21, 1999

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

588

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI

86. Tarakhovsky A. 1997. Xid and Xidlike immunodeficiencies from a signaling point of view. Curr. Opin. Immunol. 9:319–23 87. Park H, Wahl MI, Afar DEH, Turck CW, Rawlings DJ, Tam C, Scharenberg AM, Kinet J-P, Witte ON. 1996. Regulation of Btk function by a major autophosphorylation site within the SH3 domain. Immunity 4:515–25 88. Andreotti AH, Bunnell SC, Feng S, Berg LJ, Schreiber SL. 1997. Regulatory intramolecular association in a tyrosine kinase of the Tec family. Nature 385:93–97 89. Matsushita M, Yamadori T, Kato S, Takemoto Y, Inazawa J, Baba Y, Hashimoto S, Sekine S, Arai S, Kunikata T, Kurimoto M, Kishimoto T, Tsukada S. 1998. Identification and characterization of a novel SH3-domain binding protein, Sab, which prefentially associates with Bruton’s Tyrosine kinase (Btk). Biochem. Biophys. Res. Comm. 245:337–43 90. Salim K, Bottomley MJ, Querfurth E, Zvelebil MJ, Gout I, Scaife R, Margolis RL, Gigg R, Smith CIE, Driscoll PC, Waterfield MD, Panayotou G. 1996. Distinct specificity in the recognition of phosphoinositides by the pleckstrin homology domains of dynamin and Bruton’s tyrosine kinase. EMBO J. 15:6241–50 91. Li Z, Wahl MI, Eguinoa A, Stephens LR, Hawkins PT, Witte ON. 1997. Phosphatidylinositol 3-kinase-γ activates Bruton’s tyrosine kinase in concert with Src family kinases. Proc. Natl. Acad. Sci. USA 94:13820–25 92. August A, Sadra A, Dupont B, Hanafusa H. 1997. Src-induced activation of inducible T cell kinase (ITK) requires phosphatidylinositol 3-kinase activity and the pleckstrin homology domain of inducible T cell kinase. Proc. Natl. Acad. Sci. USA 94:11227–32 93. Law C-L, Chandran KA, Sidorenko SP, Clark EA. 1996. Phospholipase C-γ 1 interacts with conserved phosphotyrosyl residues in the linker region of Syk and is a substrate for Syk. Mol. Cell. Biol. 16:1305–15 94. Takata M, Homma Y, Kurosaki T. 1995. Requirement of phospholipase C-γ 2 activation in surface IgM-induced B cell apoptosis. J. Exp. Med. 182:907–14 95. Fu C, Turck C, Kurosaki T, Chan AC. 1998. BLNK: A central linker protein in B cell activation. Immunity 9:93–103 96. Wienands J, Schweikert J, Wollscheid B, Jumaa H, Nielsen PJ, Reth M. 1998. SLP65: A new signaling component in B lymphocytes which requires expression of the

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

antigen receptor for phosphorylation. J. Exp. Med. 188:791–95 Zhang W, Sloan-Lancaster J, Kitchen J, Trible RP, Samelson LE. 1998. LAT: the ZAP-70 tyrosine kinase substrate that links T cell receptor to cellular activation. Cell 92:83–92 Weber JR, Orstavik S, Torgersen KM, Danbolt NC, Berg SF, Ryan JC, Tasken K, Imboden JB, Vaage JT. 1998. Molecular cloning of the cDNA encoding pp36, a tyrosine-phosphorylated adaptor protein selectively expressed by T cells and natural killer cells. J. Exp. Med. 187:1157– 61 Peterson EJ, Clements JL, Fang N, Koretzky GA. 1998. Adaptor proteins in lymphocyte antigen-receptor signaling. Curr. Opin. Immunol. 10:337–44 Fluckiger A-C, Li Z, Kato RM, Wahl MI, Ochs HD, Longnecker R, Kinet J-P, Witte ON, Scharenberg AM, Rawlings DJ. 1998. Btk/Tec kinases regulate sustained increases in intracellular Ca2+ following B-cell receptor activation. EMBO J. 17:1973–85 Liu K-Q, 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 Falasca M, Logan SK, Lehto VP, Baccante G, Lemmon MA, Schlessinger J. 1998. Activation of phospholipase Cγ by PI3-kinase-induced PH domain-mediated membrane targeting. EMBO J. 17:414– 22 O’Rourke LM, Tooze R, Turner M, Sandoval DM, Carter RH, Tybulewicz VLJ, Fearon DT. 1998. CD19 as a membraneanchored adaptor protein of B lymphocytes: costimulation of lipid and protein kinases by recruitment of Vav. Immunity 8:635–45 Sugawara H, Kurosaki M, Takata M, Kurosaki T. 1997. Genetic evidence for involvement of type 1, type 2 and type 3 inositol 1,4,5-trisphosphate receptors in signal transduction through the B cell antigen receptor. EMBO J. 16:3078–88 Dolmetsch RE, Lewis RS, Goodnow CC, Healy JI. 1997. Differential activation of transcription factors induced by Ca2+ response amplitude and duration. Nature 386:855–58 Dolmetsch RE, Xu K, Lewis R. 1998. Calcium oscillations increase the efficiency and specificity of gene expression. Nature 392:933–36 Li W-H, Llopis J, Whitney M, Zlokarnik G, Tsien RY. 1998. Cell-permeant caged

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING

108.

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

InsP3 ester shows that Ca2+ spike frequency can optimize gene expression. Nature 392:936–41 Jayaraman T, Ondrias K, Ondriasova E, Marks AR. 1996. Regulation of the inositol 1,4,5-trisphosphate receptor by tyrosine phosphorylation. Science 272:1492– 94 Gold MR, Aebersold R. 1994. Both phosphatidylinositol 3-kinase and phosphatidylinositol 4-kinase products are increased by antigen receptor signaling in B cells. J. Immunol. 152:42–50 Beckwith M, Fenton RG, Katona IM, Longo DL. 1996. Phosphatidylinositol-3kinase activity is required for the anti-Igmediated growth inhibition of a human B-lymphoma cell line. Blood 87:202– 10 Tuveson DA, Carter RH, Soltoff SP, Fearon DT. 1993. CD19 of B cells as a surrogate kinase insert region to bind phophatidylinositol 3-kinase. Science 260:986–89 Pleiman CM, Hertz WM, Cambier JC. 1994. Activation of phosphatidylinositol30 kinase by Src-family kinase SH3 binding to the p85 subunit. Science 263:1609– 12 Damen JE, Liu L, Rosten P, Humphries RK, Jefferson AB, Majerus PW, Krystal G. 1996. The 145-kDa protein induced to associate with Shc by multiple cytokines is an inositol tetraphosphate and phosphatidylinositol 3,4,5-trisphosphate 5-phosphatase. Proc. Natl. Acad. Sci. USA 93:1689–93 Helgason CD, Damen JE, Rosten P, Grewal R, Sorensen P, Chappel SM, Borowski A, Jirik F, Krystal G, Humphries RK. 1998. Targeted disruption of SHIP leads to hemopoietic perturbations, lung pathology, and a shortened life span. Genes & Dev. 12:1610–20 Kiener PA, Liboubin MN, Rohrschneider LR, Ledbetter JA, Nadler SG, Diegel ML. 1997. Co-ligation of the Antigen and Fc receptors give rise to the selective modulation of intracellular signaling in B cells. J. Biol. Chem. 272:3838–44 Bolland S, Pearse R, Kurosaki T, Ravetch JV. 1998. SHIP Modulates immune receptor responses by regulating membrane association of Btk. Immunity 8:509–16 Iritani BM, Forbush KA, Farrar MA, Perlmutter RM. 1997. Control of B cell development by Ras-mediated activation of Raf. EMBO J. 16:7019–31 Holsinger LJ, Spencer DM, Austin DJ, Schreiber SL, Crabtree GR. 1995. Signal transduction in T lymphocytes using

119.

120.

121.

122.

123.

124.

125.

126.

127.

589

a conditional allele of Sos. Proc. Natl. Acad. Sci. USA 92:9810–14 Hashimoto A, Okada H, Jiang A, Kurosaki M, Greenberg S, Clark EA, Kurosaki T. 1998. Involvement of Guanosine Triphosphatases and Phospholipase C-γ 2 in Extracellular signal-regulated kinase, c-Jun NH2-terminal kinase, and p38 mitogen-activated protein kinase activation by the B cell antigen receptor. J. Exp. Med. 188:1287–95 Jiang A, Craxton A, Kurosaki T, Clark EA. 1998. Different protein tyrosine kinases are required for B cell antigen receptor-mediated extracellular signalregulated kinase, c-Jun NH2-terminal kinase 1, and p38 mitogen-activated protein kinase activation. J. Exp. Med. 188:1297– 306 Ingham RJ, Krebs DL, Barbazuk SM, Turck CW, Hirai H, Matsuda M, Gold MR. 1996. B cell antigen receptor signaling induces the formation of complexes containing the Crk adapter proteins. J. Biol. Chem. 271:32306–14 Smit L, van der Horst G, Borst J. 1996. Sos, Vav, and C3G participate in B cell receptor-induced signaling pathways and differentially associate with Shc-Grb2, Crk, and Crk-L adaptors. J. Biol. Chem. 271:8564–69 McCormick F, Wittinghofer A. 1996. Interactions between Ras proteins and their effectors. Curr. Opin. Biotechnol. 7:449– 56 Fukazawa T, Reedquist KA, Trub T, Soltoff S, Panchamoorthy G, Druker B, Cantley L, Shoelson SE, Band H. 1995. The SH3 domain-binding T cell tyrosyl phosphoprotein p120. Demonstration of its identity with the c-cbl protooncogene product and in vivo complexes with Fyn, Grb2, and phosphatidylinositol 3-kinase. J. Biol. Chem. 270:19141–50 Tezuka T, Umemori H, Fusaki N, Yagi T, Takata M, Kurosaki T, Yamamoto T. 1996. Physical and functional association of the cbl proto-oncogene product with a Src-family protein tyrosine kinases, p53/56lyn, in the B cell antigen receptor-mediated signaling. J. Exp. Med. 183:675–80 Ota Y, Beitz LO, Scharenberg AM, Donovan JA, Kinet J-P, Samelson LE. 1996. Characterization of Cbl tyrosine phosphorylation and a Cbl-Syk complex in RBL2H3 cells. J. Exp. Med. 184:1713–23 Gold MR, Crowley MT, Martin GA, McCormick F, DeFranco AL. 1993. Targets of B lymphocyte antigen receptor signal transduction include the p21ras GTPase-

P1: SAT/spd

P2: KKK/plb

January 21, 1999

590

128.

129.

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

130.

131.

132.

133.

134.

135.

136.

137.

138.

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI activating protein (GAP) and two GAPassociated proteins. J. Immunol. 150:377– 86 Lazarus AH, Kawauchi K, Rapoport MJ, Delovitch TL. 1993. Antigen-induced B lymphocyte activation involves the p21ras and ras·GAP signaling pathway. J. Exp. Med. 178:1765–69 Settleman J, Albright CF, Forest LC, Weinberg RA. 1992. Association between GTPase activators for Rho and Ras families. Nature 359:153–54 Carpino N, Wisniewski D, Strife A, Marshak D, Kobayashi R, Stillman B, Clarkson B. 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 Tarakhovsky A, Turner M, Schaal S, Mee PJ, Duddy LP, Rajewsky K, Tybulewicz VLJ. 1995. Defective antigen receptormediated proliferation of B and T cells in the absence of Vav. Nature 374:467–70 Zhang R, Alt FW, Davidzon L, Orkin SH, Swat W. 1995. Defective signalling through the T- and B-cell antigen receptors in lymphoid cells lacking the vav proto-oncogene. Nature 374:470–73 Fischer K-D, Zmuidzinas A, Gardner S, Barbacid M, Bernstein A, Guidos C. 1995. Defective T-cell receptor signalling and positive selection of Vavdeficient CD4+ CD8+ thymocytes. Nature 374:474–77 Fischer K-D, Kong Y-Y, Nishina H, Tedford K, Marengere LEM, Kozieradzki I, Sasaki T, Starr M, Chan G, Gardener S, Nghiem MP, Bouchard D, Barbacid M, Bernstein A, Penninger JM. 1998. Vav is regulator of cytoskeletal reorganization mediated by the T-cell receptor. Curr. Biol. 8:554–62 Holsinger LJ, Graef IA, Swat W, Chi T, Bautista DM, Davidzon L, Lweis RS, Alt FW, Crabtree GR. 1998. Defects in actin-cap formation in Vav-deficient mice implicate an actin requirement for lymphocyte signal transduction. Curr. Biol. 8:563–72 Crespo P, Schuebel KE, Ostrom AA, Gutkind JS, Bustelo XR. 1997. Phosphotyrosine-dependent activation of Rac-1 GDP/GTP exchange by the vav proto-oncogene product. Nature 385: 169–72 Wu J, Motto DG, Koretzky GA, Weiss A.

139.

140.

141.

142.

143. 144.

145.

146.

147.

148.

149.

1996. Vav and SLP-76 interact and functionally cooperate in IL-2 gene activation. Immunity 4:593–602 Deckert M, Tartare-Deckert S, Couture C, Mustelin T, Altman A. 1996. Functional and physical interactions of Syk-family kinases with the Vav protooncogene product. Immunity 5:591–604 Han J, Luby-Phelps K, Das K, Shu X, Xia Y, Mosteller RD, Krishna UM, Falck JR, White MA, Broek D. 1998. Role of substrates and products of PI3-kinase in regulating activation of Rac-related guanosine triphosphatases by Vav. Science 279:558– 60 Nonoyama S, Ochs HD. 1998. Characterization of the Wiskott-Aldrich syndrome protein and its role in the disease. Curr. Opin. Immunol. 10:407–12 Symons M, Derry JMJ, Karlark B, Jiang S, Lemahieu V, McCormick F, Francke U, Abo A. 1996. Wiskott-Aldrich syndrome protein, a novel effector for the GTPasa CDC42Hs, is implicated in actin polymerization. Cell 84:723–34 Schreiber SL, Crabtree GR. 1992. The mechanism of action of cyclosporin A and FK506. Immunol. Today. 13:136–42 Choi MSK, Brines RD, Holman MJ, Klaus GGB. 1994. Induction of NFAT in normal B lymphocytes by antiimmunoglobulin or CD40 ligand in conjunction with IL-4. Immunity 1:179–87 Ventakaraman L, Francis DA, Wang Z, Liu J, Rothstein TL, Sen R. 1994. Cyclosporin-A sensitive induction of NFAT in murine B cells. Immunity 1:189–96 Genestier L, Dearden-Badet MT, Bonnefoy-Berard N, Lizard G, Revillard JP. 1994. Cyclosporin A and FK506 inhibit activation-induced death in the murine WEHI-231 B cell line. Cell Immunol. 155:283–91 Graves JD, Draves KE, Craxton A, Saklatvala J, Krebs EG, Clark EA. 1996. Involvement of stress-activated protein kinase and p38 mitogen-activated protein kinase in mIgM-induced apoptosis of human B lymphocytes. Proc. Natl. Acad. Sci. USA 93:13814–18 Zhu J, Shibasaki F, Price R, Guillemot J-C, Yano T, Dotsch V, Wagner G, Ferrara P, McKeon F. 1998. Intramolecular masking of nuclear import signal on NFAT4 by casein kinase I and MEKK1. Cell 93:851–61 Beals CR, Clipstone NA, Ho SN, Crabtree GR. 1997. Nuclear localization of NF-ATc by a calcineurin-dependent, cyclosporin-sensitive intramolecular interaction. Genes & Dev. 11:824–34

P1: SAT/spd

P2: KKK/plb

January 21, 1999

QC: KKK/mbg/plb

15:34

T1: KKK

Annual Reviews

AR078-18

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

GENETICS OF B CELL ANTIGEN RECEPTOR SIGNALING 150. Beals CR, Sheridan CM, Turck CW, Gardner P, Crabtree GR. 1997. Nuclear export of NF-ATc enhanced by glycogen synthase kinase-3. Science 275:1930–33 151. Fukuda T, Kitamura D, Taniuchi I, Maekawa Y, Benhamou LE, Sarthou P, Watanabe T. 1995. Restoration of surface IgM-mediated apoptosis in an antiIgM-resistant variant of WEHI-231 lymphoma cells by HS1, a protein-tyrosine kinase substrate. Proc. Natl. Acad. Sci. USA 92:7302–6 152. Yamanashi Y, Fukuda T, Nishizumi H, Inazu T, Higashi K, Kitamura D, Ishida T, Yamamura H, Watanabe T, Yamamoto T. 1997. Role of tyrosine phosphorylation of HS1 in B cell antigen receptor-mediated apoptosis. J. Exp. Med. 185:1387–92 153. Leitges M, Schmedt C, Guinamard R, Davoust J, Schaal S, Stable S, Tarakhovsky A. 1996. Immunodeficiency in protein kinase Cβ-deficient mice. Science 273:788–91 154. Sidorenko SP, Law C-L, Klaus SJ, Chandran KA, Takata M, Kurosaki T, Clark EA. 1996. Protein kinase Cµ (PKCµ) associates with the B cell antigen receptor complex and regulates lymphocyte signaling. Immunity 5:353–63 155. Franke TF, Kaplan DR, Cantley LC. 1997. PI3K: downstream AKTion blocks apoptosis. Cell 88:435–37 156. Hemmings BA. 1997. Akt signaling: linking membrane events to life and death decisions. Science 275:628–30 157. Stokoe D, Stephens LR, Copeland T, Gaffney PRJ, Reese CB, Painter GF, Holmes AB, McCormic F, Hawkins PT. 1997. Dual role of phosphatidylinositol3,4,5-trisphosphate in the activation of protein kinase B. Science 277:567– 70 158. Cross DAE, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA. 1995. Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature 378:785–89 159. Datta SR, Dudek H, Tao X, Masters S, Fu H, Gotoh Y, Greenberg ME. 1997. Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91:231–41 160. del Peso L, Gonzalez-Garcia M, Page C, Herrera R, Nunez G. 1997. Interleukin-3induced phosphorylation of BAD through the protein kinase Akt. Science 278:687– 89 161. Cohen P. 1997. The search for physiological substrates of MAP and SAP kinases in mammalian cells. Trends Cell Biol. 7:353–61

591

162. Jacinto E, Werlen G, Karin M. 1998. Cooperation between Syk and Rac1 leads to synergistic JNK activation in T lymphocytes. Immunity 8:31–41 163. Werlen G, Jacinto E, Xia Y, Karin M. 1998. Calcineurin preferentially synergizes with PKC-θ to activate JNK and IL2 promoter in T lymphocytes. EMBO J. 17:3101–11 164. Ono M, Bolland S, Tempst P, Ravetch JV. 1996. Role of the inositol phosphatase SHIP in negative regulation of the immune system by the receptor Fcγ RIIB. Nature 383:263–66 165. Ono M, Okada H, Bolland S, Yanagi S, Kurosaki T, Ravetch JV. 1997. Deletion of SHIP or SHP-1 reveals two distinct pathways for inhibitory signaling. Cell 90:293–301 166. Scharenberg AM, El-Hillal O, Fruman DA, Beitz LO, Z. L, Lin S, Gout I, Cantley LC, Rawlings DJ, Kinet J-P. 1998. Phosphatidylinositol-3,4,5-trisphosphate (Ptdlns-3,4,5-P3)/Tec kinase-dependent calcium signaling pathway: a target for SHIP-mediated inhibitory. EMBO J. 17: 1961–72 167. O’Rourke L, Tooze R, Fearon DT. 1997. Co-receptors of B lymphocytes. Curr. Opin. Immunol. 9:324–29 168. Blery M, Kubagawa H, Chen C-C, Vely F, Cooper MD, Vivier E. 1998. The paired Ig-like receptor PIR-B is an inhibitory receptor that recruits the protein-tyrosine phosphatase SHP-1. Proc. Natl. Acad. Sci. USA 95:2446–51 169. Maeda A, Kurosaki M, Ono M, Takai T, Kurosaki T. 1998. Requirement of tyrosine phosphatase SHP-1 and SHP-2 for PIR-B-mediated inhibitory signal. J. Exp. Med. 187:1335–60 170. Yellen AJ, Glenn W, Sukhatme VP, Cao SM, Monroe JG. 1991. Signaling through surface IgM in tolerance-susceptible immature murine B lymphocytes. Developmentally regulated difference in transmembrane signaling in splenic B cells from adult and neonatal mice. J. Immunol. 146:1446–54 171. Seyfert VL, McMohan SB, Glenn WD, Yellen AJ, Sukhatme VP, Cao X, Monroe JG. 1990. Methylation of an immediateearly inducible gene as a mechanism for B cell tolerance induction. Science 250:797–800 172. Russell DM, Dembic Z, Morahan G, Miller JFAP, B¨urki K, Nemazee D. 1991. Peripheral deletion of self-reactive B cells. Nature 354:308–11 173. Hartley SB, Crosbie J, Brink R, Kantor AA, Basten A, Goodnow CC. 1991.

P1: SAT/spd

P2: KKK/plb

January 21, 1999

592

174.

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

175.

176.

177.

178.

15:34

QC: KKK/mbg/plb

T1: KKK

Annual Reviews

AR078-18

KUROSAKI Elimination from peripheral lymphoid tissues of self-reactive B lymphocytes recognizing membrane-bound antigens. Nature 353:765–68 Okamoto M, Murakami M, Shimizu A, Ozaki S, Tsubata K, Kumagai S, Honjo T. 1992. A transgenic model of autoimmune hemolytic anemia. J. Exp. Med. 175:71– 79 Chen C, Nagy Z, Radic M, Hardy R, Huzar D, Camper SA, Weight M. 1995. The site and stage of anti-DNA B-cell deletion. Nature 373:252–55 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 Healy JI, Dolmetsch RE, Timmerman LA, Cyster JG, Thomas ML, Crabtree GR, Lewis RS, Goodnow CC. 1997. Different nuclear signals are activated by the B cell receptor during positive versus nagative signaling. Immunity 6:419–28 Cooke MP, W. HA, Shokat KM, Zeng Y, Finkelman FD, Linsely PS, Howard

179.

180.

181.

182.

M, Goodnow CC. 1994. Immunoglobulin signal transduction guides the specificity of B cell-T cell interactions and is blocked in tolerant self-reactive B cells. J. Exp. Med. 179:425–38 Wienands J, Larbolette O, Reth M. 1996. Evidence for a preformed transducer complex organized by the B cell antigen receptor. Proc. Natl. Acad. Sci. USA 93:7865–70 Scharenberg AM, Kinet J-P. 1998. Ptdlns3,4,5-P3: a regulatory nexus between tirosine kinases and sustained calcium signals. Cell 94:5–8 Okada H, Bolland S, Hashimoto A, Kurosaki M, Kabuyama Y, Iino M, Ravetch JV, Kurosaki T. 1998. Role of the inositol phosphatase SHIP in B cell receptor-induced Ca2+ oscillatory response. J. Immunol. (in press) Maeda A, Scharenberg AM, Tsukada S, Bolen JB, Kinet J-P, Kurosaki T. 1999. Paired Immunoglobulin-like receptor B (PIR-B) inhibits BCR-induced activation of Syk and Btk by SHP-1. Oncogene (in press)

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:555-592. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

Annu. Rev. Immunol. 1999. 17:593–623 c 1999 by Annual Reviews. All rights reserved Copyright °

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MECHANISMS OF PHAGOCYTOSIS IN MACROPHAGES Alan Aderem and David M. Underhill Department of Immunology, University of Washington, Box 357650 Seattle, Washington 98195, e-mail: [email protected] KEY WORDS:

receptors, signal transduction, pathogens, membrane traffic, cytoskeleton, apoptosis

ABSTRACT Phagocytosis of pathogens by macrophages initiates the innate immune response, which in turn orchestrates the adaptive response. In order to discriminate between infectious agents and self, macrophages have evolved a restricted number of phagocytic receptors, like the mannose receptor, that recognize conserved motifs on pathogens. Pathogens are also phagocytosed by complement receptors after relatively nonspecific opsonization with complement and by Fc receptors after specific opsonization with antibodies. All these receptors induce rearrangements in the actin cytoskeleton that lead to the internalization of the particle. However, important differences in the molecular mechanisms underlying phagocytosis by different receptors are now being appreciated. These include differences in the cytoskeletal elements that mediate ingestion, differences in vacuole maturation, and differences in inflammatory responses. Infectious agents, such as M. tuberculosis, Legionella pneumophila, and Salmonella typhimurium, enter macrophages via heterogeneous pathways and modify vacuolar maturation in a manner that favors their survival. Macrophages also play an important role in the recognition and clearance of apoptotic cells; a notable feature of this process is the absence of an inflammatory response.

INTRODUCTION Cells have evolved a variety of strategies to internalize particles and solutes, including pinocytosis, receptor-mediated endocytosis, and phagocytosis (reviewed in 1–4). Pinocytosis usually refers to the uptake of fluid and solutes, and it is closely related to receptor-mediated endocytosis, the specific 593 0732-0582/99/0410-0593$08.00

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

594

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

process through which macromolecules, viruses, and small particles enter cells. Pinocytosis and receptor-mediated endocytosis share a clathrin-based mechanism and usually occur independently of actin polymerization. By contrast, phagocytosis, the uptake of large particles (>0.5 µm) into cells, occurs by an actin-dependent mechanism and is usually independent of clathrin. While lower organisms use phagocytosis primarily for the acquisition of nutrients, phagocytosis in Metazoa occurs primarily in specialized phagocytic cells such as macrophages and neutrophils, and it has evolved into an extraordinarily complex process underlying a variety of critical biological phenomena. Thus, phagocytosis by macrophages is critical for the uptake and degradation of infectious agents and senescent cells, and it participates in development, tissue remodeling, the immune response, and inflammation. Monocytes/macrophages and neutrophils have been referred to as professional phagocytes and are very efficient at internalizing particles. On the other hand, most cells have some phagocytic capacity. For example, thyroid and bladder epithelial cells phagocytose erythrocytes in vivo, and numerous cell types have been induced to phagocytose particles in culture. A group of cells termed paraprofessional phagocytes by Rabinovitch (who also coined the terms professional and nonprofessional phagocytes) have intermediate phagocytic ability (3). These include retinal epithelial cells that internalize the effete ends of retinal rods (3). The major difference with respect to phagocytic capacity and efficiency of professional and nonprofessional phagocytes can probably be ascribed to the presence of an array of dedicated phagocytic receptors that increase particle range and phagocytic rate. Transfection of fibroblasts and epithelial cells with cDNAs encoding Fc receptors (FcRs) dramatically increases the phagocytic rate (and, obviously, particle range) (5), and this system has been used to dissect signaling pathways leading to particle internalization. However, it is clear that many other differences between professional and nonprofessional phagocytes exist that lead to the enhancement of both rate and efficiency of particle internalization. The study of phagocytosis requires insight into the mechanisms of signal transduction, actin-based motility, membrane trafficking, and infectious disease. While a basic description of phagocytosis has been available since the seminal studies of Metchnikoff (6), investigations conducted over the last decade have begun to unravel the molecular basis of this process. In this overview, we limit our focus to phagocytic mechanisms in macrophages. Other phagocytic cells such as neutrophils certainly use similar mechanisms, but important differences exist that may be important to the role each cell type plays in the immune response. Phagocytosis is extremely complex, and no single model can fully account for the diverse structures and outcomes associated with particle internalization. This complexity is in part due to the diversity of receptors capable of stimulating phagocytosis, and in part due to the capacity of a variety of microbes to influence

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MACROPHAGE PHAGOCYTOSIS

595

their fate as they are internalized. The fact that most particles are recognized by more than one receptor, and that these receptors are capable of cross-talk and synergy, further complicates our understanding. In addition, many phagocytic receptors have dual functions, often mediating both adhesion and particle internalization, and a complex relationship exists between these two related processes. Adhesion receptors and phagocytic receptors can both activate and inhibit each other’s function. For example, ligation of the fibronectin receptor (α5 β1 integrin) at the substrate-adherent surface of a monocyte establishes preconditions within the cell that permit the otherwise inactive complement receptor CR3 (αM β2 integrin) to mediate phagocytosis (7, 8). On the other hand, adherent cells often round up during phagocytosis, implying that there is competition for cytoskeletal and membrane components necessary for phagocytosis and adhesion. This notion is reinforced by the observation that many of the cytoskeletal components known to participate in adhesion are also enriched in the phagocytic cup. These include paxillin, talin, vinculin, α-actinin, protein kinase Cα, MARCKS and MacMARCKS (9, 10). Despite the complexity associated with different phagocytic mechanisms, a number of shared features follow: Particle internalization is initiated by the interaction of specific receptors on the surface of the phagocyte with ligands on the surface of the particle. This leads to the polymerization of actin at the site of ingestion, and the internalization of the particle via an actin-based mechanism. After internalization actin is shed from the phagosome, and the phagosome matures by a series of fusion and fission events with components of the endocytic pathway, culminating in the formation of the mature phagolysosome. Since endosome-lysosome trafficking occurs primarily in association with microtubules, phagosome maturation requires the coordinated interaction of the actin and tubulin based cytoskeletons.

RECEPTORS: MECHANISMS OF RECOGNITION A primary challenge to the innate immune system is the discrimination of a large number of potential pathogens from self, utilizing a restricted number of phagocytic receptors. This problem is compounded by the propensity of pathogens to mutate. This challenge has been met by the evolution of a variety of receptors that recognize conserved motifs on pathogens that are not found in higher eukaryotes. These motifs have essential roles in the biology of the invading agents, and they are therefore not subject to high mutation rates. Janeway has proposed calling the receptors “pattern-recognition receptors” (PRRs) and the targets for these receptors “pathogen-associated molecular patterns” (PAMPs) (11). Pathogen-associated motifs include mannans in the yeast cell wall, formylated peptides in bacteria, and lipopolysaccharides and

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

596

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

lipoteichoic acids on the surface of Gram negative and Gram positive bacteria. The recognition mechanisms leading to phagocytosis occur either cellularly or humorally. Cellular receptors that recognize these patterns include the mannose receptor (MR) and DEC 205 that recognize mannans, as well as integrins (for example CD11b/CD18) and scavenger receptors that recognize surface components on bacteria including LPS (12, 13). Humoral components that opsonize the infectious agent before being recognized by a phagocytic receptor include the mannose-binding protein, which binds mannans and is recognized by the C1q receptor, and surfactant protein A, which binds carbohydrates and is recognized by a transmembrane receptor, SPR210 (14, 15). Antibodies represent an intersection between adaptive and innate immunity: They recognize their cognate ligands on infectious agents with exquisite specificity but are bound and internalized through their generic Fc domains by the Fc family of receptors (FcRs) (16–18). The complement system lies somewhere in between: The C3bi receptor binds to the C3bi fragment that is fixed nonspecifically to the carbohydrate surface of pathogens via the alternative pathway (19, 20). Alternatively, complement is fixed to IgM that specifically recognizes epitopes on the surface of the pathogen.

Fc Receptor-Mediated Phagocytosis Most of our understanding of the signaling pathways leading to phagocytosis in macrophages comes from studies of the FcR (16–18). FcRs fall into two general classes—those involved in effector functions and those that transport immunoglobulins across epithelial barriers. There are two major classes of Fcγ receptors: receptors that activate effector functions and receptors that inhibit these functions (16, 17, 21). FcRs that mediate phagocytosis in human macrophages fall within the activation class and include Fcγ RI, Fcγ RIIA, and Fcγ RIII (Figure 1) (17). The human Fcγ RIIA is a single chain protein with an extracellular Fc binding domain, a transmembrane domain, and a cytoplasmic tail containing two YXXL ITAM motifs (for immunoglobulin gene family tyrosine activation motif) similar to those found in T cell and B cell receptors (22, 23). There is no mouse counterpart to Fcγ RIIA. Murine macrophages, as well as human macrophages, express Fcγ RIIB, an inhibitory receptor that does not contain ITAM motifs and does not participate in phagocytosis (22, 24). Ligand binding results in receptor cross-linking, and this causes tyrosine phosphorylation of the ITAMs (see below). Fcγ RI and Fcγ RIIIA have extracellular Fc binding domains similar to the Fcγ RIIA, but lack ITAMs on their cytoplasmic tails (22). For proper expression and signaling, these receptors must interact with a dimer of γ subunits (Fcγ RI and Fcγ RIIIA), or ζ subunits (Fcγ RIIIA), small transmembrane proteins that contain the ITAMs needed for signal transduction (Figure 1) (25, 26). Ligation of Fcγ receptors I or III

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MACROPHAGE PHAGOCYTOSIS

597

Figure 1 Fcγ receptors signal phagocytosis via their phosphorylated ITAM domains. Receptor cross-linking stimulates src family kinases to phosphorylate tyrosine (Y) residues within the ITAM domain of Fcγ RIIA or within the dimerized γ subunits of Fcγ RI or Fcγ RIIIA. The tyrosine kinase syk is then recruited to the phosphorylated ITAM domain, and upon its activation, it is thought to mediate particle internalization by activating PI3-kinase and phospholipase C.

results in their cross-linking and in the tyrosine phosphorylation of the ITAM domains of their γ subunits (22). Deletion of the gene encoding the γ subunit of Fcγ receptor in mice results in macrophages that are unable to express Fcγ R I or III, since these receptors are not transported to the surface of cells in the absence of their signaling subunit, and macrophages from these mice are unable to phagocytize IgG-coated particles (26). The role of the ITAM motifs of the γ subunit of the Fcγ receptor have also been analyzed in COS cells, where all three members of the Fcγ R family

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

598

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

are capable of promoting phagocytosis (5). Since COS cells are not professional phagocyctes, and all cells have some capacity to phagocytose (as discussed above), it is likely that much more than the presence or absence of FcRs is responsible for efficient phagocytosis by professional phagocytes. Indeed, COS cells expressing FcRs phagocytose IgG-opsonized particles much less efficiently than macrophages. However, many of the early signaling events may be reconstituted in COS cells. In transfected COS-1 cells, Fcγ RIIIA or I mediates phagocytosis of IgG-opsonized particles, but only when coexpressed with the γ chain, and the ITAM motif of γ is required for a competent phagocytic signal (5). The ζ chain of the T cell receptor contains sequences homologous to the γ chain, including the conserved YXXL motifs, and can substitute for the γ chain in Fcγ RIIIA-dependent signaling of phagocytosis (27). However, the ζ chain is considerably less efficient in mediating Fcγ RIIIA-dependent phagocytosis than is the γ chain, and mutational analysis demonstrates that the functional differences between the γ and ζ subunits are due to the internal amino acids of the YXXL (27, 5). Cross-linking of Fcγ RIIIA results in tyrosine phosphorylation of the γ subunit, and mutation of either tyrosine of the two YXXL motifs of the γ subunit ITAM eliminates both tyrosine phosphorylation and phagocytosis (5, 27, 28). The protein tyrosine kinase responsible for this initial phosphorylation is thought to be a member of the src family (29, 30). Subsequently, a second protein tyrosine kinase, p72Syk, is recruited to the phosphorylated ITAM domains (22, 30–33). This results in the activation of the Syk kinase, which in turn triggers a plethora of pathways leading to transcriptional activation, cytoskeletal rearrangement, and the release of inflammatory mediators. This model is supported by the observation that a chimera containing the extracellular domain of CD16 (Fcγ RIII), fused to the transmembrane stalk of CD7 and containing p72syk intracellularly, is capable of signaling phagocytosis of IgG opsonized particles in transfected COS cells (34). A competent phagocytic stimulus was independent of Syk SH2 domains but required an active Syk kinase (34). The related tyrosine kinase ZAP70 could substitute for Syk in this system, whereas members of the src family of tyrosine kinases could not (34). These studies have been extended in DT40 lymphocytes, a chicken cell line that has been valuable in dissecting signaling pathways because it undergoes a high rate of homologous recombination and therefore permits gene deletions at high frequency. DT40 cells, expressing a fusion protein consisting of the extracellular domain of human Fcγ RIIIA and the ITAM-containing γ subunit of the Fc receptor, are capable of localized actin polymerization when the chimeric receptors are clustered (35). Actin assembly is dependent upon an intact ITAM, absent in cells lacking Syk and exacerbated in cells overexpressing Syk (35), suggesting an absolute requirement for the Syk tyrosine kinase

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MACROPHAGE PHAGOCYTOSIS

599

in ITAM-dependent actin assembly in DT40 cells. The requirement for Syk can probably be extended to all hematopoetic cells, since FcR-mediated actin assembly and phagocytosis is abrogated in macrophages derived from the fetal livers of Syk null mice (36). Further evidence for the involvement of Syk in phagocytosis is also derived from the COS cell system. Upon cross-linking of Fcγ RIIIA/γ and Fcγ RI/γ , Syk is phosphorylated and enhances the phagocytosis of IgG-opsonized erythrocytes; this activity is dependent on the γ chain (37, 5). Both SH2 domains of Syk are necessary for functional association with the γ subunit, and Syk is unable to induce either Fcγ RI or Fcγ RIIIA mediated phagocytosis by γ chain mutants in which YXXL tyrosine is replaced by phenylalanine (5, 37). How the Syk tyrosine kinase stimulates actin assembly is unknown, although it is likely that PI 3-kinase is involved (see below). There are clearly a number of problems with this model. First, it only applies to Fcγ R-mediated phagocytosis. Thus, while macrophages from sykmice are incapable of Fcγ R-mediated phagocytosis, phagocytosis of latex particles, yeast, and E. coli is unimpaired (36). Second, macrophages derived from mice deficient in the three members of the Src-family kinases known to be expressed in these cells, Hck, Fgr, and Lyn, exhibit poor Syk activation when the Fcγ R is ligated but are still capable of Fcγ R-mediated phagocytosis, albeit at a slightly slower rate (36). Either a small activation of Syk is sufficient to support Fcγ R-mediated phagocytosis, or Syk participates by another means, perhaps by serving as an adapter. An answer to this conundrum might be found in the interesting observation that the c-fgr tyrosine kinase actually suppresses phagocytosis in macrophages (H Gresham, C Willman, unpublished data). These investigators found that while a fgr-negative murine macrophage line phagocytosed normally, Fcγ RI, Fcγ RII/Fcγ RIII, and C3bimediated phagocytosis was suppressed when the cells were transfected with wild-type c-fgr. While actin rearrangement and phagocytosis is suppressed, c-fgr has no effect on receptor expression or on attachment of the opsonized particle. The suppressive effect of c-fgr is independent of its kinase activity, implying that inhibition of phagocytosis may be mediated through an adapter function. DOWNSTREAM EFFECTORS OF FC RECEPTOR-MEDIATED PHAGOCYTOSIS The mechanism by which Fcγ Rs stimulate the polymerization of actin and induce the formation of phagosomes is not known, although PI-3 kinase, the rho family of GTPases, protein kinase C (PKC), and motor proteins appear to participate.

PI-3 kinase Recent evidence suggests that PI 3-kinase participates in the signaling cascade of phagocytic receptors. PI 3-kinase catalyzes phosphorylation at the D-3 position of the inositol ring of phosphatidylinositol (PI), PI(4)P and PI(4,5)P2, and is activated by many tyrosine kinase receptors that trigger the

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

600

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

polymerization of actin (38). In addition, there is compelling evidence in yeast that PI 3-kinase participates in membrane trafficking (39). Cross-linking of Fcγ RI and RII increases PI 3-kinase activity, and FcR-mediated phagocytosis is prevented by wortmannin or LY294002, specific inhibitors of PI 3-kinase (40, 41). In addition, stimulation of Fcγ RIIA in platelets causes the association of the receptor with PI 3-kinase (42). Elegant studies by Swanson and colleagues indicate that wortmannin and LY294002 PI do not inhibit actindependent formation of the phagocytic cup, but instead prevent the phagosome from sealing behind the particle (41). De Franco and coworkers confirmed these data and further demonstrated that macrophages from syk null mice are similarly capable of polymerizing actin beneath the Fcγ R-induced phagocytic cup but are unable to complete internalization (36). This suggests that PI 3-kinase may participate in a syk-dependent signaling pathway critical for Fcγ R-mediated phagocytosis. GTPases Members of the Rho family of GTPases have been shown to regulate the actin cytoskeleton in response to a variety of extracellular signals (43). In 3T3 cells, various members of the Rho family act hierarchically during cell spreading: Cdc42 partipates in the formation of filopodia and in the activation of Rac; Rac stimulates membrane ruffling and activates Rho, and Rho stimulates the formation of focal adhesions and stress fibers (44). Recent evidence demonstrates that the Rho family also participates in phagocytosis. Microinjection of the J774 mouse macrophage cell line with the Rho-specific inhibitor C3 exotoxin inhibits Fcγ R-mediated phagocytosis by preventing receptor clustering, a prerequisite for efficient particle binding and internalization (45). By contrast, inhibition of Rac1 and Cdc42, by expression of their dominant negative forms in the RAW mouse macrophage cell line, does not affect particle binding to FcRs, but inhibits phagocytosis by preventing the accumulation of F-actin in the phagocytic cup (46). The precise mechanism by which these GTPases regulate F-actin structure has not yet been defined, but a variety of cytoskeletal regulators including PIP 5-kinase and myosin II have been implicated (43). Members of the ARF family of GTPases have a role in most membrane trafficking events. ARF6 has been implicated in endocytosis, membrane recycling and regulated exocytosis (47). Expression of a mutant form of ARF6 that is incapable of hydrolyzing GTP causes profound rearrangement of F-actin in HeLa cells (48), and inhibits Fcγ R-mediated phagocytosis in a macrophage cell line (49). Protein kinase Cts Protein kinase C (PKC) also appears to have a role in phagocytosis (10, 50, 51), previous studies demonstrated that PKC is activated upon ligation of the Fcγ R in human monocytes (50) and localizes the α isozyme of PKC to nascent phagosomes in macrophages (10). The involvement of PKC

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MACROPHAGE PHAGOCYTOSIS

601

in phagocytosis is tantalizing since its major substrate, MARCKS, is known to regulate actin structure at the membrane (52). MARCKS is rapidly phosphorylated during particle uptake, and MARCKS and PKCα are recruited to the forming zymosan phagosome with kinetics similar to those of F-actin (10). MARCKS cross-links F-actin, and this activity is prevented by PKC-dependent phosphorylation and by calcium/calmodulin (53). Since the association of MARCKS with membranes is also regulated by PKC-dependent phosphorylation, it is an ideal candidate to regulate actin structure on the phagosome in response to signals from both PKC and calcium/calmodulin. This is supported by the observation that inhibitors of PKC prevent phagocytosis and block the accumulation of PKCα, MARCKS, F-actin, and a number of other cytoskeletal proteins beneath bound zymosan (10). MacMARCKS, another member of the MARCKS family, also associates with zymosan phagosomes (54, 55), and a mutant form of MacMARCKS appears to block phagocytosis when expressed in a macrophage cell line (55). The significance of this observation is unclear since macrophages derived from MacMARCKS null mice phagocytose zymosan normally (56). Motor proteins It is not clear whether actin polymerization alone is sufficient to drive pseudopod extension and particle internalization, or whether this also requires molecular motors. It has long been known that myosin II accumulates on the phagocytic cups of macrophages and neutrophils ingesting yeast, implying that it might act as a mechanical motor during particle internalization (57). Myosin I, myosin V, and myosin IX also colocalize with F-actin on forming phagosomes, suggesting that it too might facilitate ingestion (10, 58). Despite these colocalization studies and the observation that the broad spectrum myosin inhibitor BDM prevents phagocytosis (58) (D Underhill, unpublished observations), there is as yet no information on the specific roles of myosin isoforms in phagocytosis.

Complement Receptor-Mediated Phagocytosis Complement proteins, present in serum, opsonize bacteria for phagocytosis by the C3b or C3bi receptors (CRs) on macrophages. Several receptors that participate in phagocytosis of complement-opsonized particles, including CR1, CR3, and CR4 are expressed on macrophages (19, 20). CR1 is a single chain transmembrane protein consisting of a large extracellular lectin-like complementbinding domain and a short 43 amino acid cytosolic domain. CR1 binds C3b, C4b, and C3bi, and is thought to participate mainly in particle binding (59). CR3 and CR4 are integrin family members made up of heterodimers of different α chains (αm for CR3 and αx for CR4) and a shared β chain (β2 ) (19). These two receptors bind specifically to C3bi and are responsible for particle internalization.

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

602

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

While FcRs are constitutively active for phagocytosis (22), the CRs of resident peritoneal macrophages bind but do not internalize particles in the absence of additional stimuli (7, 8). Particle ingestion by CRs can be induced by PKC activators such as PMA, as well as by TNF-α, granulocyte/macrophage colonystimulating factor (GM-CSF), or attachment to laminin- or fibronectin-coated substrata (7, 8, 60). Although all types of phagocytosis require actin polymerization at the site of ingestion (2), results of electron microscopy (EM) studies demonstrate that IgG- and complement-opsonized particles are internalized differently by macrophages (9, 61). During Fcγ R-mediated phagocytosis, veils of membrane rise above the cell surface and tightly surround the particle before drawing it into the body of the macrophage (9, 61) (Figure 2A). Silverstein and colleagues have demonstrated that Fcγ R-mediated ingestion occurs by a zippering process, in which Fcγ Rs in the macrophage plasma membrane interact sequentially with IgG molecules distributed over the surface of the ingested particle (1). On the other hand, EM data indicate that CR-mediated phagocytosis is a relatively passive process that occurs by a variation of the classic zipper model; complement-opsonized particles appear to sink into the cell with elaboration of small, if any, pseudopodia (9, 61) (Figure 2B). Moreover, the phagosome membrane is less tightly opposed to complement-opsonized particles, with point-like contact areas separating regions of looser membrane. These point-like contact areas are enriched with a variety of cytoskeletal proteins including F-actin, vinculin, α-actinin, paxillin, and phosphotyrosine-containing proteins, and their formation is blocked by inhibitors of PKC, but not by inhibitors of protein tyrosine kinases (although tyrosine phosphorylation increases the efficiency of phagocytosis) (9) (Figure 3E and F, and data not shown). By contrast, all of these proteins are diffusely distributed on phagosomes containing IgG-coated particles (Figure 3C and D), and Fcγ R-mediated phagocytosis is blocked by both PKC and tyrosine kinase inhibitors (9). Thus, the signals required for particle ingestion and the arrangement of cytoskeletal proteins on the phagosome surface vary depending upon which phagocytic receptor is engaged. Moreover, complement receptor (CR)-mediated internalization requires intact −−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→ Figure 2 Different particles are internalized by distinct phagocytic mechanisms. Cryo-EM sections of peritoneal macrophages that are in the process of ingesting IgG-opsonized particles (A) or complement-opsonized particles (B). Note that pseudopodia protrude from the macrophage surface to engulf the IgG-opsonized particle, whereas the complement-coated particle sinks directly into the cell. Arrows in B indicate vesicles directly beneath the forming complement phagosome that are absent beneath the FcR phagosome in a. [Reprinted with permission from The Journal of Experimental Medicine (LAH Allen, A Aderem. 1996. J. Exp. Med. 184:627–37).] (C ) L. pneumophila is internalized into human monocytes by coiling phagocytosis. [Reprinted with permission from Cell (MA Horwitz. 1984. Cell 36:27–33).]

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: JER/spd/mbg

January 21, 1999

P2: PSA/vks

15:45

QC: KKK

Annual Reviews AR078-19

MACROPHAGE PHAGOCYTOSIS

603

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

604

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

Figure 3 Vinculin and paxillin associate differently with different types of phagosomes. Zymosancontaining phagosomes (A and B) are not stained with antibodies to vinculin (A) or paxillin (B). The arrowheads indicate the position of the phagosomes. FcR-mediated phagosomes (C and D) are enriched with vinculin (C ) and paxillin (D) in a uniform pattern, while complement receptormediated phagosomes (E and F ) are coated with vinculin (E ) and paxillin (F ) in discrete foci. [Reprinted with permission from The Journal of Experimental Medicine (LAH Allen, A Aderem. 1996. J. Exp. Med. 184:627–37).]

microtubules and is accompanied by the accumulation of vesicles beneath the forming phagosome (Figure 2B, arrows), suggesting that membrane trafficking plays a key role in CR-mediated phagocytosis (9). An additional difference between FcR- and CR-mediated phagocytosis relates to their capacity to trigger the release of inflammatory mediators. FcRinduced phagocytosis is tightly coupled to the production and secretion of proinflammatory molecules such as reactive oxygen intermediates and arachidonic acid metabolites (62, 63). By contrast, CR-mediated phagocytosis does not elicit the release of either of these classes of inflammatory mediators (62, 63).

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

MACROPHAGE PHAGOCYTOSIS

605

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Mannose Receptor-Mediated Phagocytosis The mannose receptor (MR) on macrophages recognizes mannose and fucose on the surfaces of pathogens and mediates phagocytosis of the organisms (13). The high affinity of this receptor for branched mannose and fucose oligosaccharides, prototypic PAMPS as described above, makes the MR a phagocytic receptor with broad pathogen specificity. The MR is a single chain receptor with a short cytoplasmic tail and an extracellular domain including 8 lectin-like carbohydrate-binding domains. The lectin-like carbohydrate-binding domains share homology with other C-type lectins including the mannose-binding protein, collectins, DEC 205, and the phospholipase A2 receptor (64, 65). The cytoplasmic tail is crucial to both the endocytic and phagocytic functions of the receptor, but little is known about the signals that lead to phagocytosis (13, 64). During mannose receptor-mediated phagocytosis of zymosan, the actin cytoskeleton is mobilized around the nascent phagosomes, and proteins such as F-actin, talin, PKCα, MARCKS, and Myosin I are recruited (9). However, in contrast to FcR- and CR-mediated phagocytosis, vinculin and paxillin are not recruited to MR phagosomes (Figure 3A and B), reinforcing the notion that different phagocytic receptors send different signals to the actin cytoskeleton and initiate different mechanisms of internalization (9). In addition to the phagocytic signals mediating particle internalization, proinflammatory signals are generated upon MR ligation. Specifically, IL-1β, IL-6, GM-CSF (66), TNF-α (67, 68), and IL-12 (69) are all produced. Therefore, like phagocytosis mediated by the FcR, but not the complement receptor, MR-mediated phagocytosis is a pro-inflammatory process. However, at least one report suggests that release of additional proinflammatory cytokines including MIP-1β and MIP-2 is not mediated by MR signaling (66).

MATURATION OF THE PHAGOSOMAL VACUOLE Soon after internalization, F-actin is depolymerized from the phagosome, and the newly denuded vacuole membrane becomes accessible to early endosomes (4). Through a series of fusion and fission events, the vacuolar membrane and its contents mature, fusing with late endosomes and ultimately lysosomes, to form a phagolysosome. The rates of phagosome-lysosome fusion vary dramatically depending on the nature of the ingested particle; during FcR- and MR-mediated uptake the phagosome fuses with lysosomes within 30 min, while phagosomes containg latex particles might not fuse with lysosomes for hours (70–74). The rate at which phagosomes mature may be related to the nature of the interaction between the particle surface and the phagosomal membrane; for example, phagosomal

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

606

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

membranes are in close apposition to Mycobacteria, and those phagosomes do not fuse with lysosomes (72). It has been proposed that the nature of the particle surface modifies the rate of vacuole maturation and that the hydrophobic surface of a Mycobacterium inhibits recycling or maturation of phagosomal membranes, an event required for the fusion of phagosomes and membranes (72). An important and poorly understood aspect of phagosome maturation relates to the mechanism by which the lumenal contents of the phagosome are sorted. When lysosomes are preloaded with different sizes of fluorescent probes, the smaller probes report phagosome-lysosome fusion earlier than do the larger molecules (75). This size selective transfer was further investigated and led to the suggestion that transfer occurred via narrow, aqueous bridges that only permit limited content exchange (76, 77). Desjardins has suggested that transient and incomplete fusion between compartments may preserve the specific identity of each organelle, a mechanism he has called “kiss-and-run” (78). Formation of the phagolysosome is not a terminal event, and the structure has the capacity to continue fusing with lysosomes (73). Once again, the surface of the particle plays a critical role in deciding the fate of the phagosome: For example, while many IgG-coated particles remained accessible to exogenously added probes, polystyrene beads are removed from the circulation (73).

Proteins that Regulate Vacuole Maturation The annexins are a family of proteins that bind phospholipids in a calciumdependent manner and appear to function in membrane traffic and membranecytoskeleton interactions. While annexins I, II, III, and V associate with phagosomes all the time, annexin IV preferentially binds more mature vacuoles (79). The significance of the annexins in phagocytosis is not known. The small molecular mass GTPases rab5, rab7, and rap1 sequentially associate with phagosomes as they mature (74, 80). Since homotypic fusion between early endosomes is dependent on the presence of rab5 on both organelles (81), it is likely that this mechanism also facilitates the fusion of newly formed phagosomes displaying rab5 with early endosomes. This principle can be extended. Since rab7 plays a role similar to that of rab5 further down the endocytic pathway (82), it is likely that the loss of rab5 and acquisition of rab7 by the maturing phagosome permits the fusion of the organelle with rab7-enriched late endosomes. Studies with pathogens that subvert vacuole maturation support a role for rab proteins in the process. A role for rab5 in phagosome-endosome fusion has been implicated in studies using a hemolysin-deficient strain of Listeria monocytogenes. Within 30 min after internalization by a macrophage, Listeria normally lyses the phagosomal membrane and replicates in the cytosol (83). In an in vitro fusion assay, immunodepletion of rab5 blocked phagosome-

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MACROPHAGE PHAGOCYTOSIS

607

endosome fusion, and rab5 was found specifically to accumulate on Listeriacontaining vacuoles (84). A number of studies have demonstrated the presence on phagosomes of a variety of proteins associated with the SNARE mechanism of membrane fusion (85), including synaptobrevins and syntaxins, but although these proteins undoubtedly play a role in phagosome-endosome fusion, the data seem to suggest that the recognition mechanisms underlying phagosome maturation are directed by the rab proteins (77, 86). As the phagosome matures, both its membrane composition and contents are modified by its interaction with the endocytic compartment. This is well illustrated by the distribution of acid hydrolases in the cell; early endosomes contain the bulk of cathepsin H, and this enzyme is therefore acquired more rapidly by the early phagosome than are the other lysosomal enzymes (87). By contrast, late endosomes contained the bulk of cathepsin S, and this enzyme is therefore acquired later during phagosome maturation. As phagosomes mature they move into the cell on microtubules, and this trafficking gives them the opportunity to interact with various components of the endosomal system (88). Griffiths and colleagues (88) have reconstituted bidirectional phagosome movement along microtubules in vitro and have defined a number of properties of the system. Early phagosomes move more slowly than their more mature counterparts, and most phagosomes are minus-end directed. Immunodepletion studies clearly demonstrated a role for dynein and dynactin in minusend movement of phagosomes, while plus-end movement was mediated by kinesin and its membrane receptor kinectin (88). A similar role for kinesin has been demonstrated in the radial extension of tubular lysosomes in macrophages (89).

PHAGOCYTOSIS OF PATHOGENS Even though one of the major functions of phagocytosis is to mediate the ingestion and sterilization of infectious agents, many pathogens such as Salmonella typhimurium, Legionella pneumophila, and Mycobacterium tuberculosis have evolved mechanisms for survival and even growth inside macrophage vacuoles. Once again, phagocytosis of bacteria involves a large variety of heterogenous mechanisms.

Salmonella Typhimurium Internalization Occurs by Macropinocytosis Salmonella typhimurium is a facultative intracellular pathogen that is capable of replicating in intracellular compartments in macrophages (90). The macrophage receptors that bind S. typhimurium are not known, although internalization

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

608

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

appears to be associated with an actin-dependent mechanism. After binding to the surface of a macrophage, virulent S. typhimurium induces generalized membrane ruffling that results in the internalization of the bacterium into a compartment resembling a macropinosome (91). This nascent vacuole is enormous relative to the size of the bacterium and has been called a “spacious phagosome” (91). Nonvirulent mutant strains of S. typhimurium bind to the surface of macrophages and do not induce membrane ruffling or macropinocytosis (92); instead, these bacteria are phagocytosed into a vacuole with a tightly opposed membrane. The signaling mechanisms leading to membrane ruffling and the formation of the spacious phagosome in macrophages are unknown, but recent data demonstrate that upon binding to epithelial cells, S. typhimurium utilize a type III secretion system to inject the host cell with a protein, Sop E, that has GDP/GTP exchange activity specific for Rac and Cdc42 (93). SopE is required for efficient entry into epithelial cells (94), but it is not known if SopE-mediated activation of Rho family proteins is required for entry into macrophages. Once internalized, the spacious vacuole containing virulent bacteria persists, and the bacteria multiply (91). There are conflicting reports as to the maturation of the spacious vacuole; while some suggest that the spacious phagosome acquires lysosomal markers such as LAMP-1 and cathepsin L, others suggest that the spacious phagosome remains immature and does not fuse with lysosomes (95, 96). Since one group used bone marrow–derived macrophages, while the other used a macrophage cell line, the difference in vacuole fate may be attributable to differences in cell type.

Legionella pneumophila Is Internalized by Coiling Phagocytosis L. pneumophila is a facultative intracellular pathogen that invades and replicates in macrophages (97, 98). A bacterial surface protein, MOMP (major outermembrane protein), fixes complement component C3 to the surface of the parasite, thereby facilitating binding to the macrophage surface through complement receptors (99). Removing complement from the media, or blocking CR3, with a specific antibody prevents bacterial adhesion to the macrophage surface (99). After binding, the parasite induces the formation of an extended host cell pseudopod that spirals around the bacterium forming a structure termed a “coiling phagosome” (Figure 2C ) (100). Although CR3 is enriched in coiling phagosomes, there is no evidence that these receptors signal coiling phagocytosis. Since the structure of this phagosome is very different than that induced by CR3 (discussed above), it is likely that an additional, as yet uncharacterized, L. pneumophila signals must be required. Once internalized, the outer portions of the macrophage membrane coil disintegrate, leaving the bacterium in a phagosome with a single, tightly opposed

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

MACROPHAGE PHAGOCYTOSIS

609

membrane (100, 101). This phagosome has unique properties; the vacuole does not acidify below pH 6.1 and does not fuse with endosomes or lysosomes (102, 103). Instead, the phagosome initially fuses with smooth vesicles, ultimately maturing into a ribosome-studded vacuole composed of endoplasmic reticulum-derived membranes (104, 105). Finally, the bacterium divides within this vacuole and ruptures the host.

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Mycobacterium tuberculosis Vacuole A plethora of macrophage receptors have been implicated in binding and internalization of M. tuberculosis (106). As with other bacteria, complement fixes to the surface of the organism through the alternate pathway, allowing deposition of complement proteins C3b and C3bi, which are recognized by CR1 and CR3 (107). In the absence of factors required for activation of the alternate pathway, a surface component of M. tuberculosis resembling complement component C4b can bind directly to C2b and form a C3 convertase analogous to the one formed in the classical complement cascade (108). This C3 convertase catalyzes the deposition of C3b onto the surface of the organism and facilitates binding to CR1. Blocking CR drastically reduces binding and invasion of M. tuberculosis but does not abolish it, suggesting that other receptors participate in their uptake. Consistent with this, M. tuberculosis has also been demonstrated to bind to the mannose receptor and the scavenger receptor (109). In addition, surfactant protein A enhances macrophage binding and uptake of M. tuberculosis, probably by the surfactant protein A receptors (110). Phagocytosis of either Erdman or H37Ra M. tuberculosis in the presence of autologous nonimmune serum is associated with an increase in phospholipase D activity in human monocyte–derived macrophages, and inhibition of phospholipase D prevents the uptake of the bacterium (111). M. tuberculosis uptake is also associated with the tyrosine phosphorylation of multiple macrophage proteins, and tyrosine kinase inhibitors suppress the phagocytosis of the bacterium (111). Once internalized, M. tuberculosis resides in a membrane-bound vacuole that resists lysosomal fusion (112) and is only mildly acidified (113). Phagosomes containing M. avium also fail to acidify below pH 6.5, and this appears to be due to the specific exclusion of the vesicular proton-ATPase (113). The mycobacterial vacuole is not completely sequestered, however, and it is capable of interacting selectively with early endosomes (114, 115). Transferrin cycles in and out of the mycobacterial vacuole, demonstrating that the phagosome is part of a dynamic system (116, 117). Macrophages activated with interferon-γ and bacterial lipopolysaccharide (LPS) prior to ingestion of mycobacteria are able to acidify the phagosomes containing the bacteria to pH 5.3 (118). Analysis of these vacuoles demonstrate that they accumulate the proton-ATPase and

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

610

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

are no longer accessible to transferrin, suggesting a substantially more mature phagosome. Electron microscopy demonstrates that macrophage activation is accompanied by a coalescence of vacuoles containing single bacteria, into large vacuoles containing many mycobateria, and this is accompanied by a substantial decrease in bacterial viability (118). Kinetic measurements indicate that acidification of the vacuole precedes the drop in microbial viability.

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

PHAGOCYTOSIS OF APOPTOTIC CELLS Apoptosis, or programmed cell death, is a process crucial in the development and homeostasis of all multicellular organisms (119). Even though an enormous number of cells are continuously undergoing apoptosis in tissues of higher organisms, these dying cells are rarely observed in vivo due to their efficient engulfment and degradation by phagocytic cells such as macrophages (119). Since phagocytosis of apoptotic cells is a normal, ongoing process, the mechanisms enabling macrophages to recognize, bind, internalize, and degrade apoptotic cells need to function without activating the proinflammatory responses of the macrophage, as happens during phagocytosis through many of the other phagocytic receptors (119–121). Indeed, human macrophages ingesting apoptotic neutrophils fail to secrete the chemoattractants IL-8 and MCP-1 normally associated with macrophage activation, although they do secrete transforming growth factor β1, prostaglandin E2 and platelet-activating factor, compounds that dampen inflammatory responses (122). On the other hand, mouse peritoneal macrophages ingesting apoptotic T cells have been observed to secrete the proinflammatory chemokine MIP-2, suggesting that some degree of macrophage activation may occur under certain circumstances (123). In order to phagocytose an apoptotic cell, receptors on the macrophage must see a ligand found on apoptotic cells that is not present on healthy cells. Ligands fitting these criteria that have been implicated in the recognition of apoptotic cells include phosphatidylserine in the outer leaflet of the plasma membrane, changes in the pattern of glycosylation of cell surface proteins, and surface charge (119, 121). Recent studies make it clear that there are many receptors (both defined and undefined) that participate in this process, and that understanding how these receptors work in concert will be a challenging task. Known receptors demonstrated to participate in phagocytosis of apoptotic cells by macrophages include class A scavenger receptors (124), a class B scavenger receptor, CD36, which acts in conjunction with the vitronectin receptor (125, 126), and CD14 (127) (see Figure 4).

Class A Scavenger Receptors Scavenger receptors (SRs) are a family of structurally diverse receptors having broad ligand specificity that includes LDL, phosphatidylserine, and polyanionic

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MACROPHAGE PHAGOCYTOSIS

611

Figure 4 Phagocytosis of an apoptotic cell by a macrophage. Recognition of the apoptotic cell is mediated by a variety of receptors including lectins, CD14, scavenger receptor A (SR-A), and CD36 in conjunction with the vitronectin receptor (αv β3 ). Ligands on the apoptotic cell that are recognized by these receptors include sugars, phosphatidylserine (PS), and surface-bound thrombospondin (TSP). The implicated molecules and signals are defined in the text.

compounds (128). Based on primary sequence information, scavenger receptors have been divided into at least six groups, and receptors from several of these groups are expressed on macrophages. Scavenger receptors SR-AI and SR-AII are alternatively spliced products of the same gene. These receptors are homotrimeric glycoproteins with a short (∼ =50 amino acids) amino terminal cytosplasmic domain and a carboxy terminal extracellular ligand-binding domain, including a characteristic collagenous coiled region. SR-A binds acetylated and oxidized low-density lipoprotein, and polyanionic compounds such as maleylated bovine serum albumin and polyinosinic acid (129). Direct evidence for involvement of SR-As in macrophage phagoctosis of apoptotic cells comes from studies on the phagocytosis of apoptotic thymocytes by thymus-derived macrophages in vitro (124). Internalization of the thymocytes could be substantially inhibited with a monoclonal antibody to SR-A or by polyanionic ligands (124). In support of these

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

612

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

data, phagocytosis of thymocytes was inhibited 50% in thymic macrophages derived from SR-A null mice. Significantly, the relative number of apoptotic thymocytes in the SR-A null animals was not substantially increased, indicating that other receptors are sufficient for normal apoptotic cell clearance in the thymus (121).

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Class B Scavenger Receptors and the Vitronectin Receptor A class B scavenger receptor, CD36, has also been implicated in phagocytosis of apoptotic cells (119). CD36 is an 88-kDa membrane glycoprotein expressed on a variety of cells including platelets, monocytes, endothelial cells, and erythroblasts (130). CD36 has been identified as one of the receptors for collagen type I, thrombospondin, oxidized LDL, and phosphatidylserine (131). The role of CD36 in the internalization of apoptotic cells has best been illustrated by the observation that a monoclonal antibody to CD36 substantially inhibits the phagocytosis of aged human neutrophils by human blood-derived macrophages (132). Interestingly, murine CD36 does not mediate the uptake of apoptotic cells, and amino acids 155-183 of the human receptor were able to confer apoptotic cell binding on the mouse molecule (133). Since CD36 has a very short (4 amino acids) carboxy terminal cytoplasmic domain, it seems likely that it interacts with other transmembrane signaling molecules to stimulate phagocytosis. Thrombospondin has been suggested to bridge apoptotic cells, CD36, and the vitronectin receptor, resulting in a phagocytically active ternary complex (125, 126). Thus, the vitronectin receptor (αv β3 ) may stimulate the polymerization of actin that results in CD36-dependent internalization of apoptotic cells. If this model is correct, thrombospondin is acting as an opsonin of apoptotic cells, although the mechanism by which it interacts with apoptotic cells is obscure. This cross-talk between CD36 and the vitronectin receptor illustrates the extraordinary complexity underlying phagocytosis: Here two receptors that have independent functions unrelated to phagocytosis cooperate to orchestrate a phagocytic event. Conflicting evidence exists for the importance of CD36 in phagocytizing apoptotic cells in vivo. On the one hand, blood monocytes from SLE patients demonstrate a decrease in CD36 levels paralleled by a deficiency in the phagocytosis of apoptotic cells. On the other hand, monocyte-derived macrophages from CD36-deficient patients show no defect in the phagocytosis of apoptotic neutrophils (121). This result undoubtedly reflects the considerable redundancy that underlies the uptake of apoptotic cells.

CD14 In addition to the above receptors, CD14 has also been implicated in recognition and internalization of apoptotic cells. A monoclonal antibody that specifically inhibited internalization of PMNs by monocyte-derived macrophages

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MACROPHAGE PHAGOCYTOSIS

613

was determined to recognize CD14, a molecule also known to transduce LPS signals (127). Expression of CD14 in COS cells was sufficient to mediate phagocytosis of apoptotic lymphocytes, although the uptake was relatively inefficient. Interestingly, although apoptotic cells bind to CD14 at a site close to the LPS-binding site, they do not elicit biological responses, such as TNF-α production, that are induced in macrophages by LPS (127). CD14 is a GPI-linked membrane protein and therefore must interact with other proteins to mediate signaling. Proinflammatory signaling stimulated by LPS has recently been demonstrated to be mediated by Toll-like receptor 2 (TLR2), a transmembrane receptor with homology to the IL-2 receptor (134). As for CD14, LPS signaling through TLR2 requires the serum component, LPS binding protein (LBP), and CD14 augments LPS response through TLR2 (134). This suggests that TLR2 is the transmembrane protein that associates with the LPS/LBP/CD14 ternary complex to mediate the pro-inflammatory response in macrophages. It is possible, although not yet proven, that CD14 does not interact with TLR2 during the binding of apoptotic cells, perhaps because of the absence of LBP, and that this contributes to the lack of inflammatory response during internalization. Although macrophages express many receptors capable of recognizing determinants on apoptotic cells, it is interesting that these receptors often have functions not involving phagocytosis. In addition to redundancy, this might imply that a hierarchy of recognition mechanisms for apoptotic cells exists. Such a hierarchy might be predicated on the environment in which the events are occurring. In particular, the activation state of the macrophage might dictate the receptor system used, and this, in turn, might influence whether phagocytosis of the apoptotic cells is accompanied by the elaboration of anti-inflammatory or pro-inflammatory mediators. Thus, phagocytosis of apoptotic neutrophils by human macrophages actively inhibits LPS-induced production of a spectrum of pro-inflammatory mediators including IL-1β, IL-8, IL-10, GM-CSF, TNF-α, leukotriene C4, and thromboxane B2 (122). The inhibition of production of inflammatory cytokines appears to be mediated by PGE2 and TGF-β1. The receptor systems used to bind and internalize apoptotic cells differ depending on the activation-state of the macrophage. The uptake of apoptotic lymphocytes by activated macrophages can be inhibited by PS and N-acetylglucosamine, suggesting the involvement of lectin-like receptors (135), while the integrin-binding peptide, RGDS, and cationic amino acids and sugars have little blocking activity. The opposite appears true for unactivated macrophages; phagocytosis of apoptotic lymphocytes by unactivated macrophages can be blocked by RGDS and cationic amino acids and sugars, while PS and N-acetylglucosamine have little effect (135). This has been proposed to suggest a role for the vitronectin receptor in phagocytosis of apoptotic cells by unactivated macrophages but not by activated macrophages. In both cases,

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

614

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

an antibody to CD14 inhibits uptake (127). Thus, activated and nonactivated macrophages phagocytize the same apoptotic cells via different mechanisms. In addition, the same macrophage will internalize different apoptotic cells by different mechanisms. This is supported by the observation that ligation of CD44 augments apoptotic neutrophil uptake but not uptake of apoptotic lymphocytes (136). Conversely, the uptake of apoptotic lymphocytes by peritoneal macrophages can be blocked by phosphatidyl serine, while the uptake of apoptotic neutrophils is unaffected by this phospholipid (137).

C. elegans as a Genetic System The complex interactions and downstream signals resulting in the uptake of apoptotic cells may best be deciphered using genetic systems applicable to phagocytosis. A variety of C. elegans mutants defective in apoptotic cell clearance have been described, and while this phagocytosis is performed by neighboring cells in the worm, much of what has been learned will likely be useful in dissecting out the pathway in macrophages. Mutations in ced-1, ced-2, ced5, ced-6, ced-7, and ced-10 result in a large increase in persistent apoptotic bodies in C. elegans. Genetic analysis suggests that the six engulfment genes fall into two groups: ced-1, ced-6, and ced-7 are in one group, while ced-2, ced-5, and ced-10 are in the other (138). Single and double mutants within the same group show relatively weak defects, while double mutants between the two groups show severe phagocytic defects (138). Thus, the two groups of genes might be involved in two distinct, but partially redundant pathways in phagocytosis. Ced-5 encodes a protein with homology to the mammalian protein DOCK180 and the Drosophila protein myoblast city (MBC) (139). These proteins, collectively known as the CDM family, appear to regulate cytoskeletal-membrane interactions and, in particular, membrane extension. DOCK180 interacts with the adaptor protein crk and has been implicated in intergrin-mediated signaling and cell movement (140). Expression of human DOCK180 in C. elegans rescues the cell migration defects, but does not restore the uptake of apoptotic cells, sugesting that the proteins are at least partially interchangeable (139). Drosophila MBC is necessary for myoblast fusion and for the migration of a population of epithelial cells (141). Ced-7 has protein sequence homology to ABC transporters, proteins that translocate a wide variety of substrates across membranes (142). Interestingly, ABC1, a member of the ABC transporters, has been identified in macrophages, and the ability of macrophages to ingest apoptotic thymocytes, but not yeast cells, is impaired when the macrophages are treated with antibodies to ABC1 (143). Ced-7 acts in both the target cell and the engulfing cell, and it has been suggested to be important for the interaction between the cells (142).

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

MACROPHAGE PHAGOCYTOSIS

615

Overexpression of Ced-6 can partially restore phagocytosis in ced-1 or ced-7 deficient cells, suggesting that it acts downstream of these two genes (144). Ced-6 appears to be an adaptor molecule since it contains a phosphotyrosinebinding (PTB) domain at its N-terminus and a proline-rich domain capable of interacting with an SH3 domain at its C-terminus (144).

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Implications for Immunity Bhardwaj and coworkers recently reported that dendritic cells, but not macrophages, are capable of stimulating class I–restricted CD8+ cytotoxic T lymphocytes, by efficiently presenting antigen derived from phagocytized apoptotic cells (145). This pathway might account for the in vivo phenomenon of crosspriming, whereby antigens derived from tumor cells or transplants are presented by host antigen presenting cells (146, 147). In addition, this pathway could facilitate the presentation of self-antigens, resulting in the breaking of tolerance and the activation of autoimmune disease (148). The very different outcomes of ingestion of apoptotic cells by macrophages or dendritic cells once again highlight the heterogeneous nature of phagocytosis and suggest that understanding the differences in phagocytic mechanisms between these cell types could lead to new therapeutic strategies for a number of inflammatory diseases.

CONCLUSIONS It is clear that phagocytosis in macrophages is a diverse process; the signals leading to actin polymerization and particle internalization depend on the specific receptors that mediate the process and on additional modifying signals that can be generated by complex particles. Complex particles, such as bacteria, can activate multiple receptors whose signaling pathways may interact in intricate and unpredictable ways. In addition, living bacteria have the capacity to modify signaling pathways within eukaryotic cells. For example, S. typhimurium can introduce a GDP:GTP exchange factor into cells that modifies the way that Rho family of GTPases signals the actin cytoskeleton (93), while Yersinia species introduce a broad spectrum tyrosine phosphatase (90). In addition to the complexity relating to the different receptor systems and the capacity of microbes to modify phagocytosis, it is important to recognize that the formation of the phagosome can be a heterogenous process, even when a single cell ingests two identical particles. While all phagocytosis involves actin remodeling around the phagocytic cup, we have observed that certain cytoskeletal proteins that decorate a particular phagosome are absent from otherwise identical phagosomes in the same cell. These differences are not temporal but are likely to arise from stochastic, or even chaotic, processes. For example, actin cross-linking can be achieved using any of the diverse number

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

616

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

of actin cross-linking proteins that are expressed in the cell. Thus, if a specific actin cross-linking protein happens to have been enriched in a specific region of the cell, for example the leading edge of a motile cell, it may be used during phagocytosis at the leading edge while another protein may serve the same function on a phagosome formed at the trailing edge. It is clear that the molecular dissection of phagocytosis represents a daunting task; more than 100 actin-binding proteins have been identified, and many of these are expressed in the same cell. Two common approaches that have been used to establish the role of specific gene products in a particular biological phenomenon have been to express dominant negative forms of the protein, or to delete the gene encoding the protein. However, the interpretation of the results is complicated by the observation that these two approaches do not always yield the same phenotype. Genetically tractable organisms that allow the dissection of phagocytosis will contribute greatly to our ability to make sense of this complexity. C. elegans has already extended our understanding of the phagocytosis of apoptotic cells, even though this organism does not have macrophages. Dictyostelium discoideum is a free living, haploid organism that phagocytizes bacteria and that has been used to great effect in dissecting cell motility. Gerish and coworkers have recently turned their attention to phagocytosis in D. discoideum, and they have almost immediately elucidated new pathways and molecules that are reiterated in mammalian systems (149–151). Ultimately, phagocytosis, like most other problems in biology, will have to be analyzed as a complex system rather than a linear series of isolated enzymatic reactions. In this guise, phagocytosis provides a window into the coordinate functioning of the actin and tubulin based cytoskeletons, and it could serve as a model system for analyzing diverse biological phenomena including synaptic transmission, mitogenesis, and morphogenesis. ACKNOWLEDGMENTS We thank Drs. Joel Swanson, Alan Ezekowitz, Jeffrey Ravetch, Marcus Horwitz, Dan Clemens, Steven Greenberg, Sergio Grinstein, Gareth Griffiths, Michel Desjardins, Michel Rabinovitch, Keith Joiner, David Russell, Eric Brown, David Sibley, Hattie Gresham, and Cheryl Willman for communicating results and manuscripts prior to publication. This work was supported by grants AI25032 and AI32972 from the National Institutes of Health. DU is an Irvington Institute Postdoctoral fellow. This review is dedicated to the memory of Dr. Zanvil Cohn, who laid the foundation for most of the studies reported here. Visit the Annual Reviews home page at http://www.AnnualReviews.org

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

MACROPHAGE PHAGOCYTOSIS

617

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Literature Cited 1. Silverstein SC. 1995. Phagocytosis of microbes: insights and prospects. Trends Cell Biol. 5:141–42 2. Allen L-AH, Aderem A. 1996. Mechanisms of phagocytosis. Curr. Opin. Immunol. 8:36–40 3. Rabinovitch M. 1995. Profesional and non-professional phagocytes: an introduction. Trends Cell Biol. 5:85–87 4. Swanson JA, Baer SC. 1995. Phagocytosis by zippers and triggers. Trends Cell Biol. 5:89–93 5. Indik ZK, Park JG, Hunter S, Schreiber AD. 1995. The molecular dissection of Fc gamma receptor mediated phagocytosis. Blood 86:4389–99 6. Metchnikoff E. 1905. Immunity in Infective Diseases. Cambridge, London, Glasgow: Cambridge Univ. Press 7. Pommier CG, Inada S, Fries LF, Takahashi T, Frank MM, Brown EJ. 1983. Plasma fibronectin enhances phagocytosis of opsonized particles by human peripheral blood monocytes. J. Exp. Med. 157:1844–54 8. Wright SD, Craigmyle LS, Silverstein SC. 1983. Fibronectin and serum amyloid P component stimulate C3b- and C3bimediated phagocytosis in cultured human monocytes. J. Exp. Med. 158:1338–43 9. Allen LAH, Aderem A. 1996. Molecular definition of distinct cytoskeletal structures involved in complement- and Fc receptor-mediated phagocytosis in macrophages. J. Exp. Med. 184:627–37 10. Allen LH, Aderem A. 1995. A role for MARCKS, the alpha isozyme of protein kinase C and myosin I in zymosan phagocytosis by macrophages. J. Exp. Med. 182:829–40 11. Janeway CAJ. 1992. The immune system evolved to discriminate infectious nonself from noninfectious self. Immunol. Today 13:11–16 12. Sastry K, Ezekowitz RA. 1993. Collectins: pattern recognition molecules involved in first line host defense [published erratum appears in Curr. Opin. Immunol. 1993 Aug; 5(4):566]. Curr. Opin. Immunol. 5:59–66 13. Stahl PD, Ezekowitz RA. 1998. The mannose receptor is a pattern recognition receptor involved in host defense. Curr. Opin. Immunol. 10:50–55 14. Epstein J, Eichbaum Q, Sheriff S, Ezekowitz RA. 1996. The collectins in innate immunity. Curr. Opin. Immunol. 8: 29–35

15. Tenner AJ, Robinson SL, Ezekowitz RA. 1995. Mannose binding protein (MBP) enhances mononuclear phagocyte function via a receptor that contains the 126,000 M(r) component of the C1q receptor. Immunity 3:485–93 16. Ravetch JV, Clynes RA. 1998. Divergent roles for Fc receptors and complement in vivo. Annu. Rev. Immunol. 16:421–32 17. Ravetch JV. 1997. Fc receptors. Curr. Opin. Immunol. 9:121–25 18. Unkeless JC, Shen Z, Lin CW, DeBeus E. 1995. Function of human Fc gamma RIIA and Fc gamma RIIIB. Semin. Immunol. 7:37–44 19. Sengelov H. 1995. Complement receptors in neutrophils. Crit. Rev. Immunol. 15:107–31 20. Carroll MC. 1998. The role of complement and complement receptors in induction and regulation of immunity. Annu. Rev. Immunol. 16:545–68 21. Unkeless JC, Jin J. 1997. Inhibitory receptors, ITIM sequences and phosphatases. Curr. Opin. Immunol. 9:338–43 22. Ravetch JV. 1994. Fc receptors: rubor redux. Cell 78:553–60 23. Reth M. 1989. Antigen receptor tail clue. Nature 338:383–84 24. Odin JA, Edberg JC, Painter CJ, Kimberly RP, Unkeless JC. 1991. Regulation of phagocytosis and [Ca2+]i flux by distinct regions of an Fc receptor. Science 254:1785–88 25. Anderson P, Caligiuri M, O’Brien C, Manley T, Ritz J, Schlossman SF. 1990. Fc gamma receptor type III (CD16) is included in the zeta NK receptor complex expressed by human natural killer cells. Proc. Natl. Acad. Sci. USA 87:2274– 78 26. Takai T, Li M, Sylvestre D, Clynes R, Ravetch JV. 1994. FcR gamma chain deletion results in pleiotrophic effector cell defects. Cell 76:519–29 27. Park JG, Murray RK, Chien P, Darby C, Schreiber AD. 1993. Conserved cytoplasmic tyrosine residues of the gamma subunit are required for a phagocytic signal mediated by Fc gamma RIIIA. J. Clin. Invest. 92:2073–79 28. Mitchell MA, Huang MM, Chien P, Indik ZK, Pan XQ, Schreiber AD. 1994. Substitutions and deletions in the cytoplasmic domain of the phagocytic receptor Fc gamma RIIA: effect on receptor tyrosine phosphorylation and phagocytosis. Blood 84:1753–59

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

618

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

29. Ghazizadeh S, Bolen JB, Fleit HB. 1994. Physical and functional association of Src-related protein tyrosine kinases with Fc gamma RII in monocytic THP-1 cells. J. Biol. Chem. 269:8878–84 30. Greenberg S. 1995. Signal transduction of phagocytosis. Trends Cell Biol. 5:93– 99 31. Agarwal A, Salem P, Robbins KC. 1993. Involvement of p72syk, a protein-tyrosine kinase, in Fc gamma receptor signaling. J. Biol. Chem. 268:15900–5 32. Kiener PA, Rankin BM, Burkhardt AL, Schieven GL, Gilliland LK, Rowley RB, Bolen JB, Ledbetter JAl. 1993. Crosslinking of Fc gamma receptor I (Fc gamma RI) and receptor II (Fc gamma RII) on monocytic cells activates a signal transduction pathway common to both Fc receptors that involves the stimulation of p72 Syk protein tyrosine kinase. J. Biol. Chem. 268:24442–48 33. Ghazizadeh S, Bolen JB, Fleit HB. 1995. Tyrosine phosphorylation and association of Syk with Fc gamma RII in monocytic THP–1 cells. Biochem. J. 74:669–674 34. Greenberg S, Chang P, Wang D, Xavier R, Seed B. 1995. Clustered syk tyrosine kinase domains trigger phagocytosis. Proc. Natl. Acad. Sci. USA 93:1103–7 35. Cox D, Chang P, Kurosaki T, Greenberg S. 1996. Syk tyrosine kinase is required for immunoreceptor tyrosine activation motif–dependent actin assembly. J. Biol. Chem. 271:16597–602 36. Crowley MT, Costello PS, Fitzer-Attas CJ, Turner M, Meng F, Lowell C, Tybulewicz VL, DeFranco AL. 1997. A critical role for Syk in signal transduction and phagocytosis mediated by Fcgamma receptors on macrophages. J. Exp. Med. 186:1027–39 37. Indik ZK, Park JG, Pan XQ, Schreiber AD. 1995. Induction of phagocytosis by a protein tyrosine kinase. Blood 85:1175– 80 38. Toker A, Cantley LC. 1997. Signalling through the lipid products of phosphoinositide-3-OH kinase. Nature 387:673–76 39. De Camilli P, Emr SD, McPherson PS, Novick P. 1996. Phosphoinositides as regulators in membrane traffic. Science 271:1533–39 40. Ninomiya N, Hazeki K, Fukui Y, Seya T, Okada T, Hazeki O, Ui M. 1994. Involvement of phosphatidyl inositol 3-kinase in Fcgamma receptor signaling. J. Biol. Chem. 269:22732–37 41. Araki N, Johnson MT, Swanson JA. 1996. A role for phosphoinositide 3-kinase in the completion of macropinocytosis and

42.

43. 44.

45.

46.

47.

48.

49.

50.

51.

52. 53.

phagocytosis by macrophages. J. Cell Biol. 135:1249–60 Chacko GW, Brandt JT, Coggeshall KM, Anderson CL. 1996. Phosphoinositide 3-kinase and p72syk noncovalently associate with the low affinity Fc gamma receptor on human platelets through an immunoreceptor tyrosine-based activation motif. Reconstitution with synthetic phosphopeptides. J. Biol. Chem. 271: 10775–81 Hall A. 1998. Rho GTPases and the actin cytoskeleton. Science 279:509–14 Ridley AJ, Paterson HF, Johnston CL, Diekmann D, Hall A. 1992. The small GTP-binding protein rac regulates growth factor–induced membrane ruffling. Cell 70:401–10 Hackam DJ, Rotstein OD, Schreiber A, Zhang W, Grinstein S. 1997. Rho is required for the initiation of calcium signaling and phagocytosis by Fcgamma receptors in macrophages. J. Exp. Med. 186:955–66 Cox D, Chang P, Zhang Q, Reddy PG, Bokoch GM, Greenberg S. 1997. Requirements for both Rac1 and Cdc42 in membrane ruffling and phagocytosis in leukocytes. J. Exp. Med. 186:1487–94 Roth MG, Sternweis PC. 1997. The role of lipid signaling in constitutive membrane traffic. Curr. Opin. Cell Biol. 9:519– 26 Radhakrishna H, Klausner RD, Donaldson JG. 1996. Aluminum fluoride stimulates surface protrusions in cells overexpressing the ARF6 GTPase. J. Cell Biol. 134:935–47 Zhang Q, Cox D, Tseng C-C, Donaldson JG, Greenberg S. 1998. A requirement for ARF6 in Fcγ receptor-mediated phagocytosis in macrophages. J. Biol. Chem. 273:19977–81 Zheleznyak A, Brown EJ. 1992. Immunoglobulin-mediated phagocytosis by human monocytes requires protein kinase C activation. J. Biol. Chem. 267:12042–48 Allen LA, Aderem A. 1995. Protein kinase C regulates MARCKS cycling between the plasma membrane and lysosomes in fibroblasts. EMBO J. 14:1109– 20 Aderem A. 1992. The MARCKS brothers: a family of protein kinase C substrates. Cell 71:713–16 Hartwig JH, Thelen M, Rosen A, Janmey PA, Nairn AC, Aderem A. 1992. MARCKS is an actin filament crosslinking protein regulated by protein kinase C Rand calcium-calmodulin. Nature 356:618–22

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MACROPHAGE PHAGOCYTOSIS 54. Li J, Aderem A. 1992. MacMARCKS, a novel member of the MARCKS family of protein kinase C substrates. Cell 70:791– 801 55. Zhu Z, Bao Z, Li J. 1995. MacMARCKS mutation blocks macrophage phagocytosis of zymosan. J. Biol. Chem. 270: 17652–55 56. Underhill DM, Chen J, Allen L-AH, Aderem A. 1998. MacMARCKS is not essential for phagocytosis in macrophages. J. Biol. Chem. 273:33,619–23 57. Stendahl OI, Hartwig JH, Brotschi EA, Stossel TP. 1980. Distribution of actin-binding protein and myosin in macrophages during spreading and phagocytosis. J. Cell Biol. 84:215–24 58. Swanson JA, Johnson MT, Beningo K, Post P, Mooseker M, Araki N. 1998. A contractile activity that closes phagosomes in macrophages. J. Cell Sci. In press 59. Brown EJ. 1991. Complement receptors and phagocytosis. Curr. Opin. Immunol. 3:76–82 60. Wright SD, Griffin FM Jr. 1985. Activation of phagocytic cells’ C3 receptors for phagocytosis. J. Leuk. Biol. 38:327–39 61. Kaplan G. 1977. Differences in the mode of phagocytosis with Fc and C3 receptors in macrophages. Scand. J. Immunol. 6:797–807 62. Aderem AA, Wright SD, Silverstein SC, Cohn ZA. 1985. Ligated complement receptors do not activate the arachidonic acid cascade in resident peritoneal macrophages. J. Exp. Med. 161:617–22 63. Wright SD, Silverstein SC. 1983. Receptors for C3b and C3bi promote phagocytosis but not the release of toxic oxygen from human phagocytes. J. Exp. Med. 158:2016–23 64. 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 65. Taylor ME, Conary JT, Lennartz MR, Stahl PD, Drickamer K. 1990. Primary structure of the mannose receptor contains multiple motifs resembling carbohydrate-recognition domains. J. Biol. Chem. 265:12156–62 66. Yamamoto Y, Klein TW, Friedman H. 1997. Involvement of mannose receptor in cytokine interleukin-1beta (IL-1beta), IL-6, granulocyte-macrophage colonystimulating factor responses, but not in chemokine macrophage inflammatory

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

619

protein 1beta (MIP-1beta), MIP-2, KC responses, caused by attachment of Candida albicans to macrophages. Infect. Immun. 65:1077–82 Stein M, Gordon S. 1991. Regulation of tumor necrosis factor (TNF) release by murine peritoneal macrophages: role of cell stimulation and specific phagocytic plasma membrane receptors. Eur. J. Immunol. 21:431–37 Garner RE, Rubanowice K, Sawyer RT, Hudson JA. 1994. Secretion of TNF-alpha by alveolar macrophages in response to Candida albicans mannan. J. Leuk. Biol. 55:161–68 Shibata Y, Metzger WJ, Myrvik QN. 1997. Chitin particle-induced cell-mediated immunity is inhibited by soluble mannan: Mannose receptor-mediated phagocytosis initiates IL-12 production. J. Immunol. 159:2462–67 Pitt A, Mayorga LS, Stahl PD, Schwartz AL. 1992. Alterations in the protein composition of maturing phagosomes. J. Clin. Invest. 90:1978–83 Racoosin EL, Swanson JA. 1993. Macropinosome maturation and fusion with tubular lysosomes in macrophages. J. Cell Biol. 121:1011–20 de Chastellier C, Thilo L. 1997. Phagosome maturation and fusion with lysosomes in relation to surface property and size of the phagocytic particle. Eur. J. Cell Biol. 74:49–62 Oh YK, Swanson JA. 1996. Different fates of phagocytosed particles after delivery into macrophage lysosomes. J. Cell Biol. 132:585–93 Desjardins M, Huber LA, Parton RG, Griffiths G. 1994. Biogenesis of phagolysosomes proceeds through a sequential series of interactions with the endocytic apparatus. J. Cell Biol. 124:677–88 Wang YL, Goren MB. 1987. Differential and sequential delivery of fluorescent lysosomal probes into phagosomes in mouse peritoneal macrophages. J. Cell Biol. 104:1749–54 Berthiaume EP, Medina C, Swanson JA. 1995. Molecular size-fractionation during endocytosis in macrophages. J. Cell Biol. 129:989–98 Desjardins M, Nzala NN, Corsini R, Rondeau C. 1997. Maturation of phagosomes is accompanied by changes in their fusion properties and size-selective acquisition of solute materials from endosomes. J. Cell Sci. 110:2303–14 Desjardins M. 1995. Biogenesis of phagolysosomes: the “kiss and run” hypothesis. Trends Cell Biol. 5:183–86

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

620

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL

79. Diakonova M, Gerke V, Ernst J, Liautard JP, van-der VG, Griffiths G. 1997. Localization of five annexins in J774 macrophages and on isolated phagosomes. J. Cell Sci. 110:1199–1213 80. Pizon V, Desjardins M, Bucci C, Parton RG, Zerial M. 1994. Association of Rap1a and Rap1b proteins with late endocytic/phagocytic compartments and Rap2a with the Golgi complex. J. Cell Sci. 107:1661–70 81. Gorvel J-P, Chavrier P, Zerial M, Gruenberg J. 1991. Rab5 controls early endosome fusion in vitro. Cell 64:915–25 82. Feng Y, Press B, Wandinger-Ness A. 1995. Rab 7: an important regulator of late endocytic membrane traffic. J. Cell Biol. 131:1435–52 83. Tilney LG, Portnoy DA. 1989. Actin filaments and the growth, movement, spread of the intracellular bacterial parasite, Listeria monocytogenes. J. Cell Biol. 109:1597–1608 84. Alvarez DC, Barbieri AM, Ber’on W, Wandinger NA, Stahl PD. 1996. Phagocytosed live Listeria monocytogenes influences Rab5-regulated in vitro phagosome-endosome fusion. J. Biol. Chem. 271:13834–43 85. Hay JC, Scheller RH. 1997. SNAREs and NSF in targeted membrane fusion. Curr. Opin. Cell Biol. 9:505–12 86. Hackam DJ, Rotstein OD, Bennett MK, Klip A, Grinstein S, Manolson MF. 1996. Characterization and subcellular localization of target membrane soluble NSF attachment protein receptors (t-SNAREs) in macrophages. Syntaxins 2, 3, 4 are present on phagosomal membranes. J. Immunol. 156:4377–83 87. Claus V, Jahraus A, Tjelle T, Berg T, Kirschke H, Faulstich H, Griffiths G. 1998. Lysosomal enzyme trafficking between phagosomes, endosomes, lysosomes in J774 macrophages. Enrichment of cathepsin H in early endosomes. J. Biol. Chem. 273:9842–51 88. Blocker A, Severin FF, Burkhardt JK, Bingham JB, Yu H, Olivo JC, Schroer TA, Hyman AA, Griffiths G. 1997. Molecular requirements for bi-directional movement of phagosomes along microtubules. J. Cell Biol. 137:113–29 89. Hollenbeck PJ, Swanson JA. 1990. Radial extension of macrophage tubular lysosomes supported by kinesin. Nature 346:864–66 90. Finlay BB, Cossart P. 1997. Exploitation of mammalian host cell functions by bacterial pathogens. Science 276:718–25 91. Alpuche-Aranda CM, Raccoosin EL,

92.

93.

94.

95.

96.

97.

98.

99.

100.

101.

Swanson JA, Miller SI. 1994. Salmonella stimulate macrophage macropinocytosis and persist within spacious phagosomes. J. Exp. Med. 179:601–8 Alpuche-Aranda CM, Berthiaume EP, Mock B, Swanson JA, Miller SI. 1995. Spacious phagosome formation within mouse macrophages correlates with Salmonella serotype pathogenicity and host susceptibility. Infect. Immun. 63: 4456–62 Hardt WD, Chen LM, Schuebel KE, Bustelo XR, Galan JE. 1998. S. typhimurium encodes an activator of Rho GTPases that induces membrane ruffling and nuclear responses in host cells. Cell 93:815–26 Wood MW, Rosqvist R, Mullan PB, Edwards MH, Galyov EE. 1996. SopE, a secreted protein of Salmonella dublin, is translocated into the target eukaryotic cell via a sip-dependent mechanism and promotes bacterial entry. Mol. Microbiol. 22:327–38 Oh YK, Alpuche AC, Berthiaume E, Jinks T, Miller SI, Swanson JA. 1996. Rapid and complete fusion of macrophage lysosomes with phagosomes containing Salmonella typhimurium. Infect. Immun. 64:3877–83 Rathman M, Barker LP, Falkow S. 1997. The unique trafficking pattern of Salmonella typhimurium–containing phagosomes in murine macrophages is independent of the mechanism of bacterial entry. Infect. Immun. 65:1475–85 Shuman HA, Horwitz MA. 1996. Legionella pneumophila invasion of mononuclear phagocytes. Curr. Top. Microbiol. Immunol. 112:99–112 Vogel JP, Andrews HL, Wong SK, Isberg RR. 1998. Conjugative transfer by the virulence system of Legionella pneumophila. Science 279:873–76 Bellinger KC, Horwitz MA. 1990. Complement component C3 fixes selectively to the major outer membrane protein (MOMP) of Legionella pneumophila and mediates phagocytosis of liposomeMOMP complexes by human monocytes. J. Exp. Med. 172:1201–10 Horwitz MA. 1984. Phagocytosis of the Legionnaires’ disease bacterium (Legionella pneumophila) occurs by a novel mechanism: engulfment within a pseudopod coil. Cell 36:27–33 Marra A, Horwitz MA, Shuman HA. 1990. The HL-60 model for the interaction of human macrophages with the Legionnaires’ disease bacterium. J. Immunol. 144:2738–44

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MACROPHAGE PHAGOCYTOSIS 102. Horwitz MA. 1983. The Legionnaires’ disease bacterium (Legionella pneumophila) inhibits phagosome-lysosome fusion in human monocytes. J. Exp. Med. 158:2108–26 103. Horwitz MA, Maxfield FR. 1984. Legionella pneumophila inhibits acidification of its phagosome in human monocytes. J. Cell Biol. 99:1936–43 104. Horwitz MA, Silverstein SC. 1980. Legionnaires’ disease bacterium (Legionella pneumophila) multiplies intracellularly in human monocytes. J. Clin. Invest. 66: 441–50 105. Swanson MS, Isberg RR. 1995. Association of Legionella pneumophila with the macrophage endoplasmic reticulum. Infect. Immun. 63:3609–20 106. Ernst JD. 1998. Macrophage receptors for Mycobacterium tuberculosis. Infect. Immun. 66:1277–81 107. Schlesinger LS, Bellinger-Kawahara CG, Payne NR, Horwitz MA. 1990. Phagocytosis of Mycobacterium tuberculosis is mediated by human monocyte complement receptors and complement component C3. J. Immunol. 144:2771–80 108. Schorey JS, Carroll MC, Brown EJ. 1997. A macrophage invasion mechanism of pathogenic mycobacteria. Science 277:1091–93 109. Schlesinger LS. 1993. Macrophage phagocytosis of virulent but not attenuated strains of Mycobacterium tuberculosis is mediated by mannose receptors in addition to complement receptors. J. Immunol. 150:2920–30 110. Zimmerli S, Edwards S, Ernst JD. 1996. Selective receptor blockade during phagocytosis does not alter the survival and growth of Mycobacterium tuberculosis in human macrophages. Am. J. Respir. Cell Mol. Biol. 15:760–70 111. Kusner DJ, Hall CF, Schlesinger LS. 1996. Activation of phospholipase D is tightly coupled to the phagocytosis of Mycobacterium tuberculosis or opsonized zymosan by human macrophages. J. Exp. Med. 184:585–95 112. Hart PD, Armstrong JA, Brown CA, Draper P. 1972. Ultrastructural study of the behavior of macrophages toward parasitic mycobacteria. Infect. Immun. 5:803– 7 113. Sturgill-Koszycki S, Schlesinger PH, Chakraborty P, Haddix PL, Collins PL, Fok AK, Allen RD, Gluck SL, Heuser J, Russell DG. 1994. Lack of acidification in Mycobacterium phagosomes produced by exclusion of the vesicular proton-ATPase. Science 263:678–81

621

114. Clemens DL, Horwitz MA. 1995. Characterization of the Mycobacterium tuberculosis phagosome and evidence that phagosomal maturation is inhibited. J. Exp. Med. 181:257–70 115. Russell DG, Dant J, Sturgill Koszycki S. 1996. Mycobacterium avium– and Mycobacterium tuberculosis–containing vacuoles are dynamic, fusion-competent vesicles that are accessible to glycosphingolipids from the host cell plasmalemma. J. Immunol. 156:4764–73 116. Clemens DL, Horwitz MA. 1996. The Mycobacterium tuberculosis phagosome interacts with early endosomes and is accessible to exogenously administered transferrin. J. Exp. Med. 184:1349–55 117. Sturgill-Koszycki S, Schaible UE, Russell DG. 1996. Mycobacterium-containing phagosomes are accessible to early endosomes and reflect a transitional state in normal phagosome biogenesis. EMBO J. 15:6960–68 118. Schaible UE, Sturgill KS, Schlesinger PH, Russell DG. 1998. Cytokine activation leads to acidification and increases maturation of Mycobacterium avium–containing phagosomes in murine macrophages. J. Immunol. 160:1290–96 119. Savill J. 1997. Recognition and phagocytosis of cells undergoing apoptosis. Br. Med. Bull. 53:491–508 120. Savill J. 1998. Apoptosis. Phagocytic docking without shocking. Nature 392: 442–43 121. Platt N, da-Silva RP, Gordon S. 1998. Recognizing death: the phagocytosis of apoptotic cells. Trends Cell Biol. 8:365– 72 122. 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, PAF. J. Clin. Invest. 101:890–98 123. Uchimura E, Kodaira T, Kurosaka K, Yang D, Watanabe N, Kobayashi Y. 1997. Interaction of phagocytes with apoptotic cells leads to production of proinflammatory cytokines. Biochem. Biophys. Res. Commun. 239:799–803 124. Platt N, Suzuki H, Kurihara Y, Kodama T, Gordon S. 1996. Role for the class A macrophage scavenger receptor in the phagocytosis of apoptotic thymocytes in vitro. Proc. Natl. Acad. Sci. USA 93:12456–60 125. Savill J, Hogg N, Ren Y, Haslett C. 1992. Thrombospondin cooperates with

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

622

126.

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

127.

128.

129. 130.

131.

132.

133.

134.

135.

136.

137.

15:45

QC: KKK

Annual Reviews

AR078-19

ADEREM & UNDERHILL CD36 and the vitronectin receptor in macrophage recognition of neutrophils undergoing apoptosis. J. Clin. Invest. 90: 1513–22 Savill J, Dransfield I, Hogg N, Haslett C. 1990. Vitronectin receptor-mediated phagocytosis of cells undergoing apoptosis. Nature 343:170–73 Devitt A, Moffatt OD, Raykundalia C, Capra JD, Simmons DL, Gregory CD. 1998. Human CD14 mediates recognition and phagocytosis of apoptotic cells. Nature 392:505–9 Krieger M, Herz J. 1994. Structures and functions of multiligand lipoprotein receptors: macrophage scavenger receptors and LDL receptor-related protein (LRP). Annu. Rev. Biochem. 63:601–37 Pearson AM. 1996. Scavenger receptors in innate immunity. Curr. Opin. Immunol. 8:20–28 Greenwalt DE, Lipsky RH, Ockenhouse CF, Ikeda H, Tandon NN, Jamieson GA. 1992. Membrane glycoprotein CD36: a review of its roles in adherence, signal transduction, transfusion medicine. Blood 80:1105–15 Rigotti A, Acton SL, Krieger M. 1995. The class B scavenger receptors SRBI and CD36 are receptors for anionic phospholipids. J. Biol. Chem. 270:16221– 24 Savill J, Fadok V, Henson P, Haslett C. 1993. Phagocyte recognition of cells undergoing apoptosis. Immunol. Today 14:131–36 Navazo MD, Daviet L, Savill J, Ren Y, Leung LL, McGregor JL. 1996. Identification of a domain (155-183) on CD36 implicated in the phagocytosis of apoptotic neutrophils. J. Biol. Chem. 271:15381–85 Yang R-B, Mark MR, Gray A, Huang A, Xie MH, Zhang M, Goddard A, Wood WI, Gurney AL, Godowski PJ. 1998. Toll-like receptor-2 mediates lipopolysacharide-induced cellular signalling. Nature 395:284–88 Pradhan D, Krahling S, Williamson P, Schlegel RA. 1997. Multiple systems for recognition of apoptotic lymphocytes by macrophages. Mol. Biol. Cell 8:767– 78 Hart SP, Dougherty GJ, Haslett C, Dransfield I. 1997. CD44 regulates phagocytosis of apoptotic neutrophil granulocytes, but not apoptotic lymphocytes, by human macrophages. J. Immunol. 159:919– 25 Fadok VA, Voelker DR, Campbell PA, Cohen JJ, Bratton DL, Henson PM. 1992.

138.

139.

140.

141.

142.

143.

144.

145.

146.

147.

148.

149.

Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J. Immunol. 148:2207– 16 Ellis RE, Jacobson DM, Horvitz HR. 1991. Genes required for the engulfment of cell corpses during programmed cell death in Caenorhabditis elegans. Genetics 129:79–94 Wu YC, Horvitz HR. 1998. C. elegans phagocytosis and cell-migration protein CED-5 is similar to human DOCK180. Nature 392:501–4 Hasegawa H, Kiyokawa E, Tanaka S, Nagashima K, Gotoh N, Shibuya M, Kurata T, Matsuda M. 1996. DOCK180, a major CRK-binding protein, alters cell morphology upon translocation to the cell membrane. Mol. Cell Biol. 16:1770–76 Erickson MR, Galletta BJ, Abmayr SM. 1997. Drosophila myoblast city encodes a conserved protein that is essential for myoblast fusion, dorsal closure, cytoskeletal organization. J. Cell Biol. 138:589–603 Wu YC, Horvitz HR. 1998. The C. elegans cell corpse engulfment gene ced-7 encodes a protein similar to ABC transporters. Cell 93:951–60 Luciani MF, Chimini G. 1996. The ATP binding cassette transporter ABC1, is required for the engulfment of corpses generated by apoptotic cell death. EMBO J. 15:226–35 Liu QA, Hengartner MO. 1998. Candidate adaptor protein CED-6 promotes the engulfment of apoptotic cells in C. elegans. Cell 93:961–72 Albert ML, Sauter B, Bhardwaj N. 1998. Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs. Nature 392:86–89 Bevan MJ. 1977. Priming for a cytotoxic response to minor histocompatibility antigens: antigen specificity and failure to demonstrate a carrier effect. J. Immunol. 118:1370–74 Huang AY, Golumbek P, Ahmadzadeh M, 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 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 Peracino B, Borleis J, Jin T, Westphal M, Schwartz JM, Wu L, Bracco E, Gerisch G, Devreotes P, Bozzaro S. 1998. G protein

P1: JER/spd/mbg

P2: PSA/vks

January 21, 1999

15:45

QC: KKK

Annual Reviews

AR078-19

MACROPHAGE PHAGOCYTOSIS

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

beta subunit-null mutants are impaired in phagocytosis and chemotaxis due to inappropriate regulation of the actin cytoskeleton. J. Cell Biol. 141:1529–37 150. Niewohner J, Weber I, Maniak M, Muller TA, Gerisch G. 1997. Talin-null cells of Dictyostelium are strongly defective in adhesion to particle and substrate surfaces

623

and slightly impaired in cytokinesis. J. Cell Biol. 138:349–61 151. Maniak M, Rauchenberger R, Albrecht R, Murphy J, Gerisch G. 1995. Coronin involved in phagocytosis: dynamics of particle-induced relocalization visualized by a green fluorescent protein tag. Cell 83:915–24

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:593-623. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:625–56 c 1999 by Annual Reviews. All rights reserved Copyright °

POPULATION BIOLOGY OF HIV-1 INFECTION: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues A. T. Haase Department of Microbiology, University of Minnesota, Minneapolis, Minnesota 55455; e-mail: [email protected] KEY WORDS:

in situ hybridization and in situ PCR, quantitative image analysis, follicular dendritic cells, virus production, treatment, immune depletion and restoration

ABSTRACT Human immunodeficiency virus-1 (HIV-1) is usually transmitted through sexual contact and in the very early stages of infection establishes a persistent infection in lymphatic tissues (LT). Virus is produced and stored at this site in a dynamic process that slowly depletes the immune system of CD4+ T cells, setting the stage for AIDS. In this review, I describe the changes in viral and CD4+ T cell populations in LT over the course of infection and after treatment. I present recent evidence that productively infected CD4+ T cells play an important role in establishing persistent infection from the onset, and that the LT are the major reservoir where virus is produced and stored on follicular dendritic cells (FDCs). I discuss the methods used to define the size of viral and CD4+ T cell populations in LT and the nature of virus-host cell interactions in vivo. These experimental approaches have identified populations of latently and chronically infected cells in which virus can elude host defenses, perpetuate infection, and escape eradication by highly active antiretroviral treatment (HAART). I discuss the dramatic impact of HAART on suppressing virus production, reducing the pool of stored virus, and restoring CD4+ T cell populations. I discuss the contributions of thymopoiesis and other renewal mechanisms, lymphatic homeostasis and trafficking to these changes in CD4+ T cell populations in LT, and conclude with a model of immune depletion and repopulation based on the limited regenerative capacity of the adult and the uncompensated losses of productively infected cells that treatment stems. The prediction of this model is that immune regeneration will be slow, variable,

625 0732-0582/99/0410-0625$08.00

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

626

QC: PSA

Annual Reviews

AR078-20

HAASE

and partial. It is nonetheless encouraging to know that even in late stages of infection, control of active replication of HIV-1 provides an opportunity for the immune system to recover from the injuries inflicted by infection.

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTRODUCTION AND OVERVIEW HIV-1 usually gains entry to its human host by crossing mucosal surfaces (1) and is subsequently disseminated throughout the lymphatic tissues. These then become the reservoir where virus is produced and stored throughout the long course of infection (2–11). In the first few weeks of infection, virus and viral antigens appear and rise often to quite high levels in the blood stream, and a majority of infected individuals have a brief illness that resembles infectious mononucleosis (12–17). The primary stage of HIV-1 infection generally resolves, virus and viral antigens fall to lower levels coincident with the cellular immune response (18), and, in most infected individuals, a relatively asymptomatic, clinically latent stage ensues that lasts for several years. Over the course of infection, the CD4+ T cell count in blood slowly declines, and at the AIDS-defining level of 200 cells/mm3, opportunistic tumors and infections (OIs) supervene that eventually claim the lives of most infected individuals (19). The rate of progression to AIDS and death is correlated with the levels of viral RNA in the blood stream (20), which in turn reflects virus production in LT (11) by a dynamic chain mechanism of viral replication, spread and infection of new cells (21, 22). This natural history of infection and grim outcome have been dramatically altered recently by treatment with combinations of antiretroviral drugs that break the chain of the cycles of de novo infections required to maintain virus production. Under optimal circumstances in previously untreated individuals, highly active antiretroviral therapy (HAART) can suppress replication of HIV-1 to undetectable levels in the bloodstream and LT reservoir (23). Although longlived latently infected cells are not eradicated by HAART (24–26), control of active HIV-1 infection translates into increases in CD4+ T cells in blood and LT (27, 28), partial restoration of immune function (25, 27, 29, 30), and a decrease in OIs and deaths due to AIDS (31). The natural history of infection and encouraging results with HAART are the major topics of this review. The focus is largely on LT and CD4+ T cells because these cells play such an important role in the immune system and in AIDS and because LT is the principal site where HIV-1 is produced and persists and is the daily residence of all but a small fraction of the CD4+ T cell population. The review was written almost exclusively with the intention of emphasizing in vivo analyses; and from a demographic perspective, with the rationale that

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

POPULATION BIOLOGY OF HIV-1 INFECTION

627

studies of the size, growth, distribution, migration, and birth and death rates of viral and CD4+ T cell populations will continue to have significant impact on understanding the pathogenesis of HIV-1 infection.

INFECTED CELL DEMOGRAPHICS AND DYNAMICS AND VIRUS-CELL INTERACTIONS IN LT Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Infection of CD4+ T Cells at the Earliest Stages of Infection When do CD4+ T cells first become infected? In the current prevailing view, HIV-1 is thought to cross mucosal barriers by infecting dendritic cells (DCs) or macrophages (Mφs) (32–34). These cells then convey infection to LT where at some point CD4+ T cells become infected (Figure 1). In reality, the types of cells in which virus initially replicates and the timing of productive infection of CD4+ T cells are controversial issues. The evidence favoring productive infection of DCs and Mφs at a mucosal portal of entry is based on detecting replication of the HIV relative, simian immunodeficiency virus (SIV), in Mφs or DCs at later rather than the initial stages of infection (32). In studies of the initial stages of sexual mucosal transmission of SIV to rhesus macaque monkeys, an animal model that closely approximates HIV-1 infection where the initial stages of infection can be analyzed in detail, the conclusion that DCs in the submucosa are infected after vaginal inoculation rests on ambiguous morphological criteria for identification of infected cells (33, 34). Moreover, putative infected DCs were shown to harbor only viral DNA. In our studies of sexual mucosal transmission of SIV we found productively infected Mφs, DCs, and CD4+ T cells at the portal of entry, but the majority (∼≥80%) were CD4+ T cells. In Figure 2A (see color figure insert) the arrows point to a productively infected intraepithelial lymphocyte and a T cell in the lamina propria in the endocervix of a monkey three days after inoculation intravaginally with SIV. The brown immunohistochemical stain with antibodies to CD3 unequivocally identifies the cells as T cells. SIV RNA, indicated by the collections of black silver grains overlying the two cells, was detected in the same section by in situ hybridization with an 35S labeled SIV specific riboprobe and radioautography. The majority of the productively infected cells in the second week of infection in the draining and distant lymph nodes (LNs) and other LT were also CD4+ T cells (ZQ Zhang, manuscript in preparation). Similarly, Veazey et al (35) found that in primary SIV infection, virus replicates predominantly in CD4+ T cells in gut-associated LT (GALT). In primary HIV-1 infection we also recently found that most (>90%) productively infected cells in LT biopsies were CD4+ T cells (TS Schacker, manuscript submitted). In Figure 2B, the arrows point to three brown-stained T cells with

P2: PSA/SPD

March 3, 1999 10:10

628 Annual Reviews

Figure 1 The depicted virus-cell interactions at the portal of entry and in LT relevant to transmission (A), virus production, storage, and persistence (B) are described in the text. (Figure adapted in part and reproduced with permission from IM Roitt, J Brostoff, DK Male. 1989. Immunology, London: Gower Medical Publishing, p. 2.12.)

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: APR/ARY QC: PSA

AR078-20

HAASE

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SAT/VKS P2: NBL/plb

February 27, 1999 16:35

QC: NBL/tkj T1: NBL

Annual Reviews AR078-01

P1: SAT/VKS

P2: NBL/plb

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

February 27, 1999

16:35

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Figure 2 Arrows point to T cells with SIV RNA at the portal of entry (A) in sexual mucosal transmission and to T cells with HIV-1 RNA, (B) in LT in the early stages of HIV-1 infection. C-F distinguish virus production in viral DNA+ cells with spliced mRNAs (C,E) outside the GC from virus storage by FDCs in the GC indicated by the diffuse ISH signal from probes to unspliced RNA bound to virion RNA (D,F). G,H illustrate the well-localized propagation of two distinguishable clones of SIV (color coded) in subadjacent sections of spleen. Arrows point to areas where propagation of one clone of virus but not the other was detected. [Modified and reproduced with permission from J Embretson et al, 1993. Nature 363:559–62 (C,D), TA Reinhart et al, 1997, Nature Med. 3:218–21 (E,F), TA Reinhart et al 1998. J. Virol. 72:113–20 (G,H)]

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

POPULATION BIOLOGY OF HIV-1 INFECTION

629

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

collections of black silver grains that show that the cells contain HIV RNA. Immunohistochemical staining has again been combined with in situ hybridization with an 35S-labeled HIV specific riboprobe and radioautography to unambiguously demonstrate productive infection of T cells in sections of a lymph node from an individual in the primary stage of HIV-1 infection. Thus, in both SIV and HIV-1 good reasons exist to think that productive infection of CD4+ T cells from the very beginning plays a pivotal role in sustaining infection and driving the pathological processes that result in immune depletion.

The LT Reservoir and the Host Cell Range of Productive Infection In the subsequent course of infection, from the clinically latent stage to AIDS, the LT are a major reservoir of virus and infected cells in HIV-1 infection, and the major reservoir for SIV. For the latter, it has been possible to compare exhaustively and systematically the frequency of productively infected mononuclear cells (CD4+ T cells and Mφs) and viral burden, defined operationally as the number of copies of viral RNA in LT and in other organ systems such as the central nervous system (CNS), lung, and liver that might be additional reservoirs and sanctuaries (36). After intravaginal or intravenous inoculation of SIV, virus and infected cells are quickly disseminated throughout the LT (35–39; ZQ Zhang, manuscript in preparation). An explosive increase of several hundred fold of productively infected CD4+ T cells in GALT (35) and in lymph nodes and spleen occurs between the first and second week of infection (ZQ Zhang, manuscript in preparation). We found no evidence of infection in the CNS or other organ systems in the acute stage of infection. At later stages, the LT is still the major reservoir, with a direct correlation between the size of the population of productively infected cells and AIDS (36). In the animals that developed AIDS we occasionally found productively infected cells in nearly every organ system including the liver, kidney, adrenal and other endocrine organs, and many productively infected cells in the lung and the CNS. At the latter two sites, however, nearly all of the productively infected cells are Mφs, or cells in the Mφ lineage, e.g. microglial cells, the resident Mφs in the CNS. Similarly in HIV-1 infection, from the early to late stages of infection, the majority of productively infected cells in LT are CD4+ T cells (TS Schacker, manuscript submitted). In the preterminal stages of infection where CD4+ T cells have been severely depleted, Mφs may become important sources of HIV-1 (40), and throughout the course of infection HIV-1 replicates in syncytia of DCs and perhaps CD4+ T cells in the lymphoepithelia of tonsils and adenoids (41, 42). The frequency of infected DCs, however, in spleen (43) and other LT (11) is 10 to 100 times less than infection of CD4+ T cells. HIV-1 DNA has also been detected in CD8+ T

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

630

QC: PSA

Annual Reviews

AR078-20

HAASE

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

cells in the circulation in the late stages of infection (44). The infected CD8+ T cells may be the progeny of CD4+/CD8+ T cells infected in the thymus (45–48), or activated CD8+ T cells that are infected when they upregulate expression of CD4 (49, 50). It is not known at present what contribution these other host cells make to virus production. What we do know at this point is that productive infection of CD4+ T cells predominates throughout much of the course of infection and that the LT reservoir develops quickly and remains the major site where virus is produced and stored.

Virus-Cell Relationships in LT The exploded view (Figure 1) of the LT reservoir in established infection depicts virus host cell interactions and outcomes that are important in the ongoing production, storage, and persistence of HIV-1. Virus production can be attributed for the most part throughout infection to a relatively small population of infected CD4+ T cells with lesser contributions from infected Mφs and chronically infected cells. These populations are shown in the paracortical region, but productively infected cells are also frequently found in the germinal centers (11, 51). In the exploded view (Figure 1) FDCs are depicted as a storage site for HIV-1, serving their roles as a site where antigen can be stored and processed to induce and maintain immunologic memory and production of specific antibodies (52–56). Virus coated with complement, or immune complexes with antibody and complement, are bound to the membrane of the FDC by interaction with the C3 complement receptor (57–59). Although budding viral particles apparently associated with FDCs have been described (60), they are rare and most recent analyses support the conclusion that HIV-1 does not replicate in FDCs (9, 61, 62). In Figure 2C all of the cells in a LN section that harbor a single copy of the 10 kbp HIV-1 provirus have been identified by in situ PCR (9). Most of the cells that contain HIV-1 DNA are CD4+ T cells in the follicular mantle of the germinal complex (GC) and in the paracortex. After histochemical staining of the CD4+ T cells, HIV-1 DNA in cells in the section shown in Figure 2C was amplified with HIV-1-specific primers and detected by in situ hybrdization with a radiolabeled HIV-1-specific probe and radioautography. When the section is illuminated with epipolarized light, specular reflection from silver grains over the nuclei of cells that contain HIV-1 DNA imparts a greenish tint to the nuclei. Far fewer cells with HIV-1 are in the GC, and most of these cells are identified immunohistochemically as CD4+ T cells, not FDCs. By contrast, most of the viral RNA detectable in a subjacent section is in virions diffusely distributed over the network of FDC processes within the GC (Figure 2D). As further evidence that FDCs are not productively infected, we showed (62) that FDCs lack the spliced viral RNA found in productively infected cells. Probes

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

POPULATION BIOLOGY OF HIV-1 INFECTION

631

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

that specifically detect spliced SIV transcripts only hybridize to productively infected cells outside the GC (Figure 2E), whereas probes that detect all classes of spliced and unspliced SIV RNA identify viral RNA in virions bound to the RDC network inside the GC as well as productively infected cells outside the GC (Figure 2F). Although the FDCs do not produce virus by these criteria, they have the remarkable ability to restore infectivity to HIV complexed with neutralizing antibody (63). The pool of virus stored on FDCs is thus another way that virus might thwart immune defenses, and another hurdle for eradication, because HAART will not affect virus produced before treatment, was begun.

Further Definition of Virus Host Relationships and a Census of Infected Cells by Single Cell Amplification Techniques The development of in situ amplification techniques with sufficient sensitivity to detect a single copy of a lentivirus genome (64) has made it possible to count all infected cells irrespective of the level of viral gene expression. By combining in situ amplification and immunohistochemical staining with antibodies to cell type–specific markers, the proportion of infected cells in a particular population such as CD4+ T cells can be estimated (9). From the recent estimates of the total size of the population of CD4+ T cells in HIV-1 infection (28), we can derive estimates of the number of infected CD4+ T cells in LT. Of the 1011 CD4+ T cells in LT in the clinically latent to late stages of infection, about 20% of the subset recognized by OPD4 antibody harbor one copy of at least close-to-full-length HIV-1 DNA (since the region of the genome that was amplified and detected was at the 50 end of the provirus) (9). This subset constitutes about half of the population so the population of infected CD4+ T cells is ∼1010. Only 1:100 to 1:400 of the CD4+ T cells in this infected population were found to harbor transcriptionally active genomes (9). The steadystate population size of productively infected cells is therefore between about 2.5 × 107 to ∼108. This estimate is in excellent agreement with the numbers of productively infected cells determined by quantitative image analysis (11). The population of latently infected CD4+ T cells with replication competent but transcriptionally silent viral genomes is even smaller, currently estimated to be about 5 × 106 cells (65). These are the long-lived resting memory CD4+ T cells (66) that represent a major hurdle for eradication of HIV-1 with current therapies (24–26). Since the sum of the latently and productively infected cells with replication-competent genomes is 2 × 108, most of the infected CD4+ T cell population must harbor defective or incomplete proviruses, a conclusion in accord with most (21, 66, 67), but not all (68), studies. The CD4+ T cells with defective proviruses might be a viral genetic graveyard and a dead end pathway for infection, but if the cells produce and present sufficient antigen

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

632

QC: PSA

Annual Reviews

AR078-20

HAASE

for detection and elimination by immune surveillance mechanisms, they could also be another drain on the CD4+ T population.

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Another Approach to Census Taking In Vivo: Quantitative Image Analysis of Viral Populations in LT Virus production and storage have been measured by quantitative image analysis (QIA) of radioautographs of sections of LT hybridized to a 35S labelled riboprobe collection with sequences complementary to 90% of the RNA genome (11). The number of silver grains over a productively infected cell or in virions overlying FDCs is proportional to the number of copies of viral RNA that bound probe (Figure 3A; see color figure insert). The FDC and productively infected cell pools can be distinguished by the pattern of grains (diffuse over FDC; concentrated over productively infected cells, see e.g. Figure 2E,F). From the specific activity of the probe, exposure time, and efficiency of formation of silver grains, one can back-calculate from grain counts to an absolute number of copies of viral RNA. This calculation has been validated by comparing copy numbers determined in this way with copy numbers determined by independent methods (11). To enumerate silver grains in the requisite large scale, we analyzed images of radioautographs with a computer software program (Metamorph). In the upper panel on the left side of Figure 3B the arrow points to a cell with HIV-1 RNA in a color image of a LT section illuminated with transmitted and epipolarized light. In the middle panel in the black and white image of the section illuminated only with epipolarized light, the white silver grains over the cells and background stand out more clearly. In the lower left panel a threshold value has been set so that only the greater light reflected from the silver grain will be measured. This is indicated by a red overlay. With the tracing tool of the program (indicated by the white trace around the cell) the program measures the thresholded areas of the silver grains over the cell. By manually count of silver grains over cells or GCs, a conversion factor relating the measured area of the silver grains (in pixels) to the number of silver grains can be introduced into the program to record the latter automatically. Similarly in the right panel of Figure 3B, the red overlay indicates the thresholded area of silver grains in a GC that results from binding of the radiolabeled riboprobe to viral RNA in virions associated with the FDC network. Manual counting again establishes a conversion factor to obtain grain counts. The grain counts are then used to calculate the number of copies of viral RNA per cell or over FDCs in GCs. By measuring the area of the sections with the computer program’s tracing tool (red trace in Figure 3C), and multiplying it by the thickness of the section and a predetermined density, the number of copies of viral RNA in the FDC pool can be expressed per gm LT. Similarly, the frequency of productively infected cells can be expressed per gm LT, and the total number of copies of viral RNA in productively infected cell pool can be calculated from

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SAT/VKS P2: NBL/plb

February 27, 1999 16:35

QC: NBL/tkj T1: NBL

Annual Reviews AR078-01

P1: SAT/VKS

P2: NBL/plb

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

February 27, 1999

16:35

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Figure 3 Quantitative image analysis of viral populations in LT. (A) The number of silver grains (red dots) overlaying a cell with viral RNA is proportional to the number of copies of viral RNA that bound the radiolabeled probe. The copy number can therefore be back-calculated from the grain count. (B) This task has been facilitated by quantitative image analysis as described in the text. In C, the traced area of the section is used to determine its weight. Population sizes are expressed per gram of tissue, which can be extrapolated to total body estimates. (Modified and reproduced with permission from Haase AT et al 1996. Science 274:985–89.)

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

QC: PSA

Annual Reviews

AR078-20

POPULATION BIOLOGY OF HIV-1 INFECTION

633

the frequency/gm multiplied by the mean number of copies of viral RNA per productively infected cell. The total population size is extrapolated for a 70-kg individual by assuming that 1% of the weight is in LT (69) and that tonsil is representative of other LT. The validity of the latter assumption is supported by comparisons of tonsil with lymph node and spleen (11).

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

VIRUS PRODUCTION AND STORAGE IN LT The FDC Pool Measurement of viral pools in LT in the presymptomatic stages of infection revealed first of all the extraordinary size of the FDC pool of ∼1011 copies of viral RNA (11), corresponding to 5 × 1010 virions and 5 × 1014 copies of p24, the major viral capsid protein (Table 1) from virions alone, not counting free viral antigen bound to FDCs. The FDC pool of HIV-1 therefore is the dominant antigen in the immune system and is the major source of viral RNA in RNA isolated from LT, exceeding the viral RNA in productively infected cells by >50-fold. The level of HIV-1 RNA in virions associated with FDCs is also larger than plasma levels by two orders of magnitude, consistent with the conclusions that plasma levels greatly underestimate viral burden and that the LT are major reservoirs where HIV-1 is stored. In the later stage of infection (CD4+ T cells in blood >200/mm3) there are comparable quantities of virus stored in the FDC pool (23), and the same conclusions hold, whereas in the earlier stages the FDC pool is somewhat smaller and is closer in size to the viral load in productively infected cells. Even in the first weeks of infection, however, substantial quantities (>1010 virions) have already accumulated in the FDC pool (TS Schacker, manuscript in preparation). The rapidity with which the FDC pool and the pool of latently infected cells is established in the early stages of HIV-1 infection (70) underscores the difficulties in purging HIV-1 even if HAART is begun at the earliest recognizable stage of infection. Table 1 Changes in viral populations in the course of infection

Stage of infection

Blood Frequency of Number of copies of HIV-1 RNA CD4+ productively T cell infected Productively Intracellular ‘FDC’ Blood 3 1 2 count/mm cells infected cell 2 pool pool pool

Acute/early Presymptomatic Late 1

Total body estimates. Per cell. 3 Blood volume of 5 L. 2

393 402 194

5 × 108 4 × 107 8 × 107

55 74 50

3 × 106 3 × 109 4 × 109

1010 1011 1011

4 × 1010 9 × 108 2 × 109

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

634

QC: PSA

Annual Reviews

AR078-20

HAASE

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Productive Infection The frequency of productively infected cells (≥20 copies/HIV-1 RNA per cell, Table 1) varies between ∼107 to ∼108 cells with a mean of about 50–100 copies of viral RNA per cell. Both the mean copy number per cell and the number of cells with the highest concentration of viral RNA (< ∼200 copies per cell) are much lower than cells infected with HIV-1 in vitro (mean 1100 copies/cell, maximum ∼4000 copies/cell) (11), possibly because of premature loss of productive life in vivo. In LT where infections are initiated asynchronously, cells at the later stages of the viral life cycle, because of their higher levels of antigen production, will have a greater probability of detection and elimination by virus specific CTLs. They might therefore be eliminated before they can progress to the same late stage of the lentivirus life cycle where infected cells in culture make more than 90% of the total viral RNA and progeny (71). Alternatively and additionally, cytokines (72–75) and other CD8+ cell-associated factors can suppress HIV-1 gene expression and may generally damp down viral production in vivo.

Total Body Virus Production in LT Despite lower levels of viral RNA in productively infected cells in vivo, the virus these cells do produce in LT is sufficient to account for all of the virus in the body. Total body production was first estimated at >109 to >1010 virions per day from clearance of virus from peripheral blood, extrapolated to total extracellular fluid (22, 76). Similar estimates of total daily production of virus were obtained from direct measurements of the population of productively infected cells in LT and the estimate of virus per cell (11, 23), supporting the conclusion that the LT are the major site of virus production.

Localized Propagation of Infection in LT The pattern of productive infection in LT is consistent with the hypothesis that infection is propagated by transmission of virus from one cell to another activated T cell in its vicinity (78). The localized character of virus production was first appreciated when foci of infected cells were identified in LT (4), and from evidence of founder effects in the population of virus in individually dissected white pulps (79) and subsequently by direct tracking of individual SIV clonotypes with genotypic probes (80). The latter approach revealed spatially restricted but overlapping patterns of virus production of individual viral clonotypes in white pulps in distinct regions of the spleens of infected animals. In Figure 2G, H overlays of the sites where productively infected cells of each clonotype were found show that two clonotypes of SIV may replicate in T cells in the same white pulp but in distinct areas on a microscopic scale or in individual white pulps in which one but not the other viral clonotype is replicated (indicated by the arrows).

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

POPULATION BIOLOGY OF HIV-1 INFECTION

635

IMPACT OF HAART ON VIRAL AND INFECTED CELL POPULATIONS IN LT

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Two Populations of Productively Infected Cells Are Eliminated at Different Rates We have examined the effects of HAART on the LT reservoir (23) in a cohort of previously untreated individuals where there would be less likelihood that the interpretation of the results would be clouded by replication of drug resistant virus. In tonsillar biopsies obtained after two days of treatment with inhibitors of HIV-1 protease and reverse transcriptase, most of the productively infected cells with ≥75 copies of HIV-1 RNA per cell were no longer detectable. This was the expected result given the dynamic nature of infection in vivo where infection is propagated locally and asynchronously. Infected cells that are in the later stages of the viral life cycle will have the highest concentration of viral RNA and the shortest life expectancy as they succumb to the cytopathic effects of viral replication and/or are eliminated by CTLs. From the rate at which these cells disappeared when HAART blocked new rounds of infection, we estimated that there were 8 × 107 cells in this population of productively infected cells in LT that are lost each day at a relatively late stage of HIV-1 infection (the mean CD4+ T cell count in peripheral blood of this cohort was 194/mm3) (Table 2). I will return to this estimate in discussing the contribution this loss makes to depletion of CD4+ T cells, but I stress at this point that production of virus by this population is largely responsible for maintaining the pretreatment levels of virus in the circulation and FDC pool. The initially rapid decline of virus in blood and the FDC pool (see below) reflects the loss of virus production from this source when HAART interrupts the propagation of infection in LT. Since most of the virus before treatment was in the FDC pool, dividing the turnover of virus in the FDC pool by the initial turnover rate for productively infected cells gives a direct estimate of virus yield of about 180 virions/cell and virus production of 1.4 × 1010/day (Table 2). In subsequent tonsillar biopsies in the ensuing weeks Table 2 Viral population dynamics and the response to treatment Turnover of productively infected cells ≥75 copies Initial decay HIV-1 RNA rate FDC pool 8 × 107 cells/day

1.5 × 1010 virions/day

Virus yield 180 virions/cell

Virus production 1.4 × 1010 virions/day

Reduction of viral load in the FDC pool with treatment >2,500x

(The data used to construct Tables 1 and 2 were taken from reference 11, 23 and Schacker et al, manuscript submitted.)

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

636

QC: PSA

Annual Reviews

AR078-20

HAASE

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

there were still cells with ≥20 but <75 copies of viral RNA, and this population shrank at a slower rate. The decrease in productively infected cells in this population is similar to the second phase decay of virus in peripheral blood. This has been thought to be due to continued virus production by a cell such as a macrophage that does not succumb as quickly to infection and continues to produce virus (77). However, this longer lived infected cell turned out to be not Mφ, (23) but rather probably CD4+ T cells that are chronically infected, produce small amounts of virus, and as a consequence, have longer lifespans.

FDC Pool The FDC pool of HIV-1 had been expected to turn over slowly like other antigens over a period of many months or longer (53), but surprisingly the level of viral RNA declined at rates that paralleled the decline in numbers of productively infected cells. In the model of the kinetics of clearance of virus from the FDC pool shown in Figure 4, two populations of infected CD4+ T cells produce the virus in LT that maintains the levels of virus in blood, extracellular fluid, and tissues. Immune complexes of virus and antibody are deposited on FDCs, where they are processed conventionally and slowly by known pathways involving formation of immune complex aggregates called iccosomes (54, 55). The model also incorporates speculative pathways such as complement mediated virolysis (81) and phagocytosis by Mφs (82) capable of rapidly degrading virus and viral RNA in the FDC pool. The production of virus, storage, and clearance are assumed to be in equilibrium before treatment. After HAART is instituted, the infected cells that produce most of the virus and have the shortest lifespan quickly disappear, which is mirrored in the rapid decline of virus in circulation and in the FDC pool. The rate at which these levels decline then slows because of continued production and deposition of virus by the cells with a longer life span that are not immediately affected by HAART.

Control But Not Eradication of Infection After six months of HAART, Cavert et al documented in this cohort a more than 2500-fold reduction in the FDC pool of virions (23); after a year of treatment no viral RNA could be detected at the limit of about 104 copies/gm LT (W Cavert, personal communication) (Table 2). Productively infected cells with ≥20 copies of viral RNA per cell had also disappeared, but there were cells in some cases in LT that had a few copies of viral RNA (≤5). The extent of viral replication and significance of these cells is uncertain. They could be chronically infected cells that maintain a low level of virus production and evolution; or they might be cells in which the virus life cycle is incomplete because the levels of expression of viral regulatory genes (e.g. rev, tat) are insufficient to support later stages in the viral life cycle (83–86). Persistence of virus in these cells, and in cells

March 3, 1999

P2: PSA/SPD

10:10 Annual Reviews

POPULATION BIOLOGY OF HIV-1 INFECTION

Figure 4 Hypothetical mechanisms of virus production and clearance. Most HIV-1 is produced in continuing de novo cycles of infection and reinfection of CD4+ T cells with short life spans, but smaller amounts of virus are produced by cells with longer life spans. Most of the virus in LT is in immune complexes stored on FDCs cleared by slow and possibly fast processes such as virolysis and phagocytosis. Treatment (Rx ) interrupts the cycles of infection, levels of production and virus levels decline in the blood and FDC pool with biphasic kinetics related to the different life spans. See text for full explanation.

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: APR/ARY QC: PSA

AR078-20

637

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

638

QC: PSA

Annual Reviews

AR078-20

HAASE

with replication competent or defective proviruses account for the slow decay and substantial pool of viral DNA that is detectable in LT even after 2.5 years of effective suppression of viral replication (25). The inhibitory effects of HAART on viral replication and levels of virus in blood and LT have been consistent in a number of studies (23, 25, 87–89). As long as treatment suppresses replication to the limits of detection in blood, little if any viral RNA will be detectable in the FDC pool, in total RNA extracted from LT, or in cells isolated from LT; and there will be no productively infected cells with ≥20 or even ≥5 copies/cell of viral RNA. However, cells with lower levels of viral RNA can still be detected (23, 90) and latently infected CD4+ T cells have been detected in peripheral blood even after 2.5 years of treatment (25). The chronically infected cells, or replication of less fit drug resistant mutant viruses, result in the continued evolution of genotypic drug resistance (91). The chronically and latently infected cells, and residual virus in the FDC pool, represent potential sources of virus to restart infection if treatment for any reason is ineffective or is discontinued. In these circumstances, infection is quickly reestablished with the reappearance of productively infected cells and viral RNA in the FDC pool with similar kinetics to acute infection (90). The persistence of virus and recrudescence of infection have thus far largely frustrated efforts to eradicate HIV-1 or to discontinue the expensive and complicated treatment regimens required to suppress active replication. As is discussed in the next section, control of infection is nonetheless clearly beneficial in slowly repopulating LT and blood with CD4+ T cells.

CD4+ T CELL DEMOGRAPHICS IN HEALTH AND HIV-1 INFECTION Thymopoiesis and Other Sources of Naive Cells The four sections of Figure 5 entitled Renewal, Trafficking, Proliferation and Apoptosis, and Infection are intended as a conceptual framework and an abridged primer for the discussion that follows on CD4+ T cell renewal and homeostasis in a healthy adult human, and the possible consequences of HIV-1 infection on these processes. Normally, T cell progenitors in the bone marrow emigrate to the thymus where they rearrange their T cell receptors (TCR). T cells with TCRs specific for self-antigen are deleted, whereas TCRs specific for foreign antigens are retained. Differentiation takes place in the cortical regions of a thymic epithelial space (TES), comprised mainly of cytokeratin+ thymic epithelia, Hassall’s corpuscles (HCs), and CD3+ CD4+ CD8+ thymocytes. CD3+ CD4+ CD8− and CD3+ CD4− CD8+ “na¨ıve” T cells that have survived selection concentrate in the medulla and are exported to LT. These naive CD4+ T cells have the high molecular weight isoform of a phosphatase, CD45 (RA),

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

639

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POPULATION BIOLOGY OF HIV-1 INFECTION

Figure 5 Overview of CD4+ T cell depletion in HIV-1 infection and repopulation with treatment. Each titled and shaded section identifies a process that is described in detail in the text. Under Trafficking, the relative size of the arrow reflects the greater exchange of cells between blood and spleen. Under Infection, the Rx indicates that HAART interrupts the cycle of de novo infections. The ? to the thymus indicates the possibility that the sources of renewal may be directly or indirectly compromised in infection and relieved by treatment.

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

640

QC: PSA

Annual Reviews

AR078-20

HAASE

whereas memory CD4+ T cells have a lower molecular weight isoform (RO) (92–94). The TES is surrounded by a perivascular space (PVS) in which there are large numbers of lymphocytes and macrophages, granulocytes and fat cells. Thymopoiesis of naive T cells occurs within the TES, whereas the PVS is the site where lipomatous atrophy ordinarily occurs (95), or lymphoid hyperplasia in HIV-1 infection (95; B Haynes, L Haley, manuscript submitted). Normally the thymus continues to grow during the first year of life and does not change much in overall size thereafter, even at advanced ages. The TES, however, begins to involute at a relatively rapid and linear rate from 1 to 40 years of age, and then at a much slower rate, so that even after the age of 100 small amounts of TE tissue may persist (95; B Haynes, L Haley, manuscript submitted). The apparent overall diminution in thymic tissue is much less because during the first 25 years of life the PVS expands. Between the ages of 25 and 40, both the total PVS and TES decline at a rate of about 5% per year and 0.1% thereafter, with close to complete replacement of LT by fat. The “typical” human thymus from age 20 to 50, consists of fatty tissue within a fibrous capsule with islands or stripes of TES surrounded by the PVS (95; B Haynes, L Haley, manuscript submitted). Decreasing renewal capacity not surprisingly accompanies involution of the TES. When T cells have been depleted by irradiation and chemotherapy in patients with cancer (98) or with monoclonal antibodies to CD4 in patients with multiple sclerosis (99), on cessation of treatment regeneration of naive CD4+ T cells is inversely related to age. In infants, rough estimates of repopulation rates for naive CD4+ T cells (98, 99) are of the order of >109 cells/day, whereas by the age of 20 the rate is 6–7 × 107/day. In both adults and children, the growth rate for total CD4+ T cells is >109 (100, 101), but this initial expansion is almost exclusively in the RO+ subset and is antigen driven. The resulting population is prone to activation induced programmed cell death (AICD) (102) and has a contracted and skewed TCR repertoire (103, 104). Restoration of CD4+ T cell counts in peripheral blood to >600/mm3 occurs in only about 50% of adults in the second year following cancer chemotherapy, and the increase in naive CD45+ CD4+ T cells is slow and delayed (104). I return shortly to the striking parallels in the repopulation of CD4+ T cells in AIDS patients responsive to HAART. Theoretically, there are other possible sources of naive CD4+ T cells. Memory RO+ CD4+ T cells can revert to RA+ cells (105), but this conversion occurs on average on a time scale >3 years (106), and in studies both in mice and humans the RO → RA conversion pathway is minor (104, 107). In mice extrathymic pathways exist for T cell development (108, 109), but there is no evidence that these pathways contribute in any substantial way to T cell renewal in humans.

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

POPULATION BIOLOGY OF HIV-1 INFECTION

641

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Lymphocyte Homeostasis and Niche Size Naive CD4+ T cells have receptors such as L-selectin (62L+) for homing to LT where they may be activated by antigens and cytokines to become RO+ cells. The latter population expands by proliferation and contracts by programmed cell death and removal to leave a surviving population comprised of longlived resting memory cells. In the adult, the total population is maintained at a relatively constant level independent of the supply of naive cells by homeostatic mechanisms (110, 111) that involve competition for cytokines, other growth and transcription factors (112), and by continued contact between cells and antigen-MHC complexes (113, 114). The early seeding of LT with naive cells thus provides the immune system with a population of cells that can respond to a broad range of antigens, and proliferation and persistence of mature T cells maintain T cell populations, immunological memory, and the ability to quickly mount an effective immune response. The competitive homeostatic mechanisms and dynamic processes of cell division and death maintain a quasi– steady state population in which a newly produced lymphocyte survives only if another cell is lost (111), and expansion and contraction of populations thus reflect temporary pertubations in the balance between division and survival vs activation-induced death and removal. The concept of a “niche size” for T cell populations also derives from the enormous proliferative capacity of T cells (115) and competitive homeostatic mechanisms that operate to maintain relatively stable populations. After T cells are depleted by some means, T cells proliferate in the regenerative process to refill the niche to about the same size as it was before the system was perturbed. I will shortly discuss the evidence that the LT niche is impaired in its ability to support full expansion of CD4+ T cell populations in LT of HIV-1 infected individuals after HAART.

Lymphocyte Trafficking The segment of Figure 5 entitled Trafficking illustrates the exchange of naive and memory T cells between blood and LT (116, 117). This occurs about 50 times a day, with about half of the daily recirculation of T cells between blood and spleen. Only a small fraction of the total population of T cells is in the circulation at any time (2%). Naive and differentiated memory/RO+ effector T cells also differ in that only the latter traffic through the lung, CNS, liver, and peripheral tissues, in keeping with their role in countering microbial pathogens throughout the body.

Normal Values In Figure 6 I have reduced the processes shown in Figure 3 to a schematic skeleton of the inputs and outputs and a balance sheet of credits and debits to the total body pool of CD4+ T cells in LT. I summarize in the balance sheet

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: APR/ARY P2: PSA/SPD

March 3, 1999 10:10

642

QC: PSA

Annual Reviews

HAASE

AR078-20

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POPULATION BIOLOGY OF HIV-1 INFECTION

643

the more recent direct measurements, by quantitative image analysis analogous to grain counts, of LT CD4+ T cell populations in normal adults and in HIV-1 infected individuals. In the normal young adult (<30 years of age) the total population of CD4+ T cells in LT is 2 × 1011 (28, 118), compared to 5 × 109 cells in peripheral blood. The 98%/2% distribution in the two compartments is in excellent agreement with extrapolations derived from studies of rat T cell populations (119). Both blood (120, 121) and LT (28, 122) have comparable fractions (45%) of putatively naive CD45RA+ CD4+ T cells. About 0.5% of the cells are cycling by the criteria that they are Ki67+ (a marker for cells in G1 through S, M and G2), and only about half as many cells are in the late stages of apoptosis (TUNEL+). Under the assumption that the population size is at steady state, and that the cell and programmed death cycles occur over a 24-h period (123), the unmeasured fraction of cells in the earlier stages of apoptosis and programmed removal pathway is assumed to make up the difference to maintain the system in equilibrium.

Redistribution of CD4+ T Cells in HIV-1 and SIV Infection In the late stages of HIV-1 infection (blood CD4+ T cell counts ≤ ∼200/mm3), there are fewer CD4+ T cells in blood vs LT (1%/99%) compared to seronegative controls (2%/98%), and the total number of CD4+ T cells in LT is relatively higher than would be expected from a greater proportional reduction of counts in blood in HIV-1 (28, 124) and SIV infection (125). Both observations immediately point to altered trafficking of CD4+ T cells in the chronically activated state associated with infection as part of the explanation for the lower CD4+ ←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 6 Balance sheet of the inputs and outputs that affect the CD4+ T cell population in LT and blood. See text for explanation and the basis for the estimates shown at the right. Excess capacity is defined by repopulation rates in a steady state model where the input of new naive cells is normally balanced by activation and other processes that diminish the pool of naive cells. When the balance is perturbed as in MS patients with antibody to CD4 (98), which presumably eliminates mature and progenitor cells, input falls to 0, and the pool is depleted at a rate equal to normal output rate plus the additional rate of losses mediated by antibody. After antibody treatment is discontinued, input resumes at a pretreatment rate equal to output. The increases over the baseline therefore equal the extra regenerative capacity. The repopulation rate estimated from the blood from the data in reference 98 was calculated by dividing the blood (B) CD4+ T cell count/mm3 × 5 × 106 mm3 total blood volume × LT/B = 49 by 2 (∼50% naive CD4+ T cells) X number of days over which the increase occurred. The estimates for LT were based on direct measurements of repopulation rates and % naive CD4+ T cells in LT (28). CD4+ T cells were classified as naive or memory cells based on immunohistochemical staining as CD45RA+ or CD45RO+. Although in peripheral blood, naive cells can be more accurately immunohistotyped by staining for the 62L homing receptor, presumably the CD45RA+ cells in LT must have this receptor if they are in LT.

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

644

QC: PSA

Annual Reviews

AR078-20

HAASE

T cell counts in blood. Altered trafficking also likely accounts for the immediate effects of treatment in blood CD4+ T cell counts, as had been proposed previously (126–130). Following HAART, the total CD4 count increases initially at a rate of about 109 cells/day in both blood and LT (Figure 7) (28, 130), comparable to the first estimates from CD4+ T cell turnover in peripheral blood (21, 22). Most of the initial increase is in the RO+ subset (27, 28, 130, 131). Thereafter, however, the rise in CD4+ RO+ cells soon reaches a plateau and the slow linear increase in the naive subset in blood (26, 130) and LT (28) becomes increasingly significant. In LT, about 70% of the increase in total CD4+ T cells from 1011 to 1.2 × 1011 after 6 months of treatment is the result of increases in the CD4+ RA+ subset (28). These findings support a repopulation model in which in the initial stages, mature CD4+ T cells, perhaps in the gut, lung, and other tissue compartments, are rapidly redistributed to blood and peripheral lymph nodes following treatment. This redistribution (126–130), rather than the proposed relief from the high turnover (22) of productively infected cells, is the likely explanation for the immediate rebound of CD4+ T cells counts in blood and LT.

Proliferation, Apoptosis and Productive Infection In this cohort, the fraction of proliferating cells also fell, and the TUNEL+ fraction remained unchanged so that a change in the balance between cell division and death cannot account for the early rise in CD4+ T cells in LT or blood. Recent measurements of the proliferating CD4+ T cell population in LT (28, 118) estimate in the normal adult that 0.4 to 0.5% of the CD4+ T cells are ki67+. In the early stages of HIV-1 infection, the Ki67+ CD4+ T cell population prior to treatment (118) matched that of HIV-1 seronegative controls but increased after six months of treatment. In the late stages of infection, we found about a three-fold increase in Ki67+ CD4+ T cells prior to treatment, and this fraction declined to normal levels after six months of HAART (28). It would have been difficult, if not impossible, to predict the actual proportion of proliferating CD4+ T cells in LT from the proportion in blood because of the altered migration of lymphocytes in the state of chronic activation induced by infection. Nonetheless, in several studies using Ki67 staining (132), 3HdTr (133) or BrDu labelling (134) CD4+ T cell turnover in blood in HIV-1 and SIV infection (135) was also found to be elevated two- to threefold. Telomere length measurements are also compatible with this but not greater increases in CD4+ T cell turnover (136, 137). We also found that the proportion of CD4+ T cells in the later stages of apoptosis (TUNEL+) in LT is about twofold higher in HIV-1 infected individuals than in the normal controls (28). This was not unexpected as repeated stimulation through the CD3/TCR causes AICD (102), and increased

March 3, 1999

P2: PSA/SPD

10:10 Annual Reviews

Figure 7 Comparison of repopulation of blood and LT with CD4+ T cells and na´ıve and memory subsets. Mean values for the same set of patients treated with HAART for 6 months described in references 28, 130.

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: APR/ARY QC: PSA

AR078-20

POPULATION BIOLOGY OF HIV-1 INFECTION

645

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

646

QC: PSA

Annual Reviews

AR078-20

HAASE

apoptosis of CD4+, CD8+ T cells, and B cells in LT in HIV-1-infected individuals proportional to the extent of activation had been documented previously in LT (138). Some of the increased apoptosis in the CD4+ T cell subset may be the result of cross-linking of CD4 by gp120 shed from infected cells or virions that prime CD4+ T cells for AICD. Abundant gp120 (as well as other viral and nonviral antigens) in LT, and gp120 has been detected by immunoelectron microscopy in the membranes of apoptotic CD4+ T cells isolated from lymph nodes of HIV-1-infected individuals (139, 140). This mechanism of CD4+ T cell depletion is in addition to losses due to productive infection, as apoptosis occurs predominantly in bystander cells, and not in the productively infected cells themselves (141). The population of apoptotic CD4+ T cells undergoing removal is unknown because many cells at early stages of cell death will have been phagocytosed and are no longer recognizable as TUNEL+ CD4+ T cells. Under the same steady state assumptions applied to normal adults, the sum of cell death and removal, plus T cells outside the surveyed LT, is assumed to equal proliferation. The necessity of these ad hoc assumptions weakens the analysis of mechanisms of immune depletion and repopulation, and it underscores the importance of developing methods to directly measure the total losses to programmed cell death and removal.

Depletion of Naive CD4+ T Cells in HIV-1 Infection and Repopulation with Treatment Although the total number of CD4+ T cells was relatively preserved in LT compared to blood (28, 124), by the late stages of infection the pool was halved. There were disproportionately large decreases in LT of naive CD45RA+ CD4+ T RO+ cell compartments. Following treatment in LT and in peripheral blood (28, 130), the naive population of CD4+ T cells in LT increased at a slow and linear rate of about 8 × 107 cells/day. This directly measured rate is close to the estimate of repopulation rates in adults recovering from cancer chemotherapy or treatment with antibody to CD4 (98, 99). After 12–14 months of suppression of active infection with HAART, the proportion of CD45RA+ CD4+ T cells had recovered to about half the normal level (28). In peripheral blood, naive CD4+ T cells also increased slowly, but only after several months of treatment (27). Immune responses to opportunistic pathogens, e.g. CMV and TB, also improved after a year of treatment in patients in whom treatment effects sustained reductions in the levels of HIV-1 (29, 30). Perturbations in the CD4+ T cell repertoire in the late stages of infection generally persist for at least the first six months of treatment (142), although some early normalization of the repertoire has been reported (143).

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

POPULATION BIOLOGY OF HIV-1 INFECTION

647

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Renewal Mechanisms and Prospects for Immune Reconstitution The treatment-associated increase in CD45RA+ CD4+ T cells in blood and LT (27, 28), normalization of the T cell repertoire and function (29, 30), and correlation between detection of thymic tissue by CT scan and a higher percentage and absolute number of circulating CD45RA+ CD62L+ CD4+ T cells (144) are all consistent with the conclusion that functional thymic tissue may be preserved in HIV-1-infected adults. Several other recent observations are also consistent with this conclusion. Haynes et al (manuscript submitted) found functioning thymic tissue in two HIV-1-infected individuals who had had CD4+ T cell counts under 50 cells/mm3 in the preceding three years. The frequency of productively infected cells (96, 97, 145) is relatively low, and histology of TES in thymic tissue in the asymptomatic stage of HIV-1 infection (96, 97) is relatively normal; transient renewal of thymopoiesis has been documented in HIV-1-infected human thymic implants following antiviral therapy (146). The counter-arguments, from postmortem studies of thymus from the terminal stages of HIV-1 infection, are that the majority of the thymic epithelial component is usually destroyed, HCs are greatly reduced, and in many cases, there is no evidence of active thymopoiesis (147–149; BT Haynes, manuscript submitted). Increases in peripheral blood CD4+ T cells, including the naive subset, have also been documented after treatment of a thymectomized HIV-1-infected individual (BT Haynes, manuscript submitted), although another report (150) provided no evidence of regeneration of CD45RA+ cells in a thymectomized individual after bone marrow transplantation. In addition, the thymocyte precursors and epithelium are infected and depleted in human thymus explanted into SCID mice and infected with HIV-1 (45, 151); multilineage hematopoiesis is indirectly inhibited in HIV-1 infection by unknown mechanisms that alter the survival and differentiation of CD34+ progenitor cells (152–155). These observations collectively point to damage to the bone marrow–thymus pathway for regeneration of naive CD4+ T cell in HIV-1 infection.

Depletion and Repopulation of Naive CD4+ T Cells: Supply Side Economics and Trickle Down Deficits I conclude this chapter with some speculations on the mechanisms of CD4+ T cell depletion in HIV-1 infection, particularly of naive cells, the contributions of productive infection to these losses, and the mechanisms of and expectations for repopulation. I begin with the concept (28, 102–104, 136, 156) that the excess capacity (Figure 6) for renewal of naive CD4+ T cells, inferred from repopulation rates in peripheral blood, or measured directly in LT, is extremely limited in the adult. Of the approximately 1011 naive CD4+ T cells, the capacity to

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

648

QC: PSA

Annual Reviews

AR078-20

HAASE

increase the supply of new naive cells beyond the normal turnover is only about 7 × 107 cells/day or <0.1% of the total population. To underscore the marginal excess immune regenerative capacity (28), we have drawn a literary analogy to the Red Queen, who in Through the Looking Glass, explains to Alice that “it takes all the running you can do, to keep in the same place” (157). In HIV-1 infection, activation draws off naive cells into the RO+ pool at an increased rate, and a portion of this population as well as of activated memory cells are susceptible and succumb to HIV-1 infection. The excess capacity in renewal sources can partially compensate for the losses to productive infection of 2 × 107 cells/day in the presymptomatic stages (11), but as infection progresses and the daily losses to productive infection in the late stages become greater (8 × 107 cells/day), deficits accumulate in the naive pool in proportion to their entry into and representation in the pool of productively infected cells. An average imbalance of only about 2 × 107 cells/day between supply and losses will result in the deficit of 6.5 × 1010 naive cells (Figure 6) over the average incubation period to AIDS of about 10 years. We refer to this mechanism of naive CD4+ T cell depletion as a trickle down mechanism because the drain of productive infection on the pool of CD4+ T cells is a relative “trickle” compared to the much larger number of cells lost to programmed cell death. It can nonetheless slowly deplete the pool of naive CD4+ T cells because it is an uncompensated loss, whereas the losses to apoptosis can be balanced by increased proliferation.

Residual Ailments in the Niche and Partial Recovery of CD4+ T Cell Populations Most of the increase of ∼1010 CD4+ CD45RO+ T cells in LT and blood occurs within a few weeks of initiating HAART and then occurs at a slow rate (LT) (28) or does not change (blood) (130). This initial repopulation of LT and blood with RO+ CD4+ T cells is likely for reasons already discussed to be largely the result of redistribution. It is also likely that there is residual damage to the LT niche’s ability to support recovery of LT populations of mature CD4+ T cells to preinfection levels. After six months of HAART we found that the fraction of proliferating CD4+ T cells had actually decreased to normal levels rather than increasing to repair the deficits as would be expected if homeostatic mechanisms were altogether intact. The TUNEL+ population was also higher than in uninfected individuals as further evidence of abnormalities in LT milieu that impede recovery. Later positive balances between proliferation and programmed cell death may result in continued increases in the total population of CD4+ T cells. There is also a possibility that the RO+ population will continue to grow because thymic emigrants are activated and enter the pool of proliferating RO+ cells, as Mackall et al have shown in regeneration of T cells in old mice (158).

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POPULATION BIOLOGY OF HIV-1 INFECTION

649

The expectation from the analysis in this and the preceding section is that recovery of the immune system with effective control of HIV-1 infection will not be unlike that in adults following cancer chemotherapy or treatment with antibody to CD4 (98, 99). The initial increases will be in the memory subset with slow recovery of naive cells to a variable extent related to age and preexisting damage to renewal sources. The bad news is that immune regeneration in these settings is predictably slow, variable, and partial. The good news is that even in relatively late stages of HIV-1 infection the capacity for renewal has not been irreparably damaged in many cases and even partial restoration of immune function translates into better control of OIs and decreased death rates due to AIDS (31). The challenges for the future are to prevent infection in the first place, to continue to develop more effective and less costly treatments, and to explore ways to further enhance rehabilitation of the immune system. ACKNOWLEDGMENTS I thank Melodie Bahan and Tim Leonard for preparation of the manuscript and figures, and I am grateful to Jan Anderson, Brigitte Autran, Winston Cavert, Tom Fehniger, Mark Feinberg, Zvi Grossman, Barton Haynes, Keith Henry, Marc Jenkins, Richard A Koup, Crystal Mackall, Joseph M McCune, Angela McLean, Frank Miedema, William Paul, Alan S Perelson, Paul Racz, Timothy Schacker, Lawrence Steinman, Klara Tenner-Racz, Stephen Wietgrefe, Steven Wolinsky, and Zhi-Qiang Zhang for their helpful comments and discussion. Work from my laboratory cited in this review was supported by NIH grants AI28246 and AI38565. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Quinn TC. 1996. Global burden of the HIV pandemic. Lancet 348:99–106 2. Tenner-Racz K, Racz P, Gluckman JC, Popovic M. 1988. Cell-free HIV in lymph nodes of patients with persistent generalized lymphadenopathy and AIDS. N. Engl. J. Med. 318:49–50 3. Tenner-Racz K, Racz P, Gartner S, Ramsauer J, Dietrich M, Gluckman JC, Popovic M. 1989. Ultrastructural analysis of germinal centers in lymph nodes of patients with HIV-1-induced persistent generalized lymphadenopathy: evidence for persistence of infection. In Progress in AIDS Pathology, ed. H Rotterdam, SC Sommers, P Racz, PR Meyer, pp 29–40. New York: Field and Wood

4. Tenner-Racz K, Racz P, Schmidt H, Dietrich M, Kern P, Louie A, Gartner S, Popovic M. 1988. Immunohistochemical, electron microscopic and in situ hybridization evidence for the involvement of lymphatics in the spread of HIV-1. AIDS 2:299–309 5. Parmentier HK, van Wichen D, Sie-Go DMDS, Goudsmit J, Borleffs JCC, Schuurman HJ. 1990. HIV-1 infection and virus production in follicular dendritic cells in lymph nodes. Am. J. Pathol. 137:247–51 6. Pantaleo G, Grazioso C, Butini L, Pizzo PA, Schnittman SM, Kotler DP, Fauci AS. 1991. Lymphoid organs function as major reservoirs for human immunodefi-

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

650

7.

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

8.

9.

10.

11.

12.

13.

14.

15.

16.

QC: PSA

Annual Reviews

AR078-20

HAASE ciency virus. Proc. Natl. Acad. Sci. USA 88:9838–42 Fox CH, Tenner-Racz K, Racz P, Firpo A, Pizzo PA, Fauci AS. 1991. Lymphoid germinal centers are reservoirs of human immunodeficiency virus type 1 RNA. J. Infect. Dis. 164:1051–57 Spiegel H, Herbst H, Niedobitek G, Foss HD, Stein H. 1992. Follicular dendritic cells are a major reservoir for human immunodeficiency virus type 1 in lymphoid tissues facilitating infection of CD4+ T– helper cells. Am. J. Path. 140:15–22 Embretson J, Zupancic M, Ribas JL, Burke A, Racz P, Tenner–Racz K, Haase AT. 1993. Massive covert infection of helper T lymphocytes and macrophages by HIV during the incubation period of AIDS. Nature 362:359–62 Pantaleo G, Graziosi C, Demarest JF, Butini L, Montroni M, Fox CH, Orenstein JM, Kotler DP, Fauci AS. 1993. HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease. Nature 362:355–9 Haase AT, Henry K, Zupancic M, Sedgewick G, Faust RA, Melroe H, Cavert W, Gebhard K, Staskus K, Zhang ZQ, Daily PJ, Balfour HH, Erice A, Perelson AS. 1996. Quantitative image analysis of HIV-1 infection in lymphoid tissue. Science 274:985–9 Daar ES, Moudgil T, Meyer RD, Ho DD. 1991. Transient high levels of viremia in patients with primary human immunodeficiency virus type 1 infection. N. Engl. J. Med. 324:961–4 Clark SJ, Saag MS, Decker WD, Campbell–Hill S, Roberson JL, Veldkamp PJ, Kappes JC, Hahn BH, Shaw GM. 1991. High titers of cytopathic virus in plasma of patients with symptomatic primary HIV-1 infection. N. Engl. J. Med. 324:954–60 Cooper DA, Gold J, Maclean P, Donovan B, Finlayson R, Barnes TG, Michelmore HM, Brooke P, Penny R. 1985. Acute AIDS retrovirus infection: definition of a clinical illness associated with seroconversion. Lancet 1:537–40 Ho DD, Sarngadharan MG, Resnick L, DiMarzo–Veronese F, Rota TR, Hirsch MS. 1985. Primary human T–lymphotropic virus type III infection. Ann. Intern. Med. 103:880–83 Henrard DR, Daar E, Farzadegan H, Clark SJ, Phillips J, Shaw GM, Busch MP. 1995. Virologic and immunologic characterization of symptomatic and asymptomatic primary HIV-1 infection. J. Acquir. Immune Defic. Syndr. 9:305–10

17. Schacker TW, Hughes JP, Shea T, Coombs RW, Corey L. 1998. Biological and virologic characteristics of primary HIV infection. Ann. Intern. Med. 128:613–20 18. Koup RA, Safrit JR, Cao Y, Andrews CA, McLeod G, Borkowsky W, Farthing C, Ho DD. Temporal association of cellular immune responses with the initial control of viremia in primary human immunodeficiency virus type 1 syndrome. J. Virol. 68:4650–55 19. Pantaleo G, Graziosi C, Fauci AS. 1993. The immunopathogenesis of human immunodeficiency virus infection. N. Engl. J. Med. 328:327–35 20. Mellors JW, Rinaldo CR, Gupta P, White RM, Todd JA, Kingsley LA. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 272:1167–70 21. Wei X, Ghosh SK, Taylor ME, Johnson VA, Emini EA, Deutsch P, Lifson JD, Bonhoeffer S, Nowak MA, Hahn BH, Saag MS, Shaw GM. 1995. Viral dynamics in human immunodeficiency virus type 1 infection. Nature 373:117–22 22. Ho DD, Neumann AU, Perelson AS, Chen W, Leonard JM, Markowitz M. 1995. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 373:123–26 23. Cavert W, Notermans DW, Staskus K, Wietgrefe SW, Zupancic M, Gebhard K, Henry K, Zhang ZQ, Mills R, McDade H, Schuwirth CM, Goudsmit J, Danner SA, Haase AT. 1997. Kinetics of response in lymphoid tissues to antiretroviral therapy of HIV-1 infection. Science 276:960–64 24. Finzi D, Hermankova M, Pierson T, Carruth LM, Buck C, Chaisson RE, Quinn TC, Chadwick K, Margolick J, Brookmeyer R, Gallant J, Markowitz M, Ho DD, Richman DD, Siliciano RF. 1997. Identification of a reservoir for HIV-1 in patients on highly active antiretroviral therapy. Science 278:1295–1300 25. Wong JK, Hezareh M, Gunthard HF, Havlir DV, Ignacio CC, Spina CA, Richman DD. 1997. Recovery of replication– competent HIV despite prolonged suppression of plasma viremia. Science 278: 1291–95 26. Chun TW, Stuyver L, Mizell SB, Ehler LA, Mican JAM, Baseler M, Lloyd AL, Nowak MA, Fauci AS. 1997. Presence of an inducible HIV-1 latent reservoir during highly active antiretroviral therapy. Proc. Natl. Acad. Sci. USA 94:13193–97 27. Autran B, Carcelain G, Li TS, Blanc C, Mathez D, Tubiana R, Katlama C, Debre P, Leibowitch J. 1997. Positive effects of

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

QC: PSA

Annual Reviews

AR078-20

POPULATION BIOLOGY OF HIV-1 INFECTION

28.

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

29.

30.

31.

32.

33.

34.

35.

36.

combined antiretroviral therapy on CD4+ T cell homeostasis and function in advanced HIV disease. Science 277:112–16 Zhang ZQ, Notermans DW, Sedgewick G, Cavert W, Wietgrefe S, Zupancic M, Gebhard K, Henry K, Boies L, Chen Z, Jenkins M, Mills R, McDade H, Goodwin C, Schuwirth CM, Danner SA, Haase AT. 1998. Kinetics of CD4+ T cell repopulation of lymphoid tissues after treatment of HIV-1 infection. Proc. Natl. Acad. Sci. USA 95:1154–59 Li TS, Tubiana R, Katlama C, Calvez, Ait Mohand H, Autran B. 1998. Long lasting recovery in CD4+ T cell function mirrors viral load reduction after highly active anti–retroviral therapy in patients with advanced HIV disease. Lancet 351:1682–86 Komanduri KV, Viswanathan MN, Wieder ED, Schmidt DK, Bredt BM, Jacobson MA, McCune JM. 1998. Restoration of cytomegalovirus–specific CD4+ T–lymphocyte responses after ganciclovir and highly active antiretroviral therapy in individuals infected with HIV-1. Nature Med. 4:953–56 Palella FJ, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, Aschman DJ, Holmberg SD, HIV Outpatient Study Investigators. 1998. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N. Engl. J. Med. 338, 853–60 Miller CJ, Alexander NJ, Vogel P, Anderson J, Marx PA. 1992. Mechanism of genital transmission of SIV: a hypothesis based on transmission studies and the location of SIV in the genital tract of chronically infected female rhesus macaques. J. Med. Primatol. 21,65–68 Spira AI, Marx PA, Patterson BK, Mahoney J, Koup RA, Wolinsky SM, Ho DD. 1996. Cellular targets of infection and route of viral dissemination after an intravaginal inoculation of simian immunodeficiency virus into rhesus macaques. J. Exp. Med. 183,215–25 Marx PA, Spira AI, Gettie A, Dailey PJ, Veazey RS, Lackner AA, Mahoney CJ, Miller CJ, Claypool LE, Ho DD, Alexander NJ. 1996. Progesterone implants enhance SIV vaginal transmission and early virus load. Nature Med. 2:1084–89 Veazey RS, DeMaria M, Chalifoux LV, Shvetz DE, Pauley DR, Knight HL, Rosenzweig M, Johnson RP, Desrosiers RC, Lackner AA. 1998. Gastrointestinal tract as a major site of CD4+ T cell depletion and viral replication in SIV infection. Science 280:427–31 Reinhart TA, Rogan MJ, Huddleston D,

37.

38.

39.

40.

41.

42.

43.

44.

45.

651

Rausch DM, Eiden LE, Haase AT. 1997. Simian immunodeficiency virus burden in tissue and cellular compartments during clinical latency and AIDS. J. Infect. Dis. 176:1198–1208 Reimann KA, Tenner-Racz K, Racz P, Montefiori DC, Yasutomi Y, Lin W, Ransil BJ, Letvin NL. 1994. Immunopathogenic events in acute infection of rhesus monkeys with simian immunodeficiency virus of macaques. J. Virol. 68: 2362–70 Chakrabarti L, Cumont MC, Montagnier L, Hurtrel B. 1994. Variable course of primary simian immunodeficiency virus infection in lymph nodes: relation to disease progression. J. Virol. 68:6634–42 Chakrabarti L, Isola P, Cumont MC, Claessens-Maire MA, Hurtrel M, Montagniew L, Hurtrel B. 1994. Early states of simian immunodeficiency virus infection in lymph nodes: evidence for high viral load and successive populations of target cells. Am. J. Pathol. 144:1226–36 Orenstein JM, Fox C, Wahl SM. 1997. Macrophages as a source of HIV during opportunistic infections. Science 276: 1857–61 Frankel SS, Wenig BM, Burke AP, Mannan P, Thompson LDR, Abbondanzo SL, Nelson AM, Pope M, Steinman RM. 1996. Replication of HIV-1 in dendritic cell–derived syncytia at the mucosal surface of the adenoid. Science 272:115–17 Frankel SS, Tenner–Racz K, Racz P, Wenig BM, Hansen CH, Heffner D, Nelson AM, Pope M, Steinman RM. 1997. Active replication of HIV-1 at the lymphoepithelial surface of the tonsil. Am. J. Pathol. 151:89–96 McIlroy D, Autran B, Cheynier R, Wain– Hobson S, Clauvel JP, Oksenhendler E, Debre P, Hosmalin A. 1995. Infection frequency of dendritic cells and CD4+ T lymphocytes in spleens of human immunodeficiency virus–positive patients. J. Virol. 69:4737–45 Livingstone WJ, Moore M, Innes D, Bell JE, Simmonds P, the Edinburgh Heterosexual Transmission Study Group. 1996. Frequent infection of peripheral blood CD8–positive T–lymphocytes with HIV1. Lancet 348:649–54 Stanley SK, McCune JM, Kaneshima H, Justement JS, Sullivan M, Boone E, Baseler M, Adelsberger J, Bonyhadi M, Orenstein J, Fox CH, Fauci AS. 1993. Human immunodeficiency virus infection of the human thymus and disruption of the thymic microenvironment in the SCID– hu mouse. J. Exp. Med. 178:1151–63

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

652

QC: PSA

Annual Reviews

AR078-20

HAASE

46. Jamieson BD, Uittenbogaart CH, Schmid I, Zack JA. 1997. High viral burden and rapid CD4+ cell depletion in human immunodeficiency virus type 1–infected SCID–hu mice suggest direct viral killing of thymocytes in vivo. J. Virol. 71:8245– 53 47. Kitchen SG, Uittenbogaart CH, Zack JA. 1997. Mechanism of human immunodeficiency virus type 1 localization in CD4–negative thymocytes: differentiation from a CD4–positive precursor allows productive infection. J. Virol. 71: 5713–22 48. Lee S, Goldstein H, Baseler M, Adelsberger J, Golding H. 1997. Human immunodeficiency virus type 1 infection of mature CD3hiCD8+ thymocytes. J. Virol. 71:6671–76 49. Flamand L, Crowley RW, Lusso P, Colombini–Hatch S, Margolis DM, Gallo RC. 1998. Activation of CD8+ T lymphocytes through the T cell receptor turns on CD4 gene expression: implications for HIV pathogenesis. Proc. Natl. Acad. Sci. USA 95:3111–16 50. Yang LP, Riley JL, Carroll RG, June CH, Hoxie J, Patterson BK, Ohshima Y, Hodes RJ, Delespess G. 1998. Productive infection of neonatal CD8+ T lymphocytes by HIV-1. J. Exp. Med. 187:1139–44 51. Hufert FT, van Lunzen J, Janossy G, Bertram S, Schmitz J, Haller O, Racz P, von Laer D. 1997. Germinal centre CD4+ T cells are an important site of HIV replication in vivo. AIDS 11:849–57 52. Klaus GGB, Humphrey JH, Kunkl A, Dongworth DW. 1980. The follicular dendritic cell: its role in antigen presentation in the generation of immunological memory. Immunol. Rev. 53:3–28 53. Tew JG, Kosco MH, Szakal AK. 1989. The alternative antigen pathway. Immunol. Today 10:229–32 54. Szakal AK, Kosco MH, Smith JP, Tew JG. 1991. Kinetic and ultrastructural aspects of the antigen transport-FDC-iccosome-B cell axis. In Accessory Cells in HIV and Other Retroviral Infections, ed. P Racz, CD Dijkstra, JC Gluckman, pp 29–43. Basel, Switzerland: Karger 55. Szakal AK, Kosco MH, Tew JG. 1988. A novel in vivo follicular dendritic celldependent iccosome–mediated mechanism for delivery of antigen to antigen– processing cells. J. Immunol. 140:341–53 56. Burton GF, Conrad DH, Szakal AK, Tew JG. 1993. Follicular dendritic cells and B cell costimulation. J. Immunol. 150:31– 38 57. Spear GT, Landay AL, Sullivan BL, Dit-

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

tel B, Lint TF. 1990. Activation of complement on the surface of cells infected by human immunodeficiency virus. J. Immunol. 144:1490–96 Spear GT, Takefman DM, Sullivan BL, Landay AL, Zolla-Pazner S. 1993. Complement activation by human monoclonal antibodies to human immunodeficiency virus. J. Virol. 67:53–59 Joling P, Bakker LJ, Van Strijp JAG, Meerloo T, de Graaf L, Dekker MEM, Goudsmit J, Verhoef J, Schuurman HJ. 1993. Binding of human immunodeficiency virus type-1 to follicular dendritic cells in vitro is complement dependent. J. Immunol. 150:1065–73 Armstrong JA, Horne R. 1984. Follicular dendritic cells and virus–like particles in AIDS-related lymphadenopathy. Lancet 2:370–72 Tsunoda R, Hashimoto K, Baba M, Shigeta S, Sugai N. 1996. Follicular dendritic cells in vitro are not susceptible to infection by HIV-1. AIDS 10:595–602 Reinhart TA, Rogan MJ, Viglianti GA, Rausch DM, Eiden LE, Haase AT. 1997. A new approach to investigating the relationship between productive infection and cytopathicity in vivo. Nature Med. 3:218– 21 Heath SL, Tew JG, Tew JG, Szakal AK, Burton GF. 1995. Follicular dendritic cells and human immunodeficiency virus infectivity. Nature 377:740–44 Haase AT, Retzel EF, Staskus KA. 1990. Amplification and detection of lentiviral DNA inside cells. Proc. Natl. Acad. Sci. USA 87:4971–75 Chun TW, Carruth L, Finzi D, Shen X, DiGuiseppe JA, Taylor H, Hermankova M, Chadwick K, Margolick J, Quinn TC, Kuo YH, Brookmeyer R, Zeiger MA, Barditch-Crovo P, Siliciano RF. 1997. Quantification of latent tissue reservoirs and total body viral load in HIV-1 infection. Nature 387:183–98 Chun TW, Finzi D, Margolick J, Chadwick K, Schwartz D, Siliciano RF. 1995. In vivo fate of HIV-1-infected T cells: quantitative analysis of the transition to stable latency. Nature Med. 1:1284–90 Sanchez G, Xu X, Chermann JC, Hirsch I. 1997. Accumulation of defective viral genomes in peripheral blood mononuclear cells of human immunodeficiency virus type 1-infected individuals. J. Virol. 71:2233–40 Brinchmann JE, Albert J, Vartdal F. 1991. Few infected CD4+ T cells but a high proportion of replication-competent provirus copies in asymptomatic human immuno-

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

QC: PSA

Annual Reviews

AR078-20

POPULATION BIOLOGY OF HIV-1 INFECTION

69.

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

deficiency virus type 1 infection. J. Virol. 65:2019–23 Drinker CK, Yoffey JM. 1941. Lymphatics, lymph, and lymphoid tissue: their physiological and clinical significance. In Lymphatics, Lymph, and Lymphoid Tissue, pp. 11–18. Cambridge, MA: Harvard Univ. Press Chun TW, Engel D, Berrey MM, Shea T, Corey L, Fauci AS. 1998. Early establishment of a pool of latently infected, resting CD4+ T cells during primary HIV-1 infection. Proc. Natl. Acad. Sci. USA 95: 8869–73 Haase AT, Stowring L, Harris JD, Traynor B, Ventura P, Peluso R, Brahic M. 1982. Visna DNA synthesis and the tempo of infection in vitro. Virology 119:399–410 Walker CM, Moody DJ, Stites DP, Levy JA. 1986. CD8+ lymphocytes can control HIV infection in vitro by suppressing virus replication. Science 234:1563–66 Pal R, Garzino-Demo A, Markham PD, Burns J, Brown M, Gallo RC, DeVico AL. 1997. Inhibition of HIV-1 infection by the β-chemokine MDC. Science 278:695–98 Borrow P, Lewicki H, Hahn BH, Shaw GM, Oldstone MBA. 1994. Virus-specific CD8+ cytotoxic T-lymphocyte activity associated with control of viremia in primary human immunodeficiency virus type I infection. J. Virol. 68:6103–10 Kannagi M, Masuda T, Hattori T, Kanoh T, Nasu K, Yamamoto N, Harada S.1990. Interference with human immunodeficiency (HIV) replication by CD8+ T cells in peripheral blood leukocytes of asymptomatic HIV carriers in vitro. J. Virol. 64:3399–406 Perelson AS, Neumann AU, Markowitz M, Leonard JM, Ho DD. 1996. HIV-1 dynamics in vivo: Virion clearance rate, infected cell life-span, and viral generation time. Science 271:1582–86 Perelson AS, Essunger P, Cao Y, Vesanen M, Hurley A, Saksela K, Markowitz M, Ho DD. 1997. Decay characteristics of HIV-1-infected compartments during combination therapy. Nature 387:188–91 Grossman Z, Feinberg MB, Paul WE. 1998. Multiple modes of cellular activation and virus transmission in HIV infection: a role for chronically and latently infected cells in sustaining viral replication. Proc. Natl. Acad. Sci. USA 95:6314–19 Cheynier R, Hanrichwark S, Hadida F, Pelletier E, Oksenhendler E, Autran B, Wain-Hobson S. 1994. HIV and T cell expansion in splenic white pulps is accompanied by infiltration of HIV-1 specific cytotoxic T lymphocytes. Cell 78:2705–14

653

80. Reinhart TA, Rogan MJ, Amedee AM, Murphey–Corb M, Rausch DM, Eiden LE, Haase AT. 1998. Tracking members of the simian immunodeficiency virus deltaB670 quasispecies population in vivo at single–cell resolution. J. Virol. 72:113–20 81. Spear GT, Sullivan BL, Landay AL, Lint TF. 1990. Neutralization of human immunodeficiency virus type 1 by complement occurs by viral lysis. J. Virol. 64: 5869–73 82. Smith JP, Kosco MH, Tew JG, Szakal AK. 1988. Thy-1 positive tangible body macrophages (TBM) in mouse lymph nodes. Anat. Rec. 222:380–90 83. Pomerantz RJ, Trono D, Feinberg MB, Baltimore D. 1990. Cells nonproductively infected with HIV-1 exhibit an aberrant pattern of viral RNA expression: a molecular model for latency. Cell 61:1271–76 84. Seshamma T, Bagasra O, Trono D, Baltimore D, Pomerantz RJ. 1992. Blocked early-stage latency in the peripheral blood cells of certain individuals infected with human immunodeficiency virus type 1. Proc. Natl. Acad. Sci. USA 89:10663–67 85. Pomerantz RJ, Seshamma T, Trono D. 1992. Efficient replication of human immunodeficiency virus type 1 requires a threshold level of Rev: potential implications for latency. J. Virol. 66:1809–13 86. Adams M, Sharmeen L, Kimpton J, Romeo JM, Garcia JV, Peterlin BM, Groudine M, Emerman M. 1994. Cellular latency in human immunodeficiency virus–infected individuals with high CD4 levels can be detected by the presence of promoter-proximal transcripts. Proc. Natl. Acad. Sci. USA 91:3862–66 87. Lafeuillade A, Poggi C, Profizi N, Tamalet C, Costes O. 1996. Human immunodeficiency virus type 1 kinetics in lymph nodes compared with plasma. J. Infect. Dis. 174:404–7 88. Lafeuillade A, Poggi C, Profizi N, Tamalet C, Profizi N. 1997. Human immunodeficiency virus type 1 dynamics in different lymphoid tissue compartments. J. Infect. Dis. 176:804–6 89. Tamalet C, Lafeuillade A, Fantini J, Poggi C, Yahi N. 1997. Quantification of HIV-1 viral load in lymphoid and blood cells: assessment during four-drug combination therapy. AIDS 11:895–901 90. Wong JK, Gunthard HF, Havlir DV, Zhang JQ, Haase AT, Ignacio CC, Kwok S, Emini E, Richman DD. 1997. Reduction of HIV-1 in blood and lymph nodes following potent antiretroviral therapy and the virologic correlates of treatment failure.

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

654

QC: PSA

Annual Reviews

AR078-20

HAASE

Proc. Natl. Acad. Sci. USA 94:12574–79 91. Gunthard HF, Wong JK, Ignacio CC, Guatell JC, Riggs NL, Havlir DV, Richman DD. 1998. Human immunodeficiency virus replication and genotypic resistance in blood and lymph nodes after a year of potent antiretroviral therapy. J. Virol. 72:2422–28 92. Mackall CL, Hakim FT, Gress RE. 1997. T-cell regeneration: all repertoires are not created equal. Immunol. Today 18:245–51 93. Mackall CL, Gress RE. 1997. Thymic aging and T-cell regeneration. Immunol. Rev. 160:91–102 94. Mackall CL, Granger L, Sheard MA, Cepeda R, Gress RE. 1993. T-cell regeneration after bone marrow transplantation: differential CD45 isoform expression on thymic-derived versus thymic-independent progeny. Blood 82:2585–94 95. Steinmann GG. 1986. Changes in the human thymus during aging. Curr. Top. Pathol. 75:43–88 96. Prevot S, Audouis J, Andre-Bougaran J, Griffais R, Le Tourneau A, Fournier JG, Biebold J. 1992. Thymic pseudotumorous enlargement due to follicular hyperplasia in a human immunodeficiency virus sero–positive patient. Am. J. Clin. Pathol. 97:420–25 97. Burke AP, Anderson D, Benson W, Turnicky R, Mannan P, Liang YH, Smialek J, Virmani R. 1995. Localization of human immunodeficiency virus 1 RNA in thymic tissues from asymptomatic drug addicts. Arch. Pathol. Lab. Med. 119:36–41 98. Mackall CL, Fleisher TA, Brown MR, Andrich MP, Chen CC, Feuerstein IM, Horowitz ME, Magrath IT, Shad AT, Steinberg SM, Wexler LH, Gress RE. 1995. Age, thymopoiesis, and CD4+ T– lymphocyte regeneration after intensive chemotherapy. N. Engl. J. Med. 332:143– 49 99. Lindsey JW, Hodgkinson S, Mehta R, Mitchell D, Enzmann D, Steinman L. 1994. Repeated treatment with chimeric anti-CD4 antibody in multiple sclerosis. Ann. Neurol. 36:183–89 100. Mackall CL, Fleisher TA, Brown MR, Andrich MP, Chen CC, Feuerstein IM, Magrath IT, Wexler LH, Dimitrov DS, Gress RE. 1997. Distinctions between CD8+ and CD4+ T-cell regenerative pathways result in prolonged T-cell subset imbalance after intensive chemotherapy. Blood 89:3700–7 101. Mackall CL, Hakim FT, Gress RE. 1997. Restoration of T-cell homeostasis after Tcell depletion. Semin. in Immunol. 9:339– 46

102. Hakim FT, Cepeda R, Kaimei S, Mackall CL, McAtee N, Zujewski J, Cowan K. Gress RE. 1997. Constraints on CD4 recovery postchemotherapy in adults: thymic insufficiency and apoptotic decline of expanded peripheral CD4 cells. Blood 90:3789–98 103. Mackall CL, Bare CV, Granger LA, Sharrow SO, Titus JA, Gress RE. 1996. Thymic-independent T cell regeneration occurs via antigen-driven expansion of peripheral T cells resulting in a repertoire that is limited in diversity and prone to skewing. J. Immunol. 156:4609–16 104. Mackall CL, Gress RE. 1997. Pathways of T-cell regeneration in mice and humans: implications for bone marrow transplantation and immunotherapy. Immunol. Rev. 157:61–72 105. Bell EB, Sparshott SM. 1990. Interconversion of CD45R subsets of CD4 T cells in vivo. Nature 348:163–66 106. McLean AR, Michie CA. 1995. In vivo estimates of division and death rates of human T lymphocytes. Proc. Natl. Acad. Sci. USA 92:3707–11 107. Picker LJ, Treer JR, Ferguson-Darnell, Collins PA, Buck D, Terstappen LWMM. 1993. Control of lymphocyte recirculation in man. J. Immunol. 150:1105–21 108. Clegg CH, Rulffes JT, Wallace PM, Haugen HS. 1996. Regulation of an extrathymic T-cell development pathway by oncostatin M. Nature 384:261–63 109. Zheng B, Han S, Zhu Q, Goldsby R, Kelsoe G. 1996. Alternative pathways for the selection of antigen-specific peripheral T cells. Nature 384:263–66 110. Butcher EC, Picker LJ. 1996. Lymphocyte homing and homeostasis. Science 272:60–66 111. Freitas AA, Rocha BB. 1993. Lymphocyte lifespans: homeostasis, selection and competition. Immunol. Today 14:25–29 112. Kuo CT, Veselits ML, Leiden JM. 1997. LKLF: a transcriptional regulator of single-positive T cell quiescence and survival. Science 277:1986–90 113. Tanchot C, Lemonnier FA, Berarnau B, Freitas AA, Rocha B. 1997. Differential requirements for survival and proliferation of CD8 naive or memory T cells. Science 276:2057–62 114. Markiewicz MA, Girao C, Opferman JT, Sun J, Hu Q, Agulnik AA, Bishop CE, Thompson CB, Ashton-Rickardt PG. 1998. Long–term T cell memory requires the surface expression of selfpeptide/major histocompatibility complex molecules. Proc. Natl. Acad. Sci. USA 95:3065–70

P1: APR/ARY

March 3, 1999

P2: PSA/SPD

10:10

QC: PSA

Annual Reviews

AR078-20

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POPULATION BIOLOGY OF HIV-1 INFECTION 115. Rocha B, Dautigny N, Pereira Pablo. 1989. Peripheral T lymphocytes: expansion potential and homeostatic regulation of pool sizes and CD4/CD8 ratios in vivo. Eur. J. Immunol. 19:905–11 116. Westermann J, Pabst R. 1990. Lymphocyte subsets in the blood: a diagnostic window on the lymphoid system? Immunol. Today 11:406–10 117. Pabst R. 1988. The spleen in lymphocyte migration. Immunol. Today 9:43–45 118. Fleury S, de Boer RJ, Rizzardi GP, Wolthers KC, Otto SA, Welbon CC, Graziosi C, Knabenhans C, Soudeyns H, Bart PA, Gallant S, Corpataux JM, Gillet M, Meylan P, Schnyder P, Meuwly JY, Spreen W, Glauser MP, Miedema F, Pantaleo G. 1998. Limited CD4+ T cell renewal in early HIV-1 infection: effect of highly active antiretroviral therapy. Nature Med. 4:794–801 119. Trepel F. 1974. Number and distribution of lymphocytes in man. A critical mass. Klin. Wschr. 52:511–15 ˜ 120. Chou CC, Gudeman V, OORourke S, Isacescu V, Detels R, Williams GJ, Mitsuyasu, Giorgi JV. 1994. Phenotypically defined memory CD4+ cells are not selectively decreased in chronic HIV disease. J. Acquir. Immune Defic. Syndr. 7: 665–75 121. Roederer M, Dubs JG, Anderson MT, Raju PA, Herzenberg LA, Herzenberg LA. 1995. CD8 naive T cell counts decrease progressively in HIV–infected adults. J. Clin. Invest. 95:2061–66 122. Janossy G, Bofill M, Rowe D, Muir J, Beverley PCL. 1989. The tissue distribution of T lymphocytes expressing different CD45 polypeptides. J. Immunol. 66: 517–25 123. Collins JA, Schandl CA, Young KK, Vesely J, Willingham MC. 1997. Major DNA fragmentation is a late event in apoptosis. J. Histochem. Cytochem. 45:923–34 124. Rosok BI, Bostad L, Voltersvik P, Bjerknes R, Olofsson J, Asjo B, Brinchmann JE. 1996. Reduced CD4 cell counts in blood do not reflect CD4 cell depletion in tonsillar tissue in asymptomatic HIV-1 infection. AIDS 10:F35–F38 125. Rosenberg YJ, Zack PM, White BD, Papermaster SF, Elkins WR, Eddy GA, Lewis MG. 1993. Decline in the CD4+ lymphocyte population in the blood of SIV-infected macaques is not reflected in lymph nodes. AIDS Res. Hum. Retroviruses 9:639–46 126. Mosier DE. 1995. HIV results in the frame. CD4+ cell turnover. Nature 375: 193–94

655

127. Sprent J, Tough D. 1995. HIV results in the frame. CD4+ cell turnover. Nature 375:194 128. Grossman Z, Herberman RB. 1997. T-cell homeostasis in HIV infection is neither failing nor blind: modified cell counts reflect an adaptive response of the host. Nature Med. 3:486–90 129. Rosenberg YJ, Anderson AO, Pabst R. 1998. HIV-induced decline in blood CD4/CD8 ratios: viral killing or altered lymphocyte trafficking? Immunol. Today 19:10–17 130. Pakker NG, Notermans DW, de Boer RJ, Roos MT, de Wolf F, Hill A, Leonard JM, Danner SA, Miedema F, Schellekens PT. 1998. Biphasic kinetics of peripheral blood T cells after triple combination therapy in HIV-1 infection: a composite of redistribution and proliferation. Nature Med. 4:145–46 131. Lafeuillade A, Chouraqui M, Hittinger G, Poggi C, Delbeke E. 1997. Lymph node expansion of CD4+ lymphocytes during antiretroviral therapy. J. Infect. Dis. 176:1378–82 132. Sachsenberg N, Perelson AS, Yerly S, Schockmel GA, Leduc D, Hirschel B, Perrin L. 1998. Turnover of CD4+ and CD8+ T lymphocytes in HIV-1 infection as measured by Ki–67 antigen. J. Exp. Med. 187:1295–1303 133. Lane HC, Depper JM, Greene WC, Whalen G, Waldmann TA, Fauci AS. 1985. Qualitative analysis of immune function in patients with the acquired immunodeficiency syndrome. Evidence for a selective defect in soluble antigen recognition. N. Engl. J. Med. 313:79–84 134. Rosenzweig M, DeMaria M, Harper DM, Friedrich S, Jain RK, Johnson RP. 1998. Increased rates of CD4+ and CD8+ T lymphocyte turnover in simian immunodeficiency virus–infected macaques. Proc. Natl. Acad. Sci. USA 95:6388–93 135. Mohri H, Bonhoeffer S, Monard S, Perelson AS, Ho DD. 1998. Rapid turnover of T lymphocytes in SIV–infected rhesus macaques. Science 279:1223–27 136. Wolthers KC, Wisman GBA, Otto SA, de Roda Husman AM, Schaft N, de Wolf F, Goudsmit J, Coutinho RA, van der Zee AG, Meyaard L, Miedema F. 1996. T cell telomere length in HIV-1 infection: no evidence for increased CD4+ T cell turnover. Science 274:1543–47 137. Palmer LD, Weng N, Levine BL, June CH, Lane C, Hodes RJ. 1997. Telomere length, telomerase activity, and replicative potential in HIV infection: analysis of CD4+ and CD8+ T cells from HIV–

P1: APR/ARY

P2: PSA/SPD

March 3, 1999

10:10

656

138.

139.

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

140.

141.

142.

143.

144.

145.

146.

QC: PSA

Annual Reviews

AR078-20

HAASE discordant monozygotic twins. J. Exp. Med. 185:1381–86 Muro-Cacho CA, Pantaleo G, Fauci AS. 1995. Analysis of apoptosis in lymph nodes of HIV-infected persons. J. Immunol. 154:5555–66 Banda NK, Bernier J, Kurahara DK, Kurrle R, Haigwood N, Sekaly RP, Finkel TH. 1992. Crosslinking CD4 by human immunodeficiency virus gp120 primes T cells for activation-induced apoptosis. J. Exp. Med. 176:1099–106 Sunila I, Vaccarezza M, Pantaleo G, Fauci AS, Orenstein JM. 1997. gp120 is present on the plasma membrane of apoptotic CD4 cells prepared from lymph nodes of HIV-1-infected individuals: an immunoelectron microscopic study. AIDS 11:27– 32 Finkel TH, Tudor–Williams G, Banda NK, Cotton MF, Curiel T, Monks C, Baba TW, Ruprecht RM, Kupfer A. 1995. Apoptosis occurs predominantly in bystander cells and not in productively infected cells of HIV- and SIV-infected lymph nodes. Nature Med. 1:129–34 Connors M, Kovacs JA, Krevat S, Gea– Banacloche JC, Sneller MC, Flanigan M, Metcalf JA, Walker RF, Falloon J, Baseler M, Stevens R, Feuerstein I, Masur H, Lane HC. 1997. HIV infection induces changes in CD4+ T-cell phenotype and depletions within the CD4+ T-cell repertoire that are not immediately restored by antiviral or immune–based therapies. Nature Med. 3:703–4 Gorochov G, Neumann AU, Kereveur A, Parizot C, Li T, Katlama C, Karmochkine M, Raguin G, Autran B, Debre P. 1998. Perturbation of CD4+ and CD8+ T-cell repertoires during progression to AIDS and regulation of the CD4+ repertoire during antiviral therapy. Nature Med. 4:215–21 McCune JM, Loftus R, Schmidt DK, Carroll P, Webster D, Swor-Yim LB, Francis IR, Gross BH, Grant RM. 1998. High prevalence of thymic tissue in adults with human immunodeficiency virus–1 infection. J. Clin. Invest. 101:2301–8 Schuurman HJ, Krone WJA, Broekhuizen R, van Baarlen J, van Veen P, Goldstein AL, Huber J, Goudsmit J. 1989. The thymus in acquired immune deficiency syndrome. Am. J. Pathol. 134:1329–38 Withers-Ward ES, Amado RG, Koka PS, Jamieson BD, Kaplan AH, Chen ISY, Zack JA. 1997. Transient renewal of thymopoiesis in HIV-infected human thymic implants following antiviral therapy. Nature Med. 3:1102–9

147. Linder J. 1987. The thymus gland in secondary immunodeficiency. Arch. Pathol. Lab. Med. 111:1118–22 148. Seemayer TA, Laroche AC, Russo P, Malebranche R, Arnoux E, Guerin JM, Pierre G, Dupuy JM, Gartner JG, Lapp WS, Spira TJ, Elie R. 1984. Precocious thymic involution manifest by epithelial injury in the acquired immune deficiency syndrome. Hum. Pathol. 15:469–74 149. Grody WW, Fligiel S, Naeim F. 1985. Thymus involution in the acquired immune deficiency syndrome. Am. J. Clin. Pathol. 84:85–95 150. Heitger A, Neu N, Kern H, PanzerGrumayer ER, Greinix H, Nachbaur D, Niederwieser D, Fink FM. 1997. Essential role of the thymus to reconstitute naive (CD45RA+) T-helper cells after human allogeneic bone marrow transplantation. Blood 90:850–57 151. Bonyhadi ML, Rabin L, Salimi S, Brown DA, Kosek J, McCune JM, Kaneshima H. 1993. HIV induces thymus depletion in vivo. Nature 363:728–32 152. Jenkins M, Hanley MB, Moreno MB, Wieder E, McCune JM. 1998. Human immunodeficiency virus–1 infection interrupts thymopoiesis and multilineage hematopoiesis in vivo. Blood 91:2672– 78 153. Zauli G, Re MC, Visani G, Furlini G, Mazza P, Vignoli M, La Placa M. 1992. Evidence for a human immunodeficiency virus type 1-mediated suppression of uninfected hematopoietic (CD34+) cells in AIDS patients. J. Infect. Dis. 166:710– 16 154. Koka PS, Fraser JK, Bryson Y, Bristol GC, Aldrovandi GM, Daar ES, Zack JA. 1998. Human immunodeficiency virus inhibits multilineage hematopoiesis in vivo. J. Virol. 72:5121–27 155. Davis BR, Schwartz DH, Marx JC, Johnson CE, Berry JM, Lyding J, Merigan TC, Zander A. 1991. Absent or rare human immunodeficiency virus infection of bone marrow stem/progenitor cells in vivo. J. Virol. 65:1985–90 156. Walthers HG, Schuitemaker H, Miedema F. 1998. Rapid CD4+ T cell turnover in HIV-1 infection: a paradigm revisited. Immunol. Today 19:44–48 157. Carroll L. 1871. Through the Looking– Glass. New York: MacMillan 158. Mackall CL, Punt JA, Morgan P, Farr AG, Gress RE. 1998. Thymic function in young/old chimeras: substantial thymic T cell regenerative capacity despite irreversible age–associated thymic involution. 28:1886–93

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:625-656. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:657–700 c 1999 by Annual Reviews. All rights reserved Copyright °

CHEMOKINE RECEPTORS AS HIV-1 CORECEPTORS: Roles in Viral Entry, Tropism, and Disease Edward A. Berger1, Philip M. Murphy2 and Joshua M. Farber3 1Laboratory

of Viral Diseases, 2Laboratory of Host Defenses and 3Laboratory of Clinical Investigation, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland 20892; e-mail: Edward [email protected]

KEY WORDS:

AIDS, genetics, G protein-coupled receptors, pathogenesis, therapeutics

ABSTRACT In addition to CD4, the human immunodeficiency virus (HIV) requires a coreceptor for entry into target cells. The chemokine receptors CXCR4 and CCR5, members of the G protein-coupled receptor superfamily, have been identified as the principal coreceptors for T cell line-tropic and macrophage-tropic HIV-1 isolates, respectively. The updated coreceptor repertoire includes numerous members, mostly chemokine receptors and related orphans. These discoveries provide a new framework for understanding critical features of the basic biology of HIV-1, including the selective tropism of individual viral variants for different CD4+ target cells and the membrane fusion mechanism governing virus entry. The coreceptors also provide molecular perspectives on central puzzles of HIV-1 disease, including the selective transmission of macrophage-tropic variants, the appearance of T cell line-tropic variants in many infected persons during progression to AIDS, and differing susceptibilities of individuals to infection and disease progression. Genetic findings have yielded major insights into the in vivo roles of individual coreceptors and their ligands; of particular importance is the discovery of an inactivating mutation in the CCR5 gene which, in homozygous form, confers strong resistance to HIV-1 infection. Beyond providing new perspectives on fundamental aspects of HIV-1 transmission and pathogenesis, the coreceptors suggest new avenues for developing novel therapeutic and preventative strategies to combat the AIDS epidemic.

657 0732-0582/99/0410-0657$08.00

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

658

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INTRODUCTION The discovery that specific chemokine receptors function together with CD4 as coreceptors for the human immunodeficiency virus (HIV) emerged from two seemingly distinct puzzles in HIV-1 research: the specificity of HIV-1 entry into different target cell types and the control of HIV-1 infection by soluble suppressor factors. The molecular solutions to these problems led to a convergence of research on a viral pathogen that devastates the human immune system and on the chemokine regulatory network that orchestrates the immune system’s response to pathogen invasion. Within a remarkably short period the coreceptor discoveries have engendered new perspectives on the major problems of HIV-1 disease, including the mechanisms of HIV-1 transmission and disease progression, and possibly new modes of therapeutic intervention. This review focuses on coreceptors used by HIV type 1 (HIV-1 ), the major cause of AIDS worldwide and the subject of the research that led to the original coreceptor discoveries. HIV-2 and simian immunodeficiency virus have also been shown to use chemokine receptors and related orphans as coreceptors; for these subjects the reader is referred to recent review articles (1, 2).

IDENTIFICATION OF CHEMOKINE RECEPTORS AS HIV-1 CORECEPTORS Historical Perspective: HIV-1 Tropism Suggests Existence of Coreceptors HIV-1 enters target cells by direct fusion of the viral and target cell membranes. The fusion reaction is mediated by the viral envelope glycoprotein (Env), which binds with high affinity to CD4, the primary receptor on the target cell surface (reviewed in 3). By a similar or identical mechanism, cells expressing Env can fuse with CD4+ target cells, sometimes leading to the formation of giant cells (syncytia). The notion that a coreceptor is required for HIV-1 entry stemmed from the awareness that CD4 expression is not sufficient to explain HIV-1 tropism for different target cells in vitro (see 4 for review and citations). Two related phenomena led to this conclusion. The first series of findings, initially reported in the mid-1980s and extended through the early 1990s, was based on curious results with recombinant human CD4. The receptor was found to render cells permissive for Env-mediated fusion/entry/infection, but only when expressed on a human cell type. Experiments with cell hybrids supported the conclusion that this restriction was due to the requirement for a cofactor (coreceptor) of unknown identity that is specific to human cells, rather than to the presence of a fusion inhibitor on the nonhuman cells.

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIV-1 CORECEPTORS

659

The second phenomenon concerned the distinct tropisms of different HIV-1 isolates for various CD4+ human target cell types in vitro. All HIV-1 strains infect and replicate in activated primary CD4+ T lymphocytes; the tropism distinctions emerge when other target cells are examined. Some isolates show efficient infectivity for continuous CD4+ T cell lines, but poor infectivity for primary macrophages; this phenotype was originally observed with isolates that had been adapted in the laboratory to replicate in T cell lines and was subsequently observed with some clinical isolates. Such viruses are designated T cell line–tropic (TCL-tropic); they are generally syncytium-inducing in assays using a highly permissive T cell line. Other HIV-1 strains show the opposite preference, infecting primary macrophages much more efficiently than continuous T cell lines; these are designated macrophage-tropic (M-tropic) or nonsyncytium-inducing. Isolates that replicate efficiently in both target cell types are designated dual-tropic. These in vitro tropism phenotypes were first revealed in the late 1980s and were soon demonstrated to have profound implications for HIV-1 transmission and pathogenesis (see 5 for original citations). The viral isolates obtained from peripheral blood of individuals shortly after infection and during the asymptomatic phase are predominantly M-tropic; as the infection progresses to AIDS, TCL-tropic viruses can be isolated from many (but not all) patients. TCLtropic strains typically display higher cytopathic effects in vitro, suggesting that they may have a particularly important role in the decline of CD4+ T cells in vivo, which is the hallmark of AIDS. In the early to mid 1990s, TCL- versus M-tropism was shown to result primarily from the fusion specificities of the corresponding Envs; again, experiments with cell hybrids revealed that these specificities derived from the requirement for an additional factor (coreceptor) in the permissive cell type rather than from an inhibitor in the nonpermissive type. In the resulting model, TCL- versus M-tropism of different HIV-1 isolates was postulated to reflect the preferential fusogenic activity of the corresponding Envs for distinct coreceptors that are differentially expressed on these target cell types. Thus, by the mid-1990s, it was clear that the key to the HIV-1 entry/tropism problem rested on identification of these coreceptors. This was finally achieved in 1996.

Identification of CXCR4 and CCR5 as HIV-1 Coreceptors CORECEPTOR FOR TCL-TROPIC HIV-1 The first HIV-1 coreceptor was identified using an unbiased functional cDNA cloning strategy based on the ability of a cDNA library to render a CD4-expressing murine cell permissive for fusion with cells expressing Env from a TCL-adapted strain (6). A single cDNA was isolated, and sequence analysis indicated that the protein product is a member of

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

660

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

the superfamily of the seven transmembrane domain G protein–coupled receptors, the largest receptor superfamily in the human genome. G protein–coupled receptor superfamily. The cDNA previously had been isolated and sequenced by several laboratories investigating G protein–coupled receptors, but no ligands or functional activities had been found; the protein was thus considered an “orphan” receptor. Because of its new-found activity in HIV-1 Env-mediated fusion, the protein was named “fusin” (6). Its role as a coreceptor was based on both gain-of-function experiments demonstrating that coexpression of fusin along with CD4 rendered nonhuman cells permissive for Env-mediated cell fusion and infection, and loss-of-function experiments showing that anti-fusin antibodies potently inhibited fusion and infection of primary human CD4+ T lymphocytes. Most importantly, both types of analyses indicated that fusin functioned for TCL-tropic, but not M-tropic, HIV-1 strains. Fusin thus fit the criteria for the TCL-tropic HIV-1 coreceptor. Among known members of the G protein–coupled receptor superfamily, fusin has the strongest sequence homology with peptidergic receptors, including chemokine receptors. Chemokines are small proteins (∼70–90 amino acid residues) with chemotactic activity for leukocytes; they play prominent roles in leukocyte activation and trafficking to sites of inflammation (7). Receptors for chemokines (8) comprise a subfamily within the G protein–coupled receptors superfamily. All known human chemokines fit within four classes based on the cysteine motifs near the N-terminus. The two major classes are the CXC chemokines, in which the first two cysteines are separated by a single residue, and the CC chemokines, in which the first two cysteines are adjacent. Two minor classes have also been defined, each containing one known example: the C class with only a single cysteine residue in the motif and the CX3C class with three residues separating the first two cysteines. Nearly all the receptors are selective for one class of chemokines; however, there is redundancy in the system, as individual receptors can bind multiple chemokines within a class, and many chemokines can function with more than one receptor. CORECEPTOR FOR M-TROPIC HIV-1 The discovery of fusin, a putative chemokine receptor, as the coreceptor for TCL-tropic HIV-1 strains provided an impetus and direction for identifying the coreceptor for M-tropic isolates. The focus was narrowed to CC chemokine receptors by a link with an earlier study directed at a seemingly unrelated problem, namely the identity of soluble HIV-1 suppressor factor(s) released by CD8+ T lymphocytes. This phenomenon was first described in the late 1980s (reviewed in 9). Biochemical identification of the molecule(s) in question promised to shed light on the natural mechanisms controlling HIV-1 infection in vivo, and could potentially lead to new modes of prevention and treatment. The first success was achieved in 1995 with the demonstration that the CC chemokines RANTES, MIP-1α, and MIP-1β are

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIV-1 CORECEPTORS

661

major suppressive factors released by CD8+ T lymphocytes (10) (although additional factors may contribute to the entire CD8 soluble suppressor phenomenon; see 9). Particularly intriguing was the fact that these CC chemokines potently suppressed infection by M-tropic HIV-1 strains but had little effect on a TCL-tropic strain. Since fusin, the coreceptor for TCL tropic strains, was a putative chemokine receptor, an obvious mechanism for infection inhibition by CC chemokines was suggested: They bind to and block a chemokine receptor that functions as a coreceptor for M-tropic HIV-1. Fortuitously, at approximately the same time, a chemokine receptor was identified with precisely the corresponding specificity for RANTES, MIP-1α, and MIP-1β (11–13); it was designated CCR5, in keeping with the standard nomenclature system for chemokine receptors (fifth receptor for CC chemokines). Within the span of one week, five independent reports demonstrated that CCR5 is a major coreceptor for M-tropic HIV-1 strains (14–18); the evidence was based on both gain-of-function studies with recombinant CCR5, and loss-of-function studies using CCR5 chemokine ligands as blocking agents. FUSIN AS A CHEMOKINE RECEPTOR (CXCR4) The coreceptor discoveries were soon followed by the demonstration that fusin is indeed a chemokine receptor, specific for the functionally equivalent CXC chemokines SDF-1α and SDF1β (19, 20), which are formed by alternative splicing. Fusin was therefore renamed CXCR4 (fourth receptor for CXC chemokines). SDF-1 was shown to be a selective inhibitor of TCL-tropic HIV-1 strains. PRIMARY HIV-1 ISOLATES AND GENETIC DIVERSITY The initial descriptions of coreceptor activity for CXCR4 and CCR5 were made with well-characterized prototypic HIV-1 strains. In subsequent studies, virtually all primary HIV-1 isolates were found to use one or both coreceptors (21–25). Isolates from different geographic regions display little relationship between genetic subtype and coreceptor usage (21, 23, 25, 26), although some distinctions have been reported (27). Thus, viruses from all clades can use CCR5 and/or CXCR4, and a marked correlation is observed between biological phenotype (including tropism) and coreceptor usage. A SIMPLE MECHANISTIC MODEL FOR HIV-1 TROPISM The essential pieces of the tropism puzzle thus appeared to be in place. In a minimal model, the tropism of different HIV-1 strains can be explained by two considerations: the abilities of the corresponding Envs to use CXCR4 and/or CCR5, and the expression patterns of these coreceptors on different CD4+ target cells (Figure 1). Thus, TCL-tropic strains preferentially use CXCR4, M-tropic strains prefer CCR5, and dual-tropic strains can use both; continuous T cell lines abundantly express CXCR4, primary macrophages express CCR5, and primary T cells express both.

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

662

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

Figure 1 Model for coreceptor usage and HIV-1 tropism. TCL-tropic strains are specific for CXCR4 and can infect continuous CD4+ T cell lines and primary CD4+ T cells. M-tropic strains are specific for CCR5 and can infect primary macrophages and primary CD4+ T cells. Dual-tropic strains can use both CXCR4 and CCR5, and can infect continuous CD4+ T cell lines, macrophages, and primary T cells. A NEW HIV-1 NOMENCLATURE SYSTEM In keeping with this model, it has been suggested that the designation of HIV-1 phenotype be revised to indicate coreceptor usage, rather than the less biochemically defined characteristics of target cell tropism or syncytium-inducing properties (28). Accordingly, throughout this review, HIV-1 variants are designated as either X4 (CXCR4-specific, generally corresponding to TCL-tropic and syncytium-inducing), R5 (CCR5-specific, generally corresponding to M-tropic and nonsyncytium-inducing), or R5X4 (using both CCR5 and CXCR4, generally corresponding to dual-tropic). While this designation provides a useful framework, not all features of HIV-1 tropism are fully explained by differential usage of CXCR4 and CCR5 (see below).

The Expanding Coreceptor Repertoire OTHER CHEMOKINE RECEPTORS AND RELATED ORPHANS Additional complexity results from the findings that HIV-1 coreceptor activity is not limited to CXCR4 and CCR5. Studies with recombinant proteins have demonstrated coreceptor activity for several other human chemokine receptors and related orphans. Table 1 summarizes our current knowledge of the HIV-1 coreceptor repertoire and includes the chemokine receptors CCR2b (18), CCR3 (17, 18),

+d

?

?

?

c

+

+

+ + + +

?

?

? ? ? ?

? ? + ?

?

?

? ? ? ?

? ? ? ?

+

? ?

Genetic

HIV-1 CORECEPTORS

b

See text for references. PBL, peripheral blood lymphocytes; Mo/MDM, monocytes/monocyte-derived macrophages. +, expressed in cells or direct evidence obtained; ?, unknown, sometimes due to conflicting data. d Expressed only in CMV-infected cells.

a

+d

+

Other human receptors BLTR Leukotriene B4 Viral chemokine receptors US28 MIP-1α, MIP-1β, RANTES, MCP-1

? + + +

Human orphan receptors APJ ? ChemR23 ? GPR15/BOB ? STRL33/Bonzo ?

+ + + +

+

? +

Primary cells

Annual Reviews

? + ? ?

+ ? + +

+ ? + +

? ? + +

+

+ +

Recombinant protein

QC: KKK/uks

? + + ?

+

+

+

+ +

+ +

Other CD4+ cells

+c +

Mo/MDM

In vivo

P2: PSA/plb

Human chemokine receptors CCR2B MCP-1, -2, -3, -4 CCR3 Eotaxin-1, -2, RANTES MCP-2, -3, -4 CCR5 RANTES, MIP-1α, MIP-1β, MCP-2 CCR8 I-309 CCR9 CC chemokines CXCR4 SDF-1α, -1β Fractalkine CX3CR1

PBLb

In vitro

9:53

Ligands

Expression patterns

Evidence for coreceptor function

March 3, 1999

Coreceptors

Table 1 HIV-1 coreceptor repertoirea

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SKH/spd/vks T1: KKK

AR078-21

663

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

664

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

CCR8 (29–31), CCR9 (32), and CX3CR1 (formerly designated CMKBRL1 or V28) (29, 33); the chemokine receptor–like orphans STRL33/Bonzo (34, 35), GPR15/BOB (35, 36), and Apj (32, 37). Not all members of the human chemokine receptor family can function as HIV-1 coreceptors; absence of activity has been noted in most studies for CCR1, CCR4, and CCR6, for CXC chemokine receptors other than CXCR4, and for several other chemokine receptor–like orphans. In addition to these human proteins, HIV-1 coreceptor activity has been detected for US28, a CC chemokine receptor encoded by human cytomegalovirus (29, 38). Also interesting in this regard are the HIV-1 inhibitory effects of vMIP-I and vMIP-II, chemokine-like proteins encoded by Kaposi’s sarcoma–associated herpesvirus (39, 40). These findings highlight the potential interplay of concurrent viral infections during HIV-1 pathogenesis. DEVIATIONS FROM THE SIMPLE CHEMOKINE RECEPTOR/CORECEPTOR PARADIGM

The notion that coreceptor activity is restricted to chemokine receptors and related orphans has been challenged by the recent reports of coreceptor function for BLTR, the leukotriene B4 receptor (41), and for Chem R23 (42), an orphan receptor more closely related to complement and formyl peptide receptors than to chemokine receptors. While in most cases the HIV-1 inhibitory activity of chemokines can be attributed to their specific blocking of the corresponding coreceptors (10, 14– 17, 19, 20, 30, 33), at least one puzzling exception has been noted. The CC chemokine designated MDC (macrophage-derived chemokine) has been purified based on its ability to block both R5 and X4 HIV-1 replication in PBMCs (43, 44). This activity is controversial, since it has been reproduced by one group using synthetic MDC (45), but not by others using synthetic and recombinant forms of the protein (46, 47). The mechanism of MDC inhibition is undefined, since its only known receptor is CCR4 (48) for which coreceptor activity has not been detected; moreover the HIV-blocking activity of MDC is observed in PBMCs, which R5 and X4 viruses clearly infect via CCR5 and CXCR4, respectively. These results raise the possibility that the infectionblocking activity of MDC occurs by a mechanism other than simple blocking of an HIV-1 coreceptor.

Major Questions About Tropism and the Coreceptor Repertoire The breadth of the HIV-1 coreceptor repertoire poses important interrelated questions: 1. How well does CXCR4 versus CCR5 usage account for the selectivity of HIV-1 for different CD4+ target cell types?

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

HIV-1 CORECEPTORS

665

2. What are the relative efficiencies of the various coreceptors? 3. Is the activity of one coreceptor influenced by the presence of others on the same target cell? 4. How is the coreceptor repertoire used by diverse HIV-1 isolates?

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

5. Which coreceptors are significant on relevant human target cells? 6. Which coreceptors are significant for HIV-1 transmission and disease progression? The failure of CD4-expressing nonhuman cell lines to allow HIV-1 entry/fusion/infection is clearly explained by the coreceptors. Murine CXCR4 can function for X4 strains, but it is not expressed on most murine cell lines (6, 49–51). Murine CCR5 does not display coreceptor activity (52–55). However, ambiguities arise regarding the tropism of individual HIV-1 isolates for different CD4+ human target cells. The situation is relatively clear with infection of continuous T cell lines. These targets generally express abundant levels of CXCR4 and only rarely CCR5; they are susceptible to infection by X4 and R5X4 strains. The macrophage problem is much more complex. Consistent with the model, CCR5 expression can account for the susceptibility of these cells to most R5 isolates (16, 56–60). However, CCR5 usage does not appear to be sufficient for macrophage infection, based on the isolation of HIV-1 strains that can use CCR5 yet fail to infect macrophages (61). Conflicting results have also been obtained regarding the effects of CCR5 chemokine ligands in macrophages, with some groups reporting the expected inhibition of fusion/entry/infection by R5 viruses (16, 62, 63), and others not observing this effect (15, 64, 65). One proposed explanation is that HIV-1 inhibition by CC chemokines is facilitated by chemokine interaction with cell surface heparan sulfate, and that this interaction occurs minimally with macrophages (66). To further complicate the matter, the effect of CC chemokines on HIV-1 replication in macrophages is reportedly highly dependent on the time of addition relative to virus, with stimulation occurring when they are added before infection, and inhibition occurring when they are added during or after infection (67). Perhaps even more puzzling is the resistance of macrophages to infection by some CXCR4-using strains, since several groups have demonstrated CXCR4 expression on macrophages and its functionality as a coreceptor for some HIV-1 isolates (25, 65, 68–71). As yet there is no simple resolution of these dilemmas, but several confounding issues have been noted, including variations in macrophage isolation and culture methods, donor-dependent macrophage differences, temporal modulation of CXCR4 levels during cell culture, possible

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

666

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

differences in coreceptor display on different cell types, varying sensitivities of different assay systems for coreceptor function, chemokine effects at multiple stages in the HIV-1 replication cycle, possible coreceptor-dependent post-entry effects on viral replication, and the fact that the resistance of macrophages to T cell line–adapted isolates is not absolute. A major in vitro criterion for identifying members of the HIV-1 coreceptor repertoire is the ability of the recombinant proteins to confer HIV-1 permissiveness to a CD4-expressing target cell. However, numerous experimental variables have confounded efforts to assess the relative activities of each coreceptor. For one, the recombinant expression efficiencies can vary widely between different coreceptors. Thus, in several studies CCR3 has been reported as a minor coreceptor, based on the relatively inefficient activity observed with recombinant CCR3 compared to CCR5, and the limited number of HIV-1 isolates that could function with CCR3 (14, 15, 17, 18, 21–23). However, these findings were made under conditions in which recombinant CCR3 expression either was not monitored or was found to be very low compared to recombinant CCR5; in subsequent studies where CCR3 was expressed more efficiently, this coreceptor demonstrated activity comparable to CCR5 and/or CXCR4 and functioned with a broad range of isolates (25, 29). A related problem is the diversity of assay systems used to evaluate coreceptor activity (virus entry, Env-mediated cell fusion, productive infection); it is questionable whether the quantitative readout is proportional to the number of available coreceptor molecules in any of these assays, and whether the limiting molecular determinants are the same in each. Complexities associated with the varying dependencies of different HIV-1 strains on levels of both CD4 and coreceptor have been noted (72). These findings raise the issue of the “relevant” levels of coreceptor, which would seem to be the amounts that are endogenously expressed on natural human CD4+ target cells. The reagents necessary to obtain such quantitative information (e.g. monoclonal antibodies, labeled chemokine ligands) have been used for CXCR4, CCR5, and CCR3, but less so for the other coreceptors, particularly the orphans. Another important in vitro criterion in demonstrating the coreceptor activity of CXCR4 and CCR5 is based on the ability of coreceptor-specific blocking reagents to inhibit HIV-1 in natural human CD4+ target cells (10, 14–16, 19, 20). In only a few instances has this been achieved for other coreceptors, e.g. for CCR3 in microglia (73) and in monocyte-derived dendritic cells (74). Again these approaches have been hampered by the limited availability of specific coreceptor-blocking agents. Moreover, the possible presence of multiple endogenous coreceptors on a natural target cell type complicates attempts to determine the relative importance of each. Also critical for assessing the potential significance of an individual coreceptor is its expression pattern on diverse CD4-positive cells. It is possible

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

HIV-1 CORECEPTORS

667

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

that virus replication in a specialized tissue compartment might select for viral variants, with enhanced usage of a coreceptor that is preferentially enriched in that compartment. Focus must therefore be extended to HIV-1 isolates from various specialized tissue compartments (e.g. thymus, genital tract, placenta). The most powerful evidence for the significance of a given coreceptor is based on correlation of genetic variation in coreceptors with HIV-1 disease. Such analyses have provided convincing evidence of a central role for CCR5 in vivo. These findings are detailed in later sections.

CORECEPTORS IN THE HIV-1 FUSION/ENTRY MECHANISM Model for Coreceptor Function in Fusion/Entry ENV STRUCTURE The HIV-1 Env consists of two noncovalently associated subunits generated by cleavage of the gp160 precursor: (a) the heavily glycosylated external gp120 subunit, derived from the N-terminal portion of gp160 and containing the CD4 binding site, and (b) the membrane-spanning gp41 subunit, derived from the C-terminal portion of the precursor and containing at its N-terminus a hydrophobic fusion peptide, which is directly involved in membrane fusion (3). Native Env expressed on the surface of the virion or the infected cell is a trimeric structure containing three gp120/gp41 complexes associated via noncovalent interactions within gp41. SEQUENTIAL CONFORMATIONAL CHANGES Env can be envisioned as a fusogenic machine, catalyzing direct pH-independent fusion between the virion membrane displaying Env and the target membrane displaying CD4 plus coreceptor. It seems logical that Env is activated only at the right time and place, i.e. when the virion encounters the target cell. This suggests that Env might be activated by receptor binding; indeed, numerous studies have demonstrated CD4-induced changes in Env conformation (3). The coreceptors represent new players in the fusion process, leading to more complex models involving multiple, and probably sequential, protein-protein interactions and conformational changes. In the most favored model (Figure 2), CD4 binding induces conformational change(s) in gp120 that exposes, creates, or stabilizes the coreceptor-binding determinants; the gp120 interaction with the seven transmembrane domain coreceptor then induces a further conformational change(s) in Env that results in activation of gp41, presumably by exposing and extending its fusion peptide so that it can insert into the plasma membrane of the target cell. Several lines of experimental evidence support this model of receptor-induced conformational changes: (a) Env-mediated fusion generally requires the presence of

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

668

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

Figure 2 Model for coreceptor usage in HIV-1 entry. Upon binding to CD4, gp120 undergoes a conformational change that exposes, creates, or stabilizes the coreceptor-binding determinants. The interaction of gp120 with coreceptor triggers Env to undergo another conformational change, leading to extension of gp41 and insertion of the fusion peptide into the target cell membrane. Only a monomer of gp120/gp41 is shown for simplicity, though the native Env is a trimer.

CD4 as well as coreceptor on the target cell (6, 14–18); (b) complexes between CD4, gp120, and coreceptor have been isolated (75); associations between coreceptor and CD4 are enhanced in the presence of gp120, though they apparently occur to some extent in its absence (75–78); (c) soluble gp120 binds weakly to coreceptor expressed on cells, but the affinity is greatly increased upon CD4 binding (79–82); (d ) treatment of Env-expressing cells with soluble CD4 activates them to fuse with coreceptor-positive CD4-negative target cells (K Salzwedel, ED Smith, EA Berger, unpublished); (e) soluble CD4 treatment renders fusion/entry/infection more susceptible to inhibition by monoclonal antibodies directed against the coreceptor-binding determinants of gp120 (83; also K Salzwedel, ED Smith, EA Berger, unpublished); ( f ) X-ray crystallographic structure analysis of a ternary complex containing gp120 bound to CD4 and a monoclonal antibody against the coreceptor-binding region (84), combined with mutational (85) and antigenic (86) analyses, suggests that the gp120 captured in the bound state has undergone dramatic conformational changes and that both the CD4 binding site and the coreceptor-interacting regions probably exist in very different conformations prior to CD4 binding. FUSION INDEPENDENCE FROM CORECEPTOR SIGNALING AND INTERNALIZATION

The effector functions of chemokines are mediated by coupling of their receptors to complex intracellular signaling pathways mediated by G proteins. However, Env-mediated fusion can occur without such signaling or coreceptor internalization (52, 53, 87–91). Moreover, inhibition of fusion/entry/infection by chemokine ligands of the coreceptors does not require receptor activation/signaling and occurs by at least two mechanisms: downmodulation of the coreceptor and direct blocking of the Env/coreceptor interaction (26, 89, 92–96). CD4-INDEPENDENT CORECEPTOR USAGE The model described above accounts for the major features of the CD4-dependent mechanism, which is undoubtedly

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIV-1 CORECEPTORS

669

the major route of HIV-1 fusion/entry/infection. However, in certain cases, Env/coreceptor interactions have been reported in the absence of CD4. These phenomena, which have been observed in assays of both Env-mediated fusion/entry/infection and gp120/coreceptor binding, were first described with HIV-2 (97–100) and then with SIV (99, 101, 102) and HIV-1 (82, 103, 104). It has been suggested that the gp120 molecules displaying CD4-independent interaction with coreceptor have already undergone (at least partially) the conformational changes normally induced by CD4. These CD4-independent entry pathways are relatively inefficient; moreover, Envs capable of mediating CD4-independent interaction with coreceptor retain the ability to bind CD4, and CD4 binding still markedly enhances functional interaction between gp120 and coreceptor. The significance of CD4-independent Env/coreceptor interactions in vivo is thus questionable. However, an intriguing hypothesis deduced from these findings is that the evolutionary predecessor of HIV-1 strictly used the chemokine receptors for entry, and that the CD4 requirement evolved later as a means of conferring greater target cell specificity as well as protecting the coreceptor-binding region from the humoral immune system. According to this notion, the designation of CD4 as the primary receptor and the chemokine receptor as the coreceptor is a reflection of both the historical sequence of their discoveries and the kinetic sequence by which they function. These designations do not imply that the chemokine receptor plays only a secondary role; to the contrary, the coreceptor is believed to be essential for triggering the fusogenic activity of Env, and may represent the primordial receptor for primate immunodeficiency retroviruses.

Structural Correlates of the Env-Coreceptor Interaction The gp120/coreceptor interaction is extremely intricate, involving multiple discontinuous regions on each protein. The gp120 molecule contains five variable loops designated V1-V5, interspersed with five relatively conserved regions designated C1-C5 (3). This framework has provided a basis for defining coreceptor interaction sites on gp120 (17, 24, 84–86, 105–114). In keeping with the long-appreciated role of the V3 loop as a critical determinant of Env fusogenicity and tropism, this region has a major role in gp120’s activity and specificity for coreceptor binding. In particular, basic residues at fixed positions on either side of the conserved tip determine usage of CXCR4. V3 is critical for gp120 binding to coreceptor; in the fusion process, V3 functions not alone but rather in concert with other gp120 regions including V1, V2, and C4. Recent X-ray crystallographic structural determinations (84) coupled with mutagenic (85) and antigenic (86) analyses have led to a model in which coreceptor interacts with the V3 loop and a conserved “bridging sheet” composed of the V1/V2 stem and an antiparallel,

DETERMINANTS ON gp120

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

670

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

four-stranded structure including sequences in the C4 region. As noted above, the coreceptor binding site appears to be exposed, created, or stabilized upon CD4 binding. DETERMINANTS ON CORECEPTORS The coreceptor is topologically arranged with seven transmembrane segments, the N-terminus and three loops extracellular, and the C-terminus and three loops intracellular (Figure 2). Regions of coreceptor involved in Env interaction have been studied by analyzing chemokine receptor chimeras and site-directed mutants, comparing chemokine receptor homologues from different species, and assessing the blocking activities of chemokine ligands and anti-coreceptor antibodies (51–55, 68, 99, 106, 115– 131). The results are exceedingly complex and, in some cases, contradictory. The extracellular regions of the coreceptors have been the focus of most studies, with the assumption that Env probably makes initial direct contacts with these regions. However, the transmembrane and/or cytoplasmic regions of the coreceptor also critically influence activity, perhaps by affecting display of the extracellular regions. Each extracellular region has been implicated in coreceptor function, with several studies suggesting a particularly important role for the N-terminal segment. Interestingly, the effects of a particular coreceptor modification or anti-coreceptor agent can vary markedly for different Envs; in many cases the activities cluster with the class of Env (R5, X4, or R5X4). It has also been suggested that evolution of the viral quasispecies within the infected host (from R5 to R5X4 or X4, see below) is associated with changes in those extracellular regions of coreceptor that are most critical.

Major Questions About Coreceptors in the HIV-1 Fusion/Entry Mechanism The coreceptors provide a new focus for mechanistic questions about Envmediated fusion: 1. What are the precise determinants of gp120 and coreceptor involved in intermolecular contacts? 2. How do the interacting determinants vary with different isolates using the same receptor, and a given isolate using different coreceptors? 3. What are the precise conformational changes in Env induced by interaction with CD4 and coreceptor? 4. Does the coreceptor undergo essential conformational changes upon interaction with Env? 5. Do molecular interactions between CD4 and coreceptor have significance for fusion/entry?

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

HIV-1 CORECEPTORS

671

6. How does Env interaction with CD4 and coreceptor function in the context of the trimeric Env structure?

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

7. Are there additional target cell molecular components involved in fusion/ entry? A critical focus for future work will involve detailed characterization of the intricate conformational changes associated with the mechanism described above. The recent success in determining the X-ray crystallographic structure of a core fragment of gp120 complexed to CD4 and a Fab against the coreceptor binding region (84), coupled with the X-ray (132, 133) and NMR (134) structural determinations of gp41, provide a framework for future analyses. Particularly critical are the structures of gp120 prior to CD4 binding, and complexed to coreceptor; the latter problem will undoubtedly be extremely difficult in view of the challenges associated with crystallization of proteins containing multiple membrane-spanning regions. In considering the regions on gp120 and coreceptor involved in fusion, it is particularly puzzling that many Envs can function with a wide variety of coreceptors that have minimal sequence homology. For example, the dualtropic 89.6 strain can use CCR5, CCR2, CCR3, CCR8, CXCR4, CX3CR1, and STRL33/Bonzo (18, 29, 33, 34); several R5 strains (e.g. Ba-L, JR-FL), though they cannot use CXCR4, can function quite efficiently with other coreceptors such as CCR3 (25, 29). Thus, in spite of the findings that individual amino acid substitutions in both Env and coreceptor can dramatically affect activity, the interactions may involve structural features not revealed by simple considerations of amino acid sequence. Also of interest is the possibility that gp120 binding might induce conformational changes in the coreceptor, which in turn trigger Env to initiate the final step in the fusion mechanism. Finally, little is known about how the CD4 and coreceptor interactions occur in the context of the trimeric (and possibly higher ordered) Env structure. Preliminary information indicates that an individual gp120 subunit must interact with both CD4 and coreceptor; these interactions can then promote activation of gp41 subunits which can be on other members of the oligomeric complex (K Salzwedel, EA Berger, unpublished).

CORECEPTORS IN HIV-1 DISEASE Each member of the HIV-1 coreceptor repertoire was discovered using in vitro model systems; thus far there is limited information on which ones actually play a role in vivo in HIV-1 pathogenesis. Coreceptor parameters relevant to this question include the range of viral isolates recognized, the pattern of expression on target cells and tissues, the ability to mediate infection of primary cells,

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

672

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

and allelic polymorphisms associated with altered clinical outcome. By these criteria, the strongest evidence for a role in HIV-1 disease is for CCR5.

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Evolution of Viral Tropism in Disease The coreceptors put a molecular face on one of the most puzzling phenomena of HIV-1 disease, namely the evolution of viral tropism in vivo in a specific pattern (reviewed in 5). M-tropic variants are found early after infection and throughout all stages of disease; TCL-tropic variants are detected in many infected persons, but only at late disease stages (Figure 3). The failure to find TCLtropic variants in the newly infected individual occurs even when such variants are present in the transmitting source; moreover this pattern is observed whether transmission is by mucosal or parenteral routes. Following the coreceptor discoveries, it was determined that R5 viruses predominate at early stages, with X4 and R5X4 variants appearing at late stages. In a longitudinal study of vertical HIV-1 transmission, disease progression in infants was associated with loss of viral sensitivity to CC chemokines and emergence of CXCR4-using variants (24). While there is as yet no precise understanding of the selective factors underlying early R5 restriction and the late R5 → X4 evolution in

Figure 3 Temporal evolution of HIV-1 tropism during HIV-1 pathogenesis. HIV-1 transmission is restricted to R5 HIV-1 variants, which persist throughout the asymptomatic period as well as after the onset of AIDS. In many but not all individuals, viral tropism broadens to include X4 and R5X4 variants near the time when AIDS-defining symptoms are first observed.

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIV-1 CORECEPTORS

673

many individuals, and the contributions of these phenomena to pathogenesis, the results suggest an important role for CCR5 during initial viral transmission, and for CXCR4 and/or possibly other coreceptors in late stages of disease progression. CCR5 and CXCR4 are both expressed on known cell and tissue targets of HIV-1, consistent with roles in disease transmission and progression. In particular, CCR5 expression has been documented on cell and tissue types that may be important targets in the establishment of initial infection, including CD4+ T cells (13, 58, 135), monocyte/macrophages (12, 136, 137), dendritic cells (74, 138, 139), Langerhans cells (140, 141), and the mucosa of rectum and colon (137) as well as vagina and cervix (137, 141a). CXCR4 is expressed in many of the same cells and tissues as CCR5 (135, 137–141a).

Role of Coreceptors in HIV-1 Transmission THE CCR5 132 MUTATION: PROOF OF A CRITICAL ROLE FOR CCR5 IN HIV-1 TRANSMISSION The realization that CCR5 is the molecular factor mediating entry

of the preferentially transmitted M-tropic HIV-1 variants led to a focus on this coreceptor as a possible determinant of transmission. Definitive evidence came from the discovery of a mutant CCR5 allele designated CCR5 132 and the association of this allele with resistance to HIV-1 infection. CCR5 132 was discovered independently by several groups using different methods, including direct sequencing (142, 143), single stranded conformational polymorphism analysis (144), and heteroduplex mobility shift analysis (145) of CCR5 alleles. CCR5 132 has a 32 base pair deletion in the region of the open reading frame encoding the second extracellular loop, causing a frame shift and premature stop codon in transmembrane domain 5. The truncated protein product is not expressed on the cell surface (57, 142, 146). Homozygous CCR5 132 and HIV-1 resistance The first report of CCR5 132 described the mutation in homozygous form in two homosexual men who remained uninfected with HIV-1 despite repeated high-risk exposure (the so-called exposed-uninfected or EU phenotype) (142). Molecular epidemiological studies revealed highly statistically significant and reciprocol distortions in expected genotypic frequencies in HIV-1-infected versus ELL populations (143–145, 147, 148). Thus, the 132/132 genotype was found to be significantly enriched in several EU cohorts; conversely, 132/132 homozygotes were not found among several thousand HIV-1-infected individuals tested. Most importantly, these genotypic distortions were noted in cohorts of individuals exposed via mucosal or parenteral routes, including homosexual men, intravenous drug users, and hemophiliacs. In vitro experiments provided a

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

674

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

dramatic correlation to the population data: PBMCs and PBL from EU 132/ 132 homozygotes were shown to be susceptible to infection with X4 viruses, but completely resistant to R5 viruses (56, 57, 142, 143, 149). Heterozygous CCR5 132 and HIV-1 resistance Despite the unequivocal protective effect of CCR5 132 in homozygotes, no significant effect of the heterozygous genotype (+/132) on transmission has been found in most studies of homosexual, heterosexual, vertical, blood, and blood product transmission of HIV-1 (144, 145, 147, 150–157). Exceptions, all of which suggest a modest protective effect, include (a) a large European study that found a 35% reduction of +/132 in HIV-1-infected versus HIV-1-uninfected individuals (143); (b) a small study of HIV-1 discordant couples that found an increased frequency of +/132 in seronegative heterosexual, but not homosexual, individuals versus their seropositive partners (158); and (c) a study of 34 vertically infected Austrian infants in which the frequency of +/132 was significantly reduced (159). Identification of an EU who was heterozygous for CCR5 132 and whose PBMC were resistant in vitro to R5 HIV-1 initially suggested another exception. However, further analysis revealed that the second allele, designated CCR5 m303, harbored a chain-terminating single nucleotide polymorphism (T → A) at nucleotide 303 of the open reading frame (160), and did not encode a functional coreceptor. CCR5 m303 is inherited as a single mendelian trait; only three of 209 healthy blood donors tested were heterozygous for this allele, which is not an allelic frequency sufficient for meaningful epidemiologic studies using existing HIV-1 cohorts. Incomplete protection by CCR5 132/132 Rare HIV-1-infected 132/132 individuals have been reported (148, 161, 162), thus demonstrating that CCR5 is not absolutely required for HIV-1 transmission. In all cases, disease was progressive and the viral isolates appeared to be syncytium-inducing. In one of these individuals, X4 virus was exclusively and persistently detected (163); although isolates were not available at the time of seroconversion, this result raises the speculation that CXCR4 was the coreceptor responsible for initiating infection. Origin of CCR5 132 CCR5 132 is common in Caucasians and found at lower frequencies in the Middle East and India, but only sporadically among native Africans, Amerindians and East Asians (142–145, 147, 164–166). The allele frequency in North American Caucasians is typically ∼10%; heterozygotes and homozygotes represent ∼20% and ∼1% of this population, respectively, consistent with Hardy-Weinberg expectations and the absence of any effect of 132 on reproductive fitness. Consistent with this, 132/132 homozygotes who

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIV-1 CORECEPTORS

675

have been examined have no obvious health problems, suggesting that normal CCR5 function is dispensable, perhaps because of compensating function by other chemokine receptors with a similar leukocyte distribution. Haplotype analysis indicates that the 132 allele originated quite recently, ∼700 years ago (range 275–1, 875 years) at a single point in northeastern Europe (165, 166). A cline of CCR5 132 allele frequencies in a north to south gradient has been found in Europe, with the highest frequencies in Finnish and Mordvinian populations (16%), and the lowest in Sardinia (4%). These properties of CCR5 132 suggest that it was rapidly enriched in Caucasians because it conferred an advantage against some relatively recent and strong selective factor, possibly a catastrophic epidemic. Based on the time and place of fixation and the historical record, bubonic plague has been proposed as a plausible selective factor (165). Together the data provide a particularly compelling example of Darwinian evolution, or natural selection working on variation, and a population-based proof of the importance of CCR5 in HIV-1 transmission. CHEMOKINE BLOCKADE OF TRANSMISSION Although the 132/132 genotype is enriched in EU populations, the great majority of EUs do not have this genotype and their PBMCs are infectable with R5 HIV-1, indicating that the EU phenotype is heterogenous and that other resistance mechanisms must exist (145, 149). Indeed, genetic mechanisms unrelated to coreceptors have been proposed to influence susceptibiltiy to HIV-1 infection and disease (reviewed in 167). However, at least one other mechanism mediated by the coreceptor/ chemokine system may influence transmission. Several reports have suggested an association between the EU phenotype, both in homosexuals and in hemophiliacs, and high levels of the endogenous CC chemokines MIP-1α, MIP-1β and RANTES; this suggests that a “chemokine condom” could contribute to clinical resistance in some cases (149, 168). In fact, elevated secretion of CC chemokines by PBMCs from 132/132 homozygotes has been noted, suggesting a negative feedback loop controlling chemokine production (149). It has been proposed that resistance to HIV-1 infection may arise from a combination of high levels of inhibitory chemokines and low level expression of CCR5 (169).

Role of Coreceptors in HIV-1 Disease Progression Primary HIV-1 infection is associated with a burst of plasma viremia and an acute febrile illness, characterized by nonspecific symptomatology and spontaneous resolution. Plasma viremia then drops to relatively low levels, and the individual enters a prolonged asymptomatic period known as clinical latency. Numerous factors have been proposed to control viral replication during this period, including neutralizing antibody, virus-specific cytotoxic T cells, cytokines, chemokines, and availability of HIV-1 coreceptors (170).

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

676

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

The rate of disease progression is very heterogeneous among individuals, and is affected by many factors including age, associated diseases, immune activation, nutritional status, and viral strain. However, even when these factors are normalized, progression rates differ dramatically; in the extreme some rare individuals, known as long-term nonprogressors, do not appear to progress at all. These observations raise the possibility that progression rate may be influenced by factors that affect HIV-1 coreceptor levels or function. Directly correlating in vivo coreceptor levels with clinical progression rate is problematic, mainly due to the variability found in the levels on primary cells from different individuals. Associating in vivo levels of blocking chemokines with rates of disease progression is also difficult, due to problems in quantitating levels in plasma and tissues. A more informative approach has been to correlate genetic polymorphisms that may affect gene expression with clinical progression rate, using survival analysis and disease category analysis. Survival analysis requires large numbers of well-characterized HIV-1 seroconvertors (individuals for whom it is known accurately both the time of infection and the time when AIDS-defining criteria were met) (144, 157, 171, 172). Results from seroprevalent cohorts (individuals found to be HIV-1 seropositive when first tested), which have been used in some studies, can be misleading due to bias introduced by unintended exclusion of rapid progressors. Results are analyzed by Cox proportional hazards or KaplanMeier techniques using various endpoints, including the time to AIDS and death. Disease category analysis tests for distortion in expected genotypic frequencies in cohorts of HIV-1-infected subjects with specific outcomes, such as long-term survival (144, 145). Fortunately, several relatively large, well-characterized and well-managed cohorts of seroconvertors were initiated during the early years of the HIV-1 epidemic. So far four coreceptor/chemokine genetic polymorphisms have been identified and correlated with delayed HIV-1 disease progression rate using these cohorts: CCR5 132 (144, 145, 147, 152), CCR5 59029 G/A (173), CCR2-64I (157, 174, 175), and SDF-1 30 UTR-801G-A (abbreviated SDF-1 30 A) (157, 176). Several polymorphisms in the gene encoding CXCR4 have been found, but none has proven informative (177). Table 2 summarizes the allelic polymorphisms in coreceptors and their ligands that have been linked to HIV-1 disease. CCR5 59029 G/A, CCR2-64I, and SDF-1 30 A are single nucleotide polymorphisms (SNP) in the CCR5 promoter, CCR2 ORF and SDF-1 30 untranslated region (UTR), respectively. The mechanisms by which the genotypes produce their associated clinical effects are not clearly established. In this regard, it is important to note that CCR5 132, CCR2-64I, and CCR5 59029 G/A are in linkage disequilibrium, due to the adjacent location of CCR2 and CCR5 genes

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-21

HIV-1 CORECEPTORS

677

Table 2 Inherited polymorphisms in genes for chemokines and HIV-1 coreceptors that alter susceptibility to HIV-1 infection and progression Phenotypes Molecule

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CCR5

CCR2 SDF-1

Polymorphism

Site

Type

Freqa

CCR5 132 m303 59029G/A CCR2-64I SDF1-30 A

ORF ORF Pro ORF 30 UTR

Del SNP SNP SNP SNP

10% 1% 50% 10% 21%

−/−

+/−

Mechanism

R R DP ND DP

DP ND — DP —

Truncation Truncation ND ND ND

a The allelic frequencies listed are approximations based on genotyping North American Caucasians. Frequencies vary by racial group (see text for details and references). Abbreviations: Freq, Allelic frequency; −/−, homozygous for the given allele; +/−, heterozygous for the given allele; ORF, open reading frame; SNP, single nucleotide polymorphism; Del, deletion; Pro, promoter; 30 UTR, 30 untranslated region of the mRNA; R, HIV-1 resistance; DP, delayed progression to AIDS relative to other genotypes; ND, not determined. NE, no effects; ?, not determined.

within a chemokine receptor gene cluster on human chromosome 3p21-p24, that also includes CCR1, CCR3, CCR4, CCR8, and CX3CR1 (178). This must be kept in mind when considering statistical associations between particular polymorphisms and clinical outcome, since the identified polymorphism may just be a marker linked to other polymorphisms that have a direct effect on disease outcome. THE CCR5 132 ALLELE Too few HIV-1-infected 132/132 homozygotes have been found to meaningfully analyze the effect of this genotype on progression, except to say that individuals with this genotype have been identified who have progressed to AIDS. In adult populations of HIV-1-infected seroconvertors who are +/132, there is a delay of 0–2 years in mean time to AIDS compared to wild-type controls, depending on the AIDS definition and particular cohorts that are used (144, 145, 147, 152). In the largest study, seroconvertors were pooled from three different cohorts, including hemophiliacs and homosexual men, and a ∼2 year delay was observed (144). In addition, several cohorts of longterm nonprogressors have been reported in which a 50% increase in +/132 genotypic frequency relative to other HIV-1-infected populations and normals was observed, although immunologic and viral parameters broadly overlapped in individuals with similar rates of progression and discordant CCR5 genotypes (144, 145, 179). Significant protection from progression by this genotype has been difficult to demonstrate in studies of seroconvertors that have relied on only a single cohort. Perhaps contributing to delayed progression among heterozygotes, the mean values of CCR5 by flow cytometry of peripheral blood T cells from HIV-1 negative CCR5 +/132 individuals are significantly lower than would be

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

678

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

predicted by a simple gene dosage effect (∼10% of wild type controls, with considerable overlap) (58). There is biochemical evidence suggesting that this may be partly due to a transdominant effect of 132 through production of dysfunctional heterodimers composed of normal and truncated receptor subunits, which become trapped in the endoplasmic reticulum (143, 146). Functional consequences of the CCR5 +/132 genotype have been demonstrated both in vitro and in vivo. Thus, R5 HIV-1 entry and replication have been reported to be reduced in CCR5 +/132 versus +/+ PBMCs and monocyte-derived macrophages infected in vitro (58, 180); R5 HIV-1 exhibits delayed replication in SCID-hu mice reconstituted with CCR5 +/132 versus +/+ PBLs (181); and median viral load has been reported to be lower in CCR5 +/132 versus +/+ HIV-1-infected individuals (172). Taken together, these results suggest that CCR5 levels can be limiting, and that the +/132 genotype slows disease progression by causing reduced viral replication through reduced expression of CCR5. There is some evidence that the CCR5 +/132 genotype does not operate throughout the course of disease. In fact, paradoxically, it has been asssociated with accelerated progression to death after AIDS defining criteria were met (182). THE CCR2-64I ALLELE The CCR2-64I polymorphism (174) causes a conservative amino acid change, valine to isoleucine, at position 64 in the first transmembrane domain of CCR2, a region that has complete amino acid sequence conservation with CCR5. The allele is found in all racial groups tested at the following frequencies: Caucasians, 10%; African Americans, 15%; Hispanics, 17%; and Asians, 25% (174). CCR2-64I and CCR5 132 occur on separate haplotypes, meaning they are never inherited together on the same chromosome. CCR2-64I has no effect on initial HIV transmission. However, several studies have shown that seroconvertors bearing the CCR2-64I allele progress to AIDS significantly slower (by ∼2–3 years) than do CCR2 +/+ HIV-1 seroconvertors (157, 174, 175, 183). Like CCR5 132, CCR2-64I is enriched among long-term nonprogressors and reduced in rapid progressors (174, 175, 183). Moreover, the two alleles appear to exert an additive protective effect on progression rate in individuals who carry both (174, 175). The effect of CCR2-64I has not been observed in seroprevalent cohorts (175, 184), probably because it is masked by exclusion of rapid progressors from these cohorts (175). Also, one example of variability among seroconvertor cohorts has been reported in which the association between CCR2-64I and delayed progression was found for African-Americans but not for Caucasians in one study (157), and in the converse pattern in another (174). A study of African sex workers found that the CCR2-64I allele correlated with delayed progression to AIDS (185).

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIV-1 CORECEPTORS

679

Consistent with its association with delayed progression in seroconvertor cohorts, CCR2-64I has also been associated with significantly lower viral load at 9–12 months after seroconversion (175); the early viral load is a “set point” highly predictive of progression rate (186). Nevertheless, there is substantial doubt that the mechanism of action of CCR2-64I directly involves CCR2, since this coreceptor can be used by relatively few HIV-1 isolates in vitro (18, 29) and has not been consistently demonstrated to mediate HIV-1 infection in primary cells (14, 69, 115). Moreover, the CCR2-64I mutation does not affect coreceptor expression either in primary cells or in transfected cells, nor does it appear to affect chemokine binding or HIV-1 coreceptor activity (187). Instead, it has been suggested that CCR2-64I may be linked to another polymorphism that affects CCR5 expression or function (174). One candidate has been proposed, a C → T SNP at position 59653 in the CCR5 promoter (numbering as in GenBank #U95626) which is in 100% linkage with CCR2-64I, i.e. the two mutations are always inherited together (157, 175); however, so far there is no evidence that this polymorphism affects CCR5 expression (175). Instead, and quite surprisingly, an association has been found between the +/CCR2-64I genotype and reduced levels of CXCR4 on PBMCs from healthy donors (187). The mechanism cannot involve cis genetic effects since the gene encoding CXCR4 is on a different chromosome (2q) than CCR2 (3p). THE CCR5 59029 G/A POLYMORPHISM CCR5 59029 G/A is a G versus A SNP at base pair 59029 in the CCR5 promoter. Neither allele can be considered as “wild type,” since both are very common in all racial groups; the G allele is found at 43%–68% frequency depending on race (173). Haplotype analysis indicates that the 59029 A allele is in complete linkage disequilibrium with both CCR5 132 and CCR2-64I; that is, all chromosomes bearing either CCR5 132 or CCR2-64I also have 59029 A, although most chromosomes bearing 59029 A lack both CCR5 132 and CCR2-64I. In the one cohort study reported so far, the Multicenter AIDS Cohort Study (MACS) of homosexual and bisexual men, in individuals selected for absence of CCR5 132 and CCR2-64I the mean time to AIDS for 59029 G/G individuals was 3.8 years longer than for 59029 A/A individuals, p = 0.004 (173). This is the largest distortion of HIV-1 progression rate found so far for any polymorphism tested in a single cohort. For comparison, in the same cohort the +/64I and +/132 genotypes were each associated with an ∼1 year delay in mean time to AIDS, neither of which reached statistical significance (DA McDermott and PM Murphy, unpublished data). Consistent with the epidemiologic difference, promoter fragments differing in sequence only at 59029 G versus A had differential activity in a reporter gene assay, with the 59029 A promoter ∼50% more active than 59029 G

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

680

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

(173). This suggests that cells from 59029 G/G individuals may have decreased transcription of the CCR5 gene and decreased expression of CCR5, although this has not yet been demonstrated directly. Many other CCR5 promoter and open reading frame SNPs have been found, but associations with disease outcome have not been reported (157, 175, 177, 188, 189). The polymorphism designated SDF1-30 A is a G → A transition at bp 809 of the 30 -UTR of the mRNA encoding one of the two known chemokine ligands for CXCR4, SDF-1β. The second ligand, designated SDF-1α, produced by alternative splicing of a common SDF-1 gene, does not contain the SDF-1 30 A polymorphism. SDF-1 30 A is found in all racial groups tested, with high allele frequency in Caucasians (21%), Hispanics (16%), and Asians (26%), and relatively low frequency in African Americans (6%) (176). A clear picture has not yet emerged for the role of SDF-1 30 A in HIV-1 disease. In the initial report (176), which was based on analysis of 639 seroconvertors pooled from two cohorts of homosexual men (MACS and San Francisco City Cohort), one cohort of hemophiliacs (Multicenter Hemophilia Cohort Study), and one cohort of IV drug abusers (the ALIVE cohort), a strong association was described between the homozygous 30 A/30 A genotype and delayed onset of AIDS. A statistically significant effect was even observed in the MACS cohort tested separately from the others. In the pooled analysis, the effect was reported to be increased in later stages of HIV-1 infection, twice as strong as that found for CCR2-64I or CCR5 132 separately, and additive, possibly even synergistic, to them. In contrast, in a study of 470 seroconvertors from a single cohort (Tri-Service HIV-1 Natural History Study), the 30 A/30 A was associated with accelerated disease progression (157). Neither study found an altered rate of disease progression in heterozygotes, yet counterintuitively the first study found a significant increase in +/30 A in a group of 79 high-risk ELLs from the MACS, suggesting a protective effect from initial infection. The significance of this is unclear report however, since the same, did not find distortion of expected genotypic frequencies for either +/30 A or 30 A/30 A in a second group of 435 ELLs or in HIV-1 + individuals. The frequency of 30 A/30 A in the group of 79 ELLs was not reported. The differences between these two studies could result from differences in cohort composition or definition, or the low number of homozygotes available for statistical analysis (30 A/30 A individuals represent only ∼5% of Caucasians). It is quite possible that effects of SDF-1 30 A on progression rate could vary in different cohorts through differential environmental factors acting on the same postulated mechanism (176), namely posttranscriptional modulation of SDF-1 levels leading to effects on CXCR4 coreceptor activity.

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE SDF-1 30 A ALLELE

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

HIV-1 CORECEPTORS

681

Major Questions About Coreceptors in HIV-1 Disease With the coreceptor discoveries, some of the most perplexing questions about HIV-1 disease can now be posed in molecular terms:

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

1. What are the mechanisms underlying the predominance of R5 HIV-1 variants during establishment of initial infection as well as throughout the asymptomatic period? 2. What are the mechanisms involved in the emergence of X4 and R5X4 variants at later stages? 3. Why are CXCR4-using viruses detected in some AIDS individuals but not in others? 4. Is there a causal relationship between CXCR4 usage and the decline of CD4+ T cells and, if so, what are the mechanisms? 5. Does signaling induced by Env interaction with coreceptors play any role in HIV-1 disease? 6. Does Env impairment of normal chemokine-mediated signaling play any role in HIV-1 disease? 7. Are coreceptor levels limiting for transmission and disease progression? 8. To what extent does regulation of the levels of coreceptors and/or their chemokine ligands modulate infection susceptibility and disease progression rates? 9. What additional genetic influences on HIV-1 disease are mediated by the coreceptor/chemokine system, and what are their mechanisms of action? 10. Which members of the coreceptor repertoire are important for HIV-1 disease? The selective mechanisms observed during virus transmission are particularly bewildering. CCR5, the coreceptor used by HIV-1 variants detected shortly after transmission, is expressed on target cells and tissues that may be important for the establishment of initial infection, (i.e. CD4+ T cells, monocyte/macrophages, dendritic cells, Langerhans cells, vagina, cervix, rectum, and colon). Yet CXCR4, the coreceptor used by HIV-1 variants that are rarely detected at early stages of infection, is also expressed at many of these sites. The findings that CXCR4 is expressed at lower levels than CCR5 in colonic (137) and cervical (141a) mucosa may provide a partial explanation for the

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

682

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

predominance of R5 variants after sexual transmission. It is also possible that in order to gain a foothold after entering the body, the infection must become established in some uncharacterized compartment that is enriched in CCR5 but not CXCR4, or that has high levels of blocking chemokine ligands for CXCR4 but not CCR5. However, none of these considerations would explain the enigma of the low transmission frequency of R5X4 viruses, since these can readily use CCR5. It is presently unknown whether this apparent negative selection against CXCR4-using viruses is due to the interaction of the respective Envs with CXCR4 per se, or to other properties of the Envs that accompany functionality with CXCR4. An intriguing speculation is that those very features that confer CXCR4 functionality to Env might render the corresponding virus or infected cell susceptible to elimination by immune mechanisms (humoral and/or cellular). Similarly perplexing is the enrichment for CXCR4-using (R5X4 and X4) viruses during the course of disease progression. There are many possible selective mechanisms, including the elevation of CXCR4 and depression of CCR5 levels upon activation of CD4+ T lymphocytes (190, 191), the presence of R5-blocking CC chemokines at sites of virus replication (192, 193), enhanced production of CC chemokine caused by X4 virus infection (194), the ability of CC chemokines to stimulate replication of CXCR4-using viruses by inducing colocalization of CXCR4 with CD4 (77), limiting levels of CCR5, e.g. as occurs in CCR5 +/132 heterozygotes (195), and upregulation of CXCR4 induced by CC chemokines (196) and associated with bacterial infection (197). Just as the basis for the emergence of CXCR4-using variants is not understood, it is similarly unclear why CXCR4-using viruses are detected in only a subset of AIDS subjects (5). Since this notion is based mainly on analyses of isolates from peripheral blood, it is possible that the fraction would be significantly higher if variants are examined from a broader range of tissues. It is also possible that the frequency of CXCR4-using variants might be underestimated due to difficulties in detection; this concern is raised by the finding that in cells from some individuals, TCL-tropic viruses are only observed upon culture in the presence of CC chemokines (77). Another fundamental problem is the relationship of coreceptor usage to depletion of CD4+ T cells. R5X4 and X4 HIV-1 strains are generally more cytopathic than R5 strains in vitro, raising the possibility that CXCR4 usage might contribute to target cell killing, directly or indirectly. An ex vivo correlate has come from analysis of lymphoid tissue histocultures, where CD4+ T cell depletion occurred much more with X4 than with R5 viruses (198). The in vivo kinetics of plasma viremia in SCID-hu mice reconstituted with human peripheral blood leukocytes was found to correlate with the phenotype of the innoculated virus, with R5 strains exhibiting high viral titers and low CD4+

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-21

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIV-1 CORECEPTORS

683

T cell depletion, and X4 viruses the converse (199). A critical question for future studies is to determine whether these phenomena reflect an inherently greater cytopathicity of X4 strains, or simply a greater fraction of CD4+ T cells expressing CXCR4 compared to CCR5. A related problem is the possible physiologic significance of Env/coreceptor interactions beyond those involved in virus entry. Soluble Env has been demonstrated to induce signaling events via interaction with CCR5 (200) or CXCR4 (201), and apoptosis has been observed upon gp120 engagement of CXCR4 (202, 203). Furthermore, gp120 reportedly antagonizes CXCR4 and CCR5 signaling induced by their respective chemokine ligands, in a CD4-dependent fashion (204). Interaction between gp41 and coreceptors has also been suggested, based on the ability of gp41 to induce downmodulation of chemoattractant receptors (including HIV-1 coreceptors) on monocytes (205). The relevance of these processes for HIV-1 disease is presently unknown. While the importance for HIV-1 disease has clearly been established for CCR5, and to a lesser extent for CXCR4, the data are much less compelling for other members of the coreceptor repertoire. The genetic evidence has defined a central role for CCR5 in transmission; however, it does not rule out the possibility that other coreceptors may act in concert with CCR5, perhaps in a specialized physiological compartment critical for establishment of clinical infection. Other coreceptors may also be important for tissue-specific virus replication that might have profound consequences for the progression and manifestations of disease. This notion suggests the importance of analyzing coreceptor usage by primary virus isolates obtained from diverse target cells and tissues. An intriguing example is CCR8, which is expressed at minimal levels on peripheral blood T cells and monocytes, but at high levels in thymus (206, 207); perhaps this coreceptor plays an important role in the profound thymic abnormalities associated with HIV-1 infection (208). Similarly, STRL33/BONZO is abundantly expressed in placenta (34, 35), raising the possibility that it might play a role in vertical transmission.

CORECEPTOR-BASED THERAPEUTIC STRATEGIES The coreceptor discoveries have engendered new concepts to combat HIV-1 disease, at the levels of both treatment of infected subjects and prevention of transmission. Most are still at the developmental stage, although a few have progressed to clinical trials.

Coreceptor Blocking Agents Several classes of coreceptor blocking agents have been described. For each of these, the fact that different HIV-1 strains “see” a given coreceptor differently

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

684

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

(see sections above) raises the possibility that escape variants capable of using that coreceptor in the presence of the blocking agent will be selected for. CHEMOKINES AND DERIVATIVES The most straightforward coreceptor blocking agents are the natural chemokines that bind to and inhibit fusion/entry/infection mediated by the corresponding coreceptors, as has been shown with the ligands for CXCR4 (19, 20), CCR5 (10, 14–16), CCR3 (17), CCR8 (30), and CX3CR1 (33). However, concern has been raised that native chemokines may have unwanted activities due to their coupling to signaling pathways; thus, the CC chemokine stimulation of HIV-1 replication in macrophages under certain conditions is dependent on G protein signaling (67), as is the CC chemokine enhancement of TCL-tropic virus replication in T cells (77). Derivatized chemokine variants (93, 95, 209–212) and chemokine-based synthetic peptides (213, 214) have been reported with properties that may be favorable for anti-HIV therapy, including reduced agonist activity, enhanced blocking of fusion/entry/infection, and improved selectivity for the desired coreceptor. A variant of MIP-1α was found to be well-tolerated in Phase II cancer trials aimed at protecting hematopoietic stem cells during chemotherapy (212); however, no significant effects on viral load or CD4 counts were observed in a Phase I trial in HIV-infected subjects (LG Czaplewski, unpublished). ANTI-CORECEPTOR ANTIBODIES Anti-coreceptor monoclonal antibodies that inhibit HIV-1 entry represent another class of blocking agent. Murine monoclonal antibodies with HIV-1 inhibitory activity have been described for CXCR4 (68), CCR5 (58, 117), and CCR3 (73). Such antibodies can be humanized and tested for potential clinical utility. LOW MOLECULAR WEIGHT COMPOUNDS Perhaps the most promising class of blockers is low molecular weight compounds that bind directly to the coreceptors and inhibit their function. In general, members of the G protein–coupled receptor superfamily have proven to be ripe targets for pharmacologic intervention, and the chemokine receptors were already under investigation for development of antiinflammatory drugs prior to their implication in HIV-1 entry. Several low molecular weight blockers specific for CXCR4 have been described (215–217), as has been a more broadly active compound that blocks CCR5, CXCR4, and CCR3 (218). One CXCR4-blocking agent, the bicyclam AMD3100, was shown to inhibit HIV-1 replication in SCID-hu mice (219) prior to the realization that it is a specific inhibitor of entry via CXCR4 (216).

Ex Vivo Modulation of Coreceptor Expression An interesting therapeutic approach involves treatments that modulate coreceptor expression. An example is ex vivo activation of CD4+ T lymphocytes with

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

HIV-1 CORECEPTORS

685

antibodies to CD3 and CD28 adsorbed on beads (190). Cells treated in this fashion are refractory to infection with R5 strains but susceptible to X4 virus, consistent with the pattern of coreceptor expression (negligible CCR5, high CXCR4). This strategy is being investigated in the context of immune reconstitution with syngeneic CD4+ T lymphocytes, since the cells are refractory to infection by the R5 viruses that predominate in individuals at earlier stages of infection. Phase I clinical trials have been initiated (220).

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Gene Therapy Approaches Several strategies are suggested in the context of gene therapy, with the goal of depleting coreceptors from the surface of the target cells. One approach is to target coreceptor RNA (e.g. using ribozymes, antisense, DNA-enzymes). Some progress has been reported with a hammer-head ribozyme and a DNA-enzyme targeted to CCR5 mRNA; expression of the latter agent in target cells reduced their capacity to fuse with cells expressing an R5 Env in vitro (221). Alternatively, the coreceptor protein can be targeted. An intriguing strategy involves the expression of a so-called “intrakine,” a genetically engineered chemokine with a carboxy-terminal endoplasmic reticulum retention sequence; the intrakine traps the newly synthesized coreceptor and prevents its expression on the surface, thereby rendering the target cell refractory to HIV-1 infection. Intrakines have been described for downregulation of CXCR4 (222) and CCR5 (223).

Genetically Engineered Enveloped Virus Particles The coreceptors have also been used to design genetically engineered enveloped virus particles displaying CD4 and coreceptor in place of their native envelope glycoproteins. This concept has been applied to rabies virus (224), vesicular stomatitus virus (225), and an HIV-1 vector (226). The engineered viruses fuse selectively with cells productively infected with HIV-1 and expressing surface Env. The idea is that the engineered virus will kill the HIV-infected cells, either by a viral cytopathic effect or by delivering an antiviral gene.

Major Questions about Coreceptor-Based Therapeutic Strategies As with any novel therapeutic modality, the approaches described above are subject to obvious concerns about efficacy, toxicity, and practicality. Several interrelated questions arise in view of the current understanding of the coreceptor system: 1. Will the loss of normal chemokine receptor function of a specific coreceptor be tolerated?

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

686

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

2. Will impairment of CCR5 usage accelerate disease progression by enhancing selection of X4 and R5X4 variants?

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

3. How many members of the coreceptor repertoire must be blocked in order to achieve a therapeutic effect? Coreceptor-based therapeutic strategies have the appeal of targeting relatively invariant host determinants, in contrast with anti-HIV-1 agents directed against components of the rapidly mutating virus population. The limited in vivo data demonstrating slower HIV-1 disease progression in CCR5 +/132 heterozygotes and in CCR5 59029 G/G homozygotes provide support for a potential beneficial effect of targeting CCR5. Moreover, these data argue that the positive effects of lower CCR5 levels outweigh negative effects, including possible enhanced selection for CXCR4-using variants. One caveat, however, is that blocking CCR5 in the face of established HIV-1 disease may not be equivalent to expressing limiting CCR5 levels from the time of disease onset. There are reasonable grounds for optimism that interference with coreceptor function might be well tolerated. Individual components of the chemokine/ receptor system have limited rather than pleiotropic physiologic effects, in contrast with elements of other cytokine/receptor systems; moreover, the redundancy of the chemokine/receptor system offers the possibility that blockade of one will be compensated by the activity of others. Again, the genetic data for CCR5 are encouraging, since no adverse medical consequences have been noted in CCR5 132/132 homozygotes. However, the concern remains that individuals with CCR5 genetic defects may have compensating developmental changes in other components of the chemokine/receptor system, and that such compensations may not occur when CCR5 activity is modulated by the treatment modalities noted above. The situation with CXCR4 may be more problematic. SDF-1 30 A homozygosity has been associated with slower HIV-1 disease progression, but the effect is controversial and the mechanism is unknown. Moreover, there is cause for concern about undesired side effects of blocking CXCR4 function. Knockout mice lacking either SDF-1 (227) or CXCR4 (228, 229) die during embryogenesis, with evidence of hematopoietic, cardiac, vascular, and cerebellar defects. Thus, the SDF-1/CXCR4 interaction appears to be reciprocally monogamous, and may participate in wide-ranging developmental functions. From an optimistic viewpoint, it is possible that CXCR4 function is dispensable after embryogenesis, and that impairment of its activity would be tolerated.

CONCLUSION The discovery of HIV-1 coreceptors is an example of how basic research can link previously unrelated fields in totally unexpected ways to create a powerful

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIV-1 CORECEPTORS

687

new research paradigm with immediate clinical relevance. We now know that chemokine receptors, long studied for their roles in leukocyte trafficking in anti-microbial host defense, have been converted to pro-HIV-1 factors. As a result, we now have a more refined model for HIV-1 entry into target cells, a molecular basis for target cell tropism, new insights into the mechanisms underlying diverse aspects of HIV-1 disease, and new targets for therapeutic intervention. Demonstration of HIV-1 disease-modifying polymorphisms in CCR5 and other coreceptor genes represents a landmark in our understanding of the influence of human genetic factors on outcome in HIV-1 infection, and it supports genetic analysis as a general approach for understanding the heterogeneity of outcome characteristic of all infectious diseases. The observation that protective coreceptor mutations and polymorphisms are well-tolerated provides the proof of principle needed to justify development of coreceptor-based preventive and therapeutic interventions for the AIDS epidemic, and reason for optimism that they may succeed. ACKNOWLEDGMENTS We dedicate this review to the memory of Meta Ann Snyder, who, despite her illness, worked tirelessly to help others with AIDS. The insightful comments of Dr. Paolo Lusso are gratefully acknowledged. We regret that limitations of space prevented citation of all relevant articles in this field. NOTE ADDED IN PROOF

A newly described CCR5 promoter allele, designated P1, has been reported to show an epidemiological association with rapid progression to AIDS (230). Alleles at position 59029 were not assessed in that report; it is possible they may track the same phenotype due to linkage with P1. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Unutmaz D, Kewalramani VN, Littman DR. 1998. G protein-coupled receptors in HIV and SIV entry: new perspectives on lentivirus-host interactions and on the utility of animal models. Semin. Immunol. 10:225–36 2. Marx PA, Chen ZW. 1998. The function of simian chemokine receptors in the replication of SIV. Semin. Immunol. 10:215–23 3. Wyatt R, Sodroski J. 1998. The HIV-1 envelope glycoproteins: fusogens, anti-

gens, and immunogens. Science 280: 1884–88 4. Berger EA. 1997. HIV entry and tropism: the chemokine receptor connection. AIDS 11(Suppl. A):S3–S16 5. Miedema F, Meyaard L, Koot M, Klein MR, Roos MTL, Groenink M, Fouchier RAM, Van’t Wout AB, Tersmette M, Schellenkens PTA, Schuitemaker H. 1994. Changing virus-host interactions in the course of HIV-1 infection. Immunol. Rev. 140:35–72

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

688

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

6. Feng Y, Broder CC, Kennedy PE, Berger EA. 1996. HIV-1 entry cofactor: functional cDNA cloning of a seventransmembrane, G protein-coupled receptor. Science 272:872–77 7. Baggiolini M, Dewald B, Moser B. 1997. Human chemokines: an update. Annu. Rev. Immunol. 15:675–705 8. Murphy PM. 1996. Chemokine receptors: structure, function, and role in microbial pathogenesis. Cytokine Growth Fact. Rev. 7:47–64 9. Levy JA, Mackewicz CE, Barker E. 1996. Controlling HIV pathogenesis: the role of the noncytotoxic anti-HIV response of CD8(+) T cells. Immunol. Today 17:217–24 10. Cocchi F, DeVico AL, Garzino-Demo A, Arya SK, Gallo RC, Lusso P. 1995. Identification of RANTES, MIP-1α, and MIP-1β as the major HIV-suppressive factors produced by CD8+ T cells. Science 270:1811–15 11. Samson M, Labbe O, Mollereau C, Vassart G, Parmentier M. 1996. Molecular cloning and functional expression of a new human CC-chemokine receptor gene. Biochemistry 35:3362–67 12. Combadiere C, Ahuja SK, Tiffany HL, Murphy PM. 1996. Cloning and functional expression of CC CKR5, a human monocyte CC chemokine receptor selective for MIP-1α, MIP-1β, and RANTES. J. Leukocyte Biol. 60:147–52 13. Raport CJ, Gosling J, Schweickart VL, Gray PW, Charo IF. 1996. Molecular cloning and functional characterization of a novel human CC chemokine receptor (CCR5) for RANTES, MIP-1β, and MIP-1α. J. Biol. Chem. 271:17161–66 14. Deng HK, Liu R, Ellmeier W, Choe S, Unutmaz D, Burkhart M, Di Marzio P, Marmon S, Sutton RE, Hill CM, Davis CB, Peiper SC, Schall TJ, Littman DR, Landau NR. 1996. Identification of a major co-receptor for primary isolates of HIV-1. Nature 381:661–66 15. Dragic T, Litwin V, Allaway GP, Martin SR, Huang YX, Nagashima KA, Cayanan C, Maddon PJ, Koup RA, Moore JP, Paxton WA. 1996. HIV-1 entry into CD4+ cells is mediated by the chemokine receptor CC-CKR-5. Nature 381:667–73 16. Alkhatib G, Combadiere C, Broder CC, Feng Y, Kennedy PE, Murphy PM, Berger EA. 1996. CC CKR5: a RANTES, MIP-1α MIP-1β receptor as a fusion cofactor for macrophage-tropic HIV-1. Science 272:1955–58 17. Choe H, Farzan M, Sun Y, Sullivan N,

18.

19.

20.

21. 22.

23.

24.

25.

26.

Rollins B, Ponath PD, Wu L, Mackay CR, Larosa G, Newman W, Gerard N, Gerard C, Sodroski J. 1996. The betachemokine receptors CCR3 and CCR5 facilitate infection by primary HIV-1 isolates. Cell 85:1135–48 Doranz BJ, Rucker J, Yi YJ, Smyth RJ, Samson M, Peiper SC, Parmentier M, Collman RG, Doms RW. 1996. A dualtropic primary HIV-1 isolate that uses fusin and the beta-chemokine receptors CKR-5, CKR-3, and CKR-2b as fusion cofactors. Cell 85:1149–58 Bleul CC, Farzan M, Choe H, Parolin C, Clark-Lewis I, Sodroski J, Springer TA. 1996. The lymphocyte chemoattractant SDF-1 is a ligand for LESTR/fusin and blocks HIV-1 entry. Nature 382:829–33 Oberlin E, Amara A, Bachelerie F, Bessia C, Virelizier JL, ArenzanaSeisdedos F, Schwartz O, Heard JM, Clark-Lewis I, Legler DF, Loetscher M, Baggiolini M, Moser B. 1996. The CXC chemokine SDF-1 is the ligand for LESTR/fusin and prevents infection by T-cell-line-adapted HIV-1. Nature 382:833–35 Zhang LQ, Huang YX, He T, Cao YZ, Ho DD. 1996. HIV-1 subtype and second-receptor use. Nature 383:768 Connor RI, Sheridan KE, Ceradini D, Choe S, Landau NR. 1997. Change in coreceptor use correlates with disease progression in HIV-1-infected individuals. J. Exp. Med. 185:621–28 Bjorndal A, Deng H, Jansson M, Fiore JR, Colognesi C, Karlsson A, Albert J, Scarlatti G, Littman DR, Feny¨o EM. 1997. Coreceptor usage of primary human immunodeficiency virus type 1 isolates varies according to biological phenotype. J. Virol. 71:7478–87 Scarlatti G, Tresoldi E, Bjorndal A, Fredriksson R, Colognesi C, Deng HK, Malnati MS, Plebani A, Siccardi AG, Littman DR, Fenyo EM, Lusso P. 1997. In vivo evolution of HIV-1 coreceptor usage and sensitivity to chemokine-mediated suppression. Nature Med. 3:1259–65 Bazan HA, Alkhatib G, Broder CC, Berger EA. 1998. Patterns of CCR5, CXCR4 and CCR3 usage by envelope glycoproteins from human immunodeficiency virus type 1 primary isolates. J. Virol. 72:4485–91 Trkola A, Paxton WA, Monard SP, Hoxie JA, Siani MA, Thompson DA, Wu L, Mackay CR, Horuk R, Moore JP. 1998. Genetic subtype–independent inhibition of human immunodeficiency

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

HIV-1 CORECEPTORS

27.

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

28.

29.

30.

31.

32.

33.

34.

35.

virus type 1 replication by CC and CXC chemokines. J. Virol. 72:396– 404 Tscherning C, Alaeus A, Fredriksson R, Bjorndal A, Deng HK, Littman DR, Feny¨o EM, Albert J. 1998. Differences in chemokine coreceptor usage between genetic subtypes of HIV-1. Virology 241:181–88 Berger EA, Doms RW, Feny¨o EM, Korber BTM, Littman DR, Moore JP, Sattentau QJ, Schuitemaker H, Sodroski J, Weiss RA. 1998. A new classification for HIV-1. Nature 391:240 Rucker J, Edinger AL, Sharron M, Samson M, Lee B, Berson JF, Yi Y, Margulies B, Collman RG, Doranz BJ, Parmentier M, Doms RW. 1997. Utilization of chemokine receptors, orphan receptors, and herpesvirus-encoded receptors by diverse human and simian immunodeficiency viruses. J. Virol. 71:8999– 9007 Horuk R, Hesselgesser J, Zhou Y, Faulds D, Halks-Miller M, Harvey S, Taub D, Samson M, Parmentier M, Rucker J, Doranz BJ, Doms RW. 1998. The CC chemokine I-309 inhibits CCR8dependent infection by diverse HIV-1 strains. J. Biol. Chem. 273:386–91 Jinno A, Shimizu N, Soda Y, Haraguchi Y, Kitamura T, Hoshino H. 1998. Identification of the chemokine receptor TER1/CCR8 expressed in brain-derived cells and T cells as a new coreceptor for HIV-1 infection. Biochem. Biophys. Res. Commun. 243:497–502 Choe H, Farzan M, Konkel M, Martin K, Sun Y, Marcon L, Cayabyab M, Berman M, Dorf ME, Gerard N, Gerard C, Sodroski J. 1998. The orphan seven-transmembrane receptor ApJ. supports the entry of primary Tcell-line-tropic and dualtropic human immunodeficiency virus type 1. J. Virol. 72:6113–18 Combadiere C, Salzwedel K, Smith ED, Tiffany HL, Berger EA, Murphy PM. 1998. Identification of CX3CR1: a chemotactic receptor for the human CX3C chemokine fractalkine, and a fusion coreceptor for HIV-1. J. Biol. Chem. 273:23799–804 Liao F, Alkhatib G, Peden KWC, Sharma G, Berger EA, Farber JM. 1997. STRL33, a novel chemokine receptorlike protein, functions as a fusion cofactor for both macrophage-tropic and T cell line-tropic HIV-1. J. Exp. Med. 185:2015–23 Deng H, Unutmaz D, Kewalramani VN,

36.

37.

38.

39.

40.

41.

42.

689

Littman DR. 1997. Expression cloning of new receptors used by simian and human immunodeficiency viruses. Nature 388:296–300 Farzan M, Choe H, Martin K, Marcon L, Hofmann W, Karlsson G, Sun Y, Barrett P, Marchand N, Sullivan N, Gerard N, Gerard C, Sodroski J. 1997. Two orphan seven-transmembrane segment receptors which are expressed in CD4positive cells support simian immunodeficiency virus infection. J. Exp. Med. 186:405–11 Edinger AL, Hoffman TL, Sharron M, Lee BH, Yi YJ, Choe W, Kolson DL, Mitrovic B, Zhou YQ, Faulds D, Collman RG, Hesselgesser J, Horuk R, Doms RW. 1998. An orphan seventransmembrane domain receptor expressed widely in the brain functions as a coreceptor for human immunodeficiency virus type 1 and simian immunodeficiency virus. J. Virol. 72: 7934– 40 Pleskoff O, Treboute C, Brelot A, Heveker N, Seman M, Alizon M. 1997. Identification of a chemokine receptor encoded by human cytomegalovirus as a cofactor for HIV-1 entry. Science 276:1874–78 Kledal TN, Rosenkilde MM, Coulin F, Simmons G, Johnsen AH, Alouani S, Power CA, Luttichau HR, Gerstoft J, Clapham PR, Clark-Lewis I, Wells TNC, Schwartz TW. 1997. A broad-spectrum chemokine antagonist encoded by Kaposi’s sarcoma-associated herpesvirus. Science 277:1656–59 Boshoff C, Endo Y, Collins PD, Takeuchi Y, Reeves JD, Schweickart VL, Siani MA, Sasaki T, Williams TJ, Gray PW, Moore PS, Chang Y, Weiss RA. 1997. Angiogenic and HIVinhibitory functions of KSHV-encoded chemokines. Science 278:290–94 Owman C, Garzino-Demo A, Cocchi F, Popovic M, Sabirsh A, Gallo RC. 1998. The leukotriene B-4 receptor functions as a novel type of coreceptor mediating entry of primary HIV-1 isolates into CD4-positive cells. Proc. Natl. Acad. Sci. USA 95:9530–34 Samson M, Edinger AL, Stordeur P, Rucker J, Verhasselt V, Sharron M, Govaerts C, Mollereau C, Vassart G, Doms RW, Parmentier M. 1998. ChemR23, a putative chemoattractant receptor, is expressed in monocyte-derived dendritic cells and macrophages and is a coreceptor for SIV and some primary HIV-1 strains. Eur. J. Immunol. 28:1689–1700

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

690

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

43. Pal R, Garzino-Demo A, Markham PD, Burns J, Brown M, Gallo RC, DeVico AL. 1997. Inhibition of HIV-1 infection by the beta-chemokine MDC. Science 278:695–98 44. DeVico AL, Pal R, Markham PD, Garzino-Demo A, Gallo RC. 1998. βchemokine MDC and HIV-1 infection. Science 281:487a 45. Struyf S, Proost P, Sozzani S, Mantovani A, Wuyts A, De Clercq E, Van Damme J. 1998. Enhanced anti-HIV-1 activity and altered chemotactic potency of NH2-terminally processed macrophagederived chemokine (MDC) imply an additional MDC receptor. J. Immunol. 161:2672–75 46. Lee B, Rucker J, Doms RW, Tsang M, Hu X, Dietz M, Bailer R, Montaner LJ, Gerard C, Sullivan N, Sodroski J, Stantchev TS, Broder CC. 1998. βchemokine MDC and HIV-1 infection. Science 281:487a 47. Arenzana-Seisdedos F, Amara A, Thomas D, Virelizier JL, Baleux F, Clark-Lewis I, Legler DF, Moser B, Baggiolini M. 1998. β-chemokine MDC and HIV-1 infection. Science 281: 487a 48. Imai T, Chantry D, Raport CJ, Wood CL, Nishimura M, Godiska R, Yoshie O, Gray PW. 1998. Macrophage-derived chemokine is a functional ligand for the CC chemokine receptor 4. J. Biol. Chem. 273:1764–1768 49. Tachibana K, Nakajima T, Sato A, Igarashi K, Shida H, Iizasa H, Yashida N, Yoshie O, Kishimoto T, Nagasawa T. 1997. CXCR4/fusin is not a speciesspecific barrier in murine cells for HIV-1 entry. J. Exp. Med. 185:1865–70 50. Bieniasz PD, Fridell RA, Anthony K, Cullen BR. 1997. Murine CXCR-4 is a functional coreceptor for T-cell-tropic and dual-tropic strains of human immunodeficiency virus type 1. J. Virol. 71:7097–7100 51. Parolin C, Barsetti A, Choe H, Farzan M, Kolchinsky P, Heesen M, Ma Q, Gerard C, Palu G, Dorf ME, Springer T, Sodroski J. 1998. Use of murine CXCR-4 as a second receptor by some T-cell-tropic human immunodeficiency viruses. J. Virol. 72:1652–56 52. Atchison RE, Gosling J, Monteclaro FS, Franci C, Digilio L, Charo IF, Goldsmith MA. 1996. Multiple extracellular elements of CCR5 and HIV-1 entry: dissociation from response to chemokines. Science 274:1924–26 53. Doranz BJ, Lu ZH, Rucker J, Zhang TY,

54.

55.

56.

57.

58.

59.

60.

Sharron M, Cen YH, Wang ZX, Guo HH, Du JG, Accavitti MA, Doms RW, Peiper SC. 1997. Two distinct CCR5 domains can mediate coreceptor usage by human immunodeficiency virus type 1. J. Virol. 71:6305–14 Picard L, Simmons G, Power CA, Meyer A, Weiss RA, Clapham PR. 1997. Multiple extracellular domains of CCR-5 contribute to human immunodeficiency virus type 1 entry and fusion. J. Virol. 71:5003–11 Ross TM, Bieniasz PD, Cullen BR. 1998. Multiple residues contribute to the inability of murine CCR-5 to function as a coreceptor for macrophage-tropic human immunodeficiency virus type 1 isolates. J. Virol. 72:1918–24 Connor RI, Paxton WA, Sheridan KE, Koup RA. 1996. Macrophages and CD4+ T lymphocytes from two multiply exposed, uninfected individuals resist infection with primary non-syncytiuminducing isolates of human immunodeficiency virus type 1. J. Virol. 70:8758– 64 Rana S, Besson G, Cook DG, Rucker J, Smyth RJ, Yi Y, Turner JD, Guo HH, Du J-D, Peiper SC, Lavi E, Samson M, Libert F, Liesnard C, Vassart G, Doms RW, Parmentier M, Collman RG. 1997. Role of CCR5 in infection of primary macrophages and lymphocytes by macrophage-tropic strains of human immunodeficiency virus: resistance to patient-derived and prototype isolates resulting from the delta CCR5 mutation. J. Virol. 71:3219–27 Wu L, Paxton WA, Kassam N, Ruffing N, Rottman JB, Sullivan N, Choe H, Sodroski J, Newman W, Koup RA, Mackay CR. 1997. CCR5 levels and expression pattern correlate with infectability by macrophage-tropic HIV-1, in vitro. J. Exp. Med. 185:1681–91 Simmons G, Wilkinson D, Reeves JD, Dittmar MT, Beddows S, Weber J, Carnegie G, Desselberger U, Gray PW, Weiss RA, Clapham PR. 1996. Primary, syncytium-inducing human immunodeficiency virus type 1 isolates are dualtropic and most can use either Lestr or CCR5 as coreceptors for virus entry. J. Virol. 70:8355–60 Naif HM, Li S, Alali M, Sloane A, Wu L, Kelly M, Lynch G, Lloyd A, Cunningham AL. 1998. CCR5 expression correlates with susceptibility of maturing monocytes to human immunodeficiency virus type 1 infection. J. Virol. 72:830– 36

P1: SKH/spd/vks

March 3, 1999

P2: PSA/plb

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIV-1 CORECEPTORS 61. Dittmar MT, McKnight A, Simmons G, Clapham PR, Weiss RA, Simmonds P. 1997. HIV-1 tropism and co-receptor use. Nature 385:495–96 62. Coffey MJ, Woffendin C, Phare SM, Streiter RM, Markovitz DM. 1997. RANTES inhibits HIV-1 replication in human peripheral blood monocytes and alveolar macrophages. Am. J. Physiol. 272:L1025–29 63. Capobianchi MR, Abbate I, Antonelli G, Turriziani O, Dolei A, Dianzani F. 1998. Inhibition of HIV type 1 BaL replication by MIP-1 alpha, MIP-1 beta, and RANTES in macrophages. AIDS Res. Hum. Retrovir. 14:233–40 64. Schmidtmayerova H, Sherry B, Bukrinsky M. 1996. Chemokines and HIV replication. Nature 382:767 65. Moriuchi H, Moriuchi M, Combadiere C, Murphy PM, Fauci AS. 1996. CD8+ T-cell-derived soluble factor(s), but not β-chemokines RANTES, MIP-1α, and MIP-1β, suppress HIV-1 replication in monocyte/macrophages. Proc. Natl. Acad. Sci. USA 93:15341–45 66. Oravecz T, Pall M, Wang J, Roderiquez G, Ditto M, Norcross MA. 1997. Regulation of anti-HIV-1 activity of RANTES by heparan sulfate proteoglycans. J. Immunol. 159:4587–92 67. Kelly MD, Naif HM, Adams SL, Cunningham AL, Lloyd AR. 1998. Dichotomous effects of beta-chemokines on HIV replication in monocytes and monocyte-derived macrophages. J. Immunol. 160:3091–95 68. McKnight A, Wilkinson D, Simmons G, Talbot S, Picard L, Ahuja M, Marsh M, Hoxie JA, Clapham PR. 1997. Inhibition of human immunodeficiency virus fusion by a monoclonal antibody to a coreceptor (CXCR4) is both cell type and virus strain dependent. J. Virol. 71: 1692–96 69. Yi Y, Rana S, Turner JD, Gaddis N, Collman RG. 1998. CXCR-4 is expressed by primary macrophages and supports CCR5-independent infection by dualtropic but not T-tropic isolates of human immunodeficiency virus type 1. J. Virol. 72:772–77 70. Verani A, Pesenti E, Polo S, Tresoldi E, Scarlatti G, Lusso P, Siccardi AG, Vercelli D. 1998. CXCR4 is a functional coreceptor for infection of human macrophages by CXCR4-dependent primary HIV-1 isolates. J. Immunol. 161: 2084–88 71. Schmidtmayerova H, Aflano M, Nuovo G, Bukrinsky M. 1998. Human immun-

72.

73.

74.

75.

76.

77.

78.

79.

80.

691

odeficiency virus type 1 T-lymphotropic strains enter macrophages via a CD4and CXCR4-mediated pathway: Replication is restricted at a post-entry level. J. Virol. 72:4633–42 Kozak SL, Platt EJ, Madani N, Ferro FE Jr, Peden K, Kabat D. 1997. CD4, CXCR-4, and CCR-5 dependencies for infections by primary patient and laboratory-adapted isolates of human immunodeficiency virus type 1. J. Virol. 71:873–82 He J, Chen Y, Farzan M, Choe H, Ohagen A, Gartner S, Busciglio J, Yang X, Hofmann W, Newman W, Mackay CR, Sodroski J, Gabuzda D. 1997. CCR3 and CCR5 are co-receptors for HIV-1 infection of microglia. Nature 385:645– 49 Rubbert A, Combadiere C, Ostrowski M, Arthos J, Dybul M, Machado E, Cohn MA, Hoxie JA, Murphy PM, Fauci AS, Weissman D. 1998. Dendritic cells express multiple chemokine receptors used as coreceptors for HIV entry. J. Immunol. 160:3933–41 Lapham CK, Ouyang J, Chandrasekhar B, Nguyen NY, Dimitrov DS, Golding H. 1996. Evidence for cell-surface association between fusin and the CD4gp120 complex in human cell lines. Science 274:602–5 Ugolini S, Moulard M, Mondor I, Barois N, Demandolx D, Hoxie J, Brelot A, Alizon M, Davoust J, Sattentau QJ. 1997. HIV-1 gp120 induces an association between CD4 and the chemokine receptor CXCR4. J. Immunol. 159:3000– 8 Kinter A, Catanzaro A, Monaco J, Ruiz M, Justement J, Moir S, Arthos J, Oliva A, Ehler L, Mizell S, Jackson R, Ostrowski M, Hoxie J, Offord R, Fauci AS. 1998. CC-chemokines enhance the replication of T-tropic strains of HIV-1 in CD4+ T cells: role of signal transduction. Proc. Natl. Acad. Sci. USA 95: 11880–85 Iyengar S, Hildreth JEK, Schwartz DH. 1998. Actin-dependent receptor colocalization required for human immunodeficiency virus entry into host cells. J. Virol. 72:5251–55 Wu L, Gerard NP, Wyatt R, Choe H, Parolin C, Ruffing N, Borsetti A, Cardoso AA, Desjardin E, Newman W, Gerard C, Sodroski J. 1996. CD4-induced interaction of primary HIV-1 gp120 glycoproteins with the chemokine receptor CCR-5. Nature 384:179–83 Trkola A, Dragic T, Arthos J, Bin-

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

692

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

81.

82.

83.

84.

85.

86.

87.

88.

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER ley JM, Olson WC, Allaway GP, Cheng-Mayer C, Robinson J, Maddon PJ, Moore JP. 1996. CD4-dependent antibody-sensitive interactions between HIV-1 and its co-receptor CCR-5. Nature 384:184–87 Hill CM, Deng H, Unutmaz D, Kewalramani VN, Bastiani L, Gorny MK, ZollaPazner S, Littman DR. 1997. Envelope glycoproteins from human immunodeficiency virus types 1 and 2 and simian immunodeficiency virus can use human CCR5 as a coreceptor for viral entry and make direct CD4-dependent interactions with this chemokine receptor. J. Virol. 71:6296–304 Bandres JC, Wang QF, O’Leary J, Baleaux F, Amara A, Hoxie JA, ZollaPazner S, Gorny MK. 1998. Human immunodeficiency virus (HIV) envelope binds to CXCR4 independently of CD4, and binding can be enhanced by interaction with soluble CD4 or by HIV envelope deglycosylation. J. Virol. 72:2500– 4 Sullivan N, Sun Y, Sattentau Q, Thali M, Wu D, Denisova G, Gershoni J, Robinson J, Moore J, Sodroski J. 1998. CD4induced conformational changes in the human immunodeficiency virus type 1 gp120 glycoprotein: Consequences for virus entry and neutralization. J. Virol. 72:4694–4703 Kwong PD, Wyatt R, Robinson J, Sweet RW, Sodroski J, Hendrickson WA. 1998. Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 393:648–59 Rizzuto CD, Wyatt R, HernandezRamos N, Sun Y, Kwong PD, Hendrickson WA, Sodroski J. 1998. A conserved HIV gp120 glycoprotein structure involved in chemokine receptor binding. Science 280:1949–53 Wyatt R, Kwong PD, Desjardins E, Sweet RW, Robinson J, Hendrickson WA, Sodroski JG. 1998. The antigenic structure of the HIV gp120 envelope glycoprotein. Nature 393:705–11 Farzan M, Choe H, Martin KA, Sun Y, Sidelko M, Mackay CR, Gerard NP, Sodroski J, Gerard C. 1997. HIV-1 entry and macrophage inflammatory protein1β-mediated signaling are independent functions of the chemokine receptor CCR5. J. Biol. Chem. 272:6854–57 Gosling J, Monteclaro FS, Atchison RE, Arai H, Tsou CL, Goldsmith MA, Charo IF. 1997. Molecular uncoupling of C-C chemokine receptor 5–induced chemo-

89.

90.

91.

92.

93.

94.

95.

96.

taxis and signal transduction from HIV1 coreceptor activity. Proc. Natl. Acad. Sci. USA 94:5061–66 Alkhatib G, Locati M, Kennedy PE, Murphy PM, Berger EA. 1997. HIV-1 coreceptor activity of CCR5 and its inhibition by chemokines: independence from G protein signaling and importance of coreceptor downmodulation. Virology 234:340–48 Aramori I, Zhang J, Ferguson SS, Bieniasz PD, Cullen BR, Caron MG. 1997. Molecular mechanism of desensitization of the chemokine receptor CCR-5: receptor signaling and internalization are dissociable from its role as an HIV-1 co-receptor. EMBO. J. 16:4606–16 Hu H, Shioda T, Hori T, Moriya C, Kato A, Sakai Y, Natsushima K, Uchiyama T, Nagai Y. 1998. Dissociation of ligandinduced internalization of CXCR-4 from its co-receptor activity for HIV-1 Envmediated membrane fusion. Arch. Virol. 143:851–61 Amara A, Gall S, Schwartz O, Salamero J, Montes M, Loetscher P, Baggiolini M, Virelizier J-L, Arenzana-Seisdedos F. 1997. HIV coreceptor downregulation as antiviral principle: SDF-1α-dependent internalization of the chemokine receptor CXCR4 contributes to inhibition of HIV replication. J. Exp. Med. 186:139– 46 Crump MP, Gong JH, Loetscher P, Rajarathnam K, Amara A, ArenzanaSeisdedos F, Virelizier JL, Baggiolini M, Sykes BD, Clark-Lewis I. 1997. Solution structure and basis for functional activity of stromal cell-derived factor-1; dissociation of CXCR4 activation from binding and inhibition of HIV-1. EMBO J. 16:6996–7007 Signoret N, Oldridge J, PelchenMatthews A, Klasse PJ, Tran T, Brass LF, Rosenkilde MM, Schwartz TW, Holmes W, Dallas W, Luther MA, Wells TN, Hoxie JA, Marsh M. 1997. Phorbol esters and SDF-1 induce rapid endocytosis and down modulation of the chemokine receptor CXCR4. J. Cell. Biol. 139:651–664 Mack M, Luckaw B, Nelson PJ, Cihak J, Simmons G, Clapham PR, Signoret N, Marsh M, Stangassinger M, Borlat F, Wells TN, Schlondorff D, Proudfoot AE. 1998. Aminooxypentane-RANTES induces CCR5 internalization but inhibits recycling: a novel inhibitory mechanism of HIV infectivity. J. Exp. Med. 187:1215–24 Tarasova NI, Stauber RH, Michejda CJ.

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

HIV-1 CORECEPTORS

97.

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

98.

99.

100.

101.

102.

103.

104.

1998. Spontaneous and ligand-induced trafficking of CXC-chemokine receptor 4. J. Biol. Chem. 273:15883–86 Endres MJ, Clapham PR, Marsh M, Ahuja M, Turner JD, McKnight A, Thomas JF, Stoebenau-Haggarty B, Choe S, Vance PJ, Wells TNC, Power CA, Sutterwala SS, Doms RW, Landau NR, Hoxie JA. 1996. CD4-independent infection by HIV-2 is mediated by Fusin/CXCR4. Cell 87:745–56 Reeves JD, McKnight A, Potempa S, Simmons G, Gray PW, Power CA, Wells T, Weiss RA, Talbot SJ. 1997. CD4-independent infection by HIV-2 (ROD/B): use of the 7-transmembrane receptors CXCR-4, CCR-3, and V28 for entry. Virology 231:130–34 Potempa S, Picard L, Reeves JD, Wilkinson D, Weiss RA, Talbot SJ. 1997. CD4independent infection by human immunodeficiency virus type 2 strain ROD/B: the role of the N-terminal domain of CXCR-4 in fusion and entry. J. Virol. 71:4419–24 Reeves JD, Schulz TF. 1997. The CD4independent tropism of human immunodeficiency virus type 2 involves several regions of the envelope protein and correlated with a reduced activation threshold for envelope-mediated fusion. J. Virol. 71:1453–65 Martin KA, Wyatt R, Farzan M, Choe H, Marcon L, Desjardin E, Robinson J, Sodroski J, Gerard C, Gerard NP. 1997. CD4-independent binding of SIV gp120 to rhesus CCR5. Science 278:1470–73 Edinger AL, Markowski JL, Doranz BJ, Margulies BJ, Lee B, Rucker J, Sharron M, Hoffman TL, Berson JF, Zink MC, Hirsch VM, Clements JE, Doms RW. 1997. CD4-independent, CCR5dependent infection of brain capillary endothelial cells by a neurovirulent simian immunodeficiency virus strain. Proc. Natl. Acad. Sci. USA 94:14742– 47 Hesselgesser J, Halks-Miller M, DelVecchio V, Peiper SC, Hoxie J, Kolson DL, Taub D, Horuk R. 1997. CD4-independent association between HIV-1 gp120 and CXCR4: functional chemokine receptors are expressed in human neurons. Curr. Biol. 7:112–21 Dumonceaux J, Nisole S, Chanel C, Quivet L, Amara A, Baleux F, Briand F, Hazan U. 1998. Spontaneous mutations in the env gene of the human immunodeficiency virus type 1 NDK isolate are associated with a CD4-independent entry phenotype. J. Virol. 72:512–19

693

105. Cocchi F, DeVico AL, Garzino-Demo A, Cara A, Gallo RC, Lusso P. 1996. The V3 domain of the HIV-1 gp120 envelope glycoprotein is critical for chemokinemediated blockade of infection. Nature Med. 2:1244–47 106. Bieniasz PD, Fridell RA, Aramori I, Ferguson SS, Caron MG, Cullen BR. 1997. HIV-1-induced cell fusion is mediated by multiple regions within both the viral envelope and the CCR-5 co-receptor. EMBO J. 16:2599–2609 107. Pleskoff O, Sol N, Labrosse B, Alizon M. 1997. Human immunodeficiency virus strains differ in their ability to infect CD4+ cells expressing the rat homolog of CXCR-4 (fusin). J. Virol. 71: 3259–62 108. Speck RF, Wehrly K, Platt EJ, Atchison RE, Charo IF, Kabat D, Chesebro B, Goldsmith MA. 1997. Selective employment of chemokine receptors as human immunodeficiency virus type 1 coreceptors determined by individual amino acids within the envelope V3 loop. J. Virol. 71:7136–39 109. Cho MW, Lee MK, Carney MC, Berson JF, Doms RW, Martin MA. 1998. Identification of determinants on a dualtropic human immunodeficiency virus type 1 envelope glycoprotein that confer usage of CXCR4. J. Virol. 72:2509–15 110. Ross TM, Cullen BR. 1998. The ability of HIV type 1 to use CCR-3 as a coreceptor is controlled by envelope V1/V2 sequences acting in conjunction with a CCR-5 tropic V3 loop. Proc. Natl. Acad. Sci. USA 95:7682–86 111. Smythe RJ, Yi YJ, Singh A, Collman RG. 1998. Determinants of entry cofactor utilization and tropism in a dualtropic human immunodeficiency virus type 1 primary isolate. J. Virol. 72:4478–84 112. Wang WK, Dudek T, Zhao YJ, Brumblay HG, Essex M, Lee TH. 1998. CCR5 coreceptor utilization involves a highly conserved arginine residue of HIV type 1 gp120. Proc. Natl. Acad. Sci. USA 95: 5740–45 113. Xiao LZH, Owen SM, Goldman I, Lal AA, deJong JJ, Goudsmit J, Lal RB. 1998. CCR5 coreceptor usage of nonsyncytium-inducing primary HIV-1 is independent of phylogenetically distinct global HIV-1 isolates: delineation of consensus motif in the V3 domain that predicts CCR5 usage. Virology 240:83– 92 114. Hoffman TL, Stephens EB, Narayan O, Doms RW. 1998. HIV type I envelope determinants for use of the CCR2b,

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

694

115.

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

116.

117.

118.

119.

120.

121.

122.

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER CCR3, STRL33, and APJ. coreceptors. Proc. Natl. Acad. Sci. USA 95:11360–65 Frade JMR, Llorente M, Mellado M, Alcami J, Gutierrez-Ramos JC, Zaballos A, Real G, Martinez-A C. 1997. The amino-terminal domain of the CCR2 chemokine receptor acts as coreceptor for HIV-1 infection. J. Clin Invest 100:497–502 Rucker J, Samson M, Doranz BJ, Libert F, Berson JF, Yi YJ, Smyth RJ, Collman RG, Broder CC, Vassart G, Doms RW, Parmentier M. 1996. Regions in betachemokine receptors CCR5 and CCR2b that determine HIV-1 cofactor specificity. Cell 87:437–46 Wu L, Larosa G, Kassam N, Gordon CJ, Heath H, Ruffing N, Chen H, Humblias J, Samson M, Parmentier M, Moore JP, Mackay CR. 1997. Interaction of chemokine receptor CCR5 with its ligands: multiple domains for HIV-1 gp120 binding and a single domain for chemokine binding. J. Exp. Med. 186: 1373–81 Alkhatib G, Berger EA, Murphy PM, Pease JE. 1997. Determinants of HIV1 coreceptor function on CC chemokine receptor 3: importance of both extracellular and transmembrane/cytoplasmic regions. J. Biol. Chem. 272:20420–26 Lu Z, Berson JF, Chen Y, Turner JD, Zhang T, Sharron M, Jenks MH, Wang Z, Kim J, Rucker J, Hoxie JA, Peiper SC, Doms RW. 1997. Evolution of HIV1 coreceptor usage through interactions with distinct CCR5 and CXCR4 domains. Proc. Natl. Acad. Sci. USA 94: 6426–31 Edinger AL, Amedee A, Miller K, Doranz BJ, Endres M, Sharron M, Samson M, Lu ZH, Clements JE, Murphey-Corb M, Peiper SC, Parmentier M, Broder CC, Doms RW. 1997. Differential utilization of CCR5 by macrophage and T cell tropic simian immunodeficiency virus strains. Proc. Natl. Acad. Sci. USA 94:4005–10 Strizki JM, Turner JD, Collman RG, Hoxie J, Gonzalez-Scarano F. 1997. A monoclonal antibody (12G5) directed against CXCR-4 inhibits infection with the dual-tropic human immunodeficiency virus type 1 isolate HIV-1(89.6) but not the T-tropic isolate HIV-1(HxB). J. Virol. 71:5678–83 Alkhatib G, Ahuja SS, Light D, Mummidi S, Berger EA, Ahuja SK. 1997. CC chemokine receptor 5-mediated signaling and HIV-1 co-receptor activity share common structural determinants:

123.

124.

125.

126.

127.

128.

129.

130.

131.

critical residues in the third extracellular loop support HIV-1 fusion. J. Biol. Chem. 272:19771–76 Kuhmann SE, Platt EJ, Kozak SL, Kabat D. 1997. Polymorphisms in the CCR5 genes of African green monkeys and mice implicate specific amino acids in infections by simian and human immunodeficiency viruses. J. Virol. 71:8642–56 Willett BJ, Picard L, Hosie MJ, Turner JD, Adema K, Clapham PR. 1997. Shared usage of the chemokine receptor CXCR4 by the feline and human immunodeficiency viruses. J. Virol. 71:6407– 15 Brelot A, Heveker N, Pleskoff O, Sol N, Alizon M. 1997. Role of the first and third extracellular domains of CXCR-4 in human immunodeficiency virus coreceptor activity. J. Virol. 71:4744–51 Picard L, Wilkinson DA, McKnight A, Gray PW, Hoxie JA, Clapham PR, Weiss RA. 1997. Role of the amino-terminal extracellular domain of CXCR-4 in human immunodeficiency virus type 1 entry. Virology 231:105–11 Dragic T, Trkola A, Lin SW, Nagashima KA, Kajumo F, Zhao L, Olson WC, Wu LJ, Mackay CR, Allaway GP, Sakmar TP, Moore JP, Maddon PJ. 1998. Aminoterminal substitutions in the CCR5 coreceptor impair gp120 binding and human immunodeficiency virus type 1 entry. J. Virol. 72:279–85 Rabut GEE, Konner JA, Kajumo F, Moore JP, Dragic T. 1998. Alanine substitutions of polar and nonpolar residues in the amino-terminal domain of CCR5 differently impair entry of macrophageand dualtropic isolates of human immunodeficiency virus type 1. J. Virol. 72:3464–68 Wang ZX, Berson JF, Zhang TY, Cen YH, Sun Y, Sharron M, Lu ZH, Peiper SC. 1998. CXCR4 sequences involved in coreceptor determination of human immunodeficiency virus type-1 tropism. Unmasking of activity with Mtropic Env glycoproteins. J. Biol. Chem. 273:15007–15 Farzan M, Choe H, Vaca L, Martin K, Sun Y, Desjardins E, Ruffing N, Wu LJ, Wyatt R, Gerard N, Gerard C, Sodroski J. 1998. A tyrosine-rich region in the N terminus of CCR5 is important for human immunodeficiency virus type 1 entry and mediates an association between gp120 and CCR5. J. Virol. 72:1160– 64 Reeves JD, Heveker N, Brelot A, Alizon M, Clapham PR, Picard L. 1998.

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

HIV-1 CORECEPTORS

132.

133.

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

134.

135.

136.

137.

138.

139.

140.

141.

The second extracellular loop of CXCR4 is involved in CD4-independent entry of human immunodeficiency virus type 2. J. Gen. Virol. 79:1793–99 Chan DC, Fass D, Berger JM, Kim PS. 1997. Core structure of gp41 from the HIV envelope glycoprotein. Cell 89: 263–73 Weissenhorn W, Dessen A, Harrison SC, Skehel JJ, Wiley DC. 1997. Atomic structure of the ectodomain from HIV-1 gp41. Nature 387:426–30 Caffrey M, Cai ML, Kaufman J, Stahl SJ, Wingfield PT, Covell DG, Gronenborn AM, Clore GM. 1998. Threedimensional solution structure of the 44 kDa ectodomain of SIV gp41. EMBO J. 17:4572–84 Bleul CC, Wu L, Hoxie JA, Springer TA, Mackay CR. 1997. The HIV coreceptors CXCR4 and CCR5 are differentially expressed and regulated on human T lymphocytes. Proc. Natl. Acad. Sci. USA 94:1925–30 Rottman JB, Ganley KP, Williams K, Wu L, Mackay CR, Ringler DJ. 1997. Cellular localization of the chemokine receptor CCR5. Correlation to cellular targets of HIV-1 infection. Am. J. Pathol. 151:1341–51 Zhang LQ, He T, Talal A, Wang G, Frankel SS, Ho DD. 1998. In vivo distribution of the human immunodeficiency virus simian immunodeficiency virus coreceptors: CXCR4, CCR3, and CCR5. J. Virol. 72:5035–45 Granelli-Piperno A, Moser B, Pope M, Chen D, Wei Y, Isdell F, O’Doherty U, Paxton W, Koup R, Mojsov S, Bhardwa JN, Clark-Lewis I, Baggiolini M, Steinman RM. 1996. Efficient interaction of HIV-1 with purified dendritic cells via multiple chemokine coreceptors. J. Exp. Med. 184:2433–38 Ayehunie S, Garcia-Zepeda EA, Hoxie JA, Horuk R, Kupper TS, Luster AD, Ruprecht RM. 1997. Human immunodeficiency virus-1 entry into purified blood dendritic cells through CC and CXC chemokine coreceptors. Blood 90:1379– 86 Dittmar MT, Simmons G, Hibbitts S, O’Hare M, Louisirirotchanakul S, Beddows S, Weber J, Clapham PR, Weiss RA. 1997. Langerhans cell tropism of human immunodeficiency virus type 1 subtype A through F isolates derived from different transmission groups. J. Virol. 71:8008–13 Zaitseva M, Blauvelt A, Lee S, Lapham CK, Klaus-Kovtun V, Mostowski H,

141a.

142.

143.

144.

145.

146.

695

Manischewitz J, Golding H. 1997. Expression and function of CCR5 and CXCR4 on human Langerhans cells and macrophages: implications for HIV primary infection. Nature Med. 3:1369– 75 Patterson BK, Landay A, Andersson J, Brown C, Behbahani H, Jiyamapa D, Burki Z, Stanlaslawski D, Czerniewski MA, Garcia P. 1998. Repertoire of chemokine receptor expression in the female genital tract: implications for human immunodeficiency virus transmission. Am. J. Pathol. 153:481–90 Liu R, Paxton WA, Choe S, Ceradini D, Martin SR, Horuk R, Macdonald ME, Stuhlmann H, Koup RA, Landau NR. 1996. Homozygous defect in HIV1 coreceptor accounts for resistance of some multiply-exposed individuals to HIV-1 infection. Cell 86:367–77 Samson M, Libert F, Doranz BJ, Rucker J, Liesnard C, Farber CM, Saragosti S, Lapoumeroulie C, Cognaux J, Forceille C, Muyldermans G, Verhofstede C, Burtonboy G, Georges M, Imai T, Rana S, Yi YJ, Smyth RJ, Collman RG, Doms RW, Vassart G, Parmentier M. 1996. Resistance to HIV-1 infection in Caucasian individuals bearing mutant alleles of the CCR-5 chemokine receptor gene. Nature 382:722–25 Dean M, Carrington M, Winkler C, Huttley GA, Smith MW, Allikmets R, Goedert JJ, Buchbinder SP, Vittinghoff E, Gomperts E, Dornfield S, Vlahov D, Kaslow R, Saah A, Rinaldo C, Detels R, Hemophilia Growth and Development Study, Multicenter AIDS Cohort Study, Multicenter Hemophilia Cohort Study, San Francisco City Cohort, ALIVE Study, O’Brien SJ. 1996. Genetic restriction of HIV-1 infection and progression to AIDS by a deletion allele of the CKR5 structural gene. Science 273:1856–62 Zimmerman PA, Buckler-White A, Alkhatib G, Spalding T, Kubofcik J, Combadiere C, Weissman D, Cohen O, Rubbert A, Lam G, Vaccarezza M, Kennedy PE, Kumaraswami V, Giorgi JV, Detels R, Hunter J, Chopek M, Berger EA, Fauci AS, Nutman TB, Murphy PM. 1997. Inherited resistance to HIV-1 conferred by an inactivating mutation in CC chemokine receptor 5: studies in populations with contrasting clinical phenotypes, defined racial background, and quantified risk. Mol. Med. 3:23–35 Benkirane M, Jin DY, Chun RF,

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

696

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

147.

148.

149.

150.

151.

152.

153.

154.

155.

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER Koup RA, Jeang KT. 1997. Mechanism of transdominant inhibition of CCR5-mediated HIV-1 infection by ccr5delta32. J. Biol. Chem. 272:30603– 6 Huang YX, Paxton WA, Wolinsky SM, Neumann AU, Zhang LQ, He T, Kang S, Ceradini D, Jin ZQ, Yazdanbakhsh K, Kunstman K, Erickson D, Dragon E, Landau NR, Phair J, Ho DD, Koup RA. 1996. The role of a mutant CCR5 allele in HIV-1 transmission and disease progression. Nature Med. 2:1240–43 O’Brien TR, Winkler C, Dean M, Nelson JR, Carrington M, Michael NL, White GC II. 1997. HIV-1 infection in a man homozygous for CCR5 delta 32. Lancet 349:1219 Paxton WA, Martin SR, Tse D, O’Brien TR, Skurnick J, VanDevanter NL, Padian N, Braun JF, Kotler DP, Wolinsky SM, Koup RA. 1996. Relative resistance to HIV-1 infection of CD4 lymphocytes from persons who remain uninfected despite multiple high-risk sexual exposures. Nature Med. 2:412–17 Edelstein RE, Arcuino LA, Hughes JP, Melvin AJ, Mohan KM, King PD, McLellan CL, Murante BL, Kassman BP, Frenkel LM. 1997. Risk of motherto-infant transmission of HIV-1 is not reduced in CCR5/delta32ccr5 heterozygotes. J. Acquired Immune Defic. Syndr. Hum. Retrovirol. 16:243–46 Rousseau CMS, Just JJ, Abrams EJ, Casabona J, Stein Z, King MC. 1997. CCR5del32 in perinatal HIV-1 infection. J. Acquired Immune Defic. Syndr. Hum. Retrovirol. 16:239–242 Michael NL, Chang G, Louie LG, Mascola JR, Dandero D, Birx DL, Sheppard HW. 1997. The role of viral phenotype and CCR-5 gene defects in HIV-1 transmission and disease progression. Nature Med. 3:338–40 Misrahi M, Teglas JP, N’Go N, Burgard M, Mayaux MJ, Rouzioux C, Delfraissy JF, Blanche S. 1998. CCR5 chemokine receptor variant in HIV-1 mother-tochild transmission and disease progression in children. J. Am. Med. Assoc. 279:317–18 Malo A, Rommel F, Bogner J, Gruber R, Schramm W, Goebel FD, Riethmuller G, Wank R. 1998. Lack of protection from HIV infection by the mutant HIV coreceptor CCR5 in intravenously HIV infected hemophilia patients. Immunobiology 198:485–88 Mangano A, Prada F, Roldan A, Picchio G, Bologna R, Sen L. 1998. Distribu-

156.

157.

158.

159.

160.

161.

162.

163.

164.

165.

tion of CCR-5 delta32 allele in Argentinian children at risk of HIV-1 infection: its role in vertical transmission. AIDS 12:109–10 Shearer WT, Kalish LA, Zimmerman PA. 1998. CCR5 HIV-1 vertical transmission. J. Acquired Immune Defic. Syndr. 17:180–81 Mummidi S, Ahuja SS, Gonzalez E, Anderson SA, Santiago EN, Stephan KT, Craig FE, O’Connell P, Tryon V, Clark RA, Dolan MJ, Ahuja SK. 1998. Genealogy of the CCR5 locus and chemokine system gene variants associated with altered rates of HIV-1 disease progression. Nature Med. 4:786–93 Hoffman TL, MacGregor RR, Burger H, Mick R, Doms RW, Collman RG. 1997. CCR5 genotypes in sexually active couples discordant for human immunodeficiency virus type 1 infection status. J. Infect. Dis. 176:1093–96 Mandl CW, Aberle SW, Henkel JH, Puchhammer-Stockl E, Heinz FX. 1998. Possible influence of the mutant CCR5 allele on vertical transmission of HIV-1. J. Med. Virol. 55:51–55 Quillent C, Oberlin E, Braun J, Rousset D, Gonzalez-Canali G, Metais P, Montagnier L, Virelizier J-L, ArenzanaSeisdedos F, Beretta A. 1998. HIV-1 resistance phenotype conferred by combination of two separate inherited mutations of CCR5 gene. Lancet 351:14–18 Biti R, Ffrench R, Young J, Bennetts B, Stewart G, Liang T. 1997. HIV-1 infection in an individual homozygous for the CCR5 deletion allele. Nature Med. 3:252–53 Theodorou I, Meyer L, Magierowska M, Katlama C, Rouzioux C. 1997. HIV-1 infection in an individual homozygous for CCR5 delta 32. Lancet 349:1219–20 Michael NL, Nelson JAE, Kewalramani VN, Chang G, O’Brien SJ, Mascola JR, Volsky B, Louder M, White GC, Littman DR, Swanstrom R, O’Brien TR. 1998. Exclusive and persistent use of the entry coreceptor CXCR4 by human immunodeficiency virus type 1 from a subject homozygous for CCR5 Delta 32. J. Virol. 72:6040–47 Martinson JJ, Chapman NH, Rees DC, Liu YT, Clegg JB. 1997. Global distribution of the CCR5 gene 32-basepair deletion. Nature Genet 16:100–3 Stephens JC, Reich DE, Goldstein DB, Shin HD, Smith MW, Carrington M, Winkler C, Huttley GA, Allikmets R, Schriml L, Gerrard B, Malasky M, Ramos MD, Morlot S, Tzetis M, Oddoux

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

HIV-1 CORECEPTORS

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

166.

167.

168.

169.

170. 171.

172.

C, di Giovine FS, Nasiolas G, Chandler D, Aseev M, Hanson M, Kalaydjieva L, Glavac D, Gasparini P, Kanavakis E, Claustres M, Kambouris M, Ostrer H, Duff G, Baranov V, Sibul H, Metspalu A, Goldman D, Martin N, Duffy D, Schmidtke J, Estivill X, O’Brien SJ, Dean M. 1998. Dating the origin of the CCR5-Delta 32 AIDS-resistance allele by the coalescence of haplotypes. Am. J. Hum. Genet. 62:1507–15 Libert F, Cochaux P, Beckman G, Samson M, Aksenova M, Cao A, Czeizel A, Claustres M, de la Rua C, Ferrari M, Ferrec C, Glover G, Grinde B, Guran S, Kucinskas V, Lavinha J, Mercier B, Ogur G, Peltonen L, Rosatelli C, Schwartz M, Spitsyn V, Timar L, Beckman L, Parmentier M, Vassart G. 1998. The Delta ccr5 mutation conferring protection against HIV-1 in Caucasian populations has a single and recent origin in Northeastern Europe. Hum. Mol. Genet. 7:399–406 Simonsen JN, Fowke KR, MacDonald KS, Plummer FA. 1998. HIV pathogenesis: mechanisms of susceptibility and disease progression. Curr. Opin. Microbiol. 1:423–29 Zagury D, Lachgar A, Chams V, Fall LS, Bernard J, Zagury JF, Bizzini B, Gringeri A, Santagostino E, Rappaport J, Feldman M, O’Brien SJ, Burny A, Gallo RC. 1998. C-C chemokines, pivotal in protection against HIV type 1 infection. Proc. Natl. Acad. Sci. USA 95:3857–61 Paxton WA, Liu R, Kang S, Wu LJ, Gingeras TR, Landau NR, Mackay CR, Koup RA. 1998. Reduced HIV-1 infectability of CD4(+) lymphocytes from exposed-uninfected individuals: association with low expression of CCR5 and high production of beta-chemokines. Virology 244:66–73 Fauci AS. 1996. Host factors and the pathogenesis of HIV-induced disease. Nature 384:529–34 Eugen-Olsen J, Iversen AKN, Benfield TL, Koppelhus U, Garred P. 1998. Chemokine receptor CCR2b 64I polymorphism and its relation to CD4 T-cell counts and disease progression in a Danish cohort of HIV-infected individuals. J. Acquired Immune Defic. Syndr. 18:110– 16 Meyer L, Magierowska M, Hubert JB, Rouzioux C, Deveau C, Sanson F, Debre P, Delfraissy JF, Theodorou I, the SEROCO Study Group. 1997. Early protective effect of CCR-5 delta 32 heterozygosity on HIV-1 disease progres-

173.

174.

175.

176.

177.

178.

697

sion: relationship with viral load. AIDS 11:F73–F78 McDermott DH, Zimmerman PA, Guignard F, Kleeberger CA, Leitman SF, Murphy PM. 1998. CCR5 promoter polymorphism and HIV-1 disease progression. Lancet 352:866–70 Smith MW, Dean M, Carrington M, Winkler C, Huttley GA, Lomb DA, Goedert JJ, O’Brien TR, Jacobson LP, Kaslow R, Buchbinder S, Vittinghof E, Vlahov D, Hoots K, Hilgartner MW, Hemophilia Growth and Development Study (HGDS), Multicenter AIDS Cohort Study (MACS), Multicenter Hemophilia Cohort Study (MHCS), San Francisco City Cohort (SFCC), ALIVE Study, O’Brien SJ. 1997. Contrasting genetic influence of CCR2 and CCR5 variants on HIV-1 infection and disease progression. Science 277:959– 64 Kostrikis LG, Huang Y, Moore JP, Wolinsky SM, Zhang LQ, Guo Y, Deutsch L, Phair J, Neumann AU, Ho DD. 1998. A chemokine receptor CCR2 allele delays HIV-1 disease progression and is associated with a CCR5 promoter mutation. Nature Med. 4:350–53 Winkler C, Modi W, Smith MW, Nelson GW, Wu X, Carrington M, Dean M, Honjo T, Tashiro K, Yabe D, Buchbinder S, Vittinghoff E, Goedert JJ, O’Brien TR, Jacobson LP, Detels R, Donfield S, Willoughby A, Gomperts E, Vlahov D, Phair J, ALIVE Study, Hemophilia Growth and Development Study (HGDS), Multicenter AIDS Cohort Study (MACS), Multicenter Hemophilia Cohort Study (MHCS), San Francisco City Cohort (SFCC), O’Brien SJ. 1998. Genetic restriction of AIDS pathogenesis by an SDF-1 chemokine gene variant. Science 279: 389–93 Cohen OJ, Paolucci S, Bende SM, Daucher M, Moriuchi H, Moriuchi M, Cicala C, Davey RT, Baird B, Fauci AS. 1998. CXCR4 and CCR5 genetic polymorphisms in long-term nonprogressive human immunodeficiency virus infection: lack of association with mutations other than CCR5-Delta 32. J. Virol. 72:6215–17 Samson M, Soularue P, Vassart G, Parmentier M. 1996. The genes encoding the human CC-chemokine receptors CC-CKR1 to CC-CKR5 (CMKBR1CMKBR5) are clustered in the p21. 3p24 region of chromosome 3. Genomics 36:522–526

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

698

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER

179. Cohen OJ, Vaccarezza M, Lam GK, Baird BF, Wildt K, Murphy PM, Zimmerman PA, Nutman TB, Fox CH, Hoover S, Adelsberger J, Baseler M, Arthos J, Davey RT Jr, Dewar RL, Metcalf J, Schwartzentruber DJ, Orenstein JM, Buchbinder S, Saah AJ, Detels R, Phair J, Rinaldo C, Margolick JB, Fauci AS. 1997. Heterozygosity for a defective gene for CC chemokine receptor 5 is not the sole determinant for the immunologic and virologic phenotype of HIV-infected long-term nonprogressors. J. Clin. Invest. 100:1581–89 180. Blaak H, Vantwout AB, Brouwer M, Cornelissen M, Kootstra NA, Albrechtvanlent N, Keet RPM, Goudsmit J, Coutinho RA, Schuitemaker H. 1998. Infectious cellular load in human immunodeficiency virus type 1 (HIV-1)infected individuals and susceptibility of peripheral blood mononuclear cells from their exposed partners to nonsyncytium-inducing HIV-1 as major determinants for HIV-1 transmission in homosexual couples. J. Virol. 72:218–24 181. Picchio GR, Gulizia RJ, Mosier DE. 1997. Chemokine receptor CCR5 genotype influences the kinetics of human immunodeficiency virus type 1 infection in human PBL-SCID mice. J. Virol. 71:7124–27 182. Garred P, Eugen-Olsen J, Iversen AK, Benfield TL, Svejgaard A, Hofmann B. 1997. Dual effect of CCR5 delta 32 gene deletion in HIV-1-infected patients. Copenhagen AIDS Study Group. Lancet 349:1884 183. Rizzardi GP, Morawetz RA, Vicenzi E, Ghezzi S, Poli G, Lazzarin A, Pantaleo G. 1998. CCR2 polymorphism and HIV disease. Nature Med. 4:252–53 184. Michael NL, Louie LG, Rohrbaugh AL, Schultz KA, Dayhoff DE, Wang CE, Sheppard HW. 1997. The role of CCR5 and CCR2 polymorphisms in HIV-1 transmission and disease progression. Nature Med. 3:1160–62 185. Anzala AO, Ball TB, Rostron T, O’Brien SJ, Plummer FA, Rowland-Jones S, Univ. Nairobi Collab. HIV Res. 1998. CCR2-64I allele and genotype association with delayed AIDS progression in African women. Lancet 351:1632–33 186. Mellors JW, Rinaldo CR, Gupta P, White RM, Todd JA, Kingsley LA. 1996. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 272:1167–70 187. Lee B, Doranz BJ, Doms RW. 1998. Influence of the CCR2-V64I polymor-

188.

189.

190.

191.

192.

193.

194.

195.

196.

phism on human immunodeficiency virus type 1 coreceptor activity and on chemokine receptor function of CCR2b, CCR3, CCR5 and CXCR4. J. Virol. 72: 2509–15 Mummidi S, Ahuja SS, McDaniel BL, Ahuja SK. 1997. The human CC chemokine receptor 5 (CCR5) gene: multiple transcripts with 50 -end heterogeneity, dual promoter usage, and evidence for polymorphisms within the regulatory regions and noncoding exons. J. Biol. Chem. 272:30662–71 Carrington M, Kissner T, Gerrard B, Ivanov S, O’Brien SJ, Dean M. 1997. Novel alleles of the chemokine-receptor gene CCR5. Am. J. Hum. Genet. 61: 1261–67 Carroll RG, Riley JL, Levine BL, Feng Y, Kaushal S, Ritchey DW, Bernstein W, Weislow OS, Brown CR, Berger EA, June CH, St. Louis DC. 1997. Differential regulation of HIV-1 fusion cofactor expression by CD28 costimulation of CD4+ T cells. Science 276:273–76 Valentin A, Lu WH, Rosati M, Schneider R, Albert J, Karlsson A, Pavlakis GN. 1998. Dual effect of interleukin 4 on HIV-1 expression: implications for viral phenotypic switch and disease progression. Proc. Natl. Acad. Sci. USA 95:8886–91 Tedla N, Palladinetti P, Kelly M, Kumar RK, DiGirolamo N, Chattophadhay U, Cooke B, Truskett P, Dwyer J, Wakefield D, Lloyd A. 1996. Chemokines and T lymphocyte recruitment to lymph nodes in HIV infection. Am. J. Pathol. 148:1367–73 Trumfheller C, Tenner-Racz K, Racz P, Fleischer B, Frosch S. 1998. Expression of macrophage inflammatory protein (MIP)-1 alpha, MIP-1 beta, and RANTES genes in lymph nodes from HIV+ individuals: correlation with a Th1-type cytokine response. Clin. Exp. Immunol. 112:92–99 Margolis LB, Glushakova S, Grivel JC, Murphy PM. 1998. Blockade of CC chemokine receptor 5 (CCR5)-tropic human immunodeficiency virus-1 replication in human lymphoid tissue by CC chemokines. J. Clin. Invest. 101:1876– 80 D’Aquila RT, Sutton L, Savara A, Hughes MD, Johnson VA. 1998. CCRS/1 ccr5 heterozygosity: a selective pressure for the syncytium-inducing human immunodeficiency virus type 1 phenotype. J. Infect. Dis. 177:1549–53 Dolei A, Biolchini A, Serra C, Cur-

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

HIV-1 CORECEPTORS

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

197.

198.

199.

200.

201.

202.

203.

204.

reli S, Gomes E, Dianzani F. 1998. Increased replication of T-cell-tropic HIV strains and CXC-chemokine receptor-4 induction in T cells treated with macrophage inflammatory protein (MIP)-1 alpha, MIP-1 beta and RANTES beta-chemokines. AIDS 12: 183–90 Ostrowski MA, Justement SJ, Catanzaro A, Hallahan CA, Ehler LA, Mizell SG, Kumar PN, Mican JA, Chun TW, Fauci AS. 1998. Expression of chemokine receptors CXCR4 and CCR5 in HIV-1infected and uninfected individuals. J. Immunol. 161:3195–3201 Glushakova S, Baibakov B, Zimmerberg J, Margolis LB. 1997. Experimental HIV infection of human lymphoid tissue: correlation of CD4+ T cell depletion and virus syncytium-inducing/nonsyncytium-inducing phenotype in histocultures inoculated with laboratory strains and patient isolates of HIV type 1. AIDS Res. Hum. Retroviruses 13:461–71 Picchio GR, Gulizia RJ, Wehrly K, Chesebro B, Mosier DE. 1998. The cell tropism of human immunodeficiency virus type 1 determines the kinetics of plasma viremia in SCID mice reconstituted with human peripheral blood leukocytes. J. Virol. 72:2002–9 Weissman D, Rabin RL, Arthos J, Rubbert A, Dybul M, Swofford R, Venkatesan S, Farber JM, Fauci AS. 1997. Macrophage-tropic HIV and SIV envelope proteins induce a signal through the CCR5 chemokine receptor. Nature 389:981–85 Davis CB, Dikic I, Unutmaz D, Hill CM, Arthos J, Siani MA, Thompson DA, Schlessinger J, Littman DR. 1997. Signal transduction due to HIV-1 envelope interactions with chemokine receptors CXCR4 or CCR5. J. Exp. Med. 186:1793–98 Hesselgesser J, Taub D, Baskar P, Greenberg M, Hoxie J, Kolson DL, Horuk R. 1998. Neuronal apoptosis induced by HIV-1 gp120 and the chemokine SDF1 alpha is mediated by the chemokine receptor CXCR4. Curr. Biol. 8:595–98 Herbein G, Mahlknecht U, Batliwalla F, Gregersen P, Pappas T, Butler J, O’Brien WA, Verdin E. 1998. Apoptosis of CD8(+) T cells is mediated by macrophages through interaction of HIV gp120 with chemokine receptor CXCR4. Nature 395:189–94 Madani N, Kozak SL, Kavanaugh MP, Kabat D. 1998. gp120 Envelope gly-

205.

206.

207.

208. 209.

210.

211.

212.

213.

699

coproteins of human immunodeficiency viruses competitively antagonize signaling by coreceptors CXCR4 and CCR5. Proc. Natl. Acad. Sci. USA 95:8005–10 Ueda H, Howard OMZ, Grimm MC, Su SB, Gong WH, Evans G, Ruscetti FW, Oppenheim JJ, Wang JM. 1998. HIV-1 envelope gp41 is a potent inhibitor of chemoattractant receptor expression and function in monocytes. J. Clin. Invest. 102:804–12 Tiffany HL, Lautens LL, Gao J-L, Pease J, Locati M, Combadiere C, Modi W, Bonner TI, Murphy PM. 1997. Identification of CCR8: a human monocyte and thymus receptor for the CC chemokine I-309. J. Exp. Med. 186:165–70 Napolitano M, Zingoni A, Bernardini G, Spinetti G, Nista A, Storlazzi CT, Rocchi M, Santoni A. 1996. Molecular cloning of TER1, a chemokine receptorlike gene expressed by lymphoid tissues. J. Immunol. 157:2759–63 McCune JM. 1997. Thymic function in HIV-1 disease. Semin. Immunol. 9:397– 404 Arenzana-Seisdedos F, Virelizier J-L, Rousset D, Clark-Lewis I, Loetscher P, Moser B, Baggiolini M. 1996. HIV blocked by chemokine antagonist. Nature 383:400 Simmons G, Clapham PR, Picard L, Offord RE, Rosenkilede NM, Schwartz TW, Buser R, Wells TNC, Proudfoot AE. 1997. Potent inhibition of HIV-1 infectivity in macrophages and lymphocytes by a novel CCR5 antagonist. Science 276:276–79 Struyf S, De Meester I, Scharpe S, Lanaerts JP, Menten P, Wang JM, Proost P, Van Damme J. 1998. Natural truncation of RANTES abolishes signaling through the CC chemokine receptors CCR1 and CCR3, impairs its chemotactic potency and generates a CC chemokine inhibitor. Eur. J. Immunol. 28:1262–71 Clemons MJ, Marshall E, Durig J, Watanabe K, Howell A, Miles D, Earl H, Kiernan J, Griffiths A, Towlson K, DeTakats P, Testa NG, Dougal M, Hunter MG, Wood LM, Czaplewski LG, Millar A, Dexter TM, Lord BI. 1998. Randomized phase-II study of BB-10010 (macrophage inflammatory protein-1 alpha) in patients with advanced breast cancer receiving 5-fluorouracil, adriamycin, and cyclophosphamide chemotherapy. Blood 92:1532–40 Heveker N, Montes M, Germeroth L,

P1: SKH/spd/vks

P2: PSA/plb

March 3, 1999

9:53

700

214.

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

215.

216.

217.

218.

219.

220.

221.

QC: KKK/uks

Annual Reviews

T1: KKK

AR078-21

BERGER, MURPHY & FARBER Amara A, Trautman A, Alizon M, Schneider-Mergener J. 1998. Dissociation of the signalling and antiviral properties of SDF-1-derived small peptides. Curr. Biol. 8:369–76 Nishiyama Y, Murakami T, Kurita K, Yamamoto N. 1998. Synthesis of some peptides corresponding to the active region of RANTES for chemotaxis and evaluation of their anti-human immunodeficiency virus-1 activity. Chem. Pharmacolog. Bull. 45:2125–27 Doranz BJ, Grovit-Ferbas K, Sharron MP, Mao SH, Goetz MB, Daar ES, Doms RW, O’Brien WA. 1997. A smallmolecule inhibitor directed against the chemokine receptor CXCR4 prevents its use as an HIV-1 coreceptor. J. Exp. Med. 186:1395–1400 Schols D, Struyf S, Van Damme J, Este JA, Henson G, De Clercq E. 1997. Inhibition of T-tropic HIV strains by selective antagonization of the chemokine receptor CXCR4. J. Exp. Med. 186:1383– 88 Murakami T, Nakajima T, Koyanagi Y, Tachibana K, Fujii N, Tamamura H, Yoshida N, Waki M, Matsumoto A, Yoshie O, Kishimoto T, Yamamoto N, Nagasawa T. 1997. A small molecule CXCR4 inhibitor that blocks T cell linetropic HIV-1 infection. J. Exp. Med. 186:1389–93 Howard OMZ, Openheim JJ, Hollingshead MG, Covey JM, Bigelow J, McCormack JJ, Buckheit RW, Clanton DJ, Turpin JA, Rice WG. 1998. Inhibition of in vitro and in vivo HIV replication by a distamycin analogue that interferes with chemokine receptor function: a candidate for chemotherapeutic and microbicidal application. J. Med. Chem. 41:2184–93 Datema R, Rabin L, Hincenbergs M, Moreno MB, Warren S, Linquist V, Rosenwirth B, Seifert J, McCune JM. 1996. Antiviral efficacy in vivo of the anti-human immunodeficiency virus bicyclam SDZ SID 791 (JM 3100), an inhibitor of infectious cell entry. Antimicrob. Agents Chemother. 40:750–54 Levine BL, Cotte J, Carroll RG, Riley JL, Small CS, Francomano TN, Weislow OS, St. Louis DC, Bernstein W, June CH. 1997. A phase I dose-escalation study of polyclonal CD4 T cell ex vivo expansion for immune system restoration of HIV infection. J. Aller. Clin. Immunol. 99, Part 2, Suppl.:1625 Goila R, Banerjea AC. 1998. Sequence

222.

223.

224.

225.

226.

227.

228.

229.

230.

specific cleavage of the HIV-1 coreceptor CCR5 gene by a hammer-head ribozyme and a DNA-enzyme: inhibition of the coreceptor function by DNAenzyme. FEBS Lett. 436:233–38 Chen JD, Bai X, Yang AG, Cong Y, Chen SY. 1997. Inactivation of HIV1 chemokine co-receptor CXCR-4 by a novel intrakine strategy. Nature Med. 3:1110–16 Yang AG, Bai X, Huang XF, Yao C, Chen S. 1997. Phenotypic knockout of HIV type 1 chemokine coreceptor CCR5 by intrakines as potential therapeutic approach for HIV-1 infection. Proc. Natl. Acad. Sci. USA 94:11567–72 Mebatsion T, Finke S, Weiland F, Conzelmann KK. 1997. A CXCR4/CD4 pseudotype rhabdovirus that selectively infects HIV-1 envelope proteinexpressing cells. Cell 90:841–47 Schnell MJ, Johnson JE, Buonocore L, Rose JK. 1997. Construction of a novel virus that targets HIV-1-infected cells and controls HIV-1 infection. Cell 90:849–57 Endres MJ, Jaffer S, Haggarty B, Turner JD, Doranz BJ, O’Brien PJ, Kolson DL, Hoxie JA. 1997. Targeting of HIV- and SIV-infected cells by CD4chemokine receptor pseudotypes. Science 278:1462–64 Nagasawa T, Hirota S, Tachibana K, Takakura N, Nishikawa S, Kitamura Y, Yoshida N, Kikutnai H, Kishimoto T. 1996. Defects of Bcell lymphopoiesis and bone-marrow myelopoiesis in mice lacking the CXC chemokine PBSF/SDF-1. Nature 382: 635–38 Tachibana K, Hirota S, Iizasa H, Yoshida H, Kawabata K, Kataoka Y, Kitamura Y, Matsushima K, Nishikawa S, Kishimoto T, Nagasawa T. 1998. The chemokine rceptor CXCR4 is essential for vascularization of the gastrointestinal tract. Nature 393:591–94 Zou YR, Kottmann AH, Kuroda M, Taniuchi I, Littman DR. 1998. Function of the chemokine receptor CXCR4 in haematopoiesis and in cerebellar development. Nature 393:595–99 Martin MP, Dean M, Smith MW, Winkler C, Gerrard B, Michael NL, Lee B, Doms RW, Margolick J, Buchbinder S, Goedert JJ, O’Brien TR, Hilgartner MW, Vlahov D, O’Brien SJ, Carrington M. 1998. Genetic acceleration of AIDS progression by a promoter variant of CCR5. Science 282:1907–11

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:657-700. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999. 17:701–38 c 1999 by Annual Reviews. All rights reserved Copyright °

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE IL-4 RECEPTOR: Signaling Mechanisms and Biologic Functions Keats Nelms∗1, Achsah D. Keegan+, Jos´e Zamorano+, John J. Ryan∗2 and William E. Paul∗ ∗ Laboratory

of Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892 and +Department of Immunology, Jerome J. Holland Laboratories, American Red Cross, Rockville, MD 20855 KEY WORDS:

interleukin-4, interleukin-4 receptor, Stat-6, IRS-1/2, Janus kinases, phosphatase, FRIP

ABSTRACT Interleukin-4 is a multifunctional cytokine that plays a critical role in the regulation of immune responses. Its effects depend upon binding to and signaling through a receptor complex consisting of the IL-4Rα chain and the common gamma chain (γ c), resulting in a series of phosphorylation events mediated by receptor-associated kinases. In turn, these cause the recruitment of mediators of cell growth, of resistance to apoptosis, and of gene activation and differentiation. Here we describe our current understanding of the organization of the IL-4 receptor, of the signaling pathways that are induced as a result of receptor occupancy, and of the various mechanisms through which receptor function is modulated. We particularly emphasize the modular nature of the receptor and the specialization of different receptor regions for distinct functions, most notably the independent regulation of cell growth and gene activation.

INTRODUCTION Interleukin-4 is a pleiotropic type I cytokine produced by a subset of CD4+ T cells, designated TH2 cells, and by basophils and mast cells, in response to 1Present Address: Searle/Monsanto Corporation, 700 Chesterfield Parkway, AA4G, St. Louis, MO 63198; 2Virginia Commonwealth University, Dept Biology, 816 Park Avenue, Rm 202, Richmond, VA 23284

701 0732-0582/99/0410-0701$08.00

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

702

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

receptor-mediated activation events (1). IL-4 is also produced by a specialized subset of T cells (2), some of which express NK1.1 and appear to be specific for CD-1 (NK T cells) (3). γ /δ T cells have been reported to produce IL-4 (4), and mice lacking these cells fail to develop IL-4-dependent airway hypersensitivity upon immunization with ovalbumin in alum (5). Eosinophils have also been reported to be capable of producing IL-4 (6). IL-4 plays a central role in regulating the differentiation of antigen-stimulated naive T cells. IL-4 causes such cells to develop into cells capable of producing IL-4 and a series of other cytokines including IL-5, IL-10 and IL-13 (i.e. TH2like cells) (7, 8). It powerfully suppresses the appearance of IFNγ -producing CD4+ T cells. A second function of major physiologic importance is IL-4’s control of the specificity of immunoglobulin class switching. IL-4 determines that human B cells switch to the expression of IgE and IgG4 (9) and mouse B cells to IgE and IgG1(10, 11). Indeed, in IL-4 (12) and IL-4 receptor (13) knockout mice as well as in mice that lack a principal substrate of the IL-4 receptor, Stat-6 (14–16), IgE production is diminished by a factor of 100-fold or more. IL-4 receptor knockout mice (13) and Stat-6 knockout mice (16) are also deficient in the development of IL-4-producing T cells in mice infected with the helminthic parasite Nippostrongylus brasiliensis. These physiologic functions of IL-4 give it a preeminent role in the regulation of allergic conditions; it also plays a major role in the development of protective immune responses to helminths and other extracellular parasites. In experimental and clinical situations, it appears to be capable of ameliorating the effects of tissue-damaging autoimmunity (17). IL-4 has a variety of other effects in hematopoietic tissues. It increases the expression of class II MHC molecules in B cells (18), enhances expression of CD23 (19), upregulates the expression of the IL-4 receptor (20), and, in association with lipopolysaccharide, allows B cells to express Thy 1 (21). It also acts as a co-mitogen for B cell growth (22). Although not a growth factor by itself for resting lymphocytes, it can substantially prolong the lives of T and B lymphocytes in culture (23) and can prevent apoptosis by factor-dependent myeloid lines that express IL-4 receptors (24–28). IL-4 also has an important role in tissue adhesion and inflammation. It acts with TNF to induce expression of vascular cell adhesion molecule-1 (VCAM-1) on vascular endothelial cells (29), and it downregulates the expression of Eselectin (30). This shift in balance of expression of adhesion molecules by IL-4 is thought to favor the recruitment of T cells and eosinophils, rather than granulocytes, into a site of inflammation. An understanding of how IL-4 mediates this wide range of effects requires an analysis of the function of the IL-4 receptor. Here we review many aspects of the structure and function of the receptor, with particular emphasis on the

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

703

biochemical mechanisms through which it transmits signals. Such an analysis not only will be of relevance to the understanding of IL-4 receptor function but should also help to illuminate functions of other type I cytokine receptors and the receptors for other families of ligands.

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

THE IL-4 RECEPTOR COMPLEX IL-4 receptors are present in hematopoietic, endothelial, epithelial, muscle, fibroblast, hepatocyte and brain tissues and are usually expressed at 100 to 5000 sites per cell (31, 32), in keeping with the broad range of action of this cytokine. The receptor consists of a 140-kDa IL-4Rα chain (Figure 1) that binds IL-4 with high affinity (Kd 20 to 300 pM). Although artificial homodimerization of the IL-4Rα chain can result in the generation of biochemical signals within the cell (33–35), physiologic signaling depends upon IL-4-mediated heterodimerization of the IL-4Rα chain with a second chain. The gamma common chain (γ c), first identified as a component of the IL-2 receptor (36–38), appears to be the dominant chain involved in this heterodimerization in many cell types (Figure 1). Molecular binding studies have indicated that the γ c chain recognizes a complex of IL-4 and the IL-4Rα chain (39). Although the γ c chain only modestly increases the observed affinity of the IL-4R complex for IL-4, it is required for the activation of signaling pathways after binding of IL-4 (36). The IL-4Rα chain also functions as a component of the IL-13 receptor (IL-13R) (40–43). IL-13 appears not to utilize the γ c chain; rather, its receptor employs other cell surface polypeptides, the IL-13Rα and IL-13Rα 0 chains (42–45), presumably in place of γ c. A number of cell lines lacking γ c are IL-4 responsive (40, 46), raising the possibility that IL-13Rα and/or IL-13Rα 0 , which are expressed in these lines, may function, with the IL-4Rα chain, as components of the IL-4R complex. Indeed, recent studies indicate that the IL-13Rα 0 is the predominant accessory chain of the IL-4R complex in nonhematopoietic cells (43). IL-4Rα is a member of the hematopoietin receptor superfamily. Among the defining features of the members of this superfamily of receptors are shared structural motifs in the extracellular region, which consists of type III fibronectin domains (47). These motifs include conserved paired Cys residues and, in the membrane proximal region, a WSXWS motif. The latter has been proposed to be required for maintaining the receptor in a conformation favorable to cytokine binding (48). Structural alterations in the IL-4Rα extracellular region may result in altered receptor signaling capabilities. Indeed, a variant of the human IL-4Rα chain containing a Ile50Val substitution was isolated from atopic individuals and has been shown to enhance signal transduction resulting in the increased production of IgE (49).

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SKH/SPD P2: PSA/ARY

January 21, 1999

704

15:59

QC: PSA/UKS

NELMS ET AL

T1: PSA

Annual Reviews AR078-22

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

705

The murine IL-4Rα is 785 amino acids long with a 553 amino acid cytoplasmic region (50). The cytoplasmic region contains sequences found in other members of the hematopoietin receptor family as well as residues that are highly conserved between IL-4Rα chains of different species. In particular, there are five tyrosine residues within the IL-4Rα cytoplasmic region whose position and surrounding sequences are highly conserved, suggesting that these sequences are functionally important (Figure 1). A short proline-rich sequence in the membrane proximal region of the IL-4Rα, termed a “box1 motif,” is found in a number of hematopoietin receptor family members. A mutational analysis of the gp130 chain of the IL-6 receptor demonstrated the importance of this sequence for the function of the receptor (51). An acidic region adjacent to the box1 motif is similar to a region of the IL-2 receptor β that has been shown to interact with Src-family kinases (52).

Activation of Signal Transduction by the IL-4R Insight into the initiation of signal transduction by hematopoietin receptors has come from elegant structural studies of the growth hormone (GH) receptor (53, 54). These studies demonstrated that a single GH molecule cross-links two GH receptor molecules, resulting in the cross-activation of kinases associated with the cytoplasmic domain of the GH receptor. The erythropoietin (EPO) receptor and c-Kit, like the GH receptor, are homodimerized by their respective cytokine ligands. Studies of chimeric models made with the cytoplasmic domain of the IL-4Rα and the extracellular domains of the EPO or c-Kit receptors have also indicated that stimulation of cells expressing EPO- or c-Kit-IL-4Rα chimeras with their respective ligands induces IL-4R signaling pathways in cells. Thus, IL-4Rα cross-linkage appears capable of initiating signal transduction (33–35). As noted above, IL-4, rather than homodimerizing the IL-4Rα ←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 1 IL-4R structure and function. The IL-4R complex is composed of the IL-4Rα and γ c receptor subunits that associate with the Jak1 and Jak3 kinases, respectively. The cytoplasmic domain of the human IL-4Rα contains five conserved Tyr residues. The most membrane proximal, Y497, is within the I4R motif that is critical for generating proliferative signals. The second, third, and fourth Tyr residues (Y575, Y603, and Y631) are within a highly conserved sequence motif critical for the activation of Stat-6. The C-terminal Tyr, Y713, is within a ITIM motif that may serve as a docking site for different phosphatases. The ability or inability (−) of deletion (d437, d557, and d657) and point (Y497F, Y2,3,4F and Y5F) mutants of the IL-4Rα to fully activate (+) or to partially activate (+/−) cellular proliferation, gene expression, protection from apoptosis, or the IRS-1/2, Shc, FRIP, Stat-6 and SHIP signaling pathways is summarized. In two situations, Stat-6 phosphorylation and gene activation in response to occupancy of Y1F mutants, there is heterogeneity among stably transfected cell lines, with some lines displaying full activation and others displaying virtually none. Such differences are not determined by numbers of receptors and ∗ remain to be explained. These results are designated as + .

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

706

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

chain, causes hetereodimerization of this chain with the γ c chain leading to the activation of IL-4R signaling pathways (55). Ligand induced dimerization (or multimerization) of cytokine receptors results in the activation of tyrosine kinases that phosphorylate cellular substrates and initiate signaling cascades (47). Neither the IL-4Rα or the γ c chain has endogenous kinase activity; rather the IL-4R (and all members of the hematopoietin receptor family) require receptor-associated kinases for the initiation of signal transduction. The Janus-family (Jak) tyrosine kinases are critical in the initiation of signaling by hematopoietin receptors (56, 57). Three members of the Janus kinase family, Jak-1, Jak-2, and Jak-3, have been demonstrated to be activated in response to IL-4R engagement and to associate with components of the IL-4R complex (58–60). Jak-1 has been proposed to associate with the IL-4Rα chain while Jak-3 associates with the γ c chain (Figure 1) (61, 62). In certain cell lines, Jak-2 has also been demonstrated to associate with the IL-4Rα (60). IL-4 engagement of the IL-4Rα chain results in tyrosine phosphorylation of Jak-1 and Jak-3. Analysis of members of the Ba/F3 pro-B cell line expressing mutant human IL-4Rα chains suggests that the membrane proximal region containing the box1 motif and the acidic region may be required for IL-4-mediated responses (63). The importance of this region may reflect the fact that it is a potential site of interaction with Jak1. In addition to these Jak-family kinases, the Src-family kinase Fes has also been reported to associate with the IL-4Rα and to be activated in response to IL-4 stimulation (64). Activation of IL-4R-associated kinases leads to the tyrosine phosphorylation of the IL-4Rα chain itself, a process that occurs rapidly after IL-4R engagement (65). The five conserved Tyr residues in the cytoplasmic region of the IL-4Rα are potential sites of phosphorylation and of subsequent interaction with downstream signaling proteins through Src-homology 2 (SH2) or phosphotyrosinebinding (PTB) domains within these molecules. A critical point that remains to be clarified is which kinases actually catalyze the IL-4-induced phosphorylation of receptor tyrosines and of tyrosines on substrates that dock to the receptor. The identification of tyrosine residues critical for activation of signaling pathways and the subsequent analysis of molecules that interact with these residues have led to the biochemical characterization of pathways activated by IL-4R engagement. Independent studies in which truncation and deletion mutants of the human IL-4Rα chain were expressed in different hematopoietic cells showed that the region between residues 437 and 557, containing one conserved Tyr residue (Y497) (numbering according to 66), was required for IL-4-mediated activation of proliferation (Figure 1) (63, 67–69). Mutant receptors lacking this region were unable to transmit signals that normally result in phosphorylation of a set of key PTB-domain-containing substrates (see

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

707

below). In most instances, IL-4-mediated proliferative responses did not occur in factor-dependent myeloid progenitor cell lines (32D.IRS-1) expressing these mutant receptors; such responses were greatly diminished in Ba/F3 cells expressing these receptors. Although IL-4Rα sequences C-terminal to residue 557 do not appear to be essential for IL-4-stimulated proliferative responses, further analyses of deletion mutants have indicated that these sequences are important in induction of IL-4-responsive genes (Figure 1). In particular, the three conserved Tyr residues, Y575, Y603, and Y631, that lie within this region are critical for transducing signals that result in activation of a series of IL-4-induced genes (35, 70). Thus, on a first level of analysis, the IL-4Rα chain cytoplasmic region appears to have three functionally distinct domains, one that acts as an interaction site for the Janus kinase, one required for activation of proliferative pathways, and a third involved in the activation of pathways leading to induction of gene expression. Whether this represents an evolutionary process in which distinct segments mediating distinct functions are independently acquired is uncertain. The segregation of functions into distinct regions of the receptor would be consistent with this view. However, the regions of the receptor responsible for stimulation of proliferation and for gene activation are encoded in a single exon, which might suggest that the receptor achieved its current form long ago. In the next section, we discuss in more detail the signaling pathways initiated by IL-4 receptor engagement.

IL-4R SIGNALING PATHWAYS Initial experiments directed at characterizing the signaling pathways activated by IL-4R engagement compared the pattern of cellular proteins phosphorylated in response to IL-4 and IL-3 in hematopoietic cell lines (71). Strikingly, a 170-kDa phosphoprotein was uniquely phosphorylated in response to IL-4. In further studies this protein, initially termed the IL-4 phosphorylation substrate (4PS), was shown to be related to insulin receptor substrate-1 (IRS-1), the primary substrate phosphorylated in response to treatment of nonhematopoietic cells with insulin or IGF-1 (72, 73). The gene encoding the 4PS phosphoprotein has a high degree of homology to IRS-1; accordingly, IRS-2 was adopted as the formal designation for 4PS (74).

The IRS-1/2 Signaling Pathway The importance of IRS-1 and IRS-2 in responses to IL-4 was demonstrated using the factor-dependent myeloid progenitor cell line 32D; 32D cells do not express detectable levels of IRS-1 or IRS-2 (73, 74). Whereas other myeloid cell lines that expressed IRS-2 proliferated in response to IL-4 stimulation, 32D cell lines did not, suggesting a possible role for IRS-1/2 in IL-4-mediated

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

708

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

proliferation. Stable transfectants of 32D cells were prepared that expressed IRS-1 or IRS-2; these transfectants phosphorylated the IRS substrate they expressed when stimulated with IL-4 and showed IL-4-dependent cell growth. These observations led to the conclusion that IRS-1/2 molecules link the IL-4R to signaling pathways involved in cellular proliferation. INTERACTION OF IRS-1/2 WITH THE I4R MOTIF OF IL-4Rα How is IRS-1/2 activated in response to IL-4? The truncation mutants indicated the importance of the sequence between amino acids 437 and 557 of the IL-4Rα in this process. The sequence surrounding the single Tyr in this interval is 488PL-(x)4NPxYxSxSD502, which is highly homologous to sequences in the cytoplasmic regions of the insulin and IGF-1 receptors that also activate the IRS-1/2 signaling pathways (Figure 1) (67). Mutation of the central Tyr residue, Y497, of this motif, to Phe greatly diminished the ability of mutant receptors to signal proliferation in response to IL-4 and blocked IRS-1/2 phosphorylation (Figure 1) (67). Additionally, chimeric receptors consisting of a truncated IL-2 receptor β (IL-2Rβ) molecule linked to a protein segment containing the IL-4Rα I4R motif were able to activate IRS-1/2 phosphorylation on IL-2 stimulation, while the truncated IL-2Rβ alone did not, nor did a chimeric receptor expressing an I4R motif with a Y497F mutation (75). It had been previously demonstrated that mutation of the homologous Tyr (Y960) in the insulin receptor diminished insulin-directed cell activation (76). Thus, this sequence, being critical for transducing signals through the insulin receptor and the IL-4Rα, was termed the insulin IL-4 receptor, or I4R, motif. Moreover, the importance of this central Tyr suggested that the I4R motif, once phosphorylated at Y497, is a site of interaction with IRS-1/2. Direct evidence for an interaction between the phosphorylated I4R motif and IRS-1/2 molecules came from co-precipitation experiments and analysis of this interaction in the yeast two-hybrid system. Immunoprecipitation of the IL4Rα from FDC-P1 hematopoietic cells transfected with IRS-1 co-precipitated IRS-1 after IL-4 stimulation (67). Similarly a GST-fusion protein that expressed a 368 amino acid region of the IL-4Rα containing the I4R motif was capable of precipitating phosphorylated IRS-1 from IL-4-stimulated cell extracts (67). However, this GST-fusion protein precipitated IRS-1 less efficiently than did the full receptor, possibly due to inefficient phosphorylation of the fusion protein in cell extracts. Studies by Gustafson and colleagues utilizing a modified yeast-two hybrid system, in which the inclusion of the insulin receptor kinase in the bait allowed phosphorylation of the central tyrosine of the I4R motif, indicated that the IRS-1/2 molecules bind to phosphorylated I4R motifs through an N-terminal PTB domain (Figure 2A) (77, 78). This PTB domain has a three-dimensional

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

709

structure composed of a β–sandwich capped by a C-terminal α helix, similar to the PTB domain first described in the adapter molecule Shc (79–82). PTB domains bind to phosphopeptides with a core NxxY sequence, similar to that found in the I4R motif (83). Phosphopeptides derived from the I4R-motif of the IL-4Rα (LVIAGNPApYRS) inhibited the binding of a phosphopeptide derived from the I4R-motif of the insulin receptor (LYASSNPApYLSASDV) to the PTB domain of IRS-1 (84). The structural basis for the interaction between a phosphopeptide derived from the I4R-motif of the IL-4Rα and the IRS-1 PTB domain has been determined by nuclear magnetic resonance spectroscopy (81). This analysis showed that a cleft formed between the β–sandwich and the C-terminal α-helix of the IRS-1 PTB-domain served as the binding site for the phosphopeptide. The phosphopeptide (LVIAGNPApYR) inserts in the cleft rather like a shepherd’s crook. The NPApY sequence forms a Type I β turn making up the hook, and the amino terminal hydrophobic residues make up the extended cane that lies in a groove parallel to the α-helix. Amino acid residues at the −8 (L) and −6 (I) positions relative to Y497 in the I4R-motif make contact with residues in the PTB domain of IRS-1. A detailed mutational analysis of the I4R-motif in the human IL-4Rα confirmed the critical nature of L489 and I491 and identified additional residues necessary for regulating IL-4R signaling (85). Cell lines expressing the Y497F mutant consistently failed to activate the IRS-1/2 pathway and did not proliferate in response to IL-4, reflecting the requirement for I4R motif phosphorylation in PTB binding. Mutagenesis of P488 to A also greatly diminished the tyrosine phosphorylation of IRS-2 in response to IL-4 while mutation of a P488 to G resulted in a receptor competent to signal IRS-2 phosphorylation. The tolerance of the P to G change suggests that P488 controls the availability of the I4R-motif to PTB-domain containing proteins since this residue would lie just outside the binding cleft described by NMR. As predicted from the NMR structural data, mutation of both L489 and I491 to A also greatly diminished the tyrosine phosphorylation of IRS-2 to IL-4. However, mutation of only one of these residues to A did not affect signaling function, indicating that potential interactions from either the Leu or Ile is sufficient to make the receptor competent to recruit and phosphorylate IRS2 in these cells. These results indicate the important role of P488, L489, I491, and Y497 of the I4R-motif in regulating IRS-recruitment/activation. In contrast to the N-terminal residues of the I4R motif, mutagenesis of residues downstream of Y497 such as R498 or F500 to A had no effect on IL-4-induced biochemical or biological responses, although, as described below, these sequences may be important in the weak Stat-6 activating capacity of the I4R motif of the IL-4R.

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SKH/SPD P2: PSA/ARY

January 21, 1999

710

15:59

QC: PSA/UKS

NELMS ET AL

T1: PSA

Annual Reviews AR078-22

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

711

The N-terminal region of IRS-1/2 also contains a pleckstrin homology (PH) domain, similar to those found in a large number of signaling molecules. These PH domains have tertiary structures quite similar to that of PTB domains (80). Although the function of the PH domain is still being defined, it likely plays a role in localizing proteins to the plasma membrane by interacting directly with phosphatidylinositol membrane lipids that are generated through the activation of the phosphoinositide-3-kinase signaling pathway (discussed below) (86). IRS1/2 becomes phosphorylated as a result of interaction with phosphorylated IL-4Rα, presumably through the action of receptor-associated kinases (Figure 2). Indeed, in vitro experiments have indicated that Jak1, Jak2, and Jak3 are capable of directly phosphorylating IRS-1 (87). IRS-1/2 molecules are multiply phosphorylated in response to stimulation by a number of cytokines in addition to IL-4, including IL-2, IL-7, IL-9, and IL-15, indicating the presence of IRS-1/2 docking sites within these receptor complexes as well as the ability of different kinases to mediate IRS-1/2 phosphorylation (88). Analysis of cell lines lacking specific Jaks have indicated that Jak1 is critical for IL-4-stimulated induction of IRS-1 phosphorylation (89–91). This likely occurs through the direct action of Jak1 on IRS-1; however, this has not been proven directly. Jak2 and Tyk2 also mediate IRS-1 phosphorylation in certain cell lines (91). IRS-1/2 molecules each have approximately 20 potential sites for tyrosine phosphorylation (74, 92). A number of these sites are bound by specific SH2 domains indicating that IRS-1/2 act as cytosolic docking proteins capable of linking a variety of SH2 domain signaling molecules to phosphorylated receptors (74, 92, 93). Among the molecules that interact with phosphorylated IRS-1/2 molecules are the regulatory subunit of phosphoinositide-3-kinase (PI-3-K) and the adapter molecule, Grb-2 (Figure 2). These interactions lead to the activation of the PI-3-K and Ras/MAPK signaling pathways, respectively. THE PHOSPHOINOSITIDE-3-KINASE PATHWAY Several biochemically distinct forms of PI-3-K have been characterized, but the primary form activated by IL-4 is a complex of two subunits, a 85-kDa regulatory (p85) and a 110-kDa

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 2 Activation of signaling pathways through the I4R motif of the IL-4Rα. A. The PI-3kinase pathway can be activated through the interaction of IRS-1/2 molecules with the phosphorylated I4R motif of the IL-4Rα. This interaction leads to the phosphorylation of IRS-1/2 molecules by IL-4R associated kinases, the interaction of the p85 subunit of PI-3-kinase with IRS-1/2, the production of phosphoinositides, and the activation of downstream effectors (e.g. Akt, PKC). B. The phosphorylated I4R motif can act as a docking site for the adapter Shc. This may lead to the activation of small GTPases such as Ras in certain cell types. The activation of Ras by IL-4 is not seen consistently in all cell types and the importance of Shc and Ras activation in IL-4 proliferative responses remains to be fully delineated.

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

712

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

catalytic (p110) subunit (Figure 2A). The p85 subunit contains tandem SH2 domains in the C-terminus and an N-terminal SH3 domain (94, 95). The SH2 domains of the p85 subunit flank a 104 amino acid sequence that mediates the interaction with the p110 catalytic subunit (96, 97). The p85 subunit thus acts as an adapter molecule linking the p110 subunit to tyrosine phosphorylated molecules. IL-4 stimulation leads to the binding of the p85 subunit of PI-3-K to phosphorylated IRS-1/2 molecules (Figure 2A) (74, 93). IRS-1 and IRS-2 have four and ten potential sites of p85 subunit binding, respectively. Interaction of the p85 subunit with phosphorylated IRS-1/2 molecules results in a conformational change in the PI-3-K complex leading to the activation of the p110 catalytic subunit (98). The PI-3-K complex also interacts with Fes kinase after IL-4R engagement (99). Once activated, the p110 catalytic subunit is capable of phosphorylating membrane lipids as well as Ser/Thr residues of proteins (98). The lipid kinase activity mediates the transfer of phosphate from ATP to the D3 position of inositol in phosphotidylinositol in the cellular membrane (Figure 2A) (100, 101). Several forms of phosphorylated phosphotidylinositol have been identified, but the most important biologically appear to be phosphotidylinositol-(3,4,5)-triphosphate and phosphotidylinositol-(3,4)bisphosphate. These are produced within seconds of stimulation (100, 102). Their rapid production led to the hypothesis that these molecules act as second messenger molecules for IL-4 function. Indeed, phosphoinositides have since been implicated in the activation of a number of downstream kinases including different forms of protein kinase C (δ, ε, and η isozymes) and the Akt kinase (also known as protein kinase B) that play a key role in cell survival (Figure 2A) (103, 104). IL-4 has been demonstrated to enhance the survival of hematopoietic cells (24–26). Thus, it could be hypothesized that activation of the PI-3-K pathway by IL-4 may enhance cell survival through the production of phosphoinositides and the subsequent activation of kinases critical for cell survival. This hypothesis is supported by the finding that inhibitors of PI-3-K, such as Wortmannin, block the ability of IL-4 to prevent apoptosis in hematopoietic cells (26). In contrast to the lipid kinase activity of the p110 catalytic subunit of PI-3-K, the importance of the Ser/Thr kinase activity has not yet been fully defined. However, PI-3-K has been shown to catalyze the Ser/Thr phosphorylation of IRS- 1 (105). Since the Ser/Thr phosphorylation of IRS-1 has been suggested to diminish the interaction of IRS-1/2 with the I4R motif of the insulin receptor (106), it is possible that IRS-1/2 activation of PI-3-K may result in the Ser/Thr phosphorylation of IRS-1/2 by p110 and the inhibition of further IRS1/2 activation. Thus, activation of the PI-3-K Ser/Thr kinase activity may result in a negative feedback loop that contributes to the regulation of the IRS-1/2 signaling pathway.

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

713

THE Ras/MAPK PATHWAY Activation of the IRS-1/2 signaling pathway is associated with the activation the Ras/MAPK pathway in response to a number of cytokines, including insulin; IL-4 activation of the Ras/MAPK pathway is not consistently observed (107, 108). Phosphorylated IRS-1/2 has been proposed to interact with the SH2 domain of the adapter Grb2 (Figure 2A) (93). Grb2 is constitutively complexed to the guanine nucleotide exchange protein Sos (109). The primary function of Sos is to catalyze the exchange of GDP in inactive Ras for GTP, producing the active GTP-bound form of Ras (109, 110). IRS-1/2 interaction with phosphorylated receptors results in the accumulation of phosphorylated IRS-1/2 molecules at the cellular membrane where Ras is located. The subsequent interaction of phosphorylated IRS-/2 with Grb2/Sos also increases the concentration of Sos at the membrane, leading to activation of Ras. The MAPK pathway is initiated by the Ser/Thr kinase Raf following its activation by Ras-GTP. Although the mechanism of Raf activation is not yet understood, active Raf initiates a cascade of kinase activation events that ultimately result in the phosphorylation and activation of the mitogen activated protein kinases ERK-1 and ERK-2 (111, 112). Active ERK-1/2 translocates to the nucleus and activates the expression of genes such as c-fos by phosphorylating specific transactivating factors (113). Distinct kinases with functions similar to the ERKs, such as the Jun nuclear kinase (JNK), can be activated through kinase cascades initiated by Ras as well as by other small GTPases related to Ras (111). Activation of these kinases results in the nuclear phosphorylation and activation of c-Jun as well as of other transcription factors. As noted above, although IL-4 dramatically activates IRS-1/2 phosphorylation, IL-4 activation of the Ras/MAPK pathway is not consistently observed. In particular, stimulation of a number of hematopoietic cell lines with IL-4 failed to result in detectable activation of components of the Ras/MAPK pathway (108, 114, 115). Additionally, expression of the IL-4R in an L6 myoblast line enabled IRS-1 to be phosphorylated on stimulation with IL-4 and led to its association of Grb2/Sos (116). Nonetheless, MAPK and cellular proliferation were not activated in response to IL-4 in these myoblast lines, while insulin stimulation resulted in Grb2/Sos association with IRS-1, MAPK activation and proliferation. In these myoblast lines, phosphorylation of Shc, which also acts as an adapter molecule between receptors and Grb2/Sos, did correlate with the activation of the Ras/MAPK pathway. Insulin induced Shc phosphorylation whereas IL-4 did not. Thus, IRS-1 phosphorylation and association with Grb2/Sos is not sufficient for the activation of cellular proliferation by IL-4 in certain cells; activation of the Ras/MAPK pathway may require the activation of other signaling molecules such as Shc. Other studies have shown that IL-4 stimulation does lead to activation of the Ras/MAPK pathway and to Shc phosphorylation in certain cell types including

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

714

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

B cells and keratinocytes (Figure 2B) (117, 118). In addition, we have observed activation of Shc in the myeloid progenitor cell line 32D (119) and in the IL-4 responsive cell line CT.4R (K Nelms, unpublished data). Thus, activation of the Ras/MAPK pathway by IL-4 may critically depend on cell type and more specifically on the array of signaling molecules expressed in these cells. Recently, a homolog of the PTB-domain protein p62dok has been cloned using the yeast two-hybrid system with a bait consisting of the I4R motif from the IL-4Rα chain linked to the insulin receptor kinase (120). This molecule, designated the IL-four receptor interacting protein (FRIP), is rapidly phosphorylated in T cells and some myeloid cells stimulated with IL-4. FRIP, like p62dok, binds to the N-terminal SH2-domain of Ras-GAP. FRIP may thus link activated IL-4Rα molecules with RasGAP, leading to the activation of the endogenous GTPase activity of Ras and to the inactivation of the Ras pathway. FRIP is discussed in greater detail below.

Other Adapter Molecules As discussed above, Shc may play a pivotal role in the ability of certain cells to activate the Ras/MAPK pathway in response to IL-4. Shc shares some structural and functional characteristics with IRS-1/2. Shc contains two distinct domains capable of binding tyrosine-phosphorylated receptor sequences. The C-terminal region of the Shc protein contains an SH2 domain while the N-terminal domain contains a PTB domain (121, 122). The PTB and SH2 domains of Shc mediate its interaction with phosphorylated receptor molecules. The PTB domain of Shc is very similar in structure to the PTB domain of IRS-1/2 that mediates its interaction with the I4R motif of the IL-4Rα (82). Once this interaction occurs, Shc itself is phosphorylated at Tyr317, which then serves as a docking site for the SH2 domain of Grb2 (123, 124). In this way, Shc may link the Grb2/Sos complex to phosphorylated receptors and thus catalyze Ras activation (Figure 2B). In addition to the IL-4-mediated induction of Shc phosphorylation in certain cell types, a number of other cytokines and growth factors have been demonstrated to activate Shc phosphorylation (108, 117, 118). The adapter molecule Cbl, encoded by the proto-oncogene c-Cbl , may also play a role in the activation of signaling pathways by IL-4. Like IRS-1/2 and Shc, Cbl is phosphorylated in response to IL-4 (125) and cytokine-induced phosphorylation of Cbl has been demonstrated to link Grb2/Sos to receptor complexes and thus may play a role in activation of the Ras/MAPK pathway (125–127). IL-4-induced phosphorylation of Cbl also leads to its association with the p85 subunit of PI-3-K. Cbl has an N-terminal PTB domain that mediates its interaction with the ZAP-70 tyrosine kinase (128). Cbl also contains a proline-rich sequence that can interact with SH3 domains of different molecules (126, 127, 129).

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

715

The relative importance of the activation of Cbl phosphorylation in response to IL-4R engagement has not yet been determined. Likewise, the roles of newly identified members of the IRS signaling molecule family, IRS-3 and IRS-4, in IL-4R signaling processes have not yet been defined (130, 131). Because of the structural similarity and likely functional redundancy of IRS proteins, it will be of interest to evaluate the function of IRS-3 and IRS-4 in cells lacking the IRS-1 and IRS-2 molecules.

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Stat-6 Activation Pathway While the I4R motif region of the IL-4Rα is critical for activating pathways involved in regulating the proliferation of cells to IL-4, additional regions of the cytoplasmic tail are required for activation of IL-4-induced gene expression. Indeed, analyses of IL-4Rα deletion mutants have indicated that the region between residues 557 and 657 of the human IL-4Rα is critical for the induction of signaling pathways leading to the expression of IL-4-responsive genes (Figure 1) (70). M12.4.1 mouse B lymphoma cells expressing a truncated human IL-4Rα that terminates at residue 657 are as active as cells expressing the full-length receptor in IL-4-induced expression of CD23, class II MHC, or germline ε Ig H chain mRNA (Iε) (70). However, receptors truncated at residue 557 (1557) are greatly diminished in their ability to induce expression of these genes in response to challenge with human IL-4. These studies have been further supported by the demonstration that a chimeric receptor consisting of truncated IL-2Rβ coupled to IL-4Rα sequences from residues 557 to 657 induces CD23 expression in response to IL-2 (75). Thus, we have termed the IL-4Rα region between residues 557–657 the gene regulation domain. The gene regulation domain contains three conserved Tyr residues (Y575, Y603, and Y631), which represent potential sites of phosphorylation and subsequent association of SH2-containing proteins (Figure 1). Studies of IL-4Rα receptors specifically mutated at these Tyr residues have indicated that any one or any two can be mutated to Phe without ablating the capacity of the receptor to fully induce gene expression in response to IL-4; however, it is generally necessary to express substantially more mutant receptors than wild-type receptors to make M12.4.1 cells competent to optimally express IL-4-inducible genes. Mutation of all three Tyr residues results in a receptor with a very limited capability to activate gene expression (Figure 1). Thus, the gene regulation domain requires at least one functional Tyr residue for activity. An important development in understanding the mechanism by which IL-4 and other cytokines rapidly activate gene expression has been the identification and characterization of molecules termed signal transducers and activators of transcription, or Stats. One or more Stat molecules are activated by each

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

716

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

member of the hematopoeitin receptor superfamily and the related set of receptors for interferon-related molecules (132). Elegant experiments utilizing mutant cell lines that lack specific Jak kinases have shown that Jak activation is required for Stat activation (133). Thus, the Stat activation pathway is often referred to as the Jak-Stat pathway. Stat-6 is the primary Stat activated in response to IL-4 stimulation. It is critical in the activation or enhanced expression of many IL-4-responsive genes, including those for class II major histocompatibility molecules, CD23, germline immunoglobulin ε and γ 1, and IL-4Rα chain (14, 15, 35, 70, 134, 135). Stats act as direct connections between the cytokine receptor and the transcription apparatus. The mechanism of Stat-6 activation reflects the general model proposed for all Stat activation events (Figure 3). IL-4R engagement results in the activation of Jak1 and Jak3 and phosphorylation of specific tyrosine residues in the receptor cytoplasmic region. Stat-6 then binds to the phosphorylated receptor through a highly conserved SH2 domain, enabling the activated kinases to phosphorylate Stat-6 at a C-terminal tyrosine residue (132, 136). Once phosphorylated, the Stat-6 molecule disengages from the receptor and forms homodimers through interaction of its SH2 domain with the C-terminal phosphotyrosine residue of a second Stat-6 molecule. The dimerized Stat-6 complexes translocate to the nucleus where they bind to specific DNA motifs in the promoter of responsive genes. The DNA motifs bound by different STATs bear remarkable similarity to each other and reflect a dyad symmetry. Stat-6 appears to bind in particular to the sequence TTC-N4-GAA (137, 138). The importance of the Stat-6 activation pathway in the expression of IL-4responsive genes was examined using IL-4Rα mutants deficient in their ability to activate IL-4-responsive genes. The Stat-6 activation capability of the different IL-4Rα mutants matched the ability of these receptors to stimulate IL-4-responsive gene expression (Figure 1). In particular IL-4Rα mutants that lacked the gene regulation domain or had Tyr to Phe mutations at each of the Tyr residues in this region (Y575F, Y603F, and Y631F) induced little or no Stat-6 phosphorylation and DNA binding (70). This suggested that phosphorylation of Tyr residues in the gene regulation domain is critical for Stat-6 activation and −−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→ Figure 3 The Stat-6 activation pathway. After IL-4R engagement and phosphorylation, Stat-6 binds to phosphotyrosine residues in the gene regulation domain, becomes phosphorylated, disengages from the IL-4Rα cytoplasmic tail, dimerizes and translocates to the nucleus. Activation of gene transcription by Stat-6 may require cooperative interactions with additional transcription factors (e.g. C/EBPα) or phosphorylation by kinases activated in the Ras/MAP kinase cascade (e.g. ERK1/2). Alternately spliced forms of Stat-6 have deletions in the N-terminal (Stat-6b) or SH2 (Stat-6c) regions and may play a role in Stat-6 regulation. Stat-6c can act as a dominant negative form of Stat-6.

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

717

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

718

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

that activation of Stat-6 is a critical step leading to expression of IL-4-responsive genes. Each of the three Tyr residues in the gene regulation domain are equidistantly spaced 27 amino acids from one another. This spacing is conserved in the rat and mouse IL-4Rα chains, while the spacing between other Tyr residues is not (139). The residues immediately surrounding each Tyr possess a GYK/QXF sequence, even though there is little sequence conservation in the remainder of the gene regulation domain (Figure 1). The three conserved Tyr motifs in the gene regulation domain have been proposed to be docking sites for the SH2 domain of Stat-6. The central Tyr of the I4R motif is within a sequence, AYRSF, that is similar to those in the gene regulation domain. Indeed, this I4R motif sequence appears to be a weak Stat-6 binding site. The 1557 truncation mutant that lacks the gene activation domain nonetheless can signal weak Stat-6 and IL-4-responsive gene activation (Figure 1) (70). Replacing the AYRSF sequence in the 1557 truncation mutant with the sequence EYLSA, drawn from the insulin receptor I4R motif, results in the loss of the weak gene activating function of the mutant receptor. The requirements for forming a Stat-6 binding site were examined by replacing the NPAYRSF sequence surrounding Y497 in the 1557 truncation mutant with sequences derived from those surrounding Y575, Y603, and Y631; as shown above, each of these Ys is part of a sequence that conveys Stat-6 activation function to the IL-4Rα chain. Substituting the juxta-Y575 sequence (EAGYKAF) into the 1557 truncation mutant resulted in a receptor that had strikingly enhanced capacity to induce Stat-6 DNA-binding activity, CD23 expression, and class II MHC upregulation when expressed in M12.4.1 cells (140). This response, however, was significantly reduced in comparison to the response obtained with the full-length IL-4Rα. Furthermore, transfer of shorter sequences, such as GYKAF, did not confer Stat-6 activating potential, even though this “core” sequence contains those residues in which the three Stat-6 sites are homologous. These results suggest that the overall structure of the IL-4Rα chain or the presence of multiple Stat-6 binding sites within the gene activation domain is required for the full Stat-6-activating function. Indeed, transfer of this domain, containing all three Stat-6 sites in the appropriate context and spacing, to a truncated IL-2Rβ chain resulted in maximal activation of Stat-6 and CD23 in response to IL-2 (75) (AD Keegan, unpublished observations). The superiority of the longer sequence has not been fully explained. These studies thus indicate that the division of the IL-4Rα chain into “domains” that principally regulate growth (residues 437 to 557) and gene activation (residues 557 to 657), respectively, while largely correct is imperfect. Indeed, the IRS1/2 pathway leads to phosphorylation of the DNA-binding

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

719

protein HMG-I (Y) that participates in the regulation of Iε expression in response to IL-4 (141, 142). An additional complexity is the observation (85) that altering certain residues in the I4R motif (within the “growth domain”) affected not only the IRS-1/2 pathway, but also the Stat-6 activation pathway (largely a function of the “gene activation domain”). In particular, mutation of P488 to A in the full-length human IL-4Rα greatly diminished the tyrosine phosphorylation of Stat-6, as well as that of IRS-2, and abolished the induction of CD23 and Stat-6 DNA-binding activity in response to IL-4. In contrast, a P488G mutant was competent to signal these responses to IL-4. Mutating both L489 and I491 to A also diminished the tyrosine phosphorylation of Stat-6 and abolished induction of CD23 and of Stat-6 DNA-binding activity in response to human IL-4. These observations have not yet been completely explained. It is possible that changing residues in the I4R-motif disrupts the overall receptor structure. However, 125I-huIL-4 cross-linking data and Jak3 tyrosine phosphorylation studies show no gross alterations in the capacity of the receptor to bind ligand or to activate Jak3, making this possibility unlikely. Another possibility is that the changes in the I4R-motif disrupt important protein structures in the gene activation domain or make the Ys in this region unavailable to kinases. A third possibility is that the structure of the I4R-motif must be maintained to recruit PTB-domain containing proteins that participate in the recruitment and/or tyrosine phosphorylation of Stat-6. Deletion and mutational analyses of the Stat-6 protein itself have defined domains and residues within Stat-6 that are required for DNA binding and transcriptional activation. Deletions in the C-terminus of Stat-6 blocked its ability to activate transcription (136). Similar deletions in Stat-1 abrogate its ability to activate transcription (143). The effect of these C-terminal deletions may reflect the importance of this region for transcriptional activation or the presence of critical residues. Indeed, mutation of a Tyr residue (Y641) in this C-terminal region whose position is conserved between different STAT molecules also blocked Stat-6 function and DNA binding (136). This C-terminal Tyr residue in Stat-6 is thus predicted to be the site of Jak phosphorylation and the target of SH2 domains of other Stat-6 molecules. In addition to the C-terminal deletions, amino acid substitutions in the DNAbinding domain that blocked binding activity inhibited Stat-6 transcriptional activation (136). Similarly, mutation of a conserved Arg residue in the SH2 domain (R562), predicted to be critical for phosphotyrosine binding, also abolished activation of transcription and DNA binding by Stat-6 (136). This mutation likely prevents receptor interaction and dimerization of the mutated Stat-6 molecules. Naturally occurring deletion mutants of Stat-6, termed Stat-6b and Stat-6c, that result from alternate splicing have also been characterized (144). Stat-6b

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

720

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

contains an N-terminal truncation that attenuates but does not block its function, in agreement with in vitro deletion studies (136). Co-expression of Stat-6b with full-length Stat-6 did not alter the activation or function of the full-length Stat-6 molecule (144) as expected from other studies performed with N-terminal deletion mutants of Stat-6 (Figure 3) (136). In contrast to Stat-6b, Stat-6c contains a deletion in the SH2 domain and does not become phosphorylated in response to IL-4 stimulation. Nonetheless, transfection of Stat-6c prevented FDC-P2 cells from expressing I-Ad, CD16/CD32 and CD23 in response to IL-4 and diminished IL-4-mediated cellular proliferation. Stat-6c appeared to mediate these inhibitory functions by preventing the dimerization of full-length Stat-6 (Figure 3) (144). This is in contrast to studies done with a Stat-6 mutant containing a point mutation in the SH2 domain whose overexpression did not alter the activation of endogenous Stat-6 molecules (136). The mechanism by which Stat-6c inhibits Stat-6 activation as well as the importance of the Stat-6 splice variants in vivo remain to be elucidated. The exact mechanism by which Stats activate transcription is still being determined. It is likely that Stat molecules themselves activate the basic transcriptional machinery, but Stat molecules form complexes with other wellcharacterized transcription factors such as c-Jun and SP1 and thus may activate transcription through cooperative interaction with these factors (145, 146). Cooperative action with the transcription factor C/EBPα and NF-κB appears to be particularly important in the transcriptional activation of the immunoglobulin ε gene by Stat-6 (136, 147, 148). Additional signaling pathways may also contribute to general activation of Stat function as indicated by the observation that activation of the Ras/MAPK pathway is required for the full function of some Stat molecules (149, 150). In particular, a serine residue in the C-terminal region of Stat-1α was shown to be phosphorylated in response to activation of the Ras/MAPK pathway, presumably through the action of MAPKs such as ERK1/2 (149). Mutation of this serine resulted in a threefold reduction in Stat-1α-induced transcriptional activation. Serine phosphorylation of Stat-3 was required for optimal DNA binding (151). This is in contrast to the findings with Stat-1α, in which serine phosphorylation did not seem to alter DNA binding. Phosphorylation of Stat-6 on Ser/Thr residues has not yet been shown to play a role in Stat-6 function. However, it is likely that Ser/Thr phosphorylation is a general phenomenon in regulating Stat function, so its role in the regulation of Stat-6 should be carefully examined. The critical importance of Stat-6 activation in vivo has been demonstrated in Stat-6 knockout mice. These mice have undetectable serum levels of IgE and respond to infection with N. brasiliensis or injection of anti-IgD with increases in IgE that are less than 1% of wild-type mice. They fail to develop CD4+ T cells

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

721

of the TH2 type in response to N. brasiliensis infection (135). Furthermore, they fail to expel the parasite (152). In each respect, they resemble IL-4Rα chain knockout mice (13). When lymphocytes from Stat-6 knockout mice are studied in vitro, they fail to show switching to IgE and IgG1 in response to LPS plus IL-4; they also fail to show IL-4-mediated enhancement in expression of CD23, class II MHC molecules and IL-4 receptors. Furthermore, their CD4+ T cells fail to respond to immobilized anti-CD3 and IL-4 with the development of TH2 cells and IL-4 fails to prevent the differentiation of naive CD4+ T cells into IFNγ -producing cells. The specific immunodeficiency of Stat-6−/− mice likely results both from a block in IL-4-dependent Th2 cell development and from an inability of B cells to target the Cε gene for class switching both because IL-4 is not being produced and because Stat-6−/− cells would be insensitive to the switch-stimulating effects of IL-4 even if it were produced. Analyses of cells from Stat-6 deficient mice have shown that IL-4’s action as a co-mitogen for B and T cells is not ablated. IL-4 is able to protect B and T lymphocytes from Stat-6−/− mice from spontaneous apoptosis (153) (J Zamorano, J Austrian, H-Y Wang, AD Keegan, submitted for publication). However, depending upon the circumstances of the stimulation, lymphocytes from Stat-6 knockout mice can display a moderate or even striking diminution in IL-4-dependent DNA synthesis. The impairment of IL-4-mediated growth effects in cells from Stat-6 knockout mice may be a consequence of reduced expression of factors in Stat-6−/− cells required for IL-4-induced proliferation. Indeed, expression of the IL-4Rα chain (15) and IRS-2 (154) are diminished in Stat-6−/− cells. The altered expression of other factors in Stat-6−/− cells may also diminish the ability of these cells to proliferate to IL-4. In particular, IL-4 results in the accumulation of the cyclin-dependent kinase p27Kip1 in Con A-stimulated Stat-6−/− cells when compared to control cells (155). Increased levels of p27Kip1 lead to a decrease in cdk2-associated kinase activity and thus inhibit the progression of cells from the G1 to S phases of the cell cycle (155). In contrast to IL-4 stimulation, Stat-6−/− and control T cells proliferated similarly in response to IL-2 stimulation, indicating that levels of p27Kip1 are not elevated by all cytokines in Stat-6−/− cells. It is likely that the altered expression of these and other proteins in Stat-6−/− cells all contribute to the reduced level of proliferation of these cells to IL-4. A polymorphism of the human IL-4Rα chain, Q576R, in the core Stat-6binding sequence surrounding Y575, has recently been reported and found in 3 of 3 patients with the hyper-IgE syndrome and 4 of 7 patients with severe atopic dermatitis (156). Among 50 adults, it was present in 13 of 20 subjects with atopy and only 5 of 30 without atopy. Cells from individuals expressing

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

722

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

the R576 allele responded to IL-4 with higher levels of expression of CD23 than did cells from individuals homozygous for Q576. The association of the R576 allele with increased IgE production and enhanced induction of CD23 implies that the IL-4Rα chain containing R576 signals more vigorously upon IL-4 engagement. It might have been anticipated that a substitution of R for Q at 576 would enhance the activity of the sequence as a Stat-6 docking site. This expectation is based on the fact that the sequences surrounding Y603 and Y631 are GYKXF; these two sites were identified by Hou et al (157) in their studies of peptide inhibition of Stat-6 dimerization. Large peptides containing these core sequences inhibited Stat-6 dimerization, leading the authors to identify those sites as Stat-6 docking sites. They failed to detect the sequence surrounding Y575, which has a GYQXF core; Ryan et al were able to demonstrate that the Y575 site was a Stat-6 docking site by mutational analysis. Thus, the presence of a K (or the closely related residue R) rather than a Q in the Y+1 position might be anticipated to improve the Stat-6 activation function of the receptor. However, the authors of the report showed that the substitution of R for Q at this site diminished the affinity of the peptide for SHP-1, suggesting that the R575 form of the receptor might be less subject to the action of phosphatases and thus might signal more vigorously. Somewhat surprisingly, unpublished observations involving both R576Q and Y575F mutant receptors transfected into M12 lymphoma cells revealed no differences in their capacity to induce CD23 expression in response to IL-4 (AD Keegan, JJ Ryan, unpublished observations).

MODULATION OF IL-4 RECEPTOR SIGNALING PATHWAYS Recent evidence has emphasized the importance of regulatory pathways that function to modulate intracellular signals initiated by IL-4 and other cytokines. Just as the IRS-1/2 and Jak/Stat signaling pathways are activated through several different cytokine and growth factor receptors, certain negative regulatory pathways also appear to be involved in the regulation of signaling by different cytokine receptors.

General Signal Modulation: The Role of Phosphotyrosine Phosphatases The tyrosine phosphorylation and interaction of signaling proteins represent the foundation of many signaling pathways. General control of tyrosine phosphorylation of signaling molecules is accomplished through the action of phosphotyrosine phosphatases (PTP). The SH2-containing phosphatases SHP-1 and SHP-2 and the SH2-containing inositol-5-phosphatase (SHIP) have been recognized

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

723

to be critical modulators of cytokine signaling (158). Although specific roles played by SHP-1/2 and SHIP in IL-4 signaling have yet to be fully delineated, inhibition of phosphatase activity can result in Jak1 and Stat-6 activation, suggesting a modulatory role of these enzymes in IL-4R signaling (159). SHP-1 and SHP-2 are highly related and share a number of structural characteristics including tandem SH2 domains located in the N-terminal region. These SH2 domains are thought to be critical for linking SHP-1/2 to phosphorylated receptors and proteins, leading to their dephosphorylation. In particular, a specific sequence motif (8xYxxL), termed an immunoregulatory tyrosine-based inhibitory motif or ITIM, is found in the cytoplasmic domains of the Fcγ RIIb1 immunoglobulin receptor, KIR and CTLA-4 molecules and functions in the generation of negative regulatory signals by these receptors (158, 160–163). When phosphorylated, ITIMs serve as docking sites for the SH2 domains of SHP-1/2 and SHIP. The IL-4R also has a sequence related to the ITIM motif in its C-terminus that may play a role in regulating IL-4-stimulated signal transduction by interacting with SHP-1/2 or SHIP (Figure 4) (discussed below). Indeed, Marsh et al have reported that SHP-1 is associated with the IL-4Rα chain in unstimulated cells (164). SHP-2 has also been shown to constitutively associate with Jak1 and Jak3 and to co-precipitate with IRS-1, Grb2 and the p85 subunit of PI-3-kinase after cytokine stimulation (165, 166). However, SHP-2 did not associate with IRS-1 or IRS-2 after IL-4 treatment (167). Recent crystallographic evidence has indicated that the tandem SH2 domains of SHP-1/2 play a pivotal role in the regulation of SHP-1/2 activity. The structure of SHP-2 indicates that its N-terminal SH2 domain binds to the catalytic domain and in so doing blocks both the phosphatase active site and the phosphopeptide recognition site (168); the C-terminal SH2 domain is still capable of binding to phosphopeptides. Thus, the activation of SHP-2 has been proposed to require the sequential interaction of its SH2 domains with phosphoproteins. SHP-2 binding would be initiated through the interaction of the C-terminal SH2 domain with appropriate target sequences, localizing the N-terminal SH2 domain in the proximity of other tyrosine phosphorylated sequences. These phosphorylated sequences would then achieve a high enough local concentration to compete for binding to the N-terminal SH2, thus freeing the phosphatase active site and activating phosphatase activity. While SHP-2 is expressed in many tissues, SHP-1 is expressed primarily in hematopoietic tissue. The importance of SHP-1 in cytokine signaling has been indicated from analysis of mice homozygous for the motheaten allele that present marked hyperproliferation of hematopoietic cells (169). The motheaten phenotype results from point mutations in the SHP-1 gene that cause aberrant splicing of the SHP-1 transcript. Activation of macrophages from motheaten mice by interferon-α results in a dramatic increase in Jak1 phosphorylation

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SKH/SPD P2: PSA/ARY

January 21, 1999

724

15:59

QC: PSA/UKS

NELMS ET AL

T1: PSA

Annual Reviews AR078-22

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

725

leading to the suggestion that SHP-1 acts directly on this kinase (170). In contrast, the activity of a second Jak kinase activated by interferon-α, Tyk2, is not elevated in macrophages from motheaten mice. This demonstrates that SHP-1 is critical to the regulation of Jak-1 activity and suggests that SHP-1 may modulate the activity of Jak-1 that is induced by heterodimerization of the IL-4Rα chain and γ c as a result of binding of IL-4. In contrast to SHP-1/2 that act on phosphoproteins, SHIP acts on the 50 phosphates of PtdIns(3,4,5)P3 and thus appears to regulate the PI-3-kinase pathway by dephosphorylating the products of this enzyme (158). This phosphatase activity, however, does not necessarily result in the negative regulation of the PI-3-kinase pathway. In particular, the formation of PtdIns(3,4)P2, a critical activator of the anti-apoptotic kinase Akt, results from the dephosphorylation of PtdIns(3,4,5)P3 by SHIP (103). SHIP can interact with phosphorylated ITIM motifs such as that in the C-terminus of the IL-4Rα (158, 171). IL-4 will stimulate the tyrosine phosphorylation of SHIP (119). However, the Y713F mutant of the human IL-4Rα chain expresses the capacity to signal SHIP phosphorylation. IL-4 can also induce the association of SHIP with Shc as has been observed with other cytokines (172, 173). Therefore, SHIP could potentially be recruited to the IL-4 receptor complex by at least two different mechanisms, by direct docking to the ITIM site at Y713 or by indirect recruitment through binding to Shc at the I4R-motif (Y497). The function of the C-terminal ITIM of the IL-4Rα and, more generally, the role of phosphatases in IL-4 signaling pathways have yet to be delineated. Initial deletion studies indicated that the C-terminal region containing the ITIM motif was dispensable for short-term proliferation, gene induction, and tyrosine phosphorylation of Stat-6 and IRS-1/2 in response to IL-4 (67, 70). In contrast, more recent studies implicate the C-terminal region in certain aspects of IL-4 signaling. In particular, SHP-1 interacts with the IL-4Rα after IL-4 treatment and promotes the dephosphorylation of the p85 subunit of PI-3 kinase (164). Phosphopeptide pull-down experiments have indicated that the SHP-1 interaction with the IL-4Rα may involve Y575 in the gene regulation domain as well as Y713 in the C-terminal ITIM (156) (P Rothman, personal communication). ←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 4 Modulation of IL-4R signaling pathways. General modulation of IL-4R signaling pathways may result from the activation of phosphatases such as SHP-1, SHP-2, and SHIP that interact with phosphorylated Tyr residues of the IL-4Rα. These phosphatases may attenuate signals by dephosphorylating proteins such as the IL-4Rα. Activation of Stat-6 results in the expression of the SOCS/CIS/JAB/SSI-1 family of inhibitor proteins that attenuate the Jak-Stat pathway. Activation of small GTPases such as Ras can be modulated through the action of RasGAP, which is recruited to phosphorylated IL-4Rα through its interaction with the phosphorylated adapter FRIP.

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

726

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

In spite of these interaction data, receptors with Y575F or Y713F mutations do not show enhanced tyrosine phosphorylation of IRS-1/2 or Stat-6, as would be expected if SHP-1 docking to these sites was blocked (67, 70, 119). Additionally, Y713 and its surrounding amino acid sequence play a positive role in signaling the protection of cells from apoptosis by IL-4 (119). Thus, a precise delineation of the roles of phosphatases in regulating IL-4 signaling pathways and the mechanism of their activation by the IL-4R will require further investigation.

Negative Regulation of the Jak-Stat Pathway A second regulatory pathway has been described that plays a specific role in the modulation of the Jak-Stat pathway. The regulatory components of this pathway are a series of related SH2-domain proteins whose expression is induced in response to cytokine-induced Stat activation. These molecules, termed CIS (for cytokine-induced SH2), SOCS-1, 2, 3 (for suppressors of cytokine signaling), JAB (for Jak binding), and SSI-1 (for Stat-induced Stat-inhibitor), are expressed within 1 h of cytokine stimulation although the kinetics differed somewhat between the different genes (174–176). The level of expression of each molecule appears to differ depending on the activating cytokine. IL-4 in particular increases CIS and SOCS-3 expression predominantly in bone marrow–derived cells but also stimulates SOCS-1 and SOCS-2 expression to a lesser extent (174). SSI-1 was also induced by IL-4 in the CT.4S cell line (175). The mechanism by which the CIS/SOCS/JAB/SSI molecules act remains to be elucidated, but they appear to interact directly with and inhibit active Jaks (Figure 4). JAB in particular was cloned based on its capacity to bind the phosphorylated kinase domain of Jak2 (176). This interaction is likely to occur through the SH2 domain. The action of CIS/SOCS/JAB/SSI molecules appears to be specific for Janus kinases as evidenced by the fact that JAB expression diminished Jak1, Jak2, and Jak3 kinase activity while SOCS-1 and SSI-1 specifically abrogated the activation of Stat-3 and gp130 phosphorylation in response to IL-6 (174–176). In contrast, the overall induction of phosphorylation of cellular substrates was unaffected by CIS/SOCS/JAB/SSI expression, indicating that not all tyrosine kinases were inhibited. Since the expression of CIS/SOCS/JAB/SSI is dependent on Stat activation and results in the inactivation of Jaks, the CIS/SOCS/JAB/SSI pathway represents a classical negative feedback loop that specifically modulates the Jak-Stat activation pathway. It is has not yet been determined if CIS/SOCS/JAB/SSI proteins play a role in the regulation of IL-4R signaling pathways other than the Stat-6 activation pathway. Due to their Jak-inhibitory activity, CIS/SOCS/JAB/SSI expression might be hypothesized to result in a general downregulation of the IRS-1/2 and other pathways through inhibition of Jak-1/3. Indeed, it has been shown that

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

727

cell lines expressing a Jak-1 mutation fail to phosphorylate IRS-1 in response to IL-4 stimulation (89, 90). However, the contribution of other IL-4R-associated kinases such as Fes in the activation the IRS-1/2 and other pathways has not yet been elucidated. Thus, if Fes or other non-Jak kinases can function to initiate these IL-4R signaling pathways under certain circumstances or in certain cell types, expression of CIS/SOCS/JAB/SSI may not fully inhibit IL-4R signaling.

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Negative Regulation of the Ras/MAP Kinase Pathway The importance of the Ras/MAPK pathway in the response of cells to IL-4 has not yet been fully resolved. Both IL-4 and insulin stimulation result in the phosphorylation of IRS-2 and its interaction with the adapter molecule Grb2, which in turn can provide a link to the Ras pathway through its potential association with SOS. IL-4 activation of Ras appears to occur in a cell type– dependent manner. By contrast, insulin activation of Ras appears to be more general. It has not yet been determined whether IL-4 stimulates the activation of other small GTPases related to Ras such as Rho, Rac, and Rap1. A central protein involved in the regulation of small GTPases is the 120 kDa Ras GTPase activating protein (RasGAP) that binds to the active, GTP-bound form of Ras and activates its GTPase activity, catalyzing the formation of inactive Ras-GDP (177, 178). Because of its GTPase activating function, RasGAP is thought to function primarily as a negative regulator of Ras activation (110). An understanding of the processes that lead to RasGAP action on Ras has been enhanced by the identification of two molecules, p62dok and FRIP, that interact with RasGAP. The p62dok (downstream of kinases) molecule has long been observed as a 62-kDa phosphoprotein that co-precipitates with RasGAP. The Dok gene was cloned after purification of the p62dok protein from both Abelson murine leukemia virus (AbMuLV)-transformed cell lines and human chronic myelogenous leukemia cells (179, 180). p62dok has domains common to a number of adapter molecules such as N-terminal PH and PTB domains. In addition, it has a Cterminal region that contains consensus binding sites for the N-terminal SH2 domain of RasGAP (179, 180). The specific function of p62dok remains to be determined, but it has been shown to be phosphorylated in response to stimulation by cytokines including stem cell factor, IL-3 and IL-4 (120, 179). A second molecule highly homologous to p62dok was cloned based on its ability to bind to the phosphorylated I4R motif of the IL-4R in the yeast two-hybrid system. This molecule, termed FRIP (interleukin-Four Receptor Interacting Protein) has been demonstrated to be phosphorylated in response to different cytokines including IL-4, IL-3, IL-2, and insulin (120). FRIP is highly homologous to Dok, with a 35% overall amino acid identity and 48% identity in the PTB domain. This molecule was also cloned independently from

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

728

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

Bcr-Abl-transformed cells based on its interaction with RasGAP and was termed p56dok-2 (181). In contrast to p62dok, which is expressed in a wide variety of tissues, FRIP expression is limited to hematopoietic cells. It is expressed at particularly high levels in T cells and has not been detected in B cells. IL-4stimulation induced the tyrosine phosphorylation of FRIP; in extracts from IL-4-treated 32D myeloid progenitor cells, a GST-fusion protein containing the N-terminal SH2 domain of RasGAP precipitated phosphorylated FRIP, indicating that such stimulation could increase the interaction of FRIP with RasGAP. Activation of other cytokine receptors such as the IL-2, IL-3, and insulin receptors has also been demonstrated to activate FRIP phosphorylation, presumably due to the presence of NxxY PTB-domain docking motifs in the cytoplasmic domains of these receptors. Thus, FRIP may play a role in the regulation of signaling by a number of cytokines in addition to IL-4. Mutation of the central Tyr of the I4R motif in the IL-4R to Phe inhibited the IL-4-stimulated phosphorylation of FRIP. The mutation of two Arg residues in the PTB domain of FRIP also dramatically diminished FRIP phosphorylation and prevented its cytokine-induced interaction with RasGAP (K Nelms, unpublished observations). These two Arg residues are homologous to such residues in the PTB domain of IRS-1 that interact directly with the phosphotyrosine residue of the I4R motif of the IL-4R. Based on these results, we postulate that upon IL-4 stimulation, FRIP interacts with the phosphorylated I4R motif of IL-4Rα, becomes phosphorylated by receptor-associated tyrosine kinases, and is then bound by the N-terminal SH2 domain of RasGAP (Figure 4). In this way, FRIP can link RasGAP to activated receptor complexes. Such FRIP/RasGAP complexes may function by interacting locally with Ras-GTP and increasing its hydrolysis, thus inactivating the Ras/MAPK pathway. It remains to be determined whether these complexes function with FRIP still bound to the IL-4Rα chain or whether the complex diffuses away from the receptor and perhaps remains locally concentrated as a result of the binding of its PH domain to phosphoinositides deposited in the cell membrane as a result of the activation of PI-3 kinase by the receptor. Evidence supporting this model has come from the observation that the Frip gene is linked to the hairless locus on mouse chromosome 14 (120). T cells isolated from mice homozygous for the hairless allele (hr/hr) express three- to fivefold lower levels of FRIP mRNA and protein compared to control T cells. hr/hr mice develop splenomegaly, lymphadenopathy and leukemia (182). Their purified T cells respond to anti-CD3 plus IL-2 or IL-4 with a three to fivefold higher level of cytokine-induced proliferation in comparison to control (+/hr) T cells. Additionally, overexpression of FRIP in 32D and A.E7 cells decreased MAPK activation and Ras/MAPK pathway-dependent AP-1 transactivation in response to IL-2, respectively (120). Together, these observations strongly

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

729

suggest that FRIP plays an important role in negatively regulating cytokineinduced activation of the Ras/MAPK pathway. The contribution of FRIP to the regulation of other pathways regulated by other small GTPases remains to be determined. It also remains to be determined whether FRIP and p62dok function similarly or if the significant sequence differences in the C-terminal regions of these proteins lead to different functionalities. Additional mechanisms may be involved in the negative regulation of Ras activation in response to IL-4 and other cytokines. Evidence for one such mechanism has come from the study of a state of T cell nonresponsiveness, termed anergy, that is induced through stimulation of the T cell receptor in the absence of co-stimulation. Inhibition of Ras activation is important in maintaining T cell anergy and may result in part from the hyperphosphorylation of the Cbl adapter (183). Cbl hyperphosphorylation leads to its interaction with the Crk/C3G guanine-nucleotide exchange complex that activates the Ras-related protein Rap1 but not Ras. Like Ras, Rap1 is a small GTPase but is not associated with the plasma membrane. Rap1 inhibits activation of the Ras/MAPK pathway probably by acting as a cytoplasmic competitor for Ras effectors such as Raf. Thus, Cbl hyperphosphorylation can result in Rap1 hyperactivation and thus prevent cell proliferation by blocking the activation of the Ras/MAPK pathway. Since IL-4 also induces the phosphorylation of Cbl, it is possible that Cbl phosphorylation may contribute to the regulation of the Ras/MAPK pathway activated by IL-4. It is interesting to note that FRIP, like Cbl, is hyperphosphorylated in anergic T cells (K Nelms, J Powell, WE Paul, RH Schwartz, unpublished observations). Cbl hyperphosphorylation can account for inhibition of downstream activation of the Ras/MAPK pathway, but it does not account for the inability of anergic T cells to induce Ras activation as measured by accumulation of active Ras-GTP. Thus, hyperphosphorylation of FRIP in anergic T cells could result in heightened levels of membrane-associated RasGAP and specifically could inhibit Ras-GTP accumulation in anergic T cells.

CONCLUSION The recent and dramatic expansion in our knowledge of the mechanisms underlying cytokine signaling pathways has led to a better understanding of how cytokines elicit their diverse biological effects. The signaling pathways that are activated by IL-4R engagement, such as the IRS-1/2 and Jak-Stat pathways, mirror those activated by a number of other cytokines. Nevertheless, the activation of these pathways results in a unique array of cellular responses to IL-4. In the case of IL-4, specificity is in part achieved through the activation of Stat-6, an event that, among type I cytokine receptors, has been demonstrated

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

730

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL

to occur only through engagement of the IL-4Rα. An important challenge for the future will be to determine how the activation of similar signaling pathways by different cytokines results in varied biological responses. Specific cellular responses to IL-4 may also result from the unique character of the IL-4R. Indeed, the IL-4Rα appears to have a distinct domain structure that results in the activation of a specific array of signaling pathways. Functionally distinct domains of the IL-4Rα are required for IL-4 binding, the activation of receptor associated kinases, proliferative pathways, and gene expression. The association of particular functions with particular regions of the receptor suggests that the receptor may have acquired different functions evolutionarily by adding segments with particular functions. Genetic polymorphisms in two of the functional domains have been identified that result in heightened responsiveness to IL-4 and a susceptibility to atopy (49, 156). This emphasizes that gaining a fuller understanding of the signaling processes initiated by IL-4 can make a important contribution to determining the pathogenesis of allergic, anti-parasitic, and autoimmune diseases and may suggest potential opportunities for therapy. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Seder RA, Paul WE. 1994. Acquisition of lymphokine-producing phenotype by CD4+ T-cells. Annu. Rev. Immunol. 12:635–73 2. 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 3. Chen HJ, Paul WE. 1997. Cultured NK1.1(+)CD4(+) T cells produce large amounts of IL-4 and IFN-gamma upon activation by anti-CD3 or CD1. J. Immunol. 159:2240–49 4. Ferrick DA, Schrenzel MD, Mulvania T, Hsieh B, Ferlin WG, Lepper H. 1995. Differential production of interferongamma and interleukin-4 in response to TH1-stimulating and TH2-stimulating pathogens by gamma-delta T-cells in vivo. Nature 373:255–57 5. Zuany-Amorim C, Ruffie C, Haile S, Vargaftig BB, Pereira P, Pretolani M. 1998. Requirement for gamma delta T cells in allergic airway inflammation. Science 280:1265–67 6. Dubucquoi S, Desreumaux P, Janin A, Klein O, Goldman M, Tavernier J, Capron M, Capron A. 1994. Interleukin-5 se-

7.

8.

9.

10.

cretion by eosinophils–association with granules and immunoglobulin-dependent secretion. J. Exp. Med. 179:703–8 Seder RA, Paul WE, Davis MM, De ST, Groth BF. 1992. The presence of interleukin-4 during in vitro priming determines the lymphokine-producing potential of CD4+ T-cells from T-cell receptor transgenic mice. J. Exp. Med. 176: 1091–98 Hsieh CS, Heimberger AB, Gold JS, O’Garra A, Murphy KM. 1992. Differential regulation of T-helper phenotype development by interleukin-4 and interleukin-10 in an alpha-beta-T-cellreceptor transgenic system. Proc. Natl. Acad. Sci. USA 89:6065–69 Gascan H, Gauchat JF, Roncarolo MG, Yssel H, Spits H, de Vries JE. 1991. Human B cell clones can be induced to proliferate and to switch to IgE and IgG4 synthesis by interleukin 4 and a signal provided by activated CD4+ T cell clones. J. Exp. Med. 173:747–50 Coffman RL, Ohara J, Bond MW, Carty J, Zlotnik A, Paul WE. 1986. B cell stimulatory factor-1 enhances the IgE response of lipopolysaccharide-activated B cells. J. Immunol. 136:4538–41

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS 11. Vitetta ES, Ohara J, Myers CD, Layton JE, Krammer PH, Paul WE. 1985. Serological, biochemical, and functional identity of B cell-stimulatory factor 1 and B cell differentiation factor for IgG1. J. Exp. Med. 162:1726–31 12. Kuhn R, Rajewsky K, Muller W. 1991. Generation and analysis of interleukin-4 deficient mice. Science 254:707–10 13. Noben-Trauth N, Shultz LD, Brombacher F, Urban JF Jr, Gu H, Paul WE. 1997. An interleukin 4 (IL-4)-independent pathway for CD4+ T cell IL-4 production is revealed in IL-4 receptor-deficient mice. Proc. Natl. Acad. Sci. USA 94:10838– 43 14. Shimoda K, van Deursen J, Sangster MY, Sarawar SR, Carson RT, et al. 1996. Lack of IL-4-induced Th2 response and IgE class switching in mice with disrupted Stat-6 gene. Nature 380:630–33 15. Kaplan MH, Schindler U, Smiley ST, Grusby MJ. 1996. Stat-6 is required for mediating responses to IL-4 and for development of Th2 cells. Immunity 4:313– 19 16. Akira S, Kishimoto T. 1997. NF-IL6 and NF-kappa B in cytokine gene regulation. Adv. Immunol. 65:1–46 17. O’Garra A, Steinman L, Gijbels K. 1997. CD4(+) T-cell subsets in autoimmunity. Curr. Opin. Immunol. 9:872–83 18. Noelle R, Kramme RP, Ohara J, Uhr JW, Vitetta ES. 1984. Increased expression of Ia antigens on resting B cells: an additional role for B-cell growth factor. Proc. Natl. Acad. Sci. USA 81:6149–53 19. Defrance T, Aubry JP, Rousset F, Vanbervliet B, Bonnefoy JY, Arai N, Takebe Y, Yokota T, Lee F, Arai K, et al. 1987. Human recombinant interleukin 4 induces Fc epsilon receptors (CD23) on normal human B lymphocytes. J. Exp. Med. 165:1459–67 20. Ohara J, Paul WE. 1988. Up-regulation of interleukin 4/B-cell stimulatory factor 1 receptor expression. Proc. Natl. Acad. Sci. USA 85:8221–25 21. Snapper CM, Hornbeck PV, Atasoy U, Pereira GM, Paul WE. 1988. Interleukin 4 induces membrane Thy-1 expression on normal murine B cells. Proc. Natl. Acad. Sci. USA 85:6107–11 22. Howard M, Farrar J, Hilfiker M, Johnson B, Takatsu K, Hamaoka T, Paul WE. 1982. Identification of a T cell-derived B cell growth factor distinct from interleukin 2. J. Exp. Med. 155:914–23 23. Hu-Li J, Shevach EM, Mizuguchi J, Ohara J, Mosmann T, Paul WE. 1987. B cell stimulatory factor 1 (interleukin 4) is

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

731

a potent costimulant for normal resting T lymphocytes. J. Exp. Med. 165:157–72 Dancescu M, Rubio-Trujillo M, Biron G, Bron D, Delespesse G, Sarfati M. 1992. Interleukin 4 protects chronic lymphocytic leukemic B cells from death by apoptosis and upregulates Bcl-2 expression. J. Exp. Med. 176:1319–26 Illera VA, Perandones CE, Stunz LL, Mower J, Ir DA, Ashman RF. 1993. Apoptosis in splenic B lymphocytes: regulation by protein kinase C and IL-4. J. Immunol. 151:2965–73 Zamorano J, Wang HY, Wang LM, Pierce JH, Keegan AD. 1996. IL-4 protects cells from apoptosis via the insulin receptor substrate pathway and a second independent signaling pathway. J. Immunol. 157:4926–34 Boise LH, Minn AJ, June CH, Lindsten T, Thompson CB. 1995. Growth factors can enhance lymphocyte survival without committing the cell to undergo cell division. Proc. Natl. Acad. Sci. USA 92:5491– 95 Minshall C, Arkins S, Straza J, Conners J, Dantze R, Freund GG, Kelley KW. 1997. IL-4 and insulin-like growth factor-I inhibit the decline in Bcl-2 and promote the survival of IL-3-deprived myeloid progenitors. J. Immunol. 159:1225–32 Thornhill MH, Wellicoms SM, Mahiouz DL, Lanchbury JS, Kyan-Aung U, Haskard DO. 1991. Tumor necrosis factor combines with IL-4 or IFNgamma to selectively enhance endothelial cell adhesiveness for T cells. The contribution of vascular cell adhesion molecule1-dependent and -independent binding mechanisms. J. Immunol. 146:592–98 Bennett BL, Cruz R, Lacson RG, Manning AM. 1997. Interleukin-4 suppression of tumor necrosis factor alpha-stimulated E-selectin gene transcription is mediated by STAT-6 antagonism of NF-kappaB. J. Biol. Chem. 272:10212–19 Ohara J, Paul WE. 1987. Receptors for B-cell stimulatory factor-1 expressed on cells of haematopoietic lineage. Nature 325:537–40 Lowenthal JW, Castle BE, Christiansen J, Schreurs J, Rennick D, Arai N, Hoy P, Takebe Y, Howard M. 1988. Expression of high affinity receptors for murine interleukin 4 (BSF-1) on hemopoietic and nonhemopoietic cells. J. Immunol. 140:456– 64 Lai SY, Molden J, Liu KD, Puck JM, White MD, Goldsmith MA. 1996. Interleukin-4-specific signal transduction events are driven by homotypic interac-

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

732

34.

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

35.

36.

37.

38.

39.

40.

41.

42.

43.

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL tions of the interleukin-4 receptor alpha subunit. EMBO J. 15:4506–14 Fujiwara H, Hanissian SH, Tsytsykova A, Geha RS. 1997. Homodimerization of the human interleukin 4 receptor alpha chain induces Cepsilon germline transcripts in B cells in the absence of the interleukin 2 receptor gamma chain. Proc. Natl. Acad. Sci. USA 94:5866–71 Reichel M, Nelson BH, Greenberg PD, Rothman PB, Lu B, Reichel M, Fisher DA, Smith JF, Rothman P. 1997. The IL-4 receptor alpha-chain cytoplasmic domain is sufficient for activation of JAK-1 and STAT-6 and the induction of IL-4-specific gene expression. J. Immunol. 158:5860– 67 Russell SM, Keegan AD, Harada N, Nakamura Y, Noguchi M, Leland P, Friedmann MC, Miyajima A, Puri RK, Paul WE, et al. 1993. Interleukin-2 receptor gamma chain: a functional component of the interleukin-4 receptor. Science 262:1880–3 Leonard WJ, Noguchi M, Russell SM. 1994. Sharing of a common gamma chain, gamma c, by the IL-2, IL-4, and IL-7 receptors: implications for X-linked severe combined immunodeficiency (XSCID). Adv. Exp. Med. Biol. 365:225– 32 Kondo M, Takeshita T, Ishii N, Nakamura M, Watanabe S, Arai K, Sugamura K. 1993. Sharing of the interleukin-2 (IL-2) receptor gamma chain between receptors for IL-2 and IL-4. Science 262:1874–77 Letzelter F, Wang Y, Sebald W. 1998. The interleukin-4 site-2 epitope determining binding of the common receptor gamma chain. Eur. J. Biochem. 257:11-20 Obiri NI, Debinski W, Leonard WJ, Puri RK. 1995. Receptor for interleukin 13. Interaction with interleukin 4 by a mechanism that does not involve the common gamma chain shared by receptors for interleukins 2, 4, 7, 9, and 15. J. Biol. Chem. 270:8797–8804 Obiri NI, Leland P, Murata T, Debinski W, Puri RK. 1997. The IL-13 receptor structure differs on various cell types and may share more than one component with IL-4 receptor. J. Immunol. 158:756–64 Miloux B, Laurent P, Bonnin O, Lupker J, Caput D, Vita N, Ferrara P. 1997. Cloning of the human IL-13R alpha1 chain and reconstitution with the IL4R alpha of a functional IL-4/IL-13 receptor complex. FEBS Lett. 401:163–66 Murata T, Taguchi J, Puri RK. 1998. Interleukin-13 receptor alpha’ but not alpha chain: a functional component of

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

interleukin-4 receptors. Blood 91:3884– 91 Hilton DJ, Zhang JG, Metcalf D, Alexander WS, Nicola NA, Willson TA. 1996. Cloning and characterization of a binding subunit of the interleukin 13 receptor that is also a component of the interleukin 4 receptor. Proc. Natl. Acad. Sci. USA 93:497–501 Aman MJ, Tayebi N, Obiri NI, Puri RK, Modi WS, Leonard WJ. 1996. cDNA cloning and characterization of the human interleukin 13 receptor alpha chain. J. Biol. Chem. 271:29265–70 Dawson CH, Brown BL, Dobson PR. 1997. A 70-kDa protein facilitates interleukin-4 signal transduction in the absence of the common gamma receptor chain. Biochem. Biophys. Res. Commun. 233:279–82 Miyajima A, Kitamura T, Harada N, Yokota T, Arai K. 1992. Cytokine receptors and signal transduction. Annu. Rev. Immunol. 10:295–331 Livnah O, Stura EA, Johnson DL, Middleton SA, Mulcahy LS, Wrighton NC, Dower WJ, Jolliffe LK, Wilson IA. 1996. Functional mimicry of a protein hormone by a peptide agonist: the EPO receptor complex at 2.8 A [see comments]. Science. 273:464–71 Mitsuyasu H, Izuhara K, Mao X-Q, Gao P-S, Arinobu Y, Enomoto T, Kawai M, Sasaki S, Dake Y, Hamasaki N, Shirakawa T, Hopkin JM. 1998. Ile50Val variant of IL-4Ralpha upregulates IgE synthesis and associates with atopic asthma. Nature Med. 19:119–20 Mosley B, Beckmann MP, March CJ, Idzerda RL, Gimpel SD, VandenBos T, Friend D, Alpert A, Anderson D, Jackson J, et al. 1989. The murine interleukin-4 receptor: molecular cloning and characterization of secreted and membrane bound forms. Cell 59:335–48 Murakami M, Narazaki M, Hibi M, Yawata H, Yasukawa K, Hamaguchi M, Taga T, Kishimoto T. 1991. Critical cytoplasmic region of the interleukin 6 signal transducer gp130 is conserved in the cytokine receptor family. Proc. Natl. Acad. Sci. USA 88:11349–53 Minami Y, Kono T, Yamada K, Kobayashi N, Kawahara A, Perlmutter RM, Taniguchi T. 1993. Association of p56(lck) with IL-2 receptor beta chain is critical for the IL-2 induced activation of p56(lck). EMBO J. 12:759–68 de Vos AM, Ultsch M, Kossiakoff AA. 1992. Human growth hormone and extracellular domain of its receptor: crys-

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS

54.

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

55.

56. 57. 58.

59.

60.

61.

62.

63.

64.

tal structure of the complex. Science 255: 306–12 Cunningham BC, Ultsch M, De Vos AM, Mulkerrin MG, Clauser KR, Wells JA. 1991. Dimerization of the extracellular domain of the human growth hormone receptor by a single hormone molecule. Science 254:821–25 Kammer W, Lischke A, Moriggl R, Groner B, Ziemiecki A, Gurniak CB, Berg LJ, Friedrich K. 1996. Homodimerization of interleukin-4 receptor alpha chain can induce intracellular signaling. J. Biol. Chem. 271:23634–37 Ihle JN. 1995. The Janus protein tyrosine kinase family and its role in cytokine signaling. Adv. Immunol. 60:1–35 Taniguchi T. 1995. Cytokine signaling through nonreceptor protein tyrosine kinases. Science 268:251–55 Witthuhn BA, Silvennoinen O, Miura O, Lai KS, Cwik C, Liu ET, Ihle JN. 1994. Involvement of the Jak-3 Janus kinase in signalling by interleukins 2 and 4 in lymphoid and myeloid cells. Nature 370:153– 57 Johnston JA, Kawamura M, Kirken RA, Chen Y-Q, Blake TB, Shibuya K, Ortaldo R, McVicar DW, O’Shea JJ. 1994. Phosphorylation and activation of the Jak-3 kinase in response to interleukin-2. Nature 370:151–53 Murata T, Noguchi PD, Puri RK. 1996. IL-13 induces phosphorylation and activation of JAK2 Janus kinase in human colon carcinoma cell lines: similarities between IL-4 and IL-13 signaling. J. Immunol. 156:2972–78 Miyazaki T, Kawahara A, Fujii H, Nakagawa Y, Minami Y, Liu Zj, Oishi I, Silvennoinen O, Witthahn BA, Ihle JN, et al. 1994. Functional activation of Jak1 and Jak3 by selective association with IL-2 receptor subunits. Science 266:1045–47 Russell SM, Johnston JA, Noguchi M, Kawamura M, Bacon CM, Friedmann M, Berg M, McVicar DW, Witthuhn BA, Silvennoinen O, et al. 1994. Interaction of IL-2R beta and gamma-c chains with Jak1 and Jak3: implications for XSCID and XCID. Science 266:1042–45 Deutsch HH, Koettnitz K, Chung J, Kalthoff FS. 1995. Distinct sequence motifs within the cytoplasmic domain of the human IL-4 receptor differentially regulate apoptosis inhibition and cell growth. J. Immunol. 154:3696–3703 Izuhara K, Feldman RA, Greer P, Harada N. 1994. Interaction of the c-fes protooncogene product with the interleukin-4 receptor. J. Biol. Chem. 269:18623–29

733

65. Smerzbertling C, Duschl A. 1995. Both interleukin-4 and interleukin-13 induce tyrosine phosphorylation of the 140 kD subunit of the interleukin-4 receptor. J. Biol. Chem. 270:966–70 66. Galizzi JP, Zuber CE, Harada N, Gorman DM, Djossou O, Kastelein R, Banchereau J, Howard M, Miyajima A. 1990. Molecular cloning of a cDNA encoding the human interleukin 4 receptor. Int. Immunol. 2:669–75 67. Keegan AD, Nelms K, White M, Wang LM, Pierce JH, Paul WE. 1994. An IL-4 receptor region containing an insulin receptor motif is important for IL-4mediated IRS-1 phosphorylation and cell growth. Cell 76:811–20 68. Koettnitz K, Kalthoff FS. 1993. Human interleukin-4 receptor signaling requires sequences contained within two cytoplasmic regions. Eur. J. Immunol. 23:988–91 69. Seldin DC, Leder P. 1994. Mutational analysis of a critical signaling domain of the human interleukin 4 receptor. Proc. Natl. Acad. Sci. USA 91:2140–44 70. Ryan JJ, McReynolds LJ, Keegan A, Wang LH, Garfein E, Rothman P, Nelms K, Paul WE. 1996. Growth and gene expression are predominantly controlled by distinct regions of the human IL-4 receptor. Immunity 4:123–32 71. Wang L-M, Keegan AD, Paul WE, Heidaran MA, Gutkind JA, Pierce JH. 1992. IL-4 activates a distinct signal transduction cascade from IL-3 in factor-dependent myeloid cells. EMBO J. 11:4899–4908 72. Wang LM, Keegan AD, Li W, Lienhard GE, Pacini S, Gutkind JS, Myers MG, Sun XJ, White MF, Aaronson SA, et al. 1993. Common elements in interleukin-4 and insulin signaling pathways in factor-dependent hematopoietic cells. Proc. Natl. Acad. Sci. USA 90: 4032–36 73. Wang LM, Myers MG Jr, Sun XJ, Aaronson SA, White M, Pierce JH. 1993. IRS-1: essential for insulin- and IL-4-stimulated mitogenesis in hematopoietic cells. Science 261:1591–94 74. Sun XJ, Wang LM, Zhang Y, Yenush L, Myers MGJ, Glasheen E, Lane WS, Pierce JH, White MF. 1995. Role of IRS-2 in insulin and cytokine signalling. Nature 377:173–77 75. Wang HY, Paul WE, Keegan AD. 1996. IL-4 function can be transferred to the IL-2 receptor by tyrosine containing sequences found in the IL-4 receptor alpha chain. Immunity 4:113–21 76. White MF, Livingston JN, Backer JM,

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

734

77.

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

78.

79.

80.

81.

82.

83.

84.

85.

86.

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL Lauris V, Dull TJ, Ullrich A, Kahn CR. 1988. Mutation of the insulin receptor at tyrosine 960 inhibits signal transmission but does not affect its tyrosine kinase activity. Cell 54:641–49 Gustafson TA, He W, Craparo A, Schaub CD, O’Neill TJ. 1995. Phosphotyrosinedependent interaction of SHC and insulin receptor substrate 1 with the NPEY motif of the insulin receptor via a novel nonSH2 domain. Mol. Cell. Biol. 15:2500–8 O’Neill TJ, Craparo A, Gustafson TA. 1994. Characterization of an interaction between insulin receptor substrate 1 and the insulin receptor by using the twohybrid system. Mol. Cell. Biol. 14:6433– 42 Kavanaugh WM, Williams LT. 1994. An alternative to SH2 domains for binding to tyrosing phosphorylated proteins. Science 266:1862–65 Zhou M-M, Ravichandran KS, Olejniczak ET, Petros AM, Meadows RP, Harlan JE, Wade WS, Burakoff SJ, Fesik SW. 1995. Structure and ligand recognition of the phosphotyrosine binding domain of Shc. Nature 92:7784–88 Eck MJ, Dhe-Paganon S, Trub T, Nolte RT, Shoelson SE. 1996. Structure of the IRS-1 PTB domain bound to the juxtamembrane region of the insulin receptor. Cell 85:695–705 Zhou MM, Huang B, Olejniczak ET, Meadows RP, Shuker SB, Miyazaki M, Trub T, Shoelson SE, Fesik SW. 1996. Structural basis for IL-4 receptor phosphopeptide recognition by the IRS-1 PTB domain. Nat. Struct. Biol. 3:388–93 Trub T, Choi WE, Wolf G, Ottinger E, Chen Y, Weiss M, Shoelson SE. 1995. Specificity of the PTB domain of Shc for beta turn-forming pentapeptide motifs amino- terminal to phosphotyrosine. J. Biol. Chem. 270:18205–8 Wolf G, Trub T, Ottinger E, Groninga L, Lynch A, White M, Miyazaki M, Lee J, Shoelson SE. 1995. The PTB domains of IRS-1 and Shc have distinct but overlapping binding specificities. J. Biol. Chem. 270:27407–10 Wang HY, Zamorano J, Keegan AD. 1998. A role for the insulin-interleukin-4 receptor motif of the IL-4 receptor alphachain in regulating activation of the insulin receptor substrate-2 and signal transducer and activator or transcription 6 pathways. Analysis by mutagenesis. J. Biol. Chem. 273:9898–9905 Lemmon MA, Ferguson KM, Sigler PB, Schlessinger J. 1995. Specific and highaffinity binding of inositol phosphates to

87.

88.

89.

90.

91.

92.

93.

94. 95.

96.

an isolated pleckstrin homology domain. Proc. Natl. Acad. Sci. USA 92:10472–76 Yin TG, Keller SR, Quelle FW, Witthuhn BA, Tsang MLS, Lienhard GE, Ihle JN, Yang YC. 1995. Interleukin-9 induces tyrosine phosphorylation of insulin-receptor substrate-1 via Jak tyrosine kinases. J. Biol. Chem. 270:20497– 502 Johnston JA, Wang LM, Hanson EP, Sun XJ, White MF, Oakes SA, Pierce JH, O’Shea JJ. 1995. Interleukins 2, 4, 7, and 15 stimulate tyrosine phosphorylation of insulin receptor substrates 1 and 2 in T cells. Potential role of JAK kinases. J. Biol. Chem. 270:28527–30 Wang HY, Zamorano J, Yoerkie JL, Paul WE, Keegan AD. 1997. The IL-4-induced tyrosine phosphorylation of the insulin receptor substrate is dependent on JAK1 expression in human fibrosarcoma cells. J. Immunol. 158:1037–40 Chen XH, Patel BK, Wang LM, Frankel M, Ellmore N, Flavell RA, LaRochelle WJ, Pierce JH. 1997. Jak1 expression is required for mediating interleukin4-induced tyrosine phosphorylation of insulin receptor substrate and Stat-6 signaling molecules. J. Biol. Chem. 272: 6556–60 Burfoot MS, Rogers NC, Watling D, Smith JM, Pons S, Paonessaw G, Pellegrini S, White MF, Kerr IM. 1997. Janus kinase-dependent activation of insulin receptor substrate 1 in response to interleukin-4, oncostatin M, and the interferons. J. Biol. Chem. 272:24183–90 Sun XJ, Rothenberg P, Kahn CR, Backer JM, Araki E, Wilden PA, Cahill DA, Goldstein BJ, White MF. 1991. Structure of the insulin receptor substrate IRS-1 defines a unique signal transduction protein. Nature 352:73–77 Sun XJ, Crimmins DL, Myers MG, Miralpeix M, White MF. 1993. Pleiotropic insulin signals are engaged by multisite phosphorylation of IRS-1. Mol. Cell. Biol. 13:7418–28 Kapeller R, Cantley LC. 1994. Phosphatidylinositol 3-kinase. Bioessays 16:565– 76 Kapeller R, Prasad KV, Janssen O, Hou W, Schaffhausen BS, Rudd CE, Cantley LC. 1994. Identification of two SH3binding motifs in the regulatory subunit of phosphatidylinositol 3-kinase. J. Biol. Chem. 269:1927–33 Klippel A, Escobedo JA, Hu Q, Williams LT. 1993. A region of the 85-kilodalton (kDa) subunit of phosphatidylinositol 3kinase binds the 110-kDa catalytic sub-

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS unit in vivo. Mol. Cell. Biol. 13:5560–66 97. Dhand R, Hara K, Hiles I, Bax B, Gout I, Panayotou G, Fry MJ, Yonezawa K, Kasuga M, Waterfield MD. 1994. PI 3kinase: structural and functional analysis of intersubunit interactions. EMBO J. 13:511–21 98. Dhand R, Hiles I, Panayotou G, Roche S, Fry MJ, Gout I, Totty NF, Truong O, Vicendo P, Yonezawa K, et al. 1994. PI 3-kinase is a dual specificity enzyme: autoregulation by an intrinsic protein-serine kinase activity. EMBO J. 13:522–33 99. Izuhara K, Feldman RA, Greer P, Harada N. 1996. Interleukin-4 induces association of the c-fes proto-oncogene product with phosphatidylinositol-3 kinase. Blood 88:3910–18 100. Stephens LR, Jackson TR, Hawkins PT. 1993. Agonist-stimulated synthesis of phosphatidylinositol(3,4,5)-trisphosphate: a new intracellular signalling system? Biochim. Biophys. Acta 1179:27–75 101. Toker A, Cantley LC. 1997. Signalling through the lipid products of phosphoinositide-3- OH kinase. Nature 387:673–76 102. Auger KR, Serunian LA, Soltoff SP, Libby P, Cantley LC. 1989. PDGFdependent tyrosine phosphorylation stimulates production of novel polyphosphoinositides in intact cells. Cell 57:167–75 103. Franke TF, Kaplan DR, Cantley LC, To ker A. 1997. Direct regulation of the Akt proto- oncogene product by phosphatidylinositol-3,4-bisphosphate. Science 275:665–68 104. Franke TF, Kaplan DR, Cantley LC. 1997. PI3K: downstream AKTion blocks apoptosis. Cell 88:435–37 105. Lam K, Carpenter CL, Ruderman NB, Friel JC, Kelly KL. 1994. The phosphatidylinositol 3-kinase serine kinase phosphorylates IRS-1. Stimulation by insulin and inhibition by Wortmannin. J. Biol. Chem. 269:20648–52 106. Paz K, Hemi R, LeRoith D, Karasik A, Elhanany E, Kanety H, Zick Y. 1997. A molecular basis for insulin resistance. Elevated serine/threonine phosphorylation of IRS-1 and IRS-2 inhibits their binding to the juxtamembrane region of the insulin receptor and impairs their ability to undergo insulin-induced tyrosine phosphorylation. J. Biol. Chem. 272:29911–18 107. Duronio V, Welham MJ, Abraham S, Dryden P, Schrader JW. 1992. p21ras activation via hemopoietin receptors and c-kit requires tyrosine kinase activity but not tyrosine phosphorylation of p21ras GTPase-activating protein. Proc. Natl. Acad. Sci. USA 89:1587–91

735

108. Welham MJ, Duronio V, Leslie KB, Bowtell D, Schrader JW. 1994. Multiple hemopoietins, with the exception of interleukin-4, induce modification of Shc and mSos1, but not their translocation. J. Biol. Chem. 269:21165–76 109. Chardin P, Camonis JH, Gale NW, van Aelst L, Schlessinger J, Wigler MH, BarSagi D. 1993. Human Sos1: a guanine nucleotide exchange factor for Ras that binds to GRB2. Science 260:1338–43 110. Downward J. 1996. Control of ras activation. Cancer Surv. 27:87–100 111. Denhardt DT. 1996. Signal-transducing protein phosphorylation cascades mediated by Ras/Rho proteins in the mammalian cell: the potential for multiplex signalling. Bochem. J. 318:729–47 112. Marais R, Marshall CJ. 1996. Control of the ERK MAP kinase cascade by Ras and Raf. Cancer Surv. 27:101–25 113. Davis RJ. 1995. Transcriptional regulation by MAP kinases. Mol. Reprod. Dev. 42:459–67 114. Welham JM, Duriono V, Schrader JW. 1994. Interleukin-4 dependent proliferation dissociates p44 erk1, erk2, and p21 ras activation from cell growth. J. Biol. Chem. 269:5865–73 115. Welham MJ, Learmonth L, Bone H, Schrader JW. 1995. Interleukin-13 signal transduction in lymphohemopoietic cells. Similarities and differences in signal transduction with interleukin-4 and insulin. J. Biol. Chem. 270:12286–96 116. Pruett W, Yuan Y, Rose E, Batzger AG, Harada N, Sloknik EY. 1995. Association between GRB2/Sos and insulin receptor substrate 1 is not sufficient for activation of extracellular signal-regulated kinases by interleukin-4: implications for Ras activation by insulin. Mol. Cell. Biol. 15:1778–85 117. Wery S, Letourneur M, Bertoglio J, Pierre J. 1996. Interleukin-4 induces activation of mitogen-activated protein kinase and phosphorylation of shc in human keratinocytes. J. Biol. Chem. 271:8529–32 118. Crowley MT, Harmer SL, DeFranco AL. 1996. Activation-induced association of a 145-kDa tyrosine-phosphorylated protein with Shc and Syk in B lymphocytes and macrophages. J. Biol. Chem. 271:1145– 52 119. Zamorano J, Keegan AD. 1998. Regulation of apoptosis by tyrosine-containing domains of IL-4R alpha: Y497 and Y713, but not the Stat-6-docking tyrosines, signal protection from apoptosis. J. Immunol. 161:859–67 120. Nelms K, Snow AL, Hu-Li J, Paul WE.

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

736

121. 122.

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

123.

124.

125.

126.

127.

128.

129.

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL 1998. FRIP, a hematopoietic cell-specific rasGAP interacting protein phosphorylated in response to cytokine stimulation. Immunity 9:13–24 Pawson T. 1995. Protein modules and signalling networks. Nature 373:573–80 Songyang Z, Shoelson SE, Chaudhuri M, Gish G, Pawson T, Haser WG, King F, Roberts T, Ratnofsky S, Lechleider RJ, et al. 1993. SH2 domains recognize specific phosphopeptide sequences. Cell 72: 767–78 Rozakis-Adcock M, McGlade J, Mbamalu G, Pelicci G, Daly R, Li W, Batzer A, Thomas S, Brugge J, Pelicci PG, et al. 1992. Association of the Shc and Grb2/Sem5 SH2-containing proteins is implicated in activation of the Ras pathway by tyrosine kinases. Nature 360:689– 92 Salcini AE, McGlade J, Pelicci G, Nicoletti I, Pawson T, Pelicci PG. 1994. Formation of Shc-Grb2 complexes is necessary to induce neoplastic transformation by overexpression of Shc proteins. Oncogene 9:2827–36 Ueno H, Sasaki K, Honda H, Nakamoto T, Yamagata T, Miyagawa K, Mitani K, Yazaki Y, Hirai H. 1998. c-Cbl is tyrosinephosphorylated by interleukin-4 and enhances mitogenic and surival signals of interleukin-4 receptor by linking with the phosphatidylinositol 30 -kinase pathway. Blood 91:46–53 Fukazawa T, Reedquist KA, Trub T, Soltoff S, Panchamoorthy G, Druker B, Cantley L, Shoelson SE, Band H. 1995. The SH3 domain-binding T cell tyrosyl phosphoprotein p120. Demonstration of its identity with the c-cbl protooncogene product and in vivo complexes with Fyn, Grb2, and phosphatidylinositol 3-kinase. J. Biol. Chem. 270:19141–50 Panchamoorthy G, Fukazawa T, Miyake S, Soltoff S, Reedquist K, Druker B, Shoelson S, Cantley L, Band H. 1996. p120cbl is a major substrate of tyrosine phosphorylation upon B cell antigen receptor stimulation and interacts in vivo with Fyn and Syk tyrosine kinases, Grb2 and Shc adaptors, and the p85 subunit of phosphatidylinositol 3-kinase. J. Biol. Chem. 271:3187–94 Lupher ML Jr, Reedquist KA, Miyake S, Langdon WY, Band H. 1996. A novel phosphotyrosine-binding domain in the N-terminal transforming region of Cbl interacts directly and selectively with ZAP70 in T cells. J. Biol. Chem. 271:24063–68 Fukazawa T, Miyake S, Band V, Band H. 1996. Tyrosine phosphorylation of

130.

131.

132. 133.

134.

135. 136.

137.

138. 139.

140.

141.

Cbl upon epidermal growth factor (EGF) stimulation and its association with EGF receptor and downstream signaling proteins. J. Biol. Chem. 271:14554–59 Lavan BE, Fantin VR, Chang ET, Lane WS, Keller SR, Lienard Ge. 1997. A novel 160-kDa phosphotyrosine protein in insulin-treated embryonic kidney cells is a new member of the insulin receptor substrate family. J. Biol. Chem. 272: 21403–7 Lavan BE, Lane WS, Lienard GE. 1997. The 60-kD phosphotyrosine protein in insulin-treated adipocytes is new member of the insulin receptor substrate family. J. Biol. Chem. 272:11439–43 Darnell JE. 1997. STATs and gene regulation. Science 277:1630–35 Velazquez L, Fellous M, Stark GR, Pellegrini S. 1992. A protein tyrosine kinase in the interferon alpha/beta signaling pathway. Cell 70:313–22 Delphin S, Stavnezer J. 1995. Regulation of antibody class switching to IgE: characterization of an IL-4-responsive region in the immunoglobulin heavy-chain. Takeda K, Tanaka T, Shi W, Matsumoto M, Minami M, Kashiwamura S, germline epsilon promoter. Ann. NY Acad. Sci. 764:123–35 Nakanishi K, Yoshida N, Kishimoto T, Akira S. 1996. Essential role of Stat-6 in IL-4 signalling. Nature 380:627–30 Mikita T, Campbell D, Wu PG, Williamson K, Schindler U. 1996. Requirements for interleukin-4-induced gene expression and functional characterization of Stat-6. Mol. Cell. Biol. 16:5811–20 Schindler U, Wu P, Rothe M, Brasseur M, McKnight SL. 1995. Components of a Stat recognition code: evidence for two layers of molecular selectivity. Immunity 2:689–97 Ihle JN. 1996. STATs: signal transducers and activators of transcription. Cell 84:331–34 Richter G, Hein G, Blakenstein T, Diamantstein T. 1995. The rat interleukin 4 receptor: coevolution of ligand and receptor. Cytokine 7:237–41 Ryan JJ, McReynolds LJ, Huang H, Nelms K, Paul WE. 1998. Characterization of a mobile Stat-6 motif in the human IL-4 receptor. J. Immunol. 161:1811– 21 Kim J, Reeves R, Rothman P, Boothby M. 1995. The non-histone chromosomal protein HMG-I(Y) contributes to repression of the immunoglobulin heavy chain germ-line epsilon RNA promoter. Eur. J. Immunol. 25:798–808

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

IL-4 RECEPTOR SIGNALING MECHANISMS AND FUNCTIONS 142. Wang D, Zamorano J, Keegan AD, Boothby M. 1997. HMG-I(Y) phosphorylation status as a nuclear target regulated through insulin receptor substrate-1 and the I4R motif of the interleukin-4 receptor. J. Biol. Chem. 272:25083–90 143. Muller M, Laxton C, Briscoe J, Schindler C, Improta T, Darnell JE Jr, Stark GR, Kerr IM. 1993. Complementation of a mutant cell line: central role of the 91 kDa polypeptide of ISGF3 in the interferonalpha and -gamma signal transduction pathways. EMBO J. 12:4221–28 144. Patel BKR, Pierce JH, LaRochelle WJ. 1998. Regulation of interleukin 4mediated signaling by naturally occurring dominant negative and attenuated forms of human Stat-6. Proc. Natl. Acad. Sci. USA 95:172–77 145. Schaefer TS, Sanders LK, Nathans D. 1995. Cooperative transcriptional activity of Jun and Stat3 beta, a short form of Stat3. Proc. Natl. Acad. Sci. USA 92: 9097–9101 146. Look DC, Pelletier MR, Tidwell RM, Roswit WT, Holtzman MJ. 1995. Stat1 depends on transcriptional synergy with Sp1. J. Biol. Chem. 270:30264–67 147. Delphin S, Stavnezer J. 1995. Characterization of an interleukin 4 (IL-4) responsive region in the immunoglobulin heavy chain germline epsilon promoter: regulation by NF-IL-4, a C/EBP family member and NF-kappa B/p50. J. Exp. Med. 181:181–92 148. Shen CH, Stavnezer J. 1998. Interaction of Stat-6 and NF-kappaB: direct association and synergistic activation of interleukin-4-induced transcription. Mol. Cell. Biol. 18:3395–3404 149. Wen Z, Zhong Z, Darnell JE, Jr. 1995. Maximal activation of transcription by Stat1 and Stat3 requires both tyrosine and serine phosphorylation. Cell 82:241– 50 150. David M, Petricoin Er, Benjamin C, Pine R, Weber MJ, Larner AC. 1995. Requirement for MAP kinase (ERK2) activity in interferon alpha- and interferon betastimulated gene expression through STAT proteins. Science 269:1721–23 151. Zhang X, Blenis J, Li HC, Schindler C, Chen-Kiang S. 1995. Requirement of serine phosphorylation for formation of STAT-promoter complexes. Science 267:1990–94 152. Urban JF Jr, Noben-Trauth N, Donaldson DD, Madden KB, Morris SC, Collins M, Finkelman FD. 1998. IL-13, IL-4Ralpha, and Stat-6 are required for the expulsion of the gastrointestinal nematode par-

153.

154. 155.

156.

157.

158.

159.

160.

161.

162.

163.

164.

737

asite Nippostrongylus brasiliensis. Immunity 8:255–64 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 Huang H, Paul WE. 1998. Impaired interleukin 4 signaling in T helper type 1 cells. J. Exp. Med. 187:1305–13 Kaplan MH, Daniel C, Schindler U, Grusby MJ. 1998. Stat proteins control lymphocyte proliferation by regulating p27(Kip1) expression. Mol. Cell. Biol. 18:1996–2003 Hershey GK, Friedrich MF, Esswein LA, Thomas ML, Chatila TA. 1997. The association of atopy with a gain-of-function mutation in the alpha subunit of the interleukin-4 receptor. N. Engl. J. Med. 337:1720–25 Hou JZ, Schindler U, Henzel WJ, Ho TC, Brasseur M, McKnight SL. 1994. An interleukin-4-induced transcription factor-IL-4 Stat. Science 265:1701–6 Scharenberg AM, Kinet JP. 1996. The emerging field of receptor-mediated inhibitory signaling: SHP or SHIP? Cell 87:961–64 Haque SJ, Wu Q, Kammer W, Friedrich K, Smith JM, Kerr IM, Stark GR, Williams BRG. 1997. Receptor-associated constitutive protein tyrosine phosphatase activity controls the kinase function of JAK1. Proc. Natl. Acad. Sci. USA 94:8563–68 D’Ambrosio D, Hippen KL, Minskoff SA, Mellman I, Pani G, Siminovitch KA, Cambier JC. 1995. Recruitment and activation of PTP1C in negative regulation of antigen receptor signaling by Fc gamma RIIB1. Science 268:293–97 Burshtyn DN, Scharenberg AM, Wagtmann N, Rajagopalan S, Berrada K, Yi T, Kinet JP, Long EO. 1996. Recruitment of tyrosine phosphatase HCP by the killer cell inhibitor receptor. Immunity 4:77-85 Muta T, Kurosaki T, Misulovin Z, Sanchez M, Nussenzweig MC, Ravetch JV. 1994. A 13-amino-acid motif in the cytoplasmic domain of Fc gamma RIIB modulates B-cell receptor signalling. Nature 369:340 Daeron M, Latour S, Malbec O, Espinosa E, Pina P, Pasmans S, Fridman WH. 1995. The same tyrosine-based inhibition motif, in the intracytoplasmic domain of Fc gamma RIIB, regulates negatively BCR-, TCR-, and FcR-dependent cell activation. Immunity 3:635–46 Imani F, Rager KJ, Catipovic B, Marsh DG. 1997. Interleukin-4 induces phos-

P1: SKH/SPD

P2: PSA/ARY

January 21, 1999

738

165.

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

166.

167.

168.

169.

170.

171.

172.

173.

15:59

QC: PSA/UKS

T1: PSA

Annual Reviews

AR078-22

NELMS ET AL phatidylinositol 3-kinase (p85) dephosphorylation. Implications for the role of SHP-1 in the IL-4-induced signals in human B cells. J. Biol. Chem. 272:7927– 31 Kuhne MR, Pawson T, Lienhard GE, Feng GS. 1993. The insulin receptor substrate 1 associates with the SH2-containing phosphotyrosine phosphatase Syp. J. Biol. Chem. 16:11479–81 Gadina M, Stancato LM, Bacon CM, Larner AC, O’Shea JJ. 1998. Involvement of SHP-2 in multiple aspects of IL-2 signaling: evidence for a positive regulatory role. J. Immunol. 160:4657–61 Sun XJ, Pons S, Wang LM, Zhang Y, Yenush L, Burks D, Myers MG Jr, Glasheen E, Copeland NG, Jenkins NA, Pierce JH, White MF. 1997. The IRS-2 gene on murine chromosome 8 encodes a unique signaling adapter for insulin and cytokine action. Mol. Endocrinol. 11:251–62 Hof P, Pluskey S, Dhe-Paganon S, Eck MJ, Shoelson SE. 1998. Crystal structure of the tyrosine phosphatase SHP-2. Cell 92:441–50 Shultz LD, Schweitzer PA, Rajan TV, Yi T, Ihle JN, Matthews RJ, Thomas ML, Beier DR. 1993. Mutations at the murine motheaten locus are within the hematopoietic cell protein-tyrosine phosphatase (Hcph) gene. Cell 73:1445–54 David M, Chen HE, Goelz S, Larner AC, Neel BG. 1995. Differential regulation of the alpha/beta interferonstimulated Jak/Stat pathway by the SH2 domain-containing tyrosine phosphatase SHPTP1. Mol. Cell. Biol. 15:7050–58 D’Ambrosia D, Fong DC, Cambier JC. 1996. The SHIP phosphatase becomes associated with Fc gamma RIIB1 and is tyrosine phosphorylated during negative signaling. Immunol. Lett. 54:77–82 Damen JE, Liu L, Rosten P, Humphries RK, Jefferson AB, Majerus PW, Krystal G. 1996. The 145-kDa protein induced to associate with Shc by multiple cytokines is an inositol tetraphosphate and phosphatidylinositol 3,4,5-triphosphate 5-phosphatase. Proc. Natl. Acad. Sci. USA 93:1689–93 Ware MD, Rosten P, Damen JE, Liu L, Humphries RK, Krystal G. 1996. Cloning and characterization of human SHIP, the 145-kD inositol 5-phosphatase that

174.

175.

176.

177. 178.

179.

180.

181.

182.

183.

associates with SHC after cytokine stimulation. Blood 88:2833–40 Starr R, Willson TA, Viney EM, Murray LJ, Rayner JR, Jenkins BJ, Gonda TJ, Alexander WS, Metcalf D, Nicola NA, Hilton DJ. 1997. A family of cytokineinducible inhibitors of signalling. Nature 387:917–21 Naka T, Narazaki M, Hirata M, Matsumoto T, Minamoto S, Aono A, Nishimoto N, Kajita T, Taga T, Yoshizaki K, Akira S, Kishimoto T. 1997. Structure and function of a new STAT-induced STAT inhibitor. Nature 387:924–29 Endo TA, Masuhara M, Yokouchi M, Suzuki R, Sakamoto H, Mitsui K, Matsumoto A, Tanimura S, Ohtsubo M, Misawa H, Miyazaki T, Leonor N, Taniguchi T, Fujita T, Kanakura Y, Komiya S, Yoshimura A. 1997. A new protein containing an SH2 domain that inhibits JAK kinases. Nature 387:921–24 Boguski MS, McCormick F. 1993. Proteins regulating Ras and its relatives. Nature 366:643–54 Trahey M, McCormick F. 1987. A cytoplasmic protein stimulates normal N-ras p21 GTPase, but does not affect oncogenic mutants. Science 238:542–45 Carpino N, Wisniewski D, Strife A, Marshak D, Kobayashi R, Stillman B, Clarkson B. 1997. p62(dok): a constitutively tyrosine-phosphorylated, GAPassociated 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 Di Cristofano A, Carpino N, Dunant N, Friedland G, Kobayashi R, Strife A, Wisniewski D, Clarkson B, Pandolfi PP, Resh MD. 1998. Molecular cloning and characterization of p56dok-2 defines a new family of RasGAP-binding proteins. J. Biol. Chem. 273:4827–30 Meier H, Myers DD, Huebner RJ. 1963. Genetic control by the hr-locus of susceptibility and resistance to leukemia. Proc. Natl. Acad. Sci. 63:759–66 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

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:701-738. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:739–79 c 1999 by Annual Reviews. All rights reserved Copyright °

DEGRADATION OF CELL PROTEINS AND THE GENERATION OF MHC CLASS I-PRESENTED PEPTIDES Kenneth L. Rock1 and Alfred L. Goldberg2 1Department

of Pathology, University of Massachusetts Medical School, Worcester, Massachusetts 01655; 2Department of Cell Biology, Harvard Medical School, 240 Longwood Avenue, Boston, Massachusetts 02115; e-mail: [email protected] KEY WORDS:

antigenic peptide, antigen presentation, antigen processing, protein degradation, cytotoxic T lymphocyte, proteasome, proteolysis, ubiquitin

ABSTRACT Major histocompatibility complex (MHC) class I molecules display on the cell surface 8- to 10-residue peptides derived from the spectrum of proteins expressed in the cells. By screening for non-self MHC-bound peptides, the immune system identifies and then can eliminate cells that are producing viral or mutant proteins. These antigenic peptides are generated as side products in the continual turnover of intracellular proteins, which occurs primarily by the ubiquitin-proteasome pathway. Most of the oligopeptides generated by the proteasome are further degraded by distinct endopeptidases and aminopeptidases into amino acids, which are used for new protein synthesis or energy production. However, a fraction of these peptides escape complete destruction and after transport into the endoplasmic reticulum are bound by MHC class I molecules and delivered to the cell surface. Herein we review recent discoveries about the proteolytic systems that degrade cell proteins, how the ubiquitin-proteasome pathway generates the peptides presented on MHC-class I molecules, and how this process is stimulated by immune modifiers to enhance antigen presentation.

Presentation of Peptides on Major Histocompatibility Complex Class I Molecules The immune system provides continual surveillance against viral infections and cancers by monitoring whether cells are synthesizing foreign or mutant 739 0732-0582/99/0410-0739$08.00

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

740

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

proteins. In this process, major histocompatibility complex (MHC) class I molecules bind oligopeptide fragments derived from a cell’s expressed proteins and display them on the cell surface. Under normal physiological conditions, all of the class I-presented peptides are derived from normal autologous sequences to which the immune system is nonreactive owing to self-tolerance. However, if the cell is infected by viruses or is expressing mutant gene products, then nonnative (“foreign”) peptides will be displayed and will stimulate cytotoxic T lymphocytes (CTL) to kill the affected cell. The MHC class I molecule is a heterodimeric protein that is assembled in the endoplasmic reticulum (ER) (for review, see 1). Before binding peptides, the newly synthesized class I molecules are retained in the ER, through interactions with the molecular chaperones calnexin, calreticulin, and ERp57 (2–4), as well as with other associated proteins such as tapasin (5). In the absence of an appropriate peptide, the class I heterodimers are intrinsically unstable and readily dissociate at physiological temperature. The binding of an 8- to 10-residue peptide stabilizes the interaction between the class I heavy and light chains and allows the heterodimer to be transported via the “default exocytic” pathway to the plasma membrane (1). Presumably, this peptide-driven mechanism prevents the transport of empty class I molecules to the cell surface where they might bind extracellular peptides. This mechanism thus insures that the spectrum of class I-presented peptides accurately reflects the proteins expressed within a cell. In the ER, the class I heavy chain folds to form a peptide-binding groove that can accommodate oligopeptides of 8–10 residues (depending on the MHC haplotype). These peptides are bound primarily through their N- and C-terminal residues (and a few of the side chains of central residues). Although the bound peptides are rather uniform in length, they can vary greatly in sequence, especially in those residues not involved in binding the class I molecule (6). A limited number of different MHC molecules thus present a large number of peptides to circulating lymphocytes. Consequently, the cells, while continually degrading intracellular proteins, must produce some peptides of a very specific length and with a characteristic C terminus. This review discusses our current understanding of how cells generate these antigenic peptides.

Most MHC Class I-Presented Peptides Are Generated in the Cytoplasm To understand the origin of the peptides presented on MHC class I molecules, it was first necessary to learn where these peptides were produced in cells. Experiments with agents that blocked proteolysis in endocytic compartments, such as weak bases (which prevent the acidification of endosomes and lysosomes) or inhibitors of the lysosomal proteases (such as leupeptin or E64), did not interfere

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

741

with class I presentation (7, 8). Thus, in contrast to class II-presented epitopes, which are derived primarily from endocytosed proteins (9), most class I-presented peptides must be generated outside of endosomes or lysosomes. The finding that proteins normally resident in the cytoplasm stimulated CTL responses suggested that at least some antigenic peptides were produced in this compartment (10, 11). Moreover, if an antigenic protein was artificially targeted to the cytosol (e.g. by preventing its translocation into the ER by deletion of its signal sequence or by loading the protein into the cytoplasm), epitopes from these proteins were presented on class I molecules (12–14). Definitive evidence that the cytoplasm is the major site of generation of class I-presented peptides came from the analysis of mutant cells that lacked the transporter of antigen presentation (TAP) (15). This integral ER protein is a member of the ABC family of ATP-dependent transporter proteins and functions to transfer oligopeptides (most efficiently of 7 to ∼15 residues in length) from the cytoplasm to the ER (16). In TAP-deficient cells, which lack this continual supply of peptides from the cytoplasm, most class I molecules cannot bind peptides and are retained in the ER (16). Although most class I-presented peptides are generated in the cytoplasm (or possibly the nucleus) and are supplied to class I molecules via TAP, a small number of peptides are presented on class I molecules even in TAP-deficient cells. Some of these peptides were isolated from human class I allele HLA-A2 and the murine allele Qa1b and were shown to be derived from signal sequences that direct newly synthesized proteins into the ER (17, 18). Presumably, these peptides are released by the “signal peptidase” that cleaves these targeting sequences from membrane or secreted proteins during their transport into the ER. In addition, there may be other TAP-independent mechanisms that at low rates allow some peptides to be taken up into the ER from the cytoplasm (19). There are many examples of peptides from membrane or secreted proteins presented on class I molecules. Since these proteins are cotranslationally transported into the ER, it had been assumed that most of these molecules should not be available to proteases in the cytoplasm. However, the presentation of some, but not all, of these antigens, surprisingly, still requires the TAP transporter, and presumably, therefore, these antigenic peptides are generated by protein degradation in the cytosol (8, 20, 21). Recently, it has become clear that a significant fraction of many membrane-bound or secretory proteins [e.g. the cystic fibrosis transporter (22, 22a), T cell receptor subunits (23), and individual MHC subunits (24)] after translocation into the ER are transported back into the cytosol, probably by the SEC61 complex (25), and degraded by cytosolic proteasomes. This process serves as a quality control system that prevents accumulation in the ER of incorrectly folded or unassembled proteins (26). Most likely, this process accounts for their TAP-dependent presentation on class I

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

742

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

molecules (27). Alternatively, some fraction of these antigens may fail to be transferred efficiently into the ER, or possibly defective transcription (28) or translation (29–31) may result in synthesis of truncated products that lack a functional leader sequence. MHC class I molecules can also present peptides derived from mitochondrial proteins (32). Mitochondria contain their own pathways for protein degradation, which differ from those present in the cytoplasm (33, 33a,b). Although they lack the ubiquitin-proteasome pathway, in the mitochondrial matrix there are several ATP-dependent proteolytic complexes homologous to those described in Escherichia coli (33–35). It is uncertain whether the class I-presented mitochondrial peptides are generated in the mitochondria and/or in the cytosol, where some precursors of mitochondrial proteins may be degraded before they are translocated into the organelle. Generation in the mitochondria, which probably is occurring for ND1 (maternally transmitted) antigen [because it contains an N-formylated N terminus (36)] would necessitate the existence of a peptide exit system from the organelle, and no such transporter has been described as yet.

Pathways for Degrading Proteins in Cells Most presented peptides are generated in the cell by the same proteolytic pathways as are responsible for the continuous turnover of nuclear and cytosolic proteins. These degradative pathways evolved in lower eukaryotes long before the emergence of the immune system and play many essential roles in cells. Nearly all proteins in mammalian cells are continually being degraded and replaced by new synthesis (37). The rates of degradation of individual cell constituents vary widely, with half-lives ranging from 10 minutes to several days or even weeks. In addition, cells eliminate particularly rapidly unfolded, damaged, or mutant proteins with highly abnormal conformations whose accumulation might be harmful to cells (37–39). Moreover, many processes in cells, such as progress through the cell cycle or gene transcription, and metabolic pathways are controlled by rapid degradation of key regulatory molecules (40–42). Finally, the hydrolysis of endogenous proteins can provide the organism with a source of amino acids for synthesis of new proteins or for energy metabolism, especially in poor nutritional states. Recent studies have helped identify the pathways that are responsible for these proteolytic processes. Extracellular proteins taken up by endocytosis (e.g. foreign antigens or circulating polypeptides), as well as some cellular proteins, particularly membrane receptors, are degraded in lysosomes by the acid optimal proteases (cathepsins) (43). This lysosomal process accounts for a small fraction (<10–20%) of normal protein turnover and does not contribute to MHC class I antigen presentation in most cells (8, 44; see above). In contrast, the bulk of cellular proteins

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

743

are degraded by a proteolytic system in the cytoplasm and nucleus. Early studies demonstrated that this nonlysosomal pathway requires ATP (45), which was quite surprising because peptide hydrolysis is a thermodynamically favored reaction, and no protease was known to require ATP. A nonlysosomal degradative system was then demonstrated in soluble cell extracts that completely degraded cell proteins to amino acids in an ATP-dependent process (46). Analysis of this ATP requirement led to the elucidation of the ubiquitin-proteasome pathway, in which ATP is essential for covalent marking of substrates to be degraded and for the function of the very large proteolytic complex, the 26S proteasome (34, 38, 47).

Ubiquitin and Protein Degradation Many cellular proteins to be degraded in the cytoplasm are first covalently modified by linkage to the polypeptide cofactor ubiquitin (38, 48). Initially, ubiquitin is activated in an ATP-dependent reaction by an enzyme (E1) that forms a thiol ester with ubiquitin’s C terminus and then transfers the ubiquitin as a thiol ester adduct to one of the cell’s many ubiquitin-carrier proteins (E2) (49, 50). In a process catalyzed by a ubiquitin-protein ligase, E3, the carboxyl terminus of ubiquitin is then conjugated via an isopeptide bond to the epsilon amino group of lysine on a protein substrate or to a ubiquitin moiety already bound to the protein (49, 50). In this manner, a polyubiquitin chain is synthesized, which targets the protein for rapid degradation by the 26S proteasome (see below). Mammalian cells contain a single E1, but 10–15 different E2s and a large number (>12) of E3s (50, 51). The cell can selectively degrade specific proteins while sparing others (50) because the individual E2 and E3 enzymes preferentially ubiquitinate certain proteins. Thus, by using different combinations of E2s and E3s, cells are able to selectively eliminate specific regulatory molecules, such as cyclin-dependent kinases (52) or IkB, the inhibitor of the NFκB transcription factor (41, 53). Much has been learned recently about the functions of individual E2s and E3s, through genetic studies in yeast and biochemical studies of ubiquitin conjugation in cell-free extracts (50, 51). Unfortunately, studies of protein ubiquitination in mammalian cells are still difficult, and selective inhibitors are not available. Analysis of mutant cells that have a thermolabile E1 enzyme has confirmed that ubiquitination is necessary for the degradation of a large number of proteins in intact mammalian cells. When these mutants are cultured at the nonpermissive temperature, the rapid breakdown of short-lived, normal and abnormal cell proteins is markedly reduced (54). Interestingly, under these same conditions, the degradation of long-lived cell proteins is inhibited much less, which may indicate that these substrates are degraded largely by an ubiquitinindependent process. However, because these cells do not tolerate for long

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

744

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

periods the high temperatures (41–43◦ C) that are required to inactivate E1, it is difficult or impossible to reliably test whether ubiquitin is necessary for the breakdown of their long-lived proteins, which (by definition) turn over very slowly (at a rate of ≤1% per h). Moreover, even at 43◦ C, E1 is not completely inactivated in these cells, and its residual activity might be sufficient to allow these proteins to be ubiquitinated. Therefore, although ubiquitin is clearly required for regulated proteolysis, it is still unclear whether ubiquitin conjugation is essential in the degradation of most cell proteins, especially the long-lived cell components. In cell-free systems, some proteins, like the short-lived enzyme ornithine decarboxylase and many denatured polypeptides (e.g. casein), can be degraded by 26S proteasomes without being conjugated to ubiquitin (55, 56). Moreover, eubacteria and archaebacteria, which lack ubiquitin, contain simpler forms of the 20S proteasome that efficiently degrade cell proteins in an ATP-dependent process, in association with ring-shaped ATPase complexes (57, 58). These structures thus appear to be the evolutionary precursors of the eukaryotic 19S proteasomeregulatory complex (PA700) (see below; 58).

Role of Ubiquitin in Class I Antigen Presentation The first evidence that the ubiquitin-proteasome pathway is the source of peptides presented on MHC class I molecules was through studies of the consequences of inhibiting protein ubiquitination in intact cells (59–61). When two different mutant cell lines with a thermolabile ubiquitin-activating enzyme (E1) were microinjected with ovalbumin, the presentation of the peptide SIIFNEKL, derived from ovalbumin, was impaired at the nonpermissive temperature (60, 62). In contrast, if this antigenic peptide was introduced directly into the cytoplasm (instead of ovalbumin), its presentation was not inhibited upon inactivation of E1. Therefore, ubiquitination was necessary to generate the presented peptide from ovalbumin and not for any subsequent step in the class I pathway. Furthermore, the generation of this same peptide from full-length ovalbumin was recently demonstrated in cell-free extracts and shown to require both ATP and ubiquitin (63). Another approach, indicating a requirement for ubiquitin conjugation for antigen presentation, involved modifying the antigenic proteins, so that they were more rapidly conjugated to ubiquitin. One feature of a protein that can lead to its rapid ubiquitination is the nature of its N-terminal residue (“the N-end rule”) (64). Proteins with large, bulky or charged amino termini are rapidly ubiquitinated and degraded, whereas the same proteins with the normal N-terminal methionines or other small N-terminal residues are quite stable. In most (61, 65–67) but not all cases (68), the more rapidly ubiquitinated form of the antigen was presented on class I molecules much more rapidly and/or induced stronger

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

745

immune responses. In fact, within 30 minutes after microinjection, peptides from a very short-lived protein appeared on MHC molecules, whereas with the long-lived form (t1/2 = 20 h), many hours were required before maximum presentation. Moreover, blocking the potential ubiquitin conjugation sites by methylating all of the epsilon amino groups on the protein inhibited its presentation (61). These results, together with the data from the E1 mutants (60, 62), indicated that ubiquitination was critical for the presentation of at least some antigens and suggested that protein degradation can be a rate-limiting step in this process. Although the degradation of many unfolded proteins seems to require ubiquitination in vivo, there are several examples in which presented peptides can be generated in an ubiquitination-independent manner. Surprisingly, denaturation and alkylation of ovalbumin allow this antigen to be presented despite inactivation of E1 (62) and to be presented, to some extent, when all potential ubiquitin conjugation sites are blocked by methylation of lysine-amino groups. Although the presentation of this substrate does not require ubiquitination, this process is still completely inhibited by proteasome inhibitors. Ubiquitin-independent degradation of reduced and alkylated ovalbumin has been confirmed with purified 26S proteasomes (161). Therefore, it is possible that, for ovalbumin, ubiquitination serves only to accelerate substrate binding to the 26S proteasome. Also, presented peptides can be generated by proteasomes from extended peptides that have no lysines for ubiquitin conjugation (70; A Craiu & KL Rock, unpublished data). There are several other reports in which the presentation of antigens in these temperature-sensitive mutants was not reduced upon shift to the nonpermissive temperature to inactivate E1 (62, 71; KL Rock, unpublished data). Unfortunately, such negative results are difficult to interpret, because the inactivation of E1 is incomplete under these conditions, and sufficient residual E1 activity remains to allow ubiquitin conjugation to many substrates. Moreover, at the nonpermissive temperature, these cells compensate by overproducing ubiquitin and other heat-shock proteins (72). Definitive tests of whether ubiquitination is required for a protein’s degradation and antigen presentation will require the development of better methods to block this pathway in cells.

20S Proteasomes—The Proteolytic Core Particle In cells, ubiquitinated polypeptides are rapidly degraded by the 26S proteasome complex, a very large particle found in the nucleus and cytoplasm of all eukaryotic cells (57, 73–75). The proteolytic component of these complexes is the 20S core particle, a cylindrical structure made up of 28 subunits, arranged in four stacked rings (57, 76). The central two rings are each composed of seven homologous, but distinct, ß-subunits, which surround a central chamber

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

746

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

where proteolysis occurs. In eukaryotes, three of the ß-subunits in each ring contain proteolytic active sites (see below). The outer two α rings of the 20S proteasome are made up of seven different, but homologous, α-subunits, which surround a central opening through which substrates appear to enter. In the X-ray analysis of the archaebacterial proteasomes, the central openings in the ˚ in diameter, and therefore only unfolded polypeptides can α rings are only 13 A enter the 20S particle (78). Presumably, this organization, in which substrate entry is restricted and the active sites are isolated within the central lumen of the particle, allows efficient hydrolysis of proteins that are transported into the particle, while protecting other cell components from nonspecific proteolytic attack. The nature and location of the proteasome’s active sites were long uncertain, since these subunits are not homologous in sequence to any other protease. The active sites were initially localized in X-ray diffraction studies by their ability to bind peptide aldehyde inhibitors (78; see below), which are substrate analogs that form transition state complexes with the active-site nucleophilic residue (57, 78, 79). This finding and related ones involving site-directed mutagenesis (80, 81, 82) uncovered a novel proteolytic mechanism, in which the nucleophilic residue attacking the peptide bond is the hydroxyl group on the N-terminal threonine residue on these ß-subunits. This critical threonine residue was also found to be modified by the irreversible proteasome inhibitors lactacystin (83), vinyl sulfones (84), and dicloroisocoumarin (85). This threoninebased catalytic mechanism is distinct from that of any other known protease. Proteolytic enzymes have been traditionally classified by the nature of their active sites into serine proteases (e.g. complement or pancreatic enzymes), cysteine proteases (e.g. lysosomal enzymes), metalloproteinases (e.g. collagenases), or aspartic proteinases (e.g. human immunodeficiency virus protease). It remains uncertain, however, why the proteasome has evolved such a novel catalytic mechanism. Interestingly, replacement of the N-terminal threonine by a serine also allows efficient hydrolysis of standard peptide substrates (80), but markedly reduces the degradation of proteins, for reasons that are unclear (69). When isolated, these subunits lack proteolytic activity, and thus proteolysis appears to require the precise conformations found in the ß ring to be active (86). The three active ß-subunits are synthesized as longer precursors, and during the assembly of the particle, the N-terminal peptide extensions are cleaved off by an autocatalytic mechanism, revealing the catalytic N-terminal–threonine residues (57, 81, 87). This process occurs in the final stages of assembly of the 20S particle. In fact, smaller precursors (∼300 K) of the mature proteasome have been isolated that are inactive as proteases, apparently because of the presence of this N-terminal extension on the ß-subunits (81, 88, 89). In the

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

MHC CLASS I ANTIGEN PRESENTATION

747

four other ß-subunits, these extended peptides are not removed during assembly, and therefore they remain inactive. As discussed below, three additional active ß-subunits are induced by interferon-γ (IFN-γ ) in mammalian cells and are also found in such 300 K precursors (57, 90). These homologous subunits are then preferentially incorporated into new 20S and 26S proteasomes, often termed “immunoproteasomes,” and alter the peptidase activities of the particles (90; see below).

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Peptidase Activities of the 20S Proteasome A number of studies with specific fluorogenic peptide substrates or selective inhibitors (initially by Orlowski and colleagues; 90a) indicated that the mammalian proteasome is a “multicatalytic protease” that exhibits three distinct peptidase activities. These activities have now been associated with specific ß-subunits, through mutagenesis (81, 91, 92), transfection (93, 94) , and especially X-ray diffraction studies of the homologous yeast particles (79). One is termed the “chymotrypsin-like activity,” because it preferentially hydrolyzes model peptides after large hydrophobic residues, one is the “trypsin-like” activity and cleaves peptides after basic residues, and one hydrolyzes after acidic residues. This later site has been commonly referred to as the “peptidyl postglutamyl cleaving activity,” but this designation is misleading because this activity actually cuts after aspartyl residues ≤30-fold faster than after glutamates, and thus it resembles the caspases in specificity (69). A variety of observations indicate that the chymotrypsinlike site is the ratelimiting one in protein degradation. For example, mutations that inactivate this site or pharmacological inhibitors of this activity markedly retard protein hydrolysis, whereas inactivation or inhibition of the other two sites has only minor effects on proteolysis (26, 81, 95). Mammalian proteasomes have also been reported to exhibit two additional activities that cleave certain model peptides after branched-chain amino acids (termed BrAAP) or after small neutral residues (termed SNAP) (96). However, the biochemical basis of these two activities is unclear and controversial. Structural and biochemical studies have found only three subunits with N-terminal threonine residues (53, 57). Also, recent studies suggest that the “BrAAP substrates” are actually hydrolyzed in part by the chymotryptic site and partly by the “postacidic cleaving activity” (97), and the “BrAAP” and “SNAP” activities are clearly evident only after modification of the proteasomes (e.g. by dichloroisocoumarin) (96, 97). As discussed below, the precise cleavage specificity of these different sites has major implications for antigen presentation but remains poorly defined. Recent observations indicate that these three active sites do not function independently in protein hydrolysis (69). Occupancy of the chymotryptic site by a peptide substrate has been found to allosterically activate cleavages by

P1: JER/MBG

P2: NBL/spd

February 11, 1999

748

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

the postacidic site, although occupancy of the latter site strongly inhibits the chymotryptic site (69). These observations suggest that proteins are digested in a highly ordered process involving cycles of activation-inactivation of the rate-limiting chymotryptic site. On this basis, it has been proposed that, in each cycle, the chymotryptic site clips off pieces of the incoming polypeptide, which are then further fragmented by the other active sites to oligopeptides, small enough to diffuse out of the particle (69).

20S Proteasome α-Subunits: Roles in Assembly and Controlling Substrate Entry The functions of the α-subunits are less clear; in fact, E. coli and related bacteria contain a proteasome homolog, the ATP-dependent HsIUV complex, which contains two ß-rings and no α-subunits (98). In the proteasomes from archaebacteria, the seven α-subunits are identical and can self-assemble into rings (80, 86). These rings lack proteolytic activity, but they are required as a scaffold in the subsequent self-assembly of the associated ß-rings (86). The α rings are also the sites for binding of the PA28 (11S) or PA700 (19S) regulatory complexes. Although a clear opening in the α rings was found in the X-ray analysis of archaebacterial proteasomes (78), X-ray diffraction of the 20S yeast proteasomes has failed to show any such opening in the α ring (79). So, presumably the entry site for substrates can exist in either of two forms, open or closed, each of which was captured in one of the crystal forms studied. Physiological activators of the proteasome, such as the 19S regulatory complex, PA28, and ATP hydrolysis, may control proteolysis by retarding or facilitating substrate entry into the 20S particle. Accordingly, when 20S proteasomes are isolated rapidly by gentle approaches, they are in an inactive, latent state, but can be readily activated by a variety of treatments, such as dialysis, against buffers lacking glycerol, low concentrations of detergents (e.g. 0.01% sodium dodecyl sulfate), high Mg2+ levels, polylysine, and heat (53, 161). Presumably, such treatments induce the open conformation of this channel. It remains unclear exactly what factors normally prevent the spontaneous activation of 20S proteasomes in cells, since this process occurs readily in vitro at 37◦ C. This activation could be potentially damaging to the cells, and interestingly, it is strongly retarded by several factors found typically in the intracellular milieu, such as K+ and Na+ ions and high-protein concentrations (KM Woo & AL Goldberg, unpublished data).

26S Proteasome: The 19S Regulatory Complex Plus the 20S Core Particle The 26S proteasome is a 2000-kDa complex that contains ∼50 different subunits, and its proteolytic activity requires concomitant ATP hydrolysis. In this large complex, the 20S (650-kDa) proteasome is capped at each end by a 19S

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

749

(700-kDa) regulatory complex (53, 74). This structure is much more labile and difficult to isolate than the 20S core particles. Each 19S cap contains ∼20 distinct subunits (57, 74, 76), most of which are essential for viability in yeast (53, 99). It has also been termed PA700, proteasome activator of 700 K, because, in the presence of ATP, it associates with the 20S particle and markedly stimulates all three of its peptidase activities (100). Polymorphic forms of the 19S complex have not been found, although purified preparations also contain a large fraction of asymmetric complexes composed of only one 19S and one 20S complex (53, 99). The biological importance of the symmetrical or asymmetrical complexes in vivo is still uncertain. This 19S regulator can be dissociated in yeast mutants into two subcomplexes, a “lid” component and a “base” that binds to the 20S core (101). The base contains six distinct, but homologous, ATPases, which presumably function as molecular chaperones to unfold polypeptide substrates and thread them into the 20S proteasome (53, 57). The 19S complex is critical in recognition of ubiquitinated substrates and contains sites for tight binding of the polyubiquitin chain as well as an isopeptidase activity that, during protein breakdown, disassembles the ubiquitin chain and releases free ubiquitin molecules (53, 57, 102). In the absence of PA700, the 20S particles cannot degrade ubiquitinated proteins (56). Purified 20S particles can degrade some proteins, but do so much more slowly than the 26S complex and must first be activated (e.g. by sodium dodecyl sulfate) (56). Therefore, it is still unknown whether the 20S proteasome ever functions by itself in protein breakdown in vivo. The 26S proteasome is the only activity in cells that degrades ubiquitin-protein conjugates, but it can also hydrolyze rapidly some nonubiquitinated proteins [e.g. ornithine decarboxylase (55) and various unfolded (e.g. casein) (55) and denatured proteins (161)]. Unlike the 20S, the 26S particle uses ATP as a cofactor in hydrolyzing proteins, and this requirement presumably reflects key roles of the ATPases within the 19S cap structures (57). Therefore, the 26S form of the proteasome clearly functions in vivo in the energy-dependent degradation of ubiquitinated and certain nonubiquitinated polypeptides (discussed below). Proteins are digested by 26S (and 20S) proteasomes by a highly processive manner (85, 161); i.e. once a protein enters the particle, it is digested all the way to small peptides, ranging in length from 4 to 24 residues. Thus, unlike traditional proteases, these particles do not simply cleave a polypeptide chain and release large fragments. However, in one example, the NFκB precursor p105, the digestion of the substrate by 26S particles ceases in the middle, releasing an N-terminal (50-kDa) fragment that functions in the cells as a transcription factor (41). Although other such examples may be discovered, the primary function of 20S and 26S particles is to hydrolyze completely polypeptides to short oligopeptides.

P1: JER/MBG

P2: NBL/spd

February 11, 1999

750

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

11S (PA28) Regulatory Complexes Another intriguing regulator of proteasome function is PA28 (also called REG), a 200-kDa (11S) conical structure that can also form a cap at either end of the 20S particle (103, 104). PA28 markedly stimulates the three peptidase activities of the 20S particle (104, 105), but, surprisingly, it does not influence rates of protein degradation by 20S or 26S particles. Electron microscope tomography (106) and X-ray diffraction studies (107) reveal that PA28 contains a large ˚ through which, presumably, substrates may enter central channel (of 20 A) the complex or peptide products may diffuse out. When isolated from cells, PA28 is a heterodimeric complex composed of three pairs of alternating α- and β- 28-kDa subunits (108). PA28 α-subunits can also assemble into a homopolymeric ring, which also stimulates peptide hydrolysis, although to a lesser degree than the heteropolymeric native structure (109, 110). PA28 α- and ß-subunits are highly homologous to each other (∼50% identical), as well as to the nuclear protein Ki (also called PA28 γ ) (111, 111a). PA28 γ -subunits, the expression of which is suppressed by IFNγ (111a), can also assemble into a complex that stimulates primarily the trypsin-like activity of proteasomes (112). Despite extensive biochemical and structural studies, the functions of PA28 in protein breakdown or peptide metabolism in vivo remain uncertain, largely because of its failure to enhance protein digestion. Unlike the association of the 20S particle with PA700, its binding to PA28 does not require ATP and is quite weak (113). A role of PA28 in antigen presentation seems likely, since PA28 α and β are both induced by the immune modulator IFN-γ , and one group has reported a stimulation of class I presentation in cells transfected with PA28 alpha genes (114; see below). One intriguing and potentially very important recent finding is that interferon-treated cells contain a mixed heterocomplex, which includes a 19S cap, the core 20S proteasome, and the PA28αβ complex (115). The evidence for this structure comes from coimmunoprecipitation, and its biochemical properties, tissue distribution, and possible function in antigen presentation await further study.

Proteasome Inhibitors and Intracellular Proteolysis Analyzing the physiological roles of the proteasome in intact mammalian cells had been difficult due to the lack of suitable experimental tools. However, in the past few years, a variety of low-molecular inhibitors have been synthesized or isolated that can inhibit proteasome function in intact cells (117). The first inhibitors that were developed were peptide aldehydes such as acetyl-leu-leunorleucinal or MG132 (CBZ-leu-leu-leucinal) (8, 116). These compounds act as transition state analogs and form a hemiacetal adduct with the proteasome’s active sites, primarily the chymotryptic and postacidic sites. Because these compounds also can inhibit to various degrees the activity of cysteine proteases,

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

751

additional control experiments are essential to perform to rule out the possible effects on calpain and lysosomal cathepsins in studies using these aldehydes in vivo as proteasome inhibitors (8, 117). The great advantage of the peptide aldehydes for experimental studies is that they are inexpensive, and their effects are readily reversible. Subsequently, the natural product from Actinomyces lactacystin (83), its active derivative clastolactacystin ß-lactone (118), and also peptide-vinyl sulfones (84) were found to inhibit proteasomes by forming a covalent (slowly hydrolyzed) adduct with the threonine hydroxyl group. Unlike other proteasome inhibitors, lactacystin does not appear to inhibit any other protease (83, 119). These inhibitors can enter most living cells and inhibit proteasome function (117). As a result, they block the rapid degradation of short-lived, normal and abnormal proteins. In mammalian cells (8, 119), but not yeast cells (117a), they also inhibit the degradation of most long-lived proteins, which comprise the bulk of cell proteins. By blocking proteasome function, the peptide aldehydes may reduce intracellular proteolysis by 80–90% (8, 119); by contrast, blocking of lysosomal function at most decreases this process by 10–20%. Therefore, the proteasome constitutes the primary site for degrading proteins in mammalian cells. Cells tolerate these inhibitors remarkably well and do not begin to show any signs of measurable toxicity until 8–15 h (or longer) of exposure, depending on the cell. After 1- to 2-h exposure, the cells induce the heat-shock response (120, 121), and eventually apoptosis occurs, presumably due to the essential roles of the ubiquitin-proteasome pathway in the cell (117, 122).

Role of Proteasomes in Antigen Presentation In addition to the studies that demonstrate a role of ubiquitination and, by extrapolation, of 26S proteasomes in antigen presentation, there is extensive evidence implicating these particles in the generation of antigenic peptides. The first clue was the discovery that two proteasome subunits were encoded in the MHC, and subsequent studies demonstrated that incorporation of these subunits enhanced the presentation of some antigens (see below). A number of studies have also examined the ability of purified 20S proteasomes to generate antigenic peptides from proteins or longer peptides. In many cases, these particles have been found to generate some of the presented peptides from proteins (e.g. ovalbumin or ß-galatosidase) or from longer synthetic oligopeptides corresponding to antigenic regions of viral proteins (123–125). However, the physiological relevance of these in vitro observations remains uncertain; for example, it is unclear if any mixture of proteases (e.g. pronase or pancreatic extracts) might not with time also generate these antigenic epitopes. Also, these experiments typically have been carried out under highly artificial, unphysiological conditions (see below), and the efficiency of epitope generation in this

P1: JER/MBG

P2: NBL/spd

February 11, 1999

752

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

process cannot be estimated. Although such experiments (discussed below) showed the potential of the proteasome to generate presented peptides, other approaches were necessary to prove that proteasomes serve this role in vivo. Perhaps the most definitive evidence was the demonstration that proteasome inhibitors selectively blocked the generation of the majority of class I-presented peptides in cells (8). Each of these lines of evidence is discussed below.

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Proteasome Inhibitors Block Antigen Presentation in Living Cells The proteasome inhibitors have provided a powerful tool to assess the role of this particle in class I antigen presentation. In living cells, the peptide aldehyde inhibitors and lactacystin–ß-lactone were found to completely block the presentation of peptides from several different proteins, including ovalbumin, ß-galactosidase, and influenza viral antigens (8, 116, 119, 126). In contrast, when the 8- or 9-residue antigenic peptides from these proteins (which do not require any further proteolytic cleavages) were introduced into the cytoplasm by injection or were expressed from minigenes, then blocking proteasome function did not inhibit presentation (8, 119). Therefore, these inhibitors block antigen presentation by interfering with the proteolytic generation of the presented peptide and do not affect subsequent steps in this process. The proteasome inhibitors also block the generation of most of the cell’s antigenic peptides, which was demonstrated by monitoring the assembly of newly synthesized class I molecules in cells (8, 119). These heterodimers are not stable in the absence of a bound peptide and are retained in the ER until appropriate peptides are available (see above). In cells treated with peptide aldehyde inhibitors or lactacystin, the generation of the stable heterodimers of several class I molecules and their transport out of the ER were markedly inhibited (8, 119). In extracts of these inhibitor-treated cells, the class I molecules could be stabilized by the addition of the missing antigenic peptides. Therefore, the inhibitors appear to block the assembly of stable MHC class I molecules by causing a general reduction in the supply of appropriate peptides. This inhibition has been seen for many (8, 126a,b), but not all, MHC class I molecules (126b). The conclusion that the proteasome generates the majority of class Ipresented peptides is consistent with the observation (8, 119) that this particle is also responsible for degrading the bulk of cellular proteins (see above), which is precisely the pool of proteins that are monitored by the immune system via class I presentation. It is now firmly established that the inhibition of antigen presentation by these agents is a consequence of their blocking proteasome function and does not result from inhibition of another protease or some nonspecific effect. The extent of inhibition of protein degradation by the peptide aldehydes and lactacystin

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

753

correlated closely with the decrease in antigen presentation (8, 116, 119). Also, lactacystin can block antigen presentation almost completely (119), and it does not appear to inhibit any other known protease (83, 119), aside from the proteasome. In addition, the peptide aldehyde inhibitors and lactacystin are unrelated in structure and mode of action, yet cause a similar inhibition of antigen presentation (8, 119) . Moreover, a large series of different di- and tripeptide aldehydes with a wide range of Kis for proteasomes display the same rank order of potency in inhibiting intracellular protein breakdown and antigen presentation (8, 116, 119). In contrast, the potency of the peptide aldehydes against cysteine proteases did not correlate with the inhibition of antigen presentation. Also arguing against involvement of cysteine proteases was the finding that selective inhibitors of such enzymes (e.g. E64) did not affect antigen presentation. Furthermore, a distinct type of proteasome inhibitor, the vinyl sulfones, also can reduce antigen presentation (84). These observations together strongly support the conclusion that a fraction of peptides generated by the proteasomes during normal protein turnover are the sources for class I peptides.

Proteasome Subunits That Are Interferon Inducible and Encoded in the Major Histocompatibility Complex In the early 1980s, antibodies raised against proteins encoded in the MHC were used to identify two polymorphic polypeptides, which were named descriptively “low-molecular-weight proteins” (or LMPs) 2 and 7 and found to be subunits of a large complex (127, 128). Lymphoid tissues constitutively expressed LMP2 and LMP7, whereas in other cells, their expression was strongly induced with IFN-γ (128). Almost a decade later, it was pointed out that the LMP complex resembled proteasomes (129), and immunoprecipitation experiments with antiLMP and antiproteasome sera demonstrated that the LMPs were indeed subunits of a subset of proteasomes (130, 131). At about the same time, cloning and sequencing of LMP2 and LMP7 genes revealed that they were proteasome subunits of the ß-type (ß1i and ß5i) (132–134). They are encoded in the MHC in close juxtaposition to the TAP genes, whereas all other proteasome ß-genes are encoded elsewhere in the genome (73, 135). A third IFN-γ –inducible ß-subunit, MECL1, was found subsequently, but it is not encoded in the MHC (89, 136). More recently, the particles containing the interferon-induced subunits were given the convenient name “immunoproteasomes” to indicate their special role in antigen presentation. When LMP2, LMP7, and MECL1 are expressed in cells, e.g. after IFN-γ stimulation, they are incorporated into newly assembling proteasomes in place of the constitutively expressed homologous subunits; LMP 2 (also called ß1i) replaces ß1 (subunit Y, delta, ring 10), LMP7 (ß5i) replaces ß5 (LMP 17, subunit X, epsilon, ring 12), and MECL1 (ß2i, LMP 10) replaces ß2 (LMP19,

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

754

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

subunit Z) (135, 137–140). Although interferon suppresses the incorporation of normal subunits into new proteasomes, it does not greatly reduce the mRNA content for ß5 (X), ß1 (Y), or ß2 (Z). So, the cytokine appears to regulate primarily the assembly process by favoring incorporation of newly synthesized LMP2, LMP7, and MECL1 to the exclusion of the homologous normal subunits (which must then be degraded). Due to the coordinated incorporation of these new subunits, it is likely that after interferon treatment cells contain a mixture of particles that are primarily composed of either ß5 (X), ß1 (Y), and ß2 (Z) or LMP2 (ß1i), LMP7 (ß5i), and MECL1 (ß2i) (140a). These interferon-induced subunits, like their normal homologs, all contain N-terminal threonine residues. Thus, they are all catalytically active, but their kinetic properties and substrate preference have been shown to differ in important ways, as discussed below. The first studies to evaluate whether LMP2 and LMP7 play a role in class I antigen presentation analyzed a mutant human B-cell line (0.174) or a derivative line (T2), which had a homozygous deletion in the MHC that eliminated LMP and TAP genes. When these cells were transfected with TAP genes to restore the missing transporter, class I presentation initially appeared normal (141– 143). These results indicated that LMP subunits were not absolutely required for generating the class I-presented peptides. At the time, it was concluded incorrectly that they play no role in this process. However, subsequent studies with knock-out mice and cell lines lacking these subunits have indicated an important influence on proteasome function. Analysis of mutant mice lacking LMP7 revealed a defect in the presentation of a cellular antigen (HY) and a ≤50% reduction in class I molecules, which suggests a major decrease in overall supply of peptides for antigen presentation (144). LMP2-deficient mice were found to be defective in presenting some (e.g. influenza nucleoprotein), but not all (e.g. Sendai virus), viral antigens, although the total level of class I molecules on the cells appeared normal (145). Similar results were obtained with an LMP2-deficient cell line (146). Moreover, when the original TAPtransfected 0.174 cell line was re-analyzed in greater detail and other cell lines deficient in LMP gene expression were tested, defects in antigen presentation were found (147). Thus, LMP genes clearly are important for the optimal presentation of some antigens, but because animals lacking both LMP2 and LMP7 have not yet been generated, it is still unclear whether these subunits influence the generation of most presented peptides or only a small subset of antigens. In other words, although it is established that the immunoproteasomes can promote antigen presentation, their quantitative importance in presentation of different antigens remains uncertain and may depend on the specific cell type and physiological (or pathological) conditions. The biochemical consequences of LMP2 and LMP7 incorporation were investigated by analyzing the function of 20S proteasomes isolated from cells that

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

755

lacked these subunits owing to gene deletions and cell lines that overexpressed these subunits, after IFN-γ treatment or transfection of subunit genes. Several studies using specific fluorogenic peptide substrates have found that interferon specifically stimulated the maximal rates of peptide-bond cleavage (Vmax ) after hydrophobic and basic amino acids (138, 148, 149), whereas it reduced cleavage after acidic residues. Moreover, opposite changes in peptide cleavage rates were seen in cell lines lacking LMP subunits (138, 148, 149) and in proteasomes of LMP2 knock-out mice (150). The LMP subunits were demonstrated in both 20S and 26S particles, and when isolated, both types of particles showed enhanced chymotryptic and tryptic activities and reduced postacidic cleavages (148). More recently, the immunoproteasomes were shown to cleave more rapidly model peptides after branched chain residues, which was attributed to enhanced BrAAP activity (156). The augmented peptidase activities appear to be caused by catalytic properties of the newly induced subunits, although the suppression of others (e.g. postacidic cleavages) seems to be caused by the relative loss of normal subunits that bear this activity. Like interferon treatment, transfection of the LMP2 gene into the normal or mutant cell suppressed the post-acidic activity, and this effect was proportional to the LMP2 content of the proteasomes (93). On the other hand, LMP2 deletion or overexpression of the subunit it replaces, delta, stimulated the postacidic activity (94, 150). Consistent with these findings, the subunit that is homologous to delta in yeast proteasomes appears to catalyze the postacidic activity (151). In contrast, transfection of LMP7 stimulated the proteasome’s chymotrypsin-like and trypsin-like activities, whereas the overexpression of the normal homolog of LMP7 (ß5, epsilon) lowered these activities (93, 94). Interestingly, mutation in the yeast subunit that is most homologous to LMP7 prevents the chymotrypsin-like activity (91). These differences in activities of the immunoproteasome may also be augmented by the presence of the IFN-γ –induced activator PA28 (152). There has been some controversy in the literature concerning the precise effects of the IFN-γ –induced subunits on peptide hydrolysis (124). Although one group initially failed to demonstrate such interferon-induced changes in their peptidase activities (153), in subsequent studies, enhanced chymotryptic and tryptic activities were observed (152). Others have found that deletion of LMP subunits or IFN-γ treatment alters activities against model peptides, but found different changes in the peptidase activities, for reasons that are still unclear (154, 155). On the other hand, recent analysis of all the peptide products released during digestion of polypeptides has indicated a consistent increase in C-terminal hydrophobic residues (primarily branched-chain amino acids) and a sharp decrease in acidic ones, exactly in accord with findings with small peptide substrates (156).

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

756

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

Because the incorporation of LMP and MECL1 subunits was not found to alter the overall rates of protein degradation by the 20S and 26S particles (148), these subunits seem only to affect the sites of polypeptide cleavage and result in the generation of a different set of peptides, presumably ones that are more favorable for antigen presentation (90, 148). Based on the results with fluorogenic substrates, it seems likely that one way the efficiency of this process is enhanced is by proteasomes generating a higher fraction of peptides that end in hydrophobic and basic C-terminal residues than are produced by normal 20S and 26S particles. It is noteworthy that the great majority of presented epitopes are of this form (90, 157). Thus, the LMPs appear to enhance the rates of cleavage carried out ordinarily by proteasomes lacking these subunits. This important observation can explain why LMPs are not essential for antigen presentation by cultured cells (141–143), but may simply enhance the efficiency of this process.

Are 20S Proteasomes, 26S Particles, and/or PA28 Complexes Involved in Antigen Presentation? Because most protein breakdown in vivo requires metabolic energy (34, 37) and the only known ATP-dependent protease in the nucleus or cytoplasm is the 26S proteasome (51, 76), it seems likely that the 26S proteasome plays a primary role in the generation of antigenic peptides. Moreover, since presentation of some (and perhaps many or most) antigens is dependent on ubiquitin conjugation and ubiquitinated proteins are degraded exclusively by the 26S complex, this structure clearly must generate a large fraction of, and perhaps most, class I-presented peptides. In particular, the degradation of most short-lived proteins requires ubiquitination; therefore the 26S proteasome, at the minimum, is responsible for their presentation. What is presently unclear is whether this structure plays an exclusive role in this process. When isolated gently, native 20S particles are relatively inactive in hydrolyzing proteins or peptides (as discussed elsewhere in this review). Although association with PA28αβ, the interferon-induced 11S activator of the proteasome, markedly stimulates rates of peptide cleavage by 20S particles, the 20S-PA28 complex is still relatively inactive in degrading proteins. Moreover, since most intracellular protein breakdown is ATP dependent, it seems unlikely that these complexes would make the rate-limiting cleavages in many antigens. Thus, it is uncertain whether the 20S proteasome ever functions in vivo, by itself or with PA28, on protein substrates to generate antigenic peptides. However, it is still quite possible that the 20S structure, with or without PA28, might play a role in further cleaving the peptides generated by the 26S particle (see below). More specifically, as discussed above, if PA28 associates with 20S proteasomes and PA700 in some trimeric complex (115), such a structure may be particularly important in enhancing peptide generation after IFNγ treatment. Because PA28αβ is coordinately induced together with LMPs and MECL1, it seems likely that it plays an important role in enhanced

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

757

class I presentation. The addition of the PA28α activator to purified 20S proteasomes has been reported to enhance the generation of some presented peptides from longer polypeptides (see below). In addition, it has been reported that cells transfected with PA28-α show enhanced MHC class I presentation of several antigenic peptides (114). Although these findings suggest a role for PA28-proteasome complexes in antigen presentation, several other groups, including our own, have not observed such a stimulation with other antigens, and at least one group has not been able to replicate that original report (L Eisenlohr, personal communication). Therefore, the role of PA28 in antigen presentation clearly requires further study.

Proteasome Cleavage Sites and the Generation of Presented Peptides In Vitro To bind tightly, peptides must fit entirely within the groove in the MHC class I molecules and must therefore be of an exact size, typically 8–10 residues, depending on the MHC molecule (158). It remains uncertain whether the proteasome generates antigenic peptides of this size directly. One initially attractive proposal was that the mean length of the peptides produced by the proteasomes might be eight residues, which actually corresponds to the length of a peptide stretched between adjacent active sites of the symmetric proteasomes of archaebacteria (159). It was therefore proposed that the distance between adjacent active sites served as a “molecular ruler,” determining the length of products (159, 160). This model could thus account for production of the octapeptides and nonapeptides used primarily in class I antigen presentation. However, subsequent studies indicated that the archaebacterial proteasomes do not function by such a mechanism (95). Moreover, the mammalian proteasome also cannot function in this way, because it contains only six active sites, which are located at various distances from one another, and the adjacent sites have distinct proteolytic specificities. Furthermore, the finding that the products of archaebacterial and mammalian 20S and 26S proteasomes vary from 4–24 residues (161, 161a) and are distributed as a log-normal function of length (161) is inconsistent with their cleaving polypeptides based on a molecular ruler. Thus, product sizes cannot depend on the nature or location of proteolytic sites; instead, protein breakdown appears to continue randomly until the pieces are small enough to diffuse out of the particle. Studies with mammalian 20S and 26S proteasomes degrading several different polypeptide substrates indicate that only about 10% of the peptides produced were octapeptides. About 70% were smaller than that length (i.e. too small to function in antigen presentation), and about 20% were larger than residues and would have to be trimmed further to bind tightly to MHC molecules (161). One approach to examine whether the proteasome can generate antigenic peptides directly has been to analyze the products generated by purified, activated

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

758

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

20S proteasomes in vitro. After prolonged incubations (≤24 h), some peptides able to stimulate specific T cells are generated from ovalbumin or ß-galactosidase, but with a very low yield (123). However, a very large number of oligopeptides are generated during protein breakdown, and, to characterize the cleavage products from proteasomes in greater detail, most investigators have used smaller substrates, typically consisting of an antigenic peptide flanked by five or more residues, or a small polypeptide (e.g. insulin B chain.) Proteasomes can generate the appropriate presented peptide from these oligopeptide substrates (124, 125, 162). However, cleavages also occur in many other sites both within and around the antigenic sequence, and longer species containing the antigenic epitope are also produced. In prolonged incubations, mammalian proteasomes have been reported to cleave after every amino acid within the insulin B chain, although the rates of cleavage at each position vary greatly (163, 164). Based on this finding and limited experiments substituting residues around the cleavage sites, it has even been argued that, in contrast to the simple classification of active sites based on the use of fluorogenic substrates, the residue preceding the cleavage site (P1 position) does not alone determine where proteasomes cut larger substrates (164). However, such a conclusion seems quite premature. Recently, evidence demonstrates that 20S immunoproteasomes isolated from spleen do display a P1 preference when digesting carboxyamidomethylated lysozyme (156). In general, their preferred cleavages appeared to be made after a sequence motif of hydrophobic—small neutral or polar residue—branched-chain amino acids (at the P1 position). These cleavages would be consistent with the so-called BrAAP activity and the enhanced capacity of the immmunoproteasomes to cleave model peptides after hydrophobic residues. Moreover, even with small peptide substrates and peptide inhibitors, there is strong evidence that, in addition to P1 residues, the p4 or p5 residues strongly influence active-site preference (165, 166). Furthermore, because the activity of the proteasome’s different active sites is typically studied with model peptides containing a C-terminal fluorescent-reporter group, the influence of residues C terminal to the cleavage site (commonly termed P10 , P20 , etc) on rates of peptide-bond cleavage are unknown. In fact, the influence of the different side chains at each position (P1, P2, etc, and P10 , P20 , etc) on cleavage rates by the different active sites has never been systematically investigated, either for normal proteasomes or the immunoproteasomes. Additional evidence that preference for the P1 position is not absolute comes from the finding that archaebacterial proteasomes, which contain only one type of active site, rapidly cleave chymotryptic substrates, but also more slowly hydrolyze model peptides after acidic and even basic residues (at 5–10% the rate of the chymotryptic activity) (85).

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

759

The effects of LMP proteins and PA28 on cleavage patterns have been examined for some long oligopeptide antigens. Purified proteasomes that lack LMPs and PA28 are capable of generating presented peptides, as can proteasomes from invertebrates or archaebacteria (167). In some experiments, incorporation of LMP proteins has been reported to alter the sites where cleavages occurred in polypeptides. In other experiments, the rate of the generation of antigenic peptides was enhanced, but the nature of the cleavage sites was not altered (114, 124, 154, 156, 167). Similarly, PA28 has been reported to alter the pattern of peptide bonds cut by 20S proteasomes (168) and to favor multiple (or “dual”) cleavages in a substrate so as to enhance the generation of the correct peptide for antigen presentation (169, 170). However, other reports suggest that PA28 may increase the generation of antigenic peptides without altering the sites of cleavage (167). These differing results probably reflect the use of different substrates and very different experimental conditions. Although the in vivo relevance of these various experiments is uncertain (see below), the interferoninduced proteasome subunits and PA28 do seem to consistently alter the rates and nature of peptides generated from model substrates. For many reasons, it seems unlikely that such simple experiments with purified 20S particles accurately reflect the in vivo process of peptide generation. The conditions used in such differ widely from the reactions that occur in intact cells. In the in vitro experiments, substrates were not ubiquitinated, and such studies have used 20S proteasomes instead of the 26S particle. Moreover, to obtain sufficient rates of hydrolysis, the 20S proteasomes were activated, for example, with SDS, and recent data indicate that the peptides released during protein breakdown by these activated particles vary depending on the mode of activation, and are not identical to those generated by the 26S complexes (161). It is also unclear whether the 10- to 30-residue fragments typically used as substrates in such studies are cleaved at the same positions as when the same sequence is presented in the context of a protein. In the 26S complex, the many ATPases of the 26S proteasome and the ubiquitin on the substrate may influence the mode of substrate entry into the 20S particle. Also, PA700 activates the different peptidase sites of the 20S core particle in distinct fashions. Finally, most studies have used extremely long incubations (as long as 24 h), during which time the properties of proteasomes change, and the initial products of proteasome digestion can reenter the particle and be degraded further, an event that is quite unlikely to occur in vivo, because the cytoplasm is rich in peptidases that quickly convert generated peptides to free amino acids. Also, most experiments have been of a qualitative nature, and, therefore, it is impossible to determine whether an antigenic peptide was generated by (or destroyed by) the proteasomes during the initial degradation of a protein or in later proteolytic cycles. Such multiple cycles may be avoided if protein substrates are present in

P1: JER/MBG

P2: NBL/spd

February 11, 1999

760

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

excess, and linear conditions are maintained (95). However, the recent demonstration in cell-free extracts of the complete ubiquitin–proteasome-mediated degradation of ovalbumin to the presented peptide suggests that it should be possible to reconstitute and analyze the entire intracellular process in vitro in a physiologically relevant manner (63).

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Do Proteasomes In Vivo Generate the N and C Termini of Presented Peptides? Recently, surprising findings were made concerning the nature of the enzymes that generate the N- and C-terminal residues of a presented peptide through studies of the presentation of a series of N- or C-terminally extended antigenic peptides in intact cells (171). When constructs consisting of the antigenic peptide SIINFEKL with one to five additional carboxy-terminal flanking residues were introduced into the cytoplasm, the appropriate eight-residue antigenic peptide was generated and presented on MHC class I molecules. Production of this peptide was inhibited by proteasome inhibitors, even when the expressed peptide contained only a single C-terminal–flanking residue. Thus, the proteasome appears to make the cleavages that generate the C terminus of this antigen. Because similar results have also been obtained on expression of extended versions of two other antigenic peptides (XY Mo & KL Rock, unpublished results), the proteasome seems to generate the C-terminal residues of many (and possibly most) presented peptides. This conclusion is consistent with earlier predictions, based on the data with model peptides, that the interferon-induced subunits alter the C termini of proteasomal products (90, 148, 156). Moreover, these results and earlier observations with a C-terminally extended influenza peptide that was poorly presented (172, 173) suggest that in the cytosol there are few proteases that can rapidly remove the C-terminal residues from antigenic peptides or other proteasome products. This conclusion has been supported by the failure to find significant carboxypeptidase activity in cytosolic extracts (174; see below). Although these results imply that the proteasome generates the correct C terminus of antigenic peptides, very different results were obtained upon expression or electroporation of SIINFEKL constructs that were extended with N-terminal–flanking sequences (171). When introduced into the cells, these constructs were presented more rapidly (and efficiently) than constructs with C-terminal extensions. Moreover, their presentation was not inhibited at all by proteasome inhibitors. Similar results have been observed with N-extended versions of other antigenic peptides (XY Mo & KL Rock, unpublished results). Therefore, the N and C termini of antigenic peptides can be generated by distinct enzymes in vivo (171). Several observations make it unlikely that the removal of these N-terminal–flanking residues is mediated by a proteasome site that is relatively insensitive to the inhibitors. For example, the presentation of the N-extended constructs is not affected by very high concentrations of lactacystin

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

761

and tripeptide aldehydes, which can inhibit all of the proteasome’s active sites. Moreover, we have been able to identify an aminopeptidase in cytosolic extracts, which efficiently removes the N-terminal–flanking sequences (174; see below). In addition, acetylation of the amino terminus of an extended construct, which prevents trimming by aminopeptidases (but not endopeptidases), prevents the presentation of these peptides (XY Mo, P Cascio, & KL Rock, unpublished results). It thus appears likely that in vivo, there are multiple enzymes that can generate the N terminus of a presented peptide, and their relative importance may depend on the protein’s sequence. During breakdown of some proteins, the proteasome may generate the final N terminus as well as the C termini of the presented peptide. Such dual cleavages by the proteasome could be the major source of presented peptides from certain proteins. Alternatively, for other proteins, the proteasome may release primarily N-extended versions of the antigenic peptide, and then cytosolic aminopeptidases (or endopeptidases) may trim them to the presented 8 mer or 9 mer. In fact, when an N-terminally extended peptide was expressed in the cytoplasm in a form that was almost certainly too long (32 residues) to be transported by TAP into the ER (175, 176), it was appropriately trimmed, and the correct epitope presented on the surface. This process was not blocked by proteasome inhibitors (171). Moreover, several exopeptidases that can trim the N terminus of peptides are demonstrable in proteasome-free cytosolic extracts (174; see below). In addition, N-terminally-extended peptides that are directed into the ER by a colinear signal sequence are also efficiently trimmed to the appropriate size (171, 173, 177). Proteasomes are not present in the ER lumen, and this trimming was not blocked by proteasome inhibitors. Because identical results were obtained in TAP-deficient cells, it is unlikely that these peptides were transported back into the cytoplasm and their N termini removed there. Thus, proteases capable of trimming the N termini of proteasome products exist in both the ER and cytoplasm. Further studies are needed to identify the responsible proteases and to define their relative importance in the presentation of different epitopes. Presumably, some mechanism retards this trimming process when the presented 8 mer or 9 mer is generated. Perhaps further cleavages by leucine aminopeptidase are not favored structurally, or, once the optimal size is reached, binding to an unknown molecular chaperone, to TAP, or mostly likely to the appropriate MHC molecules may prevent further aminopeptidase action.

Cytosolic Proteases That May Generate the N Termini of Antigenic Peptides To define the enzymes that remove the N-terminal–flanking residues from antigenic peptides, studies were carried out with model peptides in proteasomefree cell extracts. The ovalbumin-derived peptide QLESIINFEKL, which is

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

762

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

trimmed to SIINFEKL in cells, was incubated in soluble extracts from HeLa or U937 cells, after removal of proteasomes. When analyzed by high-performance liquid chromatography, the 11-residue peptide decreased with time owing to the sequential removal of the amino-terminal residues. Thus, an exopeptidase shown to be a metalloaminopeptidase was trimming the 11mer to the 10 mer, 9 mer, and eventually the 8 mer. Remarkably, cytosolic extracts from IFN-γ – treated cells trimmed the N-extended antigenic peptides more rapidly than did extracts from controls. Thus, IFN-γ induces an N-terminal exopeptidase activity, which was purified from these extracts and identified as leucine aminopeptidase (174). This enzyme is induced by IFN-γ , simultaneously with the appearance or LMP subunits, and after several days, the cells contain two- to fourfold higher levels of this aminopeptidase than do controls. Moreover, purified leucine aminopeptidase trimmed this precursor to SIINFEKL in a fashion similar to that of the extracts from interferon-treated cells. Because IFN-γ increases the expression of many components of the pathway for class I presentation, including the various MHC molecules, TAP, LMP2 and 7, MECL1, and tapasin, the finding that leucine aminopeptidase is coordinately induced makes it likely that this enzyme plays a role in trimming precursors to many antigenic peptides. Although cells also contain a number of other aminopeptidases, they appear less active against the 11 mer than leucine aminopeptidase, and their activities were not induced by IFN-γ . However, it is possible that these exopeptidases preferentially trim precursors to other antigenic peptides. In these cell-free extracts, most of the loss of the 11-residue peptide could be accounted for by N-terminal trimming, and there was little, if any, carboxypeptidase activity that could trim the C-terminal residues on these peptides or C-terminal–extended versions. This finding correlates well with the findings described above that in cells the trimming of C-terminally extended peptides was exclusively mediated by proteasomes (171). Moreover, it is consistent with the proposal that the interferon-induced proteasome subunits enhance the efficiency of generation of hydrophobic or basic C-terminal residues appropriate for antigen presentation.

Are There Proteasome-Independent Mechanisms for Generating Antigenic Peptides? Several studies have suggested that cells contain additional cytosolic proteases that may also generate presented peptides. Because proteasome inhibitors fail to completely prevent the assembly of class I molecules, it has been proposed that some peptides may also be supplied to newly assembling class I molecules by a distinct proteolytic system (8, 119, 178). Furthermore, in certain studies, antigen presentation from some proteins was not even reduced by treatment

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

763

with proteasome inhibitors (178–181). The interpretation of these observations, however, depends on whether proteasome activity was totally blocked under these conditions, and, in these experiments, complete inhibition of proteasomemediated protein breakdown was assumed, but not measured. In some of these studies, adequate concentrations of the inhibitors may not have been used to completely block proteasome function. Perhaps the strongest evidence for an alternate proteolytic system in the cytoplasm comes from studies in which cells have been grown chronically in the presence of a vinyl-sulfone proteasome inhibitor (182). Under these conditions, a new proteolytic activity seems to be induced, which allows the cell to survive and to generate peptides for class I presentation. It has been proposed that this activity may be related to the tricorn protease of archaebacteria (183). However, such an enzyme has not yet been demonstrated in mammalian cells, although some evidence for a similar activity has been obtained in Schizosaccharomyces pombe (184). Because most proteasome subunits and ubiquitin-dependent proteolysis are essential for viability in yeast (53), it is very difficult to conceive of a eukaryotic cell functioning without the ubiquitin-proteasome pathway, and it remains possible that protein breakdown in the vinyl-sulfone–treated cells is by an altered form of the proteasome exhibiting a novel vinyl-sulfone–resistant mode of protein hydrolysis. There is strong evidence that at least some presented peptides can be generated in the ER. In TAP-deficient cells, peptides derived from a signal sequences are generated and presented on class I molecules (17, 18). Presumably, these peptides are generated in the lumen of the ER by the signal peptidase, perhaps in conjunction with other lumenal proteases. Similarly, some peptides from Jaw 1, an ER-resident protein, and from HIV gp120, which is transported into the ER, are presented independently of TAP function and are presumably generated by ER-lumenal proteases (20, 21). Finally, in phagocytes such as macrophages, some class I-presented peptides are generated from proteins degraded in phagolysosomes (185, 186). It has been proposed that these peptides are regurgitated into the extracellular fluids where they bind to class I molecules on the cell surface. However, antigens or fragments arising in the endocytic compartments of macrophages and dendritic cells can also escape into the cytoplasm, where they can be degraded by proteasomes and presented via the conventional class I pathway (186, 187). This pathway for class I presentation of antigens from extracellular polypeptides has not been observed in other cell types.

Intracellular Destruction of Antigenic Peptides Upon incubation with presented peptides, purified 20S proteasomes can cleave within the internal sequences of many antigenic peptides (124, 125, 164). If this process occurs during the degradation of cell proteins in vivo, it clearly may

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

764

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

limit the presentation of some antigens. For example, 20S proteasomes cleave within the ovalbumin-derived peptide (NKVVRFDKL), and this effect may explain why this particular peptide is subdominant (i.e. is a weak determinant that does not generally elicit CTL responses) (125). Similarly, proteasomes cleave within a CTL epitope faster than the wild type from a murine-leukemia virus that contains a single amino acid change epitope, and this variant is presented on class I molecules less efficiently than on the wild-type proteins (188). Another line of evidence suggesting that proteasomes may destroy some presented peptides is the finding that proteasome inhibitors enhance the presentation of certain peptides from protein antigens (180, 181). We have even found examples in which the presentation of an antigen is enhanced by low doses of a proteasome inhibitor, but blocked by high concentrations (XY Mo & KL Rock, unpublished data). Caution is warranted in interpreting these interesting observations, because the binding of a substrate or an inhibitor to one active site can activate the proteasome’s other sites (69). Therefore, proteasome inhibitors might augment the presentation of a peptide by either preventing its destruction or by stimulating its generation. It is noteworthy that, in vivo, the overwhelming majority of peptides released by the proteasome are degraded rapidly to amino acids. Therefore, in addition to the proteasome, there must be multiple enzymes in the nucleus and cytosol that can destroy oligopeptides. This process occurs very rapidly in vivo, and, consequently, free peptides cannot be detected in the cytoplasm. In fact, when antigenic peptides are produced in cells, they cannot be detected unless they become bound to class I molecules (189) or perhaps to chaperones, such as HSC70 or grp96 (190, 191), which may protect them from further degradation. Therefore, a major challenge is to understand how antigenic peptides can escape the fate of all other peptides released by the proteasome. The cellular proteases that rapidly digest most proteasome-derived oligopeptides to amino acids are poorly characterized, but this process almost certainly involves further fragmentation by endopeptidases followed by attack on the small pieces by aminopeptidases. Accordingly, blocking aminopeptidases with bestatin in cell extracts leads to the accumulation of di- or tripeptides derived from the ubiquitin-proteasome pathway (192). Presented peptides are not inherently resistant to this degradative process, because when seven different antigenic peptides were incubated in cytosolic extracts, they were all rapidly degraded by a metalloprotease-dependent process (193). These observations also suggest that the antigenic peptides are also not largely protected by binding to cytosolic chaperones (at least in cell extracts). The cytosolic enzyme that cleaves most of the antigenic peptides has been identified as thimet oligopeptidase (193; C Dax & AL Goldberg, manuscript in preparation). The physiological functions of this metalloendopeptidase had been unclear; it had been

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

765

believed to play a role in neuropeptide metabolism (194, 195), and it is related to the yscD oligopeptidase in yeast, which has been proposed to play a role in the final intracellular steps in protein turnover (196). In addition to cleaving rapidly these antigenic peptides in cell extracts, this endopeptidase rapidly degrades proteasome products of ≤20 residues in length to pieces that are further digested by aminopeptidases to amino acids (C Dax & AL Goldberg, manuscript in preparation). However, in cytosolic extracts, some other as yet unidentified metallopeptidases also contribute to the rapid digestion of proteasome products and other antigenic peptides (193). Of particular interest was the finding that the activity and content of thimet oligopeptidase were reduced in HeLa cells after IFNγ treatment (193). This enzyme thus is controlled in opposite fashion to leucine aminopeptidase and to antigen presentation. Down-regulation of this degradative enzyme suggests that, after interferon treatment, antigenic peptides released from the proteasome are likely to be more stable in the cytosol. Presumably, therefore, more peptides would be available for antigen presentation (i.e. for transport by TAP or N-terminal trimming by leucine aminopeptidase). Much remains to be learned about these final post-proteasomal steps, which have been a neglected aspect of this degradative pathway. These peptidases are of particular immunological interest, because they should limit the availability of peptides for antigen presentation unless there are mechanisms to protect the initial peptides from destruction. The efficiency of this whole process is quite unclear. It is unknown with what frequency the correct peptide, or an extended version of it, is generated by proteasomes, or what fraction of the generated appropriate peptides can escape digestion by cytosolic peptidases. Interestingly, in one case, as few as 3 (and in another case, as few as 35) antigen molecules had to be degraded to yield one peptide presented on a class I molecule (197, 198). This relatively high yield would not be expected if peptides were free to attack by cytosolic exo- and endopeptidases. One strong hint that cytosolic proteases might limit antigen presentation is the finding, described above, that the expression of thimet oligopeptidase is suppressed by IFN-γ . Another important implication of these observations is that the relative susceptibility of antigenic peptides or potentially antigenic ones to such endoor exoproteases may be a determinant of immunodominance.

How Does the Primary Sequence of an Antigen Affect the Generation of Presented Peptides? It is possible to identify many potential peptides, within foreign proteins, that have the appropriate motifs to bind to MHC class I molecules (199). However, most often the immune system responds to very few of these potential epitopes, and, in many cases, the failure to present such epitopes is caused by a failure of

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

766

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

the cells to generate them (199). To understand this failure, greater knowledge must be obtained not only about the function of the 26S proteasomes and immunoproteasomes and how they select sites for cleavage in protein substrates, but also about the specificities and cleavage preferences of the endopeptidases and aminopeptidases that may trim or destroy the proteasome products. If the sequence “motifs” they recognize could be defined, then we could better predict what peptides would get presented. Such information could have obvious practical applications; for example, antigens might be able to be engineered to make epitopes more available in vaccines. Although the picture is still very incomplete, several studies have examined how sequences in and around the presented peptides influence their cleavage in vivo and presentation on class I molecules. Most studies to date have examined the influence on antigen presentation of residues flanking a presented peptide. Early studies made chimeric-fusion proteins containing an antigenic peptide within another protein. In some cases, the new residues flanking the peptide did not affect antigen presentation (200–202), whereas in other cases, these flanking sequences increased or decreased presentation (203, 204). For one of the epitopes whose presentation was affected, TYQRTRALV from influenza nucleoproteins, mutating the residue on either side of this peptide to a proline markedly inhibited presentation (70). Since proteasome inhibitors do not block the generation of this peptide, these changes may prevent the activity of some other protease. Most subsequent studies have modified flanking residues in minigene constructs, and, in at least one case, different results were obtained than with the same mutations in a full-length construct (70). Again, alterations in the flanking sequence can promote or decrease antigen presentation, and, in some cases, these changes correlate with the ability of purified proteasomes to cleave the altered sequence in extended peptides (125, 180). In a few studies, alterations in the C-terminal–flanking region, but not the N-terminal extension, influenced presentation (205, 206). Interestingly, purified proteasomes are unable to generate the correct C-terminal cleavage to generate an epitope from a p53 tumor suppressor protein that has a point mutation in the P10 flanking residue, and this mutation inhibits the presentation of this same antigenic peptide in cells (207). Also, as discussed above, sequence alteration within an antigenic epitope from a strain of murineleukemia virus results in more rapid cleavage within the peptide by purified proteasomes and less efficient MHC class I presentation by cells (188). It has also been proposed that proline residues in an epitope are important for promoting presentation, because they seem to reduce proteasomal destruction of an epitope (208). Apart from these limited examples, there have been no systematic investigations of how residues in a presented peptide influence its production by isolated proteasomes and the efficiency of its presentation. Such

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION

767

studies are difficult, in part, because changes in sequence can also alter MHC binding and T-cell recognition. Taken together, these results indicate that both internal and flanking residues can influence the generation of presented peptides. However, much more work is needed to define the key residues that influence antigen presentation and to determine whether they do so by altering protein ubiquitination, rates of degradation by proteasomes, sites of polypeptide chain cleavage, and susceptibility to degradation (or trimming) by cytosolic peptidases. Interestingly, even sequences that are not directly adjacent to an epitope can influence antigen presentation. One clear example of this effect is with the EBNA1 protein of Epstein-Barr virus, which fails to stimulate CTL responses. This protein contains an unusual repeating gly-ala sequence that allows ubiquitin conjugation, but somehow prevents degradation by 20S proteasomes, and thus may help the virus to evade detection by the immune system (209, 210). Interestingly, transfer of this sequence to a normally immunogenic antigen blocks its presentation. In addition to preventing degradation by the proteasome, sequences that are more distant from the antigenic epitopes may affect presentation in other ways. As discussed above N-terminal modifications that enhance ubiquitination and degradation can greatly accelerate presentation of a specific epitope (61). Also, because the proteasome degrades proteins in a highly processive manner, sequences that determine where initial cleavages are made might constrain where subsequent cleavages occur. Despite all the progress in our understanding of class I presentation, we still are unable to predict immunogenic regions within protein antigens or to explain why some sequences are immunodominant. This goal will obviously require greater understanding of the molecular mechanisms for protein breakdown within the proteasome and especially the factors determining the sites of polypeptide cleavage, but it will also require detailed information about the specificities of the peptidases that act subsequently to trim or to destroy antigenic peptides. Once we better understand these issues it should be possible to better predict what peptides will get presented, and it may be possible to optimize the sequences of epitopes to enhance their bioavailability in vaccines.

Summary and Outstanding Issues To screen for the appearance of viruses or transformed cells, the immune system has taken advantage of the phylogenetically older catabolic system that functions in the continual turnover of cell proteins. The turnover of most cell proteins occurs via the ubiquitin-proteasome pathway, and it is now well established through studies with mutants in ubiquitin conjugation, proteasome inhibitors, LMP-deficient cells and mice, and model studies with isolated proteasomes that this particle is the source of the great majority of class I-presented

P1: JER/MBG

P2: NBL/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

768

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG

peptides. The proteasome degrades proteins processively to oligopeptides, ranging from 4–20 residues long, and generates some with C termini appropriate for MHC binding. If antigenic peptides with N-terminal extensions are generated, they can be trimmed by aminopeptidases in the cytosol and ER to the presented octa- or nonapeptide. However, successful presentation also requires that the peptide must escape destruction by cytosolic peptidases. Several mechanisms have evolved that modify this basic process to enhance the generation of class I-presented peptides. IFN-γ stimulates this process in part by inducing three ß-subunits (LMP2, 7, and MECL-1) that replace homologous subunits in new proteasomes. The resulting “immunoproteasomes” cleave peptides after hydrophobic and basic residues at enhanced rates, which should increase the production of peptides capable of binding to class I molecules. IFN-γ also induces PA28, a ring-shaped activator of peptide hydrolysis by the 20S particle, as well as leucine aminopeptidase, which can trim the N termini of longer peptides to generate the presented peptide. In addition, IFN-γ reduces the content of thimet oligopeptidase, an endopeptidase, which in cell extracts appears to destroy many antigenic peptides. Although there has been remarkable progress in delineating the steps involved in antigen presentation and their mode of regulation, many fundamental questions remain unresolved. It is uncertain whether proteasome-mediated degradation and antigen presentation in general require ubiquitination and the 26S form of the proteasome. What is the precise role of PA28 activator in this process, and how do ß-subunits actually influence the cleavages made in a protein? How often are the 8 mers and 9 mers generated directly by proteasomes, and how often are they produced from longer proteasome products by cytosolic or ER aminopeptidases? What determines whether an individual peptide released by the proteasome is rapidly hydrolyzed to amino acids or escapes destruction and is transported into the ER for class I presentation? Also, largely unknown are the variety of enzymes catalyzing these post-proteasome steps, even though their substrate specificities are likely to be of appreciable immunological importance. Finally, it is uncertain whether other immune modifiers, like IFN-γ , also can stimulate these (or some other) steps in antigen presentation. On the other hand, in this multistep pathway, there are multiple possible steps by which viruses may have evolved mechanisms to inhibit in order to avoid immune detection (211). Insights into all of these questions are not only of appreciable scientific importance but may also have valuable practical applications, such as making it possible to optimize epitopes to enhance their efficacy as vaccines. ACKNOWLEDGMENTS We are grateful to our many colleagues who communicated results that were not yet published. We thank Aurora Scott for valuable assistance in the

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

MHC CLASS I ANTIGEN PRESENTATION

769

preparation of this manuscript. This work was supported by grants from the National Institutes of Health to KLR and from the National Institutes of Health, Human Frontier Science Program, and Muscular Dystrophy Association to ALG. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Literature Cited 1. York I, Rock KL. 1996. Antigen processing and presentation by the class I major histocompatibility complex. Annu. Rev. Immunol. 14:369–96 2. Williams D, Watts T. 1995. Molecular chaperones in antigen presentation. Curr. Opin. Immunol. 7:77–84 3. Hughes EA, Cresswell P. 1998. The thiol oxidoreductase ERp57 is a component of the MHC class I peptide-loading complex. Curr. Biol. 8:709–12 4. Lindquist JA, Jensen ON, Mann M, Hammerling GJ. 1998. ER-60, a chaperone with thiol-dependent reductase activity involved in MHC class I assembly. EMBO J. 17:2186–95 5. Ortmann B, Copeman J, Lehner PJ, Sadasivan B, Herberg JA, Grandea AG, Riddell SR, Tampe R, Spies T, Trowsdale J, Cresswell P. 1997. A critical role for tapasin in the assembly and function of multimeric MHC class I-TAP complexes. Science 277:1306–9 6. Madden DR. 1995. The three-dimensional structure of peptide-MHC complexes. Annu. Rev. Immunol. 13:587– 622 7. Braciale TJ, Morrison LA, Sweetser MT, Sambrook J, Gething M-J, Braciale VL. 1987. Antigen presentation pathways to class I and class II MHC-restricted T lymphocytes. Immunol. Rev. 98:95–114 8. Rock KL, Gramm C, Rothstein L, Clark K, Stein R, Dick L, Hwang D, Goldberg AL. 1994. Inhibitors of the proteasome block nonlysosomal protein degradation and the generation of peptides presented on MHC-class I molecules. Cell 78:761– 71 9. Germain RN, Margulies DH. 1993. The biochemistry and cell biology of antigen processing and presentation. Annu. Rev. Immunol. 11:403–50 10. Bennink JR, Yewdell JW, Gerhard W. 1982. A viral polymerase involved in recognition of influenza virus-infected

11.

12.

13.

14.

15.

16.

17.

18.

19.

cells by a cytotoxic T-cell clone. Nature 296:75–76 Gooding LR, O’Connell KA. 1983. Recognition by cytotoxic T lymphocytes of cells expressing fragments of the SV40 tumor antigen. J. Immunol. 131:2580–86 Townsend ARM, Bastin J, Gould K, Brownlee GG. 1986. Cytotoxic T lymphocytes recognize influenza hemagglutinin that lacks a signal sequence. Nature 324:575–77 Moore MW, Carbone FR, Bevan MJ. 1988. Introduction of soluble protein into the class I pathway of antigen processing and presentation. Cell 54:777– 85 Yewdell JW, Bennick JR, Hosaka Y. 1988. Cells process exogenous proteins for recognition by cytotoxic T lymphocytes. Science 239:637–40 ¨ Townsend A, Ohlen C, Bastin J, Ljunggren H-G, Foster L, K¨arre K. 1989. Association of class I major histocompatibility heavy and light chains induced by viral peptides. Nature 340:443–48 Momburg F, Hammerling GJ. 1998. Generation and TAP-mediated transport of peptides for major histocompatibility complex class I molecules. Adv. Immunol. 68:191–256 Henderson RA, Michel H, Sakaguchi K, Shabanowitz J, Appella E, Hunt DF, Engelhard VH. 1992. HLA-A2.1associated peptides from a mutant cell line: a second pathway of antigen presentation. Science 255:1264–66 Kurepa Z, Hasemann C, Forman J. 1998. Qa1b binds conserved class I leader peptides derived from several mammalian species. Submitted for publication Zweerink HJ, Gammon MC, Utz U, Sauma SY, Harrer T, Hawkins JC, Johnson RP, Sirotina A, Hermes JD, Walker BD, et al. 1993. Presentation of endogenous peptides to MHC class I-restricted

P1: JER/MBG

P2: NBL/spd

February 11, 1999

770

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

20.

21.

22. 22a.

23.

24.

25.

26. 27.

28.

29.

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG cytotoxic T lymphocytes in transport deletion mutant T2 cells. J. Immunol. 150:1763–71 Hammond SA, Johnson RP, Kalams SA, Walker BD, Takiguchi M, Safrit JT, Koup RA, Siliciano RF. 1995. An epitope-selective transporter associated with antigen presentation (TAP)-1/2independent pathway and a more general TAP-1/2-dependent antigen-processing pathway allow recognition of the HIV1 envelope glycoprotein by CD8+ CTL. J. Immunol. 154:6140–56 Snyder HL, Bacik I, Bennink JR, Kearns G, Behrens TW, Bachi T, Orlowski M, Yewdell JW. 1997. Two novel routes of transporter associated with antigen processing (TAP)-independent major histocompatibility complex class I antigen processing. J. Exp. Med. 186:1087–98 Ward CL, Omura S, Kopito RR. 1995. Degradation of CFTR by the ubiquitinproteasome pathway. Cell 83:121–27 Jensen TJ, Loo MA, Pind S, Williams DB, Goldberg AL, Riordan JR. 1999. Multiple proteolytic systems, including the proteasome, contribute to CFTR processing. Cell 83:129–35 Huppa JB, Ploegh HL. 1997. The alpha chain of the T cell antigen receptor is degraded in the cytosol. Immunity 7:113– 22 Hughes EA, Hammond C, Cresswell P. 1997. Misfolded major histocompatibility complex class I heavy chains are translocated into the cytoplasm and degraded by the proteasome. Proc. Natl. Acad. Sci. USA 94:1896–901 Sommer T, Wolf DH. 1997. Endoplasmic reticulum degradation: reverse protein flow of no return. FASEB J. 11:1227–33 Hilt W, Wolf D. 1996. Proteasomes: destruction as a programme. Trends Biochem. Sci. 21:96–102 Mosse C, Meadows L, Luckey C, Kittlesen D, Huczko E, Slingluff C, Shabanowitz J, Hunt D, Engelhard V. 1998. The class I antigen-processing pathway for the membrane protein tyrosinase involves translation in the endoplasmic reticulum and processing in the cytosol. J. Exp. Med. 187:37–48 Van Pel A, Boon T. 1989. T cellrecognized antigenic peptides derived from the cellular genome are not protein degradation products but can be generated directly by transcription and translation of short subgenic regions. A hypothesis. Immunogenetics 29:75–79 Shastri N, Nguyen V, Gonzalez F.

30.

31.

32.

33.

33a.

33b. 34.

35.

36.

37.

38. 39. 40.

1995. Major histocompatibility class I molecules can present cryptic translation products to T-cells. J. Biol. Chem. 270:1088–91 Yewdell J, Anton L, Bennink J. 1996. Defective ribosomal products (DRiPs): a major source of antigenic peptides for MHC class I molecules? J. Immunol. 157:1823–26 Bullock T, Eisenlohr L. 1996. Ribosomal scanning past the primary initiation codon as a mechanism for expression of CTL epitopes encoded in alternative reading frames. J. Exp. Med. 184:1319– 29 Loveland B, Wang CR, Yonekawa H, Hermel E, Lindahl KF. 1990. Maternally transmitted histocompatibility antigen of mice: a hydrophobic peptide of a mitochondrially encoded protein. Cell 60:971–80 Desautels M, Goldberg AL. 1982. Demonstration of an ATP-dependent, vanadate-sensitive endoprotease in the matrix of rat liver mitochondria. J. Biol. Chem. 257:11673–79 Kuzela S, Goldberg AL. 1994. The mitochondrial ATP-dependent protease from rat liver and yeast. Methods Enzymol., pp. 376–83 Langer T, Neupert W. 1996. Regulated protein degradation in mitochondria. Experientia 52:1065–76 Goldberg AL. 1992. The mechanism and functions of ATP-dependent proteases in bacterial and animal cells. Eur. J. Biochem. 203:9–23 Gottesman S, Maurizi M. 1992. Regulation by proteolysis: energy-dependent proteases and their targets. Microbiol. Rev. 56:592–621 Lindahl KF, Byers DE, Dabhi VM, Hovik R, Jones EP, Smith GP, Wang CR, Xiao H, Yoshino M. 1997. H2-M3, a full-service class Ib histocompatibility antigen. Annu. Rev. Immunol. 15:851– 79 Goldberg A, St. John A. 1976. Intracellular protein degradation in mammalian and bacterial cells. Annu. Rev. Biochem. 45:747–803 Rechsteiner M. 1987. Ubiquitin-mediated pathways for intracellular proteolysis. Annu. Rev. Cell Biol. 3:1–30 Hershko A, Ciechanover A. 1992. The ubiquitin system for protein degradation. Annu. Rev. Biochem. 61:761–807 Murray A, Kirschner M. 1989. Dominoes and clocks: the union of two views of the cell cycle. Science 246:614– 21

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION 41. Palombella V, Rando O, Goldberg A, Maniatis T. 1994. The ubiquitinproteasome pathway is required for processing the NF-κB1 precursor protein and the activation of NF-κB. Cell 78:773–85 42. Orian A, Whiteside S, Israel A, Stancovski I, Schwartz AL, Ciechanover A. 1995. Ubiquitin-mediated processing of NF-κB transcriptional activator precursor p105. Reconstitution of a cellfree system and identification of the ubiquitin-carrier protein, E2, and a novel ubiquitin-protein ligase, E3, involved in conjugation. J. Biol. Chem. 270:21707– 14 43. Dice J. 1987. Molecular determinants of protein half-lives in eukaryotic cells. FASEB J. 1:349–57 44. Morrison LA, Lukacher A, Braciale VL, Fan DP, Braciale TJ. 1986. Differences in antigen presentation to MHC class I and class II–restricted influenza virusspecific cytolytic T lympyhocyte clones. J. Exp. Med. 163:903–21 45. Gronostajski R, Pardee A, Goldberg A. 1985. The ATP dependence of the degradation of short- and long-lived proteins in growing fibroblasts. J. Biol. Chem. 260:3344–49 46. Etlinger J, Goldberg A. 1977. A soluble ATP-dependent proteolytic system responsible for the degradation of abnormal proteins in reticulocytes. Proc. Natl. Acad. Sci. USA 74:54–58 47. Hershko A, Ciechanover A. 1992. The ubiquitin system for protein degradation. Annu. Rev. Biochem. 61:761–807 48. Hershko A. 1988. Ubiquitin-mediated protein degradation. J. Biol. Chem. 263: 15237–40 49. Finley D, Chau V. 1991. Ubiquitination. Annu. Rev. Cell Biol. 7:25–69 50. Ciechanover A. 1994. The ubiquitinproteasome proteolytic pathway. Cell 79:13–21 51. Hershko A, Ciechanover A. 1998. The ubiquitin system. Annu. Rev. Biochem. 67:425–79 52. Hershko A, Ganoth D, Sudakin V, Dahan A, Cohen LH, Luca FC, Ruderman JV, Eytan E. 1994. Components of a system that ligates cyclin to ubiquitin and their regulation by the protein kinase cdc2. J. Biol. Chem. 269:4940–46 53. Coux O, Tanaka K, Goldberg A. 1996. Structure and functions of the 20S and 26S proteasomes. Annu. Rev. Biochem. 65:801–4 54. Ciechanover A, Finley D, Varshavsky A. 1984. Ubiquitin dependence of selective

55.

56.

57.

58.

59.

60.

61.

62.

63.

64. 65.

771

protein degradation demonstrated in the mammalian cell cycle mutant ts85. Cell 37:57–66 Murakami Y, Matsufuji S, Kameji T, Hayashi S, Igarashi K, Tamura T, Tanaka K, Ichihara A. 1992. Ornithine decarboxylase is degraded by the 26S proteasome without ubiquitination. Nature 360:597–99 Driscoll J, Goldberg A. 1990. The proteasome (multicatalytic protease) is a component of the 1500-kDa proteolytic complex which degrades ubiquitinconjugated proteins. J. Biol. Chem. 265:4789–92 Baumeister W, Walz J, Zuhl F, Seemuller E. 1998. The proteasome: paradigm of a self-compartmentalizing protease. Cell 92:367–80 Zwickl P, Woo K, Ng D, Klenk H-P, Goldberg A. 1998. A novel 650 kDATPase complex, related to the eukaryotic 26S ATPases, activates protein breakdown by 20S proteasomes from Arhaea. Submitted for publication Townsend A, Bastin J, Gould K, Brownlee G, Andrew M, Coupar B, Boyle D, Chan S, Smith G. 1988. Defective presentation to class I-restricted by cytotoxic T lymphocytes in vacciniainfected cells is overcome by enhanced degradation of antigen. J. Exp. Med. 168:1211–24 Michalek M, Grant E, Gramm C, Goldberg A, Rock K. 1993. A role for the ubiquitin-dependent proteolytic pathway in MHC class I-restricted antigen presentation. Nature 363:552–54 Grant E, Michalek M, Goldberg A, Rock K. 1995. Rate of antigen degradation by the ubiquitin-proteasome pathway influences MHC class I presentation. J. Immunol. 155:3750–58 Michalek M, Grant E, Rock K. 1996. Chemical denaturation and modification of ovalbumin alters its dependence on ubiquitin conjugation for class I antigen presentation. J. Immunol. 157:617–24 Ben-Shahar S, Cassouto B, Novak L, Porgador A, Reiss Y. 1997. Production of a specific major histocompatibility complex class I-restricted epitope by ubiquitin-dependent degradation of modified ovalbumin in lymphocyte lysate. J. Biol. Chem. 272:21060–66 Varshavsky A. 1992. The N-end rule. Cell 69:725–35 Tobery T, Siliciano R. 1997. Targeting of HIV-1 antigens for rapid intracellular degradation enhances cytotoxic T lymphocyte (CTL) recognition and the

P1: JER/MBG

P2: NBL/spd

February 11, 1999

772

66.

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

67.

68.

69.

70.

71.

72.

73. 74. 75. 76.

77. 78.

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG induction of de novo CTL responses in vivo after immunization. J. Exp. Med. 185:909–20 Wu Y, Kipps T. 1997. Deoxyribonucleic acid vaccines encoding antigens with rapid proteasome-dependent degradation are highly efficient inducers of cytolytic T lymphocytes. J. Immunol. 159:6037–43 Rodriguez F, Zhang J, Whitton J. 1997. DNA immunization: Ubiquitination of a viral protein enhances cytotoxic Tlymphocyte induction and antiviral protection but abrogates antibody induction. J. Virol. 71:8497–503 Goth S, Nguyen V, Shastri N. 1996. Generation of naturally processed peptide/MHC class I complexes is independent of the stability of endogenously synthesized precursors. J. Immunol. 157: 1894–904 Kisselev A, Akopian T, Castillo V, Goldberg A. 1998. Certain proteoytic sites in mammalian 20S proteasomes allosterically regulate activities of other sites, suggesting an ordered mechanism for protein breakdown. Submitted for publication Yellen-Shaw AJ, Wherry EJ, Dubois GC, Eisenlohr LC. 1997. Point mutation flanking a CTL epitope ablates in vitro and in vivo recognition of a full-length viral protein. J. Immunol. 158:3227–34 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 temperaturesensitive cell lines. J. Immunol. 154: 511–19 Finley D, Ciechanover A, Varshavsky A. 1984. Thermolability of ubiquitinactivating enzyme from the mammalian cell cycle mutant ts85. Cell 37:43–55 Goldberg A, Rock K. 1992. Proteolysis, proteosomes and antigen presentation. Nature 357:375–79 Rechsteiner M, Hoffman L, Dubiel W. 1993. The multicatalytic and 26S proteases. J. Biol. Chem. 268:6065–68 Rivett A. 1993. Proteasomes: multicatalytic proteinase complexes. Biochem. J. 291:1–10 Coux O, Tanaka K, Goldberg ALl. 1996. Structure and functions of the 20S and 26S proteasomes. Annu. Rev. Biochem. 65:801–47 Deleted in proof Lowe J, Stock D, Jap B, Zwickl P, Baumeister W, Huber R. 1995. Crystal structure of the 20S proteasome from the

79.

80.

81.

82.

83.

84.

85.

86. 87.

88.

89.

˚ resoluarchaeon T. acidophilum at 3.4 A tion [see comments]. Science 268:533– 39 Groll M, Ditzel L, Lowe J, Stock D, Bochtler M, Bartunik HD, Huber R. 1997. Structure of 20S proteasome from ˚ resolution [see comments]. yeast at 2.4 A Nature 386:463–71 Seemuller E, Lupas A, Stock D, Lowe J, Huber R, Baumeister W. 1995. Proteasome from Thermoplasma acidophilum: a threonine protease. Science 268:579– 82 Arendt CS, Hochstrasser M. 1997. Identification of the yeast 20S proteasome catalytic centers and subunit interactions required for active-site formation. Proc. Natl. Acad. Sci. USA 94:7156–61 Seemuller E, Lupas A, Zuhl F, Zwickl P, Baumeister W. 1995. The proteasome from Thermoplasma acidophilum is neither a cysteine nor a serine protease. FEBS Lett. 359:173–78 Fenteany S, Standaert RF, Lane WJ, Choi S, Corey EJ, Schreiber SL. 1995. Inhibition of proteasome activities and subunit-specific amino-terminal threonine modification by lactacystin. Science 268:726–31 Bogyo M, McMaster J, Gaczynska M, Tortorella D, Goldberg A, Ploegh H. 1997. Covalent modification of the active site threonine of proteasomal beta subunits and the Escherichia coli homolog HslV by a new class of inhibitors. Proc. Natl. Acad. Sci. USA 4:6629–34 Akopian TN, Kisselev AF, Goldberg AL. 1997. Processive degradation of proteins and other catalytic properties of the proteasome from Thermoplasma acidophilum. J. Biol. Chem. 272:1791–98 Zwickl P, Kleinz J, Baumeister W. 1994. Critical elements in proteasome assembly. Nat. Struct. Biol. 1:765–70 Schmidtke G, Kraft R, Kostka S, Henklein P, Frommel C, Lowe J, Huber R, Kloetzel PM, Schmidt M. 1996. Analysis of mammalian 20S proteasome biogenesis: the maturation of beta-subunits is an ordered two-step mechanism involving autocatalysis. EMBO J. 15:6887–98 Schmidtke G, Schmidt M, Kloetzel PM. 1997. Maturation of mammalian 20S proteasome: purification and characterization of 13S and 16S proteasome precursor complexes. J. Mol. Biol. 268:95– 106 Nandi D, Woodward E, Ginsburg DB, Monaco JJ. 1997. Intermediates in the formation of mouse 20S proteasomes:

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

MHC CLASS I ANTIGEN PRESENTATION

90.

90a.

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

91.

92.

93.

94.

95.

96.

97.

implications for the assembly of precursor beta subunits. EMBO J. 16:5363–75 Beninga J, Goldberg A. 1998. Function of the proteasome in antigen presentation. In Uniquitin and the Biology of the Cell, ed. J Peters, J Harris, D Finley, pp. 191–221. New York: Plenum Orlowski M. 1990. The multicatalytic proteinase complex, a major extralysosomal proteolytic complex. Biochemistry 29:10289–97 Heinemeyer W, Gruhler A, Mohrle V, Mahe Y, Wolf DH. 1993. PRE2, highly homologous to the human major histocompatibility complex-linked RING10 gene, codes for a yeast proteasome subunit necessary for chrymotryptic activity and degradation of ubiquitinated proteins. J. Biol. Chem. 268:5115–20 Enenkel C, Lehmann H, Kipper J, Guckel R, Hilt W, Wolf DH. 1994. PRE3, highly homologous to the human major histocompatibility complexlinked LMP2 (RING12) gene, codes for a yeast proteasome subunit necessary for the peptidylglutamyl-peptide hydrolyzing activity. FEBS Lett. 341:193–96 Gaczynska M, Rock KL, Spies T, Goldberg A. 1994. Peptidase activities of proteasomes are differentially regulated by the MHC-encoded genes LMP2 and LMP7. Proc. Natl. Acad. Sci. USA 91: 9212–17 Gaczynska M, Goldberg A, Tanaka K, Hendil K, Rock K. 1996. Proteasome subunits X and Y alter peptidase activities in opposite ways to the interferongamma-induced subunits LMP2 and LMP7. J. Biol. Chem. 271:17275–80 Kisselev AF, Akopian TN, Goldberg AL. 1998. Range of sizes of peptide products generated during degradation of different proteins by archaeal proteasomes. J. Biol. Chem. 273:1982–89 Orlowski M, Cardozo C, Michaud C. 1993. Evidence for the presence of five distinct proteolytic components in the pituitary multicatalytic proteinase complex. Properties of two components cleaving bonds on the carboxyl side of branched chain and small neutral amino acids. Biochemistry 32:1563–72 McCormack T, Cruikshank A, Grenier L, Melandri F, Nunes S, Plamondon L, Stein R, Dick L. 1998. Kinetic studies of the branched chain amino acid preferring peptidase activity of the 20S proteasome: development of a continuous assay and inhibition by tripeptide aldehydes and clasto-lactacystin betalactone. Biochemistry 37:7792–800

773

98. Rohrwild M, Pfeifer G, Santarius U, Muller SA, Huang HC, Engel A, Baumeister W, Goldberg AL. 1997. The ATP-dependent HslVU protease from Escherichia coli is a four-ring structure resembling the proteasome. Nat. Struct. Biol. 4:133–39 99. Glickman MH, Rubin DM, Fried VA, Finley D. 1998. The regulatory particle of the Saccharomyces cerevisiae proteasome. Mol. Cell. Biol. 18:3149–62 100. DeMartino GN, Moomaw CR, Zagnitko OP, Proske RJ, Chu-Ping M, Afendis SJ, Swaffield JC, Slaughter CA. 1994. PA700, an ATP-dependent activator of the 20S proteasome, is an ATPase containing multiple members of a nucleotide-binding protein family. J. Biol. Chem. 269:20878–84 101. Glickman MH, Rubin DM, Coux O, Wefes I, Pfeifer G, Cjeka Z, Baumeister W, Fried VA, Finley D. 1998. A subcomplex of the proteasome regulatory particle required for ubiquitin-conjugate degradation and related to the COP9signalosome and eIF3. Cell 94:615–23 102. Pickart CM. 1997. Targeting of substrates to the 26S proteasome. FASEB J. 11:1055–66 103. Chu-Ping M, Slaughter CA, DeMartino GN. 1992. Purification and characterization of a protein inhibitor of the 20S proteasome (macropain). Biochim. Biophys. Acta 1119:303–11 104. Dubiel W, Pratt G, Ferrell K, Rechsteiner M. 1992. Purification of an 11S regulator of the multicatalytic protease. J. Biol. Chem. 267:22369–77 105. Ma CP, Slaughter CA, DeMartino GN. 1992. Identification, purification, and characterization of a protein activator (PA28) of the 20S proteasome (macropain). J. Biol. Chem. 267:10515– 23 106. Gray CW, Slaughter CA, DeMartino GN. 1994. PA28 activator protein forms regulatory caps on proteasome stacked rings. J. Mol. Biol. 236:7–15 107. Knowlton JR, Johnston SC, Whitby FG, Realini C, Zhang Z, Rechsteiner M, Hill CP. 1997. Structure of the proteasome activator REGalpha (PA28alpha). Nature 390:639–43 108. Song X, Mott JD, von Kampen J, Pramanik B, Tanaka K, Slaughter CA, DeMartino GN. 1996. A model for the quaternary structure of the proteasome activator PA28. J. Biol. Chem. 271:26410–17 109. Kuehn L, Dahlmann B. 1996. Reconstitution of proteasome activator PA28

P1: JER/MBG

P2: NBL/spd

February 11, 1999

774

110.

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

111.

111a.

112.

113.

114.

115.

116.

117. 117a.

118.

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG from isolated subunits: optimal activity is associated with an alpha,betaheteromultimer. FEBS Lett. 394:183– 86 Realini C, Dubiel W, Pratt G, Ferrell K, Rechsteiner M. 1994. Molecular cloning and expression of a gamma-interferoninducible activator of the multicatalytic protease. J. Biol. Chem. 269:20727– 32 Ahn JY, Tanahashi N, Akiyama K, Hisamatsu H, Noda C, Tanaka K, Chung CH, Shibmara N, Willy PJ, Mott JD, et al. 1995. Primary structures of two homologous subunits of PA28, a gammainterferon-inducible protein activator of the 20S proteasome. FEBS Lett. 366:37– 42 Tanahashi N, Yokota K-y, Ahn JY, Chung CH, Fujiwara T, Takahashi E-I, DeMartino GN, Slaughter CA, Toyonaga T, Yamamura K-I, Shimbara N, Tanaka K. 1997. Genes Cells 2:195–211 Realini C, Jensen C, Zhang Z-G, Jhonston S, Knowlton J, Hill C, Rechsteiner M. 1997. Characterization of recombinant REGalpha, REGbeta and REGgamma proteasome activators. J. Biol. Chem. 272:25483–92 Kania MA, Demartino GN, Baumeister W, Goldberg AL. 1996. The proteasome subunit, C2, contains an important site for binding of the PA28 (11S) activator. Eur. J. Biochem. 236:510–16 Groettrup M, Soza A, Eggers M, Kuehn L, Dick TP, Schild H, Rammensee HG, Koszinowski UH, Kloetzel PM. 1996. A role for the proteasome regulator PA28 alpha in antigen presentation. Nature 381:166–68 Hendil KB, Khan S, Tanaka K. 1998. Simultaneous binding of PA28 and PA700 activators to 20S proteasomes. Biochem. J. 332:749–54 Harding C, France J, Song R, Farah J, Chatterjee S, Iqbal M, Siman R. 1995. Novel dipeptide aldehydes are proteasome inhibitors and block the MHCI antigen-processing pathway. J. Immunol. 155:1767–75 Lee DH, Goldberg A. 1998. Proteasome inhibitors: valuable new tools for cell biologists. Trends Cell Biol. 8:397–403 Lee DH, Goldberg A. 1996. Selective inhibitors of the proteasome and vacuola, pathways in Saccharomyces cerevisiae. J. Biol. Chem. 271:287280–84 Dick LR, Cruikshank AA, Grenier L, Melandri FD, Nunes SL, Stein RL. 1996. Mechanistic studies on the inactivation of the proteasome by lactacystin: a

119.

120.

121.

122.

123.

124.

125.

126.

126a.

central role for clasto-lactacystin betalactone. J. Biol. Chem. 271:7273–76 Craiu A, Gaczynska M, Akopian T, Gramm C, Fenteany G, Goldberg A, Rock KL. 1997. Lactacystin and clastolactacystin beta-lactone modify multiple proteasome beta-subunits and inhibit intracellular protein degradation and major histocompatibility complex class I antigen presentation. J. Biol. Chem. 272:13437–45 Bush KT, Goldberg AL, Nigam SK. 1997. Proteasome inhibition leads to a heat-shock response, induction of endoplasmic reticulum chaperones, and thermotolerance. J. Biol. Chem. 272:9086– 92 Mathur M, Morimoto R. 1998. Heat shock response and protein degradation: regulation of HSF2 by the ubiquitinproteasome pathway. Mol. Cell. Biol. 18:5091–8 Meriin AB, Gabai VL, Yaglom J, Shifrin VI, Sherman MY. 1998. Proteasome inhibitors activate stress kinases and induce Hsp72. Diverse effects on apoptosis. J. Biol. Chem. 273:6373–79 Dick LR, Aldrich C, Jameson SC, Moomaw CR, Pramanik BC, Doyle CK, DeMartino GN, Bevan MJ, Forman JM, Slaughter CA. 1994. Proteolytic processing of ovalbumin and betagalactosidase by the proteasome to a yield antigenic peptides. J. Immunol. 152:3884–94 Boes B, Hengel H, Ruppert T, Multhaup G, Koszinowski U, Kloetzel P. 1993. Interferon gamma stimulation modulates the proteolytic activity and cleavage site preference of 20S mouse proteasomes. J. Exp. Med. 179:901–9 Niedermann G, Butz S, Ihlenfeldt H, Grimm R, Lucchiari M, Hoschutzky H, Jung G, Maier B, Eichmann K. 1995. Contribution of proteasome-mediated proteolysis to the hierarchy of epitopes presented by major histocompatibility complex class I molecules. Immunity 2:289–99 Cerundolo V, Benham A, Braud V, Mukherjee S, Gould K, Macino B, Neefjes J, Townsend A. 1997. The proteasomespecific inhibitor lactacystin blocks presentation of cytotoxic T lymphocyte epitopes in human and murine cells. Eur. J. Immunol. 27:336–41 Benham AM, Neejes JJ. 1997. Proteasome activity limits the assembly of MHC class I molecules after IFN-gamma stimulation. J. Immunol. 159:5896–904

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

MHC CLASS I ANTIGEN PRESENTATION 126b. Benham A, Gromme M, Neefjes J. 1998. Allelic differences in the relationship between proteasome activity and MHC class I peptide loading. J. Immunol. 161:83–89 127. Monaco JJ, McDevitt HO. 1984. H-2linked low-molecular weight polypeptide antigens assemble into an unusual macromolecular complex. Nature 309:797–99 128. Monaco J, McDevitt H. 1986. The LMP antigens: a stable MHC-controlled multisubunit protein complex. Hum. Immunol. 15:416–26 129. Parham P. 1990. Antigen processing. Transporters of delight [news; comment]. Nature 348:674–75 130. Brown MG, Driscoll J, Monaco JJ. 1991. Structural and serological similarity of MHC-linked LMP and proteasome (multicatalytic proteinase) complexes [see comments]. Nature 353:355–57 131. Ortiz-Navarrete V, Seelig A, Gernold M, Frentzel S, Kloetzel P, Hammerling G. 1991. Subunit of the ‘20S’ proteasome (multicatalytic proteinase) encoded by the major histocompatibility complex. Nature 353:662–64 132. Martinez CK, Monaco JJ. 1991. Homology of proteasome subunits to a major histocompatibility complex-linked LMP gene. Nature 353:664–67 133. Glynne R, Powis S, Beck S, Kelly A, Kerr L, Trowsdale J. 1991. A proteasome-related gene between the two ABC transporter loci in the class II region of the human MHC. Nature 353:357–60 134. Kelly A, Powis SH, Glynne R, Radley E, Beck S, Trowsdale J. 1991. Second proteasome-related gene in the human MHC class II region. Nature 353:667– 68 135. Tanaka K. 1998. Molecular biology of the proteasome. Biochem. Biophys. Res. Commun. 247:537–41 136. Hisamatsu H, Shimbara N, Saito Y, Kristensen P, Hendil KB, Fujiwara T, Takahashi E, Tanahashi N, Tamura T, Ichihara A, Tanaka K. 1996. Newly identified pair of proteasomal subunits regulated reciprocally by interferon gamma. J. Exp. Med. 183:1807–16 137. Yang Y, Waters JB, Fruh K, Peterson PA. 1992. Proteasomes are regulated by interferon gamma: implications for antigen processing. Proc. Natl. Acad. Sci. USA 89:4928–32 138. Aki M, Shimbara N, Takashina M, Akiyama K, Kagawa S, Tamura T, Tanahashi N, Yoshimura T, Tanaka K,

139.

140.

140a.

141.

142.

143.

144.

145.

146.

147.

775

Ichihara A. 1994. Interferon-gamma induces different subunit organizations and functional diversity of proteasomes. J. Biochem. 115(2):257–69 Akiyama K, Yokota K, Kagawa S, Shimbara N, Tamura T, Akioka H, Nothwang HG, Noda C, Tanaka K, Ichihara A. 1994. cDNA cloning and interferon gamma down-regulation of proteasomal subunits X and Y. Science 265:1231–34 Fruh K, Gossen M, Wang K, Bujard H, Peterson PA, Yang Y. 1994. Displacement of housekeeping proteasome subunits by MHC-encoded LMPs: a newly discovered mechanism for modulating the multicatalytic proteinase complex. EMBO J. 13:3236–44 Griffin TA, Nandi D, Cruz M, Fehling HJ, Kaer LV, Monoco JJ, Colbert RA. 1998. Immunoproteasome assembly: cooperative incorporation of interferon gamma (IFN-gamma)-inducible subunits. J. Exp. Med. 187:97–104 Arnold D, Driscoll J, Androlewicz M, Hughes E, Cresswell P, Spies T. 1992. Proteasome subunits encoded in the MHC are not generally required for the processing of peptides bound by MHC class I molecules. Nature 360:171–74 Momburg F, Ortiz-Navarrete V, Neefjes J, Goulmy E, van de Wa Y, Spits H, Powis S, Butcher G, Howard J, Walden P, et al. 1992. Proteasome subunits encoded by the major histocompatibility complex are not essential for antigen presentation. Nature 360:174–77 Yewdell J, Lapham C, Bacik I, Spies T, Bennink J. 1994. MHC-encoded proteasome subunits LMP2 and LMP7 are not required for efficient antigen presentation. J. Immunol. 152:1163–70 Fehling H, Swat W, Laplace C, Kuhn R, Rajewsky K, Muller U, von Boehmer H. 1994. MHC class I expression in mice lacking the proteasome subunit LMP-7. Science 265:1234–37 Van Kaer L, Ashton-Rickardt P, Ploegh H, Tonegawa S. 1992. TAP1 mutant mice are deficient in antigen presentation, surface class I molecules, and CD48+ T cells. Cell 71:1205–14 Sibille C, Gould KG, Willard-Gallo K, Thomson S, Rivett AJ, Powis S, Butcher GW, De Baetselier P. 1995. LMP2+ proteasomes are required for the presentation of specific antigens to cytotoxic T lymphocytes. Curr. Biol. 5:923–30 Cerundolo V, Kelly A, Elliott T, Trowsdale J, Townsend A. 1995. Genes encoded in the major histocompatibility complex affecting the generation of

P1: JER/MBG

P2: NBL/spd

February 11, 1999

776

148.

149.

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

150.

151.

152.

153.

154.

155.

156.

157.

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG peptides for TAP transport. Eur. J. Immunol. 25:554–62 Gaczynska M, Rock K, Goldberg A. 1993. Gamma-interferon and expression of MHC genes regulate peptide hydrolysis by proteasomes. Nature 365:264–67 Driscoll J, Brown M, Finley D, Monaco J. 1993. MHC-linked LMP gene products specifically alter peptidase activities of the proteasome. Nature 365:262–64 Van Kaer L, Ashton-Rickardt PG, Eichelberger M, Gaczynska M, Nagashima K, Rock KL, Goldberg AL, Doherty PC, Tonegawa S. 1994. Altered peptidase and antiviral activities in LMP2 mutant mice. Immunity 1:533–41 Hilt W, Enenkel C, Gruhler A, Singer T, Wolf DH. 1993. The PRE4 gene codes for a subunit of the yeast proteasome necessary for peptidylglutamyl-peptidehydrolyzing activity. Mutations link the proteasome to stress- and ubiquitindependent proteolysis. J. Biol. Chem. 268:3479–86 Ustrell V, Realini C, Pratt G, Rechsteiner M. 1995. Human lymphoblast and erythrocyte multicatalytic proteases: differential peptidase activities and responses to the 11S regulator. FEBS Lett. 376:155–58 Ustrell V, Pratt G, Rechsteiner M. 1995. Effects of interferon gamma and major histocompatibility complex-encoded subunits on peptidase activities of human multicatalytic proteases. Proc. Natl. Acad. Sci. USA 92:584–88 Kuckelkorn U, Frentzel S, Kraft R, Kostka S, Groettrup M, Kloetzel PM. 1995. Incorporation of major histocompatibility complex–encoded subunits LMP2 and LMP7 changes the quality of the 20S proteasome polypeptide processing products independent of interferon-gamma. Eur. J. Immunol. 25:2605–11 Stohwasser R, Kuckelkorn U, Kraft R, Kostka S, Kloetzel PM. 1996. 20S proteasome from LMP7 knock out mice reveals altered proteolytic activities and cleavage site preferences. FEBS Lett. 383:109–13 Cardozo C, Kohanski RA. 1998. Altered properties of the branched chain amino acid-preferring activity contribute to increased cleavages after branched chain residues by the “immunoproteasome”. J. Biol. Chem. 273:16764–70 Rammensee H, Falk K, Rotzschke O. 1993. Peptides naturally presented by MHC class I molecules. Annu. Rev. Immunol. 11:213–44

158. Stern LJ, Wiley DC. 1994. Antigenic peptide binding by class I and class II histocompatibility proteins. Behring Inst. Mitt.: 1–10 159. Stock D, Ditzel L, Baumeister W, Huber R, Lowe J. 1995. Catalytic mechanism of the 20S proteasome of thermoplasma acidophilum revealed by X-ray crystallography. Cold Spring Harbor Symp. Quant. Biol. 60:525–32 160. Wenzel T, Eckerskorn C, Lottspeich F, Baumeister W. 1994. Existence of a molecular ruler in proteasomes suggested by analysis of degradation products. FEBS Lett. 349:205–9 161. Kisselev AF, Akopian TN, Goldberg AL. 1998. Range of sizes of peptide products generated during degradation of different proteins by mammalian 20S and 26S proteasomes. J. Biol. Chem. In press 161a. Nussbaum AK, Dick TP, Keilholz W, Schirle M, Stevanovic S, Dietz K, Heinemeyer W, Groll M, Wolf DH, Huber R, Rammensee H-G, Schild H. 1998. Cleavage motifs of the yeast 20S proteasome β subunits deduced from digests of enolast 1. Proc. Natl. Acad. Sci. USA 95:12504–9 162. Eggers M, Boes-Fabian B, Ruppert T, Kloetzel PM, Koszinowski UH. 1995. The cleavage preference of the proteasome governs the yield of antigenic peptides. J. Exp. Med. 182:1865–70 163. Dick L, Moomaw C, DeMartino G, Slaughter C. 1991. Degradation of oxidized insulin B chain by the multiproteinase complex macropain (proteasome). Biochemistry 30:2725–34 164. Ehring B, Meyer TH, Eckerskorn C, Lottspeich F, Tampe R. 1996. Effects of major-histocompatibility-complexencoded subunits on the peptidase and proteolytic activities of human 20S proteasomes. Cleavage of proteins and antigenic peptides. Eur. J. Biochem. 235:404–15 165. Bogyo M, Shin S, McMaster JS, Ploegh HL. 1998. Substrate binding and sequence preference of the proteasome revealed by active-site-directed affinity probes. Chem. Biol. 5:307–20 166. Akopian T, Gilbert B, Rando R, Goldberg A. 1998. Potent peptide aldehyde inhibitors of the proteasome indicate an important influence on the P4-P5 positions on peptide binding. Submitted for publication 167. Niedermann G, Grimm R, Geier E, Maurer M, Realini C, Gartmann C, Soll J, Omura S, Rechsteiner MC, Baumeister

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

MHC CLASS I ANTIGEN PRESENTATION

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

168.

169.

170.

171.

172.

173.

174.

175.

176.

177.

W, Eichmann K. 1997. Potential immunocompetence of proteolytic fragments produced by proteasomes before evolution of the vertebrate immune system. J. Exp. Med. 186:209–20 Groettrup M, Ruppert T, Kuehn L, Seeger M, Standera S, Koszinowski U, Kloetzel PM. 1995. The interferongamma-inducible 11S regulator (PA28) and the LMP2/LMP7 subunits govern the peptide production by the 20S proteasome in vitro. J. Biol. Chem. 270:23808–15 Dick T, Ruppert T, Groettrup M, Kloetzel P, Kuehn L, Koszinowski U, Stevanovic S, Schild H, Rammensee H. 1996. Coordinated dual cleavages induced by the proteasome regulator PA28 lead to dominant MHC ligands. Cell 86:253–62 Shimbara N, Nakajima H, Tanahashi N, Ogawa K, Niwa S, Uenaka A, Nakayama E, Tanaka K. 1997. Double-cleavage production of the CTL epitope by proteasomes and PA28: role of the flanking region. Genes Cells 2:785–800 Craiu A, Akopian T, Goldberg A, Rock KL. 1997. Two distinct proteolytic processes in the generation of an MHC class I-presented peptide. Proc. Natl. Acad. Sci. USA 94:10850–55 Eisenlohr LC, Bacik I, Bennink JR, Bernstein K, Yewdell JW. 1992. Expression of a membrane protease enhances presentation of endogenous antigens to MHC class I-restricted T lymphocytes. Cell 71:963–72 Snyder H, Yewdell J, Bennink J. 1994. Trimming of antigenic peptides in an early secretory compartment. J. Exp. Med. 180:2389–94 Beninga J, Rock K, Goldberg A. 1998. Interferon-gamma can stimulate post-proteasomal trimming of the N-termini of antigenic peptides by inducing leucine aminopeptidase. J. Biol. Chem. 273:18734–42 Androlewicz M, Cresswell P. 1994. Human transporters associated with antigen processing possess a promiscuous peptide-binding site. Immunity 1:7–14 Momburg F, Roelse J, Hammerling G, Neefjes J. 1994. Peptide size selection by the major histocompatibility complexencoded peptide transporter. J. Exp. Med. 179:1613–23 Elliott T, Willis A, Cerundolo V, Townsend A. 1995. Processing of major histocompatibility class I-restricted antigens in the endoplasmic reticulum. J. Exp. Med. 181:1481–91

777

178. Luckey C, King G, Marto J, Venketeswaran S, Maier B, Crotzer V, Colella T, Shabanowitz J, Hunt D, Engelhard V. 1998. Proteasomes can either generate or destroy MHC class I epitopes: evidence for nonproteasomal epitope generation in the cytosol. J. Immunol. 161:112–21 179. Vinitsky A, Anton L, Snyder H, Orlowski M, Bennink J, Yewdell J. 1997. The generation of MHC class I-associated peptides is only partially inhibited by proteasome inhibitors. J. Immunol. 159:554–64 180. Yellen-Shaw A, Eisenlohr L. 1997. Regulation of class I-restricted epitope processing by local or distal flanking sequence. J. Immunol. 158:1727–33 181. Anton LC, Snyder HL, Bennink JR, Vinitsky A, Orlowski M, Porgador A, Yewdell JW. 1998. Dissociation of proteasomal degradation of biosynthesized viral proteins from generation of MHC class I-associated antigenic peptides. J. Immunol. 160:4859–68 182. Glas R, Bogyo M, McMaster JS, Gaczynska M, Ploegh HL. 1998. A proteolytic system that compensates for loss of proteasome function. Nature 392:618–22 183. Tamura T, Tamura N, Cejka Z, Hegerl R, Lottspeich F, Baumeister W. 1996. Tricorn protease—the core of a modular proteolytic system [see comments]. Science 274:1385–89 184. Osmulski P, Gaczynska M. 1998. A new large proteolytic complex distinct from the proteasome is present in the cytosol of fission yeast. Curr. Biol. 8:1023–26 185. Pfeifer JD, Wick MJ, Roberts RL, Fidlay K, Normark SJ, Harding CV. 1993. Phagocytic processing of bacterial antigens for class I MHC presentation to T cells. Nature 361:359–61 186. Rock KL. 1996. A new foreign policy: MHC class I molecules monitor the outside world. Immunol. Today. In press 187. Kovacsovics-Bankowski M, Rock KL. 1995. A phagosome-to-cytosol pathway for exogenous antigens presented on MHC class I molecules. Science 267:243–46 188. Ossendorp F, Eggers M, Neisig A, Ruppert T, Groettrup M, Sijts A, Mengede E, Kloetzel PM, Neefjes J, Koszinowski U, Melief C. 1996. A single residue exchange within a viral CTL epitope alters proteasome-mediated degradation resulting in lack of antigen presentation. Immunity 5:115–24 189. Falk K, Rotzschke O, Rammensee HG.

P1: JER/MBG

P2: NBL/spd

February 11, 1999

778

190.

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

191.

192.

193.

194.

195.

196.

197.

198.

199.

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

ROCK AND GOLDBERG 1990. Cellular peptide composition governed by major histocompatibility complex class I molecules. Nature 348:248– 51 Udono H, Srivastava PK. 1993. Heat shock protein 70-associated peptides elicit specific cancer immunity. J. Exp. Med. 178:1391–96 Lammert E, Arnold D, Nijenhuis M, Momburg F, Hammerling GJ, Brunner J, Stevanovic S, Rammensee HG, Schild H. 1997. The endoplasmic reticulumresident stress protein gp96 binds peptides translocated by TAP. Eur. J. Immunol. 27:923–27 Botbol V, Scornik OA. 1979. Intermediates in the degradation of abnormal globin. Bestatin permits the accumulation of the same peptides in cell-free extracts as in intact reticulocytes. J. Biol. Chem. 254:11254–57 Beninga J, Akopian T, Rock KL, Goldberg A. 1998. Many MHC class Ipresented antigenic peptides are digested in cell extracts by thimet oligopeptidase (EC 3.4.24.15), whose activity is suppressed by interferongamma. Submitted for publication Camargo AC, Gomes MD, Reichl AP, Ferro ES, Jacchieri S, Hirata IY, Juliano L. 1997. Structural features that make oligopeptides susceptible substrates for hydrolysis by recombinant thimet oligopeptidase. Biochem. J. 324:517–22 Barrett AJ, Brown MA, Dando PM, Knight CG, McKie N, Rawlings ND, Serizawa A. 1995. Thimet oligopeptidase and oligopeptidase M or neurolysin. Methods Enzymol. 248:529–56 Buchler M, Tisljar U, Wolf DH. 1994. Proteinase yscD (oligopeptidase yscD). Structure, function and relationship of the yeast enzyme with mammalian thimet oligopeptidase (metalloendopeptidase, EP 24.15). Eur. J. Biochem. 219:627–39 Villanueva MS, Fischer P, Feen K, Pamer EG. 1994. Efficiency of MHC class I antigen processing: a quantitative analysis. Immunity 1:479–89 Sijts AJ, Neisig A, Neefjes J, Pamer EG. 1996. Two Listeria monocytogenes CTL epitopes are processed from the same antigen with different efficiencies. J. Immunol. 156:683–92 Deng Y, Yewdell JW, Eisenlohr LC, Bennink JR. 1997. MHC affinity, peptide liberation, T cell repertoire, and immunodominance all contribute to the paucity of MHC class I-restricted pep-

200.

201.

202.

203.

204.

205.

206.

207.

208.

tides recognized by antiviral CTL. J. Immunol. 158:1507–15 Chimini G, Pala P, Sire J, Jordan BR, Maryanski JL. 1989. Recognition of oligonucleotide-encoded T cell epitopes introduced into a gene unrelated to the original antigen. J. Exp. Med. 169:297– 302 Hahn Y, Braciale V, Braciale T. 1991. Presentation of viral antigen to class I major histocompatibility complex-restricted cytotoxic T lymphocyte. Recognition of an immunodominant influenza hemagglutinin site by cytotoxic T lymphocyte is independent of the position of the site in the hemagglutinin translation product. J. Exp. Med. 174:733–36 Zaghouani H, Krystal M, Kuzu H, Moran T, Shah H, Kuzu Y, Schulman J, Bona C. 1992. Cells expressing an H chain Ig gene carrying a viral T cell epitope are lysed by specific cytolytic T cells. J. Immunol. 148:3604–9 Del Val M, Schlicht HJ, Ruppert T, Reddehase MJ, Koszinowski UH. 1991. Efficient processing of an antigenic sequence for presentation by MHC class I molecules depends on its neighboring residues in the protein. Cell 66:1145–53 Eisenlohr L, Yewdell J, Bennink J. 1992. Flanking sequences influence the presentation of an endogenously synthesized peptide to cytotoxic T lymphocytes. J. Exp. Med. 175:481–87 Shastri N, Serwold T, Gonzalez F. 1995. Presentation of endogenous peptide/MHC class I complexes is profoundly influenced by specific Cterminal flanking residues. J. Immunol. 155:4339–46 Bergmann C, Yao Q, Ho C, Buckwold S. 1996. Flanking residues alter antigenicity and immunogenicity of multi-unit CTL epitopes. J. Immunol. 157:3242–49 Theobald M, Ruppert T, Kuckelkorn U, Haubler A, Biggs J, Levine A, Huber C, Koszinowski U, Kloetzel P-M, Sherman L. 1998. The sequence alteration associated with a mutational hotspot in p53 protects cells from lysis by cytotoxic T lymphyocytes specific for a flanking peptide epitope. Submitted for publication Shimbara N, Ogawa K, Hidaka Y, Nakajima H, Yamasaki N, Niwa S, Tanahashi N, Tanaka K. 1998. Contribution of proline residue for efficient production of MHC class I ligands by proteasomes [In Process Citation]. J. Biol. Chem. 273:23062–71

P1: JER/MBG

P2: NBL/spd

February 11, 1999

12:59

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-23

MHC CLASS I ANTIGEN PRESENTATION

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

209. Levitskaya J, Coram M, Levitsky V, Imreh S, Steigerwald-Mullen PM, Klein G, Kurilla MG, Masucci MG. 1995. Inhibition of antigen processing by the internal repeat region of the Epstein-Barr virus nuclear antigen-1. Nature 375:685– 88 210. Levitskaya J, Sharipo A, Leonchiks A, Ciechanover A, Masucci MG. 1997. Inhibition of ubiquitin/proteasomedependent protein degradation by the

779

Gly-Ala repeat domain of the EpsteinBarr virus nuclear antigen 1. Proc. Natl. Acad. Sci. USA 94:12616–21 211. Rinat-Yehudar R, Groettrup M, Soza A, Kloetzel P, Ehrlich R. 1996. LMPassociated proteolytic activities and TAP-dependent peptide transport for class I molecules are suppressed in cell lines transformed by the highly oncogenic adenovirus 12. J. Exp. Med. 183:499–514

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:739-779. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:781–828 c 1999 by Annual Reviews. All rights reserved Copyright °

THE CENTRAL EFFECTORS OF CELL DEATH IN THE IMMUNE SYSTEM Jeffrey C. Rathmell and Craig B. Thompson Gwen Knapp Center for Lupus and Immunology Research, Department of Medicine, and Howard Hughes Medical Institute, University of Chicago, Chicago, Illinois 60637; e-mail: [email protected] KEY WORDS:

apoptosis, caspase, Bcl-2, Bcl-xL, Apaf-1, death domain, death effector domain, caspase recruitment domain, mitochondria, CD95, TNFR-I

ABSTRACT The immune system relies on cell death to maintain lymphoid homeostasis and avoid disease. Recent evidence has indicated that the caspase family of cysteine proteases is a central effector in apoptotic cell death and is absolutely responsible for many of the morphological features of apoptosis. Cell death, however, can occur through caspase-independent and caspase-dependent pathways. In the case of cells that are irreversibly neglected or damaged, death occurs even in the absence of caspase activity. In contrast, healthy cells require caspase activation to undergo cell death induced by surface receptors. This review summarizes the current understanding of these two pathways of cell death in the immune system.

INTRODUCTION Apoptotic cell death plays a central role in shaping the repertoire of circulating mature lymphocytes. Lymphocytes that fail to properly rearrange antigen receptors or that generate antigen receptors that are not positively selected undergo apoptosis early in their development. Lymphocytes that have or gain an autoreactive specificity in the process of V(D)J rearrangement are eliminated as a consequence of negative selection. After emigration from the central lymphoid organs, immature lymphocytes that have successfully rearranged antigen 781 0732-0582/99/0410-0781$08.00

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

782

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

receptors must survive an additional selection process to enter the recirculating pool of mature lymphocytes. Lymphocytes that are specific for peripheral selfantigen also can undergo cell death. During a peripheral immune response, excess lymphocytes must be eliminated after antigen clearance. Furthermore, if lymphocytes suffer irreversible DNA damage or are neglected by removal from necessary survival factors at any time in their lifespan, they are eliminated. Thus, there are many instances in their lifespan when lymphocytes may undergo programmed cell death. The combined effect of these numerous cell death checkpoints is the establishment and maintenance of lymphoid homeostasis. Homeostasis, however, can be maintained only where there is proper regulation of cell death. If cell death occurs at too low a frequency, autoimmunity or cancer can ensue, whereas leukopenia and immunodeficiency may result from excessive cell death (1). Cells undergoing programmed cell death were first visualized in humans in 1890, when William Councilman described vacuolated acidophilic bodies in liver tissue from yellow fever patients (2). Despite observations of similar acidophilic bodies in other types of liver damage, it was unclear for an additional 70 years what Councilman bodies represented. Not until electron microscopy was used to analyze acidophilic bodies was it shown that such bodies were, in fact, dying and dead cells (3–5). A study contemporary with the early electronmicroscopic works on acidophilic bodies showed that these cell death morphologies were not unique to liver damage but also occurred in the normal developing mouse during the fusion of the palate (6). Subsequent to these analyses, Kerr and colleagues provided a thorough morphological description of this form of cell death and proposed that it be called apoptosis (7). In the past several years it has become clear that the morphology of cell death is determined by the activity of a group of cysteine proteases known as caspases. When activated, caspases cleave a variety of proteins to elicit the apoptotic morphology. Cells that have been metabolically or genetically damaged, however, can undergo cell death regardless of caspase activity. Caspases may be activated in these types of cell deaths, but they determine only the morphology and mode of death, not its occurrence. On the other hand, healthy cells can undergo cell death induced by surface receptors only if caspases are activated. In the absence of caspases, these deaths do not occur. Thus, cell death can be divided into two general categories: one that is caspase independent and another that is caspase dependent.

APOPTOSIS AND NECROSIS Cell death can occur through apoptotic or necrotic death pathways. Apoptosis is an orderly process that proceeds through several morphological phases (7, 8). Dying cells release contacts with neighboring cells and become detached from

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

783

the surrounding tissue. There is a marked condensation of both the nucleus and the cytoplasm, causing the cells to shrink significantly in size. Mitochondria release cytochrome c into the cytoplasm (9, 10) and undergo a loss of membrane potential, 19 m, and mitochondrial permeability transition (PT), which might be caused by the opening of specialized PT pores (11–14). The nuclear envelope and the nucleolus break apart as chromatin condenses and is cleaved intranucleosomally into regular fragments with unit lengths of ∼180 base pairs, the same length as fragments produced by nuclease treatment of intact chromatin, within hours of the induction of cell death (15–17). The plasma membrane also begins to bleb, or form convoluted invaginations and protrusions. The blebs in the plasma membrane can divide the cell into smaller apoptotic bodies that can contain condensed or morphologically normal organelles. Dying cells also signal their apoptotic state to their surroundings to aid in recognition of the apoptotic bodies by phagocytes. Phosphatidylserine molecules flip from a cytoplasmic to an extracellular orientation after the mitochondrial PT (18, 19), owing to the decreased function of ATP-dependent amino-phospholipid translocase and the calcium-dependent activity of a nonspecific lipid scramblase (20, 21). Additionally, the proinflammatory cytokines interleukin (IL)-1 α and β (IL-1α/β), as well as IL-18, normally cytosolic in unprocessed forms, can be processed and released during apoptosis (22). Finally, apoptotic bodies are rapidly phagocytosed and degraded in lysosomes. Although the apoptotic process has long been considered immunologically silent, it is now known that peptides from phagocytosed apoptotic bodies can be presented by dendritic cells (23, 24) to elicit further immunological activity. The progress through these morphological changes is quite rapid, occurring in ≤24 h. Importantly, throughout the entire process the plasma membranes on the dying cell and the individual apoptotic bodies remain intact. There is no spillage of cellular contents as cells condense and fragment into apoptotic bodies. In contrast to apoptosis, necrosis is a disorderly mode of cell death (8). Rather than condensing and shrinking as apoptotic cells do, necrotic cells take up water and swell. This swelling causes the plasma membrane to burst and release the cytoplasmic contents into the cell’s surroundings. Organelles also swell rather than condense. Nuclei do not condense, nor is chromatin processed in the same way as in apoptotic cells. DNA degradation can still occur, but it occurs later and leaves a more continuous sizing of fragments than the DNA laddering induced in apoptosis (8, 25, 26). Furthermore, the DNA degradation in necrosis is induced by a different mechanism, relying on serine proteases rather than cysteine proteases to induce endonuclease activity (26), and the ends of DNA fragments do not contain single-base 30 overhangs, which can occur in apoptosis (27). The processing and release of IL-1 is also different in necrotic cells (22). Like apoptotic cell death, necrotic cell death can lead to the release of IL-1α/β, but only IL-1α is processed.

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

784

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

The observation that apoptosis is an orderly process with particular biochemical characteristics, whereas necrosis is a disorderly process in which dying cells appear to simply fall apart, suggested that apoptosis is energy dependent. Indeed, depletion of ATP in Jurkat and mouse kidney proximal tubular cells has shown that ATP must be present for death to occur by apoptosis (28, 29). If ATP is absent or at sufficiently low concentrations, cells die by necrosis. In addition, specific events in apoptosis may be more energy dependent than others. Experiments with isolated rat thymocyte nuclei incubated in the presence of lysates from apoptotic Jurkat T cells with or without the addition of ATP or ATP analogs, for example, showed that, although the addition of ATP was required for nuclear condensation, it was not required for DNA fragmentation (30).

Immunological Consequences of Apoptosis and Necrosis Apoptosis and necrosis have distinct repercussions in the immune system. Apoptosis, owing to the maintenance of plasma membrane integrity, restricts the dispersion of cytoplasmic contents. In this way, apoptosis limits the spread of intracellular pathogens. The sequestration of cytoplasmic contents into discrete apoptotic bodies also limits inflammation. Inflammation is not necessarily absent in apoptosis because apoptotic cells can release the proinflammatory cytokines IL-1 and IL-18 into their environment. It is, however, a controlled and regulated form of inflammation. Furthermore, the ability of phagocytes to present peptides from engulfed apoptotic cells (23, 24) may allow the activation of antigen-specific T cells to enhance specific immune responses or regulatory T cells to cause immune deviation or suppression. Indeed, apoptosis of inflammatory cells was found to be required for the induction of tolerance to herpes simplex virus type 1 after virus injection into the eye (31). Necrosis can have far different immunological effects due to the loss of plasma membrane integrity. Infected cells release intracellular pathogens into their environment, thus spreading infection. Furthermore, the release of cytoplasmic contents can cause a less controlled inflammation, which could result in additional tissue damage, a less specific immune response, and activation of autoreactive lymphocytes. It is important that, within a given damaged tissue, both apoptosis and necrosis can occur. The type of death may depend on the cell’s available energy stores as well as the rapidity and extent of damage. In these cases, the balance of apoptosis to necrosis may influence the extent and duration of the immune response.

MOLECULAR REQUIREMENTS FOR APOPTOSIS In a remarkable example of how the study of simple organisms can provide insight into biological processes of mammalian cells, the genetic analysis of the nematode Caenorhabditis elegans has revealed a number of key molecular

P1: JER/mbg

P2: ARS/ary

January 22, 1999

QC: ARS

10:9

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

785

components for initiation and regulation of apoptosis which appear to be conserved throughout evolution. C. elegans generates 1090 somatic cells in the course of its development, and, of these, 131 are destined to undergo programmed cell death. The observation of C. elegans mutants with cell death defects has enabled the identification of a genetic pathway consisting of 11 genes (32). Of these 11 genes, 3 were identified as key regulators of apoptosis in all somatic cells, ced-3, ced-4, and ced-9. ced-3 and ced-4 are both required to initiate cell death because loss-of-function mutations in either of these genes block all 131 somatic cell deaths. ced-9 has an antagonistic function to ced-3 and ced-4 and, instead of promoting death, promotes survival. Partial-lossof-function alleles of ced-9 are lethal owing to excessive cell death, whereas gain-of-function alleles of ced-9 lead to an excess of cells. Although genetic analysis had ordered the pathway with ced-9 upstream of ced-4 and ced-3 (32), little was understood about the function of these molecules until their molecular cloning. Cloning revealed that ced-3 encoded a cysteine protease homologous to IL-1β–converting enzyme (ICE; now caspase 1) (33), a previously described enzyme that exists in a pro form and requires processing to become active (34, 35). ced-4 encoded a protein with no apparent homologies (36). ced-9 was structurally and functionally homologous to the antiapoptotic mammalian gene bcl-2 (37, 38). Further insight came from recent biochemical experiments that showed that CED-4 can physically interact with CED-3 and stimulate its processing to the active proteolytic form (39–42). CED-4 also binds CED-9, which then inhibits its ability to promote CED-3 processing (39–41, 43, 44). These findings illustrate three critical classes of molecules involved in the regulation of apoptosis both in worms and in mammals. One class is represented by CED-3 and the caspase family of cysteine proteases, which are necessary to elicit the downstream events of apoptosis. The remaining two classes function to regulate CED-3 activity. CED-4 and its newly identified human homolog, Apaf-1 (45), act as adaptor molecules and facilitate the activation of CED-3– related cysteine proteases. CED-9 and Bcl-2 family members act to inhibit cell death by acting upstream in the apoptotic pathway to inhibit the activation of CED-3–related cysteine proteases.

CASPASES As the C. elegans system demonstrates, cysteine proteases play an important role in apoptosis. The first such mammalian protease, ICE (now caspase 1), was identified as a regulatory molecule for the proinflammatory cytokine IL1β (34, 35). The importance of ICE as a potential cell death effector molecule, however, was not appreciated until ced-3 was cloned and identified as a homolog

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

786

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

to ICE (33). Subsequent studies demonstrated that ICE itself could cause apoptotic cell death when overexpressed in fibroblasts (46). The homology between ICE and CED-3 was highest in the C-terminal regions, which are processed to form the p10 and p20 subunits, with the active-site residues QACRG conserved between three species of nematode CED-3 and mouse and human ICE (33). An additional gene that had been previously cloned among a group of developmentally downregulated genes in the mouse brain, Nedd-2, was also found to share homology with CED-3 (33, 47, 48). Like ICE, Nedd-2 could cause the induction of apoptosis when overexpressed in fibroblasts (48). A number of additional mammalian ICE-related proteins have been identified and the family members have been renamed caspase-1 through -13. The term caspase refers to the cysteine protease mechanism of family members and their aspartic acid specificity (49). An updated list of caspases and their alternative names is shown in Table 1. A defining feature of these family members is their specificity for cleavage after aspartic acid residues. This specificity is quite unusual. Only the serine protease granzyme B and its family members have similar cleavage specificity. The basic primary structure of each caspase consists of three domains (Figure 1): an N-terminal prodomain, a domain that

Figure 1 Caspase domains. Caspases contain three primary domains that are processed in a twostep mechanism to produce an active enzyme.

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

APOPTOSIS IN THE IMMUNE SYSTEM

787

Table 1 Caspases Caspase number 1 2

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

3

Other names ICE ICH-1 Nedd-2 CPP32

6

Yama Apopain TX ICH-2 ICErel-II ICErel-III TY Mch2

7

Mch3

4

5

10

ICE-LAP3 CMH-1 MACH FLICE Mch5 ICE-LAP6 Mch6 Mch4

11

FLICE-2 ICH-3

12 13

ERICE

8

9

Species Human, mouse Human, mouse Human, mouse

Human

Human Human, mouse Human, mouse

Human, mouse

Human Human

Mouse, probable homolog to caspase 4 Mouse Human

Reference 34, 35 81 47, 48 320 117 118 132 321 322 322 323 324 95 325 326 327 95 53 54 55 328 133 55 97 82 95 95 96

becomes the large subunit (17–21 kDa), and a domain that becomes the small enzyme subunit (10–14 kDa). Some have a brief spacer between the large and small subunits. On processing, the small protease subunit domain is first separated by proteolytic cleavage from the prodomain/large protease subunit domain polypeptide, which is then cleaved to release the large protease subunit domain and to allow the large and small subunits to heterodimerize and form an active enzyme complex consisting of homodimers of large-/small-enzyme subunit heterodimers (50–52). The homologies between caspases are retained primarily in the large and small enzyme domains, with all sharing the invariant QACxG catalytic motif in the large subunit. Caspase prodomains are divergent,

P1: JER/mbg

P2: ARS/ary

January 22, 1999

788

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

and they vary widely in both their length and sequence, with caspases 3, 6, and 7 having short prodomains and caspases 1, 2, 4, 5, 8, 9, 10, 11, 12, and 13 having long prodomains. Caspase prodomains appear to promote proteinprotein interactions. The prodomain of caspase 8, for example, contains two death effector domains (DEDs) that can bind to the adaptor molecule FADD (53–55).

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Caspase-Independent and Caspase-Dependent Cell Death Pathways The use of potent and specific peptide inhibitors of caspase activity has shown that cells can die by caspase-independent mechanisms as well as by caspasedependent mechanisms. Owing to their inability to maintain cellular metabolism, cells that have suffered irreversible damage will die regardless of caspase activity. Although caspases are often activated in this type of cell death, they function only to allow death to proceed by apoptosis. If caspase activity is blocked, the cell death resembles necrosis rather than apoptosis. Overexpression of the pro-apoptotic Bcl-2 family member Bax (56, 57), for example, has been found to lead to caspase activation and cell death (58). The inhibitor benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone (zVAD-fmk) was sufficient to block Bax-induced caspase activity but was nevertheless unable to block cell death (58). Cell death was unusual, however, in that, although mitochondria lost their membrane potential, DNA was not degraded and the cell membranes became permeable. The latter two phenotypes are reminiscent of necrotic rather than apoptotic cell death. A number of other death stimuli that normally induce apoptosis, including treatment of cells with drugs that induce mitochondrial PT, such as protoporphyrin IX or mClCCP (59), inhibition of ubiquitination (60), serum deprivation (61), treatment of cells with staurosporine (61), overexpression of oncogenes such as myc and E1A (62), DNA damage (62), overexpression of Bak (62), or IL-3 withdrawal (13; M Vander, C Heiden, CB Thompson, unpublished data), are also able to induce a form of cell death that could not be blocked by caspase inhibitors. In each of these cases, cells died through a process most similar to necrosis, without nuclear changes indicative of apoptosis (59–62). Despite the necrosislike character of these cell deaths, plasma membranes were found to undergo blebbing, a process normally reserved for apoptotic cells. In this case, however, blebbing was excessively longlived and vigorous and was followed by plasma membrane permeabilization (61, 62). Although blebbing was not blocked by the caspase inhibitor zVADfmk, it was significantly diminished in the presence of myosin light-chain kinase (MLCK) inhibitors, the Rho inhibitor Clostridium botulinum C3 exoenzyme, or drugs that disrupt the actin cytoskeleton (61). These data show that blebbing is dependent on MLCK and Rho, but it is unclear whether this pathway is initiated

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

789

in a non–caspase-dependent manner or whether a caspase family member can retain sufficient activity in the presence of zVAD-fmk to activate Rho and MLCK. Unlike irreversibly damaged cells, which die even in the absence of caspase activation, caspase-dependent cell death occurs when otherwise healthy cells are instructed to undergo programmed cell death. The best examples of caspasedependent cell deaths are those initiated by the death domain (DD)-containing receptors of the tumor necrosis factor receptor (TNFR) family (63–66). The DDs of these receptors bind to adaptor proteins that directly link cell surface receptors to intracellular caspases. Ligation of receptors is sufficient to activate caspases and initiate cell death. In these cases, caspase activation is required for cell death, because cell death does not occur when caspase activity is blocked (67–69).

Caspase Knockouts To date, caspases 1, 2, 3, 8, 9, and 11 have been the subjects of gene-targeting experiments. Of these knockout mice, deficiencies in caspases 3, 8, and 9 provide the most striking demonstration of the requirement for caspases in apoptosis. Mice lacking caspase 3 had severe neurological defects owing to failed apoptosis that resulted in fatality by 5 weeks of age (70, 71). Caspase 3-null mouse brains were characterized by multiple supplemental indentations and ectopic cell masses. These cell masses were located between the cerebral cortex, the hippocampus, and the striatum and were not tumors, because they lacked actively cycling cells (70). Furthermore, pyknotic clusters of dead and dying cells normally at the sites of morphological change, such as the interventricular junction, were not present in caspase 3-deficient mice (70). Despite these severe neurological apoptosis defects, the immune systems of caspase 3-deficient mice were largely intact. Developing caspase 3-null T and B cells appeared unaffected, with the only noticeable phenotype being a small reduction in thymocyte number. Caspase 3-null thymocytes, themselves, appeared to differentiate normally and were as susceptible to apoptosis induced by anti-CD95 antibody, dexamethasone, C2-ceramide, staurosporine, and γ irradiation as were wild-type thymocytes (70, 71). The most significant immune defects were found in peripheral T cells and developing neutrophils. Mature caspase 3-null T cells were less susceptible to activation-induced cell death, antiCD3ε–induced cell death, and anti-CD95–induced death than were wild-type T cells (71), and neutrophils from caspase 3-deficient mice failed to undergo apoptosis when treated with cycloheximide (71). Most interestingly, caspase 3-deficient cells showed an aberrant apoptotic morphology (71). Caspase 3-null cells undergoing apoptosis did not demonstrate the nuclear changes typical in apoptosis of wild-type cells. Chromosomal

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

790

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

condensation did not occur, and DNA remained intact (71). When viewed by electron microscope, these dying cells were highly vacuolated and had swollen mitochondria, both characteristic of necrosis (71). The successful redistribution of phosphatidylserine on the plasma membrane (71), however, indicated that caspase 3-deficient cells were dying of an altered form of apoptosis, rather than necrosis. These data strongly implicate caspase 3 in the pathways that control nuclear condensation, DNA fragmentation, and mitochondrial condensation in apoptosis. Consistent with the proposed requirement for caspase 3 in mediating the nuclear changes that occur in apoptosis are the lack of nuclear condensation and DNA fragmentation when apoptosis was induced in the presence of selective caspase 3 inhibitors (72). The subcellular localization of pro-caspase 3 is also consistent with its proposed role in mitochondria, because unprocessed caspase 3 is found both in the cytosol and in the mitochondrial intermembrane space in resting cells (73). Mice were also severely affected by a deficiency of caspase 9 (74, 75). These mice suffered from early lethality and had severe perturbations of the cerebral cortex and forebrain that were reminiscent of caspase 3 deficiency. The lethality of caspase 9 deficiency, however, occurred much earlier than with that of caspase 3-deficient mice. Whereas caspase 3-deficient mice died at ∼5 weeks of age, <8% of all live births from caspase 9 +/− matings were homozygous for the disrupted allele. Furthermore, those caspase 9-deficient mice that did survive to birth typically died within 3 days. In embryonic stem cells, mouse embryonic fibroblasts, and lymphocytes from rag-2 −/− blastocyst chimeras, caspase 9 deficiency was found to block activation of caspases 2, 3, 7, and 8 and cell death by γ -irradiation and chemotherapeutic drugs. In contrast, cell death initiated by anti-CD95 or anti-CD3ε occurred normally in the absence of caspase 9. Caspase 9 was also not required in target cells for lysis by cytotoxic T cells. Caspase 8-deficient mice suffered from hyperemia, or accumulation of blood, throughout the animal, notably in the abdominal area, and defective heart development (76). Mice deficient in caspase 8 also suffered from early lethality. In this case, homozygosity for the disrupted allele caused embryonic lethality at day 12.5. Importantly, TNFRs, CD95, and DR3 were unable to induce cell death, indicating that despite the association of these receptors with caspases 2 and 10, they depend on caspase 8 for death induction in vivo. Unlike deficiencies of caspases 3, 8, or 9, deficiencies in caspases 2, 1, or 11 did not lead to early lethality (77–80). Caspase 2 was initially identified based on high expression in embryonic brain and down regulation in adult brain (47). It was therefore expected that, much like the caspase 3- and 9-null mice, mice lacking caspase 2 would suffer severe neurological defects owing to insufficient apoptosis of developing neurons. Surprisingly, caspase 2 deficiency did not result in significant brain developmental abnormalities (79).

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

791

Excess doxyrubicin-resistant germ cells were, however, present in adult females (79). Interestingly, caspase 2-deficient mice displayed several phenotypes of enhanced rather than reduced cell death, such as reduced numbers of facial motor neurons and decreased survival of sympathetic neurons deprived of nerve growth factor (79). This apparent discrepancy may be caused by the existence of two alternatively spliced forms of caspase 2, a short form (caspase 2S) that can protect from cell death and a long form (caspase 2L) that promotes death (81). Indeed, caspase 2L was found to be the predominant form of caspase 2 expressed in female germ cells, whereas both caspases 2S and 2L were expressed in brain (79). The elimination of both forms of caspase 2 in the knockout, therefore, affected each tissue differently, depending on its normal balance of caspase 2L to 2S. In addition to demonstrating in vivo the role of caspase 2 in apoptosis, these data also show that caspases can act in vivo to prevent cell death and point to another mechanism by which caspases might control apoptosis. Despite the in vitro induction of apoptosis in fibroblasts by transfection of caspase 1 or caspase 11 (46, 82), mice deficient in either of these two caspases had very minor defects in apoptosis (77, 78, 80). Developmentally, these mice were normal, and they did not suffer early lethality. Caspase 1-deficient thymocytes were resistant to anti-CD95 antibody but underwent normal apoptosis in response to dexamethasone and ionizing irradiation (77, 78). Despite the limited extent of their apoptotic defects, both caspase 1- and caspase 11-deficient mice had significant defects in cytokine processing in response to lipopolysaccharide, because they were unable to produce mature IL-1β (77, 78, 80). The export of mature IL-1α, TNF-α, IL-6, and IL-18 [interferon-γ –inducing factor (IGIF)] was also found to be reduced in caspase 1-deficient mice (77, 78, 83, 84). Most likely as a consequence of the failure to process IL-1β, caspase 1and caspase 11-deficient mice were resistant to endotoxic shock induced by injection of lipopolysaccharide (78). The similarities of these phenotypes suggest that they act through a common pathway. Indeed, caspase 1 activation is dependent on caspase 11 activity (80). It appears that caspase 11 acts solely as an activator of caspase 1, whose activity is then necessary for the proper processing of a number of cytokines. Because some caspases, such as caspases 3, 8, and 9, are required for apoptosis to proceed, other caspases, therefore, may be required not to mediate apoptosis itself but rather to stimulate the production of inflammatory cytokines that might accompany it. It is interesting that the caspase knockouts were found to have such minimal immune defects. In caspase 3 and 9 knockouts, this may be caused by their early lethality and the subtlety of any lymphocyte apoptosis defects relative to the severity of neurological defects. Caspase function in lymphocytes may also be redundant. Further analysis of described caspase knockouts, generation of additional caspase knockouts, interbreeding to create double caspase knockouts,

P1: JER/mbg

P2: ARS/ary

January 22, 1999

792

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

and the use of inhibitors may begin to reveal cell death defects in the lymphoid lineages. Alternatively, the critical role for caspases in the developing brain may reflect a need to eliminate healthy neurons, whereas, in the immune system, much of the death may be of cells that are damaged or neglected and cannot be rescued. In these cases, lymphocytes would die regardless of caspase activation but perhaps by necrotic or aberrant apoptotic morphology. In this context, it may be interesting to analyze caspase knockouts for occurrence of necrosislike lymphocyte deaths.

Caspase Activation Depending on the caspase, two or three protealytic cleavages must occur for caspases to become activated. For example, upon activation initiated by cytochrome c (85), CD95 (86), or B-cell antigen receptor (86), caspase 3 is cleaved in two steps. The first cleavage occurs between the large and small enzyme subunits followed by the separation of the prodomain from the large subunit. Importantly, the two cleavages are blocked by different peptide inhibitors (85). These findings suggest that different caspases may account for each cleavage, with a caspase 8-, 9-, or 10-like activity resulting in the initial cut and autoproteolysis responsible for the second cut. Caspases 1 and 3 have been crystallized, and their structures demonstrate two additional factors that should be considered in caspase activation (50–52). First, caspase 1 and caspase 3 crystals were found to be tetramers, with homodimers of large and small enzyme subunit heterodimers (50–52). Such tetramers may be the result of dimerization of pro-caspase molecules, which has been reported to be sufficient for caspase processing (87–91). Second, each individual heterodimer of large and small subunits was found to be arranged such that the distance from the C terminus of the large subunit to the N terminus of the small subunit was significantly greater than the distance between these domains in the unprocessed forms. The caspase 1 large-/small-subunit heterodimer held the large-subunit C terminus and small-subunit N terminus 6.5 nm apart (50), whereas the corresponding distance for caspase 3 was 4.8 nm (52). The generation of such large separations of heterodimer subunits would require either a significant conformational change on processing or that the heterodimer subunits originate from different pro-caspase molecules. In the latter case, two pro-caspase molecules could dimerize and, on processing, swap subunits. Noncaspase molecules that stimulate caspase processing, such as CED-4 and Apaf-1, may do so by stabilizing the formation of pro-caspase dimers and the conformational transition of the large and small subunits. CED-4 (39–41) and Apaf-1 (92) normally require ATP and, thus, provide a potential source of transition energy for caspase conformational change and dimerization. The lack of requirement for ATP by mutant Apaf-1, which lacks a C-terminal domain,

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

APOPTOSIS IN THE IMMUNE SYSTEM

793

however, suggests that ATP is required for Apaf-1 rather than caspase activation (91).

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Caspase Substrates Substrate specificities and phylogenetic relationships allow the division of caspases into three groups (93). Group I, the ICE-like caspases, include caspases 1, 4, and 5. These enzymes prefer substrates with bulky hydrophobic residues at P4, the fourth amino acid N terminal to the cleavage site, and have a consensus site of (YWL)EHD (93, 94). Group II enzymes, the CPP32-like caspases, have a strong preference for Asp at P4 and have a consensus site of DExD (93, 94). Group II includes caspases 2, 3, and 7, as well as CED-3 (93). Group III enzymes prefer branched-chain aliphatic amino acids at P4, have a consensus site of (IVL)(QE)xD (93, 94), and include caspases 6, 8, and 9. Although their substrate specificities have been less well characterized, the phylogenetic relationships of other caspases allow their tentative assignments to these groups. Caspases 11, 12, and 13 are similar to group I caspases (82, 95, 96), and caspase 10 is similar to the group III caspases (55, 97). The activation of caspases in apoptosis does not lead to indiscriminate proteolytic degradation of the dying cell (98, 99). Rather, specific subsets of proteins are cleaved to produce the apoptotic phenotype. These substrates include proteins involved in cell structure, signaling, cell cycle control, and DNA repair among others. Substrate cleavage by caspases does not always lead to protein degradation, because a number of substrates, particularly those involved in signaling, become constitutively activated on proteolytic separation from regulatory domains. A current list of known caspase substrates is provided in Table 2. It is important to note that a large number of caspase substrates are thought to be cleaved by caspase 3. Whereas this is certainly true for some substrates, other caspases with similar specificities may play these roles in vivo. Caspase 3-deficient MCF-7 breast carcinoma cells, for example, fail to cleave only spectrin on apoptosis, in an analysis of the putative caspase 3 substrates spectrin, PARP, Rb, PAK2, DNA-PK, gelsolin, and ICAD/DFF-45 (100). Each group of substrates appears to play a particular role in the apoptotic process. Proteolysis of structural proteins leads to the breakdown of the nuclear matrix and disassembly of cytoskeletal contacts with the plasma membrane. Cleaved signaling proteins act to turn on death-promoting pathways and turn off survival pathways (99). Death-promoting pathways include the activation of SAPK (101) by constitutive activation of the SAPK activator MEKK1 by caspase 3 (102, 103). Survival pathways affected include the inhibition of ERK activity (101) by the inactivation of Raf-1 (99) and activation of protein phosphatase 2a (PP2a) (104). Consistent with the need to turn off ERK in apoptosis is the finding that apoptosis induced by CD95 is inhibited by constitutive ERK

P1: JER/mbg

P2: ARS/ary

January 22, 1999

QC: ARS

10:9

794

Annual Reviews

AR078-24

RATHMELL & THOMPSON

Table 2 Caspase substrates

Category

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Structural

Substrate

Site (caspase)

Nuclear lamin A

VEID (6)

Nuclear lamin B

VEID

Nuclear lamin C

VEID

Gas2

SRVD

α-Spectrin

DETD, DSLD (3)

Focal adhesion kinase Gelsolin

DQTD (7) VSWD (6) DQTD (3)

Keratin 18

VEVD (6, 3, 7)

Actin β-Catenin

ELPD (3) LVVD (1) NYQDD, LLNDED (3)

NuMA

3, 4, 6, 7

Functional result Degraded

Proposed effect in apoptosis

Disassembly of nuclear matrix, required for chromosome condensation and nuclear shinkage Degraded Disassembly of nuclear matrix, required for chromosome condensation and nuclear shinkage Disassembly of nuclear matrix Activated Disassembly of actin filaments Degraded Disassemble cytoskeletonmembrane contacts Inactivated Disassembly of focal adhesions Activated Disassembly of actin filaments, required for rapid blebbing and DNA cleavage Degraded Disassembly of intermediate filaments Degraded Disassembly of actin filaments Degraded Disassemble cytoskeleton at cell-cell contacts Degraded Disassemble chromatin/ nuclear matrix contacts

Reference 329–331

98, 331

331 332 333, 334

335, 336 337

338

339, 340 341

72, 98

(Continued )

P1: JER/mbg

P2: ARS/ary

January 22, 1999

QC: ARS

10:9

Annual Reviews

AR078-24

795

APOPTOSIS IN THE IMMUNE SYSTEM Table 2 (Continued )

Category

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Signaling

Site (caspase)

Substrate Vimentin

3

MEKK1 cPLA2

DTVD (3) DELD (3)

Stat1

MELD

PKC δ

DMQD (3)

PKC θ

DEVD (3)

D4-GDI

DELD (3) LLGD (1)

PAK2

DSHVD (3)

RasGAP Raf1

DTVD? (3)

Akt-1 Cbl Cbl-b PP2a NF-κB p65

NF-κB p50 Cell cycle

DEQD (3) ASVD, DTDD, DCRD (3) DVSD (3)

DNA replication DEVD, complex C large DLVD, subunit IETD (3) mdm2 DVPD (3) DNA topoisomerase I/II p110 PITSLRE YVPD kinase (1, 3)

Functional result

Proposed effect in apoptosis

Degraded

Disassembly of intermediate filaments Activated Activate SAPK Activated Generate arachodonic acid Inactivated Inhibit IFN signaling Activated Nuclear condensation Activated Nuclear fragmentation Inactivated Deregulate Rho GTPase function Activated Required for morphological changes Inactivated Inactivated Inhibit ERK activation Inactivated Inhibit survival pathway Inactivated Inactivated Activated Decrease active ERK Inactivated Inhibit survival gene induction

Reference 342

102, 103 343

344 345 346 347

348, 349

99 99 99 99 99 104 110

Inactivated

Inhibit survival gene induction

110

Inactivated

Inhibit DNA replication

113, 114

Altered

Inhibit p53?

350 98 351

(Continued )

P1: JER/mbg

P2: ARS/ary

January 22, 1999

QC: ARS

10:9

796 Table 2

Annual Reviews

AR078-24

RATHMELL & THOMPSON (Continued )

Category

Substrate Rb Wee1

Site (caspase)

Functional result

Proposed effect in apoptosis

DEAD (3)

Inactivated

Release E2F-1 Increase cdc2 and cdk2 activity

PARP

DEVD (3)

Inactivated

DNA-PK

DEVD, DDVD, DWVD, DFND (3)

Inactivated

DNA nicks unrecognized Inhibit repair of DNA strand breaks

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CDC27 DNA repair

Interleukin 1β Interleukin 16 Interleukin 18

YVAD (1) SSTD (3) LESD (1)

Presenilin 1 and 2

hnRNP C1 and C2

AQRD, DSYD (3) 1, 3 1, 3, 7, 8 1, 3, 7, 8 1, 3 DESD, VGAD 7, 3, 6

U1-70 kDa

DGPD (3)

lCAD/DFF-45

DEPD, DAVD (3)

Bcl-2

DAGD (3)

Huntingtin Androgen receptor Atrophin-1 Ataxin-3 Rabaptin-5

Bcl-xL

Bid

111 352

352

hsRAD51

Other

Reference

LQTD (8)

115–118 119–121

Inactivated

Inhibit repair of DNA strand breaks

122

Activated Activated Activated

Inflammation Inflammation Inflammation, induce interferon-γ

34, 35 123 83, 84

137

Inactivated

Inhibit endosome fusion Inactivated Inhibit RNA splicing Inactivated Inhibit RNA splicing Inactivated Required to allow CAD/CPAN to degrade DNA Altered Function similar to Bax to promote death Altered Function similar to Bax to promote death Inactivated Bcl-2 and Bcl-xL

135, 136 136 136 136 353 124 119, 125 126–128

130

131

278, 279

(Continued )

P1: JER/mbg

P2: ARS/ary

January 22, 1999

QC: ARS

10:9

Annual Reviews

AR078-24

797

APOPTOSIS IN THE IMMUNE SYSTEM Table 2

(Continued )

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Category

Site (caspase)

Substrate Sp1

2?, 3?

SREBP-1,2

3

UBF NOR-90

p28 Bap31 Nedd4 Caspases

Functional result

Proposed effect in apoptosis

Inactivated

Inhibit transcription Activated Promote cholesterol synthesis and LDL uptake? Inactivated Inhibit transcription, nucleolar organization

8, 1 DQPD (1, 3, 6, 7) Caspases and Activated Granzyme B

Regulate ubiquitination Required to elicit downstream effects in apoptosis

Reference 354 355

98

356 357 358, 359

activity (105). Other survival pathways that are inhibited include Akt-1 (99), which normally inactivates the death-promoting Bad (106, 107), and NF-κB (108–110). Proteolysis of cell cycle regulatory proteins, like Rb (111, 112), appears to act to block cell cycle checkpoints, while simultaneously inhibiting DNA synthesis via the inactivation of the DNA replication complex C large subunit (113, 114). DNA repair is also inhibited by the degradation of PARP (115– 118), DNA-PK (119–121), and hsRAD51 (122). A number of other substrates are cleaved to elicit the remaining characteristics of apoptosis. Inflammatory cytokines, IL-1β (34, 35), IL-16 (123), and IL-18 are processed (83, 84). RNA splicing is inhibited by the degradation of hnRNP C1 and C2 (124), as well as the U1 70-kDa protein (119, 125). Importantly, ICAD/DFF45 is degraded (126–128), allowing the apoptotic endonuclease CAD/CPAN to degrade DNA (127–129). The cell survival proteins Bcl-2 and Bcl-xL can also be cleaved, reversing their function to promote cell death rather than survival (130, 131). Caspases themselves can serve as substrates for other caspases, as well as for themselves via autoprocessing. Pro-caspase 1 is a substrate for caspases 4 and 11 (80, 82, 132). Pro-caspases 9 and 6 are substrates for caspase 3 (72, 133). Pro-caspases 3 and 7 are substrates for caspases 6, 8, and 10 (55, 72, 134). Procaspase 3 is also a substrate for caspase 1 (117). Pro-caspase 13 is a substrate for caspase 8 (96). The activation of pro-caspase substrates by caspase-mediated cleavage leaves no doubt as to its importance as a mechanism for the amplification and diversification of caspase activity (72, 134).

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

798

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

Caspase substrates are involved with surprising frequency in autoimmune and degenerative diseases. A number of substrates have been found to be autoantigens in autoimmune disorders (98). Although the immunological significance of this is unclear, the recent observation that phagocytes can present peptides from engulfed apoptotic bodies (23, 24) suggests a possible mechanism. It is possible that cleavage allows alternate processing of the caspase substrates by the phagocyte to reveal cryptic epitopes that could stimulate autoreactive T cells. Newly revealed cryptic epitopes may also be exposed to B cells if plasma membranes on apoptotic bodies permeabilize and allow cytosolic contents to escape before engulfment. Caspase substrates implicated in disease also include several proteins with polyglutamine tracts caused by CAG expansion, such as huntingtin, atrophin-1, ataxin-3, and the androgen receptor (135, 136). Additionally, the Alzheimer’s disease–associated presenilins 1 and 2 can be substrates for caspase 3 (137). It will be important to establish the relationship between caspase cleavage and the etiology of these diseases.

Caspase Inhibitors Owing to the importance of caspases in apoptosis, it is not surprising that inhibitors of caspase activity have evolved. For viruses, such inhibitors block both cell death and the release of inflammatory cytokines so as to allow the virus to complete its life cycle and infect more cells. Among the best described viral caspase inhibitors are the cowpox virus product cytokine response modifier A (CrmA) and the baculovirus antiapoptotic protein p35. CrmA is an early cowpox gene product that strongly inhibits inflammation and recruitment of macrophages to the site of infection (138). Overexpression of CrmA has also been shown to block apoptosis induced by growth factor withdrawal, CD95, or TNF (139, 140). Inhibition of both inflammation and cell death by CrmA is likely to be caused by its ability to inhibit caspases. CrmA has been found to directly inhibit the proteolytic activity of caspases 1 and 8 and, to a lesser extent, caspase 6 (141, 142). Other caspases, such as 3 and 7, are poorly affected by CrmA (118, 142). The baculovirus antiapoptosis product p35 inhibits a wider variety of caspases, including caspases 1, 2, 3, and 4, as well as CED-3, but has no effect on granzyme B (143–145). Mechanistically, CrmA and p35 act as competitive inhibitors. They can each be cleaved by the caspases they inhibit (144–146) and then remain firmly bound to the enzyme, thus blocking its ability to cleave additional substrates (144, 146). Mammalian cells also produce endogenous inhibitors of apoptosis that are homologous to an additional apoptosis-inhibiting baculovirus gene iap (inhibitor of apoptosis) (147, 148). To date, four human iap genes have been identified: NAIP (149); cIAP-1 (150)/hIAP-2 (149)/MIHB (151); cIAP-2 (150)/hIAP-1 (149)/MIHC (151); and hILP (152)/XIAP (149)/MIHA (151). IAP molecules

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

799

have been shown to protect against apoptosis by binding and inhibiting the activation and activity of specific caspases. XIAP, cIAP-1, and cIAP-2 inhibit caspases 3, 7, and 9 but not caspases 1, 6, and 8 (153–155). NAIP, however, has been found to be unable to inhibit caspases 1, 3, 6, 7, or 8 (154). IAP molecules are characterized by three N-terminal baculovirus IAP repeats (BIRs) and one C-terminal zinc fingerlike RING domain, except NAIP, which lacks the C-terminal RING domain. Both N-terminal BIRs and a C-terminal RING have been reported to be required for viral iap to protect against cell death in insects (147). In contrast, dissection of XIAP determined that the RING domain is not necessary and only BIR2 was required to bind and inhibit caspases 3 and 7 (153, 156). In addition to direct inhibition of caspases through physical interaction, IAPs can affect signal transduction pathways that may have important effects on cell survival. Although the precise function of the interaction is uncertain, cIAP-1 and -2 have each been shown to bind the TNFR-associated factors (TRAFs) 1 and 2 (150, 151, 154), important molecules in signal transduction from TNFR family members. This interaction occurs through the IAP BIR domains and the TRAF-N domain of TRAFs 1 and 2 (150) and can lead to the recruitment of IAP molecules to the TNFR-2 signaling complex (150), which may affect its signaling outcome. XIAP and NAIP do not associate with any of the six described TRAF molecules (154). IAP molecules can also directly activate signaling pathways. XIAP expression has been reported to be sufficient to cause the selective activation of SAPK1 (JNK1) by a mechanism independent of the SAPK1 activators MEKK1 and MKK4 (157). Inhibition of XIAP-induced SAPK1 activity by a dominant negative JNK1 was then found to block the ability of XIAP to protect against caspase 1-induced death (157). The IAP molecule, cIAP-2, has also been found to be sufficient to induce NF-κB activity (158). This induction of NF-κB may also be critical to the antiapoptotic function of cIAP-2, as indicated by its inability to suppress TNF cytotoxicity when NF-κB activation was blocked by dominant negative IκB. An additional class of caspase inhibitors are the FLIPs (FLICE-inhibitory proteins). These caspase inhibitors were first identified in herpesvirus and molluscipoxvirus (159, 160). Like caspases 8 and 10, FLIPs each contained DEDs. By binding the DED-containing adaptor molecule FADD or caspase 8, FLIPs were able to block the recruitment of caspase 8 to the death receptorsignaling complex and interfere with the early signaling events of CD95, DR-3, DR-4, and TNFR-I that lead to apoptosis. A human homolog of these viral genes, FLIP (161)/Casper (162)/I-FLICE (163)/FLAME-1 (164)/CASH (165)/CLARP (166)/MRIT (167)/Usurpin (168), was rapidly identified by a number of groups. There are two predominant splice variants of FLIP, including FLIPL, which consists of two DEDs and a caspaselike domain, and FLIPS, which harbors only the two DEDs. FLIP has homology to caspases 8 and 10 and may

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

800

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

have arisen from a gene duplication. It is not a caspase, however, because the conserved caspase active-site sequence of QACxG is replaced by QNYxV in human and mouse FLIP. As a result of these substitutions, FLIP has no detectable enzymatic activity. FLIP has been described both as protecting from apoptosis (161, 163, 164, 168) and inducing apoptosis (162, 166, 167), whereas another group found that splice variants and cell type were critical in determining whether FLIP would enhance or protect from apoptosis (165). Consistent with FLIP as a caspase inhibitor is the observation that FLIP mRNA expression levels decrease as T cells become susceptible to CD95-induced apoptosis (161, 169). FLIP has been reported to associate with a number of different proteins that may mediate its biological activity. Some reports say FLIP binds the adaptor protein FADD and can be recruited to receptor signaling complexes (161, 162, 165, 167), whereas others maintain FLIP does not bind FADD or does so only weakly (163, 164, 166). FLIP does not bind the TNFR-1–associated adaptor protein TRADD or the serine/threonine kinase RIP (162, 165). It has, however, been reported to directly bind Bcl-xL (167), TRAF1 (162), and TRAF2 (162). FLIP can also bind and be proteolytically processed by caspases (161, 162, 164). Caspases 3, 6, 7, and 8 can bind and cleave after LEVD376 (341 in the FLAME-1–predicted amino acid sequence). The N-terminal cleavage product (equivalent to the prodomain and large enzyme subunit of caspases) was shown to maintain the ability to interact with caspase 8 (161). Mutation of D376 to block processing, however, did not block FLIP’s ability to induce death when experimental conditions favored FLIP toxicity (162). The role of FLIP in caspase regulation is clearly quite complex. Its prosurvival or proapoptosis function varies on splice variant, cell type, and expression level. It may act to prevent caspase activation by interfering with caspase 8’s recruitment to death receptor complexes, perhaps by acting as a competitive inhibitor substrate similar to p35 and CrmA. Alternatively, it may act to promote death by causing the oligomerization and subsequent activation of caspases. Its interactions with other cell death regulatory proteins, such as Bcl-xL, and TRAFs 1 and 2, point to possible additional roles in apoptosis regulation.

PATHWAYS TO CASPASE ACTIVATION IN LYMPHOCYTE DEATH Stimulation of Caspase Activity in Damaged or Neglected Cells Redistribution of cytochrome c from mitochondria to cytosol is an early event in the apoptosis of damaged or neglected cells and results in the activation of caspases (9, 10, 13, 14). This caspase activation was recently shown to depend on the presence of Apaf-1 (apoptotic protease-activating factor), a mammalian

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

801

homolog of CED-4 (45). Apaf-1 was biochemically purified, along with cytochrome c (Apaf-2) (170) and caspase 9 (Apaf-3) (92), as a factor necessary for the activation of caspases in a cell-free system. Just as CED-4 is required for efficient CED-3 processing (39–41), Apaf-1 was found to be required for caspase 9 processing. When cytochrome c is released from mitochondria, Apaf-1 binds caspase 9 in a cytochrome c- and ATP-dependent manner (92). Apaf-1 then stimulates the activation of caspase 9, which can, in turn, process caspase 3 to elicit downstream apoptotic events. In addition to interacting with cytochrome c and caspase 9, Apaf-1 has been immunoprecipitated with Bcl-xL and caspases 4 and 8 (171, 172). The association of Bcl-xL with Apaf-1 most likely occurs through the BH4 domain of Bcl-xL (173) and inhibits association of Apaf-1 with caspase 9 and its subsequent activation. Apaf-1 has three primary domains (45, 92). An 85-amino-acid N-terminal domain has homology to prodomains of CED-3 and other caspases. This domain, the caspase recruitment domain (CARD) (174), is important in the physical interaction of Apaf-1 with caspases (92). CARD domains have also been identified in the DD adaptor molecules RAIDD (175)/CRADD (174, 176), c-IAP1 and -2 (174), RICK (177)/Rip2 (178)/CARDIAK (179), and ARC (180). The CARD domain of Apaf-1 is followed by a domain of 320 amino acids with homology to CED-4 that may function as a nucleotide-binding sequence and a C-terminal domain containing 12 WD-40 repeats, a motif that may participate in protein-protein interactions (181). Although the role of the CED-4 homology domain is uncertain, the WD-40–containing domain appears to be critical in Apaf-1 regulation, as its deletion has been reported to allow Apaf-1 to stimulate caspase 9 processing in the absence of cytochrome c or ATP (91). Interestingly, domains similar to the CED-4 homology domain are also found in numerous plant disease resistance genes (182). In plants, these disease resistance genes are thought to encode receptors for specific pathogen components to elicit the hypersensitive response, a form of programmed cell death that blocks further infection (182, 183).

Stimulation of Caspase Activity in Healthy Cells In addition to activation of caspases through mitochondrial release of cytochrome c, caspases can be activated directly by cell surface receptors to initiate cell death. This pathway of caspase activation is invoked to eliminate otherwise healthy and undamaged but potentially dangerous cells, such as autoreactive lymphocytes. The best described receptors that activate caspases are members of the TNFR family. The TNFR family can be roughly divided into receptors that promote cell survival and receptors that promote cell death (65). Receptors that oppose death and promote survival include CD27 (184), CD30 (185), CD40 (186), TNFR-II (187), Ox40 (188), 4-1BB (189), and p75 NGFR (190). Receptors in this family that can promote cell death include CD95 (Fas/APO-1)

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

802

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

(191, 192), TNFR-I (193, 194), DR3 (195)/Wsl-1 (196), DR4 (66), and DR5 (66). Death induced by TNFR family members leads to and depends on the activation of caspases (67–69). Although all family members have homologous cysteine-rich repeats in their extracellular domains, the cell-death–inducing receptors also share an 80-aminoacid domain known as the DD (63–66). DDs are critical in the activation of caspases by TNFR family members. Point mutations, such as the lpr cg mutation of CD95 (197), or deletions of the CD95 or TNFR-I DDs abrogate the ability of the receptors to induce apoptosis (63, 64, 198). The structure of the DD of CD95 has been solved by NMR and consists of six amphipathic α-helices, which are arranged antiparallel to one another (199). They have no enzymatic activity but instead function to bind DD-containing adaptor proteins on receptor ligation. DD-harboring adaptor proteins that are recruited to DDs on TNFR family members activate caspases or other downstream signaling pathways (Figure 2). FADD (200)/MORT1 (201) binds directly to the DD of CD95. Through the

Figure 2 Activation of caspases by cell surface receptors. CD95 and TNFR-I can recruit caspases to their receptor complexes by three primary mechanisms involving the interactions of death domains, death effector domains, and caspase recruitment domains.

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

803

interaction of a second domain on FADD, a DED, FADD can then recruit caspases 8 (53–55) and 10 (55, 97), which each have two DEDs, to the signaling complex. Oligomerization of the caspases then stimulates their activation (88–90), and the caspase cascade is initiated (72). The TRADD adaptor protein binds the DD of TNFR-I and is sufficient, upon overexpression, to induce NF-κB activity as well as apoptosis (202). TRADD is able to activate these disparate pathways by its ability to bind the additional adaptor molecules TRAF2 and FADD (203). The binding of TRAF2 can lead to the activation of NF-κB and SAPK (204, 205), and the binding of FADD leads to caspase activation. TRADD can also bind the DD-containing RIP (206), which has itself been found to associate with CD95 (207). RIP is a serine/threonine kinase which, like TRADD, can induce NF-κB or apoptosis (206, 207). The primary function of RIP may be to activate NF-κB, because cells deficient in RIP have been found to be unable to induce this transcription factor in response to TNF-α (208, 209). The role of the kinase domain of RIP is uncertain, but expression of a truncated RIP containing only the DD potently inhibited TNF-α–induced NF-κB activity while still inducing apoptosis (206). RIP also binds to the DD protein RAIDD (175)/CRADD (176). RAIDD has an N-terminal CARD domain that allows a specific interaction with caspase 2 but not caspases 1, 3, 4, 6, 7, or 9 (175, 176). Overexpression of RAIDD allows caspase 2 to be recruited to the TNFR-I signaling complex and promotes apoptosis. The significance of RAIDD/caspase 2 interactions in TNFR-I–induced apoptosis, however, is unclear owing to the lack of TNFR-I–mediated cell death in caspase 8-deficient mice (76) and the lack of a relevant phenotype in caspase 2-null mice (79). Thus, caspases can be activated within cell surface receptor complexes by the molecular interactions of DDs, DEDs, and CARDs in FADD/caspase 8 or 10, TRADD/ FADD/caspase 8 or 10, and TRADD/RIP/RAIDD/caspase 2. Several other molecules have been found to associate directly with CD95 to influence its activation of caspases. FAP-1 is a phosphatase that binds CD95 and inhibits apoptosis (210). The selective expression of FAP-1 in Th2 CD4+ T cells may explain the greater sensitivity of Th1 cells to CD95-induced apoptosis (211). Faf-1 binds to CD95 but not the lpr cg mutant of CD95 to potentiate cell death through an unknown mechanism (212). Daxx lacks a DD, yet it binds to the DD of CD95 and causes the activation of SAPK and potentiates CD95-induced apoptosis (213).

REGULATION OF MITOCHONDRIAL HOMEOSTASIS AND CYTOCHROME C RELEASE BY BCL-2 FAMILY PROTEINS The functional and sequence homologies of CED-9 with Bcl-2 and its family members (37, 38) point to the Bcl-2 family as critical regulators of caspase

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

804

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

activation and apoptosis. Bcl-2 was originally cloned as a gene from chromosome 18, which became constitutively overexpressed on translocation adjacent to the immunoglobulin locus on chromosome 14 in human follicular lymphoma (214–217). How Bcl-2 may function as an oncogene was unclear until Vaux et al (218) demonstrated that the expression of Bcl-2 provided a survival signal that allowed growth factor–dependent cells to persist in the absence of growth factor. Thus, by preventing cell death, constitutive Bcl-2 expression perturbed the homeostatic regulation of cell number and allowed cells to accumulate and become cancerous. In addition to protecting against growth factor withdrawalinduced death, Bcl-2 has also been shown to provide protection against a wide variety of death stimuli, such as DNA damage, metabolic inhibition, oncogene expression, and glucocorticoid treatment (219). Bcl-2 can also protect against CD95-induced cell death in some cell types. The identification of additional Bcl-2–related genes has now defined a family of Bcl-2–like molecules. Some mammalian Bcl-2 family members, like Bcl-2 itself, promote survival and antagonize death stimuli, such as Bcl-xL (220), Bcl-w (221), Mcl-1 (222, 223), and A-1 (224)/Bfl1 (225). The Epstein-Barr virus protein BHRF1 (226) and the adenovirus E1B 19-kDa proteins (227–229) are also survival-promoting Bcl-2 homologs. Another group of Bcl-2–related proteins promotes death rather than survival. The mammalian proapoptotic Bcl-2 family members include Bax (56), Bak (230–232), Bcl-xS (220), Bok (233)/Mtd (234), Bad (235), Nbk (236)/Bik (237), Bid (238), Bim (239), Hrk (240), Blk (241), and BNIP3 (242, 243). Shared among these family members are homologies in four domains, the Bcl-2 homology (BH) domains 1–4. Although Bcl-w and A-1/Bfl1 lack an unstructured loop between BH3 and BH4 present in Bcl-2, Bcl-xL, and Mcl-1 (244), all the mammalian survivalpromoting family members contain BHs 1, 2, 3, and 4 (reviewed in 245). The death-promoting Bcl-2 family members can be divided into three classes. Class I consists of proteins with BH1, 2, 3, and 4 and includes Bax, Bak, and Mtd/Bok. Class II consists of the only protein with only BH3 and BH4, Bcl-xS. Class III family members have only the BH3 domain and include the death promoting Bik, Hrk, Bim, Blk, Bad, Bid, and BNIP3. With the exception of Bad and Bid, all family members also have a transmembrane tail that can target them to similar subcellular localizations such as outer mitochondrial membrane, endoplasmic reticulum, and outer nuclear membrane (246–249).

Regulation of Bcl-2 Family Member Activity by Dimerization An important characteristic of Bcl-2 family members is their ability to form homodimers and heterodimers. Indeed, it was through such interactions that a number of family members were identified. Structurally, homodimers of the

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

805

antiapoptotic Bcl-2 or Bcl-xL involve the BH1 and BH2 domains (250). Proapoptotic BH3-containing molecules, such as Bax or Bak, can bind Bcl-2 or Bcl-xL through the interaction of their BH3 domains and the hydrophobic cleft created by BH1, BH2, and BH3 of Bcl-2 or Bcl-xL (231, 251–253). There has been much debate as to the role of dimerization in the function of Bcl-2 family members. It appears, though, that a rheostat model of the balance of homodimers of survival-promoting to homodimers of class I death-promoting family members influences cell fate to survive or die (56, 254, 255). In this way, the expression of death-promoting molecules like Bax and survival-promoting molecules like Bcl-2 would form a continuum. When Bax is expressed preferentially, Bax homodimers would be formed and death would be promoted. When Bcl-2 is expressed preferentially, Bcl-2 homodimers would form and survival would be promoted. In contrast, BH3-only proteins, such as Bad and Bid, do not homodimerize and have no intrinsic activities. By heterodimerizing with Bcl-2 or Bcl-xL, however, BH3-only proteins can inhibit their antiapoptotic activities, thus promoting cell death by decreasing the ratio of active antiapoptotic to active proapoptotic Bcl-2 family members (251, 256). BH3only proteins may also bind death-promoting family members to stimulate their activity (245).

Regulation of Bcl-2 Family Member Activity by Phosphorylation Although Bcl-2 family members have been typically shown to be regulated by expression levels and dimerization patterns, post-translational modification can also control their effects on cell fate. The best defined example of such modification is the regulation of cell death in growth factor–dependent cell lines by phosphorylation of the BH3 domain containing Bcl-2 family member Bad (106). In the presence of the growth factor IL-3, Bad was found to be serine phosphorylated by the kinase Akt (107) and associated with the cytosolic protein 14-3-3 (106). On removal of IL-3, however, Bad became dephosphorylated and dissociated from 14-3-3. The BH3 domain of free Bad was then available to bind Bcl-xL, thus blocking the function of Bcl-xL (256, 257) and releasing the proapoptotic Bax (106). Bcl-2 itself can also be phosphorylated on the 60-amino-acid-loop domain that separates α1 from α2 (258). The treatment of cells with the chemotherapeutic drug Taxol or the phosphatase inhibitor okadaic acid resulted in serine phosphorylation of Bcl-2, which was then unable to interfere with apoptosis (259). A likely candidate for this phosphorylation would be the serine/threonine kinase Raf-1 owing to its association with Bcl-2 (260–262). In addition, the SAPK pathway (263) and the cyclic AMP-dependent protein kinase A (264) have also been implicated in Bcl-2 phosphorylation.

P1: JER/mbg

P2: ARS/ary

January 22, 1999

806

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Bcl-2 Family Members Exhibit Ion Channel Activity Although it was known that Bcl-2 and Bcl-xL acted in the same pathway (265) and that this pathway could be antagonized by the death agonists, such as Bax (56), little was known about the biochemical nature of their function before the solution of the structure of Bcl-xL (244). Bcl-xL comprises two central hydrophobic α-helices surrounded by five amphipathic helices. BH1, -2, and -3 are spatially close and form a hydrophobic cleft to which the BH3 domains of proapoptotic family members can bind (251). The BH4 domain consists of a side α-helix, α1 (244). An additional 60-amino-acid-loop domain between α1 and α2 is unstructured and nonessential for the proper folding of the remainder of the molecule or its antiapoptotic function (244, 258). Importantly, the overall structure of Bcl-xL is similar to that of bacterial pore-forming proteins such as the A subunit of diphtheria toxin and the colicin family of bacterial toxins (244), a family of molecules that can insert into membranes and form ion channels. Thus, it was predicted that the two central α-helices of Bcl-xL could insert into membranes to form ion channels. The antiapoptotic Bcl-xL (266) and Bcl-2 (267–269), as well as the proapoptotic Bax (267, 269), have indeed been shown to form ion channels in lipid membranes. The channels generated by Bcl-2 have a preference for monovalent cations, such as K+, and show optimal conductance at acidic pH (267–269). Bcl-xL channels are cation selective at physiological pH but lose selectivity at acidic pH (266). These channels were also found to be reversibly inhibited by lumenal calcium (270). In contrast, Bax channels function over a wide range of pH but preferentially conduct anions (267, 269).

Bcl-2 Family Members Regulate Mitochondria in Apoptosis Mitochondria undergo loss of cytochrome c and PT in apoptosis (13, 14). Mitochondrial PT is blocked by Bcl-2 (11, 12, 19) and induced by Bax (271). The release of cytochrome c from mitochondria is also blocked by Bcl-2 (9, 10, 13) and induced by Bax (272). In most instances, the release of cytochrome c, however, appears to occur before mitochondrial PT and loss of membrane potential in cells undergoing apoptosis (13, 14). This suggests that the ability of Bcl-2 to block the mitochondrial PT is secondary to its ability to cause the retention of cytochrome c. Recent studies have suggested one mechanism by which Bcl-2 proteins may regulate mitochondria and cytochrome c release (Figure 3). The removal of growth factor from growth factor–dependent cell lines causes a decrease in metabolic activity (273), which is accompanied by a decrease in mitochondrial membrane potential, a block in mitochondrial ATP/ADP exchange, and a decrease in cellular ATP (274). Although Bcl-2 cannot overcome the metabolic slowdown and decrease in mitochondrial membrane potential, the Bcl-2 family

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

807

Figure 3 Proposed mechanism by which growth factor withdrawal leads to caspase activation. Removal from growth factor causes cells to metabolically arrest and blocks ADP/ATP exchange, resulting in a lack of ADP in the mitochondrial matrix. This lack of substrate for ATP synthase causes a buildup of protons in the intermembrane space, which can lead to swelling and outer membrane rupture, thus releasing cytochrome c. Bcl-xL can inhibit this process by maintaining ADP/ATP exchange.

member Bcl-xL promotes the continued mitochondrial exchange of ADP for ATP, thus maintaining oxidative phosphorylation and sustaining cellular ATP levels at lower rates of glycolysis (274). Maintenance of ATP synthase function could prevent the transient mitochondrial hyperpolarization and swelling (13) that may be caused by a lack of matrix ADP as a substrate to relieve the proton gradient created by the electron transport system. If unchecked by Bcl-2 or Bcl-xL, hyperpolarization can result in mitochondrial swelling and rupture of the outer mitochondrial membrane, thus releasing cytochrome c into the cytoplasm (13). Whereas the mechanism by which Bcl-xL allows continued ADP transport into mitochondria after growth factor withdrawal is uncertain, it is possible that Bcl-xL can act as an ion-conducting channel to relieve the electrical gradient caused by proton accumulation. By dissipating this accumulating electrical potential, Bcl-xL would allow the voltage-gated adenine nucleotide channel to remain in an open state and continue to allow ADP/ATP exchange between cytosol and mitochondria to sustain ATP production and cell survival. The role of pH selectivity of Bcl-2 family member channel activities correlates with their proposed role in regulating mitochondrial polarization. The increased dimerization of Bcl-2 family members in acidic pH (275) and the preference of Bcl-2 and Bcl-xL to form cation-selective channels and Bax to form anion-selective channels suggests that prosurvival and proapoptotic

P1: JER/mbg

P2: ARS/ary

January 22, 1999

808

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Bcl-2 family members function in opposite fashions to regulate the polarization of the intermitochondrial space. Bcl-2 and Bcl-xL may act as channels to dissipate cations and inhibit hyperpolarization and voltage-gated adenine nucleotide channel closure on increased intermembrane-space pH and metabolic arrest. Bax, on the other hand, may function as a channel to promote anion influx to the intermembrane space, causing further H+ accumulation and matrix swelling. Alternatively, Bax may function as a channel to release cytochrome c directly. In this way Bcl-2 and Bcl-xL could inhibit and Bax would promote cytochrome c redistribution.

INTEGRATION OF CELL DEATH PATHWAYS Activations of caspases by cytosolic cytochrome c and Apaf-1 or by cell surface receptors are not mutually exclusive occurrences. If cells are moderately damaged, only a small fraction of mitochondria may progress to release cytochrome c, and, therefore, caspase 9 activation may be insufficient to cause cell death. Nevertheless, these moderately damaged cells may be sensitized to death by death receptors. Likewise, insufficient caspase activation by cell surface receptors can be amplified by the release of cytochrome c from mitochondria and Apaf-1-mediated caspase 9 activation. Indeed, although CD95 is able to induce caspase activity directly in its receptor complex, the release of cytochrome c from mitochondria is often a key component in CD95-induced apoptosis. It is possible to divide cells into two types based on the different roles played by mitochondria (276). Type I cells were found to generate a large amount of active caspase 8 immediately after CD95 ligation. Type II cells, however, did not have significant amounts of active caspases 8 or 3 until after the loss of 19 m. The overexpression of pro-caspase 3 in a type II cell was sufficient to convert it to a type I cell, suggesting that the amount of caspase activity generated in the cytosol determined whether cells would undergo a mitochondria-independent (type I) or -dependent (type II) pathway of apoptosis (276). If cells are type II, then the need for 19 m loss to allow sufficient amplification of the caspase cascade for apoptosis to proceed suggests a potential role for Bcl-2 proteins in CD95-induced apoptosis. Bcl-2 overexpression can indeed block the release of cytochrome c from mitochondria in anti-CD95–treated cells (277). Recently, a mechanism for the involvement of mitochondria in CD95induced apoptosis was identified. The BH3-only-containing Bcl-2 family member, Bid (238), is normally a cytosolic protein. After CD95 ligation, however, it is cleaved and binds Bcl-2 and Bcl-xL on the mitochondrial outer membrane, causing their inactivation (278, 279), which may lead to the loss of integrity of the outer mitochondrial membrane and release of cytochrome c into the cytosol where Apaf-1 and caspase 9 can amplify the caspase cascade to cause apoptosis.

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

809

In addition to preventing outer mitochondrial membrane rupture and redistribution of cytochrome c, Bcl-2 and Bcl-xL can also regulate caspase activity and amplification downstream of cytochrome c release and caspase activation. At low levels, microinjection of cytochrome c into Bcl-2–expressing cells has been reported to result in only 30% of the cell death observed in control cells (280). Likewise, Bcl-2 expression could protect cells from Bax-induced death and caspase activation despite the release of cytochrome c (281). Bcl-xL was also able to inhibit the loss of 19 m and cell death of anti-CD95–treated Jurkat T cells in which caspases had been activated sufficiently to cleave PARP (220). There are three possible explanations for the effect of survival-promoting Bcl-2 family members on apoXptosis after cytochrome c release or caspase activation. First, caspases may become activated but at only a low level. The high expression of transfected Bcl-2 or Bcl-xL may then be sufficient to block the further amplification of the apoptotic signal mediated through cytochrome c release and subsequent apoptosis. Second, Bcl-xL has been suggested to have the ability to bind cytochrome c (282). By binding cytochrome c, Bcl-xL could sequester it from Apaf-1/caspase 9, thereby blocking the efficient activation of the caspase cascade after release of cytochrome c from mitochondria. Third, Bcl-2 family members may directly regulate caspases. Both Bcl-2 and Bcl-xL have been shown to be caspase substrates and their cleavage within the loop between BH4 and BH1 was found to reverse their function from a prosurvival to a proapoptotic one (130, 131). The mutation of the caspase cleavage site in Bcl-xL, however, was found to impair rather than enhance the ability of Bcl-xL to protect against apoptosis (131). These data suggest that Bcl-xL may bind and inhibit caspases until the point when caspase activity becomes sufficient to cleave Bcl-xL and cause it to become proapoptotic.

CELL DEATH PATHWAYS IN THE IMMUNE SYSTEM Neglect It has been proposed that all cells require the continuous presence of survival factors to avoid death by neglect (283). Lymphocytes have been shown to receive survival signals both from antigen receptors and their microenvironment. Lymphocytes that do not receive sufficient survival signals from these sources may become unable to generate ATP and will undergo death by neglect when energy stores are sufficiently reduced (29). If mitochondrial damage is extreme, cells will die regardless of caspase activity. Activation of the caspase cascade by mitochondrial release of cytochrome c will, however, allow an apoptotic morphology. During development, lymphocytes require signals from their antigen receptors to be positively selected and continue development. In the absence of these signals, due to failed V(D)J recombination, generation of a nonfunctional

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

810

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

receptor, or mutations in antigen receptor signaling pathways, lymphocytes die of neglect. Those lymphocytes that successfully mature to enter the periphery must then migrate to their appropriate niche in the secondary lymphoid organs, the periarteriolar sheath (PALS) for T cells or the B-cell follicles for B cells, to survive and enter the recirculating pool of long-lived mature lymphocytes. If lymphocytes fail to compete for entry into these limited pools of lymphocytes, they do not receive proper survival signals and they quickly die of neglect (284–286). The mechanism of this selection is unclear, but B cells that receive sufficient antigen receptor signals have diminished ability to enter B-cell follicles (287–289). Nevertheless, antigen receptors appear to be required for a necessary survival signal. Conditional knockout of surface immunoglobulin has shown that mature B cells cannot survive in the absence of antigen receptors (290). T cells also require antigen receptor-initiated signals to survive, because CD8 T cells adoptively transferred into TAP-deficient mice fail to survive (291, 292). Antigen-specific lymphocytes also may undergo death by neglect after a productive immune response when cytokines become scarce. Ectopic overexpression of Bcl-2 and Bcl-xL in transgenic mice to allow continued mitochondrial retention of cytochrome c in the absence of growth factor has allowed the observation of neglected cells and has highlighted the importance of neglect as a mechanism of cell death in the immune system. The immune repertoire is indeed altered by Bcl-2 or Bcl-xL transgenes to include additional cells (293–296). Most likely these cells were developing lymphocytes that were not positively selected but nevertheless survived owing to Bcl-2 or Bcl-xL expression (295, 297). In addition to allowing more cells to survive and enter the periphery, mature B-cell life spans are extended by Bcl-2 expression (298–300), and B cells that fail to enter the B-cell follicles survive significantly longer when expressing Bcl-2 (287). The large discrepancy between the number of lymphocytes produced each day and the number that survive even in the presence of Bcl-2 or Bcl-xL transgenes, however, demonstrates that Bcl-2 and Bcl-xL cannot completely protect against a lack of positive selection or growth factor. Rather, Bcl-2 and Bcl-xL appear to reduce the thresholds of positively selecting and growth factor stimuli required for survival.

Cell-Death–Inducing Receptors The elimination of healthy cells by cell surface receptor-induced activation of caspases is also important in the immune system. Lymphocytes subject to this type of cell death are generally dangerous, such as autoreactive cells or cells that have been activated in a bystander fashion. In these cases, cells are not damaged or neglected and, in the absence of caspase activation, would survive. Activation of caspases by receptor pathways, therefore, is crucial to the initiation of death in these cells, and the absence of caspase activation can lead to disease, such as

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

811

autoimmunity. The best example of this is the elimination of mature autoreactive lymphocytes by activation of caspase 8 by CD95 or TNFR-I. In the absence of CD95 or its ligand, as is the case in the lymphoproliferation (lpr)- and generalized lymphoproliferative-disorder (gld ) mutant mice, autoreactive T and B cells are not eliminated in the periphery, and mice suffer from autoimmunity and lymphadenopathy (198, 301). In the absence of sufficient costimulation, autoreactive T cells normally depend on CD95 for activation-induced cell death (302), and self-reactive B cells can be eliminated by CD95 on presentation of antigen to CD4+ T cells (303, 304). The similar phenotypes of humans harboring mutations in CD95 highlight the importance of this pathway of cell death (305, 306).

Antigen Receptors Antigen receptors can also cause caspase activation and cell death of both developing and mature lymphocytes (86, 307–312). Whether antigen receptor activation of caspases involves mitochondrial release of cytochrome c or a mitochondria-independent pathway is unknown. Any mitochondrial damage induced by antigen receptors is not in itself sufficient for cell death, however, because unlike cells that have suffered irreversible mitochondrial damage, antigen receptor–induced cell death depends on caspase activity (86, 307–312). Further supporting a lack of involvement or only a limited role for mitochondria in antigen receptor–induced cell death, expression in transgenic mice of Bcl-2 and Bcl-xL, which can inhibit mitochondrial release of cytochrome c, has, in most instances, failed to significantly affect antigen receptor–induced cell death (293, 297, 313–316). In one case, however, Bcl-xL was reported to allow B cell survival in the presence of negatively selecting antigen when Bcl-2 would not (317). Alternatively, antigen receptors may activate caspases in a mitochondrialindependent fashion similar to TNFR family members. Because antigen receptors are not known to link directly to caspases as TNFR family members can, however, it is unclear how they may activate caspases independent of mitochondria. It is possible that intermediate pathways may be involved in antigen receptor–induced caspase activation. Indeed, the general caspase inhibitor zVAD-fmk, while able to block complete processing, was unable to block the initial cleavage of caspase 3 between prodomain/p17 and p12 (86) after stimulation through the B-cell antigen receptor. Additional pathways that have been implicated in antigen receptor–induced cell death that may be involved in caspase activation include nur77 (318, 319) and p38 MAPK (311). These two pathways of caspase activation are not mutually exclusive. Antigen receptors may cause limited mitochondrial release of cytochrome c as well as lead directly to caspase activation. In this case, each pathway would sensitize cells to the other and enhance overall likelihood of sufficient caspase activation to cause cell death.

P1: JER/mbg

P2: ARS/ary

January 22, 1999

812

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CONCLUSIONS Our understanding of apoptosis has increased dramatically over the past few years. The search for the “central executioner” of apoptosis has given way to the study of caspases and their regulation. Caspases are required for critical aspects of apoptosis, such as the rapid deconstruction of cellular structure, initiating the breakdown of DNA, and sending signals to alert the cell’s environment that it is dying and must be phagocytosed. Caspases are not required for cells to die. In the absence of caspase activity, irreversibly damaged or neglected cells will still die, but they will die in an unorganized manner resembling necrosis, releasing their cytosolic contents into their environment. The activation of caspases, however, is required for the induction of apoptosis in otherwise healthy cells. Among the most critical issues in the study of apoptosis, therefore, is how caspases get activated to allow the orderly death of damaged cells and how they become activated to eliminate healthy cells. There are two fundamental ways by which caspases are known to become activated, by the release of cytochrome c from mitochondria and by the ligation of death receptors (Figure 4). In the former case, cells that are damaged or

Figure 4 Caspase activation by neglect or surface receptors. See text for details.

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM

813

neglected become metabolically arrested, which can result in hyperpolarization of the mitochondrial membranes. Bcl-2 and its survival-promoting family members can halt this hyperpolarization, perhaps by acting as ion channels to dissipate excess charge or to promote more efficiently coupled respiration. The efforts of prosurvival Bcl-2 family members may be thwarted, however, if growth factor withdrawal has caused BAD dephosphorylation and binding to Bcl-2 or if the proapoptotic Bcl-2 family members, such as Bax or Bak, are highly expressed. If the Bcl-2 family is insufficient to block hyperpolarization, then the mitochondria swell transiently, and the outer membrane ruptures, releasing cytochrome c. Cytochrome c binds Apaf-1, which recruits and activates caspase 9, which, in turn, activates caspase 3, thus initiating the caspase cascade. Death receptors, on the other hand, can directly activate caspases in healthy cells in their receptor complexes. Caspases 8 and 10, for example, can be directly recruited to CD95 and TNFR-I receptors by FADD and TRADD/FADD complexes. Their oligomerization in the receptor complexes causes processing and the caspase cascade is initiated. In some instances, however, death receptors do not activate a sufficient number of caspases and the cascade must be amplified through the involvement of mitochondria. This can be accomplished by the cleavage of BID, which then acts to inhibit Bcl-2 and Bcl-xL, causing mitochondria to release cytochrome c. The combined caspase activation from death receptor complexes and cytochrome c can then result in the death of the cell. Although much progress has been made toward understanding caspase activation and function, numerous issues remain unresolved. How are caspases physically activated to allow the formation of homodimers or heterodimers, as the caspase 1 and 3 structures suggest? How do caspase-interacting proteins such as Apaf-1 or FADD enhance this activation? Is Apaf-1 an initial representative of a family of molecules with similar functions or does it lack mammalian functional homologs? What are the energy requirements for caspase activation and activity? Which caspase substrates account for which phenotypes of apoptosis? How do other pathways, such as those initiated through antigen receptors, cause caspase activation? How do the different death pathways interact? Importantly, what causes their initiation in the immune system and how may their disruption lead to disease? As some of the above issues become resolved, no doubt many more will arise. The study of the central components of apoptosis and indeed the nature of apoptosis itself will continue to be a fascinating area of inquiry for many years to come. ACKNOWLEDGMENTS We thank Mathew Vander Heiden and Maria-Luisa Alegre, Ken Frauwirth, and Ameeta Kelekar for useful discussions and critical reading of this manuscript. JCR is a fellow of the Irvington Institute for Immunological Research, and CBT is an investigator in the Howard Hughes Medical Institute.

P1: JER/mbg

P2: ARS/ary

January 22, 1999

814

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON Visit the Annual Reviews home page at http://www.AnnualReviews.org

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Literature Cited 1. Thompson CB. 1995. Apoptosis in the pathogenesis and treatment of disease. Science 267:1456–62 2. Councilman WT. 1890. Report on the etiology and prevention of yellow fever. In Public Health Bulletin, ed. GM Sternberg, 2:151–59. Washington, DC: US Govt. Print. Off. 3. Klion FM, Schaffner F. 1965. The ultrastructure of acidophilic “Councilmanlike” bodies in the liver. Am. J. Pathol. 48:755–67 4. Kerr JFR. 1969. An electron-microscope study of liver cell necrosis due to heliotrine. J. Pathol. 97:557–62 5. Kerr JFR. 1971. Shrinkage necrosis: a distinct mode of cellular death. J. Pathol. 105:13–20 6. Farbman AI. 1968. Electron microscope study of palate fusion in mouse embryos. Dev. Biol. 18:93–116 7. Kerr JFR, Wyllie AH, Currie AR. 1972. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br. J. Cancer 26:239–57 8. Duvall E, Wyllie AH. 1986. Death and the cell. Immunol. Today 7:115–19 9. Kluck RM, Bossy-Wetzel E, Green DR, Newmeyer DD. 1997. The release of cytochrome c from mitochondria: a primary site for Bcl-2 regulation of apoptosis. Science 275:1132–36 10. Yang J, Liu X, Bhalla K, Kim CN, Ibrado AM, et al. 1997. Prevention of apoptosis by Bcl-2: release of cytochrome c from mitochondria blocked. Science 275:1129–32 11. Marchetti P, Hirsch T, Zamzami N, Castedo M, Decaudin D, et al. 1996. Mitochondrial permeability transition triggers lymphocyte apoptosis. J. Immunol. 157:4830–36 12. Zamzami N, Susin SA, Marchetti P, Hirsch T, Gomez-Monterrey I, et al. 1996. Mitochondrial control of nuclear apoptosis. J. Exp. Med. 183:1533–44 13. Vander Heiden MG, Chandel NS, Williamson EK, Schumacker PT, Thompson CB. 1997. Bcl-xL regulates the membrane potential and volume homeostasis of mitochondria. Cell 91: 627–37 14. Bossy-Wetzel E, Newmeyer DD, Green D. 1998. Mitochondrial cytochrome c

15.

16.

17.

18.

19.

20.

21.

22.

23.

release in apoptosis occurs upstream of DEVD-specific caspase activation and independently of mitochondrial transmembrane depolarization. EMBO J. 17:37– 49 Skalka M, Matyasova J, Cejokova M. 1976. DNA in chromatin of irradiated lymphoid tissues degrades in vivo into regular fragments. FEBS Lett. 72:271–74 Wyllie AH. 1980. Glucocorticoid-induced thymocyte apoptosis is associated with endogenous endonuclease activation. Nature 284:555–56 Wyllie AH, Morris RG, Smith AL, Dunlop D. 1984. Chromatin cleavage in apoptosis: association with condensed chromatin morphology and dependence on macromolecular synthesis. J. Pathol. 142:67–77 Martin SJ, Reutelingsperger CPM, McGahon AJ, Rader JA, van Schie RCAA, et al. 1995. Early redistribution of plasma membrane phosphotidylserine is a general feature of apoptosis regardless of initiating stimulus: inhibition by overexpression of Bcl-2 and Abl. J. Exp. Med. 182:1545–56 Castedo M, Hirsch T, Susin SA, Zamzami N, Marchetti P, et al. 1996. Sequential acquisition of mitochondrial and plasma membrane alterations during early lymphocyte apoptosis. J. Immunol. 157:512– 21 Verhoven B, Schlegel RA, Williamson P. 1995. Mechanisms of phosphatidylserine exposure, a phagocyte recognition signal, on apoptotic T lymphocytes. J. Exp. Med. 182:1597–601 Bratton DL, Fadok VA, Richter DA, Kailey JM, Guthrie LA, et al. 1997. Appearance of phosphatidylserine on apoptotic cells requires calcium-mediated nonspecific flip-flop and is enhanced by loss of the aminophospholipid translocase. J. Biol. Chem. 272:26159–65 Hogquist KA, Nett MA, Unanue ER, Chaplin DD. 1991. Interleukin 1 is processed and released during apoptosis. Proc. Natl. Acad. Sci. USA 88:8485–89 Bellone M, Iezzi G, Rovere P, Galati G, Ronchetti A, et al. 1997. Processing of engulfed apoptotic bodies yields T cell epitopes. J. Immunol. 159:5391–99

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM 24. Albert ML, Sauter B, Bhardwaj N. 1998. Dendritic cells acquire antigen from apoptotic cells and induce class Irestricted CTLs. Nature 392:86–89 25. Bicknell GR, Cohen GM. 1995. Cleavage of DNA to large kilobase pair fragments occurs in some forms of necrosis as well as apoptosis. Biochem. Biophys. Res. Comm. 207:40–47 26. Dong Z, Sikumar P, Weinberg JM, Venkatachalam MA. 1997. Internucleosomal DNA cleavage triggered by plasma membrane damage during necrotic cell death: involvement of serine but not cysteine proteases. Am. J. Pathol. 151:1205– 13 27. Didenko VV, Hornsby PJ. 1996. Presence of double-strand breaks with singlebase 30 overhangs in cells undergoing apoptosis but not necrosis. J. Cell Biol. 135:1369–76 28. Leist M, Single B, Castoldi AF, K¨uhnle S, Nicotera P. 1997. Intracellular adenosine triphosphate (ATP) concentration: a switch in the decision between apoptosis and necrosis. J. Exp. Med. 185:1481–86 29. Lieberthal W, Menza SA, Levine JS. 1998. Graded ATP depletion can cause necrosis or apoptosis of cultured mouse proximal tubular cells. Am. J. Physiol. 274:F315–27 30. Kass GEN, Eriksson JE, Weis M, Orrenius S, Chow SC. 1996. Chromatin condensation during apoptosis requires ATP. Biochem. J. 318:749–52 31. Griffith TS, Herndon JM, Green DR, Ferguson TA. 1996. CD95-induced apoptosis of lymphocytes in an immune privileged site induces immunological tolerance. Immunity 5:7–16 32. Hengartner MO, Horvitz HR. 1994. Programmed cell death in Caenorhabditis elegans. Curr. Opin. Gen. Dev. 4:581–86 33. Yuan J, Shaham S, Ledoux S, Ellis HM, Horvitz HR. 1993. The C. elegans cell death gene ced-3 encodes a protein similar to mammalian interleukin-1β-converting enzyme. Cell 75:641–52 34. Cerrati DP, Kozlosky CJ, Mosley B, Nelson N, Van Ness K, et al. 1992. Molecular cloning of the interleukin-1β converting enzyme. Science 256:97–100 35. Thornberry NA, Bull HG, Calaycay JR, Chapman KT, Howard AD, et al. 1992. A novel heterodimeric cysteine protease is required for interleukin-1β processing in monocytes. Nature 356:768–74 36. Yuan J, Horvitz HR. 1992. The Caenorhabditis elegans cell death gene ced-4 encodes a novel protein and is expressed during the period of extensive pro-

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

815

grammed cell death. Development 116: 309–20 Vaux DL, Weissman IL, Kim SK. 1992. Prevention of programmed cell death in Caenorhabditis elegans by human Bcl-2. Science 258:1955–57 Hengartner MO, Horvitz HR. 1994. C. elegans cell death gene ced-9 encodes a functional homolog of mammalian protooncogene bcl-2. Cell 76:665–76 Chinnaiyan AR, O’Rourke K, Lane BR, Dixit VM. 1997. Interaction of CED4 with CED-3 and CED-9: a molecular framework for cell death. Science 275: 1122–26 Seshagiri S, Miller LK. 1997. Caenorhabditis elegans CED-4 stimulates CED-3 processing and CED-3-induced apoptosis. Curr. Biol. 7:455–60 Chinnaiyan AR, Chaudhary D, O’Rourke K, Koonin EV, Dixit VM. 1997. Role of CED-4 in the activation of CED-3. Nature 388:728–29 Irmler M, Hofmann K, Vaux D, Tschopp J. 1997. Direct physical-interaction between the Caenorhabditis elegans ‘death proteins’ CED-3 and CED-4. FEBS Lett. 406:189–90 Spector MS, Desnoyers S, Hoeppner DJ, Hengartner MO. 1997. Interaction between the C. elegans cell-death regulators CED-9 and CED-4. Nature 385:653–56 Wu D, Wallen HD, Nu˜nez G. 1997. Interaction and regulation of subcellular localization of CED-4 by CED-9. Science 275:1126–29 Zou H, Henzel WJ, Liu X, Lutschg A, Wang X. 1997. Apaf-1, a human protein homologous to C. elegans CED-4, participates in cytochrome c-dependent activation of caspase-3. Cell 90:405–13 Miura M, Zhu H, Rotello R, Hartweig EA, Yuan J. 1993. Induction of apoptosis in fibroblasts by IL-1β-converting enzyme, a mammalian homolog of the C. elegans cell death gene ced-3. Cell 75:653–60 Kumar S, Tomooka Y, Noda M. 1992. Identification of a set of genes with developmentally down-regulated expression in the mouse brain. Biochem. Biophys. Res. Commun. 185:1155–61 Kumar S, Kinoshita M, Noda M, Copeland NG, Jenkins NA. 1994. Induction of apoptosis by the mouse Nedd2 gene, which encodes a protein similar to the product of the Caenorhabditis elegans cell death gene ced-3 and the mammalian IL-1β-converting enzyme. Genes Dev. 8:1613–26 Alnemri ES, Livingston DJ, Nicholson DW, Salvesen G, Thornberry NA,

P1: JER/mbg

P2: ARS/ary

January 22, 1999

816

50.

51.

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

52.

53.

54.

55.

56.

57.

58.

59.

60.

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON et al. 1996. Human ICE/CED-3 protease nomenclature. Cell 87:171 Walker NPC, Talanian RV, Brady KD, Dang LC, Bump NJ, et al. 1994. Crystal structure of the cysteine protease interleukin-1β-converting enzyme: a (p20/p10)2 homodimer. Cell 78:343–52 Wilson KP, Black JF, Thomson JA, Kim EE, Griffith JP, et al. 1994. Structure and mechanism of interleukin-1β converting enzyme. Nature 370:270–75 Rotonda J, Nicholson DW, Fazil KM, Gallant M, Gareau Y, et al. 1996. The three dimensional structure of apopain/CPP32, a key mediator of apoptosis. Nat. Struct. Biol. 3:619–25 Boldin MP, Goncharov TM, Goltsev YV, Wallach D. 1996. Involvement of MACH, a novel MORT1/FADD-interacting protease, in Fas/APO-1- and TNF receptorinduced cell death. Cell 85:803–15 Muzio M, Chinnaiyan AM, Kischkel FC, O’Rourke K, Shevchenko A, et al. 1996. FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-inducing signaling complex. Cell 85:817–27 Fernandes-Alnemri T, Armstrong RC, Krebs J, Srinivasula SM, Wang L, et al. 1996. In vitro activation of CPP32 and Mch3 by Mch4, a novel human apoptotic cysteine protease containing two FADDlike domains. Proc. Natl. Acad. Sci. USA 93:7464–69 Oltvai ZN, Milliman CL, Korsmeyer SJ. 1993. Bcl-2 heterodimerizes in vivo with a conserved homolog, bax, that accelerates programed cell death. Cell 74:609– 19 Kitanaka C, Namiki T, Noguchi K, Mochizuki T, Kagaya S, et al. 1997. Caspase-dependent apoptosis of COS-7 cells induced by Bax overexpression: differential effects of Bcl-2 and Bcl-xL on Bax-induced caspase activation and apoptosis. Oncogene 15:1763–72 Xiang J, Chao DT, Korsmeyer SJ. 1996. BAX-induced cell death may not require interleukin 1β-converting enzymelike proteases. Proc. Natl. Acad. Sci. USA 93:14559–63 Hirsch T, Marchetti P, Susin SA, Dallaporta B, Zamzami N, et al. 1997. The apoptosis-necrosis paradox. Apoptogenic proteases activated after mitochondrial permeability transition determine the mode of cell death. Oncogene 15:1573–81 Monney L, Otter I, Olivier R, Ozer HL, Haas AL, et al. 1998. Defects in the ubiquitin pathway induce caspase-

61.

62.

63. 64.

65. 66.

67.

68.

69.

70.

71.

72.

73.

independent apoptosis blocked by Bcl-2. J. Biol. Chem. 273:6121–31 Mills J, Stone NL, Erhardt J, Pittman RN. 1998. Apoptotic membrane blebbing is regulated by myosin light chain phosphorylation. J. Cell Biol. 140:627–36 McCarthy NJ, Whyte MKB, Gillbert CS, Evan GI. 1997. Inhibition of Ced-3/ICErelated proteases does not prevent cell death induced by oncogenes, DNA damage, or the Bcl-2 homologue Bak. J. Cell Biol. 136:215–27 Itoh N, Nagata S. 1993. A novel protein domain required for apoptosis. J. Biol. Chem. 268:10932–37 Tartaglia LA, Ayres M, Wong GH, Goeddel DV. 1993. A novel domain within the 55 kd TNF receptor signals cell death. Cell 74:845–53 Nagata S. 1997. Apoptosis by death factor. Cell 88:355–65 Chaudhary PM, Eby M, Lasmin A, Bookwalter A, Murray J, et al. 1997. Death receptor 5, a new member of the TNF family, and DR4 induce FADD-dependent apoptosis and activate the NF-κB pathway. Immunity 7:821–30 Enari M, Hug H, Nagata S. 1995. Involvement of an ICE-like protease in Fas-mediated apoptosis. Nature 375:78– 81 Los M, Van de Craen M, Penning LC, Schenk H, Westendorp M, et al. 1995. Requirement of an ICE/CED-3 protease for Fas/APO-1-mediated apoptosis. Nature 375:81–83 Longthorne VL, Williams GT. 1997. Caspase activity is required for commitment to Fas-mediated apoptosis. EMBO J. 16:3805–12 Kuida K, Zheng TS, Na S, Kuan C, Yang D, et al. 1996. Decreased apoptosis in the brain and premature lethality in CPP32deficient mice. Nature 384:368–72 Woo M, Hakem R, Soengas MS, Duncan GS, Shahinian A, et al. 1998. Essential contribution of caspase 3/CPP32 to apoptosis and its associated nuclear changes. Genes Dev. 12:806–19 Hirata H, Takahashi A, Kobayashi S, Yonehara S, Sawai H, et al. 1998. Caspases are activated in a branched protease cascade and control distinct downstream processes in Fas-induced apoptosis. J. Exp. Med. 187:587–600 Mancini M, Nicholson DW, Roy S, Thornberry NA, Peterson EP, et al. 1998. The caspase 3 precursor has a cytosolic and mitochondrial distribution: implications for apoptotic signaling. J. Cell Biol. 140:1485–95

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM 74. Hakem R, Hakem A, Duncan GS, Henderson JT, Woo M, et al. 1998. Differential requirement for caspase 9 in apoptotic pathways in vivo. Cell 94:339–52 75. Kuida K, Haydar TF, Kuan C-Y, Gu Y, Taya C, et al. 1998. Reduced apoptosis and cytochrome c-mediated caspase activation in mice lacking caspase 9. Cell 94:325–37 76. Varfolomeev EE, Schuchmann M, Luria V, Chiannilkulchai N, Beckmann JS, et al. 1998. Targeted disruption of the mouse Caspase 8 gene ablates cell death induction by the TNF receptors, Fas/Apo1, and DR3 and is lethal prenatally. Immunity 9:267–76 77. Kuida K, Lippke JA, Ku G, Harding MW, Livingston DJ, et al. 1995. Altered cytokine export and apoptosis in mice deficient in Interleukin-1β converting enzyme. Science 267:2000–3 78. Li P, Hamish A, Banerjee S, Franklin S, Herzog L, et al. 1995. Mice deficient in IL-1β-converting enzyme are defective in production of mature IL-1β and resistant to endotoxic shock. Cell 80:401–11 79. Bergeron L, Perez GI, Macdonald G, Shi L, Sun Y, et al. 1998. Defects in regulation of apoptosis in caspase-2-deficient mice. Genes Dev. 12:1304–14 80. Wang S, Miura M, Jung Y, Zhu H, Yuan J. 1998. Murine Caspase-11, an ICEinteracting protease, is essential for the activation of ICE. Cell 92:501–9 81. Wang L, Miura M, Bergeron L, Zhu H, Yuan J. 1994. Ich-1, an Ice/ced-3-related gene, encodes both positive and negative regulators of programmed cell death. Cell 78:739–50 82. Wang S, Miura M, Jung Y, Zhu H, Gagliardini V, et al. 1996. Identification and characterization of Ich-3, a member of the interleukin-1β converting enzyme (ICE)/Ced-3 family and an upstream regulator of ICE. J. Biol. Chem. 271:20580– 87 83. Ghayur T, Banerjee S, Hugunin M, Butler D, Herzong L, et al. 1997. Caspase-1 processes IFN-γ -inducing factor and regulates LPS-induced IFN-γ production. Nature 386:619–23 84. Gu Y, Kuida K, Tsutsui H, Ku G, Hsiao K, et al. 1997. Activation of Interferon-γ inducing factor mediated by Interleukin-1β converting enzyme. Science 275:206–9 85. Han Z, Hendrickson EA, Bremner TA, Wyche JH. 1997. A sequential two-step mechanism for the production of the mature p17:p12 form of Caspase-3 in vitro. J. Biol. Chem. 272:13432–36 86. Lens SMA, den Drijver BFA, P¨otgens

87.

88.

89.

90.

91.

92.

93. 94.

95.

96. 97.

98.

817

AJG, Tesselaar K, van Oers MHJ, et al. 1998. Dissection of pathways leading to antigen receptor-induced and Fas/CD95induced apoptosis in human B cells. J. Immunol. 160:6083–92 Butt AJ, Harvey NL, Parasivam G, Kumar S. 1998. Dimerization and autoprocessing of the Nedd2 (Caspase-2) precursor requires both the prodomain and the carboxyl-terminal regions. J. Biol. Chem. 273:6763–68 MacCorkle RA, Freeman KW, Spencer DM. 1998. Synthetic activation of caspases: artificial death switches. Proc. Natl. Acad. Sci. USA 95:3655–60 Martin DA, Siegel RM, Zheng L, Lenardo MJ. 1998. Membrane oligomerization and cleavage activates the caspase-8 (FLICE/MACHα1) death signal. J. Biol. Chem. 273:4345–49 Muzio M, Stockwell BR, Stennicke HR, Salvesen GS, Dixit VM. 1998. An induced proximity model for caspase-8 activation. J. Biol. Chem. 273:2926–30 Srinivasula SM, Ahmad M, FernandesAlnemri T, Alnemri ES. 1998. Autoactivation of Procaspase-9 by Apaf1-mediated oligomerization. Mol. Cell 1:949–57 Li P, Nijhawan D, Budihardjo I, Srinivasula S, Ahmad M, et al. 1997. Cytochrome c and dATP-dependent formation of Apaf-1/Caspase-9 complex initiates an apoptotic protease cascade. Cell 91:479–89 Nicholson DW, Thornberry NA. 1997. Caspases: killer proteases. TIBS 22:299– 306 Talanian RV, Quinlan C, Trautz S, Hackett MC, Mankovich JA, et al. 1997. Substrate specificities of caspase family proteases. J. Biol. Chem. 272:9677–82 Van de Craen M, Vandenabeele P, Declercq W, Van den Brande I, Van Loo G, et al. 1997. Characterization of seven murine caspase family members. FEBS Lett. 403:61–69 Humke EW, Dixit VM. 1998. ERICE, a novel FLICE-activatable caspase. J. Biol. Chem. 272:15702–7 Vincenz C, Dixit VM. 1997. Fas-associated death domain protein interleukin1beta-converting enzyme 2 (FLICE2), an ICE/Ced-3 homologue, is proximally involved in CD95- and p55-mediated death signaling. J. Biol. Chem. 272:6578–83 Casiano CA, Martin SJ, Green DR, Tan EM. 1996. Selective cleavage of nuclear autoantigens during CD95 (Fas/APO-1)mediated T cell apoptosis. J. Exp. Med. 184:765–70

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

818

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

99. Widmann C, Gibson S, Johnson GL. 1998. Caspase-dependent cleavage of signaling proteins during apoptosis: A turnoff mechanism for anti-apoptotic signals. J. Biol. Chem. 273:7141–47 100. Janicke RU, Ng P, Sprengart ML, Porter AG. 1998. Caspase-3 is required for alpha-fodrin cleavage but dispensable for cleavage of other death substrates in apoptosis. J. Biol. Chem. 273:15540–45 101. Xia Z, Dickens M, Raingeaud J, Davis RJ, Greenberg ME. 1995. Opposing effects of ERK and JNK-p38 MAP kinases on apoptosis. Science 270:1326–31 102. Deak JC, Cross JV, Lewis M, Qian Y, Parrott LA, et al. 1998. Fas-induced proteolytic activation and intracellular redistribution of the stress-signaling kinase MEKK1. Proc. Natl. Acad. Sci. USA 95:5595–600 103. Widmann C, Gerwins P, Johnson NL, Jarpe MB, Johnson GL. 1998. MEK Kinase 1, a substrate for DEVD-directed caspases is involved in genotoxin-induced apoptosis. Mol. Cell. Biol. 18:2416–29 104. Santoro MF, Annand RR, Robertson MM, Peng Y, Brady MJ, et al. 1998. Regulation of protein phosphatase 2A activity by caspase-3 during apoptosis. J. Biol. Chem. 273:13119–28 105. Holmstr¨om TH, Chow SC, Elo I, Coffey ET, Orrenius S, et al. 1998. Suppression of Fas/APO-1-mediated apoptosis by Mitogen-Activated Kinase signaling. J. Immunol. 160:2626–36 106. Zha J, Harado 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-XL. Cell 87:619–28 107. 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 91:231–41 108. Beg AA, Baltimore D. 1996. An essential role for NF-κB in preventing TNF-αinduced cell death. Science 274:782–84 109. Van Antwerp DJ, Martin SJ, Kafri T, Green DR, Verma IM. 1996. Suppression of TNF-α-induced apoptosis by NF-κB. Science 274:787–89 110. Ravi R, Bedi A, Fuchs EJ, Bedi A. 1998. CD95 (Fas)-induced caspasemediated proteolysis of NF-κB. Cancer Res. 58:882–86 111. Tan X, Martin SJ, Green DR, Wang JYJ. 1997. Degradation of retinoblastoma protein in tumor necrosis factor- and CD95-induced cell death. J. Biol. Chem. 272:9613–16 112. Gottleib E, Oren M. 1998. p53 facil-

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

itates pRb cleavage in IL-3-deprived cells: novel pro-apoptotic activity of p53. EMBO J. 17:3587–96 Rh´eaume E, Cohen LY, Uhlmann F, Lazure C, Alam A, et al. 1997. The large subunit of replication factor C is a substrate for caspase-3 in vitro and is cleaved by a caspase-3-like protease during Fasmediated apoptosis. EMBO J. 16:6346– 54 Ubeda M, Habener JF. 1997. The large subunit of the DNA Replication Complex C (DSEB/RF-C140) cleaved and inactivated by caspase 3 (CPP32/YAMA) during Fas-induced apoptosis. J. Biol. Chem. 272:19562–68 Kaufmann SH, Desnoyers S, Ottaviano Y, Davidson NE, Poirier GG. 1993. Specific proteolytic cleavage of Poly(ADPribose) Polymerase: an early marker of chemotherapy-induced apoptosis. Cancer Res. 53:3976–85 Lazebnik YA, Kaufmann SH, Desnoyers S, Poirier GG, Earnshaw WC. 1994. Cleavage of poly(ADP-ribose) polymerase by a proteinase with properties like ICE. Nature 371:346–47 Tewari M, Quan LT, O’Rourke K, Desnoyers S, Zeng Z, et al. 1995. Yama/ CPP32β, a mammalian homolog of CED3, is a CrmA-inhibitable protease that cleaves the death substrate Poly(ADPRibose) Polymerase. Cell 81:801–9 Nicholson DW, Ali A, Thornberry NA, Vaillancourt JP, Ding CK, et al. 1995. Identification and inhibition of the ICE/CED-3 protease necessary for mammalian apoptosis. Nature 376:37–43 Casciola-Rosen L, Nicholson DW, Chong T, Rowan KR, Thornberry NA, et al. 1996. Apopain/CPP32 cleaves proteins that are essential for cellular repair: a fundamental principle of apoptotic death. J. Exp. Med. 183:1957–64 Han Z, Malik N, Carter T, Reeves WH, Wyche JH, et al. 1996. DNA-dependent Protein Kinase is a target for a CPP32like apoptotic protease. J. Biol. Chem. 271:25035–40 Song Q, Lees-Miller SP, Kumar S, Zhang N, Chan DW, et al. 1996. DNA-dependent protein kinase catalytic subunit: a target for an ICE-like protease in apoptosis. EMBO J. 15:3238–46 Flygare J, Armstrong RC, Wennborg A, Orsan S, Hellgren D. 1998. Proteolytic cleavage of HsRad51 during apoptosis. FEBS Lett. 427:247–51 Zhang Y, Center DM, Wu MH, Cruikshank WW, Yuan J, et al. 1998. Processing and activation of pro-interleukin-16

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

APOPTOSIS IN THE IMMUNE SYSTEM

124.

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

125.

126.

127.

128.

129.

130.

131.

132.

133.

134.

135.

by caspase-3. J. Biol. Chem. 273:1144– 49 Waterhouse N, Kumar S, Song Q, Strike P, Sparrow L, et al. 1996. Heteronuclear ribonucleoproteins C1 and C2, components of the spliceosome, are specific targets of Interleukin 1β-converting enzyme-like proteases in apoptosis. J. Biol. Chem. 271:29335–41 Casciola-Rosen LA, Miller DK, Anhalt GJ, Rosen A. 1994. Specific cleavage of the 70-kDa protein component of the U1 small nuclear ribonucleoprotein is a characteristic biochemical feature of apoptotic cell death. J. Biol. Chem. 269:30757–60 Liu X, Zou H, Slaughter C, Wang X. 1997. DFF, a heterodimeric protein that functions downstream of caspase-3 to trigger DNA fragmentation during apoptosis. Cell 89:175–84 Enari M, Sakahira H, Yokoyama H, Okawa K, Iwamatsu A, et al. 1998. A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature 391:43–50 Sakahira H, Enari M, Nagata S. 1998. Cleavage of CAD inhibitor in CAD activation and DNA degradation during apoptosis. Nature 391:96–99 Halenbeck R, MacDonald H, Roulston A, Chen TT, Conroy L, et al. 1998. CPAN, a human nuclease regulated by the caspase-sensitive inhibitor DFF45. Curr. Biol. 8:537–40 Cheng EH, Kirsch DG, Clem RJ, Ravi R, Kastan MB, et al. 1997. Conversion of Bcl-2 to a Bax-like death effector by caspases. Science 278:1966–68 Clem RJ, Cheng EH, Karp CL, Kirsch DG, Ueno K, et al. 1998. Modulation of cell death by Bcl-XL through caspase interaction. Proc. Natl. Acad. Sci. USA 95:554–59 Faucheu C, Diu A, Chan AWE, Blanchet A, Miossec C, et al. 1995. A novel human protease similar to the interleukin1β converting enzyme induces apoptosis in transfected cells. EMBO J. 14:1914–22 Srinivasula SM, Fernandes-Alnemri T, Zangrilli J, Robertson R, Armstrong RC, et al. 1996. The Ced-3/Interleukin 1β converting enzyme-like homolog Mch6 and the Lamin-cleaving enzyme Mch2α are substrates for the apoptotic mediator CPP32. J. Biol. Chem. 271:27099–106 Orth K, O’Rourke K, Salvesen GS, Dixit VM. 1996. Molecular ordering of apoptotic mammalian CED-3/ICE-like proteases. J. Biol. Chem. 271:20977–80 Goldberg YP, Nicholson DW, Rasper DM, Kalchman MA, Koide HB, et al.

136.

137.

138.

139.

140.

141.

142.

143.

144.

145.

146.

147.

819

1996. Cleavage of hungtingtin by apopain, a proapoptotic cysteine protease, is modulated by the polyglutamine tract. Nature Gen. 13:442–49 Wellington CL, Ellerby LM, Hackam AS, Margolis RL, Trifiro MA, et al. 1998. Caspase cleavage of gene products associated with triplet expansion disorders generates truncated fragments containing the polyglutamine tract. J. Biol. Chem. 273:9158– 67 Loetscher H, Deuschle U, Brockhaus M, Reinhardt D, Nelboeck P, et al. 1997. Presenilins are processed by caspase-type proteases. J. Biol. Chem. 272:20655–59 Palumbo GJ, Pickup DJ, Fredrickson TN, McIntyre LJ, Buller RM. 1989. Inhibition of an inflammatory response is mediated by a 38-kDa protein of cowpox virus. Virology 172:262–73 Gagliardini V, Fernandez P-A, Lee RKK, Drexler HCA, Rotello RJ, et al. 1994. Prevention of vertebrate neuronal death by the crmA gene. Science 263:826–28 Tewari M, Dixit VM. 1995. Fas- and tumor necrosis factor-induced apoptosis is inhibited by the poxvirus crmA gene product. J. Biol. Chem. 270:3255–60 Ray CA, Black RA, Kronheim SR, Greenstreet TA, Sleath PR, et al. 1992. Viral inhibition of inflammation: Cowpox virus encodes an inhibitor of the Interleukin-1β converting enzyme. Cell 69:597–604 Zhou Q, Snipas S, Orth K, Muzio M, Dixit VM, et al. 1997. Target protease specificity of the viral serpin CrmA: analysis of five caspases. J. Biol. Chem. 272:7797– 800 Clem RJ, Fechheimer M, Miller LK. 1991. Prevention of apoptosis by a baculovirus gene during infection of insect cells. Science 254:1388–90 Bump NJ, Hackett M, Hugunin M, Seshagiri S, Brady K, et al. 1995. Inhibition of ICE family proteases by Baculovirus antiapoptotic protein p35. Science 269:1885– 88 Xue D, Horvitz HR. 1995. Inhibition of the Caenorhabditis elegans cell-death protease CED-3 by a CED-3 cleavage site in baculovirus p35 protein. Nature 377:248–51 Komiyama T, Ray CA, Pickup DJ, Howard AD, Thornberry NA, et al. 1994. Inhibition of Interleukin-1β converting enzyme by the Cowpox virus serpin CrmA. J. Biol. Chem. 169:19331–37 Clem RJ, Miller LK. 1994. Control of programmed cell death by the Baculovirus genes p35 and iap. Mol. Cell. Biol. 14: 5212–22

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

820

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

148. Crook NE, Clem RJ, Miller LK. 1993. An apoptosis-inhibiting baculovirus gene with a zinc finger-like motif. J. Virol. 67:2168–74 149. Liston P, Roy N, Tamai K, Lefebvre C, Baird S, et al. 1996. Suppression of apoptosis in mammalian cells by NAIP and a related family of IAP genes. Nature 379:349–53 150. Rothe M, Pan M-G, Henzel WJ, Ayres TM, Goeddel DV. 1995. The TNFR2TRAF signaling complex contains two novel proteins related to baculovirus inhibitor of apoptosis proteins. Cell 83:1243–53 151. Uren AG, Pakusch M, Hawkins CJ, Puls KL, Vaux DL. 1996. Cloning and expression of apoptosis inhibitory protein homologs that function to inhibit apoptosis and/or bind tumor necrosis factor receptor-associated factors. Proc. Natl. Acad. Sci. USA 93:4974–78 152. Duckett CS, Nava VE, Gedrich RW, Clem RJ, Van Dongen JL, et al. 1996. A conserved family of cellular genes related to the baculovirus iap gene and encoding apoptosis inhibitors. EMBO J. 15:2685– 94 153. Deveraux QL, Takahashi R, Salvesen GS, Reed JC. 1997. X-linked IAP is a direct inhibitor of cell-death proteases. Nature 388:300–4 154. Roy N, Deveraux QL, Takahashi R, Salvesen GS, Reed JC. 1997. The c-IAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J. 16:6914– 25 155. Deveraux QL, Roy N, Stennicke HR, Van Arsdale T, Zhou Q, et al. 1998. IAP’s block apoptotic events by caspase-8 and cytochrome c by direct inhibition of distinct caspases. EMBO J. 17:2215–23 156. Takahashi R, Deveraux Q, Tamm I, Welsh K, Assa-Munt N, et al. 1998. A single BIR domain of XIAP sufficient for inhibiting caspases. J. Biol. Chem. 273:7787–90 157. Sanna MG, Duckett CS, Richter BWM, Thompson CB, Ulevitch RJ. 1998. Selective activation of JNK1 is necessary for the anti-apoptotic activity of hILP. Proc. Natl. Acad. Sci. USA 95:6015–20 158. Chu Z-L, McKinsey TA, Liu L, Gentry JJ, Malim MH, et al. 1997. Suppression of tumor necrosis factor-induced cell death by inhibitor or apoptosis c-IAP2 is under NF-κB control. Proc. Natl. Acad. Sci. USA 94:10057–62 159. Bertin J, Armstrong RC, Ottilie S, Martin DA, Wang Y, et al. 1997. Death effector domain-containing herpesvirus and poxvirus proteins inhibit both Fas- and

160.

161.

162. 163.

164.

165.

166.

167.

168.

169.

170.

171.

TNFR1-induced apoptosis. Proc. Natl. Acad. Sci. USA 94:1172–76 Thome M, Schneider P, Hofmann K, Fickenscher H, Meinl E, et al. 1997. Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature 386:517–21 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 Shu H, Halpin DR, Goeddel DV. 1997. Casper is a FADD- and caspase-related inducer of apoptosis. Immunity 6:751–63 Hu S, Vincenz C, Ni J, Gentz R, Dixit VM. 1997. I-FLICE, a novel inhibitor of tumor necrosis factor receptor-1 and CD-95-induced apoptosis. J. Biol. Chem. 272:17255–57 Srinivasula SM, Ahmad M, Ottilie S, Bullrich F, Banks S, et al. 1997. FLAME1, a novel FADD-like anti-apoptotic molecule that regulates Fas/TNFR1induced apoptosis. J. Biol. Chem. 272: 18542–45 Goltsev YV, Kovalenko AV, Arnold E, Varfolomeev EE, Brodianskii VM, et al. 1997. CASH, a novel caspase homologue with death effector domains. J. Biol. Chem. 272:19641–44 Inohara N, Koseki T, Hu Y, Chen S, N´un˜ ez G. 1997. CLARP, a death effector domain-containing protein, interacts with caspase-8 and regulates apoptosis. Proc. Natl. Acad. Sci. USA 94:10717–22 Han DKM, Chaudhary PM, Wright ME, Friedman C, Trask BJ, et al. 1997. MRIT, a novel death-effector domaincontaining protein, interacts with caspases and BclXL and initiates cell death. Proc. Natl. Acad. Sci. USA 94:11333–38 Rasper DM, Vaillancourt JP, Hadano S, Houtzager VM, Seiden I, et al. 1998. Cell death attenuation by ‘Usurpin’, a mammalian DED-caspase homologue that precludes caspase-8 recruitment and activation by the CD-95 (Fas, Apo-1) receptor complex. Cell Death Differ. 5:271–88 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 Liu X, Naekyung K, Yang J, Jemmerson R, Wang X. 1996. Induction of apoptotic program in cell-free extracts: requirement for dATP and cytochrome c. Cell 86:147–57 Hu Y, Benedict MA, Wu D, Inohara N, N´un˜ ez G. 1998. Bcl-xL interacts with Apaf-1 and inhibits Apaf-1-dependent

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

APOPTOSIS IN THE IMMUNE SYSTEM

172.

173.

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

174.

175. 176.

177.

178.

179.

180.

181.

182.

183.

184.

caspase-9 activation. Proc. Natl. Acad. Sci. USA 95:4386–91 Pan G, O’Rourke K, Dixit VM. 1998. Caspase-9, Bcl-XL, and Apaf-1 form a ternary complex. J. Biol. Chem. 273: 5841–45 Huang DCS, Adams JM, Cory S. 1998. The conserved N-terminal BH4 domain of Bcl-2 homologues is essential for inhibition of apoptosis and interaction with CED-4. EMBO J. 17:1029–39 Hofmann K, Bucher P, Tschopp J. 1997. The CARD domain: a new apoptotic signalling motif. Trends Biochem. Sci. 22:155–56 Duan H, Dixit VM. 1997. RAIDD is a new ‘death’ adaptor molecule. Nature 385:86– 89 Ahmad M, Srinivasula SM, Wang L, Talanian RV, Litwack G, et al. 1997. CRADD, a novel human apoptotic adaptor molecule for caspase-2, and FasL/tumor necrosis factor receptorinteracting protein RIP. Cancer Res. 57: 615–19 Inohara N, del Peso L, Koseki T, Chen S, N´un˜ ez G. 1998. RICK, a novel protein kinase containing a caspase recruitment domain, interacts with CLARP and regulates CD95-mediated apoptosis. J. Biol. Chem. 273:12296–300 McCarthy JV, Ni J, Dixit VM. 1998. RIP2 is a novel NF-kappaB-activating and cell death inducing kinase. J. Biol. Chem. 273:16968–75 Thome M, Hofmann K, Burns K, Martino F, Bodmer J-L, et al. 1998. Identification of CARDIAK, a RIP-like kinase that associates with caspase-1. Curr. Biol. 8:885–88 Koseki T, Inohara N, Chen S, Nunez G. 1998. ARC, an inhibitor of apoptosis expressed in skeletal muscle and heart that interacts selectively with caspases. Proc. Natl. Acad. Sci. USA 95:5156–60 Neer EJ, Schmidt CJ, Nambudripad R, Smith TF. 1994. The ancient regulatoryprotein family of WD-repeat proteins. Nature 371:297–300 van der Biezen EA, Jones JDG. 1997. The NB-ARC domain: a novel signalling motif shared by plant resistance gene products and regulators of cell death in animals. Curr. Biol. 8:R226–27 Parker JE, Coleman MJ. 1997. Molecular intimacy between proteins specifying plant-pathogen recognition. Trends Biochem. Sci. 22:291–96 Camerini D, Walz G, Loenen WAM, Borst J, Seed B. 1991. The T cell activation antigen CD27 is a member of

185.

186.

187.

188.

189. 190.

191.

192.

193.

194.

195.

196.

821

the nerve growth factor/tumor necrosis factor receptor gene family. J. Immunol. 147:3165–69 D¨urkop H, Latza U, Hummel M, Eitelbach F, Seed B, et al. 1992. Molecular cloning and expression of a new member of the nerve growth factor receptor family that is characteristic for Hodgkin’s disease. Cell 68:421–27 Stamenkovic I, Clark EA, Seed B. 1989. A B-lymphocyte activation molecule related to the nerve growth factor receptor and induced by cytokines in carcinomas. EMBO J. 8:1403–10 Smith CA, Davis T, Anderson D, Solam L, Beckmann MP, et al. 1990. A receptor for tumor necrosis factor defines an unusual family of cellular and viral proteins. Science 248:1019–23 Mallett S, Fossum S, Barclay AN. 1990. Characterization of the MRC OX40 antigen of activated CD4 positive T lymphocytes—a molecule related to nerve growth factor receptor. EMBO J. 9:1063– 68 Kwon BS, Weissman SM. 1989. cDNA sequences of two inducible T-cell genes. Proc. Natl. Acad. Sci. USA 86:1963–67 Johnson D, Lanahan A, Buck CR, Sehgal A, Morgan C, et al. 1986. Expression and structure of the human NGF receptor. Cell 47:545–54 Itoh N, Yonehara S, Ishii A, Sameshima M, Hase A, et al. 1991. The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis. Cell 66:233–43 Oehm A, Behrmann I, Falk W, Pawlita M, Maier G, et al. 1992. Purification and molecular cloning of the APO-1 cell surface antigen, a member of the tumor necrosis factor/nerve growth factor receptor superfamily: sequence identity with the Fas antigen. J. Biol. Chem. 267:10709–15 Loetscher H, Pan Y-CE, Lahm H-W, Gentz R, Brockhaus M, et al. 1990. Molecular cloning and expression of the human 55 kd tumor necrosis factor receptor. Cell 61:351–59 Schall TJ, Lewis M, Koller KJ, Lee A, Rice GC, et al. 1990. Molecular cloning and expression of a receptor for human tumor necrosis factor. Cell 61:361–70 Chinnaiyan AM, O’Rourke K, Yu G-L, Lyons RH, Garg M, et al. 1996. Signal transduction by DR-3, a death domaincontaining receptor related to TNFR-1 and CD95. Science 274:990–92 Kiston J, Raven T, Jiang Y-P, Goeddel DV, Giles KM, et al. 1996. A death-domain-

P1: JER/mbg

P2: ARS/ary

January 22, 1999

822

197.

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

198.

199.

200.

201.

202.

203.

204.

205.

206.

207.

208.

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON containing receptor that mediates apoptosis. Nature 384:372–75 Matsuzawa A, Moriyama T, Kaneko T, Tanaka M, Kimura M, et al. 1990. A new allele of the lpr locus, lpr cg, that complements the gld gene in induction of lymphadenopathy in the mouse. J. Exp. Med. 171:519–31 Watanabe-Fukunaga R, Brannan CI, Copeland NG, Jenkins NA, Nagata S. 1992. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356:314– 17 Huang B, Eberstadt M, Olejniczak ET, Meadows RP, Fesik SW. 1996. NRM structure and mutagenesis of the Fas (APO-1/CD95) death domain. Nature 384:638–41 Chinnaiyan AM, O’Rourke K, Tewari M, Dixit VM. 1995. FADD, a novel death domain-containing protein, interacts with the death domain of Fas and initiates apoptosis. Cell 81:505–12 Boldin MP, Varfolomeev EE, Pancer Z, Mett IL, Camonis JH, et al. 1995. A novel protein that interacts with the death domain of Fas/APO1 contains a sequence motif related to the death domain. J. Biol. Chem. 270:7795–98 Hsu H, Xiong J, Goeddel DV. 1995. The TNF Receptor 1-associated protein TRADD signals cell death and NF-κB activation. Cell 81:495–504 Hsu H, Shu H, Pan M, Goeddel DV. 1996. TRADD-TRAF2 and TRADD-FADD interactions define two distinct TNF receptor 1 signal transduction pathways. Cell 84:299–308 Rothe M, Sarma V, Dixit VM, Goeddel DV. 1995. TRAF2-mediated activation of NF-κB by TNF receptor 2 and CD40. Science 269:1424–27 Liu Z, Hsu H, Goeddel DV, Karin M. 1996. Dissection of TNF receptor 1 effector functions: JNK activation is not linked to apoptosis while NF-κB activation prevents cell death. Cell 87:565–76 Hsu H, Huang J, Shu HB, Baichwal V, Goeddel DV. 1996. TNF-dependent recruitment of the protein kinase RIP to the TNF receptor-1 signaling complex. Immunity 4:387–96 Stanger BZ, Leder P, Lee T, Kim E, Seed B. 1995. RIP: a novel protein containing a death domain that interacts with Fas/Apo1 (CD95) in yeast and causes cell death. Cell 81:513–23 Ting AT, Pimentel-Muinos FX, Seed B. 1996. RIP mediates tumor necrosis factor receptor 1 activation of NF-kappaB

209.

210.

211.

212.

213.

214.

215.

216.

217.

218.

219. 220.

221.

but not Fas/APO-1-initiated apoptosis. EMBO J. 15:6189–96 Kelliher MA, Grimm S, Ishida Y, Kuo F, Stanger BZ, et al. 1998. The death kinase RIP mediates the TNF-induced NFkappaB signal. Immunity 8:297–303 Sato T, Irie S, Kitada S, Reed JC. 1995. FAP-1: a protein tyrosine phosphatase that associates with Fas. Science 268:411–15 Zhang X, Brunner T, Carter L, Dutton RW, Rogers P, et al. 1997. Unequal death in T helper cell (Th)1 and Th2 effectors: Th1, but not Th2, effectors undergo rapid Fas/FasL-mediated apoptosis. J. Exp. Med. 185:1837–49 Chu K, Niu X, Williams LT. 1995. A Fasassociated protein factor, FAF1, potentiates Fas-mediated apoptosis. Proc. Natl. Acad. Sci. USA 97:11894–98 Yang X, Khosravi-Far R, Chang HY, Baltimore D. 1997. Daxx, a novel Fasbinding protein that activates JNK and apoptosis. Cell 89:1067–76 Bakhshi A, Jensen JP, Goldman P, Wright JJ, McBride W, et al. 1985. Cloning the chromosomal breakpoint of t(14;18) human lymphomas: clustering around JH on chromosome 14 and near a transcriptional unit on 18. Cell 41:899–906 Cleary ML, Sklar J. 1985. Nucleotide sequence of a t(14;18) chromosomal breakpoint in follicular lymphoma and demonstration of a breakpoint-cluster region near a transcriptionally active locus on chromosome 18. Proc. Natl. Acad. Sci. USA 82:7439–43 Tsujimoto Y, Gorham J, Cossman J, Jaffe E, Croce CM. 1985. The t(14;18) chromosome translocations involved in B-cell neoplasms result from mistakes in VDJ joining. Science 229:1390–93 Cleary ML, Smith SD, Sklar J. 1986. Cloning and structural analysis of cDNAs for bcl-2 and a hybrid bcl-2/immunoglobulin transcript resulting from the t(14;18) translocation. Cell 47:19–28 Vaux DL, Cory S, Adams JM. 1988. Bcl-2 gene promotes haemopoietic cell survival and cooperates with c-myc to immortalize pre-B cells. Nature 335:440–42 Yang E, Korsmeyer SJ. 1996. Molecular thanatopsis: a discourse on the BCL2 family and cell death. Blood 88:386–401 Boise LH, Thompson CB. 1997. Bcl-xL can inhibit apoptosis in cells that have undergone Fas-induced protease activation. Proc. Natl. Acad. Sci. USA 94:3759–64 Gibson L, Holmgreen SP, Huang DCS, Bernard O, Copeland NG, et al. 1996. bcl-w, a novel member of the bcl-2 family,

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

APOPTOSIS IN THE IMMUNE SYSTEM promotes survival. Oncogene 13:665–75 222. Kozopas KM, Yang T, Buchan HL, Zhou P, Craig RW. 1993. MCL1, a gene expressed in programmed myeloid cell differentiation, has sequence similarity to BCL2. Proc. Natl. Acad. Sci. USA 90:3516–20 223. Zhou P, Qian L, Kozopas KM, Craig RW. 1997. Mcl-1, a Bcl-2 family member, delays the death of hematopoietic cells under a variety of apoptosis-inducing conditions. Blood 89:630–43 224. Lin EY, Orlofsky A, Berger MS, Prystowsky MB. 1993. Characterization of A1, a novel hemopoietic-specific earlyresponse gene with sequence similarity to bcl-2. J. Immunol. 151:1979–88 225. Choi SS, Park IC, Yun JW, Sung YC, Hong SI, et al. 1995. A novel Bcl-2 related gene, Bfl-1, is overexpressed in stomach cancer and preferentially expressed in bone marrow. Oncogene 11:1693–98 226. Henderson S, Huen D, Rowe M, Dawson C, Johnson G, et al. 1993. Epstein-Barr virus-coded BHRF1 protein, a viral homologue of Bcl-2, protects human B cells from programmed cell death. Proc. Natl. Acad. Sci. USA 90:8479–83 227. Rao L, Debbas M, Sabbatini P, Hockenberry D, Korsmeyer S, et al. 1992. The adenovirus E1A proteins induce apoptosis, which is inhibited by the E1B 19-kDa and Bcl-2 proteins. Proc. Natl. Acad. Sci. USA 89:7742–46 228. Boyd J, Malstrom S, Subramanian T, Venkatesch L, Schaeper U, et al. 1994. Adenovirus E1B 19 kDa and Bcl-2 proteins interact with a common set of cellular proteins. Cell 79:341–51 229. Chiou SK, Tseng CC, Rao L, White E. 1994. Functional complementation of the adenovirus E1B 19-kilodalton protein with Bcl-2 in the inhibition of apoptosis in infected cells. J. Virol. 68:6553–66 230. Farrow SN, White JHM, Martinou I, Raven T, Pun K-T, et al. 1995. Cloning of a bcl-2 homologue by interaction with adenovirus E1B 19K. Nature 374:731–33 231. Chittenden T, Harrington EA, O’Connor R, Flemington C, Lutz RJ, et al. 1995. Induction of apoptosis by the Bcl-2 homologue Bak. Nature 374:733–36 232. Kiefer MC, Brauer MJ, Powers VC, Wu JJ, Umansky SR, et al. 1995. Modulation of apoptosis by the widely distributed Bcl2 homologue Bak. Nature 374:736–39 233. Hsu S, Kaipia A, McGee E, Lomeli M, Hsueh AJW. 1997. Bok is a pro-apoptotic Bcl-2 protein with restricted expression in reproductive tissues and heterodimerizes with selective anti-apoptotic Bcl-2

234.

235.

236.

237.

238.

239.

240.

241.

242.

243.

244.

823

family members. Proc. Natl. Acad. Sci. USA 94:12401–6 Inohara N, Ekhterae D, Garcia I, Carrio R, Merino J, et al. 1998. Mtd, a novel Bcl-2 family member activates apoptosis in the absence of heterodimerization with Bcl-2 and Bcl-XL. J. Biol. Chem. 273:8705–10 Yang E, Zha J, Jockel J, Boise LH, Thompson CB, et al. 1995. Bad, a heterodimeric partner for Bcl-xL and Bcl-2, displaces Bax, and promotes cell death. Cell 80:285–91 Han J, Sabbatini P, White E. 1996. Induction of apoptosis by human Nbk/Bik, a BH3-containing protein that interacts with E1B 19K. Mol. Cell. Biol. 16:5857– 64 Boyd JM, Gallo GJ, Elangovan B, Houghton AB, Malstrom S, et al. 1995. Bik, a novel death-inducing protein shares a distinct sequence motif with Bcl-2 family proteins and interacts with viral and cellular survival-promoting proteins. Oncogene 11:1921–28 Wang K, Yin X-M, Chao DT, Milliman CL, Korsmeyer SJ. 1996. BID: a novel BH3 domain-only death agonist. Genes Dev. 10:2859–69 O’Connor L, Strasser A, O’Reilly LA, Hausmann G, Adams JM, et al. 1998. Bim: a novel member of the Bcl-2 family that promotes apoptosis. EMBO J. 17:384–95 Inohara N, Ding L, Chen S, N´un˜ ez G. 1997. Harakiri, a novel regulator of cell death, encodes a protein that activates apoptosis and interacts selectively with the survival-promoting proteins Bcl-2 and Bcl-XL. EMBO J. 16:1686–94 Hegde R, Srinivasula SM, Ahmad M, Fernandes-Alnemri T, Alnemri ES. 1998. Blk, a BH3-containing mouse protein that interacts with Bcl-2 and Bcl-xL, is a potent death agonist. J. Biol. Chem. 273:7783–86 Chen G, Ray R, Dubik D, Shi L, Cizeau J, et al. 1997. The E1B 19K.Bcl-2-binding protein Nip3 is a dimeric mitochondrial protein that activates apoptosis. J. Exp. Med. 186:1975–83 Yasuda M, Theodorakis P, Subramanian T, Chinnadurai G. 1998. Adenovirus E1B-19K/BCL-2 interacting protein BNIP3 contains a BH3 domain and a mitochondrial targeting sequence. J. Biol. Chem. 273:12415–21 Muchmore SW, Sattler M, Liang H, Meadows RP, Harlan JE, et al. 1996. Xray and NMR structure of human Bcl-xL, an inhibitor of programmed cell death. Nature 381:335–41

P1: JER/mbg

P2: ARS/ary

January 22, 1999

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

824

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON

245. Kelekar A, Thompson CB. 1998. Bcl2 family proteins: the role of the BH3 domain in apoptosis. Trends Cell Biol. 8:324–30 246. Krajewski S, Tanaka S, Takayama S, Schibler MJ, Fenton W, et al. 1993. Investigation of the subcellular distribution of the bcl-2 oncoprotein: residence in the nuclear envelope, endoplasmic reticulum, and outer mitochondrial membranes. Cancer Res. 53:4701–14 247. Nguyen M, Millar DG, Yong VW, Korsmeyer SJ, Shore GC. 1993. Targeting of Bcl-2 to the mitochondrial outer membrane by a COOH-terminal signal anchor sequence. J. Biol. Chem. 268:25265–68 248. Akao Y, Otsuki Y, Kataoka S, Ito Y, Tsujimoto Y. 1994. Multiple subcellular localization of bcl-2: detection in nuclear outer membrane, endoplasmic reticulum membrane, and mitochondrial membranes. Cancer Res. 54:2468–71 249. Hsu Y-T, Wolter KG, Youle RJ. 1997. Cytosol-to-membrane redistribution of Bax and Bcl-XL during apoptosis. Proc. Natl. Acad. Sci. USA 94:3669–72 250. Yin XM, Oltvai ZN, Korsmeyer SJ. 1994. BH1 and BH2 domains of Bcl-2 are required for inhibition of apoptosis and heterodimerization with Bax. Nature 369:321–23 251. Sattler M, Liang H, Nettesheim D, Meadows RP, Harlan JE, et al. 1997. Structure of Bcl-xL/Bak peptide complex reveals how regulators of apoptosis interact. Science 275:983–86 252. Simonen M, Keller H, Heim J. 1997. The BH3 domain of Bax is sufficient for interaction of Bax with itself and with other family members and it is required for induction of apoptosis. Eur. J. Biochem. 249:85–91 253. Zha H, Aime-Sempe C, Sato T, Reed JC. 1996. Proapoptotic protein Bax heterodimerizes with Bcl-2 and homodimerizes with Bax via a novel domain (BH3) distinct from BH1 and BH2. J. Biol. Chem. 271:7440–44 254. Oltvai Z, Korsmeyer SJ. 1994. Checkpoints of dueling dimers foil death wishes. Cell 79:189–92 255. Sedlak TW, Oltvai ZN, Wang K, Boise LH, Thompson CB, et al. 1995. Multiple Bcl-2 family members demonstrate selective dimerizations with Bax. Proc. Natl. Acad. Sci. USA 92:7834–38 256. Kelekar A, Chang BS, Harlan JE, Fesik SW, Thompson CB. 1997. Bad is a BH3 domain-containing protein that forms an inactivating dimer with Bcl-xL. Mol. Cell. Biol. 17:7040–46

257. Zha J, Harada H, Osipov K, Jockel J, Waksman G, et al. 1997. BH3 domain of BAD is required for heterodimerization with Bcl-XL and pro-apoptotic activity. J. Biol. Chem. 272:24101–4 258. Chang BS, Minn AJ, Muchmore SW, Fesik SW, Thompson CB. 1997. Identification of a novel regulatory domain in BclxL and Bcl-2. EMBO J. 16:968–77 259. Haldar S, Jena N, Croce CM. 1995. Inactivation of Bcl-2 by phosphorylation. Proc. Natl. Acad. Sci. USA 92: 260. Wang HG, Miyashita T, Takayama S, Sato T, Torigoe T, et al. 1994. Apoptosis regulation by interaction of Bcl-2 protein and Raf-1 kinase. Oncogene 9:2751– 56 261. Wang H-G, Rapp UR, Reed JC. 1996. Bcl2 targets the protein kinase Raf-1 to mitochondria. Cell 87:629–38 262. Blagosklonny MV, Schulte T, Nguyen P, Trepel J, Neckers LM. 1996. Taxolinduced apoptosis and phosphorylation of Bcl-2 protein involves c-Raf-1 and represents a novel c-Raf-1 signal transduction pathway. Cancer Res. 56:1851–54 263. Maundrell K, Antonsson B, Magnenat E, Camps M, Muda M, et al. 1997. Bcl-2 undergoes phosphorylation by c-Jun Nterminal kinase/stress-activated protein kinases in the presence of the constitutively active GTP-binding protein Rac1. J. Biol. Chem. 272:25238–42 264. Srivastava RK, Srivastava AR, Korsmeyer SJ, Nesterova M, Cho-Chung YS, et al. 1998. Involvement of microtubules in the regulation of Bcl2 phosphorylation and apoptosis through cyclic AMP-dependent protein kinase. Mol. Cell. Biol. 18:3509– 17 265. Chao DT, Linette GP, Boise LH, White LS, Thompson CB, et al. 1995. Bcl-xL and Bcl-2 repress a common pathway of cell death. J. Exp. Med. 182:821–28 266. Minn AJ, V´elez P, Schendel SL, Liang H, Muchmore SW, et al. 1997. Bcl-xL forms an ion channel in synthetic lipid membranes. Nature 385:353–57 267. Antonsson B, Conti F, Ciavatta A, Montessuit S, Lewis S, et al. 1997. Inhibition of Bax channel-forming activity by Bcl-2. Science 277:370–72 268. Schendel SL, Xie Z, Oblatt Montal M, Matsuyama S, Montal M, et al. 1997. Channel formation by antiapoptotic protein Bcl-2. Proc. Natl. Acad. Sci. USA 94:5113–18 269. Schlesinger PH, Gross A, Yin X-M, Yamamoto K, Saito M, et al. 1997. Comparison of the ion channel characteristics of proapoptotic BAX and antiapoptotic

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

APOPTOSIS IN THE IMMUNE SYSTEM

270.

271.

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

272.

273.

274.

275.

276.

277.

278.

279.

280.

281.

282.

BCL-2. Proc. Natl. Acad. Sci. USA 94: 11357–62 Lam M, Bhat MB, N´un˜ ez G, Ma J, Distelhorst CW. 1998. Regulation of Bclxl channel activity by calcium. J. Biol. Chem. 273:17307–10 Pastorino JG, Chen S-T, Tafani M, Snyder JW, Farber JL. 1998. The overexpression of Bax produces cell death upon induction of the mitochondrial permeability transition. J. Biol. Chem. 273:7770–75 Jurgensmeier JM, Deveraux Q, Ellerby L, Bredesen D, Reed JC. 1998. Bax directly induces release of cytochrome c from isolated mitochondria. Proc. Natl. Acad. Sci. USA 95:4997–5002 Garland JM, Halestrap A. 1997. Energy metabolism during apoptosis: Bcl-2 promotes survival in hematopoietic cells induced to apoptose by growth factor withdrawal by stabilizing a form of metabolic arrest. J. Biol. Chem. 272:4680–88 Vander Heiden MG, Chandel NS, Schumacker PT, Thompson CB. 1998. Bcl-xL prevents cell death following growth factor withdrawal by facilitating mitochondrial ATP/ADP exchange. Mol. Cell. In press Xie Z, Schendel S, Matusuyama S, Reed JC. 1998. Acidic pH promotes dimerization of Bcl-2 family proteins. Biochemistry 37:6410–18 Scaffidi C, Fulda S, Srinivasan A, Friesen C, Li F, et al. 1998. Two CD95 (APO-1/ Fas) signaling pathways. EMBO J. 17: 1675–87 Adachi S, Cross AR, Babior BM, Gottlieb RA. 1997. Bcl-2 and the outer mitochondrial membrane in the inactivation of cytochrome c during Fas-mediated apoptosis. J. Biol. Chem. 272:21878–82 Li H, Zhu H, Xu C, Yuan J. 1998. Cleavage of BID by caspase 8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cell 94:491–501 Luo X, Budihardjo I, Zou H, Slaughter C, Wang X. 1998. Bid, a Bcl2 interacting protein, mediates cytochrome c release from mitochondria in response to activation of cell surface death receptors. Cell 94:481–90 Zhivotovsky B, Orrenius S, Brustugun OT, Doskeland SO. 1998. Injected cytochrome c induces apoptosis. Nature 391:449–50 Ross´e T, Olivier R, Monney L, Rager M, Conus S, et al. 1998. Bcl-2 prolongs cell survival after Bax-induced release of cytochrome c. Nature 391:496–99 Kharbanda S, Pandey P, Schofield L, Israels S, Roncinske R, et al. 1997. Role

283. 284. 285. 286.

287.

288.

289.

290.

291.

292.

293.

294.

295.

825

for Bcl-xL as an inhibitor of cytosolic cytochrome c accumulation in DNA damage-induced apoptosis. Proc. Natl. Acad. Sci. USA 94:6939–42 Raff MC. 1992. Social controls on cell survival and death. Nature 356:397–400 Sprent J. 1973. Circulating T and B lymphocytes of the mouse. I. Migratory properties. Cell. Immunol. 7:10–39 Sprent J, Basten A. 1973. Circulating T and B lymphocytes of the mouse. II. Lifespan. Cell. Immunol. 7:40–59 Sprent J, Schaefer M, Hurd M, Surh CD, Ron Y. 1991. Mature murine B and T cells transferred to SCID mice can survive indefinitely and many maintain a virgin phenotype. J. Exp. Med. 174:717–28 Cyster JG, Hartley SB, Goodnow CC. 1994. Competition for follicular niches excludes self-reactive cells from the recirculating B-cell repertoire. Nature 371:389–95 Cyster JG, Goodnow CC. 1995. Antigeninduced exclusion from follicles and anergy are separate and complementary processes that influence peripheral B cell fate. Immunity 3:691–701 Fulcher DA, Lyons AB, Korn SL, Cook MC, Koleda C, et al. 1996. The fate of self-reactive B cells depends primarily on the degree of antigen receptor engagement and availability of T cell help. J. Exp. Med. 183:2313–28 Lam K-P, K¨uhn R, Rajewsky K. 1997. In vivo ablation of surface immunoglobulin on mature B cells by inducible gene targeting results in rapid cell death. Cell 90:1073–83 Markiewicz MA, Girao C, Opferman JT, Sun J, Hu Q, et al. 1998. Long-term T cell memory requires the surface expression of self-peptide/major histocompatibility complex molecules. Proc. Natl. Acad. Sci. USA 95:3065–70 Tanchot C, Lemonnier FA, P´erarnau B, Freitas AA, Rocha B. 1997. Differential requirements for survival and proliferation of CD8 na¨ıve or memory T cells. Science 276:2057–61 Sentman CL, Shutter JR, Hockenbery D, Kanagawa O, Korsmeyer SJ. 1991. Bcl2 inhibits multiple forms of apoptosis but not negative selection in thymocytes. Cell 67:879–88 Yeh TM, Korsmeyer SJ, Teale JM. 1991. Skewed B cell VH family repertoire in Bcl-2-Ig transgenic mice. Int. Immunol. 3:1329–33 Tao W, Teh S-J, Melhado I, Jirik F, Korsmeyer SJ, et al. 1994. The T cell receptor repertoire of CD4−8+ thymocytes

P1: JER/mbg

P2: ARS/ary

January 22, 1999

826

296.

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

297.

298.

299.

300.

301.

302. 303.

304.

305.

306.

307.

308.

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON is altered by overexpression of the BCL2 protooncogene in the thymus. J. Exp. Med. 179:145–53 Fang W, Mueller DL, Pennell CA, Rivard JJ, Li YS, et al. 1996. Frequent aberrent immunoglobulin gene rearrangements in pro-B cells revealed by a bcl-xL transgene. Immunity 4:291–99 Strasser A, Harris AW, von Boehmer H, Cory S. 1994. Positive and negative selection of T cells in T-cell receptor transgenic mice expressing a bcl-2 transgene. Proc. Natl. Acad. Sci. USA 91:1376–80 N´un˜ ez G, Hockenbery D, McDonnell TJ, Sorensen CM, Korsmeyer SJ. 1991. Bcl-2 maintains B cell memory. Nature 353:71– 73 Strasser A, Whittingham S, Vaux DL, Bath ML, Adams JM, et al. 1991. Enforced BCL2 expression in B-lymphoid cells prolongs antibody responses and elicits autoimmune disease. Proc. Natl. Acad. Sci. USA 88:8661–65 Fulcher DA, Basten A. 1997. Influences on the lifespan of B cell subpopulations defined by different phenotypes. Eur. J. Immunol. 27:1188–99 Cohen PL, Eisenberg RA. 1991. Lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu. Rev. Immunol. 9:243–69 Nagata S, Golstein P. 1995. The Fas death factor. Science 267:1449–56 Rathmell JC, Cooke MP, Ho WY, Grein J, Townsend SE, et al. 1995. CD95 (Fas)dependent elimination of self-reactive B cells upon interaction with CD4+ T cells. Nature 376:181–84 Rathmell JC, Townsend SE, Xu JC, Flavell RA, Goodnow CC. 1996. Expansion or elimination of B cells in vivo: dual roles for CD40- and Fas (CD95)-ligands modulated by the B cell antigen receptor. Cell 87:319–29 Fisher G, Rosenberg F, Straus S, Dale J, Middelton L, et al. 1995. Dominant interfering Fas gene mutations impair apoptosis in a human autoimmune lymphoproliferative syndrome. Cell 81:935–46 Rieux-Laucat F, Le Deist F, Hivroz C, Roberts A, Debatin K, et al. 1995. Mutations in Fas associated with human lymphoproliferative syndrome and autoimmunity. Science 268:1347–49 Sarin A, Wu M-L, Henkart PA. 1996. Different interleukin-1β converting enzyme (ICE) family protease requirements for the apoptotic death of T lymphocytes triggered by diverse stimuli. J. Exp. Med. 184:2445–50 Alam A, Braun MY, Hartgers F, Lesage S,

309.

310.

311.

312.

313.

314.

315.

316.

317.

318.

319.

Cohen L, et al. 1997. Specific activation of the cysteine protease CPP32 during negative selection of T cells in the thymus. J. Exp. Med. 186:1503–12 Clayton LK, Ghendler Y, Mizoguchi E, Patch RJ, Ocain TD, et al. 1997. T-cell receptor ligation by peptide/MHC induces activation of a caspase in immature thymocytes: the molecular basis of negative selection. EMBO J. 16:2282–93 Andjelic S, Liou H-C. 1998. Antigen receptor-induced B lymphocyte apoptosis mediated via a protease of the caspase family. Eur. J. Immunol. 28:570–81 Graves JD, Draves KE, Craxton A, Krebs EG, Clark EA. 1998. A comparison of signaling requirements for apoptosis of human B lymphocytes induced by the B cell receptor and CD95/Fas. J. Immunol. 161:168–74 Rickers A, Brockstedt E, Mapara MY, Otto A, D¨orken B, et al. 1998. Inhibition of CPP32 blocks surface IgM-mediated apoptosis and D4-GDI cleavage in human BL60 Burkitt lymphoma cells. Eur. J. Immunol. 28:296–304 Strasser A, Harris AW, Cory S. 1991. BCL-2 transgene inhibits T cell death and perturbs thymic self-censorship. Cell 67:889–99 Siegel RM, Katsumata M, Miyashita T, Louie DC, Greene MI, et al. 1992. Inhibition of thymocyte apoptosis and negative antigenic selection in bcl-2 transgenic mice. Proc. Natl. Acad. Sci. USA 89:7003–7 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 Lang J, Arnold B, Hammerling G, Harris AW, Korsmeyer S, et al. 1997. Enforced Bcl-2 expression inhibits antigenmediated clonal elimination of peripheral B cells in an antigen dose-dependent manner and promotes receptor editing in autoreactive, immature B cells. J. Exp. Med. 186:1513–22 Fang W, Weintraub BC, Dunlap B, Garside P, Pape KA, et al. 1998. Self-reactive B lymphocytes overexpressing Bcl-xL escape negative selection and are tolerized by clonal anergy and receptor editing. Immunity 9:35–45 Woronicz JD, Calnan B, Ngo V, Winoto A. 1994. Requirement for the orphan steroid receptor Nur77 in apoptosis of Tcell hybridomas. Nature 367:277–80 Liu Z-G, Smith SW, McLaughlin KA, Schwartz LM, Osborne BA. 1994.

P1: JER/mbg

P2: ARS/ary

January 22, 1999

10:9

QC: ARS

Annual Reviews

AR078-24

APOPTOSIS IN THE IMMUNE SYSTEM

320.

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

321.

322.

323.

324.

325.

326.

327.

328.

329.

Apoptotic signals delivered through the T-cell receptor of a T-cell hybrid require the immediate-early gene nur77. Nature 367:281–84 Fernandes-Alnemri T, Litwack G, Alnemri ES. 1994. CPP32, a novel human apoptotic protein with homology to Caenorhabditis elegans cell death protein Ced-3 and mammalian Interleukin1β-converting enzyme. J. Biol. Chem. 269:30761–64 Kamens J, Paskind M, Hugunin M, Talanian RV, Allen H, et al. 1995. Identification and characterization of ICH-2, a novel member of the Interleukin-1βconverting enzyme family of cysteine proteases. J. Biol. Chem. 270:15250–56 Munday NA, Vaillancourt JP, Ali A, Casano FJ, Miller DK, et al. 1995. Molecular cloning and pro-apoptotic activity of ICErelII and ICErelIII, members of the ICE/CED-3 family of cysteine proteases. J. Biol. Chem. 270:15870–76 Faucheu C, Blanchet A, Collard-Dutilleul V, Lalanne J, Dui-Hercend A. 1996. Identification of a cysteine protease closely related to interleukin-1β-converting enzyme. Eur. J. Biochem. 236:207–13 Fernandes-Alnemri T, Litwack G, Alnemri ES. 1995. Mch2, a new member of the apoptotic Ced3/Ice cysteine protease gene family. Cancer Res. 55:2737–42 Fernandes-Alnemri T, Takahashi A, Armstrong R, Krebs J, Fritz L, et al. 1995. Mch3, a novel human apoptotic cysteine protease highly related to CPP32. Cancer Res. 55:6045–52 Duan H, Chinnaiyan AM, Hudson PL, Wing JP, He W, et al. 1996. ICE-LAP3, a novel mammalian homologue of the Caenorhabditis elegans cell death protein Ced-3 is activated during Fas- and tumor necrosis factor-induced apoptosis. J. Biol. Chem. 271:1621–25 Lippke JA, Gu Y, Sarnecki C, Caron PR, Su MS. 1996. Identification and characterization of CPP32/Mch2 homolog 1, a novel cysteine protease similar to CPP32. J. Biol. Chem. 271:1825–28 Duan H, Orth K, Chinnaiyan AM, Poirier GG, Froelich CJ, et al. 1996. ICE-LAP6, a novel member of the ICE/Ced-3 gene family, is activated by the cytotoxic T cell protease granzyme B. J. Biol. Chem. 271:16720–24 Takahashi A, Alnemri ES, Lazebnik YA, Fernandes-Alnemri T, Litwack G, et al. 1996. Cleavage of lamin A by Mch2α but not CPP32: Multiple interleukin 1βconverting enzyme-related proteases with distinct substrate recognition properties

330.

331.

332.

333.

334.

335.

336.

337.

338.

339.

340.

827

are active in apoptosis. Proc. Natl. Acad. Sci. USA 93:8395–400 Orth K, Chinnaiyan AM, Garg M, Froelich CJ, Dixit VM. 1996. The CED-3/ICE-like protease Mch2 is activated during apoptosis and cleaves the death substrate lamin A. J. Biol. Chem. 271:16443–46 Rao L, Perez D, White E. 1996. Lamin proteolysis facilitates nuclear events during apoptosis. J. Cell Biol. 135:1441– 55 Brancolini C, Benedetti M, Schneider C. 1995. Microfilament reorganization during apoptosis: role of Gas2, a possible substrate for ICE-like proteases. EMBO J. 14:5179–90 Cryns VL, Bergeron L, Zhu H, Li H, Yuan J. 1996. Specific cleavage of α-Fodrin during Fas- and Tumor Necrosis Factorinduced apoptosis is mediated by an Interleukin-1β-converting enzyme/Ced-3 protease distinct from the Poly (ADPribose) polymerase protease. J. Biol. Chem. 271:31277–82 Nath R, Raser KJ, Stafford D, Hajimohammadreza I, Posner A, et al. 1996. Non-erythroid α-spectrin breakdown by calpain and interleukin 1β-convertingenzyme-like protease(s) in apoptotic cells: contributory roles of both protease families in neuronal apoptosis. Biochem. J. 319:683–90 Wen L, Fahrni JA, Troie S, Guan J, Orth K, et al. 1997. Cleavage of focal adhesion kinase by caspases during apoptosis. J. Biol. Chem. 272:26056–61 Levkau B, Herren B, Koyama H, Ross R, Raines EW. 1998. Caspasemediated cleavage of focal adhesion kinase pp125FAK and disassembly of focal adhesions in human endothelial cell apoptosis. J. Exp. Med. 187:579–86 Kothakota S, Azuma T, Reinhard C, Klippel A, Tang J, et al. 1997. Caspase-3generated fragment of gelsolin: effector of morphological change in apoptosis. Science 278:294–98 Caul´ın C, Salvesen GS, Oshima RG. 1997. Caspase cleavage of Keratin 18 and reorganization of intermediate filaments during epithelial cell apoptosis. J. Cell Biol. 138:1379–94 ¨ T, Testa P, Zhong L, BredKayalor C, Ord sen DE. 1996. Cleavage of actin by interleukin 1β-converting enzyme to reverse DNase I inhibition. Proc. Natl. Acad. Sci USA 93:2234–38 Mashima T, Naito M, Noguchi K, Miller DK, Nicholson DW, et al. 1997. Actin cleavage by CPP-32/apopain during the

P1: JER/mbg

P2: ARS/ary

January 22, 1999

828

341.

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

342.

343.

344. 345.

346.

347.

348.

349.

350.

10:9

QC: ARS

Annual Reviews

AR078-24

RATHMELL & THOMPSON development of apoptosis. Oncogene 14:1007–12 Brancolini C, Lazarevic D, Rodriguez J, Schneider C. 1997. Dismantling cell-cell contacts during apoptosis is coupled to a caspase-dependent proteolytic cleavage of β-catenin. J. Cell Biol. 139:759–71 Hashimoto M, Inoue S, Ogawa S, Conrad C, Muramatsu M, et al. 1998. Rapid fragmentation of vimentin in human skin fibroblasts exposed to tamoxifen: a possible involvement of caspase 3. Biochem. Biophys. Res. Commun. 247:401–6 Wissing D, Mouritzen H, Egeblad M, Poirier GG, J¨aa¨ ttel¨a M. 1997. Involvement of caspase-dependent activation of cytosolic phospholipase A2 in tumor necrosis factor-induced apoptosis. Proc. Natl. Acad. Sci. USA 94:5073–77 King P, Goodbourn S. 1998. STAT1 is inactivated by a caspase. J. Biol. Chem. 273:8699–704 Ghayur T, Hugunin M, Talanian RV, Ratnofsky S, Quinlan C, et al. 1996. Proteolytic activation of Protein Kinase C δ by an ICE/CED 3-like protease induces characteristics of apoptosis. J. Exp. Med. 184:2399–404 Datta R, Kojima H, Yoshida K, Kufe D. 1997. Caspase-3-mediated cleavage of protein kinase C theta in induction of apoptosis. J. Biol. Chem. 272:20317–20 Na S, Chuang T, Cunningham A, Turi TG, Hanke JH, et al. 1996. D4-GDI, a substrate of CPP32, is proteolyzed during Fas-induced apoptosis. J. Biol. Chem. 271:11209–13 Lee N, Reinhard C, Halenbeck R, Roulston A, Shi T, et al. 1997. Activation of hPAK62 by caspase cleavage induces some of the morphological and biochemical changes of apoptosis. Proc. Natl. Acad. Sci. USA 94:13642–47 Rudel T, Bokoch GM. 1997. Membrane and morphological changes in apoptotic cells regulated by caspase-mediated activation of PAK2. Science 276:1571–74 Chen L, Marechal V, Moreau J, Levine

351.

352.

353.

354.

355.

356.

357.

358. 359.

AJ, Chen J. 1997. Proteolytic cleavage of the mdm2 oncoprotein during apoptosis. J. Biol. Chem. 272:22966–73 Beyaert R, Kidd VJ, Cornelis S, Van de Craen M, Denecker G, et al. 1997. Cleavage of PITSLRE kinases by ICE/CASP-1 and CPP32/CASP-3 during apoptosis induced by Tumor Necrosis Factor. J. Biol. Chem. 272:11694–97 Zhou BB, Yuan J, Kirschner MW. 1998. Caspase-dependent activation of cyclindependent kinases during fas-induced apoptosis in jurkat cells. Proc. Natl. Acad. Sci. USA 95:6785–90 Cosulich SC, Horiuchi H, Zerial M, Clarke PR, Woodman PG. 1997. Cleavage of Rabaptin-5 blocks endosome fusion during apoptosis. EMBO J. 16:6182– 91 Piedrafita FJ, Phahl M. 1997. Retinoidinduced apoptosis and Sp1 cleavage occur independently of transcription and require caspase activation. Mol. Cell. Biol. 17:6348–58 Wang X, Zelenski NG, Yang J, Sakai J, Brown MS, et al. 1996. Cleavage of sterol regulatory element binding proteins (SREBPs) by CPP32 during apoptosis. EMBO J. 15:1012–20 Ng FW, Nguyen M, Kwan T, Branton PE, Nicholson DW, et al. 1997. p28 Bap31, a Bcl-2/Bcl-XL- and procaspase8-associated protein in the endoplasmic reticulum. J. Cell Biol. 139:327–38 Harvey KF, Harvey NL, Michael JM, Parasivam G, Waterhouse N, et al. 1998. Caspase-mediated cleavage of the ubiquitin-protein ligase Nedd4 during apoptosis. J. Biol. Chem. 273:13524–30 Salvesen GS, Dixit VM. 1997. Caspases: intracellular signaling by proteases. Cell 91:443–46 Andrade F, Roy S, Nicholson D, Thornberry N, Rosen A, et al. 1998. Granzyme B directly and efficiently cleaves several downstream caspase substrates: Implications for CTL-induced apoptosis. Immunity 8:451–60

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:781-828. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999. 17:829–74 c 1999 by Annual Reviews. All rights reserved Copyright °

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

SELECTION OF THE T CELL REPERTOIRE Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, and Pamela S. Ohashi Ontario Cancer Institute, 610 University Avenue, Toronto, Ontario Canada M5G 2M9; e-mail: [email protected] KEY WORDS:

positive selection, negative selection, peptides, thymocyte development, diversity

ABSTRACT Advances in gene technology have allowed the manipulation of molecular interactions that shape the T cell repertoire. Although recognized as fundamental aspects of T lymphocyte development, only recently have the mechanisms governing positive and negative selection been examined at a molecular level. Positive selection refers to the active process of rescuing MHC-restricted thymocytes from programmed cell death. Negative selection refers to the deletion or inactivation of potentially autoreactive thymocytes. This review focuses on interactions during thymocyte maturation that define the T cell repertoire, with an emphasis placed on current literature within this field.

INTRODUCTION T Cell Development T cell ontogeny is characterized by the ordered expression of several surface molecules, including the coreceptors CD4 and CD8 (Figure 1) (1, 2). In general, thymocyte maturation can be divided into three broad categories based on coreceptor surface expression: 1. an early double negative (DN) CD4−CD8− stage, 2. a predominant double positive (DP) CD4+CD8+ stage, and 3. mature CD4+ or CD8+ single positive (SP) cell stage. Immature DN thymocytes upregulate the coreceptors following TCR β locus rearrangement and preTCR or β-selection (3). TCR α chain rearrangement is initiated and TCR αβ heterodimers are expressed on the cell surface at the DP stage. At this point, these thymocytes 829 0732-0582/99/0410-0829$08.00

P2: SKH/KKK/ary

January 25, 1999

830

12:31

QC: KKK/uks

Annual Reviews

Figure 1 Basic stages of thymocyte development. T cell development is characterized by the general steps outlined. Alternate pathways for positive selection have been described (282).

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SKH/spd T1: KKK

AR078-25

SEBZDA ET AL

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

831

become eligible for both positive and negative selection. T cells that express MHC class II restricted receptors are positively selected to the CD4 lineage, while T cells expressing class I–restricted TCRs are generally selected to the CD8 lineage. T cells expressing receptors that strongly react with relatively abundant thymic self-antigens are functionally eliminated or clonally deleted from the mature T cell repertoire. In addition to positive and negative selection, numerous studies have examined the progression from DN to DP thymocytes as well as CD4 versus CD8 lineage commitment. These topics are beyond the scope of this review and have been covered in recent articles (3–5) (Robey et al, this volume, pp. 283–95).

Thymocyte Positive and Negative Selection Since the phenomenon of MHC restriction was first described, a great deal of research has focused on how T cells learn to recognize antigen in the context of self-MHC (6, 7). Seminal studies using chimeric animals demonstrated that T cells become educated within the thymus (8–11) and that self-reactive T cells are also eliminated through an MHC-restricted mechanism (12, 13). Thus, the mature T cell repertoire is established in association with self-MHC molecules. The events surrounding thymocyte selection were further clarified with the generation of TCR transgenic mice. Analysis of developing thymocytes expressing defined TCRs clearly showed that MHC class I–restricted thymocytes were skewed toward the CD8+ SP lineage (14, 15), whereas class II–restricted TCR transgenic thymocytes were selected to the CD4+ lineage (16, 17). Selfreactive transgenic thymocytes were clonally deleted (18, 19), supporting earlier studies that suggested this mechanism as a form of thymic tolerance (20–22). These key findings have since been viewed as central elements of thymocyte development. Continuing research has been directed toward identifying molecules and signaling events that distinguish positive and negative selection, in order to understand the full complexity of thymocyte maturation.

Early Experiments Suggesting the Role of Peptides in Thymocyte Selection Early studies have shown that peptide/MHC ligands present during development directly alter the mature T cell repertoire (23). Nikolic-Zugic and Bevan examined the T cell response to ovalbumin, using mutants of H-2Kb that have defined alterations in the MHC molecule. Using a series of bone marrow chimeras, they could demonstrate a direct correlation between the ability of the MHC molecule to present the ovalbumin peptide and the ability of the MHC to select a T cell repertoire that could respond to the ovalbumin peptide. This study supported previous findings that suggested that the endogenous positively selecting peptide was closely related to the TCR-specific antigenic ligand (24).

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

832

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

These and other experiments demonstrated that the peptide/MHC molecules present in the thymus influence the final T cell repertoire (25–28). Analysis of peptide-specific thymocyte selection became feasible with the generation of gene-targeted mice lacking molecules associated with class I expression. MHC class I molecules are stable at the cell surface as a trimolecular complex of heavy chain, β 2 microglobulin (β 2m), and peptide. Formation of this antigen-presenting complex initially occurs within the endoplasmic reticulum (eR), where processed cytoplasmic peptide is inserted into a binding cleft formed by properly folded class I heavy chain. Disruption of transporter associated with antigen processing (TAP) genes prevents peptide transfer from the cytosol into the eR, resulting in surface expression of unstable empty MHC class I molecules. Unstable MHC class I surface expression is also found in the absence of β 2m. However, the addition of specific peptides with exogneous β 2m rescues MHC class I surface expression and results in stable class I molecules presenting defined ligands. Both of these systems are useful in vitro models for controlling thymocyte-stromal cell interactions by providing a means of associating defined peptides with MHC class I molecules. Studies have shown that mature CD8+ thymocyte development was dramatically reduced in β 2m or TAP-deficient mice (29–31). Likewise, CD8 maturation was arrested in vitro due to a lack of positively selecting ligands. However, experiments by Hogquist et al (32) demonstrated that the addition of exogenous β 2m and mixtures of peptides to β 2m-deficient thymic lobes in culture, restored positive selection. Heterogenous splenocyte acid extracts and bacterially synthesized mixtures of random peptides were more effective at inducing positive selection than a combination of two MHC-binding peptides, suggesting that positive selection was peptide specific. Similar results were obtained using TAP-deficient mice in fetal thymic organ culture (33). Collectively these studies suggested that the peptides present during thymocyte selection had an impact on the final T cell repertoire.

MODELS FOR THYMOCYTE SELECTION One of the fundamental principles of thymocyte development that has yet to be fully understood is the mechanisms that distinguish positive and negative selection events during T cell ontogeny. Individual thymocytes usually express a single TCR heterodimer at any one time, and the fate of the thymocyte is largely determined by TCR mediated interactions. Recent models have focused upon the role of peptides during selection (34–38). At issue is the nature of the selecting ligand; either qualitatively distinct ligands promote positive or negative selection, or particular TCR interactions trigger an integrated intracellular signal that meets the quantitative thresholds required for T cell selection.

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

833

Figure 2 Principles of the different models for T cell selection. (A) The qualitative model predicts that peptides generate unique signals (designated by different arrows) are generated that direct positive selection versus negative selection. (B) The avidity model predicts that low avidity interactions lead to positive selection, whereas higher avidity interactions lead to negative selection. The strength of these signals (designated by thin and thick arrows) is integrated intracellularly through specific TCR interactions. In this model, a gradient of cumulative signals define thresholds for positive or negative selection. MHC molecules are solid shapes, while TCR are designated by open shapes.

The Qualitative/Peptide Model The qualitative model proposes that mutually exclusive qualitatively different peptides promote positive and negative selection. Positive selection is achieved through defined TCR mediated interactions that generate unique signals (36, 39, 40) (Figure 2). This model was primarily supported by studies demonstrating that nonstimulatory antagonist peptides could promote positive selection. Antagonist peptides were originally shown to inhibit the subsequent activation of mature T cells using stimulatory agonist peptides (41). Using the OT-1 TCR transgenic mouse model specific for ovalbumin and H-2Kb, a correlation was drawn between positively selecting ligands and antagonist peptides (42, 43). In addition, agonist peptides were shown to promote clonal deletion. Other studies using an antibody mediated model of thymocyte selection also implied that antagonist-like interactions promoted the maturation of CD8+ T cells

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

834

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

Figure 2 (Continued)

(44). The qualitative model was also supported by studies showing that interactions with altered peptide ligands transmit distinct intracellular signals, leading to altered ζ -chain phosphorylation and a lack of ZAP-70 activity (45–50). These and other experiments (51, 52) are consistent with the idea that qualitatively different signals define thymocyte fate. However, several experiments have questioned the direct association between antagonist peptides and positive selection. For instance, antagonist peptides have been shown to induce clonal deletion (53), inhibit negative selection (54), or inhibit T cell development (55). Another study demonstrated that mature T cell function was not antagonized by the presence of the positively selecting ligand (56). Furthermore, results from an in vivo model, where defined peptides

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

835

were expressed in the thymus using adenoviral vectors, did not show a correlation between antagonist activity and positive selection (57). Together these studies demonstrate that antagonist ligands are not strictly associated with positive selection. Studies examining intracellular signaling events associated with altered peptide ligands have called into question the concept of antagonist peptides inducing unique signals. Experiments have shown that a panel of altered peptide ligands ranging from agonist to antagonist ligands can transmit a gradient of intracellular signals (58). In addition, the antagonist-like phosphorylation profiles could be mimicked with low concentrations of nominal antigen. Therefore, antagonists, partial agonists, and agonists may represent relative positions within a continuous spectrum of T cell stimulation. Recent studies have further supported the idea that altered peptide ligands form a gradient from T cell agonists to antagonists (50, 59). By the use of an MHC class I–restricted TCR transgenic system, a hierarchy of peptides capable of inducing a range of effector functions has been defined (60, 61). These studies found that the degree of TCR triggering and internalization paralleled the ability of a given peptide to induce various effector functions, as previously reported (62, 63). Ligands that stimulated full effector functions induced maximal TCR downregulation. Interestingly, peptide variants that clearly exhibited agonist-like functions, such as the capacity to induce T cell proliferation and cytokine secretion and to trigger moderate TCR internalization, inhibited maximal TCR downregulation when T cells were cocultured with this ligand and the nominal antigen (61). In all cases, subordinate ligands in this hierarchy were able to inhibit TCR downregulation and antagonize the generation of various effector functions. Thus, peptides may not fall within absolute categories and may simply reflect the assays chosen to define ligand attributes. For a specific TCR, a gradient of ligands can be described in terms of agonist qualities. However, these same ligands exhibit an inverse hierarchy with regard to antagonist functions. From this perspective, it remains difficult to rationalize the existence of antagonist-specific signals.

The Quantitative/Avidity Model An avidity model proposes that positive selection is the result of low avidity thymocyte interactions, whereas high avidity interactions elicit negative selection (35, 37, 64). Molecules that affect thymocyte-stromal cell avidity, such as coreceptors and adhesion molecules, will influence the duration or strength of TCR signaling and therefore have an impact on thymocyte fate. In addition, studies have demonstrated that glucocorticoids can also influence T cell selection (65, 66). However, these factors are relatively constant between thymocytes populations, suggesting that the key parameters that govern cell fate are TCR

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

836

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

affinity for peptide/MHC and the overall avidity of the TCR for these ligands expressed on thymic stromal cells. In this model, multiple TCR-peptide/MHC interactions are integrated to form a signaling gradient that defines cell fate. Numerous weak TCR-ligand binding events or limited high-affinity interactions provide sufficient avidity to induce an integrated signaling cascade that reaches a positively selecting threshold. Stronger signals are required to traverse a tolerizing threshold. Since thymocyte signals are integrated, a concentration dependent overlap is possible in terms of positively and negatively selecting ligands. Direct evidence indicating that thymocyte selection may be governed by a quantitative avidity model came from studies using TCR transgenic mouse lines bred into different MHC class I–deficient backgrounds. Two groups initially showed that low concentrations of a strong peptide agonist induced positive selection, whereas clonal deletion occurred with high concentrations of the same ligand (67, 68). Importantly, recent experiments have extended these findings in vivo using class I and class II models (69, 70). These studies indicate that different concentrations of the same peptide can mediate both positive and negative selection. Using fetal thymic organ culture, studies have also shown that abundant amounts of “weak” peptide/MHC ligands can mediate positive selection of defined TCR transgenic thymocytes (42, 43, 71, 72). Collectively, these findings are compatible with an avidity model. In models using a fixed concentration of selecting peptides, the avidity model would predict that low-affinity interactions would favor positive selection, while relatively higher affinity interactions favor negative selection. Direct affinity measurement of the ligands used in the ova-specific TCR transgenic mouse model demonstrated that positively selecting peptides had a lower affinity than negatively selecting ligands (73). The positively selecting peptide variants fell within a one log range of affinity, starting threefold below the affinity for thymocyte deletion. Studies have also examined the correlation between affinity and thymocyte fate using the 2C TCR transgenic model. Although the majority of data corresponds to an avidity model, both Tonegawa’s and Teh’s group have independently suggested that exceptions may be found (72, 74). In the 2C system, unfortunately the affinity measurements cannot be directly compared as the data were generated from different sources using different methodologies. Collectively however, the data suggest that a higher affinity ligand (dEV-8/Kb) required a higher peptide concentration to induce positive and negative selection than a lower affinity ligand (p2Ca/Kb). This is contrary to the avidity model, where lower affinity ligands should require a higher peptide concentration to achieve a given integrated signaling threshold. These results suggest that other parameters may also influence thymocyte fate.

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

837

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

One of the main arguments against a quantitative avidity model were reports that T cells selected by agonist ligands were nonfunctional (75, 76). However, several groups have demonstrated that functional T cells can be generated in the presence of agonist peptides using both class I and class II models of thymocyte selection (57, 70, 71, 77–79). In some cases (70, 71, 79), stronger agonist peptides or increased ligand concentrations were required to elicit a functional T cell response, and it is possible that the previous work had not adequately stimulated the selected cells to obtain a measurable response (for further discussion see section: Thymocyte Tuning to Promote Survival).

Perspectives on Selection Models QUALITATIVE VERSUS QUANTITATIVE MODELS The relationship between positively and negatively selecting peptides distinguishes a qualitative model from a quantitative model. A qualitative model stresses the distinction between these two forms of selection through the peptide-mediated generation of unique signals, whereas an avidity model views selection as a gradient based on TCRpeptide/MHC affinity in combination with integrated signaling events. Both models can explain in vitro results such as the correlation between TCR-ligand affinity and selection outcome. Likewise, the observation that low-affinity peptides induce positive selection more efficiently than high-affinity ligands is consistent with either model. However, studies reporting an overlap between positively and negatively selecting ligands clearly favors an avidity model. Collectively, the majority of studies support the quantitative avidity model of thymocyte development (see above references and 80, 81). LIMITATIONS ASSOCIATED WITH CURRENT MODELS There are limits to many of the models used to examine positive and negative selection. Approaches that manipulate MHC assembly and expression often lead to the suboptimal expression of MHC molecules on the cell surface (42, 76). As a consequence the avidity between thymocytes and stromal cells is also reduced. Therefore, in order to compensate for lost avidity, relatively higher affinity peptides may be required to mediate positive selection. For this reason, results from these model systems should not be interpreted in terms of peptide homology relative to endogenous naturally selecting ligands. Likewise these experiments are not meant to address the agonist/antagonist nature of endogenous positively selecting peptides. Instead, these studies are relevant because they demonstrate that integrated TCR-mediated signals that meet a positive selection threshold are weaker than signals that lead to negative selection. It remains possible that peptides ranging from agonists to antagonists mediate positive selection under normal conditions, but this has yet to be convincingly demonstrated.

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

838

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

As a consequence of identifying positively and negatively selecting peptides that have a relatively strong ability to interact with the TCR, it has been possible to examine the affinities for these peptide/MHC ligands. However, under more natural circumstances, it is likely that many positively selecting peptides will have no measurable affinity (82) and will have minimal homology to the antigenic ligand (83, 84). TCR AFFINITY AND OTHER FACTORS? A primary feature of the quantitative model is the integration of signaling events derived from TCR interactions as well as other surface molecules during T cell development. Unfortunately, we do not clearly understand the requirements and kinetics of TCR signaling during this process. Positive selection may or may not require TCR multimerization to induce an intracellular signal. Although the affinity of TCR peptide/MHC interactions is clearly important, other parameters may also influence cell fate. The off rate or half-life of the TCR peptide/MHC interaction may be more relevant in terms of TCR dimerization/oligomerization (85). Likewise, these factors may define the ability of thymocytes to recruit and form multimeric complexes with other surface/coreceptor molecules, which may significantly contribute to the selection of the T cell repertoire.

MOLECULES THAT INFLUENCE POSITIVE AND NEGATIVE SELECTION TCR-peptide/MHC interactions must at some point initiate signals that ultimately distinguish between positive and negative selection. Interactions leading to positive selection may generate distinct intracellular signals. Strong TCR-peptide/MHC interactions that lead to negative selection may result in the propagation of several distinct signaling cascades, some of which may be absent during positive selection. Recent studies supporting this possibility have examined the importance of a conserved structural motif found in the TCRα chain and its role in mediating TCR signals. Interestingly, mutations in this conserved region abrogated positive selection of defined TCR transgenic thymocytes, while no effect on negative selection was seen (86). Immunoprecipitation experiments showed that the TCR was not associated with the CD3δ or ζ chains, providing a mechanism to explain the differential signaling in the absence of the TCRα motif. This work supports the concept that positive selection results from unique TCR signaling. Alternatively, identical pathways may be induced by both positive and negative selection, and the relative strength or duration of these signaling pathways may determine cell fate. To further understand the underlying mechanisms that discriminate between these selection events, extensive research has begun to identify common or unique components associated with TCR-mediated signaling pathways (for

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

839

reviews, see 87–90). Recent studies examining factors such as TCR internalization, proximal TCR signaling molecules, the ERK (MAPK) pathway, the PLC-γ pathway, transcription factors, and accessory surface molecules have been shown to influence thymocyte selection (Table 1, Figure 3).

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Regulation of TCR Internalization Recent studies have shown that productive interactions between the TCR and peptide/MHC ligands result in TCR oligomerization and internalization (62, 91, 92). The degree of TCR internalization in mature cells has been correlated with the induction of effector function: strong TCR downregulation results in the induction of a spectrum of full effector responses, whereas weak TCR downregulation leads to the induction of a subset of T cell responses (50, 60, 63). A similar relationship between TCR internalization and thymocyte fate has been reported, where maximal TCR downregulation is correlated with negative selection, and suboptimal TCR downregulation mediated positive selection (79). This suggests that modulation of TCR surface expression correlates with intracellular signals that directly influence thymocyte fate. One consequence of receptor internalization is the attenuation of further signaling (93, 94). The outcome of prolonged TCR expression on thymocyte fate was examined by inhibiting a clathrin-dependent endocytosis pathway (95). The small GTPase, Rab5, regulates endocytosis. Mice expressing a dominantnegative form of Rab5 exhibited reduced TCR endocytosis and increased TCR signaling, resulting in enhanced positive and negative selection. Thus, factors that affect TCR surface expression levels can have an impact on selection events. T cell interactions with APCs induce actin polymerization and sequential morphological changes in lymphocytes, both of which are important for TCR serial triggering and internalization (96). Recent results indicate that the protooncogene Vav enhances cytoskeletal reorganization and TCR clustering, possibly through actin-cap formation (97, 98), and therefore may be necessary for both TCR signaling and T cell ontogeny. Vav-deficient mice generate limited numbers of T cells due to a defect in positive selection (99–103). Negative selection is also impaired, although inefficient deletion can occur (102, 103). In addition, these Vav–deficient T lymphocytes are hypoproliferative when stimulated through the TCR (99–101). Stimulation with phorbol ester and calcium ionophore restores proliferation, demonstrating that Vav regulates an upstream event in TCR-mediated signaling.

TCR Proximal Signals Involved in Selection Several membrane-proximal tyrosine kinases have been identified as critical components during T cell ontogeny, including the lymphocyte-specific tyrosine kinase, lck. This src family member is important in the early (DN to DP) and late (DP to SP) transitional stages of thymocyte development (104). Experiments

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

840

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

Table 1 Molecules influencing T cell development

Molecule/model

Positive selection

Negative selection

Reference

nt

↓↓

nt

248

OK OK OK

↓↓↓ ↓↓↓ ↓↓↓

nt ↓↓↓ nt

283

↓↓ Restored OK

↓↓ ↓ ↓

nt ↓ ↓

OK

Restored

Restored

80

CD5−/− F5 TCR CD5−/− H-Y TCR CD5−/− P14 TCR

OK OK OK

↑ ↑ ↓

OK ↑ ↑

266

CD30−/− H-Y TCR

OK

OK

↓↓

169

CD40L blockade

OK

OK

170

CD40L blockade AND TCR

OK

OK

Endo SAg ↓↓ Exo SEB OK Endo PCC ↓↓ Exo PCC OK

CD45 exon6−/− CD45−/−

↓ ↓

↓↓ ↓↓

SAg OK ↓

109 110

fyn−/− fyn−/− H-Y TCR fyn−/− 2C TCR

OK OK OK

OK ↓ ↑

OK ↓ nt

285, 286 108

IL7Rα−/− IL7Rα−/− Bcl-2Tg

↓↓ Restored

↓↓ Restored

nt

287, 288 241, 242

IRF-1−/− H-Y TCR IRF-1−/− P14 TCR

OK

↓↓CD8+ ↓↓CD8+

↓↓ ↓

149

Itk−/− AND TCR Itk−/− H-Y TCR

OK OK

↓↓ ↓

nt OK

124

lck DN H-Y TCR lck DN DO10 TCR

OK

↓↓ ↓↓

H-Y TCR SAg ↓ SAg ↓

105

lck−/− lck−/− fyn CA lck−/− fyn−/−

↓↓ OK ↓↓↓

↓↓ Partially restored —

nt OK —

289 107 106, 107

Mek CA Mek DN Mek DN

↑ ↓↓ OK

nt nt ↓↓

nt nt OK

129 127

Rab5 DN P14 TCR

OK

↑

↑↑

95

baxα Tg CD3δ−/− CD3δ−/− H-Y TCR CD3δ−/− AND TCR

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DN to DP

CD3ζ −/− CD3ζ −/− x ζ -0 ITAMtg CD3ζ −/− H-Y TCR ζ -1 ITAMtg CD3ζ −/− H-Y TCR ζ -3 ITAMtg

111–115 284 80

(Continued )

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

841

POSITIVE AND NEGATIVE THYMOCYTE SELECTION Table 1 (Continued )

DN to DP

Positive selection

Negative selection

Raf DN Raf CA

OK OK

↓↓ ↑

nt nt

126

Ras DN Ras/Mek DN tg

OK OK

↓↓ ↓↓↓

OK OK

125 128

Syk−/−

OK

OK

OK

290, 291

TNF-RI−/− RII−/− AND TCR TNF-RI−/− RII−/− H-Y TCR

OK

OK

OK

168

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Molecule/model

Reference

OK

vav−/− Rag chimera

↓

↓↓

nt

vav−/− vav−/− F5 TCR vav−/− BM3.6 TCR vav−/− A1 TCR vav−/− P14

OK OK

OK

↓↓ ↓↓ ↓↓ ↓↓ ↓↓

SAg ↓ nt OK nt ↓

103

Zap70−/− Zap70−/− Syk-overexpression Zap70−/− Syk−/−

↓ OK ↓↓↓

↓↓↓ Restored —

↓↓↓ nt —

116 121 120

99–101 102

CA = constitutively active, DN = dominant negative, nt = not tested, ↓ = impaired, ↓↓↓ = severely impaired, ↑ = enhanced, ↑↑↑ = severely enhanced, H-Y; male antigen/Db, P14; LCMV gp33/Db, AND; pigeon cytochrome C/I-Ek, BM3.6; anti-H-2kb/H-2k, A1 anti-H-Y/I-Ek, F5; influenza NP/Db, 2C; allo Ld/Kb, D010; OVA/I-Ad, SAg = superantigen

have shown that expression of a dominant negative form of lck in DP thymocytes prevents positive selection and impairs superantigen-mediated deletion (105). Studies examining another src family tyrosine kinase, fyn, suggest that this molecule may compensate for some lck functions (106, 107). A gain-offunction fyn mutant partially restored the DP to SP transitional stage in lck−/− thymocytes (107). In addition, TCR transgenic mice that do not express fyn have impaired maturation from HSAhi to HSAlo SP cells, and a slight reduction in negative selection (108). These studies indicate that while lck appears to play a prominent role in TCR-mediated signaling and positive and negative selection, a limited amount of cooperation and redundancy exists between the src family kinases. The activity of both lck and fyn is regulated by the transmembrane phosphatase CD45 (89). CD45 plays an important role in positive selection. In addition, studies using CD45 deficient mice have also shown that anti-CD3 induced negative selection is impaired, although deletion of superantigen reactive T cells proceeds normally (109, 110).

P2: SKH/KKK/ary

January 25, 1999

842

12:31

QC: KKK/uks

Annual Reviews

Figure 3 Signaling cascades influencing thymocyte development. Molecules implicated in thymocyte development are outlined. However, there is evidence for cross talk between pathways and not all interactions and pathways have been shown.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

P1: SKH/spd T1: KKK

AR078-25

SEBZDA ET AL

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

843

Both lck and fyn phosphorylate the immunoreceptor tyrosine-based activation motifs (ITAM) found within the CD3 complex. Each of the CD3 molecules γ , δ, and ε contain one ITAM, while the associated ζ chain contains three intracellular motifs. Mice rendered ζ -chain deficient have a reduction of DP and SP thymocytes and reduced TCR surface expression (111–115). Studies have examined the influence of the ζ -ITAM motifs in positive and negative thymocyte selection (80). TCR transgenic ζ -deficient mice reconstituted with ζ transgenes expressing 0, 1, or 3 ITAMs demonstrated that 3 ITAM motifs were required to restore normal positive and negative selection. These data elegantly showed the quantitative impact of TCR signals, suggesting that the ζ chain serves an important role in amplifying TCR-mediated signals. The tyrosine kinase ZAP-70 is rapidly activated after productive TCR engagement. Studies have shown that both positive and negative selection are severely blocked in ZAP-70-deficient mice (116). In humans, a rare genetic disorder resulting in an absence of functional ZAP-70 impairs thymocyte development such that CD8+ T cells are absent and CD4+ T lymphocytes are nonfunctional (117–119). Similar to the limited redundancy exhibited by the src family members, a related kinase, syk, may compensate for some ZAP-70 functions (120). Overexpression of syk can restore thymocyte development and mature T cell function in ZAP-70-deficient mice (121). Recently a downstream effector of activated ZAP-70 has been identified (122, 122a). This integral membrane protein, termed linker for activation of T cells (LAT), has been shown to be phosphorylated by active ZAP-70, which in turn leads to the binding of known TCR-mediated signaling molecules, including Grb2, PLC-γ 1, and the p85 subunit of PI3K (122). The exact role of LAT in vivo is unknown; however, due to its association with critical signaling molecules, it is likely to have a dramatic impact on T cell ontogeny. Another tyrosine kinase, Itk, has also been implicated in TCR signaling. Itk, a member of the Tec family, is phosphorylated after TCR-mediated signals and affects PLC-γ 1 phosphorylation, calcium mobilization, and the production of IL-2 (123). In Itk-deficient mice, thymocyte maturation was noticeably reduced in nontransgenic (123) and in TCR transgenic mice (124). Studies using the male-specific H-Y TCR transgenic mice have shown that negative selection was not affected by the absence of Itk (124).

ERK (MAPK) Pathway The extracellular regulated kinases (ERKs) or mitogen-activated protein kinase (MAPK) pathways have been generally implicated in differentiation and proliferation of cells (Figure 3). Grb2, a potential substrate of LAT, has been shown to link the guanine nucleotide exchange factor Sos with TCR-mediated signals. In turn, Sos promotes Ras GTP exchange, which results in the activation of this small GTPase. Active Ras then recruits the ser/thr kinase Raf to the plasma

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

844

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

membrane, where this molecule is activated. Raf can then phophorylate and activate the dual specificity kinase Mek, which subsequently phosphorylates its downstream effector, Erk. This pathway is necessary for positive selection. Dominant negative forms of Ras (125), Raf-1 (126), Mek-1 (127), or Ras and Mek-1(128) inhibited positive selection but not negative selection. Constitutive expression of active Raf-1 enhanced positive selection (126), again confirming the importance of this kinase during T cell maturation. Therefore, experiments that have modified the ERK pathway suggest that this cascade is solely involved in positive thymocyte selection. However, studies by others have shown that this pathway is also important in the transition from the DN to DP stage (129). A disruption in this early transition was not observed in the previous reports (125, 127, 128), suggesting that the dominant negative molecules did not effectively outcompete endogenous molecules under these conditions. During intense signaling events, such as those postulated to occur during negative selection, the ERK pathway may not have been sufficiently blocked. For this reason, further experiments addressing these issues are warranted.

The PLC-γ1 Pathway Another major signaling pathway associated with TCR stimulation is regulated by the lipid-specific enzyme, phospholipase C-γ 1 (PLC-γ 1). This molecule is responsible for the hydrolysis of phosphoinositol (4, 5) bisphosphate [PI(4, 5)P2] into inositol triphosphate [I(1, 4, 5)P3] and diacylglycerol (DAG). DAG activates protein kinase C (PKC), which participates in mature T cell activation (Figure 3). Experiments using pharmacological inhibitors of PKC such as staurosporine and Go6976 suggest that PKC plays a role during positive selection (130, 131), although conflicting reports using a different inhibitor Ro31.8425 have been published (132). Other experiments have used the phorbol ester PMA to induce PKC activity and have demonstrated that PMA together with calcium ionophores can induce positive selection using an in vitro model (133). Studies have also shown that PMA-induced activation of PKC leads to the reduction of both RAG and terminal deoxynucleotidyl transferase (TdT) expression (134, 135). Since these genes are normally downregulated upon positive selection, this provides another link between PKC activation and positive selection. Other work has shown that the inhibition of PKC also blocked clonal deletion (136), indicating that this pathway may be utilized during both positive and negative selection. The other component of PI(4, 5)P2 hydrolysis, IP3, leads to an increase in intracellular calcium and the activation of the ser/thr phosphatase calcineurin. Inhibition of calcineurin using cyclosporin A (CsA) and FK506 has been shown to block positive selection in vivo and in vitro (132, 137–140). The role of the calcineurin pathway during negative selection is controversial. Several studies using inhibitors that block calcineurin activity suggest that this pathway does

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

845

not play a role in clonal deletion (132, 136, 140), whereas other work suggests that suboptimal antigen-specific clonal deletion can be blocked (141). Further studies reported that CsA or FK506 does in fact impair negative selection (137– 139, 142). In addition, recent work has demonstrated that a relative increase in calcium flux can be observed when comparing conditions that mediate positive versus negative selection (79). Thus, using pharmacological agents, it may be easier to block positive selection and “weak” negative selection events than to completely inhibit strongly tolerizing events. Interestingly, the study by Urdahl et al reported that CsA converted negatively selecting signals to positively selecting signals in vivo (139). Currently defined signaling pathways would suggest that calcineurin may influence positive and negative selection through alterations in nuclear factor of activated T cells (NF-AT) activity. The activation of calcineurin results in the dephosphorylation of NF-AT, which leads to the translocation of this molecule into the nucleus where it can act in synergy with AP-1 to transactivate a variety of genes. Although pharmacological inhibition of calcineurin has been shown to alter thymocyte selection, mice deficient for calcineurin do not have any obvious defects in thymocyte development. Curiously, T cells from calcineurin-deficient mice still remain susceptible to FK506 and CsA (143). Gene-deficient mice for the NF-AT family members NF-ATp (NFAT1) and NF-ATc (NFAT2) also do not show an obvious block in the DP–SP transition (144–146). These findings may be explained by work of others showing that NF-AT activity is not inducible in the majority of DP thymocytes (147). Therefore, other calcium-dependent pathways, possibly including additional members of the NF-AT family, may be involved in thymocyte selection.

Transcription Factors The roles of several transcription factors have been examined during thymocyte selection. The absence of IRF-1 (interferon regulatory factor-1) leads to a partial block in CD8+ thymocyte development (148). Studies have shown that this transcription factor is activated by a CD45-dependent pathway after TCR stimulation of thymocytes. Using TCR transgenic mice that do not express IRF-1, Penninger et al demonstrated that IRF-1 is involved in both positive and negative selection of CD8+ thymocytes (149). The activity of transcription factors belonging to the early growth response (Egr) family has been correlated with positive selection (150). Messenger RNA from Egr-1, 2 and 3 is induced in DP thymocytes by anti-CD3 cross-linking, via a ras-dependent pathway. Expression of Nur77, an orphan member of the nuclear hormone receptor superfamily, has been correlated with the induction of thymocyte apoptosis (151, 152). However, nur77 gene–deficient mice did not display an apparent defect in thymocyte deletion (153), suggesting that either nur77 was not

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

846

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

required for this process in vivo, or that a closely related molecule may substitute for its action in knockout mice. In support of this latter possibility, a transgenic dominant negative mutant inhibited antigen-stimulated negative selection, while constitutive expression of wild-type nur77 induced apoptosis (154, 154a). This enhanced cell death required FasL, suggesting that nur77 was influencing thymocyte apoptosis by regulating FasL expression (155).

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Role of the TNFR Family, Costimulation, and Other Molecules in Thymic Apoptosis An increasing number of receptors have been shown to influence thymocyte selection. The TNFR superfamily member Fas (CD95) has been implicated in several models examining clonal deletion (156). Initial studies using Fas-(lpr) or FasL-deficient (gld) mice demonstrated that thymocytes could undergo TCRmediated deletion, suggesting that a Fas-dependent pathway was not involved in thymocyte death (157–159). However, using a sensitive technique to detect apoptotic thymocytes, Fas-FasL interactions were found to participate in limited negative selection (160). Moreover, recent experiments suggest that thymocyte deletion becomes Fas-dependent at high antigen concentrations (161). This dose-dependent criteria may explain why defects in negative selection were not originally observed in the lpr and gld mice. A downstream effector molecule linked to Fas-mediated apoptosis is CPP32 or caspase-3 (162). This cysteine protease was activated during anti-CD3 and peptide-induced thymocyte apoptosis (163). Moreover, addition of cysteine protease inhibitors blocked peptide or anti-CD3–induced apoptosis (163–165). However, caspase-3-deficient thymocytes displayed normal sensitivity toward anti-CD3-induced deletion (166, 167), suggesting that other caspase-3-like enzymes may be involved in this process (165). Future experiments should determine if other caspases similar to caspase-3 specifically regulate negative selection. Like Fas, the contributions of other members of the TNFR family have been examined in thymocyte deletion. TCR transgenic mice deficient for TNF receptors RI and RII showed that ligand-specific clonal deletion was not altered in vivo. However, anti-CD3-mediated deletion was impaired in the absence of these receptors (168). Elimination of another TNFR family member, CD30, results in impaired antigen-specific negative selection, although superantigenmediated deletion is not affected (169). Likewise, inhibition of CD40-CD40L interactions, either through the use of anti-CD40L mAb or the production of CD40L-deficient mice, demonstrated that under certain conditions CD40L may participate in clonal deletion (170, 171). These studies show that several TNFR family members may contribute to negative selection in particular circumstances.

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

847

Another phenotype of CD40L−/− mice is the reduction of B7-2 expression on thymic tissue (170). The expression of B7 on thymic medullary epithelium has previously been correlated with thymocyte deletion (172). While B7-1 and -2 interactions with CD28 have been shown to augment mature T cell activation, the role of these costimulatory molecules during thymocyte selection is less clear. Studies have shown that costimulation is not strictly required for positive or negative selection (173–175). In addition, in vitro deletion assays using CD28 blocking antibodies or cell lines lacking costimulatory molecules support the premise that negative selection is B7-CD28 independent (176, 177). However, other studies indicate that TCR-mediated in vitro thymocyte deletion can be enhanced with the addition of anti-CD28 mAb (178–181), suggesting that this molecule may be involved in negative selection (182). To this extent, studies investigating SEK, a component of the SAPK pathway that has been implicated in TCR and CD28 signaling, has produced controversial results regarding its role during thymocyte deletion (184, 185). Together, these reports suggest that CD28-induced costimulation is not necessary for positive selection but may augment thymocyte deletion. Another signaling molecule, Jak3 has been implicated in negative selection. Jak3 is associated with cytokine receptor signaling, including IL-2, IL-4, IL-7, IL-9, and IL-15 (186). T cells from Jak3-deficient mice do not undergo normal superantigen-mediated deletion in the thymus, and peripheral T cells display an abnormal activated phenotype (187). However, studies examining mice deficient for the γ c chain, which is a common component of these cytokine receptors, suggested that antigen-specific deletion was not impaired (188). It is not clear how these findings are related to other models of negative selection. In general, the significance and contribution of these molecules toward thymocyte selection remain unclear. It is possible that these surface receptors trigger signals that overlap with TCR-mediated signaling cascades, thus modifying signaling pathways. Alternatively, some of these molecules may generate TCR-independent signals involved in promoting death or survival. In either case, it is evident that a variety of surface receptors affect T cell maturation in addition to the TCR itself.

Summary of Molecules that Influence Thymocyte Selection These studies have shown that TCR proximal molecules such as lck, CD45, ZAP-70, and Vav are generally important components of both positive and negative selection. Further downstream, both the ERK pathway and PLC-γ pathway have been associated with positive selection, while PKC and calcineurin have been implicated in negative selection in some models. In addition, a role for various molecules of the TNFR family and costimulatory signals have been associated with negative selection. Although significant advances have been

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

848

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

made, it is still not clear how the intracellular pathways determine thymocyte fate.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

WHAT IS THE FREQUENCY OF POSITIVELY SELECTED CELLS THAT UNDERGO NEGATIVE SELECTION? Several recent studies have tried to quantitate the number of positively selected cells that are functionally self-reactive and normally undergo clonal deletion. Using a variety models, a range from 5% to 75% of the positively selected cells were shown to be potentially autoreactive (189–194). However, it may not be relevant to directly compare the frequencies of self-reactive cells using these models, because “different” T cell repertoires may be selected in each situation. Let us first consider the ways that these repertoires may be established. As thymocytes mature, the TCR density increases on the cell surface (195– 197). Since cell fate is largely dependent upon thymocyte-stromal cell avidity, this suggests that thymocytes that have initially undergone positive selection may be susceptible to negative selection via the same selecting ligand. Studies have shown that there is a window of T cell maturation when the T cells are still sensitive to deletion signals (18, 19, 156, 198–200) (Figure 1). In addition, the frequency of the selecting peptides and TCR affinity are two factors that must be considered when accounting for the establishment of T cell “sub-repertoires”. Those T cells selected by rare peptide/MHC complexes (Figure 4A) −−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→ Figure 4 Peptide interactions generate different T cell repertoires. TCR-mediated signaling is indicated by + ; where + or ++ is sufficient for positive selection and +++ surpasses the threshold for negative selection. Unselected TCRlo DP thymocytes will interact with thymic epithelial (TE) cells. (A) T cells expressing receptors for low-abundance peptides will mature as outlined in lineage A. After initial positive selection signals are received (+), the T cell will upregulate the TCR surface levels and continue to interact with TE cells to complete the positive selection process. TCRhi DP thymocytes will go on to interact with TE cells and bone marrow (BM) derived cells. The cumulative signal may increase due to higher levels of TCR expression when the T cell contacts TE cells. However, no increase in signals may occur after interactions with BM cells if the peptide is not expressed by BM cells. These cells will go on to survive and form part of the T cell repertoire. (B) Lineage B cells are selected by abundant self-peptides and potentially receive an increased positive selection stimulus (++). As the TCR level increases, the positive selection stimulus also increases (++++) due to the enhanced avidity. This leads to deletion of thymocytes specific for highly abundant peptides on the TE and/or BM. (C) Lineage C expresses very low affinity TCR specific for abundant peptides, and as a consequence receives a (+) signal for positive selection. Upregulation of the TCR levels increases the cumulative signal (++), but this is still within the threshold for positive selection. This cell will go on to form part of the mature repertoire. Different T cell sub-repertoires will be generated through these different interactions. Many other interactions will also result in the maturation of T cells that form part of the final repertoire. These would include positive selection of a T cell through interactions with mixtures of peptides (204).

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

849

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

850

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

may initially receive a minimal signal (+) that is sufficient to initiate positive selection events. Even as the TCR levels increase, the cumulative signaling threshold that is mediated through TCR specific interactions may not increase beyond the positive selection threshold (++). These cells may go on to mature and form part of the T cell sub-repertoire A. T cells will also be selected by abundant peptide/MHC complexes (Figure 4B). In some cases, these may already have integrated signals close to the threshold for negative selection (++) (Figure 4B). By increasing the TCR density, it is likely that further encounter with the selecting peptide/MHC ligand will increase the integrated signaling threshold (++++) and lead to deletion of the cell. This may or may not occur as the developing thymocyte begins to encounter bone marrow–derived stromal cells. It is possible that the deleting bone marrow–derived cell and selecting epithelial cell may present a similar set of abundant peptides, since previous studies have identified similar peptides associated with class II molecules in the spleen and thymus (201). A different scenario is also considered in Figure 4C that is relevant to these experiments. It is also possible that thymocytes expressing a very low affinity TCR are selected by abundant ligands (+). As the TCR density increases, the cumulative signals perceived by the T cell may increase slightly (++), but not to the level of the threshold required for negative selection. This cell will go on to survive as sub-repertoire C and form part of the mature T cell repertoire. Three types of experiments have recently addressed the frequency of positively selected T cells that are autoreactive. One approach has examined T cells that have been positively selected in the presence of a diverse peptide/MHC, but have not been negatively selected by MHC-positive bone marrow–derived cells. Laufer et al generated transgenic mice expressing I-Ab on thymic epithelial cells using the keratin 14 promoter. In this model, the bone marrow–derived cells did not express I-Ab and therefore could not delete thymocytes reactive to I-Ab plus self-peptides. Limiting dilution analysis showed that 5% of the cells positively selected by thymic epithelial cell peptides plus I-Ab responded to splenic peptides plus self-I-Ab (189). Surprisingly the bulk proliferative response to splenic peptides plus self-I-Ab was stronger than against allogeneic MHC. Using a chimeric model, host animals expressing MHC class I and II were reconstituted with bone marrow that did not express class I or class II molecules. Results showed that approximately twice as many CD4+ or CD8+ cells were present in the absence of negatively selecting bone marrow–derived cells. From these numbers, an estimated 50–66% of positively selected cells undergo deletion (190). Both of these studies quantitate the number of thymocytes expressing self-reactive TCR that would normally undergo negative selection as outlined in lineage B (Figure 4B). Other models have examined the frequency of T cells reactive to syngeneic MHC after positive selection by a single ligand (191–193). Although the

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

851

frequency of self-reactive T cells was measured using widely different assays, a concensus of 65–75% of the repertoire was reactive to self-MHC plus a variety of peptides. Evidence from these reports also suggests that the repertoire generated in the presence of a single ligand is different from the repertoire generated by diverse peptide/MHC interactions [(69, 191–193, 202) see below for further discussions]. According to Figure 4, these experiments are examining the T cell sub-repertoires that are generated in the presence of a single ligand, therefore, they are quantitating the amount of negative selection that occurs in the T cell sub-repertoire that is derived from ‘lineage B and C’ type interactions. Using a third approach, experiments by Zerrahn et al have examined the frequency of self-MHC reactive T cells within an immature unselected thymocyte repertoire (203). Their data shows that 5% of these cells react to self, and they extrapolate to suggest that a minimum of 5% to 30% of the repertoire may be negatively selected in a normal mouse. A related approach has investigated the frequency of “unselected” MHC null thymocytes that respond to MHC molecules by upregulating CD69 or CD5 (194). Using this assay, approximately 15%–20% of naive thymocytes were able to react with MHC molecules. However, it is not clear what fraction would have been initially negatively selected or what fraction would have been positively selected and then potentially negatively selected. These studies have examined the broad T cell repertoire, generated from many types of interactions, and may more accurately reflect the percentage of self-reactive cells in the total repertoire. These studies suggest that a range of 5% to 75% of the cells that undergo positive selection are also eligible for negative selection. In order for the estimated frequencies of negative selection to be properly evaluated, we should have a better understanding of how the T cell repertoire is shaped and the “composition” of the total T cell repertoire. For example, we do not know what fraction of the repertoire comes from interactions descending from lineage A or C, in Figure 4. It is also likely that a proportion of the repertoire is generated through interactions on a mixture of peptides (lineage M) (204). Therefore at one extreme, it is possible that the full T cell repertoire is equally represented by sub-repertoires generated though all these types of interactions. At the other extreme, it is also possible that the T cell repertoire is composed of unequal proportions of T cells from these pathways. Thus, in order to understand the relevance of the frequencies of deletion that occur in these models, one has to have an idea of the percentage of T cells that are formed through the interactions of pathway A or B or M. Clearly, further studies are required to resolve this issue. A curious prediction can be made by the model outlined in Figure 4, concerning the relevance of TCR upregulation during thymocyte maturation. According to the model, it is likely that a significant proportion of the repertoire is generated by interactions similar to lineage A and C. In the case of lineage B,

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

852

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

relatively high avidity T cells selected by abundant peptides will subsequently be negatively selected once the TCR levels increase. TCR upregulation may be an important event during the development of the T cell repertoire for several reasons. First, it may serve to efficiently purge the repertoire of clones reactive to abundant self–peptides. If the TCR levels were the same during development as in the periphery, then T cells would be tolerized only to a given level of TCR signaling thresholds. However, since the TCR is upregulated at least tenfold during maturation, the additional increase of integrated signaling events will eliminate other potentially autoreactive cells. This would provide a reasonable margin of safety against autoreactive T cells specific for commonly presented self-antigens. Experimental evidence exists to support this hypothesis (205). Secondly, the process of TCR upregulation may efficiently eliminate a potentially redundant repertoire of T cells, since T cells selected by the same ligand may have similar cross reactivities. This would also allow more space for T cells that are selected by the less abundant peptides, resulting in positively selected T cells that are capable of different flexible TCR interactions. This process would result in an overall increase in diversity of the T cell repertoire.

PROPERTIES OF POSITIVE SELECTION Multiple interactions during development lead to the maturation of a small fraction of thymocytes. Several questions pertaining to these events have been addressed over the last few years: What is the nature and specificity of the interactions during positive selection that are required to generate a diverse functional repertoire? What is the nature of the natural, positively selecting ligand? Does positive selection induce survival signals that are independent of differentiation and maturation signals? The following sections review the current studies that have addressed these topics.

Positive Selection Requires Multiple Interactions One physical parameter of T cell maturation involves sequential thymocytestromal cell interactions during development. In vivo studies using thymocytes that have initially received positively selecting signals (CD4+CD8+TCRint/hi or CD4+CD8+CD69+) have shown that these thymocytes require further interactions with appropriate MHC ligands to complete differentiation events associated with positive selection (206–208). Recent studies examining TCR internalization indicate that positively selected thymocytes potentially receive continual TCR-mediated signals (79). Strongly stimulating peptides that efficiently mediated the deletion of TCR transgenic thymocytes caused a rapid and maximal degree of TCR internalization. Weaker ligands that promoted efficient positive selection were unable to trigger maximal levels of TCR internalization

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

853

on thymocytes after TCR engagement, indicating that maturing thymocytes were capable of accruing continual TCR stimulation. Thus, thymocyte selection should be viewed as a continuous process, in which developing T cells have several opportunities to communicate with their microenvironment.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Flexibility of TCR Interactions during Positive Selection The T cell repertoire is sufficiently diverse to provide a cellular response to a wide array of foreign pathogens. However, the diversity of the repertoire is dependent upon interactions that occur during thymocyte selection. If positive selection were an event that depended upon strict peptide/MHC interactions, the repertoire would be severely limited by the peptide/MHC complexes expressed on thymic epithelial cells. MHC molecules have been estimated to present 103–104 different self-peptides when an average of 105–106 MHC molecules are expressed on the surface (209). However, thymic epithelial cells express approximately tenfold less MHC levels (72). Therefore, it may not be possible to develop a diverse T cell repertoire if positive selection is mediated by strict TCR-peptide/MHC interactions. Instead, TCR-peptide/MHC recognition during positive selection would be predicted to be a partially degenerate process that may involve multiple interactions with a mixture of peptides (52, 204, 210). SELECTION IN THE PRESENCE OF LIMITED PEPTIDE/MHC LIGANDS Several experiments have examined the influence of peptide diversity on the T cell repertoire. Using TAP−/− fetal thymic organ cultures, Ashton-Rickardt et al found that a single synthetic peptide/class I complex could generate a population of polyclonal CD8+ T cells (33). Positive selection in TAP-1-deficient mice was also analyzed. Since only a few different peptides could be isolated from class I molecules from the surface of a TAP-deficient human cell line (211, 212), positive selection in this model would occur on a limited spectrum of peptides. A small population of functional CD8+ T cells were selected by TAP-deficient thymic stroma. These CD8+ T cells displayed similar Vβ diversity as wildtype animals, although holes in the repertoire could be detected (213). Similar studies demonstrated that a population of CD8+ T cells can be detected in MHCdeficient mice following sufficient priming with antigenic ligands (214–217). These and other experiments (218) suggest that a degree of degeneracy exists between TCR peptide/MHC ligand interactions during positive selection. A variety of studies investigating thymocyte maturation using class II-restricted models also supported the possibility that TCR interactions during positive selection are flexible (57). Independent groups addressed the contribution of peptides in generating a diverse repertoire by examining H2-M deficient mice (219–221). H2-M expediates the exchange of MHC class II-associated invariant chain peptides (CLIP) for antigenic peptides (222–224). Thus, in an H2-M

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

854

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

H-2b deficient background, class II presentation is limited to I-Ab associated with a dominant CLIP peptide, since H-2b mice do not express I-Eb molecules (225). Using an alternative approach, the I-Ab β chain was covalently linked to a defined peptide and expressed as a transgene in an appropriate MHC class II–defective background (69, 191). These studies showed that single or limited ligands selected a relatively broad, though not complete, spectrum of CD4+ T cells (20%–50% of the normal CD4+ T cell number). Recent experiments suggest that a minimum of 105 different Vβ rearrangements are selected under these conditions (226). In addition, the CD4+ T cells selected in H2-M mice or in the presence of a single ligand are able to respond to a variety of antigens (192, 193, 202). To examine peptide specificity during positive selection, class II-restricted TCR transgenic thymocyte maturation was examined in H-2M-deficient mice. Five different class II–restricted transgenic TCR failed to mature in the absence of H-2M (192, 193, 227). Similarly, two of three class II-restricted transgenic TCR were not selected when bred onto an invariant chain–deficient background (228). Studies have also shown that a different repertoire of Vα or Vβ CD4+ T cells were selected in the presence of a single peptide/class II complex (69, 202). These studies largely support the idea that distinct T cell repertoires are selected by limited peptide/class II complexes, compared with the natural class II ligands. Another molecular approach has also been used to investigate the contribution of peptides toward TCR diversity (229). Sant’Angelo et al investigated the repertoire of immature and mature thymocytes from TCR β chain transgenic mice. By examining the VαJα junctional diversity in both populations, they clearly demonstrated that specific selection events have promoted the maturation of thymocytes expressing a distinct subset of TCRαβ heterodimers. They also compared the junctional diversity of the VαJα regions after positive selection in H-2M-deficient mice. These mice present a limited array of peptides in the context of class II MHC and also show an altered combination of VαJα junctional sequences. Sequence analysis of CDR3 sequences from TCRβ chains selected by a single peptide/class II complex also suggest that the sequences contain an imprint of the selecting ligand (226). These studies provide strong evidence that the mature T cell repertoire is shaped by peptide directed interactions (38, 230). Although degenerate interactions must occur to create a moderate T cell repertoire in models where limited peptide–class I and –class II molecules are present, it is likely that this flexibility relies in part on the ability of TCR gene rearrangements to generate a receptor that has sufficient avidity to be positively selected. Results show that holes in the repertoire exist when T cells are selected in the presence of minimal ligands. Collectively, these experiments strongly

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

855

suggest that the peptides present during selection have an impact on the final T cell repertoire. Similar to the flexibility observed during antigen-specific recognition (231), degenerate TCR interactions during positive selection seem to be required to generate a diverse T cell repertoire (210).

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

DIFFERENCES IN THE CONTRIBUTION OF PEPTIDES DURING POSITIVE SELECTION: INSIGHTS FROM TCR TRANSGENIC MODELS Although studies examining TCR transgenic thymocyte development in class I-deficient FTOC models have largely supported an avidity model of development, conflicting details concerning the contribution of the peptide have emerged. While some studies using the OT-1 ova-specific TCR support the role of antagonist ligands during selection, (42, 43), other studies using the P14 LCMV-specific TCR have shown that agonists are able to positively select CD8+ T cells (67, 68, 71). Furthermore, studies using the 2C alloreactive TCR transgenic model demonstrated that a variety of unrelated peptides can mediate positive selection (232). These latter findings suggest that the 2C TCR is relatively insensitive to the selecting peptide and that degenerate interactions are sufficient for positive selection. Since the majority of these studies were done in β 2m–deficient FTOC, the levels of MHC expression rescued by exogenous peptides should be similar. Therefore, variations in the role of the peptide during positive selection may be associated with different properties of the transgenic TCRs. One trivial explanation for these differences may be related to transgenic artifacts. It is possible that transgenic TCRαβ expression level varied between the three models. This would lead to differences in thymocyte-stromal cell activity, and hence affect the requirements for a transgenic T cell to surpass a positively selecting threshold. Alternatively, TCR flexibility may be defined by the number and strength of contact residues between TCR and the peptide and MHC molecules (as previously discussed in 192, 210) (Figure 5). Crystallographic studies have provided detailed information regarding the dynamics of TCR peptide/MHC interactions (233–235a). Recent studies have compared the structure of two different TCR that recognize the same viral peptide/class I complexes. Although the positions of the TCR residues that contact the peptide/MHC are conserved, only one of the 17 amino acids are identical (235). These findings emphasize the flexibility of TCR interactions. Studies have also shown that the TCR is aligned diagonally across the peptide/MHC surface (233–235a). Although the TCR generally interacts in a parallel plane with the surface of the peptide/MHC molecule, comparisons of different TCR peptide/MHC complexes have shown that the TCR may position itself at slightly different angles, altering the contribution of the complementarity-determining regions (CDR) in ligand recognition (233, 234, 235a). Other studies have shown

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

856

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

Figure 5 The affinity for a given TCR may be composed of different numbers of contact sites with peptide or MHC molecules. Arrows designate different contact points that contribute to the affinity of the TCR for peptide/MHC ligands. (A) would be representative for the 2C transgenic TCR model, (B) the OT-1 TCR transgenic model, and (C ) the P14 TCR transgenic model.

that the TCR exhibits poor shape complementarity with the surface of the peptide/MHC ligand (234). Together, the structural analysis suggests that there is a degree of plasticity and flexibility in direct TCR peptide/MHC contacts during ligand recognition. In this context, it is possible that the alloreactive H-2Ld-specific 2C TCR is especially biased toward MHC recognition and thus requires relatively fewer peptide interactions to achieve the threshold for positive selection. The OT-1 and P14 TCR may be more dependent upon peptide interactions for positive selection. Therefore, the flexible nature of the peptide interactions involved in positive selection may reflect the type of TCR that is selected. Thymocytes expressing TCR that have more contact residues and affinity for the MHC molecules may not depend on peptide-specific interactions, whereas TCR that contacts several peptide residues may be more dependent upon specific peptide interactions.

Natural Peptides for Positive Selection Self-peptides that mediate TCR transgenic positive selection have recently been identified (83, 84). Using a sensitive screening technique to initially identify H-2Kb-bound peptides capable of stimulating OVA-specific TCR transgenic thymocytes, Hogquist et al (83) found that few eluted peptide fractions actually formed significant TCR interactions. These assays equated peptide-induced thymocyte stimulation with potential T cell selection and suggested that a limited

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

857

number of self-ligands could promote positive selection. Another report also examined the effect of self-peptides on two distinct transgenic receptors (84). This study showed that peptides derived from the mouse self-antigens, histone H2A.1 (aa76-84) and brain protein E46 (aa100-108), both of which were eluted from a thymic epithelial cell line, could induce minimal positive selection of F5 TCR transgenic thymocytes specific for influenza 1968 virus nucleoprotein presented by H-2Db. These same peptides had no affect on the P14 TCR transgenic thymocytes specific for lymphocytic choriomeningitis virus glycoprotein (LCMV) in the context of H-2Db. In contrast, the self-peptide, mouse ribonucleotide reductase M1 (aa634-642), solely promoted LCMV-specific TCR transgenic thymocyte maturation. At the same time, both groups reported that the endogenous selecting ligands had minimal sequence homology with the nominal antigens, indicating that a limited number of apparently unrelated peptides may be able to induce positive selection. It is worth noting that in both studies, self-ligands were originally identified by sequencing peptide fractions derived from cell lines. This technique favors selection of abundant self-ligands associated with the appropriate MHC molecules. Since thymocyte selection seems to be influenced by thymocytestromal cell avidity, and positive selection can only be detected in the absence of efficient clonal deletion, this means that positively selecting, abundant proteins should be of a relatively low affinity. Functional assays using TCR transgenic T cells demonstrated that these selecting ligands were poorly stimulating, indicating that these peptides may exhibit low affinity for the transgenic TCRs. If high-affinity self-ligands were to induce TCR positive selection, it is assumed that these peptides would have to be present at a much lower concentration. Experimental techniques may not have favored the identification of this type of ligand. Intuitive reasoning suggests that higher affinity ligands should be closer in sequence to the nominal antigen than lower affinity peptides, which have fewer restraints in terms of TCR-contact residues. Therefore, if less abundant self-ligands are capable of efficient positive selection, it is predicted that these peptides will express limited sequence degeneracy. It will be interesting to see if future work can detect higher affinity self-ligands, and if so, how these peptides relate to the wild-type antigen.

Molecules Associated with Thymocyte Survival Recent experiments suggest that one aspect of positive selection is the generation of thymocyte survival signals. Several factors have been implicated in this process including the proto-oncogene, Bcl-2. This molecule is upregulated during positive selection and enhances thymocyte survival (236–240). Bcl-2 expression parallels IL-7Rα expression, and together, these molecules are involved in cell survival during positive selection (241, 242). Although reports

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

858

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

suggest that Bcl-2 augments positive selection (243, 244), upregulation of this molecule alone does not promote thymocyte differentiation (243). Some of the inconsistencies derived from experiments using Bcl-2 transgenic mice may be attributed to prolonged, aberrent survival of DP thymocytes at a stage when further endogenous TCR α locus rearrangement is possible (244). Hence studies examining TCR transgenic T cell development may have correlated TCR β chain expression on mature T lymphocytes with enhanced transgenic positive selection, when in fact T cell maturation was the result of novel endogenous TCR rearrangements. The role of Bcl-2 during negative selection remains controversial (236, 244–246). Current research suggests that Bcl-2 functions as a survival signal by forming heterodimers with Bax and thus inhibiting Bax-induced apoptosis (247). Since the ratio of Bcl-2 to Bax defines cell fate, overexpression of Bax is predicted to favor cell death. Indeed, transgenic expression of Bax resulted in defective thymocyte survival and a reduction in positive selection (248), consistent with previous studies suggesting that survival signals are required during T cell maturation. Reports have also indicated that other members of the Bcl-2 family influence thymocyte fate. For instance, overexpression of Bcl-xL led to increased thymocyte viability and resistance to various apoptotic signals, although it was unable to block superantigen or self-antigen induced clonal deletion (249). Furthermore, Bcl-xL is expressed at the DP stage of thymocyte development when Bcl-2 is downregulated, suggesting that these survival factors may have independent roles during selection. In addition, another isoform of Bcl-x termed Bcl-xγ has recently been identified (250). It is expressed in DP thymocytes and mature, activated T cells. Further studies are required to determine the significance of these Bcl family members during thymocyte maturation. Several other studies have provided evidence that other molecules such as the Rho GTPases and transcription factors Ets-1 and E2F-1 are important in thymocyte proliferation and survival (251–254). However, further research should delineate potential relationships between these molecules and a more complete understanding of thymocyte survival signals.

Thymocyte Tuning to Promote Survival Thymocyte fate in terms of positive selection versus clonal deletion may be defined by integrated signaling thresholds. Recent data suggest that thymocytes may use a variety of strategies to attempt to modify or tune these thresholds in order to survive (210, 255). Surface molecules may modify TCR-transduced intracellular signals and thus influence thymocyte development. A well-documented example of this involves the coreceptors CD4 and CD8. Both molecules augment TCR signaling,

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

859

possibly by recruiting the tyrosine kinase lck to the CD3 complex (256). Furthermore, these coreceptors contribute to thymocyte stromal cell avidity, such that an increase (257, 258) or decrease (259–263) of CD4 or CD8 directly affects class I– or class II–restricted thymocyte maturation, respectively. Studies have also suggested that CD8 levels may be lowered or switched from an αβ heterodimer to an αα homodimer in order to decrease avidity (43, 76, 77, 264). Although little evidence for coreceptor modification is found in nontransgenic models in vivo, this work demonstrates that thymocytes attempt to survive by modifying signaling thresholds. Another molecule that influences TCR signaling is the 67-kDa glycoprotein, CD5 (Ly-1), which is expressed on the surface of T cells and a subset of B lymphocytes. Interestingly, thymocytes isolated from CD5-deficient mice are hyperresponsive to ConA stimulation or anti-CD3 plus PMA, whereas PMA/ionophore induced proliferation is similar to wild-type controls. However, mature T cells from CD5-deficient mice proliferate normally, suggesting that CD5 plays a role in modulating thymocyte signals (265). Further analysis of thymocyte development using various TCR transgenic models demonstrated that the absence of CD5 significantly alters positive selection (Table 1) (266). These studies suggest that CD5 acts as a negative regulator of TCR-mediated signal transduction. Current studies by Love and coworkers examined CD5 surface expression in various TCR transgenic models (267). Results have shown that thymocytes expressing a TCR that undergoes efficient positive selection also express high levels of CD5, whereas poorly selected TCR transgenic thymocytes express relatively lower levels of CD5. Introduction of TCR transgenic thymocytes into a partially deleting background results in a further increase in CD5 expression. This model provides evidence that CD5 modulates TCR transgenic cell survival by negatively regulating T cell signaling. Intracellular molecules have been identified that modify T cell signaling. The cytosolic tyrosine phosphatase SHP-1 (PTP-1C, SHPTP1) negatively regulates thymocyte and T lymphocyte signaling (268, 269). Studies using the naturally arising mutants of SHP-1, motheaten or motheaten-viable mice, have reported that both thymocytes and T cells from these mice are hyperresponsive to T cell signals (269, 270). However, further analysis examining the fate of defined TCR transgenic thymocytes in the absence of SHP-1 has not been reported. Therefore it remains unclear whether this phosphatase has a direct influence on thymocyte selection and tuning. The ability of various molecules to modulate TCR-mediated signals, together with the apparent flexibility of TCR interactions during selection, may have a significant impact on mature T cell specificity. Several studies suggest that ligands present during thymocyte selection alter the response of mature T cells

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

860

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

(57, 70, 71, 79, 271). When the P14 TCR transgenic model was used, transgenic thymocytes positively selected in FTOC responded against the nominal antigen p33 while remaining tolerant toward the selecting peptide A4Y (71). What makes this interesting is the fact that the positively selecting ligand was an agonist peptide that normally stimulated TCR transgenic responses. Substantial clonal deletion was not detected in these assays, suggesting that a nonapoptotic form of tolerance was being invoked. Control CD8+ transgenic thymocytes selected in the presence of endogenous peptides responded to both A4Y and p33, demonstrating that the presence of A4Y was required to induce tolerance. In this system, as in other studies, the developing thymocytes were modified using a nondeletional mechanism such that the mature T cells remained tolerant toward the selecting ligand (as reviewed in 272). At the same time, these T cells were able to respond against higher affinity ligands, suggesting that the resting threshold can be modified during maturation. Studies using other models have recently supported T cell tuning during development (70, 230). 2C TCR transgenic T cells positively selected by low levels of H-2Ld were unable to respond to antigen presenting cells expressing endogenous levels of the self-peptide p2Ca complexed with H-2Ld (70). However, the addition of exogneous peptide p2Ca could elicit a functional cytotoxic response from H-2Ld-selected T cells. This suggests that 2C thymocytes were modified so that a response to the selecting ligands at low concentrations was not detected, while higher avidity interactions activated functional 2C transgenic T cells. Together these studies support a mechanism whereby T cells actively modulate thresholds during development in order to avoid clonal deletion and instead undergo positive selection.

T CELL SELECTION AND SUSCEPTIBILITY TO AUTOIMMUNE DISEASE Since the interactions that occur during thymocyte development define the T cell repertoire, it is possible that the thymus may imprint protection or susceptibility toward autoimmune diseases (210, 273–277). Recent studies have shown that low concentrations of agonist peptides can promote the maturation of thymocytes that are responsive toward the same ligand (57, 70, 77, 78). Therefore, the possibility exists that low amounts of potential autoantigens may promote selection of a T cell repertoire biased toward autoimmunity. Alternatively, it is possible that T cells are selected that protect an individual against autoimmune disease, by either promoting the maturation of regulatory cells or establishing a repertoire that does not cross-react with a particular autoantigen. Studies have suggested that mouse strains prone to autoimmune disease have defects in their thymic microenvironment (278). To examine the influence of thymocyte selection on autoimmunity, Thomas-Vaslin et al transplanted

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

861

thymic epithelium from nonobese diabetic (NOD) mice into control C57Bl/6 athymic nude animals. Results from this system indicated that thymic epithelium from NOD mice was sufficient to transfer susceptibility to spontaneous insulitis (279). Recent studies by Ohteki et al have demonstrated that a peptide that is the target for autoimmunity can also positively select autoreactive T cells (280). These findings are consistent with the possibility that thymic selection in certain MHC haplotypes may bias a T cell repertoire toward autoimmunity. Other studies support the idea that positive selection confers protection toward autoimmune disease. Mathis’s group has examined a TCR transgenic mouse model expressing a receptor specific for an unknown β-islet antigen and I-Ag7 (281). In a homozygous g7/g7 background, the TCR transgenic mice developed strong insulitis and approximately 50–60% of the mice went on to develop diabetes. However, breeding with C57Bl/6 mice generated TCR transgenic g7/b mice that were protected against diabetes, although delayed insulitis occurred. Further analysis showed that the protective effect was due to the selection of T cells expressing endogenous TCR by I-Ab molecules. Collectively, these studies suggest that thymocyte selection may alter the T cell repertoire and play a role in autoimmune disease susceptibility. However, research into this area has only begun, and further experiments are needed to address the role of peptides in generating a T cell repertoire that is predisposed to autoimmunity.

CONCLUDING REMARKS Although our understanding of thymocyte development continues to improve, the molecular mechanism that distinguishes positive and negative selection remains unresolved. Extensive research investigating the selection of the T cell repertoire has also unveiled new questions and complexities involved in T cell maturation. Evidence suggests that the T cell repertoire is formed by a balance between peptide-directed interactions and flexible interactions. Together this interplay is able to create a functional diverse T cell repertoire. However, one has to consider whether the repertoire generated in various gene-deficient or transgenic models reflects the same specificities that are generated in the normal animal. Early research into this area suggests that this is not always the case. The plasticity of the immune system may allow the generation of distinct functional repertoires, given different conditions and limitations. Future experiments will undoubtedly provide answers and reveal new problems in understanding the mechanisms that generate a diverse T cell repertoire and how this influences immunity and autoimmunity. ACKNOWLEDGMENTS We would like to thank June Galicia for help in the preparation of the manuscript and Juan Carlos Zuniga-Pfl¨ucker for critical comments and discussion.

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

862

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL Visit the Annual Reviews home page at http://www.AnnualReviews.org

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Literature Cited 1. Boyd RL, Hugo P. 1991. Towards an integrated view of thymopoiesis. Immunol. Today 12:71–79 2. Anderson G, Moore NC, Owen JJ, Jenkinson EJ. 1996. Cellular interactions in thymocyte development. Annu. Rev. Immunol. 14:73–99 3. von Boehmer H, Fehling HJ. 1997. Structure and function of the pre-T cell receptor. Annu. Rev. Immunol. 15:433– 52 4. Zuniga-Pfl¨ucker JC, Lenardo MJ. 1996. Regulation of thymocyte development from immature progenitors. Curr. Opin. Immunol. 8:215–24 5. Di Santo JP, Rodewald H-R. 1998. In vivo roles of receptor tyrosine kinases and cytokine receptors in early thymocyte development. Curr. Opin. Immunol. 10:196–207 6. Zinkernagel RM, Doherty PC. 1974. Restriction of in vitro T cell mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system. Nature 248:701–2 7. Shearer GM. 1974. Cell mediated cytotoxicity to trinitrophenyl-modified syngeneic lymphocytes. Eur. J. Immunol. 4:257 8. Bevan MJ. 1977. In a radiation chimera, host H-2 antigens determine immune responsiveness of donor cytotoxic cells. Nature 269:417–18 9. Zinkernagel RM, Callahan GN, Althage A, Cooper S, Klein PA, Klein J. 1978. On the thymus in the differentiation of “H-2 self-recognition” by T cells: Evidence for dual recognition? J. Exp. Med. 147:882–96 10. Zinkernagel RM, Callahan GN, Klein J, Dennert G. 1978. Cytotoxic T cells learn specificity for self H-2 during differentiation in the thymus. Nature 271:251–53 11. Fink PJ, Bevan MJ. 1978. H-2 antigens of the thymus determine lymphocyte specificity. J. Exp. Med. 148:766– 75 12. Matzinger P, Zamoyska R, Waldmann H. 1984. Self tolerance is H-2 restricted. Nature 308:738–41 13. Rammensee H-G, Bevan MJ. 1984. Evidence from in vitro studies that tolerance to self antigens is MHC-restricted. Nature 308:741–44

14. Teh HS, Kisielow P, Scott B, Kishi H, Uematsu Y, Bl¨uthmann H, von Boehmer H. 1988. Thymic major histocompatibility complex antigens and the αβ T cell receptor determine the CD4/CD8 phenotype of T cells. Nature 335:229–33 15. Sha WC, Nelson ChA, Newberry RD, Kranz DM, Russell JH, Loh DY. 1988. Selective expression of an antigen receptor on CD8-bearing T lymphocytes in transgenic mice. Nature 335:271–74 16. Berg LJ, Pullen AM, Fazekas de StGB, Mathis D, Benoist C, Davis MM. 1989. Antigen/MHC-specific T cells are preferentially exported from the thymus in the presence of their MHC ligand. Cell 58:1035–1046 17. Kaye J, Hsu ML, Sauron ME, Jameson SC, Gascoigne NR, Hedrick SM. 1989. Selective development of CD4+ T cells in transgenic mice expressing a class II MHC-restricted antigen receptor. Nature 341:746–49 18. Kisielow P, Bl¨uthmann 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 19. Sha WC, Nelson CA, Newberry RD, Kranz DM, Russell JH, Loh DY. 1988. Positive and negative selection of an antigen receptor on T cells in transgenic mice. Nature 336:73–76 20. Kappler JW, Roehm N, Marrack P. 1987. T cell tolerance by clonal elimination in the thymus. Cell 49:273–80 21. Kappler JW, Staerz UD, White J, Marrack P. 1988. Self tolerance eliminates T cells specific for Mls-modified products of the major histocompatibility complex. Nature 332:35–40 22. MacDonald HR, Schneider R, Lees RK, Howe RC, Acha-Orbea H, Festenstein H, Zinkernagel RM, Hengartner H. 1988. T cell receptor Vβ use predicts reactivity and tolerance to Mlsa-encoded antigens. Nature 332:40–45 23. Nikolic-Zugic J, Bevan MJ. 1990. Role of self-peptides in positively selecting the T-cell repertoire. Nature 344:65– 67 24. Bevan MJ, H¨unig T. 1981. T cells respond preferentially to antigens that are

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

25.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35. 36.

similar to self H-2. Proc. Natl. Acad. Sci. USA 78:1843–47 Singer A, Mizuochi T, Munitz TI, Gress RE. 1986. Role of self antigens in the selection of the developing T cell repertoire. In Progress in Immunology VI, ed. B Cinader, RG Miller, pp. 60-66. Orlando, FL: Academic Jacobs H, von Boehmer H, Melief CJM, Berns A. 1990. Mutations in the major histocompatibility complex class I antigen presenting groove affect both negative and positive selection of T cells. Eur. J. Immunol. 20:2333–37 Sha WC, Nelson CA, Newberry RD, Pullen JK, Pease LR, Russell JH, Loh DY. 1990. Positive selection of transgenic receptor-bearing thymocytes by Kb antigen is altered by Kb mutations that involve peptide binding. Proc. Natl. Acad. Sci. USA 87:6186–90 Ohashi PS, Zinkernagel RM, Luescher IF, Hengartner H, Pircher H. 1993. Enhanced positive selection of a transgenic TCR by a restriction element that does not permit negative selection. Int. Immunol. 5:131–38 Zijlstra M, Bix M, Simister NE, Loring JM, Raulet DH, Jaenisch R. 1990. Beta 2-microglobulin deficient mice lack CD4+8+ cytolytic T cells. Nature 344:742–46 Koller BH, Marrack P, Kappler JW, Smithies O. 1990. Normal development of mice deficient in beta 2M, MHC class I proteins, and CD8+ T cells. Science 248:1227–30 Van Kaer L, Ashton-Rickardt PG, Ploegh HL, Tonegawa S. 1992. TAP1 mutant mice are deficient in antigen presentation, surface class I molecules, and CD4−CD8+ T cells. Cell 71:1205–14 Hogquist KA, Gavin MA, Bevan MJ. 1993. Positive selection of CD8+ T cells induced by major histocompatibility complex binding peptides in fetal thymic organ culture. J. Exp. Med. 177:1469–73 Ashton-Rickardt PG, Van Kaer L, Schumacher TNM, Ploegh HL, Tonegawa S. 1993. Peptide contributes to the specificity of positive selection of CD8+ T cells in the thymus. Cell 73:1041–49 Schumacher TNM, Ploegh HL. 1994. Are MHC-bound peptides a nuisance for positive selection? Immunity 1:721– 23 Ashton-Rickardt PG, Tonegawa S. 1994. A differential-avidity model for T-cell selection. Immunol. Today 15:362–66 Jameson SC, Hogquist KA, Bevan MJ.

37.

38. 39.

40. 41.

42.

43. 44.

45.

46.

47.

48.

49.

50.

863

1995. Positive selection of thymocytes. Annu. Rev. Immunol. 13:93–126 Williams O, Tanaka Y, Tarazona R, Kioussis D. 1997. The agonist-antagonist balance in positive selection. Immunol. Today 18:121–26 Janeway CA Jr. 1998. A tale of two T cells. Immunity 8:391–94 Mannie MD. 1991. A unified model for T cell antigen recognition and thymic selection of the T cell repertoire. J. Theor. Biol. 151:169–92 Janeway CA. 1993. High fives or hard clasps? Curr. Biol. 2:591–93 De Magistris MT, Alexander J, Coggeshall M, Altman A, Gaeta FCA, Grey HM, Sette A. 1992. Antigen analogmajor histcompatibility complexes act as antagonists of the T cell receptor. Cell 68:625–34 Hogquist KA, Jameson SC, Heath WR, Howard JL, Bevan MJ, Carbone FR. 1994. T cell receptor antagonist peptides induce positive selection. Cell 76:17–27 Jameson SC, Hogquist KA, Bevan MJ. 1994. Specificity and flexibility in thymic selection. Nature 369:750–52 Basson MA, Bommhardt U, Cole MS, Tso JY, Zamoyska R. 1998. CD3 ligation on immature thymocytes generates antagonist-like signals appropriate for CD8 lineage commitment, independently of T cell receptor specificity. J. Exp. Med. 187:1249–60 Sloan-Lancaster J, Shaw AS, Rothbard JB, Allen PM. 1994. Partial T cell signaling: altered phospho-zeta and lack of Zap70 recruitment in APL-induced T cell anergy. Cell 79:913–22 Madrenas J, Wange RL, Wang JL, Isakov N, Samelson LE, Germain RN. 1995. Zeta phosphorylation without ZAP-70 activation induced by TCR antagonists or partial agonists. Science 267:515–18 Sousa CR, Levine EH, Germain RN. 1996. Partial signaling by CD8+ T cells in response to antagonist ligands. J. Exp. Med. 184:149–57 La Face DM, Couture C, Anderson K, Shih G, Alexander J, Sette A, Mustelin T, Altman A, Grey HM. 1997. Differential T cell signaling induced by antagonist peptide-MHC complexes and the associated phenotypic responses. J. Immunol. 158:2057–64 Chau LA, Bluestone JA, Madrenas J. 1998. Dissociation of intracellular signalling pathways in response to partial agonist ligands of the T cell receptor. J. Exp. Med. 187:1699–1709 Hemmer B, Stefanova I, Vergelli M, Ger-

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

864

51.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

52.

53.

54.

55.

56.

57.

58.

59.

60.

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL main R, Martin R. 1998. Relationships among TCR ligand potency, thresholds for effector function elicitation, and the quality of early signaling events in human T cells. J. Immunol. 160:5807–14 Elliott JI. 1997. T cell repertoire formation displays characteristics of qualitative models of thymic selection. Eur. J. Immunol. 27:1831–37 Williams O, Tarazona R, Kioussis D, Harker N, Roderick K. 1998. Interactions with multiple peptide ligands determine the fate of developing thymocytes. Proc. Natl. Acad. Sci. USA 95:5706–11 Page DM, Alexander J, Snoke K, Appella E, Sette A, Hedrick SM, Grey HM. 1994. Negative selection of CD4+CD8+ thymocytes by T-cell receptor peptide antagonists. Proc. Natl. Acad. Sci. USA 91:4057–61 Williams O, Tanaka Y, Bix M, Murdjeva M, Littman DR, Kioussis D. 1996. Inhibition of thymocyte negative selection by T cell receptor antagonist peptides. Eur. J. Immunol. 26:532–38 Spain LM, Jorgensen JL, Davis MM, Berg LJ. 1994. A peptide antigen antagonist prevents the differentiation of T cell receptor transgenic thymocytes. J. Immunol. 152:1709–17 Ignatowicz L, Kappler J, Parker DC, Marrack P. 1996. The responses of mature T cells are not necessarily antagonized by their positively selecting peptide. J. Immunol. 157:1827–31 Nakano N, Rooke R, Benoist C, Mathis D. 1997. Positive selection of T cells induced by virus-mediated delivery of neo-peptides to the thymus. Science 275:678–83 Smyth LA, Williams O, Huby RDJ, Norton T, Acuto O, Ley SC, Kioussis D. 1998. Altered peptide ligands induce quantitatively but not qualitatively different intracellular signals in primary thymocytes. Proc. Natl. Acad. Sci. USA 95:8193–98 Preckel T, Grimm R, Martin S, Weltzien HU. 1997. Altered hapten ligands antagonize trinitrophenyl-specific cytotoxic T cells and block internalization of hapten-specific receptors. J. Exp. Med. 185:1803–13 Bachmann MF, Oxenius A, Speiser DE, Mariathasan S, Hengartner H, Zinkernagel RM, Ohashi PS. 1997. Peptideinduced T cell receptor down-regulation on naive T cells predicts agonist/partial agonist properties and strictly correlates with T cell activation. Eur. J. Immunol. 27:2195–2203

61. Bachmann MF, Speiser DE, Zakarian A, Ohashi PS. 1998. Inhibition of TCRtriggering by a spectrum of altered peptide ligands suggests the mechanism for TCR-antagonism. Eur. J. Immunol. In press 62. Valitutti S, Muller S, Cella M, Padovan E, Lanzavecchia A. 1995. Serial triggering of many T-cell receptors by a few peptide-MHC complexes. Nature 375:148–51 63. Valitutti S, Muller S, Dessing M, Lanzavecchia A. 1996. Different responses are elicited in cytotoxic T lymphocytes by different levels of T cell receptor occupancy. J. Exp. Med. 183:1917– 21 64. Sprent J, Lo D, Gao E, Ron Y. 1988. T cell selection in the thymus. Immunol. Rev. 101:173–90 65. King LB, Vacchio MS, Dixon K, Hunziker R, Margulies DH, Ashwell JD. 1995. A targeted glucocorticoid receptor antisense transgene increases thymocyte apoptosis and alters thymocyte development. Immunity 3:647–56 66. Vacchio MS, Ashwell JD. 1997. Thymus-derived glucocorticoids regulate antigen-specific positive selection. J. Exp. Med. 185:2033–38 67. Ashton-Rickardt PG, Bandeira A, Delaney JR, Van Kaer L, Pircher HP, Zinkernagel RM, Tonegawa S. 1994. Evidence for a differential avidity model of T cell selection in the thymus. Cell 76:651–63 68. Sebzda E, Wallace VA, Mayer J, Yeung RSM, Mak TW, Ohashi PS. 1994. Positive and negative thymocyte selection induced by different concentrations of a single peptide. Science 263:1615–18 69. Fukui Y, Ishimoto T, Utsuyama M, Gyotoku T, Koga T, Nakao K, Hirokawa K, Katsuki M, Sasazuki T. 1997. Positive and negative CD4+ thymocyte selection by a single MHC class II/peptide ligand affected by its expression level in the thymus. Immunity 6:401–10 70. Cook JR, Wormstall E-M, Hornell T, Russell J, Connolly JM, Hansen TH. 1997. Quantitation of the cell surface level of Ld resulting in positive versus negative selection of the 2C transgenic T cell receptor in vivo. Immunity 7:233– 41 71. Sebzda E, K¨undig TM, Thomson CT, Aoki K, Mak SY, Mayer J, Zamborelli TM, Nathenson S, Ohashi PS. 1996. Mature T cell reactivity altered by a peptide agonist that induces positive selection. J. Exp. Med. 183:1093–1104

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION 72. Delaney JR, Sykulev Y, Eisen HN, Tonegawa S. 1998. Differences in the level of expression of class I major histocompatibility complex proteins on thymic epithelial and dendritic cells influence the decision of immature thymocytes between positive and negative selection. Proc. Natl. Acad. Sci. USA 95:5235– 40 73. Alam SM, Travers PJ, Wung JL, Nasholds W, Redpath S, Jameson SC, Gasciogne NRJ. 1996. T-cell-receptor affinity and thymocyte positive selection. Nature 381:616–20 74. Motyka B, Teh H-S. 1998. Naturally occurring low affinity peptide/MHC class I ligands can mediate negative selection and T cell activation. J. Immunol. 160:77–86 75. Hogquist KA, Jameson SC, Bevan MJ. 1995. Strong agonist ligands for the T cell receptor do not mediate positive selection of functional CD8+ T cells. Immunity 3:78–86 76. Girao C, Hu Q, Sun J, Ashton-Rickardt PG. 1997. Limits to the differential avidity model of T cell selection in the thymus. J. Immunol. 159:4205–11 77. Chidgey A, Boyd R. 1997. Agonist peptide modulates T cell selection thresholds through qualitative and quantitative shifts in CD8 co-receptor expression. Int. Immunol. 9:1527–36 78. Wang R, Nelson A, Kimachi K, Grey HM, Farr AG. 1998. The role of peptides in thymic positive selection of class II major histocompatibility complexrestricted T cells. Proc. Natl. Acad. Sci. USA 95:3804–9 79. Mariathasan S, Bachmann MF, Bouchard D, Ohteki T, Ohashi PS. 1998. Degree of TCR internalization and Ca2+ flux correlates with thymocyte selection. J. Immunol. 161:6030–37 80. Shores EW, Tran T, Grinberg A, Sommers CL, Shen H, Love PE. 1997. Role of the multiple T cell receptor (TCR)-ζ chain signaling motifs in selection of the T cell repertoire. J. Exp. Med. 185:893– 900 81. Yamazaki T, Arase H, Ono S, Ohno H, Watanabe H, Saito T. 1997. A shift from negative to positive selection of autoreactive T cells by the reduced level of TCR signal in TCR-transgenic CD3deficient mice. J. Immunol. 158:1634– 40 82. Liu C-P, Crawford F, Marrack P, Kappler J. 1998. T cell positive selection by a high density, low affinity ligand. Proc. Natl. Acad. Sci. USA 95:4522–26

865

83. Hogquist KA, Tomlinson AJ, Kieper WC, McGargill MA, Hart MC, Naylor S, Jameson SC. 1997. Identification of a naturally occurring ligand for thymic positive selection. Immunity 6:389–99 84. Hu Q, Bazemore Walker CR, Girao C, Opferman JT, Sun J, Shabanowitz J, Hunt DF, Ashton-Rickardt PG. 1997. Specific recognition of thymic selfpeptides induces the positive selection of cytotoxic T lymphocytes. Immunity 7:221–31 85. Bachmann MF, Ohashi PS. 1998. The role of T cell receptor dimerization in T cell activation. Submitted 86. B¨ackstr¨om BT, M¨uller U, Hausmann B, Palmer E. 1998. Positive selection through a motif in the αβ T cell receptor. Science 281:835–38 87. Cantrell D. 1996. T cell antigen receptor signal transduction pathways. Annu. Rev. Immunol. 14:259–74 88. Wange RL, Samelson LE. 1996. Complex complexes: Signaling at the TCR. Immunity 5:197–205 89. Okumura M, Thomas ML. 1995. Regulation of immune function by protein tyrosine phosphatases. Curr. Opin. Immunol. 7:312–19 90. Qian D, Weiss A. 1997. T cell antigen receptor signal transduction. Curr. Opin. Cell Biol. 9:205–12 91. Reich Z, Boniface JJ, Lyons DS, Borochov N, Wachtel EJ, Davis MM. 1997. Ligand-specific oligomerization of T-cell receptor molecules. Nature 387:617–20 92. Bachmann MF, Salzmann M, Oxenius A, Ohashi PS. 1998. Formation of TCR dimers/trimers as a crucial step for T cell activation. Eur. J. Immunol. 28:2571– 79 93. Valitutti S, Muller S, Dessing M, Lanzavecchia A. 1996. Signal extinction and T cell repolarization in T helper cellantigen-presenting cell conjugates. Eur. J. Immunol. 26:2012–16 94. Valitutti S, Lanzavecchia A. 1997. Serial triggering of TCRs: a basis for the sensitivity and specificity of antigen recognition. Immunol. Today 18:299–304 95. Andr´e P, Boretto J, Hueber A-O, R´egnier-Vigouroux A, Gorvel J-P, Ferrier P, Chavrier P. 1997. A dominantnegative mutant of the Rab5 GTPase enhances T cell signaling by interfering with TCR down-modulation in transgenic mice. J. Immunol. 159:5253–63 96. Valitutti S, Dessing M, Aktories K, Gallati H, Lanzavecchia A. 1995. Sustained signaling leading to T cell activation

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

866

97.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

98.

99.

100.

101.

102.

103.

104.

105.

106.

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL results from prolonged T cell receptor occupancy. Role of T cell actin cytoskeleton. J. Exp. Med. 181:577–84 Fischer K-D, Yong Y-Y, Nishina H, Tedford K, Mareng`ere LEM, Kozieradzki I, Sasaki T, Starr M, Chan G, Gardener S, Nghiem MP, Bouchard D, Barbacid M, Bernstein A, Penninger JM. 1998. Vav is a regulator of cytoskeletal reorganization mediated by the T-cell receptor. Curr. Biol. 8:554–62 Holsinger LJ, Graef IA, Swat W, Chi T, Bautista DM, Davidson L, Lewis RS, Alt FW, Crabtree GR. 1998. Defects in actin-cap formation in Vav-deficient mice implicate an actin requirement for lymphocyte signal transduction. Curr. Biol. 8:563–72 Tarakhovsky A, Turner M, Schaal S, Mee PJ, Duddy LP, Rajewsky K, Tybulewicz VL. 1995. Defective antigen receptor-mediated proliferation of B and T cells in the absence of Vav. Nature 374:467–70 Zhang R, Alt F, Davidson L, Orkin SH, Swat W. 1995. Defective signalling through the T- and B-cell antigen receptors in lymphoid cells lacking the vav proto-oncogene. Nature 374:470– 73 Fischer KD, Zmuldzinas A, Gardner S, Barbacid M, Bernstein A, Guidos C. 1995. Defective T-cell receptor signalling and positive selection of Vavdeficient CD4+ CD8+ thymocytes. Nature 374:474–77 Turner M, Mee PJ, Walters AE, Quinn ME, Mellor AL, Zamoyska R, Tybulewicz VLJ. 1997. A requirement for the Rho-family GTP exchange factor Vav in positive and negative selection of thymocytes. Immunity 7:451–60 Kong Y-Y, Fischer KD, Bachmann MF, Mariathasan S, Kozieradzki I, Nghiem MP, Bouchard D, Bernstein A, Ohashi PS, Penninger JM. 1998. Vav regulates peptide-specific apoptosis in thymocytes. J. Exp. Med. 188:2099–2111 Alberola-Ila J, Takaki S, Kerner JD, Perlmutter RM. 1997. Differential signalling by lymphocyte antigen receptors. Annu. Rev. Immunol. 15:125–54 Hashimoto K, Sohn SJ, Levin SD, Tada T, Perlmutter RM, Nakayama T. 1996. Requirement for p56lck tyrosine kinase activation in T cell receptor-mediated thymic selection. J. Exp. Med. 184:931– 43 van Oers NS, Lowin-Kropf B, Finlay D, Connolly K, Weiss A. 1996. αβ T cell development is abolished in mice lack-

107.

108.

109.

110.

111.

112.

113.

114.

115.

ing both Lck and Fyn protein tyrosine kinases. Immunity 5:429–36 Groves T, Smiley P, Cooke MP, Forbush K, Perlmutter RM, Guidos CJ. 1996. Fyn can partially substitute for Lck in T lymphocyte development. Immunity 5:417– 28 Utting O, Teh S-J, Teh H-S. 1998. T cells expressing receptors of different affinity for antigen ligands reveal a unique role for p59fyn in T cell development and optimal stimulation of T cells by antigen. J. Immunol. 160:5410–19 Kishihara K, Penninger J, Wallace VA, K¨undig TM, Kawai K, Wakeham A, Timms E, Pfeffer K, Ohashi PS, Thomas ML, Furlonger C, Paige CJ, Mak TW. 1993. Normal B lymphocyte development but impaired T cell maturation in CD45-exon6 protein tyrosine phosphatase-deficient mice. Cell 74:143–56 Byth KF, Conroy LA, Howlett S, Smith AJH, May J, Alexander DR, Holmes N. 1996. CD45-null transgenic mice reveal a positive regulatory role for CD45 in early thymocyte development, in the selection of CD4+CD8+ thymocytes, and in B cell maturation. J. Exp. Med. 183:1707–18 Love PE, Shores EW, Johnson MD, Tremblay ML, Lee EJ, Grinberg A, Huang SP, Singer A, Westphal H. 1993. T cell development in mice that lack the ζ chain of the T cell antigen receptor complex. Science 261:918–21 Ohno H, Aoe T, Taki S, Kitamura D, Ishida Y, Rajewsky K, Saito T. 1993. Developmental and functional impairment of T cells in mice lacking CD3ζ chains. EMBO J. 12:4357–66 Malissen M, Gillet A, Rocha B, Trucy J, Vivier E, Boyer C, K¨ontgen F, Brun N, Mazza G, Spanopoulou E, Guy-Grand D, Malissen B. 1993. T cell development in mice lacking the CD3-ζ /η gene. EMBO J. 12:4347–55 Liu C-P, Ueda R, She J, Sancho J, Wang B, Weddell G, Loring J, Kurahara C, Dudley EC, Hayday A, Terhorst C, Huang M. 1993. Abnormal T cell development in CD3-zeta−/− mutant mice and identification of a novel T cell population in the intestine. EMBO J. 12:4863– 75 Shores EW, Ono M, Kawabe T, Sommers CL, Tran T, Lui K, Udey MC, Ravetch J, Love PE. 1998. T cell development in mice lacking all T cell receptor ζ family members (ζ , η, and FcεRIγ ). J. Exp. Med. 187:1093–1101

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POSITIVE AND NEGATIVE THYMOCYTE SELECTION 116. Negishi I, Motoyama N, Nakayama K, Senju S, Hatakeyama S, Zhang Q, Chan AC, Loh DY. 1995. Essential role for ZAP-70 in both positive and negative selection of thymocytes. Nature 376:435– 38 117. Arpaia E, Shahar M, Dadi H, Cohen A, Roifman CM. 1994. Defective T cell receptor signaling and CD8+ thymic selection in humans lacking Zap-70 kinase. Cell 76:947–58 118. Chan AC, Kadlecek TA, Elder ME, Filipovich AH, Kuo W-L, Iwashima M, Parslow TG, Weiss A. 1994. ZAP-70 deficiency in an autosomal recessive form of severe combined immunodeficiency. Science 264:1599–1601 119. Elder ME, Lin D, Clever J, Chan AC, Hope TJ, Weiss A, Parslow TG. 1994. Human severe combined immunodeficiency due to a defect in ZAP-70, a T cell tyrosine kinase. Science 264:1596– 99 120. Cheng AM, Negishi I, Anderson SJ, Chan AC, Bolen J, Loh DY, Pawson T. 1997. The Syk and ZAP-70 SH2containing tyrosine kinases are implicated in pre-T cell receptor signalling. Proc. Natl. Acad. Sci. USA 94:9797– 801 121. Gong Q, White L, Johnson R, White M, Negishi I, Thomas M, Chan AC. 1997. Restoration of thymocyte development and function in Zap-70−/− mice by the Syk protein tyrosine kinase. Immunity 7: 369–77 122. Zhang W, Sloan-Lancaster J, Kitchen J, Trible RP, Samelson LE. 1998. LAT: The ZAP-70 tyrosine kinase substrate that links T cell receptor to cellular activation. Cell 92:83–92 122a. Weber JR, Orstavik S, Torgersen KM, Danbolt NC, Berg SF, Ryan JC, Tasken K, Imboden JB, Vaage JT. 1998. Molecular cloning of the cDNA encoding pp36, a tyrosine phosphorylated adaptor protein selectively expressed by T cells and natural killer cells. J. Exp. Med. 187:1157–61 123. 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. 187:1721–27 124. Liao XC, Littman DR. 1995. Altered T cell receptor signaling and disrupted T cell development in mice lacking Itk. Immunity 3:757–69 125. Swan KA, Alberola-Ila J, Gross JA, Appleby MW, Forbush KA, Thomas JF, Perlmutter RM. 1995. Involvement of

126.

127.

128.

129.

130.

131.

132.

133.

134.

135.

136.

137.

867

p21ras distinguishes positive and negative selection in thymocytes. EMBO J. 14:276–85 O’Shea CC, Crompton T, Rosewell IR, Hayday AC, Owen MJ. 1996. Raf regulates positive selection. Eur. J. Immunol. 26:2350–55 Alberola-Ila J, Forbush KA, Seger R, Krebs EG, Perlmutter RM. 1996. Selective requirement for MAP kinase activation in thymocyte differentiation. Nature 373:620–23 Alberola-Ila J, Hogquist KA, Swan KA, Bevan MJ, Perlmutter RM. 1996. Positive and negative selection invoke distinct signaling pathways. J. Exp. Med. 184:9–18 Crompton T, Gilmour KC, Owen MJ. 1996. The MAP kinase pathway controls differentiation from doublenegative to double-positive thymocytes. Cell 86:243–51 Nakayama K, Nakauchi H. 1993. Cyclosporin A inhibits the decrease of CD4/CD8 expression induced by protein kinase C activation. Int. Immunol. 5:419–26 Ohaka Y, Kuwata T, Asada A, Zhao Y, Mukai M, Iwata M. 1997. Regulation of thymocyte lineage commitment by the level of classical protein kinase C activity. J. Immunol. 158:5707–16 Anderson G, Anderson KL, Conroy LA, Hallam TJ, Moore NC, Owen JJT, Jenkinson EJ. 1995. Intracellular signaling events during positive and negative selection of CD4+CD8+ thymocytes in vitro. J. Immunol. 154:3636–43 Takahama Y, Nakauchi H. 1996. Phorbol ester and calcium ionophore can replace TCR signals that induce positive selection of CD4 T cells. J. Immunol. 157:1508–13 Tillinghast JP, Russell JH, Fields LE, Loh DY. 1989. Protein kinase C regulation of terminal deoxynucleotidyl transferase. J. Immunol. 143:2378–83 Turka LA, Schatz DG, Oettinger MA, Chun JJM, Gorka C, Lee K, McCormack WT, Thompson CB. 1991. Thymocyte expression of RAG-1 and RAG-2: termination by T cell receptor crosslinking. Science 253:778–81 Vasquez NJ, Kane LP, Hedrick SM. 1994. Intracellular signals that mediate thymic negative selection. Immunity 1:45–56 Jenkins MK, Schwartz RH, Pardoll DM. 1988. Effects of cyclosporine A on T cell development and clonal deletion. Science 241:1655–58

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

868

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

138. Gao E-K, Lo D, Cheney R, Kanagawa O, Sprent J. 1988. Abnormal differentiation of thymocytes in mice treated with cyclosporin A. Nature 336:176–79 139. Urdahl KB, Pardoll DM, Jenkins MK. 1994. Cyclosporin A inhibits positive selection and delays negative selection in α–β TCR transgenic mice. J. Immunol. 152:2853–59 140. Wang C-R, Hashimoto K, Kubo S, Yokochi T, Kubo M, Suzuki M, Suzuki K, Tada T, Nakayama T. 1995. T cell receptor-mediated signaling events in CD4+CD8+ thymocytes undergoing thymic selection: Requirement of calcineurin activation for thymic positive selection but not negative selection. J. Exp. Med. 181:927–41 141. Kane LP, Hedrick SM. 1996. A role for calcium influx in setting the threshold for CD4+CD8+ thymocyte negative selection. J. Immunol. 156:4594–601 142. Shi Y, Sahai BM, Green DR. 1989. Cyclosporin A inhibits activation-induced cell death in T cell hybridomas and thymocytes. Nature 339:625–26 143. Zhang W, Zimmer G, Chen JZ, Ladd D, Li E, Alt FW, Wiederrecht G, Cryan J, O’Neill EA, Seidman CE, Abbas AK, Seidman JG. 1996. T cell responses in calcineurin Aα-deficient mice. J. Exp. Med. 183:413–20 144. Xanthoudakis S, Viola JPB, Shaw KTY, Luo C, Wallace JD, Bozza PT, Curran T, Rao A. 1996. An enhanced immune response in mice lacking the transcription factor NFAT1. Science 272:892–95 145. Hodge MR, Ranger AM, Charles de la Brousse F, Hoey T, Grusby MJ, Glimcher LH. 1996. Hyperproliferation and dysregulation of IL-4 expression in NF-ATp-deficient mice. Immunity 4:1– 20 146. Yoshida H, Nishina H, Takimoto H, Mareng`ere LEM, Wakeham AC, Bouchard D, Kong Y-Y, Ohteki T, Shahinian A, Bachmann MF, Ohashi PS, Penninger JM, Crabtree GR, Mak TW. 1998. The transcription factor NFATc1 regulates lymphocyte proliferation and Th2 cytokine production. Immunity 8:115–24 147. Rinc´on M, Flavell RA. 1996. Regulation of AP-1 and NFAT transcription factors during thymic selection of T cells. Mol. Cell Biol. 16:1074–84 148. Matsuyama T, Kimura T, Kitagawa M, Pfeffer K, Kawakami T, Watanabe N, K¨undig TM, Amakawa R, Kishihara K, Wakeham A, Potter J, Furlonger CL, Narendran A, Suzuki H, Ohashi PS,

149.

150.

151.

152.

153.

154.

154a.

155.

156. 157. 158.

159.

Paige CJ, Taniguchi T, Mak TW. 1993. Targeted Disruption of IRF-1 or IRF-2 results in abnormal type I IFN gene induction and aberrant lymphocyte development. Cell 75:83–97 Penninger JM, Sirard C, Mittr¨ucker H-W, Chidgey A, Kozieradzki I, Nghiem M, Hakem A, Kimura T, Timms E, Boyd R, Taniguchi T, Matsuyama T, Mak TW. 1997. The interferon regulatory transcription factor IRF-1 controls positive and negative selection of CD8+ thymocytes. Immunity 7:243–54 Shao H, Kono DW, Chen L-Y, Rubin EM, Kaye J. 1997. Induction of the early growth response (Egr) family of transcription factors during thymic selection. J. Exp. Med. 185:731–44 Liu Z-G, Smith SW, McLaughlin KA, Schwartz LM, Osborne BA. 1994. Apoptotic signals delivered through the T-cell receptor of a T-cell hybrid require the immediate-early gene nur77. Nature 367:281–84 Woronicz JD, Calnan B, Ngo V, Winoto A. 1994. Requirement for the orphan steroid receptor Nur77 in apoptosis of T-cell hybridomas. Nature 367:277–81 Lee SL, Wesselschmidt RL, Linette GP, Kanagawa O, Russell JH, Milbrandt J. 1995. Unimpaired thymic and peripheral T cell death in mice lacking the nuclear receptor NGFI-B (Nur77). Science 269:532–34 Calnan BJ, Szychowski S, Chan FKM, Cado D, Winoto A. 1995. A role for the orphan steroid receptor Nur77 in apoptosis accompanying antigen-induced negative selection. Immunity 3:273–82 Zhou T, Cheng J, Yang Z, Liu C, Su X, Bluethmann H, Mountz JD. 1996. Inhibition of Nur77/Nurr1 leads to inefficient clonal delection of self-reactive T cells. J. Exp. Med. 183:1879–92 Weih F, Rysek R-P, Chen L, Bravo R. 1996. Apoptosis of nur77/N10-transgenic thymocytes involves the Fas/Fas ligand pathway. Proc. Natl. Acad. Sci. USA 93:5533–38 Kishimoto H, Sprent J. 1997. Negative selection in the thymus includes semimature T cells. J. Exp. Med. 185:263–71 Nagata S, Golstein P. 1995. The fas death factor. Science 267:1449–55 Singer GG, Abbas AK. 1994. The Fas antigen is involved in peripheral but not thymic deletion of T lymphocytes in T cell receptor transgenic mice. Immunity 1:365–71 Sytwu H-K, Liblau RL, McDevitt HO. 1996. The roles of Fas/APO-1

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

160.

161.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

162.

163.

164.

165.

166.

167.

168.

169.

(CD95) and TNF in antigen-induced programmed cell death in T cell receptor transgenic mice. Immunity 5:17–30 Castro JE, Listman JA, Jacobson BA, Wang Y, Lopez PA, Ju S, Finn PW, Perkins DL. 1996. Fas modulation of apoptosis during negative selection of thymocytes. Immunity 5:617–27 Kishimoto H, Surh CD, Sprent J. 1998. A role for Fas in negative selection of thymocytes in vivo. J. Exp. Med. 187:1427–38 Enari M, Talanian RV, Wong WW, Nagata S. 1996. Sequential activation of ICE-like and CPP32-like proteases during Fas-mediated apoptosis. Nature 380:723–26 Alam A, Braun MY, Hartgers F, Lesage S, Cohen L, Hugo P, Denis F, Sekaly RP. 1997. Specific activation of the cysteine protease CPP32 during the negative selection of T cells in the thymus. J. Exp. Med. 186:1503–12 Sarin A, Wu M-L, Henkart PA. 1996. Different interleukin-1β converting enzyme (ICE) family protease requirements for the apoptotic death of T lymphocytes triggered by diverse stimuli. J. Exp. Med. 184:2445–50 Clayton LK, Ghendler Y, Mizoguchi E, Patch RJ, Ocain TD, Orth K, Bhan AK, Dixit VM, Reinherz EL. 1997. T-cell receptor ligation by peptide/MHC induces activation of a caspase in immature thymocytes: the molecular basis of negative selection. EMBO J. 16:2282–93 Kuida K, Zheng TS, Na S, Kuan C, Yang D, Karasuyama H, Rakic P, Flavell RA. 1996. Decreased apoptosis in the brain and premature lethality in CPP32deficient mice. Nature 384:368–72 Woo M, Hakem R, Soengas MS, Duncan GS, Shahinian A, K¨agi D, Hakem A, McCurrach M, Khoo W, Kaufman SA, Senaldi G, Howard T, Lowe SW, Mak TW. 1998. Essential contribution of caspase 3/CPP32 to apoptosis and its associated nuclear changes. Genes Dev. 12:806–19 Page DM, Roberts EM, Peschon JJ, Hedrick SM. 1998. TNF receptordeficient mice reveal striking differences between several models of thymocyte negative selection. J. Immunol. 160:120–33 Amakawa R, Hakem A, K¨undig TM, Matsuyama T, Simard JJL, Timms E, Wakeham A, Mittruecker HW, Griesser H, Takimoto H, Schmits R, Shahinian A, Ohashi PS, Penninger JM, Mak TW. 1996. Impaired negative selection of

170.

171.

172.

173.

174.

175.

176.

177.

178.

179.

869

T cells in Hodgkin’s disease antigen CD30-deficient mice. Cell 84:551–62 Foy TM, Page DM, Waldschmidt TJ, Schoneveld A, Laman JD, Masters SR, Tygrett L, Ledbetter JA, Aruffo A, Claassen E, Xu JC, Flavell RA, Oehen S, Hedrick SM, Noelle RJ. 1995. An essential role for gp39, the ligand for CD40, in thymic selection. J. Exp. Med. 182:1377–88 Foy TM, Aruffo A, Bajorath J, Buhlmann JE, Noelle RJ. 1996. Immune regulation by CD40 and its ligand GP39. Annu. Rev. Immunol. 14:591–617 Degermann S, Surh CD, Glimcher LH, Sprent J, Lo D. 1994. B7 expression on thymic medullary epithelium correlates with epithelium-mediated deletion of Vβ5+ thymocytes. J. Immunol. 152:3254–63 Shahinian A, Pfeffer K, Lee KP, K¨undig TM, Kishihara K, Wakeham A, Kawai K, Ohashi PS, Thompson CB, Mak TW. 1993. Differential T cell costimulatory requirements in CD28-deficient mice. Science 261:609–12 Walunas T, Sperling A, Khattri R, Thompson C, Bluestone J. 1996. CD28 expression is not essential for positive and negative selection of thymocytes or peripheral T cell tolerance. J. Immunol. 156:1006–13 Vukmanovic S, Stella G, King PD, Dyall R, Hogquist KA, Harty JT, NikolicZugic J, Bevan MJ. 1994. A positively selecting thymic epithelial cell line lacks costimulatory activity. J. Immunol. 152:3814–23 Tan R, Teh S-J, Ledbetter JA, Linsley PS, Teh HS. 1992. B7 costimulates proliferation of CD4−CD8+ T lymphocytes but is not required for the deletion of immature CD4+8+ thymocytes. J. Immunol. 149:3217–24 Page DM, Kane LP, Allison JP, Hedrick SM. 1993. Two signals are required for negative selection of CD4+CD8+ thymocytes1. J. Immunol. 151:1868–80 Punt JA, Osborne BA, Takahama Y, Sharrow SO, Singer A. 1994. Negative selection of CD4+CD8+ thymocytes by T cell receptor-induced apoptosis requires a costimulatory signal that can be provided by CD28. J. Exp. Med. 179:709–13 Punt JA, Havran W, Abe R, Sarin A, Singer A. 1997. T cell receptor (TCR)induced death of immature CD4+CD8+ thymocytes by two distinct mechanisms differing in their requirement for CD28 costimulation: implications for negative

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

870

180.

181.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

182.

183.

184.

185.

186.

187.

188.

189.

190.

191.

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL selection in the thymus. J. Exp. Med. 186:1911–22 Amsen D, Kruisbeek AM. 1996. CD28B7 interactions function to co-stimulate clonal deletion of double-positive thymocytes. Int. Immunol. 8:1927–36 Lesage S, Charron J, Winslow G, Hugo P. 1997. Induction of thymocyte deletion by purified single peptide/major histocompatibility complex ligands. J. Immunol. 159:2078–81 Kishimoto H, Cai Z, Brunmark A, Jackson MR, Peterson PA, Sprent J. 1996. Differing roles for B7 and intercellular adhesion molecule-1 in negative selection of thymocytes. J. Exp. Med. 184:531–37 Su B, Jacinto E, Hibi M, Kallunko T, Karin M, Ben-Neriah Y. 1994. JNK is involved in signal integration during costimulation of T lymphocytes. Cell 77:727–36 Nishina H, Fischer KD, Radvanyi L, Shahinian A, Hakem R, Rubie EA, Bernstein A, Mak TW, Woodgett JR, Penninger JM. 1997. Stress-signalling kinase Sek1 protects thymocytes from apoptosis mediated by CD95 and CD3. Nature 385:350–53 Swat W, Fujikawa K, Ganiatsas S, Yang D, Xavier RJ, Harris NL, Davidson L, Ferrini R, Davis RJ, Labow MA, Flavell RA, Zon LI, Alt FW. 1998. SEK1/MKK4 is required for maintenance of a normal peripheral lymphoid compartment but not for lymphocyte development. Immunity 8:625–34 Thomis DC, Berg LJ. 1997. The role of Jak3 in lymphoid development, activation, and signaling. Curr. Opin. Immunol. 9:541–47 Saijo K, Park SY, Ishida Y, Arase H, Saito T. 1997. Crucial role of Jak3 in negative selection of self-reactive T cells. J. Exp. Med. 185:351–56 DiSanto JP, Guy-Grand D, Fisher A, Tarakhovsky A. 1996. Critical role for the common cytokine receptor γ chain in intrathymic and peripheral T cell selection. J. Exp. Med. 183:1111–18 Laufer TM, DeKoning J, Markowitz JS, Lo D, Glimcher LH. 1996. Unopposed positive selection and autoreactivity in mice expressing class II MHC only on thymic cortex. Nature 383:81–85 van Meerwijk JPM, Marguerat S, Lees RK, Germain RN, Fowlkes BJ, MacDonald HR. 1997. Quantitative impact of thymic clonal deletion on the T cell repertoire. J. Exp. Med. 185:377–84 Ignatowicz L, Kappler J, Marrack P.

192.

193.

194.

195.

196.

197.

198.

199.

200.

201.

202.

1996. The repertoire of T cells shaped by a single MHC/peptide ligand. Cell 84:521–29 Tourne S, Miyazaki T, Oxenius A, Klein L, Fehr T, Kyewski B, Benoist C, Mathis D. 1997. Selection of a broad repertoire of CD4+ T cells in H-2Ma0/0 mice. Immunity 7:187–95 Surh CD, Lee D-S, Fung-Leung W-P, Karlsson L, Sprent J. 1997. Thymic selection by a single MHC/peptide ligand produces a semidiverse repertoire of CD4+ T cells. Immunity 7:209–19 Merkenschlager M, Graf D, Lovatt M, Bommhardt U, Zamoyska R, Fisher AG. 1997. How many thymocytes audition for selection? J. Exp. Med. 186:1149– 58 Ohashi PS, Pircher H, B¨urki K, Zinkernagel RM, Hengartner H. 1990. Distinct sequence of negative or positive selection implied by thymocyte T cell receptor densities. Nature 346:861–63 Borgulya P, Kishi H, Muller U, Kirberg J, von Boehmer H. 1991. Development of the CD4 and CD8 lineage of T cells: instruction versus selection. EMBO J. 10:913–18 Guidos CJ, Danska JS, Fathman CG, Weissman IL. 1990. T cell receptormediated negative selection of autoreactive T lymphocyte precursors occurs after commitment to the CD4 or CD8 lineages J. Exp. Med. 172:835–45 Pircher H, B¨urki K, Lang R, Hengartner H, Zinkernagel R. 1989. Tolerance induction in double specific T-cell receptor transgenic mice varies with antigen. Nature 342:559–61 White J, Herman A, Pullen AM, Kubo R, Kappler JW, Marrack P. 1989. The Vβ -specific superantigen staphylococcal enterotoxin B: stimulation of mature T cells and clonal deletion in neonatal mice. Cell 56:27–35 Ghendler Y, Hussey RE, Witte T, Mizoguchi E, Clayton LK, Bhan AK, Koyasu S, Chang H-C, Reinherz EL. 1997. Double-positive T cell receptorhigh thymocytes are resistant to peptide/major histocompatibility complex ligand-induced negative selection. Eur. J. Immunol. 27:2279–89 Marrack P, Ignatowicz L, Kappler JW, Boymel J, Freed JH. 1993. Comparison of peptides bound to spleen and thymus class II. J. Exp. Med. 178:2173–83 Ignatowicz L, Rees W, Pacholczyk R, Ignatowicz H, Kushnir E, Kappler J, Marrack P. 1997. T cells can be activated by peptides that are unrelated in

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

203.

204.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

205.

206.

207.

208.

209.

210. 211.

212.

213.

214.

sequence to their selecting peptide. Immunity 7:179–86 Zerrahn J, Held W, Raulet DH. 1997. The MHC reactivity of the T cell repertoire prior to positive and negative selection. Cell 88:627–36 Bevan MJ. 1997. In thymic selection, peptide diversity gives and takes away. Immunity 7:175–78 Lin S-Y, Ardouin L, Gillet A, Malissen M. 1997. The single positive T cells found in CD3-ζ /η−/− mice overtly react with self-major histocompatibility complex molecules upon restoration of normal surface density of T cell receptorCD3 complex. J. Exp. Med. 185:707–15 Kisielow P, Miazek A. 1995. Positive selection of T cells: rescue from programmed cell death and differentiation require continual engagement of the T cell receptor. J. Exp. Med. 181:1975–84 Wilkinson RW, Anderson G, Owen JJT, Jenkinson EJ. 1995. Positive selection of thymocytes involves sustained interactions with the thymic microenvironment. J. Immunol. 155:5234–40 Schmitt S, M¨uller K-P, Kyewski BA. 1997. Two separable T cell receptor signals reconstitute positive selection of CD4 lineage T cells in vivo. Eur. J. Immunol. 27:2139–44 Engelhard VH. 1994. Structure of peptides associated with class I and class II MHC molecules. Annu. Rev. Immunol. 12:181–207 Ohashi PS. 1996. T cell selection and autoimmunity: flexibility and tuning. Curr. Opin. Immunol. 8:808–14 Henderson RA, Michel H, Sakaguchi K, Shabanowitz J, Appella E, Hunt DF, Engelhard VH. 1992. HLA-A2. 1-associated peptides from a mutant cell line: a second pathway of antigen presentation. Science 255:1264–66 Wei ML, Cresswell P. 1992. HLAA2 molecules in an antigen-processing mutant cell contains signal sequencederived peptides. Nature 356:443–46 Sandberg J, Chambers B, Van Kaer L, K¨arre K, Ljunggren H-G. 1996. TAP1deficient mice select a CD8+ T cell repertoire that displays both diversity and peptide specificity. Eur. J. Immunol. 26:288–93 Apasov S, Sitkovsky M. 1993. Highly lytic CD8+, αβ T-cell receptor cytotoxic T cells with major histocompatibility complex (MHC) class I antigen-directed cytotoxicity in β 2-microglobulin, MHC class I-deficient mice. Proc. Natl. Acad. Sci. USA 90:2837–41

871

215. Aldrich CJ, Ljunggren H-G, Van Kaer L, Ashton-Rickardt PG, Tonegawa S, Forman J. 1994. Positive selection of selfand alloreactive CD8+ T cells in Tap-1 mutant mice. Proc. Natl. Acad. Sci. USA 91:6525–28 216. Glas R, Ohlen C, Hoglund P, K¨arre K. 1994. The CD8+ T cell repertoire in β 2-microglobulin-deficient mice is biased towards reactivity against self-major histocompatibility class I. J. Exp. Med. 179:661–72 217. Cook JR, Solheim JC, Connolly JM, Hansen TH. 1995. Induction of peptidespecific CD8+ CTL clones in β 2-microglobulin-deficient mice. J. Immunol. 154:47–57 218. Tallquist MD, Weaver AJ, Pease LR. 1998. Degenerate recognition of alloantigenic peptides on a positiveselecting class I molecule. J. Immunol. 160:802–9 219. Miyazaki T, Wolf P, Tourne S, Waltzinger C, Dierich A, Barois N, Ploegh H, Benoist C, Mathis D. 1996. Mice lacking H2-M complexes, enigmatic elements of the MHC Class II peptide-loading pathway. Cell 84:531– 41 220. Martin DW, Hicks GG, Mendiaratta SK, Leva HI, Ruley HE, Van Kaer L. 1996. H2-M mutant mice are defective in the peptide loading of Class II molecules, antigen presentation, and T cell repertoire selection. Cell 84:543–50 221. Fung-Leung W-P, Surh CD, Liljedahl M, Pang J, Letrucq D, Peterson A, Webt SR, Karisson L. 1996. Antigen presentation and T cell development in H2-Mdeficient mice. Science 271:1278–81 222. Sloan VS, Cameron P, Porter G, Gammon M, Amaya M, Mellins E, Zaller DM. 1995. Mediation by HLA-DM of dissociation of peptides from HLA-DR. Nature 375:802–6 223. Denzin LK, Cresswell P. 1995. HLADM induces CLIP dissociation from MHC class II αβ dimers and facilitates peptide loading. Cell 82:155–65 224. Sherman MA, Weber DA, Jensen PE. 1995. DM enhances peptide binding to class II MHC by release of invariant chain-derived peptide. Immunity 3:197– 205 225. Mathis D, Benoist C, Williams V, Kanter M, McDevitt HO. 1983. Several mechanisms can account for defective Eα gene expression in different mouse haplotypes. Proc. Natl. Acad. Sci. USA 80:273–77 226. Gapin L, Fukui Y, Kanellopoulos J, Sano

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

872

227.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

228.

229.

230.

231. 232.

233. 234.

235.

235a.

236.

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL T, Casrouge A, Malier V, Beaudoing E, Gautheret D, Claverie J-M, Sasazuki T, Kourilsky P. 1998. Quantitative analysis of the T cell repertoire selected by a single peptide-major histocompatibility complex. J. Exp. Med. 187:1871–83 Grubin CE, Kovats S, deRoos P, Rudensky AY. 1997. Deficient positive selection of CD4 T cells in mice displaying altered repertoires of MHC class II-bound self-peptides. Immunity 7:197–208 Tourne S, Nakano N, Viville S, Benoist C, Mathis D. 1995. The influence of invariant chain on the positive selection of single T cell receptor specificities. Eur. J. Immunol. 25:1851–56 Sant’Angelo DB, Waterbury PG, Cohen BE, Martin WD, Kaer LV, Hayday AC, Janeway CA Jr. 1997. The imprint of intrathymic self-peptides on the mature T cell receptor repertoire. Immunity 7:517–24 Liu C-P, Parker D, Kappler J, Marrack P. 1997. Selection of antigen-specific T cells by a single IEk peptide combination. J. Exp. Med. 186:1441–50 Kersh GJ, Allen PM. 1996. Essential flexibility in the T-cell recognition of antigen. Nature 380:495–98 Pawlowski TJ, Singleton MD, Loh DY, Berg R, Staerz UD. 1996. Permissive recognition during positive selection. Eur. J. Immunol. 26:851–57 Bjorkman PJ. 1997. MHC restriction in three dimensions: a view of T cell receptor/ligand interactions. Cell 89:167–70 Garcia KC, Degano M, Pease LR, Huang M, Peterson PA, Teyton L, Wilson IA. 1998. Structural basis of plasticity in T cell receptor recognition of a self peptide-MHC antigen. Science 279:1166–72 Ding Y-H, Smith KJ, Garboczi DN, Utz U, Biddison WE, Wiley DC. 1998. Two human T cell receptors bind in a similar diagonal mode to the HLA-A2/tax peptide complex using different TCR amino acids. Immunity 8:403–11 Teng M-K, Smolyar A, Tse AGD, Liu J-H, Liu J, Hussey RE, Nathenson SG, Chang H-C, Reiinherz EL, Wang JH. 1998. Crystal structure of the N15 TCR is complex with its peptide/MHC (pMHC) ligand: identification of a common docking topology with substantial variation among different TCR-pMHC complexes. Curr. Biol. 6:409–12 Strasser A, Harris AW, Cory S. 1991. Bcl-2 transgene inhibits T cell death and perturbs thymic self-censorship. Cell 67:889–99

237. Gratiot-Deans J, Merino R, Nunez G, Turka LA. 1994. Bcl-2 expression during T-cell development: Early loss and late return occur at specific stages of commitment to differentiation and survival. Proc. Natl. Acad. Sci. USA 91:10685–89 238. Veis DJ, Sentman CL, Bach EA, Korsmeyer SJ. 1993. Expression of the bcl-2 protein in murine and human thymocytes and in peripheral T lymphocytes. J. Immunol. 151:2546–54 239. Veis DJ, Sorenson CM, Shutter JR, Korsmeyer SJ. 1993. Bcl-2 deficient mice demonstrate fulminant lymphoid apoptosis, polycystic kidneys, and hypopigmented hair. Cell 75:229–40 240. Nakayama KI, Nakayama K, Dustin LB, Loh DY. 1995. T-B cell interaction inhibits spontaneous apoptosis of mature lymphocytes in Bcl-2 deficient mice. J. Exp. Med. 182:1101–9 241. Maraskovsky E, O’Reilly LA, Teepe M, Corcoran LM, Peschon JJ, Strasser A. 1997. Bcl-2 can rescue T lymphocyte development in interleukin-7 receptordeficient mice but not in mutant rag-1−/− mice. Cell 89:1011–19 242. Akashi K, Kondo M, von Freeden-Jeffry U, Murray R, Weissman IL. 1997. Bcl-2 rescues T lymphopoiesis in interleukin7 receptor-deficient mice. Cell 89:1033– 41 243. Linette GP, Grusby MJ, Hedrick SM, Hansen TH, Glimcher LH, Korsmeyer SJ. 1994. BcI-2 is upregulated at the CD4+ CD8+ stage during positive selection and promotes thymocyte differentiation at several control points. Immunity 1:197–205 244. Tao W, Teh S-J, Melhado I, Jirik F, Korsmeyer SJ, Teh H-S. 1994. The T cell receptor repertoire of CD4−8+ thymocytes is altered by overexpression of the BCL-2 protooncogene in the thymus. J. Exp. Med. 179:145–53 245. Sentman CL, Shutter JR, Hockenbery D, Kanagawa O, Korsmeyer SJ. 1991. Bcl2 inhibits multiple forms of apoptosis but not negative selection in thymocytes. Cell 67:879–88 246. Strasser A, Harris AW, von Boehmer H, Cory S. 1994. Positive and negative selection of T cells in T-cell receptor transgenic mice expressing a bcl-2 transgene. Proc. Natl. Acad. Sci. USA 91:1376– 80 247. Chao DT, Korsmeyer SJ. 1998. BCL-2 family: regulators of cell death. Annu. Rev. Immunol. 16:395–419 248. Brady HJM, Gil-Gomez G, Kirberg J,

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

POSITIVE AND NEGATIVE THYMOCYTE SELECTION

249.

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

250.

251.

252.

253.

254.

255.

256. 257.

258.

259.

260.

Berns AJM. 1996. Baxα perturbs T cell development and affects cell cycle entry of T cells. EMBO J. 15:6991–7001 Grillot DAM, Merino R, N´un˜ ez G. 1995. Bcl-xL displays restricted distribution during T cell development and inhibits multiple forms of apoptosis but not clonal deletion in transgenic mice. J. Exp. Med. 182:1973–83 Yang X-F, Weber GF, Cantor H. 1997. A novel Bcl-x isoform connected to the T cell receptor regulates apoptosis in T cells. Immunity 7:629–39 Henning SW, Galandrini R, Hall A, Cantrell DA. 1997. The GTPase Rho has a critical regulatory role in thymus development. EMBO J. 16:2397–407 Muthusamy N, Barton K, Leiden JM. 1995. Defective activation and survival of T cells lacking the Ets-1 transcription factor. Nature 377:639–42 Bories J-C, Willerford DM, Gr´evin D, Davidson L, Camus A, Martin P, St´ehelln D, Alt FW. 1995. Increased T-cell apoptosis and terminal B-cell differentiation induced by inactivation of the Ets-1 proto-oncogene. Nature 377:635–38 Field SJ, Tsai FY, Kuo F, Zubiaga AM, Kaelin WG Jr, Livingston DM, Orkin SH, Greenberg ME. 1996. E2F-1 functions in mice to promote apoptosis and suppress proliferation. Cell 85:549– 61 Grossman Z, Singer A. 1996. Tuning of activation thresholds explains flexibility in the selection and development of T cells in the thymus. Proc. Natl. Acad. Sci. USA 93:14747–52 Weiss A, Littman DR. 1994. Signal transduction by lymphocyte antigen receptors. Cell 76:263–74 Lee NA, Loh DY, Lacy E. 1992. CD8 surface levels alter the fate of α/β T cell receptor-expressing thymocytes in transgenic mice. J. Exp. Med. 175:1013– 25 Robey EA, Ramsdell F, Kioussis D. 1992. The level of CD8 expression can determine the outcome of thymic selection. Cell 69:1089–96 Marusic Galesic S, Stephany DA, Longo DL, Kruisbeek AM. 1988. Development of CD4−CD8+ cytotoxic T cells requires interactions with class I MHC determinants. Nature 333:180–83 Zuniga-Pfl¨ucker JC, McCarthy SA, Weston M, Longo DL, Singer A, Kruisbeek AM. 1989. Role of CD4 in thymocyte selection and maturation. J. Exp. Med. 169:2085–96

873

261. Fung-Leung WP, Schilham MW, Rahemtulla A, K¨undig TM, Vollenweider M, Potter J, van Ewijk W, Mak TW. 1991. CD8 is need for development of cytotoxic T cells but not helper T cells. Cell 65:443–49 262. Rahemtulla A, Fung-Leung WP, Schilham MW, K¨undig TM, Sambhara SR, Narendran A, Arabian A, Wakeham A, Paige CJ, Zinkernagel RM, Miller RG, Mak TW. 1991. Normal development and function of CD8+ cells but markedly decreased helper cell activity in mice lacking CD4. Nature 353:180– 84 263. Killeen N, Littman DR. 1993. Helper T-cell development in the absence of CD4-p56lck association. Nature 364: 729–32 264. Mitnacht R, Bischof A, Torres-Nagel N, H¨unig T. 1998. Opposite CD4/CD8 lineage decisions of CD4+8+ mouse and rat thymocytes to equivalent triggering signals: correlation with thymic expression of a truncated CD8α chain in mice but not rats. J. Immunol. 160:700–7 265. Tarakhovsky A, M¨uller W, Rajewsky K. 1994. Lymphocyte populations and immune responses in CD5-deficient mice. Eur. J. Immunol. 24:1678–84 266. Tarakhovsky A, Kanner SB, Hombach J, Ledbetter JA, M¨uller W, Killeen N, Rajewsky K. 1995. A role for CD5 in TCR-mediated signal transduction and thymocyte selection. Science 269:535– 37 267. Shores EW, Love PE. Personal communication. Submitted 268. Plas DR, Johnson R, Pingel JT, Matthews RJ, Dalton M, Roy G, Chan AC, Thomas ML. 1996. Direct regulation of ZAP-70 by SHP-1 in T cell antigen receptor signaling. Science 272: 1173–76 269. Pani G, Fischer K-D, Mlinaric-Rascan I, Siminovitch KA. 1996. Signaling capacity of the T cell antigen receptor is negatively regulated by the PTP1C tyrosine phosphatase. J. Exp. Med. 184:839–52 270. Lorenz U, Ravichandran KS, Burakoff SJ, Neel BG. 1996. Lack of SHPTP1 results in src-family kinase hyperactivation and thymocyte hyperresponsiveness. Proc. Natl. Acad. Sci. USA 93: 9624–29 271. Kawai K, Ohashi PS. 1995. Immunological function of a defined T-cell population tolerized to low-affinity self antigens. Nature 374:68–69 272. Nossal GJV. 1994. Negative selection of lymphocytes. Cell 76:229–39

P1: SKH/spd

P2: SKH/KKK/ary

January 25, 1999

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

874

12:31

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-25

SEBZDA ET AL

273. Baum H, Davies H, Peakman M. 1996. Molecular mimicry in the MHC: hidden clues to autoimmunity? Immunol. Today 17:64–70 274. Wekerle H, Bradl M, Linington C, Kaab G, Kojima K. 1996. The shaping of the brain-specific T lymphocyte repertoire in the thymus. Immunol. Rev. 149:231– 43 275. Fairchild PJ, Wraith DC. 1996. Lowering the tone: mechanisms of immunodominance among epitopes with low affinity for MHC. Immunol. Today 17:80–91 276. Le Douarin N, Corbel C, Bandeira A, Thomas-Vaslin V, Modigliani Y, Coutinho A, Sala¨un J. 1996. Evidence for a thymus-dependent form of tolerance that is not based on elimination or anergy of reactive T cells. Immunol. Rev. 149:35–53 277. Tolosa E, King LB, Ashwell JD. 1998. Thymocyte glucocorticoid resistance alters positive selection and inhibits autoimmunity and lymphoproliferative disease in MRL-lpr/lpr mice. Immunity 8:67–76 278. Watanabe Y, Naiki M, Wilson T, Godfrey D, Chiang B, Boyd R, Ansari A, Gershwin ME. 1993. Thymic microenvironmental abnormalities and thymic selection in NZB. H-2bm12 mice. J. Immunol. 150:4702–12 279. Thomas-Vaslin V, Damotte D, Coltey M, Le Douarin NM, Coutinho A, Sala¨un J. 1997. Abnormal T cell selection on thymic epithelium is sufficient to induce autoimmune manifestations in C57BL/6 athymic nude mice. Proc. Natl. Acad. Sci. USA 94:4598–603 280. Ohteki T, Nishina H, Hessel A, Sebzda E, Ho S, Hugo P, Penninger J, Ohashi PS. 1998. Positive selection of autoreactive T cells. In preparation 281. L¨uhder F, Katz J, Benoist C, Mathis D. 1998. Major histocompatibility complex class II molecules can protect from diabetes by positively selecting T cells with additional specificities. J. Exp. Med. 187:379–87 282. Akashi K, Kondo M, Weissman IL. 1998. Two distinct pathways of posi-

283.

284.

285.

286.

287.

288.

289.

290.

291.

tive selection for thymocytes. Proc. Natl. Acad. Sci. USA 95:2486–91 Dave VP, Cao Z, Browne C, Alarcon B, Fernandez-Miguel G, Lafaille J, De la Hera A, Tonegawa S, Kappes DJ. 1997. CD3δ deficiency arrests development of the αβ but not the γ δ T cell lineage. EMBO J. 16:1360–70 Shores EW, Huang K, Tran T, Lee E, Grinberg A, Love PE. 1994. Role of TCR ζ chain in T cell development and selection. Science 266:1047–1050 Stein PL, Lee H-M, Rich S, Soriano P. 1992. pp59fyn mutant mice display differential signaling in thymocytes and peripheral T cells. Cell 70:741–50 Appleby MW, Gross JA, Cooke MP, Levin SD, Qian X, Perlmutter RM. 1992. Defective T cell receptor signaling in mice lacking the thymic isoform of p59fyn. Cell 70:751–63 Bhatia SK, Tygrett LT, Grabstein KH, Waldschmidt TJ. 1995. The effect of in vivo IL-7 deprivation on T cell maturation. J. Exp. Med. 181:1399–1409 Peschon JJ, Morrissey PJ, Grabstein KH, Ramsdell FJ, Maraskovsky E, Gliniak BC, Park LS, Ziegler SF, Williams DE, Ware CB, Meyer JD, Davison BL. 1994. Early lymphocyte expansion is severely impaired in interleukin 7 receptor-deficient mice. J. Exp. Med. 180:1955–60 Molina TJ, Kishihara K, Siderovski DP, van Ewijk W, Narendran A, Timms E, Wakeham A, Paige CJ, Hartmann K-U, Veillette A, Davidson D, Mak TW. 1992. Profound block in thymocyte development in mice lacking p56lck. Nature 357:161–64 Cheng AM, Rowley RB, Pao W, Hayday A, Bolen JB, Pawson T. 1995. Syk tyrosine kinase required for mouse viability and B-cell development. Nature 378:303–6 Turner M, Mee PJ, Costello PS, Williams O, Price AA, Duddy LP, Furlong MT, Geahlen RL, Tybulewicz VLJ. 1995. Perinatal lethality and blocked B-cell development in mice lacking the tyrosine kinase Syk. Nature 378:298– 302

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:829-874. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: APL/NBL/vks

January 22, 1999

P2: APR/MBG

QC: APR/anil

T1: APR

15:24

Annual Reviews

AR078-26

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:875–904 c 1999 by Annual Reviews. All rights reserved Copyright °

REGULATION OF IMMUNE RESPONSES THROUGH INHIBITORY RECEPTORS Eric O. Long Laboratory of Immunogenetics, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Rockville, Maryland 20852; e-mail: [email protected] KEY WORDS:

C-type lectin, immunoglobulin, MHC class I, natural killer cell

ABSTRACT Major histocompatibility complex class I-specific inhibitory receptors on natural killer cells prevent the lysis of healthy autologous cells. The outcome of this negative signal is not anergy or apoptosis of natural killer cells but a transient abortion of activation signals. The natural killer inhibitory receptors fulfill this function by recruiting the tyrosine phosphatase SHP-1 through a cytoplasmic immunoreceptor tyrosine-based inhibition motif. This immunoreceptor tyrosine-based inhibition motif has become the hallmark of a growing family of receptors with inhibitory potential, which are expressed in various cell types such as monocytes, macrophages, dendritic cells, leukocytes, and mast cells. Most of the natural killer inhibitory receptors and two members of a monocyte inhibitory-receptor family bind major histocompatibility complex class I molecules. Ligands for many of the other receptors have yet to be identified. The inhibitory-receptor superfamily appears to regulate many types of immune responses by blocking cellular activation signals.

INTRODUCTION Biological responses are often controlled and cellular functions sustained by a balance of positive and negative signals. This balance is quite evident in the immune system, which is constantly kept in check by responding to activating and inhibitory signals and by subjecting cells to positive and negative selection. This regulation is manifested at the cellular level through a combination of signals from receptors and coreceptors. T cells can be turned off 875 0732-0582/99/0410-0875$08.00

P1: APL/NBL/vks

January 22, 1999

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

876

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

(to prevent responsiveness to self antigens and to terminate responses to foreign antigens) in different ways: by a lack of costimulation by CD28, by CTLA-4– mediated signals that counteract CD28 (1), by upregulation of the ligand of the death-inducing receptor fas, or by suppression mediated by cytokines such as interleukin (IL)-10 and transforming growth factor-β1 (2). Similarly, signals from the B cell receptor (BCR) are positively modulated by the CD19/CD21 complex and down modulated by CD22 (3). T- and B-cell antigen receptors are also regulated by a developmental switch in the signal outcome; immature B and T cells are deleted after antigen receptor cross-linking, whereas mature cells are induced to proliferate (4). Negative regulation is also required at the biochemical level to terminate signals delivered by receptors and to return the receptors to their basal state of activation. For instance, biochemical signals that depend on the activity of tyrosine kinases can be down modulated by the activity of tyrosine phosphatases. Growth inhibition, anergy, apoptosis, and down modulation of receptor signals are all part of the inhibitory tools used by the immune system to maintain homeostasis. In addition, the importance of another type of negative regulation used to prevent unwanted effector functions, such as activation of cytolysis and cytokine production, has been appreciated recently. This negative regulation is mediated by specialized inhibitory receptors that block signaling cascades initiated by separate activation receptors. Because they interrupt signal transduction at an early step, these inhibitory receptors leave little evidence of their action. Despite its recent characterization, this type of tight negative regulation has already been implicated in the control of a number of different kinds of cells. Furthermore, different types of receptors can mediate this inhibition. The importance of such receptors was first appreciated in natural killer (NK) cells. Because inhibitory receptors on NK cells have been extensively reviewed recently (5–7), this review is restricted mainly to recent findings on NK receptors and on the expansion of inhibitory receptors as a new superfamily.

THE INHIBITORY-RECEPTOR SUPERFAMILY The term “inhibitory-receptor superfamily” (IRS) was coined by Lanier in last year’s volume of Annual Review of Immunology (6) to describe the expanding group of receptors that block activation of a number of cell types in the immune system. Membership in the IRS is defined by a few characteristic properties. First, these receptors inhibit activation receptors in trans (Figure 1; 8, 9). Second, they do so by recruiting phosphatases through an immunoreceptor tyrosine-based inhibition motif (ITIM). Finally, inhibition can target different activation pathways but requires coligation with the activating receptor.

P1: APL/NBL/vks

January 22, 1999

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INHIBITORY RECEPTORS

877

Figure 1 Activation receptor A with an immunoreceptor tyrosine-based activation motif (ITAM) is inhibited in trans by inhibitory receptor B with an immunoreceptor tyrosine-based inhibition motif (ITIM).

The IRS is subdivided into two structural types of molecules (Figure 2). One consists of immunoglobulin (Ig)-superfamily inhibitory receptors (ISIR), the other of C-type lectin inhibitory receptors (CLIR). A feature common to most members of the IRS is that receptor isoforms with activation properties are also expressed. Different approaches have led to the recent molecular cloning of several receptors in the IRS, some identifying the gene for a function, some identifying a function for a cDNA, and others identifying substrates of phosphatases. Many of the IRS members are not conserved in evolution such that counterparts of human receptors have not been identified in the mouse and vice versa. This sequence divergence suggests a need for rapid adaptation, possibly owing to variability in the receptor ligands.

The Natural Killer Cell Paradigm The important functions of NK cells place them at the interface between the innate and adaptive immune systems. NK cells respond rapidly to interferon (IFN)-α/β and to IL-12 during infections by viruses and other intracellular pathogens. Activated NK cells produce lymphokines such as IFN-γ and tumor necrosis factor α that regulate immune defenses. In addition, NK cells kill certain tumor and virally infected cells. The cytolytic activity and the lymphokine production by NK cells are under tight regulation by inhibitory receptors that are specific for major histocompatibility complex (MHC) class I antigens. NK cells kill hematopoietic target cells unless they receive an inhibitory signal from

P1: APL/NBL/vks

January 22, 1999

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

878

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

Figure 2 Relationships among inhibitory-receptor superfamily (IRS), immunoglobulinsuperfamily inhibitory receptors (ISIR), C-type lectin inhibitory receptors (CLIR), and paired activation/inhibition receptors.

an MHC class I-specific receptor. This concept of NK cells recognizing “missing self” was originally proposed by Ljunggren & K¨arre (10). The first such receptors to be identified were the Ly49A molecule on mouse NK cells (11) and the p58 receptors on human NK cells (12). The cytotoxic activity of NK cells can be triggered by many different activation receptors. NK cells are potent mediators of antibody-mediated cellular cytotoxicity on activation by Fcγ receptors (CD16), but the receptors involved in natural killing of target cells are not well characterized. In either case, binding of an NK inhibitory receptor to MHC class I on the target cells blocks cytolysis. Triggering of NK cytotoxicity is dependent on the activity of an src-family kinase (13). Such tyrosine kinases can phosphorylate immunoreceptor tyrosine-based activation motifs (ITAM) in activating-receptor complexes. In turn, phosphorylated ITAMs recruit and activate tyrosine kinases of the ZAP70/syk family. Syk activation occurs within 5 minutes in NK cells mixed with sensitive target cells, and functional experiments have implicated syk in

P1: APL/NBL/vks

January 22, 1999

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INHIBITORY RECEPTORS

879

the triggering of NK natural cytotoxicity (14). Signals received from a combination of receptors may be necessary to achieve NK activation (15). For example, CD28 and CD11a/18 act in concert to mediate activation of an NK cell line (16). Ubiquitous receptors such as CD2, CD69, and DNAM-1 can activate NK cytotoxicity. In mouse NK cells, a receptor for B7-1 (distinct from CD28) activates cytotoxicity of target cells expressing high levels of B7-1 (17). An important role for the adhesion molecule CD11a/18 in NK activation was also suggested by the observation that redistribution of the intercellular adhesion molecule-2 (a ligand for CD11a/18) into uropods of target cells by the cytoskeletal linker protein ezrin resulted in enhanced sensitivity to killing by NK cells (18). Two activation receptors named p46 (19) and NKp44 (20) are expressed specifically in NK cells. The ligands for p46 and NKp44 are not known. Whereas p46 is expressed on both resting and activated NK cells, NKp44 is expressed only on IL-2–activated NK cells. p46 and NKp44 may cooperate in activating NK lysis of tumor target cells (20). These two receptors associate with different tyrosine-phosphorylated proteins, most likely the ITAM-containing ζ or γ chain for p46 and the DAP12 protein for NKp44 (20). Coengagement of NK inhibitory receptors with any one of the many activation receptors tested so far has resulted in inhibition of the activation signal. However, experimental evidence for such inhibition has often involved the use of antibodies to cross-link the two types of receptors. In any case, whichever NK surface molecules are engaged during interaction with target cells, their activation signals are aborted by the MHC class I-specific inhibitory receptors. These inhibitory receptors have the important function of blocking many potential NK responses to protect normal cells from being killed and to limit the production of the inflammatory cytokines IFN-γ and tumor necrosis factor α. By blocking activation signals at a very early point, these MHC class I-specific inhibitory receptors also protect NK cells from exhausting their supply of cytolytic granules. The knowledge gained from the study of IRS members in NK cells has led to the realization that this mode of negative regulation is also used by other cell types.

Inhibitory Signals The block occurs early in the activation cascade, because inositol-1,4,5triphosphate production and Ca2+ flux were not detected in NK cells that were mixed with target cells expressing MHC class I ligands of the inhibitory receptor (13). Activation of NK cytotoxicity, whether through CD16 or by target cells, requires the activity of tyrosine kinases (13). Inhibitory receptors counteract the kinase activity by recruiting a tyrosine phosphatase, in general SHP-1. Inhibitory receptors contain one or more tyrosine residues in their cytoplasmic tail that, on phosphorylation, bind the SH2 domains of SHP-1 (9). A sequence

P1: APL/NBL/vks

January 22, 1999

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

880

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

context flanking the phosphotyrosine ([V,I] xYxx[L,V]) determines specific association with SHP-1 (21). The important role of SHP-1 in the immune system is evident from studies of moth-eaten (me) mice that are deficient in SHP-1. Thymocytes from me mice exhibit prolonged tyrosine phosphorylation of the T cell receptor (TCR)-ζ and CD3-ε, as well as a 38-kDa phosphoprotein (22) that is probably the linker for activation of T cells (LAT) (23). Coligation of an IRS member with an ITAMcontaining activation receptor aborts the tyrosine kinase-dependent signal (24). Independent ligation of the activation receptor and the inhibitory receptor does not result in inhibition (25). This may be because the inhibitory receptor tail is phosphorylated by a kinase associated with the activation receptor or because the action of SHP-1, which is triggered by binding to an ITIM, requires proximity of the tyrosine kinase substrates. In NK cells activated by CD16 (which is different from natural killing), binding of an ISIR to MHC class I on target cells resulted in a diminished tyrosine phosphorylation of the FcR-associated γ chain, ZAP-70, syk, and PLCγ (13). The tyrosine kinase syk plays an important role in natural killing, and its phosphorylation is reduced during inhibition of natural cytotoxicity by MHC class I-specific inhibitory receptors (14). Another outcome of MHC class I-mediated inhibition of natural killing is the loss of PLCγ association with a tyrosine-phosphorylated protein, pp36 (26), which is most likely the recently cloned LAT (23). The direct substrate of SHP-1 during inhibition mediated by any of the IRS members is still not known. Although SHP-1 recruitment and activation by a phosphorylated ITIM is a hallmark of inhibition mediated by an IRS member, it is important to remain aware that SHP-1 is also involved in other types of negative signals. For instance, the tyrosine kinase lyn phosphorylates CD22, which in turn recruits SHP-1 (27, 28). CD22 exerts a constitutive down modulation of signals from the BCR, which can be relieved by cross-linking of CD22 (29). SHP-1 also associates with many growth and cytokine receptors after the signal has been transmitted and serves to return these receptors to a basal level of activation (30, 31). A similar motif for the binding of SHP-2 (a protein tyrosine phosphatase related to SHP-1) has been defined (32). However, this tyrosine phosphatase has been associated mostly with positive signals (33). Therefore, the presence of an ITIM sequence in a receptor does not necessarily imply an inhibitory function. In addition to the substrate specificities of the SHP-1 and SHP-2 catalytic sites, the context in which the tyrosine phosphatase is recruited must also determine the signal outcome. Indeed, SHP-2 can substitute for SHP-1 in B cells that have lost SHP-1 by homologous recombination (34). Fcγ RIIb has long been known to inhibit B cell activation when coligated with the BCR by soluble Ig bound to a multivalent antigen. This inhibitory function

P1: APL/NBL/vks

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

January 22, 1999

T1: APR

AR078-26

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INHIBITORY RECEPTORS

881

led to the first definition of an ITIM (35, 36). Although phosphotyrosine peptides corresponding to the Fcγ RIIb1 ITIM bind SHP-1 and SHIP in vitro, the inhibition mediated by Fcγ RIIb1 during coligation with the BCR requires SHIP but not SHP-1 (37, 38). Coligation results in dephosphorylation of CD19, which may reduce PI-3K activation (39, 40). By converting PI-3,4,5-P3 into PI-3,4-P2 , SHIP inhibits the sustained intracellular calcium flux by preventing the PIP3mediated activation of the Tec kinase Btk, which, in turn, activates PLCγ (41). The somatostatin receptor sst2 mediates yet another type of inhibition through SHP-1 (42). Somatostatin is an inhibitory hormone with multiple biological effects, including inhibition of cell growth, that disrupt a constitutive association of SHP-1 with sst2, thereby causing an increase in SHP-1 activity. Association of SHP-1 with the insulin receptor in response to insulin, as well as the dephosphorylation of the insulin receptor and its substrates, are enhanced by the addition of somatostatin (43). Thus, rather than recruiting SHP-1 to exert an inhibition in trans, sst2 releases SHP-1 on ligand binding.

C-TYPE LECTIN INHIBITORY RECEPTORS Ly49 The Ly49A molecule is the founding member of the CLIR family. Ly49A inhibits mouse NK cells on binding to H-2 Dd expressed on target cells (11). It is a member of a family of nine Ly49 molecules designated Ly49A–I, which are encoded within the NK complex, a stretch of 2-Mb on mouse chromosome 6 (44). Several of the Ly49 family members have been shown to bind MHC class I, with some overlap in their specificities (45). Although Ly49 belongs to the C-type lectin superfamily, the recognition of MHC class I by Ly49A is independent of the carbohydrate on class I (46). The α1 and α2 domains of MHC class I both contribute to the recognition by Ly49A (46, 47), and peptide occupancy of the peptide-binding site is required (7). A putative human counterpart of Ly49 has been cloned (48). The single Ly49-like human gene appears to be a pseudogene and may represent the remnant of a common ancestral gene. Tyrosine-phosphorylated Ly49A binds SHP-1 (8). A role for SHP-1 in the inhibitory function of Ly49A was strongly suggested by studies of NK cells from me mice, in which the Ly49A-mediated inhibition was compromised (49). Residual inhibition may have been caused by SHP-2, which can substitute for SHP-1 in SHP-1–deficient cells (34).

CD94/NKG2 The invariant CD94 chain associates with different members of the small NKG2 family, including NKG2A, which has an inhibitory function (50). The first inhibitory receptor expressed by NK cells derived from CD7+CD34+ precursors

P1: APL/NBL/vks

January 22, 1999

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

882

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

is CD94/NKG2A (51). Most human NK cells express a CD94/NKG2 heterodimer. The genes encoding CD94 and the NKG2 molecules are located in the human NK complex on chromosome 12, a region syntenic to the mouse NK complex on chromosome 6 (44). Tyrosine-phosphorylated NKG2A binds SHP-1 and SHP-2 (52, 53). The CD94/NKG2A, B, and C heterodimers are receptors for the class Ib molecule HLA-E. The specificity was determined by direct binding of HLA-E tetramers to CD94/NKG2, by CD94/NKG2A-mediated inhibition of NK cells on recognition of peptide-loaded HLA-E, and by functional reconstitution of HLA-E recognition in rat basophil leukemia cells transfected by CD94 and NKG2A genes (53–56). The HLA-E specificity explains why CD94/NKG2 provided inhibition of lysis of cells transfected with various HLA-A, -B, and -C alleles, as well as the class Ib gene HLA-G. Expression of HLA-E at the cell surface requires a peptide that is derived from the signal sequence of these various class I molecules (57, 58). Therefore, the CD94/NKG2 receptor is designed to gauge the overall level of HLA class I synthesis in cells because down regulation of HLA-E–permissive HLA allotypes will coordinately reduce HLA-E expression. Expression of only one protective HLA class I allotype in a target cell provides only partial protection from lysis by NK clones that express CD94/NKG2A (59). It is unclear why certain HLA class I molecules do not support HLA-E expression. The class Ib molecule Qa1 in the mouse binds very similar peptides derived from certain mouse class I molecules, primarily H-2d allotypes. As mouse relatives of CD94 and NKG2 have been cloned (60, 61), it will be interesting to see whether they mediate recognition of Qa1 by NK cells. Inhibition of mouse NK cells by a CD94/NKG2-like receptor could explain several cases of inhibition that were not caused by an Ly49 molecule. Ly49-deficient fetal and neonatal NK cells are inhibited by class I on target cells (62, 63). The utilization of a CD94/NKG2-like receptor for recognition of MHC class I by Ly49-deficient fetal mouse NK cells would parallel the observation of CD94/NKG2 receptors on early human NK cells (51). In addition, evidence for a non-Ly49–mediated inhibition by “empty” class I has been reported (64).

C-Type Lectin Inhibitory Receptor Candidates Other C-type lectin-like receptors have the potential to serve inhibitory functions. The NKR-P1 family of molecules in the mouse, which includes the NK1.1 antigen expressed on a specific subset of T cells, has been implicated in activation rather than inhibition. Cross-linking of NK1.1 on T cells results in production of IFN-γ by these cells (65). However, some of the NKR-P1 molecules carry a cytoplasmic ITIM (6). Furthermore, cross-linking of the sole NKR-P1 relative in human NK cells resulted in either inhibition or activation

P1: APL/NBL/vks

January 22, 1999

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INHIBITORY RECEPTORS

883

signals (66). Because NKR-P1 is dispensable for NK cell development and cytolytic function (6), its function remains to be clarified, a task that would benefit from knowledge of NKR-P1 ligands. CD72 is a B cell–specific homodimer of a type II membrane protein with a C-type lectin domain and two cytoplasmic ITIMs. BCR triggering induces SHP-1 association with tyrosine-phosphorylated CD72 (67). The functional outcome of this association is unknown. An avian CD72 relative with a single ITIM is encoded by the chB1 gene and is expressed in immature B cells of the bursa of Fabricius (68). Rat mast cells express a C-type lectin molecule with inhibitory potential, called MAFA (69). Although inhibition of mast cell degranulation was detected by antibody-mediated cross-linking, the sequence context of the single cytoplasmic tyrosine (SIYSTL) does not conform to a standard ITIM.

IMMUNOGLOBULIN SUPERFAMILY INHIBITORY RECEPTORS Killer Cell Ig-Like Receptors The first IRS members identified in humans were the HLA-C–specific receptors, called p58 (12), members of the killer cell Ig-like receptor (KIR) family with two Ig domains (KIR2D). (A nomenclature for the KIR family is described at www.ncbi.nlm.nih.gov/prow.) Human NK cells can discriminate two groups of HLA-C allotypes which are distinguished by the presence of asn77 and lys80 (e.g. HLA-Cw4) or ser77 and asn80 (e.g. HLA-Cw3) in the α1 domain. KIR2DL1 (CD158a) is specific for the asn77lys80 group, whereas KIR2DL2 and KIR2DL3 (CD158b) each recognize the ser77asn80 group and cross-react with the asn77lys80 group of HLA-C allotypes (70). Amino acids 44, 45, and 70 in the first Ig domain of KIR2D are important for binding to HLA-C and for discrimination between the two groups of HLA-C allotypes (70, 71). These amino acids are located on adjacent loops that connect β strands in the first Ig domain (72). The crystal structure of KIR2DL1 revealed that the HLA-C–binding site is probably formed by the bottom face of the first and the top face of the second Ig domains, owing to a 60◦ angle between them. This acute angle between the two Ig domains, along with the topology of each Ig domain, places KIR2D within the hematopoietic receptor family (72). All members of this receptor family (e.g. growth hormone receptor and IL-2 receptor) dimerize on ligand binding. There is no evidence so far that a homodimerization or heterodimerization of KIR2D occurs on binding to HLA-C. In solution, recombinant forms of KIR2D and HLA-C form stoichiometric 1:1 complexes (73).

P1: APL/NBL/vks

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

January 22, 1999

884

T1: APR

AR078-26

LONG

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The other well defined HLA specificity is that of a KIR with three Ig domains (KIR3D), called p70 or KIR3DL1. KIR3DL1 recognizes HLA-B allotypes that belong to the Bw4 serologic specificity. Amino acids 82 and 83 contribute to this specificity (74, 75). The specificities of several KIR members, including some with a greater sequence divergence from the rest of the KIR family, are unknown, although weak binding to some class I molecules has been described for two of them (76, 77). The KIR family is encoded by about ten genes located next to the FcαR gene on chromosome 19 (78).

Gp49 The utilization of both ISIR (KIR) and CLIR (CD94/NKG2) for inhibition of human NK cells and of only CLIR (Ly49) in mouse NK cells was puzzling. A distant structural relative of KIR in mouse cells is gp49B. Gp49B shares 33% amino acid identity in its two Ig domains with KIR2D. Gp49B, previously known as a mast cell-specific protein, has two cytoplasmic ITIMs. Two gp49 genes, encoding gp49B and a noninhibitory gp49A isoform, are linked on chromosome 10 in a region that is not syntenic with the human KIR complex (79). Gp49B is expressed by mouse NK cells, and its cytoplasmic tail is able to inhibit NK cytotoxicity (80, 81), as well as mast cell degranulation (82). Tyrosine-phosphorylated peptides corresponding to the gp49B ITIMs associate with SHP-1, SHP-2, and SHIP in vitro (79). Ligands for the gp49 proteins have not been identified.

ILT One of the most exciting developments in the field has been the realization that IRS members regulate activation of various effector cells in the immune system (Figure 3). Monocytes, macrophages, dendritic cells, and some B and NK cells express members of the ILT receptor family. In a search for new Ig superfamily members expressed in NK cells, Samaridis & Colonna isolated ILT cDNA clones (83). Two other groups in search of a human relative of mouse gp49 isolated similar cDNA clones (78, 84). Finally, expression cloning of a ligand for the cytomegalovirus (CMV)-encoded class I-like molecule UL18 led to the identification of similar cDNA clones (85). In turn, expression cloning with a soluble form of an ILT led to the identification of HLA class I molecules as its ligands (85). Three different names have been used to designate these molecules: ILT (83), monocyte Ig-like receptor [MIR (78)], and leukocyte Ig-like receptor (LIR) (85). The original ILT designation is used here. ILT receptors consist of a family of at least eight members with different numbers of Ig domains, cytoplasmic tails with or without ITIMs, and distinct expression patterns (Table 1). ILTs are expressed primarily on monocytic cells and dendritic cells but are also found on some NK and B cells. They are

P1: APL/NBL/vks

January 22, 1999

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INHIBITORY RECEPTORS

885

Figure 3 Expression of immunoglobulin-superfamily inhibitory receptors in various human cell types. Fcγ RIIb, SIRP, and CD66, but not the other receptors, are conserved in mice. Mouse-specific immunoglobulin-superfamily inhibitory receptors include paired immunoglobulinlike receptors (mostly on monocytic cells) and gp49 (on mast and NK cells).

encoded in a gene complex centromeric of the KIR complex at 19q13.4 (78). ILT2, a member with four Ig domains, binds several different HLA class I molecules, including the class Ib molecule HLA-G (86). In addition, ILT2 (LIR-1) is the only ILT family member that binds to the CMV-encoded class I analog UL18 (87). ILT2 is expressed on most myelomonocytic cells, B cells, dendritic cells, and subsets of T cells and NK cells. ILT2 inhibits killing of HLA class I-positive target cells by the NK cell line NKL and inhibits CD16mediated activation of NKL as well (86). ILT2 inhibits T cell cytotoxicity induced by the TCR and inhibits intracellular Ca2+ mobilization in B cells and monocytes triggered by the BCR and HLA-DR, respectively (86). Thus, ILT2 inhibits several types of cellular responses on interaction with HLA class I on other cells. It is not clear whether this inhibition serves to regulate the threshold of activation for B cells and monocytic cells or to favor responses to microbial organisms that are devoid of HLA class I. The ILT3 member with two Ig domains is expressed in monocytes and dendritic cells (88). Antibody-mediated

P1: APL/NBL/vks

January 22, 1999

886

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

Table 1 The immunoglobulin-like-transcript receptor family

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

cDNA clonesa

Immunoglobulin domains

Features

Expression (Reference)

ILT1

LIR-7

4

Short tail

Myeloid, some B, NK lines (83); not in monocytes, DC, NK, T, B cells (87)

ILT2

LIR-1, MIR-7

4

Binds UL18 and HLA class I

Monocytes, B, low in NK (87); not in myeloid lines (83)

ILT3

LIR-5, HM18

2

Inhibits macrophage activation

Monocytes, DC, low in NK and B (87); monocytes, DC (83)

ILT4

LIR-2, MIR-10

4

Bind HLA class I; Monocytes and B (87); inhibits FcR signal monocytes, DC, not in B lines (89)

ILT5

LIR-3, HL9

4

ILT6

LIR-4, HM43

4

Secreted

NK and B, not in monocytes (87); monocytes, not in NK (84)

LIR-6

2 or 4

Short tail

Monocytes and B, low in NK and T (87)

LIR-8

4

a

Monocytes and B, low in DC (87); mast and NK (84)

Only in NK (87)

References for cDNA clones are ILT (83, 86, 88, 89), LIR (85, 87), MIR (78), and HM/HL (84).

co–cross-linking of ILT3 with activating receptors such as CD16 or HLA-DR on monocytes blunted the intracellular Ca2+ mobilization (88). ILT4, less widely expressed than ILT2, is on monocytes, macrophages, and dendritic cells. ILT4 is another HLA class I-binding receptor (mostly HLA-A and HLA-B allotypes) that can inhibit activation signals delivered to those cells by Fc receptors or HLA-DR (89). Tyrosine-phosphorylated ILT2 and ILT4 bind SHP-1 (86, 89). The potential for negative regulation by ILT receptors is vast, because the ILT family consists of at least eight members, including a potentially secreted form. The ligands for most of the ILT receptors have not been identified.

Paired Ig-Like Receptors Like most other IRS members, the ILT receptor family is not conserved in evolution and does not have a mouse ortholog. However, in terms of tissue distribution, a mouse counterpart of human ILT is the family of molecules

P1: APL/NBL/vks

January 22, 1999

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INHIBITORY RECEPTORS

887

termed paired Ig-like receptors (PIR) or p91 (90, 91). PIR is a family of receptors with six Ig domains expressed in myeloid and B cells. Ligands of PIR have not been identified. Most PIR family members have a short cytoplasmic tail lacking an ITIM and are designated PIR-A. A single family member, PIR-B, carries a long cytoplasmic tail with one ITIM and three additional YxxL motifs (90). PIR-B was also identified by substrate trapping in macrophage lysates with a catalytically inactive SHP-1 mutant (92). Tyrosine phosphorylated PIR-B is both a ligand and a substrate of SHP-1. PIR-B associates with SHP-1 when cross-linked with the IgE receptor on RBL cells (93). Inhibition by PIR-B was impaired in chicken DT40 cells lacking both SHP-1 and SHP-2, but not in DT40 cells lacking SHP-1 or SHP-2 alone, suggesting that PIR-B can use either phosphatase for inhibition (34). The PIR genes map to a region on mouse chromosome 7 that is syntenic with the human KIR-ILT gene complex (78, 90).

Leukocyte-Associated Ig-Like Receptor Leukocyte-associated Ig-like receptor (LAIR)-1 is an IRS member with a single Ig domain that is expressed on all leukocytes (94). Natural killing of an FcR+ target cell was inhibited by cross-linking with an anti-LAIR monoclonal antibody bound to the target. LAIR-1 binds SHP-1 and SHP-2 when tyrosine phosphorylated on its ITIM in NK cells. LAIR-1 shares its size (40 kDa) and other properties with an inhibitory receptor called p40 that is expressed on hematopoietic cells and inhibits NK and T cell responses (95, 96). Despite the discordant expression of LAIR-1 and p40 on two cell lines, their similarity suggests that these inhibitory receptors may be related molecules.

Extended Ig-Superfamily Inhibitory Receptor Family All the members of the ISIR family described up to this point (KIR, gp49, ILT, PIR, LAIR) form a subfamily of the Ig superfamily, which are related by at least 40% identity in one or more of their Ig domains. This structural subfamily includes the two noninhibitory human FcαR and bovine Fcγ 2R that lack a cytoplasmic ITIM and a proline-serine–rich stem between the Ig domain and the transmembrane region that characterize ISIRs. Several receptors that do not belong to this Ig subfamily, owing to the lack of sequence similarity, nevertheless have properties that qualify them for membership in the ISIR family. The Fcγ RIIb1 inhibitory receptor has two Ig domains and a single cytoplasmic ITIM. As described above, the negative signal transmitted by Fcγ RIIb1 is distinct from that of other IRS members. Other receptors with features reminiscent of ISIR have been described in cells that are not within the immune system. The BIT/SHP substrate 1 (SHPS-1)

P1: APL/NBL/vks

January 22, 1999

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

888

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

and signal-regulatory protein (SIRP) (97–99) molecules with three Ig domains belong to a gene family of about 15 members that is conserved between human and mice. These receptors have two ITIMs and two additional tyrosines in the cytoplasmic tail. Their expression is ubiquitous, with highest levels in the brain. This gene family maps at 20 p13 (100), outside of the KIR-ILT gene complex. Molecular clones were obtained based on the association of these receptors with SHP-1 or SHP-2. SHP-2 associates with SIRP in response to growth hormone, but not in response to leukemia inhibitory factor or IFN-γ , even though the receptors for all three associate with the same kinase, JAK2. Transfection experiments suggest that JAK2 binds and phosphorylates SIRP (101). Overexpression of SHPS-1 enhanced insulin-induced activation in CHO cells expressing human insulin receptor (102). Phosphorylated in response to integrin-mediated adhesion, SHPS-1 may connect integrin signals to the rasMAPK pathway (103). On the other hand, overexpression of SIRPα1 in 3T3 fibroblasts reduced the MAPK activation in response to epidermal growth factor or insulin and inhibited cell growth and formation of cell foci on infection with a retrovirus (99). SIRP was also identified as a constitutive ligand and substrate of SHP-1 in macrophages, forming a complex with the CSF-R (92). SHP-1 forms separate complexes with PIR in the same cells (92). Although it is clear that the SIRPs regulate responses to growth factor receptors, their ability to bind either SHP-1 or SHP-2 and the lack of known ligands leave some uncertainty as to their physiologic function. CD66a, also called biliary glycoprotein, has three Ig domains and belongs to the family of carcinoembryonic antigens. CD66a is expressed on many cell types, including hemopoietic cells. Constitutive expression of CD66a inhibits tumor growth (104). CD66a associates with SHP-1 when phosphorylated on two cytoplasmic tyrosine residues that conform to the ITIM (105). However, addition of monoclonal antibodies to CD66 activates integrin-mediated signals and the respiratory burst of neutrophils (106) and the degranulation of human granulocytes (107), suggesting that CD66a may serve different functions. Although the presence of a cytoplasmic ITIM in several molecules of unknown function has been successfully used to predict an inhibitory function, it is unlikely to apply to all other molecules with ITIM-related cytoplasmic sequences. The Programmed Death-1 (PD-1) gene encodes a molecule with a single V-like Ig domain and an ITIM-like sequence in the cytoplasmic tail (108). PD-1 expression correlates with activation of thymocytes and pro-B cells. The vitamin K-dependent carboxylase is a cytosolic enzyme that carries an ITIM sequence (QEVTYANL) at its carboxy terminus (109) almost identical to that of several ISIRs. It is not known whether these molecules bind SHP-1 and whether they have an inhibitory function.

P1: APL/NBL/vks

January 22, 1999

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

INHIBITORY RECEPTORS

889

REPERTOIRES OF PAIRED ACTIVATION/INHIBITION RECEPTOR ISOFORMS

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Activating Isoforms of IRS Members Most subfamilies of receptors within the IRS include members with activating rather than inhibitory function (Figure 2). These activating isoforms have a different cytoplasmic tail lacking an ITIM and carry a positively charged residue in their transmembrane region. The exceptions are Fcγ RIIb and LAIR, each encoded by a single gene, and the noninhibitory gp49A and CD66, which do not carry the charged amino acid in the transmembrane region. The activating forms of KIR2D with a short cytoplasmic tail (KIR2DS or p50) have an in-frame stop codon that interrupts the first ITIM. The cytoplasmic tails of gp49A and gp49B differ by a frameshift such that the shorter gp49A tail ends with several different C-terminal amino acids. Similarly, frameshifts or sequence divergence near the end of the coding regions of KIR3D, ILT, PIR, and SIRP genes account for the distinct isoforms with long or short cytoplasmic tails. Different cytoplasmic tails in CD66 are generated by alternative splicing. CLIR are type II membrane proteins with their cytoplasmic tails at the amino terminus. Therefore, stop codon insertions or frameshift mutations preceding the ITIM would eliminate the protein altogether. Instead, point mutations have replaced the tyrosine in the ITIM by a phenylalanine, thereby eliminating the inhibitory function (110). The ITIM-deficient CLIR isoforms also carry a positively charged residue in the transmembrane region. Evidence was obtained for an activating function of KIR2DS (111–114), NKG2C (50), and Ly49D and Ly49H (115–117). Each one associates with a short disulfide-linked homodimer of the protein called DAP12 that carries a cytoplasmic ITAM (117–119). The basic and acidic residues in the transmembrane regions of NKG2C and DAP12, respectively, are necessary for this association (119). These receptors activate cytotoxicity and IFN-γ production by NK and T cells (113, 114, 120, 121) and proliferation of class II-restricted T cells (112). MHC class I-specific NK receptors in the rat display activating and inhibitory functions as well, probably mediated by rat Ly49 orthologs (122, 123). The KIR2DS1 and KIR2DS2 molecules bind to HLA-C much less well than their KIR2DL inhibitory counterparts (70, 71). This difference may be necessary to ensure that the inhibitory signal overrides the activation signal. The activation signal delivered by NKR-P1 is distinct in that it requires the γ chain (124). This signal results in IFN-γ production by NK1.1 T cells (65). The existence of activating isoforms in several receptor families and the different ways in which they have been selected during evolution suggest that they serve an important role. However, a unifying principle that would explain this

P1: APL/NBL/vks

January 22, 1999

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

890

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

dual diversity in subfamilies of the IRS, which is neither a cell type-, ISIR-, or CLIR-specific feature, is still missing. Six hypotheses for the role of activatingreceptor isoforms can be proposed, none of which is particularly satisfying. First, they are important in triggering NK cytotoxicity or activation of other cell types. However, such activating receptors are not necessary to activate NK cells that can respond to many different surface receptors. Second, they provide specific activation on binding to foreign structures. Without adaptive potential, such a targeted recognition would be limited. In addition, the high degree of sequence identity between KIR2DL and KIR2DS molecules, as well as functional data with cells expressing these receptor isoforms, suggests that they both recognize similar forms of HLA-C. Third, they enhance activation only in some cases. A tumor cell that has lost expression of a protective MHC class I allotype may be killed more efficiently by NK cells if the remaining MHC class I serves as ligand for an activating-receptor isoform. However, several tumor-specific activation receptors have been described on NK cells, in addition to the receptors that mediate natural killing of allogeneic or class I-deficient hematopoietic cells. Furthermore, the complex KIR repertoire (see below) is not determined by the HLA of the host, such that an activating KIR isoform does not necessarily recognize self HLA. Fourth, activating isoforms are specific for self epitopes that are induced by infection or stress. An invariant stress signal would be best recognized by an invariant receptor. An HLA-bound stress-specific epitope would require a repertoire of receptors that is MHC linked. Activating KIR isoforms fulfill neither of these conditions. Fifth, they serve as a coreceptor of the inhibitory receptor. For instance, KIR2DS or Ly49D may provide the tyrosine kinase necessary to phosphorylate KIR2DL or Ly49A, respectively. The discordant expression of these isoforms (59, 125) argues against such a function. Finally, they provide signals that are distinct from those of other activation receptors. Even though signaling through the ITAM of DAP12 may be very similar to activation by CD16 in NK cells or by the T cell receptor (via the γ or the ζ chain ITAM, respectively), the activating isoforms of IRS members may control responses other than cytotoxicity, such as proliferation or homing.

Repertoire Selection and Self Tolerance of Natural Killer Receptors The repertoire of NK receptors has been extensively covered by Lanier’s review (6). Briefly, some of the outstanding features are as follows. Individual NK cells express more than one MHC class I-specific receptor. Various combinations of receptor expression, including inhibitory and activating forms of ISIR and CLIR, are observed on NK cells. The repertoire of these NK receptors is not selected strictly by the host MHC, although the frequency and level of expression of Ly49 molecules are influenced by MHC haplotypes. In addition to

P1: APL/NBL/vks

January 22, 1999

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INHIBITORY RECEPTORS

891

the requisite inhibitory receptor for a self MHC class I, NK cells often express receptors with no class I ligand in the host. Little is known about the selective process that ensures the presence of at least one useful inhibitory receptor on every NK cell. Expression of different Ly49 family members is monoallelic (125a) and appears sequentially on mouse NK cells during the first few weeks after birth (126). The human repertoire of ISIR and CLIR expressed on NK cells is very complex and appears to be determined in part by KIR genotypes (59, 127). NK clones, in a donor that has only one known HLA class I ligand for their KIR, rely preferentially on the CD94/NKG2 receptor for inhibition by autologous cells. Conversely, another donor with several HLA class I ligands for KIR has NK clones that are inhibited most often by KIR2DL and KIR3DL (59). A surprising result was that the set of KIR genes was different in those two individuals, as determined by polymerase chain reaction (127). These two KIR haplotypes are about equally frequent in the human population but are found only as homozygous. One way to explain this unique finding is that some form of massive gene conversion duplicates the one KIR haplotype that may be more suited for recognition of self HLA class I. Elucidation of the molecular basis for these findings awaits further characterization of the genomic organization and nucleotide sequence of the KIR complex. Tolerance to self by NK cells is not due to clonal deletion but to an active process that requires the continuous presence of sensitive targets. This tolerance can be broken in vitro by culturing NK cells in the absence of sensitive targets, as shown with NK cells from a mouse that has a mosaic expression of an MHC class I transgene (128) and with self-tolerant NK cells from MHC class I-deficient mice (129). Tolerance can be induced in vivo by hemopoietic and nonhemopoietic class I-negative cells in mice reconstituted with class I-positive fetal liver cells (130). Studies of NK self tolerance in mice with a severe class I expression defect [β2 m and transporter for antigen presentation (TAP) double knockouts] suggest that additional tolerance mechanisms exist that down regulate activation of NK cells (131, 132).

FUNCTION OF INHIBITORY-RECEPTOR SUPERFAMILY MEMBERS An inhibitory function has been demonstrated for only a few members of the IRS. A negative signal induced by antibody-mediated receptor cross-linking is insufficient evidence for an inhibitory function. This approach may even be misleading in some cases. For instance, cross-linking CD22 releases it from a constitutive association with the BCR and removes its negative influence on BCR signaling (3). Identification of ligands and of the physiological context

P1: APL/NBL/vks

January 22, 1999

892

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

for inhibition is still needed for most IRS members. For some of the better characterized receptors, such as KIR and Ly49, one important role is the protection of self. In addition to providing the basis for self/nonself discrimination by T cells, MHC class I molecules also have the important duty of marking host cells to be spared by NK cells. The presence of MHC class I-specific inhibitory receptors on other cell types, mainly APC, suggests that MHC class I molecules may serve other functions as well. The following sections speculate about the potential roles of inhibitory receptors in different cell types and how such negative signaling may be exploited by infectious organisms.

Modulation of T- and Natural Killer Cell Responses Because NK cells use ligands that are expressed on healthy cells for the recognition and killing of target cells, a mechanism for the protection of healthy self is needed. The existence of two types of inhibitory receptors, CLIR and ISIR, that are coexpressed on NK cells and recognize MHC class I molecules in different ways, underlines the importance of such a protection mechanism. What then may be the role of IRS members on T cells? Selection and activation of T cells is tightly regulated by their antigen-specific receptor and coreceptors such that responses to self are largely avoided or suppressed. Nevertheless, some peripheral T cells express CLIRs, ISIRs, or both. The proportion of T cells expressing an inhibitory receptor is much greater in the γ δ subset. Whereas only 5%–10% of the αβ T cells express inhibitory receptors, a majority of CD8+ γ δ T cells, in particular the Vγ 9Vδ2 subset, in peripheral blood and in the gut, express CD94, KIR, or both (133–135). KIR expressed on human T cells can impair antigen-specific CTL functions, anti-tumor responses, and cytokine production (134–137). Ly49 modulates cytokine production in mouse T cells (138). Because KIRs are expressed preferentially on memory T cells, it has been proposed that they serve to raise the threshold of activation for secondary responses (6). Activation of CD94+ γ δ T cells is diminished on MHC class I-positive cells and can be overcome by higher doses of antigen (133). CD94/NKG2 engagement with MHC class I on APC results in enhanced SHP-1 recruitment to the TCR and diminishes the tyrosine phosphorylation of ZAP70 and the association of lck with the TCR (133). CD94 expression in T cells is induced by stimulation with antigen and IL-15 (139, 140). It would be interesting to follow the expression of IRS members on T cells during T cell development and during immune responses. The two-signal requirement for activation of naive T cells mediated by the antigen-specific TCR and CD28 may be replaced by a different type of dual signal for activation of mature effector cells: a TCR signal and a negative signal from a class I-specific inhibitory receptor. Only TCR signals that are strong enough to overcome inhibition will result in activation. Whether inhibitory receptors on T cells serve to raise the threshold of activation in secondary

P1: APL/NBL/vks

January 22, 1999

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

INHIBITORY RECEPTORS

893

responses, act as a safeguard against autoreactivity, or terminate a primary response remains to be determined. Whatever the reason for CLIR and ISIR on T cells, it comes at the expense of desirable antigen-specific responses (140).

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Regulation of Other Effector Cells A large number of receptors have been identified that inhibit cells on interaction with MHC class I, including class Ib molecules. Some class I isotypes, such as HLA-C, HLA-E, and HLA-G, may have evolved to serve primarily protective functions. Unlike HLA-A and HLA-B allotypes, all HLA-C allotypes are recognized by a KIR. Expression of HLA-A and HLA-B only would be clearly insufficient to protect target cells in each individual from NK cells that are inhibited by KIR. The only function assigned so far to HLA-E is to inhibit NK cells and some other effector cells. The characterization of new IRS members on APC, such as macrophages and dendritic cells, has recently widened the field of inhibitory receptors. Several cell types that express IRS members represent effector cells in the innate or the adaptive immune systems (NK, CTL, macrophages, and mast cells) that release effector molecules. This feature suggests that inhibitory receptors may also control other effector cells that degranulate, such as neutrophils and basophils. The role of ISIR on macrophages (human ILT and mouse PIR) may be to protect cells that produce activating cytokines from phagocytosis and to favor macrophage responses to foreign organisms. However, the specificity for class I (isotypes and allotypes) of ILT members is limited to a subset of the HLA class I repertoire, such that protection would not be assured in all individuals. Most ILT members have ligands other than MHC class I that have yet to be identified and that may well hold the clue to the function of inhibitory ILTs. One possibility is that the ligands for ILTs and PIRs are distributed in specific sites and tissues that require special protection from cells in the monocytic lineage.

Immune Privilege Certain anatomical sites receive special treatment by the immune system, referred to as immune privilege. The anterior chamber of the eye, the testes, and the fetus are among such privileged sites. One protection mechanism against attack by cytotoxic cells is provided by the surface expression of the fas ligand on the anterior segment of the eye (141). Fas-expressing lymphocytes that reach this site may be eliminated by fas-mediated apoptosis. Another useful mechanism for protection would be the expression of ligands for inhibitory receptors that are expressed on various types of potentially dangerous cells. As suggested above, ILTs and PIRs on macrophages could serve such a function. The identification of ligands for IRS members may reveal sites that are specifically protected from effector cells of the immune system.

P1: APL/NBL/vks

January 22, 1999

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

894

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

The anterior chamber of the eye produces a soluble factor, recently identified as the macrophage migration inhibitory factor (MIF), that also inhibits NK cells (142). MIF is a trimer (143) and has therefore the potential to cross-link its receptor. It would be interesting to test the binding of MIF to IRS members. The basis for immune privilege at the maternal-fetal interface is poorly understood. The presence of HLA-G on fetal cytotrophoblasts that invade the maternal decidua has suggested a role for this class Ib molecule in the inhibition of maternal NK cells (144). Several studies reported HLA-G–mediated inhibition of NK cells by CD94/NKG2 (145–148). However, it has since become clear that HLA-G provides a signal sequence-derived peptide for binding to HLA-E, which is the specific ligand for CD94/NKG2. There has been no direct identification of an HLA-G–specific inhibitory receptor. Reports of HLA-G recognition by KIR (149, 150) have not been supported by further studies (146, 147). Several of those studies on inhibition of NK cells by HLA-G suggest that a receptor for HLA-G has yet to be found. Soluble forms of ILT2, ILT4, and a p49 member of the KIR family bind weakly to HLA-G and to other class I molecules (77, 86, 89). A truncated HLA-G molecule (HLA-G2), missing the α2 domain as a result of alternative splicing, still provides protection from lysis by NK cells (151), possibly by providing a signal sequence-derived peptide that allows HLA-E expression. The high expression of HLA-G2 on some melanoma cell lines suggests that it may be used by tumor cells to escape NK responses (152). Induction of HLA-G could compensate for the frequent loss of classical class I molecules on tumor cells.

The Pathogen’s Perspective Every type of immune defense mechanism drives the evolution of evasion tactics used by pathogens. NK cells are often mentioned as a backup system to fight infections by viruses that evade class I-restricted T cell responses, such as herpesviruses (153, 154). The complex repertoire of inhibitory receptors used by NK cells makes it difficult to achieve inhibition of all NK cells by a single class I mimic. This may in fact have been the evolutionary driving force behind the diversity of inhibitory receptors. Distant relatives of MHC class I are encoded in the mouse and the human CMV, each one a member of the herpes family that interferes with MHC class I-restricted antigen presentation (154). UL18, the HLA class I relative encoded by the human CMV, has been proposed to serve as a ligand for inhibitory receptors on NK cells (155). However, careful transfection and infection experiments showed that UL18-expressing cells were, if anything, more sensitive to lysis by NK cells (156). Expression cloning of a ligand for UL18 identified the ILT2 receptor (LIR-1), expressed mainly on monocytic cells, rather than an NK receptor (85). Therefore, UL18 is more likely to inhibit monocytes,

P1: APL/NBL/vks

January 22, 1999

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

INHIBITORY RECEPTORS

895

macrophages, and dendritic cells. Such a function could affect NK cell responses indirectly by reducing the production of IL-12, a lymphokine that is a major stimulator of NK cells. This could explain the NK-mediated decrease in viral replication of a mouse CMV mutant that lacks the UL18 counterpart m144 (157). The mouse locus Cmv1 that controls resistance to CMV maps within the NK complex (158), suggesting the interesting possibility that the receptor for m144 may be a CLIR, rather than an ISIR such as ILT2. A virus trying to evade NK as well as T cell responses could achieve some success by interfering selectively with the expression of classical class I, but not with that of the class Ib HLA-E (or Qa1 in the mouse). Expression of HLA-E and Qa1 at the cell surface requires binding of a peptide derived from the signal sequences of other MHC class I molecules. Even though signal sequences are cleaved in the endoplasmic reticulum, peptide loading onto HLA-E and Qa1 requires the TAP. This requirement may in fact be a protection mechanism against viruses that interfere with TAP function, as it would prevent HLA-E and Qa1 surface expression and remove a major protection signal from NK cells. The task for the virus is to either retain selectively classical class I molecules inside the cell or provide a TAP-independent peptide for HLA-E while blocking TAP function. The AIDS virus HIV may have developed a different strategy by the nef-mediated internalization of HLA-A and HLA-B, but not HLA-C, from the cell surface (159), possibly as a way to reduce detection by both T and NK cells.

CONCLUSIONS The IRS represents an interesting aspect of the immune system that escapes the precise definition of either innate or adaptive immunity, as formulated by Janeway (160). These receptors, independent of V(D)J recombination, are clearly not typical of adaptive immunity. However, NK receptors have shown some degree of specificity for MHC-peptide complexes, which can also be the ligands for specific T cell receptors. Distribution of the receptors is clonal (an adaptive feature), but many NK cells in an individual share the same repertoire of receptors. NK cells use a unique strategy to achieve self-nonself discrimination that is based on the loss of inhibition by self. Some type of selection process must contribute to the repertoire of inhibitory receptors that are responsible for this vital discrimination. Much remains to be learned in this exploding field of inhibitory receptors. Some of the most fundamental questions have barely begun to be addressed, such as the evolution of different receptor families within the IRS, the development of complex repertoires of paired activation/inhibition receptors on different cell types, and the in vivo function of these receptors during infections.

P1: APL/NBL/vks

January 22, 1999

896

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

Mouse experimental models, including gene knockouts, are needed to address some of these questions, but the existence of multiple genes within CLIR and ISIR families complicates the task, and their usefulness will be limited by the lack of conservation with human receptors. Nevertheless, advances in this field are likely to yield new answers by the time this review is published and may eventually provide useful applications in the prevention of autoimmunity and in the reduction of allograft rejection.

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ACKNOWLEDGMENTS I thank the members of my group for their support, D. Burshtyn and S. Rajagopalan for useful comments, and J. Cambier, K. K¨arre, B. Neel, T. Takai, and T. Tsubata for providing preprints. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Thompson CB, Allison JP. 1997. The emerging role of CTLA-4 as an immune attenuator. Immunity 7:445–50 2. Van Parijs L, Abbas AK. 1998. Homeostasis and self-tolerance in the immune system: turning lymphocytes off. Science 280:243–48 3. O’Rourke LM, Tooze R, Fearon DT. 1997. Co-receptors of B lymphocytes. Curr. Opin. Immunol. 9:324–29 4. Healy JI, Goodnow CC. 1998. Positive versus negative signaling by lymphocyte antigen receptors. Annu. Rev. Immunol. 16:645–70 5. Moretta A, Bottino C, Vitale M, Pende D, Biassoni R, Mingari MC, Moretta L. 1996. Receptors for HLA class-I molecules in human natural killer cells. Annu. Rev. Immunol. 14:619–48 6. Lanier LL. 1998. NK cell receptors. Annu. Rev. Immunol. 16:359–93 7. Yokoyama WM. 1998. Natural killer cell receptors. Curr. Opin. Immunol. 10:298– 305 8. Vivier E, Daeron M. 1997. Immunoreceptor tyrosine-based inhibition motifs. Immunol. Today 18:286–91 9. Burshtyn DN, Long EO. 1997. Regulation through inhibitory receptors: lessons from natural killer cells. Trends Cell Biol. 7:473–79 10. Ljunggren HG, Karre K. 1985. Host resistance directed selectively against H-2-deficient lymphoma variants. Anal-

11.

12.

13.

14.

15. 16.

ysis of the mechanism. J. Exp. Med. 162:1745–59 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 Moretta A, Vitale M, Bottino C, Orengo AM, Morelli L, Augugliaro R, Barbaresi M, Ciccone E, Moretta L. 1993 P58 molecules as putative receptors for major histocompatibility complex (MHC) class I molecules in human natural killer (NK) cells. Anti-p58 antibodies reconstitute lysis of MHC class I-protected cells in NK clones displaying different specificities. J. Exp. Med. 178:597– 604 Leibson PJ. 1997. Signal transduction during natural killer cell activation: inside the mind of a killer. Immunity 6:655–61 Brumbaugh KM, Binstadt BA, Billadeau DD, Schoon RA, Dick CJ, Ten RM, Leibson PJ. 1997. Functional role for Syk tyrosine kinase in natural killer cell-mediated natural cytotoxicity. J. Exp. Med. 186:1965–74 Lanier LL, Corliss B, Phillips JH. 1997. Arousal and inhibition of human NK cells. Immunol. Rev. 155:145–54 Azuma M, Cayabyab M, Buck D, Phillips JH, Lanier LL. 1992. Involvement of CD28 in MHC-unrestricted cytotoxicity mediated by a human natural

P1: APL/NBL/vks

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

January 22, 1999

T1: APR

AR078-26

INHIBITORY RECEPTORS

17.

18.

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

19.

20.

21.

22.

23.

24.

25.

26.

killer leukemia cell line. J. Immunol. 149:1115–23 Chambers BJ, Salcedo M, Ljunggren HG. 1996. Triggering of natural killer cells by the costimulatory molecule CD80 (B7-1). Immunity 5:311–17 Helander TS, Carpen O, Turunen O, Kovanen PE, Vaheri A, Timonen T. 1996. ICAM-2 redistributed by ezrin as a target for killer cells. Nature 382:265–68 Sivori S, Vitale M, Morelli L, Sanseverino L, Augugliaro R, Bottino C, Moretta L, Moretta A. 1997. p46, a novel natural killer cell-specific surface molecule that mediates cell activation. J. Exp. Med. 186:1129–36 Vitale M, Bottino C, Sivori S, Sanseverino L, Castriconi R, Marcenaro E, Augugliaro R, Moretta L, Moretta A. 1998. NKp44, a novel triggering surface molecule specifically expressed by activated natural killer cells, is involved in non-major histocompatibility complexrestricted tumor cell lysis. J. Exp. Med. 187:2065–72 Burshtyn DN, Yang WT, Yi TL, Long EO. 1997. A novel phosphotyrosine motif with a critical amino acid at position2 for the SH2 domain-mediated activation of the tyrosine phosphatase SHP-1. J. Biol. Chem. 272:13066–72 Pani G, Fischer KD, Mlinaric-Rascan I, Siminovitch KA. 1996. Signaling capacity of the T cell antigen receptor is negatively regulated by the PTP1C tyrosine phosphatase. J. Exp. Med. 184:839–52 Zhang WG, Sloan-Lancaster J, Kitchen J, Trible RP, Samelson LE. 1998. LAT: the ZAP-70 tyrosine kinase substrate that links T cell receptor to cellular activation. Cell 92:83–92 Daeron M, Latour S, Malbec O, Espinosa E, Pina P, Pasmans S, Fridman WH. 1995. The same tyrosinebased inhibition motif, in the intracytoplasmic domain of FcgammaRIIB, regulates negatively BCR-, TCR-, and FcR-dependent cell activation. Immunity 3:635–46 Bl´ery M, Delon J, Trautmann A, Cambiaggi A, Olcese L, Biassoni R, Moretta L, Chavrier P, Moretta A, Da¨eron M, Vivier E. 1997. Reconstituted killer cell inhibitory receptors for major histocompatibility complex class I molecules control mast cell activation induced via immunoreceptor tyrosine-based activation motifs. J. Biol. Chem. 272:8989–96 Valiante NM, Phillips JH, Lanier LL, Parham P. 1996. Killer cell inhibitory receptor recognition of human leukocyte

27.

28.

29.

30. 31. 32.

33. 34.

35.

36.

37.

897

antigen (HLA) class I blocks formation of a pp36/PLC-gamma signaling complex in human natural killer (NK) cells. J. Exp. Med. 184:2243–50 Cornall RJ, Cyster JG, Hibbs ML, Dunn AR, Otipoby KL, Clark EA, Goodnow CC. 1998. Polygenic autoimmune traits: Lyn, CD22, and SHP-1 are limiting elements of a biochemical pathway regulating BCR signaling and selection. Immunity 8:497–508 Chan VWF, Lowell CA, DeFranco AL. 1998. Defective negative regulation of antigen receptor signaling in Lyndeficient B lymphocytes. Curr. Biol. 8:545–53 Doody GM, Justement LB, Delibrias CC, Matthews RJ, Lin J, Thomas ML, Fearon DT. 1995. A role in B cell activation for CD22 and the protein tyrosine phosphatase SHP. Science 269:242–44 Ihle JN. 1995. Cytokine receptor signalling. Nature 377:591–94 Neel BG. 1997. Role of phosphatases in lymphocyte activation. Curr. Opin. Immunol. 9:405–20 Huyer G, Li ZM, Adam M, Huckle WR, Ramachandran C. 1995. Direct determination of the sequence recognition requirements of the SH2 domains of SHPTP2. Biochemistry 34:1040–49 Tonks NK, Neel BG. 1996. From form to function: signaling by protein tyrosine phosphatases. Cell 87:365–68 Maeda A, Kurosaki M, Ono M, Takai T, Kurosaki T. 1998. Requirement of SH2-containing protein tyrosine phosphatases SHP-1 and SHP-2 for paired immunoglobulin-like receptor B (PIRB)-mediated inhibitory signal. J. Exp. Med. 187:1355–60 Amigorena S, Bonnerot C, Drake JR, Choquet D, Hunziker W, Guillet JG, Webster P, Sautes C, Mellman I, Fridman WH. 1992. Cytoplasmic domain heterogeneity and functions of IgG Fc receptors in B lymphocytes. Science 256:1808–12 Muta T, Kurosaki T, Misulovin Z, Sanchez M, Nussenzweig MC, Ravetch JV. 1994. A 13-amino-acid motif in the cytoplasmic domain of Fc gamma RIIB modulates B-cell receptor signalling. Nature 368:70–73 Gupta N, Scharenberg AM, Burshtyn DN, Wagtmann N, Lioubin MN, Rohrschneider LR, Kinet JP, Long EO. 1997. Negative signaling pathways of the killer cell inhibitory receptor and FcgammaRIIb1 require distinct phosphatases. J. Exp. Med. 186:473–78

P1: APL/NBL/vks

January 22, 1999

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

898

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

38. Ono M, Okada H, Bolland S, Yanagi S, Kurosaki T, Ravetch JV. 1997. Deletion of SHIP or SHP-1 reveals two distinct pathways for inhibitory signaling. Cell 90:293–301 39. Kiener PA, Lioubin MN, Rohrschneider LR, Ledbetter JA, Nadler SG, Diegel ML. 1997. Co-ligation of the antigen and Fc receptors gives rise to the selective modulation of intracellular signaling in B cells-Regulation of the association of phosphatidylinositol 3-kinase and inositol 50 -phosphatase with the antigen receptor complex. J. Biol. Chem. 272:3838–44 40. Hippen KL, Buhl AM, D’Ambrosio D, Nakamura K, Persin C, Cambier JC. 1997. Fc gammaRIIB1 inhibition of BCR-mediated phosphoinositide hydrolysis and Ca2+ mobilization is integrated by CD19 dephosphorylation. Immunity 7:49–58 41. Scharenberg AM, El-Hillal O, Fruman DA, Beitz LO, Li ZM, Lin SQ, Gout I, Cantley LC, Rawlings DJ, Kinet JP. 1998. Phosphatidylinositol-3,4,5trisphosphate (PtdIns-3,4,5-P3) Tec kinase-dependent calcium signaling pathway: a target for SHIP-mediated inhibitory signals. EMBO J. 17:1961– 72 42. Lopez F, Est`eve JP, Buscail L, Delesque N, Saint-Laurent N, Th´eveniau M, Nahmias C, Vaysse N, Susini C. 1997. The tyrosine phosphatase SHP1 associates with the sst2 somatostatin receptor and is an essential component of sst2-mediated inhibitory growth signaling. J. Biol. Chem. 272:24448– 54 43. Bousquet C, Delesque N, Lopez F, SaintLaurent N, Est`eve JP, Bedecs K, Buscail L, Vaysse N, Susini C. 1998. sst2 somatostatin receptor mediates negative regulation of insulin receptor signaling through the tyrosine phosphatase SHP1. J. Biol. Chem. 273:7099–106 44. Brown MG, Fulmek S, Matsumoto K, Cho R, Lyons PA, Levy ER, Scalzo AA, Yokoyama WM. 1997. A 2-Mb YAC contig and physical map of the natural killer gene complex on mouse chromosome 6. Genomics 42:16–25 45. Takei F, Brennan J, Mager DL. 1997. The Ly-49 family: genes, proteins, and recognition of class I MHC. Immunol. Rev. 155:67–77 46. Matsumoto N, Ribaudo RK, Abastado JP, Margulies DH, Yokoyama WM. 1998. The lectin-like NK cell receptor Ly-49A recognizes a carbohydrate-

47.

48.

49.

50.

51.

52.

53.

54.

55.

independent epitope on its MHC class I ligand. Immunity 8:245–54 Sundb¨ack J, Nakamura MC, Waldenstr¨om M, Niemi EC, Seaman WE, Ryan JC, K¨arre K. 1998. The alpha2 domain of H-2Dd restricts the allelic specificity of the murine NK cell inhibitory receptor Ly-49A. J. Immunol. 160:5971–78 Westgaard IH, Berg SF, Orstavik S, Fossum S, Dissen E. 1998. Identification of a human member of the Ly-49 multigene family. Eur. J. Immunol. 28:1839–46 Nakamura MC, Niemi EC, Fisher MJ, Shultz LD, Seaman WE, Ryan JC. 1997. Mouse Ly-49A interrupts early signaling events in natural killer cell cytotoxicity and functionally associates with the SHP-1 tyrosine phosphatase. J. Exp. Med. 185:673–84 Houchins JP, Lanier LL, Niemi EC, Phillips JH, Ryan JC. 1997. Natural killer cell cytolytic activity is inhibited by NKG2-A and activated by NKG2-C. J. Immunol. 158:3603–9 Mingari MC, Vitale C, Cantoni C, Bellomo R, Ponte M, Schiavetti F, Bertone S, Moretta A, Moretta L. 1997. Interleukin-15-induced maturation of human natural killer cells from early thymic precursors: selective expression of CD94/NKG2-A as the only HLA class I-specific inhibitory receptor. Eur. J. Immunol. 27:1374–80 Le Dr´ean E, V´ely F, Olcese L, Cambiaggi A, Guia S, Krystal G, Gervois N, Moretta A, Jotereau F, Vivier E. 1998. Inhibition of antigen-induced T cell response and antibody-induced NK cell cytotoxicity by NKG2A: association of NKG2A with SHP-1 and SHP-2 proteintyrosine phosphatases. Eur. J. Immunol. 28:264–76 Carretero M, Palmieri G, Llano M, Tullio V, Santoni A, Geraghty DE, L´opez-Botet M. 1998. Specific engagement of the CD94/NKG2-A killer inhibitory receptor by the HLA-E class Ib molecule induces SHP-1 phosphatase recruitment to tyrosine-phosphorylated NKG2-A: evidence for receptor function in heterologous transfectants. Eur. J. Immunol. 28:1280–91 Braud VM, Allan DSJ, O’Callaghan CA, S¨oderstr¨om K, D’Andrea A, Ogg GS, Lazetic S, Young NT, Bell JI, Phillips JH, Lanier LL, McMichael AJ. 1998. HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C. Nature 391:795–99 Borrego F, Ulbrecht M, Weiss EH, Coligan JE, Brooks AG. 1998. Recognition

P1: APL/NBL/vks

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

January 22, 1999

T1: APR

AR078-26

INHIBITORY RECEPTORS

56.

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

57.

58.

59.

60.

61.

62.

63.

64.

of human histocompatibility leukocyte antigen (HLA)-E complexed with HLA class I signal sequence-derived peptides by CD94/NKG2 confers protection from natural killer cell-mediated lysis. J. Exp. Med. 187:813–18 Lee N, Llano M, Carretero M, Ishitani A, Navarro F, Lopez-Botet M, Geraghty DE. 1998. HLA-E is a major ligand for the natural killer inhibitory receptor CD94/NKG2A. Proc. Natl. Acad. Sci. USA 95:5199–204 Braud V, Jones EY, McMichael A. 1997. The human major histocompatibility complex class Ib molecule HLA-E binds signal sequence-derived peptides with primary anchor residues at positions 2 and 9. Eur. J. Immunol. 27:1164–69 Lee N, Goodlett DR, Ishitani A, Marquardt H, Geraghty DE. 1998. HLA-E surface expression depends on binding of TAP-dependent peptides derived from certain HLA class I signal sequences. J. Immunol. 160:4951–60 Valiante NM, Uhrberg M, Shilling HG, Lienert-Weidenbach K, Arnett KL, D’Andrea A, Phillips JH, Lanier LL, Parham P. 1997. Functionally and structurally distinct NK cell receptor repertoires in the peripheral blood of two human donors. Immunity 7:739–51 Vance RE, Tanamachi DM, Hanke T, Raulet DH. 1997. Cloning of a mouse homolog of CD94 extends the family of C-type lectins on murine natural killer cells. Eur. J. Immunol. 27:3236–41 Ho EL, Heusel JW, Brown MG, Matsumoto K, Scalzo AA, Yokoyama WM. 1998. Murine Nkg2d and Cd94 are clustered within the natural killer complex and are expressed independently in natural killer cells. Proc. Natl. Acad. Sci. USA 95:6320–25 Sivakumar PV, Bennett M, Kumar V. 1997. Fetal and neonatal NK1.1+ Ly49-cells can distinguish between major histocompatibility complex class Ihi and class Ilo target cells: evidence for a Ly-49-independent negative signaling receptor. Eur. J. Immunol. 27:3100– 4 Toomey JA, Shrestha S, De la Rue SA, Gays F, Robinson JH, ChrzanowskaLightowlers ZMA, Brooks CG. 1998. MHC class I expression protects target cells from lysis by Ly49-deficient fetal NK cells. Eur. J. Immunol. 28:47–56 Su RC, Kung SKP, Gari´epy J, Barber BH, Miller RG. 1998. NK cells can recognize different forms of class I MHC. J. Immunol. 161:755–66

899

65. 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 66. Poggi A, Costa P, Morelli L, Cantoni C, Pella N, Spada F, Biassoni R, Nanni L, Revello V, Tomasello E, Mingari MC, Moretta A, Moretta L. 1996. Expression of human NKRP1A by CD34+ immature thymocytes: NKRP1A-mediated regulation of proliferation and cytolytic activity. Eur. J. Immunol. 26:1266–72 67. Adachi T, Flaswinkel H, Yakura H, Reth M, Tsubata T. 1998. Cutting edge: the B cell surface protein CD72 recruits the tyrosine phosphatase SHP-1 upon tyrosine phosphorylation. J. Immunol. 160:4662–65 68. Goitsuka R, Chen CH, Cooper MD. 1997. B cells in the Bursa of Fabricius express a novel C-type lectin gene. J. Immunol. 159:3126–32 69. Guthmann MD, Tal M, Pecht I. 1995. A secretion inhibitory signal transduction molecule on mast cells is another C-type lectin. Proc. Natl. Acad. Sci. USA 92:9397–401 70. Winter CC, Gumperz JE, Parham P, Long EO, Wagtmann N. 1998. Direct binding and functional transfer of NK cell inhibitory receptors reveal novel patterns of HLA-C allotype recognition. J. Immunol. 161:571–77 71. Biassoni R, Pessino A, Malaspina A, Cantoni C, Bottino C, Sivori S, Moretta L, Moretta A. 1997. Role of amino acid position 70 in the binding affinity of p50.1 and p58.1 receptors for HLA-Cw4 molecules. Eur. J. Immunol. 27:3095–99 72. Fan QR, Mosyak L, Winter CC, Wagtmann N, Long EO, Wiley DC. 1997. Structure of the inhibitory receptor for human natural killer cells resembles haematopoietic receptors. Nature 389:96–100 73. Fan QR, Garboczi DN, Winter CC, Wagtmann N, Long EO, Wiley DC. 1996. Direct binding of a soluble natural killer cell inhibitory receptor to a soluble human leukocyte antigen-Cw4 class I major histocompatibility complex molecule. Proc. Natl. Acad. Sci. USA 93:7178–83 74. Gumperz JE, Barber LD, Valiante NM, Percival L, Phillips JH, Lanier LL, Parham P. 1997. Conserved and variable residues within the Bw4 motif of HLAB make separable contributions to recognition by the NKB1 killer cell-inhibitory receptor. J. Immunol. 158:5237–41

P1: APL/NBL/vks

January 22, 1999

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

900

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

T1: APR

AR078-26

LONG

75. Kurago ZB, Lutz CT, Smith KD, Colonna M. 1998. NK cell natural cytotoxicity and IFN-gamma production are not always coordinately regulated: engagement of DX9 KIR+ NK cells by HLA-B7 variants and target cells. J. Immunol. 160:1573–80 76. Dohring C, Scheidegger D, Samaridis J, Cella M, Colonna M. 1996. A human killer inhibitory receptor specific for HLA-A. J. Immunol. 156:3098– 101 77. Cantoni C, Verdiani S, Falco M, Pessino A, Cilli R, Conte R, Pende D, Ponte M, Mikaelsson MS, Moretta L, Biassoni R. 1998. p49, a putative HLA class Ispecific inhibitory NK receptor belonging to the immunoglobulin superfamily. Eur. J. Immunol. 28:1980–90 78. Wagtmann N, Rojo S, Eichler E, Mohrenweiser H, Long EO. 1997. A new human gene complex encoding the killer cell inhibitory receptors and related monocyte/macrophage receptors. Curr. Biol. 7:615–18 79. Kuroiwa A, Yamashita Y, Inui M, Yuasa T, Ono M, Nagabukuro A, Matsuda Y, Takai T. 1998. Association of tyrosine phosphatases SHP-1 and SHP-2, inositol 5-phosphatase SHIP with gp49B1, and chromosomal assignment of the gene. J. Biol. Chem. 273:1070–74 80. Rojo S, Burshtyn DN, Long EO, Wagtmann N. 1997. Type I transmembrane receptor with inhibitory function in mouse mast cells and NK cells. J. Immunol. 158:9–12 81. Wang LL, Mehta IK, LeBlanc PA, Yokoyama WM. 1997. Mouse natural killer cells express gp49B1, a structural homologue of human killer inhibitory receptors. J. Immunol. 158:13–17 82. Katz HR, Vivier E, Castells MC, McCormick MJ, Chambers JM, Austen KF. 1996. Mouse mast cell gp49B1 contains two immunoreceptor tyrosine-based inhibition motifs and suppresses mast cell activation when coligated with the highaffinity Fc receptor for IgE. Proc. Natl. Acad. Sci. USA 93:10809–14 83. Samaridis J, Colonna M. 1997. Cloning of novel immunoglobulin superfamily receptors expressed on human myeloid and lymphoid cells: structural evidence for new stimulatory and inhibitory pathways. Eur. J. Immunol. 27:660–65 84. Arm JP, Nwankwo C, Austen KF. 1997. Molecular identification of a novel family of human Ig superfamily members that possess immunoreceptor tyrosinebased inhibition motifs and homology to

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

the mouse gp49B1 inhibitory receptor. J. Immunol. 159:2342–49 Cosman D, Fanger N, Borges L, Kubin M, Chin W, Peterson L, Hsu ML. 1997. A novel immunoglobulin superfamily receptor for cellular and viral MHC class I molecules. Immunity 7:273–82 Colonna M, Navarro F, Bell´on T, Llano M, Garc´ıa P, Samaridis J, Angman L, Cella M, L´opez-Botet M. 1997. A common inhibitory receptor for major histocompatibility complex class I molecules on human lymphoid and myelomonocytic cells. J. Exp. Med. 186:1809–18 Borges L, Hsu ML, Fanger N, Kubin M, Cosman D. 1997. A family of human lymphoid and myeloid Ig-like receptors, some of which bind to MHC class I molecules. J. Immunol. 159:5192–96 Cella M, Dohring C, Samaridis J, Dessing M, Brockhaus M, Lanzavecchia A, Colonna M. 1997. A novel inhibitory receptor (ILT3) expressed on monocytes, macrophages, and dendritic cells involved in antigen processing. J. Exp. Med. 185:1743–51 Colonna M, Samaridis J, Cella M, Angman L, Allen RL, O’Callaghan CA, Dunbar R, Ogg GS, Cerundolo V, Relink A. 1998. Cutting edge: human myelomonocytic cells express an inhibitory receptor for classical and nonclassical MHC class I molecules. J. Immunol. 160:3096–100 Kubagawa H, Burrows PD, Cooper MD. 1997. A novel pair of immunoglobulinlike receptors expressed by B cells and myeloid cells. Proc. Natl. Acad. Sci. USA 94:5261–66 Hayami K, Fukuta D, Nishikawa Y, Yamashita Y, Inui M, Ohyama Y, Hikida M, Ohmori H, Takai T. 1997. Molecular cloning of a novel murine cell-surface glycoprotein homologous to killer cell inhibitory receptors. J. Biol. Chem. 272:7320–27 Timms JF, carlberg K, Gu HH, Chen HY, Kamatkar S, Nadler MJS, Rohrschneider LR, Neel BG. 1998. Identification of major binding proteins and substrates for the SH2-containing protein tyrosine phosphatase SHP-1 in macrophages. Mol. Cell. Biol. 18:3838–50 Bl´ery M, Kubagawa H, Chen CC, V´ely F, Cooper MD, Vivier E. 1998. The paired Ig-like receptor PIR-B is an inhibitory receptor that recruits the protein-tyrosine phosphatase SHP-1. Proc. Natl. Acad. Sci. USA 95:2446–51 Meyaard L, Adema GJ, Chang CW, Woollatt E, Sutherland GR, Lanier LL,

P1: APL/NBL/vks

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

January 22, 1999

T1: APR

AR078-26

INHIBITORY RECEPTORS

95.

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

96.

97.

98.

99.

100.

101.

102.

103.

Phillips JH. 1997. LAIR-1, a novel inhibitory receptor expressed on human mononuclear leukocytes. Immunity 7:283–90 Poggi A, Pella N, Morelli L, Spada F, Revello V, Sivori S, Augugliaro R, Moretta L, Moretta A. 1995. p40, a novel surface molecule involved in the regulation of the non-major histocompatibility complex-restricted cytolytic activity in humans. Eur. J. Immunol. 25:369–76 Poggi A, Tomasello E, Revello V, Nanni L, Costa P, Moretta L. 1997. P40 molecule regulates NK cell activation mediated by NK receptors for HLA class I antigens and TCR-mediated triggering of T lymphocytes. Int. Immunol. 9:1271–79 Ohnishi H, Kubota M, Ohtake A, Sato K, Sano Si . 1996. Activation of proteintyrosine phosphatase SH-PTP2 by a tyrosine-based activation motif of a novel brain molecule. J. Biol. Chem. 271:25569–74 Fujioka Y, Matozaki T, Noguchi T, Iwamatsu A, Yamao T, Takahashi N, Tsuda M, Takada T, Kasuga M. 1996. A novel membrane glycoprotein, SHPS-1, that binds the SH2-domain-containing protein tyrosine phosphatase SHP-2 in response to mitogens and cell adhesion. Mol. Cell. Biol. 16:6887–99 Kharitonenkov A, Chen ZJ, Sures I, Wang HY, Schilling J, Ullrich A. 1997. A family of proteins that inhibit signalling through tyrosine kinase receptors. Nature 386:181–86 Yamao T, Matozaki T, Amano K, Matsuda Y, Takahashi N, Ochi F, Fujioka Y, Kasuga M. 1997. Mouse and human SHPS-1: molecular cloning of cDNAs and chromosomal localization of genes. Biochem. Biophys. Res. Commun. 231:61–67 Stofega MR, Wang HY, Ullrich A, Carter-Su C. 1998. Growth hormone regulation of SIRP and SHP-2 tyrosyl phosphorylation and association. J. Biol. Chem. 273:7112–17 Takada T, Matozaki T, Takeda H, Fukunaga K, Noguchi T, Fujioka Y, Okazaki I, Tsuda M, Yamao T, Ochi F, Kasuga M. 1998. Roles of the complex formation of SHPS-1 with SHP-2 in insulin-stimulated mitogen-activated protein kinase activation. J. Biol. Chem. 273:9234–42 Tsuda M, Matozaki T, Fukunaga K, Fujioka Y, Imamoto A, Noguchi T, Takada T, Yamao T, Takeda H, Ochi F, Yamamoto T, Kasuga M. 1998. Integrin-

104.

105.

106.

107.

108.

109.

110. 111.

112.

113.

901

mediated tyrosine phosphorylation of SHPS-1 and its association with SHP2-roles of Fak and Src family kinases. J. Biol. Chem. 273:13223–29 Kunath T, Ordonez-Garcia C, Turbide C, Beauchemin N. 1995. Inhibition of colonic tumor cell growth by biliary glycoprotein. Oncogene 11:2375–82 Beauchemin N, Kunath T, Robitaille J, Chow B, Turbide C, Daniels E, Veillette A. 1997. Association of biliary glycoprotein with protein tyrosine phosphatase SHP-1 in malignant colon epithelial cells. Oncogene 14:783–90 Stocks SC, Ruchaud-Sparagano MH, Kerr MA, Grunert F, Haslett C, Dransfield I. 1996. CD66: role in the regulation of neutrophil effector function. Eur. J. Immunol. 26:2924–32 Klein ML, McGhee SA, Baranian J, Stevens L, Hefta SA. 1996. Role of nonspecific cross-reacting antigen, a CD66 cluster antigen, in activation of human granulocytes. Infect. Immun. 64:4574– 79 Finger LR, Pu J, Wasserman R, Vibhakar R, Louie E, Hardy RR, Burrows PD, Billips LG. 1997. The human PD-1 gene: complete cDNA, genomic organization, and developmentally regulated expression in B cell progenitors [published erratum appears in Gene 1997 203:253;203(2):253]. Gene 197:177–87 Wu S-M, Cheung W-F, Frazier D, Stafford DW. 1991. Cloning and expression of the cDNA for human gamma-glutamyl carboxylase. Science 254:1634–36 Ryan JC, Seaman WE. 1997. Divergent functions of lectin-like receptors on NK cells. Immunol. Rev. 155:79–89 Biassoni R, Cantoni C, Falco M, Verdiani S, Bottino C, Vitale M, Conte R, Poggi A, Moretta A, Moretta L. 1996. The human leukocyte antigen (HLA)-Cspecific “activatory” or “inhibitory” natural killer cell receptors display highly homologous extracellular domains but differ in their transmembrane and intracytoplasmic portions. J. Exp. Med. 183:645–50 Mandelboim O, Davis DM, Reyburn HT, Val´es-G´omez M, Sheu EG, Pazmany L, Strominger JL. 1996. Enhancement of class II-restricted T cell responses by costimulatory NK receptors for class I MHC proteins. Science 274:2097–100 Bottino C, Sivori S, Vitale M, Cantoni C, Falco R, Pende D, Morelli L, Augugliaro R, Semenzato G, Biassoni R, Moretta L, Moretta A. 1996. A novel surface

P1: APL/NBL/vks

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

January 22, 1999

902

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

T1: APR

AR078-26

LONG molecule homologous to the p58/p50 family of receptors is selectively expressed on a subset of human natural killer cells and induces both triggering of cell functions and proliferation. Eur. J. Immunol. 26:1816–24 Campbell KS, Cella M, Carretero M, L´opez-Botet M, Colonna M. 1998. Signaling through human killer cell activating receptors triggers tyrosine phosphorylation of an associated protein complex. Eur. J. Immunol. 28:599– 609 Mason LH, Anderson SK, Yokoyama WM, Smith HRC, Winkler-Pickett R, Ortaldo JR. 1996. The Ly-49D receptor activates murine natural killer cells. J. Exp. Med. 184:2119–28 Raziuddin A, Longo DL, Mason L, Ortaldo JR, Bennett M, Murphy WJ. 1998. Differential effects of the rejection of bone marrow allografts by the depletion of activating versus inhibiting Ly49 natural killer cell subsets. J. Immunol. 160:87–94 Smith KM, Wu J, Bakker ABH, Phillips JH, Lanier LL. 1998. Cutting edge: Ly49D and Ly-49H associate with mouse DAP12 and form activating receptors. J. Immunol. 161:7–10 Lanier LL, Corliss BC, Wu J, Leong C, Phillips JH. 1998. Immunoreceptor DAP12 bearing a tyrosine-based activation motif is involved in activating NK cells. Nature 391:703–7 Lanier LL, Corliss B, Wu J, Phillips JH. 1998. Association of DAP12 with activating CD94/NKG2C NK cell receptors. Immunity 8:693–701 Moretta A, Sivori S, Vitale M, Pende D, Morelli L, Augugliaro R, Bottino C, Moretta L. 1995. Existence of both inhibitory (p58) and activatory (p50) receptors for HLA-C molecules in human natural killer cells. J. Exp. Med. 182:875–84 Mandelboim O, Kent S, Davis DM, Wilson SB, Okazaki T, Jackson R, Hafler D, Strominger JL. 1998. Natural killer activating receptors trigger interferon gamma secretion from T cells and natural killer cells. Proc. Natl. Acad. Sci. USA 95:3798–803 Rolstad B, Vaage JT, Naper C, Lambracht D, Wonigeit K, Joly E, Butcher GW. 1997. Positive and negative MHC class I recognition by rat NK cells. Immunol. Rev. 155:91–104 Dissen E, Ryan JC, Seaman WE, Fossum S. 1996. An autosomal dominant locus, Nka, mapping to the Ly-49 region

124.

125.

125a.

126.

127.

128.

129.

130.

131.

132.

of a rat natural killer (NK) gene complex, controls NK cell lysis of allogeneic lymphocytes. J. Exp. Med. 183:2197–207 Arase N, Arase H, Park SY, Ohno H, Ra C, Saito T. 1997. Association with FcRgamma is essential for activation signal through NKR-P1 (CD161) in natural killer (NK) cells and NK1.1+ T cells. J. Exp. Med. 186:1957–63 Cantoni C, Biassoni R, Pende D, Sivori S, Accame L, Pareti L, Semenzato G, Moretta L, Moretta A, Bottino C. 1998. The activating form of CD94 receptor complex: CD94 covalently associates with the Kp39 protein that represents the product of the NKG2-C gene. Eur. J. Immunol. 28:327–38 Held W, Kunz B. 1998. An allelespecific, stochastic gene expression process controls the expression of multiple Ly49 family genes and generates a diverse, MHC-specific NK cell receptor repertoire. Eur. J. Immunol. 28:2407– 16 Dorfman JR, Raulet DH. 1998. Acquisition of Ly49 receptor expression by developing natural killer cells. J. Exp. Med. 187:609–18 Uhrberg M, Valiante NM, Shum BP, Shilling HG, Lienert-Weidenbach K, Corliss B, Tyan D, Lanier LL, Parham P. 1997. Human diversity in killer cell inhibitory receptor genes. Immunity 7:753–63 Johansson MH, Bieberich C, Jay G, K¨arre K, H¨oglund P. 1997. Natural killer cell tolerance in mice with mosaic expression of major histocompatibility complex class I transgene. J. Exp. Med. 186:353–64 Salcedo M, Andersson M, Lemieux S, Van Kaer L, Chambers BJ, Ljunggren HG. 1998. Fine tuning of natural killer cell specificity and maintenance of self tolerance in MHC class I-deficient mice. Eur. J. Immunol. 28:1315–21 Wu MF, Raulet DH. 1997. Class I-deficient hemopoietic cells and nonhemopoietic cells dominantly induce unresponsiveness of natural killer cells to class I-deficient bone marrow cell grafts. J. Immunol. 158:1628–33 Dorfman JR, Zerrahn J, Coles MC, Raulet DH. 1997. The basis for selftolerance of natural killer cells in beta2microglobulin- and TAP-1-mice. J. Immunol. 159:5219–25 H¨oglund P, Glas R, M´enard C, K˚ase A, Johansson MH, Franksson L, Lemmonier F, K¨arre K. 1998. Beta2Microglobulin-deficient NK cells show

P1: APL/NBL/vks

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

January 22, 1999

T1: APR

AR078-26

INHIBITORY RECEPTORS

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

133.

134.

135.

136.

137.

138.

139.

increased sensitivity to MHC class I-mediated inhibition, but self tolerance does not depend upon target cell expression of H-2Kb and Db heavy chains. Eur. J. Immunol. 28:370–78 Carena I, Shamshiev A, Donda A, Colonna M, Libero GD. 1997. Major histocompatibility complex class I molecules modulate activation threshold and early signaling of T cell antigen receptor-gamma/delta stimulated by nonpeptidic ligands. J. Exp. Med. 186:1769–74 Fisch P, Meuer E, Pende D, Rothenfusser S, Viale O, Kock S, Ferrone S, Fradelizi D, Klein G, Moretta L, Rammensee HG, Boon T, Coulie P, van der Bruggen P. 1997. Control of B cell lymphoma recognition via natural killer inhibitory receptors implies a role for human Vgamma/Vdelta2 T cells in tumor immunity. Eur. J. Immunol. 27:3368– 79 Poccia F, Cipriani E, Vendetti S, Colizzi V, Poquet Y, Battistini L, L´opezBotet M, Fourni´e JJ, Gougeon ML. 1997. CD94/NKG2 inhibitory receptor complex modulates both anti-viral and anti-tumoral responses of polyclonal phosphoantigen-reactive Vgamma9Vdelta2 T lymphocytes. J. Immunol. 159:6009–17 De Maria A, Ferraris A, Guastella M, Pilia S, Cantoni C, Polero L, Mingari MC, Bassetti D, Fauci AS, Moretta L. 1997. Expression of HLA class Ispecific inhibitory natural killer cell receptors in HIV-specific cytolytic T lymphocytes: impairment of specific cytolytic functions. Proc. Natl. Acad. Sci. USA 94:10285–88 Bakker ABH, Phillips JH, Figdor CG, Lanier LL. 1998. Killer cell inhibitory receptors for MHC class I molecules regulate lysis of melanoma cells mediated by NK cells, gamma/delta T cells, and antigen-specific CTL. J. Immunol. 160:5239–45 Ortaldo JR, Winkler-Pickett R, Mason AT, Mason LH. 1998. The Ly-49 family: regulation of cytotoxicity and cytokine production in murine CD3+ cells. J. Immunol. 160:1158–65 Mingari MC, Ponte M, Bertone S, Schiavetti F, Vitale C, Bellomo R, Moretta A, Moretta L. 1998. HLA class I-specific inhibitory receptors in human T lymphocytes: Interleukin 15-induced expression of CD94/NKG2A in superantigenor alloantigen-activated CD8+ T cells. Proc. Natl. Acad. Sci. USA 95:1172–77

903

140. Mingari MC, Moretta A, Moretta L. 1998. Regulation of KIR expression in human T cells: a safety mechanism that may impair protective T-cell responses. Immunol. Today 19:153–57 141. Griffith TS, Brunner T, Fletcher SM, Green DR, Ferguson TA. 1995. Fas ligand-induced apoptosis as a mechanism of immune privilege. Science 270:1189–92 142. Apte RS, Sinha D, Mayhew E, Wistow GJ, Niederkorn JY. 1998. Cutting edge: role of macrophage migration inhibitory factor in inhibiting NK cell activity and preserving immune privilege. J. Immunol. 160:5693–96 143. Sun HW, Bernhagen J, Bucala R, Lolis E. 1996. Crystal structure at 2.6-∼A resolution of human macrophage migration inhibitory factor. Proc. Natl. Acad. Sci. USA 93:5191–96 144. Rouas-Freiss N, Gon¸calves RMB, Menier C, Dausset J, Carosella ED. 1997. Direct evidence to support the role of HLA-G in protecting the fetus from maternal uterine natural killer cytolysis. Proc. Natl. Acad. Sci. USA 94:11520– 25 145. P´erez-Villar JJ, Melero I, Navarro F, Carretero M, Bell´on T, Llano M, Colonna M, Geraghty DE, L´opez-Botet M. 1997. The CD94/NKG2-A inhibitory receptor complex is involved in natural killer cellmediated recognition of cells expressing HLA-G1. J. Immunol. 158:5736–43 146. Soderstrom K, Corliss B, Lanier LL, Phillips JH. 1997. CD94/NKG2 is the predominant inhibitory receptor involved in recognition of HLA-G by decidual and peripheral NK cells. J. Immunol. 159:1072–75 147. Pende D, Sivori S, Accame L, Pareti L, Falco M, Geraghty D, Le Bouteiller P, Moretta L, Moretta A. 1997. HLAG recognition by human natural killer cells. Involvement of CD94 both as inhibitory and as activating receptor complex. Eur. J. Immunol. 27:1875–80 148. Mandelboim O, Pazmany L, Davis DM, Val´es-G´omez M, Reyburn HT, Rybalov B, Strominger JL. 1997. Multiple receptors for HLA-G on human natural killer cells. Proc. Natl. Acad. Sci. USA 94:14666–70 149. Pazmany L, Mandelboim O, ValesGomez M, Davis DM, Reyburn HT, Strominger JL. 1996. Protection from natural killer cell-mediated lysis by HLA-G expression on target cells. Science 274:792–95 150. M¨unz C, Holmes N, King A, Loke YW,

P1: APL/NBL/vks

P2: APR/MBG

QC: APR/anil

15:24

Annual Reviews

January 22, 1999

904

151.

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

152.

153.

154. 155.

156.

T1: APR

AR078-26

LONG Colonna M, Schild H, Rammensee HG. 1997. HLA-G molecules inhibit NKAT3 expressing natural killer cells. J. Exp. Med. 185:385–91 Rouas-Freiss N, Marchal RE, Kirszenbaum M, Dausset J, Carosella ED. 1997. The alpha1 domain of HLA-G1 and HLA-G2 inhibits cytotoxicity induced by natural killer cells: Is HLA-G the public ligand for natural killer cell inhibitory receptors? Proc. Natl. Acad. Sci. USA 94:5249–54 Paul P, Rouas-Freiss N, Khalil-Daher I, Moreau P, Riteau B, Le Gal FA, Avril MF, Dausset J, Guillet JG, Carosella ED. 1998. HLA-G expression in melanoma: a way for tumor cells to escape from immunosurveillance. Proc. Natl. Acad. Sci. USA 95:4510–15 Biron CA. 1997. Activation and function of natural killer cell responses during viral infections. Curr. Opin. Immunol. 9:24–34 Ploegh HL. 1998. Viral strategies of immune evasion. Science 280:248–53 Reyburn HT, Mandelboim O, Val´esG´omez M, Davis DM, Pazmany L, Strominger JL. 1997. The class I MHC homologue of human cytomegalovirus inhibits attack by natural killer cells. Nature 386:514–17 Leong CC, Chapman TL, Bjorkman PJ, Formankova D, Mocarski ES, Phillips

157.

158.

159.

160.

JH, Lanier LL. 1998. Modulation of natural killer cell cytotoxicity in human cytomegalovirus infection: the role of endogenous class I major histocompatibility complex and a viral class I homolog. J. Exp. Med. 187:1681– 87 Farrell HE, Vally H, Lynch DM, Fleming P, Shellam GR, Scalzo AA, DavisPoynter NJ. 1997. Inhibition of natural killer cells by a cytomegalovirus MHC class I homologue in vivo. Nature 386:510–14 Forbes CA, Brown MG, Cho R, Shellam GR, Yokoyama WM, Scalzo AA. 1997. The Cmv1 host resistance locus is closely linked to the Ly49 multigene family within the natural killer cell gene complex on mouse chromosome 6. Genomics 41:406–13 Le Gall S, Erdtmann L, Benichou S, Berlioz-Torrent C, Liu L, Benarous R, Heard JM, Schwartz O. 1998. Nef interacts with the mu subunit of clathrin adaptor complexes and reveals a cryptic sorting signal in MHC I molecules. Immunity 8:483–95 Janeway CA Jr. 1998. Presidential address to the American Association of Immunologists. The road less traveled by: the role of innate immunity in the adaptive immune response. J. Immunol. 161:539–44

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:875-904. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:905–29 c 1999 by Annual Reviews. All rights reserved Copyright °

THE WISKOTT-ALDRICH SYNDROME PROTEIN (WASP): Roles in Signaling and Cytoskeletal Organization Scott B. Snapper 1,2,3 and Fred S. Rosen1,4 1Center

for Blood Research, 200 Longwood Avenue, Boston, Massachusetts 02115,

2Gastrointestinal Unit (Medical Services) and the Center for the Study of Inflammatory

Bowel Disease, Massachusetts General Hospital, Boston, Massachusetts 02114, 3Departments of Medicine and 4Pediatrics, Harvard Medical School, Boston, Massachusetts 02115 KEY WORDS:

Wiskott-Aldrich Syndrome, WASP, N-WASP, actin cytoskeleton, lymphocyte signaling

ABSTRACT The Wiskott-Aldrich Syndrome (WAS) is a rare X-linked primary immunodeficiency that is characterized by recurrent infections, hematopoietic malignancies, eczema, and thrombocytopenia. A variety of hematopoietic cells are affected by the genetic defect, including lymphocytes, neutrophils, monocytes, and platelets. Early studies noted both signaling and cytoskeletal abnormalities in lymphocytes from WAS patients. Following the identification of WASP, the gene mutated in patients with this syndrome, and the more generally expressed WASP homologue N-WASP, studies have demonstrated that WASP-family molecules associate with numerous signaling molecules known to alter the actin cytoskeleton. WASP/NWASP may depolymerize actin directly and/or serve as an adaptor or scaffold for these signaling molecules in a complex cascade that regulates the cytoskeleton.

WISKOTT-ALDRICH SYNDROME Clinical Features and Immunologic Abnormalities The Wiskott-Aldrich syndrome (WAS) is a rare immunodeficiency disease, which is transmitted as an X-linked recessive trait. Prominent features of the 905 0732-0582/99/0410-0905$08.00

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

906

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN

syndrome include recurrent pyogenic and opportunistic infections, eczema, and thrombocytopenia with small platelets. WAS was first described by Wiskott in 1936 when he observed thrombocytopenia, bloody diarrhea, eczema, and recurrent inner ear infections (otitis media) in three unrelated male infants (1). In 1954, Aldrich and his colleagues described affected males in five generations of a Dutch-American family and concluded that the syndrome was inherited in an X-linked recessive manner. Obligate female heterozygotes were asymptomatic (2). In 1968, Blaese et al (3) and Cooper et al (4) reported that affected males had a distinctive immunodeficiency disease. They noted T cell depletion of the thymus and paracortical areas of secondary lymphoid organs as well as peripheral blood lymphopenia, defective delayed-type hypersensitivity reactions, and failure to produce antibodies to polysaccharide (particularly isohemagglutinins) antigens (3, 4). Affected males consistently had low serum IgM, normal serum IgG, and elevated serum IgA and IgE levels (3–5). Subsequently, Ochs et al (5) documented a progressive decline in peripheral lymphocyte numbers in WAS patients after 6 years of age, reflecting a decrease in T cell numbers. Further examination of antibody responses revealed that, in addition to defective production of antibodies to polysaccharide antigens, immune responses to protein antigens are frequently blunted and not associated with isotype switching (3, 5, 6). Other pyogenic infections in addition to otitis media such as bacterial pneumonia, sinusitis and meningitis have been frequently encountered in affected males. Among the opportunistic infections, recurrent herpes simplex infection, Pneumocystis carinii pneumonia, and systemic varicella are common, as is Molluscum contagiosum, particularly in skin areas affected by eczema (7). Candidal infections are not common. The disease was considered to be fatal in the first decade of life due to infection or hemorrhage, but improved medical care has prolonged the life of affected males. Many patients now in the third and fourth decade of life are succumbing to B cell lymphomas, associated with Epstein-Barr virus, and immune-complex diseases such as chronic glomerulonephritis (7). Autoimmune hemolytic anemia, vasculitis, inflammatory bowel disease, and other autoimmune diseases have been observed in as many as 40% of affected males (7). Overall, the constellation of symptoms, signs, and laboratory findings of patients with WAS varies considerably, though all patients have thrombocytopenia and small platelets. The severest form of WAS cases, presenting with the classic triad of eczema, immunodeficiency, and thrombocytopenia, occurs less frequently (approximately 30%) than those cases presenting with only milder forms. A recent retrospective study of 154 WAS patients documented a significant number of patients with appropriate immune-responses to both protein and carbohydrate antigens, normal lymphocyte counts and functional assays, and normal immunoglobulin levels (7). The mildest form of the

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

WISKOTT-ALDRICH SYNDROME PROTEIN

907

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

disease, which manifests only as isolated thrombocytopenia with small platelets, has been designated X-linked thrombocytopenia (XLT) (8–12). Complete correction of all the abnormalities of the WAS was achieved with bone marrow transplantation 20 years ago (13). Some of the early attempts at bone marrow transplantation that were only partially successful, resulting in T cell but not B cell chimerism, nonetheless resulted in disappearance of eczema and restoration of normal immune responses, particularly to polysaccharide antigen. This suggested that B cell defects were unlikely to be the major cause of clinical immunologic symptoms (13).

Chromosomal Mapping and Identification of the WASP Gene A concerted effort was undertaken in the late 1980s to map the defective gene by examination of informative crossovers in over 75 kindred with affected and unaffected males. The gene was first localized to the pericentromeric region (14) and then was found to map to the short arm of the X-chromosome at position Xp11.22-.3 (15, 16). A high resolution map of Xp11.23 ultimately revealed the presence of polymorphic dinucleotide repeats near the defective gene, detected with the probe DXS6940 that can be used for prenatal diagnosis (17, 18). A positional cloning approach was used to identify the gene mutated in patients with WAS. This gene and its encoded protein were named WASP, for Wiskott-Aldrich syndrome protein (19). Subsequently, several reports have confirmed these findings and have demonstrated that the same gene is mutated in XLT (17, 20–22; reviewed in 23). Interestingly, in blood cells of obligate female heterozygotes who bear a WASP mutation on one of their X-chromosomes only the wild-type X-chromosome is active; the X-chromosome bearing the mutation is invariably inactivated (i.e. nonrandom X-inactivation) (24–28). Further examination of the patterns of X-chromosome inactivation in CD34+ hematopoietic stem cells of two obligate female heterozygotes revealed that these cells also exhibit nonrandom X inactivation (29). Even females who bear WASP mutations for the mild phenotype of XLT exhibit nonrandom X-chromosome inactivation in their blood cells (HD Ochs, FS Rosen, unpublished results). These observations suggested that WASP is critical for the development of hematopoietic stem cells; perhaps cells bearing wild-type WASP on the active X-chromosome have a survival advantage over those cells bearing the mutant gene on the active X-chromosome.

Signaling and Cytoskeletal Abnormalities in Lymphocytes T cells from WAS patients have both signaling and cytoskeletal abnormalities (reviewed in 30). T cell proliferative responses to antigen receptor stimulation

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

908

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN

(via anti-CD3ε antibodies) are severely depressed or absent (31). A decrease in IL-2 secretion is associated with the proliferative defect. In contrast, responses to nonspecific mitogens are often normal, suggesting a role for WASP in T cell receptor proximal signaling events (3, 5, 7, 31, 32). In B cells from WAS patients, however, the responses to antigen-receptor stimulation are less clear; both normal and abnormal proliferative responses have been described (33, 34). A unique cytoskeletal abnormality was noted in peripheral blood lymphocytes by scanning electron microscopy. Many cells had a paucity of cell surface microvilli (35). Interestingly, T cell lines derived from WAS patients that were grown in vitro in the presence of IL-2 and allogeneic stimulator cells, and thereby not exposed to the circulation, exhibited even greater cytoskeletal abnormalities than circulating peripheral blood T cells. When examined by scanning and transmission electron microscopy, many of these cells had abnormalities in cytoarchitecture, including surface blebbing, decreased microvilli, aborted mitosis, and an abnormal pattern of actin filaments (30, 32). Furthermore, antigen-receptor stimulation of T cell lines from WAS patients resulted in aberrant actin polymerization and the formation of abnormal cell shapes (36). There is also evidence that suggests that EBV transformed B-cell lines from WAS patients may have cytoskeletal defects (37). Defective regulation of the actin cytoskeleton may be responsible for the abnormal expression of cell surface glycoproteins (38). The major sialoglycoprotein of T cell membranes, CD43, localizes to microvilli (30, 39). CD43 is absent or markedly decreased in the membranes of WAS patients’ T cells (40, 41). An autosomal location precluded a causative role for CD43 in this syndrome (42, 43). Further examination has revealed that other cell surface sialoglycoproteins, such as CD75 and CD76, are also decreased in WAS patients’ blood T cells (44).

Defects in Platelets and Other Hematopoietic Cells All WASP mutations result in severe thrombocytopenia, with platelet counts typically 10% of normal (7). The platelets are also small in diameter and volume. The mean platelet diameter in normal individuals is 2.3 ± .12 µm; in WAS patients the mean platelet diameter is 1.82 ± .12 µm (45). Normal platelet volume ranges from 7.1 to 10.5 fl, whereas platelet volume in WAS patients varies from 3.8 to 5.0 fl (5). Splenectomy results in an increase in platelet number and platelet diameter (1.96 µm); however, neither parameter reaches normal values (46). The platelet abnormality appears to result from both ineffective thrombopoiesis and increased peripheral platelet destruction. The survival of normal platelets in WAS patients is normal (5, 47–49). Autologous platelet survival in WAS patients is reduced by approximately 50% (5), not sufficient to ascribe the profound thrombocytopenia to peripheral platelet destruction alone.

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

WISKOTT-ALDRICH SYNDROME PROTEIN

909

Although various defects have been reported with platelets from WAS patients, no consistent specific structural or functional abnormality has yet been appreciated (45, 49–53). Levels of procalpain, the Ca2+-dependent neutral protease involved in the cleavage of cytoskeletal proteins and the regulation of the actin cytoskeleton, are reduced in WAS platelets (54). Furthermore, recent studies have shown that intracellular Ca2+ levels are elevated in WAS platelets, which may cause aberrant Ca2+-mediated events, including calpain activation and microparticle release. Increased release of microparticles may account for the small size of WAS platelets; indeed a seven-fold increase in platelet microparticles is observed in the plasma of affected males (A Shcherbina, FS Rosen, E Remold-O’Donnell, unpublished results). Moreover, WAS platelet activation is associated with an increased expression of phosphatidylserine on the platelet surface, which may account for increased sequestration/destruction of WAS platelets by the spleen (A Shcherbina, FS Rosen, E Remold-O’Donnell, unpublished results). A defect in neutrophil and monocyte chemotaxis was noted in early studies with cells from WAS patients (5, 55); this suggested that WASP may have a more general effect on signaling pathways in hematopoietic cells by regulating cell motility and the cytoskeleton. In more recent studies, two groups have extended these studies demonstrating that monocytes from WAS patients fail to migrate and polarize their cytoskeleton in response to a variety of chemotactic agents, including formyl-methionyl-leucyl-phenylalanine (FMLP), monocyte chemoattractant protein-1 (MCP-1), macrophage inflammatory protein-1α (MIP-1α), and colony stimulating factor-1 (CSF-1) (56, 57). In contrast, neutrophil chemotaxis appeared normal in one study in response to the chemokines FMLP and CSF-1 (57).

THE WASP GENE AND PROTEIN Expression Patterns and Homologs WASP has been cloned from human and mouse; the proteins share 86% amino acid identity. Most studies have suggested that hWASP and mWASP have a pattern of RNA (19, 58, 59) (SB Snapper, unpublished data) and protein expression (60) limited to cells of the hematopoietic lineage, consistent with the cell types affected in patients with WAS. However, one study has reported a more general expression pattern, an apparent discrepancy that has not yet been resolved (61). One possible explanation for this finding might be that the rabbit polyclonal anti-WASP (N-terminus 279 amino acids) antibodies used in these studies also bind to the related neural-WASP (see below). The subcellular localization of WASP has been reported to be predominantly cytoplasmic (60, 62) with some protein found in membrane (16%) and nuclear (<3%) fractions (62).

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

910

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN

Recently, a WASP homologue, neural-WASP (N-WASP), was identified as a Grb2 interactor from bovine brain. N-WASP has approximately 50% homology at the protein level to WASP. The N-WASP expression pattern is more general with high levels of RNA expression in brain, colon, heart, lung, and testis. Protein expression has been demonstrated in brain, heart and lung (63). N-WASP has been cloned from rat, mouse (SB Snapper, unpublished results), and human (mapped to 7q31.3) (64). Northern blotting analyses suggest that murine N-WASP is also expressed in a variety of tissues including brain, colon, lung, heart, muscle, kidney, testis, liver, embryonic stem cells and lymphocytes (SB Snapper, unpublished results). WASP-like molecules are also expressed in more primitive organisms such as Saccharomyces cerevisiae, Dictyostelium discoideum and Xenopus laevis (65–67). The yeast WASP homolog, Bee1, has been characterized recently and shown to be involved in cytoskeletal organization (66, 68). Since SCAR, the novel Dictyostelium WASP-like molecule has numerous distinct uncharacterized homologs in human, mouse, and Caenorhabditis elegans, WASP may be a member of a larger family of molecules that coordinate the actin cytoskeleton (67).

Genomic Structure The genomic structure of WASP has been determined for both human and mouse. Each gene contains 12 exons, nearly identical in size and composition, which span less than 10 kb genomic DNA (19, 58, 69). In an effort to determine the basis for the tissue specific expression of WASP, the genomic sequences upstream of the transcriptional start site were investigated for putative regulatory elements. Several hematopoietic-specific binding sequences including two Ets-1 binding motifs were found (70).

Protein Structure Given the signaling and cytoskeletal abnormalities in lymphocytes from WAS patients, it seemed likely that the gene product would be involved in regulating a signaling pathway that connected cell surface activation with regulation of the cytoskeleton. Analysis of the amino acid sequence of the encoded protein confirmed this prediction. The WASP gene encodes for a 502 amino acid proline-rich protein; prolines account for >15% of the entire protein sequence (see Figure 1) (19). Dispersed among the various proline enriched regions are numerous sequences that match the Src-homology-3 (SH3) binding core consensus sequence, P-p-X-P (where the P’s are invariant prolines critically involved in binding, and the X and p tend to be hydrophobic and proline residues, respectively) (71). Proteins containing SH3-binding domains (e.g. Btk, Sos, dynamin, PI3K) and SH3 domains

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

WISKOTT-ALDRICH SYNDROME PROTEIN

911

Figure 1 Functional domains in WASP and their putative interactors. WASP protein sequence are denoted with solid black line. Functional domains are depicted as filled rectangles. Shown are WASP homology 1 (WH1)/pleckstrin homology (PH) domain; basic residues (BR), GTPase (Cdc42) binding domain (GBD); polyprolines with SH3-binding domains, WASP homology 2 (WH2)/verprolin homology (VH) domain, cofilin homology (CH) domain and acidic residues (AR). Thick arrows point to targets of these functional domains. For N-WASP, Miki et al (84) suggest that an intramolecular interaction between the BR and AR (shown by thin arrow above sequence) masks the CH domain, thereby preventing it from interacting with actin (84). Binding of GTP-loaded Cdc42 to the GBD domain release the CH domain, allowing it to depolymerize actin filaments. No data is yet available for WASP. See text for details.

(e.g. Grb 2, PLCγ 1) are involved in diverse cellular functions including cell signaling and cytoskeletal organization (72). WASP and N-WASP also contain a GTPase binding domain (GBD) or Cdc42/Rac Interactive Binding (CRIB) motif (73) that has been shown to interact with the Rho family GTPase Cdc42 and, to a lesser extent, Rac (65, 74, 75). As discussed in more detail below, the Rho family of small GTPases are members of the Ras superfamily and are known to regulate various signaling pathways including those that control cytoskeletal reorganization (76, 77). Further database analysis identified two regions in WASP, denoted as WASP Homology 1 and 2 (WH1 and WH2), located in the N- and C-terminus of the protein, respectively, which are also present in other cytoskeleton-associated proteins that contain polyproline stretches. The vasodilator-stimulated phosphoprotein (VASP), Drosophila ENA (dena) and homer contain regions homologous to WH1 (65, 78). WH1 has also been shown to overlap with a pleckstrin homology (PH) domain, perhaps mediating localization of WASP and N-WASP to the membrane upon activation (63). Homology to the WH2 region was observed in verprolin (63, 65), a yeast protein involved in regulating cytoskeletal organization (79, 80), as well as the WASP-interacting protein (WIP) (81), and the WASP-like molecule SCAR from Dictyostelium (67). The WH2/verprolin homology (VH) domain directly mediates N-WASP binding to actin (82).

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

912

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN

WASP and N-WASP each contain a cofilin homology domain at the Cterminus (63). Cofilin is an actin binding protein that when activated, accelerates the depolymerization of actin filaments (83). In support of a role for WASP-family molecules in actin-depolymerization, Miki et al (63, 84) demonstrated that full length N-WASP decreased the viscosity of F-actin and that a small fragment of N-WASP, containing the cofilin homology region, decreased the sedimentibility of F-actin. WASP contains basic and acidic regions located directly upstream of the GBD and at the extreme C-terminus, respectively (19, 63). Miki et al (84) showed that these regions interact with each other in vitro and suggested that intramolecular interactions between them prevent the cofilin homology region from depolymerizing actin. A potential nuclear localization signal has also been found within WASP that may mediate WASP trafficking between the cytoplasm and nucleus (19). The protein structures of each of the WASP homologues (WASP, N-WASP, and Bee1) are similar except that the yeast homologue Bee1 does not contain a GBD domain and that N-WASP, but not WASP, contains an IQ motif. Since IQ motifs bind Ca2+/calmodulin, this suggests that N-WASP’s function may be influenced by calcium ions (85).

Binding Partners CDC42 AND THE RHO-FAMILY GTPASES Independent studies have demonstrated that WASP interacts with the GTP-bound form of Cdc42 and to a lesser degree Rac1 (65, 74, 75). Cdc42 and Rac are members of the Rho family of small GTPases that regulate the architecture of the cytoskeleton as well as direct signals to the nucleus (76). Aspenstrom et al (75) and Symons et al (65) isolated WASP while searching for novel Cdc42 interactors, using a yeast-two hybrid approach and affinity chromatography, respectively. Kolluri et al (74) specifically sought out an interaction between WASP and Cdc42 because of the cytoskeletal abnormalities in WAS lymphocytes (32, 35) and the known role of Cdc42 in regulating the actin cytoskeleton in lymphocytes (86). The interaction between WASP and Cdc42 was formally shown to be dependent upon the GBD/CRIB motif (65, 75); however, tight binding appears to require additional WASP sequences (87). Subsequent studies demonstrated a similar interaction between N-WASP and Cdc42 (84). SH3 DOMAIN-CONTAINING PROTEINS WASP has been identified as a binding partner for various SH3 domain-containing proteins in vitro (e.g. Itk, Btk, Grb2, PLCγ 1, c-src, c-Fgr, p85α) (22, 88–92) and in vivo (Nck, Fyn, PSTPIP, Grb-2). Each of these WASP in vivo interactors has been connected to signaling pathways linked to the actin cytoskeleton.

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

WISKOTT-ALDRICH SYNDROME PROTEIN

913

Nck WASP interacts with the SH3 domain of Nck in vivo (62). Nck, a ubiquitously expressed adaptor molecule composed of one SH2 domain and three SH3 domains (93), associates with numerous proteins including signaling molecules downstream of the Rho family GTPases that activate the stress-associated protein kinase (SAPK)/Jun kinase pathway (e.g. PAK1, PRK2, NAP) (94–98). Although the function of Nck is not clear, a link to the cytoskeleton was implicated with the finding that mutations in Dock, the Drosophila Nck homologue, impair axonal guidance (99). Furthermore, Nck also interacts with the WASP interacting protein (WIP) (100), another WASP binding partner known to influence the actin cytoskeleton (81). Fyn The nonreceptor tyrosine kinase Fyn interacts in vivo with WASP via its SH3 domain. Fyn is implicated in numerous lymphocyte signaling pathways (including associations with CD2, CD28, CD43, Fyb; SKAP55, Fas, PI3 kinase, SAM68, FAK125) (101–108), with links to the actin cytoskeleton (109–111). Grb2 The adaptor molecule Grb2 is also an in vivo interacting partner for WASP and N-WASP (61, 92). Containing one SH2 domain and two SH3 domains, Grb2 associates with cytoskeletal associated proteins and has been shown to regulate the actin cytoskeleton upon growth factor receptor stimulation (112, 113). Proline serine threonine phosphatase interacting protein (PSTPIP) The cytoskeletal-associated protein PSTPIP has been found to interact with WASP (114). The SH3 domain of PSTPIP interacts with a polyproline rich region of WASP and appears to require at least WASP amino acids 350-384. PSTPIP colocalizes to the cortical actin cytoskeleton and can induce actin-containing membrane protrusions such as lamellipodia when overexpressed in NIH 3T3 cells (115). PSTPIP and the related Schizosaccharomyces pombe phosphoprotein CDC15p both may regulate cytokinesis (115). WASP-INTERACTING PROTEIN (WIP) Using a yeast two-hybrid approach, Ramesh et al (81) screened a human T cell cDNA library and identified the novel ubiquitously expressed WASP interacting protein, WIP. WIP binding requires a region of WASP including WH1 and some neighboring proline residues. Interestingly, like WASP, WIP is proline rich, binds Nck, and contains a verprolin homology region (81, 100). WIP also binds profilin, an actin binding protein that promotes actin polymerization (81). When introduced into permeabilized cells, WIP leads to actin polymerization and alters the actin cytoskeleton (81). Furthermore, WIP can complement for verprolin (VRP1) function in yeast; conditional vrp1-1 mutants have numerous cytoskeletal defects (80, 116), which are fully complemented by WIP (AK Hopper, N Ramesh, R Geha, unpublished results).

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

914

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

ACTIN An indirect relationship between WASP and actin was first suggested by WASP overexpression studies in lymphocytes, epithelial and endothelial cells. Immunofluorescence was used to demonstrate the colocalization of Factin and WASP (65). Subsequently, similar experiments with N-WASP also demonstrated that F-actin colocalizes with N-WASP when the latter is overexpressed (63). These studies were extended to show that N-WASP bound actin directly and could be co-immunoprecipitated with actin from cell lysates, an interaction requiring an intact WH2/VH domain (63, 82). PHOSPHATIDYLINOSITIOL 4,5, BISPHOSPHATE (PIP2) Miki et al (63) demonstrated that the N-WASP WH1/PH domain can interact with PIP2. In overexpression studies, a point mutation within this domain (C38W) led to reduced PIP2 binding and shifted the largely cytoplasmic staining pattern to a nuclear pattern. PIP2 interacts with numerous proteins linked to the actin cytoskeleton (e.g. cofilin, gelsolin, capZ, α-actinin, profilin, and vinculin) (117) and regulates the actin cytoskeleton directly (118, 119).

WASP MUTATIONS AND CLINICAL PHENOTYPE The elucidation of a number of potential functional domains of WASP has provided the framework to begin to study genotype/phenotype relationships in WAS patients. WASP genotypic analysis by several groups has revealed approximately 140 different mutational events in over 200 kindreds (17, 19–21, 23, 120–129). Missense and nonsense mutations have been found scattered throughout 12 exons of the gene together with deletions, insertions, and splice site mutations in many of the introns (summarized in 6). However, there is a remarkable cluster of missense mutations in the exons encoding the WH1/PH domain, specifically in exon 2 (particularly at the Arg86 codon), and, to a lesser extent, in exons 1 and 3. These mutations are associated with the mild WAS phenotype characterized by XLT. Except for the relationship between missense mutations within the WH1/PH domains and isolated thrombocytopenia, no clear associations between specific mutations within functional domains of WASP and clinical phenotype have been reported. Although nonsense mutations are often associated with no WASP expression, there is significant variability in WASP missense mutations that result in decreased protein expression and those that result in no detectable protein. As expected, mutations resulting in mild WAS phenotypes tend to be associated with detectable protein expression, whereas mutations resulting in severe WAS phenotypes are frequently associated with no protein expression (22, 60, 128, 129). Females with a WAS phenotype have been described. In two families, multiple females were affected (130, 131); however, they do not appear to have

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

WISKOTT-ALDRICH SYNDROME PROTEIN

915

mutations in WASP. One female with WAS has recently been reported who does have a WASP mutation (132). However, her mother and maternal grandmother have mutations in Xist, a gene required for random X chromosomal inactivation. Cells carrying this Xist mutation inactivate the X-chromosome bearing the mutation. The female child with the WAS inherited the mutated X-chromosome from her mother. The X-chromosome from her father had a de novo mutation in WASP. Since all her maternally derived X-chromosomes were inactivated because of the Xist mutation, only her paternal X-chromosome with the WASP mutation remained active. This novel inheritance of an immunodeficiency has not been observed in other X-linked immunodeficiency diseases and must be a very rare event (132).

WASP-DEFICIENCY: OF MICE AND MEN WASP-deficient mice containing a targeted disruption of the GBD/CRIB motif have recently been generated and share some, but not all, of the features of the human immunodeficiency (69). Despite normal overall lymphocyte development, both WASP-deficient mice and WAS patients have reduced blood lymphocyte counts (4, 5, 69). Similar to those in human studies (31), T cells from WASP-deficient mice fail to proliferate in response to antigen-receptor stimulation (via anti-CD3ε antibodies) (69), while responding normally to signals that bypass the antigen receptor (via the combination of a phorbol ester and calcium ionophore). However, in contrast to studies with human WAS T cells (31), costimulatory signals mediated by anti-CD28 antibodies, at least partially, rescue the anti-CD3ε-induced proliferative defects in T cells from WASP-deficient mice (69). Furthermore, WASP-deficient T cells in mice are defective in their ability to form antigen-receptor caps upon TCR stimulation (69), a process known to require reorganization of the actin cytoskeleton. T cells from WAS patients also show defects in the actin cytoskeleton upon antigen-receptor induced stimulation (36). In contrast to the results in mice with mutant T cells, B cells from WASP-deficient mice both proliferate normally and form antigenreceptor caps upon antigen-receptor stimulation (69). However, some studies with B cells from WAS patients demonstrate abnormal actin polymerization and antigen-receptor induced proliferative responses (33, 37). The difference in the phenotypic consequence of WASP deficiency in murine T- and B-cells may result either from different roles for WASP in T and B cell function or from functional redundancy for WASP in T but not B cells. Since the WASP-deficient mice have been raised in a specific pathogen-free facility, the extent, if any, to which these mice are immunodeficient as compared to WAS patients remains to be established (69). The majority of WASPdeficient mice develop chronic colitis by four months of age (69), a phenotype

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

916

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN

that has been seen rarely in human WAS patients (7, 133). While the initiation of colitis in WASP mutant mice may result from an as yet undetected infectious agent, chronic colitis develops in numerous genetically manipulated mice where gene disruption affects T cell function (e.g. TCR-α −/−, IL-2−/−, IL-2Rα −/−, IL-10−/−, MHC class II−/−; reviewed in 134). In contrast to studies with human WAS patients (3–5), WASP-deficient mice respond to both protein and carbohydrate antigenic stimulation and have normal immunoglobulin concentrations in the sera. Furthermore, neither hematopoietic malignancies nor eczema develops in young WASP-mutant mice (69). Clearly, the role of genetic background on the immunologic phenotypic consequences of WASP-deficiency in mice still needs to be assessed. It has been demonstrated in some, but not all, studies that WASP-deficiency results in defective migration of monocytes and neutrophils in human WAS patients (5, 55–57). Aberrant homing of murine WASP-deficient lymphocytes, neutrophils, and eosinophils has also been recently observed (SB Snapper, unpublished results). In comparison to human WAS patients, initial platelet studies with WASPdeficient mice suggest a more limited role for WASP in platelet function in mice. Platelet numbers are only modestly reduced in WASP-deficient mice and are normal in size (69). Furthermore, preliminary data suggest that WASPdeficient platelets have a normal half-life when transferred into wild-type mice (SB Snapper, D Wagner, unpublished results).

WASP, N-WASP, AND THE CYTOSKELETON: A ROLE FOR THE RHO FAMILY GTPASES WASP overexpression in epithelial cells, aortic endothelial cells, or lymphocytes leads to the formation of ectopic actin clusters that stain with phallotoxins in a largely perinuclear distribution (65). This pattern of expression requires the C-terminus of WASP, with its WH2/VH and cofilin homology regions (65). The effect of WASP on the cytoskeleton seems to be mediated, at least in part, through its interaction with the activated (GTP-bound) form of Cdc42. In addition to regulating cell transformation, activation of the stress-associated protein kinase (SAPK)/Jun-kinase pathway, and cell cycle progression, Cdc42 and the other Rho family GTPases coordinate the maintenance of cell shape, motility, and cytokinesis through their regulation of the actin cytoskeleton (76). The Rho family GTPases appear to be organized in a linear cascade to regulate actin reorganization and the subsequent formation of various actin-containing complexes: Activated Cdc42 induces filopodia formation and GTP-loading of Rac; activated Rac induces lamellipodia formation and GTP-loading of Rho; and

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

WISKOTT-ALDRICH SYNDROME PROTEIN

917

finally, activated Rho induces actin stress fiber formation and specialized adhesion complexes (135). Filopodia are composed of bundles of actin filaments that are often found protruding from the cell surface of motile cells. Lamellipodia, also associated with the cell surface, are composed of networks of polymerized actin and actin-binding proteins in broad, flat, sheet-like structures. Filopodia and lamellipodia are generally formed in response to migratory stimuli (e.g. chemoattractants). Actin stress fibers consist of bundles of actin filaments that traverse the cell and bind to focal adhesion complexes and integrins at the cell surface. The connection between WASP, Cdc42, and the cytoskeleton was initially demonstrated in microinjection studies. Dominant negative Cdc42, but not dominant negative Rac1 or RhoA, blocked the WASP-induced formation of ectopic actin clusters in aortic endothelial cells (65). Furthermore, WASP expression abrogated the constitutively active Cdc42-mediated cytoskeletal changes in these cells (65). Similar studies in COS cells showed that WASP blocked the induction of cellular protrusions by constitutively active Cdc42 (84). When overexpressed in COS cells, the distribution of N-WASP, compared to that of WASP, appears quite different. N-WASP is found in both the cytoplasm and the nucleus, colocalizing with accumulated actin filaments in cortical areas (63). A point mutation in the WH1/PH domain or deletion of the WH2/VH and cofilin homology region resulted in a nuclear pattern of N-WASP expression and abrogation of the cytoskeletal changes (63). These data suggest that regions mediating both actin binding and interaction with the membrane may be rquired for proper localization and function of the protein. In contrast to the studies with WASP, N-WASP appears to participate in the normal formation of filopodia. Co-expression of N-WASP with constitutively active Cdc42 augments the ability of Cdc42 to induce filopodia in COS cells (84). Cells co-expressing constitutively active Cdc42 with N-WASP, mutated in either the GBD domain or the cofilin homology region, failed to form filopodia. That an endogenous interaction between N-WASP and Cdc42 is required in the induction of filopodia was also demonstrated in Swiss 3T3 cells (84). Bradykinin induces filopodia in 3T3 cells in a Cdc42 dependent manner. Microinjected anti-N-WASP antibodies blocked the bradykinin-induced filopodia formation in Swiss 3T3 cells. Miki et al (84) concluded that N-WASP, but not WASP, is a downstream effector for filopodia formation. One alternative explanation for the different roles observed between WASP and N-WASP in the Cdc42-induction of filopodia may be that since WASP expression is limited to hematopoietic cells, the requisite WASP interactors that facilitate such cytoskeletal changes may not be present in COS cells or endothelial cells.

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

918

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN

A requirement for an interaction between GTP-loaded Cdc42 and WASPfamily molecules to induce cytoskeletal changes was recently challenged (136). GTP-loaded Cdc42 interacts with WASP, N-WASP, and numerous other proteins via the GBD/CRIB motif (73). A mutation within constitutively active Cdc42 (Y40C) that abolished or reduced its binding with GBD/CRIB motifcontaining proteins including WASP had no effect on the ability of the mutated Cdc42 to induce cytoskeletal change or cell cycle progression, but prevented SAPK/JNK activation (136). This study, as well as another study focusing on identical mutations in Rac (137), concluded that GBD/CRIB containing proteins, such as WASP and N-WASP, are not required for the cytoskeletal changes induced by Rho family GTPases. Since Miki et al (84) subsequently determined that the Cdc42 Y40C mutation did not abolish the ability of N-WASP to interact with Cdc42, this issue still needs further resolution. In the resting state, filament turnover is slow because the assembly end (barbed ends) of most of the intracellular F-actin filaments is largely bound to a variety of capping proteins that prevent further actin polymerization. To generate new structures, the barbed ends of actin must be either uncapped, new free ends formed as a result of actin severing/depolymerization, or nucleating proteins, that initiate actin assembly de novo, must be activated (e.g. Arp2/3 complex; reviewed in 138). A model has been proposed (see Figure 1) for GTP-loaded Cdc42 regulating the ability of the cofilin homology domain of WASP/N-WASP to initiate events leading to the formation of new actin filament containing structures (82). In the absence of GTP-bound Cdc42 binding to WASP, a basic region upstream of the GBD motif in WASP interacts with the acidic residues at the extreme C-terminus (see Figure 1; 84). It has been suggested that this interaction masks the cofilin homology region and thereby prevents the initiation of actin depolymerization (84). Upon association between GTP-Cdc42 and WASP, the binding of the basic and acidic regions is prevented, allowing the cofilin homology region to interact with actin initiating depolymerization (84). As a result of such actin depolymerization/severing, uncapped actin ends are generated, which can then serve as the nidus for subsequent actin polymerization and the formation of new actin-containing structures (e.g. filopodia) (84). Two N-WASP interactors, WIP (I Anton, N Ramesh, R Geha, unpublished results) and profilin (84), which stimulate actin polymerization (81, 139), may also contribute to the induction of actin polymerization that follows F-actin filament depolymerization. In the case of WIP, a trimolecular complex between WASP, WIP, and GTP-loaded Cdc42 has not been identified (N Ramesh, R Geha, unpublished results). It is tempting to speculate that GTP-loaded Cdc42 binding to WASP may serve as a switch for both the unmasking of the cofilin homology domain, to initiate actin depolymerization, and WIP release, to induce subsequent actin polymerization.

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

WISKOTT-ALDRICH SYNDROME PROTEIN

919

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

POTENTIAL ROLES FOR WASP IN T-CELL SIGNALING The events that integrate incoming cell surface signals into cytoskeletal changes and subsequent alterations in cellular function(s) (e.g. lymphokine secretion, motility, adhesion) in hematopoietic cells are complex. T lymphocytes undergo a dramatic change in cytoskeletal organization upon contact with antigenpresenting cells (APC) with polarization of the T cell microtubule organizing center (MTOC) toward the contact surface, changes that may be responsible for directed lymphokine secretion and cytotoxicity (140–142). Numerous reports have linked components of the T cell antigen receptor (TCR) complex with the cytoskeleton (143, 144, 144a) and have demonstrated that TCR activation is associated with actin polymerization or cytoskeletal change (145–151, 151a,b). In addition to WASP (36, 69), several potential downstream targets of TCR activation have been implicated in the signaling cascade leading to cytoskeletal reorganization including Lck, ZAP-70, Syk, Vav, protein kinase C-θ and Cdc42 (86, 146, 148, 152–156). However, the mechanisms whereby WASP and these other signaling molecules affect cytoskeletal organization and T cell activation are yet unknown. Several recent studies have shed light upon how cytoskeletal organization and T cell activation may be directly connected. Valitutti et al (157, 158) suggested that an intact cytoskeleton facilitates serial triggering of TCRs by limited MHCpeptide complexes at the contact surface between T cells and APCs. Shaw & Dustin (159) proposed that the essential requirement for T cell activation is a contact cap that allows for clustering of TCR and associated molecules (e.g. CD4, CD28, CD2, lck, fyn, Zap70) and, most importantly, the physical exclusion of the CD45 phosphatase. Recent studies by two groups demonstrated “receptor patterning” at the contact surface in T cell/APC interactions—with larger adhesive molecules such as LFA-1 found at the periphery of the T cell contact surface (denoted “peripheral supramolecular activation clusters”) and smaller adhesive molecules such as CD2 located at the center of the T cell contact cap (denoted “central supramolecular activation clusters”) (160, 161). The close proximity (<15 nm) between the T cell surface and the APC at the contact center are mediated by these smaller adhesive molecules and facilitate interactions between the TCR-MHC-peptide complex. As a result, TCR molecules become clustered at the center of the contact, overcoming the inherent low avidity of TCR-APC interactions (160, 161). Consistent with studies demonstrating T cell MTOC polarization following contact with APCs (140–142), this precise three-dimensional array of molecules that occurs upon contact between T cells and APCs must require cytoskeletal reorganization. Receptor patterning and T cell polarization seem to be linked (154, 160, 161). For example, studies by

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

920

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN

Dustin et al (161) demonstrated that the cytoplasmic domain of CD2 as well as a novel CD2-associated protein (CD2AP) is essential for both appropriate receptor patterning and MTOC polarization. Although not yet demonstrated, a role for WASP in facilitating receptor patterning and MTOC polarization seems likely. The clustering of surface receptors upon T cell receptor cross-linking (“capping”—162), a process that requires actin polymerization (163, 164), may be the “in vitro” correlate of receptor patterning and contact cap formation. Recent studies have indicated that assembly of the actin cytoskeleton and capping may have a direct role in mediating signals from the TCR. T lymphocytes deficient in either the hematopoietic cell–specific Rho family GDP-GTP exchange factor Vav or WASP have defects in actin polymerization, capping, and antigen receptor-induced proliferative responses (31, 36, 69, 155, 156). Numerous other phenotypic similarities exist between lymphocytes from WASP-deficient and Vav-deficient mice, including normal B cell development, peripheral B cell numbers, and Ig isotypes (69, 165–167). These overlapping aspects of the Vav-deficient and WASP-deficient phenotypes are consistent with the possibility that Vav and WASP are members of a common signaling cascade in T cells that regulates capping and proliferation upon antigen-receptor stimulation. The proliferation defects of both Vav and WASP-deficient T cells may result from a defect in activation of a downstream signaling pathway dependent on the capping process (69, 155, 156). Although not reported for Vav-deficient mice, the fact that murine WASP-deficient B cells cap and proliferate normally lends further support to this notion. Since both Vav and WASP influence Cdc42mediated effects on the actin cytoskeleton in vivo (65, 84, 168), the activation of signaling events leading to capping is also likely to involve Rho family GTPases. The more downstream effectors of Vav, Rho-family GTPases, and WASP that ultimately give rise to structural changes are unknown, but they are likely to involve direct WASP-mediated effects via the cofilin homology domain as well as contributions from WASP interactors (e.g. WIP, SH3 domain containing molecules), kinases, cytoskeletal-associated phosphoproteins (e.g. focal adhesion kinase, paxillin, and tensin) and other structural proteins (e.g. α-actinin, vinculin, talin, integrin, ezrin, radixin, moesin). Together these structural and signaling molecules must be activated and coordinated in a well-integrated process that leads to shape changes that regulate cell function and movement.

CONCLUSIONS WAS patients have a wide spectrum of abnormalities ranging from mild isolated thrombocytopenia to a life-threatening immunodeficiency with recurrent

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

WISKOTT-ALDRICH SYNDROME PROTEIN

921

infections, malignancies, and autoimmune disease. The identification and characterization of the defective protein, WASP, have begun to shed light upon the mechanisms responsible for the signaling and cytoskeletal defects in hematopoietic cells. WASP and its more generally expressed homolog N-WASP are intracellular signaling molecules that are clearly involved in regulating the actin cytoskeleton in mammalian cells. Over the past several years, we have witnessed an expanding number of proteins that interact in vivo with WASP; however, the mechanism(s) by which WASP/N-WASP and these associated molecules control cytoskeletal changes remain largely unknown. The elucidation of upstream signaling events that lead to the “activation” of WASP/N-WASP, such as changes in subcellular localization, phosphorylation (169), and the formation of supra-molecular complexes, remains a priority. However, with many of the biochemical and genetic tools in hand, the future seems bright. ACKNOWLEDGMENTS We apologize to the numerous investigators whose work we were unable to cite because of space limitations. S.B.S. thanks Frederick Alt and members of the Alt laboratory for a stimulating laboratory environment. We thank Narayanaswamy Ramesh and Eileen Remold-O’Donnell for sharing unpublished results and Timo Breit, Dianne Kenney, Geoff Parsons, David Fruman, Sapna Syngal, Herv´e Falet, Christoph Klein, Ching-Hui Liu, Jayanta Chaudhuri and John Hartwig for critically evaluating the manuscript. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Wiskott A. 1936. Famili¨arer, angeborener Morbus Werlhoffi? Monatsschr. Kinderheil. 68:212–16 2. Aldrich RA, Steinberg AG, Campbell DC. 1954. Pedigree demonstrating a sexlinked recessive condition characterized by draining ears, eczematoid dermatitis and bloody diarrhea. Pediatrics 13:133– 39 3. Blaese RM, Strober W, Brown RS, Waldmann TA. 1968. The WiskottAldrich syndrome. A disorder with a possible defect in antigen processing or recognition. Lancet 1:1056–61 4. Cooper MD, Chase HP, Lowman JT, Krivit W, Good RA. 1968. WiskottAldrich syndrome. An immunologic deficiency disease involving the afferent limb of immunity. Am. J. Med. 44:499– 513

5. Ochs HD, Slichter SJ, Harker LA, Von Behrens WE, Clark RA, Wedgwood RJ. 1980. The Wiskott-Aldrich syndrome: studies of lymphocytes, granulocytes, and platelets. Blood 55:243–52 6. Ochs HD, Rosen FS. 1998. The WiskottAldrich Syndrome. In Primary Immunodeficiency Diseases: A Molecular and Genetic Approach, ed. HD Ochs, CIE Smith, JM Puck, Oxford, UK: Oxford Univ. Press. In press 7. Sullivan KE, Mullen CA, Blaese RM, Winkelstein JA. 1994. A multiinstitutional survey of the Wiskott-Aldrich syndrome. J. Pediatr. 125:876–85 8. Canales ML, Mauer AM. 1967. Sexlinked hereditary thrombocytopenia as a variant of Wiskott-Aldrich syndrome. N. Engl. J. Med. 277:899–901 9. Amaya CA, Dorantes S, Toro AH, Bello

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

922

10.

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

11.

12.

13.

14.

15.

16.

17.

18.

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN A, Cuellar J. 1973. Attenuated form of Wiskott-Aldrich syndrome. J. Pediatr. 82:175–6 Chiaro JJ, Dharmkrong-at A, Bloom GE. 1972. X-linked thrombocytopenic purpura. I. Clinical and genetic studies of a kindred. Am. J. Dis. Child. 123:565– 68 Notarangelo LD, Parolini O, Porta F, Locatelli F, Lanfranchi A, Marconi M, Nespoli L, Albertini A, Craig IW, Ugazio AG. 1991. Analysis of X-chromosome inactivation and presumptive expression of the Wiskott-Aldrich syndrome (WAS) gene in hematopoietic cell lineages of a thrombocytopenic carrier female of WAS. Hum. Genet. 88:237–41 Stormorken H, Hellum B, Egeland T, Abrahamsen TG, Hovig T. 1991. Xlinked thrombocytopenia and thrombocytopathia: attenuated Wiskott-Aldrich syndrome. Functional and morphological studies of platelets and lymphocytes. Thromb. Haemost. 65:300–5 Parkman R, Rappeport J, Geha R, Belli J, Cassady R, Levey R, Nathan DG, Rosen FS. 1978. Complete correction of the Wiskott-Aldrich syndrome by allogeneic bone-marrow transplantation. N. Eng. J. Med. 298:921–27 Peacocke M, Siminovitch KA. 1987. Linkage of the Wiskott-Aldrich syndrome with polymorphic DNA sequences from the human X chromosome. Proc. Natl. Acad. Sci. USA 84: 3430–33 Kwan SP, Sandkuyl LA, Blaese M, Kunkel LM, Bruns G, Parmley R, Skarshaug S, Page DC, Ott J, Rosen FS. 1988. Genetic mapping of the Wiskott-Aldrich syndrome with two highly-linked polymorphic DNA markers. Genomics 3:39– 43 Kwan SP, Lehner T, Hagemann T, Lu B, Blaese M, Ochs H, Wedgwood R, Ott J, Craig IW, Rosen FS. 1991. Localization of the gene for the Wiskott-Aldrich syndrome between two flanking markers, TIMP and DXS255, on Xp11.22Xp11.3. Genomics 10:29–33 Kwan SP, Hagemann TL, Radtke BE, Blaese RM, Rosen FS. 1995. Identification of mutations in the Wiskott-Aldrich syndrome gene and characterization of a polymorphic dinucleotide repeat at DXS6940, adjacent to the disease gene. Proc. Natl. Acad. Sci. USA 92:4706–10 Kwan SP, Hagemann TL, Blaese RM, Rosen FS. 1995. A high-resolution map of genes, microsatellite markers, and new dinucleotide repeats from UBE1 to

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

the GATA locus in the region Xp11.23. Genomics 29:247–52 Derry JM, Ochs HD, Francke U. 1994. Isolation of a novel gene mutated in Wiskott-Aldrich syndrome [published erratum appears in Cell 79(5): following 922, 1994]. Cell 78:635–44 Zhu Q, Zhang M, Blaese RM, Derry JM, Junker A, Francke U, Chen SH, Ochs HD. 1995. The Wiskott-Aldrich syndrome and X-linked congenital thrombocytopenia are caused by mutations of the same gene. Blood 86:3797–804 Villa A, Notarangelo L, Macchi P, Mantuano E, Cavagni G, Brugnoni D, Strina D, Patrosso MC, Ramenghi U, Sacco MG, et al. 1995. X-linked thrombocytopenia and Wiskott-Aldrich syndrome are allelic diseases with mutations in the WASP gene. Nature Genetics 9:414–17 Zhu Q, Watanabe C, Liu T, Hollenbaugh D, Blaese RM, Kanner SB, Aruffo A, Ochs HD. 1997. Wiskott-Aldrich syndrome/X-linked thrombocytopenia: WASP gene mutations, protein expression, and phenotype. Blood 90:2680–89 Schwarz K, Nonoyama S, Peitsch M, de Saint Basile G, Espanol T, Rasth A, Fischer A, Freitag K, Friedrich W, Fugmann S, Hossle H, Jones A, Kinnon C, Meindl A, Notarangelo L, Wechsler A, Weiss M, Ochs H. 1996. WASPbase: a database of WAS- and XLT-causing mutations. Immunol. Today 17:496–501 Gealy WJ, Dwyer JM, Harley JB. 1980. Allelic exclusion of glucose-6phosphate dehydrogenase in platelets and T lymphocytes from a WiskottAldrich syndrome carrier. Lancet 1:63– 65 Prchal JT, Carroll AJ, Prchal JF, Crist WM, Skalka HW, Gealy WJ, Harley J, Malluh A. 1980. Wiskott-Aldrich syndrome: cellular impairments and their implication for carrier detection. Blood 56:1048–54 Fearon ER, Kohn DB, Winkelstein JA, Vogelstein B, Blaese RM. 1988. Carrier detection in the Wiskott Aldrich syndrome. Blood 72:1735–39 Greer WL, Kwong PC, Peacocke M, Ip P, Rubin LA, Siminovitch KA. 1989. Xchromosome inactivation in the WiskottAldrich syndrome: a marker for detection of the carrier state and identification of cell lineages expressing the gene defect. Genomics 4:60–67 Puck JM, Siminovitch KA, Poncz M, Greenberg CR, Rottem M, Conley ME. 1990. Atypical presentation of WiskottAldrich syndrome: diagnosis in two

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

WISKOTT-ALDRICH SYNDROME PROTEIN

29.

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

unrelated males based on studies of maternal T cell X chromosome inactivation. Blood 75:2369–74 Wengler G, Gorlin JB, Williamson JM, Rosen FS, Bing DH. 1995. Nonrandom inactivation of the X chromosome in early lineage hematopoietic cells in carriers of Wiskott-Aldrich syndrome. Blood 85:2471–77 Remold-O’Donnell E, Rosen FS, Kenney DM. 1996. Defects in WiskottAldrich syndrome blood cells. Blood 87:2621–31 Molina IJ, Sancho J, Terhorst C, Rosen FS, Remold-O’Donnell E. 1993. T cells of patients with the Wiskott-Aldrich syndrome have a restricted defect in proliferative responses. J. Immunol. 151: 4383–90 Molina IJ, Kenney DM, Rosen FS, Remold-O’Donnell E. 1992. T cell lines characterize events in the pathogenesis of the Wiskott-Aldrich syndrome. J. Exp. Med. 176:867–74 Simon HU, Mills GB, Hashimoto S, Siminovitch KA. 1992. Evidence for defective transmembrane signaling in B cells from patients with Wiskott-Aldrich syndrome. J. Clin. Invest. 90:1396–405 Henriquez NV, Rijkers GT, Zegers BJ. 1994. Antigen receptor-mediated transmembrane signaling in Wiskott-Aldrich syndrome. J. Immunol. 153:395–99 Kenney D, Cairns L, Remold-O’Donnell E, Peterson J, Rosen FS, Parkman R. 1986. Morphological abnormalities in the lymphocytes of patients with the Wiskott-Aldrich syndrome. Blood 68: 1329–32 Gallego M, Santamaria M, Pena J, Molina I. 1997. Defective actin reorganization and polymerization of Wiskott-Aldrich T cells in response to CD3-mediated stimulation. Blood 90: 3089–97 Facchetti F, Blanzuoli L, Vermi W, Notarangelo LD, Giliani S, Fiorini M, Fasth A, Stewart DM, Nelson DL. 1998. Defective actin polymerization in EBVtransformed B-cell lines from patients with the Wiskott-Aldrich syndrome. J. Pathol. 185:99–107 Remold-O’Donnell E, Van BJ, Kenney DM. 1992. Effect of platelet calpain on normal T-lymphocyte CD43: hypothesis of events in the Wiskott-Aldrich syndrome. Blood 79:1754–62 Yonemura S, Nagafuchi A, Sato N, Tsukita S. 1993. Concentration of an integral membrane protein, CD43 (leukosialin, sialophorin), in the cleav-

40.

41.

42.

43.

44.

45.

46.

47. 48.

49. 50.

923

age furrow through the interaction of its cytoplasmic domain with actin-based cytoskeletons. J. Cell Biol. 120:437–49 Parkman R, Remold-O’Donnell E, Kenney DM, Perrine S, Rosen FS. 1981. Surface protein abnormalities in lymphocytes and platelets from patients with Wiskott-Aldrich syndrome. Lancet 2:1387–89 Remold-O’Donnell E, Kenney DM, Parkman R, Cairns L, Savage B, Rosen FS. 1984. Characterization of a human lymphocyte surface sialoglycoprotein that is defective in Wiskott-Aldrich syndrome. J. Exp. Med. 159:1705–23 Pallant A, Eskenazi A, Mattei MG, Fournier RE, Carlsson SR, Fukuda M, Frelinger JG. 1989. Characterization of cDNAs encoding human leukosialin and localization of the leukosialin gene to chromosome 16. Proc. Natl. Acad. Sci. USA 86:1328–32 Shelley CS, Remold-O’Donnell E, Davis AEd, Bruns GA, Rosen FS, Carroll MC, Whitehead AS. 1989. Molecular characterization of sialophorin (CD43), the lymphocyte surface sialoglycoprotein defective in Wiskott-Aldrich syndrome [published erratum appears in Proc. Natl. Acad. Sci. USA. 1989. 86(12):4689]. Proc. Natl. Acad. Sci. USA 86:2819–23 Gerwin N, Friedrich C, Perez-Atayde A, Rosen FS, Gutierrez-Ramos JC. 1996. Multiple antigens are altered on T and B lymphocytes from peripheral blood and spleen of patients with WiskottAldrich syndrome. Clin. Exp. Immunol. 106:208–17 Kenney DM. 1990. Wiskott-Aldrich syndrome and related X-linked thrombocytopenia. Curr. Opin. Pediatr. 2:931– 34 Lum LG, Tubergen DG, Corash L, Blaese RM. 1980. Splenectomy in the management of the thrombocytopenia of the Wiskott-Aldrich syndrome. N. Engl. J. Med. 302:892–96 Krivit W, Yunis E, White JG. 1966. Platelet survival studies in Aldrich syndrome. Pediatrics 37:339–41 Pearson HA, Shulman NR, Oski FA, Eitzman DV. 1966. Platelet survival in Wiskott-Aldrich syndrome. J. Pediatr. 68:754–60 Baldini MG. 1972. Nature of the platelet defect in the Wiskott-Aldrich syndrome. Ann. N. Y. Acad. Sci. 201:437–44 Grottum KA, Hovig T, Holmsen H, Abrahamsen AF, Jeremic M, Seip M. 1969. Wiskott-Aldrich syndrome: qual-

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

924

51.

52.

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

53.

54.

55.

56.

57.

58.

59.

60.

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN itative platelet defects and short platelet survival. Br. J. Haematol. 17:373–88 Kuramoto A, Steiner M, Baldini MG. 1970. Lack of platelet response to stimulation in the Wiskott-Aldrich syndrome. N. Engl. J. Med. 282:475–79 Marone G, Albini F, di ML, Quattrin S, Poto S, Condorelli M. 1986. The Wiskott-Aldrich syndrome: studies of platelets, basophils and polymorphonuclear leucocytes. Br. J. Haematol. 62:737–45 Verhoeven AJ, van Oostrum IE, van Haarlem H, Akkerman JW. 1989. Impaired energy metabolism in platelets from patients with Wiskott-Aldrich syndrome. Thromb. Haemost. 61:10–14 Kenney DM, Reid R, Parent DW, Rosen FS, Remold-O’Donnell E. 1994. Evidence implicating calpain (Ca(2+)dependent neutral protease) in the destructive thrombocytopenia of the Wiskott-Aldrich syndrome. Br. J. Haematol. 87:773–81 Altman LC, Snyderman R, Blaese RM. 1974. Abnormalities of chemotactic lymphokine synthesis and mononuclear leukocyte chemotaxis in WiskottAldrich syndrome. J. Clin. Invest. 54: 486–93 Badolato R, Sozzani S, Malacarne F, Bresciani S, Fiorini M, Borsatti A, Albertini A, Manotovani A, Ugazio AG, Notarangelo LD. 1998. Monocytes from Wiskott-Aldrich patients display reduced chemotaxis and lack of cell polarization in response to monocyte chemoattractant protein-1 and formylmethionyl-leucyl-phenylalanine. J. Immunol. 161:1026–33 Zicha D, Allen WE, Brickell PM, Kinnon C, Dunn GA, Jones GE, Thrasher AJ. 1998. Chemotaxis of macrophages is abolished in the Wiskott-Aldrich syndrome. Br. J. Haematol. 101:659–65 Derry JM, Wiedemann P, Blair P, Wang Y, Kerns JA, Lemahieu V, Godfrey VL, Wilkinson JE, Francke U. 1995. The mouse homolog of the WiskottAldrich syndrome protein (WASP) gene is highly conserved and maps near the scurfy (sf) mutation on the X chromosome. Genomics 29:471–77 Parolini O, Berardelli S, Riedl E, BelloFernandez C, Strobl H, Majdic O, Knapp W. 1997. Expression of Wiskott-Aldrich syndrome protein (WASP) gene during hematopoietic differentiation. Blood 90:70–75 Stewart DM, Treiber-Held S, Kurman CC, Facchetti F, Notarangelo LD, Nel-

61.

62.

63.

64. 65.

66.

67.

68. 69.

70.

71.

son DL. 1996. Studies of the expression of the Wiskott-Aldrich syndrome protein. J. Clin. Invest. 97:2627–34 She J, Rockow S, Tang J, Nishimura R, Skonik EY, Chen M, Magolis B, Li W. 1997. Wiskott-Aldrich syndrome protein is associated with the adapter protein Grb2 and the epidermal growth factor receptor in living cells. Mol. Biol. Cell. 8:1709–21 Rivero-Lezcano OM, Marcilla A, Sameshima JH, Robbins KC. 1995. Wiskott-Aldrich syndrome protein physically associates with Nck through Src homology 3 domains. Mol. Cell. Biol. 15:5725–31 Miki H, Miura K, Takenawa T. 1996. N-WASP, a novel actin-depolymerizing protein, regulates the cortical cytoskeletal rearrangement in a PIP2-dependent manner downstream of tyrosine kinases. EMBO J. 15:5326–35 Fukuoka M, Miki H, Takenawa T. 1997. Identification of N-WASP homologs in human and rat brain. Gene 196:43–48 Symons M, Derry JM, Karlak B, Jiang S, Lemahieu V, McCormick F, Francke U, Abo A. 1996. Wiskott-Aldrich syndrome protein, a novel effector for the GTPase CDC42Hs, is implicated in actin polymerization. Cell 84:723–34 Li R. 1997. Bee 1, a yeast protein with homology to Wiskott-Aldrich syndrome protein, is critical for the assembly of cortical actin assembly. J. Cell Biol. 136:649–58 Bear JE, Rawls JF, Saxe CL. 1998. SCAR, a WASP-related protein, isolated as a suppressor of receptor defects in late Dictyostelium development. J. Cell Biol. 142:1325–35 Lechler T, Li R. 1997. In vitro reconstitution of cortical actin assembly sites in budding yeast. J. Cell Biol. 138:95–103 Snapper SB, Rosen FS, Mizoguchi E, Cohen P, Khan W, Liu C-H, Hagemann TL, Kwan S-P, Ferrini R, Davidson L, Bhan AK, Alt FW. 1998. WiskottAldrich syndrome protein-deficient mice reveal a role for WASP in T but not B cell activation. Immunity 9:81–90 Petrella A, Doti V, Agosti P, Giarrusso C, Vitale D, Bond HM, Cuomao P, Tassone P, Franco B, Ballabio A, Venuta S, Morrone G. 1998. A 50 regulatory sequence containing two Ets motifs controls the expression of the Wiskott-Aldrich syndrome protein (WASP) gene in human hematopoietic cells. Blood 91:4554–60 Yu H, Chen J, Feng S, Dalgarno D, Brauer A, Schreiber S. 1994. Struc-

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

WISKOTT-ALDRICH SYNDROME PROTEIN

72. 73.

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

74.

75.

76. 77. 78.

79.

80.

81.

82.

83.

84.

tural basis for the binding of prolinerich polypeptides to SH3 domains. Cell 72:933–45 Pawson T. 1995. Protein modules and signalling networks. Nature 373:573–80 Burbelo PD, Drechsel D, Hall A. 1995. A conserved binding motif defines numerous candidate target proteins for both Cdc42 and Rac GTPases. J. Biol. Chem. 270:29071–4 Kolluri R, Tolias KF, Carpenter CL, Rosen FS, Kirchhausen T. 1996. Direct interaction of the Wiskott-Aldrich syndrome protein with the GTPase Cdc42. Proc. Natl. Acad. Sci. USA 93:5615–8 Aspenstrom P, Lindberg U, Hall A. 1996. Two GTPases, Cdc42 and Rac, bind directly to a protein implicated in the immunodeficiency disorder WiskottAldrich syndrome. Curr. Biol. 6:70–75 Hall A. 1998. Rho GTPases and the actin cytoskeleton. Science 279:509–14 Reif K, Cantrell DA. 1998. Networking Rho family GTPases in lymphocytes. Immunity 8:395–401 Ponting CP, Phillips C. 1997. Identification of homer as a homologue of the Wiskott-Aldrich syndrome protein suggests a receptor-binding function for WH1 domains. J. Mol. Med. 75:769–71 Donnelly SF, Pocklington MJ, Pallotta D, Orr E. 1993. A proline-rich protein, verprolin, involved in cytoskeletal organization and cellular growth in the yeast Saccharomyces cerevisiae. Mol. Microbiol. 10:585–96 Vaduva G, Martin NC, Hopper AK. 1997. Actin-binding verprolin is a polarity development protein required for the morphogenesis and function of the yeast actin cytoskeleton. J. Cell Biol. 139:1821–33 Ramesh N, Anton IM, Hartwig JH, Geha RS. 1997. WIP, a protein associated with Wiskott-Aldrich syndrome protein induces actin polymerization and redistribution in lymphoid cells. Proc. Natl. Acad. Sci. USA 94:14671–76 Miki H, Takenawa T. 1998. Direct binding of the verprolin-homology domain in N-WASP to actin is essential for cytoskeletal reorganization. Biochem. Biophys. Res. Commun. 243:73–78 Nishida E, Maekawa S, Sakai H. 1984. Cofilin, a protein in porcine brain that binds to actin filaments and inhibits their interactions with myosin and tropomyosin. Biochemistry 23:5307–17 Miki H, Sasaki T, Takai Y, Takenawa T. 1998. Induction of filopodium formation by a WASP-related actin-

85. 86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

925

depolymerizing protein N-WASP. Nature 391:93–96 Rhoads AR, Friedberg F. 1997. Sequence motifs for calmodulin recognition. FASEB J. 11:331–40 Stowers L, Yelon D, Berg LJ, Chant J. 1995. Regulation of the polarization of T cells toward antigen-presenting cells by Ras-related GTPase CDC42. Proc. Natl. Acad. Sci. USA 92:5027–31 Rudolph MG, Bayer P, Abo A, Kuhlmann J, Vetter IR, Wittinghofer A. 1998. The Cdc42/Rac interactive binding region motif of the Wiskott-Aldrich syndrome protein (WASP) is necessary but not sufficient for tight binding to Cdc42 and structure formation. J. Biol. Chem. 273:18067–76 Banin S, Truong O, Katz DR, Waterfield MD, Brickell PM, Gout I. 1996. WiskottAldrich syndrome protein (WASp) is a binding partner for c-Src family proteintyrosine kinases. Curr. Biol. 6:981–88 Bunnell S, Henry P, Kolluri R, Kirchhausen T, Rickles R, Berg L. 1996. Identification of ITK/Tsk Src homology 3 domain ligands. J. Biol. Chem. 271: 25646–56 Cory G, MacCarthy-Morrogh L, Banin S, Gout I, Brickell P, Levinsky R, Kinnon C, Lovering R. 1996. Evidence that the Wiskott-Aldrich syndrome protein may be involved in lymphoid cell signaling pathways. J. Immunol. 157:3791–95 Finan PM, Soames CJ, Wilson L, Nelson DL, Stewart DM, Truong O, Hsuan JJ, Kellie S. 1996. Identification of regions of the Wiskott-Aldrich syndrome protein responsible for association with selected Src homology 3 domains. J. Biol. Chem. 271:26291–95 Miki H, Nonoyama S, Zhu Q, Aruffo A, Ochs HD, Takenawa T. 1997. Tyrosine kinase signaling regulates WiskottAldrich syndrome protein function, which is essential for megakaryocyte differentiation. Cell Growth Different. 8:195–202 Lehmann JM, Riethmuller G, Johnson JP. 1990. Nck, a melanoma cDNA encoding a cytoplasmic protein consisting of the src homology units SH2 and SH3. Nucleic Acids Res. 18:1048 Galisteo ML, Chernoff J, Su YC, Skolnik EY, Schlessinger J. 1996. The adaptor protein Nck links receptor tyrosine kinases with the serine-threonine kinase Pak1. J. Biol. Chem. 271:20997–1000 Lu W, Katz S, Gupta R, Mayer BJ. 1997. Activation of Pak by membrane localization mediated by an SH3 domain

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

926

96.

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

97.

98.

99.

100.

101.

102.

103.

104.

105.

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN from the adaptor protein Nck. Curr. Biol. 7:85–94 Quilliam LA, Lambert QT, MickelsonYoung LA, Westwick JK, Sparks AB, Kay BK, Jenkins NA, Gilbert DJ, Copeland NG, Der CJ. 1996. Isolation of a NCK-associated kinase, PRK2, an SH3-binding protein and potential effector of Rho protein signaling. J. Biol. Chem. 271:28772–76 Sells MA, Knaus UG, Bagrodia S, Ambrose DM, Bokoch GM, Chernoff J. 1997. Human p21-activated kinase (Pak1) regulates actin organization in mammalian cells. Curr. Biol. 7:202–10 Su YC, Han J, Xu S, Cobb M, Skolnik EY. 1997. NIK is a new Ste20-related kinase that binds NCK and MEKK1 and activates the SAPK/JNK cascade via a conserved regulatory domain. Embo. J. 16:1279–90 Garrity PA, Rao Y, Salecker I, McGlade J, Pawson T, Zipursky SL. 1996. Drosophila photoreceptor axon guidance and targeting requires the dreadlocks SH2/SH3 adaptor protein. Cell 85:639– 50 Anton IM, Lu W, Mayer BJ, Ramesh N, Geha RS. 1998. The Wiskott-Aldrich Syndrome protein-interacting protein (WIP) binds to the adaptor protein Nck. J. Biol. Chem. 273:20992–95 da Silva AJ, Li Z, 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 leukocyte protein 76 and modulates interleukin 2 production. Proc. Natl. Acad. Sci. USA 94:7493–98 Marie-Cardine A, Bruyns E, Eckerskorn C, Kirchgessner H, Meuer SC, Schraven B. 1997. Molecular cloning of SKAP55, a novel protein that associates with the protein tyrosine kinase p59fyn in human T-lymphocytes. J. Biol. Chem. 272:16077–80 Pedraza-Alva G, Merida LB, Burakoff SJ, Rosenstein Y. 1996. CD43-specific activation of T cells induces association of CD43 to Fyn kinase. J. Biol. Chem. 271:27564–68 Prasad KV, Janssen O, Kapeller R, Raab M, Cantley LC, Rudd CE. 1993. Src-homology 3 domain of protein kinase p59fyn mediates binding to phosphatidylinositol 3-kinase in T cells. Proc. Natl. Acad. Sci. USA 90:7366–70 Raab M, Cai YC, Bunnell SC, Heyeck SD, Berg LJ, Rudd CE. 1995. p56Lck and p59Fyn regulate CD28 binding to phosphatidylinositol 3-kinase, growth

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

factor receptor-bound protein GRB-2, and T cell-specific protein-tyrosine kinase ITK: implications for T-cell costimulation. Proc. Natl. Acad. Sci. USA 92:8891–95 Fusaki N, Iwamatsu A, Iwashima M, Fujisawa J. 1997. Interaction between Sam68 and Src family tyrosine kinases, Fyn and Lck, in T cell receptor signaling. J. Biol. Chem. 272:6214–19 Fortner KA, Russell JQ, Budd RC. 1998. Down-modulation of CD2 delays deletion of superantigen-responsive T cells. Eur. J. Immunol. 28:70–79 Atkinson EA, Ostergaard H, Kane K, Pinkoski MJ, Caputo A, Olszowy MW, Bleackley RC. 1996. A physical interaction between the cell death protein Fas and the tyrosine kinase p59fynT. J. Biol. Chem. 271:5968–71 Helmke S, Pfenninger KH. 1995. Growth cone enrichment and cytoskeletal association of non-receptor tyrosine kinases. Cell Motil. Cytoskeleton 30:194–207 Thomas SM, Soriano P, Imamoto A. 1995. Specific and redundant roles of Src and Fyn in organizing the cytoskeleton. Nature 376:267–71 Calautti E, Missero C, Stein PL, Ezzell RM, Dotto GP. 1995. fyn tyrosine kinase is involved in keratinocyte differentiation control. Genes Dev. 9:2279–91 Miki H, Mirua K, Matuoka K, Nakata T, Horokawa N, Orita S, Kaibuchi K, Takai Y, Takenawa T. 1994. Association of Ash/Grb2 with dynamin through the Src homology 3 domain. J. Biol. Chem. 269:5489–92 Matuoka K, Shibasaki F, Shibata M, Takenawa T. 1993. Ash/Grb-2, a SH2/SH3-containing protein, couples to signaling for mitogenesis and cytoskeletal reorganization by EGF and PDGF. EMBO J. 12:3467–73 Wu Y, Spencer SD, Lasky LA. 1998. Tyrosine phosphorylation regulates the SH3-mediated binding of the WiskottAldrich syndrome protein to PSTPIP, a cytoskeletal-associated protein. J. Biol. Chem. 273:5765–70 Spencer S, Dowbenko D, Cheng J, Li W, Brush J, Utzig S, Simanis V, Lasky LA. 1997. PSTPIP: a tyrosine phosphorylated cleavage furrow-associated protein that is a substrate for a PEST tyrosine phosphatase. J. Cell Biol. 138:845– 60 Zoladek T, Vaduva G, Hunter LA, Boguta M, Go BD, Martin NC, Hopper AK. 1995. Mutations altering the

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

WISKOTT-ALDRICH SYNDROME PROTEIN

117.

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

118.

119.

120.

121.

122.

123.

124.

125.

126.

mitochondrial-cytoplasmic distribution of Mod5p implicate the actin cytoskeleton and mRNA 30 ends and/or protein synthesis in mitochondrial delivery. Mol. Cell. Biol. 12:6884–94 Stossel TP. 1993. On the crawling of animal cells. Science 260:1086–94 Hartwig J, Bokoch G, Carpenter C, Janmey P, Taylor L, Toker A, Stossel T. 1995. Thrombin receptor ligation and activated Rac uncap actin filament barbed ends through phosphoinositide synthesis in permeabilized human platelets. Cell 82:643–53 Gilmore A, Burridge K. 1996. Regulation of vinculin binding to talin and actin by phosphatidyl-inositol-4-5bisphosphate. Nature 381:531–35 Derry JM, Kerns JA, Weinberg KI, Ochs HD, Volpini V, Estivill X, Walker AP, Francke U. 1995. WASP gene mutations in Wiskott-Aldrich syndrome and X-linked thrombocytopenia. Hum. Mol. Gen. 4:1127–35 Kolluri R, Shehabeldin A, Peacocke M, Lamhonwah AM, Teichert-Kuliszewska K, Weissman SM, Siminovitch KA. 1995. Identification of WASP mutations in patients with Wiskott-Aldrich syndrome and isolated thrombocytopenia reveals allelic heterogeneity at the WAS locus. Hum. Mol. Gen. 4:1119–26 Kwan SP, Hagemann TL, Blaese RM, Knutsen A, Rosen FS. 1995. Scanning of the Wiskott-Aldrich syndrome (WAS) gene: identification of 18 novel alterations including a possible mutation hotspot at Arg86 resulting in thrombocytopenia, a mild WAS phenotype. Hum. Mol. Gen. 4:1995–98 Wengler GS, Notarangelo LD, Giliani S, Pirastru MG, Ugazio AG, Parolini O. 1995. Mutation analysis in Wiskott Aldrich syndrome on chorionic villus DNA. Lancet 346:641–42 Wengler GS, Notarangelo LD, Berardelli S, Pollonni G, Mella P, Fasth A, Ugazio AG, Parolini O. 1995. High prevalence of nonsense, frame shift, and splice-site mutations in 16 patients with full-blown Wiskott-Aldrich syndrome. Blood 86:3648–54 Schindelhauer D, Weiss M, Hellebrand H, Golla A, Hergersberg M, Seger R, Belohradsky BH, Meindl A. 1996. WiskottAldrich syndrome: no strict genotypephenotype correlations but clustering of missense mutations in the aminoterminal part of the WASP gene product. Hum. Genet. 98:68–76 Greer WL, Shehabeldin A, Schulman

127.

128.

129.

130.

131.

132.

133.

134.

135.

136.

927

J, Junker A, Siminovitch K. 1996. Identification of WASP mutations, mutation hotspots, and genotype-phenotype disparities in 24 patients with the Wiskott-Aldrich syndrome. Hum. Genet. 98:685–90 de Saint-Basile G, Lagelouse RD, Lambert N, Schwarz K, Le Mareck B, Odent S, Schlegel N, Fischer A. 1996. Isolated X-linked thrombocytopenia in two unrelated families is associated with point mutations in the Wiskott-Aldrich syndrome protein gene. J. Pediatr. 129:56– 62 Remold-O’Donnell E, Cooley J, Shcherbina A, Hagemann TL, Kwan SP, Kenney DM, Rosen FS. 1997. Variable expression of WASP in B cell lines of Wiskott-Aldrich syndrome patients. J. Immunol. 158:4021–25 MacCarthy-Morrogh L, Gaspar HB, Wang YC, Katz F, Thompson L, Layton M, Jones AM, Kinnon C. 1998. Absence of expression of the Wiskott-Aldrich syndrome protein in peripheral blood cells of Wiskott-Aldrich syndrome patients. Clin. Immunol. Immunopathol. 88:22–27 Kondoh T, Hayashi K, Matsumoto T, Yoshimoto M, Morio T, Yata J, Tsuji Y. 1995. Two sisters with clinical diagnosis of Wiskott-Aldrich syndrome: Is the condition in the family autosomal recessive? Am. J. Med. Genet. 60:364–69 Rocca B, Bellacosa A, De CR, Neri G, Della VM, Maggiano N, Rumi C, Landolfi R. 1996. Wiskott-Aldrich syndrome: report of an autosomal dominant variant. Blood 87:4538–43 Parolini O, Ressmann G, Haas OA, Pawlowsky J, Gadner H, Knapp W, Holter W. 1998. X-linked Wiskott-Aldrich syndrome in a girl. N. Engl. J. Med. 338:291–5 Hsieh KH, Chang MH, Lee CY, Wang CY. 1988. Wiskott-Aldrich syndrome and inflammatory bowel disease. Ann. Aller. 60:429–31 Morales V, Snapper SB, Blumberg R. 1996. Probing the gastrointestinal immune function using transgenic and knockout technology. Current Opin. Gastroenterol. 12:577–83 Nobes CD, Hall A. 1995. Rho, rac, and cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell 81:53–62 Lamarche N, Tapon N, Stowers L, Burbelo P, Aspenstrom P, Bridges T, Chant J, Hall A. 1996. Rac and Cdc42

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

928

137.

138.

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

139.

140.

141.

142.

143.

144.

144a.

145.

146.

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

SNAPPER & ROSEN induce actin polymerization and G1 cell cycle progression independently of p65PAK and the JNK/SAPK MAP kinase cascade. Cell 87:519–29 Joneson T, McDonough M, Bar-Sagi D, Van Aelst L. 1997. RAC regulation of actin polymerization and proliferation by a pathway distinct from Jun kinase. Science 274:1347–56 Zigmond SH. 1998. The Arp2/3 complex gets to the point. Curr. Biol. 8: R654–R7 Pantaloni D, Carlier M-F. 1993. How profilin promotes actin filament assembly in the presence of thymosinB4. Cell 75:1007–14 Geiger B, Rosen D, Berke G. 1982. Spatial relationships of microtubuleorganizing centers and the contact area of cytotoxic T lymphoyctes and target cells. J. Cell Biol. 95:137–43 Kupfer A, Swain SL, Janeway CA, Singer SJ. 1986. The specific direct interaction of helper T cells and antigenpresenting B cells. Proc. Natl. Acad. Sci. (USA) 83:8080–3 Kupfer H, Monks CRF, Kupfer A. 1994. Small splenic B cells that bind to antigen-specific T helper (Th) cells and face the site of cytokine production in the the Th cells selectively proliferate: immunofluorescence microscopic studies of Th-B antigen presenting cell interactions. J. Exp. Med. 179:1507– 15 Rozdzial MM, Maliseen B, Finkel TH. 1995. Tyrosine-phosphorylated T cell receptor ζ chain associates with the actin cytoskeleton upon activation of mature T lymphocytes. Immunity 3:623–33 Caplan S, Zeliger S, Wnag L, Baniyash M. 1995. Cell-surface-expressed T-cell antigen-receptor ζ chain is associated with the cytoskeleton. Proc. Natl. Acad. Sci. USA 92:4768–72 Geppert T, Lipsky P. 1991. Association of various T cell-surface molecules with the cytoskeleton. Effect of cross-linking and activation. J. Immunol. 146:3298– 305 Phatak PD, Packman CH. 1994. Engagement of the T-cell antigen receptor by anti-CD3 monoclonal antibody causes a rapid increase in lymphocyte F-actin. J. Cell. Physiol. 159:365–70 Lowin-Kropf B, Shapiro VS, Weiss A. 1998. Cytoskeletal polarization of T cells is regulated by an immunoreceptor tyrosine-based activation motifdependent mechanism. J. Cell. Biol. 140:861–71

147. Parsey M, Lewis G. 1993. Actin polymerization and pseudopod reorganization accompany anti-CD3 growth arrest in Jurkat T cells. J. Immunol. 151:1881– 93 148. Pardi R, Inverardi L, Rugarli C, Bender JR. 1992. Antigen-receptor complex stimulation triggers protein kinase Cdependent CD11a/CD18-cytoskeleton association in T lymphocytes. J. Cell. Biol. 116:1211–20 149. Selliah N, Bartik MM, Carlson SL, Brooks WH, Roszman TL. 1995. cAMP accumulation in T-cells inhibits anti-CD3 monoclonal antibody-induced actin polymerization. J. Neuroimmunol. 56:107 150. Selliah N, Brooks WH, Roszman TL. 1996. Proteolytic cleavage of α-actinin by calpain in T cells stimulated with antiCD3 monoclonal antibody. J. Immunol. 156:3215–21 151. Melamed I, Downey G, Altories K, Roifman C. 1991. Microfilament assembly is required for antigen-receptor-mediated activation of human B lymphocytes. J. Immunol. 147:1139–46 151a. Geppert TD, Lipsky PE. 1990. Regulatory role of microfilaments in the induction of T4 cell proliferation and interkeukin 2 production. Cell. Immunol. 131:205–18 151b. DeBell KE, Conti A, Alava MA, Hoffman T, Bonvini E. 1992. Microfilament assembly modulates phospholipase C-mediated signal transduction by the TCR/CD3 in murine T helper lymphocytes. J. Immunol. 7:2271–80 152. Cox D, Chang P, Kuosaki T, Greenberg S. 1996. Syk tyrosine kinase is required for immunoreceptor tyrosine activation motif-dependent actin assembly. J. Cell. Biol. 140:861–70 153. Phatak P, Packman C, Lichtman M. 1988. Protein kinase C modulates actin conformation in human T lymphocytes. J. Immunol. 2929–34 154. Monks CRF, Kupfer H, Tamir I, Barlow A, Kupfer A. 1997. Selective modulation of protein kinase C-θ during T-cell activation. Nature 385:83–6 155. Holsinger LJ, Graef IA, Swat W, Chi T, Bautista DM, Davidson L, Lewis RS, Alt FW, Crabtree GR. 1998. Defects in actin cap formation in vav-deficient mice implicate an actin requirement for lymphocyte signal transduction. Curr. Biol. 8:563–72 156. Fischer K-D, Kong Y-Y, Nishina H, Tedfor D, Manengere LEM, Kozieradzki I, Sasaki T, Starr M, Chan G, Gardener S,

P1: SAT/sny/bta

P2: KKK/plb

January 20, 1999

16:9

QC: KKK/uks

T1: KKK

Annual Reviews

AR078-27

WISKOTT-ALDRICH SYNDROME PROTEIN

157.

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

158.

159.

160.

161.

162.

Hghiem MP, Bouchard D, Barbacid M, Bernstein A, Penninger JM. 1998. Vav is a regulator of cytoskeletal reorganization mediated by the T-cell receptor. Curr. Biol. 8:554–62 Valitutti S, Muller S, Cella M, Padovan E, Lanzavecchia A. 1995. Serial triggering of many T-cell receptors by a few peptide-MHC complexes. Nature 375:148–51 Valitutti S, Dessing M, Aktories H, Gallait H, Lanzavecchia A. 1995. Sustained signaling leading to T cell activation results from prolonged T cell receptor occupancy. Role of T cell actin cytoskeleton. J. Exp. Med. 181:577–84 Shaw A, Dustin M. 1997. Making the T cell receptor go the distance: a topological view of T cell activation. Immunity 6:361–9 Monks CRF, Freiberg BA, Kupfer H, Sciaky N, Kupfer A. 1998. Threedimensional segregation of supramolecular activation clusters in T-cells. Nature 395:82–6 Dustin ML, Olszowy MW, Holdorf AD, Li J, Bromley S, Desai N, Widder P, Rosenberger F, Anton van der Merwe P, Allen P, Shaw AS. 1998. A novel adaptor protein orchestrates receptor patterning and cytoskeletal polarity in T-cell contacts. Cell 94:667–77 Taylor RB, Duttus PH, Raff MC, de Petris S. 1971. Redistribution and pinocytosis of lymphocyte surface immunoglobulin molecules induced by anti-immunoglobulin antibody. Nat. New Biol. 233:225

929

163. de Petris S. 1974. Inhibition and reversal of capping by cytochalasin B, vinblastine, and colchicine. Nature 250:54–5 164. Kammer GM, Smith JA, Mitchell R. 1983. Capping of human T cell specific determinants: kinetics of capping and receptor re-expression and regulation by the cytoskeleton. J. Immunol. 130:38– 44 165. Zhang R, Alt FW, Davidson L, Orkin SH, Swat W. 1995. Defective signalling through the T- and B-cell antigen receptors in lymphoid cells lacking the vav proto-oncogene. Nature 374:470–3 166. Fischer KD, Zmuldzinas A, Gardner S, Barbacid M, Bernstein A, Guidos C. 1995. Defective T-cell receptor signalling and positive selection of Vavdeficient CD4+ CD8+ thymocytes. Nature 374:474–7 167. Tarakhovsky A, Turner M, Schaal S, Mee PJ, Duddy LP, Rajewsky K, Tybulewicz VL. 1995. Defective antigen receptor-mediated proliferation of B and T cells in the absence of Vav. Nature 374:467–70 168. Olson MF, Pasteris NG, Gorski JL, Hall A. 1996. Faciogenital dysplasia protein (FGD1) and Vav, two related proteins required for normal embryonic development, are upstream regulators of Rho GTPases. Curr. Biol. 6:1628–33 169. Oda A, Ochs HD, Druker BJ, Ozaki K, Watanabe C, Handa M, Miyakawa Y, Ikeda Y. 1998. Collagen induces tyrosine phosphorylation of Wiskott-Aldrich Syndrome Protein in human platelets. Blood 92:1852–8

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:905-929. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:931–72 c 1999 by Annual Reviews. All rights reserved Copyright °

THE HIGH-AFFINITY IGE RECEPTOR (FcεRI): From Physiology to Pathology Jean-Pierre Kinet Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, Massachusetts 02215; e-mail: [email protected] KEY WORDS:

receptor, signaling, atopy

ABSTRACT The high affinity receptor for immunoglobulin E (designated FcεRI) is the member of the antigen (Ag) receptor superfamily responsible for linking pathogenor allergen-specific IgEs with cellular immunologic effector functions. This review provides background information on FcεRI function combined with more detailed summaries of recent progress in understanding specific aspects of FcεRI biology and biochemistry. Topics covered include the coordination and function of the large multiprotein signaling complexes that are assembled when FcεRI and other Ag receptors are engaged, new information on human receptor structures and tissue distribution, and the role of the FcRβ chain in signaling and its potential contribution to atopic phenotypes.

INTRODUCTION The formation of antigen-antibody complexes plays a fundamental role in our immune defense system. Interaction of these complexes with many cells of the immune system results in a variety of critical functions, including phagocytosis, antibody-dependent cytotoxicity, modulation of antibody secretion and cell secretion. These interactions are mediated through the binding of immunoglobulin Fc domains to their corresponding Fc receptors (FcRs). One of these FcRs is FcεRI, a multimeric cell surface receptor that binds immunoglobulin E (IgE) with high affinity. The physiological function of FcεRI has remained elusive, but its role in allergic reactions is well established. The last review entirely 931 0732-0582/99/0410-0931$08.00

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

932

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

devoted to FcεRI in Annual Reviews was written in 1986 (1), at a time preceding the molecular cloning of the FcεRI α, β, and γ subunits. In 1991, my colleague Jeffrey Ravetch and I wrote a review on FcRs, based on the large amount of new information that had been generated from the molecular cloning of the immunoglobulin-binding FcRs (2). This work led to the recognition that FcRs are homologous and belong to the Ig superfamily. Another fundamental piece of information was the unexpected finding that the FcεRI γ chains (now denominated FcRγ chains) that had been found initially as a subunit of the FcεRI receptor were also subunits of at least one other Fc receptor, the low affinity IgG receptor (Fcγ RIII) (3–5) and of the T cell antigen receptor (TCR) (6). It was becoming clear that in addition to FcεRI, other Fc receptors could also be multimeric complexes and that they could share identical subunits. The 1991 review was written more to emphasize the family and modular nature of the FcRs than to elaborate on the fragmented knowledge of their individual characteristics and biological relevance. In that review, FcεRI was described as a tetrameric complex (αβγ2 ) expressed exclusively on the surface of mast cells and basophils whose function is to mediate cellular degranulation and the release of various mediators such as histamine, leukotrienes, and a number of cytokines and chemokines, the hallmark of allergic reactions. Over the last seven years, a number of important findings have dramatically improved our understanding of FcεRI biological functions. One unexpected finding has been the recognition of substantial differences between human and rodent FcεRI. While rodent FcεRI has an obligatory αβγ2 tetrameric structure, human FcεRI can be expressed as both trimeric (αγ2 ) and tetrameric (αβγ2 ) structures. The cellular distribution of human and rodent FcεRI is also different. Rodent FcεRI is only expressed on mast cells and basophils (and non-B, non-T cells), whereas expression of human FcεRI extends to monocytes, eosinophils, platelets, Langerhans cells, and dendritic cells. Other important findings include the following: the participation of FcεRI in the process of antigen presentation; the genetic demonstration that FcεRI can alter the disease evolution of some parasitic diseases; the molecular and genetic demonstration that the FcεRI β chain serves as an amplifier of FcεRI biological function; the possible association between various human FcεRI β chain polymorphisms and atopic phenotypes; the control of receptor expression by its own ligand IgE; and the possibility that the FcεRI α chain may be the target of autoantibodies in chronic urticaria. In addition, remarkable progress has been made in understanding the intracellular signaling mechanisms by which immunoreceptors, including FcεRI, are activating cells. The present review focuses on all of the advances listed above and discusses their biological implications.

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

933

DIFFERENCES BETWEEN RODENT AND HUMAN FcεRI SYSTEMS

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Rodent FcεRI: a Tetrameric Structure In rats and mice, from which it was first isolated, FcεRI is exclusively expressed in mast cells, basophils, and non-B non-T cells, and is a tetramer made of one α chain, one β chain, and two identical, disulfide-linked γ chains (7, 8). The α, β, and dimeric γ chains are maintained as a complex in the plasma membrane through a combination of hydrophobic and electrostatic noncovalent interactions involving both covalently and noncovalently bound lipids (9, 10). The α chain, a member of the immunoglobulin superfamily, comprises two extracellular immunoglobulin-related domains, one transmembrane domain containing an aspartic residue, and a short cytoplasmic tail (8, 11–13). The α chain contains seven N-linked glycosylation sites, all of which are utilized. These N-linked oligosaccharides are not intrinsically necessary for proper folding of the α chain but are required in the endoplasmic reticulum (ER) to mediate the proper interaction between the α chain and the ER folding machinery (14). Thus, the α chain is heavily glycosylated and appears as a broad, very heterogeneous band centered around the 45-kDa marker when resolved by standard SDS-PAGE; however, the weight of the α chain protein core is 27 kDa. The β subunit (FcRβ) has four transmembrane domains separating amino and carboxy terminal cytoplasmic tails (8, 15). In mast cells, FcRβ is a subunit not only of FcεRI but also of the low affinity IgG receptor (Fcγ RIII) (16). The γ subunit (FcRγ ) is a member of the γ /ζ /η family of antigen receptor subunits, which are notable for their structure, consisting essentially of a transmembrane region and cytoplasmic tail, and for their utilization among many families of antigen receptors. The γ , ζ and η chains can be found in homo- or heterodimeric forms with Fcγ RIII, Fcγ RI, FcεRI, FcαR, and the TCR depending on the cellular context (3–6, 17–24). FcRγ homodimer has also been shown recently to associate with glycoprotein VI on human platelets to form a collagen receptor (25–27).

Assembly of FcεRI When the cDNA encoding the α chain was first isolated, transfection of this cDNA failed to induce expression of a functional IgE-binding α chain on the surface of the transfected cells. An explanation for this failure awaited the subsequent cloning of the β and γ chains. The cloning of β and γ cDNAs permitted us to co-transfect the three cDNAs, resulting in the cell surface expression of a large number of IgE-binding FcεRI complexes (7, 8). Indeed, the three subunits α, β, and γ were required for efficient cell surface expression of FcεRI. A further analysis of this co-transfection requirement determined

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

934

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

that the α chain contains endoplasmic reticulum (ER) retention signals, among which is a di-lysine motif located in its cytoplasmic domain (28). In the absence of the other chains and because of the presence of these ER retention signals, the transfected α chain remains and is degraded in the ER. However, assembly of the α chain with the co-transfected β and γ chains masks the ER retention signals, allowing the αβγ2 complex to be exported to the Golgi apparatus and from there, to the plasma membrane. However, when the same experiment was attempted with human α, β, and γ chains, it became clear that requirements for surface expression of the FcεRI complex were different in the human system. Indeed, αγ2 complexes could be as easily exported to the cell surface as αβγ2 complexes, demonstrating that human γ was sufficient by itself to counterbalance the human α chain ER retention signals (29, 30). Although the precise molecular explanation for this species difference is still unclear, there is a suggestion that an additional ER retention signal may be located in the rodent α extracellular domain (see below FcεRI: a modular unit). These transfection data turned out to be more important than initially recognized. They demonstrated a species difference between rodent and human FcεRI and showed that human β, unlike rodent β, is not necessary for cell surface expression of FcεRI. Moreover they were also the first suggestion that two FcεRI isoforms (αβγ2 and αγ2 ) may in fact exist in humans.

Human FcεRI: Different Cell Distribution As mentioned above, the high affinity IgE receptor FcεRI was initially thought to be expressed exclusively on mast cells and basophils. However, G. Stingl’s group in collaboration with us reported that FcεRI is also expressed on cutaneous Langerhans cells (31). This finding, reported simultaneously by another group (32), led ourselves and our collaborators to question the selectivity of expression of FcεRI on human cells. In fact, FcεRI was detected on eosinophils of some hypereosinophilic patients (33), on monocytes of patients with a number of atopic disorders (34) and, later, on circulating dendritic cells (35) and platelets (36). While some of these findings were initially controversial, in particular FcεRI expression on subsets of eosinophils, most of them have now been reproduced by other groups (37–41). In addition, unambiguous genetic evidence now confirms the extended pattern of human FcεRI expression (see next section). It is useful to comment here on why it took so long to recognize the expression of FcεRI on cells as common as human monocytes while so many clusters of differentiation (CDs) have been detected and characterized. Two points need to be made. First, there was no available reagent until the early 1990s, when antihuman FcεRIα monoclonal antibodies (mAb) became available (34, 42). Partly for this reason, the binding of IgE, which had been detected on many of these

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIGH-AFFINITY IGE RECEPTOR

935

cells, had been interpreted as binding to CD23 (the low affinity IgE receptor, FcεRII). Second, the level of expression of FcεRI on monocytes, eosinophils, Langerhans cells, and dendritic cells is low when compared to the level of expression seen on mast cells and basophils (as judged by flow cytometry analysis) and is therefore more difficult to detect. For example, normal human monocytes express 10 to 100 times less receptors than circulating basophils. Even the upregulated levels of FcεRI on monocytes of patients with various allergic diseases (34) are still well below the levels expressed by mast cells and basophils. At present, the basis for the difference of expression levels between the cell types and the mechanisms of the regulation of these levels is not known and it is quite possible that additional cell types will be shown in the future to express FcεRI under specific circumstances.

Genetic Control of the Cell Specific Expression of Human FcεRI The difference of FcεRI cell distribution between humans and mice has now been confirmed genetically. First, a 2.9-kb fragment 50 of the initiation codon from the human FcεRIα gene is active in a reporter gene assay in mouse monocytic cell line, and as expected in mast cells and basophils but not in fibroblasts or B and T cells (A Brini, J-P Kinet, unpublished). Moreover, we generated a transgenic mouse using a 9.5-kb FcεRIα gene, whose 5-exon structure has been described (43–45), under the control of the 2.9-kb promoter and showed that the transgene induces expression of human FcεRIα not only on mast cells and basophils (46) but also on monocytes/macrophages, Langerhans cells, and eosinophils but not on B and T cells nor on neutrophils (47). Therefore, the 2.9-kb fragment contains the information necessary to achieve cell-specific expression of human FcεRI. Another group reported the generation of a trangenic mouse line, using as promoter sequence, a smaller 1.3-kb fragment upstream of the start codon (48). These authors claimed that this promoter is specific for mast cells. However, possible expression of FcεRI on monocytes, eosinophils, and Langerhans cells was not analyzed and may therefore have been missed. Alternatively, the regulatory elements governing FcεRI cellular specificity may lie in the 2.9 kb fragment in a region upstream of the 1.3 kb fragment. Analysis of the FcεRI receptor structure in the “humanized” transgenics shows that the 9.5-kb transgene recapitulates not only the appropriate cell specificity of expression of human FcεRI but also the appropriate cell type/structure observed in human cells (47), as discussed in the next section.

Human FcεRI: Two Isoforms The observation that human FcεRI cell is expressed on a variety of cells other than mast cells and basophils led to another surprising finding. In contrast to

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

936

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

the easily detectable α and γ transcripts, those for the β chain were not detected in monocytes even using the sensitive polymerase chain reaction (PCR) (34 and D Dombrowicz, J-P Kinet, unpublished observations). Our inability to detect FcRβ transcripts in monocytes, Langerhans cells, and dendritic cells suggested that the structure of FcεRI on these cells is indeed different from that on mast cells and basophils. Thus, in these cells, either FcεRI lacks the FcRβ subunit altogether and is an αγ2 complex, as on the transfectants described above, or a β-like molecule that has not yet been identified is taking the place of β in the complex. In fact, data on the detergent sensitivity of FcεRI complexes support the existence of αγ2 complexes. The αγ2 complexes are less stable in detergent than αβγ2 complexes, and as a consequence, the FcRγ2 homodimer dissociates more easily from the FcεRIα chain (49). The pattern of detergent instability usually seen with αγ2 structure was observed in our laboratory for monocyte-expressed FcεRI, suggesting that FcεRI on monocytes are αγ2 complexes (D Dombrowicz, J Kinet, unpublished). Although it is still formally possible that αγ2 complexes associate with β-like protein, which would not confer the same stability as the FcRβ chain, the data, when taken together, are best explained by the existence of αγ2 complexes. The capacity of αγ2 complexes to be expressed on transfectants without β chains, the absence of detection of FcRβ chain transcripts on human FcεRI-expressing cells, and the detergent sensibility studies provide the experimental basis in support of the existence of αγ2 isoforms on human monocytes, Langerhans cells, and dendritic cells. By contrast, mast cells and basophils that express FcRβ chain have on their surface the classical αβγ2 isoform (Figure 1). It is presumed that

Figure 1 Human FcεR 1 isoforms.

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIGH-AFFINITY IGE RECEPTOR

937

eosinophils and platelets also express αβγ2 complexes, although the existence of FcRβ in eosinophils and platelets is based so far on PCR data only. In addition, it has not yet been determined whether human mast cells, basophils, eosinophils, and platelets express only the αβγ2 isoform or a variable proportion of the two isoforms. In view of the critical importance of the FcRβ chain in signal amplification (see below), this question may turn out to be critical in our overall understanding of the human allergic response. In this regard, it is also worth noting that the experimental evidence gathered so far does not eliminate the possibility that the FcRβ chain could be expressed on the apparently FcR β chain–negative cells in different circumstances or, perhaps, in pathological conditions. More knowledge on the virtually unknown mechanisms regulating FcRβ gene and protein expression will be needed to address these questions.

FcεRI: A MODULAR UNIT The extracellular domain of the α chain contains sufficient molecular information to provide the receptor with high affinity binding, whereas the β and γ chains play no apparent role in ligand binding. This point was demonstrated by making a single chain chimera containing the extracellular region of FcεRIα (50) or, alternatively, by truncation of the extracellular and transmembrane domains of FcεRIα that results in secretion of a soluble protein (51). The chimera and the soluble FcεRIα chain both have apparently the same ligand binding characteristics as the tetrameric, plasma-membrane inserted, FcεRI complex. In these truncation/transfection experiments, another interesting difference between rodent and humans was detected (51). Soluble α chain is seen only with the human but not with the mouse or rat truncated construct. The rodent constructs result in the ER retention of the corresponding protein, an indication that, as suggested above, the extracellular domain of the rodent FcεRIα chain contains an additional ER retention signal. The FcεRIα chain seems to be entirely devoted to the ligand binding function and, until now, has not been found to contribute any intracellular signals. Truncation of its cytoplasmic tail in a reconstituted tetrameric complex does not seem to affect holo-receptor signaling (52). Thus, in both rodent and human, FcεRI can be considered as a modular structure with one binding module, the α chain and one signaling module, the βγ2 or γ2 complex. The function of the signaling chains is addressed in details below (see signal transduction).

IgE-FcεRI BINDING CHARACTERISTICS FcεRI is specific for the ε isotype but is not completely species specific. Rodent IgE binds to both rodent and human FcεRI whereas human IgE binds only to

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

938

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

human FcεRI (reviewed in 53). A large body of literature has been generated about the FcεRI-binding site on IgE (54–60). The consensus reached from these earlier studies was that the FcεRI α chain binding site on IgE was located within the Cε3 domain (53). Recent attempts have been made to define the residues and domains within Cε3 that are important for binding. Mutations, which negatively affected the binding site, were located in the three putative loops (CD, EF, and FG) of the Cε3 domain (61). A second study, which clearly disagrees with the previous one, implicates the segment 343-353 (located in the AB loop) as being important for IgE binding (62). The following studies, recently published, took advantage of efficient biophysical methods, such as surface plasmon resonance (SPR) and isothermal titration calorimetry, and used circular dichroism (CD) to verify that the overall structure of the IgE constructs was not affected by the various mutations made to localize the critical residues. One study found that Arginine 334, located at the N terminus of Cε3, is an important contact residue. The absence of involvement of Cε2 was confirmed by thermodynamic measurements that were identical for the whole IgE and for a construct including only Cε3 and Cε4 (63). CD studies showed the conservation of the secondary structures of Cε3, Cε4, and the two Ig-like domains of FcεRIα in the complex, indicating that complex formation affects only the relative position of these domains (63). This contradicts a previous report of a major conformational rearrangement of IgE upon binding (64). The reason for this contradiction is at present unclear. Incidentally, several of the studies cited here have confirmed the previously shown 1/1 stoichiometry of the IgE/FcεRI interaction in spite of the fact that there are two potential FcεRIα-binding sites per molecule of IgE. The IgE binding site on FcεRI α is entirely contained in the extracellular portion (EC) of FcεRIα (51). This EC domain contains two immunoglobulinlike domains (D1 and D2), which belong to the truncated C2 subtype of the immunoglobulin superfamily (65, 66). Earlier studies concluded that the carboxyterminal domain (D2) contained the IgE-binding site, but that the presence of the amino-terminal domain D1 was required for high affinity (67–71). Cyclic peptides homologous to the putative C-C0 region of D2 were designed on the basis of the known structure of the second EC of CD2, the prototype for the C2 subtype. These cyclic peptides act as competitive inhibitors of IgE binding, contributing about half of the free energy of binding of the entire EC (72). Lys117 in the C strand was identified as a contact residue (73). These results combined with those regarding the binding site on IgE and the modeling of IgE and FcεRIα allowed these authors to propose a model of the interaction (73). The crystal structure of FcεRIα has now been solved and is described in the addendum following this chapter.

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

939

BIOLOGICAL FUNCTION OF FcεRI

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Control of Parasite-Induced Pathology and Resistance to Parasites Despite numerous studies on the structure and function of FcεRI, the physiological relevance of FcεRI has remained an open question. FcεRI-deficient mice obtained by targeting specifically the FcεRIα or FcεRIβ gene are essentially normal in all aspects of their development and maturation (47, 74). Because IgE has been traditionally thought to play a role in the immune defense against parasites, the FcεRIα-deficient mice were used to investigate the possible role of FcεRI in the immune defense against Schistosoma mansoni (75). The natural resistance to Schistosoma mansoni as analyzed by adult worm and tissue egg burdens was similar in wild-type and FcεRIα-deficient mice. Similarly, the helminth-induced initial Th2 response appeared to be the same in both types of mice. However, clear-cut and somewhat unexpected differences in helminth-induced liver pathology were observed between the wild-type and FcεRIα-deficient mice, demonstrating a role for FcεRI in the control of the parasite-induced disease. The volume of granulomas was 25% greater, and the fibrosis 50% increased in the FcεRIα-deficient mice. The mechanism by which FcεRI controls the helminth-induced pathology is still unclear. It does not seem to be due to a variation in the Th2/Th1 balance since in situ measurements of various cytokines did not show measurable differences between both types of mice. However, non-B non-T cells from the FcεRIα-deficient showed a complete defect in their capacity to secrete IL-4 in response to anti-IgE or soluble parasite antigen, in contrast to wild-type mice. Whether and how this defect may translate to enhanced liver pathology is not clear. It is possible that IgE-FcεRI dependent secretion of IL-4 from non-B non-T participates in an inflammatory response which is beneficial to the host. In addition, IgEFcεRI dependent secretion of other inflammatory mediators from mast cells, a function which is clearly abolished in FcεRIα-deficient mice (see below), may also participate in a beneficial inflammatory response and in the control of the helminth-induced pathology. Regardless of the mechanism, this role must be played by mast cells, basophils, or non-B non-T cells because these are the only cell types expressing FcεRI in mice. However, in view of the capacity of the FcεRI-expressing eosinophils and platelets from humans to induce FcεRI-mediated killing of larvae in vitro (33, 36), it is tempting to speculate that both eosinophils and platelets, and possibly other FcεRI-expressing cells, may provide an additional protective layer in humans. As already mentioned, the resistance to infection by Schistosoma mansoni is not dependent on an IgE-FcεRI pathway. However, FcεRI may be required to

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

940

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

promote resistance to at least one other type of parasite: Haemaphysalis longicornis ticks. Studies on the resistance to cutaneous feeding of larval Haemaphysalis longicornis ticks show that mast cell–deficient mice are deficient in their capacity to resist this parasite and that this resistance depends on both the presence of IgEs and of mast cells (76–78). Since IgE interacts with mast cells mainly (and possibly exclusively) through FcεRI, these data suggest strongly that FcεRI plays a role in the mechanism of resistance. These studies show that IgE/FcεRI-mediated pathways and inflammatory cascade play a beneficial role in our defense system against parasites. However, the overall biological response that operates the control of parasite-induced pathology in one case and the resistance to the infestation in the other is not yet clearly understood and requires further study.

Type I Allergic Reactions It is beyond dispute that FcεRI, at the cellular level, plays a key role in mediating the IgE-dependent activation and degranulation of mast cell and basophils. The consequent release of mediators such as histamine is associated with the production of type I IgE-dependent allergic reactions. However, there had not been any definite analysis of the role of FcεRI in vivo and of possible redundant mechanisms supporting FcεRI function. These questions have now been addressed genetically by comparing wild-type mice with FcεRI-deficient mice obtained by specific gene targeting of either the α chain gene or of the β chain gene (47, 74). While it is easy to induce local or systemic anaphylactic reactions in wild-type mice after sequential injection of IgEs and the corresponding antigens, these reactions are completely abolished in FcεRI-deficient mice. Therefore, FcεRI is indeed absolutely required to permit IgE-mediated allergic reactions in vivo, and no other mechanisms can compensate, even partially, for its absence.

Antigen Presentation Because of the difference of cellular expression between mouse and humans, it is obvious that the specifically human function and biological significance of human FcεRI expression on monocytes, Langerhans cells, dendritic cells, eosinophils, and platelets cannot be studied in wild-type mice and FcεRIdeficient mice. To begin addressing this question, a number of functional studies have been done with human monocytes and dendritic cells, and, as already mentioned, with human eosinophils and platelets. FcεRI expressed on monocytes was shown to mediate small calcium fluxes (34) and to promote IgE-mediated antigen presentation (79). This antigen presenting function by FcεRI is even more effective on circulating dendritic cells (35). The

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIGH-AFFINITY IGE RECEPTOR

941

latter cells are efficient stimulators of both primary and FcεRI-dependent secondary T cell responses, and they therefore may be capable of priming naive T cells to IgE-reactive antigens as well as amplifying established T cell responses. Monocytes/macrophages, dendritic cells, and Langerhans cells (which are professional antigen presenting cells), all express αγ2 complexes. Therefore, it is possible that this αγ2 structure serves the antigen presenting function for example by specifically targeting the antigen-IgE-FcεRI complexes to the intracellular antigen presenting compartment. Furthermore, the antigen presenting function may operate only at a certain level of cell activation. In that case, the absence of FcRβ and of its signaling amplification function (see below) may be required to prevent full activation of the signaling machinery and to set the adequate activation level. In any case, the FcεRI-expressing cells supporting antigen presentation are not seen in rodents and for that reason, it has not been easy to assess the overall biological importance of the FcεRI-mediated antigen presenting function in host defense or in the development of allergic reactions. The “FcεRI-humanized” mice described in the section on “Genetic control of the cell specific expression of human FcεRI” (46, 47) may be a useful model in this regard.

From Parasites to Allergy? As discussed above, evidence is accumulating in mouse studies that FcεRI is playing a role in host defense against some parasites and in limiting the pathology induced by other parasites. In other words, the biological function of FcεRI in these circumstances is beneficial to the host. Therefore the acquisition by the human species of FcεRI expression on many cells and of the antigen presenting function corresponds probably to the gain of additional advantages to the host (80, 81). Epidemiological studies in Africa and South America support the concept that the IgE-FcεRI network is protective against Schistosoma mansoni (82–84). However, the reason for which this beneficial function has evolved to a function apparently harmful to the host as in the case of allergic diseases is not yet understood. In our “clean” societies, humans are no longer continuously facing parasites, and this situation may have contributed to host sensitization to innocuous antigens and the development of allergies (85). In support for this hypothesis, recent epidemiological studies (86) suggest that there is a tendency for the countries with expected endemic parasitic infestation to have lower incidence of allergic diseases. However, even if this hypothesis is true, it does not provide an explanation for the increase in incidence of allergic diseases in countries with high levels of parasitic diseases.

P1: SAT/VKS/MBG

January 25, 1999

942

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

REGULATION OF FcεRI BIOLOGICAL FUNCTION

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Regulation of FcεRI Expression Until recently, very few studies have addressed directly the question of regulation of expression of FcεRI. A correlation between the number of IgE molecules bound to basophils and the serum IgE level had been recognized long ago (87). Although the basis for this correlation was not known, it was suggested either that there was a common factor capable of regulating both IgE levels and IgE receptors or that the receptor number was modulated by the serum IgE concentration. These authors thought that the second hypothesis was more likely. Around 1983, two concomitant studies using the mast cell line (RBL-2H3) suggested that the second interpretation might indeed be correct. Incubation of RBL cells with monomeric IgE for 12 to 24 h resulted in an increase in the number of receptors at the cell surface (88, 89). This FcεRI upregulation by IgE was found to be insensitive to cyclohexamide and was therefore judged to be independent of protein synthesis. Rather, the mechanism of this upregulation was proposed to be the result of a decrease in the degradation rate of FcεRI “protected” by receptor-bound IgE, as if monomeric IgE was protecting the bound receptors from degradation during receptor recycling. Recently, Stephen Galli’s group in collaboration with us has found that the level of FcεRI expression on the mast cells of IgE-deficient mice was extremely low (80% reduced). However, the level of surface expression of FcεRI on mast cells could be upregulated (up to 32-fold) by in vitro incubation of the cells with IgE or injection of IgE in vivo (90). This FcεRI upregulation can also be observed on mouse basophils (91) and on human mast cells (92) and basophils (93). Therefore this ligand-mediated upregulation of FcεRI appears to be a general mechanism in both rodent and human. Furthermore, the structure of FcεRI (αβγ2 or αγ2 complexes) does not seem to matter. The density of FcεRI expression on monocytes (αγ2 complexes) appears higher in atopic subjects when compared to normal (34) and is most likely related to the higher level of circulating IgE (39). Similar observations have been made with eosinophils and Langerhans cells of atopic individuals (39), and treatment of atopic patients with anti-IgE that resulted in substantial decrease of the levels of serum IgE also caused a parallel decrease of FcεRI expression on basophils (94). In my view, the mechanism of this upregulation can be explained by the protection against degradation as suggested in one of the 1983 studies. In the recent in vitro study with mouse mast cells, two components of the FcεRI upregulation were identified: an early cyclohexamide-insensitive phase (confirming earlier studies) followed, a few hours later, by a more sustained component that is highly sensitive to cyclohexamide. Note that transcription of the subunits is unaffected (unpublished). In fact, the existence of these two components

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIGH-AFFINITY IGE RECEPTOR

943

corresponds exactly to what would be expected if a mechanism involving the inhibition of degradation was involved. During the first hours, IgE could upregulate FcεRI simply by protecting against degradation without being dependent on FcεRI synthesis. Later, when the pool of available FcεRI is fully utilized, further accumulation by the same mechanism should become dependent on protein synthesis. Therefore, while it is too early to exclude that other mechanisms could be involved in IgE-mediated FcεRI upregulation, the data available so far support a simple mechanism of inhibited degradation of the entire receptor complex. Importantly, this FcεRI upregulation by IgE results in critical enhancement of effector cell functions (90–92). Both serotonin and cytokine release are substantially enhanced in terms of both the sensitivity and the intensity of the response. Therefore this upregulation mechanism greatly enhances the capacity of FcεRI-expressing cells to respond to single antigens or to a multiplicity of antigens. This may be a critical mechanism to support host defense against various types and low levels of parasites, but the other side of the coin is that the same mechanism may permit allergic reactions to multiple allergens. Obviously, the regulation of FcεRI expression is not entirely dependent on ligand-mediated regulation. The mast cells from IgE-deficient mice still express low levels of FcεRI (90). This “basal” level could be under the control of cytokines. In mouse mast cells and human basophils generated from cultures of bone marrow in the presence of IL-3 (and in the absence of IgE), FcεRI expression occurs very early when the cells are still undifferentiated and before intracellular granules can be detected (95). IL-4 plays a positive role on the transcription of FcεRIα in human mast cells cultured from cord blood mononuclear cells (96) or from fetal liver (97). The rates of transcription, translation, and degradation of individual subunits have not been studied extensively, except to show that transcription of FcR γ chains (98) is enhanced by interferon-γ (22). However, the turnover of the cell surface–expressed tetrameric receptor was analyzed in the rat basophilic leukemia cell (RBL) line. The data from this study showed that surface receptors are internalized and degraded or re-expressed at the surface as a unit (89). RNA for the three subunits in mast cells is not equally abundant; β transcripts are substantially less abundant than α transcripts, while γ transcripts are the most abundant. This could indicate that the subunits are not coordinately synthesized, assuming that the rate of translation of these mRNAs is similar. Rather than controlling the synthesis of each subunit, the cell could regulate the assembly and surface expression of the receptor by controlling only the limiting subunit, as is apparently the case for the T cell receptor (99). A number of studies suggest that it is the case. Mouse mast cells lacking expression of FcεRI are deficient in FcRγ but not FcεRIα and FcRβ transcripts (100). Similarly,

P1: SAT/VKS/MBG

January 25, 1999

944

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

expression of FcεRI (αγ2 complexes) on freshly isolated epidermal Langerhans cells appears to correlate with the level of expression of FcRγ chains while the pool of intracellular FcεRIα is greater than the surface expressed FcεRIα (101). By contrast, disappearance of the FcεRI surface expression seen on epidermal Langerhans cells subjected to 24-h culture correlates with the disappearance of FcεRIα transcripts, an indication that the same cell type may use different mechanisms to regulate the cell surface expression levels of FcεRI (101).

The FcRβ Chain Amplification A novel potential way to regulate FcεRI function has been recently identified. In vitro studies from our laboratory addressing the function of FcRβ during early FcεRI signaling demonstrated that FcRβ functions to amplify cell activation signals mediated through the FcRγ subunit (49) (see paragraph below on signaling). To assess whether FcRβ could influence a host response, an in vivo model had to be constructed. As mentioned above murine αγ2 complexes are not transported to the surface and, therefore, a standard homologous recombination to eliminate β chain expression cannot produce a murine experimental system with which to study β function. However, human αγ2 (or human α/murine γ2 ) complexes are transported to the cell surface. For this reason, the function of FcR was analyzed by comparing the “humanized” transgenics described above with corresponding mice having the same human FcεRIα cell distribution but expressing only αγ2 receptors because of targeted disruption of the FcRβ gene (47). The analysis of the FcεRI-dependent responses in these mice showed that FcRβ functions as an amplifier not only of early signaling events but also of later responses such as cell degranulation and cytokine release, with a gain factor ranging from 3 to 5. In addition, FcRβ is a critical and substantial amplifier of in vivo anaphylactic responses. Therefore, these data are unequivocal genetic evidence that FcRβ has the capacity to modulate an individual’s allergic response and that this capacity is due to its function as a net amplifier of effector cell responses. The importance of FcRβ as an amplifier of the host response must be considered in the context of recent genetic/epidemiologic studies of atopic human families (see below).

FcεRI AND HUMAN DISEASES Allergic Diseases and the FcRβ Gene Many cells and molecules are involved in the production of allergic reactions, including mast cells, basophils, IgE, FcεRI, and many signal transduction molecules. Obviously, this does not imply that any of these cells or molecules are involved in the pathogenesis of allergic diseases or, in other words, in

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIGH-AFFINITY IGE RECEPTOR

945

discriminating between allergics and nonallergics. However, a number of genetic studies have suggested that the FcRβ gene may play a role in this regard. Cookson, Hopkin, and colleagues have accumulated evidence that a gene on chromosome 11 (in the 11q12-13 region) is linked with atopy (102, 103). The linkage between this region of chromosome 11 and atopy and/or bronchial hyperreactivity has been replicated by some but not all groups working with independent sets of probands, suggesting an heterogeneity in the genetic susceptibility to atopy (104–117). The Cookson and Hopkin group narrowed the critical atopy region to focus on the gene encoding the FcRβ chain (108). The FcRβ gene consist of 7 exons, spans about 10 kb, and contains an upstream TATA box. The first exon codes for the 50 untranslated region and a portion of the N-terminal cytoplasmic tail. The first transmembrane domain (TM) is encoded by exons 2 and 3, TM 2 by exons 3 and 4, TM 3 by exon 5, and TM 4 by exon 6. The seventh exon encodes the end of the C-terminal cytoplasmic tail including the FcRβ ITAM and the 30 untranslated region (118). Several coding sequence polymorphisms in the FcRβ gene that are associated with allergic phenotypes were subsequently identified (119–121). The first to be described was in the fourth transmembrane domain of FcRβ, a substitution of a leucine for isoleucine at position 181 (121). However, this polymorphism is uncommon in the general population, with subsequent analyses showing that it is generally found together with a second amino acid change (V183L), and that this double substitution is more strongly associated with atopy than I181L alone (120). Interestingly, the variant sequence Leu181/Leu183 was detected in 72% of the chromosomes analyzed in a Kuwaiti population, indicating a high prevalence in Kuwaiti Arabs compared to British, Australian, and Austrian populations (122). This difference of prevalence among different populations is likely to explain the apparent discrepancies between the studies. In the Kuwaiti Arabs study, the double variant was confirmed to be associated with asthma. However, the I181LV183L double substitution alone could not account for all of the linkage originally found to the FcRβ gene (120). A third polymorphism was identified, consisting of a glutamic acid to glycine substitution at position 237 in the β amino acid sequence, which is also strongly linked with atopy, and which is present in 5% of the general population (119). Most intriguingly, this substitution introduces a significant hydrophobicity change just outside of the critical ITAM motif (see signal transduction and the importance of ITAMs) in the β carboxy-terminal cytoplasmic tail. Since the description of the variants described above, a number of additional variants have been found by a number of groups and are currently being investigated (123). In one recent study, a RsaI polymorphism (intron 2 and exon 7) was found to be strongly associated with atopic dermatitis (124).

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

946

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

Because they link the development of atopy with a receptor (FcεRI) that can be involved in generating atopic responses, and with a subunit (FcRβ) of this receptor now known to function to enhance its signaling, these reports provide an intriguing initial step toward understanding a potential immunologic dysregulation that could lead to the development of atopy. The demonstration that FcRβ functions as a signal amplifier provides a strong rationale for examination of the described atopy-associated polymorphisms, and possibly others, in terms of their ability to modulate FcRβ amplifier function and the effects of such modulation on immunologic and allergic responses in vivo.

Autoantibodies and Chronic Urticaria Chronic urticaria (CU) is a common disorder characterized by recurrent bouts of itchy wheals. In spite of the allergic appearance of the lesions, in most cases no specific allergen is found by provocation tests that could explain skin mast cell degranulation in an IgE-FcεRI dependent manner. The fact that patient serum can reproduce the lesions when injected intradermally and induce histamine release by normal basophils in vitro demonstrates that an endogenous factor present in the serum may be responsible for triggering mast cells (125). Part of this releasing activity could be due to anti-IgE autoAbs binding to IgE molecules normally occupying FcεRI on basophils and mast cells (126, 127). However, the role of these anti-IgE autoAbs in the pathogenesis of CU is not clear because they can also be detected in the sera of normal individuals and of patients suffering from atopic dermatitis (128, 129). Another potential mechanism for this releasing activity has been provided by recent studies reporting the presence of IgG autoAbs against the FcεRI α chain in the serum of patients with CU. Among 26 CU patients whose serum caused a wheal when injected intradermally, 4 had a releasing activity contained in the IgG fraction that was inhibited by preincubating the IgG samples with a recombinant protein corresponding to the extracellular part of the α chain of FcεRI (125). In a subsequent study, Georg Stingl’s group found that 12 CU patients among 32 had IgGs capable of recognizing α in a western blot assay (128). Out of the 12 autoAbs, 8 were tested for their functional capacity, 4 of which were able to induce histamine release by normal basophils stripped of IgE, suggesting that this activity was not due to anti-IgE IgG. Unlike anti-IgE autoAbs, the anti-FcεRIα autoAbs were apparently not observed in normal controls or in patients with atopic dermatitis providing however that the sera be diluted sufficiently. In a later study involving 163 CU patients, 38 (23%) had anti-FcεRIα autoAbs as assessed by the capacity of soluble recombinant FcεRIα to inhibit the serum-induced release from basophils or skin mast cells in vitro (130). Similar functional results were obtained by Kaplan’s group with anti-FcεRIα autoAbs detected by western blotting (131). From these studies, it appeared that there was a subset

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIGH-AFFINITY IGE RECEPTOR

947

of CU patients with anti-FcεRIα autoAbs that were able to activate basophils and skin mast cells. However, it was also apparent that both the proportion of patients with the autoAbs and the proportion of autoAbs capable of degranulating basophils varied in the different studies. This variability could be explained in part by the different methods used in the respective studies (serum dilutions, western blotting, or ELISA). In addition, various factors could contribute to the heterogeneity in releasing activity among sera: (a) differences in the releasability of basophils collected from different individuals; (b) differences in the titer of the autoAbs; (c) differences in the affinity of the autoAbs for the α chain; (d ) differences in the epitopes recognized on the α chain and in the capacity of the autoAbs to cross-link FcεRI; (e) differences in the capacity of the Fc fragment of autoAbs to interact with the inhibitory Fcγ RIIB expressed on both mast cells and basophils; after cross-linking of FcεRI by the autoAbs, the Fc domain of the autoAbs could then bind to Fcγ RIIB, thereby inducing the co-cross-linking of Fcγ RIIB with FcεRI; this would result in inhibition of the release induced through FcεRI (132). Recent findings have shed new light on the question of detection of these autoAbs in CU patients. Using an ELISA assay different than the one referred to above, a group of investigators in Switzerland have shown that the anti-FcεRIα autoAbs can be detected in normal and in CU patients with about the same frequency and that there is no apparent correlation with their detection and the CU disease status (MP Horn, T Gerster, B Ochensberger, T Derer, F Kricek, J-P Kinet, M-H Jouvin, M Vogel, BM Stadler, SM Miescher, submitted). These data are consistent with our own recent results using various types of ELISAs (M-E Jouvin, J-P Kinet, unpublished). Furthermore, the same investigators came fortuitously across an interesting observation when analyzing multi-donor IgG preparations: The anti-FcεRIα autoAbs have a stronger affinity for a tetanus toxoid antigen (TTd) than for the FcεRIα antigen (MP Horn, T Gerster, B Ochensberger, T Derer, F Kricek, J-P Kinet, M-H Jouvin, M Vogel, BM Stadler, SM Miescher, submitted). This raises the possibility that the cross-reactive anti-FcεRIα autoAbs may appear after the TTd vaccination for an unknown reason. The existence of nonpathogenic anti-FcεRIα autoAbs in normal human sera raises several important issues. First, discriminating between autoAbs having or not having the capacity to induce cell degranulation is essential. Second, it is clear that a simple ELISA cannot be used to discriminate between pathogenic and nonpathogenic anti-FcεRIα autoAbs. The identification of specific epitopes potentially associated with the pathogenicity may be required for this distinction. Alternatively, well designed functional tests that do not rely on the variability of the release assays performed with human basophils may be necessary. For example, the use of transfected clonal cell lines may help in the design of well-controlled assays to assess the capacity of anti-FcεRIα autoAbs to trigger cell degranulation.

P1: SAT/VKS/MBG

January 25, 1999

948

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

SIGNAL TRANSDUCTION MECHANISMS

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

The Concept of ITAMs FcRs and other antigen receptors are thought to transduce signals through cytoplasmic tail structures known as immunoreceptor tyrosine–based activation motifs (ITAMs) (133). These activation motifs were first described by Michael Reth in 1989, who recognized an 18–amino acid motif (E/DxxYxxLxxxxxxxYxxL) present in three copies in the T cell receptor (TCR) ζ subunit, and one copy each in the TCR γ , δ, and ε subunits, B cell receptor (BCR) Igα (MB-29) and Igβ (B1) subunits, and Fc receptor FcRβ and FcRγ subunits (134). The ITAMs of FcRβ and FcRγ are represented as filled circles in Figure 1. The physiologic importance of this homology was confirmed three years later by three elegant studies in which it was shown that a copy of this motif derived from either the TCR ζ or FcRγ chain could be attached to irrelevant extracellular and transmembrane domains to form chimeric receptors able to function identically as natural antigen receptors (135–138). An excellent review on FcRs with and without ITAMs has been published recently by Marc Daeron in Annual Review of Immunology (139). For that reason, the present review focuses almost exclusively on the signaling properties of FcεRI and of its signaling chains, the FcRβ and FcRγ chains. A special attention is given to the most recent findings on FcεRI signaling reported over the last two years.

Clustering and Tyrosine Phosphorylation Signaling through FcεRI and the resulting cell activation is initiated when binding of specific multivalent antigen to receptor-bound IgE results in receptor clustering. A unique and interesting feature of FcεRI is its ability to be rapidly engaged with multi-hapten antigen and just as rapidly disengaged through the use of monovalent hapten. Disengagement of FcεRI results in immediate arrest of cell signaling, demonstrating that continuous receptor engagement is required for maintaining cell activation. The requirements for active receptor clustering in activation of the signaling machinery have been well documented and discussed extensively before in multiple reviews (140–142) and are not discussed here. For a long time, the initial events following FcεRI clustering remained unclear. The status of the phosphorylation of FcRβ and FcRγ had been analyzed in the early 1980s, but no tyrosine phosphorylation was detected (143). Tyrosine kinase activity was found coprecipitating with FcεRI as early as 1986 but was considered a contaminating activity (144). In 1990, an important study showed that tyrosine phosphorylation of various substrates including a 72 kDa was induced by FcεRI clustering, but FcεRI was not thought to be tyrosine

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIGH-AFFINITY IGE RECEPTOR

949

phosphorylated. In 1991, concomitantly with the discovery of the functionality of the ITAMs, one of our study demonstrated that (a) tyrosine phosphorylation of FcRβ and FcRγ was one of the earliest events detectable within 5 s after receptor clustering and (b) FcεRI tyrosine phosphorylation was a reversible function; as a consequence of FcεRI disengagement, FcRβ and FcRγ dephosphorylate very rapidly (145). In 1992, Joseph Bolen’s group discovered that FcεRI was activating the Lyn tyrosine kinase of the src family shortly after receptor clustering (146). These results were followed by a number of confirmatory studies and by the important demonstration that the Syk tyrosine kinase is also activated though FcεRI (147–151). Phosphorylations of ITAMs on the canonical tyrosines after antigen receptor cross-linking are in fact critical for signal transduction, as mutations of these residues abolish signaling (138). In addition, the discovery that tyrosinephosphorylated ITAMs are binding targets of src homology two (SH2) domains provided a structural basis for the activation properties of these motifs, as an inducible binding site for SH2 containing signaling proteins (152, 153). The two phosphorylated tyrosines within ITAM motifs function as binding sites for tandem SH2 domain-containing tyrosine kinases of the Syk/zap-70 family and possibly a variety of other molecules involved in signal transduction(154–157). In particular, this type of interaction between Syk and FcεRI was shown to occur and to be functionally important for downstream events (158–164).

FcRβ– and FcRγ –ITAMs Signal Differently The variety of structures found in the antigen receptor superfamily suggests that receptors with different structures would have different functions. Since the identification of the ITAM motif as the key functional signal transduction unit of antigen receptors, much work had been put into identifying differences in the activation properties of motifs from different receptors. Initially, when placed in the context of chimeric receptors, individual ITAM motifs had been found to have cell activation properties equivalent to those of their intact parent receptors and each other, albeit of different efficiencies (135–138, 165). These findings had been interpreted to mean that the multiple ITAMs present in holoreceptors provide parallel but equivalent signaling functions (154). However, this interpretation had left unexplained the impetus for evolution of complex multisubunit structures. Data from our laboratory and others suggested that the FcRβ and γ subunits play different roles in signaling (162, 166–170). One study that used a similar strategy of chimeric molecules as described above showed that the FcRβ and FcRγ could function differently in a mast cells context. FcRβ unlike FcRγ could be coprecipitated with Lyn but not with Syk, while FcRγ but not FcRβ could activate Syk (168). Another critical difference was that FcRβ was

P1: SAT/VKS/MBG

January 25, 1999

950

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

incapable of inducing cell degranulation while FcRγ could do it, although with a lesser intensity than did the corresponding endogenous tetrameric FcεRI (168). Together these data clearly indicated that FcRβ and FcRγ were functioning differently.

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

A Two-Step Activation Model: Lyn Is Upstream of Syk In addition to the data described in the previous paragraph, another critical and unexpected finding helped us to propose an initial model for FcRβ and FcRγ signaling. An FcεRI-deficient mast cell line was reconstituted by transfection with either the wild-type αβγ2 FcεRI complexes or mutated versions. Mutation of the two canonical tyrosines of the FcRγ ITAMs abolished the tyrosine phosphorylation of FcRγ chains but not of FcRβ chains, as expected. Surprisingly, the corresponding mutation of the FcRβ ITAMs abolished the tyrosine phosphorylation of both FcRβ and FcRγ , implying that FcRβ played a role in FcRγ tyrosine phosphorylation (168). Since FcRβ chimeras could bind Lyn (see above), we proposed a model of FcεRI activation in which FcRβ and FcRγ act cooperatively; Lyn is bound to FcRβ under resting conditions, Lyn phosphorylates FcRβ and FcRγ after receptor engagement with antigen, the phosphorylated FcRγ binds Syk, and FcRγ -bound Syk becomes phosphorylated and activated (168). In support of this model were other data demonstrating that Syk could be activated in vitro by binding phosphorylated FcRγ s ITAMs (159). Studies with inhibitors of Syk that could inhibit cell degranulation but not the phosphorylation of FcRβ and FcRγ also supported the position of Lyn and of the phosphorylation of FcRβ and FcRγ upstream of the activation of Syk (163, 164). To test this proposed model of activation, a reconstitution system was designed in which the expression of each molecule or a mutated version of it could be controlled (162). Using this system, several features of the proposed model were directly demonstrated including: (a) Lyn phosphorylates both FcεRI β and γ ITAMs, (b) β and γ ITAM phosphorylation by Lyn causes Syk to bind to FcεRI, and (c) Syk binding to FcεRI leads to its Lyn-dependent tyrosine phosphorylation and activation (162). In addition, the system of reconstitution was successfully used to reconstruct the dependence of these events on receptor clustering (162). The ability to control the expression levels of Lyn and Syk was critical to this reconstruction, as clustering control was present at levels of Lyn and Syk expression near those seen in hematopoietic cells but was lost at higher levels of expression. Another approach, overexpression of catalytically inactive molecules, was used for identification of Syk substrates and Syk-dependent signaling pathways (161, 171), and to corroborate data obtained from reconstitution experiments in fibroblasts in a native signaling environment (162). In particular, a Sykdominant negative could abolish Syk activation, calcium signaling, and cell

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

951

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

degranulation but not the phosphorylation of the FcRβ and FcRγ subunits. Thus, these data could confirm, in a mast cell context, the data obtained in the reconstitution system. Later, a number of genetic studies confirmed this model. In Lyn-deficient mast cells, no phosphorylation of FcRγ and FcRβ could be detected, while in Syk-deficient mast cells, the phosphorylation of both FcRγ and FcRβ were present after receptor clustering but downstream signals (phosphorylation of PLC-γ , calcium mobilization, and degranulation) were abolished (172–174).

FcRγ Functions as an Independent Activation Module and FcRβ as an Amplifier The above model was paradoxical. It gave a prominent role to FcRβ and Lyn even though the FcRγ ITAM-containing chimeras could activate Syk, calcium mobilization and cell degranulation without apparent requirement for the FcRβ ITAM. Furthermore, the model could not explain how the newly discovered human αγ2 complexes could function. In studies designed to understand the interplay between FcRβ and FcRγ subunit function in a holo-receptor context during FcεRI signaling, we discovered that FcRγ functions as an autonomous activation module, whereas FcRβ has the functional properties of an amplifier (49). The concept of FcRβ as an amplifier is based on the facts that FcRβ (a) does not have autonomous cell activation capability, (b) amplifies the intensity of cell activation signals sent by the FcRγ chain with a gain of five-to sevenfold (as measured by Syk activation and calcium mobilization), and (c) does not qualitatively alter the FcRγ -mediated cell activation signals. Furthermore, the mechanism of FcRβ amplification was shown to involve the FcRβ ITAM acting as an inducible “anchor” for Lyn, resulting in enhanced Lyn recruitment to the receptor complex after engagement, enhanced Lyn-dependent phosphorylation of FcRγ , enhanced activation of the Syk nonreceptor tyrosine kinase, and enhanced calcium mobilization (49). The notion that FcRβ works as an amplifier was subsequently challenged by binding and coprecipitating studies. FcRβ ITAM but not FcRγ ITAM binds in vitro to SH2-containing tyrosine (SHP-1 and SHP-2) and SH2-containing inositol-50 -phosphatase (SHIP) in vitro (175–177). In addition, SHP-2 coprecipitates with FcεRI after receptor engagement (176). Since these phosphatases have been implicated in negative regulating functions, these data suggested that FcRβ might have a negative regulatory function. However, affinities of the phosphorylated FcRβ and FcRγ ITAMs were measured recently using surface plasmon resonance under carefully controlled conditions. In this study (178), Syk was confirmed to bind to FcRγ with high affinity (IC50 = 23 nM) (159) and to FcRβ with 18-fold lower affinity, confirming the preference of Syk for FcRγ ITAM. Importantly, no binding of SHP-2 to FcRβ or FcRγ ITAMs was

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

952

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

observed even at 100 µM concentrations, suggesting that in physiologic conditions, SHP-2 is unlikely to bind these motifs. More functional studies on the functional role of these phosphatases need to be done to clarify their role in FcεRI signaling. That FcRβ indeed functions as an amplifier of both early and late responses has now been settled unequivocally by the genetic study described above in the paragraph on FcRβ chain amplification (47). However, although overwhelming, the evidence for the FcRβ amplification function does not exclude the possibility that FcRβ is playing an autonomous specific role in addition to being an amplifier of FcRγ -mediated functions. So far, however, no such specific role for FcRβ has been identified functionally.

Molecular Interactions Initiating Lyn and Syk Activation How is Lyn interacting with FcRβ? In vitro studies have shown that SH2 domains of Lyn bind to FcRβ ITAM (158), but our own functional data suggest that the Lyn SH2 domain is not required for FcεRI phosphorylation (S Lin, J-P Kinet, unpublished). Other studies using the yeast two-hybrid system showed that the FcRβ ITAM-containing cytoplasmic domain interacts with the SH4-containing unique domain (179). Overexpression of this domain could inhibit Lyn-dependent tyrosine phosphorylation of FcεRI, confirming the in vitro binding data (179). However, in another study, overexpression of proline-rich peptides that bind specifically Lyn SH3 domains suppressed the Lyn-mediated phosphorylation of FcεRI, the membrane translocation of Syk to FcεRI and calcium signaling (180). One possible interpretation of these results is that both domains are involved in FcRβ interaction. Alternatively, the proline-rich peptides binding to the SH3 domain could impair indirectly the SH4-binding function. How are αγ2 complexes signaling in the absence of FcRβ? As mentioned above, FcRγ is able on its own to activate Syk, calcium signaling and degranulation but at a much lower level than αβγ2 complexes (49, 168). It is critical to note that FcRγ -mediated activation of Syk in the absence of FcRβ is still dependent on the presence of the Lyn kinase (49). The latter kinase is necessary to promote the low level (difficult to detect) tyrosine phosphorylation of FcRγ and the activation of Syk. That the Lyn kinase (or another src family kinase) initiates Syk activation has now been established (181). The Lyn kinase initiates the process by phosphorylating tyrosine-518 in the activation loop of Syk, inducing Syk to autophosphorylate and to become fully active. Thus, the absence of FcRβ does not preclude Lyn, which is anchored in the inner layer of the plasma membrane by its aminoterminal myristylated glycine, from phosphorylation of FcRγ upon receptor clustering and activation of Syk. However, this process is relatively inefficient, and by anchoring the Lyn kinase, FcRβ

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

HIGH-AFFINITY IGE RECEPTOR

953

provides access to FcRγ and the Syk tyrosine kinase, substantially enhancing the whole process. The domains of Lyn required for a functional interaction with FcRγ and Syk in the absence of FcRβ have not yet been analyzed. How is Lyn activated? The classical view of the activation process of src family kinase is as follows. The csk kinase phosphorylates src family kinases at an autoregulatory c-terminal tyrosine which, when phosphorylated, binds to its own SH2 domain thereby forming an internal loop that is keeping the kinase inactive. The CD45 phosphatase then dephosphorylates this regulatory tyrosine, consequently activating the kinase (182). Evidence that CD45 is involved in the FcεRI activation has been provided in a few studies. Antibodies against CD45 inhibited FcεRI-mediated histamine release from human basophils (183). Clustering of reconstituted FcεRI in a CD45-deficient Jurkat T cell line did not result in phosphorylation of FcRβ and FcRγ and calcium mobilization. However, both functions were restored after reconstitution with a chimeric molecule containing the phosphatase domain of CD45 (184). However, in NIH 3T3 cells that lack expression of CD45, clustering of FcεRI resulted in Lyn-dependent phosphorylation of FcRβ and FcRγ , suggesting that CD45 is not required for Lyn activation (49, 162). It is possible that another phosphatase plays the role of CD45 in NIH 3T3 fibroblasts, or that the role of CD45 is to enhance the efficiency of src family kinase activation. In that case, overexpression of Lyn in NIH3T3 may result in a sufficient amount of Lyn in activated form to get the system to work. Thus, the exact molecular events regulating Lyn activation are not yet fully understood. It is clear, however, that clustering is necessary to initiate the increase in tyrosine phosphorylation of FcRβ and FcRγ (145). This process appears to be initiated by the FcεRI molecules that are adjacent in the cluster to FcεRI-bound Lyn (185, 186). This transphosphorylation activation process seems to be the only one that would work efficiently with the stoichiometry of Lyn to FcεRI being substantially lower than 1. In fact, Lyn has been shown to be limiting in an elegant study using two different FcεRI ligands (187). Activation through the first ligand inhibited subsequent tyrosine phosphorylation of FcεRI complexes activated via the second ligand. However, the disengagement of FcεRI from the first ligand induced tyrosine phosphorylation of FcεRI already clustered by the second ligand, demonstrating that the kinase responsible for FcεRI phosphorylation cannot phosphorylate efficiently both sets of FcεRI complexes and is therefore the limiting factor. Retrospectively, it is interesting to note that the NIH3T3 reconstitution of the clustering dependence of tyrosine phosphorylation events initiated by Lyn, was dependent on the level of overexpression of Lyn (162). At high levels of expression, the clustering dependence was lost, suggesting that an intrinsic part of the clustering dependence is a limited amount of Lyn.

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

954

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

Interesting data from Henry Metzger’s group have taken the observation that Lyn is limiting a step further (188). The capacity of downstream signaling by an individual FcεRI is dependent on its capacity to remain in a cluster and is therefore influenced by the ligand affinity. Consequently, low affinity ligands allow the previously engaged receptors to become rapidly disengaged and unproductive in downstream signaling. Furthermore, it appears that an excess of receptors clustered with low affinity ligands is able to sequester Lyn so that a smaller amount of receptors clustered with higher affinity ligand are unable to activate the signaling cascade. The fact that clustered receptors compete for a limited amount of Lyn is proposed to be the basis of the mechanism for ligand antagonism (188).

The Tec Family Kinases: Bruton Tyrosine Kinase (Btk) and Itk As mentioned above, the earliest events within 5 s of FcεRI clustering are receptor tyrosine phosphorylation and activation of the Lyn tyrosine kinase. Within 15 s Syk activation follows concomitantly with the initiation of calcium mobilization. Activation of another tyrosine kinase, the Bruton tyrosine kinase (Btk) has also been described after FcεRI clustering, suggesting a role in mast cell activation. Btk is a member of the recently described Btk/Tec family of tyrosine kinases (189–191). Members of this family are distinguished by the presence of a large unique amino terminal region containing a pleckstrin homology (PH) domain, single SH3 and SH2 domains, and the lack of an autoregulatory C-terminal tyrosine in their requisite kinase domains (191–193). Within the Btk/Tec family, Btk is of particular interest due to its causal role in X-linked agammaglobulinemia in humans (Xla), and in x-linked immunodeficiency in mice (Xid) (189, 194, 195). These related immune deficiencies are associated with an arrest in B cell development at the pre-B cell stage, resulting in low or absent B lineage cell numbers and hypo- or agammaglobulinemia. Unlike B cell development, mast cell development in Xla is apparently normal. However, Xid mice exhibit an impaired FcεRI-dependent anaphylactic reaction, with late phase reactions more affected than the early phases (196). Consistent with this, cultured mast cells from both Xid and Btk null mice exhibit mild impairments in FcεRI-mediated degranulation and more profound defects in FcεRI-mediated cytokine (TNF-α, IL-2, IL-6, GM-CSF) production (196). In fact, a recent study shows that Btk is required for maximal FcεRI-mediated IL-2 gene activation and that this regulation occurs through one of the MAP kinases, the c-Jun N-terminal kinase (JNK) (197). Another Tec family kinase (Itk, also called emt) is expressed in mast cells and is activated through FcεRI (198). Overexpression of Itk in Btk-deficient mast cells could partially restore the defects of histamine release or cytokine

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

955

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

production but not to the levels obtained with wild-type Btk, indicating at least a partial redundancy of these two Tec kinases (196). From that result, one could have extrapolated that the defects in cell degranulation and cytokine release would probably have been much more profound in the absence of both Tec kinases. In fact, this is exactly the situation in Btk/Itk double deficient mice, where FcεRI-mediated degranulation of mast cells and anaphylaxis are abolished (Fred Alt, personal communication).

A Novel Connection: Phosphatidylinositol 3 Kinase-Tec Kinases-PLCγ -Sustained Calcium Influx What function could Tec kinases play in signaling that could explain their biological activity in lymphocytes and mast cells? The beginning of the explanation came from an experiment in the DT40 chicken B cell line in which the gene for Btk was targeted (199). In that cell line, no other Tec kinases are expressed, and the phenotype of the cell line was dramatic: No calcium mobilization and a reduced phosphorylation of PLC-γ 2 were observed after B cell receptor (BCR) clustering. This experiment implicated Btk as a dominant player in the calcium mobilization pathway most likely through an effect, direct or indirect, on PLC-γ2 activation. However, the molecular mechanism by which Btk could play this function was unclear. In order to understand how Btk could regulate calcium mobilization, it was critical to first understand how Tec kinases are activated. Data from B cell studies and from a reconstitution system demonstrate that the basic mechanisms of activation of the Syk and Tec kinases are similar and are initiated by Lyn (181, 200), which phosphorylates a regulatory tyrosine in the activation loop of Syk (tyrosine-518) and Btk (tyrosine-551). However, for this to happen, membrane targeting of the cytoplasmic kinase is required. Membrane targeting of Syk to the phosphorylated FcRγ ITAMs is well grasped. However, the molecular interactions required for the membrane targeting of Btk were not well understood until recently. In vitro binding studies with the SH2, SH3, and PH domains of Tec kinases had identified a number of potential membrane targets. In particular, the PH domain of Btk was shown to be capable of interacting with phosphatidyl-3,4,5,-trisphosphate (PdtIns-3,4,5P3), a product of the activation of phosphatidylinositol 3 kinases (PI3 kinases) (201, 202). Numerous receptors have been shown to activate PI3 kinases (reviewed in 203, 204). To demonstrate the effect of the activation of PI3 kinases, most studies have used wortmannin or similar inhibitors of its enzymatic activity. The activation of PI3 kinase through FcεRI was demonstrated to occur in mast cell lines using such inhibitors. Inhibition of PI3 kinase caused the inhibition of FcεRI-mediated cell degranulation but did not influence the activation of

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

956

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

Lyn or Syk kinases (205–207). However, wortmannin was shown to inhibit the formation of the second messenger inositol-1,4,5-P3 (IP3), itself the product of the action of phospholipase C isoenzymes (207). All of these observations about Tec kinases and PI3 kinases fell into place when a novel pathway linking PdtIns-3,4,5-P3 to calcium signaling was described. Using a B cell line and a reconstitution system, the direct demonstration was made that PdtIns-3,4,5-P3 interacting with the PH domain serves as the membrane targeting signal and upstream activation signal of Tec kinases (208, 209). The lyn-initiated activation of membrane-targeted Tec kinases results in Tec-kinase activation and in Tec-kinase dependent PLCγ tyrosine phosphorylation (this function requires the Tec SH2 domain), which induces enhanced IP3 production. In turn, this PdtIns-3,4,5-P3-dependent IP3 production appears to be critical for maintaining the sustained calcium influx (210, 211), probably by keeping empty the specific intracellular stores that regulate the opening of the calcium regulated activated channels (CRAC) (209, 212). The capacity of the PI3 kinase-Tec-PLCγ to regulate intracellular calcium levels allows this pathway to regulate later events of the signaling cascade. For example, the peak and sustained phases of the intracellular calcium increases are required for induction of the mitogen-activated protein kinase (MAPK), JNK and p38 cascades and the calcium dependent transcription factor NF-AT and NF-κB (213, 214). A minireview on the PI3 kinase-Tec-PLCγ pathway and the mechanisms by which it may regulate the openings of the calcium regulated activated channels (CRAC) has been recently published in Cell (209); because of space limitations, we do not discuss any more details here except to say that this pathway is blocked when a SHIP-dependent inhibitory receptor such as Fcγ RIIb is engaged. Coengagement of Fcγ RIIb with FcεRI efficiently inhibits cell degranulation (132). The inhibitory function of Fcγ RIIb in B cells or mast cells is mediated by SHIP (see the accompanying chapter on Inhibitory Receptors by Eric Long), but how SHIP was producing its inhibition had remained unclear. In fact, in a B cell line and a reconstitution system, SHIP hydrolyzes Pdt Ins-3,4,5-P3 and in so doing inhibits Btk activation, PLCγ phosphorylation, IP3 production, and the sustained calcium influx (210, 211). Where is Syk in all of this? In mast cells and B cells, the absolute requirement for Syk in calcium signaling and many downstream functions has been demonstrated in many studies (158–164, 171, 173, 174). At the same time, Btk-deficient DT40 B cells (which are in fact Tec kinases deficient) do not flux calcium either, indicating a crucial role for Btk in calcium signaling as well. How could both Btk and Syk be required? The mechanism of activation of PI3 kinases has been thought to result from an interaction of the p85 subunit of PI3 kinase with src family kinases (215). We revisited recently this question by using the Syk-deficient DT40 cell line

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

957

and demonstrated that the BCR-dependent production of PdtIns-3,4,5-P3 is abolished. In the RBL mast cell line and in another B cell line A20, we could also confirm that Syk is indeed upstream of the activation of PI3-kinases (LO Beitz, DA Fruman, AM Scharenberg, T Kurosaki, LC Cantley, J-P Kinet, unpublished). Therefore, these data support a model in which Syk is upstream of PI3 kinase, which is itself upstream of Btk.

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Sphingosine Kinases Knowledge of the diverse array of biological activities mediated by sphingolipid metabolites has exploded in the past few years (216–218). One of these metabolites, sphingosine-1-phosphate (S1P), is a calcium mobilizing second messenger (219). S1P is generated by the activity of sphingosine kinase (SK), an enzyme described approximately 20 years ago, which until recently has been difficult to adequately purify (220). SK is thought to work directly on available plasma membrane sphingosine to produce S1P (221). S1P mediates calcium efflux from microsome preparations, and an S1P gated calcium channel has been described in microsomes (222–224). Rapid production of S1P has now been proposed to be the mechanism by which the platelet derived growth factor (PDGF) receptor mobilizes calcium (219). A study of RBL-2H3 cells transfected with a G-protein coupled receptor (the m1 muscarinic receptor) had produced a paradoxical result which was not consistent with PLCγ /IP3 as the sole mechanism of FcεRI-mediated calcium mobilization. Stimulation of transfected m1 receptors with carbachol produces a calcium flux similar in magnitude to that of FcεRI, but which is associated with more than 10 times the IP3 release seen with FcεRI stimulation (225, 226). This FcεRI “IP3 gap” suggested that FcεRI may also utilize another second messenger such as S1P to mobilize calcium. This possibility was tested by using a sphingosine kinase inhibitor, DL-threo-dihydrosphingosine (DHS) (227), and measuring its effect on calcium mobilization through FcεRI and through carbachol in m1-transfected RBL-2H3 cells (228). At a given concentration, DHS almost completely suppressed calcium mobilization through FcεRI, and essentially left unaffected calcium mobilization through m1 muscarinic receptors, suggesting that FcεRI mobilizes calcium primarily via activation of an SK. FcεRI-mediated SK activation was then directly addressed by measurement of the production of S1P and of the SK activity after engagement of FcεRI with antigen. These data showed that stimulation through FcεRI produced S1P and an approximate two-fold enhancement of SK activity, nearly identical to that described for the PDGF receptor (219). Finally, the S1P applied to cells induced an EGTA-insensitive calcium mobilization, an indication that it was the result of a release from intracellular stores. Together these data suggested that FcεRI utilizes a SK pathway (228).

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

958

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

A number of studies led to a reappraisal of our initial interpretation of these data. First, studies described the presence on the cell surface of G-proteincoupled receptor for S1P, which is pertussis toxin sensitive and capable of mobilization of calcium via IP3 (218, 229–231). In light of these studies, the experiment of S1P-mediated calcium was repeated in presence of pertussis toxin. The toxin could abolish the S1P-mediated calcium mobilization, indicating that it was using the G-protein receptor (O Choi, J-P Kinet, unpublished). Because the FcεRI-dependent pathway is pertussis-toxin insensitive (232–234), it is clear that the S1P/G-protein receptor pathway is not involved in FcεRIdependent calcium mobilization. However, S1P could have a dual role and could act as a second messenger also (218). Second, by analyzing multiple RBL cells in single cell calcium mobilization studies, it became clear (E Donnadieu, J-P Kinet, unpublished) that the m1 receptor-mediated calcium signal was synchronous while the FcεRI-mediated calcium response was very asynchronous. This difference of synchronicity using the same clonal cell line was unexpected and could account, at least in part, for the IP3 gap described above. Third, the high affinity IgG receptor Fcγ RI, which is signaling by using FcRγ , is also capable of activating SK and producing S1P, and this S1P production appears to be linked to calcium mobilization as in the case of FcεRI. However, the activation of SK is the result of the activation of phospholipase D, itself being activated through tyrosine kinases (235). Fourth, Reinhold Penner’s group has recently demonstrated that sphingosine, unlike S1P, is a very potent inhibitor of CRAC current (ICRAC) (236). At the end of this paper, the authors speculate that “Sphingosine might be involved in the activation mechanism of ICRAC. In the resting state, sphingosine could act as a blocker of ICRAC. Upon depletion of internal stores, metabolism of sphingosine, possibly by conversion to sphingosine-1-phosphate by sphingosine kinase, could lower the sphingosine levels and lead to the disinhibition of ICRAC.” At present, one is left with a number of data that support the notion that sphingosine kinase is involved in calcium mobilization. Sphingosine kinase is activated and S1P is produced by two FcRγ -containing receptors, FcεRI and Fcγ RI. However, the effect of this activation on later events such as calcium mobilization is so far entirely based on studies with inhibitors (with the exception of Reinhold Penner’s data) and is therefore subject to unsuspected problems or artifacts. Because the activation of ICRAC is so intimately linked to the release of special stores, a mechanism which is itself not completely understood, it is difficult to discriminate between the possible effects of SK activation on the stores and on ICRAC. For these reasons, it is likely that a better understanding of the role of SK in FcεRI signaling will await molecular characterization of SK that will permit genetic mapping of the system.

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

959

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Other Signaling Molecules What is the role of the serine/threonine phosphorylation of FcεRI? Engagement of FcεRI induces the serine phosphorylation of FcRβ and threonine phosphorylation of FcRγ (145). The phosphorylated residues are in the ITAMs of FcRβ and FcRγ (237). The threonine phosphorylation of FcRγ appears to be mediated by protein kinase C-δ (PKC-δ) and to correlate with receptor endocytosis (238). Other PKC isoforms are also activated through FcεRI. PKC-β and PKC-ε seem to play a role in the expression of c-fos and c-jun (238a). PKC-α and PKC-ε act as negative regulators of phospholipase A2 (PLA2) activity, and PKC-β as a positive modulator of secretion, PLA2 activity, and cytokine production (239). How these pathways are connected to the tyrosine kinase pathways is at present unclear. What is the role of phosphatases? The possible role of CD45 has already been discussed. Other phosphatases that are probably implicated are SHP-1, SHP-2, and SHIP (175, 177), but their precise role remains undefined. They could play a role in balancing the activation of the tyrosine kinases and/or be operating in the rapid uncoupling of FcεRI complexes following receptor disengagement with haptens. What is known, however, is that the phosphatase activity responsible for receptor disengagement is not cytoplasmic but is associated with the plasma membrane (240). A number of studies implicate various adapter molecules in FcεRI-mediated pathways whose phosphorylation depends on Syk: (a) “shc” which may be involved in the activation of mitogen-activated protein kinase (MAPK) (171) (241), (b) “vav” (163, 242), which seems to associate with Grb2, Raf-1, and MAPK (243) and may be involved in Rac1-dependent JNK activation (241), (c) “SLP-76” (244) is also tyrosine phosphorylated by Syk (245) and is implicated in the activation of PLC-γ (246). Its gene targeting demonstrates that it is essential for T cell development and function (247, 248), and (d ) “cbl” plays also a role, but it remains to be defined (249). Many more signaling molecules have been described to be involved in FcεRI signaling pathways, but their precise molecular roles and functions remain unclear. Because of space limitation and for the purpose of clarity, this review does not include all papers published on all molecules involved in FcεRI and other immunoreceptors pathways. Of the large and complex puzzle of the FcεRIdependent pathways, only a few very small pieces have started to fit together. Our future task is to find a place for each of these molecules and to integrate them in the complex puzzle of FcεRI biological function.

Updated Model of FcεRI Signaling Upon clustering of receptor-bound IgE by multivalent antigens, Lyn which is bound to FcRβ prior to receptor engagement, transphosphorylates adjacent

P1: SAT/VKS/MBG

P2: KKK/PLB

January 25, 1999

11:35

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

960

QC: PKS

Annual Reviews

AR078-28

KINET

receptors on FcRβ and FcRγ . This tyrosine phosphorylation of FcRβ ITAM recruits more Lyn, and the transphosphorylation of more adjacent receptors in the cluster goes on. The two phosphorylated tyrosines of FcRγ ITAM serve as the membrane targeting signal for the tandem-SH2 of Syk, which is now in the proximity of FcRβ-attached Lyn. Lyn phosphorylates Syk on tyrosine-518, and this results in Syk autophosphorylation at other sites and activation of other adjacent Syk molecules. In FcεRI αγ 2 complexes where FcRβ is not present, the membrane-attached Lyn operates the same function of phosphorylating FcRγ and Syk but at a much lower level. Syk then phosphorylates a number of substrates (shc, vav, SLP-76, cbl ...) and activates PI3 kinase. PI3 kinase generates PdtIns-3,4,5-P3 in the vicinity of the cluster. Then PdtIns-3,4,5-P3 becomes the membrane targeting signal for the PH domain of Tec kinases. This allows Lyn to phosphorylate tyrosine-551 in the activation loop of Btk (or other Tec kinases). Membrane attached Tec kinases autophosphorylate at other sites and become fully active. Tec kinases serve at least two functions: the function of adapter molecules via SH2 domains (additional adapters are most likely necessary) and the kinase function. As a result, PLC-γ is membrane targeted and becomes phosphorylated by both Tec kinases and Syk. IP3 is generated and the CRAC controlling stores release intracellular calcium and activate ICRAC. This store-operated function and the opening of ICRAC may involve additional controls (sphingosine and S1P?). Visit the Annual Reviews home page at http://www.AnnualReviews.org

See Addendum following, page 973: The Crystal Structure of the High-Affinity IgE Receptor (FcεRIα). Literature Cited 1. Metzger H, Alcarez G, Hohman R, Kinet JP, Pribluda V, Quarto R. 1986. The receptor with high affinity for immunoglobulin E. Annu. Rev. Immunol. 4:419–70 2. Ravetch JV, Kinet JP. 1991. Fc receptors. Annu. Rev. Immunol. 9:457–92 3. Ra C, Jouvin MH, Blank U, Kinet JP. 1989. A macrophage Fc gamma receptor and the mast cell receptor for IgE share an identical subunit. Nature 341:752–54 4. Hibbs ML, Selvaraj P, Carpen O, Springer TA, Kuster H, Jouvin MH, Kinet JP. 1989. Mechanisms for regulating expression of membrane isoforms of Fc gamma RIII (CD16). Science 246:1608– 11 5. Kurosaki T, Ravetch JV. 1989. A sin-

gle amino acid in the glycosyl phosphatidylinositol attachment domain determines the membrane topology of Fc gamma RIII [published erratum appears in Nature 1990, Jan. 25; 343 (6256):390]. Nature 342:805–7 6. Orloff DG, Ra CS, Frank SJ, Klausner RD, Kinet JP. 1990. Family of disulphide-linked dimers containing the zeta and eta chains of the T-cell receptor and the gamma chain of Fc receptors. Nature 347:189–91 7. Blank U, Ra C, Miller L, White K, Metzger H, Kinet JP. 1989. Complete structure and expression in transfected cells of high affinity IgE receptor. Nature 337:187–89 8. Ra C, Jouvin MH, Kinet JP. 1989.

P1: SAT/VKS/MBG

P2: KKK/PLB

January 25, 1999

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

9.

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

Complete structure of the mouse mast cell receptor for IgE (Fc epsilon RI) and surface expression of chimeric receptors (rat-mouse-human) on transfected cells. J. Biol. Chem. 264:15323–27 Kinet JP, Alcaraz G, Leonard A, Wank S, Metzger H. 1985. Dissociation of the receptor for immunoglobulin E in mild detergents. Biochemistry 24:4117–24 Kinet JP, Quarto R, Perez-Montfort R, Metzger H. 1985. Noncovalently and covalently bound lipid on the receptor for immunoglobulin E. Biochemistry 24:7342–48 Kinet JP, Metzger H, Hakimi J, Kochan J. 1987. A cDNA presumptively coding for the alpha subunit of the receptor with high affinity for immunoglobulin E [published erratum appears in Biochemistry 1988 Nov. 15; 27(23):8694]. Biochemistry 26:4605–10 Shimizu A, Tepler I, Benfey PN, Berenstein EH, Siraganian RP, Leder P. 1988. Human and rat mast cell high-affinity immunoglobulin E receptors: characterization of putative alpha-chain gene products. Proc. Natl. Acad. Sci. USA 85:1907–11 Kochan J, Pettine LF, Hakimi J, Kishi K, Kinet JP. 1988. Isolation of the gene coding for the alpha subunit of the human high affinity IgE receptor. Nucleic Acids Res. 16:3584 Letourneur O, Sechi S, Willette-Brown J, Robertson MW, Kinet JP. 1995. Glycosylation of human truncated Fc epsilon RI alpha chain is necessary for efficient folding in the endoplasmic reticulum. J. Biol. Chem. 270:8249–56 Kinet JP, Blank U, Ra C, White K, Metzger H, Kochan J. 1988. Isolation and characterization of cDNAs coding for the beta subunit of the high-affinity receptor for immunoglobulin E. Proc. Natl. Acad. Sci. USA 85:6483–7 Kurosaki T, Gander I, Wirthmueller U, Ravetch JV. 1992. The beta subunit of the Fc epsilon RI is associated with the Fc gamma RIII on mast cells. J. Exp. Med. 175:447–51 Lanier LL, Yu G, Phillips JH. 1989. Coassociation of CD3zeta with a receptor (CD16) for IgG Fc on human natural killer cells. Nature 342:803–5 Anderson P, Caligiuri M, Ritz J, Schlossman SF. 1989. CD3-negative natural killer cells express zeta TCR as part of a novel molecular complex. Nature 341:159–62 Anderson P, Caligiuri M, O’Brien C, Manley T, Ritz J, Schlossman SF. 1990.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

961

Fc gamma receptor type III (CD16) is included in the zeta NK receptor complex expressed by human natural killer cells. Proc. Natl. Acad. Sci. USA 87:2274–78 Kinet JP. 1992. The gamma-zeta dimers of Fc receptors as connectors to signal transduction. Curr. Opin. Immunol. 4:43–48 Ernst LK, Duchemin AM, Anderson CL. 1993. Association of the high-affinity receptor for IgG (Fc gamma RI) with the gamma subunit of the IgE receptor. Proc. Natl. Acad. Sci. USA 90:6023–27 Scholl PR, Geha RS. 1993. Physical association between the high-affinity IgG receptor (Fc gamma RI) and the gamma subunit of the high-affinity IgE receptor (Fc epsilon RI gamma). Proc. Natl. Acad. Sci. USA 90:8847–50 Pfefferkorn LC, Yeaman GR. 1994. Association of IgA-Fc receptors (Fc alpha R) with Fc epsilon RI gamma 2 subunits in U937 cells. Aggregation induces the tyrosine phosphorylation of gamma 2. J. Immunol. 153:3228–36 Morton HC, van den Herik-Oudijk IE, Vossebeld PA, Snijders A, Verhoeven AJ, Capel PJ, van de Winkel JG. 1995. Functional association between the human myeloid immunoglobulin A Fc receptor (CD89) and FcR gamma chain. Molecular basis for CD89/FcR gamma chain association. J. Biol. Chem. 270:29781–87 Gibbins J, Asselin J, Farndale R, Barnes M, Law CL, Watson SP. 1996. Tyrosine phosphorylation of the Fc receptor gamma-chain in collagen-stimulated platelets. J. Biol. Chem. 271:18095–99 Gibbins JM, Okuma M, Farndale R, Barnes M, Watson SP. 1997. Glycoprotein VI is the collagen receptor in platelets which underlies tyrosine phosphorylation of the Fc receptor gammachain. FEBS Lett. 413:255–59 Tsuji M, Ezumi Y, Arai M, Takayama H. 1997. A novel association of Fc receptor gamma-chain with glycoprotein VI and their co-expression as a collagen receptor in human platelets. J. Biol Chem. 272:23528–31 Letourneur F, Hennecke S, Demolliere C, Cosson P. 1995. Steric masking of a dilysine endoplasmic reticulum retention motif during assembly of the human high affinity receptor for immunoglobulin E. J. Cell Biol. 129:971–78 Miller L, Blank U, Metzger H, Kinet JP. 1989. Expression of high-affinity binding of human immunoglobulin E by transfected cells. Science 244:33437

P1: SAT/VKS/MBG

P2: KKK/PLB

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

962

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

30. Kuster H, Thompson H, Kinet JP. 1990. Characterization and expression of the gene for the human Fc receptor gamma subunit. Definition of a new gene family. J. Biol. Chem. 265:6448–52 31. Wang B, Rieger A, Kilgus O, Ochiai K, Maurer D, Fodinger D, Kinet JP, Stingl G. 1992. Epidermal Langerhans cells from normal human skin bind monomeric IgE via Fc epsilon RI. J. Exp. Med. 175:1353–65 32. Bieber T, de la Salle H, Wollenberg A, Hakimi J, Chizzonite R, Ring J, Hanau D, de la Salle C. 1992. Human epidermal Langerhans cells express the high affinity receptor for immunoglobulin E (Fc epsilon RI). J. Exp. Med. 175:1285–90 33. Gounni AS, Lamkhioued B, Ochiai K, Tanaka Y, Delaporte E, Capron A, Kinet JP, Capron M. 1994. High-affinity IgE receptor on eosinophils is involved in defence against parasites. Nature 367:183– 6 34. Maurer D, Fiebiger E, Reininger B, Wolff-Winiski B, Jouvin MH, Kilgus O, Kinet JP, Stingl G. 1994. Expression of functional high affinity immunoglobulin E receptors (Fc epsilon RI) on monocytes of atopic individuals. J. Exp. Med. 179:745–50 35. Maurer D, Fiebiger S, Ebner C, Reininger B, Fischer GF, Wichlas S, Jouvin MH, Schmitt-Egenolf M, Kraft D, Kinet JP, Stingl G. 1996. Peripheral blood dendritic cells express Fc epsilon RI as a complex composed of Fc epsilon RI alpha- and Fc epsilon RI gammachains and can use this receptor for IgEmediated allergen presentation. J. Immunol. 157:607–16 36. Joseph M, Gounni AS, Kusnierz JP, Vorng H, Sarfati M, Kinet JP, Tonnel AB, Capron A, Capron M. 1997. Expression and functions of the high-affinity IgE receptor on human platelets and megakaryocyte precursors. Eur. J. Immunol. 27:2212–18 37. Tanaka Y, Takenaka M, Matsunaga Y, Okada S, Anan S, Yoshida H, Ra C. 1995. High affinity IgE receptor (Fc epsilon RI) expression on eosinophils infiltrating the lesions and mite patchtested sites in atopic dermatitis. Arch. Dermatol. Res. 287:712–17 38. Humbert M, Grant JA, Taborda-Barata L, Durham SR, Pfister R, G Menz, Barkans J, Ying S, Kay AB. 1996. Highaffinity IgE receptor (Fc epsilon RI)bearing cells in bronchial biopsies from atopic and nonatopic asthma. Am. J. Respir. Crit. Care Med. 153:1931–37

39. Sihra BS, Kon OM, Grant JA, Kay AB. 1997. Expression of high-affinity IgE receptors (Fc epsilon RI) on peripheral blood basophils, monocytes, and eosinophils in atopic and nonatopic subjects: relationship to total serum IgE concentrations. J. Allergy Clin. Immunol. 99:699–706 40. Rajakulasingam K, Durham SR, O’Brien F, Humbert M, Barata LT, Reece L, Kay AB, Grant JA. 1997. Enhanced expression of high-affinity IgE receptor (Fc epsilon RI) alpha chain in human allergen-induced rhinitis with co-localization to mast cells, macrophages, eosinophils, and dendritic cells. J. Allergy Clin. Immunol. 100:78–86 41. Ying S, Barata LT, Meng Q, Grant JA, Barkans J, Durham SR, Kay AB. 1998. High-affinity immunoglobulin E receptor (Fc epsilon RI)-bearing eosinophils, mast cells, macrophages and Langerhans’ cells in allergen-induced latephase cutaneous reactions in atopic subjects. Immunology 93:281–88 42. Riske F, Hakimi J, Mallamaci M, Griffin M, Pilson B, Tobkes N, Lin P, Danho W, Kochan J, Chizzonite R. 1991. High affinity human IgE receptor (Fc epsilon RI). Analysis of functional domains of the alpha-subunit with monoclonal antibodies. J. Biol. Chem. 266:11245– 51 43. Tepler I, Shimizu A, Leder P. 1989. The gene for the rat mast cell high affinity IgE receptor alpha chain. Structure and alternative mRNA splicing patterns. J. Biol. Chem. 264:5912–15 44. Ye ZS, Kinet JP, Paul WE. 1992. Structure of the gene for the alpha-chain of the mouse high affinity receptor for IgE (Fc epsilon RI). J. Immunol. 49:897–900 45. Pang J, Taylor GR, Munroe DG, Ishaque A, Fung-Leung WP, Lau CY, Liu FT, Zhou L. 1993. Characterization of the gene for the human high affinity IgE receptor (Fc epsilon RI) alpha-chain. J. Immunol. 151:6166–74 46. Dombrowicz D, Brini AT, Flamand V, Hicks E, Snouwaert JN, Kinet JP, Koller BH. 1996. Anaphylaxis mediated through a humanized high affinity IgE receptor. J. Immunol. 157:1645–51 47. Dombrowicz D, Lin S, Flamand V, Brini AT, Koller BH, Kinet JP. 1998. Allergyassociated FcR beta is a molecular amplifier of IgE- and IgG-mediated in vivo responses. Immunity 8:517–29 48. Fung-Leung WP, De Sousa-Hitzler J, Ishaque A, Zhou L, Pang J, Ngo K, Panakos JA, Chourmouzis E, Liu FT,

P1: SAT/VKS/MBG

P2: KKK/PLB

January 25, 1999

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

49.

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

50.

51.

52.

53. 54.

55.

56.

57.

58.

59.

Lau CY. 1996. Transgenic mice expressing the human high-affinity immunoglobulin (Ig) E receptor alpha chain respond to human IgE in mast cell degranulation and in allergic reactions. J. Exp. Med. 183:49–56 Lin S, Cicala C, Scharenberg AM, Kinet JP. 1996. The Fc(epsilon)RI beta subunit functions as an amplifier of Fc(epsilon)RI gamma-mediated cell activation signals. Cell 85:985–95 Hakimi J, Seals C, Kondas JA, Pettine L, Danho W, Kochan J. 1990. The alpha subunit of the human IgE receptor (FcERI) is sufficient for high affinity IgE binding. J. Biol. Chem. 265:22079–81 Blank U, Ra CS, Kinet JP. 1991. Characterization of truncated alpha chain products from human, rat, and mouse high affinity receptor for immunoglobulin E. J. Biol. Chem. 266:2639–46 Alber G, Miller L, Jelsema CL, VarinBlank N, Metzger H. 1991. Structurefunction relationships in the mast cell high affinity receptor for IgE. Role of the cytoplasmic domains and of the beta subunit. J. Biol. Chem. 266:22613–20 Sutton BJ, Gould HJ. 1993. The human IgE network. Nature 366:421–28 Helm B, Marsh P, Vercelli D, Padlan E, Gould HJ, Geha R. 1988. The mast cell binding site on human immunoglobulin E. Nature 331:180–83 Helm B, Kebo D, Vercelli D, Glovsky MM, Gould H, Ishizaka K, Geha R, Ishizaka T. 1989. Blocking of passive sensitization of human mast cells and basophil granulocytes with IgE antibodies by a recombinant human epsilon-chain fragment of 76 amino acids. Proc. Natl. Acad. Sci. USA 86:9465–69 Schwarzbaum S, Nissim A, Alkalay I, Ghozi MC, Schindler DG, Bergman Y, Eshhar Z. 1989. Mapping of murine IgE epitopes involved in IgE-Fcε receptor interactions. Eur. J. Immunol. 19:1015–23 Weetall M, Shopes B, Holowka D, Baird B. 1990. Mapping the site of interaction between murine IgE and its high affinity receptor with chimeric Ig. J. Immunol. 145:3849–54 Nissim A, Jouvin M-H, Eshhar Z. 1991. Mapping of the high affinity Fcε receptor binding site to the third constant region domain of IgE. EMBO J. 10:101–7 Basu M, Hakimi J, Dharm E, Kondas JA, Tsien W-H, Pilson RS, Lin P, Gilfillan A, Haring P, Braswell EH, Nettleton MY, Kochan JP. 1993. Purification and characterization of human recombinant IgE-Fc fragments that bind to the human

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

963

high affinity IgE receptor. J. Biol. Chem. 268:13118–27 Nissim A, Schwarzbaum S, Siraganian R, Eshhar Z. 1993. Fine specificity of the IgE interaction with the low and high affinity Fc receptor. J. Immunol. 150:1365–74 Presta L, Shields R, O’Connell L, Lahr S, Porter J, Gorman C, Jardieu P. 1994. The binding site on human immunoglobulin E for its high affinity receptor. J. Biol. Chem. 269:26368–73 Helm BA, Sayers I, Higginbottom A, Machado DC, Ling Y, Ahmad K, Padlan EA, Wilson AP. 1996. Identification of the high affinity receptor binding region in human immunoglobulin E. J. Biol. Chem. 271:7494–7500 Keown MB, Henry AJ, Ghirlando R, Sutton BJ, Gould HJ. 1998. Thermodynamics of the interaction of human immunoglobulin E with its high-affinity receptor Fc epsilon RI. Biochemistry 37:8863–69 Sechi S, Roller PP, Willette-Brown J, Kinet JP. 1996. A conformational rearrangement upon binding of IgE to its high affinity receptor. J. Biol. Chem. 271:19256–63 Williams AF, Davis SJ, He Q, Barclay AN. 1989. Structural diversity in domains of the immunoglobulin superfamily. Cold Spring Harbor Symp. Quant. Bio. 54:637 Padlan EA, Helm BA. 1992. A modeling study of the α-subunit of human highaffinity receptor for immunoglobulin E receptor. 2:129–43 Hogarth PM, Hulett MD, Ierino FL, Tate B, Powell MS, Brinkworth RI. 1992. Identification of the immunoglobulin binding regions (IBR) of Fcγ RII and FcεRI. Immunol. Rev. 125:21–35 Hulett MD, McKenzie IFC, Hogarth PM. 1993. Chimeric Fc receptors identify immunoglobulin-binding regions in human Fcγ RII and FcεRI. Eur. J. Immunol. 23:640–45 Mallamaci MA, Chizzonite R, Griffin M, Nettleton M, Hakimi J, Tsien W-H, Kochan JP. 1993. Identification of sites on the human FcεRIα subunit which are involved in binding human and rat IgE. J. Biol. Chem. 268:22076–83 Robertson MW. 1993. Phage and Escherichia coli expression of the human high affinity immunoglobulin E receptor α-subunit ectodomain. Domain localization of the IgE-binding site. J. Biol. Chem. 268:12736–43 Scarselli E, Esposito G, Traboni C. 1993.

P1: SAT/VKS/MBG

P2: KKK/PLB

January 25, 1999

964

72.

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

73.

74.

75.

76.

77.

78.

79.

80.

81.

11:35

QC: PKS

Annual Reviews

AR078-28

KINET Receptor phage: display of functional domains of the high affinity IgE receptoron the M13 phage surface. FEBS Lett. 329:223–26 McDonnell JM, Beavil AJ, Mackay GA, Jameson BA, Korngold R, Gould HJ, Sutton BJ. 1996. Structure based design and characterization of peptides that inhibit IgE binding to its high-affinity receptor. Nature Struc. Biol. 3:419–26 Cook JP, Henry AJ, McDonnell JM, Owens RJ, Sutton BJ, Gould HJ. 1997. Identification of contact residues in the IgE binding site of human Fc epsilon RI alpha. Biochemistry 36:15579–88 Dombrowicz D, Flamand V, Brigman KK, Koller BH, Kinet JP. 1993. Abolition of anaphylaxis by targeted disruption of the high affinity immunoglobulin E receptor alpha chain gene. Cell 75:969–76 Jankovic D, Kullberg MC, Dombrowicz D, Barbieri S, Caspar P, Wynn TA, Paul WE, Cheever AW, Kinet JP, Sher A. 1997. Fc epsilonRI-deficient mice infected with Schistosoma mansoni mount normal Th2-type responses while displaying enhanced liver pathology. J. Immunol. 159:1868–75 Matsuda H, Watanabe N, Kiso Y, Hirota S, Ushio H, Kannan Y, Azuma M, Koyama H, Kitamura Y. 1990. Necessity of IgE antibodies and mast cells for manifestation of resistance against larval Haemaphysalis longicornis ticks in mice. J. Immunol. 144:259–62 Matsuda H, Fukui K, Kiso Y, Kitamura Y. 1985. Inability of genetically mast cell-deficient W/Wv mice to acquire resistance against larval Haemaphysalis longicornis ticks. J. Parasitol. 71:443– 48 Matsuda H, Nakano T, Kiso Y, Kitamura Y. 1987. Normalization of anti-tick response of mast cell-deficient W/Wv mice by intracutaneous injection of cultured mast cells. J. Parasitol. 73:155–60 Maurer D, Ebner C, Reininger B, Fiebiger E, Kraft D, Kinet JP, Stingl G. 1995. The high affinity IgE receptor (Fc epsilon RI) mediates IgEdependent allergen presentation. J. Immunol. 154:6285–90 Gounni AS, Lamkhioued B, Delaporte E, Dubost A, Kinet JP, Capron A, Capron M. 1994. The high-affinity IgE receptor on eosinophils: from allergy to parasites or from parasites to allergy? J. Allergy Clin. Immunol. 94:1214–16 Pritchard DI, Hewitt C, Moqbel R. 1997. The relationship between immunolog-

82.

83.

84.

85. 86.

87.

88.

89.

90.

91.

ical responsiveness controlled by Thelper 2 lymphocytes and infections with parasitic helminths. Parasitology 115:S33–44 Dunne DW, Butterworth AE, Fulford AJ, Kariuki HC, Langley JG, Ouma JH, Capron A, Pierce RJ, Sturrock RF. 1992. Immunity after treatment of human schistosomiasis: association between IgE antibodies to adult worm antigens and resistance to reinfection. Eur. J. Immunol. 22:1483–94 Hagan P, Blumenthal UJ, Dunn D, Simpson AJ, Wilkins HA. 1991. Human IgE, IgG4 and resistance to reinfection with Schistosoma haematobium. Nature 349:243–45 Rihet P, Demeure CE, Bourgois A, Prata A, Dessein AJ. 1991. Evidence for an association between human resistance to Schistosoma mansoni and high anti-larval IgE levels. Eur. J. Immunol. 21:2679–86 Allen JE, Maizels RM. 1996. Immunology of human helminth infection. Int. Arch. Allergy Immunol. 109:3–10 ISAAC Steering Committee. 1998. Worldwide variation in prevalence of symptoms of asthma, allergic rhinoconjunctivitis, and atopic eczema: ISAAC. The International Study of Asthma and Allergies in Childhood (ISAAC) Steering Committee [see comments]. Lancet 351:1225–32 Malveaux FJ, Conroy MC, Adkinson NF Jr, Lichtenstein LM. 1978. IgE receptors on human basophils. Relationship to serum IgE concentration. J. Clin. Invest. 62:176–81 Furuichi K, Rivera J, Isersky C. 1985. The receptor for immunoglobulin E on rat basophilic leukemia cells: effect of ligand binding on receptor expression. Proc. Natl. Acad. Sci. USA 82:1522– 25 Quarto R, Kinet JP, Metzger H. 1985. Coordinate synthesis and degradation of the alpha-, beta- and gamma-subunits of the receptor for immunoglobulin E. Mol. Immunol. 22:1045–51 Yamaguchi M, Lantz CS, Oettgen HC, Katona IM, Fleming T, Miyajima I, Kinet JP, Galli SJ. 1997. IgE enhances mouse mast cell Fc(epsilon)RI expression in vitro and in vivo: evidence for a novel amplification mechanism in IgE-dependent reactions. J. Exp. Med. 185:663–72 Lantz CS, Yamaguchi M, Oettgen HC, Katona IM, Miyajima I, Kinet JP, Galli SJ. 1997. IgE regulates mouse basophil

P1: SAT/VKS/MBG

P2: KKK/PLB

January 25, 1999

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

92.

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

93.

94.

95.

96.

97.

98.

99.

100.

101.

Fc epsilon RI expression in vivo. J. Immunol. 158:2517–21 Yano K, Yamaguchi M, de Mora F, Lantz CS, Butterfield JH, Costa JJ, Galli SJ. 1997. Production of macrophage inflammatory protein-1 alpha by human mast cells: increased anti-IgE-dependent secretion after IgE-dependent enhancement of mast cell IgE-binding ability. Lab. Invest. 77:185–93 MacGlashan D Jr, McKenzie-White J, Chichester K, Bochner BS, Davis FM, Schroeder JT, Lichtenstein LM. 1998. In vitro regulation of Fc epsilon RI alpha expression on human basophils by IgE antibody. Blood 91:1633–43 MacGlashan DW Jr, Bochner BS, Adelman DC, Jardieu PM, Togias A, McKenzie-White J, Sterbinsky SA, Hamilton RG, Lichtenstein LM. 1997. Down-regulation of Fc(epsilon)RI expression on human basophils during in vivo treatment of atopic patients with anti-IgE antibody. J. Immunol. 158: 1438–45 Rottem M, Barbieri S, Kinet JP, Metcalfe DD. 1992. Kinetics of the appearance of Fc epsilon RI-bearing cells in interleukin-3-dependent mouse bone marrow cultures: correlation with histamine content and mast cell maturation. Blood 79:972–80 Toru H, Ra C, Nonoyama S, Suzuki K, Yata J, Nakahata T. 1996. Induction of the high-affinity IgE receptor (Fc epsilon RI) on human mast cells by IL-4. Int. Immunol. 8:1367–73 Xia HZ, Du Z, Craig S, Klisch G, NobenTrauth N, Kochan JP, Huff TH, Irani AM, Schwartz LB. 1997. Effect of recombinant human IL-4 on tryptase, chymase, and Fc epsilon receptor type I expression in recombinant human stem cell factor-dependent fetal liver-derived human mast cells. J. Immunol. 159:2911– 21 Brini AT, Lee GM, Kinet JP. 1993. Involvement of Alu sequences in the cellspecific regulation of transcription of the gamma chain of Fc and T cell receptors. J. Biol. Chem. 268:1355–61 Klausner RD, Lippincott-Schwartz J, Bonifacino JS. 1990. The T cell antigen receptor: insights into organelle biology. Annu. Rev. Cell Biol. 6:403–31 Ryan JJ, Kinzer CA, Paul WE. 1995. Mast cells lacking the high affinity immunoglobulin E receptor are deficient in Fc epsilon RI gamma messenger RNA. J. Exp. Med. 182:567–74 Kraft S, Wessendorf JH, Hanau D,

102.

103.

104. 105.

106.

107.

108.

109.

110.

111.

965

Bieber T. 1998. Regulation of the high affinity receptor for IgE on human epidermal Langerhans cells. J. Immunol. 161:1000–6 Cookson WO, Sharp PA, Faux JA, Hopkin JM. 1989. Linkage between immunoglobulin E responses underlying asthma and rhinitis and chromosome 11q. Lancet 1:1292–95 Young RP, Sharp PA, Lynch Jr, Faux JA, Lathrop GM, Cookson WO, Hopkin JM. 1992. Confirmation of genetic linkage between atopic IgE responses and chromosome 11q13. J. Med. Genet. 29:236– 38 Marsh DG, Meyers DA. 1992. A major gene for allergy–factor fancy? [news]. Nat. Genet. 2:252–54 Lympany P, Welsh K, MacCochrane G, Kemeny DM, Lee TH. 1992. Genetic analysis using DNA polymorphism of the linkage between chromosome 11q13 and atopy and bronchial hyperresponsiveness to methacholine. J. Allergy Clin. Immunol. 89:619–28 Amelung PJ, Panhuysen CI, Postma DS, Levitt RC, Koeter GH, Francomano CA, Bleecker ER, Meyers DA. 1992. Atopy and bronchial hyperresponsiveness: exclusion of linkage to markers on chromosomes 11q and 6p. Clin. Exp. Allergy 22:1077–84 Rich SS, Roitman-Johnson B, Greenberg B, Roberts S, Blumenthal MN. 1992. Genetic analysis of atopy in three large kindreds: no evidence of linkage to D11S97. Clin. Exp. Allergy 22:1070–76 Sandford AJ, Shirakawa T, Moffatt MF, Daniels SE, Ra C, Faux JA, Young RP, Nakamura Y, Lathrop GM, Cookson WO, et al. 1993. Localisation of atopy and beta subunit of high-affinity IgE receptor (Fc epsilon RI) on chromosome 11q [see comments]. Lancet 341:332– 34 Shirakawa T, Hashimoto T, Furuyama J, Takeshita T, Morimoto K. 1994. Linkage between severe atopy and chromosome 11q13 in Japanese families. Clin. Genet. 46:228–32 Brereton HM, Ruffin RE, Thompson PJ, Turner DR. 1994. Familial atopy in Australian pedigrees: adventitious linkage to chromosome 8 is not confirmed nor is there evidence of linkage to the high affinity IgE receptor. Clin. Exp. Allergy 24:868–77 Watson M, Lawrence S, Collins A, Beasley R, Doull I, Begishvili B, Lampe F, Holgate ST, Morton NE. 1995. Exclusion from proximal 11q of a common

P1: SAT/VKS/MBG

January 25, 1999

966

112.

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

113.

114.

115.

116.

117.

118.

119.

120.

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET gene with megaphenic effect on atopy. Ann. Hum. Genet. 59:403–11 van Herwerden L, Harrap SB, Wong ZY, Abramson MJ, Kutin JJ, Forbes AB, Raven J, Lanigan A, Walters EH. 1995. Linkage of high-affinity IgE receptor gene with bronchial hyperreactivity, even in absence of atopy [see comments]. Lancet 346:1262–65 Doull IJ, Lawrence S, Watson M, Begishvili T, Beasley RW, Lampe F, Holgate T, Morton NE. 1996. Allelic association of gene markers on chromosomes 5q and 11q with atopy and bronchial-hyperresponsiveness. Am. J. Respir. Crit. Care Med. 153:1280–84 Wong ZY, Tsonis D, van Herwerden L, Raven J, Forbes A, Abramson MJ, Walters EH, Harrap SB. 1997. Linkage analysis of bronchial hyperreactivity and atopy with chromosome 11q13. Electrophoresis 18:1641–45 Thomas NS, Wilkinson J, Holgate ST. 1997. The candidate region approach to the genetics of asthma and allergy. Am. J. Respir. Crit. Care Med. 156:S144– 51 Folster-Holst R, Moises HW, Yang L, Fritsch W, Weissenbach J, Christophers E. 1998. Linkage between atopy and the IgE high-affinity receptor gene at 11q13 in atopic dermatitis families. Hum. Genet. 102:236–39 Amelung PJ, Postma DS, Xu J, Meyers DA, Bleecker ER. 1998. Exclusion of chromosome 11q and the Fc epsilon RI-beta gene as aetiological factors in allergy and asthma in a population of Dutch asthmatic families. Clin. Exp. Allergy 28:397–403. In press Kuster H, Zhang L, Brini AT, MacGlashan DW, JP Kinet. 1992. The gene and cDNA for the human high affinity immunoglobulin E receptor beta chain and expression of the complete human receptor. J. Biol. Chem. 267:12782–87 Hill MR, Cookson WO. 1996. A new variant of the beta subunit of the highaffinity receptor for immunoglobulin E (Fc epsilon RI-beta E237G): associations with measures of atopy and bronchial hyper-responsiveness. Hum. Mol. Genet. 5:959–62 Hill MR, James AL, Faux JA, Ryan G, Hopkin JM, le Souef P, Musk AW, Cookson WO. 1995. Fc epsilon RI-beta polymorphism and risk of atopy in a general population sample [see comments] [published erratum appears in Br. Med. J. 1995 Nov 4; 311(7014):1196]. Br. Med. J. 311:776–79

121. Shirakawa T, Li A, Dubowitz M, Dekker JW, Shaw AE, Faux JA, Ra C, Cookson WO, Hopkin JM. 1994. Association between atopy and variants of the beta subunit of the high-affinity immunoglobulin E receptor [see comments]. Nat. Genet. 7:125–29 122. Hijazi Z, Haider MZ, Khan MR, AlDowaisan AA. 1998. High frequency of IgE receptor Fc epsilon RI beta variant (Leu181/Leu183) in Kuwaiti Arabs and its association with asthma. Clin. Genet. 53:149–52. In press 123. Palmer LJ, Pare PD, Faux JA, Moffatt MF, Daniels SE, LeSouef PN, Bremner PR, Mockford E, Gracey M, Spargo R, Musk AW, Cookson WO. 1997. Fc epsilon R1-beta polymorphism and total serum IgE levels in endemically parasitized Australian aborigines. Am. J. Hum. Genet. 61:182–88 124. Cox HE, Moffatt MF, Faux JA, Walley AJ, Coleman R, Trembath RC, Cookson WO, Harper JI. 1998. Association of atopic dermatitis to the beta subunit of the high affinity immunoglobulin E receptor. Br. J. Dermatol. 138:182– 87 125. Hide M, Francis DM, Grattan CE, Hakimi J, Kochan JP, Greaves MW. 1993. Autoantibodies against the highaffinity IgE receptor as a cause of histamine release in chronic urticaria [see comments]. N. Engl. J. Med. 328:1599– 1604 126. Gruber BL, Baeza ML, Marchese MJ, Agnello V, Kaplan AP. 1988. Prevalence and functional role of anti-IgE autoantibodies in urticarial syndromes. J. Invest. Dermatol. 90:213–17 127. Grattan CE, Francis DM, Hide M, Greaves MW. 1991. Detection of circulating histamine releasing autoantibodies with functional properties of anti-IgE in chronic urticaria. Clin. Exp. Allergy 21:695–704 128. Fiebiger E, Maurer D, Holub H, Reininger B, Hartmann G, Woisetschlager M, Kinet JP, Stingl G. 1995. Serum IgG autoantibodies directed against the alpha chain of Fc epsilon RI: a selective marker and pathogenetic factor for a distinct subset of chronic urticaria patients? J. Clin. Invest. 96:2606–12 129. Marone G, Casolaro V, Paganelli R, Quinti I. 1989. IgG anti-IgE from atopic dermatitis induces mediator release from basophils and mast cells. J. Invest. Dermatol. 93:246–52 130. Niimi N, Francis DM, Kermani F, O’Donnell BF, Hide M, Kobza-Black

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

131.

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

132.

133.

134. 135.

136.

137.

138.

139. 140.

141.

142. 143.

A, Winkelmann RK, Greaves MW, Barr RM. 1996. Dermal mast cell activation by autoantibodies against the high affinity IgE receptor in chronic urticaria. J. Invest. Dermatol. 106:1001–6 Tong LJ, Balakrishnan G, Kochan JP, Kinet JP, Kaplan AP. 1997. Assessment of autoimmunity in patients with chronic urticaria. J. Allergy Clin. Immunol. 99:461–65 Daeron M, Latour S, Malbec O, Espinosa E, Pina P, Pasmans S, Fridman WH. 1995. The same tyrosine-based inhibition motif, in the intracytoplasmic domain of Fc gamma RIIB, regulates negatively BCR-, TCR-, and FcR- dependent cell activation. Immunity 3:635– 46 Cambier JC. 1995. New nomenclature for the Reth motif (orARH1/ TAM/ARAM/YXXL) [letter]. Immunol. Today. 16:110 Reth M. 1989. Antigen receptor tail clue [letter]. Nature 338:383–84 Irving BA, Weiss A. 1991. The cytoplasmic domain of the T cell receptor zeta chain is sufficient to couple to receptorassociated signal transduction pathways. Cell 64:891–901 Romeo C, Seed B. 1991. Cellular immunity to HIV activated by CD4 fused to T cell or Fc receptor polypeptides. Cell 64:1037–46 Letourneur F, Klausner RD. 1991. T-cell and basophil activation through the cytoplasmic tail of T-cell-receptor zeta family proteins. Proc. Natl. Acad. Sci. USA 88:8905–9 Romeo C, Amiot M, Seed B. 1992. Sequence requirements for induction of cytolysis by the T cell antigen/Fc receptor zeta chain. Cell 68:889–97 Daeron M. 1997. Fc receptor biology. Annu. Rev. Immunol. 15:203–34 Holowka D, Baird B. 1996. Antigenmediated IGE receptor aggregation and signaling: a window on cell surface structure and dynamics. Annu. Rev. Biophys. Biomol. Struct. 25:79–112 Metzger H, Goldstein B, Kent U, Mao SY, Pribluda C, Pribluda V, Wofsy C, Yamashita T. 1994. Quantitative aspects of receptor aggregation. Adv. Exp. Med. Biol. 365:175–83 Metzger H. 1992. The receptor with high affinity for IgE. Immunol. Rev. 125:37– 48 Perez-Montfort R, Fewtrell C, Metzger H. 1983. Changes in the receptor for immunoglobulin E coincident with receptor-mediated stimulation of

144.

145.

146.

147.

148.

149.

150.

151.

152.

153.

967

basophilic leukemia cells. Biochemistry 22:5733–37 Quarto R, Metzger H. 1986. The receptor for immunoglobulin E: examination for kinase activity and as a substrate for kinases. Mol. Immunol. 23:1215–23 Paolini R, Jouvin MH, Kinet JP. 1991. Phosphorylation and dephosphorylation of the high-affinity receptor for immunoglobulin E immediately after receptor engagement and disengagement. Nature 353:855–58 Eiseman E, Bolen JB. 1992. Engagement of the high-affinity IgE receptor activates src protein-related tyrosine kinases. Nature 355:78–80 Benhamou M, Stephan V, Robbins KC, Siraganian RP. 1992. High-affinity IgE receptor-mediated stimulation of rat basophilic leukemia (RBL-2H3) cells induces early and late protein-tyrosine phosphorylations. J. Biol. Chem. 267: 7310–14 Stephan V, Benhamou M, Gutkind JS, Robbins KC, Siraganian RP. 1992. Fc epsilon RI-induced protein tyrosine phosphorylation of pp72 in rat basophilic leukemia cells (RBL-2H3). Evidence for a novel signal transduction pathway unrelated to G protein activation and phosphatidylinositol hydrolysis. J. Biol. Chem. 267:5434–41 Hutchcroft JE, Geahlen RL, Deanin GG, Oliver JM. 1992. Fc epsilon RImediated tyrosine phosphorylation and activation of the 72-kDa protein-tyrosine kinase, PTK72, in RBL-2H3 rat tumor mast cells. Proc. Natl. Acad. Sci. USA 89:9107–11 Pribluda VS, Metzger H. 1992. Transmembrane signaling by the high-affinity IgE receptor on membrane preparations. Proc. Natl. Acad. Sci. USA 89:11446–50 Benhamou M, Ryba NJ, Kihara H, Nishikata H, Siraganian RP. 1993. Protein-tyrosine kinase p72syk in high affinity IgE receptor signaling. Identification as a component of pp72 and association with the receptor gamma chain after receptor aggregation. J. Biol. Chem. 268:23318–24 Moran MF, Koch CA, Anderson D, Ellis C, England L, Martin GS, Pawson T. 1990. Src homology region 2 domains direct protein-protein interactions in signal transduction. Proc. Natl. Acad. Sci. USA 87:8622–26 Zhou S, Cantley LC. 1995. SH2 domain specificity determination using oriented phosphopeptide library. Meth. Enzymol. 254:523–35

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

968

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

154. Weiss A, Littman DR. 1994. Signal transduction by lymphocyte antigen receptors. Cell 76:263–74 155. Bu JY, Shaw AS, Chan AC. 1995. Analysis of the interaction of ZAP-70 and syk protein-tyrosine kinases with the T-cell antigen receptor by plasmon resonance. Proc. Natl. Acad. Sci. USA 92:5106–10 156. Cambier JC. 1995. Antigen and Fc receptor signaling. The awesome power of the immunoreceptor tyrosine-based activation motif (ITAM). J. Immunol. 155:3281–85 157. Chen T, Repetto B, Chizzonite R, Pullar C, Burghardt C, Dharm E, Zhao Z, Carroll R, Nunes P, Basu M, Danho W, Visnick M, Kochan J, Waugh D, Gilfillan AM. 1996. Interaction of phosphorylated Fc epsilon RI gamma immunoglobulin receptor tyrosine activation motif-based peptides with dual and single SH2 domains of p72syk. Assessment of binding parameters and real time binding kinetics. J. Biol. Chem. 271:25308–15 158. Kihara H, Siraganian RP. 1994. Src homology 2 domains of Syk and Lyn bind to tyrosine-phosphorylated subunits of the high affinity IgE receptor. J. Biol. Chem. 269:22427–32 159. Shiue L, Zoller MJ, Brugge JS. 1995. Syk is activated by phosphotyrosinecontaining peptides representing the tyrosine-based activation motifs of the high affinity receptor for IgE. J. Biol. Chem. 270:10498–502 160. Rivera VM, Brugge JS. 1995. Clustering of Syk is sufficient to induce tyrosine phosphorylation and release of allergic mediators from rat basophilic leukemia cells. Mol. Cell. Biol. 15:1582–90 161. Hirasawa N, Scharenberg A, Yamamura H, Beaven MA, Kinet JP. 1995. A requirement for Syk in the activation of the microtubule-associated protein kinase/phospholipase A2 pathway by Fc epsilon R1 is not shared by a G protein-coupled receptor. J. Biol. Chem. 270:10960–67 162. Scharenberg AM, Lin S, Cuenod B, Yamamura H, Kinet JP. 1995. Reconstitution of interactions between tyrosine kinases and the high affinity IgE receptor which are controlled by receptor clustering. Embo J. 14:3385–94 163. Oliver JM, Burg DL, Wilson BS, McLaughlin JL, Geahlen RL. 1994. Inhibition of mast cell Fc epsilon R1mediated signaling and effector function by the Syk-selective inhibitor, piceatannol. J. Biol. Chem. 269:29697–703

164. Valle A, Kinet JP. 1995. N-acetyl-Lcysteine inhibits antigen-mediated Syk, but not Lyn tyrosine kinase activation in mast cells. FEBS Lett. 357:41–44 165. Letourneur F, Klausner RD. 1992. Activation of T cells by a tyrosine kinase activation domain in the cytoplasmic tail of CD3 epsilon. Science 255:79–82 166. Adamczewski M, Paolini R, Kinet JP. 1992. Evidence for two distinct phosphorylation pathways activated by high affinity immunoglobulin E receptors. J. Biol. Chem. 267:18126–32 167. Eiseman E, Bolen JB. 1992. Signal transduction by the cytoplasmic domains of Fc epsilon RI-gamma and TCR-zeta in rat basophilic leukemia cells. J. Biol. Chem. 267:21027–32 168. Jouvin MH, Adamczewski M, Numerof R, Letourneur O, Valle A, Kinet JP. 1994. Differential control of the tyrosine kinases Lyn and Syk by the two signaling chains of the high affinity immunoglobulin E receptor. J. Biol. Chem. 269:5918– 25 169. Wilson BS, Kapp N, Lee RJ, Pfeiffer Jr, Martinez AM, Platt Y, Letourneur F, Oliver JM. 1995. Distinct functions of the Fc epsilon R1 gamma and beta subunits in the control of Fc epsilon R1mediated tyrosine kinase activation and signaling responses in RBL-2H3 mast cells. J. Biol. Chem. 270:4013–22 170. Shiue L, Green J, Green OM, Karas JL, Morgenstern JP, Ram MK, Taylor MK, Zoller MJ, Zydowsky LD, Bolen JB, et al. 1995. Interaction of p72syk with the gamma and beta subunits of the highaffinity receptor for immunoglobulin E, Fc epsilon RI. Mol. Cell. Biol. 15:272– 81 171. Jabril-Cuenod B, Zhang C, Scharenberg AM, Paolini R, Numerof R, Beaven MA, Kinet JP. 1996. Syk-dependent phosphorylation of Shc. A potential link between Fc epsilon RI and the Ras/mitogenactivated protein kinase signaling pathway through SOS and Grb2. J. Biol. Chem. 271:16268–72 172. Nishizumi H, Yamamoto T. 1997. Impaired tyrosine phosphorylation and Ca2+ mobilization, but not degranulation, in lyn-deficient bone marrow-derived mast cells. J. Immunol. 158:2350–55 173. Costello PS, Turner M, Walters AE, Cunningham CN, Bauer PH, Downward J, Tybulewicz VL. 1996. Critical role for the tyrosine kinase Syk in signalling through the high affinity IgE receptor of mast cells. Oncogene 13:2595–605 174. Zhang J, Berenstein EH, Evans RL,

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

175.

176.

177.

178.

179.

180.

181.

182.

183.

184.

Siraganian RP. 1996. Transfection of Syk protein tyrosine kinase reconstitutes high affinity IgE receptor-mediated degranulation in a Syk-negative variant of rat basophilic leukemia RBL-2H3 cells. J. Exp. Med. 184:71–9 Osborne MA, Zenner G, Lubinus M, Zhang X, Songyang Z, Cantley LC, Majerus P, Burn P, Kochan JP. 1996. The inositol 50 -phosphatase SHIP binds to immunoreceptor signaling motifs and responds to high affinity IgE receptor aggregation. J. Biol. Chem. 271:29271–78 Kimura T, Zhang J, Sagawa K, Sakaguchi K, Appella E, Siraganian RP. 1997. Syk-independent tyrosine phosphorylation and association of the protein tyrosine phosphatases SHP-1 and SHP-2 with the high affinity IgE receptor. J. Immunol. 159:4426–34 Kimura T, Sakamoto H, Appella E, Siraganian RP. 1997. The negative signaling molecule SH2 domain-containing inositol-polyphosphate 5-phosphatase (SHIP) binds to thetyrosine-phosphorylated beta subunit of the high affinity IgE receptor. J. Biol. Chem. 272: 13991–96 Ottinger EA, Botfield MC, Shoelson SE. 1998. Tandem SH2 domains confer high specificity in tyrosine kinase signaling. J. Biol. Chem. 273:729–35 Vonakis BM, Chen H, Haleem-Smith H, Metzger H. 1997. The unique domain as the site on Lyn kinase for its constitutive association with the high affinity receptor for IgE. J. Biol. Chem. 272:24072–80 Stauffer TP, Martenson CH, Rider JE, Kay BK, Meyer T. 1997. Inhibition of Lyn function in mast cell activation by SH3 domain binding peptides. Biochemistry 36:9388–94 El-Hillal O, Kurosaki T, Yamamura H, Kinet JP, Scharenberg AM. 1997. Syk kinase activation by a src kinase-initiated activation loop phosphorylation chain reaction. Proc. Natl. Acad. Sci. USA 94: 1919–24 Trowbridge IS, Thomas ML. 1994. CD45: an emerging role as a protein tyrosine phosphatase required for lymphocyte activation and development. Annu. Rev. Immunol. 12:85–116 Hook WA, Berenstein EH, Zinsser FU, Fischler C, Siraganian RP. 1991. Monoclonal antibodies to the leukocyte common antigen (CD45) inhibit IgEmediated histamine release from human basophils. J. Immunol. 147:2670–76 Adamczewski M, Numerof RP, Koretzky GA, Kinet JP. 1995. Regulation by

185.

186.

187.

188. 189.

190.

191. 192.

193.

194.

195.

969

CD45 of the tyrosine phosphorylation of high affinity IgE receptor beta- and gamma-chains. J. Immunol. 154:3047– 55 Pribluda VS, Pribluda C, Metzger H. 1994. Transphosphorylation as the mechanism by which the high-affinity receptor for IgE is phosphorylated upon aggregation. Proc. Natl. Acad. Sci. USA 91:11246–50 Yamashita T, Mao SY, Metzger H. 1994. Aggregation of the high-affinity IgE receptor and enhanced activity of p53/56 lyn protein-tyrosine kinase. Proc. Natl. Acad. Sci. USA 91:11251–55 Torigoe C, Goldstein B, Wofsy C, Metzger H. 1997. Shuttling of initiating kinase between discrete aggregates of the high affinity receptor for IgE regulates the cellular response. Proc. Natl. Acad. Sci. USA 94:1372–77 Torigoe C, Inman JK, Metzger H. 1998. An unusual mechanism for ligand antagonism. Science 281:568–72 Tsukada S, Saffran DC, Rawlings DJ, Parolini O, Allen RC, Klisak I, Sparkes RS, Kubagawa H, Mohandas T, Quan S, et al. 1993. Deficient expression of a B cell cytoplasmic tyrosine kinase in human X-linked agammaglobulinemia. Cell 72:279–90 Vihinen M, Vetrie D, Maniar HS, Ochs HD, Zhu Q, Vorechovsky I, Webster AD, Notarangelo LD, Nilsson L, Sowadski JM, et al. 1994. Structural basis for chromosome X-linked agammaglobulinemia: a tyrosine kinase disease. Proc. Natl. Acad. Sci. USA 91:12803–7 Desiderio S, Siliciano JD. 1994. The Itk/Btk/Tec family of protein-tyrosine kinases. Chem. Immunol. 59:191–210 Tsukada S, Rawlings DJ, Witte ON. 1994. Role of Bruton’s tyrosine kinase in immunodeficiency. Curr. Opin. Immunol. 6:623–30 Rawlings DJ, Witte ON. 1994. Bruton’s tyrosine kinase is a key regulator in B-cell development. Immunol. Rev. 138:105–19 Rawlings DJ, Saffran DC, Tsukada S, Largaespada DA, Grimaldi JC, Cohen L, Mohr RN, Bazan JF, Howard M, Copeland NG, et al. 1993. Mutation of unique region of Bruton’s tyrosinekinase in immunodeficient XID mice. Science 261:358–61 Thomas JD, Sideras P, Smith CI, Vorechovsky I, Chapman V, Paul WE. 1993. Colocalization of X-linked agammaglobulinemia and X-linked immunodeficiency genes. Science 261:355–68

P1: SAT/VKS/MBG

January 25, 1999

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

970

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET

196. Hata D, Kawakami Y, Inagaki N, Lantz CS, Kitamura T, Khan WN, MaedaYamamoto M, Miura T, Han W, Hartman SE, Yao L, Nagai H, Goldfeld AE, Alt FW, Galli SJ, Witte ON, Kawakami T. 1998. Involvement of Bruton’s tyrosine kinase in Fc epsilon RI-dependent mast cell degranulation and cytokine production. J. Exp. Med. 187:1235–47 197. Hata D, Kitaura J, Hartman SE, Kawakami Y, Yokota T, Kawakami T. 1998. Bruton’s tyrosine kinase-mediated interleukin-2 gene activation in mast cells. Dependence on the c-Jun Nterminal kinase activation pathway. J. Biol. Chem. 273:10979–87 198. Kawakami Y, Yao L, Miura T, Tsukada S, Witte ON, Kawakami T. 1994. Tyrosine phosphorylation and activation of Bruton tyrosine kinase upon Fc epsilon RI cross-linking. Mol. Cell. Biol. 14:5108–13 199. Takata M, Kurosaki T. 1996. A role for Bruton’s tyrosine kinase in B cell antigen receptor-mediated activation of phospholipase C-gamma 2. J. Exp. Med. 184:31–40 200. Rawlings DJ, Scharenberg AM, Park H, Wahl MI, Lin S, Kato RM, Fluckiger AC, Witte ON, Kinet JP. 1996. Activation of BTK by a phosphorylation mechanism initiated by SRC family kinases. Science 271:822–25 201. Salim K, Bottomley MJ, Querfurth E, Zvelebil MJ, Gout I, Scaife R, Margolis RL, Gigg R, Smith CI, Driscoll PC, Waterfield MD, Panayotou G. 1996. Distinct specificity in the recognition of phosphoinositides by the pleckstrin homology domains of dynamin and Bruton’s tyrosine kinase. Embo J. 15:6241– 50 202. Rameh LE, Arvidsson A, Carraway KL 3rd, Couvillon AD, Rathbun G, Crompton A, Van Renterghem B, Czech MP, Ravichandran KS, Burakoff SJ, Wang DS, Chen CS, Cantley LC. 1997. A comparative analysis of the phosphoinositide binding specificity of pleckstrin homology domains. J. Biol. Chem. 272:22059– 66 203. Carpenter CL, Cantley LC. 1996. Phosphoinositide kinases. Curr. Opin. Cell Biol. 8:153–58 204. Vanhaesebroeck B, Leevers SJ, Panayotou G, Waterfield MD. 1997. Phosphoinositide 3-kinases: a conserved family of signal transducers. Trends Biochem. Sci. 22:267–72 205. Yano H, Nakanishi S, Kimura K, Hanai N, Saitoh Y, Fukui Y, Nonomura Y,

206.

207.

208.

209.

210.

211.

212.

213.

214.

Matsuda Y. 1993. Inhibition of histamine secretion by wortmannin through the blockade of phosphatidylinositol3kinase in RBL 2H3 cells. J. Biol. Chem. 268:25846–56 Yano H, Agatsuma T, Nakanishi S, Saitoh Y, Fukui Y, Nonomura Y, Matsuda Y. 1995. Biochemical and pharmacological studies with KT7692 and LY294002 on the role of phosphatidylinositol 3-kinase in Fc epsilon RImediated signal transduction. Biochem J. 312:145–50 Barker SA, Caldwell KK, Hall A, Martinez AM, Pfeiffer Jr, Oliver JM, Wilson BS. 1995. Wortmannin blocks lipid and proteinkinase activities associated with PI 3-kinase and inhibits a subset of responses induced by Fc epsilon R1 crosslinking. Mol. Cell. Biol. 106:45–58 Bolland S, Pearse RN, Kurosaki T, Ravetch JV. 1998. SHIP modulates immune receptor responses by regulating membrane association of Btk. Immunity 8:509–16 Scharenberg AM, Kinet JP. 1998. PtdIns-3,4,5-P3: a regulatory nexus between tyrosine kinases and sustained calcium signals. Cell 94:5–8. In press Scharenberg AM, El-Hillal O, Fruman DA, Beitz LO, Li Z, Lin S, Gout I, Cantley LC, Rawlings DJ, Kinet JP. 1998. Phosphatidylinositol-3,4,5-trisphosphate (PtdIns-3,4,5-P3)/Teckinasedependent calcium signaling pathway: a target for SHIP-mediated inhibitory signals. Embo J. 17:1961–72 Fluckiger AC, Li Z, Kato RM, Wahl MI, Ochs HD, Longnecker R, Kinet JP, Witte ON, Scharenberg AM, Rawlings DJ. 1998. Btk/Teckinases regulate sustained increases in intracellular Ca2+ following B-cell receptor activation. Embo. J. 17:1973–85 Parekh AB, Fleig A, Penner R. 1997. The store-operated calcium current I(CRAC): nonlinear activation by InsP3 and dissociation from calcium release. Cell 89:973–80 Dolmetsch RE, Lewis RS, Goodnow CC, Healy JI. 1997. Differential activation of transcription factors induced by Ca2+ response amplitude and duration [see comments] [published erratum appears in Nature 1997 Jul. 17; 388 (6639):308]. Nature 386:855–58 Healy JI, Dolmetsch RE, Timmerman LA, Cyster JG, Thomas ML, Crabtree GR, Lewis RS, Goodnow CC. 1997. Different nuclear signals are activated by the B cell receptor during positive

P1: SAT/VKS/MBG

January 25, 1999

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

HIGH-AFFINITY IGE RECEPTOR

215.

216.

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

217.

218.

219.

220.

221. 222.

223.

224.

225.

versus negative signaling. Immunity 6: 419–28 Pleiman CM, Hertz WM, Cambier JC. 1994. Activation of phosphatidylinositol-30 kinase by Src-family kinase SH3 binding to the p85 subunit. Science 263:1609–12 Spiegel S, Merrill AH Jr. 1996. Sphingolipid metabolism and cell growth regulation. Faseb J. 10:1388–97 Spiegel S, Cuvillier O, Edsall LC, Kohama T, Menzeleev R, Olah Z, Olivera A, Pirianov G, Thomas DM, Tu Z, Van Brocklyn Jr, Wang F. 1998. Sphingosine-1-phosphate in cell growth and cell death. Ann. N. Y. Acad. Sci. 845:11–18 Van Brocklyn Jr, Lee MJ, Menzeleev R, Olivera A, Edsall L, Cuvillier O, Thomas DM, Coopman PJP, Thangada S, Liu CH, Hla T, Spiegel S. 1998. Dual actions of sphingosine-1-phosphate: extracellular through the Gi-coupled receptor edg-1 and intracellular to regulate proliferation and survival. J. Cell. Biol. 142:229–40 Olivera A, Spiegel S. 1993. Sphingosine-1-phosphate as second messenger in cell proliferation induced by PDGF and FCS mitogens. Nature 365:557–60 Olivera A, Kohama T, Tu Z, Milstien S, Spiegel S. 1998. Purification and characterization of rat kidney sphingosinekinase. J. Biol. Chem. 273:12576–83 Buehrer BM, Bell RM. 1993. Sphingosine kinase: properties and cellular functions. Adv. Lipid Res. 26:59–67 Ghosh TK, Bian J, Gill DL. 1994. Sphingosine 1-phosphate generated in the endoplasmic reticulum membrane activates release of stored calcium. J. Biol. Chem. 269:22628–35 Kim S, Lakhani V, Costa DJ, Sharara AI, Fitz JG, Huang LW, Peters KG, Kindman LA. 1995. Sphingolipid-gated Ca2+ release from intracellular stores of endothelial cells is mediated by a novel Ca(2+)-permeable channel. J. Biol. Chem. 270:5266–69 Mattie M, Brooker G, Spiegel S. 1994. Sphingosine-1-phosphate, a putative second messenger, mobilizes calcium from internal stores via an inositoltrisphosphate-independent pathway. J. Biol. Chem. 269:3181–88 Choi OH, Lee JH, Kassessinoff T, Cunha-Melo Jr, Jones SV, Beaven MA. 1993. Antigen and carbachol mobilize calcium by similar mechanisms in a transfected mast cell line (RBL-2H3 cells) that expresses ml muscarinic receptors. J. Immunol. 151:5586–95

971

226. Jones SV, Choi OH, Beaven MA. 1991. Carbachol induces secretion in a mast cell line (RBL-2H3) transfected with the ml muscarinic receptor gene. FEBS Lett. 289:47–50 227. Buehrer BM, Bell RM. 1992. Inhibition of sphingosine kinase in vitro and in platelets. Implications for signal transduction pathways. J. Biol. Chem. 267:3154–59 228. Choi OH, Kim JH, Kinet JP. 1996. Calcium mobilization viasphingosine kinase in signalling by the Fc epsilon RI antigen receptor. Nature 380:634–36 229. Wu J, Spiegel S, Sturgill TW. 1995. Sphingosine-1-phosphate rapidly activates the mitogen-activated protein kinase pathway by a G protein-dependent mechanism. J. Biol. Chem. 270:11484– 88 230. Goodemote KA, Mattie ME, Berger A, Spiegel S. 1995. Involvement of a pertussis toxin-sensitive G protein in the mitogenic signaling pathways of sphingosine 1-phosphate. J. Biol. Chem. 270:10272–77 231. Lee MJ, Van Brocklyn Jr, Thangada S, Liu CH, Hand AR, Menzeleev R, Spiegel S, Hla T. 1998. Sphingosine1-phosphate as a ligand for the G protein-coupled receptor EDG-1. Science 279:1552–55 232. Matthews G, Neher E, Penner R. 1989. Second messenger-activated calcium influx in rat peritoneal mast cells. J. Physiol. 418:105–30 233. Meyer T, Holowka D, Stryer L. 1988. Highly cooperative opening of calcium channels by inositol 1,4,5-trisphosphate. Science 240:653–56 234. Saito H, Ishizaka K, Ishizaka T. 1989. Effect of nonhydrolyzable guanosine phosphate on IgE-mediated activation of phospholipase C and histamine release from rodent mast cells. J. Immunol. 143:250–58 235. Melendez A, Floto RA, Gillooly DJ, Harnett MM, Allen JM. 1998. Fc gammaRI coupling to phospholipase D initiates sphingosine kinase-mediated calcium mobilization and vesicular trafficking. J. Biol. Chem. 273:9393–402 236. Mathes C, Fleig A, Penner R. 1998. Calcium-release-activated calcium current (ICRAC) is a direct -target for sphingosine. J. Biol. Chem. 273:25020–30 237. Pribluda VS, Pribluda C, Metzger H. 1997. Biochemical evidence that the phosphorylated tyrosines, serines, and threonines on the aggregated high affinity receptor for IgE are in the immunore-

P1: SAT/VKS/MBG

January 25, 1999

972

238.

238a.

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

239.

240.

241.

242.

243.

P2: KKK/PLB

11:35

QC: PKS

Annual Reviews

AR078-28

KINET ceptor tyrosine-based activation motifs. J. Biol. Chem. 272:11185–92 Germano P, Gomez J, Kazanietz MG, Blumberg PM, Rivera. J. 1994. Phosphorylation of the gamma chain of the high affinity receptor for immunoglobulin E by receptor-associated proteinkinase C-delta. J. Biol. Chem. 269:23102– 7 Razin E, Szallasi Z, Kazanietz MG, Blumberg PM, Rivera J. 1994. Proc. Natl. Acad. Sci. USA 91:7722–26 Chang EY, Szallasi Z, Acs P, Raizada V, Wolfe PC, Fewtrell C, Blumberg PM, Rivera J. 1997. Functional effects of overexpression of protein kinase Calpha, -beta, - delta, -epsilon, and -eta in the mast cell line RBL-2H3. J. Immunol. 159:2624–32 Mao SY, Metzger H. 1997. Characterization of protein-tyrosine phosphatases that dephosphorylate the high affinity IgE receptor [published erratum appears in J. Biol. Chem. 1997 Sep. 19; 272(38):24096]. J. Biol. Chem. 272: 14067–73 Teramoto H, Salem P, Robbins KC, Bustelo XR, Gutkind JS. 1997. Tyrosine phosphorylation of the vav protooncogene product links Fc epsilon RI to the Rac1-JNK pathway. J. Biol. Chem. 272:10751–55 Margolis B, Hu P, Katzav S, Li W, Oliver JM, Ullrich A, Weiss A, Schlessinger J. 1992. Tyrosine phosphorylation of vav proto-oncogene product containing SH2 domain and transcription factor motifs [see comments]. Nature 356:71–74 Song JS, Gomez J, Stancato LF, Rivera J. 1996. Association of a p95 Vavcontaining signaling complex with the

244.

245.

246.

247.

248.

249.

Fcepsilon RI gamma chain in the RBL2H3 mast cell line. Evidence for a constitutive in vivo association of Vav with Grb2, Raf-1, and ERK2 in an active complex. J. Biol. Chem. 271:26962– 70 Jackman JK, Motto DG, Sun Q, Tanemoto M, Turck CW, Peltz GA, Koretzky GA, Findell PR. 1995. Molecular cloning of SLP-76, a 76-kDa tyrosine phosphoprotein associated with Grb2 in T cells. J. Biol. Chem. 270:7029–32 Hendricks-Taylor LR, Motto DG, Zhang J, Siraganian RP, Koretzky GA. 1997. SLP-76 is a substrate of the high affinity IgE receptor-stimulated protein tyrosine kinases in rat basophilic leukemia cells. J. Biol. Chem. 272:1363–67 Yablonski D, Kuhne MR, Kadlecek T, Weiss A. 1998. Uncoupling of nonreceptor tyrosine kinases from PLC-gamma 1 in an SLP-76-deficient T cell. Science 281:413–16 Clements JL, Yang B, Ross-Barta SE, Eliason SL, Hrstka RF, Williamson RA, Koretzky GA. 1998. Requirement for the leukocyte-specific adapter protein SLP76 for normal T cell development. Science 281:416–19 Pivniouk V, Tsitsikov E, Swinton P, Rathbun G, Alt FW, Geha RS. 1998. Impaired viability and profound block in thymocyte development in mice lacking the adaptor protein SLP-76. Cell 94:229–38 Ota Y, Beitz LO, Scharenberg AM, Donovan JA, Kinet JP, Samelson LE. 1996. Characterization of Cbl tyrosine phosphorylation and a Cbl-Syk complex in RBL-2H3 cells. J. Exp. Med. 184:1713–23

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:931-972. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973

P1: apr/mbg

P2: apr/spd

February 11, 1999

QC: apr/UKS

13:1

T1: APR

Annual Reviews

AR078-SUP

Annu. Rev. Immunol. 1999.17:973-976. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Annu. Rev. Immunol. 1999. 17:973–76 c 1999 by Annual Reviews. All rights reserved Copyright °

THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (FcεRIα ) Scott C. Garman, Jean-Pierre Kinet†, and Theodore S. Jardetzky Department of Biochemistry, Molecular Biology, and Cell Biology, Northwestern University, Evanston Illinois 60208; e-mail: [email protected]; and †Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, Massachusetts 02215

Introduction One necessary step in the activation of effector cells in the immune system occurs through binding to antibody receptors, FcRs. These receptors link the diversity of the antibody repertoire to the variable response capacity of effector cells. On mast cells, the FcR known as FcεRI binds the IgE antibody, triggering both beneficial responses to parasites and inappropriate responses to allergens. The FcεRI receptor is composed of three membrane bound chains, α, β, and γ , with a stoichiometry of 1:1:2 (see 1–4 for review). The structure of the extracellular antibody-binding portion of the human high-affinity IgE receptor, FcεRIα, has been solved by X-ray crystallography (5). The structure reveals two tandem immunoglobulin (Ig) domains that are highly bent (Figure 1; see color figure). Thus, on the surface of the mast cell, the FcεRIα molecule presents a large convex surface to its IgE ligand. The most remarkable aspect of the structure is the collection of four tryptophans at the top of the molecule, where antibody binding occurs. Carbohydrate attachment sites cover the sides of the molecule but are not found on the top of the molecule near the antibody-binding region.

Human FcεRIα Structure As shown in Figure 1, the overall structure of the FcεRIα molecule is an inverted V shape. In this view, with the first domain (D1) to the right and the second (D2) to the left, the binding site for IgE is found at the top of the molecule, and the cell membrane at the bottom. In this orientation, the polypeptide is ˚ × 55 A ˚ × 30 A. ˚ The crystal structure includes amino acids approximately 55 A 973

P1: SKH

14:51

Annual Reviews

AR085-19

Annu. Rev. Immunol. 1999.17:973-976. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

March 15, 1999

Figure 1 A ribbon diagram (14) of the human Fc²RIα molecule. Some regions described in the text are shown.

P1: apr/mbg

P2: apr/spd

February 11, 1999

Annu. Rev. Immunol. 1999.17:973-976. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

974

QC: apr/UKS

13:1

T1: APR

Annual Reviews

AR078-SUP

GARMAN, KINET & JARDETZKY

1–176 of the mature FcεRIα polypeptide sequence, where the transmembrane region begins at amino acid 177. In the FcεRIα molecule, each Ig domain is approximately 85 amino acids, smaller than typical immunoglobulin domains found in antibodies. The two ˚ 2) between them, as domains in the structure bury a large interface (1280 A 28 amino acids are involved in the contact. In D1, the residues buried in the interface include 12–18, 20, and 84–86, while in D2, the interface residues are 87–93, 95, 104, 106, 108, 110–111, 161 and 163–165. The large amount of surface area buried and the large number of interdomain residues suggest that the molecule’s function requires the acute angle between the domains. A collection of four tryptophans packs at the top of the molecule. These amino acids include TRP 87, 110, 113, and 156. They form a hydrophobic patch suitable for binding the Fc portion of the IgE antibody. The four tryptophans are found on the D1-D2 interface, on the B-C loop of D2, and on the F-G loop of D2. In a sequence comparison of homologous FcRs, two of the tryptophans (87 and 110) are conserved across 20 members of the family. The other two positions vary depending on the ligand: They are generally lysine at position 113 and glycine at position 156 in the members of the family that bind IgG. Sequence variability in the loops at the top of the molecule in part determines isotype and species specificity of different FcRs. Mutagenesis of these regions in FcεRIα affects the binding of IgE (6–10). IgE binding is disrupted upon mutagenesis of residues in the D2 C strand (residues 115, 117–118, and 120–123), C0 -E loop (residues 129 and 131), the F strand (residues 149 and 153), the F-G loop (residue 155) and the G strand (residue 159). Additionally, amino acid 87 (in the D1-D2 interface) and 128 (in the C0 -E loop) are near the IgE binding site, as point mutants at these positions interfere with binding of the IgE mutant R334A. Figure 2 (see color figure) shows an atomic representation of the FcεRIα, highlighting the regions of the molecule implicated in IgE binding by mutagenesis studies. At the bottom of the molecule, a cleft appears close to the membrane surface. This region is inaccessible to macromolecules interacting with the top surface but may interact with the extracellular portion of the FcεRI β or γ chains or other accessory molecules on the mast cell surface. There are seven N-linked carbohydrate attachment sites in the human FcεRIα molecule. They are distributed about the front and back of the molecule, but are not found on the top of the molecule (Figure 3; see color figure). Three of these sites appear ordered in the crystal. Although the oligosaccharide on FcεRIα is not required for binding IgE, the deglycosylated receptor forms aggregates in solution (11–13). When N-linked glycosylation sites from homologous members of the FcR family are mapped onto the FcεRIα structure, the sites cover the front and back surface of the molecule but are generally not found

P1: SAT/VKS

P2: NBL/plb

Annu. Rev. Immunol. 1999.17:973-976. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

February 27, 1999

16:50

QC: NBL/tkj

T1: NBL

Annual Reviews

AR078-01

Figure 2 An atomic model of the Fc²RIα molecule in the same orientation as Figure 1. Amino acid residues shown in red have been implicated in IgE mutagenesis. Four tryptophans pack at the top of the molecule.

P1: SKH

14:51

Annual Reviews

AR085-19

Annu. Rev. Immunol. 1999.17:973-976. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

March 15, 1999

Figure 3 A surface representation (15) of the Fc²RIα molecule with carbohydrate sites highlighted in blue. The asparagine resisdues are shown as well as those carbohydrate atoms visible in the electron density. The top view is the same as is Figure 1, and the bottom view is rotated 180◦ about the vertical axis.

P1: apr/mbg

P2: apr/spd

February 11, 1999

QC: apr/UKS

13:1

T1: APR

Annual Reviews

AR078-SUP

Annu. Rev. Immunol. 1999.17:973-976. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

CRYSTAL STRUCTURE OF Fc (FcεRIα)

975

at the top of the molecule. The carbohydrates may reduce the affinity of the receptor for itself, thus preventing premature aggregation on the cell surface. The two Ig domains in the FcεRIα molecule have similar protein folds. The ˚ for 71 alpha carbon domains superimpose with an RMS deviation of 1.26 A atoms. The domains each contain primarily antiparallel beta sheet with a short region of parallel beta sheet. In each domain, strands A, B, and E comprise one face and strands A0 , G, F, C, and C0 form the second. Strand A makes hydrogen bonds with strand B, but then it crosses over to other sheet to make strand A0 , which hydrogen bonds to the parallel G strand. In each domain, the crossover region between A and A0 is involved in the interdomain contacts. Each strand also lacks a D strand (typically found on the end of the ABE beta sheet) but instead contains a short C0 strand making hydrogen bonds to the C strand of the opposite beta sheet. The regions where the domains differ correlate to sites involved in IgE binding. For example, in D2 the F-G loop extends further than the homologous loop in D1. In D2, the F-G loop forms the top of the molecule and contains residues such as 155 implicated in IgE binding. The C0 -E regions differ in the two domains, and in D2 this region is needed for FcεRIα binding to IgE.

Future Directions The structure of the FcεRIα molecule provides a window into signaling via Fc receptors, and it creates a model for homologous members of the family. It also raises many molecular questions. How does the receptor bind a single IgE antibody when the antibody Fc region is dimeric? Does the FcεRIα bind across the IgE Fc dimer axis, or does it introduce a conformational change in the antibody to disrupt the symmetry axis? What are the interactions, if any, between the extracellular α chain and the extracellular portions of the β and γ chains? The molecular details of these interactions remain to be seen. Visit the Annual Reviews home page at http://www.AnnualReviews.org

Literature Cited 1. Kinet JP. 1999. The high affinity IgE receptor (FcεRI) from physiology to pathology. Annu. Rev. Immunol. 17:931–76 2. Metzger H. 1994. Immunoglobulin receptors. Handicapping the immune response. Curr. Biol. 4:644–46 3. Ravetch JV. 1994. Fc receptors: rubor redux. Cell 78:553–60 4. Sutton BJ, Gould HJ. 1993. The human IgE network. Nature 366:421–28 5. Garman SC, Kinet JP, Jardetzky TS. 1998.

Crystal structure of the human highaffinity IgE receptor. Cell 95:951–61 6. Cook JP, Henry AJ, McDonnell JM, Owens RJ, Sutton BJ, Gould HJ. 1997. Identification of contact residues in the IgE binding site of human Fc²RIα. Biochemistry 36:15579–88 7. Mallamaci MA, Chizzonite R, Griffin M, Nettleton M, Hakimi J, Tsien WH, Kochan JP. 1993. Identification of sites on the human Fc²RIα subunit which are involved in

P1: apr/mbg

P2: apr/spd

February 11, 1999

13:1

976

8.

Annu. Rev. Immunol. 1999.17:973-976. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

9.

10.

11.

QC: apr/UKS

T1: APR

Annual Reviews

AR078-SUP

GARMAN, KINET & JARDETZKY

binding human and rat IgE. J. Biol. Chem. 268:22076–83 Hulett MD, McKenzie IF, Hogarth PM. 1993. Chimeric Fc receptors identify immunoglobulin-binding regions in human Fcγ RII and Fc²RI. Eur. J. Immunol. 23: 640–45 Hulett MD, Witort E, Brinkworth RI, McKenzie IF, Hogarth PM. 1994. Identification of the IgG binding site of the human low affinity receptor for IgG Fcγ RII. Enhancement and ablation of binding by site-directed mutagenesis. J. Biol. Chem. 269:15287–93 Hulett MD, Witort E, Brinkworth RI, McKenzie IF, Hogarth PM. 1995. Multiple regions of human Fcγ RII (CD32) contribute to the binding of IgG. J. Biol. Chem. 270:21188–94 Letourner O, Sechi S, Willette-Brown J, Robertson MW, Kinet JP. 1995. Glycosylation of human truncated Fc²RI α-chain

12.

13.

14.

15.

is necessary for efficient folding in the endoplasmic reticulum. J. Biol. Chem. 270: 8249–56 Robertson MW. 1993. Phage and Escherichia coli expression of the human high affinity immunoglobulin E receptor alphasubunit ectodomain. Domain localization of the IgE-binding site. J. Biol. Chem. 268:12736–43 Scarselli E, Esposito G, Traboni C. 1993. Receptor phage. Display of functional domains of the human high affinity IgE receptor on the M13 phage surface. FEBS Lett. 329:223–26 Kraulis PJ. 1991. MOLSCRIPT: A program to produce both detailed and schematic plots of protein structures. J. Appl. Crystall. 24:946–50 Nicholls A, Sharp KA, Honig B. 1991. Protein folding and association: insights from the interfacial and thermodynamic properties of hydrocarbons. Proteins 11:281–96

Annual Review of Immunology Volume 17, 1999

CONTENTS Discovering the Origins of Immunological Competence, Jacques F. A. P. Miller Multifaceted Regulation of IL-15 Expression and Its Role in NK Cell Differentiation & Host Response to Intracellular Pathogens, T. A. Waldmann, Y. Tagaya Immunodominance in Major Histocompatibility Complex Class IRestricted T Lymphocyte Responses, Jonathan W. Yewdell, Jack R. Bennink

Annu. Rev. Immunol. 1999.17:973-976. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Integration of TCR-Dependent Signaling Pathways by Adapter Proteins, James L. Clements, Nancy J. Boerth, Jong Ran Lee, Gary A. Koretzky Evolution of Antigen Binding Receptors, Gary W. Litman, Michele K. Anderson, Jonathan P. Rast Transcriptional Regulation of T Lymphocyte Development and Function, Chay T. Kuo, Jeffrey M. Leiden Natural Killer Cells in Antiviral Defense: Function and Regulation by Innate Cytokines, Christine A. Biron, Khuong B. Nguyen, Gary C. Pien, Leslie P. Cousens, Thais P. Salazar-Mather Mature T Lymphocyte Apoptosis--Immune Regulation in a Dynamic and Unpredictable Antigenic Environment, Michael Lenardo, Francis KaMing Chan, Felicita Hornung, Hugh McFarland, Richard Siegel, Jin Wang, Lixin Zheng Immunologic Basis of Antigen-Induced Airway Hyperresponsivenes, Marsha Wills-Karp Regulation of T Cell Fate by Notch, Ellen Robey The CD1 System: Antigen Presenting Molecules for T Cell Recognition of Lipids and Glycolipids, Steven A. Porcelli, Robert L. Modlin Tumor Necrosis Factor Receptor and Fas Signaling Mechanisms, D. Wallach, E. E. Varfolomeev, N. L. Malinin, Yuri V. Goltsev, A. V. Kovalenko, M. P. Boldin Structural Basis of T Cell Recognition, K. Christopher Garcia, Luc Teyton, Ian A. Wilson Development and Maturation of Secondary Lymphoid Tissues, Yang-Xin Fu, David D. Chaplin The Structural Basis of T Cell Activation by Superantigens, Hongmin Li, Andrea Llera, Emilio L. Malchiodi, Roy A. Mariuzza The Dynamics of T Cell Receptor Signaling: Complex Orchestration and the Key Roles of Tempo and Cooperation, Ronald N. Germain, Irena Stefanová The Regulation of CD4 and CD8 Coreceptor Gene Expression During T Cell Development, Wilfried Ellmeier, Shinichiro Sawada, Dan R. Littman Genetic Analysis of B Cell Antigen Receptor Signaling, Tomohiro Kurosaki Mechanisms of Phagocytosis in Macrophages, Alan Aderem, David M. Underhill Population Biology of HIV-1 Infection: Viral and CD4+ T Cell Demographics and Dynamics in Lymphatic Tissues, A. T. Haase

1 19

51

89 109 149 189

221

255 283 297

331 369 399 435 467

523 555 593 625

Chemokine Receptors as HIV-1 Coreceptors: Roles in Viral Entry, Tropism, and Disease, Edward A. Berger, Philip M. Murphy, Joshua M. Farber The IL-4 Receptor: Signaling Mechanisms and Biologic Functions, Keats Nelms, Achsah D. Keegan, José Zamorano, John J. Ryan, William E. Paul Degradation of Cell Proteins and the Generation of MHC Class IPresented Peptides, Kenneth L. Rock, Alfred L. Goldberg The Central Effectors of Cell Death in the Immune System, Jeffrey C. Rathmell, Craig B. Thompson

Annu. Rev. Immunol. 1999.17:973-976. Downloaded from arjournals.annualreviews.org by HINARI on 09/01/07. For personal use only.

Selection of the T Cell Repertoir, Eric Sebzda, Sanjeev Mariathasan, Toshiaki Ohteki, Russell Jones, Martin F. Bachmann, Pamela S. Ohashi Regulation of Immune Responses Through Inhibitory Receptors, Eric O. Long The Wiskott-Aldrich Syndrome Protein (WASP): Roles in Signaling and Cytoskeletal Organization, Scott B. Snapper, Fred S. Rosen The High Affinity IgE Receptor (Fc Epsilon RI): From Physiology to Pathology, Jean-Pierre Kinet THE CRYSTAL STRUCTURE OF THE HUMAN HIGH-AFFINITY IgE RECEPTOR (Fc epsilon RI alpha), Scott C. Garman, Jean-Pierre Kinet, Theodore S. Jardetzky

657

701 739 781 829 875 905 931

973