Cell Fusions
Lars-Inge Larsson Editor
Cell Fusions Regulation and Control
13
Editor Lars-Inge Larsson Faculty of Life Sciences Division of Cell Biology University of Copenhagen Gronnegaardsvej 7 1870 Frederiksberg C Denmark
[email protected]
ISBN 978-90-481-9771-2 e-ISBN 978-90-481-9772-9 DOI 10.1007/978-90-481-9772-9 Springer Dordrecht Heidelberg London New York © Springer Science+Business Media B.V. 2011 No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
Preface
The fusion between the sperm and the egg marks the beginning of life and a number of subsequent cell fusions are needed to form the placenta, the skeleton, the skeletal muscles and part of our immune defence system. Cell fusions are also needed during adulthood, e.g. to regulate our calcium homeostasis. Apart from the fusion between the sperm and the egg, all of these fusions occur between cells of the same type. Such homotypic fusions generate multinucleated cells that have lost the ability to propagate but which, through their large sizes, are more powerful than single cells. This is well illustrated by skeletal muscle fibers, which are derived from fusions of multiple stem cells and can reach 1/2 m in length. Similarly, giant cells of the immune system and osteoclasts, degrading bone tissue, reach their powerful abilities of phagocytosis from the fusion of smaller progenitor cells (macrophages/monocytes). Heterotypic fusions occur between cells of different lineages and may result in cells with proliferative ability. If such fusions occur between haploid gametes, an embryo results. However, if heterotypic fusions occur between diploid or aneuploid cells, the mitotic spindle apparatus may encounter problems of sorting the supernumerary chromosomes. This may result in genomic instability and in cell death or cancer. Accordingly, cell fusion is a process that should be entered with even more care than marriage since a divorce (defuse?) is impossible. Nevertheless, as we shall learn from this volume, heterotypic cell fusions may play roles in repair of damaged tissues and may be put to use for production of monoclonal antibodies and for boosting the immune system against cancer cells. In this book, we learn of mechanisms regulating and controlling cell fusions. An important aspect in normal physiology is the matter of self recognition mechanisms, which ensure that, in the healthy individual, homotypic fusions predominate. As we shall see from the contributions on fertilization, placentation, macrophages/osteoclasts and skeletal muscle development, multiple mechanisms are involved. At the turn of the millennium, a new player entered the fusion game. It emerged that we have adopted certain fusion genes from ancient viral infections and have used them in the cell fusion machinery. The initial chapters in this book are therefore devoted to an in-depth discussion of the viral fusion machinery and how cells may utilize these proteins for their own use. This is intellectually and conceptually a challenging area in modern biology since it shortcuts traditional Darwinian
v
vi
Preface
evolution with infections. Astonishingly, similar but distinct retroviral genes have been adopted by primates, rabbits, mice and, possibly, sheep, for cell fusions. Mouse knock-down experiments indicate that a mouse endogenous retroviral gene is needed for placentation, probably because it made pre-existing fusion genes redundant. In addition to the new “viral” players, model organisms like Drosophila and Caenorhabditis have made considerable impact on our understanding of the cell fusion machinery. These mechanisms as well as methodological aspects have been dealt with in an excellent recent volume edited by Chen and are referred to here in the context of homotypic cell fusions and in the FuRMAS structure in Chapter 6. In the final chapters, we address the question of heterotypic cell fusions. Such fusions seem to gain momentum as main players in tissue repair and stem cell therapy as well as in immune therapy of tumors. In both of these settings heterotypic fusions appear to do good whereas the backside of the coin, cancer, is addressed in the last two chapters. As captain on this voyage I have had the pleasure to be able to recruit a crew consisting of the most eminent specialists in this evolving field. They have helped set sails and chartered our path from viruses over normal development to cancer and treatment. My sincerest thanks go to them, to the representatives of the shipping company, Meran Owen and Tanja van Gaans, who patiently awaited the sight of sails against the horizon, and to my family; Benedikte, Nina and Blackie, who have served as “stowaways” on board the vessel whilst the log-book was edited. Lars-Inge Larsson
Contents
1 Regulation and Control of Cell–Cell Fusions . . . . . . . . . . . . . Lars-Inge Larsson
1
2 Retroviruses and Cell Fusions: Overview . . . . . . . . . . . . . . . Anders L. Kjeldbjerg, Shervin Bahrami, and Finn Skou Pedersen
11
3 Retroviral Membrane Fusions: Regulation by Proteolytic Processing and Cellular Factors . . . . . . . . . . . . . . . . . . . . Yoshinao Kubo 4 A Comparative Portrait of Retroviral Fusogens and Syncytins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Philippe Pérot, Cécile Montgiraud, Dimitri Lavillette, and François Mallet 5 Syncytins: Molecular Aspects . . . . . . . . . . . . . . . . . . . . . Hungwen Chen and Mei-Leng Cheong 6 Role of the Actin Cytoskeleton Within FuRMAS During Drosophila Myoblast Fusion and First Functionally Conserved Factors in Vertebrates . . . . . . . . . . . . . . . . . . . Susanne-Filiz Önel, Christine Dottermusch, Angela Sickmann, Detlev Buttgereit, and Renate Renkawitz-Pohl 7 Role of CD9 in Sperm-Egg Fusion and Its General Role in Fusion Phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . Natsuko Kawano, Yuichiro Harada, Keiichi Yoshida, Mami Miyado, and Kenji Miyado
41
63
117
139
171
8 Gamete Binding and Fusion . . . . . . . . . . . . . . . . . . . . . . Young-Joo Yi, Shawn W. Zimmerman, and Peter Sutovsky
185
9 Mechanisms Regulating Human Trophoblast Fusion . . . . . . . . Berthold Huppertz and Martin Gauster
203
vii
viii
Contents
10
Macrophage Fusion: The Making of a New Cell . . . . . . . . . . . Agnès Vignery
219
11
Molecules Regulating Macrophage Fusions . . . . . . . . . . . . . Takeshi Miyamoto and Toshio Suda
233
12
Current Progress Towards Understanding Mechanisms of Myoblast Fusion in Mammals . . . . . . . . . . . . . . . . . . . Grace K. Pavlath
249
The Endogenous Envelope Protein Syncytin Is Involved in Myoblast Fusion . . . . . . . . . . . . . . . . . . . . Bolette Bjerregaard, Jan Fredrik Talts, and Lars-Inge Larsson
267
13
14
Cell Fusion and Stem Cells . . . . . . . . . . . . . . . . . . . . . . Alain Silk, Anne E. Powell, Paige S. Davies, and Melissa H. Wong
277
15
Cell Fusion and Dendritic Cell-Based Vaccines . . . . . . . . . . . Jianlin Gong and Shigeo Koido
315
16
Cancer Cell Fusion with Myeloid Cells: Implications for Energy Metabolism in Malignant Hybrids . . . . . . . . . . . . Rossitza Lazova, Ashok K. Chakraborty, and John M. Pawelek
351
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reiner Strick, Matthias W. Beckmann, and Pamela L. Strissel
395
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
427
17
Contributors
Shervin Bahrami Department of Molecular Biology, Aarhus University, DK-8000 Aarhus C, Denmark,
[email protected] Matthias W. Beckmann Department of Gynaecology and Obstetrics, Laboratory for Molecular Medicine, University-Clinic Erlangen, D-91054 Erlangen, Germany,
[email protected] Bolette Bjerregaard Division of Cell Biology, Faculty of Life Sciences, University of Copenhagen, DK-1870 Frederiksberg C, Denmark,
[email protected] Detlev Buttgereit Fachbereich Biologie, Entwicklungsbiologie, Philipps-Universität Marburg, 35043 Marburg, Germany,
[email protected] Ashok K. Chakraborty Department of Dermatology, Yale Cancer Center, Yale University School of Medicine, New Haven, CT 06520-8059, USA,
[email protected] Hungwen Chen Institute of Biological Chemistry, Academia Sinica, Taipei, Taiwan,
[email protected] Mei-Leng Cheong Institute of Biological Chemistry, Academia Sinica, Taipei, Taiwan,
[email protected] Paige S. Davies Department of Dermatology, Knight Cancer Institute, Oregon Stem Cell Center, Oregon Health and Science University, Portland, OR 97239, USA,
[email protected] Christine Dottermusch Fachbereich Biologie, Entwicklungsbiologie, Philipps-Universität Marburg, 35043 Marburg, Germany,
[email protected] Martin Gauster Institute of Cell Biology, Histology and Embryology, Medical University Graz, Austria,
[email protected] Jianlin Gong Department of Medicine, Boston University Medical School, Boston, MA 02118, USA,
[email protected]
ix
x
Contributors
Yuichiro Harada Department of Reproductive Biology, National Center for Child Health and Development, Tokyo 157-8535, Japan,
[email protected] Berthold Huppertz Cell Biology, Institute of Cell Biology, Histology and Embryology, Center for Molecular Medicine, Medical University of Graz, 8010 Graz, Austria,
[email protected] Natsuko Kawano Department of Reproductive Biology, National Center for Child Health and Development, Tokyo 157-8535, Japan,
[email protected] Anders L. Kjeldbjerg Department of Molecular Biology, Aarhus University, DK-8000 Aarhus C, Denmark,
[email protected] Shigeo Koido Department of Internal Medicine, Institute of Clinical Medicine and Research, The Jikei University School of Medicine, Tokyo, Japan,
[email protected] Yoshinao Kubo Department of AIDS Research, Institute of Tropical Medicine, Nagasaki University, Nagasaki 852-8523, Japan,
[email protected] Lars-Inge Larsson Division of Cell Biology, Anatomy, Cell Biology and Genetics, IBHV, Faculty of Life Sciences, University of Copenhagen, 1809 Frederiksberg C, Denmark,
[email protected] Dimitri Lavillette INSERM, U758, 69007 Lyon, France,
[email protected] Rossitza Lazova Department of Dermatology, Yale Cancer Center, Yale University School of Medicine, New Haven, CT 06520-8059, USA,
[email protected] François Mallet Ecole Normale Supérieure de Lyon, 69007 Lyon, France,
[email protected] Mami Miyado Department of Reproductive Biology, National Center for Child Health and Development, Tokyo 157-8535, Japan,
[email protected] Kenji Miyado Department of Reproductive Biology, National Center for Child Health and Development, Tokyo 157-8535, Japan,
[email protected] Takeshi Miyamoto Department of Orthopedic Surgery, Keio University School of Medicine, Tokyo 160-8582, Japan; Keio Kanrinmaru Project, Keio University School of Medicine, Tokyo 160-8582, Japan,
[email protected] Cécile Montgiraud Université de Lyon, UCB-Lyon1, IFR128, 69007 Lyon, France,
[email protected] Susanne-Filiz Önel Fachbereich Biologie, Entwicklungsbiologie, Philipps-Universität Marburg, 35043 Marburg, Germany,
[email protected] Grace K. Pavlath Department of Pharmacology, Emory University, Atlanta, GA 30322, USA,
[email protected]
Contributors
xi
John M. Pawelek Department of Dermatology, Yale Cancer Center, Yale University School of Medicine, New Haven, CT 06520-8059, USA,
[email protected] Finn Skou Pedersen Department of Molecular Biology, Aarhus University, DK-8000 Aarhus C, Denmark,
[email protected] Philippe Pérot Laboratoire Commun de Recherche Hospices Civils de Lyon – bioMérieux, Cancer Biomarkers Research Group, Centre Hospitalier Lyon Sud, 69495 Pierre Bénite cedex, France,
[email protected] Anne E. Powell Department of Cell and Developmental Biology, Oregon Health and Science University, Portland, OR 97239, USA,
[email protected] Renate Renkawitz-Pohl Fachbereich Biologie, Entwicklungsbiologie, Philipps-Universität Marburg, 35043 Marburg, Germany,
[email protected] Angela Sickmann Fachbereich Biologie, Entwicklungsbiologie, Philipps-Universität Marburg, 35043 Marburg, Germany,
[email protected] Alain Silk Department of Dermatology, Knight Cancer Institute, Oregon Stem Cell Center, Oregon Health and Science University, Portland, OR 97239, USA,
[email protected] Reiner Strick Department of Gynaecology and Obstetrics, Laboratory for Molecular Medicine, University-Clinic Erlangen, D-91054 Erlangen, Germany,
[email protected] Pamela L. Strissel Department of Gynaecology and Obstetrics, Laboratory for Molecular Medicine, University-Clinic Erlangen, D-91054 Erlangen, Germany,
[email protected] Toshio Suda Department of Cell Differentiation, Keio University School of Medicine, Tokyo 160-8582, Japan,
[email protected] Peter Sutovsky Division of Animal Science, University of Missouri-Columbia, Columbia, MO 65211, USA; Departments of Obstetrics, Gynecology & Women’s Health, University of Missouri-Columbia, Columbia, MO 65211, USA,
[email protected] Jan Fredrik Talts Division of Cell Biology, Faculty of Life Sciences, University of Copenhagen, DK-1870 Frederiksberg C, Denmark,
[email protected] Agnès Vignery Departments of Orthopaedics and Cell Biology, Yale School of Medicine, New Haven, CT 06510, USA,
[email protected] Melissa H. Wong Department of Dermatology, Knight Cancer Institute, Oregon Stem Cell Center, Oregon Health and Science University, Portland, OR 97239,
xii
Contributors
USA; Department of Cell and Developmental Biology, Oregon Health and Science University, Portland, OR 97239, USA,
[email protected] Young-Joo Yi Division of Animal Science, University of Missouri-Columbia, Columbia, MO 65211, USA; Research Center for Transgenic Cloned Pigs, Chungnam National University, Daejeon 305-764, South Korea,
[email protected] Keiichi Yoshida Department of Reproductive Biology, National Center for Child Health and Development, Tokyo 157-8535, Japan,
[email protected] Shawn W. Zimmerman Division of Animal Science, University of Missouri-Columbia, Columbia, MO 65211, USA,
[email protected]
Chapter 1
Regulation and Control of Cell–Cell Fusions Lars-Inge Larsson
Abstract Cell fusions are important to fertilization, fetal development and homeostasis. Retroviruses infect cells by fusing with them and recent data suggest that mammals may have adopted the retroviral fusion machinery for their own use and combined it with numerous other factors controlling cell specificity and self recognition, motility-migration, filopodia formation, signaling and membrane organization. The multifactorial aspect of the process is suggested to create a certain amount of wobble so that, in specific disease states, heterotypic cell fusions may occur. Professional phagocytes, which specialize in recognizing and eliminating injured or dying cells, appear to be particularly prone to fusion. Such fusions may be useful for repairing damaged tissues and have been harnessed in immune therapy against cancer but may also contribute to disease development and progression. Keywords Cancer · cell-cell fusion · HERV · microdomains · retrovirus · syncytin Abbreviations ADAM ASCT BMDC CD CRISP DC-STAMP env ERM F-actin FuRMAS GCM HERV
A disintegrin and a metalloprotease Alanine, serine and cysteine selective transporters Bone marrow-derived cells Cluster of differentiation Cysteine-rich secretory protein Dendritic cell-specific transmembrane protein Envelope Ezrin-radixin-moesin Filamentous (polymerized) actin Fusion restricted myogenic adhesive structure Glial cells missing Human endogenous retrovirus
L.-I. Larsson (B) Division of Cell Biology, Anatomy, Cell Biology and Genetics, IBHV, Faculty of Life Sciences, University of Copenhagen, 1809 Frederiksberg C, Denmark e-mail:
[email protected] L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_1, C Springer Science+Business Media B.V. 2011
1
2
LTR MFR NSF PCD PKA RANK RANKL SIRP-α SNAP SNARE t-SNARE v-SNARE
L.-I. Larsson
Long terminal repeat Macrophage fusion receptor N-ethylmaleimide-sensitive factor Programmed cell death Protein kinase A Receptor activator of NFκ B RANK ligand Signal regulatory protein-alpha Soluble NSF attachment protein SNAP receptors Target-SNARE Vesicle-SNARE
Cell–cell fusions are important to the beginning of life, to the sculpturing of the new individual and to the maintenance of muscle strength, immune responses and calcium homeostasis in the adult. Additionally, cell fusions participate in tissue repair and may be important to cancer development, progression and therapy. Not surprisingly, fusions must be tightly controlled in order to ensure that only the appropriate cells fuse with their right partners at the right time. Much knowledge of such regulation comes from studies of model organisms like Caenorhabditis elegans and, as so many times before, the little worm and its associates have provided us with a wealth of important data. Much of this, as well as many methodological aspects has been summarized in an admirable recent volume, edited by Chen (2008). In the present volume, we interrogate what mechanisms that may be important to cell fusions in mammals and what mechanisms that may be shared with lower organisms, including viruses. Additionally, the importance of cell fusions to the pathogenesis as well as treatment of diseases is addressed. Retroviruses, which infect cells by fusing with them, are discussed first. Retroviral infections often induce fusions between cells in the infected individual and recent data suggest that multiple mammalian species may have adopted the retroviral fusion machinery for their own use (Chapters 2 and 4). Interestingly, different species seem to have adopted different, but basically similar, fusion machineries from different types of retroviruses. In doing so they seem to have bypassed millions of years of evolution and have taken advantage of the higher mutation rates found in viral genomes. For this reason, mammalian cells may possess cell–cell fusion mechanisms unrelated to those detected in lower invertebrates. Retroviral membrane fusions are subject to a number of controlling factors as reviewed in Chapter 3. Some of these relate to proteolytic processing e.g. of the cytoplasmic tail and others to insertions into rafts/microdomains and interactions with cell adhesion molecules and the actin cytoskeleton – regulatory events also recognized in mammalian cell–cell fusions. In the process of forming new virus particles, retroviruses get their genetic information integrated into the host DNA. It is, indeed, a humbling thought that over 8% of our genome represents identifiable remnants of ancient infections. Most of this lies in ruins because our cells have found little use for the information
1
Regulation and Control of Cell–Cell Fusions
3
encoded by viral genomes. Thus, most of these sequences have mutated or undergone other changes resulting in transcriptionally inactive mumbo-jumbo. However, certain sequences, including envelopes (env), which encode the viral proteins which makes fusion – and, hence, infection – possible, have been highly conserved, suggesting that our bodies have adopted them to specific functions (Chapters 2 and 4). At the turn of the millennium, two reports provided data suggesting that the retrovirally encoded transmembrane protein, syncytin (later renamed syncytin-1) could be important to trophoblast cell–cell fusions, which are needed for forming a functioning placenta in primates, including man (Blond et al. 2000, Mi et al. 2000). Syncytin is an env protein, which was derived from a retroviral infection that entered our genome over 25 million years ago and which has been highly conserved (Chapter 4). The fusion mechanism employed by syncytin is very similar to that used by retroviruses for infection and may be inhibited by specific peptide sequences interfering with this mechanism (Chapters 2 and 5). Later, syncytin was implicated in mediating fusions between cancer cells and endothelial cells, between cancer cells themselves and in fusions between skeletal muscle-forming myoblasts (Bjerregaard et al. 2006, Strick et al. 2007, Chapters 13 and 17). Recent data also indicate an involvement of syncytin in fusions needed to form osteoclasts (Soe K, Andersen TL, HoboltPedersen AS, Bjerregaard B, Larsson LI, Delaisse JM: submitted). Thus, syncytin may possibly participate in multiple types of cell–cell fusions in humans but, as we shall see shortly, there are many players in this game. We believe that a plethora of molecules are important for determining diverse aspects of cell fusions, including specific cell migration, cell–cell recognition, signaling events, cytoskeletal and membrane microdomain organization, filopodia/nanotube formation, and, eventually, the membrane fusion event itself. In fact, syncytin is the best candidate for a true cell–cell fusogen that, so far, has been detected in primate cells (Oren-Suissa and Podbilewicz 2007). Other retroviral envelope proteins, encoded for by other endogenous retroviral sequences, are expressed in other mammals, including rabbits, sheep and mice (Dunlap et al. 2006, Dupressoir et al. 2005, Heidmann et al. 2009). Importantly, a knock-down of murine syncytin-A results in defect placentation and prevents successful reproduction (Dupressoir et al. 2009). If confirmed, this seems to suggest that the reproduction of modern mice depends upon an accidental retroviral infection acquired some 20–25 million years ago! One way to reconcile this with traditional evolution is to assume that the accidentally acquired viral env gene was so much more efficient in fusing cells so that it made a preexisting endogenous fusogen redundant. So far, the syncytin family of proteins, which, in man, includes syncytin-1, -2 and the fusogenic retroviral envelope protein P(b) are the only bona fide membrane fusogens detected (Chapters 2 and 4). The founding member, syncytin-1 has been implicated in a number of diseases including placental disorders, multiple sclerosis and cancer (Chapters 4, 5, 9 and 17). Expression of syncytin-1 and -2 are subject to a number of regulations as described in Chapters 4 and 5. Especially, the transcription factor GCM1 (glial cells missing) appears to play an important role in such regulation and GCM1 is regulated by cAMP and protein kinase A (PKA) – factors known to regulate trophoblast cell fusions (Chapter 5). Importantly, the degree of methylation of the syncytin-1
4
L.-I. Larsson
and -2 promoters affects their regulation by GCM1 and the possibility that GCM1 itself may affect methylation in the 5 -LTR of the syncytin-2 promoter in MCF-7 breast cancer cells is raised in Chapter 5. Syncytins may represent cell-unspecific fusogens with the specific aim of making membranes fuse, whereas other proteins guarantee the necessary cell-specificity and/or may serve migratory, scaffolding or signaling roles. The syncytins possess receptors, which are needed for activating the fusion machinery itself but which also represent rather ubiquitously expressed transporters. Thus, the receptors confer little cell-specificity but may form part of a larger protein scaffold, which contains other regulators conferring cell specificity. Additionally, as pointed out in Chapter 4, syncytins have many other and potentially non-redundant functions and it is possible that these molecules have many more surprises in store for us! Membrane fusion is triggered upon binding of syncytin-1 to its receptors, the neutral amino acid transporters ASCT1 and 2. Whether such binding also may trigger other cellular events remains to be seen. The situation with respect to fusion of intracellular membranes is quite different. Here, a complicated machinery of proteins controlling the v- and the t-SNARE system have been unraveled and have been found to constitute targets for various poisons and medications (Rothman 1994). This machinery controls both membrane fusions and its organelle specificity. It seems that we have a long way to go before similar models can be constructed for cell–cell fusions. Against this background, a volume such as the present may serve a useful purpose by emphasizing differences but also similarities in the machinery that make some cells fuse with their appropriate partners. As hinted to above, it is likely that cell–cell fusions depend upon multiple molecular interactions. In fact, membrane microdomains and the actin cytoskeleton as well as cell adhesion factors play decisive roles in cell–cell fusions. Thus, again we have a parallel to the mechanisms regulating retroviral fusions as discussed in Chapter 3. In Drosophila a specific structure, the FuRMAS (fusion restricted myogenic adhesive structure) – a synapse-like structure which may integrate cell fusion with cell adhesion, signaling and actin regulation exists (Chapter 6). As also discussed in this chapter, similar mechanisms may operate in e.g. zebrafish and mice. In fact, synapse-like structures, incorporating activated signaling molecules, F-actin and ezrin-radixin-moesin (ERM) proteins, which link transmembrane proteins to the actin cytoskeleton, are also observed at sites of contact between breast cancer cells known to undergo spontaneous fusion (Larsson, unpublished data). Much data suggests that the plasma membrane is subdivided into several distinct microdomains or rafts, which i. a. differ in protein and cholesterol composition (Simons and Ikonen 1997). Such rafts have been implicated in viral infection (Chapter 3) as well as in cell–cell recognition and signaling. Tetraspanins are proteins known to participate in the organization of the plasma membrane into microdomains and some tetraspanins, in particular CD9 and CD81, have been implicated in regulation of cell–cell fusions (Duelli et al. 2005, Miyado et al. 2000, Parthasarathy et al. 2009, Weng et al. 2009). CD 9 and CD81 may also bind to the actin cytoskeleton via EWI and ERM proteins (Sala-Valdés et al. 2006). Interestingly, recent data suggest that CD9 increases GCM1 expression via the
1
Regulation and Control of Cell–Cell Fusions
5
cAMP/PKA signaling pathway, resulting in the increase in syncytin expression (Muroi et al. 2009). The roles of CD9 and the interacting protein, IZUMO with respect to fertilization are discussed in Chapters 7 and 8. Interestingly, recent data show that small CD9-containing vesicles, resembling exosomes, can be released from eggs. Moreover, such exosome-like vesicles may induce fusion between spermatozoa and eggs lacking CD9, which otherwise are refractory to fusion. These data reinforce the central role of CD9 in sperm-egg cell fusions and emphasize that analogous functions may exist in other cell–cell fusion systems. Interestingly, CD9 and CD81 positive exosomes have been implicated also in cancer–cancer cell fusions in an in-vitro model (Duelli et al. 2005) and in HIV-induced syncytia formation (Weng et al. 2009). Exosomes are potentially interesting mediators of fusions as they are released from cells and, at least in theory, could fuse receptive bystander cells. This mechanism would be analogous to the viral “fusion from without”, discussed in Chapter 2. Much more studies are needed before we know whether exosomes indeed participate in “fusion from without” mechanisms under physiological circumstances. As with all other cell–cell fusions, sperm-egg cell fusions most likely depend upon multiple interacting proteins. Additional molecules that may participate in sperm-egg fusions, including v- and t-SNARE-like proteins, ADAM, CRISP and MN9, are discussed in Chapters 7 and 8. Overall, Chapters 6–8 provide a useful input into the complexities and potential redundancies that may exist in cell– cell fusions and underline the importance of membrane organization and protein scaffolding for fusions. Chapter 9 emphasizes the potential role of molecules involved in programmed cell death (PCD) like caspase-8 and phosphatidylserine in trophoblast cell fusions. Phosphatidylserine is a membrane phospholipid, which normally faces the cytosolic side of the plasma membrane. In cells destined to undergo PCD, phosphatidylserine is exposed to the outside of the plasma membrane. This is considered an “appetizer” for phagocytosis so that the dying cell can be cleared without traces or inflammatory reactions (Zhou 2007). Interestingly, phosphatidylserine is also flipped out on the outside of the plasma membrane during cell–cell fusion and experiments show that antibodies binding to phosphatidylserine can inhibit cell–cell fusion (reviewed in Chapter 9). Intriguingly, similarities between cell–cell fusions and PCD do not end here. A PCD-inducing enzyme, caspase-8, may also play a role in cell–cell fusions in the placenta. These intriguing data are placed in a comprehensive context of other molecules implicated in trophoblast cell–cell fusions and potential methodological issues regarding e.g. syncytin antibody specificities and labeling preferences in Chapter 9. Chapters 10 and 11 discuss mechanisms which regulate fusions between macrophages/monocytes to form giant cells and osteoclasts, including factors like MFR/SIRPα-CD47, CD200-CD200R, CD44, DC-STAMP and RANK/RANKL. These mechanisms are of enormous potential not least in view of the frequency of osteoporosis – a disease characterized by osteoclast hyperactivity. We learn how signaling factors like cytokines elicit intracellular signaling cascades, which ultimately stimulate osteoclast formation and bone resorption. This is guided by self recognition factors, which make the fusions cell-specific. However, osteoclasts
6
L.-I. Larsson
may also fuse with neoplastic cells like myeloma cells (Andersen et al. 2007). In Chapter 16, Lazova et al. gives numerous examples of how macrophage-like cells may fuse with cancer cells and provide new data on the potential importance of autophagy – an apoptosis-related path to cell death – for melanoma cell fusions. This disease-associated promiscuity of macrophage-cancer cell fusions could point to some kind of wobble mechanism in the specificity of cell–cell fusions and, potentially, macrophage-like, bone marrow-derived cells (BMDCs) may travel round the body, willing to fuse with and, thus, repair cells facing demise. Incidentally, this draws some interesting parallels to Chapter 9 regarding the potential role of factors regulating PCD in cell–cell fusions. Notably, phosphatidylserine, exposed on the surface of apoptotic cells attracts macrophages bearing specific receptors for this membrane phospholipid (Zhou 2007). The role of macrophages and other BMDCs in tissue repair and maintenance in the context of stem cell therapy is dealt with in Chapter 14. Again, the question arises as to how heterotypic cell–cell fusions may become triggered by diseases. Fortunately for our good looks, heterotypic fusions probably only occur in disease states and the intricate cell–cell recognition mechanisms described in Chapters 10 and 11 would normally ensure appropriate cell-specificities of fusions. As emphasized above, cell–cell fusions are probably regulated by both cell-specific and cell-unspecific factors and changes in the expression profiles of these factors may result in cell–cell liaisons, which normally are forbidden. The complexities and intricacies of cell fusions are well illustrated by fusion of myoblasts to generate skeletal muscle fibers (Chapters 6, 12 and 13). As presented by Grace Pavlath in Chapter 12, changes in myoblast elongation, migration, adhesion and cytoskeleton characterize such fusions and are regulated by a myriad of factors which are being unraveled one by one (see Tables 1–5 in Chapter 12, which elegantly orders all of these factors according to functions during the fusion process). Some of these factors are needed for the fusion per se whereas others regulate muscle differentiation. Such effects are important to distinguish when the complicated machinery of skeletal muscle formation is sorted out. Chapter 13 by Bjerregaard et al. introduces syncytin-1 as a additional new player in myoblast fusions. Chapters 14 and 15 introduce cell fusions in the therapeutic and pathogenetic scenery. In Chapter 14, Silk et al. describes that BMDCs are capable of fusing heterotypically with different types of stem cells and considers the physiologic and pathologic implications of the high degree of fusogenicity of stem cells. Heterotypic fusions between stem cells and other cells will result in hybrids that exhibit nuclear reprogramming in a context that may be useful for normal tissue repair but also may induce chromosomal aberrations, aneuploidy, in activation of tumor suppressor genes or activation of oncogenes and genomic instability in the context of cancer (Duelli et al. 2007). With the exception of the sperm-egg fusion, physiologic cell–cell fusions are homotypic since they occur between cell types of the same lineage. However, fusions between different cells – although of the same lineage – are seen during skeletal muscle formation. In Drosophila, such fusions occur between founder cells and fusion competent myoblasts and, in mammals, fusions between myoblasts resulting in nascent myotubes are followed by
1
Regulation and Control of Cell–Cell Fusions
7
fusions between myotubes and myoblasts. Moreover, during skeletal muscle repair, fusions occur between satellite cells and mature muscle fibers (Chapters 6 and 12). Homotypic fusions usually result in cells, which are multinucleated and without proliferative ability. In the last chapters, heterotypic fusions are considered. Such fusions have been shown to occur in vitro as well as in vivo and incontrovertible evidence for this is presented in Chapters 14, 16 and 17. Heterotypic fusions result in initial formation of heterokaryons, containing two or more nuclei. Such heterokaryons formed by tumor-host cell fusion are able to undergo a subsequent round of cell division which dissolves the individual nuclear envelopes and, thus, mixes the genomes (Mortensen et al. 2004). The ensuing daughter cells are synkaryons, containing one nucleus each, which stores the mixed parental genomes. During division of polyploid cells, formed by fusion, there is a high risk of missorting and damaging of chromosomes and it has been considered that such events may predispose to increased malignancy or to PCD (Duelli et al. 2007). In Chapter 17 Lazova et al. provides numerous examples of how fusions between e.g. macrophages and melanoma cells may lead to hybrids of increased malignancy that express markers characterizing both parental cells. The best example of stable inheritance of parental characteristics by hybrids are hybridoma cells (Kohler and Milstein 1975), which have found widespread use for unlimited production of monoclonal antibodies. Another promising way of exploiting cell fusions is introduced by Gong in Chapter 15, where the production of hybrids between dendritic cells (macrophages which are very good at presenting antigens to the immune system) and cancer cells is described. Such hybrids are used clinically and experimentally for harnessing the immune response towards malignant tumors. The technique has so far produced useful results in animal models and shown promise in clinical settings. In this chapter we depart from the discussion of spontaneous hybrid formation and are taught different methods for creating artificial hybrids between dendritic cells and cancer cells. It may be a sobering thought that hybrid yields can be very low even when all sails are set for inducing them! Moreover, in this setting hybrids do good by instructing the immune system to fight cancer. Tumor cell hybrids are also the focus of the Chapter 17 by Strick et al., who provide a historic overview and lists many examples of different human tumors containing multinucleated cells that may have formed either by fusion or endomitosis (defect cytokinesis). Finally, an overview of the expression of human endogenous retroviral (HERV) sequences, including syncytin, in different human tumors is presented with indications for future studies. Together, the chapters contained in the present volume present an overview of factors that have actions that transcend the cell specificity of cell–cell fusions as well as factors which confer cell specificity. It is hoped that it will enhance understanding of this multifactorial process, which include many components involving cell–cell recognition, migration, signaling, cytoskeletal reorganization, “synapseformation” and lowering the energy level for lipid bilayer fusion. By bringing all of these experts together I am trying to propagate the theory that we should look for both similarities and dissimilarities between the fusion processes in different cell types. Finally, the final chapters also address the consequences of cell–cell fusion
8
L.-I. Larsson
as a means to repair damaged tissue, as a possible danger for carcinogenesis and cancer progression but also as an aid to immune therapy of cancer, at least when exercised under controlled conditions. Acknowledgements Work by the author presented herein was supported by the Danish MRC, FTP and Lundbeck foundation.
References Andersen TL, Boissy P, Sondergaard TE et al (2007) Osteoclast nuclei of myeloma patients show chromosome translocations specific for the myeloma cell clone: a new type of cancer-host partnership? J Pathol 211:10–17 Bjerregaard B, Holck S, Christensen IJ et al (2006) Syncytin is involved in breast cancerendothelial cell fusions. Cell Mol Life Sci 63:1906–1911 Blond JL, Lavillette D, Cheynet V et al (2000) An envelope glycoprotein of the human endogenous retrovirus HERV-W is expressed in the human placenta and fuses cells expressing the type D mammalian retrovirus receptor. J Virol 74:3321–3329 Chen EH (2008) Cell fusion. Overviews and methods. Methods Mol Biol 475:1–421 Duelli DM, Hearn S, Myers MP et al (2005) A primate virus generates transformed human cells by fusion. J Cell Biol 171:493–503 Duelli DM, Padilla-Nash HM, Berman D et al (2007) A virus causes cancer by inducing massive chromosomal instability through cell fusion. Curr Biol 17:431–437 Dunlap KA, Palmarini M, Varela M et al (2006) Endogenous retroviruses regulate periimplantation placental growth and differentiation. Proc Natl Acad Sci USA 103:14390–14395 Dupressoir A, Marceau G, Vernochet C et al (2005) Syncytin-A and syncytin-B, two fusogenic placenta-specific murine envelope genes of retroviral origin conserved in Muridae. Proc Natl Acad Sci USA 102:725–730 Dupressoir A, Vernochet C, Bawa O et al (2009) Syncytin-A knockout mice demonstrate the critical role in placentation of a fusogenic, endogenous retrovirus-derived, envelope gene. Proc Natl Acad Sci USA 106:12127–12132 Heidmann O, Vernochet C, Dupressoir A et al (2009) Identification of an endogenous retroviral envelope gene with fusogenic activity and placenta-specific expression in the rabbit: a new “syncytin” in a third order of mammals. Retrovirology 6:107 Kohler G, Milstein C (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256:495–497 Mi S, Lee X, Li X, Veldman GM et al (2000) Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature 403:785–789 Miyado K, Yamada G, Yamada S et al (2000) Requirement of CD9 on the egg plasma membrane for fertilization. Science 287:321–324 Mortensen K, Lichtenberg J, Thomsen PD et al (2004) Spontaneous fusion between cancer cells and endothelial cells. Cell Mol Life Sci 61:2125–2131 Muroi Y, Sakurai T, Hanashi A et al (2009) CD9 regulates transcription factor GCM1 and ERVWE1 expression through the cAMP/protein kinase A signaling pathway. Reproduction 138:945–951 Oren-Suissa M, Podbilewicz B (2007) Cell fusion during development. Trends Cell Biol 17:537–546 Parthasarathy V, Martin F, Higginbottom A et al (2009) Distinct roles for tetraspanins CD9, CD63 and CD81 in the formation of multinucleated giant cells. Immunology 127:237–248 Rothman JE (1994) Mechanisms of intracellular protein transport. Nature 372:55–63 Sala-Valdés M, Ursa A, Charrin S et al (2006) EWI-2 and EWI-F link the tetraspanin web to the actin cytoskeleton through their direct association with ezrin-radixin-moesin proteins. J Biol Chem 281:19665–19675
1
Regulation and Control of Cell–Cell Fusions
9
Simons K, Ikonen E (1997) Functional rafts in cell membranes. Nature 387:569–572 Strick R, Ackermann S, Langbein M et al (2007) Proliferation and cell–cell fusion of endometrial carcinoma are induced by the human endogenous retroviral Syncytin-1 and regulated by TGFbeta. J Mol Med 85:23–38 Weng J, Krementsov DN, Khurana S et al (2009) Formation of syncytia is repressed by tetraspanins in human immunodeficiency virus type 1-producing cells. J Virol 83:7467–7474 Zhou Z (2007) New phosphatidylserine receptors: clearance of apoptotic cells and more. Dev Cell 13:759–760
Chapter 2
Retroviruses and Cell Fusions: Overview Anders L. Kjeldbjerg, Shervin Bahrami, and Finn Skou Pedersen
Abstract Retroviruses are a large and diverse group of enveloped animal viruses. A metastable envelope protein (ENV) on the surface of virus particles harbors a machinery for receptor-dependent fusion of biological membranes as needed for viral entry. The basic mechanism that drives fusion is widely conserved among different groups of retroviruses, whereas the precise signals that trigger the activation of this machinery vary. The exact same processes that drive viral entry may also mediate cell–cell fusion in a receptor-dependent manner. Such fusion events that may lead to the formation of giant multinucleated cells have been widely observed in cultured cells exposed to retroviruses. However, their possible contribution to the spread and pathogenesis of retroviral infections in man and animals is unclear. By way of their mode of replication via a DNA-intermediate that is stably integrated in the chromosomal DNA of the host cell, retroviruses may also establish germ-line infections that can be vertically transmitted from parents to offspring. Such remnants of retroviral infections of our ancestors constitute 8% of the human genome. Some of these human endogenous retroviruses of more than 25 million years of age have selectively maintained the coding capacity for functional envelope proteins, which provides strong evidence that these envelope genes have been co-opted to serve a beneficial function for their host. Currently, three of these old envelope genes have been found to encode proteins that can mediate cell–cell fusions and at least two of the envelope proteins have been implicated in the generation of a multi-nucleated layer of cells in the placenta. Keywords Retrovirus · exogenous retrovirus · endogenous envelope · entry receptor · receptor interference · fusion inhibitors · HERV Abbreviations AIDS ALV
Acquired immune deficiency syndrome Avian leukosis virus
F.S. Pedersen (B) Department of Molecular Biology, Aarhus University, DK-8000 Aarhus C, Denmark e-mail:
[email protected] L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_2, C Springer Science+Business Media B.V. 2011
11
12
ASCT DC-SIGN enJSRV ENV ERV FeLV GLUT1 HA HERV HIV-1 HTLV-1 HYAL2 ISU JSRV LTR mCAT-1 MFSD2 MMTV MLV MPMV Mya NHR CHR ORF Pit PRR RBD RSV SFV SIV Smit-1 SNP SP SU TM WDSV
A.L. Kjeldbjerg et al.
Alanine, serine and cysteine selective transporters Dendritic cell-specific ICAM-3-grabbing nonintegrin Endogenous JSRV Envelope protein Endogenous retrovirus Feline leukemia virus Glucose transporter 1 Hemagglutinin Human endogenous retrovirus Human immunodeficiency virus type 1 Human T-cell lymphotropic virus type 1 Hyaluronidase 2 Immunosuppressive domain Jaagsiekte sheep retrovirus Long terminal repeat Mouse cationic amino acid transporter Major facilitator superfamily domain containing 2 Mouse mammary tumor virus Murine leukemia viruse Mason-Pfizer monkey virus Million years ago Coiled-coil – N- and C-terminal heptad repeat Open reading frame Sodium-dependent phosphate symporter Proline-rich region Receptor-binding domain Rous sarcoma virus Simian foamy virus Simian immunodeficiency virus Sodium-dependent myo-inositol transporter 1 Single nucleotide polymorphism Signal peptide Surface subunit Transmembrane subunit Walleye dermal sarcoma virus
Contents 2.1 Introduction . . . . . . . . . . . . . . . . . . 2.2 Basic Features of the Retroviral Fusion Machinery 2.2.1 Entry . . . . . . . . . . . . . . . . . . 2.2.2 Receptors . . . . . . . . . . . . . . . . 2.2.3 Interference . . . . . . . . . . . . . . . 2.2.4 Membrane Fusion . . . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
13 15 15 15 16 17
2
Retroviruses and Cell Fusions: Overview
2.2.5 Fusion Inhibitors . . . . . . . . . . . . . . . . . 2.2.6 The Significance of the Coiled Coil Structures . . . 2.2.7 Control Mechanism for Fusion Activation . . . . . 2.2.8 Retroviral Fusion of Cells . . . . . . . . . . . . . 2.3 Fusion Control in Different Groups of Retroviruses . . . . 2.4 Endogenous Retroviruses and Cell Fusion . . . . . . . . 2.4.1 Origin and Classification of Endogenous Retroviruses 2.4.2 Evolutionary View of HERV env Genes . . . . . . 2.4.3 Structural Composition of HERV Envelope Proteins . 2.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . .
13
. . . . . . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
19 19 20 21 22 26 26 27 31 32 33
2.1 Introduction Retroviruses comprise a large and diverse group of viruses whose main characteristic is that their genomic information is transmitted in the form of two copies of an RNA molecule that upon infection of a target cell is copied into a DNA molecule, which in turn will be integrated into the host genome. As all viruses, retroviruses are dependent on the host cell machinery for replication. The replication cycle of a retrovirus (Fig. 2.1a) begins by entry of a viral particle into an appropriate cell, at which stage a DNA copy of the genomic RNA is completed. The DNA copy enters the nucleus and becomes integrated into the host genome. In some viral species, degradation of the nuclear membrane during cell division is necessary for the access of the viral integration machinery to the host genome. After integration, the provirus (as the integrated viral DNA is called) will remain a part of the cellular genome, which will be inherited by the daughter cells. At this stage the transcription/translation machinery of the host cell is employed to produce more viruses (Fig. 2.1a). All retroviruses contain three vital genes: gag, which encodes the structural proteins, pol, which encodes the enzymes and env, which encodes the envelope protein (Fig. 2.1b). Gag and pol encode polyproteins that form the particle which eventually buds off the producer cell. Once inside a budded virion, these polyproteins are cleaved by the viral protease enzyme to yield the individual and functional proteins/enzymes in a process called maturation. Some retroviruses such as human immunodeficiency virus type 1 (HIV-1), the causative agent of AIDS, harbor additional genes that encode proteins with regulatory roles. If the integration event happens in cells of the germ line, the provirus will become an integrated part of every cell of the offspring. Hence a large part of the vertebrate genomes are comprised of remains of retroviral integrations that are vertically transmitted as endogenous retroviruses (ERVs). This chapter reviews the basic mechanisms whereby retroviruses mediate the controlled fusion of two biological membranes needed for retroviral entry, and illustrate how these same mechanisms may drive cell–cell fusion. These sections are mainly based upon features of the retroviral genus of gammaretroviruses, including prototypic murine leukemia viruses (MLVs).
14
A.L. Kjeldbjerg et al.
Fig. 2.1 Panel a Schematic representation of a retroviral replication cycle. The first steps are entry by attachment to a surface receptor (step 1) followed by fusion (step 2). The subsequent reactions of reverse transcription (step 3) and integration (step 4) depend upon viral enzymes whereas the later steps of transcription, RNA splicing, and translation (step 5) that make the building blocks for new viruses depend upon the macromolecular synthesis machinery of the host cell. Finally, viral RNA and proteins are packed into particles that bud off the host cell (step 6) to yield a new virus particle surrounded by a lipid bilayer envelope. The maturation (step 7) that involves proteolytic cleavage by a virus-specific protease is needed for infectivity of the virus particle. Panel b Genetic map of an integrated retrovirus in the DNA form. Indicated are the three protein-encoding genes gag, pol, and env as well as nucleic acids motifs needed for the retroviral replication cycle
Then, examples are provided of the diversity of regulation of membrane fusion among retroviruses, and subsequently the involvement of retroviral infection in causing cell–cell fusion is discussed. In the final part of the chapter the fusiogenic properties of envelope genes of human endogenous retroviruses (HERVs) are summarized and their possible beneficial role for their human hosts discussed.
2
Retroviruses and Cell Fusions: Overview
15
2.2 Basic Features of the Retroviral Fusion Machinery 2.2.1 Entry Since retroviruses are enveloped viruses (which means that the virus particle is surrounded by a lipid membrane), the viral core is topologically in the same space as the cellular cytoplasm. Therefore fusion of the viral and cellular membranes is sufficient to grant the virus access to the inside of the cell. Fusion of lipid membranes does not happen spontaneously, and an active mechanism for fusion is needed. In the case of retroviruses, this is provided by the envelope protein, which is responsible for both binding of the virus to the appropriate target cell and fusion of the viral and cellular membranes. The specificity of the binding of the envelope protein to its cellular receptor is a major determinant of the viral tropism. This specificity is found in one of the two subunits that make the envelope protein, the surface subunit or SU. The other subunit, the transmembrane subunit or TM is anchored in the viral membrane by a transmembrane helix and is responsible for the fusion of the membranes upon binding of SU to its receptor. Both subunits are products of a single viral gene, which is cleaved by a host cell protease in the endoplasmic reticulum. The envelope protein is a trimer held together by interaction of the TM subunit.
2.2.2 Receptors Most viral receptors are membrane-integral proteins in the plasma membrane, encoded by the host cells, while a few viruses such as jaagsiekte sheep retrovirus (JSRV) use protein receptors anchored in the plasma membrane by lipid anchors (Miller 2003, 2008). Retroviruses, even closely related ones, use different cellular receptors. For example, murine leukemia viruses can be divided into six subgroups, ecotropic, polytropic, amphotropic, 10A1, xenotropic and M813, based on their receptor usage. Most viral receptors have several predicted transmembrane helices and in the case of MLVs function as transporters. Ecotropic MLVs utilize the mouse cationic amino acid transporter (mCAT-1) receptor and are unable to infect non-murine cells (Albritton et al. 1989, 1993, Wang et al. 1991). Amphotropic viruses use the sodiumdependent phosphate symporter Pit-2 (Kavanaugh et al. 1994) and are able to infect cells from a variety of species including humans. The 10A1 virus, closely related to the amphotropic viruses, uses two different receptors: both the amphotropic Pit-2 and the homologous Pit-1 receptors (Miller and Miller 1994) and has a similarly wide tropism. Xenotropic and polytropic viruses utilize the same receptor XPR !1 (Battini et al. 1999, Tailor et al. 1999, Yang et al. 1999) but differ in species tropism. A relatively new member of the MLV family (M813) isolated from the Southeast Asian rodent Mus cervicoloruses a sodium-dependent myo-inositol transporter 1 (Smit-1) as the cellular receptor (Hein et al. 2003, Prassolov et al. 2001). Other retroviruses, such as HIV-1, utilize a receptor for binding to the cells, but need a co-receptor for entry. The primary receptor for HIV is CD4, the natural ligand
16
A.L. Kjeldbjerg et al.
for the major histocompatibility complex class II molecule and an important factor in the generation of immune responses (Dalgleish et al. 1984, Klatzmann et al. 1984). Expression of CD4 mediates binding of the HIV virus to cells but is not sufficient for HIV infection (Chesebro et al. 1990). Other membrane proteins besides CD4, also called co-receptors, are necessary for HIV entry. Several receptors for chemokines act as co-receptors for HIV, the most important ones are CXCR4 and CCR5 (Alkhatib et al. 1996, Choe et al. 1996, Endres et al. 1996).
2.2.3 Interference Interaction of the viral envelope protein and a cellular receptor is absolutely vital for entry of the virus. Virus-infected cells are resistant to re-infection by the viruses utilizing the same receptor. A receptor can be blocked through interaction with a viral envelope protein that is expressed inside the cell (Fig. 2.2). The envelope protein binds to the receptor on the plasma membrane, thereby blocking it from interaction with viruses in the environment. This phenomenon, called receptor interference or superinfection resistance, is used by retroviruses to protect their host cells from invasion by other viruses. Mostly as a result of the aforementioned receptor interference, similar viruses have evolved to use either different receptors or even different sites on the same receptor protein. One interesting group is the polytropic/xenotropic murine leukemia viruses which belong to the gammaretroviral group. All of these viruses use the XPR1 protein as the entry receptor, but are sensitive to small variations found on the XPR1s from different species. For example the xenotropic viruses are unable to use the murine version of this receptor, but are fully functional when encountering receptors from other species. For example a xenotropic-like virus has
Fig. 2.2 Blocking of retroviral entry by receptor interference. Panel a A retrovirus binds via the envelope protein (red) to its cognate receptor (black) as the first step in the entry process. Panel b Envelope protein (red) expressed in the target cell blocks the receptor (black) and prevents entry of a new virus
2
Retroviruses and Cell Fusions: Overview
17
been implicated in prostate cancer in humans (Dong et al. 2007, Urisman et al. 2006). On the other hand polytropic viruses have a broad species tropism and can use both murine and non-murine XPR1 receptors. Interestingly, these viruses show non-reciprocal interference with each other, meaning that they can block each other with different efficiencies and even bind to different parts of the same receptor protein (Van Hoeven and Miller 2005). Few mutations in the envelope protein of polytropic MLVs can have dramatic effect on the species tropism and interference pattern of the virus (Bahrami et al. 2004). In the lentivirus HIV-1, blocking of the receptor is more elaborate. In HIV infection, receptor/envelope complexes are retained in the endoplasmic reticulum, and the viral accessory proteins Nef and Vpu stimulate the degradation of the CD4 receptor, thus the receptor expression on the cell surface is diminished (Crise and Rose 1992, Fackler and Baur 2002, Levesque et al. 2003, Willey et al. 1992a, b). Knowledge of envelope/receptor interaction has been utilized to design coreceptor blockers, a class of anti-HIV entry inhibitors that have affinity for HIV co-receptors (Shaheen and Collman 2004).
2.2.4 Membrane Fusion Membrane fusion is an integral part of many important biological activities, such as intracellular transport, endo/exocytosis or fertilization of an egg by sperm. The machineries that mediate these processes, although different, share some important characteristics, dictated by the chemistry of lipids, the main component of biological membranes (White et al. 2008) Lipid molecules of biological membranes are amphipathic. They contain a hydrophilic, usually charged head group and a hydrophobic tail of hydrocarbons. Upon exposure to water, the hydrophobic tails cluster together in what is essentially a hydrophobic environment, while the hydrophilic head groups are oriented towards water. In biological membranes, two layers of parallelly oriented lipid molecules with their tails pointing towards each other form a lipid bilayer. This arrangement constitutes the thermodynamically most stable conformation of amphipathic molecules by exposing the hydrophilic head of the molecules to the aqueous environment, while masking the hydrophobic parts from it. Fusing two membranes necessarily involves bringing two lipid bilayers into close proximity. This involves expulsion of water from hydrophilic surfaces and opposing the electrostatic repulsion from between equal charges, both of which are energetically unfavorable processes. For this reason, membrane fusion is not a spontaneous process and requires input of free energy. One of the best studied membrane fusion processes is that mediated by the influenza envelope protein hemagglutinin (HA). Unlike retroviral envelope proteins, the crystal structure of the hemagglutinin protein is known both in pre and post fusion states (Bullough et al. 1994, Wilson et al. 1981). Because of striking structural similarities between HA and retroviral ENV, it is feasible to use HA-mediated fusion as a model for retroviruses. Similarities include the presence of two subunits in both HA and ENV, one responsible for binding the receptor (HA1 and SU) and
18
A.L. Kjeldbjerg et al.
one for mediating fusion (HA2 and TM) (Katen et al. 2001, McClure et al. 1990, Nussbaum et al. 1993). While structural deviance between HA1 and SU reflects the usage of greatly different molecules as receptors, the very similar structures of HA2 and TM reflect the functional requirements that both proteins must fulfill in order to facilitate membrane fusion. HA2, like TM, has a hydrophobic stretch of amino acids at its amino-terminus followed by a heptad repeat region known to form coiled coils. Coiled coil structures are intertwined alpha helices stabilized through hydrophobic interactions on the contact side of each helix. This hydrophobic contact surface is made of the third and seventh amino acids in a repetitive sequence. The twisting of an α-helix brings these amino-acid residues to the same surface of the helix in the secondary structure. Thus
Fig. 2.3 Membrane fusion by the hemagglutinin protein (HA) of influenza virus. Panel a shows the structure of the trimer of HA2 at neutral and acidic pH, respectively. Panel b is a schematic diagram that illustrates steps of the fusion process mediated by structural changes in the influenza hemagglutinin caused by reduction in pH. Step I, The HA1 subunit binds to its receptor, wherafter the virus is internalized by endocytosis. Step II, Low pH in the endosome triggers the structural changes in HA2 shown in panel a, which leads to insertion of the amino-terminal fusion peptide into the target membrane. Step III, The helices of HA2 forld back on themselves and bring the two target membranes in close proximity. Step IV, hemifusion, which allows mixing of lipids in the two outer leaflets, but not the innter leaflets. Step V, the creation of a fusion pore through which the virus can enter the cytosol. See text for details and color schemes
2
Retroviruses and Cell Fusions: Overview
19
the primary structure of coiled coil helices is characterized by repeated hydrophobic amino acids at the third and seventh positions often referred to as a heptad repeat (Creighton 1996). As will be discussed shortly, activation of the fusion machinery must be very precisely timed. In the case of HA, fusion is triggered by the acidic pH encountered in endosomes, while most retroviruses are not dependent on pH change (Skehel et al. 1982, White et al. 2008). Fusion mediated by the influenza HA involves several sequential and extensive conformational changes induced by acidic pH (Fig. 2.3). At neutral pH the carboxyterminal portion of the HA2 heptad repeat regions (Fig. 2.3, cyan) form a coiled coil which folds back onto itself so that the amino-terminal portion (Fig. 2.3, yellow) is oriented towards the viral membrane and the fusion peptides (Fig. 2.3, black) are buried in the trimeric structure. Infection initiates when the virus binds to the target cell and is internalized into the endosomes where the pH is lower. This change in pH induces a dramatic conformational change in the HA2 protein so that the loop structure (green) that connects the amino and carboxy-terminal helices in the prefusion structure assumes a helical structure (Fig. 2.3). This extends the triple helix and inserts the hydrophobic fusion peptide into the target membrane. The process is followed by formation of a six-bundle helix where the carboxy-terminal helices fold back to form the so-called “hairpins”. This pulls the membranes together to form a hemifusion state (where only the outer leaflets of the membranes are fused) followed by full fusion and formation of a pore through which the viral core can enter the cytoplasm (Hunter 1997, White et al. 2008).
2.2.5 Fusion Inhibitors This detailed insight into the mechanism of the fusion has given rise to the rational design of fusion inhibitors. One example is the anti-HIV drug Enfuvirtide. Enfuvirtide or T20 is a peptide corresponding to the carboxy-terminal portion of the triple-helix of HIV-1 envelope protein. As illustrated in Section 2.2.4, this portion of the helix will ultimately fold back into the triple-helix structure to form the stable six-bundle helix conformation that brings the membranes close together. When added into the environment, it can interact with the intermediate (the long triple helix) conformation and inhibit formation of the final six bundle helix, thus inhibiting viral entry (Fig. 2.4) (Greenberg et al. 2004). By the same principle, specific peptide inhibitors of fusion have been made for other retroviruses (Chang et al. 2004, Lamb et al. 2008, Netter et al. 2004).
2.2.6 The Significance of the Coiled Coil Structures Membrane fusion is accompanied by conformational changes, which result in formation of coiled coil structures and hairpins. Triple-hairpin structures are remarkably stable structures, dissociating at temperatures around 90◦ C (Fass and Kim
20
A.L. Kjeldbjerg et al.
Fig. 2.4 The point of action of peptide inhibitors of retroviral fusion. The peptide inhibitor (red) is dervied from the carboxy-terminal heptad repeat (light grey) and has affinity for the aminoterminal heptad repeat (dark grey). The inhibitor blocks the association of the two heptad repeats and thereby hinders the transition from the triple-helix structure to the six-helix bundle as needed for fusion
1995, Lu et al. 1995), which suggests that a large amount of free energy is released upon their formation. Presumably the released energy is necessary to bring the membranes into close proximity and drive their destabilization which ultimately results in fusion (Bentz and Mittal 2000, White 1992). Consequently, the fusion trigger is a “one-time-only” event for every molecule: The change in the free energy of the reaction is too large for any equilibrium to exist between the “fusion-potent” and “fusion-active” conformations of the fusion proteins. Thus the fusion trigger must be timed precisely. In the case of viral fusion it must occur after the binding of the virion to target cells. As mentioned in Section 2.2.4, the trigger can be either a pH change after internalization or conformational changes in the binding subunit of the viral fusion protein upon interaction with the receptor.
2.2.7 Control Mechanism for Fusion Activation One such mechanism that is involved in control of fusion activation in gammaretroviruses is a disulfide isomerization that occurs when the envelope binds to its receptor. In these viruses, the SU and TM subunits are bound together by a disulfide bridge. One of the cysteine residues involved is found in a CXXC motif reminiscent of the active sites of protein-disulfide isomerases. In order for the fusion process to go forth, it is necessary that these two subunits dissociate. This happens when
2
Retroviruses and Cell Fusions: Overview
21
binding to the receptor presumably results in a conformational change that brings a crucial histidine residue in the right position so that a disulfide isomerase potential in the SU subunit is activated. The result is that the inter-subunit disulfide bond is broken, and one formed between the two cysteines in the CXXC motif, resulting in the dissociation of the SU from TM (Li et al. 2007, Wallin et al. 2004). This dissociation step can be facilitated by Ca2+ depletion (Wallin et al. 2004). Another interesting control mechanism found in gammaretroviruses and some other retroviruses is the so-called R-peptide. As mentioned before, the envelope protein is found interacting with the receptors on the surface of the cells where it is expressed (Fig. 2.2). Yet this interaction does not result in triggering the fusion machinery, in which case the energy stored in the envelope protein would be wasted. The reason is that the cytoplasmic tail of envelope molecules, the R-peptide, locks the fusion machinery and prohibits the cascade of conformational changes, probably by distorting the TM trimers by pulling them out of position through the R-peptides affinity for membrane lipids. Cleavage of the R-peptide during maturation of the viral particles by the viral protease releases the pressure on the individual TM subunits and enables them to form the triple helix bundle necessary for fusion (Olsen and Andersen 1999). Interestingly, addition of the R-peptide to fusion proteins of non-retroviral origin has a similar effect (Li et al. 2006).
2.2.8 Retroviral Fusion of Cells The exact same machinery that mediates the fusion of a viral and a cellular membrane may also cause fusion of cells in a receptor-dependent manner leading to the formation of giant multi-nucleated syncytia (Fig. 2.5). Similar to retroviral entry, cell fusion can also be inhibited by receptor interference caused by expression of an
Fig. 2.5 A syncytium of HEK 293 cells caused by murine leukemia virus (unpublished result from the authors’ laboratory)
22
A.L. Kjeldbjerg et al.
envelope protein. Hence, fusion of two cells may require that one of them is noninfected and thereby has an accessible receptor. Two distinct processes have been identified, fusion from within and fusion from without (Andersen 1994, Siess et al. 1996). Fusion from within is driven by viral envelope gene expression after integration of the provirus, which results in expression of the envelope protein on the cell surface whereas fusion from without is driven by envelope proteins on virus particles. In contrast to fusion from within, fusion from without does not require viral gene expression in the target cell and can be observed during the first hours after addition of virus to an uninfected cell culture (Andersen 1994). Not unexpectedly, deletion of the R-peptide enhances the efficiency of cell–cell fusion by MLV (Rein et al. 1994). Newer results indicate that envelope-receptor interaction may serve to guide and establish 3D-patterns of cell–cell contacts and suggest that such contacts involving fusiogenic envelope proteins may lead to the formation of intercellular bridges (Jin et al. 2009, Sherer and Mothes 2008). Other than abundance and post-translational modification of envelope protein and receptor, the factors that contribute to the large variability in fusion potential among cell lines are poorly understood. One contributing factor is the presence of other molecules that facilitate cell–cell interactions (Ceccaldi et al. 2006, Pantaleo et al. 1991). Retroviruses are reported to bud from lipid rafts rich in cholesterol as also reflected by the lipid composition of the membrane enveloping the virus (Metzner et al. 2008). It is conceivable that certain membrane patches may be more amenable to fusion than others, and that such variation also contributes to variability in fusiogenicity among cells (Lorizate et al. 2009). XC-cells are a rat-muscle tumor-derived cell line that is particularly susceptible to the induction of syncytia by ecotropic MLVs (Klement et al. 1969). In these cells fusion from within is efficient even in the case of envelope proteins that have retained the R-peptide. It has been suggested that the fusion of XC-cells may involve a cellular factor with a role in membrane fusion in myogenesis (Kubo et al. 2003).
2.3 Fusion Control in Different Groups of Retroviruses The activation of the metastable envelope protein of retroviruses is tightly regulated. It is not surprising that there has been a strong evolutionary selection for such precise regulation in diverse biological settings since premature activation would lead to loss of viral infectivity. While the overall membrane-fusion machinery is quite similar, the exact modes of control, which are now beginning to be uncovered, show wide specialization for different groups of retroviruses. Such control mechanisms also impact on the regulation of the cell–cell fusiogenic activity of retroviral envelope proteins in the plasma membrane of an infected cell. The family Retroviridae comprises seven genera, alpharetroviruses, betaretroviruses, gammaretroviruses, deltaviruses, epsilonretroviruses, lentiviruses, and spumaviruses (Fig. 2.6). While alpharetroviruses are restricted to birds and epsilonretroviruses to fish, the remaining five genera are all represented in
2
Retroviruses and Cell Fusions: Overview
23
Fig. 2.6 Classification of the seven genera of Retroviridae and their three classes of related human endogenous retroviruses (HERVs), class I, class II, and class III. Avian leukosis virus (ALV), Rous sarcoma virus (RSV), mouse mammary tumor virus (MMTV), Mason-Pfizer monkey virus (MPMV), jaagsiekte sheep retrovirus (JSRV), human immunodeficiency virus type 1 (HIV-1), simian immunodeficiency virus (SIV), human T-cell lymphotropic virus type I (HTLV-1), simian foamy virus (SFV), walleye dermal sarcoma virus (WDSV), murine leukemia virus (MLV), feline leukemia virus (FeLV)
mammals. If endogenous retroviruses (ERVs) are included together with their normal (exogenous) counterparts all of those five genera, betaretroviruses, gammaretroviruses, deltaretroviruses, lentiviruses and spumaviruses are represented in primates including humans. Cell–cell fusion has been documented for all genera of the family Retroviridae with the possible exception of epsilonretroviruses. The functional organization of envelope protein of selected exogenous retroviruses is shown in Fig. 2.7. The regulation of the membrane-fusion machinery of envelope proteins of gammaretroviruses has been intensively studied as described in Section 2.2. The prototype virus of this group is murine leukemia virus, but the same principles hold for closely related viruses of other species such as cats and primates. There is no unifying overall picture as to the requirement for acidic pH during the entry step of gammaretroviral infections, however there is ample evidence that several gammaretroviral envelope proteins can mediate cell–cell fusion at neutral pH (Andersen 1994). As mentioned, one hallmark of gammaretroviral envelope proteins is the covalent linkage of SU and TM by a disulfide bond and the dissociation of SU from TM as a result of a disulfide isomerization reaction during the steps towards membrane fusion. Another characteristic feature of this group, the negative regulation
24
A.L. Kjeldbjerg et al.
Fig. 2.7 Schematic organization of envelope motifs involved in the regulation of the fusion. C(X)6 CC, grey line, involved in a disulfide bond between TM and the CXXC motif in SU. SP: Signal peptide; SPH: histidine motif; RBD: Receptor-binding domain; PRR: Proline-rich region; FP: Fusion peptide. ISU: immunosuppressive domain (black box); NHR CHR: Coiled-coil – amino- and carboxyl-terminal heptad repeat structures (green); Transmembrane domain (gray box). The disulfide partner in SU of ALV is not identified. HIV-1 contains a C(X)6 C motif, however, it is not known whether it is involved in a disulfide bond. In HTLV only one heptad repeat is identified
of fusion by the carboxy-terminal R-peptide in the cytoplasmic tail of TM, which is cleaved off by the viral protease after budding, is also shared by Mason-Pfizer monkey virus, a member of the betaretroviruses genus which has been widely used experimentally to trigger cell–cell fusion (Duelli et al. 2005). The genus of deltaretroviruses includes the important human pathogen human T-cell lymphotropic virus type 1 (HTLV-1) as well as related virus isolates from simians and humans (Lairmore and Franchini 2007). HTLV-1 is associated with adult T-cell leukemia and other severe human diseases. Progression to clinical disease generally takes decades following infection and there is a lack of knowledge on the mode of virus replication and onset of disease. HTLV-1 particles are poorly infectious and believed to spread mainly through specialized cell to cell contacts termed virological synapses (Igakura et al. 2003). The envelope protein of HTLV-1 has a disulfide bridge between SU and TM and undergoes disulfide isomerization after activation by a process dependent upon similar motifs to those of MLV (Li et al. 2008) (see Fig. 2.7). The entry receptor for HTLV-1 was recently found to be the glucose transporter 1 (GLUT1) (Manel et al. 2003). GLUT1 expression is not limited to T-lymphocytes, which are the primary target for adult T-cell leukemia, and it is likely that HTLV-1 may also infect other cells in vivo. HTLV-1 is an efficient inducer of syncytia in cultured cells (Paré et al. 2005) in a process that takes place at neutral pH and requires the viral envelope protein and the receptor. Additional cell– cell interactions may facilitate fusion by HTLV-1. It has been found that the dendritic cell-specific lectin DC-SIGN (dendritic cell-specific ICAM-3-grabbing nonintegrin) will facilitate fusion with HTLV-1 infected T-cells as a result of interaction of this
2
Retroviruses and Cell Fusions: Overview
25
lectin with ICAM molecules on the T-cells (Ceccaldi et al. 2006). It is not known to what extent cell fusion is triggered by HTLV-1 infections in man, and if cell fusion plays a role in viral pathogenesis. A unique mechanism of fusion activation in two steps that use a combination of receptor-binding and low pH for activation was first discovered for avian leukosis virus (ALV) of the alpharetroviruses group (Barnard et al. 2006, Mothes et al. 2000). In contrast to other retroviruses, the fusion protein of ALV is not located at the very amino-terminus of TM (Fig. 2.7). The first step in the entry process takes place after envelope binding to the receptor at neutral pH at the cell surface. Receptor binding leads to a conformational change in the envelope protein that exposes the fusion peptide to become inserted into the target membrane. At this stage the receptor-primed intermediate conformation of the envelope protein is susceptible to the action of a fusion-inhibitory peptide derived from the carboxy-terminal heptad repeat of TM (Netter et al. 2004). However, progression to full fusion requires acidic pH. In accordance with this model cell–cell fusion mediated by the envelope protein of ALV also requires as well the cognate receptor as acidic pH. Related mechanisms in which both receptor interaction and acidic pH are required for activation of the fusion machinery have recently been proposed for the betaretroviruses mouse mammary tumor virus (MMTV) (Wang et al. 2008) and jaagsiekte sheep retrovirus (JSRV) (Côté et al. 2009). For both of these viruses cell–cell fusion was shown to require receptor presence as well as acidic pH. Lentiviruses such as HIV-1 infect only specialized target cells by an entry process that requires interaction with a receptor as well as a co-receptor as outlined earlier in this chapter. The SU and TM subunits of HIV are not linked by a disulfide bond (Fig. 2.7). However, a host cell disulfide isomerase is implicated in fusion mediated by HIV (Papandreou et al. 2010). The fusion machinery of HIV does not need low pH for activation. The exact subcellular location where the fusion takes place has been a matter of debate, however it is now believed that receptor binding takes place at the plasma membrane and that most fusion events happen after endocytosis, but before acidification of the endosome (Miyauchi et al. 2009). HIV-1 is an inducer of cell fusion in cultured cells that express the cognate receptor and co-receptor (Ji et al. 2006). The regulation of cell–cell fusion has the same hallmarks as that of infection by HIV-1, i.e. it takes place at neutral pH and it can be inhibited by coreceptor blockers and fusion inhibitors (Ji et al. 2009). However, in some instances additional cell–cell interaction may have an impact on fusion activity (Pantaleo et al. 1991). Progression of an HIV-1 infection often leads to a shift in co-receptor usage from CCR5 to CXCR4 as a result of mutations that cause amino-acid changes in the V3-loop of SU (Shankarappa et al. 1999). The emerging CXCR4-tropic viruses are very efficient in the induction of syncytia and it has been proposed that killing of lymphocytes by cell fusion (Perfettini et al. 2005) contributes to disease progression, but this issue remains unresolved. The retroviral genus of spumaviruses or foamy viruses includes many isolates from mammals including primates. There is no clear association of any disease with infection by spumaviruses. Their mode of replication exhibits several distinctive features from that of other retroviruses (Linial 2007). Structural and functional aspects
26
A.L. Kjeldbjerg et al.
of the envelope protein of a foamy virus have been investigated, but no information on the identification of a cellular receptor for foamy virus entry is available (Duda et al. 2006). Replicating foamy viruses cause the appearance of syncytia and are highly cytopathic in cell culture. Efficient formation of syncytia requires the presence of viral capsid as well as envelope protein, presumably reflecting a need for a gag-envelope interaction (Pietschmann et al. 2000). Entry of foamy viruses has been suggested to involve a low-pH step and formation of syncytia can in fact be triggered by a short exposure to acidic pH (Picard-Maureau et al. 2003).
2.4 Endogenous Retroviruses and Cell Fusion 2.4.1 Origin and Classification of Endogenous Retroviruses If retroviral infection has taken place in the germ line, the provirus will be transmitted vertically throughout generations as an endogenous retrovirus (ERV). Each independent germ line infection event will define a novel ERV family, whose copy number may increase over an evolutionary time span, either by intracellular retrotransposition (Sverdlov 1998) or by extracellular re-infection (Belshaw et al. 2004), following the normal replication cycle of a retrovirus. Chromosomal duplication may also increase the copy number of an ERV family (Boeke and Stoye 1997, Kjeldbjerg et al. 2008). Each novel germ line infection may thereby result in a few to several hundred genomic copies throughout evolution (Löwer et al. 1996). However, the rate of ERV amplification is highest shortly after the primary infection, and decreases during evolution. This may be caused by either transcriptional inactivation of viral RNA due to proviral DNA methylation (Lavie et al. 2005, O’Neill et al. 1998), loss of viral receptor affinity which prevents reinfection (Boeke and Stoye 1997), or inactivation of the provirus as a consequence of insertions, deletions, accumulated mutations or editing during retro-transposition (Esnault et al. 2005, 2006). Recombination between the long terminal repeat (LTR) regions the form a direct repeat at the termini of the provirus can result in the excision of internal viral sequences, leaving a single LTR in the genome, termed solo LTR (Löwer et al. 1996, Roeder and Fink 1980, Rotman et al. 1984). Altogether sequences related to retroviruses represent approximately 8% of the human genome (Lander et al. 2001). Endogenous retroviruses (ERVs) are classified somewhat differently from exogenous retroviruses. ERVs are divided into three classes depending on their similarity to exogenous retroviruses (Fig. 2.6), originally based on endogenous retrovirus found in humans (HERV). The HERVs, which are related to gammaretroviruses, are classified as class I, the betaretroviruses related ERVs as class II, and the spumaviruses related ERVs within class III. Endogenous elements are also found in the epsilon-, alpha- and lentiviruses genus but none are human associated (Fig. 2.6) (Gilbert et al. 2009, Goff 2007, Herniou et al. 1998). ERV classification was originally based on sequence similarity between the primer binding site (PBS) in the ERV locus and the tRNA used to prime the reverse transcription process in the host (e.g.
2
Retroviruses and Cell Fusions: Overview
27
HERV-W has a PBS matching W; tryptophan) (Boeke and Stoye 1997), however this classification is inconsistent for some ERV groups, because two ERV primed by the same tRNA do not necessarily belong to the same phylogenetic family as classified by nucleic acid homology of other parts of their genomes. The majority of the HERVs invaded the human genome at least 25 million years ago (mya) (Boeke and Stoye 1997, Shih et al. 1991), after splitting of Old World and New World monkey lineages approximately 43 mya (Steiper and Young 2006), however the oldest identified HERV-loci seem to be approximately 100 million years (my) old (Katzourakis and Tristem 2005). Nevertheless members of the HERV-K family have been active in the human genome less than 100,000 years ago as demonstrated by insertional polymorphisms (Belshaw et al. 2005, Medstrand and Mager 1998, Turner et al. 2001). However, no present-day replication-competent HERVs have been described, even though fully intact members of the HERV-K group have been reported, encoding all the viral genes (Turner et al. 2001). HERV-K related virus-like particle have been observed in germ-cell tumours (Bieda et al. 2001, Boller et al. 1983, Kurth et al. 1983), and while the HERV-K113 locus was shown to be capable of producing intact particles of retroviral morphology in cell lines derived from human germ-cell tumours (Boller et al. 2008), they were not found to be infectious. Nevertheless, trans-complementation and recombination of human HERV-K loci can generate functional and infectious HERV-K virus particles, indicating that human cells still have the potential to produce infectious retrovirus particles (Dewannieux et al. 2006, Lee and Bieniasz 2007). In contrast, other mammalian species such as mouse, cat and pig, naturally active ERVs still exist that produce infectious particles (Boeke and Stoye 1997).
2.4.2 Evolutionary View of HERV env Genes Since infection by retroviruses is usually harmful, evolutionary selection may neutralize endogenized retroviruses of infected germ lines. Nevertheless, some ERV loci still contain an open reading frame for one or more viral genes. A survey of the human genome identified 29 viral env genes with an ORF longer than 500 amino acids (Villesen et al. 2004, de Parseval et al. 2003), however, some of them appear to be pol-env fusion proteins, leaving 19 real env genes within the human genome. The 19 env genes represent 10 distinct HERV families (Fig. 2.8). Transcriptome analysis shows that all 19 envelope genes are expressed in healthy tissues, with the placenta as the organ having the maximum expression level of numerous envelope genes (Aagaard et al. 2005, Blaise et al. 2005, de Parseval et al. 2003). The env genes of the youngest family HERV-K, which is not represented in our primate relatives, are also the most abundant, represented by six full-length envelope genes. Three different families are represented by the three oldest envelope genes found in all simian primates, syncytin 2, EnvPb1, and ENV-V2 (Fig. 2.8) (Aagaard et al. 2005, Blaise et al. 2003, Kjeldbjerg et al. 2008). Characteristic of these ancient HERV envelope genes is that the envelope genes are the only preserved viral genes in an otherwise degenerated HERV locus (Fig. 2.9). Altogether, the conservation and expression of the envelope genes have led to speculation that HERV envelope genes
28
A.L. Kjeldbjerg et al.
Fig. 2.8 Evolutionary view of HERV envelope loci. a Detection of HERV solo LTRs (Hughes and Coffin 2004). b Insertionally polymorphic. c Nonsense polymorphic sites in human population (de Parseval et al. 1998) and not detected in gorilla. d Carboxy-terminal truncation. e Homologue locus in R. macaque found on chromosome 2, and a locus is identified in New World monkeys (de Parseval et al. 2005), but is not preserved
Fig. 2.9 Proviral organization of the HERV-Pb1 locus. The env gene is preserved while the gag and pol genes are broken down by insertions and deletions (Indel). Moreover, human reading frame destroying SNPs are found only outside env
2
Retroviruses and Cell Fusions: Overview
29
have been co-opted as a bona fide gene beneficial to the host. Hypothetical beneficial functions of such a gene include (i) protecting the fetus due to immunomodulatory properties via an immunosuppressive domain located in the TM subunit of the envelope (Mangeney et al. 2001, 2007), (ii) preventing present-day retroviral infections by inhibiting cell entry of related exogenous retroviruses that use a common surface receptor through receptor interference, as discussed in Section 2.2 (Best et al. 1997, Ponferrada et al. 2003), or (iii) providing cell–cell fusions by the mechanism discussed earlier in this chapter. In particular three HERV envelope proteins can induce cell–cell fusion in vitro, the HERV-W envelope gene syncytin 1 (Blond et al. 2000, Mi et al. 2000), the HERV-FRD envelope syncytin 2 (Blaise et al. 2003), and the HERV-Pb envelope EnvPb1 (Blaise et al. 2005) (Table 2.1). All three are candidates for having a beneficial function, because they are evolutionarily conserved and have undergone amino acid-conserving (purifying) selection during primate evolution (Aagaard et al. 2005, Blaise et al. 2003, Bonnaud et al. 2004). Syncytin 1 belongs to the HERV-W family and entered the genome of the common ancestor of the hominoids and Old World monkeys about 30 mya, however, it is inactivated in Old World monkeys (Fig. 2.8) (Bonnaud et al. 2005, Cáceres and Thomas 2006, Mallet et al. 2004). A systematic comparison of paralogue HERV-W env sequences reveals a syncytin 1-specific 12-bp deletion in the carboxy-terminal end that results in a four-amino-acid shortened cytoplasmic tail in the TM subunit of syncytin 1. This deletion is critical for the fusion activity of syncytin 1, by allowing syncytin 1 to mediate cell–cell fusion (Bonnaud et al. 2004), which might have played a major role in the domestication of syncytin 1 to become a bona fidegene. The deletion furthermore indicates that there may be sequences in the cytoplasmic tail of syncytin 1 that modulate its fusiogenicity in a way reminiscent of the function of the R-peptide in gammaretroviruses (see Section 2.2.7) (Ragheb and Anderson 1994, Yang and Compans 1996). However, R-peptide- like sequences cannot be identified in the cytoplasmic tail of syncytin 1, neither could any retroviral protease cleavage site be found (Blond et al. 2000). Additionally, a syncytin 1 orthologue gene is active in all hominoid species, and none of the single-nucleotide-polymorphisms (SNPs) in the human genome spoil its fusiogenic properties (de Parseval et al. 2005). The two other fusiogenic HERV envelope proteins syncytin 2 and EnvPb1 (Table 2.1) belong to the HERV-FRD and HERV-Pb families, respectively. Together with the non-fusiogenic ENV-V2 envelope they are the oldest preserved envelope
Table 2.1 HERV envelopes with cell–cell fusion capability Fusiogenic HERV-envelope
Classification
Expression
Receptor
Evolutionary age (my)
Syncytin 1 Syncytin 2 EnvPb1
Class I Class I Class I
Placenta Placenta Low expression in many tissues
ASCT1 ASCT2 MFDS2 Not known
20 40 40
30
A.L. Kjeldbjerg et al.
genes in the human genome. These entered the simian genome about 40 mya in a common ancestor to the Old and New World monkeys. Additionally, the fusiogenic potential of all simian syncytin 2 orthologues has been preserved throughout evolution, and no human SNPs alter the open reading frames of syncytin 2 and EnvPb1 (Aagaard et al. 2005, Blaise et al. 2003, Kjeldbjerg et al. 2008, de Parseval et al. 2005). The HERV-K family also contains a functional envelope protein, however, it does not seem to be capable of mediating cell–cell fusion, and only functions as a fusion protein during infection of pseudotyped virus particles (Dewannieux et al. 2005). The highest expression levels of HERV envelope genes are found in the placenta, where among others the two fusiogenic HERV envelope proteins syncytin 1 and syncytin 2 are highly expressed (Malassiné et al. 2005, 2007, Mi et al. 2000, de Parseval et al. 2003). This pattern may reflect a physiological role of the HERV envelope proteins in mediating cell–cell fusion in placenta by generating the multinuclear syncytiotrophoblast layer, which forms a barrier between the mother and the fetus. In fact, inhibition of syncytin 1 in cytotrophoblasts leads to a decrease in cell fusion in vitro (Frendo et al. 2003). The hypothesis of syncytin 1 and syncytin 2 involvement in syncytiotrophoblast formation is further supported by the finding that the receptors ASCT2 utilized by syncytin 1 and MFSD2 utilized by Syncytin 2 are expressed in cytotrophoblasts and syncytiotrophoblast (Esnault et al. 2008, Hayward et al. 2007, Lavillette et al. 2002). Another proposed function of ERV envelope proteins relate to immunosuppressive properties of the envelope protein, However only syncytin 2 among the syncytin genes has been shown to have an immunosuppressive activity (Mangeney et al. 2007). Three other HERV envelope genes also contain immunosuppressive activities: EnvPb1, ENV-V2 and ERV3 (Mangeney et al. 2007), among which ENV-V2 and ERV3 (Blaise et al. 2005, Kato et al. 1987) are expressed in placenta. Yet the impact of the immunosuppressive activity of HERV envelope genes in placenta needs to be studied further. While the HERV envelope genes, presumed to take part in the formation of the syncytiotrophoblast and protection of the fetus from the maternal immune system, are restricted to the simian species, the placental mammals originated more than 100 mya (Springer et al. 2003). However, endogenization of ERV envelope genes involved in the development of placenta seems to have taken place in a striking show of convergent evolution in placental mammals. For example, mice have acquired two placenta-specific ERV envelope genes named syncytin A and syncytin B, which, like the human syncytins, are both fusiogenic (Dupressoir et al. 2005) and predominantly expressed in the placenta. Only one of the murine syncytins has immunosuppressive properties (Mangeney et al. 2007). Knockout of the syncytin A gene prevents formation of a syncytial layer of trophoblast cells and results in miscarriage (Dupressoir et al. 2009). Additionally, two other placental mammals also seem to have endogenized an ERV env gene utilized in placenta development. The endogenous jaagsiekte sheep retrovirus (enJSRV) env gene in sheep is expressed in the beginning of day 12 in trophoblasts, whereas its receptor hyaluronidase 2 (HYAL2) is expressed from day 16, but only in the binucleate cells
2
Retroviruses and Cell Fusions: Overview
31
and the syncytiotrophoblast (Dunlap et al. 2005). Furthermore, injection of antisense oligonucleotides that blocked enJSRV envelope protein production, inhibited “trophoblast giant binucleate cell” differentiation as is normally observed on day 16, and led to pregnancy loss (Dunlap et al. 2006). However, there is so far no demonstration of cell–cell fusion mediated by enJSRV env and its exact role in placenta formation in sheep is unknown. Additionally, a fusiogenic and placenta-specific ERV envelope has also been identified in rabbits (Heidmann et al. 2009). The convergent evolution of fusiogenic ERV envelope genes in placental mammals strongly supports the notion that individual retroviral integrations can be domesticated independently of one another.
2.4.3 Structural Composition of HERV Envelope Proteins Functional and comparative studies show that the molecular cell-fusion mechanism of ERV envelope genes resembles the mechanism of viral fusion in exogenous viruses such as MLV. Many distinguishable motifs found in the TM part of the envelope protein are conserved among the different endogenous envelope proteins (Fig. 2.10). A peptide mimicking one of the heptad repeats in the TM of syncytin 1 functions as a fusion inhibitor (see Section 2.2.5) (Chang et al. 2004), indicating six-helix bundle formations during fusion of syncytin 1. Additionally, a structural
Fig. 2.10 Identification of motifs involved in regulation of the fusion of fusiogenic ERV envelope proteins, compared to the exogenous MLV. SP: Signal peptide; SPH: Histidine motif; FP: Fusion peptide; ISU: Immunosuppressive domain (black box); C(X)6 CC grey line; NHR CHR: Coiledcoil – heptad repeat structure (green); Transmembrane domain (gray box)
32
A.L. Kjeldbjerg et al.
prediction of the ectodomain of syncytin 2 resembles the structure of the corresponding domain of present-day exogenous retroviruses (Renard et al. 2005). All fusiogenic ERV envelope proteins identified contain cysteine motifs in TM and SU, similar to those involved in the formation of the disulfide bound between the two subunits in exogenous gamma-retroviruses (Chen et al. 2008, Pinter et al. 1997). Heterogeneity of the fusiogenic ERV envelope proteins is most pronounced for the SU subunit, which contains the regulatory and receptor-binding properties of the envelope proteins. However, they all feature the cysteine CXXC motif involved in disulfide bond isomerization during fusion (Wallin et al. 2004, 2005) (Fig. 2.10), and it may be speculated that this motif has the same function as in present-day exogenous retroviruses. Nevertheless, the localization of the CXXC motif separates the envelope proteins into two groups. The first group (syncytin 1, EnvPb1, syncytin Ory1) resembles the localization in exogenous retroviruses in the carboxy-terminal part of the SU, whereas the other group (syncytin 2, syncytin A, syncytin B) has an amino-terminal localization of the CXXC motif. Furthermore, syncytin 1 and syncytin 2 contain a motif reminiscent of the SPH motif found in the amino-terminal part of gammaretorviral SU, known to be involved in activation of the disulfide isomerization after receptor binding, as discussed earlier in Section 2.2.7. It has been shown that in the case of MLV this histidine residue can be substituted by a tyrosine (Qian and Albritton 2004), which is the amino acid residue found the SPY motif found in syncytin 1 and syncytin 2. However this substitution lowered the fusion activity of gammaretroviral envelope proteins. The other fusiogenic ERV envelope proteins lack the histidine motif, similar to the HTLV-1 envelope protein (Fig. 2.7). It remains to be determined whether this variable localization of CXXC and the tyrosine-substituted histidine motifs has an influence on the fusion regulation of syncytin 1 and syncytin 2, and how fusion is regulated in those ERV envelope proteins that lack a SPH motif. Nevertheless, the fusion-regulation process of the old gamma-retroviruses, represented by endogenous retroviruses, seems to be more divergent compared to present-day exogenous gammaretroviruses.
2.5 Conclusion Major advances over the past years have contributed to our detailed understanding of basic functions of the retroviral envelope protein, and a large number of specific cellular receptors for retroviral entry have been identified. Based upon this knowledge natural mechanisms of inhibition of cellular entry by retroviruses have been identified and specific types of synthetic inhibitors developed. The ability of retroviruses to form syncytia in cell culture has long been known for a number of exogenous retroviruses, while the possible role of cell–cell fusion in retroviral pathogenesis remains elusive. Some envelope proteins encoded by endogenous retroviruses of man and animals have been selectively conserved through evolution, and in some instances have retained their ability to drive cell fusion in a receptor-dependent manner possibly as a domestication process. Our knowledge of
2
Retroviruses and Cell Fusions: Overview
33
the physiological role played by such endogenous envelope proteins is still limited, but there is strong evidence that some envelope proteins are necessary for formation of the multinucleated layer of the placenta. Further studies of the normal and possible pathophysiological roles of such envelope proteins will benefit from our basic insight into the function of retroviral envelope proteins and our ability to interfere with this function.
References Aagaard L, Villesen P, Kjeldbjerg AL et al (2005) The approximately 30-million-year-old ERVPb1 envelope gene is evolutionarily conserved among hominoids and Old World monkeys. Genomics 86:685–691 Albritton LM, Kim JW, Tseng L et al (1993) Envelope-binding domain in the cationic amino acid transporter determines the host range of ecotropic murine retroviruses. J Virol 67:2091–2096 Albritton LM, Tseng L, Scadden D et al (1989) A putative murine ecotropic retrovirus receptor gene encodes a multiple membrane-spanning protein and confers susceptibility to virus infection. Cell 57:659–666 Alkhatib G, Combadiere C, Broder CC et al (1996) CC CKR5: a RANTES, MIP-1alpha, MIP1beta receptor as a fusion cofactor for macrophage-tropic HIV-1. Science 272:1955–1958 Andersen KB (1994) A domain of murine retrovirus surface protein gp70 mediates cell fusion, as shown in a novel SC-1 cell fusion system. J Virol 68:3175–3182 Bahrami S, Duch M, Pedersen FS (2004) Change of tropism of SL3-2 murine leukemia virus, using random mutational libraries. J Virol 78:9343–9351 Barbulescu M, Turner G, Seaman MI et al (1999) Many human endogenous retrovirus K (HERV-K) proviruses are unique to humans. Curr Biol 9:861–868 Barnard RJO, Elleder D, Young JAT (2006) Avian sarcoma and leukosis virus-receptor interactions: from classical genetics to novel insights into virus-cell membrane fusion. Virology 344:25–29 Battini JL, Rasko JE, Miller AD (1999) A human cell-surface receptor for xenotropic and polytropic murine leukemia viruses: possible role in G protein-coupled signal transduction. Proc Natl Acad Sci USA 96:1385–1390 Belshaw R, Dawson ALA, Woolven-Allen J et al (2005) Genomewide screening reveals high levels of insertional polymorphism in the human endogenous retrovirus family HERV-K(HML2): implications for present-day activity. J Virol 79:12507–12514 Belshaw R, Pereira V, Katzourakis A J et al (2004) Long-term reinfection of the human genome by endogenous retroviruses. Proc Natl Acad Sci USA 101:4894–4899 Bentz J, Mittal A (2000) Deployment of membrane fusion protein domains during fusion. Cell Biol Int 24:819–838 Best S, Le Tissier PR, Stoye JP (1997) Endogenous retroviruses and the evolution of resistance to retroviral infection. Trends Microbiol 5:313–318 Bieda K, Hoffmann A, Boller K (2001) Phenotypic heterogeneity of human endogenous retrovirus particles produced by teratocarcinoma cell lines. J Gen Virol 82:591–596 Blaise S, de Parseval N, Bénit L et al (2003) Genomewide screening for fusogenic human endogenous retrovirus envelopes identifies syncytin 2, a gene conserved on primate evolution. Proc Natl Acad Sci USA 100:13013–13018 Blaise S, de Parseval N, Heidmann T (2005) Functional characterization of two newly identified Human Endogenous Retrovirus coding envelope genes. Retrovirology 2:19 Blond JL, Lavillette D, Cheynet V et al (2000) An envelope glycoprotein of the human endogenous retrovirus HERV-W is expressed in the human placenta and fuses cells expressing the type D mammalian retrovirus receptor. J Virol 74:3321–3329
34
A.L. Kjeldbjerg et al.
Boeke J, Stoye J (1997) Retrotransposons, endogenous retroviruses, and the evolution of retroelements. In: Coffin J, Hughes S, Varmus H (eds) Retroviruses. Cold Spring Harbor Laboratory, Plainview, NY, pp 343–435 Boller K, Frank H, Löwer J et al (1983) Structural organization of unique retrovirus-like particles budding from human teratocarcinoma cell lines. J Gen Virol 64(Pt 12):2549–2559 Boller K, Schönfeld K, Lischer S et al (2008) Human endogenous retrovirus HERV-K113 is capable of producing intact viral particles. J Gen Virol 89:567–572 Bonnaud B, Beliaeff J, Bouton O et al (2005) Natural history of the ERVWE1 endogenous retroviral locus. Retrovirology 2:57 Bonnaud B, Bouton O, Oriol G et al (2004) Evidence of selection on the domesticated ERVWE1 env retroviral element involved in placentation. Mol Biol Evol 21:1895–1901 Bullough PA, Hughson FM, Treharne AC et al (1994) Crystals of a fragment of influenza haemagglutinin in the low pH induced conformation. J Mol Biol 236:1262–1265 Bénit L, Calteau A, Heidmann T (2003) Characterization of the low-copy HERV-Fc family: evidence for recent integrations in primates of elements with coding envelope genes. Virology 312:159–168 Ceccaldi P, Delebecque F, Prevost M et al (2006) DC-SIGN facilitates fusion of dendritic cells with human T-cell leukemia virus type 1-infected cells. J Virol 80:4771–4780 Chang C, Chen P, Chang G et al (2004) Functional characterization of the placental fusogenic membrane protein syncytin. Biol Reprod 71:1956–1962 Chen C, Chen L, Yang S et al (2008) Functional characterization of the human placental fusogenic membrane protein syncytin 2. Biol Reprod 79:815–823 Chesebro B, Buller R, Portis J et al (1990) Failure of human immunodeficiency virus entry and infection in CD4-positive human brain and skin cells. J Virol 64:215–221 Choe H, Farzan M, Sun Y et al (1996) The beta-chemokine receptors CCR3 and CCR5 facilitate infection by primary HIV-1 isolates. Cell 85:1135–1148 Creighton TE (1996) Proteins: structures and molecular properties. W.H. Freeman and Company, New York, NY Crise B, Rose JK (1992) Human immunodeficiency virus type 1 glycoprotein precursor retains a CD4-p56lck complex in the endoplasmic reticulum. J Virol 66:2296–2301 Cáceres M, Thomas JW (2006) The gene of retroviral origin Syncytin 1 is specific to hominoids and is inactive in Old World monkeys. J Hered 97:100–106 Côté M, Zheng Y, Liu S (2009) Receptor binding and low pH coactivate oncogenic retrovirus envelope-mediated fusion. J Virol 83:11447–11455 Dalgleish AG, Beverley PC, Clapham PR et al (1984) The CD4 (T4) antigen is an essential component of the receptor for the AIDS retrovirus. Nature 312:763–767 de Parseval N, Casella J, Gressin L et al (2001) Characterization of the three HERV-H proviruses with an open envelope reading frame encompassing the immunosuppressive domain and evolutionary history in primates. Virology 279:558–569 de Parseval N, Diop G, Blaise S et al (2005) Comprehensive search for intra- and inter-specific sequence polymorphisms among coding envelope genes of retroviral origin found in the human genome: genes and pseudogenes. BMC Genomics 6:117 de Parseval N, Heidmann T (1998) Physiological knockout of the envelope gene of the singlecopy ERV-3 human endogenous retrovirus in a fraction of the Caucasianpopulation. J Virol 72:3442–3445 de Parseval N, Lazar V, Casella J et al (2003) Survey of human genes of retroviral origin: identification and transcriptome of the genes with coding capacity for complete envelope proteins. J Virol 77:10414–10422 Dewannieux M, Blaise S, Heidmann T (2005) Identification of a functional envelope protein from the HERV-K family of human endogenous retroviruses. J Virol 79:15573–15577 Dewannieux M, Harper F, Richaud A et al (2006) Identification of an infectious progenitor for the multiple-copy HERV-K human endogenous retroelements. Genome Res 16: 1548–1556
2
Retroviruses and Cell Fusions: Overview
35
Dong B, Kim S, Hong S et al (2007) An infectious retrovirus susceptible to an IFN antiviral pathway from human prostate tumors. Proc Natl Acad Sci USA 104:1655–1660 Duda A, Lüftenegger D, Pietschmann T et al (2006) Characterization of the prototype foamy virus envelope glycoprotein receptor-binding domain. J Virol 80:8158–8167 Duelli DM, Hearn S, Myers MP et al (2005) A primate virus generates transformed human cells by fusion. J Cell Biol 171:493–503 Dunlap KA, Palmarini M, Adelson DL et al (2005) Sheep endogenous betaretroviruses (enJSRVs) and the hyaluronidase 2 (HYAL2) receptor in the ovine uterus and conceptus. Biol Reprod 73:271–279 Dunlap KA, Palmarini M, Varela M et al (2006) Endogenous retroviruses regulate periimplantation placental growth and differentiation. Proc Natl Acad Sci USA 103:14390–14395 Dupressoir A, Marceau G, Vernochet C et al (2005) Syncytin-A and syncytin-B, two fusogenic placenta-specific murine envelope genes of retroviral origin conserved in Muridae. Proc Natl Acad Sci USA 102:725–730 Dupressoir A, Vernochet C, Bawa O et al (2009) Syncytin-A knockout mice demonstrate the critical role in placentation of a fusogenic, endogenous retrovirus-derived, envelope gene. Proc Natl Acad Sci USA 106:12127–12132 Endres MJ, Clapham PR, Marsh M et al (1996) CD4-independent infection by HIV-2 is mediated by fusin/CXCR4. Cell 87:745–756 Esnault C, Heidmann O, Delebecque F et al (2005) APOBEC3G cytidine deaminase inhibits retrotransposition of endogenous retroviruses. Nature 433:430–433 Esnault C, Millet J, Schwartz O et al (2006) Dual inhibitory effects of APOBEC family proteins on retrotransposition of mammalian endogenous retroviruses. Nucleic Acids Res 34:1522–1531 Esnault C, Priet S, Ribet D et al (2008) A placenta-specific receptor for the fusogenic, endogenous retrovirus-derived, human syncytin-2. Proc Natl Acad Sci USA 105:17532–17537 Fackler OT, Baur AS (2002) Live and let die: Nef functions beyond HIV replication. Immunity 16:493–497 Fass D, Kim PS (1995) Dissection of a retrovirus envelope protein reveals structural similarity to influenza hemagglutinin. Curr Biol 5:1377–1383 Frendo J, Olivier D, Cheynet V et al (2003) Direct involvement of HERV-W Env glycoprotein in human trophoblast cell fusion and differentiation. Mol Cell Biol 23:3566–3574 Gilbert C, Maxfield DG, Goodman SM et al (2009) Parallel germline infiltration of a lentivirus in two Malagasy lemurs. PLoS Genet 5:e1000425 Goff S (2007) Retroviridae: the retroviruses and their replication. In: Knipe DM, Howley PM (eds) Field s virology. Lippincott Williams and Wilkins, Philadelphia, PA, pp 1999–2051 Greenberg M, Cammack N, Salgo M et al (2004) HIV fusion and its inhibition in antiretroviral therapy. Rev Med Virol 14:321–337 Hayward MD, Pötgens AJG, Drewlo S et al (2007) Distribution of human endogenous retrovirus type W receptor in normal human villous placenta. Pathology 39:406–412 Heidmann O, Vernochet C, Dupressoir A et al (2009) Identification of an endogenous retroviral envelope gene with fusogenic activity and placenta-specific expression in the rabbit: a new “syncytin” in a third order of mammals. Retrovirology 6:107 Hein S, Prassolov V, Zhang Y et al (2003) Sodium-dependent myo-inositol transporter 1 is a cellular receptor for Mus cervicolor M813 murine leukemia virus. J Virol 77:5926–5932 Herniou E, Martin J, Miller K et al (1998) Retroviral diversity and distribution in vertebrates. J Virol 72:5955–5966 Hervé CA, Forrest G, Löwer R et al (2004) Conservation and loss of the ERV3 open reading frame in primates. Genomics 83:940–943 Hughes JF, Coffin JM (2004) Human endogenous retrovirus K solo-LTR formation and insertional polymorphisms: implications for human and viral evolution. Proc Natl Acad Sci USA 101:1668–1672 Hunter E (1997) Viral entry and receptors. In: Coffin J, Hughes S, Varmus H (eds) Retroviruses. Cold Spring Harbor Laboratory, Plainview, NY, pp 71–121
36
A.L. Kjeldbjerg et al.
Igakura T, Stinchcombe JC, Goon PKC et al (2003) Spread of HTLV-I between lymphocytes by virus-induced polarization of the cytoskeleton. Science 299:1713–1716 Ji C, Kopetzki E, Jekle A et al (2009) CD4-anchoring HIV-1 fusion inhibitor with enhanced potency and in vivo stability. J Biol Chem 284:5175–5185 Ji C, Zhang J, Cammack N et al (2006) Development of a novel dual CCR5-dependent and CXCR4-dependent cell–cell fusion assay system with inducible gp160 expression. J Biomol Screen 11:65–74 Jin J, Sherer NM, Heidecker G et al (2009) Assembly of the murine leukemia virus is directed towards sites of cell–cell contact. PLoS Biol 7:e1000163 Katen LJ, Januszeski MM, Anderson WF et al (2001) Infectious entry by amphotropic as well as ecotropic murine leukemia viruses occurs through an endocytic pathway. J Virol 75:5018–5026 Kato N, Pfeifer-Ohlsson S, Kato M et al (1987) Tissue-specific expression of human provirus ERV3 mRNA in human placenta: two of the three ERV3 mRNAs contain human cellular sequences. J Virol 61:2182–2191 Katzourakis A, Tristem M (2005) Phylogeny of human/primate endogenous and exogenous retroviruses. In: Sverdlov ED (ed) Retrovirus and primate genome evolution. Landes bioscience, Georgetown, TX, pp 186–203 Kavanaugh MP, Miller DG, Zhang W et al (1994) Cell-surface receptors for gibbon ape leukemia virus and amphotropic murine retrovirus are inducible sodium-dependent phosphate symporters. Proc Natl Acad Sci USA 91:7071–7075 Kent WJ, Sugnet CW, Furey TS et al (2002) The human genome browser at UCSC. Genome Res 12:996–1006 Kjeldbjerg AL, Villesen P, Aagaard L et al (2008) Gene conversion and purifying selection of a placenta-specific ERV-V envelope gene during simian evolution. BMC Evol Biol 8:266 Klatzmann D, Champagne E, Chamaret S et al (1984) T-lymphocyte T4 molecule behaves as the receptor for human retrovirus LAV. Nature 312:767–768 Klement V, Rowe WP, Hartley JW et al (1969) Mixed culture cytopathogenicity: a new test for growth of murine leukemia viruses in tissue culture. Proc Natl Acad Sci USA 63:753–758 Kubo Y, Ishimoto A, Amanuma H (2003) N-Linked glycosylation is required for XC cellspecific syncytium formation by the R peptide-containing envelope protein of ecotropic murine leukemia viruses. J Virol 77:7510–7516 Kurth R, Löwer J, Löwer R et al (1983) Retroviruses in human tumors. Arch Geschwulstforsch 53:289–299 Lairmore FD, Franchini G (2007) Human t-cell leukemia virus types 1 and 2. In: Knipe DM, Howley PM (eds) Fields’ virology. Lippincott Williams and Wilkins, Philadelphia, PA Lamb D, Schüttelkopf AW, van Aalten DMF et al (2008) Highly specific inhibition of leukaemia virus membrane fusion by interaction of peptide antagonists with a conserved region of the coiled coil of envelope. Retrovirology 5:70 Lander ES, Linton LM, Birren B et al (2001) Initial sequencing and analysis of the human genome. Nature 409:860–921 Lavie L, Kitova M, Maldener E et al (2005) CpG methylation directly regulates transcriptional activity of the human endogenous retrovirus family HERV-K(HML-2). J Virol 79:876–883 Lavillette D, Marin M, Ruggieri A et al (2002) The envelope glycoprotein of human endogenous retrovirus type W uses a divergent family of amino acid transporters/cell surface receptors. J Virol 76:6442–6452 Lee YN, Bieniasz PD (2007) Reconstitution of an infectious human endogenous retrovirus. PLoS Pathog 3:e10 Levesque K, Zhao Y, Cohen EA (2003) Vpu exerts a positive effect on HIV-1 infectivity by downmodulating CD4 receptor molecules at the surface of HIV-1-producing cells. J Biol Chem 278:28346–28353 Li K, Zhang S, Kronqvist M et al (2007) The conserved His8 of the Moloney murine leukemia virus Env SU subunit directs the activity of the SU-TM disulphide bond isomerase. Virology 361:149–160
2
Retroviruses and Cell Fusions: Overview
37
Li K, Zhang S, Kronqvist M et al (2008) Intersubunit disulfide isomerization controls membrane fusion of human T-cell leukemia virus Env. J Virol 82:7135–7143 Li M, Li Z, Yao Q et al (2006) Murine leukemia virus R Peptide inhibits influenza virus hemagglutinin-induced membrane fusion. J Virol 80:6106–6114 Linial M (2007) Foamy viruses. In: Knipe DM, Howley PM (eds) Fields’s virology. Lippincott Williams & Wilkins, Philadelphia, PA Lorizate M, Brügger B, Akiyama H et al (2009) Probing HIV-1 membrane liquid order by Laurdan staining reveals producer cell-dependent differences. J Biol Chem 284:22238–22247 Löwer R, Löwer J, Kurth R (1996) The viruses in all of us: characteristics and biological significance of human endogenous retrovirus sequences. Proc Natl Acad Sci USA 93:5177–5184 Lu M, Blacklow SC, Kim PS (1995) A trimeric structural domain of the HIV-1 transmembrane glycoprotein. Nat Struct Biol 2:1075–1082 Malassiné A, Blaise S, Handschuh K et al (2007) Expression of the fusogenic HERV-FRD Env glycoprotein (syncytin 2) in human placenta is restricted to villous cytotrophoblastic cells. Placenta 28:185–191 Malassiné A, Handschuh K, Tsatsaris V et al (2005) Expression of HERV-W Env glycoprotein (syncytin) in the extravillous trophoblast of first trimester human placenta. Placenta 26:556–562 Mallet F, Bouton O, Prudhomme S et al (2004) The endogenous retroviral locus ERVWE1 is a bona fide gene involved in hominoid placental physiology. Proc Natl Acad Sci USA 101:1731–1736 Manel N, Kim FJ, Kinet S et al (2003) The ubiquitous glucose transporter GLUT-1 is a receptor for HTLV. Cell 115:449–459 Mangeney M, Renard M, Schlecht-Louf G et al (2007) Placental syncytins: genetic disjunction between the fusogenic and immunosuppressive activity of retroviral envelope proteins. Proc Natl Acad Sci USA 104:20534–20539 Mangeney M, de Parseval N, Thomas G et al (2001) The full-length envelope of an HERV-H human endogenous retrovirus has immunosuppressive properties. J Gen Virol 82:2515–2518 McClure MO, Sommerfelt MA, Marsh M et al (1990) The pH independence of mammalian retrovirus infection. J Gen Virol 71(Pt 4):767–773 Medstrand P, Mager DL (1998) Human-specific integrations of the HERV-K endogenous retrovirus family. J Virol 72:9782–9787 Metzner C, Salmons B, Günzburg WH et al (2008) Rafts, anchors and viruses – a role for glycosylphosphatidylinositol anchored proteins in the modification of enveloped viruses and viral vectors. Virology 382:125–131 Mi S, Lee X, Li X et al (2000) Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature 403:785–789 Miller AD (2003) Identification of Hyal2 as the cell-surface receptor for jaagsiekte sheep retrovirus and ovine nasal adenocarcinoma virus. Curr Top Microbiol Immunol 275:179–199 Miller AD (2008) Hyaluronidase 2 and its intriguing role as a cell-entry receptor for oncogenic sheep retroviruses. Semin Cancer Biol 18:296–301 Miller DG, Miller AD (1994) A family of retroviruses that utilize related phosphate transporters for cell entry. J Virol 68(12):8270–8276 Miyauchi K, Kim Y, Latinovic O et al (2009) HIV enters cells via endocytosis and dynamindependent fusion with endosomes. Cell 137:433–444 Mothes W, Boerger AL, Narayan S et al (2000) Retroviral entry mediated by receptor priming and low pH triggering of an envelope glycoprotein. Cell 103:679–689 Netter RC, Amberg SM, Balliet JW et al (2004) Heptad repeat 2-based peptides inhibit avian sarcoma and leukosis virus subgroup a infection and identify a fusion intermediate. J Virol 78:13430–13439 Nussbaum O, Roop A, Anderson WF (1993) Sequences determining the pH dependence of viral entry are distinct from the host range-determining region of the murine ecotropic and amphotropic retrovirus envelope proteins. J Virol 67:7402–7405
38
A.L. Kjeldbjerg et al.
O’Neill RJ, O’Neill MJ, Graves JA (1998) Undermethylation associated with retroelement activation and chromosome remodelling in an interspecific mammalian hybrid. Nature 393:68–72 Olsen KE, Andersen KB (1999) Palmitoylation of the intracytoplasmic R peptide of the transmembrane envelope protein in Moloney murine leukemia virus. J Virol 73:8975–8981 Pantaleo G, Butini L, Graziosi C et al (1991) Human immunodeficiency virus (HIV) infection in CD4+ T lymphocytes genetically deficient in LFA-1: LFA-1 is required for HIV-mediated cell fusion but not for viral transmission. J Exp Med 173:511–514 Papandreou M, Barbouche R, Guieu R et al (2010) Mapping of domains on HIV envelope protein mediating association with calnexin and protein disulfide isomerase. J Biol Chem. doi 10.1074/jbc.M109.066670 Paré M, Gauthier S, Landry S et al (2005) A new sensitive and quantitative HTLV-I-mediated cell fusion assay in T cells. Virology 338:309–322 Perfettini J, Castedo M, Roumier T et al (2005) Mechanisms of apoptosis induction by the HIV-1 envelope. Cell Death Differ 12(Suppl 1):916–923 Picard-Maureau M, Jarmy G, Berg A et al (2003) Foamy virus envelope glycoprotein-mediated entry involves a pH-dependent fusion process. J Virol 77:4722–4730 Pietschmann T, Zentgraf H, Rethwilm A et al (2000) An evolutionarily conserved positively charged amino acid in the putative membrane-spanning domain of the foamy virus envelope protein controls fusion activity. J Virol 74:4474–4482 Pinter A, Kopelman R, Li Z et al (1997) Localization of the labile disulfide bond between SU and TM of the murine leukemia virus envelope protein complex to a highly conserved CWLC motif in SU that resembles the active-site sequence of thiol-disulfide exchange enzymes. J Virol 71:8073–8077 Ponferrada VG, Mauck BS, Wooley DP (2003) The envelope glycoprotein of human endogenous retrovirus HERV-W induces cellular resistance to spleen necrosis virus. Arch Virol 148: 659–675 Prassolov V, Hein S, Ziegler M et al (2001) Mus cervicolor murine leukemia virus isolate M813 belongs to a unique receptor interference group. J Virol 75:4490–4498 Qian Z, Albritton LM (2004) An aromatic side chain is required at residue 8 of SU for fusion of ecotropic murine leukemia virus. J Virol 78:508–512 Ragheb JA, Anderson WF (1994) pH-independent murine leukemia virus ecotropic envelopemediated cell fusion: implications for the role of the R peptide and p12E TM in viral entry. J Virol 68:3220–3231 Rein A, Mirro J, Haynes JG et al (1994) Function of the cytoplasmic domain of a retroviral transmembrane protein: p15E-p2E cleavage activates the membrane fusion capability of the murine leukemia virus Env protein. J Virol 68:1773–1781 Renard M, Varela PF, Letzelter C, Duquerroy S, Rey FA, Heidmann T (2005) Crystalstructure of a pivotal domain of human syncytin-2, a 40 million years oldendogenous retrovirus fusogenic envelope gene captured by primates. J Mol Biol 352:1029–1034 Roeder GS, Fink GR (1980) DNA rearrangements associated with a transposable element in yeast. Cell 21:239–249 Rotman G, Itin A, Keshet E (1984) ‘Solo’ large terminal repeats (LTR) of an endogenous retrovirus-like gene family (VL30) in the mouse genome. Nucleic Acids Res 12: 2273–2282 Shaheen F, Collman RG (2004) Co-receptor antagonists as HIV-1 entry inhibitors. Curr Opin Infect Dis 17:7–16 Shankarappa R, Margolick JB, Gange SJ et al (1999) Consistent viral evolutionary changes associated with the progression of human immunodeficiency virus type 1 infection. J Virol 73:10489–10502 Sherer NM, Mothes W (2008) Cytonemes and tunneling nanotubules in cell–cell communication and viral pathogenesis. Trends Cell Biol 18:414–420 Shih A, Coutavas EE, Rush MG (1991) Evolutionary implications of primate endogenous retroviruses. Virology 182:495–502
2
Retroviruses and Cell Fusions: Overview
39
Siess DC, Kozak SL, Kabat D (1996) Exceptional fusogenicity of Chinese hamster ovary cells with murine retroviruses suggests roles for cellular factor(s) and receptor clusters in the membrane fusion process. J Virol 70:3432–3439 Skehel JJ, Bayley PM, Brown EB et al (1982) Changes in the conformation of influenza virus hemagglutinin at the pH optimum of virus-mediated membrane fusion. Proc Natl Acad Sci USA 79:968–972 Springer MS, Murphy WJ, Eizirik E et al (2003) Placental mammal diversification and the Cretaceous-Tertiary boundary. Proc Natl Acad Sci USA 100:1056–1061 Steiper ME, Young NM (2006) Primate molecular divergence dates. Mol Phylogenet Evol 41:384–394 Sverdlov ED (1998) Perpetually mobile footprints of ancient infections in human genome. FEBS Lett 428:1–6 Tailor CS, Nouri A, Lee CG, Kozak C, Kabat D (1999) Cloning and characterization of a cell surface receptor for xenotropic and polytropic murine leukemia viruses. Proc Natl Acad Sci USA 96:927–932 Turner G, Barbulescu M, Su M et al (2001) Insertional polymorphisms of full-length endogenous retroviruses in humans. Curr Biol 11:1531–1535 Urisman A, Molinaro RJ, Fischer N et al (2006) Identification of a novel Gammaretrovirus in prostate tumors of patients homozygous for R462Q RNASEL variant. PLoS Pathog 2:e25 Van Hoeven NS, Miller AD (2005) Use of different but overlapping determinants in a retrovirus receptor accounts for non-reciprocal interference between xenotropic and polytropic murine leukemia viruses. Retrovirology 2:76 Villesen P, Aagaard L, Wiuf C et al (2004) Identification of endogenous retroviral reading frames in the human genome. Retrovirology 1:32 Wallin M, Ekström M, Garoff H (2004) Isomerization of the intersubunit disulphide-bond in Env controls retrovirus fusion. EMBO J 23:54–65 Wallin M, Ekström M, Garoff H (2005) The fusion-controlling disulfide bond isomerase in retrovirus Env is triggered by protein destabilization. J Virol 79:1678–1685 Wang E, Obeng-Adjei N, Ying Q et al (2008) Mouse mammary tumor virus uses mouse but not human transferrin receptor 1 to reach a low pH compartment and infect cells. Virology 381:230–240 Wang H, Kavanaugh MP, North RA et al (1991) Cell-surface receptor for ecotropic murine retroviruses is a basic amino-acid transporter. Nature 352:729–731 White JM (1992) Membrane fusion. Science 258:917–924 White JM, Delos SE, Brecher M et al (2008) Structures and mechanisms of viral membrane fusion proteins: multiple variations on a common theme. Crit Rev Biochem Mol Biol 43:189–219 Willey RL, Maldarelli F, Martin MA et al (1992a) Human immunodeficiency virus type 1 Vpu protein induces rapid degradation of CD4. J Virol 66:7193–7200 Willey RL, Maldarelli F, Martin MA et al (1992b) Human immunodeficiency virus type 1 Vpu protein regulates the formation of intracellular gp160-CD4 complexes. J Virol 66:226–234 Wilson IA, Skehel JJ, Wiley DC (1981) Structure of the haemagglutinin membrane glycoprotein of influenza virus at 3 A resolution. Nature 289:366–373 Yang C, Compans RW (1996) Analysis of the cell fusion activities of chimeric simian immunodeficiency virus-murine leukemia virus envelope proteins: inhibitory effects of the R peptide. J Virol 70:248–254 Yang YL, Guo L, Xu S et al (1999) Receptors for polytropic and xenotropic mouse leukaemia viruses encoded by a single gene at Rmc1. Nat Genet 21:216–219
Chapter 3
Retroviral Membrane Fusions: Regulation by Proteolytic Processing and Cellular Factors Yoshinao Kubo
Abstract Retroviruses infect host cells by fusion between the viral envelope and host cell membranes mediated by retroviral envelope (Env) glycoproteins. Because of this membrane fusion activity, cells expressing the Env proteins often fuse with neighboring cells, resulting in syncytia. Retroviral membrane fusion is directly induced by the interaction between the Env proteins and cell surface receptors. In addition, many cellular factors could affect syncytium formation by the retroviral Env proteins. Because cell fusion events are also involved in the developmental processes of several tissues, studies on the regulation of the retroviral membrane fusion would contribute to the understanding of developmental biology as well as retrovirus infections. Precursor Env protein processing by furin or other cellular proteases into surface (SU) and transmembrane (TM) subunits is required for membrane fusion. Functions of the cytoplasmic tail of TM subunits to inhibit membrane fusion activity are conserved among many retroviruses. The mature SU proteins of several retroviruses are further cleaved by cathepsin proteases in acidic late endosomes to activate the membrane fusion. Fusion requires the formation of a “pre-fusion complex” at the viral entry site, and this complex includes many viral and cellular factors. Cholesterol- and sphingolipid-enriched raft microdomains of host cell and retroviral envelope membranes are both required for membrane fusion. Further, the cytoskeleton and its associated signaling factors, such as Rho, are also associated with Env-induced membrane fusion. Keywords Cathepsin · cell adhesion · cytoskeleton · endosome · envelope proteins · fusions · membrane fusions · microdomains · proteolysis · rafts · retroviruses Abbreviations AIDS ASCT
Acquired immuno deficiency syndrome Alanine, serine and cysteine selective transporters
Y. Kubo (B) Department of AIDS Research, Institute of Tropical Medicine, Nagasaki University, Nagasaki 852-8523, Japan e-mail:
[email protected] L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_3, C Springer Science+Business Media B.V. 2011
41
42
CAT1 CCR5 CXCR4 Env ERM FIV GLUT1 HERV HIV HTLV ICAM Lck LFA-1 MLV Pit ROCK SIV SU TM VCA VCAM VLA-4 WASP XMRV XPR
Y. Kubo
Cationic amino acid transporter 1 CC chemokine receptor 5 CXC chemokine receptor 4 Envelope Ezrin, radixin, moesin Feline immunodeficiency virus Glucose transporter 1 Human endogenous retrovirus Human immunodeficiency virus Human T-cell lymphotropic virus Inter-cellular adhesion molecule Leukocyte-specific protein tyrosine kinase Lymphocyte function-associated antigen 1 Murine leukemia virus Sodium-dependent phosphate symporter Rho-associated, coiled-coil containing protein kinase Simian immunodeficiency virus Surface Transmembrane Verprolin-homologous, cofilin-homologous, and acidic domain Vascular cell adhesion molecule Very late antigen-4 (Integrin alpha4beta1) Wiscott-Aldrich syndrome protein Xenotropic MLV-related virus Xenotropic/polytropic receptor
Contents 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Membrane Fusion by Retroviral Env Protein . . . . . . . . . . . 3.2.1 Cell Surface Receptor . . . . . . . . . . . . . . . . . . 3.2.2 Membrane Fusion Mechanism . . . . . . . . . . . . . . 3.3 Regulation of Retroviral Membrane Fusion by Proteolytic Processing 3.3.1 Processing of Precursor Env Polyprotein . . . . . . . . . . 3.3.2 R Peptide Cleavage . . . . . . . . . . . . . . . . . . . 3.3.3 Syncytium Formation in XC Cells by MLV . . . . . . . . . 3.3.4 Mechanism of R Peptide to Inhibit Membrane Fusion . . . . 3.3.5 Cleavage by Cathepsin Proteases . . . . . . . . . . . . . 3.4 Regulation of Retroviral Membrane Fusion by Cellular Factors . . . 3.4.1 Lipid Raft . . . . . . . . . . . . . . . . . . . . . . . . 3.4.2 Cell Adhesion Molecules . . . . . . . . . . . . . . . . . 3.4.3 Cytoskeleton-Associated Molecules . . . . . . . . . . . . 3.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
43 44 44 46 47 47 48 49 49 50 52 52 53 54 55 55
3
Retroviral Membrane Fusions
43
3.1 Introduction Retroviruses including human immunodeficiency virus (HIV), simian immunodeficiency virus (SIV), human T-cell leukemia virus (HTLV), and murine leukemia virus (MLV) are enveloped with a lipid bi-layer derived from virus-producing host cells (Fig. 3.1). Retroviruses infect host cells by fusion between the viral envelope and host cell membranes after interaction with their cell surface receptors, similar to other enveloped viruses, and this reaction is induced by the envelope (Env) glycoproteins. Because of the membrane fusion activity of the retroviral Env proteins, cells expressing the Env proteins often fuse with neighboring susceptible cells to form multi-nucleated giant cells. This is known as syncytium-formation (Fig. 3.2). It
Fig. 3.1 Structure of a retrovirus particle. Retroviral particles have a lipid bi-layer envelope (envelope membrane) that is derived from the plasma membrane of virus-infected cells. The envelope proteins (Env) are expressed on the envelope membrane. The precursor Env proteins are proteolytically cleaved into surface (SU) and transmembrane (TM) subunits
Fig. 3.2 Morphology of syncytia induced by the R peptide-truncated ecotropic MLV Env protein in mouse NIH3T3 cells
44
Y. Kubo
is generally thought that syncytium formation induced by the retroviral Env proteins reflects processes involved in viral entry into host cells. Syncytium formation induced by the HIV Env protein activates host cell apoptosis, ultimately resulting in cell death. This results in T-lymphocyte depletion, and may be associated with HIV pathogenicity (Ferri et al. 2000, Scheller and Jassoy 2001). Alternatively, such fusogenic retroviral Env proteins could also be used as a therapeutic for cancer treatment by inducing cell death of transformed cells through syncytium formation (Diaz et al. 2000, Galanis et al. 2001). Cell–cell fusions also play important roles for fertilization, and developments of syncytiotrophoblasts in placenta, osteoclasts in bone, and multi-nucleated skeletal muscle fibers. Interestingly, syncytiotrophoblast formation is induced by a human endogenous retrovirus (HERV) Env protein, named syncytin (Dupressoir et al. 2009, Mi et al. 2000). Thus, studies of how these cell fusion processes are regulated would contribute to understanding many biological phenomena, including viral infectivity, viral pathogenicity, cancer gene therapy, and tissue developments. The cell fusion process initiated by retroviral Env proteins is directly induced by the interaction of the Env proteins with their cognate cell surface receptors. Additionally, other viral and cellular factors influence Env-induced cell fusion. In this chapter, I will describe the regulation of retroviral Env-induced membrane fusion by proteolytic processing of the Env proteins and by cellular factors.
3.2 Membrane Fusion by Retroviral Env Protein 3.2.1 Cell Surface Receptor All retroviral Env glycoproteins are synthesized as a precursor polyprotein. The polyprotein is then cleaved into surface (SU) and transmembrane (TM) subunits by endoplasmic reticulum-associated cellular proteases. The SU and TM subunits bind each other via disulfide bonds (Wallin et al. 2004). Trimers of the SU-TM heterodimers are expressed on the surface of the viral particles and virus-infected cells (Sodroski 1999). The SU subunits interact with specific cell surface receptor proteins. The TM subunits anchor the Env trimer to membrane surface of viral particles and virus-infected cells, and induce membrane fusion resulting in syncytium formation and viral entry. The SU subunits of almost all HIV isolates utilize CD4 and either CXC chemokine receptor 4 (CXCR4) or CC chemokine receptor 5 (CCR5) as the cell surface receptors (Berger et al. 1999) (Table 3.1). However, HIV variants that can infect CD4-negative cells have been frequently isolated from patients with acquired immunodeficiency syndrome (AIDS) (Xiao et al. 2008, Zerhouni et al. 2004). Such CD4-independent HIV variants recognize CXCR4 or CCR5 as the sole receptor, and might be associated with hepatitis or nephropathy in AIDS patients by infection of CD4-negative liver or kidney cells (Marras et al. 2002, Xiao et al. 2008). CD4independent SIV variants are more frequently isolated from infected animals than
3
Retroviral Membrane Fusions
45
Table 3.1 Cell surface receptors of retroviruses Retroviruses
Receptors
No. of transmembrane domains
HIV
CD4 CXCR4 or CCR5 CAT1 Pit2 XPR XPR GLUT1 ASCT2
single multiple multiple multiple multiple multiple multiple multiple
ecotropic MLV amphotropic MLV polytropic MLV xenotropic MLV HTLV HERV or syncytin
CD4-independent HIV variants (Edinger et al. 1999, Puffer et al. 2002). Almost all simple retroviruses recognize multi-membrane spanning proteins, like CXCR4 and CCR5, as the entry receptor (Overbaugh et al. 2001) (Table 3.1). Therefore, the CD4-independent HIV variants are thought to be prototypes of the CD4-dependent strains. MLVs are classified into four groups according to their host ranges; ecotropic, amphotropic, polytropic, and xenotropic MLVs. The ecotropic viruses can infect only rodent cells, mouse and rat. The amphotropic and polytropic viruses can infect many species of mammals including mouse, rat, mink, and human. The xenotropic virus can infect mink and human cells, but cannot mouse cells, although the virus has been isolated from mice. Interestingly, it has been recently reported that xenotropic MLV-related virus (XMRV) is isolated from humans with prostate cancer (Schlaberg et al. 2009, Urisman et al. 2006) or chronic fatigue syndrome (Lombardi et al. 2009). The different MLV classes recognize different multimembrane-spanning proteins as the cell surface receptors (Overbaugh et al. 2001); cationic amino acid transporter 1 (CAT1) for the ecotropic MLV (Albritton et al. 1989), and sodium-dependent phosphate symporter 2 (Pit2) for the amphotropic MLV (Kavanaugh et al. 1994). The polytropic and xenotropic MLVs recognize a same cell surface receptor protein, called the xenotropic/polytropic receptor (XPR), whose function remains unknown (Battini et al. 1999, Tailor et al. 1999). The polytropic Env can interact with the cell surface receptor of many types of mammals, and the xenotropic Env recognize the same receptor proteins of many mammals except for mouse. Other retroviruses of importance utilize other cell surface proteins. Glucose transporter protein 1 (GLUT1) has been identified as the HTLV cell surface receptor (Manel et al. 2003). The HERV type W Env protein or syncytin, which induces syncytiotrophoblast formation in placenta, interacts with a sodium-dependent transporter protein of neutral amino acids, alanine, serine, and cystein (ASCT2) (Lavillette et al. 2002). Almost all of the retroviral receptors are N-linked glycoproteins, and infections by many retroviruses are suppressed by sugar chains linked to their cognate receptors (Kubo et al. 2002, 2007b, Overbaugh et al. 2001, Tailor et al. 2003). Glycosylation-defective mutants of the ecotropic MLV receptor (CAT1) transport
46
Y. Kubo
cationic amino acids into cells as efficiently as wild type CAT1 (Wang et al. 1996), indicating that the glycosylation is not required for the natural function. Whereas glycosylation of mouse CAT1 does not inhibit ecotropic MLV infection, glycosylation of the rat, hamster, and Mus dunni receptors do block infection (Kubo et al. 2002, 2004, Yoshii et al. 2008). These results suggest that glycosylation of these receptor proteins functions as innate immunity against retrovirus infections, and the host cell surface receptor proteins have evolved to suppress retrovirus infections. Many retroviruses interact with multi-membrane spanning transporter proteins as the entry receptors (Table 3.1). Because a single amino acid mutation in such multimembrane-spanning receptor proteins often abrogates their expression, and because transporter proteins have house-keeping functions to uptake essential nutrients into cells, the rate of evolution for the retroviral receptor genes would be expected to be relatively slow. If cellular proteins that have relatively higher evolution rates were used as retroviral entry receptors, the cellular proteins would easily evolve to become resistant to the viral infection, and such retroviruses would not survive. Due to the slow evolution rate, retroviruses recognizing multi-membrane spanning transporter proteins as the entry receptors continue to circulate (Yoshii et al. 2008).
3.2.2 Membrane Fusion Mechanism The SU-receptor interaction triggers a conformational change of the TM subunit (Fig. 3.3). The N-terminal hydrophobic fusion peptide of the TM subunit is exposed by this conformational change and inserts into the target cell membrane. The TM subunit adopts a hairpin-like structure through interaction of its N- and C-domains (Fig. 3.4). By this pathway of TM conformational changes, viral and cellular membranes are brought into close apposition and then are mixed. A fusion pore is then formed and expanded, allowing the retroviral core containing the viral genomes and enzymes required for replication to enter into the host cell cytoplasm. A peptide derived from the N- or C-domain of TM subunit inhibits retrovirus membrane fusion by suppressing the hairpin interaction of TM subunit N- and C-domains (Sodroski 1999).
Fig. 3.3 Mechanism of membrane fusion by the retroviral Env protein
3
Retroviral Membrane Fusions
47
Fig. 3.4 Conformational change of TM subunit
Cell–cell fusion by the retroviral Env proteins should be induced by the same mechanism as virus entry, and therefore quantitative analysis of the Env proteininduced cell fusion is often used as an indicator of the retroviral entry into host cells. However, entry of many retroviruses into host cells is dependent on low pH, but syncytium formation by these retroviral Env proteins is not (Ragheb and Anderson 1994). Further, a non-infectious mutant of MLV Env can induce syncytium formation in a specific cell line (Kubo and Amanuma 2003). Therefore, the syncytium formation capability of the retroviral Env proteins does not always reflect viral entry into host cells, indicating that syncytium formation mechanisms of retroviral Env proteins are partially different from the membrane fusion mechanism that occurs during infection and entry.
3.3 Regulation of Retroviral Membrane Fusion by Proteolytic Processing 3.3.1 Processing of Precursor Env Polyprotein As mentioned above, the retroviral Env precursor polyprotein is proteolytically cleaved by cellular proteases in endoplasmic reticulum. Furin is one of the cellular proteases that process the HIV Env precursor protein into the SU and TM subunits (Hallenberger et al. 1992). However, the HIV Env precursor protein is also cleaved in furin-defective cells, indicating that other cellular proteases are capable of digesting the HIV Env precursor protein (Gu et al. 1995, Ohnishi et al. 1994). The other cellular proteases have not been identified yet. Mutant retroviral Env proteins that cannot be cleaved into SU and TM subunits cannot induce syncytium formation in susceptible cells (Dubay et al. 1995, Freed
48
Y. Kubo
and Risser 1987, Goodman et al. 1993, Guo et al. 1990). These findings indicate that the proteolytic cleavage of retroviral Env precursor proteins is essential for membrane fusion activity.
3.3.2 R Peptide Cleavage The MLV TM subunit contains a 16-amino acid peptide at its C-terminus, called the R peptide, which is cleaved by the viral protease during virion maturation (Green et al. 1981, Henderson et al. 1984). Expression of an R peptide-truncated Env protein of ecotropic or amphotropic MLV in susceptible cells induces syncytia. In contrast, R peptide-containing Env cannot induce syncytia, indicating that the R peptide completely inhibits Env syncytium formation activity (Ragheb and Anderson 1994, Rein et al. 1994). R peptide cleavage-defective MLV Env mutant proteins have much lower infectivity than those with wild-type Env, even when the mutant Env proteins are incorporated into viral particles as efficiently as the wildtype Env (Kieman and Freed 1998, Kubo and Amanuma 2003, Kubo et al. 2007a). These results indicate that R peptide cleavage during virion maturation is required for virus entry into host cells. Expression of MLV Env proteins lacking the R peptide causes fusion of the virus-infected cells and neighboring cells, resulting in apoptotic cell death. As a consequence, production of progeny virus is reduced. These data suggest that the R peptide on Env is essential for progenitor virus production. The R peptide cleavage-defective mutants of polytropic and xenotropic MLV Env have much lower infectivity than the wild-type Env (Kubo et al. 2007a). This suggests that R peptide cleavage of polytropic and xenotropic MLV Env proteins activates the fusion activity, like ecotropic and amphotropic Env proteins. However, the R peptide truncated forms of polytropic and xenotropic Env proteins cannot induce syncytia in susceptible cells (Kubo et al. 2007a). Since the polytropic and xenotropic MLVs recognize the same cell surface receptor, the cell surface receptor could be the determinant for the inability of R peptide truncated polytropic and xenotropic Env proteins to induce syncytia. The C-terminal tails of Env proteins of spleen necrosis virus, gibbon ape leukemia virus, porcine endogenous retrovirus (Bovkova et al. 2002), and MasonPfizer monkey virus (Brody et al. 1994) are cleaved during virion maturation and inhibit membrane fusion activities of the Env proteins, like the MLV R peptide. Env proteins of feline leukemia virus and rat leukemia virus likely also have R peptide, because amino acid sequences of the cytoplasmic domains of these Env proteins are highly homologous to that of the MLV Env protein. C-terminal truncated forms of syncytin, involved in syncytiotrophoblast formation, show enhanced fusogenicity (Drewlo et al. 2006). These results suggest that the cytoplasmic tails of these retroviral Env proteins inhibit their membrane fusion activities. Cytoplasmic domains of HIV, SIV, HTLV, and feline immunodeficiency virus (FIV) Env proteins also inhibit the membrane fusion activity, though syncytium formation is not completely abolished (Hahn and Compans 1992, Kim et al. 2003,
3
Retroviral Membrane Fusions
49
Lerner and Elder 2000, Ritter et al. 1993, Zingler and Littman 1993). Unlike the MLV R peptide, the Env cytoplasmic domain of these viruses is not cleaved during virion maturation. The membrane fusion inhibition by the cytoplasmic tail of HIV Env protein is abrogated by Gag protein processing by the viral protease (Murakami et al. 2004). Interaction between the Env cytoplasmic domain and Gag precursor protein could inhibit the membrane fusion activity of the HIV Env protein. Processing of the Gag precursor protein by the viral protease could release the Gag-Env interaction resulting in activation of membrane fusion. Together, these results indicate that the function of the Env cytoplasmic domains to inhibit membrane fusion activity is conserved among many retroviruses, and suggests that this inhibitory activity is important for retroviral proliferation, although the regulation mechanisms are different among different retroviruses.
3.3.3 Syncytium Formation in XC Cells by MLV The ecotropic MLV Env protein that contains the R peptide does not induce syncytia in almost all susceptible cells as mentioned above, but specifically can induce syncytia in rat XC and fu-1 cells (Jones and Risser 1993, Wong et al. 1977). By this property of XC cells, XC cells are widely used to titrate the ecotropic MLVs (Klement et al. 1969). Ecotropic virus-infected XC cells form syncytia and die resulting in plaques. The number of plaques correlates with the number of infectious viral particles. Because XC and fu-1 cells were both derived from muscle tissue (Svoboda 1960, Wong et al. 1977), cellular factors that regulate formation of multinucleated muscle fiber cells in myogenesis might be associated with the ability to form syncytia by the R peptide-containing Env protein. Tunicamycin, a N-linked glycosylation inhibitor, suppressed XC cell-specific syncytium formation, suggesting unknown glycosylated cellular factors are involved in the cell fusion (Kubo et al. 2003a). Syncytium formation in myoblast differentiation is also affected by glycosylation modulating agents (Spearman et al. 1987). The cell–cell fusion mechanism of R peptide-containing Env in XC cells is clearly different from the membrane fusion mechanism required for the viral entry into host cells, because virus bearing R peptide-containing Env protein cannot initiate infection, even in XC cells (Kubo and Amanuma 2003). Understanding the mechanism by which the R peptide-containing MLV Env induces syncytia in XC cells could contribute to resolving mechanisms involved in myogenesis.
3.3.4 Mechanism of R Peptide to Inhibit Membrane Fusion The disulfide bond within the SU-TM complex is isomerized during membrane fusion. It has been reported that isomerization of the SU-TM disulfide was suppressed in the R peptide-containing Env protein (Loving et al. 2008), and that a monoclonal antibody against the MLV SU protein more efficiently binds to the R
50
Y. Kubo
peptide-truncated Env protein than to the R peptide-containing Env protein (Aguilar et al. 2003). These results suggest that the R peptide in the TM subunit controls the conformation of SU subunit through the disulfide bond. However, mutants of the R peptide-containing Env protein that have amino acid substitutions in the fusion peptide cannot induce syncytia in XC cells (Jones and Risser 1993), indicating that the fusion peptide of the R peptide-containing Env is exposed by the conformational change and inserted into cellular membrane of XC cells, like the R peptide-truncated Env (Fig. 3.3). In addition, the MLV R peptide inhibits the membrane fusion activity of SIV Env and influenza virus hemagglutinin proteins (Li et al. 2006, Yang and Compans 1996). It is unlikely that the MLV R peptide artificially fused to the heterologous viral spike proteins controls conformational changes of the chimeric proteins, as it does in the natural Env protein. Rather, unknown cellular factor(s) that interact with the R peptide might inhibit the membrane fusion activity of the Env protein. In XC cells, the inhibitory cellular factor might be attenuated by unknown glycosylated protein(s). Further studies are needed to understand the mechanisms by which the R peptide inhibits membrane fusion activity and the R peptide-containing Env protein induces syncytia in XC cells. The presence of cholesterol in the target cell membrane is required for HIV entry (see below). A cholesterol-binding agent, amphotericin B methyl ester, inhibits HIV infection by suppressing membrane fusion (Waheed et al. 2006). An amphotericinresistant HIV variant has been isolated, which has acquired the R peptide (Waheed et al. 2007). The Env cytoplasmic domain of this variant is cleaved by the viral protease during virion maturation, and the membrane fusion capability of the Env protein is activated. Thus, amphotericin B methyl ester could function to inhibit the conformational change of the HIV Env protein into the fusogenic form, and the cytoplasmic domain truncation of Env by the viral protease could rescue this conformational change in the presence of the cholesterol-binding agent.
3.3.5 Cleavage by Cathepsin Proteases Endosome acidification participates in infections by many retroviruses, such as ecotropic MLV (Katen et al. 2001, McClure et al.1990), avian leukosis virus (Mothes et al. 2000), mouse mammary tumor virus (Wang et al. 2008), Jaagsiekte sheep retrovirus (Bertrand et al. 2008), and equine infectious anemia virus (Brindly and Maury 2005, Jin et al. 2005). This suggests that entry of these retroviruses occurs through acidic endosomal compartments, and endosome acidification potentiates membrane fusion activity of Env (Fig. 3.5). Influenza virus infection also requires endosome acidification. Influenza virus cannot induce syncytia in susceptible cells at natural pH, but low pH treatment of influenza virus particles induces cell–cell fusion, indicating that the low pH treatment directly triggers a conformational change into a membrane fusion-inducing form. However, low pH treatment of retroviral particles does not induce cell fusion, suggesting that low pH is required but is not sufficient for the membrane fusion. Recently it has been reported that
3
Retroviral Membrane Fusions
51
Fig. 3.5 Entry pathways of HIV and MLV
ecotropic MLV infection and syncytium formation are significantly suppressed in cathepsin B-knockout cells, and in cells treated with a cathepsin B inhibitor, CA074Me (Kumar et al. 2007). These data indicate that cathepsin B is required for membrane fusion activity by the ecotropic MLV Env protein. Cathepsin L is also required for the ecotropic MLV infection in cathepsin B-negative cells, suggesting that either cathepsin B or L is sufficient for the ecotropic virus infection (Yoshii et al. 2009). Activation of these cathepsin proteases requires low pH in acidic endosomes. These findings suggest that digestion of the Env protein by cathepsin B or L protease in acidic endosomes induces a full conformational change of the ecotropic Env protein to the membrane fusion active form. HIV infection appears to occur through the endosomal compartment, as with other retroviruses (Miyauchi et al. 2009). However, inhibitors of endosome acidification do not attenuate the HIV infection, but instead enhance it (Fredricksen et al. 2002, Wei et al. 2005). This result suggests that HIV entry into the host cell cytoplasm should occur at the plasma membrane or before the HIV particle-containing endosomes progress to acidic late endosomes through endocytosis (Fig. 3.5). If so, cathepsin proteases are not required for the membrane fusion by the HIV Env protein, because cathepsin proteases are activated by endosome acidification. Indeed, cathepsin inhibitors did not block HIV infection (unpublished data). However, multiple lines of evidence show that secreted cathepsins potentiate HIV infection and induce CD4-independent HIV infection of CD4-negative mammary epithelial cells (Ei Messaoudi et al. 2000, Moriuchi et al. 2000). These results suggest that cleavage of HIV Env by secreted cathepsin proteases triggers conformational changes to the fusion-active form without binding to CD4. Cell–cell fusion needed for multi-nucleated muscle fiber cell formation is impaired in cathepsin B-knockdown cells (Gogos et al. 1996) and by the CA-074Me
52
Y. Kubo
cathepsin B inhibitor treatment (Jane et al. 2002). This result shows that cathepsin B is also required for the cell fusion in myogenesis, suggesting that the mechanism of syncytium formation in myogenesis is similar to fusion induced by the retroviral Env protein.
3.4 Regulation of Retroviral Membrane Fusion by Cellular Factors 3.4.1 Lipid Raft Cellular membranes contain many different types of lipid components and these lipids does not distribute uniformly within membranes. Cholesterol and sphingolipids are enriched in specific regions of the plasma membrane, called lipid rafts. Glycophosphatidylinositol-anchored proteins and signaling proteins are specifically localized to the raft domains, and lipid rafts play important roles for many biological events (Simons and Ikonen 1997). Lipid rafts could also be associated with the retroviral membrane fusion. Methyl-β-cyclodextrin extracts cholosterol from the plasma membrane, destroying raft structure. Treatment of target cells with methyl-β-cyclodextrin inhibits infection and cell fusion mediated by HIV and MLV Env proteins (Kamiyama et al. 2009, Lu et al. 2002, Manes et al. 2000, Popik et al. 2002, Viad et al. 2002). Treatment of target cells with a cholesterol-binding agent, amphotericin B methyl ester, also inhibits HIV infection and membrane fusion (Waheed et al. 2006, 2007). L-Cycloserine, an inhibitor of serine palmitoyltransferase that inhibits the first enzyme of the sphingolipid pathway, blocks HIV-induced syncytium formation and infection (Mizrachi et al. 1996). 1-Phenyl-2-palmitoylamino 3-morpholine 1-propanol and N-butyldeoxynojirimycin, which inhibit glucosyl transferase activity to ceramide, suppress HIV entry and membrane fusion (Fischer et al. 1995, Hug et al. 2000, Puri et al. 2004). These results reveal that raft microdomains in the host cell plasma membrane are required for membrane fusion by the HIV Env protein (Fig. 3.6). In addition, treatment of HIV particles with methyl-β-cyclodextrin inhibits their infectivity, indicating that raft microdomains of the HIV envelope membrane are also required for membrane fusion (Guyader et al. 2002). Further, treatment of host cells by a cholesterol synthesis inhibitor, statin, attenuates HIV entry (Del Real et al. 2004). Statin also inhibits protein prenylation, and the function of the signaling molecule Rho requires prenylation. Therefore, statin treatment also inhibits Rho signal transduction. It is thought that the inhibitory effect of statin on HIV entry is through suppression of Rho functions, but not by inhibiting choresterol synthesis. After the HIV Env proteins bind to host cells, the receptor proteins are gathered to the raft domains through a cytoskeleton-dependent process (Manes et al. 2000, Popik and Alce 2004, Viad et al. 2002). As a result, multiple Env-receptor interactions are induced and membrane fusion efficiently takes place in the raft domain. Therefore, the raft domains serve as sites where the membrane fusion occurs by the HIV Env proteins (Fig. 3.6).
3
Retroviral Membrane Fusions
53
Fig. 3.6 Structure of HIV particle pre-fusion complex
CD4 is specifically localized to raft microdomains prior to HIV Env binding, but CXCR4 is not (Del Real et al. 2002, Manes et al. 2000, Nguyen and Taub 2002. It has been reported that HIV infection does not occur in target cells expressing a CD4 mutant that localizes outside raft microdomains (Del Real et al. 2002). These results suggest that raft localization of CD4 is required for HIV infection. However, another CD4 mutant protein that is not localized to raft microdomains has been reported to serve as the infection receptor as efficiently as wild-type (raft localized) CD4 (Percherancier et al. 2003, Popik and Alce 2004), suggesting that raft localization of CD4 is not necessarily required for HIV entry. However, methyl-β-cyclodextrin also inhibits the HIV infection in cells expressing the nonraft localized CD4 mutant, suggesting that HIV-cell fusion occurs through raft domains regardless of CD4 localization (Kamiyama et al. 2009, Popik and Alce 2004). Recently, it has been shown that methyl-β-cyclodextrin more significantly suppresses CD4-independent CXCR4-tropic HIV infection than CD4-dependent infection, suggesting that CXCR4 recruitment to the raft domain is primarily required for the HIV-1 infection (Kamiyama et al. 2009). Further studies are required to define the role of raft domains in HIV Env mediated-membrane fusion.
3.4.2 Cell Adhesion Molecules Because syncytium formation is induced between retroviral Env-expressing cells and neighbor susceptible cells, cell–cell contact enhanced by cell adhesion molecules could affect the cell fusion activity of retroviral Env proteins. A kind of integrin family cell adhesion molecule, LFA-1, facilitates HIV Env-mediated
54
Y. Kubo
syncytium formation and viral entry (Fortin et al. 1999, Pantaleo et al. 1991). ICAM-1, a ligand of LFA-1, is incorporated into HIV particles (Beausejour and Tremblay 2004), and enhances the viral entry through the interaction with LFA-1 (Fortin et al. 1997, Paquette et al. 1998) (Fig. 3.6). The interaction between VLA4 and the cognate receptor VCAM-1 also enhances the HIV entry into host cells. Many other cell adhesion molecules, ICAM-2 (Butini et al. 1994), ICAM-3 (Biggins et al. 2007, Butini et al. 1994, Sommerfelt and Asjo 1995), LFA-3 (Kalter et al. 1991), CD44 (Rivadeneira et al. 1995) were reported to be associated with the HIV Env-mediated cell fusion or entry. All of these cell adhesion molecules enhance cell fusion activity, because the cell adhesion molecules strengthen the contact between Env- and receptor-expressing cells.
3.4.3 Cytoskeleton-Associated Molecules After a HIV particle binds to the host cells, the receptor proteins are gathered to the HIV-binding site. An actin polymerization inhibitor, cytochalasin D, attenuates the receptor clustering and the HIV entry, indicating that actin-dependent receptor clustering is required for HIV Env-mediated membrane fusion (Iyengar et al. 1998, Jolly et al. 2004, Kizhatil and Albritton 1997, Lehmann et al. 2005) (Fig. 3.6). The microtubule cytoskeleton also regulates HIV Env-mediated cell fusion and entry (Valenzuela-Fernandez et al. 2005). However, there is no evidence showing that the HIV receptor proteins directly interact with the cytoskeleton. Linker molecules between the HIV receptor and cytoskeleton are involved in the HIV Env-induced membrane fusion. Filamin A has been reported to function as the linker molecule between the HIV receptor proteins and cytoskeleton, and a dominant negative mutant of filamin A attenuated HIV Env-mediated cell fusion (Jimenez-Baranda et al. 2007) (Fig. 3.6). Ezrin, radixin, and moesin (ERM) proteins that function as linker molecules between certain membrane proteins and the cytoskeleton are also associated with the HIV Env-mediated membrane fusion and entry into host cells (Barrero-Villar et al. 2009, Kubo et al. 2008). ERM proteins have also been reported to inhibit nuclear import of the HIV core by stabilizing microtubule formation (Haedicke et al. 2008, Naghavi et al. 2007). The C-terminal verprolin-homologous, cofilin-homologous, and acidic (VCA) domain of the Wiscott-Aldrich syndrome protein (WASP) induces actin polymerization and branching, and has been shown to attenuate HIV infection (Komano et al. 2004). However, the WASP VCA peptide did not inhibit membrane fusion mediated by the HIV Env protein. It was speculated that actin networks induced by the WASP VCA peptide physically suppress nuclear import of the HIV core after virus entry into host cells. Additionally, it has been reported that actin depolymerization by cofilin is required for the nuclear import of the HIV core (Yoder et al. 2008) supporting the above speculation. These results suggest that the actin cytoskeleton positively and negatively controls membrane fusion by HIV Env and nuclear import of HIV core, respectively.
3
Retroviral Membrane Fusions
55
Protein tyrosine kinase inhibitors, genistein and herbimycin A, suppress retroviral Env-mediated cell fusion and entry into host cells, suggesting that tyrosine kinases are involved in the membrane fusion activity (Cohen et al. 1992, Kubo et al. 2003b, Stantchev et al. 2007). A tyrosine kinase, Lck, enhances HIV Env-mediated syncytium formation (Briand et al. 1997, Yoshida et al. 1992), consistent with the above results. Because tyrosine kinase activity regulates cytoskeleton distribution and movement (Huveneers and Danen 2009), retrovirus Env-induced membrane fusion could be affected by tyrosine kinases through cytoskeleton rearrangement. Cytoskeleton rearrangement is also regulated by Rho signal transduction (Huveneers and Danen 2009). Therefore, cellular factors involved in Rho signaling have been shown to be associated with retroviral Env-mediated membrane fusion (Fig. 3.6). A dominant negative mutant of RhoA inhibits HIV entry into host cells (Jimenez-Baranda et al. 2007). Additionally, inhibition of Rho GTPase by statin attenuates HIV entry (Del Real et al. 2004). ROCK, a signaling molecule in the Rho pathway, stabilizes the actin cytoskeleton, and attenuates HIV infection (Jimenez-Baranda et al. 2007). HIV binding to target cells induces activation of Rac1, another signaling molecule in the Rho pathway (Pontow et al. 2004), and a Rac-1 dominant negative mutant or a Rac-1 inhibitor suppresses HIV Env-induced cell fusion (Pontow et al. 2004, 2007). These findings reveal that Rho signal transduction is associated with HIV Env-mediated membrane fusion through cytoskeleton rearrangement.
3.5 Conclusion Before the fusion between retroviral envelope and host cell membrane is induced by Env, many cellular factors as well as the viral proteins are gathered at lipid raft microdomain (Fig. 3.6). The pre-fusion complex contains cytoskeleton, and cytoskeleton-regulating Rho signal-associated molecules. Other critical cellular factors contained in the pre-fusion complex have yet to be discovered. The cellular factors involved in the pre-fusion complex could be also involved in other cell– cell fusion reactions, including the formation of muscle fibers, syncytiotrophoblasts, and osteoclasts. The detailed understanding of retrovirus-induced membrane fusion could contribute to the development of novel anti-HIV drugs and to the elucidation of the other cell fusion mechanisms involved in tissue developments.
References Aguilar HC, Anderson WF, Cannon PM (2003) Cytoplasmic tail of Moloney murine leukemia virus envelope protein influences the conformation of the extracellular domain: implications for mechanism of action of the R peptide. J Virol 77:1281–1291 Albritton LM, Tseng L, Scadden D et al (1989) A putative murine ecotropic retrovirus receptor gene encodes a multiple membrane-spanning protein and confers susceptibility to virus infection. Cell 57:659–666
56
Y. Kubo
Barrero-Villar M, Cabrero JR, Gordon-Alonso M et al (2009) Moesin is required for HIV-1induced CD4-CXCR4 interaction, F-actin redistribution, membrane fusion and viral infection in lymphocytes. J Cell Sci 122:103–113 Battini JL, Rasko JE, Miller AD (1999) A human cell-surface receptor for xenotropic and polytropic murine leukemia viruses: possible role in G protein-coupled signal transduction. Proc Natl Acad Sci USA 96:1385–1390 Beausejour Y, Tremblay MJ (2004) Interaction between cytoplasmic domain of ICAM-1 and Pr55Gag leads to acquisition of host ICAM-1 by human immunodeficiency virus type 1. J Virol 78:11916–11925 Berger EA, Murphy PM, Farber JM (1999) Chemokine receptors as HIV-1 coreceptors: roles in viral entry, tropism, and disease. Annu Rev Immunol 17:657–700 Bertrand P, Cote M, Zheng YM et al (2008) Jaagsiekte sheep retrovirus utilizes a pH-dependent endocytosis pathway for entry. J Virol 82:2555–2559 Biggins JE, Biesinger T, Yu Kimata MT et al (2007) ICAM-3 influences human immunodeficiency virus type 1 replication in CD4+ T cells independent of DC-SIGN-mediated transmission. Virology 364:383–394 Bovkova M, Stitz J, Engelstadter M et al (2002) Identification of R-peptides in envelope proteins of C-type retroviruses. J Gen Virol 83:2241–2246 Briand G, Barbeau B, Corbeil J et al (1997) Enhancement of HIV-1-induced syncytium formation in T cells by the tyrosyl kinase p56lck. Virology 231:10–19 Brindly MA, Maury W (2005) Endocytosis and a low-pH step are required for productive entry of equine infectious anemia virus. J Virol 79:14482–14488 Brody BA, Rhee SG, Hunter E (1994) Postassembly cleavage of a retroviral glycoorprotein cytoplasmic domain removes a necessary incorporation signal and activate fusion activity. J Virol 68:462–467 Butini L, De Fougerolles AR, Vaccarezza M et al (1994) Intracellular adhesion molecules (ICAM)-1, ICAM-2, and ICAM-3 function as counter-receptors for lymphocyte functionassociated molecule 1 in human immunodeficiency virus-mediated syncytia formation. Eur J Immunol 24:2191–2195 Cohen DI, Tani Y, Tian H et al (1992) Participation of tyrosine phosphorylation in the cytopathic effect of human immunodeficiency virus-1. Science 256:542–545 Del Real G, Jimenez-Baranda S, Lacalle RA et al (2002) Blocking of HIV infection by targeting CD4 to nonraft membrane domains. J Exp Med 196:293–301 Del Real G, Jimenez-Baranda S, Mira E et al (2004) Statins inhibit HIV-1 infection by downregulating Rho activity. J Exp Med 200:541–547 Diaz RM, Bateman A, Emiliusen L et al (2000) A lentiviral vector expressing a fusogenic glycoprotein for cancer gene therapy. Gene Ther 7:1656–1663 Drewlo S, Leyting S, Kokozidou M et al (2006) C-terminal truncations of syncytin-1 (ERVWE1 envelope) that increase its fusogenicity. Biol Chem 387:1113–1120 Dubay JW, Dubay SR, Shin HJ et al (1995) Analysis of the cleavage site of the human immunodeficiency virus type 1 glycoprotein: requirement of precursor cleavage for glycoprotein incorporation. J Virol 69:4675–4682 Dupressoir A, Vernochet C, Bawa O, Harper F, Pierron G et al (2009) Syncytin-A knockout mice demonstrate the critical role in placentation of a fusogenic, endogenous retrovirus-derived, envelope gene. Proc Natl Acad Sci USA 106:12127–12132 Edinger AL, Blanpain C, Kunstman KJ et al (1999) Functional dissection of CCR5 coreceptor function through the use of CD4-independent simian immunodeficiency virus strains. J Virol 73:4062–4073 Ei Messaoudi K, Thiry LF, Liesnard C et al (2000) A human milk factor susceptible to cathepsin D inhibitors enhances human immunodeficiency virus type 1 infectivity and allows virus entry into a mammary epithelial cell lines. J Virol 74:1004–1007 Ferri KF, Jacotot E, Blanco J et al (2000) Apoptosis control in syncytia induced by the HIV type 1-envelope glycoprotein complex: role of mitochondria and caspases. J Exp Med 192:1081–1092
3
Retroviral Membrane Fusions
57
Fischer PB, Collin M, Karlsson GB et al (1995) The alpha-glucosidase inhibitor N-butyldeoxynojirimycin inhibits human immunodeficiency virus entry at the level of postCD4 binding. J Virol 69:5791–5797 Fortin JF, Barbeau B, Hedman H et al (1999) Role of the leukocyte function antigen-1 conformational state in the process of human immunodeficiency virus type 1-mediated syncytium formation and virus infection. Virology 257:228–238 Fortin JF, Cantin R, Lamontagne G et al (1997) Host-derived ICAM-1 glycoproteins incorporated on human immunodeficiency virus type 1 are biologically active and enhance viral infectivity. J Virol 71:3588–3596 Fredricksen BL, Wei BL, Yao J et al (2002) Inhibition of endosomal/lysosomal degradation increases the infectivity of human immunodeficiency virus. J Virol 11440–11446 Freed EO, Risser R (1987) The role of envelope glycoprotein processing in murine leukemia virus infection. J Virol 61:2852–2856 Galanis E, Bateman A, Johnson K et al (2001) Use of viral fusogenic membrane glycoproteins as novel therapeutic transgenes in gliomas. Hum Gene Ther 12:811–821 Gogos JA, Thompson R, Lowry W et al (1996) Gene trapping in differentiating cell lines: regulation of the lysosomal protease cathepsin B in skeletal myoblast growth and fusion. J Cell Biol 134:837–847 Goodman LJ, Kain SR, Firestone GL (1993) Trafficking of wild-type and an endoproteolytic-site mutant of the mouse mammary tumor virus glycoprotein. J Biol Chem 268:2329–2336 Green N, Shinnick TM, Witte O et al (1981) Sequence-specific antibodies show that maturation of Moloney leukemia virus envelope polyprotein involves removal of a COOH-terminal peptide. Proc Natl Acad Sci USA 78:6023–6027 Gu M, Rappaport J, Leppla SH (1995) Furin is important but not essential for the proteolytic maturation of gp160 of HIV-1. FEBS Lett 365:95–97 Guo HG, Veronese FM, Tschachler E et al (1990) Characterization of an HIV-1 point mutant blocked in envelope glycoprotein cleavage. Virology 174:217–224 Guyader M, Miyokawa E, Abrami L et al (2002) Role for human immunodeficiency virus type 1 membrane cholesterol in viral internalization. J Virol 76:10356–10364 Haedicke J, de Los Santos K, Goff SP et al (2008) The ezrin-radixin-moesin family member ezrin regulates stable microtube formation and retroviral infection. J Virol 82:4665–4670 Hahn BH, Compans RW (1992) Cytoplasmic domain truncation enhances fusion activity by the exterior glycoproteincomplex of human immunodeficiency virus type 2 in selected cell types. J Virol 66:3971–3975 Hallenberger S, Bosch V, Angliker H et al (1992) Inhibition of furin-mediated cleavage activation of HIV-1 glycoprotein gp160. Nature 360:358–361 Henderson LE, Sowder R, Copeland TD et al (1984) Quantitative separation of murine leukemia virus proteins by reversed-phase high-pressure liquid chromatography reveals newly described Gag and Env cleavage products. J Virol 52:492–500 Hug P, Lin HM, Korte T et al (2000) Glycosphingolipids promote entry of a broad range of human immunodeficiency virus type 1 isolates into cell lines expressing CD4, CXCR4, and/or CCR5. J Virol 74:6377–6385 Huveneers S, Danen EH (2009) Adhesion signaling – crosstalk between integrins, Src, and Rho. J Cell Sci 122:1059–1069 Iyengar S, Hildreth JE, Schwartz DH (1998) Actin-dependent receptor colocalization required for human immunodeficiency virus entry into host cells. J Virol 72:5251–5255 Jane DT, Morvay LC, Allen F et al (2002) Selective inhibition of cathepsin B with cellpermeable CA074Me negatively affects L6 rat myoblast differentiation. Biochem Cell Biol 80:457–465 Jimenez-Baranda S, Gomez-Mouton C, Rojas A et al (2007) Filamin-A regulates actin-dependent clustering of HIV receptors. Nat Cell Biol 9:838–846 Jin S, Zhang B, Weisz OA et al (2005) Receptor-mediated entry by equine infectious anemia virus utilizes a pH-dependent endocytic pathway. J Virol 79:14489–14497
58
Y. Kubo
Jolly C, Kashefi K, Hollinshead M et al (2004) HIV-1 cell to cell transfer across an Env-induced, actin-dependent synapse. J Exp Med 199:283–293 Jones JS, Risser R (1993) Cell fusion induced by the murine leukemia virus envelope glycoprotein. J Virol 67:67–74 Kalter DC, Gendelman HE, Meltzer MS (1991) Inhibition of human immunodeficiency virus infection in monocytes by monoclonal antibodies against leukocyte adhesion molecules. Immunol Lett 30:219–227 Kamiyama H, Yoshii H, Tanaka Y et al (2009) Raft localization of CXCR4 is primarily required for X4-tropic human immunodeficiency virus type 1 infection. Virology 386:23–31 Katen LJ, Januszeski MM, Anderson WF et al (2001) Infectious entry by amphotropic as well as ecotropic murine leukemia viruses occurs through an endocytic pathway. J Virol 75:5018–5026 Kavanaugh MP, Miller DG, Zhang W et al (1994) Cell-surface receptors for gibbon ape leukemia virus and amphotropic murine retrovirus are inducible sodium-dependent phosphate symporters. Proc Natl Acad Sci USA 91:7071–7075 Kieman RE, Freed EO (1998) Cleavage of the murine leukemia virus transmembrane Env protein by human immunodeficiency virus type 1 protease: transdominant inhibition by matrix mutations. J Virol 72:9621–9627 Kim FJ, Manel N, Boublik Y et al (2003) Human T-cell leukemia virus type 1 envelope-mediated syncytium formation can be activated in resistant mammalian cell lines by a carboxy-terminal truncation of the envelope cytoplasmic domain. J Virol 77:963–969 Kizhatil K, Albritton LM (1997) Requirement for different components of the host cell cytoskeleton distinguish ecotropic murine leukemia virus entry via endocytosis from entry via surface fusion. J Virol 71:7145–7156 Klement V, Rowe WP, Hartley JW et al (1969) Mixed culture cytopathogenicity: a new test for growth of murine leukemia viruses in tissue culture. Proc Natl Acad Sci USA 63:753–758 Komano J, Miyauchi K, Matsuda Z et al (2004) Inhibiting the Arp2/3 complex limits infection of both intracellular mature vaccinia virus and primate lentiviruses. Mol Biol Cell 15:5197–5207 Kubo Y, Amanuma H (2003) Mutational analysis of the R peptide cleavage site of Moloney murine leukemia virus envelope protein. J Gen Virol 84:2253–2257 Kubo Y, Ishimoto A, Amanuma H (2003a) N-linked glycosylation is required for XC cellspecific syncytium formation by the R peptide-containing envelope protein of ecotropic murine leukemia viruses. J Virol 77:7510–7516 Kubo Y, Ishimoto A, Amanuma H (2003b) Genistein, a protein tyrosine kinase inhibitor, suppresses the fusogenicity of Moloney murine leukemia virus envelope protein in XC cells. Arch Virol 148:1899–1914 Kubo Y, Ishimoto A, Ono T et al (2004) Determinant for the inhibition of ecotropic murine leukemia virus infection by N-linked glycosylation of the rat receptor. Virology 330:82–91 Kubo Y, Ono T, Ogura M et al (2002) A glycosylation-defective variant of the ecotropic murine retrovirus receptor is expressed in rat XC cells. Virology 303:338–344 Kubo Y, Tominaga C, Yoshii H et al (2007a) Characterization of R peptide of murine leukemia virus envelope glycoproteins in syncytium formation and entry. Arch Virol 152:2169–2182 Kubo Y, Yokoyama M, Yoshii H et al (2007b) Inhibitory role of CXCR4 glycan in CD4independent X4-tropic human immunodeficiency virus type 1 infection and its abrogation in CD4-dependent infection. J Gen Virol 88:3139–3144 Kubo Y, Yoshii H, Kamiyama H et al (2008) Ezrin, Radixin, and Moesin (ERM) proteins function as pleiotropic regulators of human immunodeficiency virus type 1 infection. Virology 375:130–140 Kumar P, Nachagari D, Fields C et al (2007) Host cell cathepsins potentiate Moloney murine leukemia virus infection. J Virol 81:10506–10514 Lavillette D, Marin M, Ruggieri A et al (2002) The envelope glycoprotein of human endogenous retrovirus type W uses a divergent family of amino acid transporters/cell surface receptors. J Virol 76:6442–6452
3
Retroviral Membrane Fusions
59
Lehmann MJ, Sherer NM, Marks CB et al (2005) Actin- and myosin-derived movement of viruses along filopodia precedes their entry into cells. J Cell Biol 170:317–325 Lerner DL, Elder JH (2000) Expanded host cell tropism and cytopathis properties of feline immunodeficiency virus strain PPR subsequent to passage through interleukin-2-independent T cells. J Virol 74:1854–1863 Li M, Li ZN, Yao C et al (2006) Murine leukemia virus R peptide inhibits influenza virus hemagglutinin-induced membrane fusion. J Virol 80:6106–6114 Lombardi VC, Ruscetti FW, Das Gupta J et al (2009) Detection of an infectious retrovirus, XMRV, in blood cells of patients with chronic fatigue syndrome. Science 326:585–589 Loving R, Ki K, Wallin M et al (2008) R-peptide cleavage potentiates fusion-controling isomerization of the intersubunit disulfide in Moloney murine leukemia virus Env. J Virol 82:2594–2597 Lu X, Xiong Y, Silver J (2002) Asymmetric requirement for cholesterol in receptor-bearing but not envelope-bearing membranes for fusion mediated by ecotropic murine leukemia virus. J Virol 76:670–679 Manes S, Del Real G, Lacalle RA et al (2000) Membrane raft microdomains mediated lateral assemblies required for HIV-1 infection. EMBO Rep 1:190–196 Manel N, Kim FJ, Kinet S et al (2003) The ubiquitous glucose transporter GLUT-1 is a receptor for HTLV. Cell 14:449–459 Marras D, Bruggeman LA, Gao F et al (2002) Replication and compartmentalization of HIV-1 in kidney epithelium of patients with HIV-associated nephropathy. Nature Med 8:522–526 McClure MO, Sommerfelt MA, Marsh M et al (1990) The pH independence of mammalian retrovirus infection. J Gen Virol 71:767–773 Mi S, Lee X, Li X et al (2000) Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature 403:785–789 Miyauchi K, Kim Y, Latinovic O et al (2009) HIV enters cells via endocytosis and dynamindependent fusion with endosomes. Cell 137:433–444 Mizrachi Y, Lev M, Harish Z et al (1996) L-cycloserine, an inhibitor of sphingolipid biosynthesis, inhibits HIV-1 cytopathic effects, replication, and infectivity. J Acquir Immune Defic Syndr Hum Retrovirol 11, 137–141 Moriuchi H, Moriuchi M, Fauci AS (2000) Cathepsin G, a neutrophil-derived serine protease, increases susceptibility of macrophages to acute human immunodeficiency virus type 1 infection. J Virol 74:6849–6855 Mothes W, Boerger AL, Narayan S et al (2000) Retroviral entry mediated by receptor priming and low pH triggering of an envelope glycoprotein. Cell 103:679–689 Murakami T, Ablan S, Freed EO et al (2004) Regulation of human immunodeficiency virus type 1 Env-mediated membrane fusion by viral proteases activity. J Virol 78:1026–1031 Naghavi MH, Valente S, Hatziioannou T et al (2007) Moesin regulates stable microtube formation and limits retroviral infection in cultured cells. EMBO J 26:41–52 Nguyen DH and Taub D (2002) CXCR4 function requires membrane cholesterol: implications for HIV infection. J Immunol 168:4121–4126 Ohnishi Y, Shioda T, Nakayama K et al (1994) A furin-defective cell line is able to process correctly the gp160 of human immunodeficiency virus type 1. J Virol 68:4075–4079 Overbaugh J, Miller AD, Eiden MV (2001) Receptor and entry cofactors for retroviruses include single and multiple transmembrane-spanning proteins as well as newly described glycophosphatidylinositol-anchored and secreted proteins. Microbiol Mol Biol Rev 65:371–389 Pantaleo G, Butini L, Graziosi C et al (1991) Human immunodeficiency virus (HIV) infection in CD4+ T lymphocytes genetically deficient in LFA-1: LFA-1 is required for HIV-mediated cell fusion but not for virus transmission. J Exp Med 173:511–514 Paquette JS, Fortin JF, Blanchard L et al (1998) Level of ICAM-1 surface expression on virus producer cells influences both the amount of virion-bound host ICAM-1 and human immunodeficiency virus type 1 infectivity. J Virol 72:9329–9336
60
Y. Kubo
Percherancier Y, Lagane B, Planchenault T et al (2003) HIV-1 entry into T cells is not dependent on CD4 and CCR5 localization to sphingolipid-enriched, detergent-resistant, raft membrane domains. J Biol Chem 278:3153–3161 Pontow S, Harmon B, Campbell N et al (2007) Antiviral activity of a Rac GEF inhibitor characterized with a sensitive HIV/SIV fusion assay. Virology 368:1–6 Pontow SE, Heyden NV, Wei S et al (2004) Actin cytoskeleton reorganizations and coreceptormediated activation of rac during human immunodeficiency virus-induced cell fusion. J Virol 78:7138–7147 Popik W, Alce TW (2004) CD4 receptor localized to non-raft membrane microdomains supports HIV-1 entry. Identification of a novel raft localization marker in CD4. J Biol Chem 279:704–712 Popik W, Alce TM, Au WC (2002) Human immunodeficiency virus type 1 uses lipid raftcolocalized CD4 and chemokine receptors for productive entry into CD4+ T cells. J Virol 76:4709–4722 Puffer BA, Pohlmann S, Edinger D et al (2002) CD4 independence of simian immunodeficiency virus Envs is associated with macrophage tropism, neutralization sensitivity, and attenuated pathogenicity. J Virol 76:2595–2605 Puri A, Rawat SS, Lin HM et al (2004) An inhibitor of glycosphingolipid metabolism blocks HIV-1 infection of primary T-cells AIDS 18:849–858 Ragheb JA, Anderson WF (1994) pH-independent murine leukemia virus ecotropic envelopemediated cell fusion: implications for the role of the R peptide and p12E TM in viral entry. J Virol 68:3220–3231 Rein A, Mirro J, Haynes JG et al (1994) Function of the cytoplasmic domain of a retroviral transmembrane protein: p15E-p12E cleavage activates the membrane fusion capability of the murine leukemia virus Env protein. J Virol 68:1773–1781 Ritter GD, Mulligan MJ, Lydy SL et al (1993) Cell fusion activity of the simian immunodeficiency virus envelope protein is modulated by the intracytoplasmic domain. Virology 197:255–264 Rivadeneira ED, Sauls DL, Yu Y et al (1995) Inhibition of HIV type 1 infection of mononuclear phagocytes by anti-CD44 antibodies. AIDS Res Hum Retroviruses 11:541–646 Scheller C, Jassoy C (2001) Syncytium formation amplifies apoptotic signals: a new view on apoptosis in HIV infection in vitro. Virology 282:48–55 Schlaberg R, Choe DJ, Brown KR et al (2009) XMRV is present in malignant prostatic epithelium and is associated with prostate cancer, especially high-grade tumors. Proc Natl Acad Sci USA 106:16351–16356 Simons K, Ikonen E (1997) Functional rafts in cell membranes. Nature 387:569–572 Sodroski J G (1999) HIV-1 entry inhibitors in the side pocket. Cell 99:243–246 Sommerfelt MA, Asjo B (1995) Intracellular adhesion molecule 3, a candidate human immunodeficiency virus type 1 co-receptor on lymphoid and monocytoid cells. J Gen Virol 76:1345–1352 Spearman MA, Jamieson JC, Wright JA (1987) Studies on the effect of glycoprotein processing inhibitors on fusion of L6 myoblast cell lines. Exp Cell Res 168:116–126 Stantchev TS, Markovic I, Telford WG et al (2007) The tyrosine kinase inhibitor genistein blocks HIV-1 infection in primary human macrophages. Virus Res 123:178–189 Svoboda J (1960) Presence of chicken tumor virus in the sarcoma of the adult rat inoculated after birth with Rous sarcoma tissue. Nature 186:980–981 Tailor CS, Lavilette D, Martin M et al (2003) Cell surface receptors for gammaretroviruses. Curr Top Microbiol Immunol 281:29–106 Tailor CS, Nouri A, Lee CG et al (1999) Cloning and characterization of a cell surface receptor for xenotropic and polytropic murine leukemia viruses. Proc Natl Acad Sci USA 96:927–932 Urisman A, Molinaro RJ, Fischer N et al (2006) Identification of a novel gammaretrovirus in prostate tumors of patients homozygous for R462Q RNASEL variant. PLoS Pathog 2:e25 Valenzuela-Fernandez A, Alvarez S, Gordon-Alonso M et al (2005) Histone deacetylase 6 regulates human immunodeficiency virus type 1 infection. Mol Biol Cell 16: 5445–5454
3
Retroviral Membrane Fusions
61
Viad M, Parolini I, Sargiacomo M et al (2002) Role of cholesterol in human immunodeficiency virus type 1 envelope protein-mediated fusion with host cells. J Virol 76:11584–11595 Waheed AA, Ablan SD, Mankowski MK et al (2006) Inhibition of HIV-1 replication by amphotericin B methyl ester: selection for resistant variants. J Biol Chem 281:28699–28711 Waheed AA, Ablan SD, Roser JD et al (2007) HIV-1 escape from the entry-inhibiting effects of a cholesterol-binding compound via cleavage of gp41 by the viral protease. Proc Natl Acad Sci USA 104:8467–8471 Wallin M, Ekstrom M, Garoff H (2004) Isomerization of the intersubunit disulphide-bond in Env controls retrovirus fusion. EMBO J 23:54–65 Wang E, Obeng-Adjei N, Ying Q et al (2008) Mouse mammary tumor virus uses mouse but not human transferrin receptor 1 to reach a low pH compartment and infect cells. Virology 381:230–240 Wang H, Klamo E, Kuhmann SE et al (1996) Modulation of ecotropic murine retroviruses by N-linked glycosylation of the surface receptor/amino acid transporter. J Virol 70:6884–6891 Wei BL, Denton PW, O’Neill E et al (2005) Inhibition of lysosome and proteasome function enhances human immunodeficiency virus type 1 infection. J Virol 79:5705–5712 Wong PK, Yuen PH, Kaufman SJ (1977) Induction of syncytia by Moloney murine leukemia virus in myoblasts defective in differentiation. J Virol 21:319–327 Xiao P, Usami O, Suzuki Y et al (2008) Characterization of a CD4-independent clinical HIV-1 that can efficiently infect human hepatocytes through CXCR4. AIDS 22:1749–1757 Yang C, Compans RW (1996) Analysis of the cell fusion activities of chimeric simian immunodeficiency virus-murine leukemia virus envelope proteins: inhibitory effects of the R peptide. J Virol 70:248–254 Yoder A, Yu D, Dong L et al (2008) HIV envelope-CXCR4 signaling activates cofilin to overcome cortical actin restriction in resting CD4 T cells. Cell 134:782–792 Yoshida H, Koga Y, Moroi Y et al (1992) The effect of p56lck, a lymphocyte specific protein tyrosine kinase, on the syncytium formation induced by human immunodeficiency virus envelope glycoprotein. Int Immunol 4:233–242 Yoshii H, Kamiyama H, Amanuma H et al (2008) Mechanisms underlying glycosylation-mediated loss of ecotropic receptor function in murine MDTF cells and implications for receptor evilution. J Gen Virol 89:297–305 Yoshii H, Kamiyama H, Minematsu K et al (2009) Cathepsin L is required for ecotropic murine leukemia virus infection in NIH3T3 cells. Virology 394:227–234 Zerhouni B, Nelson JA, Saha K (2004) Isolation of CD4-independent primary human immunodeficiency virus type 1 isolates that are syncytium inducing and acutely cytopathic for CD8+ lymphocytes. J Virol 78:1243–1255 Zingler K, Littman DR (1993) Truncation of the cytoplasmic domain of the simian immunodeficiency virus envelope glycoprotein increases Env incorporation into particles, fusogenicity, and infectivity. J Virol 67:2824–2831
Chapter 4
A Comparative Portrait of Retroviral Fusogens and Syncytins Philippe Pérot, Cécile Montgiraud, Dimitri Lavillette, and François Mallet
Abstract The strongest candidates for developmentally regulated cellular fusogens in mammals are Syncytins which contribute to cell–cell fusion leading to placental syncytiotrophoblast in higher primates, rodents, lagomorphs and sheeps. They consist of domesticated endogenous retroviral envelope glycoproteins (Env) whose fusion properties depend on the initial recognition of a specific receptor. In order to clearly understand Syncytins characteristics, we will first illustrate molecular details characterizing the maturation of class I fusion proteins by introducing envelopedriven fusion in an infectious context, i.e. virus cell fusion, exemplifying each step that lead to functional virions with the most relevant model such as HIV-1 lentivirus or MLV and type D interference group retroviruses. In a second part, we will comparatively present the current knowledge concerning Syncytins and the associated three levels of complexity. First, the placenta is probably more variable in structure than any of the mammalian organs. Second, Syncytins recognize specific and highly function-divergent/unrelated receptors. Third, some Syncytins were shown to exhibit other functions than fusion, such as proliferation, immunomodulation, receptor interference and anti-apoptotic properties. We will conclude by a brief overview of the consequences of Syncytin expression outside of its privileged tissue. Keywords Fusion · placenta · retrovirus · endogenous retrovirus · envelope · Syncytin · enJSRV · receptor · hASCT1 · hASCT2 · MFSD2 · HYAL2 Abbreviations ASCT ALV ASLV BaEV
Alanine, serine and cysteine selective transporters Avian leukosis virus Avian sarcoma leukosis virus Baboon endogenous retrovirus
F. Mallet (B) Laboratoire Commun de Recherche Hospices Civils de Lyon – bioMérieux, Cancer Biomarkers Research Group, Centre Hospitalier Lyon Sud, 69495 Pierre Bénite cedex, France e-mail:
[email protected] Cécile Montgiraud and Dimitri Lavillette contributed equally to in this chapter L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_4, C Springer Science+Business Media B.V. 2011
63
64
BLV CA CAT-1 CT cyt DRM EBV ECT en EnCa Env ER ERV ESCRT Exo FeLV FcEV FP GaLV GCM Gp GPI h HELLP HERV HIV HR HTDV HTLV HYAL2 IDO IFN Ig IL IP3 JSRV KoRV LLP LTR m M MA MAO MFSD2 MLV MMTV Mo-MLV
P. Pérot et al.
Bovine leukemia virus Capsid Cationic amino acid transporter-1 Cytotrophoblast Cytoplasmic tail Rafts? Epstein-Barr virus Extravillous cytotrophoblasts Endogenous Endometrial carcinoma Envelope Endoplasmic reticulum Endogenous retrovirus Endosomal sorting complex required for transport Exogenous Feline leukemia virus Felis catus endogenous retrovirus Fusion peptide Gibbon ape leukemia virus Glial cell missing Glycoprotein Glycosylphosphatidylinositol Human Hemolysis, elevated liver enzymes and low platelets Human endogenous retrovirus Human immunodeficiency virus Heptad repeats Human teratocarcinoma-derived virus Human T-cell leukemia virus Hyaluronidase 2 Indoleamine 2,3-dioxygenase Interferon Immunoglobulin Interleukin Inositol-3-phosphate Jaagsiekte sheep retrovirus Koala retrovirus Lentivirus lytic peptides Long terminal repeat Mouse Mesenchyme Matrix Morpholino antisense oligonucleotide Major facilitator superfamily domain containing 2 Murine leukemia virus Mouse mammary tumor virus Moloney murine leukemia virus
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
M-PMV MS MSRV MuLV MVB MYA NC NO NWM OASIS ORF OWM PCR PcRV PDI PE Pit RBD RT SERV SIV SRV-1 SNARE SP SRP ST SU T TfR1 TGF TGN Th TM tm TNF TSE UA URE
65
Mason-Pfizer monkey virus Multiple sclerosis Multiple sclerosis associated retrovirus Murine leukemia virus Multivesicular bodies Million years ago Nucleocapsid Nitric oxide New world monkeys Old astrocytes specifically induced substance Open reading frame Old world monkeys Polymerase chain reaction Papio cynocephalus retrovirus Protein disulfide isomerase Preeclampsia Sodium-dependent phosphate symporter Receptor-binding domain Reverse transcriptase Simian endogenous retrovirus Simian immunodeficiency virus Simian retrovirus-1 Soluble N-ethylmaleimide-sensitive fusion protein attachment protein receptor Signal peptide Signal recognition particle Syncytiotrophoblast Surface unit Trophoblast Transferrin receptor 1 Transforming growth factor Trans-Golgi network T helper cells Transmembrane unit Transmembrane domain Tumor necrosis factor Trophoblast specific enhancer Uterine arteries Upstream regulatory element
Contents 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Contribution of the Envelope to the Retroviral Life Cycle . . . . 4.2.1 Synthesis of Env Glycoprotein and Viral Assembly . . . . 4.2.2 Virus-Host Cell Membrane Fusion: A Multistep Mechanism 4.2.3 Rous Meets Mendel . . . . . . . . . . . . . . . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
66 68 69 75 80
66 4.3 Syncytins and Cell–Cell Fusion . . . . . . . . . . . . . . . . . . 4.3.1 Integration, Domestication Steps and Biological Functions of Endogenous Viral Glycoproteins . . . . . . . . . . . . . 4.3.2 Fusion Mechanism and Receptor Recognition . . . . . . . . . 4.3.3 Retroviral Envelopes Are Involved in the Placenta Development 4.3.4 Syncytin-1 Expression Outside of Its Privileged Tissue . . . . 4.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
P. Pérot et al.
. . . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
83 85 89 94 102 103 104
4.1 Introduction Our life begins with fusion of our mother’s oocyte with one of our father’s spermatozoids. A consequent successful pregnancy depends on the 10 m2 placental syncytiotrophoblast resulting from fusion of abundant cytotrophoblastic cells. As the embryo develops, skeletal muscle differentiation depends on the fusion of mononucleated myoblasts to form multinucleated muscle fibers. In the adult body, macrophages can fuse to form either multinucleated osteoclasts that control the maintenance of the bones or multinucleated giant cells that are important for the immune response. Last but not least, fusion as a driver of embryonic stem cell differentiation suggests a new role of cell fusion in mammalian development. Overall, cell fusion is a process in which two or more cells become one by merging their plasma membranes. Fusion, with the exception of gametes and stem cells, produces only terminally differentiated, non-proliferating tissue, and is thus mainly involved in tissue maintenance or regeneration. The fused cells (syncytia) that contain several nuclei within a single cytoplasm may be homokaryons (homotypic fusion) or heterokaryons (heterotypic fusion) as derived from the fusion of similar or different origin cell types, respectively. In any case, fusion of two separate lipid bilayers in non aqueous environment first requires that they come in close contact. Second, an intermediate stage is characterized by the merger of only the closest contacting monolayer, a process called hemifusion. Third, the fully completed fusion results in whole bilayer merging following by the opening of the pore. It remains questionable whether cell–cell fusion involves the same type of mechanisms than in other membrane fusion events, such as intracellular vesicle fusion mainly based on SNAREs proteins (soluble N-ethylmaleimide-sensitive fusion protein attachment protein receptor) and virus-cell fusion achieved by transmembrane viral fusion proteins (Chen and Olson 2005). However, we might expect that longconserved syncytial structures, such as skeletal muscle that have become integral to the body plans of multiple phyla, may be formed by mechanisms that have been mostly conserved during evolution (Mohler 2009). It is thus striking to notice that little is known about the molecular actors that are involved in regulating and completing cell–cell fusion, and of what is known there is little conservation between species, suggesting that these mechanisms might have evolved independently (Chen et al. 2007). The multiple transmembrane-domain tetraspanin protein CD9 on the egg surface and the single transmembrane-domain protein IZUMO with an extracellular
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
67
immunoglobulin (Ig)-like domain on the spermatozoid surface contribute to the fusion of the mice gametes, although it is still a matter of conjecture whether the two molecules interact directly in trans to achieve the membrane fusion reaction (Inoue et al. 2005). Transmembrane-domain proteins with extracellular Ig-like domains have been implicated in homotypic macrophage fusion in rats (Han et al. 2000) and in cell–cell tethering prior myoblast fusion in drosophila melanogaster (Bour et al. 2000). Interestingly, the strongest candidates for developmentally regulated cellular fusogens in mammals are Syncytins, a family of single-pass transmembrane proteins, which contribute to cell–cell fusion leading to placental syncytiotrophoblast in higher primates, rodents, lagomorphs and sheeps. They consist of domesticated endogenous retroviral envelope glycoproteins whose fusion properties depend on the initial recognition of a specific receptor. With the exception of retrovirus-derived Syncytins, none of the cell surface proteins identified in the various cell–cell fusion processes resemble SNAREs or class I fusion protein, (i.e. fusion does not appear to be mediated by an α-helical bundle). Though, fusogenic proteins contribute to decrease the kinetic barrier to allow the fusion of the two bilayer membranes. Viral fusion proteins do so by using the force energy released during a protein conformational change to draw together the membranes. The understanding of the Syncytins dependant cell–cell fusions will likely parallel the mechanism of at least retroviral infection. Indeed, Syncytins are host domesticated genes derived from ancient retroviruses infections of the host germ line. Syncytins appear to group relatively distinct actors that may exhibit common principles leading to membrane fusion and hence are good examples of the various scheme of evolution to establish similar but different structures (microscopic and macroscopic) with similar roles. Such a dichotomy between distinct players with common principles was indeed proposed for all fusion processes by Martens and McMahon (2008). Three classes of viral fusogens have been described. The class I and II fusion proteins are characterized by trimers of hairpins containing a central α-helical coiled-coil or β-sheets structure, respectively, while the class III fusion proteins have a mixed secondary structure (Weissenhorn et al. 2007). We will first introduce envelope-driven fusion in an infectious context, i.e. virus-cell fusion, by illustrating each step leading to functional virions with the most relevant model such as HIV-1 lentivirus or MLV and type D interference group retroviruses. The purpose is to illustrate molecular details characterizing the maturation of class I fusion proteins, defined by the following four characteristics: the cleavage of an envelope protein precursor leading to surface and transmembrane subunits, a fusion peptide located just next to the cleavage site (except avian ASLV), a trimeric complex association, and the ability to form a hairpin structure, also called a coiled-coil structure, in its active fusion conformation. The progression of these structural rearrangements slows down the kinetic barrier between hemifusion and fusion-pore formation. Intriguingly, without unequivocal evidence of infectious agent, retroviral particles were observed in physiological situation (Lyden et al. 1994) but also in pathological ones (Perron et al. 1989; Boller et al. 1993) in man. They could derived from endogenous retroviral sequences, as the human genome (Lander et al. 2001) but also the mouse genome (Waterston et al. 2002) contain a huge amount of
68
P. Pérot et al.
endogenous retroviruses, reaching 8.5 and 9% of these genomes sizes, respectively. Although human and mouse species contain essentially different retroviral families, some of their coding sequences are still intact, e.g. 18 envelopes ORFs were identified in the human genome (Blaise et al. 2003; Blaise et al. 2005) including the two human Syncytins. This huge repertoire awaiting the identification of functions will be illustrated, as it may represent a third condition beside infectious viruses and domesticated envelope Syncytins. In a second part, we will comparatively present the current knowledge on Syncytins. Outstandingly, the situation comprises at least three levels of complexity. First, the placenta is probably the more variable in structure than any of the mammalian organ. Placentas are variously classified, as regard to (i) their form, being discoid (primates, lagomorphs and rodents) or cotyledonary (ruminant), (ii) the type of layer between fetal trophoblast and maternal endometrial surface, hemochorial (primates, lagomorphs and rodents) or epithelio- and syndesmochorial (ruminants), and (iii) the structure of the maternal-fetal interdigitation, villous type or villi (primates and sheep) or labyrinth (lagomorphs and rodents) (Bernirschke K, Comparative placentation at http://placentation.ucsd.edu). The latter, a continuous syncytiotrophoblast layer that covers the entire surface of the human placental villi which floats in maternal blood, is responsible of ion and nutrient exchanges and synthesizes steroid and peptide hormones such as progesterone and human chorionic gonadotropin (hCG) required for human gestation. Second, Syncytins recognize specific and highly function-divergent/unrelated receptors. In human, Syncytin-1 recognizes hASCT1 (Blond et al. 2000) and hASCT2 (Lavillette et al. 2002) receptors while Syncytin-2 binds to MFSD2 receptor (Esnault et al. 2008). In rodents, Syncytin-A and Syncytin-B possess unidentified but distinct receptors (Dupressoir et al. 2005), and in lagomorphs Syncytin-Ory1 functionally recognizes hASCT2 (Heidmann et al. 2009). In sheep, enJSRV envelope(s) interacts with HYAL2 (Dunlap et al. 2006). This illustrates that proteins involved in cell–cell fusion, such as Syncytin partner receptors, are likely to play pleiotropic roles in other cellular processes, e.g. transport of small molecules, but also modulation of membrane structures, with specificity being achieved through the coupling of these proteins to different upstream and downstream effectors. Third, Syncytins were shown to exhibit other functions than fusion, such as proliferation (Strick et al. 2007; Larsen et al. 2009), immunomodulation (Mangeney et al. 2007), receptor interference (Blond et al. 2000; Ponferrada et al. 2003) and anti-apoptotic properties (Strick et al. 2007; Knerr et al. 2007), these functions being not shared by all these proteins. We will conclude by a brief overview of the consequences of Syncytin-1 expression outside of its privileged tissue.
4.2 Contribution of the Envelope to the Retroviral Life Cycle Retroviral classification was initially based on virion morphology observed with electronic microscopy during maturation and assembly of particles (Coffin 1992). Accordingly, retroviruses are designated A-, B-, C- and D-type. The
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
69
International Comity of Taxonomy has now established seven genera of Retroviridae based on sequence homologies: Alpharetoviruses correspond to avian type C (Avian leukosis virus, ALV) retroviruses, Betaretroviruses to type B (Mouse Mammary Tumor Virus, MMTV) and D (Simian retrovirus-1, SRV-1) retroviruses, Gammaretroviruses to mammalian type C retroviruses (Murine Leukemia Virus, MLV), Deltaretroviruses to the ancient group of HTLV-BLV (Human T-cell Leukemia Virus/Bovine Leukemia Virus), Epsilonretroviruses (Waileye Dermal Sarcoma Viruses family), Lentiviruses group which contains HIV and SIV (Human and Simian Immunodeficiency Viruses) and Spumavirus including Human Foamy Viruses (van Regenmortel et al. 2000). Retroviruses are RNA enveloped viruses. They infect cells via a cellular receptor recognition followed by the fusion of virus and cell membranes. Upon entry, the next step of the retroviral life cycle consists of a retrotranscription stage mediated by the viral reverse transcriptase protein that converts the viral genomic RNA in double strand DNA. Subsequently, the viral genetic material is targeted to the nucleus and stably integrated in the host cell genome by the viral integrase. The integrated viral DNA is named provirus and is flanked by two Long Terminal Repeats (LTR) that act as transcriptional regulatory elements. The 5 LTR contains the promoter and enhancer signals while the 3 LTR contains the polyadenylation signal terminating the transcription. All the replication competent retroviruses contain at least three genes coding for the structural proteins (gag), the enzymatic proteins (pro-pol) and the envelope glycoprotein (env). During its life cycle the virus uses the gene replication machinery of the host cell. Herein, we will focus on the characteristics of the envelope (Env) protein that is composed of one surface unit (SU) and one transmembrane unit (TM) which is itself subdivided into three domains, an ectodomain, a strict transmembrane domain (tm) and a cytoplasmic tail (cyt) (Fig. 4.1a). Env glycoprotein will undergo several modifications to generate a mature and functional glycoprotein addressed to the plasma membrane in order to contribute to the virus infection-competency. Functionally, the SU domain is involved in receptor recognition and the TM subunit contains both the fusion peptide and the heptad repeat domains involved in fusion.
4.2.1 Synthesis of Env Glycoprotein and Viral Assembly During virus production, the host cell is basically preserved since the expression of fusogenic competent glycoproteins is highly controlled. Sequentially, the Env protein synthesis is initiated by the free-ribosomes, next modifications take place in the endoplasmic reticulum and then an oligomerized precursor is transported by vesicles to the golgi apparatus. Abundant glycoprotein at the surface of the cell could induce cellular death by syncytia formation, toxicity via receptor interaction, or immune recognition. That’s why the localization and the amount of the oligomerized retroviral envelope glycoprotein at the host cellular surface are highly modulated by fine trafficking and sequestration mechanisms. The receptor interference mechanism
70
P. Pérot et al.
a
SU
TM
CXXC SP
RBD
(K/R)X(K/R)R C
YXXΦ
CX6CC
FP HR1 HR2
tm
R
cyt
leucine zipper
plasma membrane
b Endoplasmic reticulum C
Folding
Disulfide bonds formation
C
Glycosylated precursor
trimerisation C C
C C
Signal peptide
c Vesicular transport to golgi
The golgi apparatus
Ribosomes
palmytoylation
C
C
C
C
SU-TM cleavage
C
C
C
C
N-glycans maturation and O-glycans addition
Plasma cell membrane
R
cyt m tm
Vesicular transport to plasma membrane
HR2 HR1
FP
RBD RBD
Fig. 4.1 Structure and synthesis of retroviral envelope. (a) Schematic portrait of an envelope prototype. SP, signal peptide (grey). SU, surface unit (yellow): contains RBD, receptor binding domain (gold yellow) and C, C-terminal domain (light yellow) with CXXC motif (generally CC with = L,I,V,F,M or W). (K/R)X(K/R)R, furin cleavage site. TM, transmembrane unit (hatched boxes): contains FP, fusion peptide (red); leucine zipper motif with HR1 (blue) and HR2 (green) heptad repeats followed by the CX6 CC motif; tm, trans-membrane anchorage domain (red, hatched); cyt, cytoplasmic tail with C-terminal R peptide (blue) containing YXX motif. Note that the ectodomain part of the TM contains a so-called immunosuppressive domain QNRX2 LDXLX5 GXC joining the CX6 CC motif (not illustrated). (b) Schematic maturation process of the envelope glycoprotein. The successive immature forms of the envelope glycoprotein are illustrated (petrol blue). Initial glycosylation sites (branch trees with open circles), disulfide bonds (C–C), palmitoylation sites (broken lines) and final glycosylation sites (branch trees with dark circles) are indicated. Color codes and abbreviations used in the final trimeric structure expressed at the plasma cell membrane are as given in a; post translational modifications and disulfide bonds are omitted
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
71
can also limit the amount of receptors available for fusion between infected cells. Finally, Env fusion competency may be a late event that occurs during virus budding as described for MLV. 4.2.1.1 Synthesis and Maturation of Env Glycoprotein As common cellular proteins, retroviral Env translation is initiated by free ribosomes in the cytosol. The signal peptide (SP) located in the N-terminus of the Env glycoprotein is the first synthesized segment. This initial step and the following ones are illustrated in Fig. 4.1b. SP length varies depending on the retrovirus family but its composition is conserved with an hydrophobic signal, recognized by the SRP (Signal Recognition Particle), that allows the anchorage of the nascent chain to the endoplasmic reticulum (ER). The nascent protein is translocated through the membrane of ER. At the end of the synthesis, the extracellular part of Env is folded in the lumen of the ER, as for cellular membrane proteins. The release of the protein in the lumen is impaired by a stop transfer region composed of a hydrophobic sequence followed by aromatic and charged amino acid which will delimitate the membrane anchored domain or transmembrane domain (tm) (Hunter and Swanstrom 1990). This tm domain is an α–helix constituted by at least 23 amino acids (for HIV-1) and a maximum of 36 amino acids (for MMTV) but it contains unexpected residus for alpha helix structure in the context of a membrane (helix breaker amino acids like glycine and proline or positively charged tryptophane or cysteine). The Env N-extremity (ectodomain in the future virion) is then located in the lumen of ER while the C-terminal part (cyt) of the protein remains in the cytosol. ER is then the site of co- and post-translational modifications such as N-glycosylations, protein folding, disulfide bonds formation and oligomerization (Ratner 1992). After the SP cleavage, the precursor is modified by N-glycosylations. Depending on the retrovirus, the number and the location of glycosylation sites are variable. For HIV-1, the protein presents an unusual highly glycosylation with 24–32 sites and sugars account for half of the molecular weight of the Env protein (Mizuochi et al. 1990). For the other retroviruses, the number of glycosylation is around 8. Almost all glycosylations are in the SU, and except for gammaretroviruses for which there is no glycosylation in the TM, others have 1 (Betaretroviruses like BaEV) to 7 (HIV-1) glycosylations in TM reflecting the weak exposition of this subunit hidden by SU. In all cases, glycosylations are essential for the folding, the trafficking, the cleavage and the recognition of the receptor by the Env protein (Polonoff et al. 1982). For example, N-linked glycans are critical determinants for the efficient recognition of CD4 T cells by HIV-1 gp120, since mutant protein lacking one N-glycan did not effectively stimulate CD4+ T cells (Li et al. 2008a). For MLV, due to the fewer number of glycosylations than for HIV-1 Env, their roles have been more characterized and if some glycosylations are not crucial for incorporation, they are involved in the stability of the postcleavage envelope complex (Li et al. 1997). Following glycosylation events, intramolecular disulfide bonds are formed in SU and TM subunits to generate loops in the secondary structure of the envelope protein. The cysteines of the SU involved in these bonds are well conserved in a
72
P. Pérot et al.
subgroup of retroviruses suggesting a similar domain organization. These cysteines are crucial for the folding and the transport of the envelope to different cellular compartments. The substitution of some cysteines in SU and TM of HIV-1 (Bolmstedt et al. 1991; Dedera et al. 1992) or other retroviruses like MuLV (Thomas and Roth 1995; Freed and Risser 1987) leads to a non cleaved envelope glycoprotein precursor retains into the cells. Finally, the loops generated by the disulfide bonds are essential for cellular receptor recognition (MacKrell et al. 1996). In addition, the TM of γ– and some β–retroviruses (but not for lentiviruses) contains specific cysteines that are important for SU-TM intermolecular association as it will be developed later. The disulfide bond formation is part of a more global control of modifications of the retroviral Env by the cell (Braakman and van Anken 2000). This control involved protein disulfide isomerase (PDI) (Fenouillet et al. 2007), and also chaperone molecules like GRB78 Bip (Earl et al. 1991), calnexin (Li et al. 1996) and calreticulin (Otteken and Moss 1996) as described for HIV-1. The failure to pass the quality control leads to a non cleaved envelope glycoprotein precursor that is kept either in endoplasmic reticulum or in the golgi apparatus. Retroviral Env glycoprotein leaves the ER in a trimeric form to reach the Golgi where N-glycans are matured, O-glycans added (Pinter and Honnen 1988; Bernstein et al. 1994) and the cysteines at the hedge of the tm domain are palmitoylated. (Yang et al. 1995) (Fig. 4.1b). The appropriate trimeric conformation (with glycosylation and disulfide bonds) being obtained, the precursor is then cleaved at a highly conserved site (Fig. 4.1a) among retroviruses by the furin-convertase protein into its two subunits e.g. gp120 (SU) and gp41 (TM) for HIV-1. The Env complex of HIV1, as described for most retroviral Env glycoproteins, is trimeric with six individual subunits (three gp120 and three gp41 subunits). It is the TM subunit that triggers the oligomerization, as the TM (associated or not to the SU) is always detected in the oligomerized forms (Einfeld and Hunter 1988; Earl et al. 1990). The main determinant of this trimerization is a region in heptad repeat (Gallaher et al. 1989; Poumbourios et al. 1997) with high homology with leucine zipper domain (Fass et al. 1996; Weissenhorn et al. 1997; Owens and Compans 1990). Interestingly, the SU is also a trimer when it is shedded (Tucker 1991; Owens and Compans 1990) whereas soluble SU expressed alone is usually a monomer (Poumbourios et al. 1997). Hence, TM initiates the trimerization and, after that, SU can stay as a noncovalently linked fragile trimer. The trimerization gives the required environment for the fusion by masking the fusion peptide that will be later freed following receptor binding and also confers the fusogenic potential to the glycoprotein. The two SU and TM subunits are either linked in a covalent or non-covalent way. For HIV-1, the existence of the soluble gp120 protein indicates a non-covalent link between SU and TM (Kowalski et al. 1987). The regions implicated in this interaction are principally the C1 and C5 region of the SU and the leucine zipper domain and the CX5 C region of the TM (Lopalco et al. 1993; Schulz et al. 1992). For most others retroviruses a covalent link was described at one point. In all the case, except MMTV and JSRV, a disulfide bond between the SU and the TM is formed between the highly conserved CX6 CC motif of the TM and the CXXC of the SU (Sitbon et al. 1991; Schulz et al. 1992; Pinter et al. 1997). This CXXC motif is extremely rare in cellular proteins
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
73
and is similar to a motif found in the catalytic site of enzyme involved in thiol isomerisation like PDI or thioredoxin (Sanders 2000; Pinter et al. 1997). This motif in the SU has been shown to be part of an autocatalytic isomerisation function of SU to destroy the initial bond between SU and TM that was established during Env synthesis and to create an intra-SU bond inside the CXXC motif (Wallin et al. 2004; Li et al. 2008b). This disulfide bond isomerisation is crutial for the fusogenicity of gammaretrovirus (MLV) (Fenouillet et al. 2008). It can be interesting to mention that HIV-1 bond can be reconstituted after recreating the motif in SU and TM (Binley et al. 2000). 4.2.1.2 Cellular Localization of Env Glycoprotein and Viral Assembly Complex multilevel interactions have been described between Env, Gag and sorting proteins involved in traffic of molecules or vesicles inside the cells. These proteins are involved in both Env trafficking and virus budding. To limit the quantity of Env at the cell surface, the Env undergoes endocytosis and is trafficked in endosomal pathways. These cell localizations are driven by traffic peptidic motifs, like for cell proteins, that have been characterized to direct cellular transmembrane proteins into different endosomal compartment (Bonifacino and Traub 2003). Lentiviruses, including HIV-1, are unusual in having transmembrane glycoproteins with much longer cyt intracytoplasmic tail (150 amino acids) than most other retroviruses (20–50 amino acids) (Hunter and Swanstrom 1990), suggesting that this domain has one or more functions specific to lentivirus replication or persistence. Two groups of motifs have been identified in HIV-1 cyt. The first group consists of three structurally conserved amphipathic alpha-helical domains, designated as lentivirus lytic peptides 1, 2, and 3 (LLP-1, LLP-2, and LLP-3) (Xu et al. 2006). LLP domains have been implicated in various functions, including Env cell surface expression, Env fusogenicity, and Env incorporation into virus particles (Piller et al. 2000). Several studies have suggested that Env is incorporated into virions via interactions between LLP and the matrix region of Gag. The second group of motifs regulates the intracellular trafficking of Env. At steady state, Env is predominantly located in the trans-Golgi network (TGN) (Takeda et al. 2003). This intracellular distribution results from dynamic cycling of Env between the cell surface, the endosomal compartment, and the TGN. Newly synthesized Env proteins undergo endocytosis after their arrival at the cell surface. Env internalization is mediated by the interaction of Y712SPL (YXX on prototype, Fig. 4.1a), a membrane-proximal tyrosine-based signal in the gp41 cyt tail, with the adaptor protein (AP) complexes of the cellular sorting machinery, involving the clathrin adaptor AP-2 in particular (Berlioz-Torrent et al. 1999). The cytoplasmic tail of many other retroviruses also contains a YXX motif, including gammaretrovirus like MLV (Song and Hunter 2003), RD114 (Sandrin and Cosset 2006), HTLV-1 (Berlioz-Torrent et al. 1999), M-PMV (Song and Hunter 2003; Song et al. 2005) and BLV (Inabe et al. 1999; Novakovic et al. 2004). The gp41 cyt also interacts with the TGN- and endosomebased clathrin-associated adaptor AP-1 via a dileucine motif (Berlioz-Torrent et al. 1999; Wyss et al. 2001) which induces its cellular retention. Some dileucine motifs
74
P. Pérot et al.
are also present in the cytoplasmique tail of MLV, RD114, M-PMV or BLV (Sandrin and Cosset 2006; Grange et al. 2000). Finally, Y802W803, a diaromatic motif involved in the retrograde transport of Env to the TGN, was also identified in the gp41 cyt (Takeda et al. 2003) and interacts with TIP47, a protein required for the retrograde transport from cell surface to the TGN in link with matrix (MA) interaction (Lopez-Verges et al. 2006). A retrograde acid cluster motif has also been identified in RD114 Env (Bouard et al. 2007) that induces a retrograde transport of the Env from the late endosome via interaction with PACS1 complexes. It should be note that the palmitoylated cysteines located in the tm are also involved in cell distribution of Env by contributing to the association with lipid rafts (Bhattacharya et al. 2004). These rafts seem to serve as platforms for virus assembly and budding (Suomalainen 2002). For example, membranes of HIV-1 virions have a higher cholesterol rate than the original infected cell, and HIV-1 virions produced in cells with synthesis defects in sphingolipid and cholesterol are less infectious (Brugger et al. 2006). If the cellular distribution of Env is linked to interaction with trafficking cell molecules, there is evidence indicating that an interaction between the TM cytoplasmic tail and the MA domain of the viral Gag polyprotein mediates Env packaging into particles. Gag structural HIV-1 polyprotein precursor consists of MA, CA (capsid), NC (nucleocapsid), and p6 proteins. The budding ability of retroviruses requires only Gag proteins. Indeed, Gag expression in the absence of other retroviral proteins is sufficient to the liberation of Env-free virions, but these pseudo virions are released independently of the cellular poles. However, when Env is coexpressed, the budding is restricted to the basolateral membrane and mutations of the cyt tyrosine in the membrane-proximal tyrosine-based signal Y712SPL disturb the polarized release of HIV-1 in polarized epithelial cellular models (Lodge et al. 1997; Owens et al. 1991). It is unclear how viral RNA, Gag and Env proteins reach the same site of the cellular plasmic membrane for perfect assembling and budding of the virion. It should be mention that HIV-1 and SIV budding are not only polarized in epithelial cells. In lymphocytes, the release of viruses is restricted to domain of contact between two cells or even between the cell and the culture plate (Bugelski et al. 1995; Pearce-Pratt et al. 1994). This is a budding in a virologic synapse (Morita and Sundquist 2004) and it is also driven by Env cyt motif. This polarisation of budding might have a physiological importance for the cell–cell transmission of the virus. In the case of MLV virus, a model has been developed that allows a better comprehension of Gag and viral RNA trafficking. It was shown that recruitment of glycoproteins by the gammaretroviral core proteins takes place in the intracellular compartments and not at the cell surface. Moreover, gammaretroviral core proteins could relocalize Env glycoproteins in late endosomes and could allow incorporation on viral particles (Sandrin et al. 2004; Bouard et al. 2007). Finally, it was proposed that the retrovirus budding depends on the cell types but might depend of the status of the infection and condition of the cells. In T-lymphocytes, it was initially described that the assembly and budding takes places at the plasma membrane (Barre-Sinoussi et al. 1983; Gelderblom et al. 1987), but recent reports indicated also an assembly in intracellular vesicle containing virus (Grigorov et al. 2006; Joshi et al. 2009). Similarly, in macrophages, HIV-1 assemble and bud in MVB
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
75
(multivesicular bodies) (Nydegger et al. 2003; Sherer et al. 2003; von Schwedler et al. 2003) and the viruses are then released outside the cell by fusion of the MVB with plasma membrane (Trojan Horse hypothesis). This is in agreement with the suggestion that retroviruses exploit a cell-encoded pathway of intercellular vesicle traffic, exosome exchange, for both the biogenesis of retroviral particles and a lowefficiency mode of infection (Gould et al. 2003; Fang et al. 2007). However, it was recently proposed that the vesicle containing viruses might have different genesis with some vesicles coming from the plasma membrane invaginations (Welsch et al. 2007). The reason for this discrepancies are not clear but involved Gag interaction with membrane, ESCRT (endosomal sorting complex required for transport) localizations, interferon induced proteins and lipidic composition of microdomain (Ono et al. 2004). 4.2.1.3 Fusion Competency Gammaretrovirus virions assemble and bud from the infected cells as immature particles that must undergo an additional proteolytic maturation to become infectious (Brody et al. 1992; Christodoulopoulos and Cannon 2001; Green et al. 1981; Rein et al. 1994). This maturation concerns the viral protease dependent cleavage of the so-called R peptide at the C-terminus of the cytoplasmic tail (Green et al. 1981; Rein et al. 1994) (see location on Fig. 4.1a). The R peptide inhibits the fusion, and different hypotheses have been proposed. Firstly, the R peptide contains the YXXφ internalization motif and the removal of this motif following the cleavage of the R peptide might induce higher amount of envelope at the surface membrane and consequently more fusion (Song and Hunter 2003). Secondly, another explanation is that following the R peptide cleavage, the remaining cyt tail forms a membraneembedded amphiphilic alpha-helix domain destabilizing the membrane (Zhao et al. 1998; Rozenberg-Adler et al. 2008). Thirdly, it was proposed that, as the R peptide contains a palmitoylation site, its removal induces the close trimerization of the cyt tail and drastic conformational changes in the ectodomain of Env (Aguilar et al. 2003) which might influence Env fusogenicity by destabilizing the SU-TM complexes. These conformational changes are necessary for the isomerisation of the SU-TM disulfide MLV Env (Loving et al. 2008). The R peptide cleavage is the last step leading to a fusion competent infectious MLV retrovirus but this final modification does not exist in lentiviruses which harbour a long cytoplasmic tail. However, studies indicate that artificial (HTLV, HIV or SIV) or natural (SIV) shortening of the cytoplasmic tails change the conformation of ectodomain (Edwards et al. 2002; Spies et al. 1994) and increase the fusogenicity of the Env in cell–cell fusion (Kim et al. 2003; Edwards et al. 2002; Spies et al. 1994).
4.2.2 Virus-Host Cell Membrane Fusion: A Multistep Mechanism Glycoproteins from enveloped viruses evolved to combine two main features. They have the capacity to bind with a specific cellular receptor and they harbour a fusion domain (peptide fusion and transmembrane domain) that can be activated to mediate the merging (fusion) of viral and cellular membranes.
76
P. Pérot et al.
a
RBD RBD
RBD RBD
FP HR1
FP HR1
HR2
HR2
tm CELL cyt
cyt
VIRUS
R
1. R peptide cleavage
2. Ectodomain destabilisation
Receptor
Receptor
RBD RBD
RBD
RBD
HR1FP
FP HR1
HR2
HR2
cyt
VIRUS
cyt
VIRUS
3. Binding to receptor
4. Release of fusion peptide
FP
CELL RBD
RBD
HR1
TM
HR2
HR1 HR2
VIRUS
SU
FP
VIRUS
cyt
5. Assembly of HR1 et HR2 helix (Six strands helix )
cyt
6. Gathering of viral and cellular membranes induced by the formation of the six strands helix
b VIRUS
CELL 1. Virus-Cell contact
2. Hemifusion stalk
3. Hemifusion diaphragm
4. Fusion pore
5. Core release into the cell
Fig. 4.2 Virus-host cell membrane fusion. a Schematic representation of six prototype stages beginning with the fusion competency acquisition of the envelope glycoprotein (1) based on R peptide release by viral protease and ending with the gathering of viral and cellular membranes (6) induced by the anchorage of the fusion peptide into the cell membrane. Red arrow symbolizes the R peptide cleavage. b Schematic drawing of the successive steps leading to lipidic pore formation. (1) proximal leaflets of viral (green) and cell (black) membranes coming into immediate contact,
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
77
Three different classes of viral fusion proteins have been identified to date based on key structural features at pre- and post-fusion stages. Many studies mentioned that the structural transition from a pre- to a post-fusion conformation leads to a stable hairpin conformation. This concerns the class I fusion proteins, characterized by trimers of hairpins containing a central α-helical coiled-coil structure, and the class II fusion proteins characterized by trimers of hairpins composed of beta sheets structures. A third class of fusion proteins has been described recently, that also forms trimers of hairpins by combining the two structural elements alpha-helix and beta-sheet structures (Weissenhorn et al. 2007). Three main steps are described for achieving the pre- to post conformational changes. The first one, after Env activation upon receptor binding or acidification of the endosomal compartment, exposes the fusion peptide that is projected toward the top of the glycoprotein, allowing the initial interaction with the target membrane (Fig. 4.2a, drawings 1–4). The second one is the folding back of the C-terminal region onto a trimeric N-terminal region (Fig. 4.2a, drawings 5–6) that leads to the formation of a post-fusion protein structure with the outer regions zipped up against the inner trimeric core in an antiparallel coiled coil structure. The final and third step also requires further refolding of the membrane proximal and transmembrane regions in order to obtain a full-length post-fusion structure where both membrane anchors are present in the same membrane (Fig. 4.2b). 4.2.2.1 Receptor Binding and Peptide Fusion Liberation HIV-1 fusion is mediated by specific interaction of the viral envelope glycoprotein with the cell surface CD4 molecule that serves as the primary receptor, and additionally a chemokine receptor CCR5 or CXCR4 as HIV-1 co-receptors. Both receptors and the co-receptor binding sites are on gp120 although the membrane fusion is triggered by conformational changes in the transmembrane protein gp41. The viral entry can be blocked by three categories of agents (Qian et al. 2009) (i) attachment inhibitors/antagonists targeting CD4, CCR5 and CXCR4 (ii) inhibition of the post-binding conformational changes, (iii) fusion inhibition. During fusion process, heptad repeats HR1 and HR2 form a six helix bundle structure. Synthetic peptides based on the HR1 and HR2 sequences of gp41 have anti-HIV-1 properties; this is up to date the most successful HIV-1 entry inhibitors class. Receptors of type C and D retroviruses are cell membrane anchored proteins that transport small molecules. The receptor of ectopic MLV type C retroviruses is CAT-1, a cationic amino acid (like lysine or arginine) transporter (Kavanaugh et al. 1994). The receptor of amphotrophic MLV is Pit-2 and the receptor of MLV10A1 is Pit-1. Pits-1 is also the receptor for GALV and FeLV. Pit-1 and Pit-2 are two inorganic phosphate transporters. Nevertheless, in some cases, these viruses are able
Fig. 4.2 (continued) (2) hemifusion stalk with proximal leaflets fused and unfused viral (blue) and cell (red) distal leaflets, (3) unfused stalk expansion leading to the hemifusion diaphragm, (4) fusion pore forming the hemifusion diaphragm bilayer, (5) core release into the cell
78
P. Pérot et al.
to recognize the common parts of Pit-1 and Pit-2 receptors and to infect human cells by using one or the other indifferently (Tailor et al. 2000). Two receptors have been identified for the type D retroviruses interference group: ASCT1 and ASCT2, two neutral amino acid transporters and receptors for BaEV, SRV and RD114 viruses. Retroviruses of the RD114 family recognize an hypervariable region in the second bundle of ASCT2 receptor. Most of the time the viruses use the part of the transporter that is involved in the transport function of this molecule: this region is under strong selection pressure to keep the function and therefore the polymorphism is limited and the infectivity, i.e. the recognition of the receptor by the viruses, is conserved. Furthermore these type D retroviruses can use either ASCT1 or ASCT2 by recognizing the conserved domain. Two types of fusion mechanisms can occur, namely pH independent and pHdependent. In the first case, the recognition between virus and receptor directly triggers conformational changes in the glycoprotein that leads to the direct fusion between the two membranes (viral and plasma) and to the liberation of the viral genetic material. The activation of Env at neutral pH allows the fusion in vitro and in vivo of Env-expressing cells co-cultured with receptor-expressing cells. The fusion leads to the merging of cytoplasms and to the generation of multinucleated cells named syncytia. In the second case, for pH-dependent fusion, the interaction between the Env and the receptor is followed by an endocytosis of the virus-receptor complex before the acidification of the endosome triggers conformational changes in the glycoprotein. For the pH dependent virus, such a fusion can be reproduced in vitro in cell culture or in liposome-virus fusion assay after decreasing the pH in the test tube, but cannot occur in vivo. Most retroviruses use a pH-independent fusion mechanism, with a few exceptions for MMTV, JSRV and ASLV. The proposed mechanism for ASLV virions is an intermediate since entry occurs in two steps, beginning with a receptor-priming that in turn induces Env conformational changes allowing the Env to become sensitive to the low pH. This hybrid mechanism does not lead to cell–cell fusion in vivo. JSRV also uses a receptor-priming for fusion activation of Env at low pH but the mechanism is slightly different that for ASLV (Cote et al. 2009) and requires dynamin-associated endocytosis (Bertrand et al. 2008). MMTV is so far considered as a classical pH-dependent virus that uses mouse transferrin receptor 1 (TfR1) and trafficking to a low pH compartment (Wang et al. 2008). Finally, it should be note that viruses that use a pH independent mechanism of activation of Env may still enter the cell by endocytosis without any requirement for acidification activation of Env into the endosomes. So far, there are many different endocytosis pathways that have been described (Marsh and Helenius 2006; Mercer and Helenius 2009) as being used by both pH-dependant and pH-independent viruses. However, re-investigations of the entry pathways are clearly needed for many pH-independent viruses that were thought not to rely on endocytosis. For example, Nipah paramyxoviruses that can fuse cells at neutral pH seem to use macropinocytosis for entry (Pernet et al. 2009). Macropinocytosis and phagocytosis have also been proposed for HIV-1 entry even if it is unsure that this entry can lead to productive infection (Marechal et al. 2001; Trujillo et al. 2007).
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
79
Let’s describe the several critical domains which are directly involved in the fusion process. For the most part, in retrovirus-cell fusion, a fusion unit typically contains a unique transmembrane domain and a fusion peptide – a sequence of 10–30 residues that forms an amphiphilic domain usually at the N-terminus of the TM (Chernomordik and Kozlov 2003). The hydrophobic fusion peptide domain is sequestered in all previously described Env biosynthesis steps. The final acquisition of the fusion state competency is triggered by the receptor binding alone and/or a low pH surrounding the endosomes and globular head domains dissociation. This movement allows a loop-to-helix transition of a polypeptide segment of TM that was previously buried underneath the SU heads, projecting the fusion peptide ∼100 Å towards the target membrane, where it inserts irreversibly. In the case of class I fusion proteins like retroviruses, this occurs by a “spring-loaded” mechanism. This initial change is proposed to result in a “pre-hairpin intermediate”, an extended structure that is anchored both in the target membrane by the fusion peptide and in the virus membrane by the tm segment (Fig. 4.2a drawings 5 and 6). The HR2 Cterminal end of the long TM α-helix jackknifes back, reversing the direction of the viral-membrane-proximal segment of TM, which then interacts in an anti-parallel fashion with the groove formed by the N-terminal HR1 trimeric coiled coil. The final post-fusion conformation of TM is therefore a highly stable rod with the TM and fusion-peptide segments together at the same end of the molecule, a structure termed a “trimer of hairpins”. The hairpin structure brings the two membranes proximal and provides free energy to overcome the barrier of membrane merging (Melikyan 2008). Membrane fusion occurs, which leads to pore formation and release of the viral genome into the cytoplasm. In addition, compare to cellular glycoproteins, the retroviral TM ectodomain also contains a hydrophobic domain abnormally enriched in tryptophane in the juxtamembrane domain (Salzwedel et al. 1999; Suarez et al. 2000). This domain contributes to the conformational change and membrane destabilization during the fusion process of HIV-1 (Munoz-Barroso et al. 1999), and antibodies (Lorizate et al. 2006; Purtscher et al. 1994) or peptides (Moreno et al. 2006) directed against this domain inhibit the entry. This juxtamembrane domain is also critical for fusion of many envelope viruses beside retroviruses, including paramyxoviruses and coranaviruses. 4.2.2.2 Pore Formation and Fusion of the Target Membranes The hypothesis of the pore model in viral membrane fusion mechanism (Fig. 4.2b) is supported by experimental results. The first evidence for a hemifusion intermediate was achieved by studying influenza virus entry that occurs after the hemagglutinin glycoprotein binding to the host cell. The substitution of the hemagglutinin transmembrane domain by a glycosylphosphatidylinositol (GPI) revealed the importance of the transmembrane region for the fusion pore opening and expansion. Hemifusion structures are connections between outer leaflets of apposed membranes, whereas the inner leaflets remain distinct. This is a transient structure that either dissociates or gives rise to the fusion pore (Chernomordik and Kozlov 2008). Interestingly, the
80
P. Pérot et al.
helix breaker residues within the tm domain are critical for the hemifusion and pore opening step of the fusion process mediated by different retroviral Env, like HIV-1 (Owens et al. 1994) and Mo-MLV (Taylor and Sanders 1999). In addition, a hemifusion intermediate has been detected in the case of HIV-1 Env-mediated fusion (Munoz-Barroso et al. 1998) by using peptide inhibitors that target a pre-fusion or prehairpin structure such as HIV-1 gp41 T-20. Then, the pore is formed and it allows a connection between two compartments initially separated by the apposed membranes. The membrane ability to hemifuse and develop fusion pore has been found to depend on the lipid microdomain composition, e.g. cholesterol (Chernomordik and Kozlov 2003). A potential lipid dependence of virus entry processes has been first deducted from experiments on influenza virus suggesting the implication of lipid rafts (Takeda et al. 2003). For retroviruses, the tm palmitoylations which contribute to the Env localization in raft domains (Li et al. 2002) influence indirectly the fusion process (Gebhardt et al. 1984; Ochsenbauer-Jambor et al. 2001). As an alternative to lipidic pore hypothesis, a direct fusion has also been proposed. The fusion pore is a full proteic channel-like structure dependent only on the transmembrane domains of the glycoproteins. In this model, the pore is opened by the joining of two hemipores located on each membrane (Chernomordik and Kozlov 2008; Chernomordik and Kozlov 2005). After fusion pore opening and enlargement (Melikyan et al. 2005), the genetic material enters the cytoplasm of the cell and enters the nucleus.
4.2.3 Rous Meets Mendel In humans, virus-like particles without trivial evidence of inter-individual transmissibility were identified in disparate contexts such as placenta, autoimmune diseases, e.g. body fluids of multiple sclerosis patients, and cancers, e.g. seminomas, lymphomas or plasma of breast cancer patients. In the seventies, numerous electron microscopic studies have described the presence of virus related particles in placental chorionic villous tissues of humans (and primates). Further studies then revealed some retroviral characteristics of these particles such as ultrastructural features and RT activity (Lyden et al. 1994). In addition, retroviral envelopes were detected in placenta sections by immunohistochemical techniques in human (Venables et al. 1995) and in baboon (Langat et al. 1999). Retrovirus-like particles associated with reverse transcriptase (RT) activity have been described by several groups in cell cultures from patients with multiple sclerosis (MS) (Perron et al. 1989; Haahr et al. 1994). Infectious properties of these particles are at least not trivial to ascertain if not doubtful. However, using PCR techniques, a reconstructed retroviral genome was defined as Multiple Sclerosis associated RetroVirus (MSRV) (Perron et al. 1997; Komurian-Pradel et al. 1999). MSRV is closely related to the HERV-W (Human Endogenous RetroVirus) family and particularly the ERVWE1/Syncytin-1 locus (Blond et al. 1999). Though, no full length replication competent virus has been experimentally isolated (Voisset et al. 1999). Nevertheless, it has been demonstrated that MSRV particles cause
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
81
T lymphocyte-dependent death with brain haemorrhage in humanized SCID mice (Firouzi et al. 2003). MSRV envelope protein has been proposed to exert various immune properties, e.g. as triggering a superantigen effect (Perron et al. 2001) and activating innate immunity (Rolland et al. 2006). Note that HERV-H related elements were also associated with particles observed in MS (Christensen et al. 1998), dually found with viruses from the herpes group, most likely Epstein Barr Virus (EBV). Cancers of the male reproductive system appears to be a favourable context for virus-like particles detection. Thus, HTDV (Human Teratocarcinoma-Derived Virus) is only expressed in the male germ line tumor context (Boller et al. 1993), and similar particles were observed in testicular germ cell tumors (or seminomas) (Herbst et al. 1999). In both situations, HERV-K transcripts could be associated with the particles. By electron microscopy and immunogold staining, HERV-K like particles were also visualized in the plasma of individuals with lymphoma, but these particles seem to be defective, as surface spikes and free mature virus particles were never observed (Contreras-Galindo et al. 2008). In all the situations exemplified above, although nucleic acid material could be associated with the particles, it remains unclear whether such particles could result from expression of a single retroviral loci, a trans complementation process or even more complex phenomenon involving genetic material recombination. As a corollary, infectivity of these particles has not been demonstrated. A clearer view was expected from the publication of several mammalian genomes, including human (Lander et al. 2001) and mouse (Waterston et al. 2002) genomes. Genomes of mammalian species harbor a large amount of retrovirus-like sequences. These endogenous retroviruses (ERVs) are remnants of ancient retroviral infections that initially occurred in the host germline. Throughout evolutionary time, these initially stably-integrated sequences have derived into gene families by retrotransposition events, and have accumulated genetic defects as a consequence of the host domestication. This general drift basically resulted in gene silencing. Generally, the retro-elements are free-Env and are not able to dissemination between cells. Intriguingly, the human genome but also the mouse genome contains a huge amount of endogenous retroviruses, reaching 8.5 and 9% of these genomes sizes, respectively. Deciphering the human genome showed that the HERV-K family contains tens of almost complete but mutated proviruses that allow the expression of viral proteins which appears able to form retroviral particles. However, no complete proviruses able to produce replication competent and infectious viral particles have been detected. The HERV-K113 locus though to be the more recent element of the family and that contains intact ORFs for all the viral proteins does not produce any particles (Lee and Bieniasz 2007). Trans-complementation between different HERV-K(HML2) proviruses could theoretically produce infectious particles, although not demonstrated to date. Interestingly, the infectious potential of HERV-K particles was artificially restored by generating a consensus HERV-K (HML-2) provirus named Phoenix supposed to be the HERV-K family progenitor (Dewannieux et al. 2006); this consensus contained at least 20 amino acid changes
82
P. Pérot et al.
on the overall sequence as compared to individual proximal HERV-K loci. By electronic microscopy, this resurrected HERV-K forms viral particles in transfected cells. The budding of its particles is similar to γ-, δ-retroviruses or lentiviruses with no particles preassembling into the cytoplasm. As cited earlier, MSRV is closely related to the HERV-W family including the Syncytin-1 encoding ERVWE1 locus which is the only W-locus bearing a fulllength envelope. The sequencing of ERVWE1 envelope confirmed that the MSRV envelope was not encoded by the ERVWE1 locus (Mallet et al. 2004). It was thus proposed that MSRV particles (if not derived from an as yet uncharacterised exogenous retrovirus) may result from transcomplementation of dispersed HERV-W copies simultaneously activated (Dolei 2005), what appears poorly probable as regards to the HERV-W elements identified in the human genome. However, it could not be formally excluded that MSRV/HERV-W genome (associated with particles) may result from very complex recombination events involving several loci on distinct chromosomes (Laufer et al. 2009). Although complete genomes analyses did not clearly explained the mechanisms leading to the formation of endogenous retroviral particle, they uncovered the extreme plasticity of these retroviral elements. Koala retrovirus (KoRV) provides a unique opportunity to study the process of ongoing endogenisation as it still appears to be spreading through the koala population. Interestingly, infectious viral particles are produced by the endogenous form of KoRV and high levels of viraemia have been linked to neoplasia and immunosuppression (Tarlinton et al. 2008). It remains unclear how the host can react when exogenous and endogenous forms of a virus are coexisting within the genome and his environment. Studies on Koala might answer this question. Interplay between the primitive virus world and the evaluated eukaryotic one could be observed at the env level. Thus, infectious retroviruses appear to have burst from our far ancestors genome by transcomplementation of cellular retrotransposons with viral envelopes genes (Malik et al. 2000). Another type of capture exists between retroviruses of distant species, consisting in the swapping of envelopes observed for species in the same environment or linked by the food chain. For example, the RD114 virus comes from two genetic recombinations resulting in two env-captures. First, the SERV (simian endogenous retrovirus) env was captured by the PcRV (Papio cynocephalus retrovirus) leading to the BaEV (baboon endogenous retrovirus). Second, the acquisition of BaEV env by FcEV (felis catus endogenous retrovirus) led to the emergence of RD114 virus (Kim et al. 2004). Last, endogenous retroviruses as remnants of ancient retroviral infections that initially occurred in the host germline represent an intriguing heritage. More precisely, as a consequence of at least 30 distinct chapters of retroviral infection during the past 90 million years, the current human genome contains 18 coding envelope genes (de Parseval et al. 2003; Blaise et al. 2005) (Table 4.1). The most represented family is the HERV-K(HML2) family which contains six coding env genes lacking fusogenic activity. Three Env proteins belonging to HERV-W, HERV-FRD and HERV-P families, namely Syncytin-1, Syncytin-2 and EnvP(b) respectively, have fusogenic properties.
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
83
Table 4.1 Human viral heritage of envelopes open reading frames containing canonical retroviral motives Name
RNA expression
Protein expression
Fusogenity
References
envH1 envH2 envH3 envK1 envK2 envK3 envK4 envK5 envK6 envT envWa envFRDb envR envR(b) envF(c)2 envF(c)1 envV envP(b)
NEc NEc NEc NEc NEc NEc NEc NEc NEc NEc Placenta Placenta All tissue NEc NEc NEc Placenta All tissue
NEc NEc NEc NEc NEc NEc NEc NEc NEc NEc Placenta Placenta NDd NEc NEc NEc NDd NDd
NDd NDd NDd NDd NDd NDd NDd NDd NDd NDd Yese Yesf NDd NDd NDd NDd Noe Yese
Lindeskog et al. (1999) de Parseval et al. (2001) de Parseval et al. (2001) de Parseval et al. (2003) Barbulescu et al. (1999) Donner et al. (1999) Barbulescu et al. (1999) Turner et al. (2001) Turner et al. (2001) de Parseval et al. (2003) Blond et al. (1999) de Parseval et al. (2003) Cohen et al. (1985) de Parseval et al. (2003) de Parseval et al. (2003) de Parseval et al. (2003) Blaise et al. (2005) Blaise et al. (2005)
a Syncytin-1. b Syncytin-2. c No
expression. determined. e In vitro. f In vivo. d No
4.3 Syncytins and Cell–Cell Fusion In spite of ERVs have been thought to be a non-functional part of the genome for a while, the past 10 years identified open reading frames of envelope genes in human, mouse, rabbit and sheep genomes (Fig. 4.3a), and associated with transcription activity and fusogenic glycoproteins synthesis (Fig. 4.3b) likely involved in biological functions. This is the case for the two human envelopes genes ERVWE1/Syncytin-1 and ERVFRDE1/Syncytin-2, located on chromosome 7q21.2 and 6p24.1, respectively, as well as for the two Syncytins-related A and B in mice, both pairs associated with fusion steps occurring in placental development process. Recently the novel Syncytin-Ory1 was identified in rabbit given a new example of syncytin gene within a third order of mammals (Heidmann et al. 2009). The ovine species also provide a quite interesting model of endogenisation process since the exogenous and pathogenic Jaagsiekte Sheep Retrovirus (JSRV) coexists with at least 27 highly related endogenous counterparts (enJSRVs), accounting for envelope genes in the ovine genomic DNA with evidences for open reading frames (Arnaud et al. 2007). enJSRVs play a crucial role in the sheep placental morphogenesis and
84
P. Pérot et al.
Fig. 4.3 Structure, phylogeny and fusion capacities of Syncytins. a Envelopes structures of Syncytins and schematic representation of their cognate receptors. FP: fusion peptide; tm: transmembrane domain; cyt: cytoplasmic tail. Black dots indicate the predicted N-glycosylation sites. SDGGGX2DX2R, consensus motif conserved in type D retroviral interference group, is indicated in human Syncytin-1 and rabbit Syncytin-Ory1. b Demonstration of Syncytin-1 cell–cell fusion property. TELacZ cells (dark blue nucleus) expressing Syncytin-1 envelope glycoprotein
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
85
their envelope expression in the reproductive tract is mandatory for a successful pregnancy (Dunlap et al. 2006). The focus point in this chapter will be now to discuss if these endogenous viral proteins of the genome still remain fusogenic in the same way an exogenous retrovirus envelope glycoprotein is (including transcription strategy, maturation steps, receptor recognition and fusion process), and to address the question of the domestication by the host and adaptive response though integrative and evolutionary points of view.
4.3.1 Integration, Domestication Steps and Biological Functions of Endogenous Viral Glycoproteins 4.3.1.1 Integration Dating and Orthologues One starting point in the discussion about the endogenous envelopes found in genomes could be the estimation of the age of the proviruses. This can basically be done by two approaches, bringing additional informations. One way is to assess a phylogenic lineage by tracing the presence of a similar DNA sequence at the same genomic loci in the genome of different species, and to conclude by a unique hypothetical integration event into the germline of a common ancestor. Another way is to consider the divergence between the 5 and 3 LTR and assuming a molecular clock is acting randomly through the genome, to generate variations over time between two originally and identical provirus sequences (the 5 LTR and 3 LTR being identical at the time of integration in the host genome). The first striking point are the unshared properties of both families and integration times within the humans, mice, rabbits and sheeps endogenous envelopes (Fig. 4.3c). ERV-W elements have been identified in hominoidae (human, chimpanzees, gorillas, orangutans and gibbons) and Cercopithecidae (old world monkeys) (Kim et al. 1999; Voisset et al. 1999) indicating that what we call today the human ERV-W family, HERV-W, derived from an ancestral virus which entered the genome after the divergence between Catarrhini and Platyrrhini, i.e. less that 40 million years ago (MYA). The ERVWE1/Syncytin-1 locus results from a complete proviral retrotransposition event into the germ line of an ancestor before Hominoidae and Cercopithecidae divergence more than 19–25 MYA (Caceres and Thomas 2006; Bonnaud et al. 2005). A full length envelope ORF corresponding to functional envelope glycoprotein was preserved in Hominoidae but genetic drift led to truncated envelope genes in old world monkeys. In contrast, the FRD family containing the HERV-FRD envelope Syncytin-2 is found in all simians, from New
Fig. 4.3 (continued) co-cultured with indicator XC cells (light blue nucleus) expressing hASCT2 receptor generates multinucleated large syncytia (left part). TELacZ-Syncytin-1 cells co-cultured with XC cells lacking hASCT2 do not fuse (right part). c Phylogenetic tree depicting the conservation among species of the six envelope-open reading frames harbouring retroviral canonical motifs (branches of the tree are only illustrative). NWM: new world monkeys; OWN: new world monkeys
86
P. Pérot et al.
World monkeys to human, suggesting a divergence split at least 40 MYA (Blaise et al. 2003). Moreover, the human, mouse and rabbit genes are not orthologs since Syncytin-A and -B entered into the rodent lineage before speciation of Muridae about 20 MYA (Dupressoir et al. 2005), i.e. after the speciation of rodents and primates while the Syncytin-Ory1 integration took place before the divergence between Lepus and Oryctolagus/Sylvilagus around 12 MYA (Heidmann et al. 2009). The situation is also different in domesticated sheep (Ovis aries) and other species within the Caprinae subfamily, where the endogenous retroviruses JSRV (enJSRVs) start to invade the genome at least 5–7 MYA and are likely still colonizing it today as given evidences by the restricted presence of recent enJSRVs loci into the genome of only some breeds or even some animals of the same breed of domesticated sheep (Arnaud et al. 2007). In order to better understand which mechanisms may have led to a positive selection of organisms harboring embarked viral genes, many arguments in favor of a domestication scenario have been deployed, especially about the ERVWE1/Syncytin-1 locus. Indeed, the proposed evolutionary pathway occurring in Hominoidae is opposed to the genetic drift in Cercopithecus. A ∼4.3 kb region, comprising the HERV-W 5 LTR-gag-pol fraction, was deleted in Cercopithecus and was followed by a genetic drift of the Env/Syncytin-1 ORF (Bonnaud et al. 2005; Caceres and Thomas 2006). Remarkably, the Syncytin-1 ORF has been conserved in all Hominoidae, while gag and pol regions have accumulated numerous stop and frameshift mutations (Mallet et al. 2004), supporting the idea of a specific preservation. Meanwhile, the analysis of 155 individuals including Caucasians, Asians, Africans, Metis and Ashkenazi people has revealed a positional conservation of the Syncytin-1 locus and the preservation of the envelope ORF (Bonnaud et al. 2004; Mallet et al. 2004), while a close examination of 24 ERVWE1 provirus sequences has showed an unusually low polymorphism in the 5 LTR (1 base per 18 kb as compared to 1 base per 2 kb for coding sequences) (Mallet et al. 2004). An additional specificity of the ERVWE1 provirus is the MaLR-LTR trophoblast specific enhancer (TSE) located upstream the ERVWE1 provirus that is highly conserved with no polymorphism observed in the 48 human sequences analyzed. Although the envelope gene may be considered under selective pressure depending on the part of the gene we focus on, the striking feature for the ERVWE1 locus is a 12 bp deletion observed in the Syncytin-1 intracytoplasmic tail gene region and that constitutes a specific signature of this locus. This deletion is unique among all ERV-W copies in available human and chimpanzee genomes (Bonnaud et al. 2005), and is crucial for the envelope fusogenicity (Bonnaud et al. 2004; Cheynet et al. 2005). Interestingly, the comparison of the FRD/Syncytin-2 envelopes among simians has also revealed a limited number of mutations, and pseudotypes experiments demonstrated that only one mutation occurring in the transmembrane subunit of the protein can be responsible for the loss of infectivity (Blaise et al. 2004). Notably, the alignment of endogenous and exogenous JSRV envelopes reveals similar deletions in the cytoplasmic tail of enJSRVs env as compared to the exogenous one (Palmarini et al. 2001). Altogether, these elements may infer the hypothesis of a positive selection and domestication of retroviral envelopes.
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
87
4.3.1.2 Endogenous Retrovirus Envelopes Are Expressed in the Placenta and in the Testis Suggesting a Direct Involvement in Developmental Process The HERV-W family was molecularly characterized following the isolation of cDNA clones in the placenta that revealed viral sequences genes expression, especially with similarities to the avian retrovirus primer binding site (Blond et al. 1999). In 2000, protein truncation tests within this endogenous family revealed only one open reading frame (ORF) coding for a putative envelope gene associated with a functional U3 promoter (Voisset et al. 2000). One year later, Blond and Mi concomitantly associated an HERV-W envelope protein with fusion events in TE671 and BeWo cells, and the name Syncytin was proposed by Mi (Blond et al. 2000; Mi et al. 2000). Heidmann and colleagues then conducted a genome wide screening that identified a second envelope protein, belonging to the HERV-FRD family, and expressed exclusively in the human placenta. They named Syncytin-2 this putative new fusogenic Env-FRD protein (Blaise et al. 2003). A similar in silico approach was done in the murine genome, identifying the two coding envelopes genes present as unique copies and with a placenta specific expression: Syncytin-A and Syncytin-B (Dupressoir et al. 2005), and recently in the rabbit genome, identifying the Syncytin-Ory1 gene (Heidmann et al. 2009). If the situation is much more different in the ovine genome, where approximately 27 copies of endogenous betaretrovirus (enJSRVs) were detected, RT-PCR and in situ hybridization clearly indicate a conceptus (embryo/fetus and extra embryonic membranes) localization of enJSRVs env transcripts during gestation (Dunlap et al. 2006). Although human Syncytins were abundantly described in the placental tissues, initial works also mentioned a weaker but significant transcription in the testis without any protein evidence (Mi et al. 2000). Envelope-specific RT-PCR established expression in the human testis of both Syncytin-1 and Syncytin-2 (de Parseval et al. 2003), and a multiplex degenerated PCR screening for a consensus pol region has revealed a general expression of the HERV-W family in testis (Pichon et al. 2006) and epididymis (Crowell and Kiessling 2007). This is consistent with old studies that identified the epididymal epithelium as a principle reservoir for retrovirus expression in the mouse (Kiessling et al. 1989). This knowledge points out that endogenous envelopes expression are usually associated with developmental tissues, and so far raise the question of whether or not Syncytins play a direct role in the mammalian placentation.
4.3.1.3 Biological Function of ERVs Envelopes The keen interest in ERVs envelopes expressed in placentas is fed by in vivo or ex vivo demonstrations that directly link Syncytins with fusion events during placental development. Although the role of Syncytin-1 in human placentation awaits a definitive demonstration (e.g. infertility associated mutation), recent knock-out gene experiments in mice clearly achieved this goal in rodent model and demonstrated for the first
88
P. Pérot et al.
time the critical role of Syncytin-A in placenta morphogenesis. Using a homologous recombination strategy, Syncytin-A null mouse embryos exhibited growth retardation with an altered placenta labyrinth architecture and died in utero (Heidmann et al. 2009). This is consistent with previous in vitro works that used specific antibodies and antisense oligonucleotides to show a decrease in syncytia cell formation after Syncytin-A inhibition (Gong et al. 2007). In addition, the endogenous retroviruses of sheep, enJSRVs, play a fundamental role in sheep conceptus growth and trophectoderm differentiation via their envelope glycoproteins. Indeed, in vivo experiments using an enJSRV envelopes specific morpholino injection trigger the lost of pregnancy by day 20 after injection (Dunlap et al. 2006). These kind of in vivo experiments obviously cannot be performed in human. Yet, primary cultures of human villous cytotrophoblasts cells give a unique opportunity
Fig. 4.4 Involvement of Syncytins in placenta development. a Assays reporting the biological effect of Syncytins. b Ex vivo or in vivo specific inhibition of Syncytins expressions. From left to right: Syncytin-1-induced human primary trophoblasts fusion and differentiation results in syncytia formation ex vivo (a). Inhibition by specific antisense oligonucleotide largely reduces syncytia formation (b). Electron micrograph of Syncytin-A+/+ mouse placenta shows tight apposition of the syncytiotrophoblast I and II layers (ST-I; ST-II); stgc: sinusoidal trophoblast giant cells (a). Syncytin-A–/– null mouse embryo interhemal domains shows unfused trophoblast I cells (T-I) (b). Micrograph of the normal development of a sheep conceptus (a). Retarded growth of a sheep conceptus recovered after an envelope enJSRV morpholino antisense oligonucleotide (MAO-env) injection (b)
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
89
to study placenta cells as closely related as possible to tissue environment. Thus, by using specific antisense oligonucleotides and siRNA strategies, expression of Syncytin-1 mRNA and protein as well as the syncytium formation by cell fusion events were dramatically reduced (Frendo et al. 2003b; Vargas et al. 2009). In addition to that, Vargas and colleagues recently compared these results using the same targeting strategy against Syncytin-2, and interestingly showed that Syncytin-2 inhibition in primary cells culture also leads to a decrease in fusion index that is more important than for Syncytin-1 (Vargas et al. 2009). The conclusion is that Syncytin2 could also be a major determinant of trophoblast cell fusion, and in a coherent vision this underlines there should be more than one ERV envelopes proteins acting upon trophoblast cell fusion in human. Parallel procedures demonstrating the involvement of human, mouse and sheep Syncytins in placenta development are illustrated in Fig. 4.4. Note that early works identified ERV-3 (HERV-R) envelope as the first candidate for placental functions. The ERV-3 envelope protein is detected specifically in the multinucleated syncytiotrophoblast in vivo (Venables et al. 1995) and ERV-3 Env expression affects proliferation and differentiation of BeWo cells in vitro (Lin et al. 1999; Lin et al. 2000). However, the observation that approximately 1% of the Caucasian population has a mutation in ERV-3 env inducing a stop codon, and consequently resulting in a truncated envelope lacking both the fusion peptide and the immunosuppressive domain (de Parseval and Heidmann 1998) has drastically lowered the scientific efforts regarding involvement of ERV-3 in placental development. Indeed, a second hypothetical function of Syncytins is related to their putative immunosuppressive activity (see below) due to the presence of a putative immunosuppressive region conserved among murine, feline, and human retroviruses (Cianciolo et al. 1985). So far we saw that Syncytins are involved in developmental fusion process. In the next part of this chapter we aim to focus on the mechanistic comparison between exogenous and endogenous envelope glycoproteins at the synthesis and maturation steps.
4.3.2 Fusion Mechanism and Receptor Recognition 4.3.2.1 Maturation The different steps leading from a brand-new translational product in the cytosol to a functional membrane-anchored envelope glycoprotein has been discussed previously. Basically, endogenous envelopes still remain glycoproteins, engaged in the classical reticulum-golgi apparatus where post translational events occur, before to be address to the plasma membrane and to become functional. Thus, precursor synthesis and glycosylation, disulfide bonds, trimerization, peptide cleavage and the importance of the cytoplasmic tail will be illustrated here introducing specific Syncytins knowledge, in order to support our previous descriptions as well as to focus on endogenous envelope specific characteristics.
90
P. Pérot et al.
Precursor, Furin Cleavage and Glycosylation Studies in BeWo cells models have described in a fine way the maturation of Syncytin-1. Syncytin-1 is first synthesized as a 73-kDa precursor (gPr73) before to be cleaved at a conserved RNKR furin cleavage site into two subunits, SU (gp50) and TM (gp24) (Cheynet et al. 2006). Although polypeptides size of Syncytin-1 and -2 is the same (538 amino acids), sizes of SU and TM are different after processing (Chen et al. 2008). PNGase F digestion and tunicamicyn treatment predicted and confirmed seven N-glycosylation sites for the Syncytin-1. These results indicate that Syncytin-1 is a moderately glycosylated protein, with one glycosylation site in the TM subunit which is essential for correct envelope protein folding, and with highmannose N-glycans on the six glycosylation sites of the carboxy-terminal domain of the SU (Cheynet et al. 2006). Furin inhibition experiments conducted on Syncytin-2 have also established the furin to play a major role in the proteolytic cleavage of the HERV-FRD envelope proprotein (Chen et al. 2008), where the cleavage consensus sequence is also found. Interestingly, using knock down experiment, furin has been proposed to have a possible role in promoting trophoblast cell migration and invasion in human placenta (Zhou et al. 2009). Bioinformatics analyses and sequence alignments suggest that Syncytin-A and -B exhibit most features of membrane fusion proteins, including the conserved cleavage site RNKR, which separate the SU and TM subunits (Dupressoir et al. 2005; Peng et al. 2007). Finally, the same feature is observed for the Syncytin-Ory1 sequence that exhibits a RQKR site (Heidmann et al. 2009) and for the enJSRV sequences that harbour the cleavage furin motif site. Disulfide Bonds and Trimerization Considering the Syncytin-1 TM gp24 sequence, it appeared that a leucin zipper-like LX6 LX6 NX6 LX6 L and a CX6 CC motifs are present, suggesting that SU and TM may covalently link together and form homotrimers (Cheynet et al. 2006). Indeed, Syncytin-1 sequence contains a typical disulfure isomerase motif in the SU domain (CC). As previously mentioned for MLV, the first two cysteines of the CX6 CC motif can form a stable disulfide bond, leaving the third cystein free to form a disulfide bond with the CC motif (Fass et al. 1997). Mutational experiments using neutral substitution in the CX6 CC motif did not affect the protein precursor expression level, but impaired syncytia formation, suggesting that disulfide bonds likely contribute to the correct folding of the envelope. In accordance to that, Chen and colleagues demonstrated that the disulfide bridge-forming CX7 C motif of the Syncytin-2 was essential for the fusogenic activity (Chen et al. 2008). Cytoplasmic Tail and R Peptide The cytoplasmic tail region of numerous retroviral envelopes plays a critical role in the fusion triggering. In the retrovirus life cycle, the presence of an R peptide basically prevents the fusion to occur, notably because of the presence of the YXX
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
91
motif described above. During viral packaging, the R peptide is proteolytically cleaved and this event enables envelopes to cause membrane fusion, as described by mutagenesis experiments (Yang and Compans 1996). We illustrate how Syncytins used various strategies that diverge from envelopes of infectious retroviruses to adapt to their physiological functions. Surprisingly, sequences comparison of the Syncytin-1 locus with all other HERVW envelope elements revealed a 12-bp (corresponding to four LQMV amino acids) deletion in its cytoplasmic tail (Bonnaud et al. 2004). Moreover, insertion of these four amino acids into Syncytin-1 tail completely abolished the fusogenic potential (Bonnaud et al. 2004). This result argues that Syncytin-1 is constitutively fusion competent, as opposed to exogenous retroviruses envelopes, and is coherent with a domestication point of view since no viral protease open reading frames exist anymore in the human genome (Voisset et al. 2000) (Fig. 4.5). Furthermore, the role of the cytoplasmic domain of Syncytin-1 has been systematically investigated by producing a series of C-terminal truncated variants, leading to the conclusion that residues adjacent to the membrane domain are required for optimal fusion probably by forming a helical structure, while final C-terminal residues more likely act as a fusion inhibitor domain (Drewlo et al. 2006; Cheynet et al. 2005). Remarkably, a truncation mutant which shortens the cytoplasmic tail precisely at the site of the LQMV-deletion motif exhibits higher fusogenic properties than the wild-type protein (Cheynet et al. 2005). Even if no work on Syncytin-2 has been done in such a fine way to assess the fusogenic properties modulation its cytoplasmic tail, we
Fig. 4.5 Comparative evolution of Syncytins cytoplasmic tails: from viruses to genes. The first five amino acids correspond to the transmembrane domain. Experimentally determined (GaLV, MLV, exoJSRV) and putative (W Rep. and FRD Rep.) protease cleavage site (black line) and YXX signaling motif are indicated in lowercase. Comparison of the Syncytin-1 protein (Syn-1) with the HERV-W family consensus sequence obtained from Repbase (W Rep.) shows a four amino acids deletion (LQMV) in the domesticated fusogenic protein, overlapping the ancestral viral protease cleavage site. The underlined leucine indicates a C-terminal truncation mutant exhibiting hyperfusogenic activity and significant pseudotyping capacity. Comparison of the Syncytin-2 protein (Syn-2) with the Repbase FRD consensus sequence (FRD Rep.) shows a stop codon that shortens the Syncytin-2 cytoplasmic tail and no evidence of viral protease cleavage site. Alignment of enJSRV and exoJSRV shows the placenta-expressed enJSRV has accumulated mutations surrounding the protease cleavage site and lacks downstream tyrosine (Y) residue. Genebank accession numbers: MLV: M14702; GaLV: AF055060, Syncytin-1: GQ919057, Syncytin-2: HEU27240, enJSRV: enJS56A1 and exoJSRV: AF105220
92
P. Pérot et al.
identified a stop codon in the cyt of Syncytin-2, as opposed to the RepBase prototype, resulting in a shortening of the tail (Fig. 4.5). Moreover, the protease cleavage site appears absent as regard to the FRD family consensus genome. Unlike the endogenous retrovirus enJSRV, the exogenous JSRV is pathogenic for sheep and is responsible for a transmissible lung cancer in sheep via its envelope glycoprotein acting as a dominant oncoprotein (Palmarini et al. 1999). Studies on the cytoplasmic tail of JSRV envelope protein first focused on the VR3 region that was described as the least conserved region between exogenous and endogenous forms. The VR3 region includes the putative membrane-spanning domain as well as the cytoplasmic tail, and series of envelope chimeras revealed that mutations in a YXXM motif of the cytoplasmic tail of JSRV env were sufficient to inhibit its transforming abilities (Palmarini et al. 2001). Further mutational amino acid substitutions have proven the tyrosine residue to be essential for transformation of exogenous JSRV. It is noteworthy that the VR3 region of all exogenous stains of JSRV sequences exhibit this tyrosine residue whereas all the enJSRVs envelopes described so far lacked this motif critical for JSRV transformation (Fig. 4.5). However, despite differences in terms of motif and sequence, JSRV and enJSRVs envelopes use the same cellular receptor called HYAL2. Further systematic mutagenesis studies of the cytoplasmic tail of JSRV envelope TM protein have established four categories of mutants that allow the TM to be devised into subdomains with regard to the transformation efficiency. Among them, mutations in the YXXM motif have various effects including the generation of “supertransformers” while the last nine amino acids of the cytoplasmic tail appear not essentials for the envelope-induced transformation (Hull and Fan 2006). 4.3.2.2 Receptor Binding Consistent with the virology paradigms, the comprehensive search for endogenous retroviral envelope functions led to the identification of the associated receptor (or co-receptors) that allows fusion events to be complete. In 2000, state of the art was to consider three main virus receptor types, PiT-1 and PiT-2, two independent inorganic-phosphate symporters for GaLV and MLV viruses, respectively, and the RDR/Type D receptor, a neutral-amino acid transporter for the cat endogenous retrovirus RD114 and type D simian retroviruses. When we first attempted to check for the right one that could trigger cell fusion, evidence in favor of the RDR/Type D hASCT2 mammalian receptor involvement was revealed through receptor-blocked experiments in cell lines transfected with Syncytin-1 gene (formerly HERV-W Env in Blond’s report) (Blond et al. 2000). Two years later, cell–cell fusion and pseudotypes virion infection assays demonstrated that Syncytin-1 efficiently uses both hASCT2 and the related hASCT1 transporter as receptors, and could recognize the mouse mASCT2 and mASCT1 transporters lacking their N-glycosylation sites removed by mutagenesis (Lavillette et al. 2002). Very interestingly, Syncytin-Ory1 also uses the hASCT2 transporter as specific receptor (Heidmann et al. 2009). AntiRDR serum used for histochemical and flow cytometric biodistribution analyzes, further unveiled a broad expression pattern of RDR/hASCT2 in many normal tissues
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
93
including colon, testis, ovary, bone morrow and skeletal muscle (Green et al. 2004). To move forward into the comprehensive relationship between Syncytin-1 and its hASCT2 receptor, Cheynet and coworkers designed a truncated series of HERV-W SU subunits and performed cell fusion assays. This led to the identification of a minimal N-terminal 124 amino acids of the mature SU glycoprotein required as a receptor-binding domain (RBD) (Cheynet et al. 2006). Furthermore, within this domain, one especially conserved sub-domain among retroviruses belonging to the interference group, the SDGGGX2 DX2 R motif, was proved to be essential for Syncytin-1-hASCT2 interaction (Cheynet et al. 2006). The human Syncytin-2 receptor has been identified more recently. By using an old powerful approach based on a human/Chinese hamster radiation hybrid panel mapping, one candidate gene was identified to encode a putative transmembrane protein (Esnault et al. 2008). The major facilitator superfamily domain containing 2 (MFSD2) was showed to confer cell–cell fusogenicity in the presence of Syncytin-2 and infectivity to Syncytin-2 pseudotypes. MFSD2 belongs to a large family of presumptive carbohydrate transporters, and is highly conserved among mammalian genomes. Although RDR and MFSD2 are two different receptors, they belong to the ion channels and small molecules transporters category, exhibiting a classical hydrophobic profile composed of transmembrane helices. The situation is outstandingly different for the HYAL2 JSRV receptor. Interestingly, early experiments to localize the JSRV receptor gene region were also based on pseudotyping and radiation panel screening (Rai et al. 2000). After identifying a set of overlapping clones, genetic analyses confirmed that HYAL2 was the only protein that functions as a JSRV env receptor (Rai et al. 2001). If the name HYAL2 first suggests a mainly strong hyaluronidase function, studies indeed showed low but detectable hyaluronidase activity pH-dependant (Lepperdinger et al. 1998) Actually, amino acid sequence analyses established one transmembrane domain as well as an hydrophobic carboxyl terminus and upstream signal, indicating that HYAL2 is likely attached to the membrane by a glycosylphosphatidylinositol (GPI) anchor (Rai et al. 2001). As compared to other GPI-anchored proteins, HYAL2 can potentially be involved in signal transduction and mitogenic responses, strongly suggesting a role of HYAL2 in JSRV Env-mediated oncogenesis (Rai et al. 2001). No receptors for mice Syncytin-A and -B have been identified to date. However, fusogenic experiments showed that each Syncytin-A and Syncytin-B expression vectors triggered fusion in two different cell lines, respectively. This result argues in favor of a divergent receptor usage for these two envelopes proteins (Dupressoir et al. 2005). 4.3.2.3 Incorporation in Particles Although one may speculate that domesticated endogenous envelopes are not supposed to spread using viral particle carrier, in an initial attempt to test the fusion capacity of HERV-W Env/Syncytin-1, we sought to generate retroviral pseudotypes in which MLV core particles were coated with HERV-W Env glycoproteins.
94
P. Pérot et al.
The absence of infectivity of HERV-W Env pseudotypes was due to the inability of HERV-W Env glycoproteins to be incorporated on MLV particles (Blond et al. 2000). Conversely, pseudotyping of HIV-1 virions with the HERV-W envelope resulted in infectious viruses, although with a poor efficiency (An et al. 2001). Such a difference in incorporation efficiency could be attributed to the lenght of the cytoplasmic tail of Syncytin-1, which is longer than the MLV one, but closer to the HIV-1 one. On line with this, shortening of the Syncytin-1 cytoplasmic tail (Fig. 4.5) significantly enhances pseudotyping of HIV-1 viral cores (Lavillette et al. 2002).
4.3.3 Retroviral Envelopes Are Involved in the Placenta Development Syncytins or envelope glycoproteins, originated from ancient retroviral infections, are now fully integrated in mouse, rabbit, sheep and human genomes. They are naturally expressed in the placenta, following a biological protein maturation process and interacting with specific receptors. Based on these observations, the purpose is now to understand how they contribute to the placenta morphogenesis. As bona fide genes, their transcription is strongly regulated by different cellular mechanisms, they are under local and temporal expression control, and finally they cooperate to an integrative way in a biological system as a whole, involving numerous co actors. The next part aims to develop these points. 4.3.3.1 Envelope and Receptor Localization Throughout Mammalian Gestation In humans, the embryo implantation process and the placental development are driven by fetal cytotrophoblasts stem cells. Placental growth is characterized by proliferation and differentiation of the villous cytotrophoblast pool into a multinucleated syncytiotrophoblast layer upon fusion events. This polarized layer constitutes the main fetomaternal interface in direct contact with the maternal blood (Fig. 4.6a). The general attempt to finely localize Syncytin-1 protein within the
Fig. 4.6 (continued) detected either in CT or ST and MFSD2 receptor is detected in ST. b Immunohistochemical detection of Syncytin-1 protein (SC-Syn1) at the apical syncytiotrophoblast microvillus membrane of a 10 weeks gestation normal placenta (upper). Note that desmoplakin, a protein of the desmosomiale plaque involved in intercellular junctions, is absent from the syncytiotrophoblast fused tissue and lines the plasmatic membranes of the cytotrophoblasts (CT-d). The hASCT2 receptor is observed at the membrane of cytotrophoblastic cells (CT-hASCT2) underlying the syncytiotrophoblast (ST). c Transcriptional and epigenetic control of Syncytins and associate receptors during human gestation. Promoter regions are indicated as boxes and CpG schematized by circles. TSE, trophoblast specific enhancer, U3, LTR promoter, R, transcription initiation site. CpG methylation is determined by bisulfite sequencing PCR in cytotrophoblasts (CT) at different times of gestation. Each line represents an independent molecule. Methylated CpGs are schematized by black circles and unmethylated CpGs by white circles
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
95
Fig. 4.6 Local and temporal expression of Syncytins in human placenta. a Schema of human chorionic villi. Cytotrophoblastic cells (CT, in yellow) differentiate by fusion to generate the syncytiotrophoblast (ST, in green). In the anchoring villi the cytotrophoblast cells proliferate and invade the decidua. The extravillous cytotrophoblast cells (ECT) invade the uterin stroma and differentiate into multinucleated giant cells and invade also the lumen of uterine arteries (UA). M: mesenchyme. Sites of expression of Syncytins and receptors are symbolized in the lower box: Syncytin-1 protein is detected in ST and ECT, hASCT2 receptor is detected in CT and ECT, Syncytin-2 protein is
96
P. Pérot et al.
placenta led the authors to use homemade antibodies that resulted in variable conclusions. With full knowledge of that facts, we decided to focus on the hypothesis that Syncytin-1 is preferentially detected at the apical membrane of the syncytiotrophoblast (Frendo et al. 2003b; Muir et al. 2006) (Fig. 4.6b). Indeed, two convergent studies mentioned that Syncytin-1 is located on specific membrane areas, enriched in cholesterol and called detergent-resistant membrane (DRMs) or rafts (Cheynet et al. 2005; Strick et al. 2007). Moreover, in primary trophoblast cells, Syncytin-1 detection was likely to be associated with cell-to-cell contact zones (Vargas et al. 2009). According to several authors, the level of Syncytin-1 protein increases during early pregnancy but remarkably decreases in very late pregnancy (Smallwood et al. 2003; Muir et al. 2006). The hASCT2 receptor expression is restricted to the cytotrophoblast compartment (Hayward et al. 2007) (Frappart, Cheynet, Mallet, unpublished data), being largely absent in the syncytiotrophoblast (Fig. 4.6b) and no spacial or temporal changes in the hASCT2 expression has been associated with the proliferative status of cytotrophoblast cells (Hayward et al. 2007). On another hand, divergent observations mentioned Syncytin-2 expression to be either restricted to villous cytotrophoblast cells (Malassine et al. 2007) or in the syncytiotrophoblast (Chen et al. 2008). Compromising positions accordingly associated Syncytin-2 with cell-to-cell contact regions, likely at the interface between the cytotrophoblast and the syncytiotrophoblast (Malassine et al. 2007; Vargas et al. 2009). Remarkably, the level expression pattern of Syncytin-2 follows an inverse correlation compared to Syncytin-1, since a significant increase in Syncytin-2 mRNA and protein is monitoring through pregnancy time and primary trophoblast culture evolution (Chen et al. 2008; Vargas et al. 2009). Finally, the MFSD2 receptor expression is unambiguously reported at the level of the syncytiotrophoblast (Esnault et al. 2008). In addition to the villous phenotype, the cytotrophoblastic cells of the anchoring villi can proliferate and invade the endometrium to be in contact with the spiral arterioles of the mother uterus. In these cells that do not fuse, both Syncytin-1 and his receptor hASCT2 have been detected (Malassine et al. 2005; Muir et al. 2006), suggesting that the trophoblastic cell fusion is indeed a complex multifactorial process (Malassine et al. 2005). The mouse placenta is composed of spongiotrophoblasts, giant cells and a socalled labyrinth zone. In this placenta labyrinth, two layers of multinucleated syncytiotrophoblast cells, resulting from cell–cell fusion, function as the major transport surface for nutrient and gas exchange between the maternal and fetal circulation. Early in situ hybridizations showed that Syncytin-A and Syncytin-B are expressed at the level of syncytiotrophoblats, all over the labyrinth zone (Dupressoir et al. 2005). In coherence with that, latter studies more precisely indicated a clear expression of Syncytin-A in the syncytiotrophoblast while expression in trophoblast stem cells and in trophoblast giant cells could hardly be detected (Gong et al. 2007). The rabbit placenta can be divided into the maternal decidua (the uterus modifications after implantation) and the placental lobe, in which a labyrinthine structure results from the fetal invading process. At this interface, a junctional zone presumably formed by cellular cytotrophoblasts takes place and defines a broad syncytial front. In situ hybridizations on paraffin sections of rabbit placenta have shown Syncytin-Ory1 expression to be restricted at the junctional zone and limited
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
97
to the trophoblast cells surrounding the invading fetal vessels (Heidmann et al. 2009). The authors also suggest that the labeling is compatible with an expression of Syncytin-Ory1 in the cytotrophoblast just before fusion takes place and so is consistent with a role for Syncytin-Ory1 in the formation of the syncytiotrophoblast (Heidmann et al. 2009). In the sheep conceptus, trophoblast binucleated cells are in many respects analogous to the trophoblast giant cells of the human syncytiotrophoblast (Hoffman and Wooding 1993). They first appear at day 14 post-coitum and progressively form the outer layer of the fetal placental cotyledon giving rise to the syncytial plaques (Wooding 1984; Palmarini et al. 2001). The plaques then may cover the surface of the endometrial carunucles and aid in development of placentomes that are required for hematrophic nutrition exchanges from the maternal uterus. RTPCR analyses have showed that endogenous JSRV envelopes and HYLA2 were expressed in the trophoblast giant binucleated cells and in the multinucleated syncytial plaques (Dunlap et al. 2006). If both endogenous JSRV envelopes and HYAL2 are detected in placentome throughout gestation, HYAL2 expression was not detected in endometrium (Dunlap et al. 2006). Endogenous JSRV envelopes were first detected in the day 12 conceptus, whereas HYAL2 first detection appeared at day 16, in a coherent way with the initial differentiation start of the binucleated cells at day 14. 4.3.3.2 Splicing Strategy, Transcription Factors and Epigenetic Control For a brief reminder, Syncytins’s chromosomal localizations are 7q21.2 (ERVWE1/Syncytin-1) (Blond et al. 1999) and 6p24.1 (ERVFRDE1/Syncytin-2) (Blaise et al. 2003) in human, and 5qG2 (Syncytin-A) (Dupressoir et al. 2005) and 14qD1 (Syncytin-B) in mouse (Dupressoir et al. 2005). To date the rabbit genome is not available in a definitive assembly to check the Syncytin-Ory1 localization and several integrations sites for enJSRV exist in the sheep genome. Putative splice donor and acceptor sites have been identified for all of them (Blond et al. 1999), although only specific transcripts have been detected depending on the biological context. ERVWE1 produces three major singly-spliced transcripts in placenta (Blond et al. 1999). The first one, 7.4-kb long, containing the gag and pro/pol pseudogenes and env gene, is found both in testis and placenta (Mi et al. 2000), while the 3.1 kb long, strictly including the open reading frame for the envelope protein Syncytin-1, is exclusively detected in the placenta. Early northern blot experiments also detected a 1.3-kb largely-spliced transcript in placenta (Blond et al. 1999), indeed containing the cytoplasmic tail of Syncytin-1. These observations are similar to lentivirus or oncovirus transcription patterns such as human immunodeficiency virus (HIV), the mouse mammary tumor virus (MMTV) or the human T-cell leukemia virus (HTLV), in which several genomic and subgenomic transcripts derive from a single locus by alternative splice variations. Like any other classical retrovirus, endogenous retroviruses may display all the signals required for the transcription initiation and regulation within their long terminal repeat sequences (LTRs) (subdivided in three regions named U3, R and U5). Typically, the U3 region of the 5 LTR possesses a promoter activity. Fine studies
98
P. Pérot et al.
have well described how Syncytin-1 is under upstream regions control. The core promoter domain within the U3 region contains CAAT box and TATA box located upstream of the CAP site marking the beginning of the R region (Prudhomme et al. 2004). Mutant analyses have confirmed the functional role of these boxes. Moreover, the 5 end of the U3 region harbors multiple binding sites contributing to overall promoter efficiency including GATA, Sp-1, AP-2, Oct-1, and PPAR-γ/RXR. Although Sp-1 and Ap-2 binding sites remain putative, they have been found to be essential for LTR activity (Prudhomme et al. 2004). It is noteworthy that Syncytin-1 regulation elements not only include the 5 LTR but also a so-called upstream regulatory element (URE), a cellular 436 bp sequence immediately upstream the Syncytin1 proviral integration site, that define together with the 5 LTR a bipartite control element (Prudhomme et al. 2004). This URE is composed of two main domains: (i) a distal regulatory region, including the previously putative binding sites found in the promoter core as well as binding sites for the NF-κB and AP-1 important for the stimulation by TNFα, IFNγ, IL-1β, IL-6, and the inhibition by IFNβ (Mameli et al. 2007) (ii) a MaLR retrotransposon with binding sites for glucocorticoid and progesterone receptors (Bonnaud et al. 2005), and including a trophoblast specific enhancer (TSE) with putative sequences for ubiquitous Ap-2, Sp-1 and placenta-specific GCMa binding sites (Prudhomme et al. 2004). Glial cell missing (GCM) is a transcription factor family that has gradually attracted the attention of placenta researches. Originally isolated from a Drosophila melanogaster mutant line, two GCM homologues (GCMa and GCMb) have then been reported in mice, rats and humans (Keryer et al. 1998). GCMa is characterized by a zinc-coordinating DNA binding domain of β-sheets that recognizes an octomeric GCM binding motif 5 -ATGCGGGT-3 (Cohen et al. 2003). GCMa is primarily expressed in placenta in humans and highly expressed in the labyrinthine trophoblast cells in mice (Basyuk et al. 1999). Two binding sites by which GCMa can specifically transactivate Syncytin-1 have been described (Yu et al. 2002). Moreover, GCMa regulation has been linked to the cyclic AMP (cAMP) and protein kinase A signaling pathways (Chang et al. 2005; Knerr et al. 2005). In agreement with these observations, the Syncytin-1 5 LTR core promoter is cAMP-inducible (Prudhomme et al. 2004). Interestingly, a recent microarray approach that aimed to identify GCMa target genes reported Syncytin-A to be downregulated in murine GCMa-deficient placenta (Schubert et al. 2008), and siRNA GCMa inhibition in BeWo cells led to a decrease in syncytialization upon fusion events (Baczyk et al. 2009). Altogether, these data argue that GCMa acts as a major regulator in the humans and mice Syncytins expression as well as in placenta maintenance and development. To conclude with this regulation mechanisms overview, the imprinting hypothesis and the influence of the methylation level is briefly discussed. Genomic imprinting in mammals is though to be a rescue mechanism that maintains balanced growth and development through monoallelic expression of genes in placenta and embryo. Although very little is known about the regulation of most imprinted genes, in 2003 the observations that Syncytin-1 maps very closely to two neighboring maternally imprinted retroelements, SGCE and PEG10, and according to
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
99
their temporally coordinated regulation, the hypothesis that Syncytin-1 could be paternally expressed emerged (Smallwood et al. 2003). On the other hand, methylation pattern studies of the Syncytin-1 5 LTR revealed an inverse correlation between CpG methylation and locus expression indicating that demethylation of the promoter is a prerequisite for the Syncytin-1 expression in trophoblasts cells (Matouskova et al. 2006). In an attempt to generalize this epigenetic characterization, Gimenez and colleagues recently compared different HERV promoters methylation profiles and showed that Syncytin-1 and Syncytin-2 5 LTR are widely hypomethylated in cytotrophoblasts during pregnancy, although in a distinct and pregnancy-stage-dependant manner (Gimenez et al. 2009) (Fig. 4.6c). For instance, the Syncytin-2 locus remains unmethylated throughout pregnancy, whereas methylation of the Syncytin-1 locus appears increased in the last trimester. Thus, the selective and temporal unmethylation of the Syncytin-1 locus in placenta during the first trimester may allow Syncytin-1-mediated cell differentiation and fusion, while, in contrast, increased methylation at term may limit Syncytin-1 production and consequent cell fusion or putative anti-apoptotic protection (Knerr et al. 2007) in accordance with cytotrophoblast limited fusion and higher apoptosis rate. 4.3.3.3 Additional Factors The complete inhibition of cytotrophoblast fusion can’t be reached by blocking the Syncytin-1 protein (Mi et al. 2000; Frendo et al. 2003b), and as we mentioned above, extravillous cells that express Syncytin-1 and hASCT2 do not necessarily fuse. This indicates so far that Syncytin-1 plays a major role in fusion, but also strongly suggests that other elements may contribute to this event. We proposed here to review some of them, with special interest focuses on plasma membrane dynamics, cell– cell communication and immunity as a link between the various events leading to a coordinated tissue function. Human placenta only will be discussed. Before many retroviral receptors were known, observations emerged that an infected cell could not be superinfected by the same retrovirus, and sometimes even not being superinfected by a different one (Kim et al. 2004). Thus, interference groups were defined as set of retroviruses that cannot infect a cell at the same time. Indeed, all the members of an interference group utilize the same receptor for cell entry, and when a cell is infected once, cellular receptors are blocked by the envelope proteins of the first retrovirus and no longer available for infection of the second one. It is obvious that the balance between the envelope and its local receptor availabilities needs to be taken under consideration before predicting anything, as illustrated by FLV (Sommerfelt and Weiss 1990). Yet, given that Syncytin-1 is an endogenous envelope protein, we can expect the same rules to be valid. Indeed, together with spleen necrosis virus (Ponferrada et al. 2003), Syncytin-1 belongs to the RDR/Type D mammalian receptors interference group (Blond et al. 2000). Consequently, the hASCT2 receptors would no longer be available for type D retroviruses infection as soon as Syncytin-1 is co-locally expressed in massive doses. This is the core hypothesis to discuss different models that integrate local and temporal expression data (for a complete review, see (Potgens et al. 2004)). Overall it is a clear indication that the
100
P. Pérot et al.
cytotrophoblast fusion is guided through local availabilities of both envelopes and their corresponding receptors. We previously mentioned how precisely Syncytin-1 seems to be associated with a restricted part of the polarized cell membrane, i.e. mainly detected at the apical part of the syncytiotrophoblast with strong suspicions that turn around a membrane subdomain enriched in cholesterol and defined as DRMs/rafts. Plasma membranes are lipid bilayers with an asymmetrical distribution of phospholipids between the inner and the outer leaflet. They are fluidic structures that permanently reorganized themselves, through dynamic lateral diffusivity, rotations, and flippase-mediated flip-flop switches. Thus, membrane-anchored proteins are part of this flow. This would allow them to reach and stay at a plasma membrane sub-localization, as well as to contribute to their functional transmembrane stable insertion. Flippases are ATP-dependant translocase enzymes that assure the asymmetrical distribution of phospholipids between the inner and the outer leaflet of the membrane. Remarkably, the loss of asymmetry triggered by a redistribution of phosphatidylserine from the inner to the outer leaflet has been described to be a prerequisite for fusion in skeletal muscle (van den Eijnde et al. 2001) and in placenta derived BeWo cells (Lyden et al. 1993). In agreement, the loss of membrane asymmetry in a cell has been associated with early stages of the apoptosis cascade. Caspase 8 is a caspase initiator involved in early apoptosis that inactivates the flippases, resulting into the asymmetry loss. When antisense and peptide inhibition strategies against caspase 8 are used, fusion of trophoblast cells is inhibited (Black et al. 2004). In addition, cholesterol-enriched domains are associated with weak fluidic properties. Although the direct relation between membrane dynamics and fusogenic proteins still remains to be elucidated, we can speculate that the positioning of Syncytin-1 that leads to the fusion event, may occurs within a well-controlled membrane subdomain with physical properties that form a stringent envelope environment compatible with receptor binding and subsequent events. The different points we just mentioned mainly focused on the syncytiotrophoblast. Syncytin-1 membrane localization, phospholipids dynamics and early stages of apoptosis were presented as part of the multinucleated cell life. The problem here is that the syncytiotrophoblast is presumed to have a very low transcriptional activity and likely depends on the input of RNA to avoid necrosis, as indicated by different in vitro experiments (Bernirschke and Kaufmann 2000; Huppertz et al. 1999). This suggests the importance of an effective cell– cell communication and material supply systems. Gap junctions are transmembrane channels composed of connexin that provide a diffusion system for small proteins such as cAMP, IP3 or Ca2+ . In primary trophoblast culture, the inhibition of connexin 43 (Cx43) resulted in a fusion inhibition (Frendo et al. 2003b) and a decrease of Syncytin-1 mRNA expression (Frendo et al. 2003a). Moreover, the hASCT2 receptor is a Na+ -dependent amino acid transporter that can carry amino acids such as L-glutamine, L-alanine, L-leucine and L glycine, through the membrane. The clear localization of the receptor within the membrane of cytotrophoblastic cells underlying the syncytiotrophoblast (Hayward et al. 2007) (Frappart, Cheynet, Mallet, unpublished data), suggests that hASCT2 could efficiently change
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
101
the amino acid balance between the cytotrophoblast and the syncytiotrophoblast. Altogether, material flow that involves small amino acids and molecules, along with numerous electric charges balance changes (Ca2+ through gap junctions, anionic phosphatidylserine via flip flop events) appear to play a direct or indirect role in the fusion regulation. In such a model, the syncytiotrophoblast is indeed described as one element, part of a global turnover system which includes a generative pool of cytotrophoblast cells that can feed the multinuclear layer upon fusion events and thus maintain placental growth. Athough Syncytin-1 and Syncytin-2 and their receptors play major roles, they are probably not the only proteins involved in this cooperative mechanism. Given that the placenta is an extra-embryonic tissue, half paternal and half maternal genetically inherited, the past decades have gathered reproductive immunologists researches to solve the fetal allograft problem. The contact zone between mother uterus and fetus extravillous cells of spiral arterioles appears to be one of these predictive immunological conflict zones. In direct connection with our topic on retroviral fusogens, note that the simian retrovirus (SRV), that induces immunodeficiency, belongs to the same interference group that Syncytin-1, i.e. it binds to the ASCT2 amino acid receptor. The link between immune response and amino acid balance has been seriously explored, and interestingly the involvement of HERV in immune response has already been suggested (Espinosa and Villarreal 2000). We present here some points of discussion in such a way. During pregnancy, maternal tryptophan is required for the T lymphocytes activation and “immunosuppression by starvation” is the consequence of tryptophan depletion experiments (Mellor et al. 1999). Besides, a tryptophan-catabolizing enzyme, the indoleamine 2,3-dioxygenase (IDO), is particularly expressed in the syncytiotrophoblast. Thus, the lymphocyte regulation appears to be strongly mediated by the ability of the apical membrane to incorporate the tryptophan into the syncytiotrophoblast (Kudo and Boyd 2001). In other words, the tolerance towards the allograft is conditioned by the CD98/LAT1 tryptophan transporter and the resulting amino acid balance changes. Even if the Syncytin-1 and Ory-1 hASCT2 receptor only mediates the transport of small amino acids (and consequently probably not tryptophan), considerations about balance changes that could impact the immune system response are maybe not so far. Indeed, glutamine is a necessary substrate for the nucleotide synthesis of lymphocyte cells. In peripheral blood an optimal glutamine level is required to influence the switch within the sub-populations of T lymphocytes, Th1 and Th2, through a predominantly Th1 host response (Chang et al. 1999). In sepsis mice, glutamine supplementation changes the production of IL-6, IL-4 and IFN-γ, and thus may reverse or re-equilibrate the Th1/Th2 balance response during sepsis (Yeh et al. 2005). Although it is still in debate, the Th1/Th2 switch appears to play a critical role during pregnancy, especially through a Th1 bias in recurrent pregnancy loss (Chaouat 2007). This allows us to speculate a direct or indirect involvement of amino acids balance changes as a consequence of the envelope protein fixation on its receptor. Yet, another answer to the allograft tolerance during pregnancy emerged after it was reported that MSRV particles (related to the HERV-W family) induce T lymphocyte response (Perron et al. 2001), the analysis of the putative
102
P. Pérot et al.
immunosuppressive domain in the TM subunit of different Syncytins has revealed a immunosuppressive activity for Syncytins-2 and -B but surprisingly not for the Syncytins-1 and -A (Mangeney et al. 2007) We can conclude this discussion by briefly reporting different abnormalities that occur in pathological contexts in relation with some elements just mentioned above. Pre-eclampsia (PE) and HELLP syndrome (hemolysis, elevated liver enzymes and low platelets) are disorders associated with abnormal placentation, including defects in syncytiotrophoblast formation. Numerous studies have associated PE and HELLP with Syncytin-1 and Syncytin-2 significant reduction (Lee et al. 2001; Knerr et al. 2002; Chen and Olson 2005; Strick et al. 2007; Chen et al. 2008). Interestingly, a redistribution of the Syncytin-1 within the syncytiotrophoblast polarized cell layer was observed for patients with PE (Lee et al. 2001). Moreover, PE is associated with a predominant Th1 immunity type (Jianjun et al. 2010), that could hypothetically make the bridge with Syncytins defects, unbalanced amino acids flux and immunity. Hypoxia is overall important in the differentiation and fusion steps, since these conditions reduce the Syncytin-1 transcriptional level and inhibit cytotrophoblast fusion, whereas hASCT2 mRNA level remains unchanged (Kudo et al. 2003; Knerr et al. 2003; Chen and Olson 2005). Finally, higher apoptotic rates are observed in cultured cytotrophoblast cells from PE and HELLP (Strick et al. 2007).
4.3.4 Syncytin-1 Expression Outside of Its Privileged Tissue As illustrated above, the multi levels-control of expression of Syncytin-1 suggests that for all Syncytins, expression is tightly regulated to be constrained to placenta. Among Syncytins, only the expression of Syncytin-1 has been so far described outside from its privileged tissue. Syncytin-1 is expressed in astrocytes, glial cells and activated macrophages in brain regions affected by multiple sclerosis (MS). Syncytin-1 expression in astrocytes mediates neuroimmune activation and death of oligodendrocytes by inducing the release of cytotoxic redox reactants (Antony et al. 2004). In astrocytes, Syncytin1 induces the expression of OASIS (old astrocytes specifically induced substance), an endoplasmic reticulum stress sensor, which in turn increases the expression of inducible NO synthetase and concurrent suppression of cognate hASCT1 receptor, resulting in diminished myelin protein production (Antony et al. 2007). What mechanisms reactivate Syncytin-1 in the brain in MS is still not clear. It could be the result of viral infection of the brain, such as herpes simplex virus, which has previously been shown to transactivate Syncytin-1 expression, or cytokine deregulation (Perron et al. 1993). Indeed it has been shown in astrocyte cultures that MS detrimental cytokines, IFN-γ and TNF-α are able to induce Syncytin-1 expression through NF-κB activation, while MS protective IFN-β inhibits its expression (Mameli et al. 2007). In addition Syncytin-1 induction by exogenous TNF-α into the corpus callosum, a region of the brain frequently exhibiting demyelination in MS, leads to neuroinflammation, reduction of myelin proteins level and neurobehavioural deficits in Syncytin-1-transgenic mice, as observed in MS (Antony et al. 2007).
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
103
Interestingly as a parallel between MS and cancers, NO production in tumor vessels correlates with an increase of the over-all survival as well as the decrease of metastatic potency in experimental systems (Mortensen et al. 2004). On line with this, the level of Syncytin-1 expression represented a positive prognostic indicator for recurrence-free survival of breast cancer patients (Larsson et al. 2007). Conversely, increased Syncytin-1 expression was associated with decreased overall survival in rectal but not in colonic cancer patients (Larsen et al. 2009). The situation appears unclear in endometrial carcinoma (EnCa) where Syncytin-1 expression increase in normal endometrium of patients may possibly influence the development of endometriosis (Oppelt et al. 2009). Thus, the prognostic impact of Syncytin-1 expression appears to vary with the tumor type potentially, due to different functions associated with different pathways of reactivation. In breast cancers, Syncytin-1 expression was observed for about one-third of patients, and additionally, neighbouring endothelial cells were shown to express hASCT2 receptor (Bjerregaard et al. 2006). In vitro studies confirmed the involvement of Syncytin-1 in the fusion process between breast cancer cell lines and endothelial cells (Bjerregaard et al. 2006). Syncytin-1 associated cell–cell fusion was also identified in EnCa tumors in vivo, but interestingly, in vitro studies showed the implication of Syncytin-1 in both the fusion and the proliferation of EnCa cells (Strick et al. 2007). Syncytin-1 up regulation via the cAMP pathway leads to cell–cell fusion while induction by steroid hormones (estradiol) leads to proliferation. This molecular switch is apparently controlled by TGF-β1 and TGF-β3 which are induced by steroid hormones and may override Syncytin-1 mediated cell–cell fusions (Strick et al. 2007).
4.4 Conclusion Our life begins with fusion as a successful pregnancy in mammals appears to depend on Syncytin(s), retroviral members of a family of single-pass transmembrane proteins which contribute at least to cell–cell fusion necessary for placental syncytiotrophoblast morphogenesis. As we have seen, works of the last 10 years showed that Syncytins may represent extreme examples of foreign genes domestication as different elements were preserved during (parallel) evolution to assume (partly) similar roles in various species such as rodents, lagomorphs, sheep, and primates including human. These domesticated elements represent apparently a tremendous but very minor part of endogenous retroviruses (ERV) which colonized mammalian genomes. Can we consider that gift as a pay-back of retroviruses, as they emerged from our ancestors genome by transcomplementation of retrotransposons with viral envelopes (Xiong and Eickbush 1990; Malik et al. 2000), or is there a “price to pay”? From studies in animal and human cancers, there is little doubt that tumor hybrids/fused cells are generated in vivo and that at least in animals they can be a source of metastasis (Pawelek and Chakraborty 2008). Interestingly, fusion between cancer and normal cells can lead to restoration of the apoptosis cascade or to cell differentiation, inducing a reduced tumorigenicity. However cancerous cells fusion may also lead on the contrary to a more aggressive phenotype, and, if fusion occurs
104
P. Pérot et al.
with vascular endothelial cells, to metastasis. Consequences of Syncytin-1 expression in various cancers may reflect such a diversity. Although retrovirology is a 100 years old discipline, it is less than 5 years ago that a role for viruses in cell fusion and its importance in the overall evolution of cancer was proposed (Duelli and Lazebnik 2007). Altogether, these findings strongly support a comparative analysis of the modalities and consequences of infectious retroviruses and endogenous retroviral Syncytins expression on the cell–cell fusion processes. HERV expression/activation seems to be a common feature in cancers, a phenomenon that has been linked to deregulation of methylation (Schulz et al. 2006). Global hypomethylation of transposable elements may be a prerequisite of chromosomal instability. Similarly, viral induced fusion might result in the chromosomal instability observed in cancer cells (Duelli and Lazebnik 2007). Hypomethylation of the ERVWE1/Syncytin-1 in placenta and in seminoma (Gimenez et al. 2010) as compared to HERV-W family hypermethylation in placenta and hypomethylation in seminoma may reflect both situations (Gimenez et al. 2009). At the protein level, clarifying the interactions between Syncytin-1 and TGF-β may contribute to elucidate the regulation of cell– cell fusions occurring in development and in other syncytial cell tumors. Thus, the increase of cholesterol efflux from cellular membrane by TGF-β could modify membrane location and function(s) of Syncytin-1. Overall, fusion/differentiation, proliferation and suggested anti-apoptotic capacities of Syncytin-1 delineate the portrait of an oncogene. Acknowledgments We are grateful to Sarah Prudhomme, Frederick Arnaud and Juliette Gimenez for their critical comments which contributed significantly to the improvement of this chapter. We thank Danièle Evain-Brion, Thierry Heidmann, Thomas E. Spencer, and François-Loïc Cosset for providing pictures and photographs.
References Aguilar HC, Anderson WF, Cannon PM (2003) Cytoplasmic tail of Moloney murine leukemia virus envelope protein influences the conformation of the extracellular domain: implications for mechanism of action of the R Peptide. J Virol 77:1281–1291 An DS, Xie Y, Chen IS (2001) Envelope gene of the human endogenous retrovirus HERV-W encodes a functional retrovirus envelope. J Virol 75:3488–3489 Antony JM, Ellestad KK, Hammond R et al (2007) The human endogenous retrovirus envelope glycoprotein, syncytin-1, regulates neuroinflammation and its receptor expression in multiple sclerosis: a role for endoplasmic reticulum chaperones in astrocytes. J Immunol 179:1210–1224 Antony JM, van Marle G, Opii W et al (2004) Human endogenous retrovirus glycoproteinmediated induction of redox reactants causes oligodendrocyte death and demyelination. Nat Neurosci 7:1088–1095 Arnaud F, Caporale M, Varela M et al (2007) A paradigm for virus-host coevolution: sequential counter-adaptations between endogenous and exogenous retroviruses. PLoS Pathog 3:e170 Baczyk D, Drewlo S, Proctor L et al (2009) Glial cell missing-1 transcription factor is required for the differentiation of the human trophoblast. Cell Death Differ 16:719–727 Barbulescu M, Turner G, Seaman MI et al (1999) Many human endogenous retrovirus K (HERV-K) proviruses are unique to humans. Curr Biol 9:861–868 Barre-Sinoussi F, Chermann JC, Rey F et al (1983) Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune dzficiency syndrome (AIDS). Science 220:868–871
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
105
Basyuk E, Cross JC, Corbin J et al (1999) Murine Gcm1 gene is expressed in a subset of placental trophoblast cells. Dev Dyn 214:303–311 Berlioz-Torrent C, Shacklett BL, Erdtmann L et al (1999) Interactions of the cytoplasmic domains of human and simian retroviral transmembrane proteins with components of the clathrin adaptor complexes modulate intracellular and cell surface expression of envelope glycoproteins. J Virol 73:1350–1361 Bernirschke K, Kaufmann P (2000) Pathology of the human placenta. Springer, New York, NY, pp 22–70 Bernstein HB, Tucker SP, Hunter E et al (1994) Human immunodeficiency virus type 1 envelope glycoprotein is modified by O-linked oligosaccharides. J Virol 68:463–468 Bertrand P, Cote M, Zheng YM et al (2008) Jaagsiekte sheep retrovirus utilizes a pH-dependent endocytosis pathway for entry. J Virol 82:2555–2559 Bhattacharya J, Peters PJ, Clapham PR (2004) Human immunodeficiency virus type 1 envelope glycoproteins that lack cytoplasmic domain cysteines: impact on association with membrane lipid rafts and incorporation onto budding virus particles. J Virol 78:5500–5506 Binley JM, Sanders RW, Clas B et al (2000) A recombinant human immunodeficiency virus type 1 envelope glycoprotein complex stabilized by an intermolecular disulfide bond between the gp120 and gp41 subunits is an antigenic mimic of the trimeric virion-associated structure. J Virol 74:627–643 Bjerregaard B, Holck S, Christensen IJ et al (2006) Syncytin is involved in breast cancerendothelial cell fusions. Cell Mol Life Sci 63:1906–1911 Black S, Kadyrov M, Kaufmann P et al (2004) Syncytial fusion of human trophoblast depends on caspase 8. Cell Death Differ 11:90–98 Blaise S, de Parseval N, Benit L et al (2003) Genomewide screening for fusogenic human endogenous retrovirus envelopes identifies syncytin 2, a gene conserved on primate evolution. Proc Natl Acad Sci USA 100:13013–13018 Blaise S, de Parseval N, Heidmann T (2005) Functional characterization of two newly identified Human Endogenous Retrovirus coding envelope genes. Retrovirology 2:19 Blaise S, Ruggieri A, Dewannieux M et al (2004) Identification of an envelope protein from the FRD family of human endogenous retroviruses (HERV-FRD) conferring infectivity and functional conservation among simians. J Virol 78:1050–1054 Blond JL, Beseme F, Duret L et al (1999) Molecular characterization and placental expression of HERV-W, a new human endogenous retrovirus family. J Virol 73:1175–1185 Blond JL, Lavillette D, Cheynet V et al (2000) An envelope glycoprotein of the human endogenous retrovirus HERV-W is expressed in the human placenta and fuses cells expressing the type D mammalian retrovirus receptor. J Virol 74:3321–3329 Boller K, Konig H, Sauter M et al (1993) Evidence that HERV-K is the endogenous retrovirus sequence that codes for the human teratocarcinoma-derived retrovirus HTDV. Virology 196:349–353 Bolmstedt A, Hemming A, Flodby P et al (1991) Effects of mutations in glycosylation sites and disulphide bonds on processing, CD4-binding and fusion activity of human immunodeficiency virus envelope glycoproteins. J Gen Virol 72(Pt 6):1269–1277 Bonifacino JS, Traub LM (2003) Signals for sorting of transmembrane proteins to endosomes and lysosomes. Annu Rev Biochem 72:395–447 Bonnaud B, Beliaeff J, Bouton O et al (2005) Natural history of the ERVWE1 endogenous retroviral locus. Retrovirology 2:57 Bonnaud B, Bouton O, Oriol G et al (2004) Evidence of selection on the domesticated ERVWE1 env retroviral element involved in placentation. Mol Biol Evol 21:1895–1901 Bouard D, Sandrin V, Boson B et al (2007) An acidic cluster of the cytoplasmic tail of the RD114 virus glycoprotein controls assembly of retroviral envelopes. Traffic 8:835–847 Bour BA, Chakravarti M, West JM et al (2000) Drosophila SNS, a member of the immunoglobulin superfamily that is essential for myoblast fusion. Genes Dev 14:1498–1511 Braakman I, van Anken E (2000) Folding of viral envelope glycoproteins in the endoplasmic reticulum. Traffic 1:533–539
106
P. Pérot et al.
Brody BA, Rhee SS, Sommerfelt MA et al (1992) A viral protease-mediated cleavage of the transmembrane glycoprotein of Mason-Pfizer monkey virus can be suppressed by mutations within the matrix protein. Proc Natl Acad Sci USA 89:3443–3447 Brugger B, Glass B, Haberkant P et al (2006) The HIV lipidome: a raft with an unusual composition. Proc Natl Acad Sci USA 103:2641–2646 Bugelski PJ, Maleeff BE, Klinkner AM et al (1995) Ultrastructural evidence of an interaction between Env and Gag proteins during assembly of HIV type 1. AIDS Res Hum Retroviruses 11:55–64 Caceres M, Thomas JW (2006) The gene of retroviral origin Syncytin 1 is specific to hominoids and is inactive in Old World monkeys. J Hered 97:100–106 Chang CW, Chuang HC, Yu C et al (2005) Stimulation of GCMa transcriptional activity by cyclic AMP/protein kinase A signaling is attributed to CBP-mediated acetylation of GCMa. Mol Cell Biol 25:8401–8414 Chang WK, Yang KD, Shaio MF (1999) Effect of glutamine on Th1 and Th2 cytokine responses of human peripheral blood mononuclear cells. Clin Immunol 93:294–301 Chaouat G (2007) The Th1/Th2 paradigm: still important in pregnancy? Semin Immunopathol 29:95–113 Chen CP, Chen LF, Yang SR et al (2008) Functional characterization of the human placental fusogenic membrane protein syncytin 2. Biol Reprod 79:815–823 Chen EH, Grote E, Mohler W et al (2007) Cell–cell fusion. FEBS Lett 581:2181–2193 Chen EH, Olson EN (2005) Unveiling the mechanisms of cell–cell fusion. Science 308:369–373 Chernomordik LV, Kozlov MM (2003) Protein-lipid interplay in fusion and fission of biological membranes. Annu Rev Biochem 72:175–207 Chernomordik LV, Kozlov MM (2005) Membrane hemifusion: crossing a chasm in two leaps. Cell 123:375–382 Chernomordik LV, Kozlov MM (2008) Mechanics of membrane fusion. Nat Struct Mol Biol 15:675–683 Cheynet V, Oriol G, Mallet F (2006) Identification of the hASCT2-binding domain of the Env ERVWE1/syncytin-1 fusogenic glycoprotein. Retrovirology 3:41 Cheynet V, Ruggieri A, Oriol G et al (2005) Synthesis, assembly, and processing of the Env ERVWE1/syncytin human endogenous retroviral envelope. J Virol 79:5585–5593 Christensen T, Sorensen PD, Riemann H et al (1998) Expression of sequence variants of endogenous retrovirus RGH in particle form in multiple sclerosis. Lancet 352:1033 Christodoulopoulos I, Cannon PM (2001) Sequences in the cytoplasmic tail of the gibbon ape leukemia virus envelope protein that prevent its incorporation into lentivirus vectors. J Virol 75:4129–4138 Cianciolo GJ, Copeland TD, Oroszlan S et al (1985) Inhibition of lymphocyte proliferation by a synthetic peptide homologous to retroviral envelope protein. Science 230:453–455 Coffin JM (1992) Genetic diversity and evolution of retroviruses. Curr Top Microbiol Immunol 176:143–164 Cohen M, Powers M, O’Connell C et al (1985) The nucleotide sequence of the env gene from the human provirus ERV3 and isolation and characterization of an ERV3-specific cDNA. Virology 147:449–458 Cohen SX, Moulin M, Hashemolhosseini S et al (2003) Structure of the GCM domain-DNA complex: a DNA-binding domain with a novel fold and mode of target site recognition. Embo J 22:1835–1845 Contreras-Galindo R, Kaplan MH, Leissner P et al (2008) Human endogenous retrovirus K (HML-2) elements in the plasma of people with lymphoma and breast cancer. J Virol 82:9329–9336 Cote M, Zheng YM, Liu SL (2009) Receptor binding and low pH coactivate oncogenic retrovirus envelope-mediated fusion. J Virol 83:11447–11455 Crowell RC, Kiessling AA (2007) Endogenous retrovirus expression in testis and epididymis. Biochem Soc Trans 35:629–633
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
107
de Parseval N, Heidmann T (1998) Physiological knockout of the envelope gene of the singlecopy ERV-3 human endogenous retrovirus in a fraction of the Caucasian population. J Virol 72:3442–3445 de Parseval N, Casella J, Gressin L, Heidmann T (2001) Characterization of the three HERV-H proviruses with an open envelope reading frame encompassing the immunosuppressive domain and evolutionary history in primates. Virology 279:558–569 de Parseval N, Lazar V, Casella JF et al (2003) Survey of human genes of retroviral origin: identification and transcriptome of the genes with coding capacity for complete envelope proteins. J Virol 77:10414–10422 Dedera D, Gu RL, Ratner L (1992) Conserved cysteine residues in the human immunodeficiency virus type 1 transmembrane envelope protein are essential for precursor envelope cleavage. J Virol 66:1207–1209 Dewannieux M, Harper F, Richaud A et al (2006) Identification of an infectious progenitor for the multiple-copy HERV-K human endogenous retroelements. Genome Res 16:1548–1556 Dolei A (2005) MSRV/HERV-W/syncytin and its linkage to multiple sclerosis: the usability and the hazard of a human endogenous retrovirus. J Neurovirol 11:232–235 Donner H, Tonjes RR, Bontrop RE et al (1999) Intronic sequence motifs of HLA-DQB1 are shared between humans, apes and Old World monkeys, but a retroviral LTR element (DQLTR3) is human specific. Tissue Antigens 53:551–558 Drewlo S, Leyting S, Kokozidou M et al (2006) C-Terminal truncations of syncytin-1 (ERVWE1 envelope) that increase its fusogenicity. Biol Chem 387:1113–1120 Duelli D, Lazebnik Y (2007) Cell-to-cell fusion as a link between viruses and cancer. Nat Rev Cancer 7:968–976 Dunlap KA, Palmarini M, Varela M et al (2006) Endogenous retroviruses regulate periimplantation placental growth and differentiation. Proc Natl Acad Sci USA 103:14390–14395 Dupressoir A, Marceau G, Vernochet C et al (2005) Syncytin-A and syncytin-B, two fusogenic placenta-specific murine envelope genes of retroviral origin conserved in Muridae. Proc Natl Acad Sci USA 102:725–730 Earl PL, Doms RW, Moss B (1990) Oligomeric structure of the human immunodeficiency virus type 1 envelope glycoprotein. Proc Natl Acad Sci USA 87:648–652 Earl PL, Moss B, Doms RW (1991) Folding, interaction with GRP78-BiP, assembly, and transport of the human immunodeficiency virus type 1 envelope protein. J Virol 65:2047–2055 Edwards TG, Wyss S, Reeves JD et al (2002) Truncation of the cytoplasmic domain induces exposure of conserved regions in the ectodomain of human immunodeficiency virus type 1 envelope protein. J Virol 76:2683–2691 Einfeld D, Hunter E (1988) Oligomeric structure of a prototype retrovirus glycoprotein. Proc Natl Acad Sci USA 85:8688–8692 Esnault C, Priet S, Ribet D et al (2008) A placenta-specific receptor for the fusogenic, endogenous retrovirus-derived, human syncytin-2. Proc Natl Acad Sci USA 105:17532–17537 Espinosa A, Villarreal LP (2000) T-Ag inhibits implantation by EC cell derived embryoid bodies. Virus Genes 20:195–200 Fang Y, Wu N, Gan X et al (2007) Higher-order oligomerization targets plasma membrane proteins and HIV gag to exosomes. PLoS Biol 5:e158 Fass D, Davey RA, Hamson CA et al (1997) Structure of a murine leukemia virus receptor-binding glycoprotein at 2.0 angstrom resolution. Science 277:1662–1666 Fass D, Harrison SC, Kim PS (1996) Retrovirus envelope domain at 1.7 angstrom resolution. Nat Struct Biol 3:465–469 Fenouillet E, Barbouche R, Jones IM (2007) Cell entry by enveloped viruses: redox considerations for HIV and SARS-coronavirus. Antioxid Redox Signal 9:1009–1034 Fenouillet E, Lavillette D, Loureiro S et al (2008) Contribution of redox status to hepatitis C virus E2 envelope protein function and antigenicity. J Biol Chem 283:26340–26348 Firouzi R, Rolland A, Michel M et al (2003) Multiple sclerosis-associated retrovirus particles cause T lymphocyte-dependent death with brain hemorrhage in humanized SCID mice model. J Neurovirol 9:79–93
108
P. Pérot et al.
Freed EO, Risser R (1987) The role of envelope glycoprotein processing in murine leukemia virus infection. J Virol 61:2852–2856 Frendo JL, Cronier L, Bertin G et al (2003a) Involvement of connexin 43 in human trophoblast cell fusion and differentiation. J Cell Sci 116:3413–3421 Frendo JL, Olivier D, Cheynet V et al (2003b) Direct involvement of HERV-W Env glycoprotein in human trophoblast cell fusion and differentiation. Mol Cell Biol 23:3566–3574 Gallaher WR, Ball JM, Garry RF et al (1989) A general model for the transmembrane proteins of HIV and other retroviruses. AIDS Res Hum Retroviruses 5:431–440 Gebhardt A, Bosch JV, Ziemiecki A et al (1984) Rous sarcoma virus p19 and gp35 can be chemically crosslinked to high molecular weight complexes. An insight into virus assembly. J Mol Biol 174:297–317 Gelderblom HR, Hausmann EH, Ozel M et al (1987) Fine structure of human immunodeficiency virus (HIV) and immunolocalization of structural proteins. Virology 156:171–176 Gimenez J, Montgiraud C, Oriol G et al (2009) Comparative methylation of ERVWE1/Syncytin-1 and other human endogenous retrovirus LTRs in placenta tissues. DNA Res 16:195–211 Gimenez J, Montgiraud C, Pichon JP et al (2010) Custom human endogenous retroviruses dedicated microarray identifies self-induced HERV-W family elements reactivated in testicular cancer upon methylation control. Nucleic Acids Res 38:2229–2246 Gong R, Huang L, Shi J (2007) Syncytin-A mediates the formation of syncytiotrophoblast involved in mouse placental development. Cell Physiol Biochem 20:517–526 Gould SJ, Booth AM, Hildreth JE (2003) The Trojan exosome hypothesis. Proc Natl Acad Sci USA 100:10592–10597 Grange MP, Blot V, Delamarre L et al (2000) Identification of two intracellular mechanisms leading to reduced expression of oncoretrovirus envelope glycoproteins at the cell surface. J Virol 74:11734–11743 Green BJ, Lee CS, Rasko JE (2004) Biodistribution of the RD114/mammalian type D retrovirus receptor, RDR. J Gene Med 6:249–259 Green N, Shinnick TM, Witte O et al (1981) Sequence-specific antibodies show that maturation of Moloney leukemia virus envelope polyprotein involves removal of a COOH-terminal peptide. Proc Natl Acad Sci USA 78:6023–6027 Grigorov B, Arcanger F, Roingeard P et al (2006) Assembly of infectious HIV-1 in human epithelial and T-lymphoblastic cell lines. J Mol Biol 359:848–862 Haahr S, Sommerlund M, Christensen T et al (1994) A putative new retrovirus associated with multiple sclerosis and the possible involvement of Epstein-Barr virus in this disease. Ann NY Acad Sci 724:148–156 Han X, Sterling H, Chen Y et al (2000) CD47, a ligand for the macrophage fusion receptor, participates in macrophage multinucleation. J Biol Chem 275:37984–37992 Hayward MD, Potgens AJ, Drewlo S et al (2007) Distribution of human endogenous retrovirus type W receptor in normal human villous placenta. Pathology 39:406–412 Heidmann O, Vernochet C, Dupressoir A et al (2009) Identification of an endogenous retroviral envelope gene with fusogenic activity and placenta-specific expression in the rabbit: a new “syncytin” in a third order of mammals. Retrovirology 6:107 Herbst H, Kuhler-Obbarius C, Lauke H et al (1999) Human endogenous retrovirus (HERV)-K transcripts in gonadoblastomas and gonadoblastoma-derived germ cell tumours. Virchows Arch 434:11–15 Hoffman LH, Wooding FB (1993) Giant and binucleate trophoblast cells of mammals. J Exp Zool 266:559–577 Hull S, Fan H (2006) Mutational analysis of the cytoplasmic tail of jaagsiekte sheep retrovirus envelope protein. J Virol 80:8069–8080 Hunter E, Swanstrom R (1990) Retrovirus envelope glycoproteins. Curr Top Microbiol Immunol 157:187–253 Huppertz B, Frank HG, Reister F et al (1999) Apoptosis promoting caspases are sequentially activated during trophoblast differentiation: analysis of villous cytotrophoblast and syncytial fragments in vitro. Lab Invest 79 (12):1687–1702
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
109
Inabe K, Nishizawa M, Tajima S et al (1999) The YXXL sequences of a transmembrane protein of bovine leukemia virus are required for viral entry and incorporation of viral envelope protein into virions. J Virol 73:1293–1301 Inoue N, Ikawa M, Isotani A et al (2005) The immunoglobulin superfamily protein Izumo is required for sperm to fuse with eggs. Nature 434:234–238 Jianjun Z, Yali H, Zhiqun W et al (2010) Imbalance of T-cell transcription factors contributes to the Th1 type immunity predominant in pre-eclampsia. Am J Reprod Immunol 63:38–45 Joshi A, Ablan SD, Soheilian F et al (2009) Evidence that productive human immunodeficiency virus type 1 assembly can occur in an intracellular compartment. J Virol 83:5375–5387 Kavanaugh MP, Wang H, Zhang Z et al (1994) Control of cationic amino acid transport and retroviral receptor functions in a membrane protein family. J Biol Chem 269:15445–15450 Keryer G, Alsat E, Tasken K et al (1998) Cyclic AMP-dependent protein kinases and human trophoblast cell differentiation in vitro. J Cell Sci 111(Pt 7):995–1004 Kiessling AA, Crowell R, Fox C (1989) Epididymis is a principal site of retrovirus expression in the mouse. Proc Natl Acad Sci USA 86:5109–5113 Kim FJ, Battini JL, Manel N et al (2004) Emergence of vertebrate retroviruses and envelope capture. Virology 318:183–191 Kim FJ, Manel N, Boublik Y et al (2003) Human T-cell leukemia virus type 1 envelope-mediated syncytium formation can be activated in resistant Mammalian cell lines by a carboxy-terminal truncation of the envelope cytoplasmic domain. J Virol 77:963–969 Kim HS, Takenaka O, Crow TJ (1999) Cloning and nucleotide sequence of retroposons specific to hominoid primates derived from an endogenous retrovirus (HERV-K). AIDS Res Hum Retroviruses 15(6):595–601 Knerr I, Beinder E, Rascher W (2002) Syncytin, a novel human endogenous retroviral gene in human placenta: evidence for its dysregulation in preeclampsia and HELLP syndrome. Am J Obstet Gynecol 186:210–213 Knerr I, Weigel C, Linnemann K et al (2003) Transcriptional effects of hypoxia on fusiogenic syncytin and its receptor ASCT2 in human cytotrophoblast BeWo cells and in ex vivo perfused placental cotyledons. Am J Obstet Gynecol 189:583–588 Knerr I, Schubert SW, Wich C (2005) Stimulation of GCMa and syncytin via cAMP mediated PKA signaling in human trophoblastic cells under normoxic and hypoxic conditions. FEBS Lett 579:3991–3998 Knerr I, Schnare M, Hermann K et al (2007) Fusiogenic endogenous-retroviral syncytin-1 exerts anti-apoptotic functions in staurosporine-challenged CHO cells. Apoptosis 12:37–43 Komurian-Pradel F, Paranhos-Baccala G, Bedin F et al (1999) Molecular cloning and characterization of MSRV-related sequences associated with retrovirus-like particles. Virology 260:1–9 Kowalski M, Potz J, Basiripour L et al (1987) Functional regions of the envelope glycoprotein of human immunodeficiency virus type 1. Science 237:1351–1355 Kudo Y, Boyd CA (2001) The physiology of immune evasion during pregnancy; the critical role of placental tryptophan metabolism and transport. Pflugers Arch 442:639–641 Kudo Y, Boyd CA, Sargent IL et al (2003) Hypoxia alters expression and function of syncytin and its receptor during trophoblast cell fusion of human placental BeWo cells: implications for impaired trophoblast syncytialisation in pre-eclampsia. Biochim Biophys Acta 1638:63–71 Lander ES, Linton LM, Birren B et al. (2001) Initial sequencing and analysis of the human genome. Nature 409:860–921 Langat DK, Johnson PM, Rote NS et al (1999) Characterization of antigens expressed in normal baboon trophoblast and cross-reactive with HIV/SIV antibodies. J Reprod Immunol 42:41–58 Larsen JM, Christensen IJ, Nielsen HJ et al (2009) Syncytin immunoreactivity in colorectal cancer: potential prognostic impact. Cancer Lett 280:44–49 Larsson LI, Holck S, Christensen IJ (2007) Prognostic role of syncytin expression in breast cancer. Hum Pathol 38:726–731
110
P. Pérot et al.
Laufer G, Mayer J, Mueller BF et al (2009) Analysis of transcribed human endogenous retrovirus W env loci clarifies the origin of multiple sclerosis-associated retrovirus env sequences. Retrovirology 6:37 Lavillette D, Marin M, Ruggieri A et al (2002) The envelope glycoprotein of human endogenous retrovirus type W uses a divergent family of amino acid transporters/cell surface receptors. J Virol 76:6442–6452 Lee X, Keith JC Jr, Stumm N et al (2001) Downregulation of placental syncytin expression and abnormal protein localization in pre-eclampsia. Placenta 22:808–812 Lee YN, Bieniasz PD (2007) Reconstitution of an infectious human endogenous retrovirus. PLoS Pathog 3:e10 Lepperdinger G, Strobl B, Kreil G (1998) HYAL2, a human gene expressed in many cells, encodes a lysosomal hyaluronidase with a novel type of specificity. J Biol Chem 273:22466–22470 Li H, Chien PC Jr, Tuen M et al (2008a) Identification of an N-linked glycosylation in the C4 region of HIV-1 envelope gp120 that is critical for recognition of neighboring CD4 T cell epitopes. J Immunol 180:4011–4021 Li K, Zhang S, Kronqvist M et al (2008b) Intersubunit disulfide isomerization controls membrane fusion of human T-cell leukemia virus Env. J Virol 82:7135–7143 Li M, Yang C, Tong S et al (2002) Palmitoylation of the murine leukemia virus envelope protein is critical for lipid raft association and surface expression. J Virol 76:11845–11852 Li Y, Bergeron JJ, Luo L et al (1996) Effects of inefficient cleavage of the signal sequence of HIV-1 gp 120 on its association with calnexin, folding, and intracellular transport. Proc Natl Acad Sci USA 93:9606–9611 Li Z, Pinter A, Kayman SC (1997) The critical N-linked glycan of murine leukemia virus envelope protein promotes both folding of the C-terminal domains of the precursor polyprotein and stability of the postcleavage envelope complex. J Virol 71:7012–7019 Lin L, Xu B, Rote NS (1999) Expression of endogenous retrovirus ERV-3 induces differentiation in BeWo, a choriocarcinoma model of human placental trophoblast. Placenta 20:109–118 Lin L, Xu B, Rote NS (2000) The cellular mechanism by which the human endogenous retrovirus ERV-3 env gene affects proliferation and differentiation in a human placental trophoblast model, BeWo. Placenta 21:73–78 Lindeskog M, Mager DL, Blomberg J (1999) Isolation of a human endogenous retroviral HERV-H element with an open env reading frame. Virology 258:441–450 Lodge R, Lalonde JP, Lemay G et al (1997) The membrane-proximal intracytoplasmic tyrosine residue of HIV-1 envelope glycoprotein is critical for basolateral targeting of viral budding in MDCK cells. EMBO J 16:695–705 Lopalco L, Longhi R, Ciccomascolo F et al (1993) Identification of human immunodeficiency virus type 1 glycoprotein gp120/gp41 interacting sites by the idiotypic mimicry of two monoclonal antibodies. AIDS Res Hum Retroviruses 9:33–39 Lopez-Verges S, Camus G, Blot G et al (2006) Tail-interacting protein TIP47 is a connector between Gag and Env and is required for Env incorporation into HIV-1 virions. Proc Natl Acad Sci USA 103:14947–14952 Lorizate M, Cruz A, Huarte N et al (2006) Recognition and blocking of HIV-1 gp41 pretransmembrane sequence by monoclonal 4E10 antibody in a Raft-like membrane environment. J Biol Chem 281:39598–39606 Loving R, Li K, Wallin M et al (2008) R-Peptide cleavage potentiates fusion-controlling isomerization of the intersubunit disulfide in Moloney murine leukemia virus Env. J Virol 82:2594–2597 Lyden TW, Johnson PM, Mwenda JM et al (1994) Ultrastructural characterization of endogenous retroviral particles isolated from normal human placentas. Biol Reprod 51:152–157 Lyden TW, Ng AK, Rote NS (1993) Modulation of phosphatidylserine epitope expression by BeWo cells during forskolin treatment. Placenta 14:177–186 MacKrell AJ, Soong NW, Curtis CM et al (1996) Identification of a subdomain in the Moloney murine leukemia virus envelope protein involved in receptor binding. J Virol 70:1768–1774
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
111
Malassine A, Blaise S, Handschuh K et al (2007) Expression of the fusogenic HERV-FRD Env glycoprotein (syncytin 2) in human placenta is restricted to villous cytotrophoblastic cells. Placenta 28:185–191 Malassine A, Handschuh K, Tsatsaris V et al (2005) Expression of HERV-W Env glycoprotein (syncytin) in the extravillous trophoblast of first trimester human placenta. Placenta 26:556–562 Malik HS, Henikoff S, Eickbush TH (2000) Poised for contagion: evolutionary origins of the infectious abilities of invertebrate retroviruses. Genome Res 10:1307–1318 Mallet F, Bouton O, Prudhomme S et al (2004) The endogenous retroviral locus ERVWE1 is a bona fide gene involved in hominoid placental physiology. Proc Natl Acad Sci USA 101:1731–1736 Mameli G, Astone V, Arru G et al (2007) Brains and peripheral blood mononuclear cells of multiple sclerosis (MS) patients hyperexpress MS-associated retrovirus/HERV-W endogenous retrovirus, but not Human herpesvirus 6. J Gen Virol 88:264–274 Mangeney M, Renard M, Schlecht-Louf G et al (2007) Placental syncytins: Genetic disjunction between the fusogenic and immunosuppressive activity of retroviral envelope proteins. Proc Natl Acad Sci USA 104:20534–20539 Marechal V, Prevost MC, Petit C et al (2001) Human immunodeficiency virus type 1 entry into macrophages mediated by macropinocytosis. J Virol 75:11166–11177 Marsh M, Helenius A (2006) Virus entry: open sesame. Cell 124:729–740 Martens S, McMahon HT (2008) Mechanisms of membrane fusion: disparate players and common principles. Nat Rev Mol Cell Biol 9:543–556 Matouskova M, Blazkova J, Pajer P et al (2006) CpG methylation suppresses transcriptional activity of human syncytin-1 in non-placental tissues. Exp Cell Res 312:1011–1020 Melikyan GB (2008) Common principles and intermediates of viral protein-mediated fusion: the HIV-1 paradigm. Retrovirology 5:111 Melikyan GB, Barnard RJ, Abrahamyan LG et al (2005) Imaging individual retroviral fusion events: from hemifusion to pore formation and growth. Proc Natl Acad Sci USA 102:8728–8733 Mellor A, Zhou M, Conway SJ et al (1999) HLA-G transgenic mice. J Reprod Immunol 43:253–261 Mercer J, Helenius A (2009) Virus entry by macropinocytosis. Nat Cell Biol 11:510–520 Mi S, Lee X, Li X et al (2000) Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature 403:785–789 Mizuochi T, Matthews TJ, Kato M et al (1990) Diversity of oligosaccharide structures on the envelope glycoprotein gp 120 of human immunodeficiency virus 1 from the lymphoblastoid cell line H9. Presence of complex-type oligosaccharides with bisecting N-acetylglucosamine residues. J Biol Chem 265:8519–8524 Mohler WA (2009) Cell–cell fusion: transient channels leading to plasma membrane merger http://www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book=eurekah&part=A59047. Moreno MR, Giudici M, Villalain J (2006) The membranotropic regions of the endo and ecto domains of HIV gp41 envelope glycoprotein. Biochim Biophys Acta 1758:111–123 Morita E, Sundquist WI (2004) Retrovirus budding. Annu Rev Cell Dev Biol 20:395–425 Mortensen K, Christensen IJ, Nielsen HJ et al (2004) High expression of endothelial cell nitric oxide synthase in peritumoral microvessels predicts increased disease-free survival in colorectal cancer. Cancer Lett 216:109–114 Muir A, Lever AM, Moffett A (2006) Human endogenous retrovirus-W envelope (syncytin) is expressed in both villous and extravillous trophoblast populations. J Gen Virol 87:2067–2071 Munoz-Barroso I, Durell S, Sakaguchi K et al (1998) Dilation of the human immunodeficiency virus-1 envelope glycoprotein fusion pore revealed by the inhibitory action of a synthetic peptide from gp41. J Cell Biol 140:315–323 Munoz-Barroso I, Salzwedel K, Hunter E et al (1999) Role of the membrane-proximal domain in the initial stages of human immunodeficiency virus type 1 envelope glycoprotein-mediated membrane fusion. J Virol 73:6089–6092
112
P. Pérot et al.
Novakovic S, Sawai ET, Radke K (2004) Dileucine and YXXL motifs in the cytoplasmic tail of the bovine leukemia virus transmembrane envelope protein affect protein expression on the cell surface. J Virol 78:8301–8311 Nydegger S, Foti M, Derdowski A et al (2003) HIV-1 egress is gated through late endosomal membranes. Traffic 4:902–910 Ochsenbauer-Jambor C, Miller DC, Roberts CR et al (2001) Palmitoylation of the Rous sarcoma virus transmembrane glycoprotein is required for protein stability and virus infectivity. J Virol 75:11544–11554 Ono A, Ablan SD, Lockett SJ et al (2004) Phosphatidylinositol (4,5) bisphosphate regulates HIV-1 Gag targeting to the plasma membrane. Proc Natl Acad Sci USA 101:14889–14894 Oppelt P, Strick R, Strissel PL et al (2009) Expression of the human endogenous retroviruse-W envelope gene syncytin in endometriosis lesions. Gynecol Endocrinol 25:741–747 Otteken A, Moss B (1996) Calreticulin interacts with newly synthesized human immunodeficiency virus type 1 envelope glycoprotein, suggesting a chaperone function similar to that of calnexin. J Biol Chem 271:97–103 Owens RJ, Burke C, Rose JK (1994) Mutations in the membrane-spanning domain of the human immunodeficiency virus envelope glycoprotein that affect fusion activity. J Virol 68:570–574 Owens RJ, Compans RW (1990) The human immunodeficiency virus type 1 envelope glycoprotein precursor acquires aberrant intermolecular disulfide bonds that may prevent normal proteolytic processing. Virology 179:827–833 Owens RJ, Dubay JW, Hunter E et al (1991) Human immunodeficiency virus envelope protein determines the site of virus release in polarized epithelial cells. Proc Natl Acad Sci USA 88:3987–3991 Palmarini M, Gray CA, Carpenter K et al (2001) Expression of endogenous betaretroviruses in the ovine uterus: effects of neonatal age, estrous cycle, pregnancy, and progesterone. J Virol 75:11319–11327 Palmarini M, Sharp JM, De las Heras M et al (1999) Jaagsiekte sheep retrovirus is necessary and sufficient to induce a contagious lung cancer in sheep. J Virol 73:6964–6972 Pawelek JM, Chakraborty AK (2008) The cancer cell – leukocyte fusion theory of metastasis. Adv Cancer Res 101:397–444 Pearce-Pratt R, Malamud D, Phillips DM (1994) Role of the cytoskeleton in cell-to-cell transmission of human immunodeficiency virus. J Virol 68:2898–2905 Peng X, Pan J, Gong R et al (2007) Functional characterization of syncytin-A, a newly murine endogenous virus envelope protein. Implication for its fusion mechanism. J Biol Chem 282:381–389 Pernet O, Pohl C, Ainouze M et al (2009) Nipah virus entry can occur by macropinocytosis. Virology 395:298–311 Perron H, Firouzi R, Tuke PW et al (1997) Cell cultures and associated retroviruses in multiple sclerosis. Acta Neurol Scand Suppl 169:22–31 Perron H, Geny C, Laurent A et al (1989) Leptomeningeal cell line from multiple sclerosis with reverse transcriptase activity and viral particles. Res Virol 140:551–561 Perron H, Jouvin-Marche E, Michel M et al (2001) Multiple sclerosis retrovirus particles and recombinant envelope trigger an abnormal immune response in vitro, by inducing polyclonal Vbeta16 T-lymphocyte activation. Virology 287:321–332 Perron H, Suh M, Lalande B et al (1993) Herpes simplex virus ICP0 and ICP4 immediate early proteins strongly enhance expression of a retrovirus harboured by a leptomeningeal cell line from a patient with multiple sclerosis. J Gen Virol 74:65–72 Pichon JP, Bonnaud B, Mallet F (2006) Quantitative multiplex degenerate PCR for human endogenous retrovirus expression profiling. Nat Protoc 1:2831–2838 Piller SC, Dubay JW, Derdeyn CA et al (2000) Mutational analysis of conserved domains within the cytoplasmic tail of gp41 from human immunodeficiency virus type 1: effects on glycoprotein incorporation and infectivity. J Virol 74:11717–11723
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
113
Pinter A, Honnen WJ (1988) O-linked glycosylation of retroviral envelope gene products. J Virol 62:1016–1021 Pinter A, Kopelman R, Li Z et al (1997) Localization of the labile disulfide bond between SU and TM of the murine leukemia virus envelope protein complex to a highly conserved CWLC motif in SU that resembles the active-site sequence of thiol-disulfide exchange enzymes. J Virol 71:8073–8077 Polonoff E, Machida CA, Kabat D (1982) Glycosylation and intracellular transport of membrane glycoproteins encoded by murine leukemia viruses. Inhibition by amino acid analogues and by tunicamycin. J Biol Chem 257:14023–14028 Ponferrada VG, Mauck BS, Wooley DP (2003) The envelope glycoprotein of human endogenous retrovirus HERV-W induces cellular resistance to spleen necrosis virus. Arch Virol 148:659–675 Potgens AJ, Drewlo S, Kokozidou M et al (2004) Syncytin: the major regulator of trophoblast fusion? Recent developments and hypotheses on its action. Hum Reprod Update 10:487–496 Poumbourios P, Wilson KA, Center RJ et al (1997) Human immunodeficiency virus type 1 envelope glycoprotein oligomerization requires the gp41 amphipathic alpha-helical/leucine zipper-like sequence. J Virol 71:2041–2049 Prudhomme S, Oriol G, Mallet F (2004) A retroviral promoter and a cellular enhancer define a bipartite element which controls env ERVWE1 placental expression. J Virol 78:12157–12168 Purtscher M, Trkola A, Gruber G et al (1994) A broadly neutralizing human monoclonal antibody against gp41 of human immunodeficiency virus type 1. AIDS Res Hum Retroviruses 10:1651–1658 Qian K, Morris-Natschke SL, Lee KH (2009) HIV entry inhibitors and their potential in HIV therapy. Med Res Rev 29:369–393 Rai SK, DeMartini JC, Miller AD (2000) Retrovirus vectors bearing jaagsiekte sheep retrovirus Env transduce human cells by using a new receptor localized to chromosome 3p21.3. J Virol 74:4698–4704 Rai SK, Duh FM, Vigdorovich V et al (2001) Candidate tumor suppressor HYAL2 is a glycosylphosphatidylinositol (GPI)-anchored cell-surface receptor for jaagsiekte sheep retrovirus, the envelope protein of which mediates oncogenic transformation. Proc Natl Acad Sci USA 98:4443–4448 Ratner L (1992) Glucosidase inhibitors for treatment of HIV-1 infection. AIDS Res Hum Retroviruses 8:165–173 Rein A, Mirro J, Haynes JG et al (1994) Function of the cytoplasmic domain of a retroviral transmembrane protein: p15E–p2E cleavage activates the membrane fusion capability of the murine leukemia virus Env protein. J Virol 68(3):1773–1781 Rolland A, Jouvin-Marche E, Viret C et al (2006) The envelope protein of a human endogenous retrovirus-W family activates innate immunity through CD14/TLR4 and promotes Th1-like responses. J Immunol 176:7636–7644 Rozenberg-Adler Y, Conner J, Guilar-Carreno H et al (2008) Membrane-proximal cytoplasmic domain of Moloney murine leukemia virus envelope tail facilitates fusion. Exp Mol Pathol 84:18–30 Salzwedel K, West JT, Hunter E (1999) A conserved tryptophan-rich motif in the membraneproximal region of the human immunodeficiency virus type 1 gp41 ectodomain is important for Env-mediated fusion and virus infectivity. J Virol 73:2469–2480 Sanders DA (2000) Sulfhydryl involvement in fusion mechanisms. Subcell Biochem 34:483–514 Sandrin V, Cosset FL (2006) Intracellular versus cell surface assembly of retroviral pseudotypes is determined by the cellular localization of the viral glycoprotein, its capacity to interact with Gag, and the expression of the Nef protein. J Biol Chem 281:528–542 Sandrin V, Muriaux D, Darlix JL et al (2004) Intracellular trafficking of Gag and Env proteins and their interactions modulate pseudotyping of retroviruses. J Virol 78:7153–7164 Schubert SW, Lamoureux N, Kilian K et al (2008) Identification of integrin-alpha4, Rb1, and syncytin a as murine placental target genes of the transcription factor GCMa/Gcm1. J Biol Chem 283:5460–5465
114
P. Pérot et al.
Schulz TF, Jameson BA, Lopalco L et al (1992) Conserved structural features in the interaction between retroviral surface and transmembrane glycoproteins? AIDS Res Hum Retroviruses 8:1571–1580 Schulz WA, Steinhoff C, Florl AR (2006) Methylation of endogenous human retroelements in health and disease. Curr Top Microbiol Immunol 310:211–250 Sherer NM, Lehmann MJ, Jimenez-Soto LF et al (2003) Visualization of retroviral replication in living cells reveals budding into multivesicular bodies. Traffic 4:785–801 Sitbon M, d Auriol L, Ellerbrok H et al (1991) Substitution of leucine for isoleucine in a sequence highly conserved among retroviral envelope surface glycoproteins attenuates the lytic effect of the Friend murine leukemia virus. Proc Natl Acad Sci USA 88:5932–5936 Smallwood A, Papageorghiou A, Nicolaides K et al (2003) Temporal regulation of the expression of syncytin (HERV-W), maternally imprinted PEG10, and SGCE in human placenta. Biol Reprod 69:286–293 Sommerfelt MA, Weiss RA (1990) Receptor interference groups of 20 retroviruses plating on human cells. Virology 176:58–69 Song C, Hunter E (2003) Variable sensitivity to substitutions in the N-terminal heptad repeat of Mason-Pfizer monkey virus transmembrane protein. J Virol 77:7779–7785 Song C, Micoli K, Bauerova H et al (2005) Amino acid residues in the cytoplasmic domain of the Mason-Pfizer monkey virus glycoprotein critical for its incorporation into virions. J Virol 79:11559–11568 Spies CP, Ritter GD Jr, Mulligan MJ et al (1994) Truncation of the cytoplasmic domain of the simian immunodeficiency virus envelope glycoprotein alters the conformation of the external domain. J Virol 68:585–591 Strick R, Ackermann S, Langbein M et al (2007) Proliferation and cell–cell fusion of endometrial carcinoma are induced by the human endogenous retroviral Syncytin-1 and regulated by TGFbeta. J Mol Med 85:23–38 Suarez T, Gallaher WR, Agirre A et al (2000) Membrane interface-interacting sequences within the ectodomain of the human immunodeficiency virus type 1 envelope glycoprotein: putative role during viral fusion. J Virol 74:8038–8047 Suomalainen M (2002) Lipid rafts and assembly of enveloped viruses. Traffic 3:705–709 Tailor CS, Nouri A, Kabat D (2000) A comprehensive approach to mapping the interacting surfaces of murine amphotropic and feline subgroup B leukemia viruses with their cell surface receptors. J Virol 74:237–244 Takeda M, Leser GP, Russell CJ, Lamb RA (2003) Influenza virus hemagglutinin concentrates in lipid raft microdomains for efficient viral fusion. Proc Natl Acad Sci USA 100:14610–14617 Tarlinton R, Meers J, Young P (2008) Biology and evolution of the endogenous koala retrovirus. Cell Mol Life Sci 65:3413–3421 Taylor GM, Sanders DA (1999) The role of the membrane-spanning domain sequence in glycoprotein-mediated membrane fusion. Mol Biol Cell 10:2803–2815 Thomas A, Roth MJ (1995) Analysis of cysteine mutations on the transmembrane protein of Moloney murine leukemia virus. Virology 211:285–289 Trujillo JR, Rogers R, Molina RM et al (2007) Noninfectious entry of HIV-1 into peripheral and brain macrophages mediated by the mannose receptor. Proc Natl Acad Sci USA 104:5097–5102 Tucker SP, Srinivas RV, Compans RW (1991) Molecular domains involved in oligomerization of the Friend murine leukemia virus envelope glycoprotein. Virology 185:710–720 Turner G, Barbulescu M, Su M et al (2001) Insertional polymorphisms of full-length endogenous retroviruses in humans. Curr Biol 11:1531–1535 van den Eijnde SM, van den Hoff MJ, Reutelingsperger CP et al (2001) Transient expression of phosphatidylserine at cell–cell contact areas is required for myotube formation. J Cell Sci 114:3631–3642 van Regenmortel MH, Mayo MA, Fauquet CM et al (2000) Virus nomenclature: consensus versus chaos. Arch Virol 145:2227–2232
4
A Comparative Portrait of Retroviral Fusogens and Syncytins
115
Vargas A, Moreau J, Landry S et al (2009) Syncytin-2 plays an important role in the fusion of human trophoblast cells. J Mol Biol 392:301–318 Venables PJ, Brookes SM, Griffiths D et al (1995) Abundance of an endogenous retroviral envelope protein in placental trophoblasts suggests a biological function. Virology 211:589–592 Voisset C, Blancher A, Perron H et al (1999) Phylogeny of a novel family of human endogenous retrovirus sequences, HERV-W, in humans and other primates. AIDS Res Hum Retroviruses 15:1529–1533 Voisset C, Bouton O, Bedin F et al (2000) Chromosomal distribution and coding capacity of the human endogenous retrovirus HERV-W family. AIDS Res Hum Retroviruses 16:731–740 von Schwedler UK, Stuchell M, Muller B et al (2003) The protein network of HIV budding. Cell 114:701–713 Wallin M, Ekstrom M, Garoff H (2004) Isomerization of the intersubunit disulphide-bond in Env controls retrovirus fusion. Embo J 23:54–65 Wang E, Obeng-Adjei N, Ying Q et al (2008) Mouse mammary tumor virus uses mouse but not human transferrin receptor 1 to reach a low pH compartment and infect cells. Virology 381:230–240 Waterston RH, Lindblad-Toh K, Birney E et al. (2002) Initial sequencing and comparative analysis of the mouse genome. Nature 420:520–562 Weissenhorn W, Dessen A, Harrison SC et al (1997) Atomic structure of the ectodomain from HIV-1 gp41. Nature 387:426–430 Weissenhorn W, Hinz A, Gaudin Y (2007) Virus membrane fusion. FEBS Lett 581:2150–2155 Welsch S, Keppler OT, Habermann A et al (2007) HIV-1 buds predominantly at the plasma membrane of primary human macrophages. PLoS Pathog 3:e36 Wooding FB (1984) Role of binucleate cells in fetomaternal cell fusion at implantation in the sheep. Am J Anat 170:233-250 Wyss S, Berlioz-Torrent C, Boge M et al (2001) The highly conserved C-terminal dileucine motif in the cytosolic domain of the human immunodeficiency virus type 1 envelope glycoprotein is critical for its association with the AP-1 clathrin adaptor [correction of adapter]. J Virol 75:2982–2992 Xiong Y, Eickbush TH (1990) Origin and evolution of retroelements based upon their reverse transcriptase sequences. EMBO J 9:3353–3362 Xu Y, Lu H, Kennedy JP et al (2006) Evaluation of "credit card" libraries for inhibition of HIV-1 gp41 fusogenic core formation. J Comb Chem 8:531–539 Yang C, Compans RW (1996) Analysis of the cell fusion activities of chimeric simian immunodeficiency virus-murine leukemia virus envelope proteins: inhibitory effects of the R peptide. J Virol 70:248–254 Yang C, Spies CP, Compans RW (1995) The human and simian immunodeficiency virus envelope glycoprotein transmembrane subunits are palmitoylated. Proc Natl Acad Sci USA 92:9871–9875 Yeh CL, Hsu CS, Yeh SL et al (2005) Dietary glutamine supplementation modulates Th1/Th2 cytokine and interleukin-6 expressions in septic mice. Cytokine 31:329–334 Yu C, Shen K, Lin M et al (2002) GCMa regulates the syncytin-mediated trophoblastic fusion. J Biol Chem 277:50062–50068 Zhao Y, Zhu L, Benedict CA et al (1998) Functional domains in the retroviral transmembrane protein. J Virol 72:5392–5398 Zhou Z, Shen T, Zhang BH et al (2009) The proprotein convertase furin in human trophoblast: Possible role in promoting trophoblast cell migration and invasion. Placenta 30:929–938
Chapter 5
Syncytins: Molecular Aspects Hungwen Chen and Mei-Leng Cheong
Abstract Cell–cell fusion is essential for many physiological events such as egg-sperm fertilization, muscle differentiation, viral entry, and placental development. The envelope genes of human endogenous retrovirus (HERV)-W and -FRD, named syncytin-1 and -2, are primarily expressed in placenta and mediate trophoblastic fusion for formation of the multinucleated syncytiotrophoblast layer, which is essential for fetal growth. Recently, envelope genes of retroviral origin encoding fusogenic proteins have also been identified from the genomes of other mammals. These novel envelope genes, named mouse syncytin-A and -B, rabbit syncytin-Ory1, and sheep enJSRV envelope, are required for placental development and trophoblastic fusion. Although syncytins are critical for placental development, abnormal expression of syncytins is associated with placental disorders, cancers, and neurological diseases. Here we provide a general description of the biological functions of syncytins and the regulation of syncytin gene expression at the molecular level. In addition, the pathological functions of syncytins are also discussed. Keywords cAMP · cancer · cytotrophoblast · envelope proteins · GCM1 · HERV-FRD · HERV-W · membrane fusion · multiple sclerosis · preeclampsia · receptors · syncytin · syncytiotrophoblast Abbreviations AKT ASCT CBP CREB CTM en Env
Protein kinase B Alanine, serine and cysteine selective transporters CREB binding protein cAMP response element binding protein C-terminal part of TM Endogenous Envelope
H. Chen (B) Institute of Biological Chemistry, Academia Sinica, Taipei, Taiwan e-mail:
[email protected] L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_5, C Springer Science+Business Media B.V. 2011
117
118
FBW GCM GSK HA HDAV HELLP HERV HIV HR IL JSRV LTR MFSD2 MLV MS OASIS PC PI-3 K PKA RDR SCF SKP SU SUMO SynT TM
H. Chen and M.-L. Cheong
F-box/WD repeat-containing protein Glial cell missing Glycogen synthase kinase Hemagglutinin Histone deacetylase Hemolysis, elevated liver enzymes and low platelets Human endogenous retrovirus Human immunodeficiency virus Heptad repeat Interleukin Jaagsiekte sheep retrovirus Long terminal repeat Major facilitator superfamily domain containing 2 Murine leukemia virus Multiple sclerosis Old astrocytes specifically induced substance Proprotein convertase Phosphatidyl inositol-3 kinase Protein kinase A Retrovirus D-type receptor Skp, Cullin, F-box S-phase kinase-associated protein Surface unit Small Ubiquitin-like Modifier Syncytiotrophoblast Transmembrane unit
Contents 5.1 Cell Fusion in the Placenta . . . . . . . . . . . . . . . . . . 5.1.1 Development of the Placenta . . . . . . . . . . . . . . 5.1.2 Human Syncytin-1 . . . . . . . . . . . . . . . . . . . 5.1.3 Human Syncytin-2 . . . . . . . . . . . . . . . . . . . 5.1.4 Mouse Syncytin-A and -B . . . . . . . . . . . . . . . 5.2 Structure and Functional Studies of Syncytins . . . . . . . . . 5.2.1 Biosynthesis of Syncytins . . . . . . . . . . . . . . . . 5.2.2 Functional Domains and Motifs in Syncytins . . . . . . . 5.2.3 Syncytin Receptors . . . . . . . . . . . . . . . . . . 5.2.4 Mechanism of Membrane Fusion . . . . . . . . . . . . 5.3 Regulation of Syncytin Expression . . . . . . . . . . . . . . 5.3.1 GCM1 Regulation of Syncytin-1 and -2 Gene Expression . 5.3.2 Regulation of GCM1 Activity . . . . . . . . . . . . . . 5.3.3 Epigenetic Regulation of Syncytin-1 and -2 Gene Expression 5.4 Syncytins and Disease . . . . . . . . . . . . . . . . . . . . 5.4.1 Syncytins in Placental Disorders . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
119 119 120 121 122 122 123 124 126 127 128 128 129 130 131 131
5
Syncytins: Molecular Aspects
5.4.2 Syncytin-1 in Malignancies . . . . 5.4.3 Syncytin-1 in Neurological Diseases 5.5 Conclusion . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . .
119
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
132 133 134 135
5.1 Cell Fusion in the Placenta The first differentiation event in the early stage of embryonic development is the formation of the trophectoderm, which later develops into specialized placental trophoblasts essential for implantation and placentation. The primary physiological functions of placenta include mediation of fetal-maternal nutrient and gas exchange, synthesis of hormones and growth factors, and protection of the fetus from maternal immune surveillance. Recent epidemiological studies have revealed a highly significant association between undernourished fetuses and a greater risk of developing stroke, heart disease, and type II diabetes in the adult (Barker 2004). Therefore, a functional placenta is essential for both a successful pregnancy outcome and a healthy adult life.
5.1.1 Development of the Placenta The definitive structure of the human placenta becomes apparent as early as day 21 of pregnancy (Benirschke and Kaufmann 2001b). In the early gestation stage, cytotrophoblast stem cells facing the maternal decidua proliferate and fuse to form a syncytial mass. Later, vascular spaces called trophoblastic lacunae appear in the syncytium around day 8–9. The cytotrophoblast cell layer under the syncytium rapidly proliferates into lacunae and forms the primary chorionic villi. Subsequently, proliferation of the cytotrophoblasts, growth of mesenchymal cells under the cytotrophoblast cell layer, and development of blood vessels transform the primary villi into secondary and tertiary villi. The fetal placental arteries and veins develop within the chorionic villi and are surrounded by an inner layer of cytotrophoblasts and an outer layer, the multinucleated syncytiotrophoblast, formed by cell–cell fusion of underlying mononucleated cytotrophoblasts (Fig. 5.1a) (Benirschke and Kaufmann 2001a). The physiological functions of syncytiotrophoblast include gas and nutrient exchange between fetus and mother, hormone production, and immunomodulation. For mouse placental development, the trophectoderm cells overlying the inner cell mass proliferate both inwards to form the extraembryonic ectoderm and outwards to form the ectoplacental cone. As the development of the mouse placenta proceeds, the extraembryonic ectoderm expands to form the chorionic epithelium. The allantois arises from the mesoderm at the posterior end of the embryo and makes contact with the chorion later at embryonic day 8.5 (E8.5), an event termed chorioallantoic attachment (Cross et al. 2006, Hemberger and Cross 2001). The interaction of the chorion with the allantois folds into a villous tree-like structure called the labyrinth, which is analogous to the chorionic villi in human placenta. In the labyrinth, fetal blood vessels are separated from the maternal circulation by three
120
H. Chen and M.-L. Cheong
Fig. 5.1 Placental anatomy in human and mouse. (a) Structure of the human placental villus. The human placental villus is composed of mesenchymal cells, fetal blood vessels, a mononuclear cytotrophoblast layer, and a multinuclear syncytiotrophoblast layer. The latter is in contact with the maternal blood and is essential for gas and nutrient exchange between mother and fetus. Note the migration and invasion of cytotrophoblasts into the maternal endometrium, which remodels the endometrial arteries to increase maternal blood flow to the placenta. (b) Structure of murine placental labyrinth. The murine placental labyrinth contains juxtaposed maternal blood sinuses and fetal blood vessels separated by a trilaminar trophoblast cell structure, which is composed of a single layer of mononuclear trophoblast cells and two layers of multinuclear syncytiotrophoblasts (I and II)
layers of trophoblast cells, including a layer of mononuclear trophoblast cells that is in contact with the maternal blood sinusoids, and two syncytiotrophoblast layers, SynT-I and SynT-II. SynT-I also faces the maternal blood space, whereas SynT-II is in contact with the fetal blood vessel (Fig. 5.1b) (Rossant and Cross 2001, Simmons et al. 2008).
5.1.2 Human Syncytin-1 Approximately 8% of the human genome contains sequences of retroviral origin, including gag (structural proteins), pol (viral enzymes), env (envelope proteins), and long terminal repeats (LTRs). Collectively, these sequence elements are named human endogenous retroviruses (HERVs). The gag gene encodes structural proteins for viral genome encapsidation and particle formation. The pol gene encodes viral enzymes for reverse transcription of the viral genome into a double-stranded DNA and subsequent integration of this DNA into the host genome. The env gene encodes a viral envelope protein that recognizes the host cell surface receptor and mediates viral-host membrane fusion. The LTRs contain regulatory elements for
5
Syncytins: Molecular Aspects
121
transcription initiation and termination. Most HERVs are defective in replication due to accumulated mutations in their genomes. Cell–cell fusion of cytotrophoblasts is essential for formation of thesyncytiotrophoblast. Syncytin is the first identified fusogenic protein mediating trophoblastic fusion in human placenta. As a functional envelope protein of the HERV-W family, syncytin was independently identified and characterized by two groups. Blond and coworkers (Blond et al. 1999) investigated the expression of HERV sequences related to the multiple sclerosis-associated retrovirus in healthy tissues and identified a novel envelope gene of HERV-W that is highly expressed in placenta. At about the same time, using a yeast signal sequence trap for identification of novel secreted proteins, Mi and colleagues (Mi et al. 2000) identified a cDNA fragment that is also primarily expressed in human placenta and named syncytin. Syncytin turns out to be identical to the HERV-W envelope gene reported by Blond et al. Further studies demonstrated that the syncytin gene is specifically expressed in the placental syncytiotrophoblast layer and that ectopic expression of syncytin protein in non-placental cells such as COS cells promotes cell–cell fusion (Mi et al. 2000). It has long been known that elevation of cAMP level or activation of PKA can stimulate cell–cell fusion of human placental cells (Keryer et al. 1998). Correspondingly, the transcript level of syncytin gene is positively stimulated by the cAMP stimulant, forskolin, in the human BeWo placental cell line. In addition, cell–cell fusion of BeWo cells is blocked in the presence of antiserum against syncytin, suggesting that syncytin is involved in mediation of trophoblastic fusion (Mi et al. 2000). For nomenclature purpose, syncytin has been renamed syncytin-1 to distinguish it from a second HERV envelope gene named syncytin-2, described in the next subsection.
5.1.3 Human Syncytin-2 In silico analysis of human genome databases has revealed an additional 15 envelope genes predicted to encode candidate fusogenic proteins. Functional screening of these envelope genes by cell fusion assays is performed by transient expression experiments in a panel of mammalian cell lines. Of these candidate genes, one HERV-FRD envelope gene demonstrated to encode a fusogenic protein is also highly expressed in placenta. This HERV-FRD envelope gene is named syncytin-2, for encoding the second placental fusogenic protein (Blaise et al. 2003). Interestingly, similarly to syncytin-1, syncytin-2 gene expression is also upregulated in placental cells treated with forskolin. Therefore, both syncytin-1 and -2 are likely the downstream effectors in the cAMP/PKA signaling pathway that stimulates placental cell fusion. Sequence comparison indicated that the syncytin-2 sequence is highly conserved in primates from humans to New World monkeys. Phylogenetic analysis suggests that syncytin-1 and -2 genes entered the primate genome about 25 and 40 million years ago, respectively (Blaise et al. 2003). Interestingly, unlike syncytin-1, syncytin-2 possesses potential immunosuppressive activity. This is evidenced by the fact that overexpression of syncytin-2 in tumor
122
H. Chen and M.-L. Cheong
cells prevents their rejection in allogenic mouse hosts (Mangeney et al. 2007). Therefore, the immunosuppressive activity of syncytin-2 may protect the fetus against the maternal immune system. Expression of syncytin-2 transcript has been detected in placental cytotrophoblasts by in situ hybridization (Esnault et al. 2008, Kudaka et al. 2008).
5.1.4 Mouse Syncytin-A and -B About 10% of the sequences in the murine genome are of retroviral origin. A systematic in silico search of mouse genome databases has identified two murine endogenous retroviral envelope genes named syncytin-A and -B that encode placental fusogenic proteins. Evolutionarily, syncytin-A and -B entered the rodent/murine genome about 20 million years ago. Expression of syncytin-A and -B transcripts can be detected from 9.5 days postcoitum, increasing till 14.5 days postcoitum (Dupressoir et al. 2005). In situ hybridization further localized the syncytin-A and -B transcripts to the SynT-I and SynT-II layers of the labyrinth, respectively (Simmons et al. 2008). A recent gene targeting study of syncytinA demonstrates that syncytin-A is critical for formation of the SynT-I layer and for proper placental development. Poor placental vascularization was found in the syncytin-A knockout mice due to overexpansion of unfused trophoblast cells, which reduced fetal blood vessel spaces (Dupressoir et al. 2009). This study also suggests that syncytin-A and -B are not functionally redundant and syncytin-B may be indispensable for the development of the SynT-II layer. In addition, as with human syncytin-2, the ectodomain of the syncytin-B transmembrane subunit is also immunosuppressive (Mangeney et al. 2007). Further investigation is warranted to elucidate the physiological functions of syncytin-B in the mouse placenta. Endogenous retroviral envelope genes encoding placental fusogenic proteins have also been identified from the genomes of other mammals. For example, a rabbit syncytin gene named syncytin-Ory1 has recently been reported after mining the rabbit genome (Heidmann et al. 2009). In addition, an envelope gene of endogenous Jaggsiekte sheep retroviruses (enJSRVs) is crucial for ovine placental development because antisense oligonucleotides targeting the enJSRV envelope gene blocked trophoectoderm outgrowth and trophoblast giant binucleate cell differentiation (Dunlap et al. 2006).
5.2 Structure and Functional Studies of Syncytins Both syncytin-1 and -2 genes encode a polypeptide of 538 amino acids, whereas syncytin-A and -B genes encode a polypeptide of 617 and 618 amino acids, respectively (for sequence alignment, see Fig. 5.2a). Like the conventional retroviral envelope proteins, the syncytin polypeptide is also composed of functional domains and motifs that are required for conformational integrity, interaction with receptors on target cell surface, and mediation of membrane fusion.
5
Syncytins: Molecular Aspects
123
Fig. 5.2 Human and murine syncytins. (a) Protein sequence alignment of syncytin-1, -2, -A, and -B. Sequence alignment is based on MULTALIN (http://npsa-pbil.ibcp.fr/cgibin/align_multalin.pl). The CX2 C and CX6 CC motifs for disulfide bond formation are underlined. Of note, the position of CX2 C motif in syncytin-1 is away from the N-terminus of syncytin-1 polypeptide. The furin cleavage site is boxed. Conserved residues in all of four syncytins are shown in bold. Numbers on the right indicate the numbering of the last residue in each line of the indicated syncytin protein sequence. (b) Schematic representation of the domain structures of syncytins. SP, signal peptide; FP, fusion peptide; TMD, transmembrane domain. (c) Model of cell–cell fusion mediated by syncytin-1 and -2. The SU subunits of the matured syncytin-1 and -2 proteins interact with their cognate receptors, ASCT2 and MFSD2, respectively, on the target membrane. It is generally believed that the interaction induces a conformational change to expose the fusion peptide in the TM subunit and facilitate its insertion into the target membrane. An intramolecular interaction between HRA (A) and HRB (B) in the TM subunit brings the two membranes closer together to fuse with each other. Note that only one mature protein of the syncytin trimer is depicted. Topology analysis of ASCT2 and MFSD2 is based on TOPCONS (http://topcons.cbr.su.se)
5.2.1 Biosynthesis of Syncytins Syncytin polypeptides are post-translationally cleaved into two subunits, surface (SU) and transmembrane (TM) subunits during biosynthesis (Chang et al. 2004, Chen et al. 2008, Cheynet et al. 2005). Furin is a member of the proprotein convertase family of serine proteases, including PC1/3, PC2, and PC4, to name a few. It is generally believed that furin is involved in the cleavage reaction because treatment of furin inhibitor I blocks cleavage of both syncytin-1 and -2, and consequently cell–cell fusion mediated by both proteins (Chen et al. 2008). The consensus furin cleavage site is positioned after the C-terminal R residue in the sequence R/K-R or R-X2-R. Accordingly, the R-N-K-R and R-V-R-R sequences in syncytin1 and -2, respectively, and the R-R-K-P and R-P-K-R sequences in syncytin-A
124
H. Chen and M.-L. Cheong
and -B, respectively, are the predicted furin cleavage sites critical for generation of functional proteins (Fig. 5.2a, b). Indeed, mutation of the R-N-K-R motif in syncytin-1 into A-A-A-R generates mutant syncytin-1 proteins on the cell surface with no fusion activity, supporting the idea that proteolytic cleavage is essential for biosynthesis of functional syncytin-1 protein (Cheynet et al. 2005). When transiently expressed in the furin-defective LoVo cells, the processing of syncytin-1 and -2 polypeptides was not efficient enough to confer cell–cell fusion. However, enhanced processing of both syncytin precursor proteins and significant cell–cell fusion were detected when functional furin was coexpressed in LoVo cells (Chen et al. 2008). Therefore, this complementation study also demonstrates that furin is a major protease in the conversion of syncytin proproteins into mature fusogenic proteins. Moreover, cell–cell fusion events may take place only when a sufficient level of mature fusogenic protein is present on the cell surface. Transient expression experiments have been performed to study the biosynthesis of syncytin-1 in BeWo cells. This study demonstrates that syncytin-1 is subjected to N-glycosylation, which is critical for generation of functional syncytin-1 protein. This notion is also supported by the finding that syncytin-1 proteins were absent on the cell surface of transfected BeWo cells after treatment with tunicamycin, an inhibitor of N-glycosylation (Cheynet et al. 2005). Furthermore, syncytin-1 proteins undergo trimerization during biosynthesis, which may be facilitated by an N-terminal heptad repeat region in the ectodomain of the TM subunit (see Section 5.2.4).
5.2.2 Functional Domains and Motifs in Syncytins The functional domains and motifs and structure-functional relationships in syncytins have been identified and characterized by linker-scanning, site-directed, and deletion mutagenesis of syncytin-1 and -2 proteins. Random in-frame insertions of a five-amino acid linker into the syncytin-1 backbone indicates that the SU subunit and the ectodomain of TM subunit maintain a rigid conformation, which does not allow linker insertion in almost all sites tested. Exceptionally, Ser51, Val139, and Glu156 in the SU subunit and the CTM domain have a flexible enough conformation to allow linker insertion that does not abolish the fusogenicity of mutants. Deletion of the CTM domain of syncytin-1, but not syncytin-2, further enhanced its fusogenicity (Chang et al. 2004, Chen et al. 2008). A peptide region of 16 amino acids in the CTM of murine leukemia virus (MLV) envelope protein, termed R-peptide, has been demonstrated to inhibit its fusogenicity. Deletion of this R-peptide in the MLV envelope protein caused extensive fusion by formation of large syncytia. In fact, the MLV R-peptide is cleaved by a viral protease during virion maturation (Yang and Compans 1997). Apart from MLV, R-peptide cleavage has been demonstrated in other viruses such as Mason-Pfizer monkey virus and spleen necrosis virus, which are type D and C retroviruses, respectively (Bobkova et al. 2002). Since syncytin-1 has a longer CTM domain than syncytin-2, the possibility of an R-peptide-like region in the CTM domain of syncytin-1 cannot be ruled out.
5
Syncytins: Molecular Aspects
125
Critical domains for regulation of cell–cell fusion in the ectodomain of syncytin TM subunit include a fusion peptide at the N-terminus of the ectodomain and two heptad repeat regions (HRA and HRB) downstream of the fusion peptide. The fusion peptide is believed to insert into the target membrane, followed by intramolecular interaction between the two heptad repeats, to facilitate membrane fusion (see Section 5.2.4). This notion is supported by the fact that peptides derived from the heptad repeat region of an envelope protein are able to block its fusogenic activity. For example, synthetic peptides derived from the HRB of syncytin-1 and the HRA of syncytin-A have been shown to specifically inhibit cell–cell fusion mediated by syncytin-1 and -A, respectively (Chang et al. 2004, Peng et al. 2007). It is generally believed that the processed SU and TM subunits of an envelope protein initially form an unactivated complex on the cell surface. When the SU subunit binds its cognate receptor, a conformational change may occur to allow the TM subunit to resume its fusion activity. In this regard, formation and rearrangement of disulfide bond between a disulfide isomerase motif, CX2 C, in the SU subunit and a CX6 CC motif in the TM subunit are required for activation of TM after the interaction between SU and its receptor. For the unactivated MLV envelope protein, a disulfide bond is formed between the CX2 C motif and the last cysteine residue in the CX6 CC motif. Interestingly, receptor binding induces isomerization of the disulfide bond between the SU and TM subunits to form a disulfide bond in the CX2 C motif of SU (Wallin et al. 2004). As a result, the fusogenicity of TM is activated to promote membrane fusion via the fusion peptide and the heptad repeat regions (see Section 5.2.4). Because syncytin proteins contain the CX2 C and CX6 CC motifs (Fig. 5.2a, b), it is highly possible that formation and isomerization of a disulfide bond between CX2 C and CX6 CC is essential for the biosynthesis and fusion function of syncytin proteins. Indeed, mutagenesis of the last cysteine of the CX6 CC motif in syncytin-1 into alanine abolished its fusogenic activity (Cheynet et al. 2005). Furthermore, mutagenesis of the cysteine residues in the CX2 C motifs of syncytin-1 and -2 also abolished their fusogenic activities (Chen et al. 2008). Importantly, the syncytin-1 and -2 CX2 C mutants imposed a dominant negative effect on their wild-type counterparts. Moreover, mutation of the first cysteine in the CX2 C motif (C186 in syncytin-1 and C43 in syncytin-2) had a stronger dominant negative effect than mutation of the second cysteine (C189 in syncytin-1 and C46 in syncytin-2) (Chen et al. 2008). These observations suggest that the C186 and C43SU mutants may compete effectively with the wild-type SU for association with TM during biosynthesis. It is also possible that the C186SU-TM and C43SU-TM complexes fail to interact with the cognate receptors for syncytin-1 and -2 or undergo proper isomerization of disulfide bond in order to activate the fusion activity of TM. On the other hand, domain swapping experiments indicate that the SU and TM subunits of syncytin-1 and -2 are not interchangeable in terms of fusion activity. The positions of CX2 C motifs in the SU subunits of syncytin-1 and -2 are relatively different, i.e. in the middle of the syncytin-1 SU subunit and close to the N-terminus of the syncytin-2 SU subunit. It is very likely that swapping the SU and TM subunits between syncytin-1 and -2 affects the formation and isomerization of disulfide
126
H. Chen and M.-L. Cheong
bonds that are essential for biosynthesis and fusion function of syncytin-1 and -2. This speculation is further substantiated by the observation that the syncytin-1 C186 mutant had no effect on inhibition of biosynthesis and fusogenicity of syncytin-2, and neither did the syncytin-2 C43 mutant on syncytin-1 (Chen et al. 2008).
5.2.3 Syncytin Receptors Envelope-mediated membrane fusion first requires an interaction between envelope and its cognate receptor on the target membrane. Receptors for syncytin-1 and -2 have been identified. Because envelope proteins sharing sequence homology may use the same cellular receptor, it is reasonable to speculate that syncytin-1 may interact with the same receptor for those retroviral envelope proteins with sequence similarity to syncytin-1. Syncytin-1 shares sequence homology with a large interference group of retroviruses that use a type D mammalian retrovirus receptor (RDR) as their common receptor. RDR is a synonym for human sodium-dependent neutral amino acid transporter type 2 (ASCT2). Therefore, ASCT2 is a candidate receptor for syncytin-1. This speculation can be tested by receptor interference assays for the fusogenicity of syncytin-1 in an array of TE671 human cells expressing a panel of retroviral envelope proteins. If syncytin-1 binds ASCT2, cell–cell fusion mediated by syncytin-1 will be blocked in TE671 cells expressing an envelope protein that also uses ASCT2 as its receptor. This is the case and supports ASCT2 as the syncytin-1 receptor (Blond et al. 2000). Interestingly, functional studies further indicate that syncytin-1 is able to induce cell–cell fusion of CHO cells expressing ASCT2 or a related transporter ASCT1 (Lavillette et al. 2002). These observations indicate that both ASCT2 and ASCT1 are the functional receptors for syncytin-1. Structure-function studies further identified the ASCT2-binding domain in syncytin-1. Soluble full-length SU proteins were compared with the N- and C-terminal truncation mutants for binding to cells expressing ASCT2. Accordingly, a region consisting of the N-terminal 124 amino acids of the mature SU subunit was found to be the minimal receptor-binding domain of syncytin-1 (Cheynet et al. 2006). Identification of syncytin-2 receptor was much more labor-intensive. Complementation experiments were conducted to identify the cellular receptor for syncytin-2, based on the principle that cells originally refractory to syncytin-2 pseudotypes will become susceptible when a piece of human chromosome carrying the syncytin-2 receptor gene in introduced. The Chinese hamster fibroblast A23 cell line is not susceptible to syncytin-2 pseudotypes. Accordingly, a panel of human/hamster hybrid cell lines harboring different human chromosomal fragments was screened for their susceptibility to syncytin-2 pseudotypes. This screen mapped the candidate syncytin-2 receptor to chromosome 1p34.2. Functional characterization of candidate genes in this chromosome region identified the major facilitator superfamily domain containing 2 (MFSD2) as the receptor for syncytin-2 (Esnault et al. 2008). In terms of sequence and structural similarities, MFSD2, like the bacterial permeases and symporter proteins, belongs to the 10–12
5
Syncytins: Molecular Aspects
127
transmembrane domains protein superfamily. Of note, two alternatively spliced variants of human MFSD2 have been identified, with the variant 1 containing extra 13 amino acids in the first extracellular loop. Nevertheless, cell fusion assays for both MFSD2 variants do not reveal a significant difference in cell–cell fusion mediated by syncytin-2 (H.C. unpublished results). Human MFSD2 gene is highly expressed in placenta and its transcript level is upregulated in BeWo cells in response to forskolin. In situ hybridization indicates that MFSD2 gene is primarily expressed in the placental syncytiotrophoblast layer (Esnault et al. 2008). Because syncytin-2 transcripts are expressed in cytotrophoblasts, this suggests that syncytin2-expressing cytotrophoblasts are restricted to fuse into the syncytiotrophoblast layer.
5.2.4 Mechanism of Membrane Fusion The molecular mechanisms underlying the cell–cell fusion mediated by syncytin-1 and -2 are currently unknown. Nevertheless, structural studies of influenza hemagglutinin (HA) and human immunodeficiency virus (HIV) envelope protein (gp160) have provided important insights into the fusion processes mediated by retroviral envelopes. HA and gp160 proteins are arranged into a homotrimer conformation after synthesis. The HA polypeptide is cleaved into HA1 and HA2 subunits, whereas the gp160 polypeptide is cleaved into gp120 and gp41 subunits. After cleavage, the fusion peptide located at the N-terminus of HA2 and gp41 is shielded from exposure in a pocket. The fusion peptide springs out towards the target membrane when HA2 undergoes a conformational change at acidic pH in endosomes or when gp120 interacts with its cognate receptor to induce a conformational change in gp41. Subsequently, two heptad repeat regions (HRA and HRB) in HA2 and gp41, whose primary sequences are predicted to be extended amphipathic α-helices, form coiledcoils and bring the host and target membranes closer together to facilitate membrane fusion (Colman and Lawrence 2003, Frey et al. 2006). By analogy with the working model for HA and gp160, interaction between the SU subunit of syncytin and the syncytin receptor may facilitate isomerization of disulfide bonds in SU and TM subunits and cause a conformational change such that SU is released and the fusion peptide in the ectodomain of the TM subunit is exposed to insert into the target membrane. Moreover, HRA and HRB in the ectodomain of TM subunits interact with each other and bring together the apposed membranes to facilitate membrane fusion (Fig. 5.2c). As mentioned above, integration of syncytin-2-containing HERV-FRD into the primate genome occured at an early stage of primate evolution. Moreover, sequence analysis of the syncytin-2 gene demonstrates limited polymorphism in humans and high sequence identity among primate species. Theses observations suggest that the sequence of syncytin-2 gene is well conserved and may be very close to the ancestral gene. Therefore, structural analysis of syncytin-2 may help to reveal the structure of the oldest mammalian retroviral envelope protein. A central domain of 54 amino acids in the ectodomain of the syncytin-2 TM subunit, which primarily contains the
128
H. Chen and M.-L. Cheong
HRA, has been subjected to crystallographic analysis. The solution of the structure of this domain discloses a trimeric conformation with three α-helices associated with each other, which may be involved in oligomerization of syncytin-2 during biosynthesis. In addition, the backbone structure of this central domain is very similar to that of present-day retroviruses such as MLV and the human T-cell lymphotropic virus, given that the distribution of surface charges and peptide sequence are different (Renard et al. 2005). Therefore, structural similarity is well preserved in retroviral envelope proteins, and they may all undergo a similar fusion process.
5.3 Regulation of Syncytin Expression The human syncytiotrophoblast layer undergoes apoptosis and sheds into the maternal circulation. Therefore, replenishment of new syncytiotrophoblast layer is required to maintain the functional integrity of placental villi. As syncytin-1 and -2 are crucial for formation of syncytiotrophoblast layer, tight regulation of syncytin gene expression is imperative to maintain proper placental development and function.
5.3.1 GCM1 Regulation of Syncytin-1 and -2 Gene Expression The underlying mechanism for placenta-specific expression of syncytin-1 and -2 has been investigated at transcriptional and epigenetic levels. GCM (glial cell missing) was originally isolated from a fly mutant line that produces additional neurons at the expense of glial cells (Hosoya et al. 1995, Jones et al. 1995). Two GCM-like genes (GCM1 and GCM2) have been reported in mouse, rat, and human. GCM proteins share sequence homology in the N-terminal region that constitutes a DNAbinding domain called the GCM motif, which has a preferred binding sequence of 5 -(A/G)CCC(T/G)CAT-3 or its complement (Akiyama et al. 1996, Schreiber et al. 1998). Moreover, two zinc ions tightly coordinated by cysteine and histidine residues in the DNA-binding domain of GCM1 are essential for its DNA-binding activity (Cohen et al. 2003). Sequence homology is less preserved outside the GCM motif; a transactivation domain has been identified in the C terminus of GCM proteins. Ontogeny of murine GCM1 demonstrates that GCM1 gene is expressed in a subset of trophoblast cells at the chorionic plate at E7.5-8. When chorioallantoic attachment takes place at E8.5, expression of GCM1 is detected in clusters of trophoblast cells of the chorioallantoic surface, where invaginations begin, and this is accompanied by development of fetal blood vessels to form the labyrinth (Basyuk et al. 1999, Stecca et al. 2002). It is evident that GCM1 is required for this branching morphogenesis during labyrinthine development as these events are impaired in GCM1-knockout mice. Moreover, GCM1-knockout trophoblasts fail to undergo cell–cell fusion to form syncytiotrophoblasts (Anson-Cartwright et al. 2000, Schreiber et al. 2000). Expression of human GCM1 transcripts can be detected
5
Syncytins: Molecular Aspects
129
in the cytotrophoblasts and syncytiotrophoblasts of term placenta as well as in villous sprouts and cytotrophoblast cell columns of first trimester placenta, further supporting an important role for GCM1 in human placental development (Baczyk et al. 2004, Nait-Oumesmar et al. 2000). The studies of GCM1-knockout mice suggest that GCM1 may control trophoblastic fusion by regulation of syncytin-1 gene expression at the transcriptional level. Indeed, GCM1-binding sites (GBSs) have been identified upstream of the 5 -LTR of syncytin-1 promoter and are required for transactivation of syncytin-1 promoter activity by GCM1. Moreover, overexpression of GCM1 stimulates cell– cell fusion in BeWo and JEG-3 cells (Yu et al. 2002). Functional GBSs have been identified in the 5 -LTR of syncytin-2 promoter and the MFSD2 promoter, and these are also responsive to GCM1. In addition, ectopic expression of GCM1 stimulates expression of syncytin-2 and MFSD2 transcripts in BeWo cells and MCF-7 breast cancer cells (H.C. unpublished results). Correspondingly, functional assays further demonstrate that silencing GCM1 expression by RNA interference or antisense oligonucleotides suppresses BeWo cell fusion and inhibits syncytiotrophoblast formation, but promotes proliferation of cytotrophoblasts in villous explants (Baczyk et al. 2009). On the other hand, studies in GCM1-knockout mice and GCM1knockout trophoblast stem cells indicate that GCM1 is involved in regulation of syncytin-B, but not syncytin-A, gene expression (Simmons et al. 2008). Overall, these observations clearly show that GCM1 is a critical for syncytin gene expression and is required for the differentiation of human trophoblast cells.
5.3.2 Regulation of GCM1 Activity Because the cAMP/PKA signaling pathway regulates syncytin-1 and -2 expression and human placental cell fusion, it is reasonable to speculate that GCM1 is a downstream effector of the cAMP/PKA signaling pathway in stimulation of placental cell fusion. Indeed, GCM1 activity is regulated by cAMP/PKA signaling pathway at both transcriptional and post-translational levels. Elevation of cAMP level and activation of PKA increase the GCM1 transcript level in BeWo cells (Knerr et al. 2005). In addition, activation of cAMP/PKA signaling enhances the association between GCM1 and CBP, a transcriptional coactivator with histone acetyltransferase activity. As a result, CBP acetylates GCM1 at Lys367, Lys406, and Lys409 to prevent GCM1 from ubiquitination and thereby increase GCM1 stability. CBP also enhances the transcriptional activity of GCM1 and colocalizes with GCM1 at the syncytin-1 promoter (Chang et al. 2005). As the syncytiotrophoblast layer is a highly dynamic structure, its maintenance as a steady-state structure during pregnancy is crucial. Accordingly, GCM1 activity is subjected to tight regulation in order to maintain proper trophoblastic fusion in the placenta. In this regard, GCM1 is a labile protein with a half-life of about 90–120min and is subject to degradation by the ubiquitin-proteasome system (Yang et al. 2005). Ubiquitination of target proteins involves a cascade of reactions, i.e. (1) activation of ubiquitin by an E1 ubiquitin-activating enzyme in an ATP-dependent
130
H. Chen and M.-L. Cheong
process, (2) transfer of activated ubiquitin to an E2 ubiquitin-conjugating enzyme, and (3) transfer of ubiquitin from E2 to a substrate protein with or without the aid of an E3 ubiquitin-protein isopeptide ligase. The SCF E3 ligase is a complex composed of Skp1, Cullin1, Rbx1, and a substrate-recognition F-box protein. A screening of F-box proteins expressed in placenta has identified FBW2 as the functional F-box protein for mediation of GCM1 ubiquitination and degradation (Yang et al. 2005). Recognition of GCM1 by FBW2 primarily depends on phosphorylation of Ser322 in GCM1, which is mediated by GSK-3β (Chiang et al. 2009). Of interest, acetylation of GCM1 at Lys367, Lys406, and Lys409 by CBP protects GCM1 from ubiquitination. Given that there are 28 Lys residues in GCM1, just how acetylation of the three Lys residues prevents the remaining Lys residues in GCM1 from ubiquitination remains an interesting question. An additional mode of regulation of GCM1 activity is sumoylation of GCM1 Lys156, an activity mediated by Ubc9, the E2 component of the sumoylation machinery. Conjugation of SUMO-1 to Lys156 impedes the DNA-binding activity of GCM1, and therefore downregulates GCM1 activity in placenta (Chou et al. 2007). Sumoylation of endogenous GCM1 can be detected in JAR placental cells. At present, it is not clear whether sumoylation of GCM1 is a constitutive modification or is regulated by upstream signaling pathways. Overall, regulation of GCM1 activity is subjected to fine-tuning by different types of post-translational modification, which may be crucial for the functional and structural integrity of the syncytiotrophoblast layer.
5.3.3 Epigenetic Regulation of Syncytin-1 and -2 Gene Expression Epigenetic regulation of gene expression involves modification of cytosines in the context of CpG dinucleotides, i.e. methylation at the C5 of the cytosine pyrimidine ring. Methylated cytosines recruit methyl-cytosine-binding proteins, HDACs, and transcriptional corepressors to silence gene expression. Epigenetic regulation of syncytin-1 and syncytin-2 plays an important role in maintaining the placentaspecific expression pattern of both genes. Analysis of the CpG methylation pattern in the U3 region of the 5 -LTR of syncytin-1 promoter reveals hypermethylated CpGs in non-placental cells such as skin fibroblasts, HeLa cells, and breast carcinoma cells (Matouskova et al. 2006). Instead, hypomethylation of these GpGs is detected in placental tissues and BeWo cells. Of interest, treatment of HeLa cells with 5 -azacytidine and trichostatin A, which are inhibitors of DNA methylation and histone acetylation, respectively, does not significantly affect the methylation pattern of these CpGs, not to mention stimulation of syncytin-1 expression (Matouskova et al. 2006). Likewise, CpGs in the 5 -LTR of syncytin-2 promoter are hypermethylated in non-placental cells, but hypomethylated in placental tissues and BeWo cells (Gimenez et al. 2009). However, these methylated CpGs are less resistant to treatment of 5 -azacytidine and trichostatin A in non-placental cells (H.C. unpublished results). It is very likely that CpG hypomethylation in the 5 -LTRs of syncytin-1 and -2 promoters may facilitate GCM1-mediated transcriptional activation of syncytin-1
5
Syncytins: Molecular Aspects
131
and -2 in placenta. Interestingly, ectopic expression of GCM1 in MCF7 breast cancer cells promotes CpG demethylation in the 5 -LTR of syncytin-2 promoter, and therefore stimulates syncytin-2 expression and cell fusion (H.C. unpublished results). A possible mechanism underlying CpG hypomethylation in syncytin-2 promoter by GCM1 may involve active DNA demethylation. Because a GBS is within the 5 -LTR of syncytin-2 promoter, it is possible that GCM1 may recruit proteins with demethylation activity to promote CpG demethylation and then transactivate syncytin-2 gene expression.
5.4 Syncytins and Disease Expression of endogenous retroviral sequences in host species can be advantageous by induction of resistance to exogenous retrovirus infection or detrimental by causing cancer or contribution to disease development, as in multiple sclerosis and schizophrenia (Griffiths 2001). As syncnytin-1 and -2 are specifically expressed in placenta and are essential for placental function, decreased expression of syncytin-1 and -2 is found in placental disorders. Unexpectedly, aberrant expression of syncytin-1 is also found in multiple sclerosis and breast and colorectal cancers (Antony et al. 2004, Bjerregaard et al. 2006, Larsen et al. 2009).
5.4.1 Syncytins in Placental Disorders The proper interaction and transformation process between maternal decidua, myometrium, and fetal trophoblasts is essential for a successful pregnancy outcome. Abnormal placentation incurring insufficient oxygen and nutrient exchange has been reported in pregnancy disorders including preeclampsia, HELLP (hemolysis, elevated liver enzymes, and low platelets) syndrome, and fetal growth restriction. Preeclampsia is a major pregnancy-specific disorder affecting around 5% of pregnancies and is one of the major causes of maternal deaths and stillbirths worldwide. The diagnostic criteria of preeclampsia are new-onset or aggravated hypertension and proteinuria in the latter half of a pregnancy. Besides the high mortality, clinical impacts include cerebral or visual disturbances, oliguria, pulmonary edema, HELLP syndrome, fetal growth restriction, and grand mal seizures. Although the etiologic factors of preeclampsia are currently unknown, shallow trophoblast invasion and insufficient maternal vascular remodeling are observed in preeclamptic placentae (Maynard et al. 2008). It is thought that these defects impair the development of the fetal-maternal vasculature and result in placental ischemia and hypoxia, which then contributes to the pathogenesis of preeclampsia in the late second or third trimester of pregnancy. Recently, decreased expression of syncytin-1 and -2 has been reported in preeclampsia. A possible mechanism for this decreased expression in preeclampsia is disruption of the GCM1 transcription network following placental hypoxia. The activity of GSK-3β is elevated in preeclamptic placentae and hypoxic placental cells due to inhibition of the upstream
132
H. Chen and M.-L. Cheong
Fig. 5.3 Regulation of gene expression of syncytin-1 and -2 by GCM1. GCM1-responsive elements have been identified upstream of the 5 -LTR of syncytin-1 promoter and within the 5 -LTR of syncytin-2 promoter. GCM1 is a labile protein subject to ubiquitination (Ub) via FBW2, which recognizes Ser322-phosphorylated GCM1. Phosphorylation of Ser322 is mediated by GSK-3β of which activity can be suppressed by the PI-3K/Akt signaling pathway. In preeclampsia and in hypoxic placental cells, the PI-3K/Akt signaling pathway is inhibited, leading to elevation of GSK3β activity, and therefore a promotion of GCM1 degradation. GCM1 protein is stabilized in the cAMP/PKA signaling pathway by enhancement of GCM1 acetylation (Ac) by CBP. Deacetylation of GCM1 can be mediated by HDAC3. Furthermore, GCM1 is sumoylated in placental cells and conjugation of SUMO-1 (SU-1) at Lys156 in the GCM motif impedes the DNA-binding activity of GCM1. Note that GCM1 phosphorylation by GSK-3β and PKA may take place in the cytosol
PI-3K/Akt signaling pathway. As a result, more GCM1 proteins are subjected to Ser322 phosphorylation by GSK-3β, leading to GCM1 ubiquitination and degradation (Fig. 5.3). Concomitantly, syncytin-1 and -2 expression is downregulated and this may destabilize the structure of the syncytiotrophoblast layer in preeclampsia.
5.4.2 Syncytin-1 in Malignancies The most common form of breast cancer is ductal carcinoma originating from the inner lining of mammary ducts or the lobules that supply the ducts. Although the real mechanism is not clear, breast cancer cells can fuse with normal host cells in vitro and in vivo. When cancer cells migrate, they invade adjacent tissues and are prone to approach and fuse with the endothelium. Human breast cancer cell lines derived from ductal adenocarcinoma and part of ductal adenocarcinoma tissues have recently been discovered to express syncytin-1, which may function
5
Syncytins: Molecular Aspects
133
as a mediator to regulate cancer-endothelial fusion (Bjerregaard et al. 2006). The staining ratio of syncytin-1 in breast cancer tissues is positively correlated with the recurrence-free survival. Here syncytin-1 acts as a good prognostic factor. On the other hand, the expression ASCT2 in endothelial cells also increases in breast cancer tissues, but does not have a significant impact on overall or recurrence-free survival (Larsson et al. 2007). Fusions may occur when syncytin-1-expressing cancer cells make contact with ASCT2-expressing host endothelial cells. Under these circumstances, the host immune system and tumor suppressor genes may be triggered in the endothelium to eliminate the cancer cells. Colorectal cancer is the fourth most common form of cancer and the third leading cause of cancer-related death worldwide. Using a monoclonal antibody against syncytin-1 and -2, overexpression of syncytins was found to be associated with decreased survival for rectal, but not colon cancer (Larsen et al. 2009). These data suggest that syncytin expression might show different prognostic impacts in different types of tumor. Endometrial carcinoma is another major cancer in women, mostly affected at perimenopausal and postmenopausal period. Histologically, endometrioid carcinoma is the most common subtype of endometrial carcinoma, a cancer which often expresses both estrogen and progesterone receptors and is hormone-dependent. Cell–cell fusions can be detected in endometrioid carcinoma tissues by histological examination. One study (Strick et al. 2007), showed upregulated syncytin-1 expression in all benign (polyp, hyperproliferative stage, and hyperplasia) and malignant endometrial tissues, with the most significant change in endometrial carcinoma. Furthermore, syncytin-1 expression was stimulated by estrogen, progesterone, or cAMP in primary cells and cell lines from endometrial carcinoma. Such an increase in syncytin-1 gene expression may be associated with the transformation phenotype of anchorage independence, because knocking down the expression of syncytin 1 by RNA interference reduced the colony number and size in soft agar assay (Strick et al. 2007). In this study, however, statistical analysis of the prognostic impact of syncytin-1 on survival in endometrial carcinoma was not presented.
5.4.3 Syncytin-1 in Neurological Diseases Multiple sclerosis (MS) is primarily an inflammatory disorder of the brain and spinal cord in which focal lymphocytic infiltration leads to damage of myelin and axons resulting in demyelination. In most patients, clinical manifestations indicate the involvement of motor, sensory, visual, and autonomic systems (Compston and Coles 2008). A recent study demonstrated that approximately 3 times as much syncytin-1 is present in brain tissues of MS patients than the controls (Antony et al. 2004). In addition, expression of syncytin-1 in human fetal astrocytes resulted in a significant increase of proinflammatory cytokine IL-1β. The culture medium from the above condition contained redox reactants that were highly cytotoxic to human oligodendrocytes. In a mouse model of MS, syncytin-1-mediated neuroinflammation and death of oligodendrocytes were prevented by the antioxidant ferulic acid, indicating
134
H. Chen and M.-L. Cheong
syncytin-1 may be involved in the pathogenesis of active demyelination through redox reactant-mediated cellular damage in the brain. A follow-up study further demonstrated that syncytin-1 induces ER stress and the expression of old astrocyte specifically induced substance (OASIS), an ER stress sensor, in cultured astrocytes. Expression of OASIS leads to downregulation of ASCT1 expression, which may cause oligodendrocyte injury and death through dysregulated transport of trophic and toxic amino acids (Antony et al. 2007). The pathologic role of syncytin-1 has also been studied in the patients with motor neuron diseases, neurological disorders that progressively destroy cells controlling essential muscle activity such as speaking, swallowing, breathing, walking, and general movement. The levels of syncytin-1 transcript were significantly elevated in biopsies from the most affected muscles compared to the control tissues (Oluwole et al. 2007). Nevertheless, whether the role of syncytin-1 in the pathophysiology of motor neuron diseases is similar to that in MS needs further investigation.
5.5 Conclusion The multinucleated syncytiotrophoblast layer is a specialized epithelium lining the outer surface of the placental villus and is responsible for placental transport and hormone production during human pregnancy. Trophoblastic fusion is essential for differentiation of syncytiotrophoblast, which is mediated by syncytin-1 and -2 derived from HERV envelope genes and is regulated by GCM1 transcription factor and its upstream kinases such as PKA and GSK-3β. Recent studies have identified syncytin-like proteins in mouse and rabbit. Importantly, a transgenic study demonstrated that mouse syncytin-A is essential for trophoblastic fusion and syncytiotrophoblast differentiation, confirming that syncytin-like proteins are also required for placental development in other mammals. Being fusogenic proteins for trophoblastic fusion, human syncytin-2 and mouse syncytin-B possess some immunosuppressive activity via the ectodomains in their TM subunits. This additional function suggests that syncytin-2 and -B may contribute to immune tolerance of the developing embryo, though further studies are needed. Sequence comparison and phylogenetic analysis suggest that human and mouse syncytin genes are not related and were independently adopted by these species during evolution. Nevertheless, functional domains and motifs characteristic of a retroviral envelope protein have been identified in both human and mouse syncytins and these are essential for their fusogenic activities. Therefore, the fusion process mediated by syncytins may be similar to the known processes for influenza HA and HIV gp160. As the integrity of the syncytiotrophoblast layer is maintained by a dynamic balance between ongoing apoptosis and syncytial fusion, tight regulation of syncytin gene expression is imperative. Interestingly, expression of both syncytin-2 and its receptor MFSD2 genes are regulated by GCM1, and this may be a positive evolutionary selection in coordination of syncytin-2 functions in placenta. Clinically, decreased expression of syncytin-1 and -2 is detected in preeclampsia, whereas syncytin-1 overexpression is found in many other diseases. Defective
5
Syncytins: Molecular Aspects
135
placentation may impair GCM1 activity such that syncytin gene expression is downregulated contributing to the development of preeclampsia. In multiple sclerosis, overexpression of syncytin-1 in neurons and glial cells incurs an inflammatory response, ER stress, and reduced transporter activity that damage cells in the central nervous system. As the study of syncytin continues, new findings about the pathological roles of syncytins in human diseases and cancers are anticipated, which may provide new therapeutic avenues against the diseases. In addition, understanding the immunosuppressive roles of syncytin-2 and -B during pregnancy may significantly improve the success rate of pregnancy. Acknowledgments We thank Dr. Tso-Pang Yao for critical reading of this manuscript. Work in Hungwen Chen’s laboratory is supported by grants from National Science Council of Taiwan (grant 96-2311-B-001-034) and Academia Sinica of Taiwan.
References Akiyama Y, Hosoya T, Poole AM et al (1996) The gcm-motif: a novel DNA-binding motif conserved in Drosophila and mammals. Proc Natl Acad Sci USA 93:14912–14916 Anson-Cartwright L, Dawson K, Holmyard D et al (2000) The glial cells missing-1 protein is essential for branching morphogenesis in the chorioallantoic placenta. Nat Genet 25:311–314 Antony JM, Ellestad KK, Hammond R et al (2007) The human endogenous retrovirus envelope glycoprotein, syncytin-1, regulates neuroinflammation and its receptor expression in multiple sclerosis: a role for endoplasmic reticulum chaperones in astrocytes. J Immunol 179:1210–1224 Antony JM, van Marle G, Opii W et al (2004) Human endogenous retrovirus glycoproteinmediated induction of redox reactants causes oligodendrocyte death and demyelination. Nat Neurosci 7:1088–1095 Baczyk D, Drewlo S, Proctor L et al (2009) Glial cell missing-1 transcription factor is required for the differentiation of the human trophoblast. Cell Death Differ 16:719–727 Baczyk D, Satkunaratnam A, Nait-Oumesmar B et al (2004) Complex patterns of GCM1 mRNA and protein in villous and extravillous trophoblast cells of the human placenta. Placenta 25:553–559 Barker DJ (2004) Developmental origins of adult health and disease. J Epidemiol Community Health 58:114–115 Basyuk E, Cross JC, Corbin J et al (1999) Murine Gcm1 gene is expressed in a subset of placental trophoblast cells. Dev Dyn 214:303–311 Benirschke K, Kaufmann P (2001a) Architecture of normal villous trees In: Pathology of the human placenta, 4th edn. Springer, New York, NY Benirschke K, Kaufmann P (2001b) Early development of the human placenta. In: Pathology of the human placenta, 4th edn. Springer, New York, NY Bjerregaard B, Holck S, Christensen IJ et al (2006) Syncytin is involved in breast cancerendothelial cell fusions. Cell Mol Life Sci 63:1906–1911 Blaise S, de Parseval N, Benit L et al (2003) Genomewide screening for fusogenic human endogenous retrovirus envelopes identifies syncytin 2, a gene conserved on primate evolution. Proc Natl Acad Sci USA 100:13013–13018 Blond JL, Beseme F, Duret L et al (1999) Molecular characterization and placental expression of HERV-W, a new human endogenous retrovirus family. J Virol 73:1175–1185 Blond JL, Lavillette D, Cheynet V et al (2000) An envelope glycoprotein of the human endogenous retrovirus HERV-W is expressed in the human placenta and fuses cells expressing the type D mammalian retrovirus receptor. J Virol 74:3321–3329 Bobkova M, Stitz J, Engelstadter M et al (2002) Identification of R-peptides in envelope proteins of C-type retroviruses. J Gen Virol 83:2241–2246
136
H. Chen and M.-L. Cheong
Chang C, Chen PT, Chang GD et al (2004) Functional characterization of the placental fusogenic membrane protein syncytin. Biol Reprod 71:1956–1962 Chang CW, Chuang HC, Yu C et al (2005) Stimulation of GCMa transcriptional activity by cyclic AMP/protein kinase A signaling is attributed to CBP-mediated acetylation of GCMa. Mol Cell Biol 25:8401–8414 Chen CP, Chen LF, Yang SR et al (2008) Functional characterization of the human placental fusogenic membrane protein syncytin 2. Biol Reprod 79:815–823 Cheynet V, Oriol G, Mallet F (2006) Identification of the hASCT2-binding domain of the Env ERVWE1/syncytin-1 fusogenic glycoprotein. Retrovirology 3:41 Cheynet V, Ruggieri A, Oriol G et al (2005) Synthesis, assembly, and processing of the Env ERVWE1/syncytin human endogenous retroviral envelope. J Virol 79:5585–5593 Chiang MH, Liang FY, Chen CP et al (2009) Mechanism of hypoxia-induced GCM1 degradation: implications for the pathogenesis of preeclampsia. J Biol Chem 284:17411–17419 Chou CC, Chang C, Liu JH et al (2007) Small ubiquitin-like modifier modification regulates the DNA binding activity of glial cell missing Drosophila homolog a. J Biol Chem 282: 27239–27249 Cohen SX, Moulin M, Hashemolhosseini S et al (2003) Structure of the GCM domain-DNA complex: a DNA-binding domain with a novel fold and mode of target site recognition. EMBO J 22:1835–1845 Colman PM, Lawrence MC (2003) The structural biology of type I viral membrane fusion. Nat Rev Mol Cell Biol 4:309–319 Compston A, Coles A (2008) Multiple sclerosis. Lancet 372:1502–1517 Cross JC, Nakano H, Natale DR et al (2006) Branching morphogenesis during development of placental villi. Differentiation 74:393–401 Dunlap KA, Palmarini M, Varela M et al (2006) Endogenous retroviruses regulate periimplantation placental growth and differentiation. Proc Natl Acad Sci USA 103:14390–14395 Dupressoir A, Marceau G, Vernochet C et al (2005) Syncytin-A and syncytin-B, two fusogenic placenta-specific murine envelope genes of retroviral origin conserved in Muridae. Proc Natl Acad Sci USA 102:725–730 Dupressoir A, Vernochet C, Bawa O et al (2009) Syncytin-A knockout mice demonstrate the critical role in placentation of a fusogenic, endogenous retrovirus-derived, envelope gene. Proc Natl Acad Sci USA 106:12127–12132 Esnault C, Priet S, Ribet D et al (2008) A placenta-specific receptor for the fusogenic, endogenous retrovirus-derived, human syncytin-2. Proc Natl Acad Sci USA 105:17532–17537 Frey G, Rits-Volloch S, Zhang XQ et al (2006) Small molecules that bind the inner core of gp41 and inhibit HIV envelope-mediated fusion. Proc Natl Acad Sci USA 103:13938–13943 Gimenez J, Montgiraud C, Oriol G et al (2009) Comparative methylation of ERVWE1/syncytin-1 and other human endogenous retrovirus LTRs in placenta tissues. DNA Res 16:195–211 Griffiths DJ (2001) Endogenous retroviruses in the human genome sequence. Genome Biol 2:Reviews1017.1–1017.5 Heidmann O, Vernochet C, Dupressoir A et al (2009) Identification of an endogenous retroviral envelope gene with fusogenic activity and placenta-specific expression in the rabbit: a new “syncytin” in a third order of mammals. Retrovirology 6:107 Hemberger M, Cross JC (2001) Genes governing placental development. Trends Endocrinol Metab 12:162–168 Hosoya T, Takizawa K, Nitta K et al (1995) Glial cells missing: a binary switch between neuronal and glial determination in Drosophila. Cell 82:1025–1036 Jones BW, Fetter RD, Tear G et al (1995) Glial cells missing: a genetic switch that controls glial versus neuronal fate. Cell 82:1013–1023 Keryer G, Alsat E, Tasken K et al (1998) Cyclic AMP-dependent protein kinases and human trophoblast cell differentiation in vitro. J Cell Sci 111:995–1004 Knerr I, Schubert SW, Wich C et al (2005) Stimulation of GCMa and syncytin via cAMP mediated PKA signaling in human trophoblastic cells under normoxic and hypoxic conditions. FEBS Lett 579:3991–3998
5
Syncytins: Molecular Aspects
137
Kudaka W, Oda T, Jinno Y et al (2008) Cellular localization of placenta-specific human endogenous retrovirus (HERV) transcripts and their possible implication in pregnancy-induced hypertension. Placenta 29:282–289 Larsen JM, Christensen IJ, Nielsen HJ et al (2009) Syncytin immunoreactivity in colorectal cancer: potential prognostic impact. Cancer Lett 280:44–49 Larsson LI, Holck S, Christensen IJ (2007) Prognostic role of syncytin expression in breast cancer. Hum Pathol 38:726–731 Lavillette D, Marin M, Ruggieri A et al (2002) The envelope glycoprotein of human endogenous retrovirus type W uses a divergent family of amino acid transporters/cell surface receptors. J Virol 76:6442–6452 Mangeney M, Renard M, Schlecht-Louf G et al (2007) Placental syncytins: Genetic disjunction between the fusogenic and immunosuppressive activity of retroviral envelope proteins. Proc Natl Acad Sci USA 104:20534–20539 Matouskova M, Blazkova J, Pajer P et al (2006) CpG methylation suppresses transcriptional activity of human syncytin-1 in non-placental tissues. Exp Cell Res 312:1011–1020 Maynard S, Epstein FH, Karumanchi SA (2008) Preeclampsia and angiogenic imbalance. Annu Rev Med 59:61–78 Mi S, Lee X, Li X et al (2000) Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature 403:785–789 Nait-Oumesmar B, Copperman AB, Lazzarini RA (2000) Placental expression and chromosomal localization of the human Gcm 1 gene. J Histochem Cytochem 48:915–922 Oluwole SO, Yao Y, Conradi S et al (2007) Elevated levels of transcripts encoding a human retroviral envelope protein (syncytin) in muscles from patients with motor neuron disease. Amyotroph Lateral Scler 8:67–72 Peng X, Pan J, Gong R et al (2007) Functional characterization of syncytin-A, a newly murine endogenous virus envelope protein. Implication for its fusion mechanism. J Biol Chem 282:381–389 Renard M, Varela PF, Letzelter C et al (2005) Crystal structure of a pivotal domain of human syncytin-2, a 40 million years old endogenous retrovirus fusogenic envelope gene captured by primates. J Mol Biol 352:1029–1034 Rossant J, Cross JC (2001) Placental development: lessons from mouse mutants. Nat Rev Genet 2:538–548 Schreiber J, Enderich J, Wegner M (1998) Structural requirements for DNA binding of GCM proteins. Nucleic Acids Res 26:2337–2343 Schreiber J, Riethmacher-Sonnenberg E, Riethmacher D et al (2000) Placental failure in mice lacking the mammalian homolog of glial cells missing, GCMa. Mol Cell Biol 20:2466–2474 Simmons DG, Natale DR, Begay V et al (2008) Early patterning of the chorion leads to the trilaminar trophoblast cell structure in the placental labyrinth. Development 135:2083–2091 Stecca B, Nait-Oumesmar B, Kelley KA et al (2002) Gcm1 expression defines three stages of chorio-allantoic interaction during placental development. Mech Dev 115:27–34 Strick R, Ackermann S, Langbein M et al (2007) Proliferation and cell–cell fusion of endometrial carcinoma are induced by the human endogenous retroviral Syncytin-1 and regulated by TGFbeta. J Mol Med 85:23–38 Wallin M, Ekstrom M, Garoff H (2004) Isomerization of the intersubunit disulphide-bond in Env controls retrovirus fusion. EMBO J 23:54–65 Yang C, Compans RW (1997) Analysis of the murine leukemia virus R peptide: delineation of the molecular determinants which are important for its fusion inhibition activity. J Virol 71:8490–8496 Yang CS, Yu C, Chuang HC et al (2005) FBW2 targets GCMa to the ubiquitin-proteasome degradation system. J Biol Chem 280:10083–10090 Yu C, Shen K, Lin M et al (2002) GCMa regulates the syncytin-mediated trophoblastic fusion. J Biol Chem 277:50062–50068
Chapter 6
Role of the Actin Cytoskeleton Within FuRMAS During Drosophila Myoblast Fusion and First Functionally Conserved Factors in Vertebrates Susanne-Filiz Önel, Christine Dottermusch, Angela Sickmann, Detlev Buttgereit, and Renate Renkawitz-Pohl
Abstract The larval musculature of Drosophila arises by fusion of two types of myoblasts: the founder cells (FCs), which determine the identity of every individual muscle, and fusion competent myoblasts (FCMs). Cell–cell recognition and adhesion is mediated by the Ig class of transmembrane proteins. They form an adhesion ring/belt at the contact sites of FCMs and FCs/growing myotubes to establish a Fusion Restricted Myogenic Adhesive Structure (FuRMAS). FuRMAS are postulated to trigger myoblast fusion, with the formation, and dissolution of F-actin foci/plugs at the sites of cell–cell contact. Electron-dense vesicles accumulate at opposing membranes of FCMs and FCs/growing muscles, and form a pre-fusion complex (1 μm2 ). This is hypothesised to take place in the centre of the FuRMAS. The vesicles are thought to be exocytosed, followed by membrane vesiculation and removal of membrane remnants to achieve cytoplasmic continuity over an area of 12 μm2 . The FCM can then be integrated into the growing myotube. This last step depends on Arp2/3 mediated F-actin reorganisation. The data on cell adhesion, signalling and actin regulation in zebrafish, C2C12 cells and mice strongly indicate conserved factors and principles between Drosophila and vertebrate myoblast fusion. Keywords FuRMAS · myoblast fusion · actin cytoskeleton · signalling cascades Abbreviations A Ants Arf Blow CA Cdc42
Acidic domain Antisocial ADP ribosylation factor Blown fuse Cofilin-homologous, and acidic domain Cell division control protein 42 homolog
R. Renkawitz-Pohl (B) Fachbereich Biologie, Entwicklungsbiologie, Philipps-Universität Marburg, 35043 Marburg, Germany e-mail:
[email protected]
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_6, C Springer Science+Business Media B.V. 2011
139
140
CED DOCK Duf EM F-actin FC FCM FuRMAS GEF Gfl Hbs IgSF IrreC Kirre Lmd Mbc Minc Mib2 PIP2 Rac Rols Rst S2 cells Sing Sltr Sns TEM VCA vrp1 WASP Wip
S.-F. Önel et al.
Caenorhabditis elegans death gene Dedicator of cytokinesis Dumbfounded Electron microscopy Filamentous actin Founder cell Fusion competent myoblast Fusion restricted myogenic adhesive structure Guanine nucleotide exchange factor Gleefull Hibris Immunoglobulin superfamily Irregular optic Chiasma Kin of irre Lameduck Myoblast city Myoblast incompetent Mind bomb 2 Phosphatidylinositol 4,5-bisphosphat Ras-related C3 botulinum toxin substrate Rolling pebbles Roughest Schneider cells Singles bar Solitary Sticks and stones Transmission electron microscopy Verprolin-homologous, cofilin-homologous, and acidic domain Verprolin 1 Wiskott-Aldrich syndrome family protein WASP-interacting partner
Contents 6.1 Introduction to the Cell Biology and Topology of Myoblast Fusion in Drosophila . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1.1 Founder Cells, Fusion-Competent Myoblasts, Progenitors and Myofibres . . . . . . . . . . . . . . . . . . . . . . . 6.1.2 Two Phases of Myoblast Fusion . . . . . . . . . . . . . . . . . . . . . 6.2 Pre-fusion Complexes Form at Opposing Membranes, the Membranes Vesiculate, and FCMs Are Integrated into the Growing Myotube . . . . . . . . . 6.2.1 Electron-Dense Vesicles and the Pre-fusion Complex . . . . . . . . . . . 6.2.2 Electron-Dense Plaques and Vesiculating Membranes . . . . . . . . . . . 6.3 Cell Adhesion and Signalling Cascades . . . . . . . . . . . . . . . . . . . .
141 142 143 145 147 147 148
6
Role of the Actin Cytoskeleton Within FuRMAS
6.3.1 Cell Adhesion . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.2 Duf/Kirre Very Likely Acts via Rolling Pebbles in FCs and Growing Myoblasts . . . . . . . . . . . . . . . . . . . . . 6.3.3 Signalling on the FCM Side . . . . . . . . . . . . . . . . . . . . 6.4 Actin Regulation at the Site of Adhesion During Drosophila Myoblast Fusion 6.4.1 Molecular Mechanisms of F-Actin Regulation at the Site of Drosophila Myoblast Fusion . . . . . . . . . . . . . . . . . . 6.4.2 Possible Roles for Arp2/3-Based F-Actin Formation at the Site of Drosophila Myoblast Fusion . . . . . . . . . . . . . 6.4.3 Actin Regulation During Vertebrate Myoblast Fusion . . . . . . . . 6.5 The FuRMAS Model and the Topology of Myoblast Fusion . . . . . . . . 6.5.1 Fusion Pores, Membrane Vesiculation and the Size of Cytoplasmic Continuities . . . . . . . . . . . . . . . . . . . . 6.5.2 FuRMAS as Signalling Centres . . . . . . . . . . . . . . . . . . 6.6 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
141
. . .
150
. . . . . . . . .
151 152 153
. . .
154
. . . . . . . . .
156 159 161
. . . .
162 163 165 165
. . . .
. . . .
6.1 Introduction to the Cell Biology and Topology of Myoblast Fusion in Drosophila As in vertebrates, muscles of Drosophila are multinucleated and arise from fusion of myoblasts. In Drosophila, genetic screens identified genes, which when mutated, lead to failure of myoblast fusion. Recently, it turned out that a number of the essential factors identified in Drosophila are functionally conserved in vertebrates. Here, we briefly introduce the larval musculature in Drosophila before concentrating on myoblast fusion. The complete Drosophila musculature is established within a few hours, and following successful cell adhesion, a multiprotein complex, called FuRMAS (Fusion Restricted Myogenic Adhesive Structure) assembles at the site of fusion (Kesper et al. 2007, Önel and Renkawitz-Pohl 2009). F-actin plugs/foci accumulate at the fusion site (Kesper et al. 2007, Richardson et al. 2007), and alterations in the actin cytoskeleton are thought to be required for myoblast fusion (reviewed by Chen and Olson 2004, Kim et al. 2007, Önel and Renkawitz-Pohl 2009, Önel et al. 2004, Rochlin et al. 2010). Since individual fusion events take place within a matter of minutes (Beckett and Baylies 2007), a FuRMAS is a very transient structure (Kesper et al. 2007). Here, we describe the ultrastructural features of myoblast fusion, F-actin branching at the site of fusion, the essential molecules and their role during fusion. We discuss the topology at the individual fusion site and suggest that FuRMAS not only act as a platform to recruit fusion-relevant molecules, but also limit the area of membrane breakdown. We provide a comprehensive review of the key molecular players in Drosophila and functionally conserved proteins in zebrafish, C2C12 cells and mice.
142
S.-F. Önel et al.
6.1.1 Founder Cells, Fusion-Competent Myoblasts, Progenitors and Myofibres During the Drosophila life cycle, the musculature develops twice: once during embryogenesis leading to the musculature of the larvae, and a second time during metamorphosis to form the adult musculature of the holometabolic insect (reviewed by Maqbool and Jagla 2007). In this review, we focus on the well-studied formation of larval muscles during embryogenesis, since this has provided the greatest insights into myoblast fusion so far. Larval muscles comprise individual myotubes with 4–24 nuclei (Baylies et al. 1998), forming a stereotypic pattern per segment (Fig. 6.1a) that is determined by early processes within the mesoderm. A combination of ectodermal and mesodermal regulators specifies the somatic mesoderm shortly after mesoderm formation (Riechmann et al. 1997). Subsequently, NotchDelta-mediated lateral inhibition determines fusion-competent myoblasts (FCMs) and progenitor cells, the latter defined by the interplay of intrinsic and extrinsic regulators (Maqbool and Jagla 2007). The progenitor divides once more to yield two FCs, or alternatively one FC plus one adult muscle progenitor. The FCs, determining the identity of individual muscles, are localised at the external layer of the somatic mesoderm. They seed the characteristic pattern of the
Fig. 6.1 Myoblast fusion mutant Drosophila embryos lack the stereotypic muscle pattern. (a) The stereotypic pattern of larval musculature is visualized by β3 Tubulin (green) (Buttgereit et al. 1996, Leiss et al. 1988) in wild-type embryos. (b) Embryo of a fusion mutant (homozygous for blow2 ) is characterized by many unfused FCMs (green, β3 Tubulin) and FCs (red nuclei, rP298 enhancer trap in the duf/kirre gene (Nose et al. 1998), green cytoplasm β3 Tubulin)
6
Role of the Actin Cytoskeleton Within FuRMAS
143
larval musculature directed by combined sets of transcriptional regulators (reviewed by Baylies et al. 1998, Frasch 1999, Maqbool and Jagla 2007, Paululat et al. 1999, Taylor 1998). All FCMs express the Gli-related Zn-finger protein Lameduck (Lmd)/Myoblast incompetent (minc)/Gleefull (Gfl) (Duan et al. 2001, Furlong et al. 2001, Ruiz-Gomez et al. 2002). However, recent evidence suggests that FCMs are not a unique population but contain molecular diversity (discussed in Rochlin et al. 2010). After fusion Lmd/Minc/Gfl is degraded and the nuclei of the integrated FCMs become reprogrammed to the fate of the individual FC. A likely candidate for degrading Lmd/Minc/Gfl is Mind bomb 2 (Mib2, E3 ubiquitin ligase) (CarrascoRando and Ruiz-Gomez 2008) which is also required for muscle integrity and stability (Nguyen et al. 2007).
6.1.2 Two Phases of Myoblast Fusion Myoblast fusion progresses in two temporal phases (Bate 1990, Bate and Rushton 1993, Beckett and Baylies 2007). First, one FC fuses with one or two FCMs to form a muscle precursor (Fig. 6.2a). During the second phase of fusion, the precursor fuses with further FCMs until reaching the number of nuclei characteristic for an individual myofibre (for nomenclature and histology see review by Bate 1993). In addition to temporal differences, one model is that genetic differences also exist between the two phases of fusion, meaning some genes play an important role in progression from the precursor stage to the mature muscle (Berger et al.
Fig. 6.2 FCs and FCMs express Ig-class of cell adhesion molecules. (a) In the first phase of fusion a single FC (blue nucleus) fuses with FCMs (yellow nuclei) to form a precursor. After fusion the nuclei adopt the fate of the FC. Further fusion events lead to the final size of a mature muscle fibre. (b) The FC/growing myotube expresses Duf/Kirre (blue) and Rst/IrreC (green) as membrane spanning molecules. Both contain five Ig loops in the extracellular domain. (c) FCMs express Rst/IrreC (green), Sns (yellow) and Hbs (orange). Sns and Hbs are characterised by eight extracellular Ig–domains. Plasma membrane (PM); extracellular (EC), transmembrane (TM) and intracellular (IC) domains of the proteins are indicated
144
S.-F. Önel et al.
2008, Massarwa et al. 2007, Rau et al. 2001, see Önel and Renkawitz-Pohl 2009 for discussion). In the larval musculature of Drosophila, a single myotube/myofibre corresponds to an individual muscle specifically attached to their individual epidermal attachment sites yielding the stereotypic muscle pattern (see Fig. 6.1a). This pattern is severely disrupted in fusion mutants, where one can distinguish FCs, precursor cells and FCMs (Fig. 6.1b). To date, screening and analysing fusiondefective mutants have identified a number of fusion-relevant genes. Their gene products include transmembrane proteins, signalling molecules and actin regulators (summarised in Table 6.1). Ultrastructural analyses of wild-type embryos and fusion-defective mutants have provided mechanistic insights into myoblast fusion and helped create a topological model.
Table 6.1 Myoblast fusion relevant proteins of Drosophila and their functional conservation in vertebrate myoblast fusion Mice
Mouse C2C12 cells
Transmembrane proteins Duf/Kirre Kirrel
Neph1
–
Hbs
–
–
–
Rst/Irre C
–
–
–
Singles bar Sns
– Nephrin
– Nephrin
– –
Signalling molecules Arf6a –
–
Arf6
Blow
–
–
–
Cdc42b
–
Cdc42
–
Crk
Crk, Crkl
–
–
Mbc
Dock1, Dock5
Dock1, Dock5
Dock180
Mib2
–
–
–
Drosophila
Zebrafish
References Ruiz-Gomez et al. (2000), Srinivas et al. (2007), Sohn et al. (2009) Artero et al. (2001), Dowark et al. (2001) Strünkelnberg et al. (2001) Estrada et al. (2007) Bour et al. (2000), Sohn et al. (2009) Chen et al. (2003), Pajcini et al. (2008) Doberstein et al. (1997), Artero et al. (2003), Schröter et al. (2003) Schäfer et al. (2007), Vasyutina et al. (2009) Erickson et al. (1997), Balagopalan et al. (2006), Moore et al. (2007) Rushton et al. (1995), Erickson et al. (1997), Moore et al. (2007), Laurin et al. (2008), Pajcini et al. (2008) Nugyen et al. (2007), Carrasco-Rando and Ruiz-Gomez (2008)
6
Role of the Actin Cytoskeleton Within FuRMAS
145
Table 6.1 (continued) Mouse C2C12 cells
Drosophila
Zebrafish
Mice
Rac1, Rac2
Rac1
Rac1
Rols
–
–
–
Schizo/Loner
–
–
Brag-2
Actin regulators ArpC1 – Arp3 –
– –
– –
Kette/Nap-1
–
–
Nap-1
Scar
–
–
WAVE2
Vrp1/Sltr
WASP
Wip
–
–
WASP
References Hadeka-Suzuki et al. (2002), Srinivas et al. (2007), Vasyutina et al. (2009) Chen et al. (2001), Menon et al. (2001), Rau et al. (2001) Chen et al. (2003), Pajcini et al. (2008) Massarwa et al. (2007) Richardson et al. (2007), Berger et al. (2008) Schröter et al. (2004), Nowak et al. (2009) Richardson et al. (2007), Berger et al. (2008), Nowak et al. (2009) Kim et al. (2007), Massarwa et al. (2007), Berger et al. (2008) Kim et al. (2007), Massarwa et al. (2007), Schäfer et al (2007)
a Arf6 probably shares functional redundancy with another Arf-GTPase, since maternal and zygotic
loss of function mutants show wild-type muscle development (Dworak et al. 2001). seems not to be relevant for myoblast fusion in the Drosophila embryo (Schäfer et al. 2007). b Cdc42
6.2 Pre-fusion Complexes Form at Opposing Membranes, the Membranes Vesiculate, and FCMs Are Integrated into the Growing Myotube In the first phase of fusion (during stage 12), when one mononucleated FC fuses with one or two FCMs to form a precursor, myoblasts differ only slightly in size. Each myoblast is approximately 4 μm wide (light microscopy: Bate 1990 TEM: Schröter et al. 2006). The contact site of two fusing myoblasts and the area of membrane vesiculation is about 1.7–1.9 μm, with initial studies revealing no further particular features (Schröter et al. 2006). Thus, the area of membrane breakdown and cytoplasmic continuity roughly correspond to the diameter of the cells (for an example see Fig. 6.3, first phase of fusion). Thereafter, several FCMs often attach laterally to growing myotubes, with their ends extending filopodia towards
146
S.-F. Önel et al.
first phase of fusion
a N
FCMs
FC
N
second phase of fusion
b N
pre-fusion complex
N
c
electron-dense vesicle
N
electron-dense plaque
N
d
N
electron-dense plaque
membrane vesiculation
mature muscle N
Founder Cell (FC)/Fusion-Competent Myoblast (FCM)
Fig. 6.3 Characteristic features of myoblast fusion at the ultrastructural level in the first and second phase of fusion. Myoblast fusion is characterised by opposing membranes, a pre-fusion complex consisting of paired electron-dense vesicles, electron-dense plaques and vesiculating membranes. The left column illustrates the fusion events schematically at the ultrastructural level (drawings of representative photographs from transmission electron microscopy). The right column visualises the position of ultrastructural features relative to FCs (growing myotubes) and FCMs. The opposing membranes are drawn as lines, electron-dense vesicles (red in b), electron-dense plaques (green in c) and vesiculation of the membrane (a, d) with large cytoplasmatic continuities marked by arrows in left column a, d. M labels the mitochondria; N labels the nuclei (blue in FCs, yellow in FCMs)
6
Role of the Actin Cytoskeleton Within FuRMAS
147
individual attachment sites. At this point, the FCM is small in comparison to the growing myotube (Fig. 6.3, second phase of fusion). Therefore, a mechanism to restrict the area of fusion might be more important in the later rather than the initial phase of fusion. For the second phase of fusion, Doberstein et al. (1997) pioneered ultrastructural analyses of embryos at stage 13/14. The majority of fusion events occurs at this stage, progressing from precursor to multinucleated myotubes. The serial sections of Doberstein et al. (1997) revealed that several pre-fusion complexes can form simultaneously between one cell (presumably a precursor cell) and several contacting myoblasts (presumably FCMs). Myoblasts are closely aligned with their membranes directly opposite each other. Interestingly, ultrastructural analyses revealed distinct membrane-associated structures such as paired vesicles, electrondense plaques and vesiculating membranes (for a summary scheme see Fig. 6.3, second phase of fusion).
6.2.1 Electron-Dense Vesicles and the Pre-fusion Complex Electron-dense paired vesicles (40 nm in diameter) accumulate at opposing membranes. All the paired vesicles at an individual contact site comprise a pre-fusion complex. Up to 15 such paired vesicles per contact site have been detected (Doberstein et al. 1997). The number of paired vesicles varies from section to section (Doberstein et al. 1997, Kim et al. 2007); on average, sections contain 1.4 vesicles at opposing membranes (Estrada et al. 2007), which we think depends on the level of the section. Understanding the spatial distribution of vesicles of the pre-fusion complex is highly informative in terms of the topology of fusion sites. The serial sections of Doberstein et al. (1997) show that an individual pre-fusion complex spreads over an area of 1 μm2 at both sides of the opposing membranes (schematically depicted in Fig. 6.4).
6.2.2 Electron-Dense Plaques and Vesiculating Membranes Although rare, perhaps implicating a transient nature, electron-dense plaques (Fig. 6.3) with a length of 500 nm and width of 10 nm were observed at areas of intact membranes, and sometimes membrane vesiculation was seen nearby (Doberstein et al. 1997). However, neither the molecular components and function of these electron-dense plaques, nor the spaciotemporal relation between pre-fusion complexes and the electron-dense plugs are as yet clear. A frequently observed feature of fusion is that opposing membranes vesiculate, producing numerous membrane remnants. This area of membrane vesiculation often spans 3–4 μm in the second phase of fusion (for an example of an EM section see Fig. 6.5a). It is generally thought that membranes fuse by undergoing hemifusion between the outer leaflets of opposing membranes (reviewed by Martens
148 Fig. 6.4 Serial sections reveal the distribution of the pre-fusion complex over 1 μm2 . (a) Schematic drawings of an individual pre-fusion complex according to serial sections of 100 μm each (Doberstein et al. 1997, Fig. 6.3). Electron-dense vesicles accumulate at the FCM side (fusion competent myoblasts, yellow vesicles) and at the FC side (founder cell, blue vesicles)
S.-F. Önel et al.
FCM
N
FC N
N
N
M
M
Founder Cell (FC)/Fusion-Competent Cell (FCM)
and McMahon 2008). During myoblast fusion, the opposing membranes are first brought into close apposition (Fig. 6.5b). Then, we propose that multiple hemifusions occur between the outer leaflets of the FC/growing muscle and the FCM (Fig. 6.5c). Subsequent fusion of the inner leaflets leads to membrane vesiculation and thus to multiple areas of cytoplasmic continuity (Fig. 6.5d). Removal of membrane remnants is required to achieve full cytoplasmic continuity and integration of the FCM into the growing myotube. However, again the mechanism behind this is unclear. In summary, at the ultrastructural level myoblast fusion can be divided into distinct steps: e.g. cell–cell recognition and adhesion, alignment of opposing membranes, formation of pre-fusion complexes over an area of 1 μm2 and transient electron-dense plaques, vesiculation of opposing membranes, removal of membrane remnants to achieve cytoplasmic continuity, and finally integration of the FCM into the growing myotube (summarised in Fig. 6.3).
6.3 Cell Adhesion and Signalling Cascades Founder cells and fusion-competent myoblasts express a dedicated gene, and it is the dialogue between these cells that ultimately leads to myoblast fusion. This dialogue includes chemo-attraction, cell migration, recognition, adhesion and signalling. FCs/growing muscles express specific chemo-attractive signals that are proposed
6
Role of the Actin Cytoskeleton Within FuRMAS a
149 b
N
3.6 μm N
767 nm growing myotubes
FMC
c
close apposition
d
hemifusion (stalk formation)
content mixing
Fig. 6.5 Multiple hemifusions lead to vesiculation of the FC and FCM membranes. The model visualizes how hemifusion of the outer lipid-bilayers and subsequent fusion of the inner lipidbilayers of an FCM and FC/growing myotube might lead to vesiculation of the cell membranes. (a) The vesiculating membrane is visualized by transmission electron microscopy over 3.6 μm (ends marked by blue arrows), areas of cytoplasmic continuity are marked by red arrowheads. Adjacent, the individual membranes of the myoblasts are clearly distinguishable. The blue arrows also mark the position of the horizontal bars in b, c and d. The scale bar corresponds to 767 nm. In b, c and d, the processes at the membrane are shown schematically, the membranes of FCM is drawn in yellow, the membrane of the FC/growing myotube in blue. The vertical bars (blue) correspond to the positions indicated by blue arrows in (a). This scheme visualizes the phosphobilayers closely opposed. (b) Hemifusion of the outer leaflets of the membranes is proposed to lead to stalk structures. (c) Subsequent fusion of the inner leaflet of the plasma membrane of FCM and FC leads to membrane vesiculation and multiple areas of cytoplasmic continuity, which allows content mixing
150
S.-F. Önel et al.
to direct the FCMs towards the FCs. In both cell types, the relevant transmembrane molecules have multiple roles since they also mediate cell recognition, transient adhesion and signalling.
6.3.1 Cell Adhesion The contact sites between FCMs and FCs/growing muscles are characterised by the expression of four transmembrane molecules of the immunoglobulin superfamily (IgSF), Dumbfounded/Kin of irre (Duf/Kirre), Roughest/Irregular optic Chiasma (Rst/IrreC), Sticks and stones (Sns) and its paralogue Hibris (Hbs). These are involved in establishing and maintaining specific membrane contacts in various cell interactions (reviewed by Fischbach et al. 2009). During myoblast fusion, Duf/Kirre is expressed uniquely on the FC/growing muscle side (Ruiz-Gomez et al. 2000), Sns (Bour et al. 2000) and Hbs (Dworak et al. 2001) on the FCM side, while Rst/IrreC (Strünkelnberg et al. 2001) is expressed in both cell types (summarised in Fig. 6.2b, c, and Table 6.1). Duf/Kirre and Rst/IrreC share functional redundancy in FCs (Strünkelnberg et al. 2001). Expression of Duf/Kirre and Rst/IrreC at ectopic sites attract FCMs to these sites – at least in the absence of competing Duf/Kirre and Rst/IrreC in the FCs – thus it was hypothesized that Duf/Kirre and Rst/IrreC act as chemo-attractants (Ruiz-Gomez et al. 2000, Strünkelnberg et al. 2001). This idea was supported by time-lapse microscopy showing that FCMs direct their filopodia towards FCs and migrate towards them (Beckett and Baylies 2007), although the mechanism is unknown. Hbs acts in partial redundancy to Sns but less efficient (Shelton et al. 2009). Each individual muscle has its characteristic number of nuclei, resulting from a different number of fusion events. Several distinct enhancer modules are postulated to regulate duf/kirre expression in specific FCs. These modules might be a way to regulate the number of fusion events of an individual muscle at a molecular level (Guruharsha et al. 2009). The Ig repeats in the extracellular domains of all these transmembrane molecules suggest that they mediate heterologous cell adhesion between FCs/growing myotubes and FCMs. Indeed, individual transfections of Drosophila Schneider cells (S2 cells) with Sns and Duf/Kirre followed by mixing these cells leads to Sns and Duf/Kirre localising in opposing cells, and presumably as a consequence to cell adhesion, suggesting that Sns and Duf/Kirre indeed physically interact (reviewed by Abmayr et al. 2008). In the embryo, Duf/Kirre, Rst/IrreC and Sns expression is very dynamic (Kesper et al. 2007). Duf is limited to FCs and has functional redundancy to Rst/IrreC, which is reflected by enrichment of Rst/IrreC in FCs/growing myotubes (Fig. 6.6). However, Rst/IrreC is only expressed in a few FCMs at a given time point (Buttgereit and Renkawitz-Pohl, unpublished) while Sns is limited to FCMs. Duf/Kirre, Rst/IrreC and Sns appear in dots at the contact sites of FCs/growing myotubes and FCMs (shown for Duf in Fig. 6.6a). Higher magnifications reveal that Duf/Kirre, Rst/IrreC and Sns localise in a ring-shaped pattern at the contact sites of FCMs and growing myotubes (Fig. 6.6b–d). Their cytodomains are implicated in
6
Role of the Actin Cytoskeleton Within FuRMAS
151
Fig. 6.6 At their contact sites FC and FCM express Ig-class of cell adhesion molecules as a belt/ring. (a) Expression of Duf/Kirre at stage 13 of embryogenesis, a time with numerous fusion events. Anterior part of the embryo is left, dorsal site to the top. In all hemi-segments Duf/Kirre is expressed in many FCs/growing muscles in a dot like manner. In b, c and d higher magnifications of individual FCMs (circumferences are marked with dotted lines) are shown contacting growing muscles. Duf/Kirre, Rst/IrreC and Sns show a ring shaped distribution at the sites of cell contact
monitoring successful cell adhesion and inducing signalling cascades into both cell types. The cytodomains are further proposed to trigger the formation of transient F-actin plugs/foci at the contact sites (see below for details), subsequently leading to membrane vesiculation and cytoplasmic continuity (for a summary of known fusion-relevant components, see Table 6.1).
6.3.2 Duf/Kirre Very Likely Acts via Rolling Pebbles in FCs and Growing Myoblasts The intracellular domains of Duf/Kirre and Rst/IrreC are 15% homologous and contain two highly conserved motifs (Strünkelnberg et al. 2001). A cellular adaptor protein, Rolling pebbles (Rols)/Antisocial (Ants) is expressed in FCs, precursors and growing myotubes, similar to Duf/Kirre and Rst/IrreC (Chen and Olson 2001, Menon and Chia 2001, Rau et al. 2001). Myoblasts in rols/ants mutant embryos often form small syncytia, implicating that Rols is not absolutely necessary for the first fusion events (Beckett and Baylies 2007, Rau et al. 2001). Rols colocalises with Duf/Kirre in the FuRMAS (Kesper et al. 2007) and is required for the enlargement of myoblasts (Menon et al. 2005). This membrane-associated localisation of Rols is lost in the mutant Df(1)w67k30, in which both duf/kirre and rst/irreC genes are deleted (Chen and Olson 2001, Menon and Chia 2001). The myoblast fusionrelevant isoform (Pütz et al. 2005) of Rols7/Ants is a 1900 amino acid protein with multiple putative protein–protein interaction domains, such as a ring finger
152
S.-F. Önel et al.
in the N-terminal region, eight ankyrin repeats, and three tetratricopeptide repeats in the C-terminal region. Co-immunoprecipitation experiments with protein lysates from transfected cells, or yeast-two hybrid assays indicate that Duf/Kirre and Rols interact physically (Chen and Olson 2001, Kreisköther et al. 2006). Besides its role in myoblast fusion, Rols7 is a component of the Z-discs and terminal Z-discs (Kreisköther et al. 2006). Because of multiple functions interactions of Rols shown by in vitro methods need to be functionally correlated to the specific process. Also the Arf-GEF (Guanine nucleotide exchange factor) Schizo/Loner (Chen et al. 2003, Önel et al. 2004) expressed in FCs is recruited to the membrane depending on Duf/Kirre or Rst/IrreC, at least in cell culture assays (Chen et al. 2003). The GTPase Arf6 is thought to be the target for the GEF Schizo/Loner (Chen et al. 2003). Arf 6 presumably acts in redundancy to another Arf-GTPase, since lossof-function Arf6 mutants only show a cytokinesis defect in male meiosis (Dyer et al. 2007). In contrast, Richardson et al. (2007) detected Schizo/Loner also in FCMs, where it is localised outside of FuRMAS. Thus Schizo/Loner’s role in myoblast fusion requires further clarification (see Rochlin et al. 2010 for detailed discussion).
6.3.3 Signalling on the FCM Side Sns is the major cell adhesion molecule of FCMs and has been studied extensively. In S2 cells, the extracellular domain of Sns mediates heterotypic cell adhesion with Duf/Kirre or Rst/IrreC expressing cells (Galletta et al. 2004). The sns20–5 mutant isoform (formerly rost5 : Paululat et al. 1995) lacking the cytodomain of Sns allows cell adhesion t o take place, but fusion does not progress (Bour et al. 2000). This proves that the cytodomain of Sns – functionally partially redundant to Hbs (Shelton et al. 2009) – is essential for fusion to proceed. It further indicates that the cytodomain transfers the signal into the FCM after the extracellular domains of Sns and Duf/Kirre interact. The cytodomain of Sns is particularly well characterised and contains 14 putative sites for tyrosine phosphorylation. Mutagenesis of these sites and testing the rescue capabilities of such modified Sns molecules in a sns mutant background revealed considerable functional redundancies between these putative phosphorylation sites (Kocherlakota et al. 2008). There are several candidates for relaying the signal into the cell. rac1, rac2 double mutants are severely defective in myoblast fusion (Hakeda-Suzuki et al. 2002, Luo et al. 1994). An unusual GEF for Rac GTPases, Myoblast City (Mbc, also known as Dock180 in vertebrates), is required in FCMs; its function depends on its SH3 and Docker domains but is independent of its Crk-binding site. Therefore, it has been suggested that PXXP sites in the cytoplasmic domain of Sns interact with Mbc (Balagopalan et al. 2006). Rac activation is often indicative of actin polymerisation (see Section 6.4). Blown fuse (Blow) is a PH domain protein and essential for myoblast fusion (Doberstein et al. 1997). Blow is solely expressed in FCMs (Schröter et al. 2006) and is recruited to the area of F-actin foci/plugs within the
6
Role of the Actin Cytoskeleton Within FuRMAS
153
FuRMAS (Kesper et al. 2007). It also interacts genetically with Kette/Nap1/Hem (Schröter et al. 2004), implicating that Blow plays a role in F-actin regulation (see Section 6.4 and Fig. 6.9).
6.4 Actin Regulation at the Site of Adhesion During Drosophila Myoblast Fusion At the light microscopic level, fusion is accompanied by the accumulation of transient F-actin plugs/foci that appearing at the contact sites of myoblast fusion within the adhesion belt/ring formed by adhesion molecules (Fig. 6.7) (Kesper et al. 2007, Richardson et al. 2007). Recent functional data have demonstrated that dynamic regulation of filamentous actin (F-actin) is critical for myoblast fusion (reviewed by Önel 2009, Önel and Renkawitz-Pohl 2009, Richardson et al. 2008b, Rochlin et al. 2010). In duf/kirre, rst/irreC double and sns single mutants, these actin foci/plugs are significantly reduced in size (Richardson et al. 2007), indicating that the adhesion receptors may cooperate with the cytoskeleton during membrane remodelling. Furthermore, the actin foci/plugs are highly dynamic. Live-imaging data revealed that they reached a maximum size of 4.5 μm2 in approximately 2 min (Richardson et al. 2008a), subsequently dissolving rapidly. In this section, we discuss recent molecular understanding of F-actin formation during myoblast fusion and provide possible insights into the role of F-actin forming at the site of fusion. We then compare F-actin formation in Drosophila and vertebrate myoblasts to explore possible conserved mechanisms.
Fig. 6.7 An actin focus forms in an adhering FCM attached to a growing myotube. (a) Drosophila stage 15 embryo expressing GFP-actin (green) under the control of the mesoderm-specific twist promoter. Double immunostaining with the anti-β3-Tubulin antibody (red) which marks all muscles and anti-GFP to visualize the actin foci (scale bar 10 μm). An actin focus (green) is visible at the contact site between an FCM and growing myotube (yellow arrow). (b) Schematic drawing and enlargement of the actin (green) plug/focus
154
S.-F. Önel et al.
6.4.1 Molecular Mechanisms of F-Actin Regulation at the Site of Drosophila Myoblast Fusion The principles of F-actin polymerisation are well known from cell culture systems. However, until recently only little was known about the mechanism of F-actin regulation and its role at the site of fusion. In general, cells require actin nucleators to catalyse de-novo assembly of filaments. To date, three classes of nucleation factors have been identified, including the Arp2/3 complex, formins and the spire family. The evolutionary conserved Arp2/3 complex was one of the first nucleators identified and promotes formation of a new, branched actin filament on a mother filament at an angle of 70◦ (Pollard 2007, Welch and Mullins 2002). In order to perform its diverse cellular roles, the complex must first be activated by nucleation-promoting factors (NPFs), such as members of the WASP and Scar/Wave family (Millard et al. 2004). These proteins link signal transduction pathways and Arp2/3-based actin polymerisation (Stradal and Scita 2006). Recent genetic studies in Drosophila revealed that myoblast fusion depends on both Arp2/3 regulators WASP and Scar/Wave (Berger et al. 2008, Gildor et al. 2009, Massarwa et al. 2007, Richardson et al. 2007, Schäfer et al. 2007). In Drosophila, the mother supplies the egg with sufficient scar/wave and wasp mRNA to complete embryogenesis. For this reason, scar/wave and wasp zygotic mutants do not display a severe myoblast fusion defect. Defects are only apparent when both maternal and zygotic expression of scar/wave or wasp is removed (Massarwa et al. 2007, Richardson et al. 2007). Figure 6.8 illustrates the evolutionary conserved domain structure of WASP and Scar/Wave. Unlike mammals that contain two WASP and three Scar/Wave proteins, Drosophila possesses only one WASP and one Scar/Wave protein. Both proteins bind via their functional VCA domain to the Arp2/3 complex to trigger
Fig. 6.8 Schematic representation of Drosophila WASP and Scar proteins. WASP and Scar/Wave share similar regulatory and functional domains. Drosophila WASP possesses an N-terminal WASP homology domain (WH1) that can bind the WASP-interacting partner Wip (known as Vrp1 or Sltr in Drosophila); a regulatory region that contains binding sites for PIP2 (B-domain) and a GTPase-binding domain (GBD) as well as a proline-rich region (PPP) that can bind SH2-SH3 adaptor proteins. Drosophila Scar also has a B- and PPP-region. Both proteins further possess a common functional VCA domain consisting of one to two Verprolin (V) and a Cofilin (C) homologous domain as well as an acidic tail (A)
6
Role of the Actin Cytoskeleton Within FuRMAS
155
a conformational change leading to Arp2/3 complex activation. Two different mechanisms are known to regulate the activity of NPFs themselves. WASP proteins are normally folded in an auto-inhibited conformation, which masks the VCA domain. Proteins that bind to the regulatory region of the WASP protein are able to activate WASP (Fig. 6.8). Although unmasking mammalian WASP depends on the synergistic binding of phosphatidylinositol 4,5-bisphosphat (PIP2 ) and GTP-bound Cdc42, neither the PIP2 and Cdc42 binding domains (B and GDB) of WASP, nor Cdc42 itself seem to be important during Drosophila myoblast fusion (Massarwa et al. 2007, Schäfer et al. 2007). However, removing conserved A or CA domains in WASP cause severe defects in myoblast fusion (Schäfer et al. 2007). Another protein binding to the N-terminal WH1 domain of WASP (Fig. 6.8) is the WASP-interacting partner Wip, which is also known as verprolin 1 or solitary (vrp1/sltr) in Drosophila. The loss of Vrp1/Sltr/Wip function in Drosophila disrupts myoblast fusion (Berger et al. 2008, Kim et al. 2007, Massarwa et al. 2007). Furthermore, Vrp1/Sltr/Wip is exclusively expressed in FCMs at the beginning of myoblast fusion, localising at the tip of the FCM attached to the FCs/growing myotubes (Berger et al. 2008). Upon fusion Vrp1/Sltr/Wip has also been observed in growing myotubes (Massarwa et al. 2007). Unlike WASP, a protein complex keeps mammalian Scar/Wave proteins in an inactivated state. Members of this complex are Sra-1/PIR121, Nap1, Abi and HSPC300 (reviewed by Takenawa and Suetsugu 2007). The first clue that members of the Scar/Wave complex are crucial for the fusion process in Drosophila came from mutant analyses of the Drosophila Nap1 homologue Kette. kette mutant embryos show clear defects in myoblast fusion and interact genetically with the FCM-specifically expressed gene blow (Schröter et al. 2006). How the Scar/Wave complex becomes activated is not yet clear and may differ between mammalian Wave and Wave2 [reviewed by Pollard 2007). One study found that GTP-bound Rac leads to dissociation of the Wave complex and thus activation of Wave (Eden et al. 2002). For Wave2, however, it was reported that Rac-GTP binds to the Wave complex without leading to dissociation of the complex (Innocenti et al. 2004). A third group reported that Rac-GTP stimulation must be enhanced by another protein (Suetsugu et al. 2006). In Drosophila two rac (rac1 and rac2) and a rac-related (mtl) genes are known. The rac1 and rac2 double mutants as well as rac1, rac2 and mtl triple mutants show severe defects in myoblast fusion, indicating that rac1 and rac2 share functional redundancy during fusion (Hakeda-Suzuki et al. 2002). Recent data from Gildor et al. (2009) suggest that Rac and Kette, and thus the Scar/Wave signalling pathway, not only play a role during myoblast fusion, but are also important for myoblast migration occurring prior to fusion. Localisation of Rac to the plasma membrane seems to depend on Drosophila Schizo/Loner and the Arf6-GTPase (Chen et al. 2003). Additionally, activation of Rac relies on Mbc (Côté and Vuori 2006). The loss of mbc in Drosophila causes a block in myoblast fusion (Erickson et al. 1997, Rushton et al. 1995). Mbc is a member of the CDM superfamily of GEFs, other members being the vertebrate protein Dock180 and the C. elegans protein CED-5 (Wu and Horvitz 1998). Dock180 is known to form a signalling complex with
156
S.-F. Önel et al.
CrkII/CED-2 and ELMO/CED-12 (reviewed by Meller et al. 2005). Interestingly, Drosophila Elmo was identified as an interaction partner for Mbc in myoblasts (Geisbrecht et al. 2008). Moreover, the Drosophila SH2-SH3 adaptor protein Crk was shown to interact with the cell adhesion molecule Sns to activate the Wip-WASP complex (Kim et al. 2007). Some of the actin components described above apparently overlap with the F-actin plugs/foci observed at the site of fusion, e.g. Drosophila Titin, Blow, Mbc and Scar (Kesper et al. 2007, Richardson et al. 2007). Scar/Wave is virtually undetectable in kette mutant embryos, suggesting that Kette might be involved in the localisation and stabilisation of the Scar/Wave protein (Richardson et al. 2007). The same study reported that F-actin foci/plugs between adhering myoblasts in kette, scar/wavematernal/zygotic , blow, and mbc single mutants and rac1, rac2, mtl triple mutants are enlarged. This implies that the membrane breakdown of adhering myoblasts signals dissolution of the F-actin foci. In addition, experiments with double mutants shed light on the interplay between WASP and Scar/Wave in Arp2/3 activation during the fusion process (Berger et al. 2008, Schäfer et al. 2007). These experiments indicate that WASP and Scar play distinct roles in activating the Arp2/3 complex (Fig. 6.9). Phenotypic analyses of double mutants suggest that scar wip mutants stop fusing during the first phase, whereas scar wasp mutants stop fusing during the second phase of fusion. Thus, it is possible that Vrp1/Sltr/Wip may act independently of WASP together with Scar/Wave during the first fusion phase, while Scar/Wave, WASP and Vrp1/Sltr/Wip control the second fusion phase (Berger et al. 2008). In support of this notion, waspmaternal/zygotic mutant embryos were still able to form three-nucleated myotubes (Massarwa et al. 2007). Genetic epistasis experiments with kette and wasp further imply crosstalk between Scar-dependent and WASP-dependent Arp2/3 activation. Intriguingly, reducing the gene dosage of wasp in a kette mutant background suppressed the kette mutant phenotype (Schäfer et al. 2007). All actin components involved in Drosophila myoblast fusion and a model for Scar- and WASP-dependent regulation during the first and second phase of fusion are summarised in Fig. 6.9 and Table 6.1.
6.4.2 Possible Roles for Arp2/3-Based F-Actin Formation at the Site of Drosophila Myoblast Fusion Although many molecular components leading to Arp2/3-based actin polymerisation are apparently crucial for myoblast fusion, the intriguing question still remains as to the actual role of actin polymerisation in myoblast fusion. This has been investigated by transmission electron microscopy (TEM) studies on mutant embryos. Ultrastructural analyses of diverse fusion-defective mutants have shown that the characteristic ultrastructural features (Section 6.2; Fig. 6.3) are crucial for Drosophila myoblast fusion (examples in Fig. 6.10). Doberstein et al. (1997) observed that the genetic interaction partner of kette (Section 6.3.3), blow, ceases fusion after the formation of electron-dense vesicles (Fig. 6.11). In contrast,
Role of the Actin Cytoskeleton Within FuRMAS a) Initial or first phase of fusion
FCM
Arf6
Mbc Rac
Kette
Blow
157 b) Second phase of fusion
FCM
Arf6
Mbc Rac
Schizo/Loner
Scar
Blow
Kette
Schizo/Loner WASP
Scar
Vrp1/Sltr/Wip
Vrp1/Sltr/Wip? Arp 2/3
?
Kette
Crk
Sns
F-actin F-actin
Duf
PM
Rols
Arp 2/3
Rols
Arp 2/3 Scar
?
Kette
FC
Scar Rac
Rac Mbc
Crk
Arp 2/3
Duf
F-actin
Duf
F-actin
Duf
PM
PM Sns
Sns
PM
Sns
6
Arf6 Schizo/Loner
Mbc growing myotube
Arf6 Schizo/Loner
Fig. 6.9 The Arp2/3 complex is regulated differently during the first and second phase of fusion. Single and double mutant analyses suggest that the Arp2/3 complex is regulated differently during the two fusion phases. The cell adhesion molecules Duf/Kirre and Sns are required for myoblast recognition and adhesion during both fusion phases and seem to trigger a signalling cascade leading to the formation of actin foci at the site of fusion. (a) In Berger et al. (2008) we proposed that during the first fusion phase the Scar/Wave pathway induces formation of F-actin at the site of fusion. Components of this pathway known to be essential for the fusion process are the small GTPase Rac and its activator Mbc as well as the Nap-1 homologue Kette. kette is further known to interact genetically with the FCM-specifically expressed gene blow. Localisation of Rac seems to depend on Sec7-GEF Schizo/Loner by activating Arf6-GTPase. (b) During the second phase the WASP signalling pathway is also important for fusion. In the growing myotube the multidomain protein Rols/Ants interacts directly with the intracellular domain of the cell adhesion molecule Duf. Biochemical assays suggest that Sns transfers the fusion signal via the SH2-SH3 adaptor protein Crk. In turn, Crk activates the Vrp1/WIP-WASP complex. Prior to fusion Vrp1 is only present in FCMs. Whether it also plays a role in the growing myotube after fusion is not yet clear. Dosage experiments further indicate that Kette and WASP act antagonistically during fusion as indicated by black line. Thus, there might be crosstalk between the Scar and WASP signalling pathways
kette mutant embryos arrest fusion after the formation of electron-dense plaques (Fig. 6.11) (Gildor et al. 2009, Schröter et al. 2004). Doberstein et al. (1997) further observed that embryos expressing activated Rac1G12V in myoblasts sometimes form small fusion pores, leading to mixing of cytoplasmic content between fusing myoblasts. Taken together, these findings along with new EM data on kette wasp double mutant embryos from Gildor et al. (2009) may indicate that the Scar-dependent signalling pathway is required to form small fusion pores during myoblast fusion. As kette wasp double mutants, kette single mutants show no membrane breakdown (Gildor et al. 2009, Schröter et al. 2004). Phenotypic analyses of wasp and vrp1/sltr/wip mutant embryos have produced two different results. Kim et al. (2007) observed accumulation of electron-dense vesicles in vrp1/sltr/wip mutant embryos using the sltrS1946 allele. This allele, a
158
S.-F. Önel et al.
Fig. 6.10 wasp and vrp1 mutant myoblasts stop fusion during membrane breakdown whereas Arp3 mutant myoblasts fail to integrate the FCM into the growing myotube. Conventional electron microscopy of Drosophila stage 14 embryos. (a) Fusing wild-type myoblasts. The membranes of the myoblasts have fused (arrows) and membrane remnants are visible in between, numerous cytoplasmic continuities (arrow heads) have been formed and content mixing of both myoblast types has occurred. (b and c) wasp3D3-035 and wipD30 mutant myoblasts. The membranes of the myoblasts have fused at the edge of the contact sites (arrows) and the membranes in between are in the process of breaking down. (d) Arp3Schwächling mutant myoblasts show a fully open but small cytoplasmic continuity (arrows) with no visible membrane remnants
1.2 kb deletion of genomic DNA spanning three exons, produces a truncated version of the Vrp1/Sltr/Wip protein. In contrast, Massarwa et al. (2007) reported that the vrp1/sltr/wip loss-of-function allele wipD30 arrests fusion after forming multiple small fusion pores (compare Fig. 6.11). A GFP-leakage assay where GFP was expressed specifically in FCs/growing myotubes supports the idea that cytoplasmic mixing occurs in wipD30 mutant myoblasts, since GFP can only leak into attached FCMs. The existence of small fusion pores in wipD30 mutant myoblasts was confirmed by Berger et al. (2008). Similarly to wipD30 , wasp mutant myoblasts also arrest fusion after forming multiple fusion pores (Fig. 6.10b, c) (Berger et al. 2008, Massarwa et al. 2007). Based on these findings it was proposed that WASPdependent actin polymerisation coordinates the clearance of membrane remnants between fusing myoblasts (see Fig. 6.11). Interestingly, no membrane remains are seen between fused myoblasts carrying the mutant allele of the Arp2/3 subunit Arp3
6
Role of the Actin Cytoskeleton Within FuRMAS
a
159
b N
N
Blow
Sing
Kette d
c N
N
Wasp, Vrp1, Rac1G12V
N
N
Arp3schwächling
Fig. 6.11 Myoblast fusion mutants arrest fusion in different ways. (a) In sing23 mutants, prefusion complexes (grey vesicles and red arrow in a) accumulate (Estrada et al. 2007), in blow2 mutants the pre-fusion complex forms and dissolves, then fusion seems to arrest (Doberstein et al. 1997). (b) In kette mutants, the electron dense plaques are observed (c) Mutation of vrp1/wip and Rac1G12V expression lead to arrest of membrane fusion when membranes vesiculate (red arrow). (d) In Arp3schwächling mutants, membranes dissolve (red arrow), but the FCM is not integrated into growing myotubes. Drawings according to representative ultrastructural analysis (see Fig. 6.10)
(Arp3Schwächling ) (Fig. 6.10d) (Berger et al. 2008). Since Arp3schwächling mutants show cytoplasmic continuity (removal of membrane remains) between FCMs and growing myotubes, but fail to incorporate the FCM content into the growing syncytium, this may indicate that Arp2/3-based actin formation is additionally required to integrate the FCM into the growing myotube (Berger et al. 2008).
6.4.3 Actin Regulation During Vertebrate Myoblast Fusion Vertebrate skeletal muscles form like Drosophila muscles via two phases of myoblast fusion (Horsley and Pavlath 2004). In the first phase, de-novo fusion of myoblasts generates nascent myotubes that serve as scaffolds in the second
160
S.-F. Önel et al.
phase, when additional mononucleated myoblasts fuse with these myotubes until a syncytial myofibre with several nuclei is formed. Also on the cellular level, Drosophila and vertebrate muscle formation seem to be identical, in that myoblasts must recognise each other and adhere, finally leading to the breakdown of the membrane. Furthermore, the same ultrastructural features that occur during Drosophila myoblast fusion have been identified during skeletal muscle formation, namely electron-dense vesicles and electron-dense plaques (Doberstein et al. 1997, Knudsen and Horwitz 1977, Rash and Fambrough 1973). However, for a long time it seemed that skeletal muscle formation partially differs at a molecular level from Drosophila myoblast fusion. Although the cell adhesion molecules Duf/Kirre and Sns share structural similarities with Neph1 and Nephrin, Nephrin proteins were simply associated with maintaining the kidney filtration barrier (Tryggvason 1999). Only recently, the expression of Nephrin proteins and their function during fusion was also reported in vertebrate skeletal muscles (see Table 6.1) (Sohn et al. 2009, Srinivas et al. 2007). Moreover, mixed cell culture experiments suggest that Nephrins must be present in mononucleated myoblasts, but not in nascent myotubes, for secondary myotube formation (Sohn et al. 2009). This suggests that Nephrin may play a role in vertebrate myoblasts similar to Sns in FCMs. Since Arp2/3-based actin polymerisation appears essential for Drosophila myoblast fusion, this has raised the question as to whether components of the Arp2/3 machinery are also crucial for skeletal muscle formation. The first evidence that some of the molecular components are also conserved in vertebrates came from studies using zebrafish as a model organism (Table 6.1). The Rac-GTPase and its activator Mbc (Dock1 and Dock5 in zebrafish), as well as the SH2-SH3 adaptor proteins Crk and Crk-like (Crkl) play a role in forming multinucleated fasttwitch muscles in zebrafish (Moore et al. 2007, Srinivas et al. 2007). Knockdown of Dock1, Dock5, Crk and Crkl significantly compromises myoblast fusion (Moore et al. 2007). Interestingly, it was further reported that simultaneous knockdown of Crkl and Dock5 blocks fusion almost completely in zebrafish. Both human Crkl and Drosophila Crk interact with the WASP-interacting partner Wip (Kim et al. 2007, Sasahara et al. 2002), which is known to be essential for Drosophila myoblast fusion. Additionally, there is evidence that Drosophila Crk interacts with the cell adhesion molecule Sns (Kim et al. 2007). Whether zebrafish Crkl can also interact with zebrafish Nephrin and Wip still needs to be elucidated. However, there also seem to be functional differences between some of these genes during Drosophila and zebrafish myoblast fusion. For example, expression of activated Rac and overexpression of Crk and Crkl in zebrafish lead to hyperfusion in multinucleated fast-twitch muscles (Moore et al. 2007, Srinivas et al. 2007). In contrast, expression of activated Rac leads to a block in myoblast fusion Drosophila (Fig. 6.11c) (Doberstein et al. 1997, Luo et al. 1994) and no hyperfusion was reported for Crk expression in myoblasts. Up to now the conserved function of most actin components has mainly been tested in mouse C2C12 myoblasts (Table 6.1). RNAi knockdown of N-WASP and Wip reduces myoblast fusion significantly in murine C2C12 cells (Kim et al. 2007).
6
Role of the Actin Cytoskeleton Within FuRMAS
161
Furthermore, it was demonstrated that targeted knockdown of Rac effector proteins such as MBC, Schizo and Arf6 lead to fusion defects in C2C12 cells (Pajcini et al. 2008). A recent study by Nowak et al. (2009) examined the relevance of the Drosophila Kette homologue Nap1 as well as Wave1 and Wave2 for the fusion of C2C12 cells using short hairpin RNA. Large numbers of Nap1- and Wave2knockdown myoblasts remained unfused. In contrast, Wave1 did not seem to be crucial for C2C12 fusion. In the same study, accumulation of F-actin rich aggregates was observed in adhering Nap1-knockdown myoblasts. Moreover, these aggregates also contained the Mbc homologue Dock 180. Whether these aggregates correspond to enlarged F-actin foci/plugs identified in Drosophila actin mutants such as kette mutant embryos needs further investigation. At an ultrastructural level, Duan and Gallagher (2009) reported that F-actin walls form between fusing C2C12 myoblasts in culture, suggesting that this supports membrane apposition, appearance of fusion pores and regulates vesicle trafficking. Functional analysis of actin components in a developing mammalian embryo is difficult to assess since loss of most components leads to early embryonic lethality. Dock1 or Dock5 are the orthologues of Drosophila Rac1-GEF Mbc. Impaired myoblast fusion was observed for Dock1 loss-of-function mice embryos, but not for Dock5-null-mutant embryos (Laurin et al. 2008). Although homozygous Dock5mutant animals were viable and had no obvious morphological abnormalities, genetic analysis revealed functional redundancy between Dock1 and Dock5 in myoblast fusion. Another elegant study used conditional mutagenesis to demonstrate a role for Rac1 and Cdc42 in mice myogenesis (Vasyutina et al. 2009). The authors used Cre-recombinase under the control of Lbx1, which induces the recombination of floxed Rac1 and Cdc42 in myogenic precursor cells that migrate to targets like the limb, diaphragm and tongue. The fusion index in Rac1 and Cdc42 mutant mice was reduced and most myoblasts remained unfused. It is notable that in contrast to Drosophila, mutations of Rac1 in mice are sufficient to interfere with fusion (see Table 6.1). Primary cell culture experiments further revealed that although myoblasts do adhere, F-actin recruitment is reduced in Rac1 or Cdc42 mutant myoblasts. Furthermore, recruitment of Arp2/3 was different in Rac1 and Cdc42 mutant myoblasts. Arp 2/3 recruitment is remarkably reduced in rac1 mutant cells, whereas Arp 2/3 localisation to the contact sites is unaffected in Cdc42 mutant cells, indicating that Rac1 and Cdc42 may function in non-equivalent manners. In Drosophila, however, Cdc42 appears not to be important (Schäfer et al. 2007).
6.5 The FuRMAS Model and the Topology of Myoblast Fusion Live imaging of myoblasts expressing actin-GFP showed that an individual fusion event takes a matter of minutes (Richardson et al. 2008a). This means membraneassociated events such as cell adhesion, signalling, forming and dissolving the actin plugs/foci and generating the pre-fusion complex are highly dynamic. How are membrane dynamics, the ultrastructural features and cytoskeletal dynamics
162
S.-F. Önel et al.
integrated? There is evidence that F-actin regulation within the FuRMAS differs between the first and second phase of fusion (see Section 6.4 and Fig. 6.9). From a morphological point of view the first and second phase of myoblast fusion are certainly different (Fig. 6.9). During the first phase, the area of contact between FC and FCM spans nearly the width of the myoblasts (1.7–01.9 μm) (Schröter et al. 2006). Also during the second phase of fusion, the FCM is as small as in the first phase, but the myotube increases in size with every successful fusion event. This likely requires a mechanism to restrict the area of membrane breakdown in the second fusion phase, and besides signalling, might imply an additional role of Rols (Rau et al. 2001). Furthermore, FCMs attach laterally to the growing myotubes, which then migrate towards their epidermal attachment sites by extending filopodia from the distal and proximal end (Bate 1990), a process that also depends on the cytoskeleton (recently reviewed by Guerin and Kramer 2009). During the second phase of fusion, the ultrastructure data show membrane vesiculation over a large area, for example 3.6 μm2 in diameter as shown for an individual fusion site in Fig. 6.5a.
6.5.1 Fusion Pores, Membrane Vesiculation and the Size of Cytoplasmic Continuities Fusion of membranes has been particularly well studied for the entry of viruses into cells and for synaptic vesicles at neuromusculature junctions (reviewed by Chernomordik and Kozlov 2008, Martens and McMahon 2008). In these cases, the initial fusion pore is predicted to have a radius of 3 nm (Jackson 2009). In contrast, the cytoplasmic continuity (diameter of up to 4 μm) during myoblast fusion stretches over a far larger area of 12 μm2 , which is about 1,000-fold larger than in vesicle fusion. How is this cytoplasmic continuity achieved in such a short time, does fusion pore expansion progress like opening a zipper, or do multiple fusion events join to form wider cytoplasmic continuity? Doberstein et al. (1997) proposed that the vesicles of the pre-fusion complex are exocytosed and might carry a fusogen. At the ultrastuctural level membrane vesiculation was observed after this (see Section 6.2.2), leading to several small cytoplasmic bridges and mixing of the cytoplasm as shown by a diffusion assay (Gildor et al. 2009). The molecules that regulate exocytosis or induce membrane vesiculation are so far unknown. However, loss-offunction mutants existing for a number of fusion-relevant genes provided a unique opportunity to correlate the protein of interest with interruption of myoblast fusion at the ultrastrucural level. Two mutants (Fig. 6.10), blow2 and singles bar23 (sing23 ), do not allow fusion to progress towards membrane vesiculation (Doberstein et al. 1997, Estrada et al. 2007). In contrast to FCM-specific expression of blow (Schröter et al. 2004), sing is expressed in FCs/growing myotubes and FCMs (Estrada et al. 2007). blow2 loss-of-function mutants do form, but do not accumulate pre-fusion complexes, while sing23 mutants (stop codon in C-terminal part of the protein) do accumulate pre-fusion complexes. The failure to go beyond pre-fusion complexes, and the fact that wild-type Sing protein contains a MARVEL domain, a conserved
6
Role of the Actin Cytoskeleton Within FuRMAS
163
domain involved in membrane apposition (Sánchez-Pulido et al. 2002), suggest that Sing allows progression past the pre-fusion complex, possibly by mediating fusion of the electron-dense vesicles to the plasma membrane (Estrada et al. 2007).
6.5.2 FuRMAS as Signalling Centres So far we have discussed that myoblast fusion requires highly coordinated mechanisms for cell recognition, adhesion, signalling, local F-actin regulation, exocytosis, membrane merging, removal of membrane remnants and finally integration of the FCM into the growing myotube. But what is the topology of myoblast fusion at the individual site of fusion? We suggested that cell adhesion leads to establishing a FuRMAS as a signalling centre and thereby triggering myoblast fusion (Kesper et al. 2007, Önel and, Renkawitz-Pohl 2009). This structure is transient and hypothetically regulates spaciotemporal communication between FCs/growing myotubes and FCMs during fusion, i.e. events such as formation and dissolution of F-actin plugs/foci, precise apposition of vesicles of the pre-fusion complex, and finally membrane vesiculation. In Drosophila, fusion of electron-dense vesicles to the target membranes is proposed to occur before membrane merging between FCs/growing muscles and FCMs (Doberstein et al. 1997). This might require two types of fusion: first, vesicles fuse with their target membrane and second, membranes merge between opposing myoblasts. This means membranes have to be brought into tight contact. Bilayers can apparently fuse when they are in close proximity of 1 nm (Kozlovsky et al. 2004). At the pre-fusion complex stage of myoblast fusion membranes are in close apposition over a large area, 1 μm2 (see Fig. 6.4). How is this achieved? At the light microscopic level, the cell adhesion molecules form a belt/ring at the site of fusion (Fig. 6.11, Sns = yellow, Duf = blue). In a lateral view, F-actin is at the centre of the FuRMAS surrounded by a ring of Duf and SNS (Fig. 6.9). Figure 6.12 depicts our FuRMAS model for the topology of myoblast fusion with respect to the described ultrastructural features, and the role of F-actin is discussed within this context. Immuno-EM on wild-type myoblasts implicate that the electron-dense vesicles observed at the plasma membrane of adhering myoblasts emanate from the Golgi and are targeted by actin-directed migration to the plasma membrane to form pre-fusion complexes (Kim et al. 2007) (Fig. 6.12a). Based on the serial section (Fig. 6.4) of Doberstein et al. (1997) and the ring-shaped distribution of Sns and Duf/Kirre (Fig. 6.6), Kesper et al. (2007) proposed that the pre-fusion complex with its clouds of electron-dense vesicles accumulates at the opposing membranes between these cell adhesion belts (Fig. 6.12b). As mentioned above, vesicles accumulate over an area of 1 μm2 membrane vesiculation, however, is observed over an area (12 μm2 ), with a diameter of 4 μm (Fig. 6.5). At the light microscopic level, this is reflected by the expansion of the ring of adhesion molecules from 1 to 4–5 μm in diameter (Kesper et al. 2007). This suggests that the area widens to create this size of cytoplasmatic continuity, which allows integration of the FCM into the growing myotube (Fig. 6.12c). This widening and integration of the FCM
164 Fig. 6.12 The FuRMAS model of myoblast fusion. Model diagrams of the topology at the contact site between a growing myotube (blue nuclei) and an FCM (yellow nucleus). Adhesion rings/belts are formed by Duf (blue) on the growing muscle side and Sns (yellow) on the FCM side. (a) Vesicles are transported from the Golgi (G) to the opposing membranes within the FuRMAS. (b) The pre-fusion complex forms over an area of 1 μm2 . (c) Cytoplasmatic continuity is achieved over an area of 12 μm2 . Features are schematic and not drawn to scale. Plaques observed in the EM are not integrated in this scheme (see Section 6.2.2) since the topological arrangement is unclear. F-actin and signalling molecules are discussed in Fig. 6.9.
S.-F. Önel et al. a
FCM G
N
N
N G
N FC
b
growing myotube FCM N
N
N N
FC growing myotube
c
FCM N
N
N N
FC growing myotube
depend on the activation of the Arp2/3 complex (Berger et al. 2008). The Arp2/3 complex, however, is neither essential for vesicle exocytosis nor for membrane merging. The putative fusogens which mediate vesicle exocytosis and merging of the plasma membranes from FCs/growing myotubes and FCMs are unknown. The identification of such a fusogen is one of the important questions to address to in the future. It is of high interest that FuRMAS share molecular and anatomical features such as belts/rings of adhesion molecules, local actin polymerisation, and signalling centre properties with other transient cell–cell adhesions such as the immunological synapse. This synapse connects the antigen-presenting cell with a T-cell (Grakoui et al. 1999, Monks et al. 1997, Shaw and Dustin 1997). Furthermore, there are similarities to cell matrix interactions such as podosomes and invadopodia (reviewed by Linder (2007), in particular many local F-actin regulators are in common (for a detailed comparison see Önel and Renkawitz-Pohl 2009).
6
Role of the Actin Cytoskeleton Within FuRMAS
165
6.6 Outlook Myoblast fusion is a fundamental process for myogenesis in higher organisms. Cell–cell fusion, however, also occurs in several other vital processes, such as zygote formation, yeast mating (reviewed by Ydenberg and Rose 2009) hypodermis and ulca (lid of the vulva) formation in nematodes, oesteoclast development in vertebrates and placenta formation in mammals (reviewed by Oren-Suissa and Podbilewicz 2007). Hypodermis and ulca fusion starts within an epithelium and here fusogens EFF-1 and Aff1, which are both necessary and sufficient for cell– cell fusion, have been identified and functionally characterized (reviewed by Sapir et al. 2008) Besides Syncytins (section X and Y and Larsson et al. 2008) in mammals, fusogens for cell–cell fusion do not seem to be conserved in sequence. The action of fusogens is one of the last events in membrane merging after successful cell–cell recognition and adhesion, apparently signalling into the cell that the actin cytoskeleton can be modified and fusion can commence. Since 2007, several conserved molecular components of these fusion-preparing processes have been found to be conserved between Drosophila and mammalian myoblast fusion (summarised in Table 6.1 and discussed in detail in Section 6.4.3). During myoblast fusion in Drosophila, two types of fusion are required: first the electron-dense vesicles must fuse with the opposing membranes, which should be followed by fusion of the opposing membranes of FCs/growing muscles and individual FCMs. We have an advanced understanding of cell recognition, cell adhesion and F-actin regulation during fusion, but the most central and challenging question of how membranes merge and build a syncytial cell is unresolved. After birth in vertebrates, it is essential that muscles increase the number of nuclei during postnatal growth and that muscles can be repaired after injuries by myoblast fusion events. These myoblasts originate from the activation of satellite cells, the stem cells of skeletal muscles (for a recent review see Rudnicki et al. 2009). The emerging common components in Drosophila and mammalian myoblast fusion prove that ground level research in Drosophila advances investigations in mammals, which in turn opens the way to translational health-relevant research. Acknowledgments We are grateful to Katja Gessner for preparing most of the drawings and for excellent secretarial assistance, Christina Hornbruch for critical reading. This work was supported by the Deutsche Forschungsgemeinschaft with grants to S.Ö. (OE 311/4-1) and R. R.-P. (Re628/14-3, R2628/15-2) and GRK1216 as well as the EU network of Excellence MYORES.
References Abmayr SM, Zhuang S, Geisbrecht ER (2008) Myoblast fusion in Drosophila. Methods Mol Biol 475:75–97 Balagopalan L, Chen MH, Geisbrecht ER et al (2006) The CDM superfamily protein MBC directs myoblast fusion through a mechanism that requires phosphatidylinositol 3,4,5-triphosphate binding but is independent of direct interaction with DCrk. Mol Cell Biol 26:9442–9455 Bate M (1990) The embryonic development of larval muscles in Drosophila. Development 110:791–804
166
S.-F. Önel et al.
Bate M (1993) The mesoderm and its derivatives. In: Bate M, Martinez Arias A (eds) The development of Drosophila melanogaster, vol 2. Cold Spring Harbor Laboratories Press, Cold Spring Harbor Bate M, Rushton E (1993) Myogenesis and muscle patterning in Drosophila. CR Acad Sci III 316:1047–1061 Baylies MK, Bate M, Ruiz Gomez M (1998) Myogenesis: a view from Drosophila. Cell 93: 931–937 Beckett K, Baylies MK (2007) 3D analysis of founder cell and fusion competent myoblast arrangements outlines a new model of myoblast fusion. Dev Biol 309:113–125 Berger S, Schäfer G, Kesper DA et al (2008) WASP and SCAR have distinct roles in activating the Arp2/3 complex during myoblast fusion. J Cell Sci 121:1303–1313 Bour BA, Chakravarti M, West JM et al (2000) Drosophila SNS, a member of the immunoglobulin superfamily that is essential for myoblast fusion. Genes Dev 14:1498–1511 Buttgereit D, Paululat A, Renkawitz-Pohl R (1996) Muscle development and attachment to the epidermis is accompanied by expression of beta 3 and beta 1 tubulin isotypes, respectively. Int J Dev Biol 40:189–196 Carrasco-Rando M, Ruiz-Gomez M (2008) Mind bomb 2, a founder myoblast-specific protein, regulates myoblast fusion and muscle stability. Development 135:849–857 Chen EH, Olson EN (2001) Antisocial, an intracellular adaptor protein, is required for myoblast fusion in Drosophila. Dev Cell 1:705–715 Chen EH, Olson EN (2004) Towards a molecular pathway for myoblast fusion in Drosophila. Trends Cell Biol 14:452–460 Chen EH, Pryce BA, Tzeng JA et al (2003) Control of myoblast fusion by a guanine nucleotide exchange factor, loner, and its effector ARF6. Cell 114:751–762 Chernomordik LV, Kozlov MM (2008) Mechanics of membrane fusion. Nat Struct Mol Biol 15:675–683 Côté JF, Vuori K (2006) In vitro guanine nucleotide exchange activity of DHR-2/DOCKER/CZH2 domains. Methods Enzymol 406:41–57 Doberstein SK, Fetter RD, Mehta AY et al (1997) Genetic analysis of myoblast fusion: blown fuse is required for progression beyond the pre-fusion complex. J Cell Bio 136: 1249–1261 Duan H, Skeath JB, Nguyen HT (2001) Drosophila Lame duck, a novel member of the Gli superfamily, acts as a key regulator of myogenesis by controlling fusion-competent myoblast development. Development 128:4489–4500 Duan R, Gallagher PJ (2009) Dependence of myoblast fusion on a cortical actin wall and nonmuscle myosin IIA. Dev Biol 325:374–385 Dworak HA, Charles MA, Pellerano LB et al (2001) Characterization of Drosophila hibris, a gene related to human nephrin. Development 128:4265–4276 Dyer N, Rebollo E, Dominguez P et al (2007) Spermatocyte cytokinesis requires rapid membrane addition mediated by ARF6 on central spindle recycling endosomes. Development 134: 4437–4447 Eden S, Rohatgi R, Podtelejnikov AV (2002) Mechanism of regulation of WAVE1-induced actin nucleation by Rac1 and Nck. Nature 418:790–793 Erickson MR, Galletta BJ, Abmayr SM (1997) Drosophila myoblast city encodes a conserved protein that is essential for myoblast fusion, dorsal closure, and cytoskeletal organization. J Cell Biol 138:589–603 Estrada B, Maeland AD, Gisselbrecht SS (2007) The MARVEL domain protein, Singles Bar, is required for progression past the pre-fusion complex stage of myoblast fusion. Dev Biol 307:328–339 Fischbach KF, Linneweber GA, Andlauer TFM et al (2009) The irre cell recognition module (IRM) proteins. J Neurogenet 23:48–67 Frasch M (1999) Controls in patterning and diversification of somatic muscles during Drosophila embryogenesis. Curr Opin Genet Dev 9:522–529
6
Role of the Actin Cytoskeleton Within FuRMAS
167
Furlong EE, Andersen EC, Null B et al (2001) Patterns of gene expression during Drosophila mesoderm development. Science 293:1629–1633 Galletta BJ, Chakravarti M, Banerjee R et al (2004) SNS: Adhesive properties, localization requirements and ectodomain dependence in S2 cells and embryonic myoblasts. Mech Dev 121:1455–1468 Geisbrecht ER, Haralalka S, Swanson SK et al (2008) Drosophila ELMO/CED-12 interacts with Myoblast city to direct myoblast fusion and ommatidial organization. Dev Biol 314:137–149 Gildor B, Massarwa R, Shilo BZ et al (2009) The SCAR and WASp nucleation-promoting factors act sequentially to mediate Drosophila myoblast fusion. EMBO Rep 10:1043–1050 Grakoui A, Bromley SK, Sumen C et al (1999) The immunological synapse: a molecular machine controlling T cell activation. Science 285:221–227 Guerin CM, Kramer SG (2009) Cytoskeletal remodeling during myotube assembly and guidance: coordinating the actin and microtubule networks. Commun Integr Biol 2:452–457 Guruharsha KG, Ruiz-Gomez M, Ranganath HA et al (2009) The complex spatio-temporal regulation of the Drosophila myoblast attractant gene duf/kirre. PLoS One 4:e6960 Hakeda-Suzuki S, Ng J, Tzu J et al (2002) Rac function and regulation during Drosophila development. Nature 41:438–442 Horsley V, Pavlath GK (2004) Forming a multinucleated cell: molecules that regulate myoblast fusion. Cells Tissues Organs 176:67–78 Innocenti M, Zucconi A, Disanza A et al (2004) Abi1 is essential for the formation and activation of a WAVE2 signalling complex. Nat Cell Biol 6:319–327 Jackson MB (Oct 29, 2009) Minimum membrane bending energies of fusion pores. J Membr Biol (epub) Kesper DA, Stute C, Buttgereit D et al (2007) Myoblast fusion in Drosophila melanogaster is mediated through a fusion-restricted myogenic-adhesive structure (FuRMAS). Dev Dyn 236:404–415 Kim S, Shilagardi K, Zhang S et al (2007) A critical function for the actin cytoskeleton in targeted exocytosis of pre-fusion vesicles during myoblast fusion. Dev Cell 12:571–586 Knudsen KA, Horwitz AF (1977) Tandem events in myoblast fusion. Dev Biol 58:328–338 Kocherlakota KS, Wu JM, McDermott J et al (2008) Analysis of the cell adhesion molecule Sticksand-Stones reveals multiple redundant functional domains, protein-interaction motifs and phosphorylated Tyrosines that direct myoblast fusion in Drosophila melanogaster. Genetics 178:1371–1383 Kozlovsky Y, Efrat A, Siegel DP et al (2004) Stalk phase formation: effects of dehydration and saddle splay modulus. Biophys J 87:2508–2521 Kreisköther N, Reichert N, Buttgereit D et al (2006) Drosophila Rolling pebbles colocalises and putatively interacts with alpha-Actinin and the Sls isoform Zormin in the Z-discs of the sarcomere and with Dumbfounded/Kirre, alpha-Actinin and Zormin in the terminal Z-discs. J Muscle Res Cell Motil 27:93–106 Larsson LI, Bjerregaard B, Talts JF (2008) Cell fusions in mammals. Histochem Cell Biol 129: 551–561 Laurin M, Fradet N, Blangy A et al (2008) The atypical Rac activator Dock180 (Dock1) regulates myoblast fusion in vivo. Proc Natl Acad Sci USA 105:15446–15451 Leiss D, Hinz U, Gasch A, Mertz R et al (1988) Beta 3 tubulin expression characterizes the differentiating mesodermal germ layer during Drosophila embryogenesis. Development 104:525–531 Linder S (2007) The matrix corroded: podosomes and invadopodia in extracellular matrix degradation. Trends Cell Biol 17:107–117 Luo L, Liao YJ, Jan LY, et al (1994) Distinct morphogenetic functions of similar small GTPases: Drosophila Drac1 is involved in axonal outgrowth and myoblast fusion. Genes Dev 8: 1787–1802 Maqbool T, Jagla K (2007) Genetic control of muscle development: learning from Drosophila. J Muscle Res Cell Motil 28:397–407
168
S.-F. Önel et al.
Martens S, McMahon HT (2008) Mechanisms of membrane fusion: disparate players and common principles. Nat Rev Mol Cell 19:5093–5103 Massarwa R, Carmon S, Shilo BZ et al (2007) WIP/WASp-based actin-polymerization machinery is essential for myoblast fusion in Drosophila. Dev Cell 12:557–569 Meller N, Merlot S, Guda C (2005) CZH proteins: a new family of Rho-GEFs. J Cell Sci 118: 4937–4946. Menon SD, Chia W (2001) Drosophila Rolling pebbles: a multidomain protein required for myoblast fusion that recruits D-Titin in response to the myoblast attractant Dumbfounded. Dev Cell 1:691–703 Menon SD, Osman Z, Chenchill K et al (2005) A positive feedback loop between Dumbfounded and Rolling pebbles leads to myotube enlargement in Drosophila. J Cell Biol 169:909–930 Millard TH, Sharp SJ, Machesky LM (2004) Signalling to actin assembly via the WASP (WiskottAldrich syndrome protein)-family proteins and the Arp2/3 complex. Biochem J 380:1–17 Monks CR, Kupfer H, Tamir I et al (1997) Selective modulation of protein kinase C-theta during T-cell activation. Nature 385:83–86. Moore CA, Parkin CA, Bidet Y et al (2007) A role for the Myoblast city homologues Dock1 and Dock5 and the adaptor proteins Crk and Crk-like in zebrafish myoblast fusion. Development 134:3145–3153. Nguyen HT, Voza F, Ezzeddine N et al (2007) Drosophila mind bomb2 is required for maintaining muscle integrity and survival. J Cell Biol 179:219–227 Nose A, Isshiki T, Takeichi M (1998) Regional specification of muscle progenitors in Drosophila: the role of the msh homeobox gene. Development 125:215–223. Nowak SJ, Nahirney PC, Hadjantonakis AK, Baylies MK (2009) Nap1-mediated actin remodeling is essential for mammalian myoblast fusion. J Cell Sci 122:3282–3293 Önel S, Bolke L, Klämbt C (2004) The Drosophila ARF6-GEF Schizo controls commissure formation by regulating Slit. Development 131:2587–2594 Önel SF (2009) Actin regulators take the reins in Drosophila myoblast fusion. Cent Eur J Biol 4:11–18 Önel SF, Renkawitz-Pohl R (2009) FuRMAS: triggering myoblast fusion in Drosophila. Dev Dyn 238:1513–1525 Oren-Suissa M, Podbilewicz B (2007) Cell fusion during development. Trends Cell Biol 17: 537–546 Pajcini KV, Pomerantz JH, Alkan O et al (2008) Myoblasts and macrophages share molecular components that contribute to cell–cell fusion. J Cell Biol 180:1005–1019 Paululat A, Burchard S, Renkawitz-Pohl R (1995) Fusion from myoblasts to myotubes is dependent on the rolling stone gene (rost) of Drosophila. Development 121:2611–2620 Paululat A, Holz A, Renkawitz-Pohl R (1999) Essential genes for myoblast fusion in Drosophila embryogenesis. Mech Dev 83:17–26 Pollard TD (2007) Regulation of actin filament assembly by Arp2/3 complex and formins. Annu Rev Biophys Biomol Struct 36:451–477 Pütz M, Kesper DA, Buttgereit D et al (2005) In Drosophila melanogaster, the rolling pebbles isoform 6 (Rols6) is essential for proper Malpighian tubule morphology. Mech Dev 122: 1206–1217 Rash JE, Fambrough D (1973) Ultrastructural and electrophysiological correlates of cell coupling and cytoplasmic fusion during myogenesis in vitro. Dev Biol 30:166–186 Rau A, Buttgereit D, Holz A et al (2001) rolling pebbles (rols) is required in Drosophila muscle precursors for recruitment of myoblasts for fusion. Development 128:5061–5073. Richardson BE, Beckett K, Baylies MK (2008a) Live imaging of Drosophila myoblast fusion. Methods Mol Biol 475:263–274 Richardson BE, Beckett K, Nowak SJ et al (2007) SCAR/WAVE and Arp2/3 are crucial for cytoskeletal remodeling at the site of myoblast fusion. Development 134:4357–4367 Richardson BE, Nowak SJ, Baylies MK (2008b) Myoblast fusion in fly and vertebrates: new genes, new processes and new perspectives. Traffic 9:1050–1059
6
Role of the Actin Cytoskeleton Within FuRMAS
169
Riechmann V, Irion U, Wilson R (1997) Control of cell fates and segmentation in the Drosophila mesoderm. Development. 124:2915–2932. Rochlin K, Yu S, Roy S, Baylies MK (2010). Myoblast fusion: When it takes more to make one. Dev Biol. 341:66–83 Rudnicki MA, Le Grand F, McKinnell I et al (2009) The molecular regulation of muscle stem cell function. Cold Spring Harb Symp Quant Biol 73:323–331 Ruiz-Gomez M, Coutts N, Price A et al (2000) Drosophila dumbfounded: a myoblast attractant essential for fusion. Cell 102:189–198 Ruiz-Gomez M, Coutts N, Suster ML et al (2002) Myoblasts incompetent encodes a zinc finger transcription factor required to specify fusion-competent myoblasts in Drosophila. Development 129:133–141 Rushton E, Drysdale R, Abmayr SM et al (1995) Mutations in a novel gene, myoblast city, provide evidence in support of the founder cell hypothesis for Drosophila muscle development. Development 121:1979–1988 Sánchez-Pulido L, Martín-Belmonte F, Valencia A, Alonso MA (2002) MARVEL: a conserved domain involved in membrane apposition events. Trends Biochem Sci 27: 599–601 Sapir A, Avinoam O, Podbilewicz B, Chernomordik LV (2008) Viral and developmental cell fusion mechanisms: conservation and divergence. Dev Cell 14:11–21 Sasahara Y, Rachid R, Byrne MJ et al (2002) Mechanism of recruitment of WASP to the immunological synapse and of its activation following TCR ligation. Mol Cell 10: 1269–1281 Schäfer G, Weber S, Holz A et al (2007) The Wiskott-Aldrich syndrome protein (WASP) is essential for myoblast fusion in Drosophila. Dev Biol 304:664–674 Schröter RH, Buttgereit D, Beck L et al (2006) Blown fuse regulates morphogenesis but not myoblast fusion of the circular visceral muscles in Drosophila. Differentiation 74:1–14 Schröter RH, Lier S, Holz A et al (2004) Kette and blown fuse interact genetically during the second fusion step of myogenesis in Drosophila. Development 131:4501–4509 Shaw AS, Dustin ML (1997) Making the T cell receptor go the distance: a topological view of T cell activation. Immunity 6:361–369 Shelton C, Kocherlakota KS, Zhuang S et al (2009) The immunoglobulin superfamily member Hbs functions redundantly with Sns in interactions between founder and fusion-competent myoblasts. Development 136:1159–1168 Sohn RL, Huang P, Kawahara G et al (2009) A role for nephrin, a renal protein, in vertebrate skeletal muscle cell fusion. Proc Natl Acad Sci USA 106:9374–9379 Srinivas BP, Woo J, Leong WY et al (2007) A conserved molecular pathway mediates myoblast fusion in insects and vertebrates. Nat Genet 39:781–786 Stradal TE, Scita G (2006) Protein complexes regulating Arp2/3-mediated actin assembly. Curr Opin Cell Biol 18:4–10 Strünkelnberg M, Bonengel B, Moda LM et al (2001) rst and its paralogue kirre act redundantly during embryonic muscle development in Drosophila. Development 128:4229–4239 Suetsugu S, Murayama K, Sakamoto A et al (2006) The RAC binding domain/IRSp53-MIM homology domain of IRSp53 induces RAC-dependent membrane deformation. J Biol Chem 281:35347–35358 Takenawa T, Suetsugu S (2007) The WASP-WAVE protein network: connecting the membrane to the cytoskeleton. Nat Rev Mol Cell Biol 8:37–48 Taylor MV (1998) Muscle development: a transcriptional pathway in myogenesis. Curr Biol 8:R356–R358. Tryggvason K (1999) Unraveling the mechanisms of glomerular ultrafiltration: nephrin, a key component of the slit diaphragm. J Am Soc Nephrol 10:2440–2445 Vasyutina E, Martarelli B, Brakebusch C et al (2009) The small G-proteins Rac1 and Cdc42 are essential for myoblast fusion in the mouse. Proc Natl Acad Sci USA 106: 8935–8940
170
S.-F. Önel et al.
Welch MD, Mullins RD (2002) Cellular control of actin nucleation. Annu Rev Cell Dev Biol 18:247–288 Wu YC, Horvitz HR (1998) C. elegans phagocytosis and cell-migration protein CED-5 is similar to human DOCK180. Nature 393:501–504 Ydenberg CA, Rose MD (2009) Antagonistic regulation of Fus2p nuclear localization by pheromone signaling and the cell cycle. J Cell Biol 184:409–422
Chapter 7
Role of CD9 in Sperm-Egg Fusion and Its General Role in Fusion Phenomena Natsuko Kawano, Yuichiro Harada, Keiichi Yoshida, Mami Miyado, and Kenji Miyado
Abstract In fertilization, two types of sex cells or gametes – a sperm and an egg – unite in a stepwise manner to form a mother cell, which is capable of developing naturally into a new individual. Notably, the “membrane fusion” that occurs intercellularly between a sperm and an egg is essential for fertilization. A sperm factor that is delivered into the egg cytoplasm through fusion serves to activate a signaling pathway; this leads to the resumption of meiosis in the egg. In mammals, sperm-egg fusion is partly mediated by two integral membrane proteins, sperm Izumo (Inoue et al. 2005) and egg cluster of differentiation 9 (CD9) (Kaji et al. 2000, Le Naour et al. 2000, Miyado et al. 2000), and the roles played by both are critical but yet unknown. A recent study (Miyado et al. 2000) showed that CD9containing vesicles are released from wild-type eggs, and that these exosome-like vesicles induce fusion between sperm and CD9-null eggs in vitro, even though CD9null eggs are highly refractory to sperm-egg fusion. This result provides compelling evidence for the crucial involvement of CD9-containing, fusion-facilitating vesicles in sperm-egg fusion and offers new insight into both gamete fusion and other membrane fusion events. Keywords CD9 · CD81 · cell-cell fusion · egg · exosomes · fertilization · Izumo · membrane fusion · sperm · tetraspanins Abbreviations ADAM CD GM3 EGFP HIV HSP
A disintegrin and metalloprotease Cluster of differentiation Monosialo ganglioside 3 Enhanced green fluorescent protein Human immunodeficiency virus Heat shock protein
K. Miyado (B) Department of Reproductive Biology, National Center for Child Health and Development, Tokyo 157-8535, Japan e-mail:
[email protected]
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_7, C Springer Science+Business Media B.V. 2011
171
172
ICSI LEL MRP-1 M PVS Tspan Z
N. Kawano et al.
Intracytoplasmic sperm injection Large extracellular loop Motility-related protein 1 Microvilli Perivitelline space Tetraspanin Zona pellucida
Contents 7.1 Introduction . . . . . . . . . . . . . 7.2 Sperm-Egg Fusion in Fertilization . . . 7.3 CD9 and Its Role in Cell Function . . . 7.4 Tetraspanin . . . . . . . . . . . . . 7.5 Tetraspanin as a Component of Exosomes 7.6 Lessons from “Living Eggs” . . . . . . 7.7 Membrane Fusion and Exosomes . . . References . . . . . . . . . . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
172 173 175 176 178 178 179 182
7.1 Introduction Membrane fusion is defined as the merger of two neighboring membranes, which occurs intracellularly in the cytoplasm or intercellularly on cell–cell membranes (Jahn et al. 2003). Especially, intercellular membrane fusion (hereafter, simply referred to as “membrane fusion”) is very exact in relation to the formation of certain kinds of cells and tissues associated with dynamic turning events in development, e.g., myoblast fusion in muscle formation and sperm-egg fusion in fertilization. These fused cells play a central role in the formation of tissues and organs in newborns. Hence, the absence of components necessary for fusion events results in the failure of cell formation, which causes severe muscular atrophy or an extremely decreased rate of fertilization. Moreover, membrane fusion occurs on invasion of viruses into cells. For viruses to inject their genetic materials into host cells, their envelopes need to fuse with the outer membranes of the host cells (Hernandez et al. 1996). Further, when recipient cells are infected with certain types of viruses, such as human immunodeficiency virus-1 (HIV-1) and paramyxovirus, expression of the virus fusion protein at the cell–cell contact sites can induce the fusion of neighboring cells, leading to the formation of multi-nucleate immune cells (or syncytia). Such formation of abnormally fused cells severely impairs normal cell function in terms of antibody production and antigen recognition, and finally disrupts the immune system. Therefore, membrane fusion processes are critical for both the occurrence of biologically essential events and pathologically significant diseases. Thus, understanding how outer membranes meet and merge in cell–cell fusion and virus-cell fusion is a major challenge in the creation of medicines used to treat
7
Role of CD9 in Sperm-Egg Fusion
173
muscular dystrophy, infectious incurable diseases, or infertility in humans and certain endangered species, as well as in basic research for cell biology. Here, we focus on cell–cell fusion and discuss its molecular mechanisms in relation to the role of cluster of differentiation 9 (CD9) and other tetraspanins.
7.2 Sperm-Egg Fusion in Fertilization Fertilization is a complicated event that consists of cell–cell adhesion, activation of cell signaling to allow the resumption of the cell cycle, and cell–cell fusion (Fig. 7.1). In organisms, including animals, fungi, and plants, fertilization occurs when two types of sex cells come into contact and naturally form one unified cell, termed a fertilized cell. This fertilized cell subdivides into a number of cells and eventually becomes the origin of all the tissues and organs necessary for the formation of a new individual. In addition, the eggs have a unique ability to reprogramme DNA methylation patterns in both sperm genomes and their own genomes. Fertilization is a key event that is directly linked to the generation of a new individual. In fertilization, two membrane-spanning proteins necessary for fusion are expressed in the egg and sperm – a cell adhesion molecule “integrin” (Almeida et al. 1995, Chen et al. 1999) and a membrane-anchored protease “ADAM (a disintegrin and metalloprotease)” (Evans 2001) – and have been biochemically identified and
Fig. 7.1 Series of steps from sperm–egg interaction to fusion during mammalian fertilization. This is an overview of mammalian fertilization. Fertilization is divided into multiple steps: interaction of sperm-somatic cells (termed cumulus cells), binding of sperm to the extracellular matrix (termed zona pellucida), and penetration of the egg. After the sperm penetrates the zona pellucida, it can bind and fuse to the egg cell membrane. Successful fertilization requires not only that a sperm and egg fuse, but also that no more than one sperm fuses with the egg i.e., a polyspermy block occurs
174
N. Kawano et al.
immunocytochemically confirmed to localize on the outer cell membranes of each gamete. Further, antibodies against these proteins have been shown to significantly reduce the rate of sperm-egg binding and fusion in mice. Integrins were initially identified as regulators of muscle cell fusion (Rosen et al. 1992) and have also been well-identified as cell adhesion molecules along with other membrane proteins such as cadherins. Integrins, which are expressed in many types of cells and are conserved in all animals examined, mediate cell–cell and cell–matrix interaction and intercellular communication, including cell adhesion and cell–cell fusion. On the other hand, ADAMs have a characteristic domain that is homologous to an extracellular region of integrins beside a protease-like sequence (Blobel et al. 1992). Thus, the presence of domains conserved between integrins and ADAMs indicated that these two protein families play a role in sperm-egg adhesion and/or fusion (Evans 2001). However, when genetically manipulated mice were produced to facilitate the study of the relevant genes of integrins (He et al. 2003, Miller et al. 2000) and ADAMs (Okabe and Cummins 2007), both male and female mice displayed no overt anomalies in both the sperm-egg fusion and adhesion. Thereafter, other factors participating in sperm-egg fusion have emerged, and contrary to expectations, most were found to be unnecessary, like integrins and ADAMs. From these studies, it is thought that in order to ensure the continuous success of the reproduction cycle in organisms, compatible pathways tuned by overlapping functions of multiple proteins should regulate the mechanism of fertilization. There will be more than one way for a sperm and egg to fuse and the network of multiple pathways may minimize the severity of the malfunction that can occur in sperm or eggs lacking a
Fig. 7.2 Players identified in sperm-egg fusion. Izumo has been found on the sperm membrane, and Izumo-null sperm show a defect in fusion with the egg cell membrane (Inoue et al. 2005). CD9 is expressed on the egg cell membrane and functions in fusion with the sperm (Kaji et al. 2000, Le Naour et al. 2000, Miyado et al. 2000). Two membrane proteins, Izumo and CD9, are essential for sperm-egg fusion in mammals. Direct interaction between CD9 and Izumo has not been identified, and the other sperm and egg factors involved in sperm-egg fusion remain unclear. After spermegg membrane fusion occurs, a sperm-specific phospholipase C, termed PLCzeta, triggers Ca2+ oscillations and initiates egg activation in mammals (Saunders et al. 2002)
7
Role of CD9 in Sperm-Egg Fusion
175
single gene. Genetic analyses have confirmed that unlike other factors, CD9 (Kaji et al. 2000, Le Naour et al. 2000, Miyado et al. 2000) and Izumo (Inoue et al. 2005) are not exchangeable and that they are considered to be hub proteins in sperm-egg fusion (Fig. 7.2).
7.3 CD9 and Its Role in Cell Function CD9 (also known as DRAP27, MRP-1, MIC3, and Tspan-29) was initially identified as a cell surface protein on platelets and immature B cells, and has now been widely identified in immune cells, growing oocytes, and eggs (full-grown, ovulated oocytes) (Hemler 2008). The CD9 gene encoding a 24-kDa protein is transcribed in all types of cells. The CD9 protein is primarily localized on cell membranes and partially on endosomes, and it is expected to be involved in cell–cell interaction because it has also been identified as an integrin-associated protein. CD9 has also been identified as a motility-related protein 1 (MRP-1), which plays a functional role as a tumor metastasis suppressor (Miyake et al. 1991). CD9 has some structural features: two extracellular loops, four transmembrane domains, and two short cytoplasmic domains; its functional domain is expected to be a large extracellular loop (LEL) (Fig. 7.3). CD9 associates with other membrane proteins via LEL in vitro. In humans and mice, commercially available antibodies recognize the structure of the regions linked by cysteine–cysteine interaction within LEL in a species-specific manner. As mentioned above, CD9 transcripts are expressed in all the cells and tissues examined, whereas its protein is specifically translated in some
Fig. 7.3 Structural features of tetraspanin CD9. CD9 is a member of the tetraspan-membrane protein family, termed tetraspanin, and its molecular mass is 24 kDa. The structural features of CD9 include four transmembrane domains, two extracellular loops, short and large extracellular loops (SEL and LEL), and two short cytoplasmic tails. CD9 has cysteine-cysteine-glycine (CCG) residues (amino acids 152–154) as a specific motif and two other cysteines within LEL. All tetraspanins share a CCG motif and several conserved cysteine residues in LEL
176
N. Kawano et al.
categorized cells. Due to the significantly higher amounts of CD9 in mesenchymal and embryonic stem cells than in fibroblastic cells, CD9 is considered useful as a cell surface marker for the isolation of undifferentiated cells from cell pools containing fibroblastic cells in mice and humans (Akutsu et al. 2009). Despite two decades of effort, the role of CD9 in immune cells in vivo is not very well understood, although three laboratories independently generated CD9-null mice (Kaji et al. 2000, Le Naour et al. 2000, Miyado et al. 2008). Unexpectedly, the CD9-null female mice were nearly infertile. While sperm from the CD9-null males was fertile, the CD9-null eggs were unable to fuse with the sperm. CD9 has been studied as one of the crucial factors in sperm-egg fusion in vivo and in vitro. In sperm-egg fusion, a functionally essential domain of CD9 was thought to be located within the LEL and to mediate the association of CD9 with other tetraspanins and other membrane proteins (Kaji et al. 2002, Zhu et al. 2002). Even though candidate proteins have been identified in non-gamete cells (Andre et al. 2007, Charrin et al. 2003), proteins related to egg-fusing abilities have not yet been found. Thus, the functional activity of CD9 remains elusive in sperm-egg fusion.
7.4 Tetraspanin CD9 belongs to a membrane protein superfamily, collectively termed “tetraspanin,” which encompasses 35 members in mammals (including CD9, CD37, CD53, CD63, CD81, CD82, and CD151) (Hemler 2008, Le Naour et al. 2006), 30 in nematodes (Tsp15) (Moribe et al. 2004), and 30 in flies (latebloomer) (Kopczynski et al. 1996, Todres et al. 2000). CD151 forms a stable complex with integrin alpha3beta1 and regulates integrin-mediated cell-substrate adhesion, neurite growth, and cell morphology (Shigeta et al. 2003, Yamada et al. 2008). In mice, CD81 is linked to infection of hepatocytes with the malaria parasite. Malaria sporozoites, a cell form that infects new hosts, are transmitted into the liver of the mammalian host through bites from infected mosquitoes. Malaria sporozoites failed to infect CD81-null mouse hepatocytes (Silvie et al. 2003) in vivo and in vitro. These results demonstrated that CD81 is linked to sporozoite entry into hepatocytes as a host factor. Deficiency of CD81 genes also leads to a reduction in the fecundity of female mice: CD81-null eggs had impaired sperm fusion ability, while the fertility of CD9-null mice was severely impaired (Rubinstein et al. 2006, Tanigawa et al. 2008). CD81 is expressed on CD9-null eggs, whereas CD9 is expressed on CD81-null eggs at an expression rate comparable with that of wild-type eggs. From these reports, CD9 and CD81 are speculated to independently function in sperm-egg fusion. Plants have more than 60 tetraspanin members, as do animals (Chiu et al. 2007, Huang et al. 2005). Tetraspanins and tetraspanin-like members have also been identified in fungi (Lambou et al. 2008), and the molecular mass of tetraspanin-like proteins (more than 200 kDa) is greater than that of tetraspanin (20–30 kDa). Fungal plant pathogens differentiate into an infection structure that is specialized for host penetration – the appressorium. On the basis of analysis of a non-pathogenic mutant from the rice blast fungus Magnaporthe grisea, tetraspanin PLS1 (MgPLS1) has been shown to control an appressorrial function essential for the penetration of fungus into host
7
Role of CD9 in Sperm-Egg Fusion
177
leaves (Clergeot et al. 2001). Furthermore, Colletotrichum lindemuthianum PLS1 (ClPLS1), a functional homologue of MgPLS1, has been identified, and a ClPLS1null mutant has been shown to be non-pathogenic to bean leaves as a result of a defect in the formation and/or positioning of the penetration pore (Veneault-Fourrey et al. 2005). These studies suggest that fungal tetraspanins control a conserved appressorial function that determines the correct localization of the site where the penetration peg emerges. The invasion of pathogenic fungi into leaves is an event closely related to membrane fusion events. Tetraspanins are closely related to other cell–cell membrane fusion events, such as muscle (Tachibana et al. 1999) and bone formation (Takeda et al. 2003, 2008), and virus infection (Garcia et al. 2005). Especially, tetraspanins have been reported to regulate cell–cell transmission of HIV-1 (Wiley and Gummuluru 2006). The presence of tetraspanins, CD9, CD63, CD81, and CD82 at HIV-1 budding sites and their enrichment in HIV-1 virions have been demonstrated. The tetraspanin-containing exocytosed vesicle-associated HIV-1 particles from dendritic cells were 10-fold more infectious than cell-free virus particles (Fig. 7.4). However, despite two decades of effort, the physiological
Fig. 7.4 Exosome-mediated HIV-1 transinfection. Dendritic cell-mediated HIV-1 transfection can be mediated by exocytosis of the HIV-1 particles captured. After endocytosis, the captured HIV-1 particles are targeted to a multi-vesicular endosomal body (MVB) in dendritic cells. The virus particles acquire exosomal antigens including tetraspanins (black bars) through the MVB. Although some of the MVB-localized virus fraction is targeted to the lysosome and degraded, fusion of MVB with the plasma membrane results in the release of virus particles along with exosomes
178
N. Kawano et al.
activities of tetraspanin are still elusive, and its fusogenic activity corresponding to fusogenic transmembrane proteins, such as syncytin identified in human placenta (Mi et al. 2000), and virus envelope proteins (Hernandez et al. 1996) remains unidentified.
7.5 Tetraspanin as a Component of Exosomes Tetraspanin has also been identified as a common component of membrane vesicles, termed exosomes (Simons and Raposo 2009), which are released outside immune cells, e.g., reticulocytes, B cells, and dendritic cells. Although cell-cultured media are know to contain nano-sized membrane fragments (Trams et al. 1981), exosomes cannot be structurally distinguished from the debris of dead cells (Couzin 2005). Recent studies of dendritic cells and B cells have shown that the exosomes are derived from living rather than dead cells; exosomes have been proven to play a significant role in the mediation of adaptive immune reactions to pathogens and tumors through the enhancement of antigen-specific T cell responses (Kesimer et al. 2009). Besides immune cells, the exosomes have reported to be released from a wide range of normal and malignant mammalian cell types, and their diameter is estimated to range from 50 to 90 nm. Exosomes can also contain both functional mRNA and microRNA, which can be shuttled from one cell to another, thereby influencing protein synthesis in recipient cells (Valadi et al. 2007). The protein composition of exosomes varies with the origin of cells, yet the exosomes contain a ganglioside GM3, two heat shock proteins (HSP70 and HSP90), and tetraspanins as common components. In eggs, two earlier observations suggest that CD9 plays a role in the organization of the egg cell membrane: (1) CD9 is transferred from the egg to the fertilizing sperm present in the perivitelline space, suggesting the involvement of a process similar to trogocytosis – a mechanism for the cell-to-cell contact-dependent transfer of membrane fragments from antigen-presenting cells to lymphocytes as demonstrated in immune responses for pathogens (Barraud-Lange et al. 2007); (2) CD9 deficiency alters the length and density of microvilli on the egg cell membrane (Runge et al. 2007). As mentioned above, tetraspanin is known to be a common component of exosomes, yet the involvement of exosome release in sperm-egg fusion was unknown until recently.
7.6 Lessons from “Living Eggs” Our recent study demonstrated the potential of enhanced green fluorescent proteintagged CD9 (CD9-EGFP) as a reporter protein for studying sperm-egg fusion in living mouse eggs (Miyado et al. 2008). Notably, in eggs just before fertilization, we showed that CD9-EGFP is significantly accumulated within a small interspace (perivitelline space) that completely surrounds the eggs and lies between the egg cell membrane and the extracellular matrix (zona pellucida). Consistent with the images from CD9-EGFP, immunoelectron microscopic analysis of wild-type eggs revealed
7
Role of CD9 in Sperm-Egg Fusion
179
Fig. 7.5 Electron microscopic images of CD9-containing egg exosomes. After ovulation, the wildtype eggs collected from the oviducts of C57BL/6 mice were treated with an anti-mouse CD9 antibody and 5 nm of gold particles conjugated with a secondary antibody. According to immunoelectron microscopic analysis, CD9-containing vesicular structures (asterisks), lipid bilayers that are formless and distinct from microvilli (M), are estimated to have diameters ranging from 50 to 200 nm, and are accumulated within the perivitelline space (PVS), an interspace between the zona pellucida (Z) and the cell membrane of an egg. (Modified from Miyado et al. 2008)
that CD9 is not only present in the perivitelline space but also incorporated into vesicles of varying size (50–200 nm in diameter) without a sectional profile of a typical lipid bilayer (Fig. 7.5). Furthermore, in opossums (Talbot and DiCarlantonio 1984) and humans (Dandekar et al. 1992), as well as in mice, membrane vesicles have been detected electron microscopically within the perivitelline space of their eggs. We also showed that the vesicles identified in mouse eggs share CD9, GM3, and HSP90 with exosomes; they are absent in eggs lacking CD9 and are reproduced by CD9-EGFP expression restricted to the eggs (Miyado et al. 2008). Taken together, our results suggest that (1) CD9-incroporated exosome-like vesicles are produced in mouse eggs and are released outside the egg cell membrane just before fertilization; (2) CD9 is essential for formation and/or release of the exosome-like vesicles in mouse eggs (Fig. 7.6).
7.7 Membrane Fusion and Exosomes We have also confirmed that CD9-containing exosome-like vesicles (hereafter referred to as egg exosomes) render sperms capable of fusing with CD9-null eggs (Fig. 7.7) (Miyado et al. 2008). This experiment demonstrates that co-incubation with wild-type eggs reverses a defect in the fusion of CD9-null eggs; this was replicated by two independent groups. CD9-null eggs do not generally fuse with eggs, but the co-existence of wild-type eggs results in 60–70% of the CD9-null eggs
180
N. Kawano et al.
Fig. 7.6 Schematic diagram for the release of CD9-containing egg exosomes from mouse eggs. Previous studies (Kaji et al. 2000, Le Naour et al. 2000, Miyado et al. 2000) demonstrated that eggs lacking tetraspanin CD9 display a defect when fusing with sperms. Moreover, a recent study (Miyado et al. 2008) showed that CD9-containing egg exosomes are released from mature oocytes (eggs), but not from immature oocytes, and share common components with canonical exosomes, identified as intercellular transporters of proteins and RNAs. The egg exosome release is correlated with microvilli formation on the egg cell membrane. In CD9-null eggs, both the egg exosome release and microvilli formation are dramatically impaired. Therefore, the egg exosomes are thought to be released from the eggs through the formation of microvilli
fusing with at least one sperm. Considering every possibility, we have no doubt that the above co-incubation experiments can be repeated. This result shows that sperms can fuse with CD9-null eggs with impaired microvilli via the egg exosomes of wild-type eggs: this means that the egg-derived vesicles, and not the microvilli, are essential for sperm-egg fusion. The close relation between egg exosomes and spermegg fusion raises the question of how egg exosomes facilitate fusion. Exosomes (hereafter referred to as canonical exosomes) derived from dendritic cells have been proven to be capable of inducing and enhancing antigen-specific T cell responses in vivo. According to a previous report (Valadi et al. 2007), canonical exosomes can also contain both functional mRNA and microRNA, which can be shuttled from one cell to another, affecting the recipient cell’s ability to produce protein. Moreover, it has been demonstrated that HIV-1 utilizes the canonical exosome biogenesis pathway for the formation of infectious particles, and that in macrophages, HIV1 assembles into an intracellular plasma membrane domain-containing tetraspanin (i.e., CD9, CD81, CD53, or CD63) (Fig. 7.4) (Garcia et al. 2005). Thus, the canonical exosomes may play at least two closely related roles in regulating cell function:
7
Role of CD9 in Sperm-Egg Fusion
181
Fig. 7.7 Co-incubation with wild-type eggs reverses defective fusion with sperm in CD9-null eggs in vitro. CD9-containing egg exosomes are released from wild-type eggs before any interaction with the sperm. In wild-type eggs, shortly after the sperm penetrates the egg’s perivitelline space, the egg exosomes are transferred onto the acrosome-reacted sperm head, and are predicted, given the nature of canonical exosomes previously identified, to fuse into the sperm. Then, a sperm fuses with the egg cell membrane. Interaction between the sperm and the exosomes is an essential step for sperm-fusing ability. In contrast, CD9-null eggs cannot release the egg exosomes, which are correlated with the formation of microvilli on the egg cell membrane. The sperm cannot fuse to the cell membrane of the CD9-null egg. When the zona pellucida is removed from the eggs, the sperm is able to interact with the egg exosomes released from wild-type eggs and can fuse with the CD9null egg. In fact, by co-incubation with wild-type eggs, the sperm can fuse with a similar number of CD9-null and wild-type eggs. Intracytoplasmic sperm injection (ICSI) is an in vitro fertilization procedure in which a single sperm head is injected directly into an egg. This procedure is most commonly used to overcome male infertility and fusion defects in CD9-null eggs. (Modified from Miyado et al. 2008)
shuttling proteins or RNAs from one cell to another and forming infectious particles. In fertilization, these two roles identified in the canonical exosomes may be required to regulate sperm-egg fusion. Evidence obtained recently shows that the egg-derived CD9 is translocated to the head of the fertilizing sperm (Barraud-Lange et al. 2007, Miyado et al. 2008); this translocation appears to be mediated by egg exosomes, similar to that of the canonical exosomes observed in immune cells. In fact, many microvesicles and amorphous substances were observed, using an electron microscope, on the sperm surface that penetrated into the perivitelline space (Toshimori et al. 1998). Nevertheless, the physiology of egg exosomes remains poorly characterized. Our findings underscore the relevance of CD9 for healthy and pathogenic cell–cell fusion processes, and may present useful strategies for influencing the cell-to-cell spread of specific viruses and fertilization ability.
182
N. Kawano et al.
Acknowledgments We are grateful to Dr. Masaru Okabe, Dr. Kiyotaka Toshimori, Dr. Chizuru Ito, Dr. Naokazu Inoue, and Dr. Eisuke Mekada for their critical discussions.
References Akutsu H, Miura T, Machida M et al (2009) Maintenance of pluripotency and self-renewal ability of mouse embryonic stem cells in the absence of tetraspanin CD9. Differentiation 78:137–142 Almeida EA, Huovila AP, Sutherland AE et al (1995) Mouse egg integrin alpha 6 beta 1 functions as a sperm receptor. Cell 81:1095–1104 Andre M, Morelle W, Planchon S et al (2007) Glycosylation status of the membrane protein CD9P-1. Proteomics 7:3880–3895 Barraud-Lange V, Naud-Barriant N, Bomsel M et al (2007) Transfer of oocyte membrane fragments to fertilizing spermatozoa. FASEB J 21:3446–3449 Blobel CP, Wolfsberg TG, Turck CW et al (1992) A potential fusion peptide and an integrin ligand domain in a protein active in sperm-egg fusion. Nature 356:248–252 Charrin S, Le Naour F, Labas V et al (2003) EWI-2 is a new component of the tetraspanin web in hepatocytes and lymphoid cells. Biochem J 373:409–421 Chen MS, Tung KS, Coonrod SA et al. (1999) Role of the integrin-associated protein CD9 in binding between sperm ADAM 2 and the egg integrin alpha6beta1: implications for murine fertilization. Proc Natl Acad Sci USA 96:11830–11835 Chiu WH, Chandler J, Cnops G et al (2007) Mutations in the TORNADO2 gene affect cellular decisions in the peripheral zone of the shoot apical meristem of Arabidopsis thaliana. Plant Mol Biol 63:731–744 Clergeot PH, Gourgues M, Cots J et al (2001) PLS1, a gene encoding a tetraspanin-like protein, is required for penetration of rice leaf by the fungal pathogen Magnaporthe grisea. Proc Natl Acad Sci USA 98:6963–6968 Couzin J (2005) Cell biology: the ins and outs of exosomes. Science 308:1862–1863 Dandekar P, Aggeler J, Talbot P (1992) Structure, distribution and composition of the extracellular matrix of human oocytes and cumulus masses. Hum Reprod 7:391–398 Evans JP (2001) Fertilin beta and other ADAMs as integrin ligands: insights into cell adhesion and fertilization. Bioessays 23:628–639 Garcia E, Pion M, Pelchen-Matthews A et al (2005) HIV-1 trafficking to the dendritic cell-T-cell infectious synapse uses a pathway of tetraspanin sorting to the immunological synapse. Traffic 6:488–501 He ZY, Brakebusch C, Fassler R et al (2003) None of the integrins known to be present on the mouse egg or to be ADAM receptors are essential for sperm-egg binding and fusion. Dev Biol 254:226–237 Hemler ME (2008) Targeting of tetraspanin proteins–potential benefits and strategies. Nat Rev Drug Discov 7:747–758 Hernandez LD, Hoffman LR, Wolfsberg TG et al (1996) Virus-cell and cell–cell fusion. Annu Rev Cell Dev Biol 12:627–661 Huang S, Yuan S, Dong M et al (2005) The phylogenetic analysis of tetraspanins projects the evolution of cell–cell interactions from unicellular to multicellular organisms. Genomics 86:674–684 Inoue N, Ikawa M, Isotani A et al (2005) The immunoglobulin superfamily protein Izumo is required for sperm to fuse with eggs. Nature 434:234–238 Jahn R, Lang T, Sudhof TC (2003) Membrane fusion. Cell 112: 519–533 Kaji K, Oda S, Miyazaki S et al (2002) Infertility of CD9-deficient mouse eggs is reversed by mouse CD9, human CD9, or mouse CD81; polyadenylated mRNA injection developed for molecular analysis of sperm-egg fusion. Dev Biol 247:327–334 Kaji K, Oda S, Shikano T et al (2000) The gamete fusion process is defective in eggs of Cd9deficient mice. Nat Genet 24:279–282
7
Role of CD9 in Sperm-Egg Fusion
183
Kesimer M, Scull M, Brighton B et al (2009) Characterization of exosome-like vesicles released from human tracheobronchial ciliated epithelium: a possible role in innate defense. FASEB J 23:1858–1868 Kopczynski CC, Davis GW, Goodman CS (1996) A neural tetraspanin, encoded by late bloomer, that facilitates synapse formation. Science 271:1867–1870 Lambou K, Tharreau D, Kohler A et al (2008) Fungi have three tetraspanin families with distinct functions. BMC Genomics 9:63 Le Naour F, Andre M, Boucheix C et al (2006) Membrane microdomains and proteomics: lessons from tetraspanin microdomains and comparison with lipid rafts. Proteomics 6:6447–6454 Le Naour F, Rubinstein E, Jasmin C et al (2000) Severely reduced female fertility in CD9-deficient mice. Science 287:319–321 Mi S, Lee X, Li X et al (2000) Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature 403:785–789 Miller BJ, Georges-Labouesse E, Primakoff P et al (2000) Normal fertilization occurs with eggs lacking the integrin alpha6beta1 and is CD9-dependent. J Cell Biol 149:1289–1296 Miyado K, Yamada G, Yamada S et al (2000) Requirement of CD9 on the egg plasma membrane for fertilization. Science 287:321–324 Miyado K, Yoshida K, Yamagata K et al (2008) The fusing ability of sperm is bestowed by CD9containing vesicles released from eggs in mice. Proc Natl Acad Sci USA 105:12921–12926 Miyake M, Koyama M, Seno M et al (1991) Identification of the motility-related protein (MRP1), recognized by monoclonal antibody M31-15, which inhibits cell motility. J Exp Med 174:1347–1354 Moribe H, Yochem J, Yamada H et al (2004) Tetraspanin protein (TSP-15) is required for epidermal integrity in Caenorhabditis elegans. J Cell Sci 117:5209–5220 Okabe M, Cummins JM (2007) Mechanisms of sperm–egg interactions emerging from genemanipulated animals. Cell Mol Life Sci 64:1945–1958 Rosen GD, Sanes JR, LaChance R et al (1992) Roles for the integrin VLA-4 and its counter receptor VCAM-1 in myogenesis. Cell 69:1107–1119 Rubinstein E, Ziyyat A, Prenant M et al (2006) Reduced fertility of female mice lacking CD81. Dev Biol 290:351–358 Runge KE, Evans JE, He ZY et al (2007) Oocyte CD9 is enriched on the microvillar membrane and required for normal microvillar shape and distribution. Dev Biol 304:317–325 Saunders CM, Larman MG, Parrington J et al (2002) PLC zeta: a sperm-specific trigger of Ca(2+) oscillations in eggs and embryo development. Development 129:3533–3544 Shigeta M, Sanzen N, Ozawa M et al (2003) CD151 regulates epithelial cell–cell adhesion through PKC- and Cdc42-dependent actin cytoskeletal reorganization. J Cell Biol 163:165–176 Silvie O, Rubinstein E, Franetich JF et al (2003) Hepatocyte CD81 is required for Plasmodium falciparum and Plasmodium yoelii sporozoite infectivity. Nat Med 9:93–96 Simons M, Raposo G (2009) Exosomes – vesicular carriers for intercellular communication. Curr Opin Cell Biol 21:575–581 Tachibana I, Hemler ME (1999) Role of transmembrane 4 superfamily (TM4SF) proteins CD9 and CD81 in muscle cell fusion and myotube maintenance. J Cell Biol 146:893–904 Takeda Y, He P, Tachibana I et al (2008) Double deficiency of tetraspanins CD9 and CD81 alters cell motility and protease production of macrophages and causes chronic obstructive pulmonary disease-like phenotype in mice. J Biol Chem 283:26089–26097 Takeda Y, Tachibana I, Miyado K et al (2003) Tetraspanins CD9 and CD81 function to prevent the fusion of mononuclear phagocytes. J Cell Biol 161:945–956 Talbot P, DiCarlantonio G (1984) Ultrastructure of opossum oocyte investing coats and their sensitivity to trypsin and hyaluronidase. Dev Biol 103:159–167 Tanigawa M, Miyamoto K, Kobayashi S et al (2008) Possible involvement of CD81 in acrosome reaction of sperm in mice. Mol Reprod Dev 75:150–155 Todres E, Nardi JB, Robertson HM (2000) The tetraspanin superfamily in insects. Insect Mol Biol 9:581–590
184
N. Kawano et al.
Toshimori K, Saxena DK, Tanii I et al (1998) An MN9 antigenic molecule, equatorin, is required for successful sperm-oocyte fusion in mice. Biol Reprod 59:22–29 Trams EG, Lauter CJ, Salem N et al (1981) Exfoliation of membrane ecto-enzymes in the form of micro-vesicles. Biochim Biophys Acta 645:63–70 Valadi H, Ekstrom K, Bossios A et al (2007) Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol 9:654–659 Veneault-Fourrey C, Parisot D, Gourgues M et al (2005) The tetraspanin gene ClPLS1 is essential for appressorium-mediated penetration of the fungal pathogen Colletotrichum lindemuthianum. Fungal Genet Biol 42:306–318 Wiley RD, Gummuluru S (2006) Immature dendritic cell-derived exosomes can mediate HIV-1 trans infection. Proc Natl Acad Sci USA 103:738–743 Yamada M, Tamura Y, Sanzen N et al (2008) Probing the interaction of tetraspanin CD151 with integrin alpha 3 beta 1 using a panel of monoclonal antibodies with distinct reactivities toward the CD151-integrin alpha 3 beta 1 complex. Biochem J 415:417–427 Zhu GZ, Miller BJ, Boucheix C et al (2002) Residues SFQ (173-175) in the large extracellular loop of CD9 are required for gamete fusion. Development 129:1995–2002
Chapter 8
Gamete Binding and Fusion Young-Joo Yi, Shawn W. Zimmerman, and Peter Sutovsky
Abstract Successful mammalian fertilization results in the union of two gametes, a spermatozoon and a mature oocyte. Membrane fusion events are essential for at least two distinct steps of the fertilization process: (i) the vesiculation of the acrosomal surface membranes during sperm acrosomal exocytosis (AE), induced by sperm binding to the egg-coat, and (ii) fusion of the oocyte plasma membrane, the oolemma, with the sperm plasma membrane that occurs after AE and sperm-egg coat penetration. The rearrangement of sperm plasma membrane domains/membrane lipid raft formation during sperm capacitation in the female reproductive tract is a priming step that enables the fusion and vesiculation of outer acrosomal membranes during AE. The membrane fusion/vesiculation events of AE seem to share similarities with synaptic vesicle fusion, assisted by the membrane proteins of the SNARE hypothesis. The AE exposes the transmembrane receptors on the sperm head equatorial segment in preparation for sperm-oolemma adhesion and fusion. Gene ablation studies indicate that the tetraspanin family proteins CD9 and CD81 on the oolemma interact with the superglobulin family protein IZUMO on the sperm plasmalemma to mediate sperm-oolemma adhesion in mammals. The fusogenicity of IZUMO has not been established, so the involvement of this system in the actual membrane fusion part of sperm–oolemma interaction remains open. Interactions of ADAM family proteins on sperm plasma membrane with oolemma integrins appear non-essential during sperm-oolemma fusion, but integrins may play a supporting role via sustenance of the tetraspanin web in the oocyte cortex. Spermoolemma binding may be reinforced by a cast of other receptors found on the surface of the sperm head (e.g. CRISP and MN9). The present chapter reviews recent progress in the study of these fundamental factors of gamete membrane fusion during mammalian fertilization.
P. Sutovsky (B) Division of Animal Science, University of Missouri-Columbia, Columbia, MO 65211, USA; Departments of Obstetrics, Gynecology & Women’s Health, University of Missouri-Columbia, Columbia, MO 65211, USA e-mail:
[email protected]
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_8, C Springer Science+Business Media B.V. 2011
185
186
Y.-J. Yi et al.
Keywords Acrosome · CD9 · CD81 · egg · eqatorin MN9 · exocytosis · fertilization · Izumo · membrane fusion · SNAREs · sperm · vesiculation Abbreviations ADAM AE AFAF CD CEA CRISP IAM ICSI IgSF KO LEL NSF OAM PVS PSG PM PTGFRN PTP SEL SNAP SNARE t-SNARE VAMP v-SNARE ZP
A disintegrin and a metalloprotease Acrosomal exocytosis Acrosome formation associated factor Cluster of differentiation Carcinoembryonic antigen Cysteine-rich secretory protein Inner acrosomal membrane Intracytoplasmic sperm injection Immunoglobulin superfamily (protein) Knock-out Large extracellular loop N-ethylmaleimide-sensitive factor Outer acrosomal membrane Perivitelline space Pregnancy-specific glycoprotein Plasma membrane Prostaglandin F2 receptor negative regulator Protein tyrosine phosphatase Small extracellular loop Soluble NSF attachment protein SNAP receptors Target-SNARE Vesicle associated membrane protein Vesicle-SNARE Zona pellucida
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . Membrane Fusion Events During Acrosomal Exocytosis . . . . Essential Role of CD9 in Sperm-Egg Binding . . . . . . . . . IZUMO-the Candidate Sperm Partner of Oolemma Tetraspanins Integrin–Disintegrin Interactions in Sperm-Egg Binding . . . . Eqatorin MN9 and other Sperm Surface Ligands Implicated in Sperm-Oolemma Fusion . . . . . . . . . . . . 8.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.1 8.2 8.3 8.4 8.5 8.6
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
187 188 189 191 192
. . . . . . . . . . . . . . . . . . . . . . . .
194 195 197
8
Gamete Binding and Fusion
187
8.1 Introduction Successful mammalian fertilization requires the union of two gametes, an oocyte and a spermatozoon. Oocytes reach their full fertilizing potential during the gonadotropin-induced meiotic maturation in the female gonad, the ovary. In contrast, spermatozoa that have completed their morphogenesis in the seminiferous epithelium have to leave the male gonad, the testis, to acquire their fertilizing potential. They do so during sperm passage through the epididymis, an accessory gland in which the spermatozoa undergo maturation and storage. After ejaculation, spermatozoa temporarily mix with seminal plasma contributed by secretions of multiple sex accessory glands, and become fully competent for fusion with the oocyte as they move up the female reproductive tract. They reach full fertilizing ability after they transiently bind to and subsequently detach from the epithelium of the oviductal sperm reservoir during a process termed sperm capacitation. During capacitation, sperm plasma membrane undergoes an extensive remodeling thought to increase its fluidity and fuseability [reviewed by (Gadella and Visconti 2006, Suarez 2002)]. There are two important membrane fusion events involved in mammalian fertilization: The acrosomal exocytosis (AE), induced by the sperm acrosomal surface-binding to the glycoprotein-rich egg coat, zona pellucida (ZP), and the process of sperm-oolemma fusion that ensues after AE and sperm ZP-penetration. During AE, the plasma membrane overlying the sperm acrosome fuses with the underlying outer acrosomal membrane to promote the vesiculation of the acrosomal surface membranes and the release/exposure of the acrosomal matrix. The AE requires an extensive membrane domain rearrangement during capacitation, resulting in the formation of membrane lipid rafts. The process of acrosomal membrane vesiculation, giving rise to the acrosomal shroud, requires membrane vesicle fusion that may share some similarities with synaptic vesicle fusion, assisted by membrane proteins of the SNARE hypothesis, such as the synaptobrevin/VAMP and syntaxin. The completion of acrosomal exocytosis enables the spermatozoon to penetrate through ZP. At the same time, AE exposes the sperm surface antigens involved in the second major membrane fusion event of fertilization, the sperm-oolemma fusion, followed by sperm incorporation in the oocyte cytoplasm. Sperm-oolemma fusion relies on the interaction of complementary receptors on the respective sperm and oocyte plasma membranes, the sperm plasmalemma and oolemma, respectively. While oolemma integrin-sperm disintegrin interactions have been studied extensively, more definitive evidence now points to oolemma tetraspanin-sperm superglobulin mediated interactions being essential for spermoolemma adhesion and fusion. These interactions will be discussed in detail in the present chapter.
188
Y.-J. Yi et al.
8.2 Membrane Fusion Events During Acrosomal Exocytosis Acrosomal exocytosis is an irreversible exocytotic event triggered in the capacitated, hyperactivated spermatozoa by contact of the sperm acrosomal surface with the branched glycans decorating the molecules of sperm receptor protein(s) on the surface of sperm zona pellucida (Sutton-Smith et al. 2007). The sperm acrosome is a specialized organelle assembled from the Golgi complex and perinulear theca-derived components. Topologically, the acrosome is composed of the outer acrosomal membrane (OAM) and inner acrosomal membrane (IAM) surrounding the acrosomal matrix. Acrosomal exocytosis encompasses the fusion of the sperm plasma membrane with the outer acrosomal membrane, which is thought to produce the hybrid membrane vesicles of the acrosomal shroud (Yanagimachi 1994). The acrosomal matrix contains structural proteins as well as multiple proteases, which are vesicle-associated or soluble components thought to be involved in sperm-ZP binding and penetration (Buffone et al. 2008, Foster et al. 1997, Kim et al. 2001a; Olson and Winfrey 1994, Westbrook-Case et al. 1995). Some of the components of the acrosomal matrix may associate with vesicles forming the acrosomal shroud, a transient structure surrounding the ZP-bound sperm head, or with the surface of the IAM, which remains intact after AE. Some of the protein components of the acrosomal matrix that are released or exposed by AE appear to undergo post-translational modifications by proteolytic processing. For instance, proacrosin, a major component of the acrosomal matrix, is cleaved into acrosin during acrosomal exocytosis (Kim et al. 2001b). The AE, once thought to be a rapid, all-or-nothing exocytotis event is now being re-evaluated as a stepwise process, during which the layers or differentially soluble components of the acrosomal matrix are released one at the time. Kim and Gerton (2003) described four distinct stages of acrosomal exocytosis in the mouse. The first stage is characterized by an intact OAM and plasma membrane (PM). During the second stage, the OAM and PM begin vesiculation. The third stage coincides with the progressive vesiculation of OAM/PM and dispersion of soluble acrosomal matrix components from the sperm head. The fourth stage marks the completion of OAM/PM vesiculation and the loss of most of the acrosomal matrix. Once the soluble components dissipate, only the acrosomal proteins associated with the IAM participate in sperm-ZP interactions during and beyond secondary sperm-ZP binding (Buffone et al. 2008). Acrosomal exocytosis is initiated by Ca2+ influx fueling an intra-acrosomal internal calcium increase (Breitbart 2002). This mechanism is similar to the induction of synaptic vesicle-plasma membrane fusion during neurotransmitter secretion in the neurons, which is also calcium-regulated. The GTPases of the Rab family (Novick and Zerial 1997, Zerial and McBride 2001), and SNARE proteins (Chen and Scheller 2001, Gerst 1999) that participate in neurotransmitter secretion, have also been identified in spermatozoa. The SNARE hypothesis of synaptic vesicle fusion predicts that the soluble N-ethylmaleimide-sensitive factor (NSF)attachment proteins (SNAPs), including syntaxin (Bennett et al. 1992) and SNAP-25 (Oyler et al. 1989), regulate membrane vesicle fusion by associating with their specific receptors, the SNARE proteins (Terrian and White 1997, Weimbs et al.
8
Gamete Binding and Fusion
189
1997, 1998). There are two types of SNARE receptors, the t-SNAREs, on the target plasma membrane and the v-SNAREs on the secretory vesicles, such as the synaptobrevin/vesicle-associated membrane protein (VAMP) (Schulz et al. 1997). It is believed that the interaction between a t-SNARE and a v-SNARE results in the binding and fusion of vesicle membrane with the plasma membrane (Rothman 1994). Synaptobrevin/VAMP, syntaxin 1, and the SNARE hypothesisassociated calcium sensor synaptotagmin I have been detected in the acrosomes of sea urchin and mammalian spermatozoa and shown to colocalize with the acrosomal membrane vesicles during AE (Ramalho-Santos et al. 2000, Schulz et al. 1997). Both anti-syntaxin and anti-VAMP antibodies inhibited bovine ZP-induced AE and fertilization without affecting sperm-ZP binding (Ramalho-Santos et al. 2000). Multiple SNARE proteins, including VAMP2, SNAP23, SNAP25, and several syntaxins were also detected in the human sperm acrosome, where the blocking of VAMP by botulinum neurotoxin or saturation of spermatozoa with bacterially expressed SNAP proteins inhibited AE (Tomes et al. 2002). The streptolysin O-permeabilized human sperm model has been instrumental in the study of this pathway during acrosomal exocytosis (Yunes et al. 2000). The general model of SNARE involvement in AE can be summarized as follows, based on the description given by De Blas et al. (2005): Prior to AE, the SNARE proteins are locked in heteromeric cis complexes. Upon Ca2+ entry into the acrosome, Rab3 is activated and triggers NSF/SNAP-dependent disassembly of cis SNARE complexes. Monomeric SNAREs in the plasma membrane (Hohne-Zell and Gratzl 1996) and the OAM can thus reassemble into loose trans complexes that are conducive to membrane fusion. In the final step, Ca2+ is released from the acrosome, triggering the final steps of membrane fusion that require fully assembled trans SNARE complexes and synaptotagmin. This entire process of snare complex disassembly and reassembly may be triggered by dephosphorylation of the NSF by protein tyrosine phosphatase PTP1B at an early stage of AE (Zarelli et al. 2009).
8.3 Essential Role of CD9 in Sperm-Egg Binding The transmembrane 4 superfamily proteins, the tetraspanins, are widely distributed in various tissues and cell types, including the gametes. Tetraspanins contain two extracellular domains (the large and the small extracellular loops, LEL and SEL, respectively) which function as specific receptors, transmembrane anchors, interacting with transmembrane parts of other proteins (Lefevre et al. 2010). In turn, the cytoplasmic N-terminal and C-terminal tails of tetraspanins interact with cytoskeletal elements such as microfilaments and microtubules, and with cytoskeleton-associated molecules involved in signal transduction processes. The tetraspanins take their name from four transmembrane domains including several conserved amino acid residues (Lefevre et al. 2010). Tetraspanins constitute a large transmembrane complex, the tetraspanin web (Boucheix and Rubinstein 2001, Rubinstein et al. 1996), in which they associate with various cytoskeletal,
190
Y.-J. Yi et al.
transmembrane and intracellular signaling proteins (Boucheix and Rubinstein 2001, Delaguillaumie et al. 2002). Tetraspanins have been implicated in cell proliferation, adhesion, motility and signaling (Charrin et al. 2009, Hemler 2003, Lefevre et al. 2010). The tetraspanin superfamily contains 32 identified members in humans (Boucheix and Rubinstein 2001, Hemler 2003, Levy and Shoham 2005). In particular, three different tetraspanins, CD9, CD81 and CD151, have been detected, among other cell types, in the mammalian oocytes (Andria et al. 1992, Chen et al. 1999b; Neilson et al. 2000). Gene ablation studies in mice unequivocally demonstrated the essential requirement of tetraspanins for the fertilization process. The Cd9–/– female mice showed severely reduced sperm-egg adhesion/fusion (99.4% of recovered eggs failed to fuse with sperm and form pronuclei during IVF) and near complete infertility (Kaji et al. 2000, Le Naour et al. 2000, Miyado et al. 2000). Similarly, the Cd81–/– mice had reproductive impediments; however, their subfertile phenotype was milder than that of Cd9–/– mice (Higginbottom et al. 2003). Anti-CD81 antibodies suppressed sperm-oolemma binding and fusion (Takahashi et al. 2001) as did recombinant CD81 LEL (Higginbottom et al. 2003). The overexpression of Cd81 after microinjection of CD81 mRNA recovered the fusion ability of Cd9–/– eggs. Microinjections of mouse or human CD9 mRNA efficiently restored the fusion ability of Cd9–/– eggs up to 76 and 81% fertilization rate, respectively (Kaji et al. 2002). These results suggest that CD9 and CD81 play complementary, but not necessarily redundant, roles in sperm–egg fusion. Although Cd151 knockout mice did not display a fertility defect, mouse anti-CD151 antibody partially inhibited sperm–egg fusion in humans (Ziyyat et al. 2006). Recently, it has been suggested that CD9 could support plasma membrane functionality through CD9-associated, immunoglobulin superfamily protein IgSF8 (EWI-2) in the mouse oocyte (Glazar and Evans 2009). Accordingly, the expression level of EWI-2 on the Cd9–/– oocyte surface is <10% of the wild-type level. Hence, the severe reduction in EWI-2 activity may contribute to the loss of fusion ability in the Cd9–/– oocyte (He et al. 2009). In the migrating leukocytes, the EWI-2 and EWI-F proteins seem to interact directly with CD9 and CD81 and, through their direct binding with the microfilament associated ezrin-radixin-moesin proteins, act as linkers to connect tetraspanin-associated microdomains to the actin cytoskeleton (Sala-Valdes et al. 2006). Pertinent to fertilization studies in vitro, tetraspanin CD9 may also be the receptor for pregnancy-specific glycoprotein 17 (PSG17) in macrophages (Waterhouse et al. 2002). PSG17, a soluble member of the carcinoembryonic antigen (CEA) subfamily of the Ig superfamily, is produced by the placenta and binds to macrophages in a CD9-dependent manner (Waterhouse et al. 2002). In an in vitro assay, the PSG17coated beads adhered to wild type, fertilization competent mouse oocytes but not to Cd9–/– oocytes; pre-incubation of wild type oocytes with soluble, recombinant PSG17 significantly inhibited sperm–egg fusion (Ellerman et al. 2003). It is possible that PSG17 displaces a CD9-binding partner from its association with CD9 on the oocyte surface, or competes with it for CD9/oolemma binding. While a useful tool for in vitro asays, PSG17 is not expressed on the sperm surface and thus is not likely to mediate sperm–oocyte interaction in vivo (Stein et al. 2004).
8
Gamete Binding and Fusion
191
Interestingly, spermatozoa may be able to acquire CD9 from the oolemma during fertilization in a process perhaps similar to the cell–cell exchange of small membrane patches by trogocytosis (Barraud-Lange et al. 2007, Rubinstein et al. 2006). If CD9 is transferred from the oocyte membrane to the spermatozoa present in the perivitelline space (PVS), it could induce molecular reorganization of the sperm plasma membrane and convey the ability for homophilic CD9-CD9 binding to it. In support of the above observations, Miyado et al. (2008) have recently reported CD9 transfer from the oocyte to the spermatozoon by exosome-like vesicles released into the PVS. In addition, they proposed that the release of CD9-containing vesicles from eggs before fertilization supports the sperm-fusing ability, rendering wild type spermatozoa competent to fuse with Cd9–/– eggs (Miyado et al. 2008). Altogether, the above reports provide substantial evidence that CD9 has an important role in sperm-egg binding during mammalian fertilization.
8.4 IZUMO-the Candidate Sperm Partner of Oolemma Tetraspanins IZUMO, named after a Shinto fertility shrine, has been identified as a sperm surface antigen recognized by mouse antibodies having the ability to inhibit spermegg fusion (Okabe et al. 1987). The protein recognized by these antibodies was expressed exclusively in the testis, and it was detectable by immunofluorescence only in the acrosome-reacted spermatozoa. It is generally assumed that the sperm proteins that interact with oolemma receptors during fertilization are predominantly localized on the equatorial segment of the sperm head, which is covered by the posterior segment of the acrosomal cap prior to AE. Some such antigens thus may become exposed only after AE. This may hold true for IZUMO, even though its localization in the spermatozoa found in the periviteline space after penetration through the murine egg coat is not restricted to the equatorial segment (Okabe et al. 1987). The size of the IZUMO protein is 56 kDa in the mouse and 37 kDa in humans (Inoue et al. 2005). The Izumo gene encodes an immunoglobulin superfamily (IgSF), type I transmembrane protein with an extracellular immunoglobulin domain that contains one putative glycosylation site. Izumo–/– females are fertile, but Izumo–/– males are sterile, with normal mating behavior and ejaculation (Inoue et al. 2005). Spermatozoa accumulate in the perivitelline space of oocytes collected in females mated with Izumo–/– males, indicating a defect of sperm-oolemma adhesion/fusion but not of sperm migration or sperm-ZP interactions. IZUMO may also play a role in the sperm-oolemma adhesion and/or fusion in humans since a polyclonal antibody against it blocks binding of human spermatozoa to zona-free hamster oocytes (Inoue et al. 2005). It is not clear yet whether IZUMO interacts directly with CD9 or with a CD9-associated molecule. IZUMO and other IgSF proteins feature extracellular domains with a variable number of immunoglobulin (Ig)-like repeats that can participate in cell–cell adhesion (Aricescu and Jones 2007, Barclay 2003). In addition to IZUMO expression on
192
Y.-J. Yi et al.
the sperm head, two tetraspanin-interacting IgSF members, IgSF8 and Prostaglandin F2 Receptor Negative Regulator (PTGFRN; also known as EWI-F and CD9-P-1, respectively) have recently been reported to be expressed by mouse eggs (Rubinstein et al. 2006, Runge et al. 2007). IgSF8 associates with CD9 and/or CD81 in several cell types, and functions in cell aggregation, cell motility and translocation of tetraspanin proteins to filopodia (Charrin et al. 2003, Kolesnikova et al. 2004, Sala-Valdes et al. 2006, Stipp et al. 2001). It is possible that IgSF8 participates in mouse gamete interactions during fertilization, as demonstrated by discrete effects of antibody-mediated perturbation of CD9 and IgSF8. An anti-IgSF8 antibody had moderate inhibitory effects on sperm-egg binding, whereas an anti-CD9 antibody significantly inhibited sperm-egg fusion (Glazar and Evans 2009). Three novel IZUMO-related proteins showing an N-terminal domain with significant homology to the N-terminal domain of IZUMO were collectively designated as IZUMO domain proteins (Ellerman et al. 2009). While the name IZUMO1 was reserved for the protein originally described by Okabe’s group, the new proteins were named IZUMO 2, IZUMO3 and IZUMO4. IZUMO 1–3 are transmembrane proteins expressed in the testis, and IZUMO4 is a soluble protein expressed in the testis and in other tissues. Western blot analysis showed that IZUMO1, 3, and 4 formed protein complexes on the sperm head. IZUMO1 formed several larger complexes while IZUMO3 and 4 formed a single larger complex. The formation of IZUMO1 complexes was conserved among rodents, as it was observed in mouse, rat and hamster spermatozoa (Ellerman et al. 2009). Consequently, it has been proposed that IZUMO1 plays a role in gamete adhesion/fusion by organizing or stabilizing a molecular complex on the sperm plasma membrane (Ellerman et al. 2009). So far, the proteins that interact with IZUMO proteins within this complex have not been identified. IZUMO has also been explored as a potential contraceptive target. An antibody against the Ig-like domain of IZUMO reduced the rate of mouse sperm-zona free egg adhesion/fusion by ∼75%, with no effect on sperm motility (Wang et al. 2009). Although the total number of animals in this study was small, the female mice mated following an immunization with the same recombinant Iglike domain showed pregnancy rates reduced by half, and an average litter size of only 2.6 pups, compared to 8.8 pups/litter in control (Wang et al. 2009). This result indicates the Ig-like domain of IZUMO might have properties that qualify it as a target for the development of a contraceptive vaccine.
8.5 Integrin–Disintegrin Interactions in Sperm-Egg Binding In the 1990s, the disintegrin-integrin system has been a favorite candidate for spermoolemma binding and fusion machinery during mammalian fertilization. In light of recent work on tetraspanins and IZUMO, and studies of trangenic mice lacking genes of integrin-disintegrin hypothesis, it appears that integrins and disintegrins of the gamete surface play a non essential, yet facilitating role during mammalian fertilization.
8
Gamete Binding and Fusion
193
The primary candidate for the sperm component of gamete fusion machinery was Fertilin β/ADAM2, a member of ADAM (A Disintegrin and A Metalloprotease domain) family of disintegrins, the integrin binding, membrane bound metalloproteinase enzymes isolated originally from snake venom. The ADAMs have the ability to release a variety of proteins from the cell plasma membrane, for example causing apoptosis by shedding of FAS ligand (van Goor et al. 2009). At least eight different ADAM–integrin interactions have been described in literature, including the interaction of ADAM15 and ADAM23 with αγβ3 (Cal et al. 2000, Nath et al. 1999, 2000, Zhang et al. 1998), ADAM15 with α5β1 (Nath et al. 1999), ADAM15 and ADAM12 with α9β1 (Eto et al. 2000), ADAM9 with αγβ5 (Zhou et al. 2001) and fertilin β (also known as ADAM2) and ADAM9 with α6β1 (Bigler et al. 2000, Chen and Sampson 1999, Chen et al. 1999a; Evans 2001, Nath et al. 2000). ADAMs share a disintegrin domain with a conserved RGD, QCD or ECD sequence for binding to integrins, the extracellular matrix receptors on the plasma membrane. These proteins which disturb integrin-mediated adhesion, are called disintegrins (Gould et al. 1990). Several members of the ADAM have been identified in the testis. Five of them, fertilin β, cyritestin, ADAM5, ADAM16, ADAM18 can be detected in male germ cells and mature sperm [reviewed by (Evans 2001)]. Synthetic peptides and anti-disintegrin loop antibodies modeled after fertilin β and cyritestin disrupt sperm interaction with oolemma in vitro (Linder and Heinlein 1997, Yuan et al. 1997). However, male knockout mice lacking fertilin β or cyritestin encoding genes have only partially reduced fertility. Fertilin β-null spermatozoa showed decreased sperm-olemma binding, and cyritestin –/– sperm also showed low sperm binding to the zona pellucida (Cho et al. 1998, Shamsadin et al. 1999). Fertilin β is localized in the sperm head equatorial region (Yuan et al. 1997) and the soluble extracellular domains of fertilin α and fertilin β were shown to bind to the microvillar region of murine eggs (Bigler et al. 2000, Evans et al. 1997). Fertilin β-like peptides and purified soluble fertilin β protein inhibited sperm binding to olemma (Bigler et al. 2000, Evans et al. 1997, 1998, Yuan et al. 1997). Integrins, the putative binding partners of sperm disintegrins on the oolemma, are heterodimeric complexes composed of an α subunit (∼1,000 amino acids) and a β subunit (∼800 amino acids). Alpha and beta subunits can make 24 different α/β heterodimeric combinations, and specific α/β subunit combinations promote preferential binding to specific extracellular ligands. Several different integrin subunits (α2, α3, α4, α5, α6, αv, αM, β1, β2, β3, β5) have been identified in mammalian eggs (human, mouse, hamster, pig) [(Evans et al. 1995, Linfor and Berger 2000, Takahashi et al. 2000, Tarone et al. 1993); reviewed in (Evans 2001)]. Consequently, oolemma integrins are the likely receptors for sperm ADAMs, which interact with at least five integrins: αvβ3, αvβ5, α5β1, α6β1, and α9β1. The anti-α6 functionblocking monoclonal antibody GoH3 (Bigler et al. 2000, Chen et al. 1999a; Nath et al. 2000) and chemical cross-linking experiments with a synthetic peptide containing the fertilin β disintegrin loop sequence RLAQDECDVTE (Chen and Sampson 1999), showed that mouse fertilin β and ADAM9 (meltrin γ) interact with α6β1. An antibody against integrin α6β1 blocked sperm-egg fusion and binding of fertilin β peptides and recombinant fertilin β protein to mouse oocytes (Almeida
194
Y.-J. Yi et al.
et al. 1995, Chen and Sampson 1999, Chen et al. 1999b; Takahashi et al. 2001). The α6β1 integrin mediated the binding of cyritestin’s disintegrin domain to the egg surface (Takahashi et al. 2001). Despite the above data, the tissue-specific gene ablation studies did not confirm the requirement of integrins for murine fertilization. For example, oocytes lacking the α6 integrin subunit could be fertilized in vitro at normal rates (Miller et al. 2000). Also, mice producing oocytes lacking the β1 integrin subunit are fertile and exhibit normal sperm-oolemma binding and fusion in vitro (He et al. 2003). Therefore, the roles of integrins in sperm-oolemma adhesion and/or fusion need further clarification. One possibility is that integrins are necessary for the sustenance of the tetraspanin web, thus supporting the ability of CD9 and CD81 to interact with their ligands on the sperm plasma membrane. Supporting this notion, sperm–egg interactions and fertilin β binding to oolema were inhibited by anti-CD9 antibody (Chen et al. 1999b). CD9 associates with α6β1 on the egg surface (Miyado et al. 2000), and this interaction occurs through CD151 in human eggs (Neilson et al. 2000, Ziyyat et al. 2006). CD9 may control the binding of fertilin β receptor on the egg surface. This is supported by the activities of the tetraspanins CD151 and CD81 regulating the adhesion strengthening of the integrins α6β1 and α4β1 in monocytes, NIH 3T3 cells and primary murine B-cells (Feigelson et al. 2003, Lammerding et al. 2003). CD151 regulates the function of integrin α3β1 and α6β1 (Lammerding et al. 2003, Nishiuchi et al. 2005). CD81 is also believed to control the function of the integrin α4β1 (Feigelson et al. 2003). In addition to being implicated in sperm-oolemma adhesion/fusion, egg integrins and their ligands on the sperm surface may have a role in oocyte activation in mammals (Baessler et al. 2009, Campbell et al. 2000) and Xenopus (Iwao and Fujimura 1996, Shilling et al. 1998), perhaps in symphony with the sperm-borne oocyte-activating factors (“sperm factor” or SOAF) released in the oocyte cytoplasm after sperm-oolemma fusion. Such a notion is supported by experiments involving intracytoplasmic sperm injection (ICSI): When oocytes are fertilized by ICSI, the sperm-oolemma fusion step of fertilization is bypassed and oocytes are activated solely by the release of the soluble “sperm factor”. However, such a mode of activation is not 100% efficient as it fails to trigger proper anti-polyspermy defense (Maleszewski et al. 1996), a hallmark of completed oocyte activation process. Incorporation of sperm plasma membrane components such as disintegrins in the oocyte cytoplasm could be necessary to trigger a competent anti-polyspermy defense.
8.6 Eqatorin MN9 and other Sperm Surface Ligands Implicated in Sperm-Oolemma Fusion Equatorin/mMN9 antigen/MN9 is an antigen identified by monoclonal antibody mMN9 in the equatorial segment of the sperm head of various mammalian species (Toshimori et al. 1992). Equatorin is a 38- to 48-kDa protein complex in mice,
8
Gamete Binding and Fusion
195
and a 48-kDa protein in rats (Toshimori et al. 1992). Equatorin is exposed by acrosomal exocytosis (Manandhar and Toshimori 2001), suggesting that it may play a role in sperm-oolemma binding, which appears to be initiated by adhesion between equatorial segment and oolemma microvilli (Bedford and Cooper 1978, Oura and Toshimori 1990, Yanagimachi 1994). The mMN9 antibody inhibits mouse in vitro fertilization without affecting sperm motility, sperm-zona binding, or zona penetration (Toshimori et al. 1998). Recently, Yamatoya et al. (2009) reported the gene cloning and identification of the MN9 antigen as type 1 transmembrane protein with glycosylation pattern of an NO-sialoglycoprotein. Based on the mutation of equatorin’s predicted gameteinteraction domain, which is O-glycosylated, they predicted a model in which the N-terminus of equatorin faces the acrosomal lumen, while the C-terminus is localized in the cytoplasm. Equatorin localizes in the anterior acrosome and equatorial segment before AE and disperses from anterior acrosome to the center of the equatorial segment following AE. Subsequently, some equatorin remains localized on the inner acrosomal membrane after acrosomal exocytosis (Yoshida et al. 2009). It is not clear how equatorin participates in membrane adhesion and/or fusion. Sequence analysis of equatorin does not reveal any conserved domains. However, 100% identity is revealed with the previously sequenced acrosome formation associated factor (AFAF), expressed in mouse round spermatids, thought to be involved in acrosomal biogenesis during spermiogenesis (Li et al. 2006) and acrosomal exocytosis during fertilization (Hu et al. 2009). Once a KO is available, it will become clear whether equatorin is essential for spermiogenesis, ZP-induced AE and/or sperm-oolemma fusion. Similar to equatorin, the rat epididymal glycoprotein DE (37 kDa cysteine-rich secretory protein or CRISP1) undergoes redistribution in the equatorial segment during capacitation, suggesting the involvement of sperm-egg fusion (Rochwerger et al. 1992, Rochwerger and Cuasnicu 1992). However, male CRIPS1 mutants are fertile, with somewhat reduced rate of sperm-ZP penetration and sperm oolemmafusion (Da Ros et al. 2008). A related protein, CRISP2 also seems to reside on the sperm head plasma membrane and could have a compensatory effect in CRISP1 mutants. Along this line, antibodies against CRISP2 caused sperm accumulation in the periviteline space during mouse fertilization in vitro (Busso et al. 2007).
8.7 Conclusions Membrane fusion has a crucial role in at least two different steps of mammalian fertilization (Fig. 8.1). First, the acrosomal exocytosis involves the irreversible, terminal fusion of two homologous, but distinct sperm membranes, the plasma membrane and its underlying inner acrosomal membrane. This fusion event is restricted to the anterior, acrosomal compartment of the sperm head, but it does contribute to remodeling and exposure of the plasma membrane of the equatorial segment of the sperm head, which is therefore primed for adhesion to the oocyte
196
Y.-J. Yi et al.
Fig. 8.1 Membrane fusion events of mammalian fertilization. (1) Fertilizing spermatozoon binds through its plasma membrane, covering the acrosomal region of the sperm head, to the glycoprotein egg coat, zona pellucida (ZP). (2) Sperm-ZP binding induces acrosomal exocytosis; hybrid membrane vesicles form through fusion of plasma membrane with outer acrosomal membrane. These vesicles will form the acrosomal shroud. (3) Acrosomal exocytosis exposes enzymes in the acrosomal matrix, participating in sperm-ZP penetration. (4) Sperm plasma membrane, covering the sperm head equatorial segment, adheres to oolemma, the oocyte plasma membrane. (5) Sperm-oolemma fusion occurs, resulting in the intermingling of sperm and oocyte-derived plasma membrane components and diffusion of the sperm head skeleton, perinuclear theca components in the oocyte cytoplasm. This step of fertilization coincides with oocyte activation and is followed by sperm incorporation in the oocyte cytoplasm
plasma membrane, the oolemma. Acrosomal exocytosis is triggered by the binding of the sperm plasma membrane to a sperm receptor on the extracellular matrixmade egg coat, the zona pellucida, yet the oocyte does not contribute membranes to it. The mechanistic and molecular basis of membrane fusion during AE is poorly understood. Some existing evidence implicates the synaptic vesicle proteins of the SNARE hypothesis in this event. The second fusion event is heterologous in nature as it involves the fusion of the respective sperm and egg plasma membranes. On the oolemma side, the tetraspanin web with its namesake, transmembrane tetraspanins CD9 and CD81 is supported by integrins, and associated cytoskeletal and signaling proteins. Tetraspanins are the essential components of this system on the oocyte side. Tetraspanins of the oolemma interact on the sperm side with a superglobulin family protein IZUMO, which again is essential for fertility. It is not clear whether IZUMO-tetraspanin interaction mediates only the adhesion, or both adhesion and fusion of the respective gamete plasma membranes. An intriguing possibility emerged recently that tetraspanins from the oolema-derived membrane vesicles could be incorporated in the sperm plasma membrane during fertilization, possibly to mediate homophillic tetraspanin-tetraspanin adhesion. It is possible that the components of this system will yield targets for contraceptive development.
8
Gamete Binding and Fusion
197
References Almeida EA, Huovila AP, Sutherland AE et al (1995) Mouse egg integrin alpha 6 beta 1 functions as a sperm receptor. Cell 81:1095–1104 Andria ML, Barsh GS, Levy S (1992) Expression of TAPA-1 in preimplantation mouse embryos. Biochem Biophys Res Commun 186:1201–1206 Aricescu AR, Jones EY (2007) Immunoglobulin superfamily cell adhesion molecules: zippers and signals. Curr Opin Cell Biol 19:543–550 Baessler KA, Lee Y, Sampson NS (2009) Beta1 integrin is an adhesion protein for sperm binding to eggs. ACS Chem Biol 4:357–366 Barclay AN (2003) Membrane proteins with immunoglobulin-like domains–a master superfamily of interaction molecules. Semin Immunol 15:215–223 Barraud-Lange V, Naud-Barriant N, Bomsel M et al (2007) Transfer of oocyte membrane fragments to fertilizing spermatozoa. FASEB J 21:3446–3449 Bedford JM, Cooper GW (1978) Membrane fusion events in the fertilization of vertebrate eggs. In: Poste G, Nicolson GL (eds) Membrane fusion, vol 5. Elsevier/North-Holland Biomedical Press, Amsterdam Bennett MK, Calakos N, Scheller RH (1992) Syntaxin: a synaptic protein implicated in docking of synaptic vesicles at presynaptic active zones. Science 257:255–259 Bigler D, Takahashi Y, Chen MS et al (2000) Sequence-specific interaction between the disintegrin domain of mouse ADAM 2 (fertilin beta) and murine eggs. Role of the alpha(6) integrin subunit. J Biol Chem 275:11576–11584 Boucheix C, Rubinstein E (2001) Tetraspanins. Cell Mol Life Sci 58:1189–1205 Breitbart H (2002) Role and regulation of intracellular calcium in acrosomal exocytosis. J Reprod Immunol 53:151–159 Buffone MG, Foster JA, Gerton GL (2008) The role of the acrosomal matrix in fertilization. Int J Dev Biol 52:511–522 Busso D, Goldweic NM, Hayashi M et al (2007) Evidence for the involvement of testicular protein CRISP2 in mouse sperm-egg fusion. Biol Reprod 76:701–708 Cal S, Freije JM, Lopez JM et al (2000) ADAM 23/MDC3, a human disintegrin that promotes cell adhesion via interaction with the alphavbeta3 integrin through an RGD-independent mechanism. Mol Biol Cell 11:1457–1469 Campbell KD, Reed WA, White KL (2000) Ability of integrins to mediate fertilization, intracellular calcium release, and parthenogenetic development in bovine oocytes. Biol Reprod 62:1702–1709 Charrin S, Le Naour F, Labas V et al (2003) EWI-2 is a new component of the tetraspanin web in hepatocytes and lymphoid cells. Biochem J 373: 409–421 Charrin S, le Naour F, Silvie O et al (2009) Lateral organization of membrane proteins: tetraspanins spin their web. Biochem J 420:133–154 Chen H, Sampson NS (1999) Mediation of sperm-egg fusion: evidence that mouse egg alpha6beta1 integrin is the receptor for sperm fertilinbeta. Chem Biol 6:1–10 Chen MS, Almeida EA, Huovila AP et al (1999a) Evidence that distinct states of the integrin alpha6beta1 interact with laminin and an ADAM. J Cell Biol 144:549–561 Chen MS, Tung KS, Coonrod SA et al (1999b). Role of the integrin-associated protein CD9 in binding between sperm ADAM 2 and the egg integrin alpha6beta1: implications for murine fertilization. Proc Natl Acad Sci USA 96:11830–11835 Chen YA, Scheller RH (2001) SNARE-mediated membrane fusion. Nat Rev Mol Cell Biol 2:98–106 Cho C, Bunch DO, Faure JE et al (1998) Fertilization defects in sperm from mice lacking fertilin beta. Science 281:1857–1859 Da Ros VG, Maldera JA, Willis WD et al (2008) Impaired sperm fertilizing ability in mice lacking Cysteine-RIch Secretory Protein 1 (CRISP1). Dev Biol 320:12–18 De Blas GA, Roggero CM, Tomes CN et al (2005) Dynamics of SNARE assembly and disassembly during sperm acrosomal exocytosis. PLoS Biol 3:e323
198
Y.-J. Yi et al.
Delaguillaumie A, Lagaudriere-Gesbert C, Popoff MR et al (2002) Rho GTPases link cytoskeletal rearrangements and activation processes induced via the tetraspanin CD82 in T lymphocytes. J Cell Sci 115:433–443 Ellerman DA, Ha C, Primakoff P et al (2003) Direct binding of the ligand PSG17 to CD9 requires a CD9 site essential for sperm-egg fusion. Mol Biol Cell 14:5098–5103 Ellerman DA, Pei J, Gupta S et al (2009) Izumo is part of a multiprotein family whose members form large complexes on mammalian sperm. Mol Reprod Dev 76:1188–1199 Eto K, Puzon-McLaughlin W, Sheppard D et al (2000) RGD-independent binding of integrin alpha9beta1 to the ADAM-12 and -15 disintegrin domains mediates cell–cell interaction. J Biol Chem 275:34922–34930 Evans JP (2001) Fertilin beta and other ADAMs as integrin ligands: insights into cell adhesion and fertilization. Bioessays 23:628–639 Evans JP, Kopf GS, Schultz RM (1997) Characterization of the binding of recombinant mouse sperm fertilin beta subunit to mouse eggs: evidence for adhesive activity via an egg beta1 integrin-mediated interaction. Dev Biol 187:79–93 Evans JP, Schultz RM, Kopf GS (1995) Identification and localization of integrin subunits in oocytes and eggs of the mouse. Mol Reprod Dev 40:211–220 Evans JP, Schultz RM, Kopf GS (1998) Roles of the disintegrin domains of mouse fertilins alpha and beta in fertilization. Biol Reprod 59:145–152 Feigelson SW, Grabovsky V, Shamri R et al (2003) The CD81 tetraspanin facilitates instantaneous leukocyte VLA-4 adhesion strengthening to vascular cell adhesion molecule 1 (VCAM-1) under shear flow. J Biol Chem 278:51203–51212 Foster JA, Friday BB, Maulit MT (1997) AM67, a secretory component of the guinea pig sperm acrosomal matrix, is related to mouse sperm protein sp56 and the complement component 4-binding proteins. J Biol Chem 272:12714–12722 Gadella BM, Visconti PE (2006). Regulation of capacitation. In: De Jonge C, Barratt CL (eds) The sperm cell. Cambridge University Press, Cambridge, UK, pp 134–169 Gerst JE (1999) SNAREs and SNARE regulators in membrane fusion and exocytosis. Cell Mol Life Sci 55:707–734 Glazar AI, Evans JP (2009) Immunoglobulin superfamily member IgSF8 (EWI-2) and CD9 in fertilisation: evidence of distinct functions for CD9 and a CD9-associated protein in mammalian sperm-egg interaction. Reprod Fertil Dev 21:293–303 Gould RJ, Polokoff MA, Friedman PA et al (1990) Disintegrins: a family of integrin inhibitory proteins from viper venoms. Proc Soc Exp Biol Med 195:168–171 He ZY, Brakebusch C, Fassler R et al (2003) None of the integrins known to be present on the mouse egg or to be ADAM receptors are essential for sperm-egg binding and fusion. Dev Biol 254:226–237 He ZY, Gupta S, Myles D et al (2009) Loss of surface EWI-2 on CD9 null oocytes. Mol Reprod Dev 76:629–636 Hemler ME (2003) Tetraspanin proteins mediate cellular penetration, invasion, and fusion events and define a novel type of membrane microdomain. Annu Rev Cell Dev Biol 19:397–422 Higginbottom A, Takahashi Y, Bolling L et al (2003) Structural requirements for the inhibitory action of the CD9 large extracellular domain in sperm/oocyte binding and fusion. Biochem Biophys Res Commun 311:208–214 Hohne-Zell B, Gratzl M (1996) Adrenal chromaffin cells contain functionally different SNAP-25 monomers and SNAP-25/syntaxin heterodimers. FEBS Lett 394:109–116 Hu XQ, Ji SY, Li YC et al (2009) Acrosome formation-associated factor is involved in fertilization. Fertil Steril 93:1482–1492 Inoue N, Ikawa M, Isotani A et al (2005) The immunoglobulin superfamily protein Izumo is required for sperm to fuse with eggs. Nature 434:234–238 Iwao Y, Fujimura T (1996) Activation of Xenopus eggs by RGD-containing peptides accompanied by intracellular Ca2+ release. Dev Biol 177:558–567
8
Gamete Binding and Fusion
199
Kaji K, Oda S, Miyazaki S et al (2002) Infertility of CD9-deficient mouse eggs is reversed by mouse CD9, human CD9, or mouse CD81; polyadenylated mRNA injection developed for molecular analysis of sperm-egg fusion. Dev Biol 247:327–334 Kaji K, Oda S, Shikano T et al (2000) The gamete fusion process is defective in eggs of Cd9deficient mice. Nat Genet 24:279–282 Kim KS, Cha MC, Gerton GL (2001a) Mouse sperm protein sp56 is a component of the acrosomal matrix. Biol Reprod 64:36–43 Kim KS, Foster JA, Gerton GL (2001b) Differential release of guinea pig sperm acrosomal components during exocytosis. Biol Reprod 64:148–156 Kim KS, Gerton GL (2003) Differential release of soluble and matrix components: evidence for intermediate states of secretion during spontaneous acrosomal exocytosis in mouse sperm. Dev Biol 264:141–152 Kolesnikova TV, Stipp CS, Rao RM et al (2004) EWI-2 modulates lymphocyte integrin alpha4beta1 functions. Blood 103:3013–3019 Lammerding J, Kazarov AR, Huang H et al (2003) Tetraspanin CD151 regulates alpha6beta1 integrin adhesion strengthening. Proc Natl Acad Sci USA 100:7616–7621 Le Naour F, Rubinstein E, Jasmin C et al (2000) Severely reduced female fertility in CD9-deficient mice. Science 287:319–321 Lefevre B, Wolf JP, Ziyyat A (2010) Sperm-egg interaction: is there a link between tetraspanin(s) and GPI-anchored protein(s)? Bioessays 32:143–152 Levy S, Shoham T (2005) The tetraspanin web modulates immune-signalling complexes. Nat Rev Immunol 5:136–148 Li YC, Hu XQ, Zhang KY et al (2006) Afaf, a novel vesicle membrane protein, is related to acrosome formation in murine testis. FEBS Lett 580:4266–4273 Linder B, Heinlein UA (1997) Decreased in vitro fertilization efficiencies in the presence of specific cyritestin peptides. Dev Growth Differ 39:243–247 Linfor J, Berger T (2000) Potential role of alphav and beta1 integrins as oocyte adhesion molecules during fertilization in pigs. J Reprod Fertil 120:65–72 Maleszewski M, Kimura Y, Yanagimachi R (1996) Sperm membrane incorporation into oolemma contributes to the oolemma block to sperm penetration: evidence based on intracytoplasmic sperm injection experiments in the mouse. Mol Reprod Dev 44:256–259 Manandhar G, Toshimori K (2001) Exposure of sperm head equatorin after acrosome reaction and its fate after fertilization in mice. Biol Reprod 65:1425–1436 Miller BJ, Georges-Labouesse E, Primakoff P et al (2000) Normal fertilization occurs with eggs lacking the integrin alpha6beta1 and is CD9-dependent. J Cell Biol 149:1289–1296 Miyado K, Yamada G, Yamada S et al (2000) Requirement of CD9 on the egg plasma membrane for fertilization. Science 287:321–324 Miyado K, Yoshida K, Yamagata K et al (2008) The fusing ability of sperm is bestowed by CD9containing vesicles released from eggs in mice. Proc Natl Acad Sci USA 105:12921–12926 Nath D, Slocombe PM, Stephens PE et al (1999) Interaction of metargidin (ADAM-15) with alphavbeta3 and alpha5beta1 integrins on different haemopoietic cells. J Cell Sci 112:579–587 Nath D, Slocombe PM, Webster A et al (2000) Meltrin gamma(ADAM-9) mediates cellular adhesion through alpha(6)beta(1 )integrin, leading to a marked induction of fibroblast cell motility. J Cell Sci 113:2319–2328 Neilson L, Andalibi A, Kang D et al (2000) Molecular phenotype of the human oocyte by PCRSAGE. Genomics 63:13–24 Nishiuchi R, Sanzen N, Nada S et al (2005) Potentiation of the ligand-binding activity of integrin alpha3beta1 via association with tetraspanin CD151. Proc Natl Acad Sci USA 102:1939–1944 Novick P, Zerial M (1997) The diversity of Rab proteins in vesicle transport. Curr Opin Cell Biol 9:496–504 Okabe M, Adachi T, Takada K et al (1987) Capacitation-related changes in antigen distribution on mouse sperm heads and its relation to fertilization rate in vitro. J Reprod Immunol 11:91–100
200
Y.-J. Yi et al.
Olson GE, Winfrey VP (1994) Structure of acrosomal matrix domains of rabbit sperm. J Struct Biol 112:41–48 Oura C, Toshimori K (1990) Ultrastructural studies on the fertilization of mammalian gametes. Int Rev Cytol 122:105–151 Oyler GA, Higgins GA, Hart RA et al (1989) The identification of a novel synaptosomalassociated protein, SNAP-25, differentially expressed by neuronal subpopulations. J Cell Biol 109:3039–3052 Ramalho-Santos J, Moreno RD, Sutovsky P et al (2000) SNAREs in mammalian sperm: possible implications for fertilization. Dev Biol 223:54–69 Rochwerger L, Cohen DJ, Cuasnicu PS (1992) Mammalian sperm-egg fusion: the rat egg has complementary sites for a sperm protein that mediates gamete fusion. Dev Biol 153:83–90 Rochwerger L, Cuasnicu PS (1992) Redistribution of a rat sperm epididymal glycoprotein after in vitro and in vivo capacitation. Mol Reprod Dev 31:34–41 Rothman JE (1994) Mechanisms of intracellular protein transport. Nature 372:55–63 Rubinstein E, Le Naour., Lagaudriere-Gesbert C et al (1996) CD9, CD63, CD81, and CD82 are components of a surface tetraspan network connected to HLA-DR and VLA integrins. Eur J Immunol 26:2657–2665 Rubinstein E, Ziyyat A, Prenant M et al (2006) Reduced fertility of female mice lacking CD81. Dev Biol 290:351–358 Runge KE, Evans JE, He ZY et al (2007) Oocyte CD9 is enriched on the microvillar membrane and required for normal microvillar shape and distribution. Dev Biol 304:317–325 Sala-Valdes M, Ursa A, Charrin S et al (2006) EWI-2 and EWI-F link the tetraspanin web to the actin cytoskeleton through their direct association with ezrin-radixin-moesin proteins. J Biol Chem 281:19665–19675 Schulz JR, Wessel GM, Vacquier VD (1997) The exocytosis regulatory proteins syntaxin and VAMP are shed from sea urchin sperm during the acrosome reaction. Dev Biol 191:80–87 Shamsadin R, Adham IM, Nayernia K, Heinlein UA, Oberwinkler H, Engel W (1999) Male mice deficient for germ-cell cyritestin are infertile. Biol Reprod 61:1445–1451. Shilling FM, Magie CR, Nuccitelli R (1998) Voltage-dependent activation of frog eggs by a sperm surface disintegrin peptide. Dev Biol 202:113–124 Stein KK, Primakoff P, Myles D (2004) Sperm-egg fusion: events at the plasma membrane. J Cell Sci 117:6269–6274 Stipp CS, Kolesnikova TV, Hemler ME (2001) EWI-2 is a major CD9 and CD81 partner and member of a novel Ig protein subfamily. J Biol Chem 276:40545–40554 Suarez S (2002) Gamete transport. In: Hardy D (ed) Fertilization. Academic Press, San Diego, pp 3–28 Sutton-Smith M, Wong NK, Khoo KH et al (2007) Analysis of protein-linked glycosylation in a sperm-somatic cell adhesion system. Glycobiology 17:553–567 Takahashi Y, Bigler D, Ito Y et al (2001) Sequence-specific interaction between the disintegrin domain of mouse ADAM 3 and murine eggs: role of beta1 integrin-associated proteins CD9, CD81, and CD98. Mol Biol Cell 12:809–820 Takahashi Y, Yamakawa N, Matsumoto K et al (2000) Analysis of the role of egg integrins in sperm-egg binding and fusion. Mol Reprod Dev 56:412–423 Tarone G, Russo MA, Hirsch E et al (1993) Expression of beta 1 integrin complexes on the surface of unfertilized mouse oocyte. Development 117:1369–1375 Terrian DM, White MK (1997) Phylogenetic analysis of membrane trafficking proteins: a family reunion and secondary structure predictions. Eur J Cell Biol 73:198–204 Tomes CN, Michaut M, De Blas G et al (2002) SNARE complex assembly is required for human sperm acrosome reaction. Dev Biol 243:326–338 Toshimori K, Saxena DK, Tanii I et al (1998) An MN9 antigenic molecule, equatorin, is required for successful sperm-oocyte fusion in mice. Biol Reprod 59:22–29 Toshimori K, Tanii I, Araki S et al (1992) Characterization of the antigen recognized by a monoclonal antibody MN9: unique transport pathway to the equatorial segment of sperm head during spermiogenesis. Cell Tissue Res 270:459–468
8
Gamete Binding and Fusion
201
van Goor H, Melenhorst WB, Turner AJ et al (2009) Adamalysins in biology and disease. J Pathol 219:277–286 Wang M, Lv Z, Shi J et al (2009) Immunocontraceptive potential of the Ig-like domain of Izumo. Mol Reprod Dev 76:794–801 Waterhouse R, Ha C, Dveksler GS (2002) Murine CD9 is the receptor for pregnancy-specific glycoprotein 17. J Exp Med 195:277–282 Weimbs T, Low SH, Chapin SJ et al (1997) A conserved domain is present in different families of vesicular fusion proteins: a new superfamily. Proc Natl Acad Sci USA 94:3046–3051 Weimbs T, Mostov K, Low SH et al (1998) A model for structural similarity between different SNARE complexes based on sequence relationships. Trends Cell Biol 8:260–262 Westbrook-Case VA, Winfrey VP, Olson GE (1995) Sorting of the domain-specific acrosomal matrix protein AM50 during spermiogenesis in the guinea pig. Dev Biol 167:338–349 Yamatoya K, Yoshida K, Ito C et al (2009) Equatorin: identification and characterization of the epitope of the MN9 antibody in the mouse. Biol Reprod 81:889–897 Yanagimachi R (1994) Mammalian fertilization. Raven Press, New York, NY Yoshida K, Ito C, Yamatoya K et al (2009) A model of the acrosome reaction progression via the acrosomal membrane-anchored protein equatorin. Reproduction 139:533–544 Yuan R, Primakoff P, Myles DG (1997) A role for the disintegrin domain of cyritestin, a sperm surface protein belonging to the ADAM family, in mouse sperm-egg plasma membrane adhesion and fusion. J Cell Biol 137:105–112 Yunes R, Michaut M, Tomes C et al (2000) Rab3A triggers the acrosome reaction in permeabilized human spermatozoa. Biol Reprod 62:1084–1089 Zarelli VE, Ruete MC, Roggero CM et al (2009) PTP1B dephosphorylates N-ethylmaleimidesensitive factor and elicits SNARE complex disassembly during human sperm exocytosis. J Biol Chem 284:10491–10503 Zerial M, McBride H (2001) Rab proteins as membrane organizers. Nat Rev Mol Cell Biol 2:107–117 Zhang XP, Kamata T, Yokoyama K et al (1998) Specific interaction of the recombinant disintegrinlike domain of MDC-15 (metargidin, ADAM-15) with integrin alphavbeta3. J Biol Chem 273:7345–7350 Zhou M, Graham R, Russell G et al (2001) MDC-9 (ADAM-9/Meltrin gamma) functions as an adhesion molecule by binding the alpha(v)beta(5) integrin. Biochem Biophys Res Commun 280:574–580 Ziyyat A, Rubinstein E, Monier-Gavelle F et al (2006) CD9 controls the formation of clusters that contain tetraspanins and the integrin alpha 6 beta 1, which are involved in human and mouse gamete fusion. J Cell Sci 119:416–424
Chapter 9
Mechanisms Regulating Human Trophoblast Fusion Berthold Huppertz and Martin Gauster
Abstract In the human placenta, the syncytiotrophoblast is part of the villous trophoblast, which is the epithelial cover of the placental villi floating in maternal blood. The villous trophoblast is composed of two layers, the syncytiotrophoblast in direct contact to maternal blood and the underlying layer of mononucleated cytotrophoblasts. Throughout pregnancy there is continuous fusion of cytotrophoblast with the syncytiotrophoblast to maintain this highly differentiated layer until delivery. This way the syncytiotrophoblast is continuously supplied with cytoplasmic compounds derived from the fusing cytotrophoblasts. The continuous acquisition of fresh cellular components needs to be balanced by a simultaneous release of apoptotic material from the syncytiotrophoblast into the maternal circulation. In the maintenance of the syncytiotrophoblast, fusion is an essential step and was shown to be regulated by multiple factors, such as cytokines, hormones, protein kinases, transcription factors, proteases and membrane proteins. Here we focus on factors that may play roles in the trophoblast fusion process or in the preparation of the cytotrophoblasts to fuse. We will also speculate on pitfalls when studying trophoblast fusion in vitro. Keywords Placenta · trophoblast · caspase 8 · fusion Abbreviations 1T 2T 3T ADAM ASCT BW
First trimester trophoblasts Second trimester trophoblasts Third trimester trophoblasts A disintegrin and metalloprotease Alanine, serine and cysteine selective transporters BeWo cells
B. Huppertz (B) Cell Biology, Institute of Cell Biology, Histology and Embryology, Center for Molecular Medicine, Medical University of Graz, 8010 Graz, Austria e-mail:
[email protected]
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_9, C Springer Science+Business Media B.V. 2011
203
204
CSF CT EGF Env ERK ERV GCM GM-CSF hCG HERV HLA-G hPL LIF MAPK Mash-2 MIC-1 pc PKA PP13 PS ST TGF TNF VE VEGF
B. Huppertz and M. Gauster
Colony stimulating factor Cytotrophoblast Epidermal growth factor Envelope Extracellular signal-regulated kinase Endogenous retrovirus Glial cell missing Granulocyte-macrophage colony-stimulating factor Human chorionic gonadotropin Human endogenous retrovirus Human leukocyte antigen G Human placental lactogen Leukemia-inhibitory factor Mitogen-activated protein kinase Mammalian achaete/scute homolog 2 Macrophage inhibitory cytokine 1 Post conception Protein kinase A Placental protein 13 Phosphatidylserine Syncytiotrophoblast Transforming growth factor Tumor necrosis factor Villous explants Vascular endothelial growth factor
Contents 9.1 Introduction . . . . . . . . . . . . . . . . . . . . . 9.2 Human Placenta and Villous Trophoblast . . . . . . . . 9.3 Regulators of Trophoblast Fusion . . . . . . . . . . . 9.3.1 Cytokines, Growth Factors and Trophoblast Fusion 9.3.2 Protein Kinases, Transcription Factors and Trophoblast Fusion . . . . . . . . . . . . . 9.3.3 The Phosphatidylserine Flip and Trophoblast Fusion 9.3.4 Caspase 8 Activity and Trophoblast Fusion . . . . 9.3.5 Fusogenic Proteins and Trophoblast Fusion . . . . 9.4 Pitfalls in Dealing with Trophoblast Fusion In Vitro . . . 9.4.1 Phenotype of Isolated Primary Trophoblasts . . . 9.4.2 The Use of β-hCG to Determine the Extent of Trophoblast Fusion . . . . . . . . . . . . . 9.5 Conclusions . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
205 205 206 206
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
206 208 209 211 212 212
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
213 214 214
9
Mechanisms Regulating Human Trophoblast Fusion
205
9.1 Introduction Syncytial fusion of cells leads to the generation of multinucleated structures such as myoblast-derived skeletal muscle fibers (Chen and Olson 2004). By definition such a multinucleated structure is no longer termed a cell but is referred to as a syncytium. In order to syncytialize, the plasma membranes of two neighboring cells need to closely interact with each other and finally dissolve to open the way for cytoplasmic contents and organelles, allowing the exchange of contents between formerly membrane-encircled compartments. Fusion events are mostly found within a cell during membrane traffic and release of vesicles. Fusion events between two cells in the human can be found during fertilization (fusion of egg and sperm) as well as in a few other cell types (Chen and Olson 2005, Potgens et al. 2002). Membranes do not fuse easily to maintain the individuality of the intracellular compartments and of the cell itself. Hence, the fusion process requires preparation of the cell to fuse and the availability of specific fusogenic proteins.
9.2 Human Placenta and Villous Trophoblast During the very early development of a human individual, fusion processes are fundamental and mandatory for the beginning and maintenance of pregnancy. The first cell line to develop during embryogenesis is the trophoblast, which becomes the major constituent of the human placenta. During the development from the morula to the blastocyst stage the trophoblast lineage differentiates from the embryoblast. It develops the outer cover of the blastocyst around the early embryo (Benirschke et al. 2006). Implantation of the human blastocyst is followed by various stages of placental development. Between d15 and d21 post conception (pc) the early trophoblast divides into two separate differentiation pathways, to result in extravillous (invasive pathway) or villous (syncytial pathway) trophoblasts. The extravillous trophoblast develops into invasive cells that migrate towards and invade into the maternal uterine decidua basalis reaching the inner third of the myometrium. Appropriate invasion of extravillous trophoblasts is essential for fetal development since these cells transform maternal spiral arteries to the needs of the fetus and attach the placenta to the uterus. The villous trophoblast consists of two layers. Villous cytotrophoblasts remain on the basement membrane of the chorionic villi and serve as a pool of progenitor cells. These cells are covered by the syncytiotrophoblast, a multinucleated layer that lacks lateral plasma membranes. The highly differentiated syncytiotrophoblast has lost its generative capacity and does no longer show proliferative capacities. As most highly differentiated cells, it only shows moderate rates of RNA synthesis (Richart 1961, Garcia-Lloret et al. 1994, Morrish et al. 1987). Thus, there is the need for a continuous fusion of cytotrophoblasts with the syncytiotrophoblast throughout pregnancy. With syncytial fusion incorporation of cytotrophoblast-derived nuclei
206
B. Huppertz and M. Gauster
and other organelles as well as proteins and RNA into the syncytiotrophoblast is guaranteed. The multinucleated syncytiotrophoblast forms the outer surface of all chorionic villi within a placenta and comes into direct contact with maternal blood (Benirschke et al. 2006, Midgley et al. 1963, Panigel 1993). This single layer extends over the surface of all villous trees of a placenta (Benirschke et al. 2006) and is responsible for functions such as transport of oxygen, nutrients and waste products, hormone production and immune tolerance (Benirschke et al. 2006). In recent years it has become clear that coordination and appropriate control of trophoblast fusion are crucial to preserve a healthy pregnancy. Dysregulation and alterations of trophoblast fusion may be directly involved in pathological conditions such as preeclampsia (Huppertz et al. 2002, Huppertz and Kingdom 2004).
9.3 Regulators of Trophoblast Fusion 9.3.1 Cytokines, Growth Factors and Trophoblast Fusion Trophoblast fusion is influenced by cytokines and growth factors derived from the maternal as well as the fetal environment (Table 9.1). Epidermal growth factor (EGF) was described to induce syncytialization of isolated primary cytotrophoblasts leading to increased secretion of human chorionic gonadotropin (hCG) and human placental lactogen (hPL) in vitro (Morrish et al. 1987). Similar findings have been described for colony stimulating factor (CSF), granulocyte-macrophage colonystimulating factor (GM-CSF) (Garcia-Lloret et al. 1994) as well as leukemiainhibitory factor (LIF) and transforming growth factor (TGF)-α (Yang et al. 2003). Interestingly, the syncytiotrophoblast derived hCG can act in an autocrine loop to increase syncytium formation (Yang et al. 2003, Shi et al. 1993). Treatment of isolated primary first trimester cytotrophoblasts with vascular endothelial growth factor (VEGF) in vitro increased number and size of syncytia compared to untreated controls (Crocker et al. 2001). Macrophage inhibitory cytokine 1 (MIC-1), a member of the TGF-β superfamily, was suggested to be involved in trophoblast differentiation and fusion since transfection of antisense MIC-1 into term cytotrophoblasts led to inhibition of syncytialization (Li et al. 2005). So far, only a few factors have been described to reduce formation of trophoblast syncytia. Tumor necrosis factor (TNF)-α repressed hCG secretion and impaired syncytium formation of primary term trophoblasts in vitro (Leisser et al. 2006). Also transforming growth factor (TGF)-β has been shown to inhibit formation of syncytia and secretion of hCG and hPL (Morrish et al. 1991).
9.3.2 Protein Kinases, Transcription Factors and Trophoblast Fusion Environmentally derived factors may bind to respective receptors on trophoblasts, starting downstream pathways to initiate a complex cascade of cell differentiation.
9
Mechanisms Regulating Human Trophoblast Fusion
207
Table 9.1 Proteins promoting (+) or hindering (–) syncytial fusion of trophoblasts and trophoblast-derived cells Growth factors, hormones and cytokines CSF EGF GM-CSF hCG LIF MIC-1 TGF-α TGF-β TNF-α VEGF
Colony stimulating factor Epidermal growth factor Granulocyte macrophage colony stimulating factor Human chorionic gonadotropin Leukemia inhibitory factor Macrophage inhibitory cytokine 1 Transforming growth factor alpha Transforming growth factor beta Tumor necrosis factor alpha Vascular endothelial growth factor
Growth factor Growth factor Growth factor
+ + +
3T 3T 3T
[1] [2] [1]
Peptide hormone Cytokine Cytokine Growth factor Growth factor Cytokine Growth factor
+ + + + – – +
3T 3T 3T 3T 3T 3T 1T, 3T
[3, 4] [3] [5] [3] [6] [7] [8]
Mitogen activated protein kinases (MAPK) 1 and 2 Transcription factor Transcription factor
+
3T
[9]
+ –
BW 2T
[10] [11]
+
3T
[9]
+
BW
[12]
Amino acid transporter Amino acid transporter Amino acid transporter Gap junction protein Lectin Endogenous retroviral envelope proteins
? + + + + +
BW BW 3T BW 3T, BW
[13, 14] [13, 14] [15, 16] [17] [18] [19, 20]
A disintegrin and metalloprotease 12 Initiator caspase, protease involved in apoptosis & fusion Initiator caspase, protease involved in apoptosis Caspase, protease involved in keratinocyte differentiation
? +
VE
[21] [22, 23]
Protein kinases and transcription factors ERK1/2
Extracellular signal-regulated kinases 1 and 2
GCM1 Mash-2 p38
Glial cell missing homolog 1 Mammalian achaete/scute homolog 2 p38 isoforms of MAPK
PKA
Protein kinase A
Mitogen activated protein kinases (MAPK) 11–14 Protein kinase
Membrane proteins ASCT1 ASCT2 CD98 Connexin 43 Galectin 3 Syncytin 1, 2 Proteases ADAM12 Caspase 8 Caspase 10 Caspase 14
? ?
[23] [24]
Data are derived from in vitro experiments using first trimester trophoblasts (1T), second trimester trophoblasts (2T), third trimester trophoblasts (3T), BeWo cells (BW) or villous explants (VE). Some factors were suggested to play a role in trophoblast fusion without any experimental evidence so far (indicated by a “?”). References: 1. Garcia-Lloret et al. (1994); 2. Morrish et al. (1987); 3. Yang et al. (2003); 4. Shi et al. (1993); 5. Li et al. (2005); 6. Morrish et al. (1991); 7. Leisser et al. (2006); 8. Crocker et al. (2001); 9. Daoud et al. (2005); 10. Yu et al. (2002); 11. Jiang et al. (2000); 12. Knerr et al. (2005); 13. Jansson (2001); 14. Kudo and Boyd (2002); 15. Kudo and Boyd (2004); 16. Kudo et al. (2003); 17. Frendo et al. (2003); 18. Dalton et al. (2007); 19. Mi et al. (2000); 20. Frendo et al. (2003); 21. Huppertz et al. (2006); 22. Black et al. (2004); 23. Huppertz et al. (1999); 24. White et al. (2007)
208
B. Huppertz and M. Gauster
Two classical mitogen-activated protein kinases (MAPKs), p38 and the extracellular signal-regulated kinase1/2 (ERK1/2), were suggested to play important roles in regulating trophoblast differentiation and fusion. Inhibitors blocking ERK1/2 and/or p38 activities diminished differentiation and syncytium formation in cultures of primary trophoblasts (Daoud et al. 2005). Protein kinase A (PKA) also seems to regulate downstream pathways of trophoblast differentiation, since transient overexpression of the PKA catalytic subunit led to increased fusion of BeWo cells, a trophoblast-derived choriocarcinoma cell line (Knerr et al. 2005). Administration of forskolin, a reagent known to induce fusion in BeWo cells, raises intracellular cAMP concentrations and subsequently leads to upregulation of the transcription factor glial cell missing 1 (GCM1) in BeWo cells (Knerr et al. 2005). GCM1 belongs to the GCM family, a family of zinc-containing transcription factors (Lin et al. 2005) and was identified as the first transcription factor involved in trophoblast syncytialization. Expression of GCM1 in the human placenta is restricted to a subset of highly differentiated villous cytotrophoblasts (Baczyk et al. 2004), destined to fuse with the syncytiotrophoblast. One target gene for GCM1 in the human placenta is syncytin 1 (Yu et al. 2002), which is hypothesized to trigger fusion in the villous trophoblast (Mi et al. 2000).
9.3.3 The Phosphatidylserine Flip and Trophoblast Fusion Mammalian cells maintain an asymmetrical distribution of phospholipids in the two leaflets of their plasma membranes (Bevers et al. 1996). The negatively charged phospholipids such as phosphatidylserine (PS) are mostly restricted to the inner leaflet. Under specific conditions the cells redistribute PS from the inner to the outer leaflet of the plasma membrane. This “PS-flip” has been shown to be required for syncytial fusion (Lyden et al. 1993) as well as an “eat-me” signal to eliminate apoptotic cells (Savill 1998). Maintenance of the phospholipid asymmetry is regulated by three different lipid transporters: an ATP-independent bidirectional transporter (scramblase), an ATP-dependent inward flipping aminophospholipid translocase (flippase, inward translocation) and an ATP-dependent outward flipping transporter (floppase, outward translocation). The activity of a flippase transfers lipids from the outer to the inner leaflet of the bilayer, the activity of a floppase transports lipids from the inside to the outside leaflet, while a scramblase acts in both directions. In the two main multinucleated fusion systems in the human, skeletal muscle and placental trophoblast, redistribution of PS is a prerequisite for fusion (Huppertz et al. 1998, van den Eijnde et al. 2001). A transient exposition of PS has been detected in myoblasts at specific cell–cell contact sites during fusion to generate myotubes van den Eijnde et al. 2001. In the human placenta a subset of highly differentiated cytotrophoblasts displays the PS-flip without any signs of apoptosis (Huppertz et al. 1998). In the trophoblast-derived BeWo choriocarcinoma cell line the PS-flip has been shown to be essential for syncytial fusion (Lyden et al. 1993, Adler et al. 1995).
9
Mechanisms Regulating Human Trophoblast Fusion
209
The application of an antibody directed against PS inhibited syncytial fusion of BeWo cells (Lyden et al. 1993, Adler et al. 1995). Fusogenic proteins are known players in the formation of the fusion pore. Such fusion proteins have the structure of an alpha-helix and are oriented almost parallel to the lipid acyl chains of the plasma membrane. In order to enable reorientation of the fusion protein and its insertion into the lipid bilayer, the redistribution of PS is mandatory (Martin et al. 1999). The highly specific orientation of the fusion proteins is crucial for the mechanism of protein-induced membrane fusion (Decout et al. 1998).
9.3.4 Caspase 8 Activity and Trophoblast Fusion Intracellular proteases seem to play crucial roles in priming mononucleated cells for fusion in skeletal muscle as well as placental trophoblast. In skeletal muscle the activity of members of the calpain family are upregulated prior to fusion (Barnoy et al. 1998). These Ca2+ -regulated cysteine proteases play essential roles in a variety of processes including syncytial fusion (Barnoy et al. 1998, Bartoli and Richard 2005). The direct endogenous inhibitor of calpains, calpastatin, inhibits fusion of myoblasts (Barnoy et al. 2005), which makes it tempting to speculate about a role of calpains in preparing myoblasts for fusion. In the placental villous trophoblast calpains do not seem to play a role in syncytial fusion (Gauster et al. 2009b), rather this tissue makes use of a similar but different system of proteases. In the trophoblast initiator caspase 8 has been shown to be crucial for the fusion process (Black et al. 2004) (Fig. 9.1). Immunohistochemistry for
Fig. 9.1 Putative actions of active caspase 8 prior, during and after villous trophoblast fusion. (a) Prior to fusion pro-caspase 8 is conversed into its active form (active caspase 8) and cleavage of the sub-membranous cytoskeletal protein alpha-fodrin takes place in the cytotrophoblast (CT). There may be a direct action of caspase 8 on the PS-flip or it may be triggered by alpha-fodrin cleavage. To better visualize phosphatidylserine externalization (PS flip) from the inner to the outer leaflet of the plasma membrane the two leaflets are coloured green and red. (b) During fusion fragmentation and remodelling of the fodrin cytoskeleton and the PS-flip in combination with fusogenic proteins (not shown) may result in the formation of a fusion pore between the cytotrophoblast and the overlying syncytiotrophoblast (ST). (c) After fusion cellular components of the cytotrophoblast, including the nucleus as well as organelles, cytoplasmic proteins and the fragments of alpha-fodrin, are transferred into the syncytiotrophoblast. Caspase 8 is still active at the of transfer time and escorts the nucleus into the syncytium. This activity of incorporated caspase 8 needs to be inhibited, maybe by the action of anti-apoptotic proteins such as Bcl-2 and c-Flip
210
B. Huppertz and M. Gauster
the active and pro-forms of caspase 8 revealed staining in some cytotrophoblasts as well as diffuse staining in a few sites of the syncytiotrophoblast (Huppertz et al. 1999), while pro-caspase 8 is predominantly localized in the cytotrophoblast (Fong et al. 2006, De Falco et al. 2004). Functional studies in a villous explant model suggested a crucial role for caspase 8 in trophoblast fusion. Treatment of villous explants with antisense oligonucleotides or peptide inhibitors against caspase 8 resulted in multiple cytotrophoblast layers and a degenerating syncytiotrophoblast due to reduced fusion with cytotrophoblasts (Black et al. 2004, Miehe et al. 2006). Recently we have shown that active caspase 8 is present in some villous cytotrophoblasts, in rare cases of trophoblasts, that were located between the two villous trophoblast layers and in few sites of the overlying syncytiotrophoblast (Gauster et al. 2009a). In line with the general model that only highly differentiated and postmitotic cells undergo fusion, we never found active caspase 8 co-localized with the proliferation marker anti-Ki67 (Gauster et al. 2009a). Based on our and other studies, we suggest that caspase 8 is activated in highly differentiated cytotrophoblasts prior to fusion and escorts the nucleus and the other content of the fusing cell into the syncytiotrophoblast. So far, the mechanistic role of caspase 8 in trophoblast differentiation and fusion has not been adequately defined. Caspase 8 is not a classical fusogenic protein and thus its importance may be found in acting upstream, thereby preparing the cell for the upcoming fusion event. Preparing the cell for fusion could take place on several fronts such as remodelling the membrane architecture (initiating the PSflip), remodelling of the sub-membranous cytoskeleton (e.g. cleavage of α-fodrin), or by a variety of complex cell signalling processes. As described above, alterations of the aminophospholipid asymmetry and related PS externalization have been associated with caspase 8 activity in erythrocytes (Mandal et al. 2005) and a squamous cell carcinoma cell line (Ohtani et al. 2000). So far there is no proof for a direct action of caspase 8 on PS externalization during trophoblast fusion. A direct effect of caspases as direct mediators of PS externalization during trophoblast fusion was excluded using the BeWo cell system (Mombers et al. 1979). However, since the PS-flip during fusion may occur in very restricted membrane compartments (van den Eijnde et al. 2001), it may escape visualization by routine microscopy. Sub-membranous cytoskeletal proteins of the spectrin protein family (Mombers et al. 1979, Sato and Ohnishi 1983) own PS binding sites and may play a role in sequestering PS in the inner leaflet of the plasma membrane. Target proteins of caspase 8 include α-fodrin (non-erythroid spectrin) which belongs to the submembranous cytoskeleton. We have shown that α-fodrin is cleaved by caspase 8 and that expression of α-fodrin is reduced in highly differentiated cytotrophoblasts, and entirely missing in the syncytiotrophoblast (Huppertz et al. 1999). Furthermore, we could co-localize active caspase 8 with α-fodrin vesicles in newly formed syncytium after fusion, suggesting caspase 8 cleavage of α-fodrin during cellular transition from the cytotrophoblast into the syncytiotrophoblast layer (Gauster et al. 2009b).
9
Mechanisms Regulating Human Trophoblast Fusion
211
Cleavage of sub-membranous cytoskeletal proteins such as α-fodrin is known to be an important step towards syncytial fusion. The degradation of the network of such proteins alters the curvature of the plasma membrane and thus may help to facilitate fusion, as a more curved membrane is more fusogenic (Martens and McMahon 2008). Interestingly, bovine kidney epithelial cells that were microinjected with an anti-fodrin antibody, underwent cell–cell fusion (Eskelinen and Lehto 1994). This indicated a fundamental role of sub-membranous cytoskeletal proteins in preventing syncytial fusion, and cleavage of such proteins may be needed to facilitate fusion.
9.3.5 Fusogenic Proteins and Trophoblast Fusion Five criteria have been set up to define whether a molecule is a genuine fusogen (Oren-Suissa and Podbilewicz 2007): The molecule of interest (1) (2) (3) (4) (5)
must be essential for membrane fusion, should be expressed at the right place to the right time, should be able to fuse originally non fusogenic cells, should be able to fuse heterologous cells, and should be able to fuse liposomes as well.
Based on these criteria, the number of already known candidate fusogens involved in trophoblast fusion is very limited. The only fusogens described in the human trophoblasts are the syncytins, which at least meet some of the criteria described above. DNA regions of retroviral origin that were incorporated during evolution make up about 8% of the human genome (de Parseval et al. 2003). Most of those regions are not translated into mRNA or even proteins, while some regions are translated into proteins. In the human placenta retroviral elements related to trophoblast fusion are expressed and comprise the envelope genes (env regions) of ERV-3, HERV-W, and HERV-FRD (Rote et al. 2004). Syncytin-1 and syncytin-2, members of the syncytin protein family, have been proposed to be main players in fusion of the villous trophoblast, though opinions differ regarding their importance (for a more detailed analysis see Huppertz et al. 2006). Syncytin-1 is an envelope protein of the HERV-W gene (Mi et al. 2000), while syncytin-2 is an envelope protein of the HERV-FRD gene (Lee et al. 2001). A fusogenic role of syncytin-1 was shown by impaired intercellular fusion of BeWo cells and primary cytotrophoblasts after application of anti-syncytin antiserum (Mi et al. 2000) and antisense oligonucleotides against syncytin (Frendo et al. 2003). According to the criteria listed above, syncytin-1 also mediates syncytial fusion in non-fusogenic COS cells, and between non-homologous BeWo and COS cells (Mi et al. 2000). More recently, also syncytin-2 was hypothesized to be involved in fusion of trophoblast cells (Chen et al. 2008, Vargas et al. 2009).
212
B. Huppertz and M. Gauster
The specific role of the two syncytins in trophoblast fusion is still unclear, although both proteins may be candidate fusogens since they fulfil at least some of the criteria listed above. The localization of the syncytins is still rather inconsistent with syncytin-1 being described in villous cytotrophoblasts and/or syncytiotrophoblast (Yu et al. 2002, Mi et al. 2000, Lee et al. 2001, Blond et al. 2000) and syncytin-2 being detected in villous cytotrophoblasts (Esnault et al. 2008, Malassine et al. 2007, 2008). Furthermore, also the number of cells staining positive for the syncytins raises questions as the number of cytotrophoblasts reported to stain for syncytin-1 and -2 is much higher than the number of fusion events. Thus, it seems as if the real fusogens of villous trophoblast fusion might not be detected yet or other major players still need to be identified.
9.4 Pitfalls in Dealing with Trophoblast Fusion In Vitro 9.4.1 Phenotype of Isolated Primary Trophoblasts In vivo syncytial fusion in the villous trophoblast tissue only takes place between a cytotrophoblast and the overlying syncytiotrophoblast, while intercellular syncytial fusion between two neighbouring cytotrophoblasts has never been observed in the human placenta. Intercellular fusion between two mononucleated villous cytotrophoblasts should not occur during pregnancy since this would reduce the number of cytotrophoblasts serving as stem cells to maintain the villous trophoblast layer. If the continuity of fusion events with the syncytiotrophoblast stops, this multinucleated layer would become necrotic since continuous fusion is imperative for the maintenance of this layer. As described, pure preparations of cytotrophoblasts isolated from placental villi should not fuse with each other but rather only with the syncytiotrophoblast. However, syncytial fusion of isolated primary trophoblasts in vitro has been described repeatedly. Here are putative explanations why isolated primary trophoblasts fuse in vitro: (1) In vivo, villous cytotrophoblasts are able to change their phenotype dependent on the environmental conditions. As soon as the syncytiotrophoblast layer above is removed they change to an extravillous phenotype. A similar shift in phenotype may occur after isolation and in vitro the villous cytotrophoblasts shift towards an extravillous phenotype. This may also explain why isolated cytotrophoblasts do no longer proliferate in culture. In vivo some extravillous cytotrophoblasts fuse with each other to generate multinucleated giant cells, mostly found at the border between decidua and myometrium (Al-Nasiry et al. 2009). Thus, the shift of phenotype after isolation may allow primary cytotrophoblasts to fuse since they do no longer possess a villous phenotype but rather an extravillous phenotype. Unfortunately, routine staining of such cells for HLA-G expression is not performed. Since only extravillous trophoblasts stain positive for HLA-G this would be an easy test of this hypothesis.
9
Mechanisms Regulating Human Trophoblast Fusion
213
(2) The second hypothesis finds the explanation of cell–cell fusion of primary trophoblasts in the long isolation procedure. During isolation of cytotrophoblasts the syncytiotrophoblast needs to be removed and is fragmented into a large number of smaller particles (Huppertz et al. 1999). Such particles may be anuclear membrane vesicles, oligonucleated or multinucleated fragments, or may be mononucleated syncytial fragments looking very much like a cytotrophoblast. The amount of such fragments depends on the details of the isolation procedure as well as on whether trophoblast is isolated from first trimester or term placentas. The syncytial fragments may be the first fusion partners of cytotrophoblasts. As soon as a first small syncytium is generated, further fusion events will enlarge it. (3) Interestingly, the very crude trophoblast isolation procedure that has been published 20 years ago (Kliman et al. 1986) is still in use, although intensive efforts have been made to remove the syncytial fragments from the preparations of cytotrophoblasts (Guilbert et al. 2002). Today, protocols are in place to prepare cultures of highly pure mononucleated cytotrophoblasts (Guilbert et al. 2002). However, in such cultures isolated trophoblasts still show syncytial fusion even if they maintain their villous phenotype and are pure preparations of cytotrophoblasts. Here, the explanation might be as follows: Villous cytotrophoblasts are isolated by using trypsin digestion, which is known to disrupt plasma membranes. Thus, it may be possible that during trypsinization plasma membranes are opened and closed resulting in the formation of hybrids between anuclear syncytiotrophoblast fragments and cytotrophoblasts. The plasma membrane of such mononucleated hybrids partly consists of the syncytial plasma membrane. Hence, those cells are no longer ‘pure’ cytotrophoblasts although they may look identical to real cytotrophoblasts under the microscope. The hybrid cells may again be the ‘seed crystals’ for syncytial fusion between hybrids and pure cytotrophoblasts. As a result syncytial fusion between cytotrophoblast and syncytiotrophoblastic parts of the hybrids’ membranes occurs in vitro similar to the situation in vivo. However, syncytial fusion in vitro is identified and incorrectly termed fusion between two cytotrophoblasts.
9.4.2 The Use of β-hCG to Determine the Extent of Trophoblast Fusion The most extensively used model for trophoblast fusion is the choriocarcinoma cell line BeWo. Fusion of BeWo cells can be triggered by forskolin and is used to mimic in vivo syncytialisation of villous trophoblast. β-hCG is commonly used as a marker for trophoblast syncytialisation (Butler et al. 2009) as this protein is only expressed in the syncytiotrophoblast. Placental protein 13 (PP13) has also been shown to be exclusively expressed in the syncytiotrophoblast layer of the villous trophoblast (Than et al. 2009). Comparing both marker proteins, we found that β-hCG protein expression is not necessarily linked to syncytial fusion of BeWo cells (Orendi et al. unpublished results). Former studies have already shown that cell fusion is not necessarily linked to β-hCG mRNA expression (Lin et al. 1999). So far, this has not
214
B. Huppertz and M. Gauster
been shown on the protein level yet. We found that at least in BeWo cells β-hCG protein expression and syncytial fusion can be uncoupled, while this is not the case for protein expression of PP13. Protein expression of β-hCG does not necessarily demonstrate syncytial fusion and thus, β-hCG should no longer be used as a single marker to determine syncytial fusion of BeWo cells.
9.5 Conclusions Tightly controlled syncytial fusion of villous cytotrophoblasts with the syncytiotrophoblast needs to take place throughout pregnancy to constantly maintain the multinucleated layer and thus to sustain the placental barrier. Various key players have been suggested such as cytokines, hormones, protein kinases, and transcription factors. Recently syncytins 1 and 2 as well as caspase 8 have been proposed to play major roles in trophoblast fusion. However, it needs to be further elucidated what roles both protein families, the syncytins and the caspases, play in priming the cells for fusion and in generating the fusion pore complex. Conflict of interest The authors do not have any potential or actual personal, political, or financial interest in the material or information described in this paper.
References Adler RR, Ng AK, Rote NS (1995) Monoclonal antiphosphatidylserine antibody inhibits intercellular fusion of the choriocarcinoma line, JAR. Biol Reprod 53:905–910 Al-Nasiry S, Vercruysse L, Hanssens M et al (2009) Interstitial trophoblastic cell fusion and E-cadherin immunostaining in the placental bed of normal and hypertensive pregnancies. Placenta 30:719–725 Baczyk D, Satkunaratnam A, Nait-Oumesmar B et al (2004) Complex patterns of GCM1 mRNA and protein in villous and extravillous trophoblast cells of the human placenta. Placenta 25: 553–559 Barnoy S, Glaser T, Kosower NS (1998) The calpain-calpastatin system and protein degradation in fusing myoblasts. Biochim Biophys Acta 1402:52–60 Barnoy S, Maki M, Kosower NS (2005) Overexpression of calpastatin inhibits L8 myoblast fusion. Biochem Biophys Res Commun 332:697–701 Bartoli M, Richard I (2005) Calpains in muscle wasting. Int J Biochem Cell Biol 37:2115–2133 Benirschke K, Kaufmann P, Baergen R (2006) Pathology of the human placenta, 5th edn. Springer, New York, NY Bevers EM, Comfurius P, Zwaal RF (1996) Regulatory mechanisms in maintenance and modulation of transmembrane lipid asymmetry: pathophysiological implications. Lupus 5:480–487 Black S, Kadyrov M, Kaufmann P et al (2004) Syncytial fusion of human trophoblast depends on caspase 8. Cell Death Differ 11:90–98 Blond JL, Lavillette D, Cheynet V et al (2000) An envelope glycoprotein of the human endogenous retrovirus HERV-W is expressed in the human placenta and fuses cells expressing the type D mammalian retrovirus receptor. J Virol 74:3321–3329 Butler TM, Elustondo PA, Hannigan GE et al (2009) Integrin-linked kinase can facilitate syncytialization and hormonal differentiation of the human trophoblast-derived BeWo cell line. Reprod Biol Endocrinol 7:51
9
Mechanisms Regulating Human Trophoblast Fusion
215
Chen CP, Chen LF, Yang SR et al (2008) Functional characterization of the human placental fusogenic membrane protein syncytin 2. Biol Reprod 79:815–823 Chen EH, Olson EN (2004) Towards a molecular pathway for myoblast fusion in Drosophila. Trends Cell Biol 14:452–460 Chen EH, Olson EN (2005) Unveiling the mechanisms of cell–cell fusion. Science 308:369–373 Crocker IP, Strachan BK, Lash GE et al (2001) Vascular endothelial growth factor but not placental growth factor promotes trophoblast syncytialization in vitro. J Soc Gynecol Investig 8:341–346 Dalton P, Christian HC, Redman CW et al (2007) Membrane trafficking of CD98 and its ligand galectin 3 in BeWo cells–implication for placental cell fusion. FEBS J 274:2715–2727 Daoud G, Amyot M, Rassart E et al (2005) ERK1/2 and p38 regulate trophoblasts differentiation in human term placenta. J Physiol 566:409–423 Decout A, Labeur C, Goethals M et al (1998) Enhanced efficiency of a targeted fusogenic peptide. Biochim Biophys Acta 1372:102–116 De Falco M, Fedele V, Cobellis L et al (2004) Immunohistochemical distribution of proteins belonging to the receptor-mediated and the mitochondrial apoptotic pathways in human placenta during gestation. Cell Tissue Res 318:599–608 de Parseval N, Lazar V, Casella JF et al (2003) Survey of human genes of retroviral origin: identification and transcriptome of the genes with coding capacity for complete envelope proteins. J Virol 77:10414–10422 Eskelinen S, Lehto VP (1994) Induction of cell fusion in cultured fibroblasts and epithelial cells by microinjection of EGTA, GTP gamma S and antifodrin antibodies. FEBS Lett 339:129–133 Esnault C, Priet S, Ribet D et al (2008) A placenta-specific receptor for the fusogenic, endogenous retrovirus-derived, human syncytin-2. Proc Natl Acad Sci USA 105:17532–17537 Fong PY, Xue WC, Ngan HY et al (2006) Caspase activity is downregulated in choriocarcinoma:a cDNA array differential expression study. J Clin Pathol 59:179–183 Frendo JL, Cronier L, Bertin G et al (2003) Involvement of connexin 43 in human trophoblast cell fusion and differentiation. J Cell Sci 116:3413–3421 Frendo JL, Olivier D, Cheynet V et al (2003) Direct involvement of HERV-W Env glycoprotein in human trophoblast cell fusion and differentiation. Mol Cell Biol 23:3566–3574 Garcia-Lloret MI, Morrish DW, Wegmann TG et al (1994) Demonstration of functional cytokineplacental interactions:CSF-1 and GM-CSF stimulate human cytotrophoblast differentiation and peptide hormone secretion. Exp Cell Res 214:46–54 Gauster M, Siwetz M, Huppertz B (2009a) Fusion of villous trophoblast can be visualized by localizing active caspase 8. Placenta 30:547–550 Gauster M, Siwetz M, Orendi K et al (2009b) Caspases rather than calpains mediate remodelling of the fodrin skeleton during human placental trophoblast fusion. Cell Death Differ. doi:10.1038/cdd.2009.133 Guilbert LJ, Winkler-Lowen B, Sherburne R et al (2002) Preparation and functional characterization of villous cytotrophoblasts free of syncytial fragments. Placenta 23:175–183 Huppertz B, Bartz C, Kokozidou M (2006) Trophoblast fusion: fusogenic proteins, syncytins and ADAMs, and other prerequisites for syncytial fusion. Micron 37:509–517 Huppertz B, Frank HG, Kingdom JC et al (1998) Villous cytotrophoblast regulation of the syncytial apoptotic cascade in the human placenta. Histochem Cell Biol 110:495–508 Huppertz B, Frank HG, Reister F et al (1999) Apoptosis cascade progresses during turnover of human trophoblast:analysis of villous cytotrophoblast and syncytial fragments in vitro. Lab Invest 79:1687–1702 Huppertz B, Kaufmann P, Kingdom JCP (2002) Trophoblast turnover in health and disease. Fetal Maternal Med Rev 13:17–32 Huppertz B, Kingdom JC (2004) Apoptosis in the trophoblast – role of apoptosis in placental morphogenesis. J Soc Gynecol Investig 11:353–362 Jansson T (2001) Amino acid transporters in the human placenta. Pediatr Res 49:141–147
216
B. Huppertz and M. Gauster
Jiang B, Kamat A, Mendelson CR (2000) Hypoxia prevents induction of aromatase expression in human trophoblast cells in culture:potential inhibitory role of the hypoxia-inducible transcription factor Mash-2 (mammalian achaete-scute homologous protein-2). Mol Endocrinol 14:1661–1673 Kliman HJ, Nestler JE, Sermasi E et al (1986) Purification, characterization, and in vitro differentiation of cytotrophoblasts from human term placentae. Endocrinology 118:1567–1582 Knerr I, Schubert SW, Wich C et al (2005) Stimulation of GCMa and syncytin via cAMP mediated PKA signaling in human trophoblastic cells under normoxic and hypoxic conditions. FEBS Lett 579:3991–3998 Kudo Y, Boyd CA (2002) Human placental amino acid transporter genes:expression and function. Reproduction 124:593–600 Kudo Y, Boyd CA (2004) RNA interference-induced reduction in CD98 expression suppresses cell fusion during syncytialization of human placental BeWo cells. FEBS Lett 577:473–477 Kudo Y, Boyd CA, Millo J et al (2003) Manipulation of CD98 expression affects both trophoblast cell fusion and amino acid transport activity during syncytialization of human placental BeWo cells. J Physiol 550:3–9 Lee X, Keith JC, Jr, Stumm N et al (2001) Downregulation of placental syncytin expression and abnormal protein localization in pre-eclampsia. Placenta 22:808–812 Leisser C, Saleh L, Haider S et al (2006) Tumour necrosis factor-alpha impairs chorionic gonadotrophin beta-subunit expression and cell fusion of human villous cytotrophoblast. Mol Hum Reprod 12:601–609 Li H, Dakour J, Guilbert LJ et al (2005) PL74, a novel member of the transforming growth factorbeta superfamily, is overexpressed in preeclampsia and causes apoptosis in trophoblast cells. J Clin Endocrinol Metab 90:3045–3053 Lin C, Lin M, Chen H (2005) Biochemical characterization of the human placental transcription factor GCMa/1. Biochem Cell Biol 83:188–195 Lin L, Xu B, Rote NS (1999) Expression of endogenous retrovirus ERV-3 induces differentiation in BeWo, a choriocarcinoma model of human placental trophoblast. Placenta 20:109–118 Lyden TW, Ng AK, Rote NS (1993) Modulation of phosphatidylserine epitope expression by BeWo cells during forskolin treatment. Placenta 14:177–186 Malassine A, Blaise S, Handschuh K et al (2007) Expression of the fusogenic HERV-FRD Env glycoprotein (syncytin 2) in human placenta is restricted to villous cytotrophoblastic cells. Placenta 28:185–191 Malassine A, Frendo JL, Blaise S et al (2008) Human endogenous retrovirus-FRD envelope protein (syncytin 2) expression in normal and trisomy 21-affected placenta. Retrovirology 5:6 Mandal D, Mazumder A, Das P et al (2005) Fas-, caspase 8-, and caspase 3-dependent signaling regulates the activity of the aminophospholipid translocase and phosphatidylserine externalization in human erythrocytes. J Biol Chem 280:39460–39467 Martens S, McMahon HT (2008) Mechanisms of membrane fusion: disparate players and common principles. Nat Rev Mol Cell Biol 9:543–556 Martin I, Pecheur EI, Ruysschaert JM et al (1999) Membrane fusion induced by a short fusogenic peptide is assessed by its insertion and orientation into target bilayers. Biochemistry 38:9337–9347 Mi S, Lee X, Li X et al (2000) Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature 403:785–789 Miehe U, Kadyrov M, Neumaier-Wagner P et al (2006) Expression of the actin stress fiberassociated protein CLP36 in the human placenta. Histochem Cell Biol 126:465–471 Midgley AR, Pierce GB, Jr, Deneau GA et al (1963) Morphogenesis of syncytiotrophoblast in vivo: an autoradiographic demonstration. Science 141:349–350 Mombers C, Verkleij AJ, de Gier J et al (1979) The interaction of spectrin-actin and synthetic phospholipids. II. The interaction with phosphatidylserine. Biochim Biophys Acta 551:271–281
9
Mechanisms Regulating Human Trophoblast Fusion
217
Morrish DW, Bhardwaj D, Dabbagh LK et al (1987) Epidermal growth factor induces differentiation and secretion of human chorionic gonadotropin and placental lactogen in normal human placenta. J Clin Endocrinol Metab 65:1282–1290 Morrish DW, Bhardwaj D, Paras MT (1991) Transforming growth factor beta 1 inhibits placental differentiation and human chorionic gonadotropin and human placental lactogen secretion. Endocrinology 129:22–26 Ohtani T, Hatori M, Ito H et al (2000) Involvement of caspases in 5-FU induced apoptosis in an oral cancer cell line. Anticancer Res 20:3117–3121 Oren-Suissa M, Podbilewicz B (2007) Cell fusion during development. Trends Cell Biol 17:537–546 Panigel M (1993) The origin and structure of extraembryonic tissues. In: Redman C, Sargent I, Starkey P (eds) The human placenta. Blackwell Scientific Publications, London Potgens AJ, Schmitz U, Bose P et al (2002) Mechanisms of syncytial fusion: a review. Placenta 23 Suppl A:S107–S113 Richart R (1961) Studies of placental morphogenesis. I. Radioautographic studies of human placenta utilizing tritiated thymidine. Proc Soc Exp Biol Med 106:829–831 Rote NS, Chakrabarti S, Stetzer BP (2004) The role of human endogenous retroviruses in trophoblast differentiation and placental development. Placenta 25:673–683 Sato SB, Ohnishi S (1983) Interaction of a peripheral protein of the erythrocyte membrane, band 4.1, with phosphatidylserine-containing liposomes and erythrocyte inside-out vesicles. Eur J Biochem 130:19–25 Savill J (1998) Apoptosis. Phagocytic docking without shocking. Nature 392:442–443 Shi QJ, Lei ZM, Rao CV et al (1993) Novel role of human chorionic gonadotropin in differentiation of human cytotrophoblasts. Endocrinology 132:1387–1395 Than NG, Romero R, Goodman M et al (2009) A primate subfamily of galectins expressed at the maternal-fetal interface that promote immune cell death. Proc Natl Acad Sci USA 106:9731–9736 van den Eijnde SM, van den Hoff MJ, Reutelingsperger CP et al (2001) Transient expression of phosphatidylserine at cell–cell contact areas is required for myotube formation. J Cell Sci 114:3631–3642 Vargas A, Moreau J, Landry S et al (2009) Syncytin-2 plays an important role in the fusion of human trophoblast cells. J Mol Biol 392:301–318 White L, Dharmarajan A, Charles A (2007) Caspase-14: a new player in cytotrophoblast differentiation. Reprod Biomed Online 14:300–307 Yang M, Lei ZM, Rao Ch V (2003) The central role of human chorionic gonadotropin in the formation of human placental syncytium. Endocrinology 144:1108–1120 Yu C, Shen K, Lin M et al (2002) GCMa regulates the syncytin-mediated trophoblastic fusion. J Biol Chem 277:50062–50068
Chapter 10
Macrophage Fusion: The Making of a New Cell Agnès Vignery
Abstract Cell–cell fusion is a critical developmental step in most eukaryotes, as it is required for fertilization as well as for the formation of placenta, muscle and bone, where macrophages fuse to form a new cell called osteoclast. Macrophages can also fuse in chronic inflammatory reactions and in tumors where they form giant cells. While our understanding of the fusion mechanism used by viruses to infect cells has made major progress, the mechanism used by eukaryotic cells to fuse with one another remains poorly understood. Macrophages are unique in that they are mononucleate cells that fuse in rare instances to make a new cell, which suggests a regulated divergence from their routine activity. Here we discuss the formation of osteoclasts and giant cells with a special focus on the fusion mechanism of their mononucleate precursors, which belong to the mononuclear phagocyte lineage. We also discuss the unique mechanism macrophages utilize to recognize each other as self before they merge their cytoplasm into a new cell. Keywords CD200-CD200R · cell-cell fusion · DC-STAMP · fusogens · giant cells · inflammation · macrophage · monocyte · multinucleation · osteoclast · self recognition · SIRPa-CD47 Abbreviations Ae2a,b AFF-1 Bcl-2 Btk CAII CaMKIV CD CLC7
Na+-independent chloride/bicarbonate anion exchange 2 Anchor-cell fusion failure-1 B-cell leukemia/lymphoma 2 Bruton’s tyrosine kinase Carbonic anhydrase II Ca2+ /calmodulin-dependent protein kinases IV Cluster of differentiation Chloride channel 7
A. Vignery (B) Departments of Orthopaedics and Cell Biology, Yale School of Medicine, New Haven, CT 06510, USA e-mail:
[email protected]
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_10, C Springer Science+Business Media B.V. 2011
219
220
DC DC-STAMP DAP12 EFF-1 ENV FcRγ Flt3 Gab2 Grb-2 HIV-1 IKKβ IL IRAK-M LTBP-3 M-CSF MFR MITF MKP-1 MMP-9 MPS NFATc1 NFκ B oc OSTM1 pamps PLCγ PLEKHM1 RANK RANKL RBC SIRP-α TCTA TLR TNF TRAF6 TSP-1 TRP TRPV4
A. Vignery
Dendritic cell Dendritic cell-specific transmembrane protein DNAX-activating protein 12 Epithelial fusion failure Envelope protein Fc receptor common γ subunit Fms-like tyrosine kinase 3 Grb-2 – associated binding protein 2 Growth factor receptor bound protein 2 Human immunodeficiency virus type Inhibitor of κB (IκB) kinase Interleukin Interleukin-1-associated kinase in macrophages Latent TGF-beta binding proteins-3 Macrophage colony stimulating factor Macrophage fusion receptor Microphthalmia-associated transcription factor MAP kinase phosphatase-1 Matrix-metalloprotease-9 Mononuclear phagocyte system Nuclear factor of activated T-cells c1 Nuclear Factorκ B Osteoclast Osteopetrosis associated transmembrane protein 1 Pathogen-associated molecular patterns Phospholipase C γ Pleckstrin homology domain-containing family M (with RUN domain) member 1 Receptor activator of NFκ B RANK ligand Red blood cells Signal regulatory protein-alpha T-cell leukemia translocation-associated gene Toll-like receptor Tumor necrosis factor TNF receptor-associated factor 6 Thrombspondin-1 Transient receptor potential TRP cation channels, subfamily V, member 4
Contents 10.1 Macrophage Multinucleation . . . . . . . . . . . . . . . . . . . . . . . . 10.1.1 What Are Macrophages? . . . . . . . . . . . . . . . . . . . . . . . 10.1.2 Osteoclasts and Giant Cells . . . . . . . . . . . . . . . . . . . . .
221 221 222
10
Macrophage Fusion: The Making of a New Cell
10.1.3 Cellular Fusogens . . . . . . 10.1.4 Macrophage Fusion Machinery 10.1.5 Recognition of Self . . . . . 10.2 Conclusion . . . . . . . . . . . . References . . . . . . . . . . . . . . .
. . . . .
. . . . .
. . . . .
221
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
224 225 227 229 230
10.1 Macrophage Multinucleation 10.1.1 What Are Macrophages? Macrophages are mononucleate cells that play a critical role in immunity, development, homeostasis and repair of tissues. They are the guards of the immune system as they detect pathogens and novel antigens, and secure their clearance. They also process antigens to initiate and secure innate and adaptive immunity. What is unique about macrophages, in addition to their vast repertoire of functions, and their versatility, is their ubiquitous distribution among organs and tissues. Indeed macrophages literally “perfuse” tissues, yet remain separated from one another by a distance that appears relatively constant, and thereby might define a “macrophage network”. Establishing a network might allow macrophages to access and propagate information about novel antigens/pathogens at a fast pace so as to act upon in an educated, timely and efficient manner. In addition, macrophages often function as control switches for immune system balance between pro- and anti-inflammatory reactions in health and disease, as witnessed more and more by investigators studying, for instance, the cardiovascular system (Mantovani et al. 2009), the kidney (Matsumoto et al. 2010), and the digestive system (Casanova and Abel 2009). Macrophages originate from blood monocytes, which themselves originate from monoblasts that differentiate in the bone marrow from hematopoietic stem cells. Macrophages belong to the mononuclear phagocyte system (MPS), which consists of a family of myeloid precursors in the bone marrow called monoblasts, circulating blood monocytes, tissue macrophages and dendritic cells (DC). Indeed, it has become accepted by most investigators that monocytes, macrophages and some DC subsets originate from a common myeloid progenitor and that their differentiation is not linear, rather, it includes alternative pathways (for recent advances in the regulation of growth and differentiation of myeloid cells, see (Auffray et al. 2009, Chang 2009, Yona and Jung 2010). While the variety of MPS cell subtypes is being uncovered steadily, their number of functions is expanding proportionally. Yet, progress in the identification of an osteoclast or a giant cell precursor cell has been lacking, suggesting that the versatility and adaptability of macrophages is such that osteoclast and giant cell precursor cells may acquire their phenotype in situ. Most recently, Alnaeeli and collaborators (2006) reported that murine bone marrow-derived and splenic CD11c+ DC can develop into functional osteoclasts in a RANK/RANKL-dependent manner. Such CD11c+ DC-derived osteoclasts are capable of inducing bone loss after adoptive transfer in vivo. Speziani and colleagues (2007) subsequently showed that mature murine DCs, not plasmacytoid DCs, generated in vitro from Fms-like tyrosine kinase 3 (Flt3)+ bone marrow
222
A. Vignery
progenitors or ex vivo from purified spleen cells, can fuse to form osteoclasts in response to NF-κB ligand (RANKL) in vitro. The same team (Wakkach et al. 2008) reported that conventional splenic DCs cultured in the presence of macrophage colony stimulating factor (M-CSF) and RANKL differentiate into functional osteoclasts, and that transplanted splenic DCs are able to home into bone marrow and partially reverse the osteopetrotic phenotype of oc/oc mice by restoring bone resorption. It appears therefore that osteoclasts can form from a relatively wide variety of committed myeloid cells. Indeed, when alveolar macrophages from rats or mice are plated at high density and cultured in the sole presence of medium supplemented with serum, they fuse spontaneously and express osteoclast markers (Vignery 2000, Vignery 2005a, Chen et al. 2007, Vignery 2008a, b).
10.1.2 Osteoclasts and Giant Cells In order to appreciate the formation of multinucleate macrophages, which remain rare cells, it is critical to take into consideration the fact that their precursors, the macrophages, are found neither in aggregates nor in colonies in the tissues in which they reside. In specific instances, however, it is reasonable to assume that a macrophage undergoes rapid clonal expansion to respond to an unusual demand, such as a fast growing parasite colony or a foreign body/ implant. Daughter cells fuse to empower them selves with the ability to destroy or recycle the invaders or the foreign antigens. In the case of osteoclasts, which resorb bone, it is possible that the clonal expansion of macrophages takes places at sites of microfractures where bone is no longer functional, hence expose new antigenic bone components/epitopes that need to be resorbed by osteoclasts and replaced by new bone. Although the formation and activation of multinucleate macrophages, which include osteoclasts in bone and giant cells in chronic inflammatory reactions (Fig. 10.1), has been the theme of recent reviews (Takayanagi 2009, Brodbeck and Anderson 2009, Helming and
Fig. 10.1 (continued) resorbing activity, which is also dependent on M-CSF and RANKL or GM-CSF and IL-4. MITF, microphthalmia-associated transcription factor; Bcl-2, B-cell leukemia/lymphoma 2; FcRγ, Fc receptor common γ subunit; DAP12, DNAX-activating protein 12; TRAF6, tumor necrosis factor (TNF) receptor-associated factor 6; Gab2, growth factor receptor bound protein 2 (Grb-2)-associated binding protein 2; Btk, Bruton s tyrosine kinase; PLCγ, phospholipase C γ; CaMKIV, Ca2+ /calmodulin-dependent protein kinases IV; IKKβ, inhibitor of κB (IκB) kinase; TRPV4, transient receptor potential (TRP) cation channels, subfamily V, member 4; NFATc1, nuclear factor of activated T-cells c1; DC-STAMP, dendritic cell-specific transmembrane protein; CLC7, chloride channel 7; OSTM1, osteopetrosis associated transmembrane protein 1; CAII, carbonic anhydrase II; PLEKHM1, pleckstrin homology domain-containing family M (with RUN domain) member 1; TCTA, T-cell leukemia translocation-associated gene (Kotake et al. 2009); IRAK-M, Interleukin-1-associated kinase in macrophages (Li et al. 2005); MKP-1, MAP kinase phosphatase-1 (Carlson et al. 2009); LTBP-3; Latent TGF-beta binding proteins-3; Ae2a,b , Na+-independent chloride/bicarbonate anion exchange 2 (Jansen et al. 2009); Brag2/Dock 180 (Pajcini et al. 2008); MMP-9, matrix-metalloprotease-9
10
Macrophage Fusion: The Making of a New Cell
223
Fig. 10.1 Osteoclast differentiation: key transcription factors, signaling molecules and transmembrane receptors. Macrophage colony-stimulating factor (M-CSF) and granulocyte macrophagecolony stimulating factor (GM-CSF) promote the differentiation of hematopoietic stem cells into macrophage-colony forming units, which are precursor cells for osteoclasts and giant cells. While the molecules listed have been identified for the most part in osteoclasts, many have been confirmed in giant cells and in osteoclasts, such as MFR-CD47, CD200-CD200R, CD44, DC-STAMP, to name a few. Receptor activator of NF-κB (RANK) and IL-4R activation further promote the differentiation of fusogenic macrophages. Multinucleate macrophages acquire
224
A. Vignery
Gordon 2009), the functional implications of multinucleation are poorly discussed or reviewed. We had proposed in previous reviews and in one opinion (Chen et al. 2007, Vignery 2000, 2005a, 2008a) that multinucleate macrophages are highly focused and perform a limited number of tasks such as resorption of the substrate on which they have formed and induction of the repair of the resorbed tissue; in contrast to mononucleate macrophages, their mobility is limited to their birth place. Multinucleation has two main consequences in macrophages. First, it increases their size, and, consequently, it endows them with the ability to resorb large components. Indeed, their size, hence their number of nuclei, might be proportional to the size of the target to be resorbed. Using run on assays to assess RNA synthesis, Jurdic and colleagues (Boissy et al. 2002) demonstrated a strong correlation between transcription and resorption. Instead of internalizing a target, such as for instance a bacterium, and directing it to lysosomes for degradation, multinucleate macrophages use a mechanism that allows the extracellular degradation of targets, which can be assimilated to an extroversion. When in contact with their targets, multinucleate macrophages form a sealing zone or ring that seals off an extracellular compartment. The content of that compartment has a low pH that facilitates the dissolution (e.g. bone) or killing (e.g. pathogens) of the target, and the activation of lysosomal enzymes, hence is considered as an ‘extracellular lysosome’. The molecular mechanism by which multinucleate macrophages, such as osteoclasts, resorb bone, was reviewed recently by Bruzzaniti and Baron (2006). Multinucleation endows macrophages with an enhanced resorption capacity, meaning that two macrophages cannot do what one binucleate macrophage does. Hence, multinucleate macrophages are more than the sum of their parts. This capacity is best illustrated by the vast array of genes that are differentially regulated during osteoclastogenesis. Indeed, multinucleation is an essential step in the differentiation of osteoclasts as mononucleated macrophages cannot resorb bone efficiently. In diseases in which macrophages cannot fuse, such as in some forms of osteopetrosis, bones become thick and brittle. Hence, multinucleate macrophages, whether osteoclasts or giant cells, are new cells endowed with fewer but more specialized functions, which are to resorb antigens and pathogens. Thus, osteoclasts and giant cells, in their own ways, protect the body by securing its homeostasis and by isolating pathogens, respectively. After osteoclasts have completed their task, their fate remains ill defined. It is likely that osteoclasts undergo apoptosis, although the mechanisms that lead to the death of osteoclasts are poorly understood. By contrast to osteoclasts, giant cells appear to stay around for extended periods of time within granulomas. But the fate of giant cells, like that of osteoclasts, remains ill defined.
10.1.3 Cellular Fusogens In order to fuse, cells follow a sequence of events starting with cell–cell attraction, recognition, alignment and adhesion. The plasma membrane of both cells fuse via the tethering of aqueous pores, expansion of the pores leading to the merge of the
10
Macrophage Fusion: The Making of a New Cell
225
cytoplasms. Such membrane fusion is thought to be mediated by fusogenic proteins that lower the energy barrier of the fusion process (Sapir et al. 2008). Sapir and colleagues proposed that while all fusion reactions could in theory depend on destabilization of membrane bilayers and by proteins that normally carry out fusionunrelated functions, all membrane fusion events studied so far rely on fusogens that mediate the fusion of lipid bilayers. This is because viral fusogens have been well characterized, and have served as a standard to define the criteria that are necessary for a protein to be considered as a fusogen. While many viral fusion proteins have been identified, as viral fusion relies on a single protein, candidate eukaryote fusogens have been difficult to validate. The characteristics that are expected from bona fide cellular fusogens are similar to those of viral fusogens and must be necessary to induce cell fusion and localize at the membrane fusion site; such a fusogen must be sufficient to mediate fusion in heterologous cells or liposomes. In eukaryotes, only two families of molecules have been stamped as true fusogens; one is the family of Syncytins, which mediate the fusion of fetal cytotrophoblasts in some mammals, and the other is the FF family of fusogens in C. elegans, which mediate developmental fusion events. One of the FF proteins is called epithelial fusion failure (EFF-1) and is required for most cell fusion events in the epidermis, pharynx, and vulva (Mohler et al. 2002), the other protein is called anchor-cell fusion failure-1 (AFF-1) and is essential for specific fusion events leading to the formation of small syncytia, such as fusion of the anchor cell to specific uterine cells to form the nematode’s hymen, and fusion of a subset of vulval cells to form vulval rings (Sapir et al. 2007). In contrast to viral fusogens, there are limited data on the structure and function of Syncytins. Syncytins are single transmembrane proteins with characteristics of viral fusogens, including a potential amphiphilic fusion peptide, and undergo maturation steps that are reminiscent of HIV ENV. Indeed, Syncytin-1 and -2 share structural domains with several class I viral fusogens, strengthening the notion that they originated from viral fusogens, with which they share a similar fusion mechanism. In the case of FF proteins, it appears that expression of EFF-1 in only one of two bound partners is insufficient for fusion (Podbilewicz et al. 2006), and similar observations were made in vivo with AFF-1 (Sapir et al. 2007). Hence, unlike viral fusogens, which include Syncytins, FF fusogens require homotypic interaction, suggestive of a tight control in this developmental fusion event. Such a tight control might be required for the fusion of macrophages, which are for the most part mononucleate. Indeed, as we discuss below, macrophages undergo homotypic fusion, although the expression of DC-SAMP and CD36 was reported to be required in only one of the two fusion partners (Yagi et al. 2005, Helming et al. 2009).
10.1.4 Macrophage Fusion Machinery Although macrophages have long been recognized with the ability to fuse to form osteoclasts and giant cells, the mechanisms they use to recognize each other and fuse to merge their cytoplasm remain poorly characterized. Indeed, none of the
226
A. Vignery
Fig. 10.2 Macrophage recognition/fusion. Macrophages recognize each other as self when MFR/SIRPα binds to CD47 and CD200 binds to CD200R. Initially, the long form of MFR associates with CD47, followed by the short form of MFR/SIRPα (MFR-s). Then, once CD200 is expressed, it binds to CD200R, which secures plasma membranes’ interaction. The shedding of the extracellular domain of MFR might facilitate this association (Cui and Vignery, unpublished observation). The distance between macrophage plasma membranes could be reduced to 5–10 nm if MFR-s and CD47 bend upon binding. The shedding of the extracellular domain of CD44 further facilitates plasma membranes from opposite cells to get closer, and fuse. When the intracellular domain of CD44 is cleaved by a gamma secretase complex, it translocates to the nucleus to promote the activation of NF-κB. NF-κB is a transcription factor that is indispensable for osteoclastogenesis. The role of DC-STAMP remains elusive, but its expression is predominantly located intracellularly in fusogenic macrophages, and in the plasma membrane of less fusogenic donor macrophages
10
Macrophage Fusion: The Making of a New Cell
227
surface proteins identified so far by us and others, which include MFR/SIRPαCD47, CD200-CD200R, CD44 and DC-STAMP, (Fig. 10.2, and review (Vignery 2008a)) can be defined as a fusogens. In the case of CD47, two lines of investigation support its role in the formation of multinucleate macrophages/osteoclasts. One, published by Uluçkan et al. (2009), demonstrates that CD47 regulates osteoclastogenesis, and that mice that lack CD47 have increased bone mass associated with a decrease in the number of osteoclasts. The second (Kukreja et al. 2009) demonstrates a dominant role for CD47-thrombspondin-1 (TSP-1) interaction in myeloma-induced fusion of human dendritic cells, and its implication in bone diseases where metastases home to promote its resorption by osteoclasts. Myeloma is characterized by an increase in multinucleate osteoclasts in close proximity to tumor cells. Here, the authors show that dendritic cells fuse with myeloma cells to make multinucleate osteoclasts, and that perturbation of CD47-TSP-1 interaction attenuates hypercalcemia induced by parathryoid hormone. Indeed, they show that down regulation of CD47 expression by RNA interference abrogates tumor-induced osteoclast formation, which supports a role for CD47 in fusion, although the role of SIRPα is not discussed. In the case of DC-STAMP, its deletion via gene targeting in mice leads to mild osteopetrosis, which is due to a defect in the fusion of osteoclast precursor cells. Most recently, Mensah and colleagues (2010) demonstrated that DC-STAMP is expressed on osteoclast precursor cells as a dimer, which they detected by flow cytometry using a monoclonal antibody directed against the extracellular domain of DC-STAMP. They further showed that RANKL induces fusion by stimulating DCSTAMP internalization in some cells, hence generating DC-STAMPlo cells (low cell surface expression) that are the fusing partners. In contrast, DC-STAMPhi cells can only act as mononuclear donor cells. These results suggest a role for DC-STAMP intracellularly, possibly by acting as a scaffold to facilitate the functional assembly of a protein machinery that triggers the fusion of cellocytosed macrophages (see discussion below (Vignery 2005a, b)). It is therefore reasonable to assume that macrophage fusion requires cell–cell attraction, followed by expression of plasma membrane attachment molecules that mediate recognition of self in a homotypic manner. Expression of fusogens may or may not be present in both fusing partners, leading to fusion and merging of the cytoplasm from both partners to form a binucleate cell. As it is the case in some viruses, fusion might require activation of the fusogens, such as exposure of a hydrophobic peptide and conformational changes. In addition, cleavage of surface proteins might help reduce the distance between fusing partners, as it is the case for CD44 (Cui et al. 2006).
10.1.5 Recognition of Self Macrophages are master dispatchers that know how to discriminate self from nonself, and to dispose of antigens accordingly, i.e. decide whether to cellocytose a live macrophage or to phagocytose an apoptotic cell. Such an educated decision must rely on specific cell-surface determinants expressed by both fusing partners.
228
A. Vignery
In that respect, it is striking that the two pairs of receptor-ligand that we identified as putative fusion molecules, namely SIRPα-CD47 and CD200-CD200R are indeed members of the signature of self. Oldenborg and collaborators (2000) had originally suggested that SIRPα–CD47 (which we reported as macrophage fusion receptor/MFR-CD47) might be a primitive self-recognition sensor mechanism (Oldenborg et al. 2000). They demonstrated that SIRPα delivers a negative regulatory signal. They showed that red blood cells (RBCs) that lack CD47 are rapidly cleared from the bloodstream by splenic red-pulp macrophages, and that CD47, expressed on normal RBCs, prevents this elimination by binding to SIRPα. This observation was the first indication that SIRPα–CD47 belongs a primitive self-recognition mechanism. Oldenborg and colleagues went on to demonstrate that clearance and phagocytosis of opsonized (antibody-coated) RBCs is also regulated by SIRPα–CD47. This demonstration led van den Berg and colleagues to propose that the structural and functional properties shared between antigen receptors and SIRPs strongly suggest a common ancestor in primitive vertebrates. Although their proposition is attractive, it does not take into consideration the recognition of self that is required for macrophage-macrophage fusion. It is possible that the fusion of macrophages, which leads to multinucleation, requires recognition of self, and is an integral part of the innate immune defense mechanism. Blander and Medzhitov (2004) identified two modes of maturation of phagosomes, constitutive and inducible, whose differential engagement depends on the ability of the cargo to trigger Toll-like receptor (TLR) signaling. The fact that TLRs recognize pathogenassociated molecular patterns (‘pamps’), but not apoptotic cells that are routed via the constitutive pathway, suggests that the fate of the internalized cell is receptormediated and decided at an early stage. Hence, the internalized cell is not readily routed to lysosomes for degradation. This fundamental observation might explain the survival and cellular integration of cellocytosed macrophages and possibly of somatic and tumor cells, as discussed below. Most relevant to the role of CD47 as a marker of self are two recent reports by Jaiswal and colleagues (2009), and Majeti and colleagues (2009), both from the Weissman group, which demonstrated that CD47 expression is upregulated on circulating hematopoietic stem cells and leukemia cells to avoid phagocytosis, and CD47 is an adverse prognostic factor and therapeutic antibody target on human acute myeloid leukemia stem cells, respectively. While the central role plaid by CD47 as a marker of self supports a role in attachment leading to cellocytosis and survival of fusing partners, it does not take into account the induced expression of its receptor, MFR/SIRPα, which plays a central role in its downstream negative signaling to prevent macrophage activation. Thus, CD47-SIRPα interaction stands out as a potent tolerogenic mechanism, which might facilitate fusion. Regarding CD200 and its receptor CD200R, it has now become accepted that CD200 expression is central to induction of tolerance to tissue allograft (Gorczynski et al. 2009). Like SIRPα, CD200R belongs to the paired receptor families of membrane proteins, which are characterized by similar extracellular domains and varying cytoplasmic domains that transduce opposite downstream signals, i.e. stimulatory versus inhibitor signals. Paired receptors are frequently characterized by a high
10
Macrophage Fusion: The Making of a New Cell
229
degree of polymorphism and varying numbers of activating genes that are often delivered by pathogens. Indeed, polymorphism in Sirpα modulates engraftment of human hematopoietic stem cells (Takenaka et al. 2007). In contrast with SIRPα, which is expressed at low level by macrophages, and highly induced at the onset of fusion, CD200 is expressed only after induction of fusion (Cui et al. 2007), whether spontaneous or in response to RANK activation so as to allow for cellocytosis and cell survival leading to fusion of plasma membranes from fusing partners and merging of cytoplasm, which is shared by both nuclei. This suggests that resident macrophages, such as alveolar macrophages, which do not express CD200 on their plasma membrane, can dispose of apoptotic macrophages that do not express CD200 and might be recognized as non-self. Like the SIRP family, the CD200R family is a paired receptor that consists of one inhibitory member (CD200R) and several activating members (at least five in mice). CD200 interacts with CD200R with a low affinity that is typical of membrane proteins, and either fails to bind or binds with a 100-fold lower affinity to activating CD200 receptors (Akkaya and Barclay 2010). Similar to CD47, expression of CD200 might help tumor stem cells evade the immune system (Kawasaki and Farrar 2008). This is in addition to playing a role in the prevention of experimental autoimmune encephalomyelitis and collagen-induced arthritis, hence maintaining self-tolerance, CD200 is also involved in attenuating immune responses to skin grafts. Indeed, CD200 appears to play an important role in controlling autoimmunity, inflammation and adaptive immune responses. In addition, CD200 expression by tumor cells, and possibly tumor stem cells, might be a prognostic marker for cancer progression, metastasis and relapse. Hence, similar to CD47, CD200 expression might be upregulated on circulating tumor stem cells to avoid phagocytosis. Of note, CD200 is also expressed on embryonic stem cells (see http://amazonia.transcriptome.eu), and co-expressed with cancer stem cell markers, such as CD44, which incidentally plays a central role in macrophage fusion (Cui et al. 2006). Based on CD200 connection to stem cells, its role in immune tolerance and tumor immunity, Kawasaki and Farrar (2008) proposed that CD200 defines a cancer stem cell population that has the capacity to escape the immune response. Hence, both CD47 and CD200 erase the non-self components of tumor and foreign antigens to evade an immune mounted attack. CD47 and CD200, together with their inhibitory receptor SIRPα and CD200R, define a sophisticated machinery that macrophage utilize to recognize each other as self to fuse and merge their cytoplasm into a new multinucleate cell.
10.2 Conclusion While no bona fide fusogen has yet been uncovered in macrophages, a number of factors and plasma membrane molecules that mediate the attraction and the attachment of macrophages to one another to prime them for fusion have been identified. The fusion of macrophages is a rare event, which requires the regulated expression of markers of self, such as CD47 and CD200. These molecules are either not expressed (CD200) or expressed at low level in “naïve” or non-activated
230
A. Vignery
macrophages, respectively, and highly expressed at the onset of fusion. The identification of additional members of the fusion machinery will facilitate the prevention and treatment of diseases in which macrophages and other cells, such as immune and tumor cells, are involved.
References Akkaya M, Barclay AN (2010) Heterogeneity in the CD200R paired receptor family. Immunogenetics 62:15–22 Alnaeeli M, Penninger JM, Teng YTA (2006) Immune interactions with CD4+ T cells promote the development of functional osteoclasts from murine CD11c+ dendritic cells. J Immunol 177:3314–3326 Auffray C, Sieweke MH, Geissmann F (2009) Blood monocytes: development, heterogeneity, and relationship with dendritic cells. Annu Rev Immunol 27:669–692 Blander JM, Medzhitov R (2004) Regulation of phagosome maturation by signals from toll-like receptors. Science 304:1014–1018 Boissy P, Saltel F, Bouniol C et al (2002) Transcriptional activity of nuclei in multinucleated osteoclasts and its modulation by calcitonin. Endocrinology 143:1913–1921 Brodbeck WG, Anderson JM (2009) Giant cell formation and function. Curr Opin Hematol 16: 53–57 Bruzzaniti A, Baron R (2006) Molecular regulation of osteoclast activity. Rev Endocr Metab Disord 7:123–139 Carlson J, Cui W, Zhang Q et al (2009) Role of MKP-1 in osteoclasts and bone homeostasis. Am J Pathol 175:1564–1573 Casanova JL, Abel L (2009) Revisiting Crohn’s disease as a primary immunodeficiency of macrophages. J Exp Med 206:1839–1843 Chang ZÄ (2009) Recent development of the mononuclear phagocyte system: in memory of Metchnikoff and Ehrlich on the 100th anniversary of the 1908 nobel prize in physiology or medicine. Biol Cell 101:709–721 Chen EH, Grote E, Mohler W et al (2007) Cell–cell fusion. FEBS Lett 581:2181–2193 Cui W, Cuartas E, Ke J (2007) CD200 and its receptor, CD200R, modulate bone mass via the differentiation of osteoclasts. Proc Natl Acad Sci U S A 104:14436–14441 Cui W, Ke JZ, Zhang Q et al (2006) The intracellular domain of CD44 promotes the fusion of macrophages. Blood 107:796–805 Gorczynski RM., Chen Z, He W et al (2009) Expression of a CD200 transgene is necessary for induction but not maintenance of tolerance to cardiac and skin allografts. J Immunol 183:1560–1568 Helming L, Gordon S (2009) Molecular mediators of macrophage fusion. Trends Cell Biol 19:514–522 Helming L, Winter J, Gordon S (2009) The scavenger receptor CD36 plays a role in cytokineinduced macrophage fusion. J Cell Sci 122:453–459 Jaiswal S, Jamieson CH, Pang WW et al (2009) CD47 is upregulated on circulating hematopoietic stem cells and leukemia cells to avoid phagocytosis. Cell 138:271–285 Jansen IDC, Mardones P, Lecanda F et al (2009) Ae2a,b-Deficient mice exhibit osteopetrosis of long bones but not of calvaria. FASEB J 23:3470–3481 Kawasaki BT, Farrar WL (2008) Cancer stem cells, CD200 and immunoevasion. Trends Immunol 29:464–468 Kotake S, Nanke Y, Kawamoto M et al (2009) T-cell leukemia translocation-associated gene (TCTA) protein is required for human osteoclastogenesis. Bone 45:627–639 Kukreja A, Radfar S, Sun BH et al (2009) Dominant role of CD47-thrombospondin-1 interactions in myeloma-induced fusion of human dendritic cells: implications for bone disease. Blood 114:3413–3421
10
Macrophage Fusion: The Making of a New Cell
231
Li H, Cuartas E, Cui W et al (2005) IL-1 receptor-associated kinase M is a central regulator of osteoclast differentiation and activation. J Exp Med 201:1169–1177 Majeti R, Chao MP, Alizadeh AA et al (2009) CD47 is an adverse prognostic factor and therapeutic antibody target on human acute myeloid leukemia stem cells. Cell 138:286–299 Mantovani A, Garlanda C, Locati M (2009) Macrophage diversity and polarization in atherosclerosis: a question of balance. Arterioscler Thromb Vasc Biol 29:1419–1423 Matsumoto K, Fukuda N, Abe M et al (2010) Dendritic cells and macrophages in kidney disease. Clin Exp Nephrol 14:1–11 Mensah KA, Ritchlin CT, Schwarz EM (2010) RANKL induces heterogeneous DC-STAMP(lo) and DC-STAMP(hi) osteoclast precursors of which the DC-STAMP(lo) precursors are the master fusogens. J Cell Physiol 223:76–83 Mohler WA, Shemer G, del Campo JJ et al (2002) The type I membrane protein EFF-1 is essential for developmental cell fusion. Dev Cell 2:355–362 Oldenborg PA, Zheleznyak A, Fang YF et al (2000) Role of CD47 as a marker of self on red blood cells. Science 288:2051–2054 Pajcini KV, Pomerantz JH, Alkan O et al (2008) Myoblasts and macrophages share molecular components that contribute to cell–cell fusion. J Cell Biol 180:1005–1019 Podbilewicz B, Leikina E, Sapir A et al (2006) The C. elegans developmental fusogen EFF-1 mediates homotypic fusion in heterologous cells and in vivo. Dev Cell 11:471–481 Sapir A, Avinoam O, Podbilewicz B et al (2008). Viral and developmental cell fusion mechanisms: conservation and divergence. Dev Cell 14:11–21 Sapir A, Choi J, Leikina E et al (2007). AFF-1, a FOS-1-regulated fusogen, mediates fusion of the anchor cell in C. elegans. Dev Cell 12:683–698 Speziani C, Rivollier A, Gallois A et al (2007) Murine dendritic cell transdifferentiation into osteoclasts is differentially regulated by innate and adaptive cytokines. Eur J Immunol 37:747–757 Takayanagi H (2009) [Molecular determinants in osteoclast differentiation and osteoimmunology]. Rinsho Ketsueki 50:447–452 Takenaka K, Prasolava TK, Wang JC et al (2007) Polymorphism in Sirpa modulates engraftment of human hematopoietic stem cells. Nat Immunol 8:1313–1323 Uluckan O, Becker SN, Deng H et al (2009) CD47 regulates bone mass and tumor metastasis to bone. Cancer Res 69:3196–3204 Vignery A (2000) Osteoclasts and giant cells: macrophage-macrophage fusion mechanism. Int J Exp Pathol 81:291–304 Vignery A (2005a) Macrophage fusion: are somatic and cancer cells possible partners? Trends Cell Biol 15:188–193 Vignery A (2005b) Macrophage fusion: the making of osteoclasts and giant cells. J Exp Med 202:337–340 Vignery A (2008a) Macrophage fusion: molecular mechanisms. Methods Mol Biol 475:149–161 Vignery A (2008b) Methods to fuse macrophages in vitro. Methods Mol Biol 475:383–395 Wakkach A, Mansour A, Dacquin R et al (2008) Bone marrow microenvironment controls the in vivo differentiation of murine dendritic cells into osteoclasts. Blood 112:5074–5083 Yagi M, Miyamoto T, Sawatani Y et al (2005) DC-STAMP is essential for cell–cell fusion in osteoclasts and foreign body giant cells. J Exp Med 202:345–351 Yona S, Jung S (2010) Monocytes: subsets, origins, fates and functions. Curr Opin Hematol 17: 53–59
Chapter 11
Molecules Regulating Macrophage Fusions Takeshi Miyamoto and Toshio Suda
Abstract Multinuclear giant cells derived from hematopoietic stem cells or monocyte/macrophage lineage cells are subdivided into osteoclasts, bone resorbing cells, and macrophage giant cells (MGCs) including foreign body giant cells (hereafter described as FBGCs), which are induced at the site of implanted biomaterials, tumors, chronic inflammation and an infection such as tuberculosis. The most characteristic feature of these cells is multinucleation induced by the cell–cell fusion of mononuclear cells, a phenomenon first reported over 60 years ago. To date, combinations of cytokines for osteoclastogenesis or MGC formation have been identified, and osteoclasts and MGCs can be generated in the presence of specific combinations of cytokines in vitro. This makes it possible to isolate specific molecules for cell–cell fusion or to analyze the mechanisms and roles of multinucleation in osteoclasts and MGCs. Recent studies have accumulated data on molecules essential for the cell–cell fusion of osteoclasts and MGCs, and on the role of cell–cell fusion of osteoclasts and MGCs in bone homeostasis and foreign body reactions, respectively. Thus, the role of the cell–cell fusion of osteoclasts in bone homeostasis has been, at least in part, clarified. Furthermore, macrophages reportedly fuse not only with macrophages in a homophilic manner but also with somatic cells and tumors in a heterophilic manner, and the heterophilic fusion is considered involved in tissue repairs and tumor metastasis. Similar to this heterophilic cell–cell fusion, some types of virus such as human immune deficiency virus and influenza virus fuse to somatic cells during an infection. In this chapter, recent advances in the molecular understanding of cell–cell fusion in macrophages will be discussed. Keywords Bone homeostasis · CD9 · CD81 · cell-cell fusion · cytokines · DC-STAMP · foreign body reactions · giant cells · monocyte · macrophage · osteoclast · phagocytosis · RANK · RANKL
T. Miyamoto (B) Department of Orthopedic Surgery, Keio University School of Medicine, Tokyo 160-8582, Japan; Keio Kanrinmaru Project, Keio University School of Medicine, Tokyo 160-8582, Japan e-mail:
[email protected]
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_11, C Springer Science+Business Media B.V. 2011
233
234
T. Miyamoto and T. Suda
Abbreviations ADAM BCG CCR2 CD DAP12 DC DC-STAMP ER FBGC Fc FcR Fah GM-CSF ICAM1 IFN IL ITAM IP3 LFA-1 MAF M-CSF MCP-1 MGC MFF MMP NFATc1 NFκ B OCIF ODF Opg OPGL PLC RANK RANKL Src Syk TNF TRANCE TRAP TREM-2 v-ATPase VRE
A disintegrin and metalloprotease Bacillus Calmette-Guerin Chemokine (C-C motif) receptor 2 Cluster of differentiation DNAX-activating protein 12 Dendritic cell Dendritic cell-specific transmembrane protein Endoplasmic reticulum Foreign body giant cell Fragment crystallizable Fc receptor Fumarylacetoacetate hydrolase Granulocyte-macrophage colony-stimulating factor Intercellular adhesion molecule-1 Interferon Interleukin Immunoreceptor tyrosine-base activation motifs Inositol triphosphate Lymphocyte function-associated antigen 1 Macrophage activating factor Macrophage colony-stimulating factor Monocyte chemotactic protein-1 Macrophage giant cell Macrophage fusion factor Metallo matrix protease Nuclear factor of activated T-cells c1 Nuclear factor κ B Osteoclastogenesis inhibitory factor Osteoclast differentiation factor Osteoprotegerin OPG ligand Phospholipase C Receptor activator of NFκ B RANK ligand Cellular protooncogene homologous to Rous sarcoma virus Spleen tyrosine kinase Tumor necrosis factor Tumor necrosis factor-related activation induced cytokine Tartrate-resistant acid phosphatase Triggering receptor expressed on myeloid cells-2 Vacuolar adenosine triphosphatase Vitamin D response element
11
Molecules Regulating Macrophage Fusions
235
Contents 11.1 Overall . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1.1 Cell–Cell Fusion in Macrophages and Osteoclasts . . . . 11.2 Macrophage and Osteoclast Cell–Cell Fusion . . . . . . . . . . 11.2.1 MGCs . . . . . . . . . . . . . . . . . . . . . . . . 11.2.2 FBGCs as MGCs . . . . . . . . . . . . . . . . . . . 11.2.3 Phagocytosis and ER-Mediated Cell–Cell Fusion . . . . . 11.2.4 Osteoclasts . . . . . . . . . . . . . . . . . . . . . . 11.3 Differentiation of Osteoclasts . . . . . . . . . . . . . . . . . 11.3.1 Differentiation of Osteoclasts and Cell–Cell Fusion Is Induced at the Last Stage of Differentiation . . . . . . 11.3.2 Anchorage-Dependent Osteoclast Cell–Cell Fusion . . . . 11.3.3 Molecular Understanding of Cell–Cell Fusion in Macrophages and Osteoclasts . . . . . . . . . . . . 11.3.4 The Role of Cell–Cell Fusion: Described in Gene Targeted and Transgenic Mice . . . . . . . . . . . . . . . . . 11.3.5 Transcriptional Regulation of Cell–Cell Fusion in Osteoclasts and MGCs . . . . . . . . . . . . . . . 11.4 Future Directions . . . . . . . . . . . . . . . . . . . . . . . 11.4.1 Fusion of Macrophages with Cancer and Somatic Cells . . 11.5 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
235 235 236 236 236 237 237 238
. . . . . . . . 238 . . . . . . . 238 . . . . . . . . 239 . . . . . . . . 241 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
242 243 243 244 244
11.1 Overall 11.1.1 Cell–Cell Fusion in Macrophages and Osteoclasts The existence of giant multinucleated cells, possible macrophage giant cells (MGCs), in association with tumor growth was reportedly described by Muller in 1838 (Muller 1838). MGCs were also considered to have a role in the removal of foreign particulate matter (Langhans 1868) and sequestration of tubercle bacilli (Metchnikoff 1888). The multinuclear giant cells generated by foreign body reactions in response to implanted foreign materials were named foreign body giant cells (FBGCs). The formation of MGCs was considered a result of cell–cell fusion rather than abnormal cell division with a lack of cytokinesis. Macrophage fusion for multinucleation was described in 1962 by Aronson and Elberg who injected mineral oil followed by tritium-labeled thymidine into the peritoneal cavity of rabbits, and collected histiocytes. They found giant cells containing two or more nuclei. Since the giant cells could not be separated into mononuclear cells and only one of the nuclei was radio labeled, they concluded that these polynucleate cells were produced by cell–cell fusion.
236
T. Miyamoto and T. Suda
Osteoclast cell–cell fusion was described in 1963 by Jee and Nolan (Jee and Nolan 1963, Tonna 1963), who demonstrated that the origin of osteoclasts is the fusion of phagocytes. Before this, several cell types such as osteoblasts (Tonna and Cronkite1961) and osteoprogenitor cells (Kember 1960, Young 1962) were reported as the precursors of multinuclear osteoclasts. However, Jee and Nolan (1963) reached the conclusion that phagocytes are the precursor of multinuclear osteoclasts after injecting charcoal into rabbits and observing the formation of charcoal-laden osteoclasts. At 5 days post-injection, macrophages but not osteoclasts took up charcoal. At 15 and 35 days, a huge number of charcoal-laden osteoclasts had formed, but no osteoblasts containing charcoal were observed. Hancox hypothesized in 1946 that the cells in a phagocytic state fuse to form osteoclasts (Hancox 1946), and Jee and Nolan (1963) proved this hypothesis. Thus, cell–cell fusion among macrophages and osteoclasts was observed over 60 years ago. At that time, though the role of cell–cell fusion was not described, MGCs were considered to act against tumors, foreign bodies and bacteria such as tubercle bacilli, and osteoclast cell–cell fusion was considered involved in bone-resorbing activity.
11.2 Macrophage and Osteoclast Cell–Cell Fusion 11.2.1 MGCs The manipulation of macrophage fusion in vitro was tried by Galindo (1972) who treated normal rabbit alveolar macrophages with supernatants of Bacillus CalmetteGuerin (BCG)-sensitized lymph node cells. They described that macrophage fusion factor (MFF) was released from sensitized T cells upon stimulation with a specific antigen (Galindo et al. 1974). Macrophages were considered to have tumoricidal activity activated by a lymphokine named MAF (macrophage activating factor) expressed by T cells (Nathan et al. 1973), and the lymphokine was identical to IL-4 (Crawford et al. 1987). It was not clear whether MFF and MAF were identical to IL-4. Indeed, in 1988, McInnes and Rennick reported the induction of MGCs by IL-4.
11.2.2 FBGCs as MGCs FBGCs are formed during foreign body reactions induced by the implantation of biomaterials (Higgins et al. 2009, Anderson et al. 2008). To date, various cytokines or combinations of cytokines have been reported to induce the formation of FBGCs, such as IL-4 (McInnes and Rennick 1988), IL-13 (DeFife et al. 1997), GM-CSF plus IL-4, IL-3 plus IL-4, M-CSF plus IL-4, M-CSF + IL-13 (Ikeda et al. 1998, McNally and Anderson 1995, Yagi et al. 2005). Thus IL-4 is a key cytokine inducing the multinucleation of macrophages. Interestingly, IL-4, GM-CSF IL-13, and IL-3 are all strong inhibitors of osteoclast formation (Miyamoto et al. 2001, Udagawa et al.
11
Molecules Regulating Macrophage Fusions
237
1997, Ura et al. 2000), suggesting that the development of osteoclasts and MGCs is regulated differently. GM-CSF inhibits osteoclast formation by downregulating c-Fos expression in osteoclast precursor cells, and c-Fos overexpression restored the formation of multinuclear osteoclasts inhibited by GM-CSF (Miyamoto et al. 2001).
11.2.3 Phagocytosis and ER-Mediated Cell–Cell Fusion Macrophage cell–cell fusion had been considered a part of auto-phagocytosis, and phagocytotic activity is required for multinucleation. For phagocytosis, a c-type lectin such as the mannose receptor is needed to recognize foreign materials as well as IL-4-induced macrophage cell–cell fusion (McNally et al. 1996). For macrophage phagocytosis, ER was reportedly fused with plasmalemma for phagosome formation (Gagnon et al. 2002, Muller-Taubenberger et al. 2001). The ER proteins calnexin and calregulin were reportedly expressed on the surface of various cells (Okazaki et al. 2000) and foreign body giant cells (McNally and Anderson 2005). These results suggest the possible involvement of the ER in macrophage cell–cell fusion.
11.2.4 Osteoclasts For multinuclear osteoclast formation in vitro, vitamin D was added to cultures with osteoblastic stromal cells. Vitamin D was shown to induce the expression of ‘osteoclast differentiation factor (ODF)’ in osteoblasts, and since cell-to-cell contact of osteoclast precursor cells with osteoblasts was required for osteoclast formation, ODF was considered a membrane-bound molecule (Takahashi et al. 1988). Tsuda et al. identified the osteoclastogenesis inhibitory factor (OCIF), a soluble receptor belonging to the TNF receptor superfamily identical to osteoprotegerin (opg), which has a strong inhibitory effect on osteoclastogenesis induced by 1,25(OH)2 D3 , parathyroid hormone and IL-11 (Simonet et al. 1997, Tsuda et al. 1997). Then, they described that OCIF bound to osteoblasts treated with 1,25(OH)2 D3 , and that ODF was a membrane-binding molecule induced by 1,25(OH)2 D3 in osteoblasts (Yasuda et al. 1998a). In fact, ODF, also named OPGL (opg ligand) and TRANCE (tumor necrosis factor-related activation induced cytokine) and now called RANKL, was identified as a membrane-binding transmembrane ligand belonging to the TNF superfamily induced by 1,25(OH)2 D3 in osteoblasts (Lacey et al. 1998, Wong et al. 1997, Yasuda et al. 1998b). RANKL was originally identified in T cells as a dendritic cell (DC) activating factor, and the receptor, RANK, was identified in DCs (Anderson et al. 1997). The RANKL promoter has a VRE (vitamin D response element) and is activated by 1,25(OH)2 D3 (Kitazawa et al. 1999). RANK was also detected in osteoclasts, and both RANKL and RANK are essential for osteoclastogenesis (Kong et al. 1999, Dougall et al. 1999).
238
T. Miyamoto and T. Suda
11.3 Differentiation of Osteoclasts 11.3.1 Differentiation of Osteoclasts and Cell–Cell Fusion Is Induced at the Last Stage of Differentiation In vitro, M-CSF and RANKL are sufficient to generate fully differentiated multinuclear bone-resorbing osteoclasts (Teiltelbaum 2000). The role of M-CSF in osteoclastogenesis was clarified using op/op mice, which show severe osteopetrosis due to impaired osteoclast formation. Yoshida et al. (1990) identified an insertional mutation of thymidine at position 262 of the M-CSF coding sequence, which resulted in a frame shift generating a stop codon 21 base pairs downstream. This results in loss of function of M-CSF in op/op mice. The osteopetrotic phenotype and impaired osteoclastogenesis in op/op mice was restored by the injection of M-CSF (Kodama et al. 1991). Osteoclast precursor cells were isolated based on the expression of cell surface molecules (Arai et al. 1999, Miyamoto et al. 2000). Since M-CSF and RANKL are sufficient to induce osteoclastogenesis, osteoclast precursor cells were isolated by the expression of the receptors of M-CSF (c-Fms) and RANKL (RANK). Bone marrow mononuclear cells were subdivided into three populations; c-Fms+ RANK– , c-Fms+ RANK+ and c-Fms– RANK+ . Interestingly, c-Fms+ RANK– cells, which do not express RANK, had the greatest ability to differentiate into multinuclear osteoclasts in the presence of M-CSF and RANKL though RANKL is required for osteoclsat formation (Miyamoto et al. 2000). This is because M-CSF induces the expression of RANK in c-Fms+ RANK– cells and subsequent stimulation with RANKL induces osteoclastogenesis efficiently (Arai et al. 1999). For osteoclast differentiation, calcium signaling is required. The ITAM motif containing the adaptor molecules FcRγ and DAP12 regulates osteoclastogenesis by activating calcium via a Syk-PLCγ2-IP3-mediated calcium influx signaling pathway (Koga et al. 2004). The inhibition of bone resorption by bafilomycin A was reported (Sugawara et al. 1998), suggesting that calcium signaling is also required for osteoclast function.
11.3.2 Anchorage-Dependent Osteoclast Cell–Cell Fusion Osteoclasts show tight adherence to substrates and form an actin ring, a characteristic feature of osteoclasts. Osteoclasts and their precursor cells express high levels of integrins, especially integrin αv β3 and α5 β1, both of which bind to the RGD motif containing extracellular matrix proteins such as fibronectin, osteopontin and vitronectin (Miyamoto et al. 2000). Adherence and Src signaling is critical for osteoclast formation (Nakamura et al. 1998). Multinuclear osteoclast formation is anchorage-dependent, and was strongly inhibited under non-adherent conditions in semi-solid cultures (Miyamoto et al. 2000).
11
Molecules Regulating Macrophage Fusions
239
11.3.3 Molecular Understanding of Cell–Cell Fusion in Macrophages and Osteoclasts For a molecular understanding of macrophage and osteoclast cell–cell fusion, various factors have been reported. Interferon-gamma (IFNγ), intercellular adhesion molecule-1 (ICAM1) and lymphocyte function-associated antigen 1 (LFA-1) were reportedly involved in macrophage cell–cell fusion and blocking antibodies against IFNγ, ICAM-1 and LFA-1 inhibited the formation of MGCs and osteoclasts (Fais et al. 1994, Most et al. 1990, Okada et al. 2002). Vignery and her colleagues demonstrated that macrophage fusion receptor (MFR: also called SHPS-1), belonging to the immunoglobulin (Ig) superfamily of proteins, was highly expressed in macrophages at the stage of cell–cell fusion. Monoclonal antibodies against MFR and the soluble form of the extracellular domain of MFR blocked the cell–cell fusion of macrophages induced by culturing them at maximal density, and thus they concluded that MFR is involved in macrophage cell–cell fusion (Saginario et al. 1995, 1998). Similarly, CD47, a ligand of MFR, was also implicated in macrophage cell–cell fusion (Han et al. 2000). They also reported that the intracellular domain of CD44 (CD44ICD) cleaved in macrophages undergoing fusion is located in the nuclei of fusing macrophages (Cui et al. 2006). Mannose receptor was implicated in macrophage cell–cell fusion (McNally et al. 1996), however, IL-4-induced MGC formation was not altered in mannose receptor knockout mice (Helming and Gordon 2007). Abe et al. (1999) reported the possible involvement of meltrin alpha (also called ADAM12) in the multinucleation of MGCs and osteoclasts. They showed that macrophage cell–cell fusion induced by 5–10% spleen cell-conditioned medium and 10–8 M 1,25(OH)2 D3 was inhibited by the transfection of an anti-sense oligo against meltrin alpha. Connexin43, which plays a role in gap junction communication, was implicated in osteoclast-like FBGC formation in response to the implantation of nanoparticulate hydroxyapatite (Herde et al. 2007). In osteoclasts, Mbalaviele et al. (1995) showed that a neutralizing antibody against E-cadherin or synthetic peptides containing the cell adhesion recognition sequence of cadherins inhibited the cell–cell fusion of osteoclasts. Inhibition of E-cadherin-mediated cell–cell fusion of osteoclasts resulted in inhibition of bone resorption, as well (Mbalaviele et al. 1995). Recent technical advances enable us to approach the molecular mechanisms underlying cell–cell fusion by utilizing gene targeted mice. CD9 is a tetraspanin protein expressed in eggs, and was shown to be essential for cell–cell fusion of gametogenesis as sperm-egg fusion for fertilization (Chen et al. 1999, Kaji et al. 2000, Le Naour et al. 2000, Miyado et al. 2000). CD81 is also a tetraspanin protein, and was reportedly involved in myoblast fusion (Tachibana and Hemler. 1999). However, cell–cell fusion in osteoclasts was still detected in CD9 and CD81 double deficient mice suggesting that these molecules are not required for osteoclast cell–cell fusion (Takeda et al. 2003). Osteoclast cell–cell fusion was detected in CD44-deficient mice also suggesting that these molecules are not essential for the fusion (de Vries et al. 2005. Recently, DAP12, an ITAM-motif-containing adaptor
240
T. Miyamoto and T. Suda
protein, associated receptor TREM-2 and the signaling molecule Syk were reportedly involved in cell–cell fusion of MGCs (Helming et al. 2008). Reduced cell–cell fusion in MGCs was evident in DAP12 knockout or DAP12 loss-of-function mice in vivo and in vitro (Helming et al. 2008). DAP12-null mice show a severe inhibition of osteoclastogenesis (Koga et al. 2004), or a defective cytoskeletal organization and function (Zou et al. 2008). DAP12 and FcRγ double deficient mice show a severe inhibition of osteoclastogenesis compared with DAP12 or FcRγ single deficient mice, suggesting that DAP12 and FcRγ compensate for each other during osteoclastogenesis (Koga et al. 2004). Syk deficiency in osteoclasts also results in inhibition of cytoskeletal organization and function (Zou et al. 2008). Kyriakides and his colleagues demonstrated that MMP9, a matrix metaloprotease involved in matrix degradation, plays a role in cell–cell fusion of FBGCs and the degradation of foreign bodies at the site of foreign reactions (MacLauchlan et al. Table 11.1 Comparison of fusion loss among various gene deficient mice Osteoclasts
Macrophages (FBGCS)
In vivo
In vitro
Referencesa
DC-STAMPnull ATPv0d2-null MMP9-null CD44-null CD200-null CD9/CD81 double null
Complete
Complete
In vivo
In vitro
1
Reduced – No change Reduced Upregulated
Reduced Reduced – – Upregulated
– In vivo In vivo In vivo In vivo
2 3 4 5 6
CCR2-null CCR2-null MCP1-null
– Reduced Differentiation block –
Reduced Reduced –
In vivo In vivo –
In vitro In vitro In vitro In vitro In vitro (macrophage only) – In vitro In vitro
7 8 9
Reduced
In vivo
In vitro
10
Differentiation block Cytoskeletal and functional block (normal differentiation) – Cytoskeletal and functional block (normal differentiation)
–
–
In vitro
11
–
–
In vitro
12
Reduced –
– In vivo
In vitro In vitro
10 13
DAP12-null/ DAP12-KIb DAP12-null DAP12-null
Syk-null Syk-null
a References:
1: Yagi et al. (2005); 2: Lee et al. (2006); 3: MacLauchlan et al. (2009); 4: de Vries et al. (2005); 5: Cui et al. (2007); 6: Takeda et al. (2003); 7: Kyriakides et al. (2004); 8: Binder et al. (2009); 9: Miyamoto et al. (2009); 10: Helming et al. (2008); 11: Koga et al. (2004); 12: Zou et al. (2008); 13: Zou et al. (2007). b DAP12 loss of function mouse.
11
Molecules Regulating Macrophage Fusions
241
2009). MCP-1/CCL2 is also reportedly involved in the cell–cell fusion of MGCs and osteoclasts (Kim et al. 2006, Kyriakides et al. 2004). MCP-1/CCL2 was shown to be involved in osteoclast differentiation rather than cell–cell fusion itself (Binder et al. 2009, Miyamoto et al. 2009). More recently, DC-STAMP was identified as an essential molecule for cell–cell fusion of both osteoclasts and MGCs, and that DC-STAMP-deficient mice show a complete lack of cell–cell fusion in osteoclasts and MGCs in vivo and in vitro (Yagi et al. 2005). DC-STAMP was originally identified in DCs and in IL-4-treated macrophages as FIND (Hartgers et al. 2000, Staege et al. 2001). Inhibition of DCSTAMP by siRNA in the RAW264.7 osteoclast precursor cell line resulted inhibition of multinuclear osteoclast formation (Kukita et al. 2004). The osteoclasts derived from DC-STAMP-deficient mice express Cathepsin K and TRAP, both of which are osteoclast markers, indicating that cell–cell fusion was specifically inhibited in DC-STAMP-deficient osteoclasts, and thus the essential molecule for osteoclast cell–cell fusion was identified (Yagi et al. 2005). Indeed, a ruffled boader and an actin ring, both characteristic features of osteoclasts, formed in DC-STAMPdeficient osteoclasts, and DC-STAMP is specifically required for cell–cell fusion rather than differentiation (Yagi et al. 2005). Similarly, mice deficient in v-ATPase V0 subunit d2 (ATPv0d2), an isoform of vacuolar ATPase, show severe, though not complete, inhibition of cell–cell fusion in osteoclasts and MGCs (Lee et al. 2006). The molecules involved in osteoclast and MGC fusion, all of which were demonstrated using gene targeted mice, are listed in Table 11.1.
11.3.4 The Role of Cell–Cell Fusion: Described in Gene Targeted and Transgenic Mice The role of cell–cell fusion in osteoclasts and MGCs remains largely unknown. Cell–cell fusion increases the efficiency of bone resorption in osteoclasts (Lee et al. 2006, Yagi et al. 2005). Increased bone volume was shown in DC-STAMP-deficient and ATPv0d2-deficioent mice, suggesting that multinucleation of osteoclasts is involved in the regulation of physiological bone mass Lee et al. 2006, Yagi et al. 2005). CD200-deficient mice showed increased bone mass due to impaired osteoclast cell–cell fusion (Cui et al. 2007). Interestingly, the activity of osteoblasts, which do not express DC-STAMP and ATPv0d2, was upregulated in DC-STAMPdeficient and ATPv0d2-deficient mice (Iwasaki et al. 2008, Lee et al. 2006). MMP9-deficient mice showed decreased FBGC cell–cell fusion and foreign body degradation (MacLauchlan et al. 2009), suggesting that MMP9 is involved in both the cell–cell fusion and function of FBGCs. On the contrary, osteoblastic activity was downregulated in DC-STAMP-overexpressing transgenic mice, suggesting that multinucleation of osteoclasts regulates osteoblasts (Iwasaki et al. 2008, Lee et al. 2006). The molecules that play a role in cell–cell fusion of osteoclasts and MGCs are known (Fig. 11.1 and Table 11.1). Further studies are required to clarify the role of, or the mechanisms of, cell–cell fusion in osteoclasts and macrophages.
242
T. Miyamoto and T. Suda
mononuclear osteoclasts
osteoclasts
M-CSF + RANKL CD200
common progenitors
DC-STAMP ATPv0d2
fusiogenic molecules (identified by gene targeting)
MMP9 CCR2 DAP12 Syk
IL-4, GM-CSF, IL-3, IL-13
mononuclear macrophages
macrophages (FBGCs)
Fig. 11.1 Regulation of multinuclear osteoclast and macrophage giant cell formation. Multinuclear osteoclasts and macrophage giant cells are derived from common mononuclear progenitor cells in the presence of M-CSF plus RANKL or several combinations of cytokines including IL-4, GM-CSF, IL-3 and IL-13, respectively. The multinucleation of osteoclasts and macrophage giant cells is induced by the cell–cell fusion of mononuclear cells, and DC-STAMP, ATPvod2 and CD200 or DC-STAMP, ATPvod2, MMP9, CCR2, DAP12 and Syk are involved in the cell–cell fusion of osteoclasts and macrophage giant cells, respectively
11.3.5 Transcriptional Regulation of Cell–Cell Fusion in Osteoclasts and MGCs Transcriptional regulation of osteoclasts was intensively clarified. c-Fos, a component of AP1, and NFATc1 were identified as essential transcription factors for osteoclastogenesis (Asagiri et al. 2005, Grigoriadis et al. 1994, Takayanagi et al. 2002). c-Fos regulates NFATc1 expression in osteoclasts and c-Fos and NFATc1 co-operatively regulate osteoclast differentiation. c-Fos and NFATc1 bind to the DC-STAMP promoter, and both are essential for DC-STAMP expression in osteoclasts (Yagi et al. 2007). c-Fos-deficient cells failed to express DC-STAMP in the presence of M-CSF and RANKL, and FK506, a NFAT inhibitor, effectively inhibited M-CSF and RANKL-induced DC-STAMP expression (Yagi et al. 2007). Similarly, both c-Fos and NFATc1 regulated ATPv0d2 expression in osteoclasts (Kim et al. 2008). This suggests that osteoclast cell–cell fusion is a part of normal osteoclast differentiation (Fig. 11.2). Interestingly, however, the regulation of DC-STAMP expression is completely different between osteoclasts and MGCs, and DC-STAMP was detected in c-Fosdeficient cells or in the presence of FK506 in MGC-inducing conditions (Yagi et al. 2007). The precise role of PU.1 and NFκB is unclear, but both these transcription factors bound to the DC-STAMP promoter in MGCs (Fig. 11.2) (Yagi et al. 2007).
11
Molecules Regulating Macrophage Fusions
osteoclasts
AP1
NFATc1
243
Macrophage giant cells
DC-STAMP
PU.1
NFκB
DC-STAMP
Fig. 11.2 Distinct transcriptional regulation of DC-STAMP in osteoclasts and macrophage giant cells.AP1 and NFATc1 are both required for DC-STAMP expression in osteoclasts, however, these transcription factors are not required for DC-STAMP expression in macrophage giant cells. PU.1 and NFkB play a role in DC-STAMP expression in macrophage giant cells
11.4 Future Directions 11.4.1 Fusion of Macrophages with Cancer and Somatic Cells Cell–cell fusion of monocyte/macrophages with cancer cells was implicated in metastasis (Chakraborty et al. 2001, 2003, Pawelek et al. 2008a, b, Vignery 2005). Genomic hybridization was observed on macrophage-tumor cell–cell fusion (Pawelek et al. 2008a, b). Transplanted bone marrow-derived cells reportedly fuse with hepatocytes and intestinal stem cells (Rizvi et al. 2006, Vassilopoulos et al. 2003, Wang et al. 2003), and the fusion plays a role in the cure of a genetic liver disease, fumarylacetoacetate hydrolase (Fah) deficiency (Vassilopoulos et al. 2003, Wang et al. 2003). The bone marrow-derived cells that fuse with hepatocytes were monocytes (Willenbring et al. 2004). These observations support the idea that heterophilic cell–cell fusion of monocyte/macrophages with somatic cells may be of benefit for the treatment of congenital diseases. Heterophilic cell–cell fusion is also observed during the homophilic fusion of osteoclasts and MGCs, and DC-STAMP-deficient cells could fuse with wild-type cells (Yagi et al. 2005). The ‘fusion-founder’ cells fused with ‘fusion-competent’ cells. Thus, heterophilic cell–cell fusion might be the major process of cell–cell fusion by macrophages. Recently, it was reported that microvesicles, also called exosomes and microparticles, released after the fusion of multivesicular bodies and plasma membranes, played a role in intercellular communication (Pap et al. 2009, Théry et al. 2002). The membrane of microvesicles consists of proteins and lipids, and transfers information by fusing/interacting with distant or neighboring cells.Thus, it is possible that cell–cell fusion is a way of initiating signaling cascades or of providing new phenotypes to the recipient cells. Further studies are necessary to clarify the role of cell–cell fusion.
244
T. Miyamoto and T. Suda
11.5 Concluding Remarks MGCs (FBGCs) and osteoclast multinucleation is implicated in the degradation of foreign materials and resorption of bone, respectively. Thus ‘bigger is better’ for the function of MGCs and osteoclasts. Moreover, inhibition of cell–cell fusion in osteoclasts results in increased osteoblastic activity suggesting that cell–cell fusion of osteoclasts is a good target for increasing bone mass by reducing osteoclastic bone resorption and increasing osteoblastic activity. The macrophage cell–cell fusion may be a good target for inhibiting tumor metastasis. The cell–cell fusion of macrophages is beneficial for the treatment of congenital diseases by delivering various genes. Although, the molecules essential for macrophage and osteoclast cell–cell fusion have been identified, further studies are needed to clarify the mechanisms of cell–cell fusion. The 1st Gordon Research Conference ‘CELL–CELL FUSION’ was held in July, 2007, and cell–cell fusion is now a hot topic. Cell–cell fusion in drosophila myotubes, viral-cell fusion, cell–cell fusion in yeast and gametogenesis are understood better than macrophage cell–cell fusion. These studies may aid our understanding of macrophage cell–cell fusion, but defects in sperm-egg for fertilization or in myotubes are not evident in DC-STAMPdeficient and transgenic mice, suggesting that the molecular system for cell–cell fusion is tissue specific. Future studies will fully clarify the molecular mechanisms and roles of cell–cell fusion in macrophages and osteoclasts.
References Abe E, Mocharla H, Yamate T et al (1999) Meltrin-alpha, a fusion protein involved in multinucleated giant cell and osteoclast formation. Calcif Tissue Int 64:508–515 Anderson DM, Maraskovsky E, Billingsley WL et al (1997) A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature 390:175–179 Anderson JM, Rodriguez A, Cang DT (2008) Foreign body reaction to biomaterials. Semin Immunol 20:86–100 Arai F, Miyamoto T, Ohneda O et al (1999) Commitment and differentiation of osteoclast precursor cells by the sequential expression of c-Fms and receptor activator of nuclear factor kappaB (RANK) receptors. J Exp Med 190:1741–1754 Aronson M, Elberg SS (1962) Fusion of peritoneal histocytes with formation of giant cells. Nature 193:399–400 Asagiri M, Sato K, Usami T et al (2005) Autoamplification of NFATc1 expression determines its essential role in bone homeostasis. J Exp Med 202:1261–1269 Binder NB, Niederreiter B, Hoffmann O et al (2009) Estrogen-dependent and C-C chemokine receptor-2-dependent pathways determine osteoclast behavior in osteoporosis. Nat Med 15:417–424 Chakraborty AK, de Freitas Sousa J, Espreafico EM et al (2001) Human monocyte x mouse melanoma fusion hybrids express human gene. Gene 275:103–106 Chakraborty AK, Kolesnikova N, de Freitas Sousa J et al (2003) Expression of c-Met protooncogene in metastatic macrophage x melanoma fusion hybrids: implication of its possible role in MSH-induced motility. Oncol Res 14:163–174 Chen MS, Tung KS, Coonrod SA et al (1999) Role of the integrin-associated protein CD9 in binding between sperm ADAM 2 and the egg integrin alpha6beta1: implications for murine fertilization. Proc Natl Acad Sci USA 96:11830–11835
11
Molecules Regulating Macrophage Fusions
245
Crawford RM, Finbloom DS et al (1987) B cell stimulatory factor-1 (interleukin 4) activates macrophages for increased tumoricidal activity and expression of Ia antigens. J Immunol 139:135–141 Cui W, Cuartas E, Ke J et al (2007) CD200 and its receptor, CD200R, modulate bone mass via the differentiation of osteoclasts. Proc Natl Acad Sci USA 104:14436–14441 Cui W, Ke JZ, Zhang Q et al (2006) The intracellular domain of CD44 promotes the fusion of macrophages. Blood 107:796–805 de Vries TJ, Schoenmaker T, Beertsen W et al (2005) Effect of CD44 deficiency on in vitro and in vivo osteoclast formation. J Cell Biochem 94:954–966 DeFife KM, Jenney CR, McNally AK et al (1997) Interleukin-13 induces human monocyte/macrophage fusion and macrophage mannose receptor expression. J Immunol 158: 3385–3390 Dougall WC, Glaccum M, Charrier K et al (1999) RANK is essential for osteoclast and lymph node development. Genes Dev 13:2412–2424 Fais S, Burgio VL, Silvestri M et al (1994) Multinucleated giant cells generation induced by interferon-gamma. Changes in the expression and distribution of the intercellular adhesion molecule-1 during macrophages fusion and multinucleated giant cell formation. Lab Invest 71:737–744 Gagnon E, Duclos S, Rondeau C et al (2002) Endoplasmic reticulum-mediated phagocytosis is a mechanism of entry into macrophages. Cell 110:119–131 Galindo BJ (1972) Antigen-mediated fusion of specifically sensitized rabbit alveolar macrophages. Infect Immun 5:583–584 Galindo BJ, Lazdins R, Castillo O (1974) Fusion of normal rabbit alveolar macrophages induced by supernatant fluids from BCG-sensitized lymph node cells after elicitation by antigen. Infect Immun 9:212–216 Grigoriadis AE, Wang ZQ, Cecchini MG et al (1994) c-Fos: a key regulator of osteoclastmacrophage lineage determination and bone remodeling. Science 266:443–448 Han X, Sterling H, Chen Y et al (2000) CD47, a ligand for the macrophage fusion receptor, participates in macrophage multinucleation. J Biol Chem 275:37984–37992 Hancox NM (1946) On the occurrence in vitro of cells resembling osteoclasts. J Physiol 105:66–71 Hartgers FC, Vissers JL, Looman MW et al (2000) DC-STAMP, a novel multimembrane-spanning molecule preferentially expressed by dendritic cells. Eur J Immunol 30:3585–3590 Helming L, Gordon S (2007) The molecular basis of macrophage fusion. Immunobiology 212:785–793 Helming L, Tomasello E, Kyriakides TR et al (2008) Essential role of DAP12 signaling in macrophage programming into a fusion-competent state. Sci Signal 1(43) ra11, 2008 Oct 28 Herde K, Hartmann S, Brehm R et al (2007) Connexin 43 expression of foreign body giant cells after implantation of nanoparticulate hydroxyapatite. Biomaterials 28:4912–4921 Higgins DM, Basaraba RJ, Hohnbaum AC, Lee EJ, Grainger DW, Gonzalez-Juarrero M (2009) Localized immunosuppressive environment in the foreign body response to implanted biomaterials. Am J Pathol 175:161–170 Ikeda T, Ikeda K, Sasaki K et al (1998) IL-13 as well as IL-4 induces monocytes/macrophages and a monoblastic cell line (UG3) to differentiate into multinucleated giant cells in the presence of M-CSF. Biochem Biophys Res Commun 253:265–272 Iwasaki R, Ninomiya K, Miyamoto K et al (2008) Cell fusion in osteoclasts plays a critical role in controlling bone mass and osteoblastic activity. Biochem Biophys Res Commun 377:899–904 Jee WS, Nolan PD (1963) Origin of osteoclasts from the fusion of phagocytes. Nature 200: 225–226 Kaji K, Oda S, Shikano T et al (2000) The gamete fusion process is defective in eggs of Cd9deficient mice. Nat Genet 24:279–282 Kember NF (1960) Cell division in endochondral ossification: a study of cell proliferation in rat bones by the method of tritiated thymidine autoradiography. J Bone Joint Surg Br 42:824
246
T. Miyamoto and T. Suda
Kim K, Lee SH, Ha Kim J et al (2008) NFATc1 induces osteoclast fusion via up-regulation of Atp6v0d2 and the dendritic cell-specific transmembrane protein (DC-STAMP). Mol Endocrinol 22:176–185 Kim MS, Day CJ, Selinger CI et al (2006) MCP-1-induced human osteoclast-like cells are tartrateresistant acid phosphatase, NFATc1, and calcitonin receptor-positive but require receptor activator of NFkappaB ligand for bone resorption. J Biol Chem 281:1274–1285 Kitazawa R, Kitazawa S, Maeda S (1999) Promoter structure of mouse RANKL/ TRANCE/OPGL/ODF gene. Biochim Biophys Acta 1445:134–141 Kodama H, Yamasaki A, Nose M et al (1991) Congenital osteoclast deficiency in osteopetrotic (op/op) mice is cured by injections of macrophage colony-stimulating factor. J Exp Med 173:269–272 Koga T, Inui M, Inoue K et al (2004) Costimulatory signals mediated by the ITAM motif cooperate with RANKL for bone homeostasis. Nature 428:758–763 Kong YY, Yoshida H, Sarosi I et al (1999) OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397:315–323 Kukita T, Wada N, Kukita A et al (2004) RANKL-induced DC-STAMP is essential for osteoclastogenesis. J Exp Med 200:941–946 Kyriakides TR, Foster MJ, Keeney GE et al (2004) The CC chemokine ligand, CCL2/MCP1, participates in macrophage fusion and foreign body giant cell formation. Am J Pathol 165:2157–2166 Lacey DL, Timms E, Tan HL et al (1998) Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93:165–176 Langhans T (1868) Ueber riesenzellen mit wandstandigen kernen in tuberkeln und die fibrose form des tuberkels. Arch Pathol Anat 42:382 Le Naour F, Rubinstein E, Jasmin C et al (2000) Severely reduced female fertility in CD9-deficient mice. Science 287:319–321 Lee SH, Rho J, Jeong D et al (2006) v-ATPase V0 subunit d2-deficient mice exhibit impaired osteoclast fusion and increased bone formation. Nat Med 12:1403–1409 MacLauchlan S, Skokos EA, Meznarich N et al (2009) Macrophage fusion, giant cell formation, and the foreign body response require matrix metalloproteinase 9. J Leukoc Biol 85:617–626 Mbalaviele G, Chen H, Boyce BF et al (1995) The role of cadherin in the generation of multinucleated osteoclasts from mononuclear precursors in murine marrow. J Clin Invest 95:2757–2765 McInnes A, Rennick DM (1988) Interleukin 4 induces cultured monocytes/macrophages to form giant multinucleated cells. J Exp Med 167:598–611 McNally AK, Anderson JM (1995) Interleukin-4 induces foreign body giant cells from human monocytes/macrophages. Differential lymphokine regulation of macrophage fusion leads to morphological variants of multinucleated giant cells. Am J Pathol 147:1487–1499 McNally AK, Anderson JM (2005) Multinucleated giant cell formation exhibits features of phagocytosis with participation of the endoplasmic reticulum. Exp Mol Pathol 79:126–135 McNally AK, DeFife KM, Anderson JM (1996) Interleukin-4-induced macrophage fusion is prevented by inhibitors of mannose receptor activity. Am J Pathol 149:975–985 Metchnikoff E (1888) Ueber die phagocytare rolle der tuberkelriesenzellen. Arch Pathol Anat 113:63 Miyado K, Yamada G, Yamada S et al (2000) Requirement of CD9 on the egg plasma membrane for fertilization. Science 287:321–324 Miyamoto K, Ninomiya K, Sonoda KH et al (2009) MCP-1 expressed by osteoclasts stimulates osteoclastogenesis in an autocrine/paracrine manner. Biochem Biophys Res Commun 383:373–377 Miyamoto T, Arai F, Ohneda O et al (2000) An adherent condition is required for formation of multinuclear osteoclasts in the presence of macrophage colony-stimulating factor and receptor activator of nuclear factor kappa B ligand. Blood 96:4335–4343
11
Molecules Regulating Macrophage Fusions
247
Miyamoto T, Ohneda O, Arai F et al (2001) Bifurcation of osteoclasts and dendritic cells from common progenitors. Blood 98:2544–2554 Most J, Neumayer HP, Dierich MP (1990) Cytokine-induced generation of multinucleated giant cells in vitro requires interferon-gamma and expression of LFA-1. Eur J Immunol 20:1661–1667 Muller J (1838) Ueber den feineren bau die formen der krankhaften geschwulste. Berlin Muller-Taubenberger A, Lupas AN et al (2001) Calreticulin and calnexin in the endoplasmic reticulum are important for phagocytosis. EMBO J 20:6772–6782 Nakamura I, Jimi E, Duong LT et al (1998) Tyrosine phosphorylation of p130Cas is involved in actin organization in osteoclasts. J Biol Chem 273:11144–11149 Nathan CF, Remold HG, David JR (1973) Characterization of a lymphocyte factor which alters macrophage functions. J Exp Med 137:275–290 Okada Y, Morimoto I, Ura K et al (2002) Cell-to-cell adhesion via intercellular adhesion molecule1 and leukocyte function-associated antigen-1 pathway is involved in 1alpha,25(OH)2D3, PTH and IL-1alpha-induced osteoclast differentiation and bone resorption. Endocr J 49:483–495 Okazaki Y, Ohno H, Takase K et al (2000) Cell surface expression of calnexin, a molecular chaperone in the endoplasmic reticulum. J Biol Chem 275:35751–35758 Pap E, Pállinger E, Pásztói M et al (2009) Highlights of a new type of intercellular communication: microvesicle-based information transfer. Inflamm Res 58:1–8 Pawelek JM, Chakraborty AK (2008a) The cancer cell–leukocyte fusion theory of metastasis. Adv Cancer Res 101:397–444 Pawelek JM, Chakraborty AK (2008b) Fusion of tumour cells with bone marrow-derived cells: a unifying explanation for metastasis. Nat Rev Cancer 8:377–386 Rizvi AZ, Swain JR, Davies PS et al (2006) Bone marrow-derived cells fuse with normal and transformed intestinal stem cells. Proc Natl Acad Sci USA 103:6321–6325 Saginario C, Qian HY, Vignery A (1995) Identification of an inducible surface molecule specific to fusing macrophages. Proc Natl Acad Sci USA 92:12210–12214 Saginario C, Sterling H, Beckers C et al (1998) MFR, a putative receptor mediating the fusion of macrophages. Mol Cell Biol 18:6213–6223 Simonet WS, Lacey DL, Dunstan CR et al (1997) Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89:309–319 Staege H, Brauchlin A, Schoedon G et al (2001) Two novel genes FIND and LIND differentially expressed in deactivated and Listeria-infected human macrophages. Immunogenetics 53:105–113 Sugawara K, Hamada M, Hosoi S et al (1998) A useful method to evaluate bone resorption inhibitors, using osteoclast-like multinucleated cells. Anal Biochem 255:204–210 Tachibana I, Hemler ME (1999) Role of transmembrane 4 superfamily (TM4SF) proteins CD9 and CD81 in muscle cell fusion and myotube maintenance. J Cell Biol 146:893–904 Takahashi N, Akatsu T, Udagawa N et al (1988) Osteoblastic cells are involved in osteoclast formation. Endocrinology 123:2600–2602 Takayanagi H, Kim S, Koga T et al (2002) Induction and activation of the transcription factor NFATc1 (NFAT2) integrate RANKL signaling in terminal differentiation of osteoclasts. Dev Cell 3:889–901 Takeda Y, Tachibana I, Miyado K et al (2003) Tetraspanins CD9 and CD81 function to prevent the fusion of mononuclear phagocytes. J Cell Biol 161:945–956 Teiltelbaum SL (2000) Bone resorption by osteoclasts. Science 289:1504–1508 Théry C, Zitvogel L, Amigorena S (2002) Exosomes: composition, biogenesis and function. Nat Rev Immunol 2:569–579 Tonna EA (1963) Origin of osteoclasts from the fusion of phagocytes. Nature 200:226–227 Tonna EA, Cronkite EP (1961) Use of tritiated thymidine for the study of the origin of the osteoclast. Nature 190:459–460
248
T. Miyamoto and T. Suda
Tsuda E, Goto M, Mochizuki S et al (1997) Isolation of a novel cytokine from human fibroblasts that specifically inhibits osteoclastogenesis. Biochem Biophys Res Commun 234:137–142 Udagawa N, Horwood NJ, Elliott J et al (1997) Interleukin-18 (interferon-gamma-inducing factor) is produced by osteoblasts and acts via granulocyte/macrophage colony-stimulating factor and not via interferon-gamma to inhibit osteoclast formation. J Exp Med 185:1005–1012 Ura K, Morimoto I, Watanabe K et al (2000) Interleukin (IL)-4 and IL-13 inhibit the differentiation of murine osteoblastic MC3T3-E1 cells. Endocr J 47:293–302 Vassilopoulos G, Wang PR, Russell DW (2003) Transplanted bone marrow regenerates liver by cell fusion. Nature 422:901–904 Vignery A (2005) Macrophage fusion: are somatic and cancer cells possible partners? Trends Cell Biol 15:188–193 Wang X, Willenbring H, Akkari Y et al (2003) Cell fusion is the principal source of bone-marrowderived hepatocytes. Nature 422:897–901 Willenbring H, Bailey AS, Foster M et al (2004) Myelomonocytic cells are sufficient for therapeutic cell fusion in liver. Nat Med 10:744–748 Wong BR, Rho J, Arron J et al (1997) TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells. J Biol Chem 272:25190–25194 Yagi M, Miyamoto T, Sawatani Y et al (2005) DC-STAMP is essential for cell–cell fusion in osteoclasts and foreign body giant cells. J Exp Med 202:345–351 Yagi M, Ninomiya K, Fujita N et al (2007) Induction of DC-STAMP by alternative activation and downstream signaling mechanisms. J Bone Miner Res 22:992–1001 Yasuda H, Shima N Nakagawa N et al (1998a) Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): a mechanism by which OPG/OCIF inhibits osteoclastogenesis in vitro. Endocrinology 139:1329–1337 Yasuda H, Shima N Nakagawa N et al (1998b) Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95:3597–3602 Yoshida H, Hayashi S, Kunisada T et al (1990) The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345:442–444 Young RW (1962) Cell proliferation and specialization during endochondral osteogenesis in young rats. J Cell Biol 14:357–370 Zou W, Kitaura H, Reeve J et al (2007) Syk, c-Src, the alphavbeta3 integrin, and ITAM immunoreceptors, in concert, regulate osteoclastic bone resorption. J Cell Biol 176:877–888 Zou W, Reeve JL, Liu Y et al (2008) DAP12 couples c-Fms activation to the osteoclast cytoskeleton by recruitment of Syk. Mol Cell 31:422–431
Chapter 12
Current Progress Towards Understanding Mechanisms of Myoblast Fusion in Mammals Grace K. Pavlath
Abstract Myoblast fusion is critical for the formation, growth and maintenance of skeletal muscle throughout life. Myoblasts must undergo a complex series of molecular and morphological changes prior to fusing with one another. A number of extracellular, cell surface and intracellular molecules have been discovered that act to finely coordinate the cellular and molecular events that are intimately associated with, and influence the ability of muscle cells to fuse. This review discusses the molecules that regulate myoblast elongation, migration, adhesion and cytoskeletal alterations. Understanding the molecular mechanisms by which myoblast fusion is regulated is both of general biologic interest and may also lead to new strategies for manipulating myoblasts in disease, repair and aging and enhancing muscle growth. In addition, promotion of cell fusion may aid in cell therapy approaches involving transplantation of muscle stem cells. Keywords Myoblast fusion · myogenesis · myotube · skeletal muscle · myofiber · muscle growth · muscle regeneration · satellite cells Abbreviations Arf6 Arp2/3 DAPI Duf GEF IGF-1 IL IrreC Kirre MR mTOR
ADP ribosylation factor 6 Actin-related protein 2/3 4 ,6-diamidino-2-phenylindole Dumbfounded Guanine nucleotide exchange factor Insulin growth factor 1 Interleukin Irregular optic Chiasma Kin of irre Mannose receptor Mammalian target of rapamycin
G.K. Pavlath (B) Department of Pharmacology, Emory University, Atlanta, GA 30322, USA e-mail:
[email protected] L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_12, C Springer Science+Business Media B.V. 2011
249
250
mTOR NFATc2 PCR SHP-2 Sns Src WASP WAVE WGA
G.K. Pavlath
Mammalian target of rapamycin Nuclear factor of activated T cells c2 Polymerase chain reaction SH2 domain containing phosphatase-2 Sticks and stones Cellular protooncogene homologous to Rous sarcoma virus Wiskott-Aldrich syndrome protein WASP family Verprolin-homologous protein Wheat germ agglutinin
Contents 12.1 Introduction . . . . . . . . . . . . . . . . . . . . . 12.2 Biochemical Requirements for Myoblast Fusion . . . . . 12.3 Methodology for Studying Myoblast Fusion . . . . . . . 12.3.1 In Vitro Models . . . . . . . . . . . . . . . . 12.3.2 In Vivo Studies . . . . . . . . . . . . . . . . 12.4 Current Areas of Research in Myoblast Fusion . . . . . 12.4.1 Elongation and Membrane Alterations . . . . . . 12.4.2 Migration . . . . . . . . . . . . . . . . . . 12.4.3 Muscle Cell Recognition/Adhesion . . . . . . . 12.4.4 Actin Dynamics and Integrin Function . . . . . 12.4.5 Regulation of Cell Fusion with Nascent Myotubes 12.5 Future Prospects . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
250 250 251 251 253 253 254 255 256 257 258 261 262
12.1 Introduction Skeletal muscle accounts for 40–50% of lean body mass (Prior et al. 2001) and is critical for locomotion, breathing and balance. Skeletal muscle is composed of myofibers containing hundreds of nuclei in a continuous cytoplasm. These multinucleated myofibers form during embryogenesis through the fusion of myogenic precursor cells called myoblasts. Myoblast fusion is important not only for skeletal muscle formation during development, but also for post-natal muscle regeneration and growth of skeletal muscle. In addition, cell therapy approaches for various muscular disorders rely on the ability of transplanted muscle stem cells to fuse with endogenous myofibers. Thus, understanding the molecular mechanisms that regulate myoblast fusion is fundamental to the basic biology of skeletal muscle as well as to clinical applications.
12.2 Biochemical Requirements for Myoblast Fusion When cultured in high serum medium containing growth factors, myoblasts proliferate. However, upon serum and mitogen withdrawal, myoblasts differentiate and undergo an ordered process of myogenin expression, cell cycle arrest, and
12
Mechanisms of Myoblast Fusion in Mammals
251
Fig. 12.1 Time course of myoblast fusion in vitro. Primary mouse myoblasts were differentiated by placing the cells in a serum-free, low mitogen medium. Cells were fixed at the indicated times and immunostained for embryonic myosin heavy chain (brown), a marker of biochemical differentiation. Cells are elongated and mostly differentiated by 16 h. Small nascent myotubes are observed at 24 h. Subsequently, further myoblast fusion with myotubes occurs, and large mature myotubes are present at 48 h
phenotypic differentiation (Andres and Walsh 1996) to become fusion competent. Initially, cell fusion occurs between these differentiated myoblasts to form nascent myotubes with a limited number of nuclei. Later, additional fusion-competent myoblasts fuse with nascent myotubes to give rise to large myotubes with many nuclei (Fig. 12.1). Similar processes occur in vivo. Alterations in these initial differentiation events will effect the downstream formation and growth of myotubes. Thus, investigators should be careful to ascertain that markers of early biochemical differentiation such as myogenin and p21, and late biochemical differentiation such as creatine kinase or myosin heavy chain, are expressed normally before concluding that a specific molecule affects cell fusion. Some of the molecules commonly associated with myoblast fusion in the literature have either not been analyzed for, or are associated with, decreases in biochemical differentiation and are not discussed in this review.
12.3 Methodology for Studying Myoblast Fusion 12.3.1 In Vitro Models Much of the work pertaining to mammalian myoblast fusion derives from in vitro studies. The advantages of in vitro studies are the ability to (1) carefully control the environment of the cells; (2) easily manipulate cellular components with
252
G.K. Pavlath
drugs, siRNA or DNA constructs; and (3) study myoblast fusion using time-lapse microscopy. Primary muscle cells or established cell lines are both utilized in vitro to analyze many aspects of cell fusion. Mouse myoblasts are the most commonly used primary muscle cells, with human and rat muscle cells less frequently used. The mouse muscle cell line C2C12 (Blau et al. 1983) is the most frequently used cell line. A number of different assays are employed to quantitate myoblast fusion in vitro. Many of these assays are performed on fixed cultures utilizing quantitative morphological measurements on digital images obtained through microscopy. One can analyze fusion index, which is a measure of the number of nuclei in the culture that have fused into multinucleated myotubes. More detailed analyses of the myotube population can be performed by quantifying the number of nuclei within individual myotubes and either calculating average myonuclear number or the percentage of myotubes containing a certain range of nuclear number (e.g. <2 nuclei or >5 nuclei). Multiple types of assays are advised as mutations in some molecules affect both fusion index and myonuclear number (Griffin et al. 2009), whereas others can affect myonuclear number but not overall fusion index (Horsley et al. 2001, 2003, Jansen and Pavlath 2006). In addition, two populations of muscle cells can be labeled with either red or green fluorescent dyes, cocultured to allow fusion to occur, and the resulting hybrid yellow myotubes specifically analyzed. Such cell mixing experiments have provided novel insights into myoblast fusion (Horsley et al. 2003, Jansen and Pavlath 2006, Griffin et al. 2009, Sohn et al. 2009). Disadvantages to these methods include (1) the ability to only analyze a limited number of independent fields; (2) the tedium involved in performing these assays manually; and, (3) potential observer bias, which can be mitigated by blinding the person doing the analysis to the identity of the samples. Nonetheless, these methods have been widely used successfully to unravel the regulatory processes culminating in a multinucleated myotube. To counteract some of the problems with the above assays, an elegant biochemical assay was developed involving fusion-induced biochemical complementation of lacZ mutants (Mohler and Blau 1996). In this assay, two myoblast populations are each transfected with one of two plasmid constructs containing a different lacZ fragment. In myotubes formed by the fusion of these two types of myoblasts, strong β-galactosidase enzyme activity is produced that can be quantitated using a luminometer (Shinn-Thomas et al. 2008). Although one can quantitate the timecourse of fusion and the fusion rates between wild type and mutant myoblasts (Charlton et al. 1997, 2000), the number of nuclei within individual myotubes in a population cannot be measured with this assay. Time-lapse microscopy has revealed certain myoblast behaviors during fusion that cannot be recognized in the end point assays described above. These time-lapse studies can provide information regarding the velocity of muscle cell migration as well as the direction of migration (Chazaud et al. 1998, 2000, Jansen and Pavlath 2006, Bondesen et al. 2007, O’Connor and Pavlath 2007, Griffin et al. 2009, Nowak et al. 2009). In addition, dynamic changes in cell structural proteins (Nowak et al.
12
Mechanisms of Myoblast Fusion in Mammals
253
2009) or membrane structures (Mukai and Hashimoto 2008, Mukai et al. 2009, Nowak et al. 2009) can be visualized providing important clues to the process of myoblast fusion.
12.3.2 In Vivo Studies Some molecules that regulate mammalian myoblast fusion in vitro are not required for cell fusion in vivo. Such discrepancy could be due to (1) the methods used to manipulate the molecule in vitro; (2) redundancy among molecules in vivo; (3) culture conditions employed; or (4) potential effects of other cell types in vivo. Currently, many studies carefully couple in vitro experiments using primary muscle cells with in vivo studies of knockout mice, especially muscle regeneration models. Unfortunately, current technologies do not allow for visualizing myoblast fusion in real time in mammals. However, morphologic and biochemical measurements of muscle growth in response to localized injury in mice have been successfully used as a readout of myoblast fusion. The formation of regenerated myofibers is easily discerned histologically by the presence of centrally located nuclei within the myofiber and can be quantitated both by measurement of myofiber number/field and myofiber size. In addition, the number of myonuclei within myofibers can be quantitated on histologic sections by counting the number of DAPI-stained nuclei inside of the dystrophin-stained sarcolemma (Horsley et al. 2001, 2003, Jansen and Pavlath 2006). Furthermore, during muscle regeneration developmental isoforms of several proteins are re-expressed and then eventually replaced by adult isoforms when regeneration is complete. The efficiency of muscle regeneration can also be quantitated by analysis of the time course of expression of these various developmental isoforms. Studies of myoblast fusion in model organisms have been invaluable in providing new information about regulatory pathways controlling myoblast fusion and have stimulated further experiments on their vertebrate orthologues. However, a number of pathways demonstrated to control myoblast fusion in mammals are not found in lower organisms and may have evolved in higher eukaryotes to regulate the plasticity of muscle growth, allowing maintenance of muscle mass over an extended lifespan and the growth of muscles containing multiple myofibers. Furthermore, due to their larger body size and complexity in different muscle types, mammalian models such as mouse and rat allow one to study myoblast fusion in the context of muscle regeneration and hypertrophy.
12.4 Current Areas of Research in Myoblast Fusion Myoblast fusion follows an ordered set of cellular events, including elongation, cell migration, adhesion, and membrane fusion (Knudsen and Horwitz 1977). As more molecular requirements for myoblast fusion are discovered, we have learned much
254
G.K. Pavlath
about these steps that are intimately associated with, and influence the ability of two muscle cells to fuse. Below these cellular behaviors are discussed in the context of their regulatory molecules.
12.4.1 Elongation and Membrane Alterations The application of high-resolution microscopy approaches has greatly furthered our knowledge about the molecules that regulate this particular phase of myogenesis in the last few years. Differentiation leads to a shape change in myoblasts and they become elongated, spindle shaped cells that migrate towards other differentiated myoblasts to form groups of aligned cells. (Ohtake et al. 2006, Swailes et al. 2006, Nowak et al. 2009). The morphology change is permanent and elongated myoblasts only fuse with other elongated cells or myotubes (Nowak et al. 2009). This phase of myogenesis has only recently been studied at the molecular level (Table 12.1). Elongation is likely a complex interplay among multiple types of proteins as alterations in integrins, matrix remodeling enzymes and molecules that affect the cytoskeletal network lead to defects in myoblast elongation. Time-lapse microscopy indicates that elongation is followed by extension of lamellopodia and filopodia, dynamic cell extensions composed of actin filaments, which make contact with neighboring muscle cells (Yoon et al. 2007, Mukai and Hashimoto 2008, Mukai et al. 2009, Nowak et al. 2009, Stadler et al. 2010). The role of these dynamic cell structures in the fusion process is unknown. However, these structures are sites for the localization of adhesion molecules (Abramovici and Gee 2007, Mukai et al. 2009), and signaling molecules (Abramovici and Gee 2007, Mukai and Hashimoto Table 12.1 Molecules associated with shape changes and membrane alterations during myoblast fusion Molecule
Function
Process/structure
References
Diacylglycerol kinase ζ
Diacylglycerol metabolism Microtubule regulation at cell cortex Integrin-associated cytoplasmic adaptor Matrix degradation
Filopodia
Abramovici and Gee (2007) Straube and Merdes (2007)
EB3
Kindlin-2
Elongation
Elongation
Dowling et al. (2008)
Elongation
Ohtake et al. (2006)
MT1-matrix metalloprotease Non-muscle myosin 2A Protein kinase A
Associates with actin
Elongation
Swailes et al. (2006)
cAMP production
Lamellipodia
RhoE
Signaling
Elongation, alignment
Mukai and Hashimoto (2008) Fortier et al. (2008)
12
Mechanisms of Myoblast Fusion in Mammals
255
2008). These filopodia are reminiscent of axon growth cones and may respond to chemoattractants produced by other cells and be necessary to recognize other fusion competent muscle cells. In addition, filopodia may act as a zipper mechanism by which two cells are pulled in close apposition to one another for eventual fusion (Abramovici and Gee 2007). Further experiments are required to determine the full array of molecules present within filopodia and the types of molecules they may be sensing in a neighboring muscle cell.
12.4.2 Migration Migration is necessary to achieve cell–cell contact during myogenesis. Cell contact is required both to trigger differentiation (Krauss et al. 2005) and to allow fusion-competent myoblasts to fuse with one another and with nascent myotubes. Time-lapse photography in vitro indicates that myoblasts are very motile, but cell locomotion slows during differentiation (Powell 1973). Myoblasts preferentially move into some fields and out of others as myotubes begin to form (Chazaud et al. 1998) suggesting directed migration in response to chemotactic factors. Surprisingly, myotubes are also motile in vitro (Nowak et al. 2009). A number of factors regulate muscle cell migration in vitro but only a few of these have specifically been studied for their ability to affect myotube formation and growth. As seen in Table 12.2, myogenesis is influenced by both positive and negative regulators of cell migration. Positive regulators can modulate the velocity or direction of muscle cell migration (Horsley et al. 2003, Jansen and Pavlath 2006, Lafreniere et al. 2006,
Table 12.2 Molecules that regulate cell motility and influence myoblast fusion Molecule
Type of molecule
Action on migration
References
CD164
Sialomucin that forms complex with the chemokine CXCR4 Cytokine
Stimulatory
Bae et al. (2008)
Stimulatory
Endocytic C-type lectin; collagen clearance G-protein coupled receptor Prostaglandin
Stimulatory
Horsley et al. (2003), Lafreniere et al. (2006) Jansen and Pavlath (2006) Griffin et al. (2009)
Serine protease
Stimulatory
Bondesen et al. (2007) Chazaud et al. (2000)
Receptor for urokinase plasminogen activator
Stimulatory
Chazaud et al. (2000)
Interleukin 4
Mannose receptor Mouse odorant receptor 23 Prostacyclin Urokinase plasminogen activator Urokinase plasminogen activator receptor
Stimulatory Inhibitory
256
G.K. Pavlath
Griffin et al. 2009) as well as filopodia formation and membrane ruffling at the leading edge of migrating cells (Chazaud et al. 2000). Negative regulators of myoblast migration may act to enhance cell fusion by acting as a “brake” on migrating cells to facilitate cell–cell contact and adhesion (Bondesen et al. 2007). The net balance between these two classes of migratory regulators would be critical for myogenesis. Thus, cell migration during myogenesis may be modulated not only by increasing positive regulators but also by decreasing anti-migratory factors.
12.4.3 Muscle Cell Recognition/Adhesion Early studies described the early recognition/adhesion process between myoblasts as consisting of both calcium-independent and -dependent components (Gibralter and Turner 1985, Pizzey et al. 1988) as well as the necessity of glycoproteins (Knudsen 1985, Knudsen et al. 1990). Recent large scale real-time PCR profiling of the C2C12 mouse muscle cell line during differentiation revealed the expression of 276 genes encoding proteins with a role in glycosylation including biosynthetic enzymes, sugar metabolism and sugar transport (Janot et al. 2009). The 37 most highly regulated glycogenes included genes for cell adhesion molecules as well for enzymes regulating the synthesis of glycosaminoglycans and glycolipids. Interestingly, overexpression of cytotoxic T cell GalNAc transferase in skeletal muscle of mice decreased the size of myofibers in part due to a failure of myoblast fusion (Xia et al. 2002). Furthermore, dynamic clustering and dispersion of lipid rafts appears to be necessary for regulating the accumulation of adhesion-complex proteins at presumptive fusion sites (Mukai et al. 2009). Several studies have analyzed the molecules specifically involved in cell recognition and adhesion in mammalian muscle. Apparent from an analysis of Table 12.3 is that molecules identified to date that influence myoblast adhesion and fusion represent multiple classes of molecules. This diversity may allow these molecules to regulate not only myoblast recognition/adhesion, but also activation of specific intracellular signaling pathways, such as Rac1 (M-cadherin) (Charrasse et al. 2007) and cAMP (mouse odorant receptor 23) (Griffin et al. 2009). For some molecules, discrepancies exist between the results of in vitro and in vivo experiments demonstrating a purported role in myoblast fusion. This discrepancy could relate to the methods employed in vitro vs. in vivo, e.g. blocking antibodies or peptides and overexpression studies in vitro compared to genetic methods of ablating protein expression in vivo. Alternatively, a greater redundancy among specific adhesion molecules may exist in vivo. The recognition and adhesion of myoblasts prior to fusion likely involves multiple adhesion molecules but the interplay between such molecules is unknown. Following adhesion, electron microscopic studies reveal that alignment occurs through the parallel apposition of the membranes of elongated myoblasts with myotubes or other myoblasts (Wakelam 1985). Cytoplasmic vesicles are observed in close proximity to the plasma membranes where membrane union occurs in small regions between the aligned plasma membranes (Wakelam 1985). During
12
Mechanisms of Myoblast Fusion in Mammals
257
Table 12.3 Molecules that influence cell–cell adhesion and myoblast fusion Molecule
Type of molecule
In vitro fusion?
In vivo fusion?
α4β1 integrin
Integrin
Yes
Not required
M-cadherin
Cadherin
Yes
Not required
Mouse odorant receptor 23 Melanoma cell adhesion molecule (M-CAM) Neural cell adhesion molecule (NCAM)
G-protein coupled receptor Immunoglobulin superfamily
Yes
Yes
Rosen et al. (1992); Yang et al. (1996) Charrasse et al. (2006, 2007) Griffin et al. (2009)
Yes
n.d.
Cerletti et al. (2006)
Immunoglobulin superfamily
Yes
Not required
Vascular cell adhesion molecule (VCAM)
Immunoglobulin superfamily
Yes
n.d.
Knudsen et al. (1990); Charlton et al. (2000); Suzuki et al. (2003) Rosen et al. (1992)
References
n.d., not determined
development (Kalderon and Gilula 1979) or muscle regeneration (Robertson et al. 1990) unilamellar vesicles are also observed in close apposition to the fusing membranes of muscle cells. The function of these vesicles is unknown.
12.4.4 Actin Dynamics and Integrin Function Extensive cytoskeletal reorganization occurs before and after fusion (Fulton et al. 1981). Analyses of myoblast fusion during Drosophila development have been invaluable in demonstrating a role for the actin cytoskeleton and its myriad of regulatory molecules at the site of cell adhesion and fusion. Readers are referred to many recent excellent reviews for in depth discussion on this topic (Peckham 2008, Guerin and Kramer 2009, Onel 2009, Rochlin et al. 2010, Chapter 6, this volume). This is one of the few areas in myoblast fusion where we have knowledge of how multiple components act in an integrated manner. Given that orthologues of these proteins exist in mammals, there has been great interest in determining whether actin regulators and signaling molecules identified in Drosophila have conserved functions in mammalian myoblast fusion and what types of changes in the actin cytoskeleton occur in fusing cells. Visualization of actin cytostructural elements in real-time and at high resolution revealed similar dynamic changes in fusing mouse myoblasts in vitro as in Drosophila developmental myoblast fusion (Swailes et al. 2004, Duan and Gallagher 2009, Nowak et al. 2009, Stadler et al. 2010). Not surprisingly, latrunculin A or B, inhibitors of actin remodeling, decrease fusion in mouse myoblasts (Kim et al. 2007, O’Connor et al. 2008, Nowak et al. 2009).
258
G.K. Pavlath
Table 12.4 Molecules that influence actin dynamics or integrin signaling during myoblast fusion Molecule
Function
Arf6 β1 integrin Brag2 Cdc42 Dock1, 5
GTPase Recruitment of CD9 GEF; Arf6/Rac1 activation GTPase GEF; Rac1 activation
Filamin C Focal adhesion kinase
Nap1 N-WASP Rac1 Talin 1, 2 Trio
References
Chen et al. (2003) Schwander et al. (2003) Pajcini et al. (2008) Vasyutina et al. (2009) Laurin et al. (2008), Pajcini et al. (2008) Muscle specific actin cross-linking Dalkilic et al. (2006) protein Downregulation of caveolin 3 and Quach et al. (2009) β1D integrin, cytoskeletal remodeling? WAVE complex formation Nowak et al. (2009) Activation of Arp2/3 Kim et al. (2007) GTPase: WAVE complex Charrasse et al. (2007); Vasyutina formation et al. (2009) Connect integrins to the actin Conti et al. (2009) cytoskeleton GEF; Rac1 regulation Charrasse et al. (2007)
Arf6, ADP ribosylation factor 6; Arp2/3, actin-related protein 2/3; GEF, guanine nucleotide exchange factor; WASP, Wiskott-Aldrich syndrome protein; WAVE, WASP family Verprolinhomologous protein
Given the number of proteins regulating myoblast fusion that impinge on actin reorganization (Table 12.4), processes dependent on the actin cytoskeleton must play fundamental roles in cell fusion. The structure of the actin cytoskeleton at the contact site of fusing myoblasts is highly regulated by a complex signal cascade initiated by cell adhesion (Kurisu and Takenawa 2009). Various GTPases and guanine nucleotide exchange factors (GEF) are activated, which in turn impinge on WiskottAldrich Syndrome Protein (WASP) and Wasp family verprolin-homologous protein (WAVE) proteins. Activation of both WASP and WAVE is critical for the Arp2/3 complex to initiate actin polymerization. The actions of these signaling molecules and actin regulators lead to the recruitment of cytoskeletal proteins to contact sites of fusing myoblasts (Vasyutina et al. 2009), which is hypothesized to be essential to localize Golgi-derived prefusion vesicles (Kim et al. 2007) and/or fusion pore formation/expansion (Onel and Renkawitz-Pohl 2009). Integrins and integrin signaling also play roles in modulating myoblast fusion. Integrins could function by regulating formation of protein complexes or by relaying signals to the cytoskeleton (Schwander et al. 2003, Quach et al. 2009).
12.4.5 Regulation of Cell Fusion with Nascent Myotubes Two molecularly distinct phases of cell fusion occur in mammalian muscle cells, a phenomenon also observed in Drosophila myogenesis (Menon and Chia 2001, Rau et al. 2001). Initially, myoblasts fuse with one another to form small, nascent
12
Mechanisms of Myoblast Fusion in Mammals
259
Table 12.5 Molecules that regulate myoblast-myotube fusion Molecule
Role
References
EHD2
Endocytic recycling, binds to myoferlin Inhibits myostatin
Doherty et al. (2008)
Follistatin Growth hormone IGF-1 Interleukin 4 Interleukin 13 mTOR Mannose receptor Myoferlin Nephrin NFATc2 Prostaglandin F2α SHP-2
Activates NFATc2 signaling Stimulates IL13 production Cytokine; stimulates cell motility Cytokine Kinase Transmembrane endocytic receptor; collagen clearance Phospholipid binding; membrane repair Transmembrane protein; outside-in signaling? Transcription factor; regulates IL4 and IL13 expression Activates NFATc2 signaling Phosphatase; c-src regulation
Iezzi et al. (2004), Pisconti et al. (2006) Sotiropoulos et al. (2006) Jacquemin et al. (2004) Horsley et al. (2003) Jacquemin et al. (2007) Park and Chen (2005) Jansen and Pavlath (2006) Doherty et al. (2005) Sohn et al. (2009) Horsley et al. (2001) Horsley and Pavlath (2003) Fornaro et al. (2006)
IGF-1, Insulin growth factor 1; IL4, Interleukin 4; IL13, Interleukin 13; mTOR, mammalian target of rapamycin; NFATc2, Nuclear Factor of Activated T Cells c2
myotubes. Subsequently, additional myoblasts fuse with the nascent myotube leading to increased myonuclear number and cell size. The earliest evidence for the molecular distinctness of these phases of myotube formation and growth was provided by experiments utilizing wheat germ agglutinin (WGA). In the presence of WGA, the mouse muscle cell line C2C12 formed nascent myotubes but myoblastmyotube fusion was blocked (Muroya et al. 1994). Subsequently, a number of other molecules were discovered to be required only for myoblast-myotube fusion (Table 12.5). Why myoblast-myotube fusion should require unique molecules is unknown but may be related to specific challenges inherent with cell fusion to a multinucleated cell containing sarcomeres. Alternatively, these unique molecules could represent a fine-tuning mechanism for controlling the ultimate number of nuclei within a myotube/myofiber, regulating fusion to specific sites along the myotube, or directing cells to fuse with specific myotubes. In addition, since muscle regeneration in vivo is asynchronous, these molecules could specifically control growth of regenerating myofibers rather than allowing new myofibers to form and thus be a means of controlling the number of regenerated myofibers. 12.4.5.1 Nuclear Factor of Activated T Cells: Modulators and Effectors The nuclear factor of activated T cells (NFAT) family of transcription factors is comprised of four calcium-activated members, NFATc1-c4, all of which are expressed in skeletal muscle (Abbott et al. 1998, Jacquemin et al. 2007). NFATc2 plays a central role in orchestrating myoblast-myotube fusion (Horsley et al. 2001, Pavlath and Horsley 2003). NFATc2 null mice were able to normally form regenerating
260
G.K. Pavlath
myofibers after injury but these myofibers grew at a slower rate than in wild type mice and never reached the same final myofiber size as in wild type mice. Similarly, NFATc2 null myoblasts could only form small myotubes in vitro, which was due to a defect in the recruitment and/or fusion of myogenic cells with nascent myotubes (Horsley et al. 2001). Subsequent experiments demonstrated the expression of the cytokine interleukin 4 (IL4) by muscle cells was dependent on NFATc2 (Horsley et al. 2003). IL4 null and IL4 receptor alpha null myoblasts were also defective in the recruitment of myogenic cells with nascent myotubes (Horsley et al. 2003). IL4 may promote cell fusion by enhancing myoblast migration (Lafreniere et al. 2006). Another cytokine, interleukin 13 (IL13), which is similarly regulated by NFAT-dependent transcription, was also able to rescue the fusion defect observed in NFATc2 null myotubes, albeit at higher concentrations (Horsley et al. 2001). Further studies indicated IL13 specifically regulates human myoblast fusion with myotubes during IGF-1 mediated hypertrophy in vitro but not under basal conditions (Jacquemin et al. 2007). IL4 likely promotes fusion in part by regulating expression of the mannose receptor (MR), a cell surface endocytic C-type lectin that binds to sulfated glycoproteins or terminal mannose, fucose or N-acetylglucosamine residues. Like IL4 null myoblasts, MR null myoblasts formed smaller myotubes in vitro and myofibers in vivo (Jansen and Pavlath 2006). MR null myoblasts displayed a general reduction in general motility as well as an impairment of directed migration to unidentified factors released by fusing muscle cells. Collagen uptake was also decreased in MR null muscle cells supporting a role for this endocytic receptor in helping to clear extracellular matrix from the leading edge of migrating cells. Upstream activators of the NFATc2 pathway are also key players in regulating myoblast fusion with nascent myotubes. NFATc2 was required for the increase in myonuclear number due to prostaglandin F2α (Horsley and Pavlath 2003) as well as growth hormone (Sotiropoulos et al. 2006). NFATc2 may also be downstream of a c-src pathway in muscle (Fornaro et al. 2006). Fornaro et al. demonstrated that the phosphatase SHP-2 stimulates c-src leading to activation of NFAT and subsequent fusion of myoblasts with myotubes. IL-4 was decreased in SHP-2 null muscle cells. These results support the idea that NFATc2 is a target for positive regulation by SHP-2 in skeletal muscle. However, whether the phenotype of SHP-2 null mice is a consequence of disrupting NFATc2 solely as opposed to other NFAT family members needs to be formally determined.
12.4.5.2 Additional Molecules that Control Fusion with Nascent Myotubes Follistatin Follistatin is a secreted protein that interacts with several transforming growth factor-β family members and regulates their biologic activity. Importantly, follistatin inhibits the activity of myostatin, a negative regulator of skeletal muscle hypertrophy (Lee and McPherron 2001), a process that can be associated with myoblast fusion (O’Connor and Pavlath 2007). Follistatin regulates myoblast-myotube fusion
12
Mechanisms of Myoblast Fusion in Mammals
261
in response to deacetylase inhibitors (Iezzi et al. 2004). Follistatin is also produced in response to a nitric oxide/cGMP pathway that regulates myoblast fusion in vitro (Pisconti et al. 2006). mTOR The mammalian target of rapamycin (mTOR) is a serine/threonine protein kinase that plays distinct roles during myogenesis. mTOR is required both for myogenic differentiation (Cuenda and Cohen 1999) as well myoblast fusion with myotubes in vitro and in vivo (Park and Chen 2005, Ge et al. 2009). mTOR activity is necessary for the secretion of an unidentified fusion-promoting factor (Park and Chen 2005). Myoferlin Myoferilin is a calcium-dependent phospholipid binding protein with a demonstrated role in myoblast-myotube fusion (Doherty et al. 2005). Primary myoblasts from myoferlin null mice formed smaller myotubes in vitro as well as smaller myofibers in regenerating muscle. Myoferlin associated with the plasma membrane and was concentrated at sites of myoblast-myotube fusion. More recently, myoferlin was shown to interact with the endocytic recycling protein EHD2 (Doherty et al. 2008) and is hypothesized to help regulate membrane fusion at cell fusion sites. Nephrin Nephrin is a transmembrane protein originally studied in the kidney whose expression is induced in muscle cells undergoing fusion (Sohn et al. 2009). It shares structural similarities with the Drosophila protein, Sticks and Stones (Sns), which is expressed by fusion-competent myoblasts. During Drosophila muscle development, Sns interacts with kirre/duf expressed on founder myoblasts to initiate intracellular signaling leading to cell fusion (Chen and Olson 2005). Primary myoblasts from nephrin null mice fused to form nascent myotubes in vitro but further fusion of myoblasts with nascent myotubes was inhibited. Cell mixing experiments demonstrated that nephrin function is required in differentiated myoblasts not in nascent myotubes. The mechanism by which nephrin regulates fusion is unknown but it is likely to be part of a larger protein complex that initiates intracellular signaling because of its structural similarity to Sns.
12.5 Future Prospects Our knowledge of the molecules that regulate myoblast fusion in mammals is rapidly expanding but much still remains to be understood. Large scale screens are likely to identify new molecules and pathways that positively and negatively control myoblast fusion. Although multiple molecular components are already known and more are likely to be identified, emphasis needs to be placed on how these
262
G.K. Pavlath
molecules act together in a broader regulatory network. Moreover, many molecules that regulate myoblast fusion in vitro have not been functionally studied in vivo during development, regeneration or muscle growth. An emphasis should be place upon developing novel methods to study myoblast fusion in vivo. In addition, further work is needed to fully understand what time-lapse imaging of fusing muscle cells reveals about the biochemistry of myoblast fusion. Finally, we lack information about the actual fusogen molecules that regulate membrane breakdown at the final stage of fusion. Further research into the areas discussed in this review will provide insight into the biological processes of cell–cell fusion. Acknowledgements I am grateful to the many talented students and fellows who have contributed to these studies over the years in my lab. GKP is supported by grants AR047314, AR051372, AR052730 and NS069234 from the National Institutes of Health.
References Abbott KL, Friday BB, Thaloor D et al (1998) Activation and cellular localization of the cyclosporine A-sensitive transcription factor NF-AT in skeletal muscle cells. Mol Biol Cell 9:2905–2916 Abramovici H, Gee SH (2007) Morphological changes and spatial regulation of diacylglycerol kinase-zeta, syntrophins, and Rac1 during myoblast fusion. Cell Motil Cytoskeleton 64: 549–567 Andres V, Walsh K (1996) Myogenin expression, cell cycle withdrawal, and phenotypic differentiation are temporally separable events that precede cell fusion upon myogenesis. J Cell Biol 132:657–666 Bae GU, Gaio U, Yang YJ et al (2008) Regulation of myoblast motility and fusion by the CXCR4associated sialomucin, CD164. J Biol Chem 283:8301–8309 Blau HM, Chiu CP, Webster C (1983) Cytoplasmic activation of human nuclear genes in stable heterocaryons. Cell 32:1171–1180 Bondesen BA, Jones KA, Glasgow WC et al (2007) Inhibition of myoblast migration by prostacyclin is associated with enhanced cell fusion. FASEB J 21:3338–3345 Cerletti M, Molloy MJ, Tomczak KK et al (2006) Melanoma cell adhesion molecule is a novel marker for human fetal myogenic cells and affects myoblast fusion. J Cell Sci 119: 3117–3127 Charlton CA, Mohler WA, Blau HM (2000) Neural cell adhesion molecule (NCAM) and myoblast fusion. Dev Biol 221:112–119 Charlton CA, Mohler WA, Radice GL et al (1997) Fusion competence of myoblasts rendered genetically null for N-cadherin in culture. J Cell Biol 138:331–336 Charrasse S, Comunale F, Fortier M et al (2007) M-cadherin activates Rac1 GTPase through the Rho-GEF trio during myoblast fusion. Mol Biol Cell 18:1734–1743 Charrasse S, Comunale F, Grumbach Y et al (2006) RhoA GTPase regulates M-cadherin activity and myoblast fusion. Mol Biol Cell 17:749–759 Chazaud B, Bonavaud S, Plonquet A et al (2000) Involvement of the [uPAR:uPA:PAI-1:LRP] complex in human myogenic cell motility. Exp Cell Res 258:237–244 Chazaud B, Christov C, Gherardi RK et al (1998) In vitro evaluation of human muscle satellite cell migration prior to fusion into myotubes. J Muscle Res Cell Motil 19:931–936 Chen EH, Olson EN (2005) Unveiling the mechanisms of cell–cell fusion. Science 308: 369–373 Chen EH, Pryce BA, Tzeng JA et al (2003) Control of myoblast fusion by a guanine nucleotide exchange factor, loner, and its effector ARF6. Cell 114:751–762
12
Mechanisms of Myoblast Fusion in Mammals
263
Conti FJ, Monkley SJ, Wood MR et al (2009) Talin 1 and 2 are required for myoblast fusion, sarcomere assembly and the maintenance of myotendinous junctions. Development 136: 3597–3606 Cuenda A, Cohen P (1999) Stress-activated protein kinase-2/p38 and a rapamycin-sensitive pathway are required for C2C12 myogenesis. J Biol Chem 274:4341–4346 Dalkilic I, Schienda J, Thompson TG et al (2006) Loss of FilaminC (FLNc) results in severe defects in myogenesis and myotube structure. Mol Cell Biol 26:6522–6534 Doherty KR, Cave A, Davis DB et al (2005) Normal myoblast fusion requires myoferlin. Development 132:5565–5575 Doherty KR, Demonbreun AR, Wallace GQ et al (2008) The endocytic recycling protein EHD2 interacts with myoferlin to regulate myoblast fusion. J Biol Chem 283:20252–20260 Dowling JJ, Vreede AP, Kim S et al (2008) Kindlin-2 is required for myocyte elongation and is essential for myogenesis. BMC Cell Biol 9:36 Duan R, Gallagher PJ (2009) Dependence of myoblast fusion on a cortical actin wall and nonmuscle myosin IIA. Dev Biol 325:374–385 Fornaro M, Burch PM, Yang W et al (2006) SHP-2 activates signaling of the nuclear factor of activated T cells to promote skeletal muscle growth. J Cell Biol 175:87–97 Fortier M, Comunale F, Kucharczak J et al (2008) RhoE controls myoblast alignment prior fusion through RhoA and ROCK. Cell Death Differ 15:1221–1231 Fulton AB, Prives J, Farmer SR et al (1981) Developmental reorganization of the skeletal framework and its surface lamina in fusing muscle cells. J Cell Biol 91:103–112 Ge Y, Wu AL, Warnes C et al (2009) mTOR regulates skeletal muscle regeneration in vivo through kinase-dependent and kinase-independent mechanisms. Am J Physiol Cell Physiol 297: C1434–C1444 Gibralter D, Turner DC (1985) Dual adhesion systems of chick myoblasts. Dev Biol 112:292–307 Griffin CA, Kafadar KA, Pavlath GK (2009) MOR23 promotes muscle regeneration and regulates cell adhesion and migration. Dev Cell 17:649–661 Guerin CM, Kramer SG (2009) Cytoskeletal remodeling during myotube assembly and guidance: coordinating the actin and microtubule networks. Commun Integr Biol 2:452–457 Horsley V, Friday BB, Matteson S et al (2001) Regulation of the growth of multinucleated muscle cells by an NFATC2-dependent pathway. J Cell Biol 153:329–338 Horsley V, Jansen KM, Mills ST et al (2003) IL-4 acts as a myoblast recruitment factor during mammalian muscle growth. Cell 113:483–494 Horsley V, Pavlath GK (2003) Prostaglandin F2(alpha) stimulates growth of skeletal muscle cells via an NFATC2-dependent pathway. J Cell Biol 161:111–118 Iezzi S, Di Padova M, Serra C et al (2004) Deacetylase inhibitors increase muscle cell size by promoting myoblast recruitment and fusion through induction of follistatin. Dev Cell 6: 673–684 Jacquemin V, Butler-Browne GS, Furling D et al (2007) IL-13 mediates the recruitment of reserve cells for fusion during IGF-1-induced hypertrophy of human myotubes. J Cell Sci 120:670–681 Jacquemin V, Furling D, Bigot A et al (2004) IGF-1 induces human myotube hypertrophy by increasing cell recruitment. Exp Cell Res 299:148–158 Janot M, Audfray A, Loriol C et al (2009) Glycogenome expression dynamics during mouse C2C12 myoblast differentiation suggests a sequential reorganization of membrane glycoconjugates. BMC Genomics 10:483 Jansen KM, Pavlath GK (2006) Mannose receptor regulates myoblast motility and muscle growth. J Cell Biol 174:403–413 Kalderon N, Gilula NB (1979) Membrane events involved in myoblast fusion. J Cell Biol 81: 411–425 Kim S, Shilagardi K, Zhang S et al (2007) A critical function for the actin cytoskeleton in targeted exocytosis of prefusion vesicles during myoblast fusion. Dev Cell 12:571–586 Knudsen KA (1985) The calcium-dependent myoblast adhesion that precedes cell fusion is mediated by glycoproteins. J Cell Biol 101:891–897
264
G.K. Pavlath
Knudsen KA, Horwitz AF (1977) Tandem events in myoblast fusion. Dev Biol 58:328–338 Knudsen KA, Myers L, McElwee SA (1990) A role for the Ca2(+)-dependent adhesion molecule, N-cadherin, in myoblast interaction during myogenesis. Exp Cell Res 188:175–184 Krauss RS, Cole F, Gaio U et al (2005) Close encounters: regulation of vertebrate skeletal myogenesis by cell–cell contact. J Cell Sci 118:2355–2362 Kurisu S, Takenawa T (2009) The WASP and WAVE family proteins. Genome Biol 10:226 Lafreniere JF, Mills P, Bouchentouf M et al (2006) Interleukin-4 improves the migration of human myogenic precursor cells in vitro and in vivo. Exp Cell Res 312:1127–1141 Laurin M, Fradet N, Blangy A et al (2008) The atypical Rac activator Dock180 (Dock1) regulates myoblast fusion in vivo. Proc Natl Acad Sci USA 105:15446–15451 Lee SJ, McPherron AC (2001) Regulation of myostatin activity and muscle growth. Proc Natl Acad Sci USA 98:9306–9311 Menon SD, Chia W (2001) Drosophila rolling pebbles: a multidomain protein required for myoblast fusion that recruits D-Titin in response to the myoblast attractant Dumbfounded. Dev Cell 1:691–703 Mohler WA, Blau HM (1996) Gene expression and cell fusion analyzed by lacZ complementation in mammalian cells. Proc Natl Acad Sci USA 93:12423–12427 Mukai A, Hashimoto N (2008) Localized cyclic AMP-dependent protein kinase activity is required for myogenic cell fusion. Exp Cell Res 314:387–397 Mukai A, Kurisaki T, Sato SB et al (2009) Dynamic clustering and dispersion of lipid rafts contribute to fusion competence of myogenic cells. Exp Cell Res 315:3052–3063 Muroya S, Takagi H, Tajima S et al (1994) Selective inhibition of a step of myotube formation with wheat germ agglutinin in a murine myoblast cell line, C2C12. Cell Struct Funct 19:241–252 Nowak SJ, Nahirney PC, Hadjantonakis AK et al (2009) Nap1-mediated actin remodeling is essential for mammalian myoblast fusion. J Cell Sci 122:3282–3293 O’Connor RS, Mills ST, Jones KA et al (2007) A combinatorial role for NFAT5 in both myoblast migration and differentiation during skeletal muscle myogenesis. J Cell Sci 120:149–159 O’Connor RS, Pavlath GK (2007) Point:counterpoint: satellite cell addition is/is not obligatory for skeletal muscle hypertrophy. J Appl Physiol 103:1099–1100 O’Connor RS, Steeds CM, Wiseman RW et al (2008) Phosphocreatine as an energy source for actin cytoskeletal rearrangements during myoblast fusion. J Physiol 586:2841–2853 Ohtake Y, Tojo H, Seiki M (2006) Multifunctional roles of MT1-MMP in myofiber formation and morphostatic maintenance of skeletal muscle. J Cell Sci 119:3822–3832 Onel SF (2009) Actin regulators take the reins in Drosophila myoblast fusion. Central Eur J Biol 4:11–18 Onel SF, Renkawitz-Pohl R (2009) FuRMAS: triggering myoblast fusion in Drosophila. Dev Dyn 238:1513–1525 Pajcini KV, Pomerantz JH, Alkan O et al (2008) Myoblasts and macrophages share molecular components that contribute to cell–cell fusion. J Cell Biol 180:1005–1019 Park IH, Chen J (2005) Mammalian target of rapamycin (mTOR) signaling is required for a latestage fusion process during skeletal myotube maturation. J Biol Chem 280:32009–32017 Pavlath GK, Horsley V (2003) Cell fusion in skeletal muscle–central role of NFATC2 in regulating muscle cell size. Cell Cycle 2:420–423 Peckham M (2008) Engineering a multi-nucleated myotube, the role of the actin cytoskeleton. J Microsc 231:486–493 Pisconti A, Brunelli S, Di Padova M et al (2006) Follistatin induction by nitric oxide through cyclic GMP: a tightly regulated signaling pathway that controls myoblast fusion. J Cell Biol 172:233–244 Pizzey JA, Jones GE, Walsh FS (1988) Requirements for the Ca2+ -independent component in the initial intercellular adhesion of C2 myoblasts. J Cell Biol 107:2307–2317 Powell JA (1973) Development of normal and genetically dystrophic mouse muscle in tissue culture. I. Prefusion and fusion activities of muscle cells: phase contrast and time lapse study. Exp Cell Res 80:251–264
12
Mechanisms of Myoblast Fusion in Mammals
265
Prior BM, Modlesky CM, Evans EM et al (2001) Muscularity and the density of the fat-free mass in athletes. J Appl Physiol 90:1523–1531 Quach NL, Biressi S, Reichardt LF et al (2009) Focal adhesion kinase signaling regulates the expression of caveolin 3 and beta1 integrin, genes essential for normal myoblast fusion. Mol Biol Cell 20:3422–3435 Rau A, Buttgereit D, Holz A et al (2001) Rolling pebbles (rols) is required in Drosophila muscle precursors for recruitment of myoblasts for fusion. Development 128:5061–5073 Robertson TA, Grounds MD, Mitchell CA et al (1990) Fusion between myogenic cells in vivo: an ultrastructural study in regenerating murine skeletal muscle. J Struct Biol 105:170–182 Rochlin K, Yu S, Roy S et al (2010) Myoblast fusion: when it takes more to make one. Dev Biol 341:66–83 Rosen GD, Sanes JR, LaChance R et al (1992) Roles for the integrin VLA-4 and its counter receptor VCAM-1 in myogenesis. Cell 69:1107–1119 Schwander M, Leu M, Stumm M et al (2003) Beta1 integrins regulate myoblast fusion and sarcomere assembly. Dev Cell 4:673–685 Shinn-Thomas JH, Scranton VL, Mohler WA (2008) Quantitative assays for cell fusion. Methods Mol Biol 475:347–361 Sohn RL, Huang P, Kawahara G et al (2009) A role for nephrin, a renal protein, in vertebrate skeletal muscle cell fusion. Proc Natl Acad Sci USA 106:9274–9279 Sotiropoulos A, Ohanna M, Kedzia C et al (2006) Growth hormone promotes skeletal muscle cell fusion independent of insulin-like growth factor 1 up-regulation. Proc Natl Acad Sci USA 103:7315–7320 Stadler B, Blattler TM, Franco-Obregon A (2010) Time-lapse imaging of in vitro myogenesis using atomic force microscopy. J Microsc 237:63–69 Straube A, Merdes A (2007) EB3 regulates microtubule dynamics at the cell cortex and is required for myoblast elongation and fusion. Curr Biol 17:1318–1325 Swailes NT, Colegrave M, Knight PJ et al (2006) Non-muscle myosins 2A and 2B drive changes in cell morphology that occur as myoblasts align and fuse. J Cell Sci 119:3561–3570 Swailes NT, Knight PJ, Peckham M (2004) Actin filament organization in aligned prefusion myoblasts. J Anat 205:381–391 Vasyutina E, Martarelli B, Brakebusch C et al (2009) The small G-proteins Rac1 and Cdc42 are essential for myoblast fusion in the mouse. Proc Natl Acad Sci USA 106:8935–8940 Wakelam MJ (1985) The fusion of myoblasts. Biochem J 228:1–12 Xia B, Hoyte K, Kammesheidt A et al (2002) Overexpression of the CT GalNAc transferase in skeletal muscle alters myofiber growth, neuromuscular structure, and laminin expression. Dev Biol 242:58–73 Yang JT, Rando TA, Mohler WA et al (1996) Genetic analysis of alpha 4 integrin functions in the development of mouse skeletal muscle. J Cell Biol 135:829–835 Yoon S, Molloy MJ, Wu MP et al (2007) C6ORF32 is upregulated during muscle cell differentiation and induces the formation of cellular filopodia. Dev Biol 301:70–81
Chapter 13
The Endogenous Envelope Protein Syncytin Is Involved in Myoblast Fusion Bolette Bjerregaard, Jan Fredrik Talts, and Lars-Inge Larsson
Abstract Development of human skeletal muscle depends upon fusion of myoblasts to form multinucleated muscle fibers. Many factors are important to this process, but, so far, molecules directly mediating fusions have not been identified. In man, the highly conserved endogenous retroviral envelope protein syncytin-1 is the best candidate for a true fusogen. Here, we summarize data showing that syncytin-1 and its receptors ASCT-1 and -2 are expressed in human myoblasts and that syncytin-1 is involved in myoblast fusion. These data suggest a more wideranging biological role for this endogenous retroviral envelope gene that hitherto suspected. Keywords ASCT-1 · ASCT-2 · cell-cell fusion · filopodia · fusion · myoblasts · nanotubes · skeletal muscle · syncytin Abbreviations ADAM ASCT CD EFF-1 ERV F-actin GADPH GCM HERV IL MFSD2 NCAM NFATc2
A disintegrin and metalloproteinase domain Alanine, serine and cysteine-selective transporters Cluster of differentiation Epithelial fusion failure Endogenous retrovirus Filamentous actin Glyceraldehyde 3-phosphate dehydrogenase, Glial cell missing Human endogenous retrovirus Interleukin Major facilitator superfamily domain containing 2 Neural cell adhesion molecule Nuclear factor of activated T cells c2
B. Bjerregaard (B) Division of Cell Biology, Faculty of Life Sciences, University of Copenhagen, DK-1870 Frederiksberg C, Denmark e-mail:
[email protected]
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_13, C Springer Science+Business Media B.V. 2011
267
268
PCR RT VCAM
B. Bjerregaard et al.
Polymerase chain reaction Reverse transcriptase Vascular cell adhesion molecule
Contents 13.1 Introduction . . . . . . . . . . . . 13.2 Syncytin-1 and Myoblast Fusion . . . 13.3 How Does Syncytin-1 Mediate Fusion? 13.4 Conclusions and Perspectives . . . . References . . . . . . . . . . . . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
268 269 272 273 273
13.1 Introduction Myoblast fusion is required for skeletal myogenesis as it allows for the formation and growth of multinucleated myotubes. It is a multistep process, whereby myoblasts initially recognize and adhere to each other, elongate, align and form electron-dense granules at sites of cell−cell contact, where after they fuse to form myotubes (Doberstein et al. 1997, Horsley and Pavlath 2004, Chapters 6 and 12). Myotubes subsequently recruit additional myoblasts for fusion. The resulting skeletal muscle fibers are no longer able to divide and depend upon satellite cells for regeneration. In response to growth factors, satellite cells become mitotically active and proliferative myoblasts, which differentiate and fuse with pre-existing myofibers and to one another to restore normal tissue architecture (Horsley and Pavlath 2004). A number of factors have been shown to be more or less essential to skeletal myogenesis (see Chapter 12). Such factors include signaling molecules like interleukin-4 (IL-4), which recruits myoblasts (Horsley et al. 2003), adhesion molecules like β1-integrin, cadherins, neural and vascular cell adhesion molecules (N-CAMs and V-CAMs) (Schwander et al. 2003, Charrasse et al. 2007, Charlton et al. 2000, and Yang et al. 1996), membrane-organizing proteins like the CD9 and CD81 (Tachibana and Hemler 1999), ADAM (a disintegrin and metalloproteinase domain) 12 (meltrin α, Yagami-Hiromasa et al. 1995), caveolin-3 (Galbiati et al. 1999), the cytoskeleton reorganizing tyrosine phosphatase SHP-2 (Kontarides et al. 2004), and transcription factors like NFATC2 (Pavlath and Horsley 2003). However, molecules directly mediating the fusion event have not yet been identified (Sapir et al. 2008). Studies in other cell fusion systems have shown that fusogenic proteins are able to form hairpin-like, alpha-bundles, thereby bringing the membranes into close proximity and allowing for fusion (Chen and Olson 2005, Oren-Suissa and Podbilewicz 2007). This type of fusion mechanism is characteristic of both viral envelope proteins like syncytin-1 and of invertebrate fusion proteins like EFF-1, which are thus strong candidates for being bona fide fusogens (Oren-Suissa and Podbilewicz 2007).
13
The Endogenous Envelope Protein Syncytin Is Involved in Myoblast Fusion
269
The syncytins are highly conserved, human endogenous retroviral (HERV) envelope proteins, which are only expressed in man and old-world primates. They consist of the founding member syncytin-1 and syncytin-2, which both have been implicated in placental trophoblast fusion (Blond et al. 2000, Mi et al. 2000, Vargas et al. 2009). Moreover, the existence of two murine endogenous retroviral envelope genes (Syncytin-A and -B), homologous but not orthologous, to the human syncytin genes have been reported, and syncytin-A has been shown to be essential for trophoblast cell fusion (Dupressoir et al. 2005, Gong et al. 2007 and Dupressoir et al. 2009). Very recently, an envelope gene with fusogenic activity and placenta-specific expression has been identified in the rabbit, named syncytin-Ory1 (Heidmann et al. 2009). This indicates that different retroviral envelope proteins have been acquired by different genomes and may facilitate fusion in various animals. Syncytin-1 binds to a neutral amino acid transporter, ASCT-2, which is widely expressed, and uses ASCT-1 as an auxiliary receptor (Blond et al. 2000, Lavillette et al. 2002, Cheynet et al. 2006). Syncytin-2 however, binds to the Major Facilitator Superfamily Domain Containing 2 (MFSD2) receptor, which exhibits placenta specific expression (Esnault et al. 2008). Syncytin-1 was subsequently found to be involved in cancer-endothelial and cancer–cancer cell fusions (Bjerregaard et al. 2006, Strick et al. 2007). Additionally, it been shown to be a potential prognostic factor for breast and colon cancer and a potential marker for leukemias and lymphomas (Larsson et al. 2007, Larsen et al. 2009, Sun et al. 2010).
13.2 Syncytin-1 and Myoblast Fusion In order to study whether syncytins might be involved in additional cell–cell fusions, we investigated primary human muscle myoblasts from five different donors (Bjerregaard, Talts, and Larsson: submitted). RT-PCR demonstrated that exponentially growing skeletal muscle myoblasts expressed mRNAs encoding syncytin-1, ASCT-1, and ASCT-2. Moreover, quantitative RT-PCR showed that the expression of syncytin-1 increased upon induction of myoblast fusions. Thus, syncytin-1 mRNA levels increased fivefold (from 0.52 ± 0.62 to 2.51 ± 0.93 relative expression ratios, normalized to GADPH, p < 0.05) following transfer of myoblasts to differentiation medium, whereas the expression of ASCT-2 did not change significantly (from 0.69 ± 0.16 to 1.04 ± 0.26, p = 0.89; ASCT-1 was not studied in this respect). Western blotting, using a monoclonal antibody raised against the extracellular region of syncytin-1 revealed three main immunoreactive bands of 60, 50, and 35 kDa, respectively (see Fig. 13.1). The 60 kDa band, previously identified in placental and breast cancer cells (Frendo et al. 2003, Bjerregaard et al. 2006), predominated in the exponentially growing myoblasts. In contrast, upon induction of fusion, the intensity of this band decreased whilst the intensities of the 50 and 35 kDa band increased (Bjerregaard, Talts, and Larsson: submitted). Overall, the protein level of syncytin-1 increased when myoblasts were induced to fuse.
270
B. Bjerregaard et al.
Fig. 13.1 Western blot showing syncytin-1 immunoreactive bands in exponentially growing (a) and differentiating (b) human myoblasts. Note the increased immunoreactivity in the differentiating cells as well as the fact that the 60 kD band predominates in the exponentially growing cells whereas the 50 and 35 kD bands predominate in the differentiating cells. Below, reprobing for beta-actin (loading control) is shown
Syncytin-1 antisense treatment, which previously has been found to inhibit trophoblast-trophoblast and cancer-endothelial cell fusions (Frendo et al. 2003, Bjerregaard et al. 2006) markedly decreased syncytin-1 mRNA expression (to 18.4 ± 6.8% of that obtained with a scrambled control oligo, p < 0.05), but did not significantly change ASCT-2 expression (88.6 ± 12.7% of controls, p = 0.46) and Western blotting documented a pronounced decrease in syncytin-1 protein expression upon antisense treatment. In contrast, protein expression of myogenin, which is involved in myoblast differentiation, was not affected by the antisense treatment (data not shown). Antisense oligonucleotide treatment was accompanied by a significant reduction in the myoblast fusion index (the number of nuclei in fused myotubes divided by the total number of nuclei) (Fig. 13.2a–c). Peptide treatment with an syncytin-1 inhibitory peptide, which previously has been found to inhibit trophoblast and cancer-endothelial cell fusions (Bjerregaard et al. 2006, Chang et al. 2004) also significantly inhibited myoblast fusion (Fig. 13.2c). These data show that syncytin-1 and its receptors ASCT-1 and -2 are expressed by human myoblasts and suggest that syncytin-1 may act as a mediator of human skeletal myoblast fusion. Previously, syncytin-1 transcripts were detected in human muscle biopsies. Such transcripts were assumed to be derived from macrophages (Oluwole et al. 2007) because syncytin-1 previously had been detected in macrophages in the central nervous system (Antony et al. 2004). In the light of the present data, such transcripts might instead reflect expression by muscle-lineage cells, possibly engaged in a regenerative event.
13
The Endogenous Envelope Protein Syncytin Is Involved in Myoblast Fusion
271
Fig. 13.2 Effects of interference with syncytin-1 on myoblast fusions. a and b: Human myoblasts were transfected either with a scrambled control oligonucleotide (a) or with a phosphorthioateprotected syncytin-1 antisense oligonucleotide (env-W 758–777) (b). After 48 h, myoblasts were induced to fuse by exchanging the growth medium with differentiation medium and cultured for additionally 2 days before staining for determining the percentage of nuclei in myotubes relative to the total number of nuclei (± SD). Note the predominance of multinucleated myotubes in the control and of mononucleated myoblasts in the antisense treated culture. (c) Fusion indices of myoblasts pretreated with either a scrambled control oligonucleotide or with a syncytin-1 antisense oligonucleotide and then permitted to grow for 4 days in differentiation medium (antisense experiment; left) and fusion indices of myoblasts treated with either a syncytin-1 inhibitory peptide or with a control peptide and then permitted to grow for 4 days in differentiation medium (peptide experiment; right). Note that both approaches decreases fusion indices to approximately 25–30% of controls. ∗∗∗ denotes p < 0.001
Immunofluorescence staining of live (unpermeabilized) myoblasts induced to fuse showed that syncytin-1 localized in a polarized fashion to cell membranes as well as to filopodial extensions, which connected myoblasts to each other (see Fig. 13.3). As discussed in Chapter 12, filopodial contacts are observed early in the myoblast fusion process where they may play an, as yet, undefined functional role. Inhibition of proteins involved in filopodia formation decreases myoblast fusion (Yoon et al. 2007). Furthermore, recent studies using atomic force microscopy showed that filopodia connect fusing myoblasts along the entire length of membrane contact (Kim et al. 2008, Städler et al. 2010).
272
B. Bjerregaard et al.
Fig. 13.3 Live (unpermeabilized) myoblasts induced to fuse and stained with a monoclonal antibody recognizing the extracellular domain of syncytin-1. Images shown were visualized by differential interference contrast (a) and immunofluorescence (b). Note staining of filopodial extensions (exemplified with an arrow) connecting a myoblast (left) with a myotube (right)
13.3 How Does Syncytin-1 Mediate Fusion? Syncytin-1 is synthesized as a glycosylated precursor and cleaved into two mature proteins a surface unit (SU) and a transmembrane unit (TM, Cheynet et al. 2005). These SU and TM subunits are found associated as homotrimers. Previously, it has been shown that the cAMP mediated PKA signaling pathway activates syncytin gene expression by regulating GCMa (transcription factor) activity (Chang et al. 2005, Knerr et al. 2005). This might be via CD9, since overexpression of CD9 leads to increased expression level of GCMa and syncytin-1 (Muroi et al. 2009). A study by Vargas et al. (2008) indicates that syncytin expression might be regulated by tyrosine kinases, because inhibition of protein tyrosine phosphatases caused an increase in syncytin-1 mRNA and induction of cell fusion. It is not known whether syncytin-1 might be part of a larger fusion complex. Thus, future research should be focused on identifying potential interaction partners of syncytin-1. We have initiated a study focusing on the potential interaction of syncytin-1 and the actin cytoskeleton. Preliminary immunofluorescence of fixed permeabilized myoblasts induced to fuse showed a partial colocalization between filamentous actin (F-actin) and syncytin. This is of interest in view of the fact that many types of cell fusions depend upon an intact actin cytoskeleton (reviewed in Chapters 6 and 12, see also the recent study by Nowak et al. 2010). Moreover, the background to the marked upregulation of syncytin-1 synthesis and expression following induction of myoblast fusions deserves further study.
13
The Endogenous Envelope Protein Syncytin Is Involved in Myoblast Fusion
273
13.4 Conclusions and Perspectives Fusogenic endogenous retroviral (ERV) envelope proteins may have been acquired by several different species (humans, mice, and rabbits) to facilitate cell–cell fusions. Our data suggest that syncytin-1 participates in human skeletal muscle formation and indicate a more wide-ranging biological role for this retroviral envelope gene than hitherto suspected. Current data from our laboratory also suggest that syncytin-1 is expressed also during other cell–cell fusion events including fertilization and osteoclast formation.
References Antony JM, van Marle G, Opii W et al (2004) Human endogenous retrovirus glycoproteinmediated induction of redox reactants causes oligodendrocyte death and demyelination. Nat Neurosci 7:1088–1095 Bjerregaard B, Holck S, Christensen I J et al (2006) Syncytin is involved in breast cancerendothelial cell fusions. Cell Mol Life Sci 63:1906–19 Blond J, Lavillette D, Cheynet V et al (2000) An envelope glycoprotein of the human endogenous retrovirus HERV-W is expressed in the human placenta and fuses cells expressing the type D mammalian retrovirus receptor. J Virol 74:3321–3329 Chang C, Chen PT, Chang GD et al (2004) Functional characterization of the placental fusogenic membrane protein syncytin. Biol Reprod 71:1956–1962 Chang CW, Chuang HC, Yu C et al (2005) Stimulation of GCMa transcriptional activity by cyclic AMP/protein kinase A signaling is attributed to CBP-mediated acetylation of GCMa. Mol Cell Biol 25:8401–8414 Charlton CA, Mohler WA, Blau HM (2000) Neural cell adhesion molecule (NCAM) and myoblast fusion. Dev Biol 221:112–119 Charrasse S, Comunale F, Fortier M et al (2007) M-cadherin activates Rac1 GTPase through the Rho-GEF trio during myoblast fusion. Mol Biol Cell 18:1734–1743 Chen EH, Olson EN (2005) Unveiling the mechanisms of cell–cell fusion. Science 308: 369–373 Cheynet V, Ruggieri A, Oriol G et al (2005) Synthesis, assembly and processing of the Env ERVWE1/syncytin human endogenous retroviral envelope. J Virol 79:5585–5593 Cheynet V, Oriol G, Mallet F (2006) Identification of the hASCT2-binding domain of the Env ERVWE1/syncytin-1 fusogenic glycoprotein. Retrovirology 3:41 Doberstein SK, Fetter RD, Mehta AY et al (1997) Genetic analysis of myoblast fusion: blown fuse is required for progression beyond the prefusion complex. J Cell Biol 136:1249–1261 Dupressoir A, Marceau G, Vermochet C et al (2005) Syncytin- A and syncytin-B, two fusogenic placenta-specific murine envelope gtenes of retroviral origin conserved in Muridae. Proc Natl Acad Sci USA 102:725–730 Dupressoir A, Vernochet C, Bawa O et al (2009) Syncytin-A knockout mice demonstrate the critical role in placentation of a fusogenic, endogenous retrovirus-derived, envelope gene. Proc Natl Acad Sci USA 106:12127–12132 Esnault C, Priet S, Ribet D et al (2008) A placenta-specific receptor or the fusogenic, endogenous retrovirus-derived, human syncytin-2. Proc Natl Acad Sci USA. 105:17532–17537 Frendo JL, Olivier D, Cheynet V et al (2003) Direct involvement of HERV-W Env glycoprotein in human trophoblast cell fusion and differentiation. Mol Cell Biol 23:3566–3574 Galbiati F, Volonte D, Engelman JA et al (1999) Targeted down-regulation of caveolin-3 is sufficient to inhibit myotube formation in differentiating C2C12 myoblasts. Transient activation
274
B. Bjerregaard et al.
of p38 mitogen-activated protein kinase is required for induction of caveolin-3 expression and subsequent myotube formation. J Biol Chem 274:30315–30321 Gong R, Huang L, Shi J et al (2007) Syncytin-A mediates the formation of syncytiotrophoblast involved in mouse placental development. Cell Physiol Biochem 20:517–526 Heidmann O, Vernochet C, Dupressoir A et al (2009) Identification of an endogenous retroviral envelope gene with fusogenic activity and placenta-specific expression in the rabbit: a new “syncytin” in a third order of mammals. Retrovirology 6:107 Horsley V, Pavlath GK (2004) Forming a multinucleated cell: molecules that regulate myoblast fusion. Cell Tissues Organs 176:67–78 Horsley V, Jansen KM, Mills ST et al (2003) IL-4 acts as a myoblast recruitment factor during mammalian muscle growth. Cell 113:483–494 Knerr I, Schubert SW, Wich C et al (2005) Stimulation of GCMa and syncytin via cAMP mediated PKA signaling in human trophoblastic cells under normoxic and hypoxic conditions. FEBS Lett 579:3991–3998 Kim SJ, Kim S, Shin H et al (2008) Intercellular interaction observed by atomic force microscopy. Ultramicroscopy 108:1148–1151 Kontarides MI, Eminaga S, Fornaro M et al (2004) SHP-2 positively regulates myogenesis by coupling to the Rho GTPase signaling pathway. Mol Cell Biol 24:5340–5352 Larsen JM, Christensen IJ, Nielsen HJ et al (2009) Syncytin immunoreactivity in colorectal cancer: potential prognostic impact. Cancer Lett 280:44–49 Larsson LI, Holck S, Christensen IJ (2007) Prognostic role of syncytin expression in breast cancer. Hum Pathol 38:726–731 Lavillette D, Marin M, Ruggieri A et al (2000) The envelope glycoprotein of human endogenous retrovirus type W uses a divergent family of amino acid transporters/cell surface receptors. J Virol 76:6442–6452 Mi S, Lee X, Li X et al (2000) Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature 403:785–789 Muroi Y, Sakurai T, Hanashi A et al (2009) CD9 regulates transcription factor GCM1 and ERVWE1 expression through the cAMP/protein kinase A signaling pathway. Reproduction 138:945–951 Nowak SJ, Nahirney PC, Hadjantonakis AK et al (2010) Nap1-mediated actin remodeling is essential for mammalian myoblast fusion. J Cell Sci 122:3282–3293 Oluwole OSA, Yao Y, Conradi S et al (2007) Elevated levels of transcripts encoding a human retroviral envelope protein (syncytin) in muscles from patients with motor neuron disease. Amyotroph Lateral Scler 8:67–72 Oren-Suissa M, Podbilewicz B (2007) Cell fusion during development. Trends Cell Biol 17:537–546 Pavlath GK, Horsley V (2003) Cell fusion in skeletal muscle. Central role of NFATC2 in regulating muscle cell size. Cell Cycle 2:420–425 Sapir A, Avinoam O, Podbilewicz B et al (2008) Viral and developmental cell fusion mechanisms: conservation and divergence. Dev Cell 14:11–21 Schwander M, Leu M, Stumm M et al (2003) Beta-1 integrins regulate myoblast fusion and sarcomere assembly. Dev Cell 4:673–685 Städler B, Blättler TM, Franco-Obregón A (2010) Time-lapse imaging of in vitro myogenesis using atomic force microscopy. J Microsc 237:63–69 Strick R, Ackermann S, Langbein M et al (2007) Proliferation and cell–cell fusion of endometrial carcinoma are induced by the human endogenous retroviral syncytin-1 and regulated by TGFbeta. J Mol Med 85:23–38 Sun Y, Ouyang DY, Pang W et al (2010) Expression of syncytin in leukemia and lymphoma cells. Leuk Res. doi:10.1016/j.leukres.2010.03.016 Tachibana I, Hemler ME (1999) Role of transmembrane 4 superfamily (TM4SF) proteins CD9 and CD81 in muscle cell fusion and myotube maintenance. J Cell Biol 146: 893–904
13
The Endogenous Envelope Protein Syncytin Is Involved in Myoblast Fusion
275
Vargas A, Moreau J, Landry S, LeBellego F, Toufaily C, Rassart E, Lafond J, Barbeau B (2009) Syncytin-2 plays an important role in the fusion of human trophoblast cells. J Mol Biol 392:301–318 Vargas A, Moreau J, Le Bellego F, Lafond J, Barbeau B (2008) Induction of trophoblast cell fusion by a protein tyrosine phosphatase inhibitor. Placenta 29:170–174 Yagami-Hiromasa T, Sato T, Kurisaki T et al (1995) A metalloproteinase-disintegrin participating in myoblast fusion. Nature 377:652–656 Yang JT, Rando TA, Mohler WA et al (1996) Genetic analysis of alpha 4 integrin functions in the development of mouse skeletal muscle. J Cell Biol 135:829–835 Yoon S, Molloy MJ, Wu MP et al (2007) C6ORF32 is upregulated during muscle cell differentiation and induces the formation of cellular filopodia. Dev Biol 301:70–81
Chapter 14
Cell Fusion and Stem Cells Alain Silk, Anne E. Powell, Paige S. Davies, and Melissa H. Wong
Abstract Differentiation, self-renewal and the ability to readily undergo cell fusion are properties of adult and embryonic stem cells. Spontaneous fusion between stem cells, and fusion of stem cells with various differentiated cell types, has been observed in many in vitro and in vivo contexts. Stem cell fusion is implicated in many crucial functions during normal development and is increasingly being harnessed as a tool for regenerative therapies. Experimentally induced fusion between somatic and stem cells forms the basis for our current understanding of nuclear reprogramming. Additionally, the potential fusion of stem cells with cancer cells may have physiologic contributions to aspects of tumor progression and metastasis. Understanding the mechanisms of stem cell fusion might allow manipulation of these processes to affect tissue regeneration, nuclear reprogramming and cancer chemotherapy. In this chapter we consider the functional consequences of stem cell fusion in development, regeneration and disease and postulate how cell fusion might contribute to the advancement of stem cell therapies in regenerative medicine. Keywords Cancer · cell-cell fusion · hematopoietic stem cells · intestinal stem cells · intestine · mesenchymal stem cells · neural stem cells · nuclear reprogramming · regeneration · stem cells Abbreviations ADAM BMDC FAH FC FCM
A Disintegrin And Metalloprotease Bone marrow-derived cell Fumarylacetoacetate hydrolase Founder cell Fusion-competent myoblast
A. Silk (B) Department of Dermatology, Knight Cancer Institute, Oregon Stem Cell Center, Oregon Health and Science University, Portland, OR 97239, USA e-mail:
[email protected] Alain Silk and Anne E. Powell are equally contributed
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_14, C Springer Science+Business Media B.V. 2011
277
278
GFP GMP iPS cell MSC NTBC
A. Silk et al.
Green fluorescent protein Granulocytic macrophage progenitors Induced pluripotent stem cell Mesenchymal stem cell 2-(2-Nitro-4-trifluoromethylbenzoyl)-1,3-cyclohexanedione
Contents 14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 14.1.1 Understanding Stem Cell Biology for Therapeutic Applications . . . . . . . . . . . . . . . 14.1.2 Fusogenicity as a Potential Property of Embryonic and Adult Stem Cells . . . . . . . . . . . . . . . . . . 14.2 Gamete Fusion . . . . . . . . . . . . . . . . . . . . . . . . . 14.2.1 A Historic Perspective of Sperm-Egg Fusion . . . . . . . . 14.2.2 Gamete Cell Adhesion Is Facilitated by ADAM and Integrin Proteins . . . . . . . . . . . . . . . . . . 14.2.3 Tetraspanins as Oocyte Fusion Components . . . . . . . . 14.2.4 Sperm Membrane Fusion Proteins . . . . . . . . . . . . 14.2.5 Relevance of Gamete Fusion to Stem Cell Biology . . . . . 14.3 Myoblast Fusion . . . . . . . . . . . . . . . . . . . . . . . . 14.3.1 A Brief History of Myoblast Fusion . . . . . . . . . . . . 14.3.2 Drosophila as a Model to Study Myoblast Fusion . . . . . . 14.3.3 Zebrafish as a Vertebrate Myoblast Fusion Model . . . . . . 14.3.4 Relevance of Myoblast Fusion Towards Understanding Other Stem Cell Fusion . . . . . . . . . . . . . . . . . . . . 14.4 Cell Fusion with Organ Stem Cells . . . . . . . . . . . . . . . . 14.4.1 Neural Stem Cell Fusion . . . . . . . . . . . . . . . . . 14.4.2 Mesenchymal Stem Cell Fusion . . . . . . . . . . . . . 14.4.3 Intestinal Stem Cell Fusion as a Regenerative Response to Injury . . . . . . . . . . . . . . . . . . . . . . . . 14.4.4 Relevance of Tissue Stem Cell Fusion to Tissue Physiology . 14.5 Fusion of Hematopoietic Progenitors as a Source of Regenerative Repair . . . . . . . . . . . . . . . . . . . . . 14.5.1 Evidence for Hematopoietic Fusion . . . . . . . . . . . . 14.5.2 Hematopoietic Regeneration of Liver Hepatocytes . . . . . 14.5.3 Hematopoietic Regeneration of Heart Myocardium via Cell Fusion . . . . . . . . . . . . . . . . . . . . . 14.6 Cancer Stem Cell Fusion . . . . . . . . . . . . . . . . . . . . 14.6.1 Cancer Stem Cell Hypothesis . . . . . . . . . . . . . . . 14.6.2 Cell Fusion with Cancer Stem Cells . . . . . . . . . . . . 14.6.3 Genomic Instability in Cancer Stem Cell Fusion and Tumor Initiation . . . . . . . . . . . . . . . . . . 14.6.4 Fusion as a Mediator of Cancer Progression . . . . . . . .
. . . . . .
279
. . . . . . . 279 . . . . . . . 280 . . . . . . 281 . . . . . . 282 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . .
. . . .
. . . .
. . . .
283 284 285 285 286 287 287 288
. . 288 . 289 . 289 . 291
. . . . . . . 291 . . . . . . 295 . . . . . . . . . . . . . . . . . .
296 296 297
. . . . . . . . . . . . . . . . . . . . . . . . .
298 299 299 300
. . . . . . . 300 . . . . . . 301
14
Cell Fusion and Stem Cells
14.7 Insight into the Physiologic Fate of Stem Cell Fusion . . . . . . . 14.7.1 Nuclear Reprogramming Within Cell Fusion Hybrid Cells . . 14.7.2 Key Factors that Direct Nuclear Reprogramming . . . . . . 14.7.3 Directionality of Nuclear Reprogramming . . . . . . . . . 14.7.4 Identification of Discrete Factors Important for Nuclear Reprogramming . . . . . . . . . . . . . . . 14.7.5 Lessons from Stem Cell Fusion . . . . . . . . . . . . . . 14.8 Conclusion: The Biological Consequence for Stem Cell Fusigenicity References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
279
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . 305 . 305 . 307 . 309
302 302 303 304
14.1 Introduction 14.1.1 Understanding Stem Cell Biology for Therapeutic Applications A major goal of regenerative medicine is to develop cell-based therapies to effectively treat diseases resulting from specific cell type loss. Potential applications of replacing damaged cells include the treatment of diabetes, amyotrophic lateral sclerosis, deafness, blindness, spinal cord injury, heart disease, liver damage, and instances of chronic inflammation and infertility (Trounson 2009). In each case, one of the most promising current approaches is the creation of patient-specific stem cells induced to differentiate into the cell type needed for repair, followed by a functional transplant into the damaged organ. Deriving cells for tissue regeneration from the same patient in which they will be therapeutically applied circumvents current obstacles of tissue availability and antigen compatibility, which are the primary roadblocks to conventional organ and tissue transplantation. Stem cells are defined by their ability to self-renew and to differentiate into various cell types. Adult and embryonic stem cells share different degrees of differentiation potential; while embryonic stem cells can give rise to all cells of an organism, adult stem cells are generally lineage restricted and can only differentiate into a tissue-specific subset of cells. Hematopoietic stem cells, for example, can differentiate to produce only the many different lineages of blood cells, and neural stem cells only give rise to cells of the brain. Interestingly, this long-standing paradigm of lineage commitment has recently come into question. Various groups have demonstrated differentiation of neural and hematopoietic stem cells into cell types outside of their established lineage commitments and attributed these observations to “transdifferentiation”, the conversion of one cell type into another (Tosh and Slack 2002). Interestingly, these reports have led to subsequent investigations of alternative explanations for this phenomenon and identified the ability of adult and embryonic stem cells to readily undergo fusion in vitro and in vivo, highlighting that some basic biological properties of stem cells remain unclear (Ying et al. 2002). The potential for stem cell fusion to alter cellular identity has fundamental implications for the use of stem cells in cell-replacement therapy and in understanding aspects of
280
A. Silk et al.
cancer progression (Pawelek and Chakraborty 2008). In order to harness stem cell fusion for therapeutic use in a regenerative capacity, it is critical to have a profound understanding of stem cell biology. This includes a re-evaluation of the basic biological properties of stem cells and the exploration of physiological consequences generated from potential novel stem cell functions, such as cell fusion.
14.1.2 Fusogenicity as a Potential Property of Embryonic and Adult Stem Cells The ability for self-renewal and multi-lineage differentiation represents defining characteristics of stem cells; however, these are not the sole properties that distinguish stem cells from their differentiated progeny. One of the more recently appreciated capabilities of stem cells is their relatively high propensity to spontaneously fuse with other stem cells and with a wide variety of terminally differentiated cell types (Terada et al. 2002; Alvarez-Dolado et al. 2003; Vassilopoulos et al. 2003; Rizvi et al. 2006; Pawelek and Chakraborty 2008). Embryonic stem cells co-cultured with adult or embryonic neural progenitor cells undergo spontaneous fusion to produce heterokaryons, binuncleate cells in which each nucleus originates from a different parental cell. Hybrid cell proliferation results in the appearance of synkaryons, mononucleate cells with genetic material supplied by the two different parental cell types. Maintaining stem cell characteristics, these cell fusion hybrids give rise to differentiated lineages such as neurons and contracting myocytes, in vitro (Ying et al. 2002). In vivo, heterotypic fusion (between two different stem cell types) can produce embryonic-neural synkaryons, which readily differentiate into multiple embryonic lineages and contribute substantially to a variety of tissues in chimeric mice (Ying et al. 2002). Embryonic stem cells also readily undergo homotypic fusion (between two cells of the same type) (Ying et al. 2002). These important observations demonstrate the inherent fusogenicity of embryonic stem cells. Fusogenicity as a stem cell property also holds true for lineage-committed multi-potent adult stem cells, as exemplified by hematopoietic stem cells. Heterotypic fusion has been described to occur spontaneously in mice and humans, both in vitro and in vivo, between bone marrow-derived cells (BMDCs) and a number of differentiated cell types, including hepatocytes, Purkinje neurons, cardiomyocytes, embryonic stem cells, intestinal epithelial progenitors, and tumor cells (Gussoni et al. 2002; Terada et al. 2002; Alvarez-Dolado et al. 2003; Rizvi et al. 2006). It is important to note that the ability to fuse is not an exclusive property of stem cells, as evidenced by differentiated cell lineages with fusogenic potential, such as macrophages. For each stem-somatic cell fusion event however, there must be a receptive somatic cell fusion partner that expresses the appropriate fusion machinery. It is possible that, in this case, cell fusion is mediated primarily by the stem or progenitor cell. Limited evidence supports this notion, in that some somatic cells cannot undergo on their own, or even when co-cultured with stem cells (AlvarezDolado et al. 2003). This is consistent with the proposal that fusion efficiency is a
14
Cell Fusion and Stem Cells
281
functional property of both fusion partners and that differentiation of the stem cell may reduce its fusogenic potential. An unequal contribution to fusion initiation between partners is also apparent in fusion events during muscle development. Muscle-forming stem cells (myoblasts) undergo multiple rounds of fusion to produce the syncytial cells of skeletal muscle. Sustained cell fusion between myoblasts and myotubes (multinucleated, differentiated muscle cells) is required in adult animals for maintenance of muscle tissue (Rochlin et al. 2009). During muscle development and maintenance, far more fusion occurs between muscle stem cells and syncytia than between multinucleated cells, and expression of fusion promoting factors is asymmetrical between myoblasts and syncitial cells (Horsley et al. 2003). Therefore, in the case where fusion is a prelude to differentiation, there is variable expression of fusion genes and an altered contribution to fusion from differentiated cells relative to progenitors, further suggesting a link between fusogencitiy and stemness. Given the relatively recent description of spontaneous stem cell fusion, significant additional work is needed to examine the fusogenic efficiencies of various stem and somatic cell types, the mechanisms underlying stem cell fusion, and the physiological relevance of this novel stem cell property. Understanding the fundamental mechanisms driving stem cell fusion provides the potential for manipulating fusion for therapeutic benefit in tissue regeneration and potentially in interrupting disease progression. In this chapter we consider the physiologic consequences of stem cell fusion in development, regeneration and disease and translate it where possible to potential implications for stem cell therapies in regenerative medicine.
14.2 Gamete Fusion The most well-known instance of cell fusion in mammals is undoubtedly the union of gametes, sperm and oocyte, to generate the first totipotent embryonic cell of a new organism. While gametes themselves are not stem cells, fusion of gametes has been one of the longest studied examples of cell fusion in biology. Additionally, the important role of gamete fusion in generating stem cells makes an overview of this process integral to the consideration of stem cell fusion in general. Fertilization is a multi-step process which requires that the sperm efficiently penetrate the layer of cumulus cells surrounding the oocyte, digest the zona-pellucida (a glycoprotein extracellular matrix encasing the egg), and finally bind to and fuse with the oocyte plasma membrane (Primakoff and Myles 2002). This is a significantly more complex series of events than is observed in most other examples of cell fusion. For instance, sperm acquisition of membrane fusion-competence is dependent on the activation of a specialized sperm organelle, the acrosome, and the subsequent release of enzymes that facilitate breakdown of the zona-pellucida. However, this aspect of cell fusion is likely to be restricted specifically to sperm–egg interactions, while general aspects of direct cell–cell interaction and components of the membrane fusion machinery are most likely conserved among other systems and therefore
282
A. Silk et al.
represent greater relevance to a general understanding of fusion events involving embryonic and adult stem cells. Here we discuss mechanisms that function during the late stages of fertilization, with a view towards providing a background for a conceptual understanding regarding stem cell fusion. Further, we highlight experimental themes used to identify mechanisms of gamete fusion, which might provide a basis for the investigation into adult and embryonic stem cell fusion. Due in part to technical difficulties involved in studying membrane fusion in gametes, few of the molecular components involved in this process have been identified; however, the following brief historical summary of gamete fusion should provide a context in which to anticipate future directions in the study of stem cell fusion.
14.2.1 A Historic Perspective of Sperm-Egg Fusion One early intriguing hypothesis regarding the process of gamete membrane fusion was that the sperm entered the egg via phagocytosis, as was proposed based on sperm-egg size asymmetry (Szollosi and Ris 1961). However, electron microscopybased investigations in both the sea urchin (Franklin 1965) and rat (Szollosi and Ris 1961) convincingly demonstrated that fertilization did not proceed by this method, but rather by fusion of the sperm and egg membranes. This work was contemporary with studies that demonstrated fusion of organelle lipid bilayers in exocytosis and in secretory processing in the Golgi (Jahn and Sudhof 1999), and initiated the search for components of the sperm-egg membrane fusion machinery. Current models indicate that most biological fusion of lipid bilayers occurs by the same general series of steps: initial fusion of the two closest outer membranes to create a hemifusion intermediate which is resolved by inner membrane fusion (Martens and McMahon 2008). Perhaps it is surprising that no membrane components involved in organelle fusion have yet been implicated in gamete membrane fusion. Indeed, in the diverse known examples of membrane fusion, there appear to be a large number of distinct components operating to mediate the same process (Martens and McMahon 2008). The unique challenges that exist in studying gametes accentuate the difficulty to identify the specific fusion components of these cells. Perhaps it is not surprising that only within the past 15 years have the molecular identities of components involved in gamete fusion begun to be elucidated (Snell and White 1996). Early work pursued the hypothesis that gamete fusion might merely involve a single protein that possessed binding and fusion properties present on one cell and would interact with a receptor on a fusion partner cell. In fact, this view of cell fusion turns out to be overly simplistic and it is now thought that no instances of mammalian cell fusion proceed in this manner. Rather, it is believed that fusion is a multi-step process in which functionally distinct proteins are involved in cellular adhesion and subsequent membrane fusion (Yanagimachi 1994). Importantly, the involvement of cell–cell interactions as a critical precursor to membrane fusion emerged from and is exemplified by early studies delving into the mechanisms of fertilization.
14
Cell Fusion and Stem Cells
283
14.2.2 Gamete Cell Adhesion Is Facilitated by ADAM and Integrin Proteins The prototypic candidate for a mammalian fusion binding protein was identified more than 20 years ago as the sperm membrane protein complex, fertilin α/β. In an elegant study of in vitro fertilization designed to identify fusion proteins by characterizing the functional blocking potential of monoclonal antibodies generated against fusion-proteins on the sperm surface (Primakoff et al. 1987), one chain of a tight heterodimeric protein complex was identified (Blobel et al. 1990). Subsequent cloning of the associated cDNAs identified fertilins α and β, predicted type-I integral membrane proteins. These proteins, also known as ADAM1 and ADAM2 respectively, are the founding members of the ADAM (A Disintegrin And Metalloprotease) family of membrane proteins (Evans 2001). There are currently 29 known mammalian ADAM type-I transmembrane proteins with short cytoplasmic regions and extracellular “disintegrin” and metalloprotease domains. Support for the viral-like fusion mechanism grew with the structural characterization of this gameteassociated protein and identification of a sequence similar to viral ‘fusion peptides’ within the α-chain (Blobel et al. 1992). Additionally, a predicted integrin-binding ‘disintegrin’ domain at the end of the cytoplasmic tail of fertilin β provided further support (Blobel et al. 1992). Given the presence of these two putative functional domains in the fertilin heterodimer and the well-known function of mammalian integrins in control of cell adhesion (Hynes 2002), a potential dual role for fertilin in sperm-egg binding and gamete membrane fusion was immediately suggested, providing the basis for identifying the molecular contribution of fertilin to fertilization and, importantly, for identifying the oocyte integrin binding partners. Multiple integrin genes encode the 18 α and 8 β subunits, which heterodimerize in 24 known combinations (Hynes 2002). A promising initial study identified the expression of multiple integrin subunits in mouse oocytes (including α5, α6, αv, β1 and β3), localized α6β1 and αvβ3 integrins to the cell surface, and demonstrated that antibodies specific to the α6 subunit could block fertilization in vitro (Almeida et al. 1995). These early results hinted at a simple fusogenic ligandreceptor basis for gamete binding and fusion, yet it soon became clear that a virus-like model was overly simplistic. Genetic disruption of α6-, α3-, β3-, and β5- integrins had no effect on fertility in mice, and in vitro fertilization of oocytes deficient in β1-integrin also failed to identify a contribution of this protein in gamete adhesion or membrane fusion (Kaji and Kudo 2004; Primakoff and Myles 2007). Moreover, combinatorial inhibition of multiple integrins with blocking antibodies to β3 or αv subunits and β1-deficient oocytes also failed to prevent fertilization in vitro (He et al. 2003), casting doubt on the contribution of integrins to the fertilization process. By contrast, genetic disruption of the putative integrin ligand fertilin β, or the related sperm protein ADAM3 (cyritestin) resulted in a dramatic reduction in male fertility, although this resulted primarily from a loss of sperm-oocyte adhesion with little effect on membrane fusion (Cho et al. 1998; Shamsadin et al. 1999). This comprehensive series of knockout studies has informed current models for ADAM– integrin interactions in fertilization, which propose a minimally functional role for
284
A. Silk et al.
integrins, and hold that ADAM proteins are involved in sperm-egg binding but not membrane fusion. This series of experiments highlights that studies of stem cell fusion mechanisms should be designed to distinguish between cell adhesion and membrane fusion. Further, we suggest that while antibody-mediated approaches can be successful for identifying novel fusion molecules, the many caveats inherent to antibody-based inhibition require alternative approaches, such as genetic analyses, when possible to confirm relevant findings.
14.2.3 Tetraspanins as Oocyte Fusion Components The most promising recent candidate for an oocyte fusion component was initially identified serendipitously following the recognition of fertility defects in a mouse harboring a homozygous deletion of the integrin-associated tetraspanin CD9 (Kaji et al. 2000; Le Naour et al. 2000; Miyado et al. 2000). Two other members of this protein family, CD81 and CD151, have since been implicated in oocyte membrane fusion (Evans 2001). Tetraspanins are small four-transmembrane domain proteins that protrude a short distance (4–5 nm) into the extracellular space. Proteins of this family (32 members in mammals) participate in a wide variety of cellular processes, including cellular morphology, invasion, adhesion and signaling, but have been understudied relative to other abundant membrane proteins (Hemler 2005). CD9 was initially recognized as an integrin-associated protein, and was implicated in contributing to ADAM–integrin interactions (Chen et al. 1999), however it was the interest in potential hematologic functions of CD9 that drove the creation of a CD9 knockout mouse and ultimately revealed this protein as the first oocyte molecule involved in bona-fide gamete membrane fusion (Le Naour et al. 2000). Interestingly, CD9−/− mice have no obvious phenotype other than female infertility, likely noted during routine breeding to maintain the knockout allele. Despite normal mating behavior and reproductive anatomy, CD9 deficient females have severely reduced litter sizes and delays in conception relative to wild-type or heterozygous animals; there are no obvious effects on male fertility (Kaji et al. 2000; Le Naour et al. 2000). In vivo evidence along with in vitro fertilization assays subsequently demonstrated that CD9−/− oocytes were normal for sperm binding, but deficient in subsequent membrane fusion (Kaji et al. 2000; Le Naour et al. 2000). The precise molecular function of CD9 in membrane fusion, and the role of two other oocyte tetraspanins, CD81 and CD151, remains to be characterized. One interesting report suggests that CD9-containing vesicles released by oocytes interact with sperm to license membrane fusion (Miyado et al. 2008); however this has been directly contradicted by results from another group (Gupta et al. 2009b). Considering the multifunctional potential and diverse intermolecular interactions of other known tetraspanins, understanding the precise role of CD9 in gamete fusion will be a challenge, and it will be interesting to follow new developments in this field (Hemler 2005). Although its precise function remains unclear, the initial identification of CD9 in gamete fusion is an excellent example of the intelligent pursuit of an astute observation. While we should not depend on serendipity to reveal the identity of
14
Cell Fusion and Stem Cells
285
additional gamete or stem cell fusion proteins, open-minded observation in concert with direct experimental analyses should provide insight into novel cell fusion mechanisms.
14.2.4 Sperm Membrane Fusion Proteins Follow-up experiments involving proteins identified by antibody-mediated disruption have recently identified a compelling new candidate for a component of the gamete membrane fusion machinery within sperm. In the vein of general fertilization research, this discovery began with the generation of monoclonal antibodies and spanned 15 years of exploration that eventually led to the demonstration that the sperm protein Izumo is the target of a fertilization blocking antibody (Okabe et al. 1987, 1988). More recently, the definitive function of Izumo in sperm-egg fusion through genetic disruption has been identified (Inoue et al. 2005). The only reported overt phenotype of Izumo-deficient mice is sterility, which can be rescued by testis-specific transgenic re-expression of Izumo (Inoue et al. 2005). The role of Izumo in membrane fusion has been determined using in vitro assays which demonstrate that Izumo−/− sperm are able to bind normally to oocytes but are completely incapable of undergoing membrane fusion (Inoue et al. 2005; Rubinstein et al. 2006). The precise function of Izumo in membrane fusion is currently unknown, but its identification paralogs within this novel protein family and the characterization of additional Izumo-interacting proteins suggest that it does not function alone. Furthermore, it has been reported that recombinant Izumo protein does not directly interact with oocytes or prevent gamete fusion in vitro, indicating that it may not directly mediate membrane fusion (Ellerman et al. 2009). Rather, this protein may function to assemble, stabilize, or regulate a protein complex or lipid microdomain that modulates the function of other membrane-localized fusion molecules (Ellerman et al. 2009).
14.2.5 Relevance of Gamete Fusion to Stem Cell Biology The historical discovery of gamete fusion components highlights a recurrent experimental approach which might prove effective in identifying aspects of the functional machinery of stem cell fusion. This strategy includes (1) the generation of monoclonal antibodies against cell surface proteins; (2) subcellular localization of these proteins in successive stages of the cell fusion process; (3) screening these antibodies to identify function-blocking capability during in vitro fusion; (4) cloning of associated genes and knockout studies in mice to definitively identify gene function. Despite the technical limitations inherent to the study of gametes, these approaches in combination with results from other fields have successfully identified two protein families: ADAMs (ADAM 1, 2, 3) and tetraspanins (CD9, CD81, CD151) as well as many additional important molecular components such as Izumo, all of which participate in the final steps of fertilization and are prime candidates for cell
286
A. Silk et al.
Fig. 14.1 Fusion in early development. Individual gametes, sperm and oocyte, come together to fuse and form a zygote in the earliest stage of development. In a similar manner, syncytiotrophoblasts undergo repeated fusion events to produce an extensive cellular syncitium critical for placental function
fusion components in other systems (Sutovsky 2009). Furthermore, these studies have revealed that gamete fusion, and probably fusion of most mammalian stem cells, involves multiple independent proteins functioning sequentially to mediate adhesion followed by membrane fusion. Exploring the expression of ADAM and tetraspanin family members and identifying homologs should serve as a key starting point for the identification of molecules involved in other types of stem cell fusion. Genetic screening, which is limited in the gamete system, should be applied where possible to develop a more comprehensive approach over antibody-mediated disruption for the identification of novel components of the fusion machinery in adult and embryonic stem cells. Notably, cell fusion exemplified by gamete fusion, described here, and trophoblast fusion, not described, are functionally important mechanisms that are essential steps in development (Fig. 14.1). Clearly mechanisms elucidated in these settings have the potential to inform studies directed toward exploring fusion in other stem cell populations.
14.3 Myoblast Fusion Myoblasts are true adult stem cells; they undergo self renewing divisions and differentiate into mature multinucleated myocytes. The myoblast differentiation process occurs primarily as a consequence of cellular fusion, in which myoblasts fuse both with each other and with differentiated muscle cells to produce mature skeletal muscle fibers. The relative ease of myoblast culture, wide range of possible experimental manipulations, and the multiple fusion events that occur to generate each mature myocyte have allowed a significantly greater understanding of the fusion mechanisms in this system compared to that in gametes. Notably, the identification of gamete fusion components has relied primarily on functional disruption at the protein level, and focused on molecules present at the cell surface. Myoblast fusion studies, by contrast, have employed forward genetic screens to identify a wide range of secreted, cell surface, and intracellular components involved in mediating cell fusion (Richardson et al. 2008). Extensive details of myoblast fusion mechanisms
14
Cell Fusion and Stem Cells
287
are beyond the scope of this chapter, and are reviewed thoroughly in other sections of this book (Chapters 6, 12 and 13). Here we aim to briefly highlight important experimental strategies used in the study of muscle stem cell fusion and discuss how they might apply in the development of approaches to identify fusion pathways in other adult and embryonic stem cell populations.
14.3.1 A Brief History of Myoblast Fusion Although muscle fibers have long been appreciated to be composed of multinucleated cells, how these cells were created remained unclear. Opposing theories proposed cell fusion and amitosis (nuclear division in the absence of cell division) as underlying mechanisms. In the 1950s and 1960s, at the same time that gamete membrane fusion was beginning to be appreciated, the application of longterm time-lapse microscopy and advances in dissociated muscle cell culture (Capers 1960), combined with elegant studies in chimeric mice (Mintz and Baker 1967), definitively established that mature muscle fibers were formed from the fusion of multiple precursor cells (Fig. 14.2). Since that time, a variety of experimental paradigms have been employed in order to discover the identity of elements involved in the fusion apparatus.
14.3.2 Drosophila as a Model to Study Myoblast Fusion The experimental system that has yielded the greatest understanding of the molecular components of cell fusion in myoblasts has been the study of muscle formation in Drosophila melanogaster. This model system has harnessed the powerful genetics of the fly for the identification of multiple mechanistic aspects of the fusion process (Horsley and Pavlath 2004). Just as in mammals, muscle formation in Drosophila relies on precursor cell fusion. Muscles of the body wall in the Drosophila embryo arise from fusion between cells of two myoblast populations: founder cells (FCs) and fusion-competent myoblasts (FCMs). Over a 5.5 h period during embryogenesis, FCs fuse with one or two FCMs to generate bi- or tri-nucleate muscle
Fig. 14.2 Fusion in organ development. Myoblasts fuse both with each other and with differentiated muscle cells to produce mature skeletal muscle fibers. The myoblast is designated as the striped cell and the differentiated cell the solid cell
288
A. Silk et al.
progenitors. These multinucleated cells recruit additional FCMs for further fusion events to generate mature muscle cells containing between 3 and 25 nuclei (Taylor 2002). Defects in muscle formation are predicted to yield embryonic lethality, and therefore screens for fusion genes have targeted embryonic lethal mutants for analysis of disrupted muscle development. For example, one group visually examined body wall muscle organization of embryonic-lethal mutants that were generated in a transposon-based screen of a small region of the third chromosome (Rushton et al. 1995). Another effort used muscle-specific antibody staining to re-examine mutants that had been previously characterized for defects in motoneuron axon guidance (Doberstein et al. 1997), and a more recent screen examined muscle cell organization in embryonic lethal EMS-mutants generated in a muscle-specific green fluorescent protein (GFP)-expressing background (Chen and Olson 2001). These screens illustrate how steady advances in cell labeling techniques, including gross morphological analysis, antibody staining, and live-imaging of GFP-expressing muscle have allowed progressively more specific screening criterion to be applied for the identification of myoblast fusion mutants in Drosophila.
14.3.3 Zebrafish as a Vertebrate Myoblast Fusion Model Despite the large number of identified molecules participating in Drosophila myoblast fusion, only recently has the conservation of these components in other organisms been identified. The zebrafish protein Kirrel is homologous to the Drosophila protein duf and is required in myoblast fusion underlying the development of the fast-twitch muscle (Srinivas et al. 2007). Like duf, Kirrel is an Ig containing cell surface protein, hinting that an immunoglobulin fold may be characteristic of cell fusion components. Interestingly, several mammalian Kirrel homologs are known, although none are yet characterized to participate in cell fusion. However considering that muscle multinucleation is an evolutionarily conserved process, it is likely that additional homologs of Drosophila myoblast fusion genes will be discovered in other systems. The zebrafish presents as an exceptionally useful model system for the identification of novel vertebrate molecular components of myoblast fusion. Transgenic fish can be readily generated, early fish embryos are transparent and relatively easy to screen for defects in labeled populations of cells, and zebrafish have been extensively used for large scale genetic screens to identify genes involved in a wide range of developmental processes.
14.3.4 Relevance of Myoblast Fusion Towards Understanding Other Stem Cell Fusion Formation of multinucleated syncytia is a notable aspect of muscle fusion which differentiates it from other examples of cell fusion. In flies and mammals, generation of mature multinucleated muscle cells is a two step process involving
14
Cell Fusion and Stem Cells
289
fusion of mononucleated cells to form a nascent myotube, followed by additional fusion of mononucleated cells to create the final syncytium. These two stages of fusion are regulated independently, as evidenced by the existence of mutants where initial fusion events to generate binucleates are normal but additional fusion events fail (Horsley and Pavlath 2004). Regarding extrapolating mechanisms for general stem cell fusion in which the final fusion product is binucleate, priority should be given to the examination of components involved in binucleate-generating myoblast fusion events. For an in-depth understanding of myoblast fusion mechanisms, the reader is referred to the relevant Chapters (6, 12 and 13) in this book, in which it is made clear that cell fusion is more than simply a combination of cell adhesion and membrane fusion. Signaling, migration, recognition, transcriptional changes and cytoskeletal rearrangements are all aspects of the cell biological alterations accompanying the fusion of two cells. Future studies of stem cell fusion mechanisms, making use of whole organisms or perhaps carried out in vitro using genome-wide RNA interference and high-throughput imaging technologies, hold promise to uncover genes involved in these myriad processes. At this time, the degree to which fusion mechanisms will be conserved between organisms, or even between cell types is unclear; however a comprehensive picture of stem cell fusion pathways is critical for establishing a greater understanding of stem cell biology and to aid in the therapeutic application of stem cell fusion.
14.4 Cell Fusion with Organ Stem Cells The ability for a quiescent tissue stem cell population to be stimulated for cell cycle re-entry in response to tissue injury is an absolutely essential function during tissue regeneration. Largely, the mechanisms underlying this process have been assumed to be driven from the mesenchymal compartment and involve upregulation of critical developmental cell signaling pathways known to direct proliferation. While this is most likely a fundamental cue of tissue regeneration, an emerging role for cell fusion between somatic cells or tissue stem cells and circulating bone marrow-derived or mesenchymal cells provides an intriguing additional mechanism for tissue repair after injury. This section will examine the observation of cell fusion among organ stem cells and provide a potential role for this mechanism in tissue regeneration.
14.4.1 Neural Stem Cell Fusion For over 100 years, it has generally been well accepted that the adult brain displays extremely little regenerative capacity. However, as our understanding of neuronal biology expands, so do the discoveries that brain cells have the potential to be replaced after tissue injury. Interestingly, acquisition of new neurons is apparent within the adult brain of monkeys and rats (Gould and Gross 2002). Further,
290
A. Silk et al.
generation of developmentally functioning stem cells, glial cells, in response to injury has recently been described (Berninger et al. 2007). In support of a limited regenerative capacity within the brain, defined stem cell populations in the mouse forebrain have been identified both in vitro and in vivo. However these stem cells, which normally give rise to many differentiated neuronal cell types (Reynolds and Weiss 1992; Gritti et al. 1996), appear to be capable of very limited self-repair (McKay 1997; Morrison et al. 1997). A mechanism underlying stimulation of this repair process has been the focus of intense study, unfortunately with little progress. However, recent description that neuronal stem cells can undergo cell fusion and subsequent nuclear reprogramming toward a more pluripotent state provides an intriguing alternate mechanism for promoting neuronal regeneration (Ying et al. 2002). A study designed to examine the lineage commitment of newly identified neuronal stem cells using a bone marrow transplantation assay, reported conclusions that were initially misleading. The results interpreted this neuronal stem cell population to have a great plasticity and the ability to differentiate into lineages outside of the neuronal identity (Bjornson et al. 1999). In these experiments, β-galactosidaseexpressing adult neural stem cells were transplanted into sublethally-irradiated mice and a survey of the hematopoietic lineages revealed that lymphoid and the myeloid lineages expressed the donor marker, lacZ. Initial interpretation of these results suggested that neuronal stem cells could “transdifferentiate” into hematopoietic cell lineages. However, subsequent studies revealed that the process of expanding these adult neural stem cells in preparation for transplantation involved a co-culture step with embryonic stem cells. Careful analysis of the expanded neuronal stem cell populations revealed that these co-cultured cells exhibited gene expression patterns of both neural stem cells (expression of β-galactosidase) and embryonic stem cells (based upon a transgenic antibiotic resistance gene), providing surprising evidence that they were actually cell fusion hybrids between both cell types present in the culture system (Ying et al. 2002). Karyotype analysis further confirmed this observation. Because of this initial controversy for in vivo cell fusion, additional evidence was required to reinforce this observation. Using an elegant approach employing genetic recombination to demonstrate cell fusion, Alvarez-Dolado and colleagues co-cultured neurospheres from mice harboring a Cre recombinase reporter (Mao et al. 1999) with bone marrow-derived stromal cells expressing Cre recombinase (Alvarez-Dolado et al. 2003). In this scenario, reporter expression would only be detected upon the co-localization of the Cre reporter and Cre enzyme, an event that could only occur upon fusion of the two cell types. Consistent with previous observations for cell fusion, this genetic approach supported cell fusion as Cre reporter expression was detected the neuronal cells. Interestingly, using this same genetic approach cell fusion was not observed between co-cultured neurospheres and Creexpressing fibroblasts, suggesting a certain cellular or environmental requirement for fusion. Neurospheres or neuronal stem cells are not the only cell type within the brain competent to fuse. Reports of cell fusion between the Purkinje neurons, a cell type within the cerebellum, and circulating BMDCs illustrated both low levels of fusion after tissue injury (Weimann et al. 2003a, b) and enhanced levels in the
14
Cell Fusion and Stem Cells
291
presence of inflammation (Johansson et al. 2008). Collectively, these studies suggest that cell fusion is one mechanism by which neural stem cells may respond to injury.
14.4.2 Mesenchymal Stem Cell Fusion Stem cells derived from the bone marrow stroma are termed multi-potent adult progenitor cells or mesenchymal stem cells (MSCs), and represent another stem cell population that possesses the ability to fuse with and contribute to different somatic tissues. Isolated MSCs can be coaxed in vitro to differentiate down various cell lineages including endothelium, endoderm, neural ectoderm and hepatocyte-like cells (Jiang et al. 2002; Reyes et al. 2002; Schwartz et al. 2002). Intriguingly, individually isolated MSCs transplanted into either blastocysts or immune-compromised mice can give rise to multiple tissues within the resulting chimeric animals (Okumura et al. 2009). While contribution of MSC donor cells to various tissues may suggest that MSCs have a unique property to “transdifferentiate” into various lineages, it does not rule out the possibility that these cells are fusogenic and contribute to chimeric tissue by fusing with the inner cell mass of the blastocyst. Interestingly, like the Purkinje neuron, the context of tissue injury enhanced MSC engraftment into somatic tissues, potentially implicating a cell fusion mechanism (Ferrari et al. 1998; Kotton et al. 2001; Okamoto et al. 2002). Supporting a role for cell fusion among MSCs, in vitro studies revealed that MSCs co-cultured with injured small airway epithelial cells resulted in phenotypically distinct, multi-nucleated small airway epithelial cells (Spees et al. 2003). Further, time-lapse microscopy demonstrated some cell fusion hybrids also underwent nuclear fusion while other cells displayed abnormal polyploidy. Together, these studies demonstrate that MSCs have the ability to undergo fusion and potentially impact tissue regeneration after injury, similar to neuronal stem cells.
14.4.3 Intestinal Stem Cell Fusion as a Regenerative Response to Injury Unlike the bone marrow-derived MSCs or the relatively quiescent neuronal stem cells, the intestinal epithelial stem cell has a relatively high proliferative index for a stem cell population, with its division occurring as estimated one time every 24 h (Barker et al. 2007; Sato et al. 2009). As such, the intestinal epithelial stem cell provides the basis for continual renewal and repopulation of the intestinal epithelium. We have demonstrated that the intestinal stem cell fuses with circulating BMDCs in a bone marrow transplantation assay in response to gamma-irradiation induced epithelial injury (Rizvi et al. 2006). By tracking co-expression of a donor marker, green fluorescent protein (GFP), and a recipient marker (β-galactosidase or Y-chromosome) from transplanted mice, cell fusion within the epithelial compartment was readily appreciated by confocal microscopy (Fig. 14.3a–c). In this system,
292
A. Silk et al.
Fig. 14.3 Bone marrow-derived cells fuse with intestinal epithelium. (a–c) Tissue section from a β-galactosidase-expressing recipient transplanted with GFP-expressing bone marrow and analyzed by confocal microscopy for co-expression. White line indicates the epithelial/mesenchymal boundary. White asterisk denotes a villus lacking GFP expression. (a) β-gal (red) is uniformly expressed in the intestinal epithelium as detected with antibodies to β-gal and Cy-5-conjugated secondary antibody. (b) GFP expression (green) on the same tissue section as in panel (a) detected by direct fluorescence. (c) Merge of β-gal- and GFP-stained tissue depicting co-localization of both donor and host markers (yellow), indicating fusion (Rizvi et al. 2006; copyright 2006 by the National Academy of Sciences). (d–f) Genetic approach to demonstrate bone marrow/epithelial cell fusion and genetic reprogramming. (d) Schematic diagram of transplantation scheme. Whole bone marrow from mice expressing Cre recombinase driven by the intestinal epithelial-specific Villin promoter (VilCre) was transplanted into recipient mice that were homozygous for floxed Apc. (e) Resulting intestinal phenotypes were observed in transplanted mouse intestine by wholemount analysis as polyps in the distal small intestine. (f) To confirm that the phenotype was the result of recombination at the Apc allele, PCR analysis of isolated epithelium from recipient mice using primers that specifically detect the recombined Apc allele was performed. The 258 bp band was present in transplanted DSI and colon samples, indicating cell fusion by activation of Cre-recombinase (Davies et al. 2009). Bar = 25 μm
14
Cell Fusion and Stem Cells
293
cell fusion hybrids expressing GFP were detected in all of the differentiated lineages of the mouse small intestinal epithelium, providing evidence that the intestinal epithelial fusogenic partner was a stem or multi-potent progenitor cell (Rizvi et al. 2006). Corroborating this observation, GFP-expressing epithelial cells persisted for more than 17 months after transplantation, further implicating a long-lived progenitor cell as the epithelial target for fusion. Because the intestinal epithelium is rapidly renewing, cell fusion with differentiated villus cells as the epithelial target would not be appreciated in a long-term analysis. Similar to the skepticism in which stem cells could undergo cell fusion, the burden of proof for intestinal cell fusion in response to injury required multiple approaches to demonstrate this process, including one technique harnessing the genetic power of the Cre-lox recombination system. In this setting, bone marrow from a mouse harboring an intestinal-specific promoter driving Cre recombinase was transplanted into a recipient mouse harboring floxed alleles for a tumor suppressor gene. Because the Cre enzyme is not expressed within the isolated bone marrow cells, it reasons that only upon cell fusion with the epithelial compartment would the Cre recombinase enzyme become active and the recombined floxed allele be detected (Fig. 14.3d–f); (Davies et al. 2009)). Interestingly, intestinal cell fusion hybrids are mononucleated (unpublished observation), in contrast to other systems where cell fusion results in binucleated cells (Weimann et al. 2003a, b). One possible explanation for the existence of mononucleated cell fusion products within the intestinal epithelium is that the intestinal microenvironment actively selects for cells which posses a normal DNA content. This hypothesis is supported by the rationale that the stem cell progeny continue to divide for multiple rounds and must execute this process with high fidelity to avoid the propagation of errors that could impede cytokinesis. Reduction division represents one potential mechanism by which a cell fusion hybrid may reduce its DNA content. Indeed, there is precedent for unequal reduction division after fusion in the liver (Duncan et al. 2009). Therefore, it is intriguing to speculate that while cell fusion may be a mechanism that facilitates epithelial regeneration after injury, the process of unequal reduction division may contribute to tumorigenesis in the solid organs. While these studies clearly demonstrated the ability of the intestinal stem cell to fuse with circulating BMDCs, mediating factors or environmental cues are just now being identified. Surprisingly, epithelial stem cell fusion was also detected under homeostatic conditions, using a non-damage model (Davies et al. 2009). Surgical joining of two congenic mice, parabiosis, allows for shared blood supply and the ability to track tagged BMDCs in the partner mouse without subjection to gamma irradiation. Low, but significantly detectable levels of epithelial stem cell fusion were observed in the parabiotic mice, suggesting that the rapidly renewing intestinal epithelium may support a baseline level of cell fusion during homeostatic conditions. Interestingly, we showed that cell fusion is enhanced in the epithelium of mice prone to colorectal tumors (Rizvi et al. 2006) and occurs at a higher incidence compared to wild-type mice (Fig. 14.4). These data collectively suggest that microenvironmental factors characteristic of tumors may act to mediate intestinal epithelial stem cell fusion. It is well documented that intestinal tumors are characterized by both increased epithelial proliferation and an active inflammatory response (Moser et al. 1990;
294
A. Silk et al.
Fig. 14.4 Cell fusion in tissue regeneration. Bone marrow-derived cells (solid cell) have been shown to fuse with stem cells of various solid organs (striped cell), such as the intestine (epithelium), the brain (neuron), and the liver (hepatocyte) in response to injury and yield a regenerative capacity
Karin and Greten 2005; Reya and Clevers 2005; Karin 2008). Further, it is known that exposure to gamma irradiation results in an initial apoptotic response followed by proliferation of the crypt epithelial cells (Potten et al. 1990), corresponding to a stimulation of dormant stem cells to divide in a Wnt-signaling mediated fashion (Davies et al. 2008). These observations provided the basis for exploring the role of proliferation and inflammation in the cell fusion process. The extent of cell fusion reported in the continually proliferative intestinal epithelium (Rizvi et al. 2006; Davies et al. 2009) is higher than other in organs that do not self-renew or do so at a much slower rate (Johansson et al. 2008; Nygren et al. 2008), supporting the idea that a proliferative state may be required for the fusion process. Moreover, data from human bone marrow transplant patients also suggest that donor-derived intestinal epithelia arise from proliferative progenitors (Okamoto et al. 2006). Examination of intestinal epithelial cell fusion hybrids for the expression of the proliferative cell marker, Ki67, revealed that a subset of fusion hybrids reside in the proliferative portion of the intestinal crypt. Examination of the influence of epithelial proliferation on cell fusion in a genetic mouse model of inducible Wnt signaling activation revealed a rapid and significant stimulation of cell fusion (Davies et al. 2009). While these observations support epithelial proliferation as an important mediator for cell fusion, whether active engagement in the cell cycle or merely the state of differentiation (i.e. the progenitor state) is the important factor remains currently unknown. Inflammation, a second important factor in promoting cell fusion was first described in mediating fusion of Purkinje neurons (Johansson et al. 2008). Likewise, stimulation of intestinal inflammation using a genetic approach, as examined in the IL-10−/− mouse model for inflammatory bowel disease, or in a chemically induced colitis with administration of dextran sodium sulfate, resulted in an appreciable increase in intestinal epithelial cell fusion (Davies et al. 2009). Further, fusion could be suppressed in these models by administering an effective anti-inflammatory drug during the experimental course. The fact that inflammation and epithelial proliferation have been shown to be key mediators of stem cell fusion highlights the
14
Cell Fusion and Stem Cells
295
important potential physiologic impact of cell fusion on tissue repair, regeneration or homeostasis. Significantly, these two factors are often stimulated during disease processes, emphasizing the need to elucidate the physiologic impact that cell fusion may have on a tissue in the face of disease.
14.4.4 Relevance of Tissue Stem Cell Fusion to Tissue Physiology Although cell fusion among neural and intestinal progenitor cells and MSCs has now been well documented, the underlying physiologic role for cell fusion and more importantly the physiologic relevance for these fusion events remain unknown. It has been speculated that cell fusion can occur in response to injury (Fig. 14.5) and evidence that inflammatory and proliferative mediators drive this process clearly support this potential role. However, whether fusion is an active process for stimulation of tissue stem cells to rapidly enter the cell cycle in order to mediate repair, or if it is a by-product of the microenvironment is not known. Regardless, if this is an active process, it is important to determine if cell fusion hybrids have any long-term detrimental effects on the tissue by increasing its susceptibility for acquiring mutations that may lead to disease.
Fig. 14.5 Cell fusion is enhanced in tumors. (a-b) An intestinal adenoma from a GFPbone marrow transplanted tumor bearing mouse harboring ubiquitous β-galactosidase expression (ApcMin/+; ROSA26). (a) Direct fluorescence for GFP detection (green) illustrates both epithelial (bracket) and mesenchymal GFP-positive cells. (b) Confocal microscopy of tumors with detection of β-galactosidase (red) and GFP (green) depict co-expression in the epithelial compartment, indicative of cell fusion. White dashed line delineates epithelial-mesenchymal boundary. Bar = 25 μm. (c) Quantification of cell fusion in transplanted ApcMin/+ mice compared to wild-type transplanted mice
296
A. Silk et al.
14.5 Fusion of Hematopoietic Progenitors as a Source of Regenerative Repair Cells derived from the blood compartment have been demonstrated to incorporate into injured tissues and function to enhance tissue regeneration (Ferrari et al. 1998; Lagasse et al. 2000). However, the mechanism by which this process impacts physiologic repair and function is controversial. It was originally speculated that plasticity of blood cell progenitors allowed them to “transdifferentiate” into non-blood cell lineages and participate in repair of injured tissues. As such, initial transplantation experiments engrafting genetically tagged bone marrow into recipient mice attributed the presence of donor-derived cells in injured tissue to “transdifferentiation” (Brazelton et al. 2000; Jackson et al. 2001; Krause et al. 2001; Orlic et al. 2001a). These early conclusions were quickly dispelled when a closer evaluation of the donor-derived cells in the brain and liver revealed that they were tetraploid and harbored both donor and recipient markers (Wang et al. 2003; Weimann et al. 2003b). Thus, cell fusion was proposed as an alternative mechanism by which blood-derived cells may participate in tissue regeneration. Since cellular fusion plays an essential role in many biological functions including fertilization, muscle formation, development of bone, maintenance of placenta, and aspects of the innate immune response (Chen and Olson 2005), it is intriguing to speculate that fusion of blood cells with somatic cells contributes to the regenerative response to disease in a variety of organs. Supporting this hypothesis, cells from the bone marrow fuse in vitro with embryonic stem cells resulting in fusion hybrids that harbor DNA from both the bone marrow and the embryonic stem cell parental lines (Terada et al. 2002). Although the physiologic relevance of this observation remains to be clarified, it is possible that cell fusion results in the delivery of undamaged genetic material to cells that have a compromised genome and in this way critically contributes to the regenerative response to tissue injury. This section aims to describe the historical evidence of hematopoietic cell fusion and the important regenerative capacity that this biological process can have on solid organs.
14.5.1 Evidence for Hematopoietic Fusion Proper regulation of the hematopoietic system is critical to the survival of all solid organs. Beyond the functions of oxygen delivery and immune surveillance, stem cells within the blood have recently been demonstrated to incorporate into injured tissues, taking on the identity of the differentiated cells within that tissue. It has long been speculated that blood cell progenitors harbor a specialized plasticity to “transdifferentiate” into other tissue types under appropriate conditions (reviewed in Orkin and Zon (2002)) and that this plasticity might be critical to repairing injured tissue. Thus, “transdifferentiation” originally thought to explain the initial observations that bone marrow transplantation following irradiation results in both the engraftment of the blood compartment with donor marrow and incorporation of BMDCs into various solid tissues.
14
Cell Fusion and Stem Cells
297
Evidence that fusion underlies hematopoietic stem cell plasticity comes, in part, from experimental co-culture of bone marrow and embryonic stem cells (Terada et al. 2002). BMDCs from a green fluorescent protein (GFP)-expressing mouse cultured with embryonic stem cells results in the appearance of GFP-expressing embryonic stem-like cells. These GFP-positive embryonic cells contain DNA from both donors and are able to differentiate into a variety of cell types, suggesting that bone marrow cell plasticity might result from cell fusion. Additional in vivo experiments using labeled donor and recipient cells indicate that bone marrow cells engraft into solid tissues primarily by cell fusion (Alvarez-Dolado et al. 2003; Vassilopoulos et al. 2003; Rizvi et al. 2006). Indeed, hematopoietic cells have been described to fuse with many other cell types including hepatocytes, Purkinje neurons, skeletal muscle, and cardiac myocytes (Alvarez-Dolado et al. 2003). These observations have formed the basis for a number of experimental models in which blood cell transplants have been used for the repair of damaged solid tissues.
14.5.2 Hematopoietic Regeneration of Liver Hepatocytes Fusion between BMDCs and hepatocytes is the best described example of hematopoietic cell fusion with somatic cells in the context of tissue repair. Homozygous disruption of the fumarylacetoacetate hydrolase gene (FAH) in mice results in a model of human tyrosinemia type I, a debilitating metabolic disorder in which tyrosine breakdown by hepatocytes is defective. FAH−/− mice are dependent on regular administration of the drug 2-(2-Nitro-4-trifluoromethylbenzoyl)-1,3cyclohexanedione (NTBC), which prevents toxic accumulation of byproducts from failed tyrosine metabolism (Grompe et al. 1995). Transplantation of wildtype FAH+/+ bone marrow into lethally irradiated FAH−/− mice rescues the dependence on NTBC (Vassilopoulos et al. 2003; Wang et al. 2003). This rescue occurs concomitant with the appearance of FAH-positive hepatocytes containing both donor and recipient DNA which arise by fusion of wildtype transplanted bone marrow cells and endogenous FAH−/− hepatocytes (Vassilopoulos et al. 2003; Wang et al. 2003), elegantly demonstrating the therapeutic application of stem cell fusion for the cure of hepatic disease. Although bone marrow stem cells are the presumed fusion donors in this system, it remains unclear whether differentiated blood cells may also participate in hepatocyte fusion, or perhaps comprise the predominant fusogenic population. Transplant of single hematopoietic stem cells from lymphoid-deficient Rag1−/− mice into lethally irradiated FAH−/− recipients allows complete rescue, via fusion, for the FAH deficiency; indicating that B and T cells are dispensable for fusionmediated liver repair (Willenbring et al. 2004). By contrast, transplant of myeloid lineage-committed granulocytic macrophage progenitors (GMPs) or differentiated macrophages into sublethally irradiated FAH−/− mice is sufficient rescue tyrosinemia in these animals (Willenbring et al. 2004). Together, this data indicates that differentiated blood cells can fuse with somatic cells and contribute to tissue
298
A. Silk et al.
regeneration. The true relative contribution of stem and myeloid committed blood cells to the process of liver regeneration remains unknown.
14.5.3 Hematopoietic Regeneration of Heart Myocardium via Cell Fusion In addition to the liver, cardiac tissue has shown potential promise as a target for cell fusion-based repair. This was originally recognized from experiments in which cytokine mediated mobilization of stem cells from the bone marrow was observed to promote tissue repair and regeneration in a mouse model of myocardial-infarction (Orlic et al. 2001b). Further, direct injection of GFP transgenic hematopoietic stem cells into zones of myocardial infarction results in the appearance of GFP-positive myocardium covering a majority of the damaged region (Orlic et al. 2001a). While these studies did not directly demonstrate cell fusion as the mechanism of blood progenitor engraftment into injured myocardium, they supplement other strong evidence demonstrating the myogenic potential of hematopoietic stem cells (Ferrari et al. 1998). Cell fusion as a mechanism for myocardial engraftment was independently demonstrated using the Cre/loxP system, in which myocardial infarction was induced in mice carrying a Cre reporter transgene and transplanted with Crerecombinase expressing bone marrow. This genetic approach demonstrated that bone marrow-mediated myocardial engraftment occurred through cell fusion (Alvarez-Dolado et al. 2003). In support of these results, others have observed that human peripheral blood progenitor cells fuse with mouse cardiomyocytes in vivo in a xenotransplant model, although “transdifferentiation” also contributes to donorderived myocardium in this case (Zhang et al. 2004). Interestingly, the mechanism of human-mouse cell fusion following myocardial injury appears to involve the vascular cell adhesion molecule-1 (VCAM-1), a protein known to mediate chorioallantoic fusion and placenta formation in early development (Gurtner et al. 1995), providing the first insights into the molecular basis of fusion-mediated blood cell plasticity. In addition to substantial evidence demonstrating blood-derived cell fusion with cardiac tissue in mice, it is now clear that BMDCs engraft into human cardiac tissue as well. Chromosome analysis of heart tissue in four female recipients who had undergone gender-mismatched bone marrow transplant revealed the presence of the Y-chromosome in cardiomyocytes (Deb et al. 2003). Since these patients did not exhibit any signs of cardiomyopathy, these data might indicate that bone marrow cell engraftment into relatively healthy heart tissue in humans. In addition, Y-chromosome was also detected in liver hepatocytes and skeletal muscle of these women, providing further support for bone marrow stem cell plasticity in humans. While this study provides compelling evidence for fusion within the heart, it remains controversial among the field. A similar analysis of gender-mismatched heart transplants demonstrated non-cardiac donor cell engraftment into cardiac tissue with no evidence of fusion (Muller et al. 2002).
14
Cell Fusion and Stem Cells
299
Despite substantial evidence for the incorporation of blood cells into heart muscle in humans and mice, and strong support for fusion as the underlying mechanism, the functional contribution of BMDCs to myocardial maintenance and repair is unknown. The level of circulating bone marrow stem cells is not increased in response to either myocardial infarction or cryo-injury, suggesting that endogenous bone marrow stem cells contribute minimally to direct heart regeneration (Nygren et al. 2004). Further, the hematopoietic progenitors that engraft into damaged myocardium appear to retain expression of blood cell markers and fail to express cardiac muscle specific genes. Interestingly, bona fide blood cell fusion-derived cardiomyocytes can be detected at low frequency preferentially in non-infarcted areas of a damaged heart (Nygren et al. 2004). Therefore, fusion is one mechanism by which blood stem cells incorporate into myocardium, although the extent to which they contribute directly to tissue repair after injury and the therapeutic utility of this approach is unknown. Might fusion provide temporary cellular support during the early stages of nascent tissue generation following heart damage? Could this process be enhanced by stem cell therapies or modulation of cell fusion mechanisms to provide better recovery in cases of heart injury? Resolving these questions may require the construction of an experimental system in which heart repair can be affected exclusively through non-cardiac-cell engraftment into the heart, and subsequent analysis of these cells for evidence of fusion with endogenous cardiomyocytes. The potential for stem cell fusion mediated therapies in heart regeneration has sound experimental precedents but awaits more rigorous testing.
14.6 Cancer Stem Cell Fusion The concept of stem cell fusion in the context of cancer provides an intriguing potential mechanism that could explain the means by which distinct cancer cell behaviors characteristic of advanced disease are acquired. The re-emergence of the cancer stem cell theory has provided the foundation for speculation on how this cell population may facilitate disease progression, resistance to treatment and emergence of recurrent disease (Al-Hajj and Clarke 2004). In this section we will explore the potential role of cell fusion among cancer stem cell populations and provide insight into potential physiologic relevance for this process.
14.6.1 Cancer Stem Cell Hypothesis Cancer stem cells are proposed to be a specialized population of cancer cells that harbor stem cell-like properties. In this fashion, they are capable of self-renewal and propagation of differentiated cancer cell lineages leading to rapid tumor growth and heterogeneity within the tumor. Further, like endogenous stem cells, cancer stem cells have been shown to be resistant to therapeutic regimens including certain lines
300
A. Silk et al.
of chemotherapy and radiation (Gupta et al. 2009a). Cancer stem cells, or as they are alternatively called, tumor-initiating cells, have been isolated from a variety of blood and solid tumors based upon cell surface marker expression. The cells are further validated in a xenograft assay to show they are capable of recapitulating the original tumor. The existence of the cancer stem cell is controversial because these cells appear to be more abundant in some types of tumors and have been observed to be greatly expanded relative to their normal endogenous counterparts, the tissue stem cell. However, a recent thorough and in-depth examination of the cancer stem cell literature suggests that the disparate cell frequencies reflect both the various cancer types and the experimental host employed to characterize their stem behavior, thus strongly supporting their existence in the literature (Gupta et al. 2009a). Needless to say, acquisition of additional detrimental mutations within this stem cell population, as with accumulation of mutations within a tissue stem cell population, has the potential to exponentially propagate among the cancer cells and within the cancer stem cell population. In this fashion, cells are selected for that harbor acquired mutations providing a survival advantage.
14.6.2 Cell Fusion with Cancer Stem Cells While there is no direct evidence that cell fusion occurs with cancer stem cells, there are several lines of evidence that cell fusion occurs within tumors. First, cell fusion of tumor cells has been reported in cell culture and suggested to contribute to a more aggressive tumor phenotype (Larizza et al. 1984). Next, tumor cell fusion has been reported in a number of in vivo mouse systems, including intestinal tumors and in melanoma (Chakraborty et al. 2000; Rizvi et al. 2006). Finally, in humans, cell fusion has been observed in renal cell carcinoma or cervical lymphadenopathy in patients who had received bone marrow transplantation (Chakraborty et al. 2004; Yilmaz et al. 2005). While these in vitro and in vivo observations strongly support the existence of cell fusion in cancer, the important question regarding its physiologic relevance in tumorigenesis remains unanswered.
14.6.3 Genomic Instability in Cancer Stem Cell Fusion and Tumor Initiation Even in the absence of definitive evidence that cell fusion with cancer stem cells directly impacts initiation or progression of tumors, one can clearly appreciate the potential impact of the genetic alterations previously described to occur in cell fusion hybrids on the tumorigenic process. In many instances where BMDCs fuse with somatic or stem cells, the resulting hybrids are aneuploid or polyploid (Alvarez-Dolado et al. 2003; Wang et al. 2003; Weimann et al. 2003b). Abnormal chromosomal content is a hallmark of many cancer cells and this loss or gain of chromosomes is directly related to the tumorigenicity of a given population (Duesberg and Li 2003; Vogelstein and Kinzler 2004). In many cases, changes
14
Cell Fusion and Stem Cells
301
in chromosomal complement affect the cell’s ability to properly grow and divide. Although the majority of cell fusion hybrids have been reported to result in stable heterokaryons, or engage in nuclear fusion followed by a reduction division process to decrease chromosome content (Duncan et al. 2009), it is possible that an imbalance in chromosome number can lead to genomic instability. Cancer cells are characterized by genomic instability (Charames and Bapat 2003), and although a direct link between this instability and tumor initiation has long been speculated, it lacks definitive proof. Further, stem cell fusion results in gene activation or silencing. DNA demythylation as well as chromosomal and telomere reprogramming in stem cell fusion hybrids have been clearly demonstrated (Zhang et al. 2007), providing important implications of cell fusion among cancer stem cells to potentially impact tumor initiation or overall tumor biology.
14.6.4 Fusion as a Mediator of Cancer Progression Supporting the idea that cell fusion participates in cancer progression, in vitro studies from over 30 years ago demonstrated that cancer cells are more fusogenic than non-tumor cells and acquire more aggressive tumorigenic phenotypes by fusing with BMDCs (Fig. 14.6) (Goldenberg 1968; Mekler 1971). Further, many malignant and metastatic cell types share similar genetic expression profiles with migratory cells from the myeloid blood lineage (Pawelek 2005), and this bone marrowderived population has previously been implicated in cell fusion (Willenbring et al. 2004). In theory, myeloid cells could convey their migratory properties to tumorigenic cells through the mechanism of cell fusion. Interestingly, the macrophage is one such myeloid cell type that is fusogenic in nature, as it self-fuses in the
Fig. 14.6 Fusion of cancer stem cells. Both in vitro and in vivo observations strongly support the existence of cell fusion in the setting of cancer. The potential impact of the genetic alterations or migratory behaviors obtained in a cell fusion hybrid could expand mechanistic insight on both tumor initiation and metastatic spread. Solid cell represents a non-stem cell partner, striped cell represents a cancer stem cell
302
A. Silk et al.
foreign body giant cell response of the innate immune system. It is possible that tumor-associated macrophages may exploit their fusogenic capability and fuse with cancer stem cells, thereby generating tumor-macrophage hybrids that have phenotypic characteristics of a migratory macrophage along with the general transcriptome of the tumor stem cell. Indeed, malignant metastatic tumor cells acquire the capacity to migrate to distant sites and evade the immune system. This potential mechanism may be one explanation for a resulting epithelial-to-mesenchymal transition (EMT) characteristic of metastatic behavior. Despite the possibility that cell fusion with cancer stem cells may have important impact on tumor biology, the field is still nascent and requires continued exploration to validate this intriguing observation.
14.7 Insight into the Physiologic Fate of Stem Cell Fusion While cell fusion has been demonstrated in a number of systems, both in humans and animals, the most important remaining question revolves around the long-term physiologic impact from generation of cell fusion hybrids. Whether these cells integrate into normal surrounding tissue without incident or whether the heterokaryons (or hybrids) contribute to disease is not known. In vivo studies examining the potential physiologic fate of cell fusion hybrids are difficult to conduct. However, in the interim, a focus on characterizing the transcriptome is more readily achievable. Initial insight into how the nuclei of two parental cells are modified after fusion provides important information regarding the potential of these cells. Here we explore the effects of nuclear reprogramming within cell fusion hybrid cells and demonstrate how lessons learned within these experiments have impacted progress in generating induced pluripotent stem (iPS) cells.
14.7.1 Nuclear Reprogramming Within Cell Fusion Hybrid Cells Fusion between somatic cells and less differentiation cell partners has been described in numerous systems including the brain, intestine, hematopoietic system, and muscle (Bonde et al. 2010; Ying et al. 2002; Nygren et al. 2004; Cowan et al. 2005; Rizvi et al. 2006; Yu et al. 2006). This idea has recently re-emerged as a potentially intriguing mechanism to engineer a more stem-like transcriptome within the resulting cell fusion hybrid. Studies designed to explore the possibility of “transdifferentiation” elegantly revealed nuclear reprogramming in spontaneously generated cell fusion hybrids of neuronal precursors (harboring both a puromycin resistance gene and an Oct4-green fluorescent protein (GFP) transgene) co-cultured with embryonic stem cells (expressing a puromycin resistant cassette) (Ying et al. 2002). In this investigation, cell fusion was identified by the surprising observation that puromycin resistant co-cultured cells expressed GFP, representing gene expression driven off a promoter specifically expressed in germline or pluripotent cells and
14
Cell Fusion and Stem Cells
303
silenced in the central nervous system (Ying et al. 2002). Despite being tetraploid, the resulting cell fusion hybrids maintained pluripotency morphologically in culture and functionally, giving rise to chimeric mice when injected into blastocysts. This exciting observation has relevant impact on tissue regeneration or reprogramming on multiple levels. Importantly, stem cell fusion is an intriguing mechanism for mediating tissue repair. Studies examining heart muscle regeneration in the mouse after infarct suggest that fusion between myoblasts and circulating BMDC may facilitate the repair process (Nygren et al. 2004). While the topic of cell fusion within the human heart as a functional mechanism for regeneration remains controversial (Orlic et al. 2002; Kajstura et al. 2005), it is intriguing to speculate that fusion within a progenitor populations of specific tissues may result in functional regeneration to expand downstream differentiated lineages. Supporting this observation, we have shown that circulating BMDCs fuse with intestinal progenitor cells in response to gammairradiation induced epithelial injury (Rizvi et al. 2006). Although we have not yet transcriptionally defined these long-lived cell fusion hybrids, we demonstrated that they are multi-potent based on the fact that the intestine becomes populated with multi-lineage progeny from the cell fusion event. In addition, we have definitively illustrated that cell fusion is increased in the presence of an inflammatory microenvironment and when the intestinal epithelium is hyperproliferative (Davies et al. 2009), lending circumstantial evidence that cell fusion may participate in intestinal tissue regeneration. In this capacity, it is possible that BMDC fusion with the tissue progenitor population results in reprogramming toward a more stem-like state, leading to stimulation and expansion of the stem cell population and thereby facilitating rapid regeneration of the injured tissue.
14.7.2 Key Factors that Direct Nuclear Reprogramming The key factor in stem cell fusion is the ability of the progenitor cell to reprogram the less differentiated fusion partner. While this would undoubtedly have high impact on endogenous regeneration, the mechanism can also be harnessed for tissue engineering in regenerative medicine. Yu and colleagues demonstrated that isolated CD45+/CD33+/ myeloperoxidase+ myeloid progenitor cells from an Oct4-GFP transgenic mouse co-cultured with undifferentiated human embryonic stem cells resulted cell fusion hybrids expressed ES cell-specific surface antigens and marker gene expression (Yu et al. 2006). These results corroborated the initial observation of cell fusion and nuclear reprogramming in co-cultured systems (Nygren et al. 2004), as well as demonstrated the ability for the cell fusion hybrid to express donor GFP and lose myeloid antigen expression (Yu et al. 2006). These observations were similarly confirmed using a genetic approach to illustrate cell fusion and reprogramming by harnessing the bacterial Cre-lox technology (Bonde et al. 2010; Davies et al. 2009). Using a bone marrow transplant assay, we established that whole bone marrow cells isolated from a transgenic mouse harboring an intestinal-specific promoter (Villin) upstream of Cre recombinase did not express
304
A. Silk et al.
the Cre transgene until the bone marrow cell fused with the intestinal epithelium. In this experimental paradigm, the recipient mouse expressed floxed alleles of an oncogene implicated in intestinal tumor initiation, Apc. Successful recombination of the Apc allele in the epithelium was apparent by the observation of intestinal polyps and elegantly illustrated that the bone marrow cell genome had been functionally reprogrammed toward that of the less differentiated fusion partner supporting an intestinal lineage fate (Fig. 14.3d–f) (Davies et al. 2009).
14.7.3 Directionality of Nuclear Reprogramming Nuclear reprogramming of the somatic genome toward a more undifferentiated state clearly has important implications for tissue engineering and regenerative medicine. Exciting demonstrations of fusion between stem cells and differentiated somatic cells support this potential. The fusion between human ES cells and human fibroblasts resulted in reprogramming of the fibroblast genome to that of an embryonic state (Cowan et al. 2005). Interestingly, the same study demonstrated that the embryonic stem cell genome appears to be dominant, as a genome-wide analysis of the resulting cell fusion hybrids revealed transcriptional activity, allele-specific gene expression, and DNA methylation patterns mimicked an embryonic state (Cowan et al. 2005). While it has largely been thought that reprogramming may occur in a unidirectional fashion, favoring the pluripotent fusion partner at the expense of the differentiated somatic cell partner (Silva et al. 2006), recent data examining reprogramming in the resulting fusion products from embryonic carcinoma cells and neural stem cells showed otherwise (Do et al. 2009). Elegant experiments that cleverly tracked an embryonically expressed gene that is not expressed in neuronal stem cells, Xist, revealed that Xist gene expression was reprogrammed toward the somatic cell influence (Do et al. 2009). One caveat with this experimental paradigm is that both experimental cell populations existed in stem cell states, blurring the line of directionality. However, these results suggest that additional distinct factors may play an important role in directional effects of nuclear reprogramming. Indeed, transcriptional reprogramming analysis of stable heterokaryons generated by in vitro fusion of mouse muscle cells with human primary neonatal keratinocytes revealed the non-muscle nuclei were capable of activating muscle gene transcription in response to diffusible cytoplasmic muscle-specific factors (Zhang et al. 2007). In this experimental design generated heterokaryons were not engaged in the cell cycle, therefore nuclear DNA methylation must be driven by soluble cytoplasmic factors, supporting the notion that nuclear reprogramming is an active rather than passive mechanism. Cell fusion (or cell hybridization) studies have demonstrated that discrete stem cell factors can erase the developmental programming of a differentiated cell, resulting in reversion of the somatic cell to a pluripotent state (Tada et al. 2001). This is clearly illustrated in experiments examining ES cell fusion with adult thymocytes. In-depth analysis of T cell receptor and immunoglobin DNA rearrangements confirm that cell fusion did occur between the two parental cell lines. However, the
14
Cell Fusion and Stem Cells
305
inactivated X-chromosome of the female-derived thymocyte adopted some characteristics of the active X-chromosome (Tada et al. 2001), strongly supporting the idea that a stem cell has the ability to reset the epigenome of somatic cells. Although epigenetic reprogramming of the somatic nuclei has been clearly described, the underlying mechanism for this process is largely unknown.
14.7.4 Identification of Discrete Factors Important for Nuclear Reprogramming Recent studies designed to identify key factors that mediate nuclear reprogramming after stem cell fusion have led to exciting advances with translational application. The homeodoman protein Nanog, was observed to be upregulated in fusion products of ES and neural stem cells, in culture (Silva et al. 2006). Interestingly, Nanog is a transcription factor that functions in concert with Oct4 and Sox2 to maintain the pluripotency of embryonic stem cells (Silva et al. 2006). The culture experiments also revealed elevation of both Oct4 and Sox2, suggesting active reprogramming of the neural stem cell nucleus (Silva et al. 2006). Importantly, this cell fusion study revealed a key role for Nanog to operate in conjunction with other stem cell machinery in order to reset the differentiated epigenome. Stem cell fusion studies have clearly provided significant insights into potential mechanisms for in vivo regeneration and, importantly, they have set the foundation to guide somatic cell reprogramming for regenerative medicine. Spontaneous in vitro cell fusion and subsequent nuclear reprogramming has led to the identification of key nuclear factors critical for driving reversion toward a pluripotent state within the somatic-stem cell heterokaryon. In this context, cell fusion is the forerunner of induced pluripotent stem (iPS) cells which have recently emerged in the field to provide a promising avenue for regenerative medicine. Briefly, iPS cells were first produced in the laboratory of Shinji Yamanaka in 2006 in mouse cells by introducing four factors into differentiated fibroblasts, Oct4, Sox2, Klf4, and cMyc (Takahashi et al. 2007). Excitingly, the differentiated fibroblasts transformed into a pluripotent state and were able to execute developmental programs associated with embryonic stem cells. This process has been recapitulated in human cells using similar transcription factors, OCT4, SOX2, NANOG, and LIN28 (Yu et al. 2007), as well as with neuronal stem cells (Silva et al. 2008). Although the translational potential of iPS cells is tremendous, a number of obstacles remain that must be overcome before this technology can effectively be harnessed for human therapeutic purposes.
14.7.5 Lessons from Stem Cell Fusion Studies characterizing stem cell fusion hybrids have contributed to the success nuclear reprogramming essential for generation of iPS cells. Lessons from in vitro cell fusion as well as iPS studies will in turn provide insight into successful
306
A. Silk et al.
understanding the contribution of in vivo cell fusion for repair of tissues after damage from injury or disease. Thus, understanding the endogenous function of in vivo cell fusion in response to injury may also be a therapeutically applicable option. Therefore in order to harness this in vivo potential, several key factors must be investigated. First, local mediators of in vivo stem cell fusion must be defined. Evidence that inflammation mediates the process has been demonstrated in Purkinje neuron fusion as well as with intestinal epithelial progenitor cells (Johansson et al. 2008; Davies et al. 2009). However, whether the inflammatory microenvironment is crucial for recruitment of fusogenic circulating BMDCs, such as we have shown in the intestine (Fig. 14.7) (Davies et al. 2009)), or if it establishes important concentrations of cytokines that drive the fusion process is unknown. While little work has been performed to identify potentially important cytokines, it is likely that both scenarios contribute. Further, we have shown that stimulation of epithelial proliferation also enhances the in vivo fusion process (Davies et al. 2009). However, whether the proliferative status of the fusogenic partner or the level of progenitor state is the important mediator remains to be determined. The observation that cell fusion occurs with fully differentiated somatic cells may support the conclusion that a stem or progenitor status is irrelevant (Weimann et al. 2003b). However, the caveat with this interpretation is that the fusion partner is unidentified and could potentially be a progenitor within the bone marrow. Second, the underlying molecular mechanism that drives in vivo cell fusion has not been investigated. While molecular mechanisms for the fusogenic macrophage have been thoroughly examined (Vignery 2005), those for stem cell fusion machinery have not. Although it is possible that stem cells are capable of harnessing macrophage fusion machinery, it is more likely that a unique process is involved based on the fact that in vivo stem cell fusion appears to be relatively rare in non-self renewing tissue. Finally, determining whether stem cell fusion precedes nuclear fusion is controversial and may have important impact on subsequent physiologic effects of the fusion process. To date, in vivo cell fusion resulting in tissue regeneration has been reported
Fig. 14.7 Bone marrow cells home to the intestine under inflammatory conditions. (a) Schematic diagram of transplantation. Bone marrow from a GFP-expressing mouse was transplanted into an IL-10−/− recipient mouse, a well-established model for intestinal inflammation. (b–c) Tissue section of the large intestine from a recipient IL-10−/− mouse depicting GFP-expressing blood cells (green) recruited to the mesenchyme surrounding the epithelium under inflammatory conditions. (c) Nuclei are stained with Hoechst (blue). White line denotes epithelial-mesenchymal boundary. Bar = 25 μm
14
Cell Fusion and Stem Cells
307
to yield heterokaryons in the brain, liver and skeletal muscle (Wang et al. 2003; Doyonnas et al. 2004; Johansson et al. 2008). Interestingly, we have observed that in the rapidly renewing intestinal epithelium, stem cell/BMDC fusion hybrids contain a normal DNA content (unpublished observation). Together these data imply that cell fusion in a role for regenerative potential does not require reduction to a single nucleus in certain tissues. However, in some cases, a mechanism may exist to reduce the number of chromosomes to a more “normal” quantity. Regardless, the most critical aspect of the potential for in vivo stem cell fusion mediating tissue repair revolves around the long-term physiologic impact that the fusion hybrids impart on tissue homeostasis. As such, little investigation has occurred in the way of examining their potential to disrupt normal organ function. In addition, while some have speculated that cell fusion may provide the basis for disease initiation such as cancer (see Chapter 16: Cancer Host Cell Fusion), it is currently unclear which cellular and molecular mechanisms may play a role in this process.
14.8 Conclusion: The Biological Consequence for Stem Cell Fusigenicity There are two possible explanations for the relatively high degree of fusigenicity displayed by stem cells. In some types of stem cells, myoblasts for example, fusion is clearly a biologically important function for tissue development. In other stem cell populations not known to fuse under normal physiological conditions, the fusion process may be the result of misregulation of a cellular program by environmental factors. In this latter respect, fusion mechanisms utilized by ES cells or developmentally fusogenic stem cells might be upregulated in these other stem cell populations The first totipotent stem cell of all sexually reproducing multi-cellular organisms arises as the result of fusion between two gametes. In development, the cellular product of this initial fusion reaction undergoes multiple self-renewing divisions to from a growing blastocyst. The first differentiated lineage produced from a blastocyst is the extraembryonic tissue-generating trophectoderm. Within the trophectoderm lineage, trophoblast stem cells give rise to syncytiotrophoblasts which undergo repeated fusion events to produce an extensive cellular syncytium critical for placental function (Simmons and Cross 2005). Thus, the immediate precursors and early terminally differentiated products of embryonic stem cells are highly fusigenic. It is possible that retention of a gamete fusion mechanism or precocious activation of syncytiotrophoblast fusion pathways may contribute in part to the fusion properties of pluripotent embryonic cells or even multi-potent adult stem cells when they are activated to facilitate regeneration or aberrantly activated in disease. There are three primary roles of fusion involving adult and ES cells: first, it functions as a necessary aspect of normal development; second, fusion serves as a tissue maintenance mechanism during homeostasis and can be harnessed for regeneration of some damaged tissues; and third, fusion may serve as a potential component of
308
A. Silk et al.
Fig. 14.8 Gradient of cell fusion. Cell fusion occurs at various levels of development, tissue regeneration and in disease. How the underlying mechanism for fusion in each of these related contexts are similar remain to be established. Because there are similarities between development, regeneration and disease, exemplified by activation of key cell signaling pathways, this cartoon depicts a relationship between fusion in these three physiologic contexts. Further, the state of the stem cell is represented as a gradient from totipotent (dark blue) to pluripotent or multi-potent (light blue) states. Lessons learned from the many contexts of cell fusion in development, regeneration, and disease will provide a foundation for future work in understanding the molecular mechanisms of cell fusion and may provide insight into how they can be enhanced or inhibited to facilitate a beneficial physiologic state
cellular transformation and/or metastasis in cancer (Fig. 14.8). While the role of cell fusion during development is well defined and accepted, its role in regeneration after injury, in homeostatic tissue maintenance and in disease pathogenesis is more speculative and controversial. Therefore, definitively establishing the physiologic impact of cell fusion hybrids within a regenerative tissue or in disease remains a top priority. Already, important insights into characteristics of cell fusion hybrids may guide the discovery of their physiologic importance. It is known that stem cell genomes exist in a relatively open chromatin confirmation in which more genes are in an expression-permissive status than exists in differentiated cells (Turner 2008; GasparMaia et al. 2009). The epigenetic state of the stem cell genome may permit for aberrant transient or low level expression of fusion-pathway genes, sufficient to allow for rare instances of cell fusion. It is also important to consider that although fusion has only been demonstrated to be an important biological property of a subset of adult stem cells, the high degree of fusion observed in various embryonic and adult stem cell populations suggests that many stem cell types may use
14
Cell Fusion and Stem Cells
309
fusion as a mechanism to revert to a more pluripotent state. For example, cell fusion between ES cells and neural stem cells results in fusion hybrids that can access transcriptome patterns of a more pluripotent state (Do et al. 2009). Additionally, cell fusion between transformed cell lines and myeloid cells leads to more aggressive cancer phenotypes in culture, suggesting that fusion results in alterations of the transcriptome (Pawelek 2000). Supporting this idea, genomic analysis performed on in vitro-generated cell fusion hybrids indicated that they retained a donor-specific subset of transcriptional markers (Palermo et al. 2009). Despite these analyses, it is still unclear how and to what extent fusion hybrids are transcriptionally unique. It is clear that unveiling the distinctions between the cell fusion hybrid transcriptome and that of normal unfused tissue will shed light on the physiologic significance of stem cell fusion, both in a regenerative and tumorigenic context. The relationship between developmental, regenerative and disease states could be perceived as a gradient exhibiting decreasing plasticity (i.e. a gradient from totipotent to pluripotent or multipotent stem cells), as well as a decreasing gradient of regulated processes (proposed relationship depicted in Fig. 14.8). Intriguingly, development, regeneration and disease are known to harness similar cell signaling pathways and mechanisms. It is possible that fusogenicity of adult tissue stem cells may also harness similar mechanisms defined for fusion that occur during developmental processes and are repeated during disease. Although definitive underlying mechanisms for fusion remain to be established, it is intriguing that effective targeting of this process has the potential to enhance fusion for regenerative biology and inhibiting the process may be beneficial in a disease setting. Collectively, these unresolved questions promise to fuel the field of stem cell fusion biology for the future and hold promise for harnessing this knowledge for effective therapeutic design.
References Al-Hajj M, Clarke MF (2004) Self-renewal and solid tumor stem cells. Oncogene 23:7274–7282 Almeida EA, Huovila AP, Sutherland AE et al (1995) Mouse egg integrin alpha 6 beta 1 functions as a sperm receptor. Cell 81:1095–1104 Alvarez-Dolado M, Pardal R, Garcia-Verdugo JM et al (2003) Fusion of bone-marrow-derived cells with Purkinje neurons, cardiomyocytes and hepatocytes. Nature 425:968–973 Barker N, van Es JH, Kuipers J et al (2007) Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 449:1003–1007 Berninger B, Costa MR, Koch U et al (2007) Functional properties of neurons derived from in vitro reprogrammed postnatal astroglia. J Neurosci 27:8654–8664 Bjornson CR, Rietze RL, Reynolds BA et al (1999) Turning brain into blood: a hematopoietic fate adopted by adult neural stem cells in vivo. Science 283:534–537 Blobel CP, Myles DG, Primakoff P et al (1990) Proteolytic processing of a protein involved in sperm-egg fusion correlates with acquisition of fertilization competence. J Cell Biol 111:69–78 Blobel CP, Wolfsberg TG, Turck CW et al (1992) A potential fusion peptide and an integrin ligand domain in a protein active in sperm-egg fusion. Nature 356:248–252 Bonde S, Pedram M, Stultz R et al (2010) Cell fusion of bone marrow cells and somatic cell reprogramming by embryonic stem cells. FASEB J 24:364–373 Brazelton TR, Rossi FM, Keshet GI et al (2000) From marrow to brain: expression of neuronal phenotypes in adult mice. Science 290:1775–1779
310
A. Silk et al.
Capers CR (1960) Multinucleation of skeletal muscle in vitro. J Biophys Biochem Cytol 7:559–566 Chakraborty A, Lazova R, Davies S et al (2004) Donor DNA in a renal cell carcinoma metastasis from a bone marrow transplant recipient. Bone Marrow Transplant 34:183–186 Chakraborty AK, Sodi S, Rachkovsky M et al (2000) A spontaneous murine melanoma lung metastasis comprised of host x tumor hybrids. Cancer Res 60:2512–2519 Charames GS, Bapat B (2003) Genomic instability and cancer. Curr Mol Med 3:589–596 Chen EH, Olson EN (2001) Antisocial, an intracellular adaptor protein, is required for myoblast fusion in Drosophila. Dev Cell 1:705–715 Chen EH, Olson EN (2005) Unveiling the mechanisms of cell–cell fusion. Science 308:369–373 Chen MS, Tung KS, Coonrod SA et al (1999) Role of the integrin-associated protein CD9 in binding between sperm ADAM 2 and the egg integrin alpha6beta1: implications for murine fertilization. Proc Natl Acad Sci USA 96:11830–11835 Cho C, Bunch DO, Faure JE et al (1998) Fertilization defects in sperm from mice lacking fertilin beta. Science 281:1857–1859 Cowan CA, Atienza J, Melton DA et al (2005) Nuclear reprogramming of somatic cells after fusion with human embryonic stem cells. Science 309:1369–1373 Davies PS, Dismuke AD, Powell AE et al (2008) Wnt-reporter expression pattern in the mouse intestine during homeostasis. BMC Gastroenterol 8:57 Davies PS, Powell AE, Swain JR et al (2009) Inflammation and proliferation act together to mediate intestinal cell fusion. PLoS One 4:e6530 Deb A, Wang S, Skelding KA et al (2003) Bone marrow-derived cardiomyocytes are present in adult human heart: a study of gender-mismatched bone marrow transplantation patients. Circulation 107:1247–1249 Do JT, Han DW, Gentile L et al (2009) Reprogramming of Xist against the pluripotent state in fusion hybrids. J Cell Sci 122:4122–4129 Doberstein SK, Fetter RD, Mehta AY et al (1997) Genetic analysis of myoblast fusion: blown fuse is required for progression beyond the prefusion complex. J Cell Biol 136:1249–1261 Doyonnas R, LaBarge MA, Sacco A et al (2004) Hematopoietic contribution to skeletal muscle regeneration by myelomonocytic precursors. Proc Natl Acad Sci USA 101:13507–13512 Duesberg P, Li R (2003) Multistep carcinogenesis: a chain reaction of aneuploidizations. Cell Cycle 2:202–210 Duncan AW, Dorrell C, Grompe M (2009) Stem cells and liver regeneration. Gastroenterology 137:466–481 Ellerman DA, Pei J, Gupta S et al (2009) Izumo is part of a multiprotein family whose members form large complexes on mammalian sperm. Mol Reprod Dev 76:1188–1199 Evans JP (2001) Fertilin beta and other ADAMs as integrin ligands: insights into cell adhesion and fertilization. Bioessays 23:628–639 Ferrari G, Cusella-De Angelis G, Coletta M et al. (1998) Muscle regeneration by bone marrowderived myogenic progenitors. Science 279:1528–1530 Franklin LE (1965) Morphology of gamete membrane fusion and of sperm entry into oocytes of the sea urchin. J Cell Biol 25:81–100 Gaspar-Maia A, Alajem A, Polesso F et al. (2009) Chd1 regulates open chromatin and pluripotency of embryonic stem cells. Nature 460:863–868 Goldenberg DM (1968) [On the progression of malignity: a hypothesis]. Klin Wochenschr 46: 898–899 Gould E, Gross CG (2002) Neurogenesis in adult mammals: some progress and problems. J Neurosci 22:619–623 Gritti A, Parati EA, Cova L et al (1996) Multipotential stem cells from the adult mouse brain proliferate and self-renew in response to basic fibroblast growth factor. J Neurosci 16: 1091–1100 Grompe M, Lindstedt S, al-Dhalimy M et al (1995) Pharmacological correction of neonatal lethal hepatic dysfunction in a murine model of hereditary tyrosinaemia type I. Nat Genet 10: 453–460
14
Cell Fusion and Stem Cells
311
Gupta PB, Chaffer CL, Weinberg RA (2009a) Cancer stem cells: mirage or reality? Nat Med 15:1010–1012 Gupta S, Primakoff P, Myles DG (2009b) Can the presence of wild-type oocytes during insemination rescue the fusion defect of CD9 null oocytes? Mol Reprod Dev 76:602 Gurtner GC, Davis V, Li H et al (1995) Targeted disruption of the murine VCAM1 gene: essential role of VCAM-1 in chorioallantoic fusion and placentation. Genes Dev 9:1–14 Gussoni E, Bennett RR, Muskiewicz KR et al (2002) Long-term persistence of donor nuclei in a Duchenne muscular dystrophy patient receiving bone marrow transplantation. J Clin Invest 110:807–814 He ZY, Brakebusch C, Fassler R et al (2003) None of the integrins known to be present on the mouse egg or to be ADAM receptors are essential for sperm-egg binding and fusion. Dev Biol 254:226–237 Hemler ME (2005) Tetraspanin functions and associated microdomains. Nat Rev Mol Cell Biol 6:801–811 Horsley V, Jansen KM, Mills ST et al (2003) IL-4 acts as a myoblast recruitment factor during mammalian muscle growth. Cell 113:483–494 Horsley V, Pavlath GK (2004) Forming a multinucleated cell: molecules that regulate myoblast fusion. Cells Tissues Organs 176:67–78 Hynes RO (2002) Integrins: bidirectional, allosteric signaling machines. Cell 110:673–687 Inoue N, Ikawa M, Isotani A et al (2005) The immunoglobulin superfamily protein Izumo is required for sperm to fuse with eggs. Nature 434:234–238 Jackson KA, Majka SM, Wang H et al (2001) Regeneration of ischemic cardiac muscle and vascular endothelium by adult stem cells. J Clin Invest 107:1395–1402 Jahn R, Sudhof TC (1999) Membrane fusion and exocytosis. Annu Rev Biochem 68:863–911 Jiang Y, Jahagirdar BN, Reinhardt RL et al (2002) Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 418:41–49 Johansson CB, Youssef S, Koleckar K et al (2008) Extensive fusion of haematopoietic cells with Purkinje neurons in response to chronic inflammation. Nat Cell Biol 10:575–583 Kaji K, Kudo A (2004) The mechanism of sperm-oocyte fusion in mammals. Reproduction 127:423–429 Kaji K, Oda S, Shikano T et al (2000) The gamete fusion process is defective in eggs of Cd9deficient mice. Nat Genet 24:279–282 Kajstura J, Rota M, Whang B et al (2005) Bone marrow cells differentiate in cardiac cell lineages after infarction independently of cell fusion. Circ Res 96:127–137 Karin M (2008) The IkappaB kinase – a bridge between inflammation and cancer. Cell Res 18: 334–342 Karin M, Greten FR (2005) NF-kappaB: linking inflammation and immunity to cancer development and progression. Nat Rev Immunol 5:749–759 Kotton DN, Ma BY, Cardoso WV et al (2001) Bone marrow-derived cells as progenitors of lung alveolar epithelium. Development 128:5181–5188 Krause DS, Theise ND, Collector MI et al (2001) Multi-organ, multi-lineage engraftment by a single bone marrow-derived stem cell. Cell 105:369–377 Lagasse E, Connors H, Al-Dhalimy M et al (2000) Purified hematopoietic stem cells can differentiate into hepatocytes in vivo. Nat Med 6:1229–1234 Larizza L, Schirrmacher V, Pfluger E (1984) Acquisition of high metastatic capacity after in vitro fusion of a nonmetastatic tumor line with a bone marrow-derived macrophage. J Exp Med 160:1579–1584 Le Naour F, Rubinstein E, Jasmin C et al (2000) Severely reduced female fertility in CD9-deficient mice. Science 287:319–321 Mao X, Fujiwara Y, Orkin SH (1999) Improved reporter strain for monitoring Cre recombinasemediated DNA excisions in mice. Proc Natl Acad Sci USA 96:5037–5042 Martens S, McMahon HT (2008) Mechanisms of membrane fusion: disparate players and common principles. Nat Rev Mol Cell Biol 9:543–556
312
A. Silk et al.
McKay R (1997) Stem cells in the central nervous system. Science 276:66–71 Mekler LB (1971) [Hybridization of transformed cells with lymphocytes as 1 of the probable causes of the progression leading to the development of metastatic malignant cells]. Vestn Akad Med Nauk SSSR 26:80–89 Mintz B, Baker WW (1967) Normal mammalian muscle differentiation and gene control of isocitrate dehydrogenase synthesis. Proc Natl Acad Sci USA 58:592–598 Miyado K, Yamada G, Yamada S et al (2000) Requirement of CD9 on the egg plasma membrane for fertilization. Science 287:321–324 Miyado K, Yoshida K, Yamagata K et al (2008) The fusing ability of sperm is bestowed by CD9containing vesicles released from eggs in mice. Proc Natl Acad Sci USA 105:12921–12926 Morrison SJ, Shah NM, Anderson DJ (1997) Regulatory mechanisms in stem cell biology. Cell 88:287–298 Moser AR, Pitot HC, Dove WF (1990) A dominant mutation that predisposes to multiple intestinal neoplasia in the mouse. Science 247:322–324 Muller P, Pfeiffer P, Koglin J et al (2002) Cardiomyocytes of noncardiac origin in myocardial biopsies of human transplanted hearts. Circulation 106:31–35 Nygren JM, Jovinge S, Breitbach M et al (2004) Bone marrow-derived hematopoietic cells generate cardiomyocytes at a low frequency through cell fusion, but not transdifferentiation. Nat Med 10:494–501 Nygren JM, Liuba K, Breitbach M et al (2008) Myeloid and lymphoid contribution to nonhaematopoietic lineages through irradiation-induced heterotypic cell fusion. Nat Cell Biol 10:584–592 Okabe M, Adachi T, Takada K et al (1987) Capacitation-related changes in antigen distribution on mouse sperm heads and its relation to fertilization rate in vitro. J Reprod Immunol 11:91–100 Okabe M, Yagasaki M, Oda H et al (1988) Effect of a monoclonal anti-mouse sperm antibody (OBF13) on the interaction of mouse sperm with zona-free mouse and hamster eggs. J Reprod Immunol 13:211–219 Okamoto R, Matsumoto T, Watanabe M (2006) Regeneration of the intestinal epithelia: regulation of bone marrow-derived epithelial cell differentiation towards secretory lineage cells. Hum Cell 19:71–75 Okamoto R, Yajima T, Yamazaki M et al (2002) Damaged epithelia regenerated by bone marrowderived cells in the human gastrointestinal tract. Nat Med 8:1011–1017 Okumura T, Wang SS, Takaishi S et al (2009) Identification of a bone marrow-derived mesenchymal progenitor cell subset that can contribute to the gastric epithelium. Lab Invest 89:1410–1422 Orkin SH, Zon LI (2002) Hematopoiesis and stem cells: plasticity versus developmental heterogeneity. Nat Immunol 3:323–328 Orlic D, Hill JM, Arai AE (2002) Stem cells for myocardial regeneration. Circ Res 91:1092–1102 Orlic D, Kajstura J, Chimenti S et al (2001a) Bone marrow cells regenerate infarcted myocardium. Nature 410:701–705 Orlic D, Kajstura J, Chimenti S et al (2001b) Mobilized bone marrow cells repair the infarcted heart, improving function and survival. Proc Natl Acad Sci USA 98:10344–10349 Palermo A, Doyonnas R, Bhutani N et al (2009) Nuclear reprogramming in heterokaryons is rapid, extensive, and bidirectional. FASEB J 23:1431–1440 Pawelek JM (2000) Tumour cell hybridization and metastasis revisited. Melanoma Res 10:507–514 Pawelek JM (2005) Tumour-cell fusion as a source of myeloid traits in cancer. Lancet Oncol 6: 988–993 Pawelek JM, Chakraborty AK (2008) Fusion of tumour cells with bone marrow-derived cells: a unifying explanation for metastasis. Nat Rev Cancer 8:377–386 Potten CS, Owen G, Roberts SA (1990) The temporal and spatial changes in cell proliferation within the irradiated crypts of the murine small intestine. Int J Radiat Biol 57:185–199 Primakoff P, Hyatt H, Tredick-Kline J (1987) Identification and purification of a sperm surface protein with a potential role in sperm-egg membrane fusion. J Cell Biol 104:141–149
14
Cell Fusion and Stem Cells
313
Primakoff P, Myles DG (2002) Penetration, adhesion, and fusion in mammalian sperm-egg interaction. Science 296:2183–2185 Primakoff P, Myles DG (2007) Cell–cell membrane fusion during mammalian fertilization. FEBS Lett 581:2174–2180 Reya T, Clevers H (2005) Wnt signalling in stem cells and cancer. Nature 434:843–850 Reyes M, Dudek A, Jahagirdar B et al (2002) Origin of endothelial progenitors in human postnatal bone marrow. J Clin Invest 109:337–346 Reynolds BA, Weiss S (1992) Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science 255:1707–1710 Richardson BE, Nowak SJ, Baylies MK (2008) Myoblast fusion in fly and vertebrates: new genes, new processes and new perspectives. Traffic 9:1050–1059 Rizvi AZ, Swain JR, Davies PS et al (2006) Bone marrow-derived cells fuse with normal and transformed intestinal stem cells. Proc Natl Acad Sci USA 103:6321–6325 Rochlin K, Yu S, Roy S, Baylies MK (2009) Myoblast fusion: when it takes more to make one. Dev Biol 341:66–83 Rubinstein E, Ziyyat A, Wolf JP et al (2006) The molecular players of sperm-egg fusion in mammals. Semin Cell Dev Biol 17:254–263 Rushton E, Drysdale R, Abmayr SM et al (1995) Mutations in a novel gene, myoblast city, provide evidence in support of the founder cell hypothesis for Drosophila muscle development. Development 121:1979–1988 Sato T, Vries RG, Snippert HJ et al (2009) Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459:262–265 Schwartz RE, Reyes M, Koodie L et al (2002) Multipotent adult progenitor cells from bone marrow differentiate into functional hepatocyte-like cells. J Clin Invest 109: 1291–1302 Shamsadin R, Adham IM, Nayernia K et al (1999) Male mice deficient for germ-cell cyritestin are infertile. Biol Reprod 61:1445–1451 Silva J, Barrandon O, Nichols J et al (2008) Promotion of reprogramming to ground state pluripotency by signal inhibition. PLoS Biol 6:e253 Silva J, Chambers I, Pollard S et al (2006) Nanog promotes transfer of pluripotency after cell fusion. Nature 441:997–1001 Simmons DG, Cross JC (2005) Determinants of trophoblast lineage and cell subtype specification in the mouse placenta. Dev Biol 284:12–24 Snell WJ, White JM (1996) The molecules of mammalian fertilization. Cell 85:629–637 Spees JL, Olson SD, Ylostalo J et al (2003) Differentiation, cell fusion, and nuclear fusion during ex vivo repair of epithelium by human adult stem cells from bone marrow stroma. Proc Natl Acad Sci USA 100:2397–2402 Srinivas BP, Woo J, Leong WY et al (2007) A conserved molecular pathway mediates myoblast fusion in insects and vertebrates. Nat Genet 39:781–786 Sutovsky P (2009) Sperm-egg adhesion and fusion in mammals. Expert Rev Mol Med 11:e11 Szollosi DG, Ris H (1961) Observations on sperm penetration in the rat. J Biophys Biochem Cytol 10:275–283 Tada M, Takahama Y, Abe K et al (2001) Nuclear reprogramming of somatic cells by in vitro hybridization with ES cells. Curr Biol 11:1553–1558 Takahashi K, Tanabe K, Ohnuki M et al (2007) Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131:861–872 Taylor MV (2002) Muscle differentiation: how two cells become one. Curr Biol 12:R224–228 Terada N, Hamazaki T, Oka M et al (2002) Bone marrow cells adopt the phenotype of other cells by spontaneous cell fusion. Nature 416:542–545 Tosh D, Slack JM (2002) How cells change their phenotype. Nat Rev Mol Cell Biol 3:187–194 Trounson A (2009) New perspectives in human stem cell therapeutic research. BMC Med 7:29 Turner BM (2008) Open chromatin and hypertranscription in embryonic stem cells. Cell Stem Cell 2:408–410
314
A. Silk et al.
Vassilopoulos G, Wang PR, Russell DW (2003) Transplanted bone marrow regenerates liver by cell fusion. Nature 422:901–904 Vignery A (2005) Macrophage fusion: the making of osteoclasts and giant cells. J Exp Med 202:337–340 Vogelstein B, Kinzler KW (2004) Cancer genes and the pathways they control. Nat Med 10: 789–799 Wang X, Willenbring H, Akkari Y et al (2003) Cell fusion is the principal source of bone-marrowderived hepatocytes. Nature 422:897–901 Weimann JM, Charlton CA, Brazelton TR et al (2003a) Contribution of transplanted bone marrow cells to Purkinje neurons in human adult brains. Proc Natl Acad Sci USA 100:2088–2093 Weimann JM, Johansson CB, Trejo A et al (2003b) Stable reprogrammed heterokaryons form spontaneously in Purkinje neurons after bone marrow transplant. Nat Cell Biol 5:959–966 Willenbring H, Bailey AS, Foster M et al (2004) Myelomonocytic cells are sufficient for therapeutic cell fusion in liver. Nat Med 10:744–748 Yanagimachi R (1994) Mammalian fertilization. In: Knobil E, Neill J (eds) The physiology of reproduction. Raven Press New York, NY, pp 189–317 Yilmaz Y, Lazova R, Qumsiyeh M et al (2005) Donor Y chromosome in renal carcinoma cells of a female BMT recipient: visualization of putative BMT-tumor hybrids by FISH. Bone Marrow Transplant 35:1021–1024 Ying QL, Nichols J, Evans EP et al (2002) Changing potency by spontaneous fusion. Nature 416:545–548 Yu J, Vodyanik MA, He P et al (2006) Human embryonic stem cells reprogram myeloid precursors following cell–cell fusion. Stem Cells 24:168–176 Yu J, Vodyanik MA, Smuga-Otto K et al (2007) Induced pluripotent stem cell lines derived from human somatic cells. Science 318:1917–1920 Zhang F, Pomerantz JH, Sen G et al (2007) Active tissue-specific DNA demethylation conferred by somatic cell nuclei in stable heterokaryons. Proc Natl Acad Sci USA 104:4395–4400 Zhang S, Wang D, Estrov Z et al (2004) Both cell fusion and transdifferentiation account for the transformation of human peripheral blood CD34-positive cells into cardiomyocytes in vivo. Circulation 110:3803–3807
Chapter 15
Cell Fusion and Dendritic Cell-Based Vaccines Jianlin Gong and Shigeo Koido
Abstract Fusions of dendritic cells (DC) and tumor cells are increasingly used in tumor immunotherapy. The strategy for DC-tumor fusion vaccine is based on the fact that DC are the most potent antigen-presenting cells in the body, whereas tumor cells express abundant tumor antigens. The fusion of these two cell types creates a heterokaryon with both DC-derived costimulatory molecules, efficient antigenprocessing and -presentation machinery, and tumor-derived antigens. In animal and human studies, fusion-cell (FC) vaccines have been shown to possess the elements essential for processing and presenting tumor antigens to host immune cells, for inducing effective immune response, and for breaking T-cell tolerance to tumorassociated antigens. Moreover, FC vaccines provide protection against challenge with tumor cells and mediate regression of established tumors. Despite these unique features of DC-tumor fusion cells and the observation of tumor eradication in animal studies, only limited, yet encouraging, success has been seen in clinical trials. This chapter describes the methods used for preparation of DC-tumor fusion cells, summarizes the effect of FC in stimulating T cell responses, analyzes factors influencing the success or failure of FC-mediated immunotherapy and discusses recent advances in concept and techniques of DC-tumor fusions. Keywords Antitumor immunity · cancer · cell-cell fusion · dendritic cell · fusion cell vaccine · immune therapy Abbreviations APC CD CpG ODN CTL DC
Antigen presenting cell Cluster of differentiation Oligodeoxynucleotides containing CpG motif Cytotoxic T lymphocytes Dendritic cell
J. Gong (B) Department of Medicine, Boston University Medical School, Boston, MA 02118, USA e-mail:
[email protected]
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_15, C Springer Science+Business Media B.V. 2011
315
316
DLN DMEM FACS FC FCS FMG GFP GM-CSF HAU HCC HLA HVJ ICAM-1 IFN IL LFA LNC LPS MHC MMT mice MUC1 PBMC PEG PGE PyMT RCC TAP TGF-β Th1 TLR TNF Treg VEGF VSV
J. Gong and S. Koido
Draining lymph nodes Dulbecco’s modified Eagle’s medium Fluorescence activated cell sorting Fusion cell Fetal calf serum Fusogenic membrane glycoprotein Green fluorescent protein Granulocyte-macrophage colony-stimulating factor Hemagglutinating units Human hepatocellular carcinoma Human leukocyte antigen Hemagglutinating virus of Japan Inter-cellular adhesion molecule 1 Interferon Interleukin Lymphocyte function-associated antigen Lymph node cells Lipopolysaccharide Major histocompatibility complex Mice expressing PyMT oncogene and MUC Mucin 1 tumor-associated antigen Peripheral blood monocytes Polyethylene glycol Prostaglandin E Polyomavirus middle T Renal cell carcinoma Transporter associated with antigen processing Transforming growth factor-β T helper 1 cells Toll-like receptor Tumor necrosis factor Regulatory T cells Vascular endothelial growth factor Vesicular stomatitis virus
Contents 15.1 Introduction . . . . . . . . . . . . . . . . . . . . . 15.2 The Rationale for DC-Based Cell Fusion as Tumor Vaccine 15.3 Methods . . . . . . . . . . . . . . . . . . . . . . . 15.3.1 Generation of DC from Murine Bone Marrow Cells 15.3.2 Preparation of Tumor Cells . . . . . . . . . . . 15.3.3 Cell Fusion . . . . . . . . . . . . . . . . . . 15.4 Choice of Fusogen . . . . . . . . . . . . . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
317 318 321 321 323 323 324
15
Cell Fusion and Dendritic Cell-Based Vaccines
15.4.1 PEG-Mediated Fusion . . . . . . . . 15.4.2 Electrofusion . . . . . . . . . . . . 15.4.3 Virus-Mediated Fusion . . . . . . . . 15.5 Selection of Fusion Cells . . . . . . . . . . 15.6 Modifications in Cell Fusion . . . . . . . . . 15.6.1 Allogenic DC . . . . . . . . . . . . 15.6.2 Allogeneic Tumor Cells . . . . . . . 15.6.3 Fusion Cells Expressing Cytokines . . 15.6.4 DC Maturation . . . . . . . . . . . 15.7 Fusion Cell Vaccines and Antitumor Immunity 15.7.1 Animal Studies . . . . . . . . . . . 15.7.2 Clinical Trials . . . . . . . . . . . . 15.8 Promotion of Antitumor Immunity . . . . . . 15.8.1 Using Adjuvant with Fusion Vaccine . 15.8.2 Combined Approaches . . . . . . . . 15.9 Summary . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . .
317
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . .
324 325 326 327 328 328 329 329 330 331 332 336 339 339 340 341 342
15.1 Introduction It is apparent that the immune system can mount an antigen-specific response to tumors (Pardoll 2003). However, response is limited by the ability of tumor cells to evolve various strategies to evade immune eradication. In addition, the immune response to cancer can be boosted by various approaches to immunostimulation. Popular among these is the development of anticancer vaccines, in which tumor antigens are introduced into tumor-bearing animals in such a way as to lead to the production of tumor-specific cytotoxic T cells that can find and kill tumor cells in local and disseminated sites. Although desirable in its specificity and power, this approach is still evolving as we struggle to understand the parameters that will lead to production of a vaccine effective in the clinical setting (Finn 2003, Rosenberg 2004, Rosenberg et al. 2004). In this chapter we will discuss cell fusion as one of the dendritic cell-based vaccines and its deployment in cancer therapy. Cell fusion, which can be defined as two or more separate cells brought together to form a single cell occurs both naturally and under experimental conditions. It is required for propagation (Primakoff and Myles 2002), for the generation of multinucleated muscle fibers (Chen and Olson 2004) and for the replenishment and repair of organs by stem cells (Stolzing et al. 2007) and is found in certain tumors (Guccion and Enzinger 1972). The technique is used extensively in laboratory experiments for a variety of purposes, including studies of cell membranes, chromosome mapping, delivery of substances into cells, and stem cells (Frye and Edidin 1970, Furusawa et al. 1974, Islam et al. 2006, Okada and Tadokoro 1963, Okada et al. 1964, 1974, Rodriguez-Tome and Lijnzaad 1999, Singer and Nicolson 1972). The best-studied example of cell fusion, however, is the production of
318
J. Gong and S. Koido
monoclonal antibody by hybridoma using somatic cell hybridization (Kohler and Milstein 1975). In their landmark discovery, Kohler and Milstein (1975) fused murine spleen cells from immunized mice with myeloma cells using Sendai virus as fusogen, thus creating hybrid cells called hybridomas that can be propagated without limit and that produce antibodies with predefined specificity. The fusogen of Sendai virus was later changed to polyethylene glycol (PEG) to promote efficient generation of hybridomas (Galfre et al. 1977, Margulies 2005). Propelled by this discovery, the search for improved fusion partners, better fusion methods, and expanded use of the technology has exploded (Galfre et al. 1977, Margulies 2005, Milstein 1999). Through these investigations, we know that the fused cells inherit the properties of their parental cells. For example, the membranes of fused cells are integrated into a single cell, whereas the nuclei remained separate, at least in the primary hybrid cell (Davidson and Gerald 1976). In the course of cell division, the nuclear membrane disintegrates and a single large nucleus is formed containing the chromosomes of both parent cells (Kuby 1992). Random loss of chromosomes occurs if the hybrid cells continue to divide. Recent studies have explored the use of dendritic-cell (DC) fusion with tumor cells as an approach to tumor vaccine design. The fusion of DC and tumor cells, although developed for a different purpose than hybridomas, shares many similarities with such heterokaryons. For example, the fusion cell is formed by one somatic cell and one tumor cell using the same fusogen, and the fusion cell inherits the properties of the parent cells. In this context, employing the fusion of DC and tumor cells as a strategy for tumor vaccine preparation is an expanded usage of the cell-fusion technique, whose development is propelled in large part by the scientific and industrial use of hybridomas (Kohler and Milstein 1975).
15.2 The Rationale for DC-Based Cell Fusion as Tumor Vaccine Dendritic cells are the most potent antigen-presenting cells (APC) in the body and are capable of initiating primary immune responses (Steinman 1991, 2001). These cells derive their potency from constitutive and inducible expression of essential co-stimulatory ligands on the cell surface including B7, ICAM-1, LFA-1, LFA-3, and CD40 (Young et al. 1992, Inaba et al. 1994). These proteins function in concert to generate a network of secondary signals essential for reinforcing the primary antigen-specific signal in T-cell activation (Inaba et al. 1997, Thery and Amigorena 2001, Young and Inaba 1996). Many strategies have been developed to load tumor antigen onto DC and use them as tumor vaccines. DC pulsed with tumor-associated peptides or tumor lysates results in the induction of antitumor immunity and disease regression in animal and clinical studies (Celluzzi et al. 1996, Hsu et al. 1996, Loveland et al. 2006, Mayordomo et al. 1995, Nestle et al. 1998, Paglia et al. 1996). Tumor antigens have been also introduced into DC through viral vectors encoding tumor-specific genes or through transfection with liposomal DNA or RNA (Condon et al. 1996, Gong et al. 1997a, Koido et al. 2000, Specht et al. 1997, Tang et al. 2007). These loading approaches, while effective to some degree, share certain drawbacks. For example, few tumor antigens have been identified,
15
Cell Fusion and Dendritic Cell-Based Vaccines
319
and tumor cells may evade recognition through down-regulation of a single tumor antigen. As an alternative strategy, DC have been fused to tumor cells. The fusion of DC and tumor cells through chemical (Akasaki et al. 2001, Cao et al. 1999, Gong et al. 1997b, 2002, Homma et al. 2001, 2005a, Kao et al. 2003, 2005, Li et al. 2001, Lindner and Schirrmacher 2002, Liu et al. 2002, Ogawa et al. 2004, Takeda et al. 2003, Wang et al. 1998, Xia et al. 2003, 2005, Zhang et al. 2003), physical (Goddard et al. 2003, Hayashi et al. 2002, Jantscheff et al. 2002, Lindner and Schirrmacher 2002, Marten et al. 2003, Scott-Taylor et al. 2000, Shimizu et al. 2004, Siders et al. 2003, Suzuki et al. 2005, Tanaka et al. 2002, Trefzer et al. 2005, Trevor et al. 2004) or biological (Hiraoka et al. 2004, Phan et al. 2003) means creates a heterokaryon which combines DC-derived costimulatory molecules, efficient antigen-processing and -presentation machinery, and an abundance of tumor-derived antigens. Thus, the DC-tumor fusion cells combine the essential elements for presenting tumor antigens to host immune cells and for inducing effective antitumor response. Importantly, the fusion process facilitates the entry of tumor antigen into the endogenous antigen-processing pathway of the DC (Fig. 15.1). It has been shown that antigen is processed and presented along two major pathways, including: (1) endogenously synthesized intracellular proteins, such as those expressed in viral infections, through the major histocompatibility complex (MHC) class-I-restricted pathway to cytotoxic T lymphocytes (CTL); (2) the endocytic pathway. Exogenous proteins from the extracellular environment are captured and delivered to the compartments of the endosome/lysosome, where they are degraded to antigenic peptides
Fig. 15.1 A model of antigen processing and presentation by DC-tumor fusion cell. The process of cell fusion results in the integration of cytoplasm of DC and tumor cells which makes it possible to exchange cellular content and share the subcellular compartments between these two cells. Tumor antigens can be processed and presented through the antigen processing and presentation pathway of DC
320
J. Gong and S. Koido
by protease and peptidase that are complexed with MHC class II molecules and presented to CD4+ T cells (Steinman and Swanson 1995, Watts 2004). In addition, DC are also capable of presenting exogenous antigens through an endogenous pathway, a phenomenon called antigen cross-presentation (Berard et al. 2000, Heath and Carbone 2001). However, the antigen cross-presentation is generally not efficient in the absence of carrier proteins or particles (Wolkers et al. 2004). In the fusion strategy, the DC and tumor cells become one hybrid cell sharing a unified cytoplasm (Fig. 15.2). The integration of cytoplasm from DC and tumor cells renders the tumor antigens endogenous to the DC heterokaryon and, therefore, facilitates the entry of tumor antigens into the DC endogenous pathway of antigen-processing and -presentation machinery (Cao et al. 1999, Koido et al. 2004, Galea-Lauri et al. 2002). The ability of fused DC-tumor cells to process and present intracellular proteins derived from tumor cells was demonstrated by Wang et al (1998). RMA-S tumor cells are not able to process endogenous antigens due to defective TAP. After fusion with DC, the hybrid cells are able to induce antigen-specific CTL that are effective in lysis of the DC-RMA-S fusion cells, strongly suggesting that the tumor antigens are processed along the endogenous pathway, through the antigen processing machinery of DC. It is likely, moreover, that the products of fusion of two live cells can maintain the expression of their original protein repertoires, including
Fig. 15.2 Cell surface and intracellular structure of human DC-tumor fusion cell. (a) Surface structure of human DC, breast cancer cell and DC-breast tumor fusion cells was showed by scanning electron microscope (SEM, × 4,800). (b) Intracellular structure of DC, breast cancer cell and DC fused to breast cancer cells was demonstrated by transmission electron microscope (× 4,800). Arrow points to the fusion of DC and tumor cells in the enlarged panel. DC-N, DC nucleus; Tu-N, tumor nucleus
15
Cell Fusion and Dendritic Cell-Based Vaccines
321
tumor antigen (Kao et al. 2005). Thus it is conceivable that tumor antigens synthesized de novo in the heterokaryons are processed and presented through the endogenous pathway. Furthermore, fusion cells have the ability to present multiple tumor antigens, thus increasing the frequency of responding T cells and maximizing antitumor immunity (Akasaki et al. 2001, Gong et al. 1997b, 1998, Homma et al. 2001, Koido et al. 2005a, Lespagnard et al. 1998, Wang et al. 1998). Fusion cells are also advantageous in antigen presentation in vivo. Fusion of DC and tumor cell creates immunogenic cells with the ability to migrate to draining lymph nodes (DLN) where they interact with CD4+ and CD8+ T cells and induce potent antitumor immunity (Koido et al. 2002, Phan et al. 2003). In this context, the FC functions as an APC. This is important since defective DC have been observed in cancer patients (Orsini et al. 2003, Ratta et al. 2002, Satthaporn et al. 2004). Direct antigen presentation by FC can surmount this problem and bypass the defective APC of the host. In addition, the death of APC can release tumor antigen. Since FC inherit the proteome of tumor cells, it seems likely that host DC will take up the tumor antigens released from dying FC, process them and represent them to CD4+ and CD8+ T cells. Together, tumor-specific T cells can be induced directly or indirectly by FC.
15.3 Methods Cell fusion between DC and tumor cells can be achieved through chemical, physical or biological means. In our lab, we use PEG as a fusogen to fuse DC and tumor cells (Fig. 15.3). The following is the protocol that we have used to prepare the DC-tumor fusion vaccine.
15.3.1 Generation of DC from Murine Bone Marrow Cells DC can be generated from a variety of sources including bone marrow cells and peripheral blood monocytes (PBMC). For clarity, bone marrow cells from mice will be used here to describe the generation of DC. We usually use femur and tibia to obtain the bone marrow. The protocol is based on previously described method (Inaba et al. 1992) with modifications. 1. Sacrifice mice with CO2 inhalation and expose femur and tibia with long transverse incision. 2. Detach hind legs and remove muscles and connective tissues from the femur and tibia with sterile gauze. 3. Wash bones in 70% ETOH for 1 min, then wash twice with PBS. 4. Flush out bone marrow by using a syringe and 23 gauge needle with 10 ml RPMI 1640 (no serum). The bone marrow is suspended and passed through a Falcon cell filter to remove small pieces of bones and debris.
322
J. Gong and S. Koido
Fig. 15.3 The diagram of the fusion procedure
5. 6. 7. 8.
9. 10. 11. 12.
Wash cells in RPMI 1640 (no serum). Deplete red blood cells with lysing buffer. Wash bone marrow cells twice with PBS and then count cells. Resuspend cell pellet in antibody cocktail solution for 1 h on ice and then incubate them in Low-Tox-M rabbit complement solution for an additional 1 h at 370◦ C to remove/lyse lymphocytes and Ia+ cells. Wash cells and count total number of cells including dead cells and calculate the percentage of dead cells whose average is usually about 30%. Place 1 × 106 cells/ml/well in 5% FCS RPMI 1,640 medium containing 1,000 U/ml of GM-CSF for DC culture. On day 3 of culture, the cell clusters will appear. The fresh medium with 1,000 U/ml of GM-CSF is added if the medium is yellow. On day 6 of culture, the cell clusters are collected by gently dislodging with a sterile short Pasteur pipet and placed into tissues culture dish for 1 h. After
15
Cell Fusion and Dendritic Cell-Based Vaccines
323
1 h adherence, remove the non-adherent cell to another tissues culture dish and add fresh medium with 1,000 U/ml of GM-CSF into adherent cells. Repeat adherent step twice and discard the non-adherent cells including non-adherent granulocytes and dead cells. 13. Harvest floating and loosely adherent DC.
15.3.2 Preparation of Tumor Cells Tumor cells can be either freshly isolated from tumor samples or vial of frozen cell lines. The method described here is used to isolate and culture mammary tumor cells from MMT mice that develop spontaneous mammary carcinomas (Xia et al. 2003). 1. Sacrifice MMT mice with CO2 inhalation and harvest mammary tumor samples under sterile condition. 2. Mince tumor into small pieces (1–3 mm). Digest them in HBSS medium containing collagenase and DNase. 3. The digested tumor tissue is mashed through a sterile 50-μm nylon mesher in tissue culture hood. 4. Single-tumor cell suspensions are obtained by passing through a filter. 5. Culture the tumor cells in high glucose DMEM medium containing 10% FCS and antibiotics. Remove the non-adherent dead cells. 6. Incubate the cells at 370◦ C for 2–3 days. Cells are ready for fusion when they are in the logarithmic phase of growth.
15.3.3 Cell Fusion 1. DC are generated from bone marrow cells of mice as previously described and cultured in 20 ng/ml rmGM-CSF medium for 5–8 days. 2. Tumor cells are maintained in DMEM supplemented with 10% heat-inactivated FCS, 2 mM L-glutamine, 100 U/ml penicillin, 100 mg/ml streptomycin. 3. The purified DC are mixed with tumor cells at a 10:1 ratio and the mixture is washed once with no serum medium followed by spin-down to obtain cell pellets. 4. The mixed cell pellets are gently resuspended in prewarmed 50% PEG solution (1 ml per 1–5 108 cells) for 5 min at room temperature. 5. The PEG solution is diluted by slow addition and mixing of 1, 2, 4, 8 and 16 ml warm serum-free medium. 6. The cell pellets obtained after centrifuge at 1,350 rpm are resuspended in RPMI 1,640 medium supplemented with 10% heat-inactivated FCS, 2 mM L-glutamine, 10 mM nonessential amino acids, 1 mM sodium pyruvate, 10% NCTC 109, 100 U/ml penicillin, 100 mg/ml streptomycin, and 10 ng/ml murine rGM-CSF, and further cultured for 5–8 days.
324
J. Gong and S. Koido
7. After 5–8 days, DC-tumor fusion cells are loosely adherent to the culture dish, whereas tumor–tumor fusions and unfused tumor cells are attached firmly to the dish. The loosely adherent fusion cells are obtained first by the gentle pipetting. 8. The fusion efficiency is determined by dual expression of tumor antigen and DC maker (MHC class II molecules or co-stimulatory molecules).
15.4 Choice of Fusogen Investigation indicates that fusion efficiency is closely correlated to antitumor immunity (unpublished data). Therefore, effort has been made to promote fusion efficiency by choosing an optimal fusion method. Polyethylene glycol (PEG) is a chemical fusogen frequently used in the production of hybridoma for antibody production (Das and Suresh 2005, de StGroth and Scheidegger 1980, Kohler and Milstein 1975, 1976) and initial tumor–tumor fusion (McCune et al. 1982, O’Donnell et al. 1984). Dehydration by PEG facilitates the membrane contact of adjacent cells, and cell swelling caused by rehydration results in cell fusion (Ahkong and Lucy 1986, Ahkong et al. 1994, Knutton 1979). On the other hand, electricfield-mediated fusion involves alignment of cells prior to delivery of a direct current pulse in which the cell membranes are momentarily destabilized and fuse to form hybrid cells (Bischoff et al. 1982, Jaroszeski et al. 1994, Neil and Zimmermann 1993). The third option for cell fusion uses biologic means, such as gene transfer into tumor cells of a viral fusogenic membrane glycoprotein (FMG) that mediates fusion between DC and tumor cells. Although all three methods have been reported to produce therapeutic numbers of DC-tumor fusion cells in animal and/or human studies, variability of fusion efficiency exists in each method. Such differences are derived partly from those inherent in various types of tumor cells and partly from the skill of the investigator. It is evident that the method more frequently employed by the investigator produces higher fusion efficiency than one rarely used. This point is often overlooked when fusion efficiency from different fusion methods is compared. Regardless of the fusion method, quality control of the fusion partners, the DC and tumor cells, and a short-duration culture of fusion-cell product are always helpful in promotion of fusion efficiency and increased quality of FC.
15.4.1 PEG-Mediated Fusion One of the advantages in employing PEG for fusion is simplicity. In general, PEGmediated fusion leads to fusion efficiency varying from 15 to 50% in murine models (Cao et al. 1999, Homma et al. 2001, Kao et al. 2003, Lespagnard et al. 1998, Li et al. 2001, Lindner and Schirrmacher 2002, Liu et al. 2002, Takeda et al. 2003, Wang et al. 1998) and in human studies (Homma et al. 2003, 2005a, Koido et al. 2005b), although fusion efficiency in the single digits has occasionally been reported (Gottfried et al. 2002). In this approach, DC and tumor cells are mixed and washed
15
Cell Fusion and Dendritic Cell-Based Vaccines
325
once with serum-free medium. The mixed cell pellets are gently resuspended in prewarmed 50% PEG solution (1 ml per 5 × 107 cells) for a few minutes at room temperature. The PEG solution is then diluted by gradual addition and progressive mixing of 1, 2, 4, 8 and 16 ml warm serum-free medium. The cell pellets obtained after centrifugation at 1,350 rpm are resuspended in medium containing 10% heat-inactivated FCS and GM-CSF. The variable factor is the length of time the cells are exposed to PEG. We have found that there is some difference in the sensitivity of cells to PEG. It is desirable to perform a dose–response test to evaluate the conditions of PEG fusion for each type of tumor cell and to determine the optimal exposure time. Unlike electrofusion, DC-tumor fusion by PEG is an active and evolving process, and it is thus likely that the larger the initial contact surface between cells, the faster the integration of these cells. Fusion efficiency is lowest immediately after the fusion process is initiated, and 1-week culture results in more than a 10-fold increase in efficiency (Gong J., unpublished data). In addition, shortterm culture will give the fusion cells sufficient time to integrate and display the antigen in the context of MHC molecules. Fusion by PEG involves the fusion of two living cells. Thus prefusion irradiation of tumor cells may not be desirable (Ogawa et al. 2004, Krause et al. 2002).
15.4.2 Electrofusion Somatic cells have been fused by the technique of electrofusion (Bischoff et al. 1982), an approach that does not involve fusogens and has the advantage of requiring less cell manipulation. In electrofusion, DC and tumor cells are mixed, resuspended in an isotonic solution and placed in an electrofusion chamber (Gabrijel et al. 2004, Kjaergaard et al. 2003, Trevor et al. 2004) or electroplated cuvette (Scott-Taylor et al. 2000). Cells are first aligned by application of direct or alternate electric current. Alternate current can align cells midway between electrodes by a process called dielectrophoresis, and the cells form a “pearl chain,” which can be observed in the electrofusion chamber under the microscope (Gabrijel et al. 2004). As field strength is greatest along the axis of the applied current, aligning the cells with alternate current will serve to orient the sites of membrane breakdown and contribute to cell fusion (Neil and Zimmermann 1993). Scott-Taylor and associates (2000), however, found that preliminary alignment of cells in alternate current in the cuvette did not significantly augment the fusion rate. They observed that alignment by direct current, post-fusion centrifugation, or combined electrofusion with PEG can promote the yield of fusion cells. Electrofusion is driven by an exponential pulse delivered by a pulse generator (Kjaergaard et al. 2003, Trevor et al. 2004) or gene pulser (Lindner and Schirrmacher 2002, Scott-Taylor et al. 2000). Such an electric pulse, when applied across the cell membrane, will result in transient permeation of the cell membrane through the formation of micropores. The adjacent pores between the contact cells may form channels and lead to production of hybrid cells. Theoretically, electrofusion can be standardized by using controlled electrical
326
J. Gong and S. Koido
parameters. However, variations of fusion efficiency are observed, ranging from levels as high as 83.1% (Goddard et al. 2003) to lows in the single digits (Jantscheff et al. 2002, Trevor et al. 2004). Such disparities are related to the following factors. First, some cells are intrinsically more fusogenic than others (Scott-Taylor et al. 2000). Second, electrical parameters including the strength and duration of the electric pulse are critical to fusion efficiency, and they are also influenced by the distance between the plates of the cuvette, the conductivity of the medium, and the waxing of the cuvette (Scott-Taylor et al. 2000). Ideally, a preliminary experiment should be performed to determine the optimal parameters for each cell type being fused. This method is theoretically desirable for clinical use since no chemicals are involved. However, low fusion-cell yield may be a major problem (Haenssle et al. 2004), especially for the first time user. In addition, most investigators inject the fusion-cell products mediated by electrofusion into patients shortly after fusion (Haenssle et al. 2004, Krause et al. 2002, Trefzer et al. 2005), and it is likely that such fusion-cell products may not at this time be in optimal form to present tumor antigens to the host immune cells.
15.4.3 Virus-Mediated Fusion Many viruses possess envelope fusogenic membrane glycoproteins (FMG) that are critical for the initiation of infection of cells (Kinzler and Compton 2005, Negrete et al. 2005, Plattet et al. 2005). These viral proteins usually mediate the fusion of the viral particle with the membrane of the target cell by binding to specific receptors on the target cell. Viral FMG are also important for the propagation of infection. One of the mechanisms involved is the formation of large multinucleated syncytia by fusion of the infected cell with adjacent cells (Lamb 2001, Pare et al. 2005). The hyperfusogenicity of viral FMG have been explored to mediate fusions between tumor cells (Bateman et al. 2000, Ebert et al. 2004, Linardakis et al. 2002) and DC and tumor cells (Hiraoka et al. 2004, Phan et al. 2003). Phan and associates (2003) transfected murine B16 melanoma cells with vector encoding vesicular stomatitis virus VSV-G FMG and used them as fusion partners with DC. Co-culture of DC with tumor cells expressing VSV-G FMG resulted in up to 38% fusion of DC and tumor cells as determined by green fluorescent protein (GFP) expression. Remarkably, the fusion cells were able to migrate from subcutaneous injection sites to draining lymph nodes and induce antitumor immunity. Immunization of mice bearing 7-day-old pulmonary disease with 6-h hybrid cells led to survival of 60% of mice. In contrast, mice immunized with DC or FMG-expressing tumor cells alone succumbed to disease by 35–40 days. In an alternative approach, Hiraoka et al. (2004) used inactivated hemagglutinating virus of Japan (HVJ) to fuse DC and tumor cells. Co-culture of DC and tumor cells in the presence of 500 hemagglutinating units (HAU) produced fusion efficiency of approximately 30%. Immunization with HVJ-mediated fusion cells induced antitumor immunity, which was enhanced by co-administration of oligodeoxynucleotides containing CpG motif (CpG ODN) as adjuvant. These findings indicate that virus-mediated fusion may be a promising alternative for the
15
Cell Fusion and Dendritic Cell-Based Vaccines
327
generation of DC–tumor fusion vaccine. However, there are still obstacles to be overcome in its use in the clinical setting. Studies show that virus-mediated fusion leads to the formation of large multinucleated syncytia that subsequently die through nonapoptotic, autophage-like mechanisms (Bateman et al. 2002). This observation raises safety concerns about whether the vaccine can harm normal cells through the bystander killing effect. There are also stability issues for virus-mediated fusion cells. Only 6-h DC–tumor fusion cells mediated by VSV-G FMG induced effective antitumor immunity. This may be due to the cytotoxicity activated by the viral fusogen.
15.5 Selection of Fusion Cells There is currently no standardized method for selection of DC–tumor fusion cells. FACS analysis, the most commonly used technique for characterizing fusion and determining fusion frequency, is straightforward. However, one caveat for FACS analysis is its potential for false-positive results derived from cell aggregates and/or phagocytosis of tumor fragments by unfused DC. This problem will increase if a fluorescent dye is employed. In our studies, we apply several techniques to enrich the fusion cells and minimize false positives. The gentle pipetting and collection of fusion cells in the culture dish is the first step. Short-duration culture of the fusion products can significantly promote fusion efficiency and reduce cell aggregation. The cell aggregates can be further eliminated by gating out these cells before FACS analysis. Finally, cell sorting against DC and tumor markers will enrich the fusion cells more than 90% as needed (Koido et al. 2004). The confocal microscope has also been used to demonstrate DC–tumor fusion cells. Gabrijel et al. (2004) labeled DC and tumor cells with vital fluorescent dyes, fused them by electrofusion, and quantified the fusion cells by measuring the area of co-localized red and green pixels relative to all red and green pixels in the confocal image. They found that the measurement was highly correlated with manual counting under confocal microscope and more accurate than that obtained by FACS, since the confocal microscope can distinguish cell aggregates from fusion cells. This method is also however subject to false positivity derived from the use of fluorescent dye. Cell sorting was used by Li et al. (2001) and Holmes et al. (2001) to purify fusion cells in murine studies. However, cell sorting may not be practical due to the limited number of DC and tumor cells available in the clinical setting. In an effort to quantify the genuine DC– tumor fusion cells, Phan et al. (2003) transfected DC with Tyr-green fluorescent protein (GFP) reporter virus, which is controlled by the melanoma-specific mouse tyrosinase promoter, and then fused them with B16 melanoma cells by VSV-G FMG. The melanoma-specific promoter is normally silent in DC. However, upon fusion between DC and B16 cells, transcription factors from the melanoma cells should become available to transactivate the promoter. Therefore, only fusion cells express GFP. This method is theoretically sound and has the potential to characterize and select genuine DC–tumor fusion cells. However, the complexity of this method may limit its use, especially in the clinical setting.
328
J. Gong and S. Koido
15.6 Modifications in Cell Fusion 15.6.1 Allogenic DC Tumor cells have been fused to syngeneic or autologous dendtiric cells (Gong et al. 1997b, 2000a, b, Hiraoka et al. 2004, Kikuchi et al. 2001, Krause et al. 2002, Phan et al. 2003). In this case, the tumor antigens, known or unidentified, can be fully presented in the context of self HLA class I and class II molecules to autologous T cells. Theoretically, fusing tumor cells to autologous DC will facilitate the integration of DC and tumor cells and promote access of tumor-derived antigens to the endogenous class I processing pathway. In addition, when DC and tumor cells are mutually autologous, the fusion cells can acquire the ready-made MHC class I-peptide complexes already expressed on the surface of the tumor cells, thus increasing the level of tumor-antigenic peptide presentation by the FC. This approach, though effective, is contingent on the availability of autologous tumor cells, which is influenced by the treatment modality, the quality and quantity of surgical sample, and the type of tumor. As an alternative, allogeneic DC have been fused to patient or host-derived tumor cells (Galea-Lauri et al. 2002, Gong et al. 2000b, Marten et al. 2003, Tanaka et al. 2001, Trefzer et al. 2005). The advantage of this approach is that the fusion cells express both tumor-derived HLA class I molecules for the presentation of self MHC-restricted tumor peptide and allogeneic HLA class II molecules derived from DC for stimulation of allogeneic CD4+ T cells (Fabre 2001, Walden 2000). The activated allogeneic CD4+ T cells may secrete cytokines conducive to the activation of tumor-antigen-specific CD8+ T cells (Fabre 2001, Walden 2000). In addition, reports indicate that the frequency of alloreactive T cells is between 1 and 10% (Ford and Atkins 1973, Ford et al. 1975, Sherman and Chattopadhyay 1993) and these alloreactive T cells may contain T cells cross-reactive to tumor antigens, thus triggering additive antigen-specific Tcell response (Fabre 2001). In our study (Gong et al. 2000b, Tanaka et al. 2001), the allogeneic DC–tumor fusion cells stimulate T-cell proliferation more vigorously than autologous DC–tumor fusion cells. However, the T cells activated by allogeneic DC–tumor fusion cells are less effective in lysis of tumor cells, suggesting that the proliferated T-cells contained alloreactive T cells. Additional potential benefit for using allogeneic DC is that dendritic cells from healthy donors are readily available with normal function. DC from cancer patients may be defective in APC function due mainly to cancer treatments including chemotherapy. Recently, Suzuki et al. (2005) reported that allogeneic DC–tumor fusion cells are superior to autologous DC–tumor fusion cells. In contrast, Kjaergaard et al. (2003) found that fusion cells from allogeneic DC completely lack therapeutic effects even though the hybrid cells expressed the tumor-derived MHC class I molecules. In a phase I/II clinical study, Haenssle and co-workers (2004) fused MHC class I-mismatched DC to patientderived melanoma cells by electrofusion and used them as tumor vaccine. They observed limited clinical response to the vaccination. However, in this series, the cell viability and fusion efficiency were 50 and 10%, respectively. In addition, all the patients were in advanced stages of disease with severely depressed immune systems. In contrast, Trefzer et al. (2005) achieved immunologic and clinical response
15
Cell Fusion and Dendritic Cell-Based Vaccines
329
from allogeneic DC–tumor fusion vaccine with one complete response, one partial response, and six stable disease with long survival times in 17 patients with advanced melanoma. Marten et al. (2003) reported that immunologic but not clinical response was induced by either allogeneic or autologous DC–tumor fusion vaccine. Again, low fusion efficiency and cell viability may be the potential problem. Based on these conflicting findings, more investigation is needed to assess the benefit or disadvantage of using allogeneic DC.
15.6.2 Allogeneic Tumor Cells Dendritic cell–tumor cell fusion is often limited by the availability of autologous tumor cells which may not be obtainable if surgery is not a component of the treatment. Tumor biopsy, even if available, may not provide sufficient numbers of viable tumor cells due to the lengthy culture times required and the potential of contamination. To circumvent this problem, an alternative approach may be the use of established tumor-cell lines (Galea-Lauri et al. 2002, Jantscheff et al. 2002, Lundqvist et al. 2004, Parkhurst et al. 2003). This approach is based on the fact some antigens such as MUC1 are shared by most tumors (Greenlee et al. 2000) and melanoma-associated antigens by melanoma cells (Chomez et al. 2001). The potential benefit of this approach is that the tumor antigen presented through the matched HLA molecules and alloreactive T cells can be induced against the unmatched HLA molecules. We observed that it may not be necessary to match HLA between allogeneic tumor cells and autologous DC. In animal model, allogeneic tumor cells were fused to autologous DC (Koido et al. 2007c, Lee et al. 2005, 2010, Yasuda et al. 2007). The fusion cells activated T cells with activity to lyse tumor cells expressing shared tumor antigens. We fused HLA-A∗ 0201-positive DC with HLAA∗ 0201-positive allogeneic MCF7 or HLA-A∗ 0201-negative BT20 breast cancer cells. MUC1-specific T cells, induced by both fusion-cell types, lysed MUC1expressing targets and regressed established tumors in SCID mice (Koido et al. 2007c). Moreover, fusions of autologous DCs and allogeneic tumor cells were able to induce antigen-specific CTL with cytotoxic activity against autologous tumor cells (Koido et al. 2005a). This strategy has several advantages including (a) wellcharacterized TAA in allogeneic tumor cell lines, (b) unlimited source of tumor cells since allogeneic tumor cells grow well in the culture, and (c) there is no need to match the HLA type of patients and allogeneic tumor since TAA from allogeneic tumor cells can be processed and presented through the self MHC class I and class II of DC origins to T cells.
15.6.3 Fusion Cells Expressing Cytokines It has been observed that innate immunity plays a role in the generation of T-cell response to inflammatory antigens. Studies show that the pathogenassociated molecular patterns can activate NF-kappaB, resulting in the production of pro-inflammatory cytokines, such as IL-12, TNF-α and IL-1, and promoting
330
J. Gong and S. Koido
DC maturation (Kaisho and Akira 2001, Koido et al. 2007c, Michelsen et al. 2001, Schnare et al. 2001). Unlike infection by microbes, nonmutated self-antigens, the group to which most tumor-associated antigens belong, usually do not trigger innate immune responses and fail to produce pro-inflammatory chemokines and/or cytokines that are important in the breaking of T-cell tolerance and initiation of Tcell-mediated immune response. However, T-cell tolerance to nonmutated tumor antigen is not irreversible. Ample evidence indicates that tumor-antigen-specific T cells exist in cancer patients or can be induced by tumor vaccine. For example, T-cell tolerance to nonmutated tumor antigen MUC1 is reversed by FC expressing MUC1 (Gong et al. 1998). In an effort to enhance the antitumor immunity induced by FC, Cao et al. (1999) fused GM-CSF-expressing DC to B16 melanoma cells to generate B16–DC hybrid cells that express GM-CSF (B16/DC.GM). GM-CSF is one of the most crucial cytokines involved in DC maturation and functional regulation. Indeed, the B16–DC.GM fusion vaccine was able to induce a CTL response and protective immunity more potently and tended to be therapeutically more efficacious than the B16–DC vaccine, indicating a key role for GM-CSF expression. Using a similar approach, IL-2, IL-12 and/or IL-18-expressing FC were generated and corresponding enhancement of antitumor immunity was observed in mice immunized with these cytokine-producing FC (Iinuma et al. 2006, Ogawa et al. 2004). Using an alternative approach, Xia et al. (2004) fused DC to myeloma cells that express IL-18, a Th1 promoting cytokine. Immunization with IL-18-expressing FC provided protection against tumor challenge in 71% of mice. In contrast, the protection for mice immunized with FC not expressing IL-18 or irradiated myeloma cells with or without expressing IL-18 was 37, 25, or 0%, respectively. Similar findings were obtained in the fusion of DC and myeloma cells that express IL-4 (Liu et al. 2002, Xia et al. 2005); IL-4-expressing FC elicited stronger tumor-specific CTL response that was translated into protective immunity in all the vaccinated mice compared with FC without forced IL-4 expression. Recently, DC were fused with IL-12 gene-transferred cancer cells, thus generating IL-12-secreting FC (Suzuki et al. 2005). Immunization with such a gene-modified DC–tumor fusion vaccine elicited a markedly enhanced antitumor effect in the in vivo therapeutic model. Together, these experiments indicate that pro-immune cytokines are directly or indirectly involved in the induction of cellular immunity and can enhance the effectiveness of FC. Unlike purified cytokines used in an adjuvant context, cytokineexpressing FC secrete these cytokines directly at the site of FC interaction with T cells, thus creating a microenvironment conducive to the activation of functional T cells. Although promising in the animal studies, this approach has not been tested in the clinical setting for a variety of reasons, including the requirement of transfection of DC or tumor cells with vector encoding cytokine.
15.6.4 DC Maturation Dendritic cells are bone-marrow-derived leukocytes bearing a characteristic veiled morphology that excel in antigen presentation and the initiation of primary immune responses (Steinman 1991, 2001). These cells derive their potency from constitutive
15
Cell Fusion and Dendritic Cell-Based Vaccines
331
and inducible expression of essential costimulatory ligands on the cell surface including B7, ICAM-1, LFA-1, LFA-3, and CD40 (Inaba et al. 1994, Young et al. 1992) that function in concert as secondary signals essential for T-cell activation (Inaba et al. 1997, Thery and Amigorena 2001,Young and Inaba 1996). DC function depends on their level of maturation. Immature DC are located strategically at the port of potential pathogen entry where they take up antigen and migrate to secondary lymphoid tissues. During the migration, DC with captured antigen lose phagocytic capacity and acquire stimulatory ability in concert with a mature phenotype such as upregulation of costimulatory molecules. In DC-based tumor vaccine, it is recommended that mature DC be used (Shortman and Heath 2001, Steinman 2001). However, both immature and mature DC have been fused with tumor cells and antitumor immunity is induced by both types of fusion cells (Krause et al. 2002, Iinuma et al. 2004a, Liu et al. 2002, Phan et al. 2003, Takeda et al. 2003). This is probably related to the maturation of the DC moiety of the fusion cells triggered by the fusion, in a process similar to DC maturation after internalization of antigen. Shimizu et al. (2004) have found that fusions of tumor cells to either relatively immature DC from bone marrow or in-vivo-matured DC from spleens by injection of Flt-3L are effective in the treatment of three-day-old pulmonary metastasis. Phan and associates (2003) have compared the expression of MHC class II molecules in fusion cells from immature DC and from DC matured by lipopolysaccharide (LPS) and have found that MHC class II expression in some of the former hybrid cells is higher than in the latter. In this context, using immature DC as a fusion partner may be beneficial. Alternatively, some investigators used Toll-like receptor (TLR) agonists before, during, or after fusion. CpG ODN and OK-432 were used to treat DC and/or fusion cell products. Immature monocyte-derived DC stimulated by OK-432 overnight were fused to autologous tumor cells. In addition, fusion-cell preparations were stimulated with OK-432 for 2 days. These fusion cells stimulated with TLR agonists were more efficient in stimulating CD4+ and CD8+ T cells that were able to produce high levels of IFN-γ and lyse autologous tumor targets (Koido et al. 2007a, b, Ogihara et al. 2004). Du et al. (2006) used CpG ODN to stimulate DC for fusion. Their results show that fusion of CpG ODN-stimulated DC and tumor cells can induce more effective antitumor immune responses compared with fusions of non-CpG ODN-stimulated DC.
15.7 Fusion Cell Vaccines and Antitumor Immunity The strategy for DC–tumor fusion vaccine is based on the findings that DC are the most potent APC in the body, whereas tumor cells express abundant tumor antigens. The fusion of these two cell types through chemical (Akasaki et al. 2001, Cao et al. 1999, Gong et al. 1997b, 2002, Homma et al. 2001, 2005a, Kao et al. 2003, 2005, Li et al. 2001, Lindner and Schirrmacher 2002, Liu et al. 2002, Ogawa et al. 2004, Takeda et al. 2003, Wang et al. 1998, Xia et al. 2003, 2005, Zhang et al. 2003), physical (Goddard et al. 2003, Hayashi et al. 2002, Jantscheff et al.
332
J. Gong and S. Koido
2002, Lindner and Schirrmacher 2002, Marten et al. 2003, Scott-Taylor et al. 2000, Shimizu et al. 2004, Siders et al. 2003, Suzuki et al. 2005, Tanaka et al. 2002, Trefzer et al. 2005, Trevor et al. 2004) or biologic (Hiraoka et al. 2004, Phan et al. 2003) means creates a heterokaryon with both DC-derived costimulatory molecules, efficient antigen-processing and -presentation machinery, and tumor-derived antigens. Thus, the fusion cells have the essential elements for processing and presenting tumor antigens to host immune cells and for inducing effective antitumor response. The efficient antigen presentation, effectiveness of cytotoxic T lymphocyte (CTL) induction, and efficacy of antitumor immunity initiated by FC vaccines have been amply demonstrated (Galea-Lauri et al. 2002, Kao et al. 2005, Parkhurst et al. 2003, Phan et al. 2003, Shimizu et al. 2004, Trevor et al. 2004, Zhang et al. 2006). Galea-Lauri et al. (2002) have shown that FC are potent immune stimulators by comparing CTL activity induced by DC–tumor fusion cells with that induced by DC pulsed with either apoptotic tumor-cell fragments or tumor lysates: DC– tumor fusion cells induce the highest level of CTL activity against leukemia cells. In DC–tumor fusions, the tumor antigens can access the endogenous antigenprocessing pathway, while DC pulsed with apoptotic tumor-cell fragments or tumor lysates rely on antigen being cross-presented, which is usually not very efficient. Kokhaei et al. (2003), on the other hand, have observed that DC-endocytosed leukemic B-cell apoptotic bodies elicit stronger T-cell response than DC–tumor hybrids; the fusion efficiency in their study, however, was low. Parkhurst and coworkers (2003) have fused DC to allogeneic melanoma cells by electrofusion, and these fusion cells efficiently present MHC class I- and class II-restricted melanoma antigens. In contrast, DC pulsed with gp100 protein cannot cross-present MHC class I-restricted epitopes. Shimizu et al. (2004) have compared immunotherapeutic reactivities of DC loaded with a variety of antigen preparations, including DC pulsed with antigenic protein or peptide, tumor cell lysates, irradiated tumor cells, and DC– tumor fusions by electrofusion. These DC-based vaccines have been used to treat 3-day established pulmonary tumor nodules. A single intranodal vaccination with FC plus IL-12 produces a significant reduction in metastatic nodules, whereas other DC preparations are only marginally effective. The authors conclude that therapeutic vaccines from FC are superior to those involving other methods of DC loading.
15.7.1 Animal Studies Fusions of dendritic cells and tumor cells are increasingly used in tumor immunotherapy since emerging data indicate that such FC are potent immunogenic cells (Akasaki et al. 2001, Gong et al. 1997b, Phan et al. 2003, Shimizu et al. 2004, Trefzer et al. 2005) (Table 15.1). In our initial report on fusion-cell vaccines, murine MC38 adenocarcinoma cells stably transfected with human MUC1 were fused to syngeneic bone-marrow-derived DC (FC/MUC1) in the presence of polyethylene glycol (PEG). Immunization with FC/MUC1 induced antigen- and tumor-specific immune responses that eliminated established pulmonary metastases (Gong et al. 1997b). The therapeutic efficacy of DC–tumor fusion vaccine has been confirmed
Tumor
P815 Mastocytoma
B16 melanoma FBL-3 leukemia
B16 melanoma
Hepatocellular carcinoma Glioma
Plasmacytoma
DBA/2 mice
C57BL/6 mice
C57BL/6 mice
BALB/c mice
BALB/c mice
B10A mice
MC38/MUC1
C57BL/6 MUC1 transgenic mice
A. PEG-mediated fusions C57BL/6 mice MC38/MUC1
Animal
Fusion partners
Syngenic DC
Syngenic DC
Syngenic DC
Syngenic DC
Syngenic DC
Syngenic DC
Syngenic DC
Syngenic DC
DC
20–30%
39.9%
30%
12.7– 26.8%
15–25%
Data not shown
20–30%
20–30%
[6]
[5]
[4]
[3]
[2]
[1]
References
[7] Vaccination combined with rIL-12 injection prevented the growth of tumors injected s.c. and prolonged the survival of 50% mice injected i.v. with tumor cells Immunization of mice bearing multiple myeloma with fusion cells [8] prolonged the survival of mice. In addition, Immunization with FC combined with injection of IL-12 resulted in eradication of established tumors
Immunization with FC provided protection against challenge with tumor cells and eliminated 4-day established pulmonary metastasis Immunization induced cellular and humoral immune responses against MUC1, thus reversing the unresponsivenss of T cells to MUC1 tumor-associated antigen and eliminated 2 or 4-day established pulmonary metastasis Immunization with FC eliminated 3-day established intraperitoneal tumors in 50, 90 or 100% of the mice, respectively and provide long-term immunity Immunization with FC prevented tumor growth (104 cells/mouse) in 40% of mice. Adoptive transfer of lymph node cells primed by fusion cells and expanded ex vivo significantly reduced pulmonary metastatic nodules in B16 melanoma model and cured 80–100% of mice bearing 5-day established intraperitoneal FBL-3 tumor Immunization with FC protected 90% of mice against the challenge with B16 tumor cells, reduced the numbers of metastatic nodules in lungs, and eliminated 3-day pulmonary metastasis in 70% of the mice Immunization with FC rendered 80% of the mice free of the disease
Fusion efficiency Comments
Table 15.1 Results of animal studies from selected publications
15 Cell Fusion and Dendritic Cell-Based Vaccines 333
J558 myeloma cells transfected with IL-18
CT26 colon adenocarcinoma Hepatocellular carcinoma
BALB/c mice
BALB/c mice
C57BL/6, BALB/c and DBA/2 mice
C57BL/6 mice
B. Electrofusion C57BL/6 mice
Renca renal cell carcinoma B16 melanoma M3 melanoma
GL 261 glioma & melanoma B16 melanoma expressing β-gal
Spontaneous mammary carcinoma
C57Bl/6 MMT.Tg mice
BALB/c mice
Tumor
Animal
Fusion partners
>40%
5–20%
Syngenic & allogeneic DC
20–30%
35%
26%
20%
25%
Fusion efficiency
Syngenic DC
Syngenic DC
Syngenic DC
Syngenic DC
Syngenic DC
Syngenic DC
DC
[12]
[11]
[10]
[9]
References
Immunization with FC combined with IL-12 injection eradicated [13] 3-day established pulmonary or subcutaneous tumors in 100 and 80% of the mice, respectively A single vaccination with fusion cells plus IL-12 induced a [14] therapeutic immune response against 3-day established pulmonary metastases. The fusion approach is superior to other DC loading methods [15] Vaccination with fusion cell products prevented tumor growth of Renca, M3 and B16 melanoma cells in 50–75, 80–100 and 20–60% of mice, respectively. In addition, immunization with DC-Renca fusion cells eradicated 3-day established subcutaneous tumors
Prophylactic vaccination of mice that develop spontaneous mammary carcinomas with FC/MMT rendered 57–61% mice free of the disease for up to 180 days. In contrast, the control mice developed tumor between the ages of 65–108 days Immunization with FC expressing IL-18 protected 75% of the mice against challenge with tumor cells. By contrast, the protection for mice immunized with fusion cells without IL-18, irradiated tumor cells with or without expression of IL-18 was 37, 25 or 0%, respectively Immunization with FC or DC pulsed with tumor lysates prevented tumor growth in 75 and 30% of mice, respectively Immunization with FC combined with injection of IL-12 rendered 42% of mice bearing 3-day established subcutaneous tumor free of the disease
Comments
Table 15.1 (continued)
334 J. Gong and S. Koido
References
Immunization with FC combined with injection of CpG ODN [19] provides protection for 60% of mice against challenge with B16BL6 cells and 80% of mice against challenge with Renca cells
30%
Cell Fusion and Dendritic Cell-Based Vaccines
References: 1: Gong et al. (1997b); 2: Gong et al. (1998); 3: Lespagnard et al. (1998); 4: Wang et al. (1998); 5: Cao et al. (1999); 6: Homma et al. (2001); 7: Akasaki et al. (2001); 8: Gong et al. (2002); 9: Xia et al. (2003); 10: Xia et al. (2004); 11: Takeda et al. (2003); 12: Homma et al. (2005b); 13: Hayashi et al. (2002); 14: Tanaka et al. (2002); 15: Siders et al. (2003); 16: Kjaergaard et al. (2003); 17: Shimizu et al. (2004); 18: Phan et al. (2003); 19: Hiraoka et al. (2004); 20: Iinuma et al. (2006).
[20]
Treatment with single dose of 6-h hybrids led to long-term survival [18] of 60% of the mice
35–38.6%
Immunization with FC combined with injection of OX-40R mAb [16] rendered 60% of the mice bearing 3-day established subcutaneous tumor and 50% of the mice bearing 3-day intracranial tumor, respectively, free of tumors Immunization with either fusions of mature or immature DC to [17] tumor cells combined with injection of IL-12 significantly reduced the metastatic nodules in mice bearing 3-day established pulmonary tumors
Comments
D. PEG, Electrofusion or combined PEG and electrofusion A/J mice Neuroblastoma Syngenic DC Electrofusion: Immunization of mice with FC/IL-12/IL-18 provided 100% transduced 12.9% protection against challenge with tumor cells and eradicated with IL-12 and PEG: 34.1% 3-day established tumors in 60% of the mice IL-18 genes Combined fusion: 51.6%
C. Virus-mediated fusion C57BL/6 B16 melanoma Syngenic DC mice transfected with viral fusogenic membrane glycoprotein C57BL/6 & B16BL6 melanoma Syngenic DC BALB/c Renca mice
B16 melanoma cells expressing β-gal
Immature DC or 41.7–60% DC isolated from Flt-3L treated mice
Fusion efficiency
C57BL/6 mice
DC
Syngenic & 40–60% allogeneic DC
Tumor
C57BL/6 & MCA205 sarcoma BALB/c mice
Animal
Fusion partners
Table 15.1 (continued)
15 335
336
J. Gong and S. Koido
in subsequent studies in tumor models of B16 melanoma (Cao et al. 1999, Hiraoka et al. 2004, Li et al. 2001, Phan et al. 2003, Shimizu et al. 2004, Tanaka et al. 2002, Wang et al. 1998), mastocytoma P815 (Lespagnard et al. 1998), lung carcinoma (Celluzzi and Falo 1998), hepatocellular carcinoma (Homma et al. 2001, 2005b, Zhang et al. 2003), myeloma (Gong et al. 2002, Liu et al. 2002), mammary carcinoma (Lindner and Schirrmacher 2002), glioma (Akasaki et al. 2001, Hayashi et al. 2002), colon carcinoma (Kao et al. 2003, Suzuki et al. 2005, Takeda et al. 2003), renal cell carcinoma (Siders et al. 2003), and fibrosarcoma (Kjaergaard et al. 2003, Ogawa et al. 2004). In all these experiments, partial or complete tumor eradication was achieved, outcome depending on the tumor model, the age of established tumors, the protocol for preparation of the FC and/or the vaccination regimen. The pulmonary metastasis models are most frequently used with favorable results (Gong et al. 1997b, 1998, 2000c, Cao et al. 1999, Li et al. 2001, Phan et al. 2003, Tanaka et al. 2002, Wang et al. 1998). However, studies show that immunization with FC vaccine is also effective in the treatment of established intracranial and skin tumors when combined with an adjuvant (Akasaki et al. 2001, Shimizu et al. 2004). Shimizu et al. (2004) have demonstrated that intrasplenic immunization with FC combined with intraperitoneal injection of OX-40R mAb renders 60% of mice with established skin tumor and 50% of mice with three-day-old intracranial tumor free of disease.
15.7.2 Clinical Trials Based on the results from animal studies, initial phase I clinical trials with FC immunization were conducted (Table 15.2). Krause et al. (2002) fused DC to autologous melanoma cells by electrofusion and assessed the use of such FC as a vaccine in 17 patients with disseminated melanoma refractory to standard therapy. There were no serious side effects associated with the administration of the vaccine. Localized hair depigmentation occurred in one case. In the 13 evaluable patients, one patient each achieved partial response and one had stable disease for 6 months. In another patient, some of the metastases were regressing despite overall progression of the disease. The authors attributed their findings to several factors, including limited tumor-cell viability, limited numbers of DC and tumor cells, immaturity of DC, and the type of vaccination protocol. In an alternative approach, Trefzer et al. (2005) fused malignant melanoma cells with allogeneic DC. In 17 patients, they observed one complete response, one partial response, and six responses of stable disease with long survival times. In addition, 11 of 14 patients, clinical responders and non-responders alike, mounted high-frequency T-cell responses to various tumorassociated antigens. Kikuchi et al. (2001), in a phase I clinical trial in patients with malignant glioma, found that immunization with DC–glioma fusion vaccine produced partial responses in two of six patients. In a similar trial by the same group, FC in combination with IL-12 was administered to patients with malignant brain tumor, breast cancer, gastric cancer, colorectal cancer, ovarian carcinoma, or melanoma.
15
Cell Fusion and Dendritic Cell-Based Vaccines
337
Table 15.2 Results of clinical trial in patients with advanced stage of malignant tumors Fusion partners Clinical Patients (n) responsea
References
Autologous
16
[1]
Melanoma
Autologous Autologous
17
Melanoma
Allogeneic
Melanoma
Autologous Autologous
Glioma
Autologous Autologous
Glioma
Autologous Autologous
rh IL-12
12
Gastric/ Colorectal carcinoma Ovarian carcinoma Breast cancer
Autologous Autologous
rh IL-12
3
Autologous Autologous
rh IL-12
3
Autologous Autologous
rh IL-12
2
Breast cancer
Autologous Autologous
10
Renal cell carcinoma Renal cell carcinoma
Autologous Autologous
13
Allogeneic
Autologous
20
Renal cell carcinoma Renal cell carcinoma
Allogeneic
Autologous or Allogeneic Autologous
12
Tumor
DC
Tumor
Melanoma
Allogeneic
Allogeneic
Autologous
Hepatocellular Autologous Autologous carcinoma
Adjuvant
rh IL-2
11
rh IL-12
4 8
10
1
1 (CR), 1 (PR), 5 (SD), 9 (PD) 1 (PR), 1 (SD), 15 (PD) 1 (SD), 10 (PD) 4 (PD) 2 (PR), 1 (SD), 5 (PD) 3 (PR), 2 (MR), 4 (SD), 3 (PD) 1 (SD), 2 (PD) 2 (SD), 1 (PD) 1 (SD), 1 (PD) 2 (PR), 1 (SD), 7 (PD) 5 (SD), 8 (PD) 2 (PR), 8 (SD), 10 (PD) 4 (SD), 8 (PD) 1 (PR), 6 (SD), 3 (PD) 1 (PD)
[2]
[3] [4] [4]
[4, 5]
[4]
[4] [4] [6]
[6] [7]
[8] [9]
[10]
References:1: Trefzer et al. (2005); 2: Krause et al. (2002); 3: Haenssle et al. (2004); 4: Homma et al. (2005a); 5: Kikuchi et al. (2004); 6: Avigan et al. (2004); 7: Avigan et al. (2007); 8: Marten et al. (2003); 9: Zhou et al. (2009); 10: Koido et al. (2008). a CR, complete response; PR, partial response; MR, mixed response; SD, stable disease; PD, progressive disease.
338
J. Gong and S. Koido
Three of 12 patients with malignant brain tumor achieved a partial response and one patient a minor response (Homma et al. 2005a, Kikuchi et al. 2004), but the response to other types of malignant tumors was muted. The relatively favorable response to malignant brain tumors is of particular interest since the central nervous system is generally considered an immunologically privileged site as a result of the lack of lymphatic drainage and the nature of the blood–brain barrier (Homma et al. 2005a, Kikuchi et al. 2004). However, these studies demonstrate that appropriately activated T cells can cross the blood–brain barrier to access tumor and initiate tumor regression. Avigan et al. (2004) tested FC vaccine in 23 patients with metastatic breast and renal cancer. Immunologic and clinical responses to the vaccination were observed in a subset of patients. Most notably, two patients with breast cancer exhibited disease regression, including a nearly complete response of a large chest-wall mass. Five patients with renal carcinoma and one patient with breast cancer had disease stabilization. In a subsequent trial, renal cell carcinoma cells were fused to allogeneic DC using electrofusion with fusion efficiency of 20% (Avigan et al. 2007). The vaccine was well-tolerated, and vaccination resulted in antitumor immune response in 10/21 evaluable patients. Partial clinical response was demonstrated in two patients and stabilization of disease in eight patients. Marten et al. (2003) observed immunologic, but not clinical, responses in 12 patients with metastatic renal cell carcinoma (RCC) treated with DC–tumor fusion vaccine. Zhou et al. (2009) reported that DC-tumor fusion cell vaccine was safe and able to elicit immunological responses in a significant portion of patients with RCC. Haenssle et al. (2004) failed to find unequivocal beneficial effects from fusion cell products generated by fusion of HLA class I-mismatched DC from healthy donor to patient-derived melanoma cells. It is apparent that the antitumor response induced by FC vaccines in humans is muted compared with that in the animal studies. Multiple factors may be involved in the differential antitumor reaction between animal models and humans. One obvious difference is that the FC vaccine has been used in patients with a large burden of cancers and/or metastasis. There is a large body of evidence indicating that tumor exerts immunosuppression through alteration of APC, promotion of inhibitory molecules, and expansion of regulatory T cells in general and in the tumor microenvironment particular (Zou 2005). In patients with advanced stages of tumor progression, the tumor microenvironment favors the production of inhibitory molecules including vascular endothelial growth factor (VEGF), IL-6, IL-10 and transforming growth factor-β (TGF-β). These molecules work individually and in concert to promote the recruitment, differentiation, maturation and survival of inhibitory DC subsets or dysfunctional DC. These DC tend to expand the regulatory T cells, which inhibit the expansion and function of tumor-reactive effector T cells. It is likely that the tumor microenvironment plays a role in the muted clinical response to FC vaccination. Indeed, the supernatant from human hepatocellular carcinoma (HCC) cells induced functional impairment of DC as demonstrated by the down-regulation of MHC class I and class II, CD80, CD86, and CD83 molecules (Koido et al. 2008). Moreover, DC exposed to the supernatant from HCC cells secreting TGF-β failed to undergo full maturation upon stimulation with TLR 4 agonist, OK-432. These
15
Cell Fusion and Dendritic Cell-Based Vaccines
339
inhibitory molecules by tumor cells may also exert suppressive effect for DCtumor fusion vaccine. For example, tumor-derived TGF-β produced by tumor cells significantly reduced the potency of DC-tumor fusion cell vaccines (Kao et al. 2003) and the blockade of tumor-derived TGF-β reduced the Treg induction by the DC-tumor fusion cell vaccine and enhanced antitumor immunity (Zhang et al. 2008). It is conceivable that a major obstacle in the development of active immunotherapy for cancer is the immunosuppressive environment and inhibitory molecules produced by the growing tumor. In this context, concomitant immunotherapy and cytoreduction therapy, such as chemotherapy or therapy targeting the induction of regulatory T cells, may represent a promising approach. Taken together, DC–tumor fusion vaccine may have certain advantages over other DCbased vaccines. However, a strategy is needed to enhance the therapeutic efficacy of the vaccine and provide clinical benefit to patients.
15.8 Promotion of Antitumor Immunity 15.8.1 Using Adjuvant with Fusion Vaccine The induction of functional CTL requires the interaction of TCR and antigenic peptide–MHC complex presented on the APC in the context of co-stimulatory molecules. In addition, pro-inflammatory cytokines are also needed to overcome the regulation of tolerance. Fusions of DC and tumor cells generally produce proinflammatory cytokines such as IL-12 at a level comparable to that of DC (Gong et al. 2002, Homma et al. 2005a). However, the amounts of the cytokine produced by FC may be insufficient for optimal induction of sustainable CTL. In fact, a mixed T-cell response characterized by the expansion of both activated and regulatory T cells has been induced by DC–breast cancer fusion cells (Koido et al. 2008, Vasir et al. 2008). The latter may exert inhibitory effect on CTL. To address potential problems, cytokines were used as an adjuvant during the priming phase of T cells by FC. Study shows that coculture of T cells with DC–tumor fusion cells in the presence of IL-12, IL-18 or CPG ODN reduces the expansion of regulatory T cells and promotes the expansion of activated effector cells (Vasir et al. 2008). Alternatively, regulatory T cells can be removed using a low dose of chemotherapy or neoadjuvant chemotherapy (Ladoire et al. 2008, Menard et al. 2008, Zou 2006). The most widely used cytokine is IL-12 (Gong et al. 2002, Homma et al. 2005a, Akasaki et al. 2001, Hayashi et al. 2002, Iinuma et al. 2004b), which is involved in the up-regulation of co-stimulatory molecules in DC, promotion of Th1 immunity, and inhibition of induction of anergy (Coughlin et al. 1995, Grohmann et al. 1997, Mountford et al. 1996). Co-administration of IL-12 potentiated the CTL and T-cell proliferative responses to FC vaccination that were translated into better therapeutic efficacy or protection of mice against challenge with tumor cells (Gong et al. 2002, Homma et al. 2005a, Akasaki et al. 2001, Hayashi et al. 2002, Iinuma et al. 2004b). An alternative approach is the use of TLR agonists, which induce DC activation, leading to the production of pro-inflammatory cytokines and
340
J. Gong and S. Koido
the up-regulation of co-stimulatory molecules (Okamoto et al. 2004). For example, the synthetic oligodeoxynucleotides containing specific bacterial unmethylated CpG motif (CpG ODN) induce Th1 immune responses through TLR 9, resulting in the production of pro-inflammatory cytokines IL-12, TNF-α and IFN-γ that promote the development of antitumor immunity (Heckelsmiller et al. 2002). Hiraoka and colleagues (2004) injected FC admixed with CpG ODN into murine tumor models. Co-administration of FC and CpG ODN resulted in increased production of IFN-γ and IL-12, enhanced induction of tumor-specific CTL, and improved protection of mice against lethal subcutaneous tumor challenge and spontaneous lung metastasis. Importantly, long-term antitumor immunity was maintained in mice immunized with FC and CpG ODN, but not FC or CpG ODN alone. Paired, but not solitary combinations of polyinosine:polycytadilic acid (TLR3 agonist) and CpG DNA (TLR9 agonist) stimulated IL-12 secretion from DCs in vitro and synergized with vaccination to achieve potent tumor rejection Zheng et al. 2008). Another example of the use of TLR agonists in this context is OK-432, a penicillin-inactivated and lyophilized preparation of the low-virulence strain of Streptococcus pyogenes (group A). OK-432 is one of the biological response modifiers and a good manufacturing practice grade agent. It has been demonstrated that the compound promotes activation of DC through TLR4 and β2 integrin to enhance antigen-specific CTL responses to a greater extent than does a previously reported mixture consisting of TNF-α, IL-1β, IL-6, and PGE2 (Nakahara et al. 2003, Okamoto et al. 2004). It has been used without apparent side effects as an adjuvant for patients with cancer (Yamanaka et al. 2005), activates macrophages, lymphocytes, and NK cells by inducing multiple cytokines including IL-12 and IFN-γ, and polarizes the T-cell response to a Th1dominant state (Grohmann et al. 1997, Nakahara et al. 2003). These reports suggest that the combination of FC and OK-432 as an adjuvant may provide a more effective and feasible cancer vaccine and indicate the importance of innate immunity in the promotion of adaptive antitumor immunity.
15.8.2 Combined Approaches Dendritic–tumor fusion cells can be used as a vaccine to activate presumably existing, yet limited, tumor-specific CTL precursors in a host. Alternatively, the fusion cells can be used as stimulators to induce and expand ex vivo to therapeutic numbers tumor-specific CTL, which can then be transferred to the tumor-bearing host for adoptive immunotherapy. This approach, though not yet tested in the clinical setting, has potential to enhance the antitumor immunity induced by FC. Recent studies show that DC–tumor fusion cells are potent T-cell stimulators: DC fused to primary tumor cells from patients with ovarian, breast, malignant glioma, leukemia, melanoma, and multiple myeloma stimulate strong CTL activity against autologous tumor cells (Galea-Lauri et al. 2002, Goddard et al. 2003, Gong et al. 2000a, b, 2004, Kikuchi et al. 2001, Raje et al. 2004). In addition to the induction of CTL, DC–tumor fusion cells also possess properties to activate CD4+ T cells, which are required for the maintenance of CTL
15
Cell Fusion and Dendritic Cell-Based Vaccines
341
(Dranoff et al. 1993, Gong et al. 1997b, Hung et al. 1998, Marzo et al. 1999, 2000, Ossendorp et al. 1998) and in the effector/memory phase (Chaux et al. 1999, Greenberg 1991, Mumberg et al. 1999, Qin and Blankenstein 2000). These cells are needed to maintain the numbers and cytotoxic capacity of CD8+ T cells and to promote the infiltration of CD8+ T cells in tumor (Marzo et al. 1999, 2000). We observed that fusion cells migrate to draining lymph nodes after subcutaneous injection and are closely associated with CD4+ and CD8+ T cells (Koido et al. 2002). These properties significantly increase the probability of formation of the three-cell cluster composed of fusion cell, CD4+ and precursor CTL. Although this cluster is not required for the induction of CTL (Ridge et al. 1998), simultaneous recognition of cognate peptides presented by MHC class I and class II molecules on APC is more effective in the induction of CTL (Behrens et al. 2004, Bennett et al. 1997, Mitchison 1990). Based on these results, it is conceivable that T cells stimulated by FC can be used for adoptive therapy. Indeed, the efficacy of FC-activated T cells has been demonstrated in animal models (Tanaka et al. 2004, Wang et al. 1998) in which adoptive transfer of T lymph node cells isolated from FC-vaccinated mice has eliminated established pulmonary metastasis (Gong et al. 2000c, Tanaka et al. 2004). Wang and associates (1998) show that adoptive transfer of T-LNC derived from mice immunized with DC–B16 fusion vaccine into B16 tumor-bearing mice greatly reduced the number of established pulmonary metastases. Furthermore, adoptive transfer of DC/RMA-S vaccine-primed, cultured T-LNC eradicated disseminated FBL-3 tumor. Collectively, these findings indicate the therapeutic usage of T cells stimulated by FC either in vitro or in vivo and raise the possibility of dual usage of DC–tumor fusion vaccine in antitumor immunity: FC can be used as a vaccine to activate tumor-specific T cells in a host or as T-cell stimulators to generate ex vivo tumor-specific T cells for adoptive use. The dual usage of FC-based active and adoptive immunotherapy was reported by Tamai and associates (2008). Instead of vaccination with FC, T cells were selected from mice inoculated with 4T1 tumor cells and expanded ex vivo. These T cells were adoptively transferred to host bearing metastatic 4T1 tumor in conjunction with DC–tumor vaccination. Immunization with DC–tumor fusion vaccine or adoptive transfer of T-LNC alone was effective in reduction of tumor metastases but insufficient to eradicate tumor. In contrast, combined active immunization with DC–tumor vaccine and adoptive transfer of activated T-LNC rendered 8 of 17 mice free of tumors. Such FC-based active and adoptive immunotherapy may represent a promising strategy in the management of cancer with metastasis.
15.9 Summary Ample evidence indicates that DC–tumor fusion cells constitute a potent tumor vaccine. Yet, we are still searching for optimal ways to use FC-based immunotherapy to benefit patients significantly. Such a gap reflects, at least in part, the variations
342
J. Gong and S. Koido
associated with DC–tumor fusion vaccine. Fusion technique is a versatile approach in the design of tumor vaccine and can be applicable to nearly all types of tumor cells using a variety of methods. However, such versatility and wide applicability lead to variations that make it more difficult to standardize the vaccine. Although there is considerable variation in the phenotype of the parent cells and fusogenicity of different type of tumor cells, certain quality control measures are desirable. These quality controls include FC phenotype, cell viability, fusion efficiency, and stability. For example, the range of fusion cells in the fusion-cell product that constitutes a fusion vaccine needs to be defined. Such quality control may be essential in evaluation of DC–tumor fusion vaccines since a vaccine with poor quality can still elicit antitumor immunity, albeit at a lower level. The second challenge facing FC vaccine is how to benefit patients. We have observed significant therapeutic effect with FC vaccination in animal models, whereas in humans, only limited therapeutic results are obtained. A major reason is that tumor vaccines, including FC-based vaccines, are tested as a therapy in patients with advanced stages of disease and large tumor burden refractory to standard therapy. It is conceivable that in this group of patients the tumor vaccines are engaged in an uphill battle against tumor-induced and/or therapy-induced immunosuppression (Finn 2003). This explains, at least in part, why differential therapeutic efficacy of FC vaccination has been achieved in animal compared with human studies. To circumvent this problem, the FC vaccine may be tested in cancer patients in remission to prevent recurrence and metastasis. Alternatively, FC-based immunotherapy may be used in combination with other therapies including surgery and radiotherapy.
References Ahkong QF, Baldwin JM, O’Reilly R et al (1994) Interactions between metal ions and poly(ethylene glycol) in the fusion of human erythrocytes. Mol Membr Biol 11:171–179 Ahkong QF, Lucy JA (1986) Osmotic forces in artificially induced cell fusion. Biochim Biophys Acta 858:206–216 Akasaki Y, Kikuchi T, Homma S et al (2001) Antitumor effect of immunizations with fusions of Dendritic and Glioma cells in a mouse brain tumor model. J Immunother 24:106–113 Avigan D, Vasir B, Gong J et al (2004) Fusion cell vaccination of patients with metastatic breast and renal cancer induces immunological and clinical responses. Clin Cancer Res 10:4699–4708 Avigan DE, Vasir B, George DJ et al (2007) Phase I/II study of vaccination with electrofused allogeneic dendritic cells/autologous tumor-derived cells in patients with stage IV renal cell carcinoma. J Immunother 30:749–761 Bateman A, Bullough F, Murphy S et al (2000) Fusogenic membrane glycoproteins as a novel class of genes for the local and immune-mediated control of tumor growth. Cancer Res 60:1492–1497 Bateman AR, Harrington KJ, Kottke T et al (2002) Viral fusogenic membrane glycoproteins kill solid tumor cells by nonapoptotic mechanisms that promote cross presentation of tumor antigens by dendritic cells. Cancer Res 62:6566–6578 Behrens G, Li M, Smith CM et al (2004) Helper T cells, dendritic cells and CTL Immunity. Immunol Cell Biol 82:84–90 Bennett SR, Carbone FR, Karamalis F et al (1997) Induction of a CD8+ cytotoxic T lymphocyte response by cross-priming requires cognate CD4+ T cell help. J Exp Med 186:65–70
15
Cell Fusion and Dendritic Cell-Based Vaccines
343
Berard F, Blanco P, Davoust J et al (2000) Cross-priming of naive CD8 T cells against melanoma antigens using dendritic cells loaded with killed allogeneic melanoma cells. J Exp Med 192: 1535–1544 Bischoff R, Eisert RM, Schedel I et al (1982) Human hybridoma cells produced by electro-fusion. FEBS Lett 147: 64–68 Cao X, Zhang W, Wang J et al (1999) Therapy of established tumour with a hybrid cellular vaccine generated by using granulocyte-macrophage colony-stimulating factor genetically modified dendritic cells. Immunology 97:616–625. Celluzzi CM, Falo LD Jr (1998) Physical interaction between dendritic cells and tumor cells results in an immunogen that induces protective and therapeutic tumor rejection. J Immunol 160:3081–3085 Celluzzi CM, Mayordomo JI, Storkus WJ et al (1996) Peptide-pulsed dendritic cells induce antigen-specific CTL-mediated protective tumor immunity. J Exp Med 183:283–287 Chaux P, Vantomme V, Stroobant V et al (1999) Identification of MAGE-3 epitopes presented by HLA-DR molecules to CD4(+) T lymphocytes. J Exp Med 189:767–778 Chen EH, Olson EN (2004) Towards a molecular pathway for myoblast fusion in Drosophila. Trends Cell Biol 14:452–460 Chomez P, De Backer O, Bertrand M et al (2001) An overview of the MAGE gene family with the identification of all human members of the family. Cancer Res 61:5544–5551 Condon C, Watkins SC, Celluzzi CM et al (1996) DNA-based immunization by in vivo transfection of dendritic cells. Nature Med 2:1122–1128 Coughlin CM, Wysocka M, Kurzawa HL et al (1995) B7-1 and interleukin 12 synergistically induce effective antitumor immunity. Cancer Res 55:4980–4987 Das D, Suresh MR (2005) Producing bispecific and bifunctional antibodies. Methods Mol Med 109:329–346 Davidson RL, Gerald PS (1976) Improved techniques for the induction of mammalian cell hybridization by polyethylene glycol. Somatic Cell Genet 2:165–176 de StGroth SF, Scheidegger D (1980) Production of monoclonal antibodies: strategy and tactics. J Immunol Methods 35:1–21 Dranoff G, Jaffee E, Lazenby A et al (1993) Vaccination with irradiated tumor cells engineered to secrete murine granulocyte-macrophage colony-stimulating factor stimulates potent, specific, and long-lasting anti-tumor immunity. Proc Natl Acad Sci USA 90:3539–3543 Du YC, Lin P, Zhang J et al (2006) Fusion of CpG-ODN-stimulating dendritic cells with Lewis lung cancer cells can enhance anti-tumor immune responses. Tissue Antigens 67:368–376 Ebert O, Shinozaki K, Kournioti C et al (2004) Syncytia induction enhances the oncolytic potential of vesicular stomatitis virus in virotherapy for cancer. Cancer Res 64:3265–3270 Fabre JW (2001) The allogeneic response and tumor immunity. Nat Med 7:649–652 Finn OJ (2003) Cancer vaccines: between the idea and the reality. Nat Rev Immunol 3:630–641 Ford WL, Atkins RC (1973) The proportion of lymphocytes capable of recognizing strong transplantation antigens in vivo. Adv Exp Med Biol 29:255–262 Ford WL, Simmonds SJ, Atkins RC (1975) Early cellular events in a systemic graft-vs.-host reaction. II. Autoradiographic estimates of the frequency of donor lymphocytes which respond to each Ag-B-determined antigenic complex. J Exp Med 141:681–696 Frye LD, Edidin M (1970) The rapid intermixing of cell surface antigens after formation of mousehuman heterokaryons. J Cell Sci 7:319–335 Furusawa M, Nishimura T, Yamaizumi M et al (1974) Injection of foreign substances into single cells by cell fusion. Nature 249:449–450 Gabrijel M, Repnik U, Kreft M et al (2004) Quantification of cell hybridoma yields with confocal microscopy and flow cytometry. Biochem Biophys Res Commun 314:717–723 Galea-Lauri J, Darling D, Mufti G et al (2002) Eliciting cytotoxic T lymphocytes against acute myeloid leukemia- derived antigens: evaluation of dendritic cell-leukemia cell hybrids and other antigen-loading strategies for dendritic cell-based vaccination. Cancer Immunol Immunother 51:299–310
344
J. Gong and S. Koido
Galfre G, Howe SC, Milstein C et al (1977) Antibodies to major histocompatibility antigens produced by hybrid cell lines. Nature 266:550–552 Goddard RV, Prentice AG, Copplestone JA et al (2003) In vitro dendritic cell-induced T cell responses to B cell chronic lymphocytic leukaemia enhanced by IL-15 and dendritic cell-B-CLL electrofusion hybrids. Clin Exp Immunol 131:82–89 Gong J, Apostolopoulos V, Chen D et al (2000c) Selection and characterization of MUC1-specific CD8+ T cells from MUC1 transgenic mice immunized with dendritic-carcinoma fusion cells. Immunology 101:316–324 Gong J, Avigan D, Chen D et al (2000a) Activation of antitumor cytotoxic T lymphocytes by fusions of human dendritic cells and breast carcinoma cells. Proc Natl Acad Sci USA 97:2715–2718 Gong J, Chen D, Kashiwaba M et al (1997b) Induction of antitumor activity by immunization with fusions of dendritic and carcinoma cells. Nat Med 3:558–561 Gong J, Chen D, Kashiwaba M et al (1998) Reversal of tolerance to human MUC1 antigen in MUC1 transgenic mice immunized with fusions of dendritic and carcinoma cells. Proc Natl Acad Sci USA 95:6279–6283 Gong J, Chen L, Chen D et al (1997a) Induction of antigen-specific antitumor immunity with adenovirus- transduced dendritic cells. Gene Ther 4:1023–1028 Gong J, Koido S, Chen D et al (2002) Immunization against murine multiple myeloma with fusions of dendritic and plasmacytoma cells is potentiated by interleukin 12. Blood 99:2512–2517 Gong J, Koido S, Kato Y et al (2004) Induction of anti-leukemic cytotoxic T lymphocytes by fusion of patient-derived dendritic cells with autologous myeloblasts. Leuk Res 28:1303–1312 Gong J, Nikrui N, Chen D et al (2000b) Fusions of human ovarian carcinoma cells with autologous or allogeneic dendritic cells induce antitumor immunity. J Immunol 165:1705–1711 Gottfried E, Krieg R, Eichelberg C et al (2002) Characterization of cells prepared by dendritic cell-tumor cell fusion. Cancer Immun 2:15 Greenberg PD (1991) Adoptive T cell therapy of tumors: mechanisms operative in the recognition and elimination of tumor cells. Adv Immunol 49:281–355 Greenlee RT, Murray T, Bolden S, et al (2000) Cancer statistics, 2000. CA Cancer J Clin 50:7–33 Grohmann U, Bianchi R, Ayroldi E et al (1997) A tumor-associated and self antigen peptide presented by dendritic cells may induce T cell anergy in vivo, but IL-12 can prevent or revert the anergic state. J Immunol 158:3593–3602 Guccion JG, Enzinger FM (1972) Malignant giant cell tumor of soft parts. An analysis of 32 cases. Cancer 29:1518–1529 Haenssle HA, Krause SW, Emmert S et al (2004) Hybrid cell vaccination in metastatic melanoma: clinical and immunologic results of a phase I/II study. J Immunother 27:147–155 Hayashi T, Tanaka H, Tanaka J et al (2002) Immunogenicity and therapeutic efficacy of dendritictumor hybrid cells generated by electrofusion. Clin Immunol 104:14–20 Heath WR, Carbone FR (2001) Cross-presentation, dendritic cells, tolerance and immunity. Annu Rev Immunol 19:47–64 Heckelsmiller K, Beck S, Rall K et al (2002) Combined dendritic cell- and CpG oligonucleotidebased immune therapy cures large murine tumors that resist chemotherapy. Eur J Immunol 32:3235–3245 Hiraoka K, Yamamoto S, Otsuru S et al (2004) Enhanced tumor-specific long-term immunity of hemagglutinating [correction of hemaggluttinating] virus of Japan-mediated dendritic celltumor fused cell vaccination by coadministration with CpG oligodeoxynucleotides. J Immunol 173:4297–4307 Holmes LM, Li J, Sticca RP et al (2001) A rapid, novel strategy to induce tumor cell-specific cytotoxic T lymphoctye responses using instant dentritomas. J Immunother 24:122–129 Homma S, Kikuchi T, Ishiji N et al (2005a) Cancer immunotherapy by fusions of dendritic and tumour cells and rh-IL-12. Eur J Clin Invest 35:279–286 Homma S, Komita H, Sagawa Y et al (2005b) Antitumour activity mediated by CD4+ cytotoxic T lymphocytes against MHC class II-negative mouse hepatocellular carcinoma induced by dendritic cell vaccine and interleukin-12. Immunology 115:451–461
15
Cell Fusion and Dendritic Cell-Based Vaccines
345
Homma S, Matai K, Irie M et al (2003) Immunotherapy using fusions of autologous dendritic cells and tumor cells showed effective clinical response in a patient with advanced gastric carcinoma. J Gastroenterol 38:989–994 Homma S, Toda G, Gong J et al (2001) Preventive antitumor activity against hepatocellular carcinoma (HCC) induced by immunization with fusions of dendritic cells and HCC cells in mice. J Gastroenterol 36:764–771 Hsu FJ, Benike C, Fagnoni F et al (1996) Vaccination of patients with B-cell lymphoma using autologous antigen- pulsed dendritic cells. Nat Med 2:52–58 Hung K, Hayashi R, Lafond-Walker A et al (1998) The central role of CD4(+) T cells in the antitumor immune response. J Exp Med 188:2357–2368 Iinuma H, Okinaga K, Fukushima R et al (2004a) [Reduction of immunosuppression and shift to Th1 response by tumor-DC (dendritic cells) fusion vaccine]. Gan To Kagaku Ryoho 31: 1640–1642 Iinuma H, Okinaga K, Fukushima R et al (2006) Superior protective and therapeutic effects of IL-12 and IL-18 gene-transduced dendritic neuroblastoma fusion cells on liver metastasis of murine neuroblastoma. J Immunol 176:3461–3469 Iinuma T, Homma S, Noda T et al (2004b) Prevention of gastrointestinal tumors based on adenomatous polyposis coli gene mutation by dendritic cell vaccine. J Clin Invest 113:1307–1317 Inaba K, Inaba M, Romani N et al (1992) Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor. J Exp Med 176:1693–1702 Inaba K, Pack M, Inaba M et al (1997) High levels of a major histocompatibility complex II-self peptide complex on dendritic cells from the T cell areas of lymph nodes. J Exp Med 186: 665–672 Inaba K, Witmer-Pack M, Inaba M et al (1994) The tissue distribution of the B7-2 costimulator in mice: abundant expression on dendritic cells in situ and during maturation in vitro. J Exp Med 180:1849–1860 Islam MQ, Meirelles Lda S, Nardi NB et al (2006) Polyethylene glycol-mediated fusion between primary mouse mesenchymal stem cells and mouse fibroblasts generates hybrid cells with increased proliferation and altered differentiation. Stem Cells Dev 15:905–919 Jantscheff P, Spagnoli G, Zajac P et al (2002) Cell fusion: an approach to generating constitutively proliferating human tumor antigen-presenting cells. Cancer Immunol Immunother 51:367–375 Jaroszeski MJ, Gilbert R, Fallon PG, et al (1994) Mechanically facilitated cell–cell electrofusion. Biophys J 67:1574–1581 Kaisho T, Akira S (2001) Dendritic-cell function in Toll-like receptor- and MyD88-knockout mice. Trends Immunol 22:78–83 Kao JY, Gong Y, Chen CM et al (2003) Tumor-derived TGF-beta reduces the efficacy of dendritic cell/tumor fusion vaccine. J Immunol 170:3806–3811 Kao JY, Zhang M, Chen CM, et al (2005) Superior efficacy of dendritic cell-tumor fusion vaccine compared with tumor lysate-pulsed dendritic cell vaccine in colon cancer. Immunol Lett 101:154–159 Kikuchi T, Akasaki Y, Abe T et al (2004) Vaccination of glioma patients with fusions of dendritic and glioma cells and recombinant human interleukin 12. J Immunother 27:452–459 Kikuchi T, Akasaki Y, Irie M et al (2001) Results of a phase I clinical trial of vaccination of glioma patients with fusions of dendritic and glioma cells. Cancer Immunol Immunother 50:337–344 Kinzler ER, Compton T (2005) Characterization of human cytomegalovirus glycoprotein-induced cell–cell fusion. J Virol 79:7827–7837 Kjaergaard J, Shimizu K, Shu S (2003) Electrofusion of syngeneic dendritic cells and tumor generates potent therapeutic vaccine. Cell Immunol 225:65–74 Knutton S (1979) Studies of membrane fusion. III. Fusion of erythrocytes with polyethylene glycol. J Cell Sci 36:61–72 Kohler G, Milstein C (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256:495–497
346
J. Gong and S. Koido
Kohler G, Milstein C (1976) Derivation of specific antibody-producing tissue culture and tumor lines by cell fusion. Eur J Immunol 6:511–519 Koido S, Hara E, Homma S et al (2005a) Dendritic cells fused with allogeneic colorectal cancer cell line present multiple colorectal cancer-specific antigens and induce antitumor immunity against autologous tumor cells. Clin Cancer Res 11:7891–7900 Koido S, Hara E, Homma S et al (2007a) Streptococcal preparation OK-432 promotes fusion efficiency and enhances induction of antigen-specific CTL by fusions of dendritic cells and colorectal cancer cells. J Immunol 178:613–622 Koido S, Hara E, Homma S et al (2007b) Synergistic induction of antigen-specific CTL by fusions of TLR-stimulated dendritic cells and heat-stressed tumor cells. J Immunol 179:4874–4883 Koido S, Hara E, Torii A et al (2005b) Induction of antigen-specific CD4- and CD8-mediated T-cell responses by fusions of autologous dendritic cells and metastatic colorectal cancer cells. Int J Cancer 117:587–595 Koido S, Homma S, Hara E et al (2008) In vitro generation of cytotoxic and regulatory T cells by fusions of human dendritic cells and hepatocellular carcinoma cells. J Transl Med 6:51 Koido S, Kashiwaba M, Chen D et al (2000) Induction of antitumor immunity by vaccination of dendritic cells transfected with MUC1 RNA. J Immunol 165:5713–5719 Koido S, Ohana M, Liu C et al (2004) Dendritic cells fused with human cancer cells: morphology, antigen expression, and T cell stimulation. Clin Immunol 113:261–269 Koido S, Tanaka Y, Chen D et al (2002) The kinetics of in vivo priming of CD4 and CD8 T cells by dendritic/tumor fusion cells in MUC1-transgenic mice. J Immunol 168:2111–2117 Koido S, Tanaka Y, Tajiri H et al (2007c) Generation and functional assessment of antigen-specific T cells stimulated by fusions of dendritic cells and allogeneic breast cancer cells. Vaccine 25:2610–2619 Kokhaei P, Rezvany MR, Virving L et al (2003) Dendritic cells loaded with apoptotic tumour cells induce a stronger T-cell response than dendritic cell-tumour hybrids in B-CLL. Leukemia 17:894–899 Krause SW, Neumann C, Soruri A et al (2002) The treatment of patients with disseminated malignant melanoma by vaccination with autologous cell hybrids of tumor cells and dendritic cells. J Immunother 25:421–428 Kuby J (1992) Hybridoma and monoclonal antibody. In: Kuby J (ed) Immunology, vol 7. W.H. Freeman and Company, New York, NY, pp 141–156 Ladoire S, Arnould L, Apetoh L et al (2008) Pathologic complete response to neoadjuvant chemotherapy of breast carcinoma is associated with the disappearance of tumor-infiltrating foxp3+ regulatory T cells. Clin Cancer Res 14:2413–2420 Lamb RA, D K (2001) Paramyxoviridae: the viruses and their replication. Fields virology, 4th edn, 1305–1340 Lee WT, Shimizu K, Kuriyama H et al (2005) Tumor-dendritic cell fusion as a basis for cancer immunotherapy. Otolaryngol Head Neck Surg 132:755–764 Lee WT, Tan C, Koski G et al (2010) Immunotherapy using allogeneic squamous cell tumordendritic cell fusion hybrids. Head Neck in press. Lespagnard L, Mettens P, Verheyden AM et al (1998) Dendritic cells fused with mastocytoma cells elicit therapeutic antitumor immunity. Int J Cancer 76:250–258 Li J, Holmes LM, Franek KJ et al (2001) Purified hybrid cells from dendritic cell and tumor cell fusions are superior activators of antitumor immunity. Cancer Immunol Immunother 50:456–462 Linardakis E, Bateman A, Phan V et al (2002) Enhancing the efficacy of a weak allogeneic melanoma vaccine by viral fusogenic membrane glycoprotein-mediated tumor cell-tumor cell fusion. Cancer Res 62:5495–5504 Lindner M, Schirrmacher V (2002) Tumour cell-dendritic cell fusion for cancer immunotherapy: comparison of therapeutic efficiency of polyethylen-glycol versus electro-fusion protocols. Eur J Clin Invest 32:207–217
15
Cell Fusion and Dendritic Cell-Based Vaccines
347
Liu Y, Zhang W, Chan T et al (2002) Engineered fusion hybrid vaccine of IL-4 genemodified myeloma and relative mature dendritic cells enhances antitumor immunity. Leuk Res 26:757–763 Loveland BE, Zhao A, White S et al (2006) Mannan-MUC1-pulsed dendritic cell immunotherapy: a phase I trial in patients with adenocarcinoma. Clin Cancer Res 12:869–877 Lundqvist A, Palmborg A, Bidla G et al (2004) Allogeneic tumor-dendritic cell fusion vaccines for generation of broad prostate cancer T-cell responses. Med Oncol 21:155–165 Margulies DH (2005) Monoclonal antibodies: producing magic bullets by somatic cell hybridization. J Immunol 174:2451–2452 Marten A, Renoth S, Heinicke T et al (2003) Allogeneic dendritic cells fused with tumor cells: preclinical results and outcome of a clinical phase I/II trial in patients with metastatic renal cell carcinoma. Hum Gene Ther 14:483–494 Marzo AL, Kinnear BF, Lake RA et al (2000) Tumor-specific CD4+ T cells have a major “postlicensing” role in CTL mediated anti-tumor immunity. J Immunol 165:6047–6055 Marzo AL, Lake RA, Robinson BW et al (1999) T-cell receptor transgenic analysis of tumor-specific CD8 and CD4 responses in the eradication of solid tumors. Cancer Res 59:1071–1079 Mayordomo JI, Zorina T, Storkus WJ et al (1995) Bone marrow-derived dendritic cells pulsed with synthetic tumour peptides elicit protective and therapeutic antitumour immunity. Nat Med 1:1297–1302 McCune CS, O’Donnell RW, Horan PK et al (1982) Antitumor immunity induced by hybrid murine tumor cells: requirements for optimal immunization. J Natl Cancer Inst 69:647–652 Menard C, Martin F, Apetoh L et al (2008) Cancer chemotherapy: not only a direct cytotoxic effect, but also an adjuvant for antitumor immunity. Cancer Immunol Immunother 57:1579–1587 Michelsen KS, Aicher A, Mohaupt M et al (2001) The role of toll-like receptors (TLRs) in bacteriainduced maturation of murine dendritic cells (DCS). Peptidoglycan and lipoteichoic acid are inducers of DC maturation and require TLR2. J Biol Chem 276:25680–25686 Milstein C (1999) The hybridoma revolution: an offshoot of basic research. Bioessays 21:966–973 Mitchison NA (1990) An exact comparison between the efficiency of two- and three-cell-type clusters in mediating helper activity. Eur J Immunol 20:699–702 Mountford AP, Anderson S, Wilson RA (1996) Induction of Th1 cell-mediated protective immunity to Schistosoma mansoni by co-administration of larval antigens and IL-12 as an adjuvant. J Immunol 156:4739–4745 Mumberg D, Monach PA, Wanderling S et al (1999) CD4(+) T cells eliminate MHC class II-negative cancer cells in vivo by indirect effects of IFN-gamma. Proc Natl Acad Sci USA 96:8633–8638 Nakahara S, Tsunoda T, Baba T et al (2003) Dendritic cells stimulated with a bacterial product, OK-432, efficiently induce cytotoxic T lymphocytes specific to tumor rejection peptide. Cancer Res 63:4112–4118 Negrete OA, Levroney EL, Aguilar HC et al (2005) EphrinB2 is the entry receptor for Nipah virus, an emergent deadly paramyxovirus. Nature 436:401–405 Neil GA, Zimmermann U (1993) Electrofusion. Methods Enzymol 220:174–196 Nestle FO, Alijagic S, Gilliet M et al (1998) Vaccination of melanoma patients with peptide- or tumor lysate-pulsed dendritic cells. Nat Med 4:328–332 O’Donnell RW, Horan PK, Minken TJ et al (1984) Antitumor immunity induced by hybrid tumor cells: comparison between hybrids and the parental tumor. Cancer Res 44:487–492 Ogawa F, Iinuma H, Okinaga K (2004) Dendritic cell vaccine therapy by immunization with fusion cells of interleukin-2 gene-transduced, spleen-derived dendritic cells and tumour cells. Scand J Immunol 59:432–439 Ogihara T, Iinuma H, Okinaga K (2004) Usefulness of immunomodulators for maturation of dendritic cells. Int J Oncol 25:453–459 Okada Y, Kim J, Maeda Y et al (1974) Specific movement of cell membranes fused with HVJ (Sendai virus). Proc Natl Acad Sci USA 71:2043–2047
348
J. Gong and S. Koido
Okada Y, Tadokoro J (1963) The distribution of cell fusion capacity among several cell strains or cells caused by HVJ. Exp Cell Res 32:417–430 Okada Y, Yamada K, Tadokoro J (1964) Effect of antiserum on the cell fusion reaction caused by HVJ. Virology 22:397–409 Okamoto M, Furuichi S, Nishioka Y et al (2004) Expression of toll-like receptor 4 on dendritic cells is significant for anticancer effect of dendritic cell-based immunotherapy in combination with an active component of OK-432, a streptococcal preparation. Cancer Res 64:5461–5470 Orsini E, Guarini A, Chiaretti S et al (2003) The circulating dendritic cell compartment in patients with chronic lymphocytic leukemia is severely defective and unable to stimulate an effective T-cell response. Cancer Res 63:4497–4506 Ossendorp F, Mengede E, Camps M et al (1998) Specific T helper cell requirement for optimal induction of cytotoxic T lymphocytes against major histocompatibility complex class II negative tumors. J Exp Med 187:693–702 Paglia P, Chiodoni C, Rodolfo M et al (1996) Murine dendritic cells loaded in vitro with soluble protein prime cytotoxic T lymphocytes against tumor antigen in vivo. J Exp Med 183:317–322 Pardoll D (2003) Does the immune system see tumors as foreign or self? Annu Rev Immunol 21:807–839 Pare ME, Gauthier S, Landry S et al (2005) A new sensitive and quantitative HTLV-I-mediated cell fusion assay in T cells. Virology 338:309–322 Parkhurst MR, DePan C, Riley JP et al (2003) Hybrids of dendritic cells and tumor cells generated by electrofusion simultaneously present immunodominant epitopes from multiple human tumor-associated antigens in the context of MHC class I and class II molecules. J Immunol 170:5317–5325 Phan V, Errington F, Cheong SC et al (2003) A new genetic method to generate and isolate small, short-lived but highly potent dendritic cell-tumor cell hybrid vaccines. Nat Med 9:1215–1219 Plattet P, Rivals JP, Zuber B et al (2005) The fusion protein of wild-type canine distemper virus is a major determinant of persistent infection. Virology 337:312–326 Primakoff P, Myles DG (2002) Penetration, adhesion, and fusion in mammalian sperm-egg interaction. Science 296:2183–2185 Qin Z, Blankenstein T (2000) CD4+ T cell – mediated tumor rejection involves inhibition of angiogenesis that is dependent on IFN gamma receptor expression by nonhematopoietic cells. Immunity 12:677–686 Raje N, Hideshima T, Davies FE et al (2004) Tumour cell/dendritic cell fusions as a vaccination strategy for multiple myeloma. Br J Haematol 125:343–352 Ratta M, Fagnoni F, Curti A et al (2002) Dendritic cells are functionally defective in multiple myeloma: the role of interleukin-6. Blood 100:230–237 Ridge JP, Di Rosa F, Matzinger P (1998) A conditioned dendritic cell can be a temporal bridge between a CD4+ T- helper and a T-killer cell. Nature 393:474–478 Rodriguez-Tome P, Lijnzaad P (1999) The radiation hybrid database. Nucleic Acids Res 27: 115–118 Rosenberg SA (2004) Shedding light on tumor immunotherapy of cancer. New Engl J Med 350:461–1463 Rosenberg SA, Yang JC, Restifo NP (2004) Cancer Immunotherapy: moving beyond current vaccines. Nat Med 10:909–915 Satthaporn S, Robins A, Vassanasiri W et al (2004) Dendritic cells are dysfunctional in patients with operable breast cancer. Cancer Immunol Immunother 53:510–518 Schnare M, Barton GM, Holt AC et al (2001) Toll-like receptors control activation of adaptive immune responses. Nature Immunol 2:947–950 Scott-Taylor TH, Pettengell R, Clarke I et al (2000) Human tumour and dendritic cell hybrids generated by electrofusion: potential for cancer vaccines. Biochim Biophys Acta 1500:265–279 Sherman LA, Chattopadhyay S (1993) The molecular basis of allorecognition. Annu Rev Immunol 11:385–402
15
Cell Fusion and Dendritic Cell-Based Vaccines
349
Shimizu K, Kuriyama H, Kjaergaard J et al (2004) Comparative analysis of antigen loading strategies of dendritic cells for tumor immunotherapy. J Immunother 27:265–272 Shortman K, Heath WR (2001) Immunity or tolerance? That is the question for dendritic cells. Nat Immunol 2:988–989 Siders WM, Vergilis KL, Johnson C et al (2003) Induction of specific antitumor immunity in the mouse with the electrofusion product of tumor cells and dendritic cells. Mol Ther 7:498–505 Singer SJ, Nicolson GL (1972) The fluid mosaic model of the structure of cell membranes. Science 175:720–731 Specht JM, Wang G, Do MT et al (1997) Dendritic cells retrovirally transduced with a model antigen gene are therapeutically effective against established pulmonary metastases. J Exp Med 186:1213–1221 Steinman RM (1991) The dendritic cell system and its role in immunogenicity. Annu Rev Immunol 9:271–296 Steinman RM (2001) Dendritic cells and the control of immunity: enhancing the efficiency of antigen presentation. Mt Sinai J Med 68:106–166 Steinman RM, Swanson J (1995) The endocytic activity of dendritic cells. J Exp Med 182:283–288 Stolzing A, Hescheler J, Sethe S (2007) Fusion and regenerative therapies: is immortality really recessive? Rejuvenation Res 10:571–586 Suzuki T, Fukuhara T, Tanaka M et al (2005) Vaccination of dendritic cells loaded with interleukin12-secreting cancer cells augments in vivo antitumor immunity: characteristics of syngeneic and allogeneic antigen-presenting cell cancer hybrid cells. Clin Cancer Res 11:58–66 Takeda A, Homma S, Okamoto T et al (2003) Immature dendritic cell/tumor cell fusions induce potent antitumour immunity. Eur J Clin Invest 33:897–904 Tamai H, Watanabe S, Zheng R et al (2008) Effective treatment of spontaneous metastases derived from a poorly immunogenic murine mammary carcinoma by combined dendritic-tumor hybrid vaccination and adoptive transfer of sensitized T cells. Clin Immunol 127:66–77 Tanaka H, Shimizu K, Hayashi T et al (2002) Therapeutic immune response induced by electrofusion of dendritic and tumor cells. Cell Immunol 220:1–12 Tanaka Y, Koido S, Chen D et al (2001) Vaccination with allogeneic dendritic cells fused to carcinoma cells induces antitumor immunity in MUC1 transgenic mice. Clin Immunol 101:192–200 Tanaka Y, Koido S, Xia J et al (2004) Development of antigen-specific CD8+ CTL in MHC class I-deficient mice through CD4 to CD8 conversion. J Immunol 172:7848–7858 Tang CK, Lodding J, Minigo G et al (2007) Mannan-mediated gene delivery for cancer immunotherapy. Immunology 120:325–335 Thery C, Amigorena S (2001) The cell biology of antigen presentation in dendritic cells. Curr Opin Immunol 13:45–51 Trefzer U, Herberth G, Wohlan K et al (2005) Tumour-dendritic hybrid cell vaccination for the treatment of patients with malignant melanoma: immunological effects and clinical results. Vaccine 23:2367–2373 Trevor KT, Cover C, Ruiz YW et al (2004) Generation of dendritic cell-tumor cell hybrids by electrofusion for clinical vaccine application. Cancer Immunol Immunother 53:705–714 Vasir B, Wu Z, Crawford K et al (2008) Fusions of dendritic cells with breast carcinoma stimulate the expansion of regulatory T cells while concomitant exposure to IL-12, CpG oligodeoxynucleotides, and anti-CD3/CD28 promotes the expansion of activated tumor reactive cells. J Immunol 181:808–821 Walden P (2000) Hybrid cell vaccination for cancer immunotherapy. Adv Exp Med Biol 465:347–354 Wang J, Saffold S, Cao X et al (1998) Eliciting T cell immunity against poorly immunogenic tumors by immunization with dendritic cell-tumor fusion vaccines. J Immunol 161:5516–5524 Watts C (2004) The exogenous pathway for antigen presentation on major histocompatibility complex class II and CD1 molecules. Nat Immunol 5:685–692
350
J. Gong and S. Koido
Wolkers MC, Brouwenstijn N, Bakker AH et al (2004) Antigen bias in T cell cross-priming. Science 304:1314–1317 Xia D, Chan T, Xiang J (2005) Dendritic cell/myeloma hybrid vaccine. Methods Mol Med 113:225–233 Xia D, Li F, Xiang J (2004) Engineered fusion hybrid vaccine of IL-18 gene-modified tumor cells and dendritic cells induces enhanced antitumor immunity. Cancer Biother Radiopharm 19:322–330 Xia J, Tanaka Y, Koido S et al (2003) Prevention of spontaneous breast carcinoma by prophylactic vaccination with dendritic/tumor fusion cells. J Immunol 170:1980–1986 Yamanaka R, Homma J, Yajima N et al (2005) Clinical evaluation of dendritic cell vaccination for patients with recurrent glioma: results of a clinical phase I/II trial. Clin Cancer Res 11:4160–4167 Yasuda T, Kamigaki T, Kawasaki K et al (2007) Superior anti-tumor protection and therapeutic efficacy of vaccination with allogeneic and semiallogeneic dendritic cell/tumor cell fusion hybrids for murine colon adenocarcinoma. Cancer Immunol Immunother 56:1025–1036 Young JW, Inaba K (1996) Dendritic cells as adjuvants for class I major histocompatibility complex-restricted antitumor immunity. J Exp Med 183:7–11 Young JW, Koulova L, Soergel SA et al (1992) The B7/BB1 antigen provides one of several costimulatory signals for the activation of CD4+ T lymphocytes by human blood dendritic cells in vitro. J Clin Invest 90:229–237 Zhang HM, Zhang LW, Liu WC et al (2006) Comparative analysis of DC fused with tumor cells or transfected with tumor total RNA as potential cancer vaccines against hepatocellular carcinoma. Cytotherapy 8:580–588 Zhang JK, Li J, Zhang J et al (2003) Antitumor immunopreventive and immunotherapeutic effect in mice induced by hybrid vaccine of dendritic cells and hepatocarcinoma in vivo. World J Gastroenterol 9:479–484 Zhang M, Berndt BE, Chen JJ et al (2008) Expression of a soluble TGF-beta receptor by tumor cells enhances dendritic cell/tumor fusion vaccine efficacy. J Immunol 181:3690–3697 Zheng R, Cohen PA, Paustian CA et al (2008) Paired toll-like receptor agonists enhance vaccine therapy through induction of interleukin-12. Cancer Res 68:4045–4049 Zhou J, Weng D, Zhou F et al (2009) Patient-derived renal cell carcinoma cells fused with allogeneic dendritic cells elicit anti-tumor activity: in vitro results and clinical responses. Cancer Immunol Immunother 58:1587–1597 Zou W (2005) Immunosuppressive networks in the tumour environment and their therapeutic relevance. Nat Rev Cancer 5:263–274 Zou W (2006) Regulatory T cells, tumour immunity and immunotherapy. Nat Rev Immunol 6:295–307
Chapter 16
Cancer Cell Fusion with Myeloid Cells: Implications for Energy Metabolism in Malignant Hybrids Rossitza Lazova, Ashok K. Chakraborty, and John M. Pawelek
Abstract It is abundantly clear that metastasis – the migration of cancer cells from their site of origin to distant organs and tissues – is what makes cancer so deadly. It is therefore surprising that so little is known about its onset. We advocate that the century-old theory of cancer cell fusion with tumor-associated leucocytes such as macrophages is the only complete theory we have – potentially explaining most if not all aspects of metastasis, most notably its initiation. The fusion theory states that acquisition of a metastatic phenotype occurs when a healthy migratory leucocyte fuses with a primary tumor cell. The resultant hybrid adapts the white blood cell natural ability to migrate around the body, all the while continuing to go through the uncontrolled cell division of the original cancer cell. Here we review the evidence supporting these concepts. We further focus on autophagy, a common state of macrophages that is also a signature trait of experimental macrophage-melanoma hybrids in culture. We found autophagy to be widespread in pathology specimens of human malignant melanomas, suggesting that autophagy provides an alternate energy source to these tumors. It is proposed that autophagy in melanoma and other malignancies might be a reflection of fusion with myeloid cells. Thus pathways regulating autophagy as well as the fusion events themselves provide potential new targets for cancer therapy. Keywords Cancer cell fusion · metastasis · aneuploidy · autophagy · phagocytosis · cancer epigenome · macrophage · bone marrow-derived cell Abbreviations ABC ALM BMDC CCL2
ATP-binding cassette Acral lentigenous melanoma Bone marrow-derived cell Chemokine (C-C motif) ligand 2
J.M. Pawelek (B) Department of Dermatology, Yale Cancer Center, Yale University School of Medicine, New Haven, CT 06520-8059, USA e-mail:
[email protected] L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_16, C Springer Science+Business Media B.V. 2011
351
352
c-Met CSF CXCR4 DEJ EM EMT FISH GnT-V H&E HGF HIF-1 IL LAMP LC3B LPHA MC1 MC1R M-CSF MDR MIS MITF PCC PEG prcc Rcc sialyl lex SPARC SSM TAM TGF-β1 TR uPA uPAR VEGF
R. Lazova et al.
The cognate receptor for HGF Colony stimulating factor Cysteine-X-cysteine chemokine receptor 4 Dermal epidermal junction Electron microscopy Epithelial-mesenchymal transition Fluorescent in situ hybridization N-acetylglucosaminyltransferase V Hematoxylin and eosin Hepatocyte growth factor Hypoxia-inducible factor 1 Interleukin Lysosome associated membrane protein Light chain 3 isoform B (autophagy marker) Leucocytic phytohemagglutinin, Melanocortin-1 Melanocortin-1 receptor Macrophage-colony stimulating factor Multi-drug resistance Melanoma in situ Microphthalmia-associated transcription factor Premature chromosome condensation Polyethylene glycol Primary papillary rcc Renal cell carcinoma Sialyl lewisx antigen Secreted protein acidic and rich in cysteine; osteonectin; BM40 Superficial spreading melanoma Tumor associated macrophages Transforming growth factor-β1 Toll receptor Urokinase-type plasminogen activator uPA receptor Vascular endothelial growth factor
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cancer Cell Fusion In Vivo . . . . . . . . . . . . . . . . . . . . . . . Tumor Associated Macrophages as Candidates for Cancer Cell Fusion Partners BMDCs in Human Cancer and Stem Cell-Like Distribution Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . 16.5 Cancer Cell Fusion and the Hybrid Phenotype . . . . . . . . . . . . . . .
16.1 16.2 16.3 16.4
. . . . . .
353 355 359
. . . .
363 365
16
Cancer Cell Fusion with Myeloid Cells
16.6 Macrophage-Melanoma Fusion In Vitro Generates Altered Gene Expression and a Metastatic Phenotype In Vivo . . . . . . . . 16.6.1 SPARC . . . . . . . . . . . . . . . . . . . . . . 16.6.2 MCR1 and c-Met . . . . . . . . . . . . . . . . . . 16.6.3 GnT-V and β1,6-Branched Oligosaccharides . . . . . . 16.6.4 Motility-Associated Integrins . . . . . . . . . . . . . 16.6.5 Cell Surface Expression of Lysosome Associated Protein-1 (LAMP-1) . . . . . . . . . . . . . . . . 16.6.6 Autophagy and Coarse Melanin . . . . . . . . . . . . 16.6.7 Autophagy in Cutaneous Malignant Melanoma . . . . . 16.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . 16.8 Considerations for Studying Fusion In Vivo . . . . . . . . . . 16.9 Implications . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . .
353
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
365 367 368 369 369 370 370 371 383 383 384 385
16.1 Introduction The fusion theory was first proposed in the early 1900s and has attracted a lot of scientific interest over the years (Lu and Kang 2009, Pawelek 2000, 2005, Pawelek and Chakraborty 2008a, b, Pawelek et al. 2006). Its roots lie in the pioneering work of Theodore and Marcella Boveri on aberrant chromosome numbers and abnormal mitoses in sea urchin eggs and the remarkable insights of German pathologist Otto Aichel as to how this might relate to tumor progression (Aichel 1911, Boveri 2008). The Boveri’s observed that sea urchin eggs experimentally fertilized with two sets of spermatozoa underwent abnormal mitosis. They later proposed that deregulated growth of cancer cells might also be a result of chromosome imbalance (Boveri 2008). This work motivated Aichel to first propose fusion and hybridization as a mechanism for the imbalance of chromosomes in human cancer, suggesting that a combination of extra chromosomes and the “qualitative differences” in chromosomes from the two cell types could lead to the metastatic phenotype (Aichel 1911). In his 1911 article ‘About cell fusion with qualitatively abnormal chromosome distribution as cause for tumor formation’ Aichel exhorted future scientists to ‘study chromosomes from all angles’. Decades later, the same hypothesis – that metastasis is caused by leucocyte-tumor cell fusion–was proposed independently by Mekler (1968, 1971) and by Goldenberg (1968) and, Goldenberg and Gotz (1968). Several laboratories have now reported that hybrids produced by fusion in vitro or in vivo were aneuploid and of higher metastatic potential (1–5). In 1984, LaGarde and Kerbel summarized the emerging concepts: “[Tumor cell hybridization] can lead to major changes in gene expression. These processes can lead to the evolution of subpopulations of tumor cells having major losses or gains in their malignant aggressiveness and therefore represents a large-scale genetic mechanism capable of generating genotypic and phenotypic diversification . . . If the normal host cell happens to be a lymphoreticular-hematopoietic cell, it could donate this phenotype to cell types which otherwise do not normally express metastatic traits.” There is now considerable evidence to support these concepts.
354
R. Lazova et al.
The pathways of invasion and metastasis have been under intense scientific scrutiny and much is now known about the steps involved (Gupta et al. 2005, Chambers et al. 2002). However, the actual genesis of metastatic cells from within populations of non-metastatic cells of the primary tumor is not understood. What are the initiating mechanisms that cause a carcinoma or melanoma cell in the epithelium to free its adhesions to neighboring cells, adapt a migratory phenotype, cross the basal lamina into the dermis, intravasate into the blood circulatory system or lymphatics, extravasate, and form new tumors in lymph nodes and distant tissues or organs? The long-standing view is essentially Darwinian: the unstable cancer genome combined with host selective pressures generates metastatic cells in the otherwise non-metastatic primary tumor (Fidler and Kripke 1977, Nowell 1976). This view continues to provide the best framework for envisioning tumor progression. Yet it is difficult to imagine how this might occur through successive, stepwise mutations since generation of a metastatic phenotype would require activation and silencing of very large numbers of genes in the primary tumor cell (Gupta et al. 2005). One solution to this problem lies in the activation of master regulatory genes that control multiple pathways and initiate pro-metastatic cascades (Ma et al. 2007). This has been highlighted in reports that master regulators of epithelial-mesenchymal transition (EMT) in development, such as Snail, Slug, SPARC (secreted protein acidic and rich in cysteine; osteonectin; BM40), Twist, and others play analogous roles in invasion and metastasis where they activate mesoderm-associated pathways of cellular adhesion and migration (Gupta et al. 2005, Ma et al. 2007. However, the mechanisms through which master regulators such as Twist are themselves upregulated in cancer are not understood. We propose that at least in some cases this could be initiated by fusion of cancer cells with bone marrow-derived cells (BMDCs). While a transition from epithelial to mesodermal gene expression is indeed a characteristic of invasion and metastasis, the expressed genes are often remarkably similar to those associated with migratory BMDCs such as macrophages and other myeloid-lineage cells (Pawelek 2005, Pawelek et al. 2006, Chakraborty and Pawelek 2003). Fusion of migratory BMDCs and cancer cells with co-expression of both fusion partner genomes provides a potential explanation for this phenomenon as first proposed by Munzarova et al. (1992). In our opinion the fusion theory comes closer to a unifying explanation of tumor progression than any yet proposed. Fusion represents a non-mutational mechanism that could explain the aberrant gene expression patterns associated with malignant cells. Studies of macrophage-tumor cell fusions have demonstrated that genes from both parental partners are expressed in hybrid cells (Chakraborty et al. 2001a). Gene expression in such cells reflects combinations of myeloid lineage genes along with those of the cancer cell lineage, all in a background of de-regulated cell division. In fact, many molecules and traits associated with tumor progression are expressed by healthy myeloid lineage cells, for example, angiogenesis, motility, chemotaxis and tropism, immune signaling, matrix degradation and remodeling, responses to hypoxia, and multidrug resistance to chemotherapy (Pawelek 2005, Pawelek et al. 2006). Tumor fusion could also account for aneuploidy and genetic rearrangements in metastatic cells (Duelli and Lazebnik 2007, Lu and Kang 2009, Pawelek 2000, 2005, Pawelek and Chakraborty 2008a, b, Pawelek et al. 2006). It is further possible
16
Cancer Cell Fusion with Myeloid Cells
355
that tumor-BMDC fusions are a source of cancer stem cells (Bjerkvig et al. 2005, Dittmar et al. 2009). This chapter reviews the molecular and cellular pathways activated following fusion of tumor cells with BMDCs, their expression in macrophages and other BMDCs, and their similarities to those governing tumor progression in animal and human cancer.
16.2 Cancer Cell Fusion In Vivo From studies in animal and human cancers there is little doubt that tumor hybrids are generated in vivo and that at least in animals they can be a source of metastases (Pawelek 2000, 2005, Pawelek and Chakraborty 2008a, b, Pawelek et al. 2006). Cancer cells fuse with many cell types in vivo including stromal cells (Jacobsen et al. 2006), epithelial cells (Rizvi et al. 2006) and endothelial cells (Bjerregaard et al. 2006, Mortensen et al. 2004, Streubel et al. 2004). There are more than 30 reports of tumor cell fusion with host cells and many of these implicate macrophages or other BMDCs as host fusion partners (Lu and Kang 2009, Pawelek 2000, 2005, Pawelek and Chakraborty 2008a, b, Pawelek et al. 2006). Though these earlier proposals were not widely cited, by the early 1980s experimental evidence for hybridization in cancer had been reported by several laboratories (Rachkovsky et al. 1998). The results fell largely into three categories: (a) tumor-host hybrids were observed naturally within tumors; (b) some naturally-occurring hybrids found within tumors showed enhanced metastatic potential; and (c) experimental hybrids produced in the laboratory between healthy white blood cells and weakly metastatic tumor cells also showed enhanced metastatic potential. The first report of spontaneous hybrid formation came from Barski and Cornefert (1962) who mixed together two separate lines of tumor cells and co-injected them into host C3H mice. Hybrid clones of cells were isolated from the resultant tumors and were found to be tumorigenic when injected back into C3H mice. Miller et al. (1988, 1989) also observed tumor x tumor hybrids after injecting mixed populations of tumor cells into host mice. Janzen et al. (1971) co-injected a mixture of two tumorigenic cell lines, a sarcoma and a lymphoma, into host mice and found by chromosome analyses that 3% of the cells within the developing tumor were hybrids. Similarly, Hart (1984) co-injected two separate sub-lines of B16 melanoma cells into host C57/B6 mice and isolated melanoma-melanoma hybrids from within the developing tumors. Hart was unable to detect fusion hybrids between host cells and the injected B16 melanoma cells. Aviles et al. (1977), using a trypsin-Giemsa banding technique for analysis of metaphase chromosomes, was able to identify hybrid clones in each of 14 mouse L cell sarcomas growing in C3H mice. About 90% of the hybrids were tumorigenic when injected back into host mice. In the first study to demonstrate a relationship between fusion and metastasis, Goldenberg et al. (1971, 1974) injected cells of a female human astrocytic glioma into the cheek pouch of male golden hamsters. Lethal metastases rapidly developed that were found upon chromosomal analyses to be composed of human × hamster hybrids. Wiener et al. (1972, 1974a) demonstrated that a highly tumorigenic mouse
356
R. Lazova et al.
A9 sarcoma, when injected s.c. into mice regularly formed hybrids with host cells that were likely of hemopoietic origin. The hybrids were highly aggressive in vivo, in this case similar to the parental tumor cells. Kerbel et al. (1983) showed that methylcholanthrene-induced mouse sarcoma cells became highly metastatic after fusion in vivo with host cells of bone marrow origin. Larizza et al. (1984a, b), Larizza and Schirrmacher (1984) provided evidence that a highly metastatic variant of a low metastatic T-cell lymphoma was derived from spontaneous fusion with a host macrophage. The metastatic variant expressed the macrophage-specific antigen Mac-1 (CD18) that was not found on cells of the original tumor or any other of the tumors assayed except for a macrophage tumor line. They further showed that hybrids artificially generated with polyethylene glycol in vitro between the Eb T-cell lymphoma and bone marrow macrophages were highly metastatic and also expressed Mac-1. Fortuna et al. (1989, 1990) induced sarcomas in allophenic mice by treatment with methylcholanthrene and demonstrated that almost 1% of the tumor cells were hybrid in nature, with hybrid clones displaying both forms of the enzyme glucose phosphate isomerase. Clones of hybrids showed significantly increased lung colonization abilities compared to non-hybridized tumor cells isolated from the same cultures, although these traits diminished upon further in vitro passages. There is circumstantial evidence that hybrid formation occurs within human tumors. Atkin (1979) in a case study of human bladder cancer observed premature chromosome condensation (PCC) in about 6% of 284 tumor metaphases examined. PCC is a property of multinucleate cells, and Atkin postulated that this, in conjunction with increased ploidy seen in these cells, was evidence for fusions occurring between normal and malignant cells of the bladder, and that the normal cells were likely to be plasma cells that were present in the tumor in high number. Similar observations of PCC in human tumors were reported by Kovacs (1985). In a study of a freshly excised human ovarian tumor, Kerschmann et al. (1995) reported the presence of sarcomatoid cells that expressed both KP-1, a macrophage antigen, and cytokeratin, characteristic of carcinomas. The majority of the cells in the excised nodules were positive for only one or the other of the two markers. The authors concluded that this was evidence for macrophage × tumor cell hybridizations in vivo, and suggested that the presence of cholesterol crystals within the tumor might have catalyzed the fusions. Ruff and Pert (1984) reported that freshly excised human lung tumor cells, particularly those diagnosed as small cell carcinomas, expressed a number of macrophage-specific antigens, leading the authors to propose that such tumors are of hemopoietic stem cell rather than lung origin. Three reports describe spontaneous melanoma-host hybrids in mice: one of B16 melanoma cells (Hu and Pasztor 1975) and two of Cloudman S91 melanoma cells (Halaban et al. 1980, Chakraborty et al. 2000). In all three cases, the hybrids were hypermelanotic, showed increased dendricity and showed increased DNA content compared to the parental melanoma cells. All these lines showed higher tumorigenicity (Hu and Pasztor 1975, Halaban et al. 1980) and/or metastatic potential (Hu and Pasztor 1975, Chakraborty et al. 2000). We tested two of these hybrids and found that they produced ‘coarse melanin’, autophagosomes with
16
Cancer Cell Fusion with Myeloid Cells
357
multiple melanosomes. Further, when we fused mouse or human macrophages with Cloudman S91 melanoma cells in vitro, more than half of the 75 individual hybrids were of increased metastatic potential and autophagic (Rachkovsky et al. 1998, Pawelek et al. 2000, Rupani et al. 2004). Autophagy as a signature trait of cancer cell hybrids and human malignant melanomas alike is discussed later in this chapter. The hypermelanotic phenotype of our panel of macrophage-melanoma hybrids was puzzling. Why should fusion of a non-pigmented macrophage with a weaklypigmented melanoma cell yield highly pigmented hybrids? The short answer is aberrant glycosylation in the form of β1,6-branched oligosaccharides. Analyses of melanosomal proteins tyrosinase, TYRP-1, TYRP-2, and LAMP-1 in macrophagemelanoma hybrids showed that they were heavily glycosylated compared to parental melanoma cells. That LAMP-1 was one of these proteins provided the first indication that β1,6-branched oligosaccharides might be involved (Pawelek et al. 2000). LAMP-1 is one of the most heavily glycosylated of all proteins and is the chief substrate for GnT-V, (N-acetylglucosaminyltransferase V) that is rate-limiting in the formation of β1,6-branched oligosaccharides. GnT-V activates several pathways in metastastic progression. High GnT-V expression is a macrophage trait and it thus seemed likely that GnT-V might be elevated in macrophage-melanoma hybrids due to genetic input from the parental macrophage. Indeed, GnT-V and β1,6-branched oligosaccharides were elevated in the high metastatic hybrids (Chakraborty et al. 2001b). Use of glycosylation inhibitors and selective GnT-V inactivation in hybrids revealed that GnT-V expression was the underlying cause for both chemotactic motility and hyperpigmentation (Chakraborty and Pawelek 2007). It is possible that GnT-V itself induced formation of coarse vesicles. GnT-V transfection into mink lung cells induced production of LPHA-positive multilamellar vesicles and the process was dependent on autophagy (Hariri et al. 2000, Handerson and Pawelek 2003). Likewise, in macrophage-melanoma hybrids with high GnT-V expression, β1,6-branched oligosaccharides co-localized with coarse melanin vesicles (detected with the lectin LPHA) (Rupani et al. 2004). An example of melanoma cell fusion in vivo is seen in the development of a spontaneous melanoma metastasis to the lungs in a Balb c nude mouse (Chakraborty et al. 2000) (Fig. 16.1). Balb c mice are albino due to a homozygous mutation in tyrosinase (c/c), the rate-limiting enzyme in melanogenesis. Although the melanoma clone implanted into these mice was genetically wild type for tyrosinase (C/C), the cells produced little or no melanin in culture and formed amelanotic tumors in mice. Metastases, though infrequent, were generally small, amelanotic tumors in the lung, and were well-tolerated by the mice (Chakraborty et al. 2000). However, in one experiment a mouse developed a melanin-producing in transit metastasis near the site of implantation in the tail dermis. Because of this the tail was amputated and the mouse was followed to see if distant metastases developed. After 5 weeks the mouse became moribund with a massive, highly pigmented pulmonary metastasis. DNA analyses showed that cells from the metastasis had a genotype of C/c, indicating they were hybrids formed from fusion of the implanted tumor cells (C/C) with host cells (c/c). Cells from the metastasis showed an average 30–40% increase in DNA content, increased chemotaxis in vitro, activation of the glycosyltransferase
358
R. Lazova et al.
Fig. 16.1 Spontaneous in vivo fusion in melanoma (Chakraborty et al. 2000). Cells from a clone of the Cloudman S91 mouse melanoma were implanted sub-cutaneously in the tail of a Balbc nu/nu mouse. The mice were albino due to a homozygous mutation in tyrosinase (c/c), the rate-limiting enzyme in melanogenesis. Although the melanoma clone was genetically wild type for tyrosinase (C/C), the cells produced little or no melanin in culture and formed amelanotic tumors in mice. Metastases, though infrequent, were generally small, amelanotic tumors in the lung, and were well-tolerated by the mice. In one experiment (designed for other purposes) what appeared to be a melanin-producing in transit metastasis developed (Panel a, asterisk) near the site of implant (bracket). The tail was amputated and the implanted tumor was formalin fixed, embedded in paraffin, and sectioned serially. Small numbers of highly melanized, coarse melanin-producing cells were found within the implanted tumor that were not seen in cultures of the parental melanoma cells and had thus been generated in vivo (Panel b, arrows). Five weeks after removal of the tail the mouse became moribund with a massive, highly pigmented pulmonary metastasis (Panel c, asterisk). Cells from the metastasis were cloned in soft agar. DNA analyses
16
Cancer Cell Fusion with Myeloid Cells
359
GnT-V, and production of β1,6-branched oligosaccharides (see below). They also produced coarse melanin/autophagosomes. Small numbers of highly melanized, coarse melanin-producing cells were found within the original implanted tumor. These were not present in the cultured parental melanoma cells and were thus generated in vivo (Chakraborty et al. 2000). Morphologically identical cells were cultured from the metastasis and determined to be C/c hybrids with host cells, indicating that fusion and hybridization had occurred in the original implant. Histopathology studies of the original implant revealed that it was infiltrated with macrophages, supporting the possibility that macrophage-tumor fusion had occurred there.
16.3 Tumor Associated Macrophages as Candidates for Cancer Cell Fusion Partners Munzarova et al. (1992) noted that a number of macrophage-like traits are expressed by metastatic melanoma and other malignancies and proposed that metastatic melanoma cells might be macrophge-melanoma hybrids. For example, Pernick et al. (1999) showed that human melanomas are often immunoreactive for macrophage markers such as CD68, alpha-1-antitrypsin, HAM56, Mac387 and muramidase. In breast cancer, Shabo et al. (2008) showed that expression of CD163, a macrophage scavenger receptor, is related to early distant recurrence and reduced patient survival. Tumor associated macrophages (TAM’s) facilitate both cancer initiation and progression (Balkwill et al. 2005, Lin et al. 2001, Mantovani et al. 2008, Pollard 2004). Macrophages are attracted through chemotactic signals to tumors where they exert their abilities for matrix degradation, tissue remodeling, stroma deposition, tropism and neoangiogenesis. These are normally employed in functions such as wound healing, osteogenesis, and embryogenesis (Pollard 2004). Since similar microenvironments exist within tumors, it is thought that macrophages become recruited to these “wounds that never heal” (Balkwill et al. 2005); or “tissues that never cease to develop” (Pollard 2004). Indeed macrophages are recruited to existing tumors by inflammatory cytokines and growth factors normally produced following wounding or infection (e.g. chemotactic chemokine CCL2; colony stimulating factor, CSF-1; vascular endothelial growth factor, VEGF-A)
Fig. 16.1 (continued) revealed that 12 of 12 randomly picked clones had a genotype of C/c, indicating they were hybrids formed from fusion of the implanted tumor cells (C/C) with host cells (c/c). Cells from the metastasis showed an average 30–40% increase in DNA content, increased chemotaxis in vitro, activation of the glycosyltransferase, GnT-V, and production of its enzymatic product, β1,6-branched oligosaccharides. Like the pigmented cells found in the primary implant (Panel b), they also produced “coarse melanin” – autophagosomes containing melanosomes and other organelles. Similar cells were cultured from the metastasis and were also seen in histolopathology sections of the pulmonary tumor. This indicated that the coarse melanin-containing cells originated in the primary implant through host-tumor cell fusion(s) (from Chakraborty et al. 2000)
360
R. Lazova et al.
(Pollard 2004, Schioppa et al. 2003). Macrophages initiate neoplasia through release of reactive oxygen and nitrogen species that are mutagenic and carcinogenic (Lin et al. 2001). Tumor microenvironment cytokines – transforming growth factorβ1 (TGF-β1), interleukin-10 (IL-10), and macrophage-colony stimulating factor (M-CSF) – induce macrophages to differentiate from M1 to M2-type cells that produce tumor growth-promoting factors and stimulate angiogenesis (Balkwill et al. 2005). Macrophages accumulate in hypoxic regions of tumors through HIF-1-mediated upregulation of the chemokine receptor CXCR4 (Schioppa et al. 2003). The density of TAM’s has correlated with poor outcome in more than 80% of the human cancers studied, most notably in carcinomas of the breast, prostate, ovary, and cervix (Pollard 2004, Bingle et al. 2002). In these cancers potential benefits from TAM anticancer immune functions were apparently dominated by the TAM tumor-promoting abilities. This was demonstrated in a mouse mammary tumor model where mice carrying a null mutation for CSF-1 showed a marked reduction in TAM density. Mammary tumors developed normally in the null mutants but unlike those in wild type mice they tended not to metastasize (Lin et al. 2001). Thus, the presence of TAM’s was a key requirement for metastasis in this model. However, tumor progression is not completely explained by the presence of TAM’s. During transition to a metastatic phenotype, tumor cells notoriously co-opt leucocytic traits (Chambers et al. 2002, Pawelek 2005, Pawelek et al. 2006, Pollard 2004). Malignant cells are chemotactic, responding to chemokines and exhibiting their own matrix-degrading and angiogenic capabilities. Like migratory leucocytes, metastatic cells exhibit loss of homotypic adhesion, and the ability to transverse a basement membrane, migrate through the mesodermal matrix, intravasate into lymphatics or the blood circulatory system, extravasate from these vessels, and colonize lymph nodes and distant organs (Fidler 2003, Thiery 2002). But unlike normal leukocytes, cancer cells have deregulated mitotic cycles and their numbers continually increase, killing the host if left unchecked. During this process, invasive carcinomas and melanomas often lose differentiated traits such as E-cadherin expression, homotypic cell–cell adhesion, and cytokeratin or melanin production, while gaining mesodermal traits normally attributed to fibroblasts such as production of fibronectin and vimentin, loose adherence, mesenchymal motility mechanisms, and mesoderm-associated pathways such as the uPA/uPAR and HGF/cmet pathways (Friedl 2004, Kang and Massague 2004, Thiery 2003, Wang et al. 2004, Yang et al. 2004). This is known as the epithelialmesenchymal transition (EMT), and thought to be a process where cancer cells mimic the pathways through which the mesoderm is formed from the epithelium in early development (Friedl 2004, Kang and Massague 2004, Thiery 2003, Wang et al. 2004, Yang et al. 2004). A developmental connection to EMT in cancer was shown through analyses of transcription factors such as the Snail/Slug superfamily and Twist that control EMT in embryogenesis (Kang and Massague 2004). These factors regulate mesoderm formation during gastrulation, and were also associated with cancer progression (Friedl 2004, Kang and Massague 2004, Nieto 2002, Wang et al. 2004). It has thus been proposed that the complex processes in metastasis
16
Cancer Cell Fusion with Myeloid Cells
361
may be explained by the action of master regulatory genes normally associated with development (Chambers et al. 2002, Fidler 2003, Friedl 2004, Nieto 2002, Thiery 2002, Zhou et al. 2004). However, a uniform phenotype for EMT in cancer has not yet been described. Carcinomas and melanomas are notoriously heterogeneous, particularly as primary tumors (Clark et al. 1977, Ferlicot et al. 2004, Fidler and Hart 1982, Thiery 2002, Warner 1975). Many invasive and metastatic carcinomas and melanomas continue to produce cytokeratins or melanin, and not all invasive and/or metastatic carcinomas lose E-cadherin (Kerr 2004, Thiery 2003, Yang et al. 2004). Twist expression is not universal. In human breast carcinoma, Twist up-regulation is associated with invasive lobular carcinomas, but not with invasive ductal carcinomas, which make up 80% of breast cancers and which metastasize at a similar rate as the lobular (Kang and Massague 2004, Kerr 2004). If EMT defines tumor progression, why is it not expressed more uniformly? One explanation could be that EMT is transient: For example, metastases may regain differentiated traits in the process of colonizing lymph nodes or distant organs in a reversal process known as MET (mesenchymalepithelial transition) (Thiery 2003, Thiery and Morgan 2004, Wang et al. 2004, Yang et al. 2004) Another explanation could be that EMT is a consequence of tumor cell-myeloid cell fusion. Monocytes/macrophages and other myeloid cells are of mesenchymal origin, as shown in Drosophila, where double mutants in the mesoderm regulators Twist and Snail lack macrophages (Tepass et al. 1994). There is a growing list of myeloid-type traits that are shared by malignant cells. These include loss of homotypic adhesion, chemotactic motility, matrix degradation, immune signaling pathways, systemic migration, neoangiogenesis, and even multidrug resistance (Pawelek 2005, Pawelek et al. 2006). A key example is amoeboid motility, a characteristic of bone marrow-derived leukocytes, stem cells, and metastatic cancer cells alike. Amoeboid motility is required for migration through the stroma and dissemination via the circulatory system (Friedl and Wolf 2003, Wolf et al. 2003). In amoeboid motility, cells are highly deformable and because of their lack of stable focal adhesions can move at high velocities. The ability to undergo rapid shape-change allows for migration through tissue without the need to degrade matrix (Condeelis and Segall 2003, Friedl and Wolf 2003, Wolf et al. 2003). Moreover, monocytes/macrophages and malignant tumor cells both show high plasticity, such as the ability to differentiate into fibroblastic or endothelial-like cells and to exhibit vascular mimicry (Hendrix et al. 2003). This is mediated in part through FAK kinase, a monocyte/macrophageassociated enzyme whose expression is associated with both vascular mimicry and metastasis (Hess et al. 2005, McLean et al. 2005, Rovida et al. 2005). Similarly, neurotropins and neurotropin receptors are expressed by macrophages and are also associated with cancer anoikis-resistance and metastasis (Barouch et al. 2001, Geiger and Peeper 2005, Ricci et al. 2000). Melanoma and colon carcinoma cell lines express the macrophage-associated Toll Receptor-4 (TR-4) and are responsive to LPS (Molteni et al. 2006). The expression of Toll-like receptors could facilitate evasion of immune surveillance of metastatic cells (Huang et al. 2005, Molteni et al.
362
R. Lazova et al.
a
b
c
d
d
e
Fig. 16.2 Malignant melanoma cells and tumor-associated macrophages (TAM’s) stained for three metastasis-associated markers: β1,6-branched oligosaccharides, GnT-V, and matriptase. Slides were bleached to decolorize melanin and stained by the immunoperoxidase reaction with the lectin LPHA for β1,6-branched oligosaccharides (A, tumor cells; B, TAM’s), with anti-GnT-V (C, tumor cells; D, TAM’s); or anti-matriptase (E, tumor cells; F, TAM’s). TAM’s were further verified by S100/azure blue staining (not shown). All fields were from the same tumor. (Handerson T and Pawelek J, unpublished)
2006). Cancer cells and macrophages both express multidrug-resistance proteins (ABC transporters) such as p-glycoprotein and other MDR proteins that confer chemotherapeutic resistance (Breier et al. 2005, Jorajuria et al. 2004, Michot et al. 2004).
16
Cancer Cell Fusion with Myeloid Cells
363
To illustrate marker co-expression, melanoma cells and TAM’s from the same histological section of a metastatic melanoma are each shown expressing GnT-V (E.C.2.4.1.155; N-acetylglucoseaminyltransferase V), β1,6-branched oligosaccharides, and matriptase, a GnT-V substrate (Fig. 16.2). These markers play key roles in both macrophage and cancer cell migration, and all three are prognostic indicators for metastasis and poor outcome in human cancers (Fernandes et al. 1991, Handerson et al. 2005, Handerson and Pawelek 2003, Ihara et al. 2004, Kang et al. 2003). Their high expression must have been acquired at some point during or following neoplastic transformation, since normal cutaneous melanocytes were negative.
16.4 BMDCs in Human Cancer and Stem Cell-Like Distribution Patterns The first confirmation of BMDC-tumor cell fusion in humans was reported where transcriptionally active malignant nuclei and normal nuclei were observed in tumorassociated osteoclasts from myeloma patients. In the osteoclast population, 30% of the nuclei were of malignant cell origin, indicating a remarkably high incidence of osteoclast-tumor cell fusion (Andersen et al. 2007, 2009). The potential relevance of this finding to myeloma pathobiology is not yet known. Other studies below have demonstrated the presence of donor genes in carcinoma cells of secondary malignancies arising after allogeneic bone marrow transplant, however for largely technical reasons, definitive proof for or against donor-host fusion was lacking in each. In the first reported case, a renal cell carcinoma (rcc) developed in a child following a bone marrow transplant from his cancer-free brother (Chakraborty et al. 2004). A lymph node metastasis of this tumor (the only tissue available) was analyzed by laser capture microscopy of tumor cells and PCR-based analyses for donor genes. Carcinoma cells throughout the tumor contained the donor-specific A allele of the ABO blood group indicating that BMDCs had in some manner become incorporated into the tumor. The patient history of radiation and immunosuppression prior to transplant increased the likelihood that the tumor arose de novo in the patient and that donor BMDCs became incorporated via fusion with preexising tumor cells. However, because a suitable patient-specific DNA sequence was unavailable, evidence for donor and patient genes in the same cells was lacking (Chakraborty et al. 2004). In the second such case (Yilmaz et al. 2005), tumor cells from a primary papillary rcc (prcc) arising after a male to female HSC transplant were found to exhibit a trisomy 17, a common abnormality in prcc and other cancers (Salama et al. 2003). About 1% of the trisomy 17-containing tumor cells also contained the donor Y chromosome in the same nucleus (Yilmaz et al. 2005). As above, this combined with the patient history suggested that fusion had occurred between tumor cells and donor HSC cells after development of the tumor (Yilmaz et al. 2005). However the possibility that the tumor was derived solely from a donor BMDCs, without fusion, followed by growth and widespread loss of the Y was not ruled out (Lau et al. 2007). Nonetheless Y-containing carcinoma cells were rarely
364
R. Lazova et al.
Fig. 16.3 BMDC-engrafted renal carcinoma cells in pairs and clusters. H&E-stained renal carcinoma cells were co-localized to their Y chromosome-containing nuclei visualized by FISH (red). (∗ ) = trisomy 17-containing nuclei (green). Panels a and b show pairs of carcinoma cells containing one nucleus with a trisomy 17 and the other with a monosomy 17. Panel c shows a cluster of Y chromosome-containing carcinoma cells. In the H&E-stained sections, carcinoma cells were distinguished from normal cells by their large size, polygonal shape, abundant granular eosinophilic cytoplasm, and round or ovoid nuclei, usually located centrally, with fine chromatin. Verification of carcinoma cells was by pathologists. Asterisks (∗ ) denote nuclei containing a trisomy 17. Y chromosome: red; chromosome 17: green
found alone but in pairs resembling post-mitotic daughter cells and clusters suggesting a clonal origin of the cells (Fig. 16.3a–c). Should they have been daughter cells, it would also appear that the mitotic division had been asymmetric since many cases one of the Y-containing nuclei contained a trisomy 17 (∗ ) while the other contained a monosomy 17. Although it could not be proven that they were daughter cells or clonal clusters, the observations raise the possibility that they might have been mitocally-active cancer stem cells. (Bjerkvig et al. 2005, Dittmar et al. 2009, Guo and Lasky 2006) (Fig. 16.3). Also, Y-containing carcinoma cells were localized to a region covering only about 10% of the tumor, suggesting a clonal emergence of
16
Cancer Cell Fusion with Myeloid Cells
365
these cells. Supporting this, Y-containing carcinoma cells differed from the majority of carcinoma cells in this tumor through their high expression of β1,6-branched oligosaccharides, a risk factor in several cancers (below).
16.5 Cancer Cell Fusion and the Hybrid Phenotype Fusion-induced enhancement of metastasis and a differentiated trait such as melanin production is in contrast to previous studies where hybrids formed in vitro between normal epithelial cells or fibroblasts and tumorigenic cancer cells were generally suppressed in tumorigenicity compared to the parental cancer cells (Harris 1988, Herzog et al. 2007, Ramshaw et al. 1983, Sidebottom 1980, Stanbridge 1976, Weinberg 1991, Wiener et al. 1974b), with some exceptions (Levine 1995, Scaletta and Ephrussi 1965). These observations lead to the concept- and subsequent identification of a number of different tumor suppressor genes, that have been largely involved in control of progression through the cell cycle (Harris 1988, Weinberg 1991). Differentiated traits were also suppressed in such hybrids. For example, polyethylene glycol (PEG)- and Sendai virus-induced hybrids between fibroblasts and pigmented, tumorigenic melanoma cells were non-pigmented and non-tumorigenic (Davidson et al. 1966, Defendi et al. 1967, Gourdeau and Fournier 1990, Jonasson et al. 1977, Powers and Davidson 1996, Powers et al. 1994). The tendency of hybrids to lose chromosomes with successive cell divisions was exploited for chromosomal mapping of suppressor genes. However, when healthy leukocytes were used as fusion partners with cancer cells, co-activation of differentiated functions between parental genomes was seen, e.g. in leukocyte-hepatoma hybrids (Darlington et al. 1974, Malawista and Weiss 1974), leukocyte-myeloma hybrids (Giacomoni 1979), immunoglobulin-secreting hybridomas (Kohler and Milstein 1975), and macrophage-melanoma hybrids discussed herein. Thus, unlike tumor-suppressive fibroblasts and epithelial cells, hematopoietic cells enhanced malignancy and differentiation when hybridized with transformed cells. Expression of genes from both parental lineages in cancer cell hybrids could explain many properties of metastatic cells (Pawelek 2000, 2005, Pawelek and Chakraborty 2008a, b, Pawelek et al. 2006). For example, tropism to lymph nodes and organs and tissues such as bone marrow, brain, lung, and liver is a common trait of macrophages and metastatic cells alike. Likewise, the notorious multidrug resistance of malignant cells to chemotherapy due to high levels of p-glycoprotein (Gottesman and Ling 2006) could reflect that macrophages also express this phenotype (Lemaire et al. 2007).
16.6 Macrophage-Melanoma Fusion In Vitro Generates Altered Gene Expression and a Metastatic Phenotype In Vivo Tumor-BMDC fusions might explain how common gene expression patterns emerge for different tumor types. We, and others, have found that when BMDC-tumor cell hybrids were isolated in vitro with no selective pressure other than for
366
R. Lazova et al.
Fig. 16.4 Metastatic potential of macrophage-melanoma hybrid cell lines compared to parental Cloudman S91 melanoma cells. Results are shown for in vitro-generated hybrids and one spontaneous in vivo hybrid (PADA). Melanin was estimated in pelleted cells from clones on their first passage in culture and before metastatic potential was determined. A minimum of 10–20 animals was tested for each clone. In addition representative clones were tested repeatedly during continuous passage in culture for up to 4 years where 30–90 animals were tested for each clone with similar results as above. Statistical analyses of metastatic potential revealed that p values for significance vs parental melanoma cells were < 0.0001 (∗ ); < 0.01 (dagger); < 0.05 (square)
growth in drug-containing media, remarkably high numbers of them exhibited a metastatic phenotype in mice. Further, the most metastatic clones tended to be highly melanized compared to parental melanoma cells or weakly metastatic hybrids (described below) (Fig. 16.4). In two separate rounds of isolation, a total of 75 clones of PEG-fused macrophage-melanoma hybrids were isolated in vitro. About half showed increased chemotaxis in vitro and metastasis in mice (Rachkovsky et al. 1998, Rachkovsky and Pawelek 1999, Pawelek et al. 2000). Similar results were obtained in T-cell hybridomas from fusion of healthy T-lymphocytes with T-lymphoma cells (Roos et al. 1985) and in hybrids between mouse T-cell
16
Cancer Cell Fusion with Myeloid Cells
367
lymphoma cells and bone marrow-derived macrophages or spleen lymphocytes, (Larizza et al. 1984a, b). High frequency emergence of a common metastatic phenotype in vitro without host selective pressure was surprising, particularly in view of the apparently chaotic nature of aneuploidy. In fact little is known of the regulation of gene expression in hybrids at the molecular genetic level. Evidence that BMDC-tumor hybrids express many of the same genes associated with invasive and metastatic cancers and that these genes are also expressed by macrophages and other migratory BMDCs is summarized below (also reviewed in (Pawelek and Chakraborty 2008a, b)).
16.6.1 SPARC SPARC (secreted protein acidic and rich in cysteine; osteonectin; BM40) is a modulator of cell-matrix interactions during development and is a key component of wound healing, tissue repair and hard tissue formation (Bradshaw and Sage 2001, Lane and Sage 1994). SPARC modulates cellular shape and as such is a counteradhesive factor (Bradshaw and Sage 2001). SPARC binds to several proteins of the extracellular matrix and is also a chaperone aiding proper folding of collagen in the endoplasmic reticulum (Martinek et al. 2007). In development, SPARC is expressed in late gastrulation during differentiation of invaginated epithelial cells into mesoderm (Damjanovski et al. 1998). Interestingly, SPARC is important in osteoclast formation (Fugita et al. 2002, Mansergh et al. 2007). In tissue macrophages SPARC is expressed in regions of neovascularization, for example in wound repair (Reed et al. 1993) and degenerative aortic stenosis (Charest et al. 2006). High SPARC expression is associated with tumor progression and poor outcome in melanoma and a number of carcinomas including breast, colorectal, ovarian and lung (Robert et al. 2006). SPARC acts as a regulator of melanoma EMT by downregulating melanoma E-cadherin with loss of homotypic adhesion, stimulates motility, and increases expression of mesenchymal markers such as matrix metalloproteinase MMP-9 (Alonso et al. 2007). The actions of SPARC are mediated through Snail, a transcription factor in the initiation of EMT during normal development and cancer (Barrallo-Gimeno and Nieto 2005). The SPARC gene provides an example of gene regulation in BMDC-tumor fusion. In fusions between mouse macrophages or human blood monocytes and weakly metastatic mouse Cloudman S91 melanoma cells, unfused melanoma cells, macrophages and monocytes all expressed SPARC mRNA, however the levels were threefold to fourfold higher per μg total RNA in hybrids (Kerbel et al. 1983, Chakraborty and Yamaga 2003). SPARC mRNA levels were highest in hybrids of high metastatic potential and lowest in weakly metastatic hybrids and parental melanoma cells. Moreover, hybrids between human monocytes and mouse melanoma cells expressed both human and mouse SPARC mRNA (Chakraborty et al. 2001a). This indicated that genomes from cells of the two different developmental lineages were both activated. Thus, for SPARC: gene expression was enhanced by hybridization of tumor cells with macrophages; high expression was
368
R. Lazova et al.
correlated with high metastatic potential; and SPARC mRNA was produced in hybrids from the genomes of both parental fusion partners. That elevated SPARC expression was a characteristic of macrophage-melanoma hybrids provides a possible explanation for elevated SPARC and SPARC-mediated pathways in human melanoma and other cancers. It is not known whether other regulators of EMT and development in addition to SPARC were expressed in macrophage-tumor cell fusion hybrids (transcription factors Twist, Snail, and others (Gupta et al. 2005, Ma et al. 2007). However at least one, Twist, is activated in macrophages and regulates inflammatory cytokine production (Sharif et al. 2006, Sosi et al. 2003). By analogy to SPARC, this opens the possibility that Twist expression in some invasive carcinomas reflects expression of macrophage-lineage genes following macrophage-tumor cell fusion.
16.6.2 MCR1 and c-Met The melanocortin-1 (MC1, melanocyte stimulating hormone) receptor (MC1R) is activated by MC1 in healthy melanocytes and melanoma cells where, through cyclic AMP-dependent mechanisms, it activates melanogenesis and regulates proliferation along with several other actions (Carlson et al. 2007, Kanetsky et al. 2006). MC1R appears to play a role in melanoma progression at least in part through its activation of the proto-oncogene cMet, whose signalling pathway is a key regulator of metastasis in melanoma and many other cancers (Beuret et al. 2007, Boccaccio and Comoglio 2006, McGill et al. 2006). As with SPARC, gene expression for both MC1R and c-Met was increased in high metastatic macrophage-melanoma hybrids (Chakraborty et al. 1999, 2003). Moreover, each was involved in the induction of chemotactic motility in hybrids (Chakraborty et al. 1999, Rachkovsky and Pawelek 1999). Up-regulated MC1R mRNA expression in hybrids was associated with increased cellular binding of its ligand MC1, and amplified responsiveness to MC1 as shown by increased chemotactic motility, dendricity, and melanization (Pawelek et al. 2000, Rachkovsky and Pawelek 1999). Exposure of hybrids to MC1 also increased both the production of c-Met mRNA and responsiveness to HGF as a chemoattractant (Chakraborty et al. 2003). Thus the MC1/MC1R and HGF/c-Met pathways appeared to act coordinately in a positive autocrine loop to control chemotaxis and other functions in hybrid cells. This same relationship appears to be operative in malignant melanoma (Beuret et al. 2007). In melanoma, c-Met and MC1R are each regulated through the master transcription factor MITF (McGill et al. 2006) which itself is associated with tumor progression (Levy et al. 2006). Although it was not determined whether MITF was upregulated in experimental macrophage-melanoma hybrids, this appears to have been the case since mRNA’s for both c-Met and MC1R were elevated, an expected consequence of increased MITF (Bronisz et al. 2006, Garraway et al. 2005, McGill et al. 2006). High expression of MITF (Beilmann et al. 1997), c-Met (Gaasch et al. 2006, Lam et al. 2005), and MC1R (Lam et al. 2006, Manna et al. 2006, McGill et al. 2006, Taylor 2005) are all characteristics of monocytes/macrophages and other BMDCs.
16
Cancer Cell Fusion with Myeloid Cells
369
16.6.3 GnT-V and β1,6-Branched Oligosaccharides N-acetylglucosaminyltransferase V (GnT-V; Mgat5; E.C.2.4.1.155) is a Golgi complex enzyme that is highly expressed in myeloid cells and metastatic cancer cells. GnT-V and its enzymatic products, β1,6-branched oligosaccharides conjugated to N-glycoproteins, are associated with poor outcome in melanoma (Handerson et al. 2007) and carcinomas of the breast (Fernandes et al. 1991, Handerson et al. 2005), colon (Fernandes et al. 1991, Murata et al. 2000, Seelentag et al. 1998) lung (Dosaka-Akita et al. 2004) and endometrium (Yamamoto et al. 2007). β1,6-branched oligosaccharides were first purified from granulocytes (Fukuda et al. 1984). From structural analyses they are composed of poly-N-acetyllactose amines that are carriers of sialyl lewisx antigen (sialyl lex) and therein used by both leukocytes and metastatic cancer cells for binding to E-selectin and/or galectin-3 on endothelial cells during systemic migration (Sarafian et al. 1998, Sawada et al. 1993). GnT-V mRNA, protein, and/or enzymatic activity were elevated in high metastatic macrophage-melanoma hybrids formed in vitro (Chakraborty et al. 2001b), and following spontaneous host-tumor fusions in both lymphomas and melanomas in mice (Chakraborty et al. 2001b, Dennis et al. 1984, Kerbel et al. 1983). Multiple pathways in invasion and metastasis that are regulated by GnT-V were elevated in macrophage-melanoma hybrids – as seen below with motility-associated integrin subunits, cell surface expression of LAMP-1, and autophagy.
16.6.4 Motility-Associated Integrins The integrin subunits α2, α3, α5, α6, αv, β1, and β3 are all involved with migration of leucocytes and cancer cells. These same integrin subunits were significantly upregulated at the protein level in metastatic macrophage-melanoma hybrids compared to weakly metastatic hybrids and parental melanoma cells (Chakraborty et al. 2001b, Chakraborty and Pawelek 2010). Following stimulation with melanocortin-1, protein levels were further increased in high metastatic hybrids. These results correlated with findings that metastatic hybrids had acquired an MC1-inducible chemotactic phenotype that was directed toward fibronectin through the action of integrin α5β1 (Rachkovsky and Pawelek 1999). Of great interest, all the above subunits have been identified as substrates for GnT-V and their actions are strongly affected by their glycosylation status with β1,6-branched oligosaccharides (Chammas et al. 1993, Demetriou et al. 1995, Dennis et al. 1999, Guo et al. 2002, Jasiulionis et al. 1996, Leppa et al. 1995, Ochwat et al. 2004, Poche et al. 2003, Saitoh et al. 1992, Yamamoto et al. 2000, Zheng et al. 1994). For example, addition of β1-6 branched oligosaccharides onto the β1 integrin subunit by GnT-V reduced α5β1 integrin clustering and stimulated cell migration (188). Further, the above integrin subunits are each involved in metastasis. The α3β1 integrin is elevated and associated with increased migration and invasion in several types of metastatic cancers (Giannelli et al. 2002). α5β1 is a well-characterized receptor
370
R. Lazova et al.
for fibronectin that is over-expressed in metastasis (Danen et al. 1994, Galbraith et al. 2007, Natali et al. 1995). Up-regulation of αvβ3, a vitronectin receptor, was described in various cancers including malignant melanoma and glioblastoma (Danen et al. 1994, Gladson and Cheresh 1991, Natali et al. 1997, Wong et al. 1998). Expression of the β1-integrin subunit is a key component of melanoma metastasis (Juliano 1993). The above integrins and integrin subunits are also highly expressed in macrophages where they are involved with many functions, including cell adhesion and migration, signal transduction, cell–cell recognition and phagocytosis (Ammon et al. 2000, Aplin et al. 1998, Elsegood et al. 2006, Kurita-Taniguchi et al. 2002, Shinji et al. 2007).
16.6.5 Cell Surface Expression of Lysosome Associated Protein-1 (LAMP-1) LAMP-1 is a preferred substrate for GnT-V and a major carrier of sialyl lex and poly-N-acetyl-lactose amines that bind to E-selectins and galectins (Chang et al. 2004). Cell surface LAMP-1 thus mediates binding to endothelial cells by both leukocytes and cancer cells (Chang et al. 2004, Sarafian et al. 1998, Sawada et al. 1993). Macrophge-melanoma hybrids showed elevated expression of cell surface LAMP-1 (Chakraborty et al. 2001b). This was seen in high metastatic macrophagemelanoma hybrids as well as peritoneal macrophages compared to that in parental melanoma cells and low metastatic hybrids.
16.6.6 Autophagy and Coarse Melanin As mentioned, the spontaneous mouse melanoma-host hybrid shown in Fig. 16.1 showed a high level of autophagy/coarse melanin (Chakraborty et al. 2000). This was also a characteristic of another spontaneous melanoma-host hybrid described previously (“PADA”) (Pawelek et al. 2000) and of macrophage-melanoma hybrids fused in vitro (Chakraborty et al. 2000, Pawelek et al. 2000, Rachkovsky et al. 1998). EM studies revealed that melanin was localized largely to heavily melanized melanosomes packaged in autophagosomes. Autophagosomes were verified by the presence of double limiting membranes and heterogeneous morphologies. They were also strongly positive for β1,6-branched oligosaccharides, implicating a role for GnT-V in their formation. These were surprising findings because healthy melanocytes do not appear to employ GnT-V in melanogenesis and the melanosomes are not packaged in autophagosomes but exist singly in the cytoplasm. That several independently-isolated melanoma hybrids all showed high levels of autophagy/coarse melanin raised the question as to whether this trait may be a signature of BMDC-melanoma fusion in human melanoma. While coarse melanin in melanoma had been known to pathologists for more than a century, and was shown to be due to autophagy (reviewed in Handerson and Pawelek 2003), its
16
Cancer Cell Fusion with Myeloid Cells
371
frequency in human cancers had not been evaluated (Clark et al. 1977). Analyses of several hundred cases have revealed that it is a common trait, expressed by 85% or more of melanomas (Handerson and Pawelek 2003, Handerson et al. 2007). It was further determined that coarse melanin-producing melanoma cells and melanophages (macrophages with autophagolysosomal vesicles containing undigested melanin) account for the well-known hypermelanotic regions of cutaneous malignant melanoma used in clinical diagnosis. As in macrophage-melanoma hybrids, coarse melanin vesicles in human melanomas contained β1,6-branched oligosaccharides (Handerson and Pawelek 2003, Handerson et al. 2007). In cutaneous malignant melanoma, β1,6-branched oligosaccharide-positive, coarse melanin-producing melanoma cells emerge clonally as “nests” within the in situ tumor and have the capacity for invasion into the dermis (Handerson and Pawelek 2003, Handerson et al. 2007). This suggests that BMDC-tumor cell fusion could be an explanation for the appearance of these cells. These findings prompted us to explore the levels of autophagy in human malignancies. Summarized below are our first observations of autophagy in a panel of cutaneous malignant melanomas.
16.6.7 Autophagy in Cutaneous Malignant Melanoma Twelve cases of superficial spreading melanoma (SSM) with both MIS and invasive component, and one case (#13) of acral lentigenous melanoma (ALM) with residual MIS but no invasive component were selected for the study. Patients ranged in age from 22 to 84 years with a median of 50 years. Eight patients were men and five were women. The lesions were located on the back (n = 4), leg (n = 5), chest (n = 2), abdomen (n = 1), and upper arm (n = 1). The depth of the melanomas ranged from 0.4 to 1.3mm with a median of 0.7mm. Eight melanomas were Clark’s level IV, four were Clark’s level III, and one was Clark’s level I. While all 12 cases of SSM had both MIS and invasive component, only four of them, cases 1, 7, 8, and 11, could be fully evaluated comparing the staining of normal epidermis, early MIS, and florid MIS within the same section. The remaining cases showed only florid MIS with little or no normal epidermis. Case #1 was selected as a representative case to illustrate the staining patterns. Sections stained with H&E from case #1 are shown in Fig. 16.5. Normal epidermis, peripheral to the area containing melanoma in situ, displayed a normal number and distribution of melanocytes in the basal layer (Fig. 16.5a). Adjacent to the normal epidermis was an area containing an irregular proliferation of single pleomorphic melanocytes in the basal layer and above the dermal epidermal junction (DEJ), representing early MIS (Fig. 16.5b). Contiguous with that focus and more towards the center of the lesion was florid MIS comprised of irregular nests of melanocytes in the epidermis, which varied in size and shape. An underlying invasive component of melanoma in the dermis displayed large atypical melanocytes in a nested pattern (Fig. 16.5c) Normal epidermis adjacent to MIS: In areas of the epidermis adjacent to a region of MIS, normal keratinocytes and melanocytes (asterisk) could readily be
372
R. Lazova et al.
Fig. 16.5 Views from an H&E-stained sections of malignant melanoma in case #1. (a) Normal epidermis containing only a few melanocytes in the basal layer (∗ ). (b) Focus of early MIS showing an increased number of slightly pleomorphic single melanocytes at the dermal epidermal junction (DEJ) and above it. (c) Florid MIS in the epidermis with nests of melanocytes displaying irregular shapes, as well as single melanocytes, many of which are seen above the DEJ. Invasive melanoma, comprised of nests of atypical melanocytes, is present in the dermis
distinguished on H&E staining (Fig. 16.6a). Melanin was not apparent in normal melanocytes; however, it was evident that melanogenesis had occurred since adjacent keratinocytes contained brown melanin in their cytoplasm, presumably transferred from melanocytes. Normal melanocytes and keratinocytes did not stain for LC3B (Fig. 16.6b). Staining with the anti-Golgi 58k protein revealed large, globular, perinuclear Golgi complexes in both keratinocytes and melanocytes (Fig. 16.6c). There was no staining with LPHA in normal keratinocytes and melanocytes, confirming previous findings that these cells do not appear to produce β1,6-branched oligosaccharides (Fig. 16.6d) (Handerson and Pawelek 2003). Early MIS. Early MIS presents as a subtle proliferation of atypical melanocytes (melanoma cells) disposed as single units as well as in a few small nests at the
16
Cancer Cell Fusion with Myeloid Cells
373
Fig. 16.6 Medium and high power views of the staining patterns for H&E, LC3B, the Golgi 58k protein, and β1,6-branched oligosaccharides in sequential sections of normal epidermis from malignant melanoma case #1. Melanocytes are denoted with an asterisk (∗ ). Before staining, sections b–d were first bleached to decolorize melanin. Sections were subsequently stained for the specific marker and counterstained with hematoxylin. (a) An H&E stained section showing a normal number and distribution of melanocytes in the basal epidermal layer. (b) A section stained with anti-LC3B displaying negative staining with this marker. (c) A section stained with anti-Golgi 58k protein. In the higher power photomicrograph the brown globules represent the positively stained Golgi complex, sometimes overlapping and obscuring the nucleus. (d) A section stained with the lectin LPHA for β1,6-branched oligosaccharides showing negative staining
dermal epidermal junction and above it. Similar to normal melanocytes, melanoma cells in regions of early MIS did not show prominent melanin in their cytoplasm as seen with H&E staining (Fig. 16.7a). Cells of early MIS did not stain or stained only weakly for LC3B (Fig. 16.7b) and for β1,6-branched oligosaccharides (Fig. 16.7d). However, similar to normal melanocytes they did stain for the Golgi 58k protein in a globular perinuclear pattern, (Fig. 16.7c). Florid MIS. In florid MIS there is an irregular, asymmetric, and poorly circumscribed proliferation of melanoma cells. There are nests of melanoma cells that vary markedly in size and shape, which are not equidistant from one another. Single melanoma cells predominate over nests in some high power fields and there are individual melanoma cells as well as melanocytic nests above the DEJ. Melanoma cells are also seen down adnexal structures. In our study melanoma cells in florid MIS of all 13 cases produced coarse melanin to at least some extent (Fig. 16.8a). In addition, in all cases most if not all of the cells of florid MIS stained for the autophagosome marker LC3B with a heterogeneous vesicular pattern in the cytoplasm, indicating the presence of autophagosomes (Fig. 16.8b). Surprisingly, the Golgi 58k protein was distributed not in a globular perinuclear pattern, characteristic of normal cells and early MIS (above), but in a heterogeneous vesicular pattern, similar to that of coarse melanin and LC3B (Fig. 16.8c). In previous studies such a vesicular pattern for Golgi staining has been described as Golgi “fragmentation” or “vesiculation” (Dagher et al. 2003, Graves et al. 2001, Kovacs et al. 2004, Razi et al. 2009). Unlike melanocytes in normal epidermis and melanoma cells in early MIS, the nested melanoma cells in florid MIS produced β1,6-branched oligosaccharides, which, like the LC3B and Golgi 58k protein, also stained in a heterogeneous
374
R. Lazova et al.
Fig. 16.7 A region of early MIS from malignant melanoma case #1 showing crowded melanocytes (melanoma cells) near the dermal epidermal junction. (a) A section stained with H&E. (b) A section bleached to decolorize melanin, subsequently stained with anti-LC3B, and counterstained with hematoxylin, showing faint staining for LC3B. (c) A section bleached, subsequently stained with anti-Golgi 58k protein, and counterstained with azure blue with a positive perinuclear staining. (d) A section bleached, subsequently stained with biotinylated LPHA, and counterstained with hematoxylin showing negative staining
vesicular pattern (Fig. 16.8d) (Handerson and Pawelek 2003, Handerson et al. 2007). Compared to the melanoma cells in small nests of florid MIS (Fig. 16.8), melanoma cells in larger nests tended to produce more prominent coarse melanin and stain more intensely for LC3B, the Golgi 58k protein, and β1,6-branched oligosaccharides (Fig. 16.9a–d). Thus, in florid MIS most if not all of the melanoma cells had autophagosomes. The autophagosomes contained LC3B, the Golgi 58k protein, and β1,6-branched oligosaccharides. This was confirmed through co-localization studies with coarse melanin and through electron microscopy. Co-localization of the Golgi 58k protein and β1,6-branched oligosaccharides with coarse melanin. Fresh tissue containing epidermis and dermis from case #13 of acral lentiginous melanoma with residual MIS was obtained. Half of the specimen
16
Cancer Cell Fusion with Myeloid Cells
375
Fig. 16.8 Small nests of melanoma cells from florid MIS in malignant melanoma case #1. Sections b and d were bleached, stained with antibody or lectin, and counterstained with hematoxylin. Section c was not bleached, but directly stained with the antibody and counterstained with azure blue. Arrows denote coarse melanin (a) and vesicular staining pattern (b–d). (a) A section stained with H&E. (b) A section stained for LC3B. (c) A section stained for the Golgi 58k protein. (d) A section stained for β1,6-branched oligosaccharides with biotinylated LPHA
was processed for electron microscopy and the other half was fixed in formalin and processed in a standard fashion. Tissue sections were cut and stained with H&E for light microscopic examination and immuno- and lectin histochemistry was performed. Histologic examination of H&E stained sections revealed florid MIS comprised of nests of melanocytes and single melanoma cells containing coarse melanin in their cytoplasm (Fig. 16.10). As mentioned, there was no invasive melanoma component in this case. Co-localization studies in case #13 were performed by photographing coarse melanin in unstained sections and then staining the same slides for either LC3B, the Golgi 58k protein, or β1,6-branched oligosaccharides (Fig. 16.11). Sequential sections were also bleached to decolorize melanin before staining with the same results (not shown). Photographs of coarse melanin-containing melanocytes were compared to photographs of the same melanocytes after staining. Widespread co-localization was found between coarse melanin and LC3B
376
R. Lazova et al.
Fig. 16.9 Medium and high power views of large nests of melanoma cells within florid MIS from case #1. Sections b and d were bleached, stained with antibody or lectin, and counterstained with hematoxylin. Section c was not bleached, but directly stained with the antibody, and counterstained with azure blue. Arrows denote coarse melanin (a) and vesicular staining pattern (b–d). (a) A section stained with H&E. (b) A section stained for LC3B. (c) A section stained for the Golgi 58k protein. (d) A section stained for β1,6-branched oligosaccharides with biotinylated LPHA
Fig. 16.10 A region of florid MIS with melanoma cells in the epidermis producing coarse melanin (arrow) and melanophages in the superficial dermis also containing granular melanin in their cytoplasm (Case #13; H&E)
16
Cancer Cell Fusion with Myeloid Cells
377
Fig. 16.11 Co-localization of LC3B, the Golgi 58k protein, and β1,6-branched oligosaccharides to coarse melanin. The columns labeled “melanin” represent unstained sections photographed to depict coarse melanin. The sections on the right represent the same areas after staining with antibody or lectin. (a–d) Co-localization of LC3B to coarse melanin. (e–h) Co-localization of the Golgi 58k protein to coarse melanin. (i–l) Co-localization of β1,6-branched oligosaccharides to coarse melanin
(Fig. 16.11a–d), the Golgi 58k protein (Fig. 16.11e–h), and β1,6-branched oligosaccharides (Fig. 16.11i–l). Detection of autophagosomes by electron microscopy. A region of florid MIS from case #13 was analyzed by electron microscopy. Numerous vesicles surrounded by double membrane and filled with heavily melanized melanosome-like structures and other debris were seen (Fig. 16.12). The melanosomes seemed to be partially digested and might have represented only residual, undigested melanin; however,
Fig. 16.12 Electron micrographs of coarse melanin vesicles in a region of florid MIS from Case #13 (Fig. 16.6). (a) Low power view showing vesicles with heavily melanized melanosome-like structures. b and c High power views of individual vesicles. The arrows show that the vesicles are bordered by double membranes and by definition represent autophagosomes
378
R. Lazova et al.
that remains to be further elucidated. Such double-membraned vesicles containing cytoplasmic organelles are defined as autophagosomes (Klionsky et al. 2008). Thus, we conclude that coarse melanin vesicles are autophagosomes containing LC3B, the Golgi 58k protein, and β1,6-branched oligosaccharides (Fig. 16.11).
Fig. 16.13 Medium and high power views of the dermal invasive component from malignant melanoma case #1. Sections b and d were bleached, stained with antibody or lectin, and counterstained with hematoxylin. Section c was not bleached but directly stained and then counterstained with azure blue. Arrows denote coarse melanin (a) and vesicular staining pattern (b–d). (a) A section stained with H&E. (b) A section stained for LC3B. (c) A section stained for the Golgi 58k protein. (d) A section stained for β1,6-branched oligosaccharides with biotinylated LPHA
Fig. 16.14 LC3B staining in the invasive components of malignant melanoma cases #1–12. Case #2 was stained with anti-LC3B and counterstained with azure blue. The remaining cases were first bleached to decolorize any melanin, stained with anti-LC3B, and counterstained with hematoxylin
16
Cancer Cell Fusion with Myeloid Cells
379
Invasive melanoma cells. While coarse melanin in florid MIS was present in all cases to some extent, only 4 of the 12 cases showed coarse melanin in the invasive component (cases 4, 5, 7, and 11). The invasive components of the remaining cases were only lightly pigmented or amelanotic as shown for case #1 (Fig. 16.13a). Regardless of the presence or not of coarse melanin, the invasive melanoma cells in all cases revealed positive heterogeneous vesicular pattern of staining for LC3B (Fig. 16.13b), the Golgi 58k protein (Fig. 16.13c), and β1,6-branched oligosaccharides (Fig. 16.13d), seemingly identical to that seen within cells of florid MIS (cf Figs. 16.8 and 16.9). Therefore, it appeared that the vesicles produced in the invasive components were also autophagosomes but with reduced melanin content. The melanoma cells in the invasive component of all 12 cases showed positive staining, for the autophagosome marker LC3B (Fig. 16.14) and the Golgi 58k protein (Fig. 16.15). The staining for both markers displayed a heterogeneous vesicular pattern. The same pattern was also seen for β1,6-branched oligosaccharides in all 12 cases of invasive melanoma (not shown). (Handerson and Pawelek 2003). Melanophages. Melanophages are macrophages filled with melanized vesicles that appear similar under light microscopic examination to coarse melanin seen in melanoma cells. These vesicles are presumably generated from engulfment via phagocytosis of cells containing melanin, followed by transfer of the engulfed cellular debris into autophagosomes. Our study revealed that melanophages express an
Fig. 16.15 Anti-Golgi 58k protein staining in the invasive components of malignant melanoma cases #1–12. Cases 1, 2, 3, 4, 7, 11, and 12 were stained with the anti-Golgi 58k protein and counterstained with azure blue. Cases 5, 6, 8, 9, and 10 were first bleached to decolorize any melanin, subsequently stained with anti-Golgi 58k protein, and counterstained with hematoxylin
380
R. Lazova et al.
Fig. 16.16 Co-localization of granular melanin with heavily melanized vesicles in dermal melanophages (case #13). Melanin in unstained sections is displayed in the left column and the corresponding stained fields are shown in the right column. Arrows denote examples of co-localized vesicles. (a) A section stained for LC3B. (b) A section stained for the Golgi 58k protein. (c) A section stained for β1,6-branched oligosaccharides with biotinylated LPHA
autophagic phenotype that is strikingly similar to that expressed by melanoma cells of florid MIS and invasive melanoma. Co-localization studies showed that LC3B, the Golgi 58k protein, and β1,6-branched oligosaccharides were all constituents of the melanized vesicles in melanophages (Fig. 16.16). Electron micrographs of dermal melanophages in case #13 confirmed that the vesicles were autophagosomes (Fig. 16.17). These autophagosomes in the melanophages are limited by double membrane and, similar to autophagosomes
16
Cancer Cell Fusion with Myeloid Cells
381
Fig. 16.17 Electron micrographs of a dermal melanophage (Case #13). (a) Low power view showing numerous melanin-containing vesicles within the cytoplasm of a melanophage. Collagen bundles in the adjacent dermis are labeled by “c”. (b) High power view of a vesicle, which is an autophagosome, enveloped by a double membrane (arrow), and containing partially digested, heavily melanized melanosomes
in melanoma cells, contain what appears to be partially digested, melanized melanosomes, and other cellular debris. The melanophages were surrounded by collagen bundles (labeled “c”) confirming their dermal location. In summary of our initial studies of cutaneous malignant melanomas we show that autophagic tumor cells were the major, if not sole, cell type for florid MIS and for invasive cells in the dermis. In this study all melanomas were of the superficial spreading type with the exception of one case of ALM (case #13) used in EM and co-localization studies. Therefore, we cannot draw conclusions and make generalizations about other types of malignant melanoma. Using coarse melanin as a marker for autophagosomes, the autophagosomal protein LC3B, the Golgi 58k protein, and β1,6-branched oligosaccharides, all co-localized to autophagosomes. The presence of autophagosomes was additionally confirmed by electron microscopy. Tumor-associated melanophages also exhibited an autophagic phenotype, which was remarkably similar to that seen in melanoma cells. The presence of widespread autophagy in malignant melanoma can be explained by, and is consistent with, findings that most melanomas are under ER stress, an inducer of autophagy (Hersey and Zhang 2008, Rutkowski and Kaufman 2007). The results of this study provide a new view of melanoma progression, in which the metabolic energy balance of invasive cells may be dependent on autophagy. These findings suggest new paradigms for therapy involving inhibition of autophagy in order to deprive melanoma cells of this energy source. Could autophagy in human cancer result from fusions between cancer cells and macrophages or other phagocytes? In fact, macrophages express active autophagy as a part of the pathway for digestion of phagocytosed microorganisms and cells (Amer and Swanson 2005, Amer et al. 2005). Autophagy in macrophages
382
R. Lazova et al.
Fig. 16.18 A model for generation of a metastatic phenotype following fusion of a melanoma cell with a macrophage. (1) A macrophage is attracted to a non-migratory melanoma cell in situ. The epigenomes of the two cells reflect their myeloid and melanocytic lineages respectively. The melanoma cell produces “fine” or “dusty” melanin – individual melanosomes in the cytoplasm, generally with a golden-brown color. Melanoma-associated macrophages are known as melanophages because they are laden with autophagolysosomal vesicles containing melanin from injested melanoma cells, and thus at times difficult to distinguish from melanoma cells at the light microscope level. (2) The macrophage and melanoma plasma membranes form close appositional contacts, normally as a prelude to injestion and destruction of the melanoma cell. However in some cases rather than the macrophage digesting the melanoma cell, the two cells fuse. (3) Following fusion a heterokaryon is formed with the two nuclei separate in the cytoplasm. (4) Genomic hybridization occurs and a mononuclear macrophage/melanoma hybrid emerges. From studies of macrophage/melanoma hybrids generated experimentally in vitro and of melanoma/host hybrids generated spontaneously in mice, such hybrids have a deregulated cell cycle, are aneuploid and exhibit epigenomes of both parental lineages. Some exhibit the myeloid capability for chemotaxis in vitro and tropism in vivo, common characteristics of metastatic cells (from Pawelek and Chakraborty 2008a)
16
Cancer Cell Fusion with Myeloid Cells
383
is linked to phagocytosis, interestingly, another characteristic of metastatic cancers (Coopman et al. 1998, Damiani and Colombo 2003, Lugini et al. 2003, 2006, Montcourrier et al. 1994). Moreover, macrophage vesicles, like those in experimental macrophage-melanoma hybrids and cancer cells, are positive for β1,6-branched oligosaccharides (Fernandes et al. 1991, Handerson and Pawelek 2003, Handerson et al. 2007). Therefore, activation of phagocytic and autophagic pathways in human cancers could reflect expression of imprinted genes of myeloid lineage in macrophage-tumor cell fusion hybrids. We suggest that should cancer cell autophagy be linked to phagocytosis as it is in macrophages, nutrients could be continuously phagocytosed from external sources and digested through autophagy, rendering metastatic cells constitutively independent of a direct blood supply. A model for a cell fusion origin of autophagy in malignant melanoma is presented below (Fig. 16.18).
16.7 Conclusions Metastatic macrophage-melanoma hybrids show high expression of SPARC, c-Met, MC1R, integrin subunits α3, α5, α6, αv, β1, β3, cell surface LAMP-1, GnT-V, and autophagy. This is paralleled in melanoma, and in a number of other cancers in which these molecules are associated with a migratory phenotype, enhanced survival, metastasis and poor outcome. Central to the metastatic phenotype is GnT-V which, through addition of β1,6-branched oligosaccharides to several of the above proteins, causes multiple phenotypic changes including increased chemotaxis, melanogenesis, and possibly autophagy. Expression of MC1R, MITF, c-Met, motility-related integrins, cell surface LAMP-1, GnT-V and autophagy are also characteristic of monocytes/macrophages and other BMDCs. Thus, expression of these molecules in cancer could be a result of fusion of cancer cells with migratory BMDCs and co-expression of imprinted genes from both parental fusion partners. While these molecules and traits are of course not the only factors involved in tumor progression, their high expression in BMDC-tumor hybrids provides a framework for understanding how fusion can explain metastasis. While the possibility of BMDC-tumor cell fusion has yet to be tested in human melanoma, it is consistent with the known properties of this highly aggressive malignancy.
16.8 Considerations for Studying Fusion In Vivo To prove fusion and genomic hybridization requires identification of genes or chromosomes from both of the putative fusion partners in the same cell or cells. Hence fusion has been well-documented in tumor xenografts in animals where hybrids were identified by the presence of both tumor and host genes. Little is yet known of the extent of cancer cell fusion in humans. While a few human cases have recently been reported (Andersen et al. 2007, 2009, Chakraborty et al. 2004, Yilmaz et al.
384
R. Lazova et al.
2005) only one of these involving macrophage-myeloma fusion in osteoclast formation definitively proved fusion (Andersen et al. 2007, 2009). The use of myeloma clone-specific immunoglobulin rearrangements as parental markers of myeloma cells can thus be used to further investigate questions of fusion in myeloma. Other studies have suggested that incorporation of BMDCs into tumor cells can occur through differentiation or neoplastic transformation without fusion (Avital et al. 2007, Cogle et al. 2007, Houghton et al. 2004). It is possible that both mechanisms are operative in cancer as well as in healthy tissue regeneration and repair, and this remains to be resolved. The use of allogeneic HSC transplants in medicine followed by the unfortunate development of secondary malignancies provides a potential source of pathology material for study (Chakraborty et al. 2004, Yilmaz et al. 2005). However, such cases are in limited supply and it will take some time to determine the extent of fusion in human cancer by this technique alone. Further, while this technique can verify fusion of BMDCs and cancer cells, the cellular nature of the BMDC cannot at this point be determined. Another problem is that the frequency of cancer cell fusion may be quite low, as it is in culture (roughly 1 in 105–107 non-fused cells) making fusion events difficult if not impossible to follow in vivo (Rachkovsky et al. 1998). Also, depending on the time when a particular tumor is analyzed, the number of hybrid cells could range from none, should hybridization not have occurred, to 100% if hybrids had overgrown a preexisting tumor or initiated a new tumor, e.g. a metastasis. Further, hybrid cells in a tumor could result from a single progenitor hybrid or from multiple hybrids formed from separate fusions. It is thus difficult to study the molecular mechanisms of cancer cell fusion in vivo, or to estimate its frequency. Until more progress is made in these and other areas, the impact of BMDC incorporation into human tumors, whether by fusion or other mechanisms, remains to be determined.
16.9 Implications Two of the hybrid-associated features described above, enhanced migration and autophagy, could together have important implications for the initiation of metastasis. Remarkably, both features may be activated through GnT-V-mediated addition of β1,6-branched oligosaccharides. For the primary carcinoma or melanoma cell, a migratory phenotype would imply loss of adhesion to adjoining cells in the epidermis, activation of matrix proteinases, induction of chemotaxis and tropism, and major cytoskeletal changes. But cancer cells often exist under hypoxic conditions. In such situations, autophagy could play a survival role by providing a nutrient source independent of the vasculature. Interestingly, the hypoxic regions of tumors have long been thought to select for more aggressive cancer cells that can survive under limited nutrient and oxygen supply (Graeber et al. 1996). Should cancer cell autophagy be linked to phagocytosis as it is in macrophages (see below), nutrients could be continuously phagocytosed from external sources (cellular debris, matrix fragments, etc.) and digested in autophagolysosomes, rendering metastatic cells constitutively independent of a direct blood supply.
16
Cancer Cell Fusion with Myeloid Cells
385
Tumor cell-BMDC fusion as a source of metastatic cells would imply that prevention of fusion, or of early, rate-limiting post-fusion events might prevent metastasis (e.g. Parris 2008). With better understanding should come better strategies for targeting vulnerable steps in fusion and the generation of hybrids. Postfusion events and hybrid formation could present other fruitful areas of focus, for example molecular steps governing the integration of parental fusion partner genes into hybrid genomes, or those involved with activation of master regulatory genes that are rate-limiting in development of a migratory phenotype. Early post-fusion cells are also likely to express unique antigenic profiles making them susceptible to immunotherapy. The cancer cell-BMDC fusion theory presents a unifying explanation for tumor progression. It seems that this theory is not only possible but likely to be correct to at least some degree, with the remaining question being how extensively does it contribute to progression of human cancers? There are many areas to consider regarding the therapeutic implications of fusion itself. Should it be determined that fusion indeed underlies metastasis, or at least some aspects of it, then new therapeutic paradigms would surely emerge, for example in prevention of fusion itself or destruction of fused cells based on unique molecular signatures. Based on the information gathered by several laboratories to date, we would urge more scientists to enter this most important and interesting area of cancer research. Acknowledgments We gratefully acknowledge the many and invaluable contributions of David Bermudes, Jean Bolognia, Douglas Brash, Dennis Cooper, Lynn Margulis, Josh Pawelek, James Platt, Michael Rachkovsky, Stefano Sodi, and Yesim Yilmaz. Supported in part by a gift from Amway, Inc.
References Aichel O (1911) “Über Zellverschmelzung mit Qualitativ Abnormer Chromosomenverteilung als Ursache der Geschwulstbildung” In: Roux W (ed) Vorträge und Aufsätze über Entvickelungsmechanik Der Organismen, Chapter XIII, Wilhelm Engelmann, Leipzig, pp 92–111 Alonso SR, Tracey L, Ortiz P et al (2007) A high-throughput study in melanoma identifies epithelial-mesenchymal transition as a major determinant of metastasis. Cancer Res 67:3450–3460 Amer AO, Byrne BG, Swanson MS (2005) Macrophages rapidly transfer pathogens from lipid raft vacuoles to autophagosomes. Autophagy 1:53–58 Amer AO, Swanson MS (2005) Autophagy is an immediate macrophage response to Legionella pneumophila. Cell Microbiol 7:765–778 Ammon C, Meyer SP, Schwarzfischer L et al (2000) Comparative analysis of integrin expression on monocyte-derived macrophages and monocyte-derived dendritic cells. Immunology 100: 364–369 Andersen TL, Boissy P, Sondergaard TE et al (2007) Osteoclast nuclei of myeloma patients show chromosome translocations specific for the myeloma cell clone: a new type of cancer-host partnership? J Pathol 211:10–17 Andersen TL, Søe K, Sondergaard TE et al (2009) Myeloma cell-induced disruption of bone remodelling compartments leads to osteolytic lesions and generation of osteoclast-myeloma hybrid cells. Br J Haematol 148:551–561
386
R. Lazova et al.
Aplin AE, Howe A, Alahari SK et al (1998) Signal transduction and signal modulation by cell adhesion receptors: the role of integrins, cadherins, immunoglobulin-cell adhesion molecules, and selectins. Pharmacol Rev 50:197–263 Atkin NB (1979) Premature chromosome condensation in carcinoma of the bladder: presumptive evidence for fusion of normal and malignant cells. Cytogenet Cell Genet 12:217–219 Aviles D, Jami J, Rousset JP et al (1977) Tumor x host cell hybrids in the mouse: chromosomes from the normal cell parent maintained in malignant hybrid tumors. J Natl Cancer Inst 58: 1391–1397 Avital I, Moreira AL, Klimstra DS et al (2007) Donor-derived human bone marrow cells contribute to solid organ cancers developing after bone marrow transplantation. Stem Cells 25:2903–2909 Balkwill F, Charles KA, Mantovani A (2005) Smoldering and polarized inflammation in the initiation and promotion of malignant disease. Cancer Cell 7:211–217 Barouch R, Appel E, Kazimirsky G et al (2001) Macrophages express neurotrophins and neurotrophin receptors. Regulation of nitric oxide production by NT-3. J Neuroimmunol 112:72–77 Barrallo-Gimeno A, Nieto MA (2005) The Snail genes as inducers of cell movement and survival: implications in development and cancer. Development 132:3151–3161 Barski G, Cornefert FR (1962) Characteristics of “hybrid”-type clonal cell lines obtained from mixed cultures in vitro. J Natl Cancer Inst 28:801–821 Beilmann M, Odenthal M, Jung W et al (1997) Neoexpression of the c-met/hepatocyte growth factor-scatter factor receptor gene in activated monocytes. Blood 90:4450–4458 Beuret L, Flori E, Denoyelle C et al (2007) Up-regulation of MET expression by alpha-melanocytestimulating hormone and MITF allows hepatocyte growth factor to protect melanocytes and melanoma cells from apoptosis. J Biol Chem 282:14140–14147 Bingle L, Brown NJ, Lewis CE (2002) The role of tumour-associated macrophages in tumour progression: implications for new anticancer therapies. J Pathol 196:254–265 Bjerkvig R, Tysnes BB, Aboody KS et al (2005) Opinion: the origin of the cancer stem cell: current controversies and new insights. Nat Rev Cancer 5:899–904. Erratum in: Nat Rev Cancer 5:995 Bjerregaard B, Holck S, Christensen IJ et al (2006) Syncytin is involved in breast cancerendothelial cell fusions. Cell Mol Life Sci 63:1906–1911 Boccaccio C, Comoglio PM (2006) Invasive growth: a MET-driven genetic programme for cancer and stem cells. Nat Rev Cancer 6:637–645 Boveri T (2008) Concerning the Origin of Malignant Tumors. Translated and annotated by Harris H. J Cell Sci 121(Supplement 1):1–84. Bradshaw AD, Sage EH (2001) SPARC, a matricellular protein that functions in cellular differentiation and tissue response to injury. J Clin Invest 107:1049–1054 Breier A, Barancik M, Sulova Z et al (2005) P-glycoprotein – implications of metabolism of neoplastic cells and cancer therapy. Curr Cancer Drug Targets 5:457–68 Bronisz A, Sharma SM, Hu R et al (2006) Microphthalmia-associated transcription factor interactions with 14-3-3 modulate differentiation of committed myeloid precursors. Mol Biol Cell 17:3897–3906 Carlson JA, Linette GP, Aplin A et al (2007) Melanocyte receptors: clinical implications and therapeutic relevance. Dermatol Clin 25:541–557 Chakraborty A, Lazova R, Davies S et al (2004) Donor DNA in a renal cell carcinoma metastasis from a bone marrow transplant recipient. Bone Marrow Transplant 34:183–186 Chakraborty AK, de Freitas Sousa J, Espreafico EM et al (2001a) Human monocyte x mouse melanoma fusion hybrids express human gene. Gene 275:103–106 Chakraborty AK, Funasaka Y, Ichihashi M et al (1999) Upregulation of mRNA for the melanocortin-1 receptor but not for melanogenic proteins in macrophage x melanoma fusion hybrids exhibiting increased melanogenic and metastatic potential. Pigm Cell Res 12: 355–366
16
Cancer Cell Fusion with Myeloid Cells
387
Chakraborty AK, Funasaka Y, Ichihashi M, Pawelek JM (2009) Upregulation of alpha and beta integrin subunits in metastatic macrophage-melanoma fusion hybrids. Melanoma Res 19:343–349 Chakraborty AK, Kolesnikova N, Sousa JDF et al (2003) Expression of c-Met proto-oncogene in metastatic macrophage x melanoma fusion hybrids: implication of its possible role in MSHinduced motility. Oncol Res 14:163–174 Chakraborty AK, Pawelek JM (2003) GnT-V, macrophages, and cancer metastasis: A common link. Clin Exp Metastasis 20:365–373 Chakraborty AK, Pawelek JM (2007) Beta1,6-branched oligosaccharides regulate melanin content and motility in macrophage-melanoma fusion hybrids. Melanoma Res 17:9–16 Chakraborty AK, Pawelek JM, Ikeda Y et al (2001b) Macrophage fusion up-regulates N-acetylglucosaminyltransferase V, β1,6 branching, and metastasis in Cloudman S91 mouse melanoma cells. Cell Growth and Diff 12:623–630. 2001 Chakraborty AK, Sodi S, Rachkovsky M et al (2000) A spontaneous murine melanoma lung metastasis comprised of host x tumor hybrids. Cancer Res 60:2512–2519 Chakraborty AK, Yamaga S (2003) Differential gene expression in genetically matched mouse melanoma cells with different metastatic potential. Gene 315:165–175 Chambers AF, Groom AC, MacDonald, IC (2002) Dissemination and growth of cancer cells in metastatic sites. Nat Rev Cancer 2:563–572 Chammas R, Veiga SS, Travassos LR et al (1993) Functionally distinct roles for glycosylation of alpha and beta integrin chains in cell matrix interactions. Proc Natl Acad Sci USA 90: 1795–1799 Chang MH, Hua CT, Isaac EL et al (2004) Transthyretin interacts with the lysosome-associated membrane protein (LAMP-1) in circulation. Biochem J 382:481–489 Charest A, Pépin A, Shetty R et al (2006) Distribution of SPARC during neovascularisation of degenerative aortic stenosis. Heart 92:1844–1849 Clark WH, Mastangelo MJ, Ainsworth AM et al (1977) Current concepts of the biology of human cutaneous malignant melanoma. Adv Cancer Res 24:267–338 Cogle CR, Theise ND, Fu D et al (2007) Bone marrow contributes to epithelial cancers in mice and humans as developmental mimicry. Stem Cells 25:1881–1887 Condeelis J, Segall JE (2003) Intravital imaging of cell movement in tumours. Nat Rev Cancer 3:921–930 Coopman PJ, Do MT, Thompson EW et al (1998) Phagocytosis of cross-linked gelatin matrix by human breast carcinoma cells correlates with their invasive capacity. Clin Cancer Res 4: 507–515 Dagher G, Donne N, Klein C, et al (2003) HDL-mediated cholesterol uptake and targeting to lipid droplets in adipocytes. J Lipid Res 44:1811–1820 Damiani MT, Colombo MI (2003) Microfilaments and microtubules regulate recycling from phagosomes. Exp Cell Res 289:152–161 Damjanovski S, Huynh MH, Motamed K et al (1998) Regulation of SPARC expression during early Xenopus development: evolutionary divergence and conservation of DNA regulatory elements between amphibians and mammals. Dev Genes Evol 207:453–461 Danen EH, Ten Berge PJ, Van Muijen GN et al (1994) Emergence of alpha 5 beta 1 fibronectinand alpha v beta 3 vitronectin-receptor expression in melanocytic tumour progression. Histopathology 24:249–256 Darlington GJ, Bernhard HP, Ruddle FH (1974) Human serum albumin phenotype activation in mouse hepatoma – human leukocyte cell hybrids. Science 185:859–862 Davidson RL, Ephrussi RLB, Yamamoto K (1966) Regulation of pigment synthesis in mammalian cells, as studied by somatic hybridization. Proc Nat Acad Sci USA 56:1437–1440 Defendi V, Ephrussi B, Koprowski H et al (1967) Properties of hybrids between polyomatransformed and normal mouse cells. Proc Nat Acad Sci USA 57:299–305 Demetriou M, Nabi IR, Coppolino M et al (1995). Reduced contact-inhibition and substratum adhesion in epithelial cells expressing GlcNAc-transferase V. J Cell Biol 130:383–392
388
R. Lazova et al.
Dennis J, Waller CA, Schirrmacher V (1984) Identification of asparagine-linked oligosaccharides involved in tumor cell adhesion to laminin and type IV collagen. J Cell Biol 99:1034–1044 Dennis JW, Granovsky M, Warren CE (1999) Glycoprotein glycosylation and cancer progression. Biochim Biophys Acta 1473:21–34 Dittmar T, Nagler C, Schwitalla S et al (2009) Recurrence cancer stem cells–made by cell fusion? Med Hypotheses 73:542–547 Dosaka-Akita H, Miyoshi E, Suzuki O et al (2004) Expression of N-acetylglucosaminyltransferase V is associated with prognosis and histology in non-small cell lung cancers. Clin Cancer Res 10:1773–1779 Duelli D, Lazebnik Y (2007) Cell-to-cell fusion as a link between viruses and cancer. Nat Rev Cancer 7:968–976 Elsegood CL, Zhuo Y, Wesolowski GA et al (2006) M-CSF induces the stable interaction of cFms with alphaVbeta3 integrin in osteoclasts. Int J Biochem Cell Biol 38:1518–1529 Ferlicot S, Vincent-Salomon A, Medioni J et al (2004) Wide metastatic spreading in infiltrating lobular carcinoma of the breast. Eur J Cancer 40:336–341 Fernandes B, Sagman U, Auger M et al (1991) β1,6-branched oligosaccharides as a marker of tumor progression in human breast and colon neoplasia. Cancer Res 51:718–723 Fidler IJ (2003) The pathogenesis of cancer metastasis: the seed and soil hypothesis revisited. Nat Rev Cancer 3:453–458 Fidler IJ, Hart IR (1982) Biological diversity in metastatic neoplasms: origins and implications. Science 217:998–1003 Fidler IJ, Kripke ML (1977) Metastasis results from preexisting variant cells within a malignant tumor. Science 197:893–895 Fortuna MB, Dewey MJ, Furmanski P (1989) Cell fusion in tumor development and progression: occurrence of cell fusion in primary methylcholanthrene-induced tumorigenesis. Int J Cancer 44:731–737 Fortuna MB, Dewey MJ, Furmanski P (1990) Enhanced lung colonization and tumorigenicity of fused cells isolated from primary MCA tumors. Cancer Lett 55:109–114 Friedl P (2004) Prespecification and plasticity: shifting mechanisms of cell migration. Curr Opin Cell Biol 16:14–23 Friedl P, Wolf K (2003) Proteolytic and non-proteolytic migration of tumour cells and leucocytes. Biochem Soc Symp 70:277–285 Fugita T, Shiba H, Sakata M et al (2002) SPARC stimulates the synthesis of OPG/OCIF, MMP-2 and DNA in human periodontal ligament cells. J Oral Pathol Med 31:345–352. Erratum in: J Oral Pathol Med 31:04 (2002) Fukuda M, Spooncer E, Oates JE et al (1984) Structure of sialylated fucosyl lactosaminoglycan isolated from human granulocytes. J Biol Chem 25:10925–10935 Gaasch JA, Bolwahnn AB, Lindsey JS (2006) Hepatocyte growth factor-regulated genes in differentiated RAW 264.7 osteoclast and undifferentiated cells. Gene 369:142–152 Galbraith CG, Yamada KM, Galbraith JA (2007) Polymerizing actin fibers position integrins primed to probe for adhesion sites. Science 315:992–995 Garraway LA, Widlund HR, Rubin MA et al (2005) Integrative genomic analyses identify MITF as a lineage survival oncogene amplified in malignant melanoma. Nature 436:117–122 Geiger TR, Peeper DS (2005) The neurotrophic receptor TrkB in anoikis resistance and metastasis: a perspective. Cancer Res 65:7033–7036 Giacomoni D (1979) Tumorigenicity and intracisternal A-particle expression of hybrids between murine myeloma and lymphocytes. Cancer Res 39:4481–4484 Giannelli G, Astigiano S, Antonaci S et al (2002) Role of the alpha3beta1 and alpha6beta4 integrins in tumor invasion. Clin Exp Metastasis 19:217–230 Gladson CL, Cheresh DA (1991) Glioblastoma expression of vitronectin and the alpha v beta 3 integrin. Adhesion mechanism for transformed glial cells. J Clin Invest 88:1924–1932 Goldenberg DM (1968) [On the progression of malignity: a hypothesis.] Klin Wschr 46:898 (German)
16
Cancer Cell Fusion with Myeloid Cells
389
Goldenberg DM, Bhan RD, Pavia RA (1971) In vivo human-hamster somatic cell fusion indicated by glucose-6-phosphate dehydrogenase and lactate dehydrogenase profiles. Cancer Res 31:1148–1152 Goldenberg DM, Gotz H (1968) On the ‘human’ nature of highly malignant heterotransplantable tumors of human origin. Eur J Cancer 4:547–548 Goldenberg DM, Pavia RA, Tsao MC (1974) In vivo hybridization of human tumour and normal hamster cells. Nature 250:649–651 Gottesman MM, Ling V (2006) The molecular basis of multidrug resistance in cancer: the early years of P-glycoprotein research. FEBS Lett 580:998–1009 Gourdeau H, Fournier REK (1990) Genetic analysis of mammalian cell differentiation. Ann Rev Cell Biol 6:69–94 Graeber TG, Osmanian C, Jacks T et al (1996) Hypoxia-mediated selection of cells with diminished apoptotic potential in solid tumours. Nature 379:88–91 Graves TK, Patel S, Dannies PS (2001) Misfolded growth hormone causes fragmentation of the Golgi apparatus and disrupts endoplasmic reticulum-to-Golgi traffic. J Cell Sci 114:3685–3694 Guo HB, Lee I, Kamar M et al (2002) Aberrant N-glycosylation of β1 integrin causes reduced α5β1 integrin clustering and stimulates cell migration. Cancer Res 62:6837–6845 Guo W, Lasky JL 3rd, Wu H (2006) Cancer stem cells. Pediatr Res 59:59R–64R Gupta PB, Mani S, Yang J, Hartwell K et al (2005) The evolving portrait of cancer metastasis. Cold Spring Harb Symp Quant Biol 6:291–297 Halaban R, Nordlund J, Francke U et al (1980) Supermelanotic hybrids derived from mouse melanomas and normal mouse cells. Somatic Cell Genet 6:29–44 Handerson T, Camp R, Harigopal M et al (2005) β1,6-Branched oligosaccharides are associated with metastasis and predict poor outcome in breast carcinoma. Clin Can Res 11:2969–2973 Handerson T, Pawelek J (2003) β1,6-branched oligosaccharides and coarse vesicles: A common and pervasive phenotype in melanoma and other human cancers. Cancer Res 63:5363–5369 Handerson T, Berger A, Harigopol M et al (2007) Melanophages reside in hypermelanotic, aberrantly glycosylated tumor areas and predict improved outcome in primary cutaneous malignant melanoma. J Cutaneous Pathol 34:667–738 Hariri M, Millane G, Guimond M et al (2000) Biogenesis of multilamellar bodies via autophagy. Mol Biol Cell 11:255–268 Harris H (1988) The analysis of malignancy by cell fusion: the position in 1988. Cancer Res 48:3302–3306 Hart IR (1984) Tumor cell hybridization and neoplastic progression In: Nicolson, GL, Milas L (eds) Cancer invasion and metastasis: biologic and therapeutic aspects. Raven Press, New York, NY, pp 133–143 Hendrix MJ, Seftor EA, Hess A et al (2003) Vasculogenic mimicry and tumour-cell plasticity: lessons from melanoma. Nat Rev Cancer 3:4114–4121 Hersey P, Zhang XD (2008) Adaptation to ER stress as a driver of malignancy and resistance to therapy in human melanoma. Pigment Cell Melanoma Res 21:358 Herzog EL, Van Arnam J, Hu B et al (2007) Lung-specific nuclear reprogramming is accompanied by heterokaryon formation and Y chromosome loss following bone marrow transplantation and secondary inflammation. FASEB J 21:2592–12601 Hess AR, Postovit LM, Margaryan NV et al (2005) Focal adhesion kinase promotes the aggressive melanoma phenotype. Cancer Res 65:9851–9860 Houghton J, Stoicov C, Nomura S et al (2004) Gastric cancer originating from bone marrowderived cells. Science 306:1568–1571 Hu F, Pasztor LM (1975) In vivo hybridization of cultured melanoma cells and isogenic normal mouse cells. Differentiation 4: 93–97 Huang B, Zhao J, Li H et al (2005) Toll-like receptors on tumor cells facilitate evasion of immune surveillance. Cancer Res 65:5009–5014
390
R. Lazova et al.
Ihara S, Miyoshi E, Nakahara S et al (2004) Addition of beta1-6 GlcNAc branching to the oligosaccharide attached to Asn 772 in the serine protease domain of matriptase plays a pivotal role in its stability and resistance against trypsin. Glycobiology 14:139–146 Jacobsen BM, Harrell JC, Jedlicka P et al (2006) Spontaneous fusion with, and transformation of mouse stroma by, malignant human breast cancer epithelium. Cancer Res 66:8274–8279 Janzen HW, Millman PA, Thurston OG (1971) Hybrid cells in solid tumors. Cancer 27:455–459 Jasiulionis MG, Chammas R, Ventura AM et al (1996) α6β1-integrin, a major cell surface carrier of β1–6-branched oligosaccharides, mediates migration of EJ-ras-transformed fibroblasts on laminin-1 independently of its glycosylation state. Cancer Res 56:1682–1689 Jonasson J, Povey S, Harris H (1977) The analysis of malignancy by cell fusion. VII. Cytogenetic analysis of hybrids between malignant diploid cells and of tumours derived from them. J Cell Sci 24:217–254 Jorajuria S, Dereuddre-Bosquet N, Becher F.et al (2004) ATP binding cassette multidrug transporters limit the anti-HIV activity of zidovudine and indinavir in infected human macrophages. Antivir Ther 9:519–528 Juliano RL (1993) The role of beta 1 integrins in tumors. Semin Cancer Biol 4:277–283 Kanetsky PA, Rebbeck TR, Hummer AJ et al (2006) Population-based study of natural variation in the melanocortin-1 receptor gene and melanoma. Cancer Res. 66:9330–9337 Kang JY, Dolled-Filhart M, Ocal IT et al (2003) Tissue microarray analysis of hepatocyte growth factor/Met pathway components reveals a role for Met, matriptase, and hepatocyte growth factor activator inhibitor 1 in the progression of node-negative breast cancer. Cancer Res 63:1101–1105 Kang Y, Massague J (2004) Epithelial-mesenchymal transitions: twist in development and metastasis. Cell 118:277–279 Kerbel RS, Lagarde AE, Dennis JW et al (1983) Spontaneous fusion in vivo between normal host and tumor cells: possible contribution to tumor progression and metastasis studied with a lectin-resistant mutant tumor. Mol Cell Biol 3:523–538 Kerr CA (2004) A twist to metastasis. Lancet Oncol 5:465 Kerschmann RL, Woda BA, Majno G (1995) The fusion of tumor cells with host cells; Reflections on an ovarian tumor. Perspect Biol Med 38:467–472 Klionsky DJ, Abeliovich H, Agostinis P et al (2008) Guidelines for the use and interpretation of assays for monitoring autophagy in higher eukaryotes. Autophagy 4:151–175 Kohler G, Milstein C (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256:495–497 Kovacs G (1985) Premature chromosome condensation: evidence for in vivo cell fusion in human malignant tumours. Int J Cancer 36:637–641 Kovacs WJ, Schrader M, Walter I et al (2004) The hypolipidemic compound cetaben induces changes in Golgi morphology and vesicle movement. Histochem Cell Biol 122:95 Kurita-Taniguchi M, Hazeki K, Murabayashi N et al (2002) Molecular assembly of CD46 with CD9, alpha3-beta1 integrin and protein tyrosine phosphatase SHP-1 in human macrophages through differentiation by GM-CSF. Mol Immunol 38:689–700 Lagarde, AE, Kerbel RS (1984) Somatic cell hybridization in vivo and in vitro in relation to the metastatic phenotype. Biochim Biophys Acta 823:81–110 Lam CW, Getting SJ, Perretti M (2005) In vitro and in vivo induction of heme oxygenase 1 in mouse macrophages following melanocortin receptor activation. J Immunol 174:2297–2304 Lam CW, Perretti M, Getting SJ (2006) Melanocortin receptor signaling in RAW264.7 macrophage cell line. Peptides 27:404–412 Lane TF, Sage EH (1994) The biology of SPARC, a protein that modulates cell-matrix interactions. FASEB J. 8:163–173 Larizza L, Schirrmacher V (1984) Somatic cell fusion as a source of genetic rearrangement leading to metastatic variants. Cancer Metastasis Rev 3: 193–222 Larizza L, Schirrmacher V, Stöhr M et al. (1984a) Inheritance of immunogenicity and metastatic potential in murine cell hybrids from the T-lymphoma ESb08 and normal spleen lymphocytes. J Natl Cancer Inst 72:1371–1381
16
Cancer Cell Fusion with Myeloid Cells
391
Larizza L, Schirrmacher V, Graf L et al (1984b) Suggestive evidence that the highly metastatic variant ESb of the T-cell lymphoma Eb is derived from spontaneous fusion with a host macrophage. Int J Cancer 34:699–707 Lau LC, Tan PH, Chong TW et al (2007) Cytogenetic alterations in renal tumors: a study of 38 Southeast Asian patients. Cancer Genet Cytogenet 175:1–7 Lemaire S, Van Bambeke F, Mingeot-Leclercq MP et al (2007) Modulation of the cellular accumulation and intracellular activity of daptomycin towards phagocytized Staphylococcus aureus by the P-glycoprotein (MDR1) efflux transporter in human THP-1 macrophages and madin-darby canine kidney cells. Antimicrob Agents Chemother 51:2748–2757 Leppa S, Heino J, Jalkanen M (1995) Increased glycosylation of beta 1 integrin affects the interaction of transformed s115 mammary epithelial cells with laminin-1. Cell Growth Differ 6:853–861 Levine AJ (1995) Tumor suppressor genes. In: Mendelsohn J, Howley PM, Israel MA, Liotta LA (eds) The molecular basis of cancer. Philadelphia, PA: WB Saunders Co., pp 86–104 Levy C, Khaled M, Fisher DE (2006) MITF: master regulator of melanocyte development and melanoma oncogene. Trends Mol Med 12:406–414 Lin EY, Nguyen AV, Russell RG et al. (2001) Colony-stimulating factor 1 promotes progression of mammary tumors to malignancy. J Exp Med 193:727–740 Lu X, Kang Y (2009) Cell fusion as a hidden force in tumor progression. Cancer Res 69:8536–8539 Lugini L, Lozupone F, Matarrese P et al (2003) Potent phagocytic activity discriminates metastatic and primary human malignant melanomas: a key role of ezrin. Lab Invest 83:1555–1567 Lugini L, Matarrese P, Tinari A et al (2006) Cannibalism of live lymphocytes by human metastatic but not primary melanoma cells. Cancer Res 66:3629–3638 Ma L, Teruya-Feldstein J, Weinberg RA (2007) Tumour invasion and metastasis initiated by microRNA-10b in breast cancer. Nature 449:682–688 Malawista SE, Weiss MC (1974) Expression of differentiated function in hepatoma cell hybrids: high frequency of induction of mouse albumin production in rat hepatoma-mouse lymphoblast hybrids. Proc Natl Acad Sci USA 71:927–931 Manna SK, Sarkar A, Sreenivasan Y (2006) Alpha-melanocyte-stimulating hormone downregulates CXC receptors through activation of neutrophil elastase. Eur J Immunol 36:754–769 Mansergh FC, Wells T, Elford C et al (2007) Osteopenia in Sparc (osteonectin)-deficient mice: characterization of phenotypic determinants of femoral strength and changes in gene expression. Physiol Genom 32:64–73 Mantovani A, Allavena P, Sica A et al (2008) Cancer-related inflammation. Nature 454:436–444 Martinek N, Shahab J, Sodek J et al (2007) Is SPARC an evolutionarily conserved collagen chaperone? J Dent Res 86:296–305 McGill GG, Haq R, Nishimura EK et al (2006) c-Met expression is regulated by Mitf in the melanocyte lineage. J Biol Chem 281:10365–10373 McLean GW, Carragher NO, Avizienyte E et al (2005) The role of focal-adhesion kinase in cancer – a new therapeutic opportunity. Nat Rev Cancer 5:505–515 Mekler LB (1968) [A general theory of oncogenesis.] Materials of symposia on general immunology. The club of immunologists of NF Gamaleya. Inst Epidemiol Microbiol 3:91–100 (Russian) Mekler LB (1971) [Hybridization of transformed cells with lymphocytes as 1 of the probable causes of the progression leading to the development of metastatic malignant cells] Vestn Acad Med Nauk SSR (Bulletin of the USSR Acad Med Sci) 26:80–89 (Russian) Michot JM, Van Bambeke F, Mingeot-Leclercq MP et al (2004) Active efflux of ciprofloxacin from J774 macrophages through an MRP-like transporter. Antimicrob Agents Chemother 48: 2673–2682 Miller FR, McInerney D, Rogers C et al. (1988) Spontaneous fusion between metastatic mammary tumor subpopulations. J Cell Biochem 36:129–136 Miller FR, Mohamed AN, McEachern D (1989) Production of a more aggressive tumor cell variant by spontaneous fusion of two mouse tumor subpopulations. Cancer Res 49:4316–4321
392
R. Lazova et al.
Molteni M, Marabella D, Orlandi C et al (2006) Melanoma cell lines are responsive in vitro to lipopolysaccharide and express TLR-4. Cancer Lett 235:75–83 Montcourrier P, Mangeat PH, Valembois C et al (1994) Characterization of very acidic phagosomes in breast cancer cells and their association with invasion. J Cell Sci 107:2381–2391 Mortensen K, Lichtenberg J, Thomsen PD et al (2004) Spontaneous fusion between cancer cells and endothelial cells. Cell Mol Life Sci 61:2125–2131 Munzarova M, Lauerova L, Capkova J (1992) Are advanced malignant melanoma cells hybrids between melanocytes and macrophages? Melanoma Res 2:127–129 Murata K, Miyoshi E, Kameyama M et al (2000) Expression of N-Acetylglucosaminyltransferase V in colorectal cancer correlates with metastasis and poor prognosis. Clin Cancer Res 6: 1772–1777 Natali PG, Hamby CV, Felding-Habermann B et al (1997) Clinical significance of alpha(v)beta3 integrin and intercellular adhesion molecule-1 expression in cutaneous malignant melanoma lesions. Cancer Res 57:1554–1560 Natali PG, Nicotra MR, Di Filippo F et al (1995) Expression of fibronectin, fibronectin isoforms and integrin receptors in melanocytic lesions. Br J Cancer 71:1243–1247 Nieto MA (2002) The snail superfamily of zinc-finger transcription factors. Nat Rev Mol Cell Biol 3:1551–1566 Nowell PC (1976) The clonal evolution of tumor cell populations. Science 194:23–28 Ochwat D, Hoja-Lukowicz D, Litynska A (2004) N-glycoproteins bearing beta1–6 branched oligosaccharides from the A375 human melanoma cell line analysed by tandem mass spectrometry. Melanoma Res 14:479–485 Parris GE (2008) 2-Deoxy-d-glucose as a potential drug against fusogenic viruses including HIV. Med Hypotheses 70:776–782 Pawelek JM (2005) Tumour-cell fusion as a source of myeloid traits in cancer. Lancet Oncol 6: 988–993 Pawelek JM (2000) Tumour cell hybridization and metastasis revisited. Melanoma Res 10:507–514 Pawelek JM, Chakraborty A, Yilmaz Y et al (2006) Co-opting macrophage traits in cancer progression: a consequence of tumor cell fusion? Contrib Microbiol 13:138–155. Pawelek JM, Chakraborty AK (2008a) Fusion of tumour cells with bone marrow-derived cells: a unifying explanation for metastasis. Nat Rev Cancer 8:377–386 Pawelek JM, Chakraborty AK (2008b) The cancer cell – leukocyte fusion theory of metastasis. Adv Cancer Res 101:397–444 Pawelek JM, Chakraborty AK, Rachkovsky ML et al (2000) Altered N-glycosylation in macrophage x melanoma fusion hybrids. Cell Mol Biol 45:1011–1027 Pernick NL, DaSilva M, Gangi MD et al (1999)“Histiocytic markers” in melanoma. Mol Pathol 12:1072–1077 Poche E, Litysk A, Amoresano A et al (2003) Glycosylation profile of integrin α3β1 changes with melanoma progression. Biochim Biophys Acta 1643:113–123 Pollard JW (2004) Tumour-educated macrophages promote tumour progression and metastasis. Nat Rev Cancer 4:71–78 Powers TP, Davidson RL (1996) Coordinate extinction of melanocyte-specific gene expression in hybrid cells. Somatic Cell Mol Gen 22:41–56 Powers TP, Shows TB, Davidson RL (1994) Pigment-cell-specific genes from fibroblasts are transactivated after chromosomal transfer into melanoma cells. Mol Cell Biol 14: 1179–1190 Rachkovsky M, Pawelek J. (1999) Acquired melanocyte stimulating hormone-inducible chemotaxis following macrophage fusion with Cloudman S91 melanoma cells. Cell Growth Differ 10:515–524 Rachkovsky MS, Sodi S, Chakraborty A et al (1998) Enhanced metastatic potential of melanoma x macrophage fusion hybrids. Clin Exp Metastasis 16: 299–312 Ramshaw IA, Carlsen S, Wang H et al (1983) The use of cell fusion to analyse factors involved in tumour cell metastasis. Int J Cancer 32:471–478
16
Cancer Cell Fusion with Myeloid Cells
393
Razi M, Chan EY, Tooze SA (2009) Early endosomes and endosomal coatomer are required for autophagy. J Cell Biol 185:305 Reed MJ, Puolakkainen P, Lane TF et al (1993) Differential expression of SPARC and thrombospondin 1 in wound repair: immunolocalization and in situ hybridization. J Histochem Cytochem 41:1467–1477 Ricci A, Greco S, Mariotta S et al (2000) Neurotrophin and neurotrophin receptor expression in alveolar macrophages: an immunocytochemical study. Growth Factors 18:193–202 Rizvi AZ, Swain JR, Davies PS et al (2006) Bone marrow-derived cells fuse with normal and transformed intestinal stem cells. Proc Natl Acad Sci USA 103:6321–6325 Robert G, Gaggioli C, Bailet O et al (2006) SPARC represses E-cadherin and induces mesenchymal transition during melanoma development. Cancer Res 66:7516–7523 Roos E, La Riviere G, Collard JG et al (1985) Invasiveness of T-cell hybridomas in vitro and their metastatic potential in vivo. Cancer Res 45:6238–6243 Rovida E, Lugli B, Barbetti V et al (2005) Focal adhesion kinase is redistributed to focal complexes and mediates cell spreading in macrophages in response to M-CSF. Biol Chem 386: 919–929 Ruff MR, Pert DB (1984) Small cell carcinoma of the lung: Macrophage-specific antigens suggest hemopoietic stem cell origin. Science 225:1034–1036 Rupani R, Handerson T, Pawelek JM (2004) Co-localization of β1,6-branched oligosaccharides and coarse melanin in macrophage-melanoma fusion hybrids and human melanoma cells in vitro. Pigment Cell Res 17:281–288 Rutkowski DT, Kaufman RJ (2007) That which does not kill me makes me stronger: adapting to chronic ER stress. Trends Biochem Sci 32:469 Saitoh O, Wang WC, Lotan R et al (1992) Differential glycosylation and cell surface expression of lysosomal membrane glycoproteins in sublines of a human colon cancer exhibiting distinct metastatic potentials. J Biol Chem 267:5700–5711 Salama ME, Worsham MJ, DePeralta-Venturina M (2003) Malignant papillary renal tumors with extensive clear cell change: a molecular analysis by microsatellite analysis and fluorescence in situ hybridization. Arch Pathol Lab Med 127:1176–1181 Sarafian V, Jadot M, Foidart JM et al (1998) Expression of Lamp-1 and Lamp-2 and their interactions with galectin-3 in human tumor cells. Int J Cancer 75:105–111 Sawada R, Lowe JB, Fukuda M (1993) E-selectin-dependent adhesion efficiency of colonic carcinoma cells is increased by genetic manipulation of their cell surface lysosomal membrane glycoprotein-1 expression levels. J Biol Chem 268:12675–12681 Scaletta LJ Ephrussi B (1965) Hybridization of normal and neoplastic cells in vitro. Nature 205:1169 Schioppa T, Uranchimeg B, Saccani A et al (2003) Regulation of the chemokine receptor CXCR4 by hypoxia. J Exp Med 198:1391–1402 Seelentag WK, Li WP, Schmitz SF et al (1998) Pronostic value of beta 1,6-branched oligosaccharides in human colorectal carcinoma. Cancer Res 58:5559–5564 Shabo IM, Stål O, Olsson H et al (2008) Breast cancer expression of CD163, a macrophage scavenger receptor, is related to early distant recurrence and reduced patient survival. Int J Cancer 23:780–786 Sharif MN, Sosic D, Rothlin CV et al (2006) Twist mediates suppression of inflammation by type I IFNs and Axl. J Exp Med 203:1891–1901 Shinji H, Kamada M, Seki K et al (2007) Expression and distribution of very late antigen-5 in mouse peritoneal macrophages upon ingestion of fibronectin-bound Staphylococcus aureus. Microbiol Immunol 51:163–171 Sidebottom E (1980) The analysis of malignancy by cell fusion. In Vitro 16:77–86 Sosi D, Richardson JA, Yu K et al (2003) Twist regulates cytokine gene expression through a negative feedback loop that represses NF-kappaB activity. Cell 112:169–180 Stanbridge EJ (1976) Suppression of malignancy in human cells. Nature 260:17–20 Streubel B, Chott A, Huber D et al (2004) Lymphoma-specific genetic aberrations in microvascular endothelial cells in B-cell lymphomas. N Engl J Med 351:250–259
394
R. Lazova et al.
Taylor AW (2005) The immunomodulating neuropeptide alpha-melanocyte-stimulating hormone (alpha-MSH) suppresses LPS-stimulated TLR4 with IRAK-M in macrophages. J Neuroimmunol 162:43–50 Tepass U, Fessler LI, Aziz A et al. (1994) Embryonic origin of hemocytes and their relationship to cell death in Drosophila. Development 120:1829–1837 Thiery JP (2002) Epithelial-mesenchymal transitions in tumour progression. Nat Rev Cancer 2:442–454 Thiery JP (2003) Epithelial-mesenchymal transitions in development and pathologies. Curr Opin Cell Biol 15:740–746 Thiery JP, Morgan M (2004) Breast cancer progression with a twist. Nat Med 10:777–778 Wang D, Shen Q, Chen YQ et al (2004) Collaborative activities of macrophage-stimulating protein and transforming growth factor-beta1 in induction of epithelial to mesenchymal transition: roles of the RON receptor tyrosine kinase. Oncogene 23:1668–1680 Warner TFCS (1975) Cell hybridization: An explanation for the phenotypic diversity of certain tumours. Med Hypothesis 1:51–57 Weinberg AS (1991) Tumor suppressor genes. Science 254:1138–1146 Wiener F, Fenyš EM, Klein G, Harris H (1972) Fusion of tumour cells with host cells. Nature 238:155–159 Wiener F, Fenyš EM, Klein G (1974a) Tumor-host hybrids in radiochimeras. Proc Nat Acad Sci USA 71:148–152 Wiener F, Klein G, Harris H (1974b) The analysis of malignancy by cell fusion. J Cell Sci 15: 177–183 Wolf K, Mazo I, Leung H et al (2003) Compensation mechanism in tumor cell migration: mesenchymal-amoeboid transition after blocking of pericellular proteolysis. J Cell Biol 160:267–277 Wong NC, Mueller BM, Barbas CF et al (1998) Alphav integrins mediate adhesion and migration of breast carcinoma cell lines. Clin Exp Metastasis 16:50–61 Yamamoto E, Ino K, Miyoshi E et al (2007) Expression of N-acetylglucosaminyltransferase V in endometrial cancer correlates with poor prognosis. Br J Cancer 97:1538–1544 Yamamoto H, Swoger J, Greene S et al (2000) β1,6,N-acetyl-glucosamine bearing N- glycans in human gliomas; implications for role in regulating invasivity. Cancer Res 60:134–142. Yang J, Mani SA, Donaher JL et al (2004) Twist, a master regulator of morphogenesis, plays an essential role in tumor metastasis. Cell 117:927–939 Yilmaz Y, Lazova R, Qumsiyeh M, Cooper, D et al (2005) Donor Y chromosome in renal carcinoma cells of a female BMT recipient: visualization of putative BMT-tumor hybrids by FISH. Bone Marrow Transplant 35:1021–1024 Zheng M, Fang H, Hakomori S (1994) Functional role of N-glycosylation in alpha 5 beta 1 integrin receptor. De-N-glycosylation induces dissociation or altered association of alpha 5 and beta 1 subunits and concomitant loss of fibronectin binding activity. J Biol Chem 269:12325–12331 Zhou BP, Deng J, Xia W et al (2004) Dual regulation of Snail by GSK-3beta-mediated phosphorylation in control of epithelial-mesenchymal transition. Nat Cell Biol 6:931–940
Chapter 17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer Reiner Strick, Matthias W. Beckmann, and Pamela L. Strissel
Abstract The overall focus of this review is the characterization and functional role of cell–cell fusions in connection with human endogenous retroviruses (HERV) in cancer. Examples of multinucleated cells presented include placental syncytiotrophoblasts, muscle myotubes, bone osteoclasts involved in normal human development and cell–cell fusions detected in tumors. Examples of multinucleated cells in various cancers include germ cell tumors, glioblastoma, melanoma, lung, breast, ovarian and endometrial carcinomas. The role of different HERVenvelope proteins mediating fusion or regulation of cells in tumors is highlighted. Although multinucleated cells are detected in various tumors, their origin, functional role and overall cellular fate are ambiguous. The effect of multiple cancer cells fusing and in contrast cancer cells fusing with somatic cells is also discussed. Understanding tumorigenesis has to ultimately link the knowledge between the function and action of multinucleated cells, cell fusion, HERVs, retroviruses and cell signalling pathways. Keywords Cancer · cell-cell fusions · HERV · multinucleated cells · polyploidy · retrovirus · syncytin · virus Abbreviations ASCT BMD BMDC CD eff-1 EM env
Alanine, serine and cysteine selective transporters Bone marrow derived Bone marrow derived cells Cluster of differentiation Epithelial fusion failure 1 Electron microscopy Envelope
R. Strick (B) Department of Gynaecology and Obstetrics, Laboratory for Molecular Medicine, University-Clinic Erlangen, D-91054 Erlangen, Germany e-mail:
[email protected]
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9_17, C Springer Science+Business Media B.V. 2011
395
396
gag HA HELLP HERV HIV HTLV IF IHC IUGR MFSD2 MSRV PE Pol RIA RNAi SU TGF TM
R. Strick et al.
Group-specific antigen Hemagglutinin Hemolysis elevated liver enzymes low platelet count Human endogenous retrovirus Human immune deficiency virus Human T-cell lymphotropic virus Immunofluorescence Immunohistochemistry Intrauterine growth restriction Major facilitator superfamily domain containing 2 Multiple sclerosis retrovirus Pre-eclampsia Polymerase Radioimmunoassay RNA interference Surface unit Transforming growth factor Transmembrane unit
Contents 17.1 Development and Polyploidy . . . . . . . . . . . . . . . . 17.1.1 Short History of Cell–Cell Fusions . . . . . . . . . . 17.1.2 Cell–Cell Fusions in Development, Differentiation and Viral-Induced . . . . . . . . . . . . . . . . . 17.1.3 Cell–Cell Fusions During Tumorigenesis . . . . . . . 17.2 Human Endogenous Retroviruses (HERVs) . . . . . . . . . . 17.2.1 HERV Expression in Human Cancers . . . . . . . . . 17.2.2 HERVs in Cancer Cell–Cell Fusions: Driver or Passenger 17.3 Cell–Cell Fusions in Cancer: Functional Role or Dead-End . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
396 398
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
399 404 408 412 413 414 417
17.1 Development and Polyploidy One text book fact imperative for human somatic cells, which constitute the entire organism, is diploidy, where genetic information on chromosomes consists of two homologous copies of each autosome and two sex chromosomes. In the normal process of mitosis the total number of chromosomes is a diploid set with 2n members. One exception to the diploid rule mentioned in every textbook are germ cells, like eggs and sperms, which are generated by meiosis and have a haploid chromosome number (n). The phenomenon of somatic polyploidy, meaning the existence of extra copies of all chromosomes is mostly found in the world of plants, invertebrates and protista and in some vital human developmental processes like muscle, bone and placenta formation. In addition, the existence of polyploid somatic cells
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
397
or aneuploidy (a loss or gain of certain chromosomes) in human malignancies is long-established. Although polyploid cells found in human tissues exist for many scientists, they mostly represent an exotic or odd research field and in vitro studies are mostly dismissed as artefactual. Therefore, this review will describe first, examples of polyploid human cells and show important differences in order to understand the different types of polyploidy; and second, depict the occurrence of cell–cell fusions as one cause for polyploidy in carcinoma; and third, describe human endogenous retroviruses (HERV) involved in cell–cell fusions and tumorigenesis. In order to understand polyploidy and the cellular causes for this phenomenon a short excurse into cell division and the prevention of it has to be made. Normal cell division is characterized by nuclear division (karyogenesis) and cytosolic separation (cytokinesis). Incomplete cytokinesis can be a programmed step in normal development producing tetraploid progeny (Glotzer 2001), or it can be part of cancer progression where tetraploid cells occur in tumors and further create genomic instability (Ganem et al. 2007). Following tetraploidization of cells subsequent divisions result in aneuploidy. Especially in the absence of functional p53 these cells are associated with a predisposition of tumorigenesis (for review: Margolis 2005). The suppression or slowing down of the spindle apparatus relative to the chromosomes usually leads to the formation of endopolyploid nuclei, where many copies of the same chromosomes are found in one nucleus (Fig. 17.1). The most common mechanism giving rise to endopolyploidy is endoreduplication, which means that two or more DNA replications take place without an intervening mitosis. If a cell divides after endoreduplication, the chromosomes appear as diplochromosomes or as bundles of multiple chromatids. In endomitosis, which is rarer than endoreduplication, chromosomes condense as in normal prophase, but the
Fig. 17.1 Three main possibilities of a cell (2n) leading to polyploidy are detected in different mammalian tissues: endoreduplication, endomitosis and cell–cell fusions. These polyploidies take place with a fusion of nuclei (karyogamy) or without (multinucleation). Multinucleated cells with nuclei from different origins are only found after cell–cell fusions
398
R. Strick et al.
nuclear membrane does not disintegrate. In endo-metaphase the sister chromosomes lie side-by-side giving the nuclei their characteristic appearance. After the sister chromosomes separate in endo-anaphase, they revert through endo-telophase into interphase. Polytenization is a modification of endoreduplication. The chromosomes are extended and the paired strands replicate again and again. The tendency to somatic pairing in diptera aligns the homologous chromomeres which form the bands in the polytene chromosomes. In other organisms, pairing of the homologous strands is more or less loose; consequently, banding is much less clear or even absent. Restitution, mostly from metaphase or anaphase, implies that the daughter chromosomes are included in the same nucleus and are products of defect spindle mechanisms or spindle poisons such as colchicine (C-mitosis). In cancer cells where mitosis similar to C-mitosis occasionally takes place spontaneously, is mostly the result of the formation of micronuclei and not polyploidy. Some well known examples describing the above mechanisms are megakaryocytes which produce platelets in mammals and represent normal polyploid cells due to a failure of karyogenesis and cytokinesis (Ravid et al. 2002, Fig. 1). Newer experiments further showed that the endomitotic process of megakaryocytes was a late failure of the cytokinesis due to a defect in the Rho/Rock-pathway (Lordier et al. 2008). Cardiomyocytes also have an incomplete cytokinesis leading to binucleated cells (Engel et al. 2006). Hepatocytes show high percentages of tetraploidy and octoploidy with mono- and binucleation in normal cells especially after liver damage and hepatitis due to incomplete cytokinesis, but not during carcinoma (Gupta 2000, Margall-Ducos et al. 2007). In normal adult human liver 30–40% represent polyploid cells (Kudryavtsev et al. 1993). Interestingly, tetraploid cells were able to proliferate resulting in mononucleated cells (Guidotti et al. 2003). In contrast to human trophoblasts, differentiation of mouse trophoblast stem cells into trophoblast giant cells is due to endoreduplication and suppression of cyclin-dependent kinase 1 by p57 (Ullah et al. 2008).
17.1.1 Short History of Cell–Cell Fusions After the establishment of the “cell theory” by Schleiden (1838) and Schwann (1839) that the basic structural element of all plants and animals is the cell, one of the first descriptions of multinucleated giant cells and the hypothesis of cell–cell fusion events was 1868 by Langhans in tuberculoid lesions. Today these specific multinucleated giant cells are morphologically classified as Langhans’ giant cells commonly found in immune granulomas and are characterized with less than 20 nuclei per cell arranged in a kind of a circle (Roger et al. 1972). Multinucleated cells were since then not only found in tuberculoid granulomas but also in other types of granulomas, like sarcoidosis (Okamoto et al. 2003), schistosomiasis (Boros 1989), sarcoid-like type granuloma e.g. colitis or rectal carcinoma (Williams and Williams 1967) and foreign body fusion events, like biodegradation of polymeric medical devices like sutures (Anderson 2000). The foreign body giant cells have in contrast to Langhans’ giant cells more than 20 nuclei per cell and are irregularly
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
399
arranged in the cell. All of these multinucleated giant cells are most likely due to cell fusions of macrophages recruited to the site of inflammation (Helming and Gordon 2007). Other early descriptions of multiple mitosis and giant nuclei in tumors of a larynx were done by Hansemann (1890). Also Winge (1927) described giant cells with up to 40–50 nuclei, but smaller in size than normal nuclei, in human tumors, but also in tumors of mice induced with tar. Interestingly, early detections of multinucleated cells in normal tonsils were described, but especially induced after a measles infection (Finkeldey 1931). The direct connection of cell fusions after viral infections was later shown for e.g. fusion of glioblastoma cells after cytomegalovirus infections (Navarro et al. 1993) and astrocytoma cells following measles virus infection (Duprex et al. 1999). It is important to note that viral induced cell fusions do not need a virus infection per se since the Mason Pfizer monkey virus envelope gene was enough to induce cell fusions. These fused cells failed to proliferate and divide, except in case of inactivated p53 (Duelli et al. 2005). The link between virus-carcinoma and also virus-carcinoma-multinucleated cells will be important for understanding the etiology of virus induced malignancy. One example of virus-carcinoma-multinucleated cells is the human papilloma viruses induced cervix carcinoma and associated giant cells (Therman et al. 1983). Many novel virus-carcinoma connections were found and will be found, e.g. the recent connection that over 80% of Merkel cell carcinomas were associated with a newly discovered polyoma virus (Feng et al. 2008), as well as the connection of virus induced carcinoma with multinucleated cells. A connection between cell–cell fusion and tumorigenesis was already proposed 100 years ago. Aichel (1911) suggested that fusion (in the sense of fusion of gametes) of two somatic cells produces benign tumors, whereas fusion of a resident somatic cell with a leukocyte could produce a malignant tumor. The difference between these fusions and the different outcomes according to Aichel was due to the correct or incorrect (re)distribution of chromosomes, where a fusion between a resident somatic cell and a leukocyte would result in aneuploidy and evolve into malignancy. He did not support that cells acquire attributes, but that attributes of single (fusing) cells led to malignant hybrids. Since then, a plethora of publications described higher ploidies and aneuploidy in malignant cells. The tumor progression is mostly associated with chromosome or gene amplification, demonstrating that extragenetic dosage is advantageous. This is possible due to chromosome non-disjunction, endoreduplication and cell–cell fusions (for review: Larizza and Schirrmacher 1984, Hanahan and Weinberg 2000, Margolis 2005, Ganem et al. 2007, Larsson et al. 2008).
17.1.2 Cell–Cell Fusions in Development, Differentiation and Viral-Induced Cell–cell fusions can be transient, as in the case of the sperm and egg heterokaryon, which lead to a fusion of nuclei (synkaryon) and subsequent cell divisions, or they can be permanent leading to syncytia and multinuclear cells (Fig. 17.1).
400
R. Strick et al.
These syncytia have essential functions in human somatic tissues, as in the case of placenta, muscle and bone formation. Generally membrane fusions are energetically unfavourable and are therefore mediated by specific proteins to overcome repulsive hydration forces and promote hydrophobic attractive forces (Helm et al. 1992). Many proteins mediating the molecular mechanisms involved in cell–cell fusion during development of different organisms are known as further discussed in section 2 (for review: Oren-Suissa and Podbilewicz 2007, Larsson et al. 2008). Membrane fusions can also be associated with other diseases, like giant cells during inflammations, entry of enveloped viruses as well as virus induced cell–cell fusions. Also a lack of normal developmental cell–cell fusions can be associated with incorrect ossification like osteopetrosis (Everts et al. 2009), minute myotubes leading to centronuclear myopathy (Wöckel et al. 1998) and significantly reduced expression levels of the HERV-W envelope gene, Syncytin-1, found associated with pregnancy disorders like PE/IUGR and HELLP/IUGR (Langbein et al. 2008). Several organisms are used as cell–cell fusion models, like Caenorhabditis elegans (C. elegans), where during embryonic and post-embryonic development nearly one third of all cells fuse and ultimately generate the syncytial hypodermis, vulva, excretory gland, male tail, anchor cell, uterus and pharynx (Gattegno et al. 2007). Several genes were identified as promoting and inhibiting cell–cell fusion. For example, the membrane protein epithelial fusion failure 1 (eff-1) was identified as essential for cell–cell fusions in C. elegans and may act as a viral fusogen (Mohler et al. 2002). Eff-1 mutants demonstrated that cell–cell fusion in C. elegans limited cell migration, restricted cell fate determination and was essential for organogenesis and morphogenesis. 17.1.2.1 Cytotrophoblasts-Syncytiotrophoblasts During early gestation two trophoblast populations exist: (1) extravillous trophoblasts invade the maternal decidua basalis in order to attach the placenta to the uterus and to adapt maternal arteries for blood supply and (2) villous trophoblasts which differentiate and fuse with each other into a multinuclear syncytiotrophoblast. After the short period of forming the syncytiothrophoblast, maintenance of this multinuclear state is achieved through fusion of villous cytotrophoblasts into the existing syncytiotrophoblast. Interestingly, experiments showed that the syncytiotrophoblast does not replicate and is relatively transcriptionally inactive (for review: Huppertz 2009). Although syncytiotrophoblast maintenance is ongoing through cell fusion with villous cytotrophoblasts, apoptotic material from the syncytiotrophoblast is deposited in so called syncytial knots and released into the maternal blood (Huppertz 2009). 17.1.2.2 Myoblasts-Myotubes Mononuclear myoblasts fuse with each other to form bi- or trinucleated nascent myotubes. Additional rounds of cell fusion between myoblasts and nascent myotubes result in the formation of large, mature myotubes with hundreds or
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
401
thousands of nuclei (Horsley and Pavlath 2004 and for review: Chen et al. 2007). Interestingly, muscle founder cells attract fusogenic myoblasts which migrate to each other, adhere over filopodia and fuse (Kim et al. 2007). 17.1.2.3 Osteoclasts Mononuclear pre-osteoclasts derived from haematopoietic cells, migrate to the bone surface and fuse to multinuclear osteoclasts which acquire after attachment and polarization bone resorbing (osteolytic) activities. Up to eight nuclei are normally in human osteoclasts, although osteoclasts of some diseases, like Paget disease, have over 100 nuclei (Roodman and Windle 2005). Many of the molecules involved in regulation of haematopoietic cells to osteoclasts (osteoclastogenesis) are known and all nuclei of osteoclasts are transcriptionally active (Bar-Shavit 2007). In addition to the migratory activity of osteoclasts on the bone, it was shown that osteoclasts were also able to migrate through cell layers (transmigration) (Saltel et al. 2006). 17.1.2.4 Unique Cell–Cell Fusions A unique situation is the vertebrate lens, where evidences suggest that epithelial fiber cells fuse during terminal differentiation. However, the fusion of the lens syncytium is only partial, where the cytoplasm of fused fiber cells remains partly portioned by a membrane (Shestopalov and Bassnett 2000). The function of the lens syncytium is physiological in order to combine metabolically active cells at the surface with quiescent cells at the centre of the lens and on the other hand physically by enhancing the transparency and correction of spherical aberrations (Shi et al. 2009). In male germ line development, the committed spermatid precursors fail to complete cytokinesis and form syncytia until the end of differentiation (Guo and Zheng 2004), but multinucleated spermatogonia due to true cell–cell fusions are found in elderly males (Miething 1995). 17.1.2.5 Experimental Stem Cell Fusions Stem cell fusions with somatic cells lead to pluripotent cells with stem cell character. Evidences for this statement were from cell fusions of mouse embryonic stem cells with adult mouse spleen cells, which resulted in hybrid cells with retained high pluripotency (Matveeva et al. 1998). Using cell hybridization of adult thymocytes with embryonic stem cells showed pluripotency in vivo and epigenetic changes similar to the stem cell (Tada et al. 2001). Similar fusion experiments were done with human embryonic stem cells and human fibroblasts which produced pluripotent hybrids with embryonic programming (Cowan et al. 2005). Considering the differentiation of fused stem cells Terada et al. (2002) demonstrated that mouse bone marrow cells spontaneously fused with embryonic stem cells in vitro and these cells differentiated to the phenotype of the recipient cells. In view of brain cells fusing with stem cells, Ying et al. (2002) detected cell fusions between mouse brain cells
402
R. Strick et al.
and pluripotent embryonic stem cells, which resulted in brain-embryonic stem cell hybrids with an altered phenotype, but with a full pluripotent character. In addition, introducing stem cells into traumatic rat brains led to cell–cell fusions of grafted stem cells with brain cells (Horváth et al. 2006). 17.1.2.6 Virus Induced Cell Fusions Well characterized are the fusion proteins of enveloped viruses, which are mostly sufficient to facilitate fusion in cell culture. Most viral fusion proteins are integral membrane glycoproteins which form trimers or tetramers (like influenza HAtrimers). Essential of viral fusion proteins are the fusion peptides or fusogens, which are hydrophobic amino acids within a membrane-anchored transmembrane subunit. Differentiated are class I fusogens, like HA of influenza virus with coiled-coil helical domains and class II fusogens, like the E-glycoprotein of the Tick Borne Encephalitis virus with a ß-barrel domain, which undergoes structural changes (Harris and Watkins 1965, Skehel and Wiley 1998, Lescar et al. 2001). Interestingly, like in viruses, fusion peptides were also found in cell–cell fusion proteins, like PH-30 of sperms (Blobel et al. 1992). The presence of multinucleated giant cells are even used today for differential diagnosis of chickenpox (varicella), herpes and herpes zoster from small pox, where small pox does not form multinucleated cells (Koranda 2004). In the case of HIV1, infected cells expressing the env of HIV-1 on their surface can initiate cell–cell fusion between infected and uninfected cells. Detection of syncytia seems to be symptomatic for the late HIV-1 pathogensis (Blaak et al. 2000) and are found esp. in the brain and adenoid (Budka 1986, Frankel et al. 1996). One other example of virus induced cell–cell fusions associated with cancer was the detection of virus-like particles and syncytia found in ascites of an HIV-1-positive patient with ovarian carcinoma (Rakowicz-Szulczynska et al. 1999). Lastly, a case report showed syncytia with virus-like inclusions in giant tumor of bone (De Chiara et al. 1998). 17.1.2.7 Bone Marrow Derived Cells (BMDC) Earlier experiments of mice with specific liver enzyme mutations transplanted with wild-type mice bone marrow cells resulted in wild-type liver cells due to cell–cell fusion and a change of the liver cell expression profile to those of bone marrow cells (Vassilopoulos et al. 2003). In addition, similar experiments showed that cell–cell fusions occurred between two diploid cells and diploid and tetraploid cells (Wang et al. 2003 and Fig. 17.2). However, as mentioned throughout the literature the phenotypic and functional changes of BMDC are the basis of a dispute: if these cells are changed due to differentiation based upon cell–cell fusion or solely transdifferentiation without fusion (for review: Pawelek 2005). BMDCs can be reprogrammed at new locations, by transdifferentiation and function as new cells (Krause et al. 2001); on the other hand BMDCs can fuse with cells and result in nuclei reprogramming where mRNA and pre-exisiting proteins in the cytoplasm directly change the outcome of cell fusion. Importantly, the explanation is probably not exclusively
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
403
Fig. 17.2 Different possible outcomes of normal diploid cells (2n) after cell–cell fusion. Shown are preserved multinucleation with and without cell cycle arrest and division, as well as later fusions of nuclei and subsequent divisions leading to aneuploidy or apoptosis
transdifferention or cell–cell fusion, the existence of one does not exclude the other (Kodama et al. 2003, Ianus et al. 2003). Transdifferentiation was shown by several researchers. For example, Tran et al. (2003) demonstrated that BMDCs could differentiate into buccal epithelial cells; Ianus et al. (2003), showed in mice that BMDCs switched to pancreatic endocrine cells and Kodama et al. (2003), showed that splenocytes could regenerate pancreatic cells in mice without fusion. Although initially, some reviews focused on transdifferentiation of BMDCs as examples of plasticity of adult stem cells, presently with more evidences in the literature, the tone now represents cell–cell fusion of BMDCs as the main cause for switches of cell lineages (Rodi´c et al. 2004). In 2003 and 2004 heterotypic cell–cell fusion between BMDCs and resident somatic cells, such as hepatocytes, cardiomyocytes and skeletal muscle cells was demonstrated in several studies (Alvarez-Dolado et al. 2003, Nygren et al. 2004). Specifically BMD myeloid cells were shown to fuse with muscle fibers (Camargo et al. 2003). One explanation, for the observed cell type switching could be due to specific cell types. As noted above, although BMDC transdifferentiation to buccal epithelial cells or pancreatic cells occurs; these cell types do not fuse during development. On the other hand, cells undergoing BMDC-cell fusions represented cell types with a capacity to fuse during development, like liver cells, cardiomyocytes and skeletal muscle cells. Interestingly, in mice cell fusions of cardiac and skeletal muscle cells (cardiomyocytes with myoblasts) were also demonstrated (Reinecke et al. 2004).
404
R. Strick et al.
BMD macrophages generate syncytia under physiological conditions and reside everywhere, like Kupffer cells in the liver and microglia in the brain. More specific myelomonocytic cells were found to spontaneously fuse with hepatocytes (Camargo et al. 2004). Further evidences for cell–cell fusions were recently shown by irradiation experiments, which induced frequent heterotypic cell fusions between myeloid and lymphoid cells and hepatocytes (Nygren et al. 2008). In addition, chronic inflammation led to increased cell fusions of BMDCs with Purkinje cells (binucleated cells) in the cerebellum (Johansson et al. 2008). The technique of cre-lox recombination and lethally irradiated mice grafted with BMDC from mice expressing specific detectable proteins (green fluorescent protein and -Galactosidase) demonstrated after cre-lox recombination that these mice expressed these marker proteins in the brain, heart and liver, but not in skeletal muscle, gut, kidney or lung (Alvarez-Dolado et al. 2003). These experiments showed that fused cells developed into normal liver cells and in the cerebellum where binucleated cells were detected but with morphologically different nuclei (Alvarez-Dolado et al. 2003). Especially, reports of BMDCs fusing with brain cells, like Purkinje cells are known (Weimann et al. 2003, Johansson et al. 2008). A recent report even showed that cell fusion of BMDCs contributed to pericytes after induced strokes in mice (Piquer-Gil et al. 2009). On the other hand, maybe cell–cell fusion also serves as homeostasis in special organs like brain, heart and liver, like clearing of these cells and replenishing them with BMDC. Brain and liver cells also have a very slow turnover and could make them targets for cell fusions.
17.1.3 Cell–Cell Fusions During Tumorigenesis Numerous publications have described fusions of tumor cells with other tumor cells or with somatic cells. Unfortunately, many times a precise description of these so called fused cells was not given, so that the definition of true cell–cell fusion or endomitosis cannot be given, e.g. patient case reports. Additionally, the status of multinucleation or fused nuclei (karyogamy) could not be given either. When cell characterizations were stated they were given as examples of known cell–cell fusions like osteoclast-like or trophoblast-like cells. Experiments involving grafting human cancer cells into rodents produced tumors that partially or completely consisted of host-human hybrids and host tumor cells that appeared to be derived by spontaneous fusion (Goldenberg et al. 1974, Pathak et al. 1997). The role of fused cells explaining metastasis, esp. of BMDC or tumor associated macrophages and cancer cells was supported by spontaneous fusion of tumor cells with normal cells in mice followed by chromosome segregation resulting in tumor progression and metastasis (Kerbel et al. 1983, Larizza et al. 1984, Busund et al. 2003). Table 17.1 describes a variety of tumors where cell–cell fusions have been described. Below are some examples of cell–cell fusions in different tumors expanded on from the literature. Breast cancer: Cell–cell fusions between breast cancer cells were described by Miller et al. (1988). Further it was shown that human breast cancer cell lines can
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
405
Table 17.1 Selected examples of multinucleated cells found in human tumors Tumors
Multinucleated cells described as
References
Breast carcinoma Breast carcinoma Breast carcinoma Breast carcinoma (DCIS) Ovary
Osteoclast-like Stromal-like Osteoclast-like Giant Giant cell
Uterus/leiomyosarcoma
Osteoclast-like; fusion of spindle-cells Syncytiotrophoblast-like Giant, trophoblast-like
Athanasou et al. (1989) Factor et al. (1977) Krishnan and Longacre (2006) Coyne (2007) Veliath et al. (1975), Yasunaga et al. (2008) Watanabe et al. (1996)
Uterus Uterus Germ cell tumors Seminoma Testicular teratoma Liver Liver Gallbladder Lung Lung Pancreas
Syncytiotrophoblast-like Syncytiotrophoblast-like, giant
Bone Soft tissue (PEComa) Melanoma Thyroid
Epithelial syncytial giant Osteoclast-like Osteoclast-like Giant Osteoclast-like Giant, pleomorphic, osteoclast-like Osteoclast-like Giant Monster cells, pleomorphic Giant
Hodgkin lymphoma Glioblastoma
Endomitotic Giant
Strick et al. (2007) Jones et al. (1991), Pesce et al. (1991) Ulbright (2005) Miettinen et al. (1985) Ulbright (2005) Atra et al. (2007) Munoz et al. (1980) Akatsu et al. (2006) Laforga (1999) Nakahashi (1987) Deckard-Janatpour et al. (1998); Robinson et al. (1977) Werner (2006) Folpe et al. (2005) Boyd et al. (2005) Albores-Saavedra and Wu (2006) Drexler et al. (1989) Homma et al. (2006)
fuse with endothelial cells in vitro and in vivo (mice) (Mortensen et al. 2004) and fusion could be inhibited with an inhibitory peptide against the HERV-W envelope protein, Syncytin-1 (Bjerregaard et al. 2006). Cell–cell fusion of breast cancer cell lines expressing Syncytin-1 was repressed with antisense RNA against Syncytin-1 (Bjerregaard et al. 2006). Syncytin-1 was found in 38% of breast tumors by immunohistochemistry and showed a positive prognostic factor for breast cancer (Larsson et al. 2007). The question of bone metastasis and osteolytic activity of breast cancer could be answered with the findings that tumor associated macrophages from breast cancer patients were capable of differentiation into multinucleated osteoclasts in the presence of breast cancer fibroblasts and tumor cells (Lau et al. 2007). Similar results were found with metastatic melanomas (Lau et al. 2006). Grafting human cells from pleural effusion from a patient with breast cancer into mice resulted in cell hybrids with human and mice chromosomes due to spontaneous cell fusion. These human cancer-mouse stromal cells were tumorigenic (Jacobsen et al. 2006). Recent results showed that bone metastatic breast cancer
406
R. Strick et al.
cells induce migration of multinucleated osteoclasts in vitro (Saltel et al. 2006), indicating the stimulative activity of cancer cells. Germ cell tumors: Germ cell tumors are a very heterogeneous class and have for male and female very different incidences. For example, 95% of ovarian germ cell tumors are teratomas, but only 4% of testicular germ cell tumors. On the other hand seminomas count for 50% of testicular germ cell tumors and dysgerminoma for 2% of ovarian germ cell tumors (for review: Ulbright 2005). Choriocarcinomas are one of the rarest germ cell tumors. Seminomas are found with multinucleated cells (Miettinen et al. 1985) as well as testicular teratomas (Ulbright 2005). Giant neoplastic cells were often found in pure seminomas, but giant mononuclear cells less frequent in spermatocytic seminomas (von Hochstetter et al. 1985). Some germinomas and esp. choriocarcinomas were found with distinct syncytiotrophoblast cells (Mostofi and Sesterhenn 1985). Especially, choriocarcinoma cell lines are widely used as ideal test cells for cell–cell fusion studies. Fusion of teratocarcinoma cells with adult mice thymocytes produced different tissue types and carcinoma in mice indicating pluripotency (Miller and Ruddle 1976). Melanoma: Tumor cells with giant nuclei, so called monster cells, were previously described in basal cell carcinomas, dermatofibromas, pleomorphic fibromas and cutaneous melanomas. In the latter multinucleated cells were also detected and statistical association with ulceration and depth of invasion found (Boyd et al. 2005). Glioblastoma: One example of the difficulty to find the cause for multinucleation of giant cells is shown for giant cell glioblastoma: Giant cell glioblastomas are rare among glioblastomas, and are characterized by many multinucleated giant cells. A molecular analysis of the cell cycle kinase cdc2 and Rho-kinases showed inactivity in the giant cell glioblastoma with the conclusion that these cells entered into early mitosis but not through late mitosis demonstrating nuclear division but no cytokinesis and therefore endomitosis (Maeda et al. 2003). In another report, giant cell glioblastomas were found with little mitotic potential and p53 mutations concluding some cell fusions (Takeuchi et al. 2006). Finally, the existence of CD98 positive giant cell glioblastomas with low proliferative potential speaks for the development of these cells by cell fusion and syncytial formation (Takeuchi et al. 2008). Endometrial carcinoma: In contrast to endometroid endometrial carcinomas, which count for over 80% of the total endometrial carcinomas, non-endometrioid endometrial carcinomas are not derived from benign lesions. Non-endometrioid endometrial carcinomas show a high array of different histological subtypes, like serous, mucinous, clear cell, squamous and poorly differentiated tumors with trophoblastic elements and giant cells (for review: Clement and Young 2004). These trophoblastic and giant cell carcinomas have areas of multinucleated cells, which are described as syncytiotrophoblast-like and are considered highly aggressive (Jones et al. 1991, Pesce et al. 1991). However, multinucleated cells were also identified in primary endometrioid endometrial carcinomas (Strick et al. 2007). Lung carcinoma: Large cell lung carcinomas, which count for 10–16% of lung cancers, also include so called giant cell carcinomas (Yesner 1985, Laforga 1999).
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
407
Interestingly, giant cell carcinomas of the lung are specified as the most malignant lung carcinoma and include multinucleated and spindle form cells, which have phagocytotic activities (Razzuk et al. 1976). Hodgkin lymphoma: Hodgkin lymphoma shows several interesting findings in the course of the disease (for review: Küppers 2009). One phenomenon is the finding of multinucleated Hodgkin-Reed-Sternberg cells, which are derived from mononuclear Hodgkin cells probably by endomitosis (Jin and Woodgett 2005). Earlier experiments with a cell line from Hodgkin disease showed spontaneous formations of multinucleated cells, which had the same surface antigens as the monocleated cells and were still proliferative with mitosis, however without telophase (Drexler et al. 1989). BMDC fusions in cancer: Several examples are shown that BMDC-resident somatic cell fusions also play a role in multiple tumorigenesis (Liu et al. 2006). For example, transplanted BMDCs from mice showed cell fusion with normal and transformed intestinal cells, esp. with intestinal progenitor or stem cells, demonstrating a role of BMDCs in intestinal cell regeneration and/or tumorigenesis (Rizvi et al. 2006). Induction of inflammation and epithelial proliferation in mice induced cell fusions of BMDC with intestinal cells, proposing that increased inflammation and proliferation could lead to cell fusions and intestinal diseases and cancer (Davies et al. 2009). On the other hand, a connection between bone marrow stem cells and gastric cancer was shown, however the mechanism was unclear (Houghton et al. 2004). The latter example explained that the inflammatory environment probably favored the development of cancer been linked to homing and engraftment in peripheral tissue by BMDCs (e.g. Helicobacter). Human cancer cell fusions following transplantations: Only a few reports of cancer cell–cell fusions in humans are documented: Fusion of myeloma cells and osteoclasts (Andersen et al. 2007) and cell–cell fusions in two cancer patients occurring after organ transplantations (Chakraborty et al. 2004, Yilmaz et al. 2005). Additional evidence of solid cancers after allogenic bone marrow transplantation demonstrated the presence of donor-derived malignant cells (Avital et al. 2007). Signal transduction pathways and cell fusions: Several genes were identified as promoting or inhibiting cell–cell fusion, which were classified as transcription factors regulated by e.g. the Ras- and Wnt-pathways but also developmental genes, like the Homeo box-genes (for review: Shemer and Podbilewicz 2003). To date other signalling pathways have been implicated in the regulation of cell–cell fusion, for example protein kinase B or AKT2. Interestingly, AKT2 over expression has been found in many cancer types (Cheng et al. 1996). Introducing and activating a conditional AKT2 gene in human epithelial kidney cells resulted in multinucleated cells, caused by both endomitosis and cell–cell fusion (Jin and Woodgett 2005). Over expressing either the dual-specificity-tyrosine-phosphorylation-regulated kinase 1A (MNB/DYRK1A) or AURORA-A in Hela-cells resulted in multinucleated cells (Funakoshi et al. 2003, Meraldi et al. 2002). Interestingly, AURORA-A elevated protein levels is often found in cancers (Zhou et al. 1998), esp. in breast cancer (Miyoshi et al. 2001).
408
R. Strick et al.
17.2 Human Endogenous Retroviruses (HERVs) HERVs are derived from exogenous retroviral infected germ cells which integrated into the genome. Calculations showed that infection of the germline by the HERV precursors could have been more than 40 million years ago like for HERV-FRD or less than 200,000 years ago, like for HERV-K113 (Blaise et al. 2003, Turner et al. 2001). More than 8% of the human genome is considered retroviral origin, like retrotransposons, HERVs and elements with HERV origin (Bannert and Kurth 2004). More than 200 families of LTR-containing retroelements are defined by Repbase (Jurka 2000). HERVs are grouped into class I, II or III according to the sequence similarity to exogenous retroviruses: Class I HERVs, like HERV-W, -H, -R, -P, -T and -FRD are related to gammaretroviruses such as the murine leukaemia virus; class II HERVs, like HERV-K are related to betaretroviruses such as the mouse mammary tumor virus and class III HERVs, like HERV-L are related to spumaretroviruses such as the human foamy virus (for review: Nelson et al. 2003 and Tristem 2000). HERVs like exogenous retroviruses code for at least three genes: gag (group-specific antigen) encoding the structural protein of the core, pol (polymerase) encoding the viral enzymes like reverse transcriptase and env for the viral envelope. Many HERV sequences integrated through evolution adjacent to gene promoters and regulate transcription of cellular genes with functional consequences for normal tissues and malignancies (for review: Bannert and Kurth 2004). In contrast to pure retrotransposons, HERVs contain the env gene, coding for a viral membrane protein. HERV families are further distinguished due to the t-RNA codon for amino acids used for initiation of transcription. As an example, the primer binding site for a specific tRNA used to initiate reverse transcription is the tRNA(Trp) for HERV-W (Tristem 2000). Like exogenous retroviruses, HERVs consist of a provirus with the following assembly: LTR-gag-pol-env-LTR. Most of the HERV genes are nonfunctional due to recombinations, mutations and deletions; however 17 gag, 13 pol and 18 (29) env from all different HERVs encoded full-length open reading frames and could produce functional proteins (de Parseval et al. 2003, Villesen et al. 2004, Blaise et al. 2005). A few families, esp. the HERV-K family have been shown to form viral but non-infectious particles (Löwer et al. 1984, Bieda et al. 2001, Büscher et al. 2005). However, using the env protein from HERV-K108 infectivity to simian immunodeficiency virus pseudotypes was shown, demonstrating a full functional env gene (Dewannieux et al. 2005). Especially env genes of different HERVs were examined for expression. The env gene of HERV-W, called Syncytin-1 was found expressed in placentas and plays an essential role in cell–cell fusion of cytotrophoblasts to multinuclear syncytiotrophoblasts (Mi et al. 2000, Blond et al. 1999). So far, three env genes from different HERVs were found fusogenic in vitro assays: Syncytin-1, the env of HERV-FRD, also called Syncytin-2 and the env of HERVP(b). The regulation of Syncytin-1 for cell specific expression was found to be dependent on the methylation in the U3 part of the 5’LTR (although no CpG-islands were found) (Matousková et al. 2006). Syncytin-1 on chromosome 7q21.2 belongs to the class I fusion genes like other exogenous virus (paramyxovirus, influenza,
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
409
filovirus, ebola virus, HTLV, lentivirus, HIV-1) and is cellularly processed and proteolytically cleaved into a surface unit (SU) and transmembrane unit (TM). The 538 amino acid long Syncytin-1 is glycosylated and makes homotrimers (Cheynet et al. 2005). Syncytin-1 and Syncytin-2 were found up-regulated after cAMP-stimulation, either with Forskolin, which induces the adenylate cyclase or directly with e.g. SPcAMP indicating a protein kinase A regulation (Prudhomme et al. 2004, Strick et al. 2007, Chen et al. 2008). Furthermore, when Syncytin-1 was intrinsically or transiently activated by cAMP cell–cell fusion increased. Multiple cells and cell lines (human and non-human) demonstrated the fusogenicity of Syncytin-1 and on the other hand RNAi and inhibitory antibody experiments identified Syncytin-1 as the cause for cell–cell fusion in vitro. In addition, the receptors for Syncytin-1 were identified as ASCT-1 and -2, which bind to the N-terminal SU-domain of Syncytin-1 (Cheynet et al. 2006). Additionally, the C-terminal, called the cytoplasmatic region was identified as negatively regulative for cell–cell fusion (Chang et al. 2004, Drewlo et al. 2006). Analyses of genomic polymorphisms of Syncytin-1 and Syncytin-2 identified multiple changes and even non-synonymous changes, but none of them impaired cell–cell fusion capabilities in vitro, leading to the assumption that Syncytin-1 and -2 have an essential role for humans, especially in human reproduction (Mallet et al. 2004, de Parseval et al. 2005). In addition, Syncytin-1 can function as a normal env protein and also with other retroviruses, pseudotyping HIV-1 (An et al. 2001, Lavillette et al. 2002). Although, Syncytin-1 expression was proven essential for mediating cytotrophoblast fusion to the syncytiotrophoblast in placenta it was also detected in other normal and tumorigenic tissues and cell lines, however HERV-W gag and pol expression was only detected in selected tissues and cancer cells (Yi et al. 2004 and Table 17.2). In addition to cell–cell function, many HERV env proteins also contribute to immunomodulatory actions, like autoimmunity caused by superantigens or immunosuppression (for review: Balada et al. 2009). Several neuropathological diseases, like schizophrenia, rheumatoid arthritis and multiple sclerosis have been attributed to the role of Syncytin-1 and another env gene of the HERV-W family member called the multiple sclerosis retrovirus (MSRV) (Antony et al. 2007, Antony et al. 2004, Karlsson et al. 2001). The other fusogenic env protein is Syncytin-2 of HERV-FRD on chromosome 6p24.1, with MFSD2, a putative carbohydrate transporter as the receptor (Esnault et al. 2008, Blaise et al. 2003). In contrast to Syncytin-1, Syncytin-2 expression is more restricted to placental cells like villous cytotrophoblasts (Malassiné et al. 2007). The third env protein, which was shown to be fusogenic in an ex vivo assay was HERV-P(b) on chromosome 14q32.12. To date, not much is known about the expression of this env in cancer, but it was shown to be widely expressed in human tissues (Blaise et al. 2005). Interestingly, not much is known about the regulation of HERV expression. However, one early regulation shown for HERV was steroid hormones. The production of HERV-K virus particles and provirus in the breast cancer cell line T47D was estradiol and progesterone-dependent (Keydar et al. 1984, Ono et al. 1987). T47D cells also induced HERV-K gag, pol and env expression after estradiol and progesterone stimulation, and may be important for breast carcinogenesis. Golan
410
R. Strick et al. Table 17.2 Selection of HERV expression found in different human cancers
HERV
Gene
Method
Tissue or cells
References
HERV-W HERV-W
Syncytin-1 Syncytin-1
PCR RT-PCR
Yi et al. (2001) Yi et al. (2004)
HERV-W
gag
RT-PCR
HERV-W
pol
RT-PCR
HERV-W
Provirus
EST mapping
Ovar carcinoma cell lines All tissues and cancer cell lines Brain, testis, placenta, spleen All tissues, except heart and uterus Choriocarcinoma
HERV-W
Syncytin-1
Breast cancer and cells
HERV-W
Syncytin-1
HERV-W
Syncytin-1
Immunoblot + IHC Immunoblot + RT-PCR IHC
HERV-W
pol
Microarray, PCR
Control and breast cancer
HERV-W
Syncytin-1
IHC
Colo-rectal carcinoma
HERV-S HERV-T
pol pol
RT-PCR PCR
HERV-R
ERV-3, env
Northern
HERV-R
ERV-3, env
HERV-R
ERV-3, env
cDNA hybridization Northern
Cancer cell lines Carcinoma, cancer cell lines Breast cancer, glioma, rhabdo-myosarcoma and cell lines Osteosarcoma cell line
HERV-R
ERV-3, env
HERV-R
ERV-3, env
HERV-R/ ERV9 HERV-R
pol
RT-PCR, northern Tissue microarray Microarray, PCR
env
RT-PCR
HERV-P HERV-P
gag, pol, env env
RT-PCR RT-PCR
HERV-K
Virus particles
EM
HERV-K
Virus particles
Centrifugations, RIA
HERV-K
ERV-3, env
HERV-K
gag, virus particles
cDNA hybridization IF, EM
Endometrial carcinoma + cell lines Breast cancer
Yi et al. (2004) Yi et al. (2004) Stauffer et al. (2004) Bjerregard et al. (2006) Strick et al. (2007) Larsson et al. (2007) Frank et al. (2008) Larsen et al. (2009) Yi et al. (2004) Yi and Kim (2007) Cohen et al. (1988)
Leib-Mösch et al. (1990) Lung cancer Andersson et al. (1998) Prostate cancer Wang-Johanning et al. (2003) Ovarian cancer Wang-Johanning et al. (2007) Control and breast cancer Frank et al. (2008) Liver+ lung tumor Ahn and Kim (2009) Cancer cell lines Yi et al. (2007) Colon and liver tumors Ahn and Kim (2009) Teratocarcinoma cell line Bronson et al. (1979) Breast cancer line T47D Keydar et al. (1984), Ono et al. (1987) Breast/gastric/melanoma Leib-Mösch cancer cell lines et al. (1990) Teratocarcinoma cell lines Boller et al. (1993)
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
411
Table 17.2 (continued) HERV
Gene
Method
Tissue or cells
References
HERV-K
gag, env
Germ cell tumors, choriocarcinomas
Herbst et al. (1996)
HERV-K
Virus particles
In situ hybridization, immunoblot EM
Teratocarcinoma cell lines
HERV-K
env
RT-PCR
Breast cancer
HERV-K
gag
RT-PCR, northern Leukaemia
HERV-K
Np9, gag
RT-PCR, immunoblot
Bieda et al. (2001) Wang-Johanning et al. (2001)+ (2003) Depil et al. (2002) Armbruester et al. (2002)
HERV-K
Provirus
EST mapping
HERV-K HERV-K
Virus particles, provirus env
RT-PCR, immunoblot, IF, Tissue microarray
HERV-K
mel
PCR
HERV-K
RT
IHC
HERV-K
pol
microarray, PCR
HERV-K
RT
HERV-K
PCR, Northern, IHC gag, env, PCR, immunoblot, virus particles EM
HERV-K
env
Tissue array
HERV-K
Virus particles
EM and RNA
HERV-K
env
RT-PCR
HERV-H
env
RT-PCR
HERV-H
Provirus
EST mapping
HERV-H
env
RT-PCR
HERV-H
gag
Signal sequencing
HERV-H
env
RT-PCR
Germ cell and mammary tumors, leukaemia and cell lines Cancer of testis, uterus, Stauffer et al. stomach, brain, skin and (2004) normal brain and skin Melanoma Büscher et al. (2005) Ovarian cancer Wang-Johanning et al. (2007) Pancreatic cancer SchmitzWinnenthal et al. (2007) Breast cancer and normal Golan et al. breast (2008) Normal and breast cancer Frank et al. (2008) Normal and breast tumors Golan et al. (2008) ContrerasPlasma of patients with Galindo et al. lymphoma and breast (2008) cancer Breast cancer Wang-Johanning et al (2008) Melanoma cell lines Serafino et al. (2009) Testicular cancer Ahn and Kim (2009) T-cell leukaemia cell lines Lindeskog and Blomberg (1997) Cancer of prostate, colon, Stauffer et al. testes, bone marrow, (2004) intestine, bladder, cervix Human tissues and cancer Yi et al. (2006) cell lines Colon carcinoma Alves et al. (2008) Cancer of liver, lung, testes Ahn and Kim (2009)
412
R. Strick et al. Table 17.2 (continued)
HERV
Gene
Method
Tissue or cells
HERVFRD HERV-F
pol
Microarray, PCR
Control and breast cancer
HERV-F
pol
HERV-E
env
HERV-E
env
HERV-E
provirus
HERV-E
env
HERV-E
env
HERV-E
pol
HERVE/ER
gag, pol, env
pol
References
Frank et al. (2008) RT-PCR Cancer cell lines Yi and Kim (2004) Microarray, PCR Control and breast cancer Frank et al. (2008) immunoblot Endothelial, colon and Turbeville et al. prostate carcinoma (1997) RT-PCR, northern Prostate cancer Wang-Johanning et al. (2003) EST mapping Control and breast cancer Stauffer et al. (2004) Tissue microarray Ovarian cancer Wang-Johanning et al. (2007) RT-PCR Most tissues and all cancer Yi and Kim cell lines (2007) Microarray, PCR Control and breast cancer Frank et al. (2008) RT-PCR Leukaemia cell lines Prusty et al. (2008)
IHC = immunohistochemistry, IF = immunofluorescence, EM = electron microscopy, RIA = radioimmuno assay
et al. (2008) found that HERV-K RT was more induced in breast cancer biopsies than normal tissue and that HERV-K RT positive tumors showed both a shorter overall and metastasis-free survival. In addition to HERV-K, Syncytin-1 was also found steroid hormone inducible in endometrioid endometrial carcinoma cell lines due to an estrogen response element in the LTR of HERV-W, which bound specifically to estrogen receptor alpha (Strick et al. 2007). Endometrioid endometrial carcinoma represents a steroid hormone dependent tumor and the regulation between steroid hormone proliferation and cell–cell fusion was found to be TGF-β1/β3 dependent (Strick et al. 2007).
17.2.1 HERV Expression in Human Cancers Next to the detection of the fusogenic HERVs [HERV-W, -FRD and P(b)] in human cancers and cancer cell lines, multiple non-fusogenic HERVs were identified in different human cancers (Table 17.2). Especially, gag-pol- and env-gene expression from four HERV families was found associated with different cancers and cancer cell lines: HERV-E, HERV-H, HERV-R and esp. HERV-K (Table 17.2). In particular, not only gene expression of HERV-K families were found associated with different human cancers (cell lines), but also HERV-K virus particles were found. For example, HERV-K gene expression and particles were found associated with advanced
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
413
state melanoma and related with the switch from adherent to non-adherent cell growth. RNAi-experiments specific for HERV-K prevented this transition (Serafino et al. 2009). To date, infectious HERV-K has never been detected even after co-cultivation with target cells. In addition, several reports showed that human patients exhibited serological response to HERVs, like HERV-K gag and env in melanoma, prostate and testicular cancer (Hahn et al. 2008, Ishida et al. 2008, Goedert et al. 1999). The fusogenic Syncytin-1 of HERV-W was found in breast, endometrial and colon cancer. Interestingly, in breast cancer Syncytin-1 was expressed in 38% of cases and correlated with a positive prognostic role, whereas in a retrospective analysis of colorectal cancers over-expressed Syncytin-1 was associated with a decreased overall survival of rectal but not of colon carcinomas (Larsson et al. 2007, Larsen et al. 2009). Taken together strong in vitro evidences point to a direct involvement of single HERV genes in cancer.
17.2.2 HERVs in Cancer Cell–Cell Fusions: Driver or Passenger HERV gene expression and over expression are detected in primary human tumors, but evidence to prove a direct connection of HERV env genes with cell–cell fusions is ongoing (Table 17.2). So far, as stated before only three HERV env genes, Syncytin-1 (HERV-W), Syncytin-2 (HERV-FRD) and the env of HERV-P(b) were shown to be involved in cell–cell fusions in vitro (Blaise et al. 2003, Blaise et al. 2005, Strick et al. 2007). The main candidate of fusogenic env gene involved in cancer cell fusions is Syncytin-1 in choriocarcinoma, breast cancer and endometrial carcinomas. Several choriocarcinoma cell lines showed Syncytin-1 and Syncytin-2 and their receptors ASCT-1/-2 and MFSD2 expression and some of them cell–cell fusions (Kudo and Boyd 2002, Borges et al. 2003, Esnault et al. 2008) For breast cancer cell lines expressing Syncytin-1 cell fusions occurred between the cancer cells but also with endothelial cells. These cell–cell fusions were inhibited by antisense RNA as well as with an inhibitory peptide against Syncytin-1 (Bjerregaard et al. 2006). Expression of Syncytin-1 and the receptor ASCT-1 in breast cancer tumors and endothelial cells were also detected proposing that cell–cell fusion of breast cancer cells and endothelial cells could also occur in patients. This is similar to nude mouse endothelial cells fusing with human breast cancer cell lines (Mortensen et al. 2004, Larsson et al. 2007). In another human tumor, endometrioid endometrial carcinoma, Syncytin-1 and ASCT-1 and -2 receptor expressions were found significantly higher than in control endometrium. Cell–cell fusion experiments with endometrial carcinoma cell lines were demonstrated as dependent upon Syncytin-1 expression using RNAi experiments. Multinucleated cells in endometrial carcinomas were found in paraffin sections proposing that Syncytin-1 expression could be connected with cell fusions in this carcinoma (Strick et al. 2007). Similar experiments were done in ovarian carcinomas and a direct connection of Syncytin-1 with cell–cell fusions was found in vitro (Strick et al. in preparation). Similar to e.g. Mason Pfizer virus env gene, which was enough to induce cell fusions (Duelli et al.
414
R. Strick et al.
2005) many experiments with Syncytin-1 demonstrated the sufficiency of the env (or even parts of env) to mediate cell fusions as well as C-terminal truncations. For example, truncations up to 55 or 58 C-terminal amino acids of the 538 amino acid Syncytin-1 increased fusogenicity in human cancer cells (Drewlo et al. 2006, Chang et al. 2004), but in contrast C-terminal truncations of 67 or 82 amino acids inhibited cell fusions (Cheynet et al. 2005, Chang et al. 2004).
17.3 Cell–Cell Fusions in Cancer: Functional Role or Dead-End Polyploidy and multinucleation are hallmarks of many types of human malignancies, leading to genetic instability and aneuploidy and development of cancer (Hanahan and Weinberg 2000). In addition, highly malignant cells generally show higher ploidy levels than their less malignant progenitor tumor cells (review: Larizza and Schirrmacher 1984). Cancer cells may fuse spontaneously with several types of resident somatic cells, esp. with BMDC and these hybrids are proposed to play an important role in metastasis (Pawelek and Chakraborty 2008, Liu et al. 2006, Jacobsen et al. 2006, Vignery 2005). Several consequences of fused cancer cells in tumorigenesis are possible: (1) Cancer cell fusions result in karyogamy, cellular divisions and new tumorigenic functions; (2) Cancer cell fusions stay multinuclear and progress into cell cycle arrest or divide; or (3) the cancer cell hybrids apoptose (Fig. 17.3). In favour for the first possibility are the examples of detected tetraploid cells in tumors and their functional role in the progress of tumorigenesis (Ganem et al. 2007, Margolis 2005). On the other hand, many examples of tumors are known of the existence of multinuclear cells in tumors, however if they are permanent or transient is not known (Table 17.1). Then again many evidences are in favour of the third possibility, but with modifications. According to some references are cell fusions due to intact viruses doomed to either apoptose or cell cycle arrest (Zhivotovsky and Kroemer 2004). Another possibility is the phenomenon of mitotic catastrophy, which was defined as a type of cell death occurring during mitosis. Besides e.g. microtubule poisons and over-duplication of centrosomes, mitotic catastrophy can also be induced by fusion of mitotic cells with interphase cells, but also be the destiny of common multinucleated cells, which undergo karyogamy and metaphase and result into apoptosis or division with an asymmetric cell division (Castedo et al. 2004 and Fig. 17.3). One checkpoint regulator for a direct apoptosis after karyogamy of multinucleated cells was Chk2 (Castedo et al. 2004). Although many sources speak for apoptosis or cell cycle halt of fused cancer cells, the existence of multinucleated cells in tumors, like giant cell glioblastoma etc. are established and not just a rare accident and play even a role for prognosis. The fate of cancer cell-somatic cell hybrids could depend upon the existing cellular RNA and proteins, which are brought into the hybrid and could result to: (1) a gain of functional tumor suppressor genes with the consequence of reduced malignancy as shown with fusions after somatic cell hybridizations or inactivated viruses (Harris et al. 1969, Harris 1988, Stanbridge 1976, Wiener et al. 1974a), or (2) a loss of tumor suppressors (for review: Anderson and Stanbridge 1993). It was shown that malignant and normal human cell fusions resulted into a suppression of tumor
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
415
Fig. 17.3 Possible consequences of cell–cell fusions between normal somatic (2n) or normal stem cells (2n) with tumor cells. Two main outcomes are distinguished: with and without fusion of nuclei (karyogamy versus multinucleation). Different reports showed that karyogamy after cell– cell fusion and repeated cell divisions resulted in loss or gain of specific chromosomal genes (aneuploidy). Several reports also proposed the possibility of novel so called cancer stem cells after cell–cell fusions and karyogamy, which can be self renewing or lead to new tumor cells. This scenario could also be conceivable with multinucleated cells
formation, whereas cancer cell–cell fusions kept their malignancy (Stanbridge 1976, Wiener et al. 1974b). On the other hand cell–cell fusions between cancer cells and BMDCs, esp. macrophages have an important impact on tumorigenesis, like metastasis and proliferation. For example, the detection of rectal cancer cells expressing macrophage specific proteins (CD163) probably due to cancer cell fusions with macrophages had very high metastatic activities (Shabo et al. 2009), or cancer-macrophage fusion hybrids caused higher proliferation and vascularisation of tumors (Busund et al. 2002). Important is the possibility of even partial reprogramming of the somatic cell or even of the cancer cell after fusion. This was found esp. for BMDC-somatic cell hybrids after cell fusions, because tissue-specific genes normally not found in haematopoietic cells were found in skeletal muscle cells (Camargo et al. 2003, Gussoni et al. 1999, Ferrari et al. 1998), brain cells (Weimann et al. 2003) and liver cells (Vassilopoulos et al. 2003, Lagasse et al. 2000); probably due to reprogramming. This kind of reprogramming after cell–cell fusion could be particularly important if one partner is a BMDC or a stem cell, cells which are highly plastic, and the other partner is a resident somatic cell. A cell–cell fusion could then generate new hybrid pluripotent cells, which could have the potential to differentiate tissue or regenerate damaged tissue. Conversely, if this reprogramming is deregulated, this would provide a mechanism for cancer development.
416
R. Strick et al.
A deregulation of reprogramming could lead to an up-regulation of genes involved in invasion and metastasis like fusion of migratory bone marrow cells and cancer cells (for review: Pawelek and Chakraborty 2008). On the other hand the new fused hybrid cells could also be capable of being a cancer stem cell as suggested by Bjerkvig et al. (2005). This is in particular interesting for the tumor development and metastasis. Cancer stem cells were already demonstrated in leukaemia, breast cancer (Al-Hajj and Clarke 2004) and brain cancer (Singh et al. 2004, Galli et al. 2004). More evidences speak also for a functional role of cell fusions with regard to tumor metastasis, esp. the sustainment of the different tropisms of the single cells after fusion. It was shown that spontaneous cell fusion between bone-tropic and lung-tropic cancer cells generated stable hybrids with dual metastasis tropism to both organs without nuclear reprogramming in vitro and in vivo (Lu and Kang 2009). In addition to the acquisition of metastasis properties after cell fusion, the latter report also demonstrated a high level of chromosomal and phenotypic stability in hybrids during long-term passage in vitro and in vivo. These results demonstrated that the cancer cell hybrids maintained the genomic, transcriptomic and phenotypic characteristics of the original mononuclear cells. Chromosome stability after cell fusion was also demonstrated with other cell systems, e.g. fusion beween HeLa and fibroblasts (Stanbridge et al. 1982), between mouse mammary tumor cells (Miller et al. 1988) or between embryonal stem cells and fibroblasts (Cowan et al. 2005). Considering the genomic maintenance, but also the equality of different cells, cell fusion of human and mouse cells demonstrated expression of both parental partners in the hybrid (Chakraborty et al. 2001). Therefore, cell fusion of cancer cells with different metastatic abilities or fusion of cancer cells with somatic cells could increase the metastatic spreading. Significant for the treatment of cancer could also be the treatment resistance of cancer cell hybrids, because survival in other organs could be possible due to new tropisms and outright acquired chemoresistance after cell fusion. Many fundamental cell–cell fusion questions about cell–cell fusions are not completely answered up to date, for example: • How are cell–cell fusions regulated? • Which specific proteins (ligands-receptors) drive the cell–cell fusion process and which roles do HERVs play? • Do fused cells have transcriptional and translational changes compared to the same but non-fused cells? • What determines the dominance of one nucleus after cell–cell fusion and why are in some cases nuclei equal after cell–cell fusions? • What role does the existing RNA and proteins of each fusing cytoplasma play? • What factors prevent and support karyogamy after cell–cell fusion? • Which multinucleated cells are in cell cycle arrest (with or without transcriptionally active) or divide? • The unravelling of the phenomenon of cell–cell fusions, in normal human development and in carcinogenesis will help to explain the aetiology of some cancers, metastasis, relapse, and possibly treatment resistances.
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
417
References Ahn K, Kim HS (2009) Structural and quantitative expression analyses of HERV gene family in human tissues. Mol Cells 28:99–103 Aichel O (1911) Über Zellverschmelzung mit qualitativ abnormer Chromosomenverteilung als Ursache der Geschwulstbildung. Wilhelm Engelmann, Leipzig Akatsu T, Kameyama K, Kawachi S et al (2006) Gallbladder carcinoma with osteoclast-like giant cells. J Gastroenterol 41:83–87 Albores-Saavedra J, Wu J (2006) The many faces and mimics of papillary thyroid carcinoma. Endocr Pathol 17:1–18 Al-Hajj M, Clarke MF (2004) Self-renewal and solid tumor stem cells. Oncogene 23:7274–7782 Alvarez-Dolado M, Pardal R, Garcia-Verdugo JM et al (2003) Fusion of bone marrow-derived cells with Purkinje neurons, cardiomyocytes and hepatocytes. Nature 425:968–973 Alves PM, Lévy N, Stevenson BJ et al (2008) Identification of tumor-associated antigens by largescale analysis of genes expressed in human colorectal cancer. Cancer Immunol 8:11 An DS, Xie YM, Chen IS (2001) Envelope gene of the human endogenous retrovirus HERV-W encodes a functional retrovirus envelope. J Virol 75:3488–3489 Andersen TL, Boissy P, Sondergaard TE et al (2007) Osteoclast nuclei of myeloma patients show chromosome translocations specific for the myeloma cell clone: a new type of cancer-host partnership? J Pathol 211:10–17 Anderson JM (2000) Multinucleated giant cells. Curr Opin Hematol 7:40–47 Anderson MJ, Stanbridge EJ (1993) Tumor suppressor genes studied by cell hybridization and chromosome transfer. FASEB J 7:826–833 Andersson AC, Svensson AC, Rolny C, Andersson G, Larsson E (1998) Expression of human endogenous retrovirus ERV3 (HERV-R) mRNA in normal and neoplastic tissues. Int J Oncol 12:309–313 Antony JM, van Marle G, Opii W et al (2004) Human endogenous retrovirus glycoproteinmediated induction of redox reactants causes oligodendrocyte death and demyelination. Nat Neurosci 7:1088–1095 Antony JM, Zhu Y, Izad M et al (2007) Comparative expression of human endogenous retrovirusW genes in multiple sclerosis. AIDS Res Hum Retroviruses 23:1251–1256 Armbruester V, Sauter M, Krautkraemer E et al (2002) A novel gene from the human endogenous retrovirus K expressed in transformed cells. Clin Cancer Res 8:1800–1807 Athanasou NA, Wells CA, Quinn J et al (1989) The origin and nature of stromal osteoclastlike multinucleated giant cells in breast carcinoma: implications for tumour osteolysis and macrophage biology. Br J Cancer 59:491–498 Atra A, Al-Asiri R, Wali S et al (2007) Hepatocellular carcinoma, syncytial giant cell: a novel variant in children: a case report. Ann Diagn Pathol 11:61–63 Avital I, Moreira AL, Klimstra DS et al (2007) Donor-derived human bone marrow cells contribute to solid organ cancers developing after bone marrow transplantation. Stem Cells 25:2903–2909 Balada E, Ordi-Ros J, Vilardell-Tarrés M (2009) Molecular mechanisms mediated by human endogenous retroviruses (HERVs) in autoimmunity. Rev Med Virol 19:273–286 Bannert N, Kurth R (2004) Retroelements and the human genome: new perspectives on an old relation. Proc Natl Acad Sci USA 101:14572–14579 Bar-Shavit Z (2007) The osteoclast: a multinucleated, hematopoietic-origin, bone-resorbing osteoimmune cell. J Cell Biochem 102:1130–1139 Bieda K, Hoffmann A, Boller K (2001) Phenotypic heterogeneity of human endogenous retrovirus particles produced by teratocarcinoma cell lines. J Gen Virol 82:591–596 Bjerkvig R, Tysnes BB, Aboody KS et al (2005) Opinion: the origin of the cancer stem cell: current controversies and new insights. Nat Rev Cancer 5:899–904 Bjerregaard B, Holck S, Christensen IJ et al (2006) Syncytin is involved in breast cancerendothelial cell fusions. Cell Mol Life Sci 63:1906–1911
418
R. Strick et al.
Blaak H, vant Wout AB, Brouwer M (2000) In vivo HIV-1 infection of CD45RA(+)CD4(+) T cells is established primarily by syncytium-inducing variants and correlates with the rate of CD4(+) T cell decline. Proc Natl Acad Sci USA 97:1269–1274 Blaise S, de Parseval N, Bénit L et al (2003) Genomewide screening for fusogenic human endogenous retrovirus envelopes identifies syncytin 2, a gene conserved on primate evolution. Proc Natl Acad Sci USA 100:13013–13018 Blaise S, de Parseval N, Heidmann T (2005) Functional characterization of two newly identified Human Endogenous Retrovirus coding envelope genes. Retrovirology 2:19 Blobel CP, Wolfsberg TG, Turck CW et al (1992) A potential fusion peptide and an integrin ligand domain in a protein active in sperm-egg fusion. Nature 356:248–252 Blond JL, Besème F, Duret L et al (1999) Molecular characterization and placental expression of HERV-W, a new human endogenous retrovirus family. J Virol 73:1175–1185 Boller K, König H, Sauter M et al (1993) Evidence that HERV-K is the endogenous retrovirus sequence that codes for the human teratocarcinoma-derived retrovirus HTDV. Virology 196:349–353 Borges M, Bose P, Frank HG et al (2003) A two-colour fluorescence assay for the measurement of syncytial fusion between trophoblast-derived cell lines. Placenta 24:959–964 Boros DL (1989) Immunopathology of Schistosoma mansoni infection. Clin Microbiol Rev 2:250–269 Boyd AS, Wu H, Shyr Y (2005) Monster cells in malignant melanoma. Am J Dermatopathol 27:208–210 Bronson DL, Fraley EE, Fogh J et al (1979) Induction of retrovirus particles in human testicular tumor (Tera-1) cell cultures: an electron microscopic study. J Natl Cancer Inst 63:337–339 Budka H (1986) Multinucleated giant cells in brain: a hallmark of the acquired immune deficiency syndrome (AIDS). Acta Neuropathol 69:253–258 Büscher K, Trefzer U, Hofmann M et al (2005) Expression of human endogenous retrovirus K in melanomas and melanoma cell lines. Cancer Res 65:4172–4180 Busund LT, Killie MK, Bartnes K et al (2002) Spontaneously formed tumorigenic hybrids of Meth A sarcoma and macrophages grow faster and are better vascularized than the parental tumor. Int J Cancer 100:407–413 Busund LT, Killie MK, Bartnes K et al (2003) Spontaneously formed tumorigenic hybrids of Meth A sarcoma cells and macrophages in vivo. Int J Cancer 106:153–159 Camargo FD, Finegold M, Goodell MA (2004) Hematopoietic myelomonocytic cells are the major source of hepatocyte fusion partners. J Clin Invest 113:1266–1270 Camargo FD, Green R, Capetanaki Y et al (2003) Single hematopoietic stem cells generate skeletal muscle through myeloid intermediates. Nat Med 9:1520–1527 Castedo M, Perfettini JL, Roumier T et al (2004) Cell death by mitotic catastrophe: a molecular definition. Oncogene 23:2825–2837 Chakraborty A, Lazova R, Davies S et al (2004) Donor DNA in a renal cell carcinoma metastasis from a bone marrow transplant recipient. Bone Marrow Transplant 34:183–186 Chakraborty AK, de Freitas Sousa J, Espreafico EM et al (2001) Human monocyte x mouse melanoma fusion hybrids express human gene. Gene 275:103–106 Chang C, Chen PT, Chang GD et al (2004) Functional characterization of the placental fusogenic membrane protein syncytin. Biol Reprod 71:1956–1962. Chen CP, Chen LF, Yang SR et al (2008) Functional characterization of the human placental fusogenic membrane protein syncytin 2. Biol Reprod 79:815–823 Chen EH, Grote E, Mohler W et al (2007) Cell–cell fusion. FEBS Lett 581:2181–2193 Cheng JQ, Ruggeri B, Klein WM et al (1996) Amplification of AKT2 in human pancreatic cells and inhibition of AKT2 expression and tumorigenicity by antisense RNA. Proc Natl Acad Sci USA 93:3636–3641 Cheynet V, Oriol G, Mallet F (2006) Identification of the hASCT2-binding domain of the Env ERVWE1/syncytin-1 fusogenic glycoprotein. Retrovirology 3:41 Cheynet V, Ruggieri A, Oriol G et al (2005) Synthesis, assembly, and processing of the Env ERVWE1/syncytin human endogenous retroviral envelope. J Virol 79:5585–5593
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
419
Clement PB, Young RH (2004) Non-endometrioid carcinomas of the uterine corpus: a review of their pathology with emphasis on recent advances and problematic aspects. Adv Anat Pathol 11:117–142 Cohen M, Kato N, Larsson E (1988) ERV3 human endogenous provirus mRNAs are expressed in normal and malignant tissues and cells, but not in choriocarcinoma tumor cells. J Cell Biochem 36:121–128 Contreras-Galindo R, Kaplan MH, Leissner P et al (2008) Human endogenous retrovirus K (HML-2) elements in the plasma of people with lymphoma and breast cancer. J Virol 82:9329–9336 Cowan CA, Atienza J, Melton DA, Eggan K (2005) Nuclear reprogramming of somatic cells after fusion with human embryonic stem cells. Science 309:1369–1373 Coyne JD (2007) DCIS and LCIS with multinucleated giant cells-a report of 4 cases. Histopathology 50:669–671 Davies PS, Powell AE, Swain JR et al (2009) Inflammation and proliferation act together to mediate intestinal cell fusion. PLoS One 4:e6530 De Chiara A, Apice G, Fazioli F et al (1998) Multicentric giant cell tumor with viral-like inclusions associated with Paget s disease of bone: a case treated by steroid therapy. Oncol Rep 5:317–320 de Parseval N, Diop G, Blaise S et al (2005) Comprehensive search for intra- and inter-specific sequence polymorphisms among coding envelope genes of retroviral origin found in the human genome: genes and pseudogenes. BMC Genomics 6:117 de Parseval N, Lazar V, Casella JF et al (2003) Survey of human genes of retroviral origin: identification and transcriptome of the genes with coding capacity for complete envelope proteins. J Virol 77:10414–10422 Deckard-Janatpour K, Kragel S, Teplitz RL et al (1998) Tumors of the pancreas with osteoclastlike and pleomorphic giant cells: an immunohistochemical and ploidy study. Arch Pathol Lab Med 122:266–272 Depil S, Roche C, Dussart P et al (2002) Expression of a human endogenous retrovirus, HERV-K, in the blood cells of leukemia patients. Leukemia 16:254–259 Dewannieux M, Blaise S, Heidmann T (2005) Identification of a functional envelope protein from the HERV-K family of human endogenous retroviruses. J Virol79:15573–15577 Drewlo S, Leyting S, Kokozidou M et al (2006) C-Terminal truncations of syncytin-1 (ERVWE1 envelope) that increase its fusogenicity. Biol Chem 387:1113–1120 Drexler HG, Gignac SM, Hoffbrand AV et al (1989) Formation of multinucleated cells in a Hodgkin’s-disease-derived cell line. Int J Cancer 43:1083–1090 Duelli DM, Hearn S, Myers MP et al (2005) A primate virus generates transformed human cells by fusion. J Cell Biol 171:493–503 Duprex WP, McQuaid S, Hangartner L et al (1999) Observation of measles virus cell-to-cell spread in astrocytoma cells by using a green fluorescent protein-expressing recombinant virus. J Virol 73:9568–9575 Engel FB, Schebesta M, Keating MT (2006) Anillin localization defect in cardiomyocyte binucleation. J Mol Cell Cardiol 41:601–612 Esnault C, Priet S, Ribet D et al (2008) A placenta-specific receptor for the fusogenic, endogenous retrovirus-derived, human syncytin-2. Proc Natl Acad Sci USA 105:17532–17537 Everts V, de Vries TJ, Helfrich MH (2009) Osteoclast heterogeneity: lessons from osteopetrosis and inflammatory conditions. Biochim Biophys Acta 1792:757–765 Factor SM, Biempica L, Ratner I et al (1977) Carcinoma of the breast with multinucleated reactive stromal giant cells. A light and electron microscopic study of two cases. Virchows Arch A Pathol Anat Histol 374:1–12 Feng H, Shuda M, Chang Y et al (2008) Clonal integration of a polyomavirus in human Merkel cell carcinoma. Science 319:1096–1100 Ferrari G, Cusella-De Angelis G, Coletta M et al (1998) Muscle regeneration by bone marrowderived myogenic progenitors. Science 279:1528–1530
420
R. Strick et al.
Finkeldey W (1931) Über Riesenzellbefunde in den Gaumenmandeln, zugleich ein Beitrag zur Histopathologie der Mandelveränderungen im Maserninkubationsstadium. Virchows Arch 281:321–329 Folpe AL, Mentzel T, Lehr HA et al (2005) Perivascular epithelioid cell neoplasms of soft tissue and gynecologic origin: a clinicopathologic study of 26 cases and review of the literature. Am J Surg Pathol 29:1558–1575 Frank O, Verbeke C, Schwarz N et al (2008) Variable transcriptional activity of endogenous retroviruses in human breast cancer. J Virol 82:1808–1818 Frankel SS, Wenig BM, Burke AP et al (1996) Replication of HIV-1 in dendritic cell-derived syncytia at the mucosal surface of the adenoid. Science 272:115–117 Funakoshi E, Hori T, Haraguchi T et al (2003) Overexpression of the human MNB/DYRK1A gene induces formation of multinucleate cells through overduplication of the centrosome. BMC Cell Biol 4:12 Galli R, Binda E, Orfanelli U et al (2004) Isolation and characterization of tumorigenic, stem-like neural precursors from human glioblastoma. Cancer Res 64:7011–7021 Ganem NJ, Storchova Z, Pellman D (2007) Tetraploidy, aneuploidy and cancer. Curr Opin Genet Dev 17:157–162 Gattegno T, Mittal, A, Valansi, C et al (2007) Genetic control of fusion pore expansion in the epidermis of Caenorhabditis elegans. Mol Biol Cell 18:1153–1166 Glotzer M (2001) Animal cell cytokinesis. Annu Rev Cell Dev Biol 17:351–386 Goedert JJ, Sauter ME, Jacobson LP et al (1999) High prevalence of antibodies against HERVK10 in patients with testicular cancer but not with AIDS. Cancer Epidemiol Biomarkers Prev 8:293–296 Golan M, Hizi A, Resau JH et al (2008) Human endogenous retrovirus (HERV-K) reverse transcriptase as a breast cancer prognostic marker. Neoplasia 10:521–533 Goldenberg DM, Pavia RA, Tsao MC (1974) In vivo hybridisation of human tumour and normal hamster cells. Nature 250:649–651 Guidotti JE, Brégerie O, Robert A et al (2003) Liver cell polyploidization: a pivotal role for binuclear hepatocytes. J Biol Chem 278:19095–19101 Guo GQ, Zheng GC (2004) Hypotheses for the functions of intercellular bridges in male germ cell development and its cellular mechanisms. J Theor Biol 229:139–146 Gupta S (2000) Hepatic polyploidy and liver growth control. Semin Cancer Biol 10:161–171 Gussoni E, Soneoka Y, Strickland CD et al (1999) Dystrophin expression in the MDX mouse restored by stem cell transplantation. Nature 401:390–394 Hahn S, Ugurel S, Hanschmann KM et al (2008) Serological response to human endogenous retrovirus K in melanoma patients correlates with survival probability. AIDS Res Hum Retroviruses 24:717–723 Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell 100:57–70 Hansemann D (1890) Ueber asymmetrische Zelltheilung in Epithelkrebsen und deren biologische Bedeutung. Arch Pathol Anat etc. Berl (Virchow’s Arch) 119:299–326 Harris H (1988) The analysis of malignancy by cell fusion: the position in 1988. Cancer Res 48:3302–3306 Harris H, Miller OJ, Klein G et al (1969) Suppression of malignancy by cell fusion. Nature 223:363–368 Harris H, Watkins JF (1965) Hybrid cells derived from mouse and man: artificial heterokaryons of mammalian cells from different species. Nature 205:640–646 Helm CA, Israelachvili JN, McGuiggan PM (1992) Role of hydrophobic forces in bilayer adhesion and fusion. Biochemistry 31:1794–805 Helming L, Gordon S (2007) The molecular basis of macrophage fusion. Immunobiology 212:785–793 Herbst H, Sauter M, Mueller-Lantzsch N (1996) Expression of human endogenous retrovirus K elements in germ cell and trophoblastic tumors. Am J Pathol 149: 1727–1735
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
421
Homma T, Fukushima T, Vaccarella S et al (2006) Correlation among pathology, genotype, and patient outcomes in glioblastoma. J Neuropathol Exp Neurol 65:846–854 Horsley V, Pavlath GK (2004) Forming a multinucleated cell: molecules that regulate myoblast fusion. Cells Tissues Organs 176:67–78 Horváth EM, Lacza Z, Csordás A et al (2006) Graft derived cells with double nuclei in the penumbral region of experimental brain trauma. Neurosci Lett 396:182–186 Houghton J, Stoicov C, Nomura S et al (2004) Gastric cancer originating from bone marrowderived cells. Science 306:1568–1571 Huppertz B. (2009) Review: biology of the placental syncytiotrophoblast – myths and facts. Placenta. 2009 Dec 29. [Epub ahead of print] Ianus A, Holz GG, Theise ND et al (2003) In vivo derivation of glucose-competent pancreatic endocrine cells from bone marrow without evidence of cell fusion. J Clin Invest 111:843–850 Ishida T, Obata Y, Ohara N et al (2008) Identification of the HERV-K gag antigen in prostate cancer by SEREX using autologous patient serum and its immunogenicity. Cancer Immun 8:15 Jacobsen BM, Harrell JC, Jedlicka P (2006) Spontaneous fusion with, and transformation of mouse stroma by, malignant human breast cancer epithelium. Cancer Res 66:8274–8279 Jin J, Woodgett JR (2005) Chronic activation of protein kinase Bbeta/Akt2 leads to multinucleation and cell fusion in human epithelial kidney cells: events associated with tumorigenesis. Oncogene 24:5459–5470 Johansson CB, Youssef S, Koleckar K et al (2008) Extensive fusion of haematopoietic cells with Purkinje neurons in response to chronic inflammation. Nat Cell Biol 10:575–583 Jones MA, Young RH, Scully RE (1991) Endometrial adenocarcinoma with a component of giant cell carcinoma. Int J Gynecol Pathol 10:260–270 Jurka J (2000) Repbase update: a database and an electronic journal of repetitive elements. Trends Genet 16:418–20 Karlsson H, Bachmann S, Schröder J et al (2001) Retroviral RNA identified in the cerebrospinal fluids and brains of individuals with schizophrenia. Proc Natl Acad Sci USA 98:4634–4639 Kerbel RS, Lagarde AE, Dennis JW et al (1983) Spontaneous fusion in vivo between normal host and tumor cells: possible contribution to tumor progression and metastasis studied with a lectin-resistant mutant tumor. Mol Cell Biol 3:523–538 Keydar I, Ohno T, Nayak R et al (1984) Properties of retrovirus-like particles produced by a human breast carcinoma cell line: immunological relationship with mouse mammary tumor virus proteins. Proc Natl Acad Sci USA 81:4188–4192 Kim S, Shilagardi K, Zhang S et al (2007) A critical function for the actin cytoskeleton in targeted exocytosis of prefusion vesicles during myoblast fusion. Dev Cell 12:571–586 Kodama S, Kühtreiber W, Fujimura S et al (2003) Islet regeneration during the reversal of autoimmune diabetes in NOD mice. Science 302:1223–1227 Koranda FC (2004) Images in clinical medicine. Use of multinucleated giant cells to diagnose a viral eruption. N Engl J Med 350:e6 Krause DS, Theise ND, Collector MI et al (2001) Multi-organ, multi-lineage engraftment by a single bone marrow-derived stem cell. Cell 105:369–377 Krishnan C, Longacre TA (2006) Ductal carcinoma in situ of the breast with osteoclast-like giant cells. Hum Pathol 37:369–372 Kudo Y, Boyd CA (2002) Changes in expression and function of syncytin and its receptor, amino acid transport system B(0) (ASCT2), in human placental choriocarcinoma BeWo cells during syncytialization. Placenta 23:536–541 Kudryavtsev BN, Kudryavtseva MV, Sakuta GA et al (1993) Human hepatocyte polyploidization kinetics in the course of life cycle. Virchows Arch B Cell Pathol Incl Mol Pathol 64: 387–393 Küppers R (2009) The biology of Hodgkin’s lymphoma. Nat Rev Cancer 9:15–27 Laforga JB (1999) Giant cell carcinoma of the lung. Report of a case with cytohistologic and clinical correlation. Acta Cytol 43:263–267
422
R. Strick et al.
Lagasse E, Connors H, Al-Dhalimy M et al (2000) Purified hematopoietic stem cells can differentiate into hepatocytes in vivo. Nat Med 6:1229–1234 Langbein M, Strick R, Strissel PL et al (2008) Impaired cytotrophoblast cell–cell fusion is associated with reduced Syncytin and increased apoptosis in patients with placental dysfunction. Mol Reprod Dev 75:175–183 Langhans T (1868) Über Riesenzellen mit wandständigen Kernen in Tuberkeln und die fibröse Form des Tuberkels. Virchows Arch (Pathol Anat) 42:382–404 Larizza L, Schirrmacher V (1984) Somatic cell fusion as a source of genetic rearrangement leading to metastatic variants. Cancer Metastasis Rev 3:193–222 Larizza L, Schirrmacher V, Pfluger E (1984) Acquisition of high metastatic capacity after in vitro fusion of a nonmetastatic tumor line with a bone marrow-derived macrophage. J Exp Med 160:1579–1584 Larsen JM, Christensen IJ, Nielsen HJ et al (2009) Syncytin immunoreactivity in colorectal cancer: potential prognostic impact. Cancer Lett 280:44–49 Larsson LI, Bjerregaard B, Talts JF (2008) Cell fusions in mammals. Histochem Cell Biol 129:551–561. Larsson LI, Holck S, Christensen IJ (2007) Prognostic role of syncytin expression in breast cancer. Hum Pathol 38:726–731 Lau YS, Danks L, Sun SG et al (2007) RANKL-dependent and RANKL-independent mechanisms of macrophage-osteoclast differentiation in breast cancer. Breast Cancer Res Treat 105:7–16. Lau YS, Sabokbar A, Giele H et al (2006) Malignant melanoma and bone resorption. Br J Cancer 94:1496–503. Lavillette D, Marin M, Ruggieri A et al (2002) The envelope glycoprotein of human endogenous retrovirus type W uses a divergent family of amino acid transporters/cell surface receptors. J Virol 76:6442–6452. Leib-Mösch C, Brack-Werner R, Werner T et al (1990) Endogenous retroviral elements in human DNA. Cancer Res 50:5636S–5642S. Lescar J, Roussel A, Wien MW et al (2001) Fusion glycoprotein shell of Semliki Forest virus: an icosahedral assembly primed for fusogenic activation at endosomal pH. Cell 105: 137–148 Lindeskog M, Blomberg J (1997) Spliced human endogenous retroviral HERV-H env transcripts in T-cell leukaemia cell lines and normal leukocytes: alternative splicing pattern of HERV-H transcripts. J Gen Virol 78:2575–2585 Liu C, Chen Z, Zhang T et al (2006) Multiple tumor types may originate from bone marrow-derived cells. Neoplasia 8:716–724 Lordier L, Jalil A, Aurade F et al (2008) Megakaryocyte endomitosis is a failure of late cytokinesis related to defects in the contractile ring and Rho/Rock signaling. Blood 112:3164–3174 Löwer R, Löwer J, Frank H et al (1984) Human teratocarcinomas cultured in vitro produce unique retrovirus-like viruses. J Gen Virol 65:887–898 Lu X, Kang Y (2009) Efficient acquisition of dual metastasis organotropism to bone and lung through stable spontaneous fusion between MDA-MB-231 variants. Proc Natl Acad Sci USA 106:9385–9390 Maeda K, Mizuno M, Wakabayashi T et al (2003) Morphological assessment of the development of multinucleated giant cells in glioma by using mitosis-specific phosphorylated antibodies. J Neurosurg 98:854–859 Malassiné A, Blaise S, Handschuh K et al (2007) Expression of the fusogenic HERV-FRD Env glycoprotein (syncytin 2) in human placenta is restricted to villous cytotrophoblastic cells. Placenta 28:185–191 Mallet F, Bouton O, Prudhomme S et al (2004) The endogenous retroviral locus ERVWE1 is a bona fide gene involved in hominoid placental physiology. Proc Natl Acad Sci USA 101:1731–1736 Margall-Ducos G, Celton-Morizur S, Couton D et al (2007) Liver tetraploidization is controlled by a new process of incomplete cytokinesis. J Cell Sci 120:3633–3639
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
423
Margolis RL (2005). Tetraploidy and tumor development. Cancer Cell 8:353–354 Matousková M, Blazková J, Pajer P et al (2006) CpG methylation suppresses transcriptional activity of human syncytin-1 in non-placental tissues. Exp Cell Res 312:1011–1020 Matveeva NM, Shilov AG, Kaftanovskaya EM et al (1998) In vitro and in vivo study of pluripotency in intraspecific hybrid cells obtained by fusion of murine embryonic stem cells with splenocytes. Mol Reprod Dev 50:128–138 Meraldi P, Honda R, Nigg EA (2002) Aurora-A overexpression reveals tetraploidization as a major route to centrosome amplification in p53–/– cells. EMBO J 21:483–492 Mi S, Lee X, Li X et al (2000) Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature 403:785–789 Miething A (1995) Multinuclearity of germ cells in the senescent human testis originates from a process of cell–cell fusion. J Submicrosc Cytol Pathol 27:105–113 Miettinen M, Virtanen I, Talerman A (1985) Intermediate filament proteins in human testis and testicular germ-cell tumors. Am J Pathol 120:402–410 Miller FR, McInerney D, Rogers et al (1988) Spontaneous fusion between metastatic mammary tumor subpopulations. J Cell Biochem 36:129–136 Miller RA, Ruddle FH (1976) Pluripotent teratocarcinoma-thymus somatic cell hybrids. Cell 9:45–55 Miyoshi Y, Iwao K, Egawa C et al (2001) Association of centrosomal kinase STK15/BTAK mRNA expression with chromosomal instability in human breast cancers. Int J Cancer 92:370–373 Mohler WA, Shemer G, del Campo JJ et al (2002) The type I membrane protein EFF-1 is essential for developmental cell fusion. Dev Cell 2:355–362 Mortensen K, Lichtenberg J, Thomsen PD et al (2004) Spontaneous fusion between cancer cells and endothelial cells. Cell Mol Life Sci 61:2125–2131 Mostofi FK, Sesterhenn IA (1985) Pathology of germ cell tumors of testes. Prog Clin Biol Res 203:1–34 Munoz PA, Rao MS, Reddy JK (1980) Osteoclastoma-like giant cell tumor of the liver. Cancer 46:771–779 Nakahashi H (1987) Undifferentiated carcinoma of the lung with osteoclast-like giant cells. Jpn J Surg 17:199–203 Navarro D, Paz P, Tugizov S et al (1993) Glycoprotein B of human cytomegalovirus promotes virion penetration into cells, transmission of infection from cell to cell, and fusion of infected cells. Virology 197:143–158 Nelson PN, Carnegie PR, Martin J et al (2003) Demystified. Human endogenous retroviruses. Mol Pathol 56:11–18 Nygren JM, Jovinge S, Breitbach M et al (2004) Bone marrow-derived hematopoietic cells generate cardiomyocytes at a low frequency through cell fusion, but not transdifferentiation. Nat Med 10:494–501 Nygren JM, Liuba K, Breitbach M et al (2008) Myeloid and lymphoid contribution to nonhaematopoietic lineages through irradiation-induced heterotypic cell fusion. Nat Cell Biol 10:584–592 Okamoto H, Mizuno K, Horio T (2003) Monocyte-derived multinucleated giant cells and sarcoidosis. J Dermatol Sci 31:119–128 Ono M, Kawakami M, Ushikubo H (1987) Stimulation of expression of the human endogenous retrovirus genome by female steroid hormones in human breast cancer cell line T47D. J Virol 61:2059–2062 Oren-Suissa M, Podbilewicz B (2007) Cell fusion during development. Trends Cell Biol 17:537–546 Pathak S, Nemeth MA, Multani AS et al (1997) Can cancer cells transform normal host cells into malignant cells? Br J Cancer 76:1134–1138 Pawelek JM (2005) Tumour-cell fusion as a source of myeloid traits in cancer. Lancet Oncol 6:988–993
424
R. Strick et al.
Pawelek JM, Chakraborty AK (2008) Fusion of tumour cells with bone marrow-derived cells: a unifying explanation for metastasis. Nat Rev Cancer 8:377–386 Pesce C, Merino MJ, Chambers JT et al (1991) Endometrial carcinoma with trophoblastic differentiation. An aggressive form of uterine cancer. Cancer 68:1799–1802 Piquer-Gil M, García-Verdugo JM, Zipancic I et al (2009) Cell fusion contributes to pericyte formation after stroke. J Cereb Blood Flow Metab 29:480–485 Prudhomme S, Oriol G, Mallet F (2004) A retroviral promoter and a cellular enhancer define a bipartite element which controls Env ERVWE1 placental expression. J Virol 78:12157–12168 Prusty BK, zur Hausen H, Schmidt R et al (2008) Transcription of HERV-E and HERV-Erelated sequences in malignant and non-malignant human haematopoietic cells. Virology 382: 37–45 Rakowicz-Szulczynska EM, McIntosh DG, Smith ML (1999) Giant syncytia and virus-like particles in ovarian carcinoma cells isolated from ascites fluid. Clin Diagn Lab Immunol 6:115–126 Ravid K, Lu J, Zimmet JM, Jones MR (2002) Roads to polyploidy: the megakaryocyte example. J Cell Physiol 190:7–20 Razzuk MA, Urschel HC Jr, Albers JE et al (1976) Pulmonary giant cell carcinoma. Ann Thorac Surg 21:540–545 Reinecke H, Minami E, Poppa V et al (2004) Evidence for fusion between cardiac and skeletal muscle cells. Circ Res 94:e56–e60 Rizvi AZ, Swain JR, Davies PS et al (2006) Bone marrow-derived cells fuse with normal and transformed intestinal stem cells. Proc Natl Acad Sci USA 103:6321–6325 Robinson LR, Damjenov I, Brezina P (1977) Multinucleated giant cell neoplasm of pancreas. Arch Pathol Lab Med 101:590–593 Rodi´c N, Rutenberg MS, Terada N (2004) Cell fusion and reprogramming: resolving our transdifferences. Trends Mol Med 10:93–96 Roger V, Nasiell M, Nasiell K et al (1972) Cytologic findings indicating pulmonary tuberculosis. II. The occurrence in sputum of epithelioid cells and multinucleated giant cells in pulmonary tuberculosis, chronic non-tuberculous inflammatory lung disease and bronchogenic carcinoma. Acta Cytol 16:538–542 Roodman GD, Windle JJ (2005) Paget disease of bone. J Clin Invest 115:200–208. Saltel F, Chabadel A, Zhao Y et al (2006) Transmigration: a new property of mature multinucleated osteoclasts. J Bone Miner Res 21:1913–1923 Schleiden MJ (1838) Beiträge zur Phytogenesis. Archiv für Anatomie, Physiologie und wissenschaftliche Medicin 13:137–176 Schmitz-Winnenthal FH, Galindo-Escobedo LV, Rimoldi D et al (2007) Potential target antigens for immunotherapy in human pancreatic cancer. Cancer Lett 252:290–298 Schwann T (1839) Mikroskopische Untersuchungen über die Übereinstimmung in der Struktur und dem Wachstum der Thiere und Pflanzen. Sander (Reimer), Berlin Serafino A, Balestrieri E, Pierimarchi P et al (2009) The activation of human endogenous retrovirus K (HERV-K) is implicated in melanoma cell malignant transformation. Exp Cell Res 315:849–862 Shabo I, Olsson H, Sun XF et al (2009) Expression of the macrophage antigen CD163 in rectal cancer cells is associated with early local recurrence and reduced survival time. Int J Cancer 125:1826–1831 Shemer G, Podbilewicz B (2003) The story of cell fusion: big lessons from little worms. Bioessays 25:672–682 Shestopalov VI, Bassnett S (2000) Expression of autofluorescent proteins reveals a novel protein permeable pathway between cells in the lens core. J Cell Sci 113:1913–1921 Shi Y, Barton K, De Maria A et al (2009) The stratified syncytium of the vertebrate lens. J Cell Sci 122:1607–1615 Singh SK, Hawkins C, Clarke ID et al (2004) Identification of human brain tumour initiating cells. Nature 432:396–401
17
Cell–Cell Fusions and Human Endogenous Retroviruses in Cancer
425
Skehel JJ, Wiley DC (1998) Coiled coils in both intracellular vesicle and viral membrane fusion. Cell 95:871–874 Stanbridge EJ (1976) Suppression of malignancy in human cells. Nature 260:17–20 Stanbridge EJ, Der CJ, Doersen CJ et al (1982) Human cell hybrids: analysis of transformation and tumorigenicity. Science 215:252–259 Stauffer Y, Theiler G, Sperisen P et al (2004) Digital expression profiles of human endogenous retroviral families in normal and cancerous tissues. Cancer Immunol 4:2 Strick R, Ackermann S, Langbein M et al (2007) Proliferation and cell–cell fusion of endometrial carcinoma are induced by the human endogenous retroviral Syncytin-1 and regulated by TGFbeta. J Mol Med 85:23–38 Tada M, Takahama Y, Abe K et al (2001) Nuclear reprogramming of somatic cells by in vitro hybridization with ES cells. Curr Biol 11:1553–1558 Takeuchi H, Kubota T, Kitai R et al (2008) CD98 immunoreactivity in multinucleated giant cells of glioblastomas: an immunohistochemical double labeling study. Neuropathology 28:127–131 Takeuchi H, Sato K, Ido K et al (2006) Mitotic activity of multinucleated giant cells with glial fibrillary acidic protein immunoreactivity in–18 glioblastomas: an immunohistochemical double labeling study. J Neurooncol 78:15 Terada N, Hamazaki T, Oka M et al (2002) Bone marrow cells adopt the phenotype of other cells by spontaneous cell fusion. Nature 416:542–545 Therman E, Sarto GE, Buchler DA (1983) The structure and origin of giant nuclei in human cancer cells. Cancer Genet Cytogenet 9:9–18 Tran SD, Pillemer SR, Dutra A et al (2003) Differentiation of human bone marrow-derived cells into buccal epithelial cells in vivo: a molecular analytical study. Lancet 361:1084–1088 Tristem M (2000) Identification and characterization of novel human endogenous retrovirus families by phylogenetic screening of the human genome mapping project database. J Virol 74:3715–3730 Turbeville MA, Rhodes JC, Hyams DM et al (1997) Characterization of a putative retroviral envrelated human protein. Pathobiology 65:123–128 Turner G, Barbulescu M, Su M et al (2001) Insertional polymorphisms of full-length endogenous retroviruses in humans. Curr Biol 11:1531–1535 Ulbright TM (2005) Germ cell tumors of the gonads: a selective review emphasizing problems in differential diagnosis, newly appreciated, and controversial issues. Mod Pathol 18: S61–S79 Ullah Z, Kohn MJ, Yagi R et al (2008) Differentiation of trophoblast stem cells into giant cells is triggered by p57/Kip2 inhibition of CDK1 activity. Genes Dev 22:3024–3036 Vassilopoulos G, Wang PR, Russell DW (2003) Transplanted bone marrow regenerates liver by cell fusion. Nature 422:901–904 Veliath AJ, Sandaran V, Aurora AM (1975) Ovarian giant cell tumor with cystadenocarcinoma. Arch Pathol 99:488–491 Vignery A (2005) Macrophage fusion: are somatic and cancer cells possible partners? Trends Cell Biol 15:188–193 Villesen P, Aagaard L, Wiuf C et al (2004) Identification of endogenous retroviral reading frames in the human genome. Retrovirology 1:32 von Hochstetter AR, Sigg C, Saremaslani P et al (1985) The significance of giant cells in human testicular seminomas. A clinico-pathological study. Virchows Arch A Pathol Anat Histopathol 407:309–322 Wang X, Willenbring H, Akkari Y et al (2003) Cell fusion is the principal source of bone-marrowderived hepatocytes. Nature 422:897–901 Wang-Johanning F, Frost AR, Jian B et al (2003) Detecting the expression of human endogenous retrovirus E envelope transcripts in human prostate adenocarcinoma. Cancer 98:187–197 Wang-Johanning F, Frost AR, Jian B et al (2003) Quantitation of HERV-K env gene expression and splicing in human breast cancer. Oncogene 22:1528–1535
426
R. Strick et al.
Wang-Johanning F, Frost AR, Johanning GL et al (2001) Expression of human endogenous retrovirus k envelope transcripts in human breast cancer. Clin Cancer Res 7:1553–1560 Wang-Johanning F, Liu J, Rycaj K et al (2007) Expression of multiple human endogenous retrovirus surface envelope proteins in ovarian cancer. Int J Cancer 120:81–90 Wang-Johanning F, Radvanyi L, Rycaj K et al (2008) Human endogenous retrovirus K triggers an antigen-specific immune response in breast cancer patients. Cancer Res 68:5869–5877 Watanabe K, Hiraki H, Ohishi M et al (1996) Uterine leiomyosarcoma with osteoclast-like giant cells: Histopathological and cytological observations. Pathol Int 46:656–660 Weimann JM, Johansson CB, Trejo A et al (2003) Stable reprogrammed heterokaryons form spontaneously in Purkinje neurons after bone marrow transplant. Nat Cell Biol 5:959–966 Werner M (2006) Giant cell tumour of bone: morphological, biological and histogenetical aspects. Int Orthop 30:484–489 Wiener F, Klein G, Harris H (1974a) The analysis of malignancy by cell fusion. V. Further evidence of the ability of normal diploid cells to suppress malignancy. J Cell Sci 15:177–183 Wiener F, Klein G, Harris H (1974b) The analysis of malignancy by cell fusion. VI. Hybrids between different tumour cells. J Cell Sci 16:189–198 Williams WJ, Williams D (1967) “Residual bodies” in sarcoid and sarcoid-like granulomas. J Clin Pathol 20:574–577 Winge O (1927) Cytologische Untersuchungen über die Natur maligner Tumoren I. Z. Zellforschg 6 Wöckel L, Ketelsen UP, Stötter M et al (1998) Abundant minute myotubes in a patient who later developed centronuclear myopathy. Acta Neuropathol 95:547–551 Yasunaga M, Ohishi Y, Nishimura I et al (2008) Ovarian undifferentiated carcinoma resembling giant cell carcinoma of the lung. Pathol Int 58:244–248 Yesner R (1985) Large cell carcinoma of the lung. Semin Diagn Pathol 2:255–269 Yi JM, Kim HM, Kim HS (2004) Expression of the human endogenous retrovirus HERV-W family in various human tissues and cancer cells. J Gen Virol 85:1203–1210 Yi JM, Kim HM, Kim HS (2006) Human endogenous retrovirus HERV-H family in human tissues and cancer cells: expression, identification, and phylogeny. Cancer Lett 231:228–239. Yi JM, Kim HS (2004) Expression analysis of endogenous retroviral elements belonging to the HERV-F family from human tissues and cancer cells. Cancer Lett 211:89–96 Yi JM, Kim HS (2007) Expression and phylogenetic analyses of human endogenous retrovirus HC2 belonging to the HERV-T family in human tissues and cancer cells. J Hum Genet 52: 285–296 Yi JM, Kim HS (2007) Molecular phylogenetic analysis of the human endogenous retrovirus E (HERV-E) family in human tissues and human cancers. Genes Genet Syst 82:89–98 Yi JM, Kim TH, Huh JW et al (2004) Human endogenous retroviral elements belonging to the HERV-S family from human tissues, cancer cells, and primates: expression, structure, phylogeny and evolution. Gene 342:283–292 Yi JM, Lee WH, Kim HM et al (2001) Identification of new endogenous retroviral sequences belonging to the HERV-W family in human cancer cells. Intervirology 44:333–338 Yi JM, Schuebel K, Kim HS (2007) Molecular genetic analyses of human endogenous retroviral elements belonging to the HERV-P family in primates, human tissues, and cancer cells. Genomics 89:1–9 Yilmaz Y, Lazova R, Qumsiyeh M et al (2005) Donor Y chromosome in renal carcinoma cells of a female BMT recipient: visualization of putative BMT-tumor hybrids by FISH. Bone Marrow Transplant 35:1021–1024 Ying QL, Nichols J, Evans EP et al (2002) Changing potency by spontaneous fusion. Nature 416:545–548 Zhivotovsky B, Kroemer G (2004) Apoptosis and genomic instability. Nat Rev Mol Cell Biol 5:752–762 Zhou H, Kuang J, Zhong L et al (1998) Tumour amplified kinase STK15/BTAK induces centrosome amplification, aneuploidy and transformation. Nat Genet 20:189–193
Index
A Abi, 155 Acetylation, 130, 132 Acrosin, 188 Acrosomal exocytosis (AE), 187–189, 191, 195–196 Acrosomal matrix, 187–188, 196 Acrosomal shroud, 187–188, 196 Acrosome, 181, 187–189, 191, 195, 281 Actin, 2, 4, 54–55, 139–165, 190, 238, 241, 254, 257–258, 270, 272 Actin filaments, 154, 254 Actin foci/plugs, 152–153, 156, 161 Actin ring, 238, 241 ADAM, 5, 173, 193, 268, 283–286 ADAM12, 193, 207, 239 ADP ribosylation factor (Arf), 145, 152, 258 AKT2, 407 Allantois, 119 Allogeneic DC, 328–329, 334–336, 338 Allogeneic tumor cells, 329 Alpharetroviruses, 22, 25, 69 Amitosis, 287 Amoeboid motility, 361 Amphotericin B, 50, 52 Amphotropic, 15, 45, 48 Anchor-cell fusion failure-1 (AFF-1), 225 Aneuploidy, 6, 354, 367, 397, 399, 403, 414–415 Antigen cross-presentation, 320 Antigen presentation, 321, 330, 332 Antigen-presenting cells, 164, 178, 318 Antisense, 31, 88–89, 100, 122, 129, 206, 210–211, 270–271, 405, 413 Antisocial, 151 Ants, 151, 157 AP-1, 73, 98
Ap-2, 73, 98 Apoptosis, 6, 44, 99–100, 103, 128, 134, 193, 207–208, 224, 403, 414 Appressorium, 176 Arp2/3, 157, 161 ASCT1, 4, 29, 78, 126, 143, 207, 269–270, 409, 413 ASCT2, 29–30, 45, 78, 101, 123, 126, 133, 207, 269–270 Astrocytes, 102, 133–134 Autologous DC, 328–329 Autophagosomes, 356–357, 359, 370, 373–374, 377–381 Autophagy, 6, 357, 369–384 Avian leukosis virus (ALV), 23–25, 50, 69 B Basolateral membrane, 74 Betaretroviruses, 22–26, 69, 71, 408 BeWo cells, 87, 89–90, 98, 100, 121, 124, 127, 129–130, 207–211, 213–214 Blastocyst, 205, 291, 303, 307 Blood–brain barrier, 338 Blow, 144, 152–153, 155–157, 159, 162 Blown fuse, 152 BM40, 354, 367 Bone marrow-derived cells (BMDCs), 6, 280, 290–291, 293, 296–301, 303, 306, 354–355, 363–365, 367–368, 383–384, 402–404, 407, 415 Bone mass, 227, 241, 244 Bone resorption, 5, 222, 238–239, 241, 244 β1,6-Branched oligosaccharides, 357, 359, 362–363, 365, 369–381, 383–384 Breast cancer, 4, 80, 103, 129, 131–133, 269, 320, 329, 336–339, 359, 361, 404–405, 407, 409–413, 416
L.-I. Larsson (ed.), Cell Fusions, DOI 10.1007/978-90-481-9772-9, C Springer Science+Business Media B.V. 2011
427
428 C Ca2+ , 21, 100–101, 174, 188–189, 209, 222 CAAT box, 98 C2C12 cells, 141, 144–145, 160–161 Cadherin, 174, 268 Caenorhabditis elegans, 2, 400 Calcium, 2, 188–189, 238, 256, 259, 261 Calcium signaling, 238 Calpain, 209 CAMP, 3, 5, 98, 100, 103, 121, 129, 132–133, 208, 254, 256, 272, 409 Cancer cell fusion, 5–6, 269, 351–385, 407, 413–415 cells, 3–7, 103–104, 129, 131–133, 243, 269, 299–301, 320, 329–330, 351–385, 398, 404–407, 409–416 stem cells, 229, 299–302, 355, 364, 415–416 vaccines, 317, 340 Capacitation, 187, 195 Cardiomyocytes, 280, 298–299, 398, 403 Caspase, 5, 8, 100, 207, 209–211, 214 Cathepsin, 50–52, 241 CD151, 176, 190, 194, 284–285 CD200, 5, 223, 226–229, 240–242 CD200R, 5, 223, 226–229 CD4, 15–17, 44–45, 51, 53, 71, 77, 320–321, 328, 331, 340–341 CD4+ T cells, 71, 320, 328, 340–341 CD47, 5, 223, 226–229, 239 CD8+ T cells, 321, 328, 331, 341 CD81, 4–5, 176–177, 180, 190, 192, 194, 196, 239–240, 268, 284, 285 CD9, 4–5, 66–67, 171–181, 189–192, 194, 196, 239–240, 258, 268, 272, 284–285 CD9-EGFP, 178–179 C. elegans, 155, 225, 400 Cell adhesion molecule, 2, 4, 53–54, 141, 143, 148–153, 156–157, 160–161, 163, 165, 173–174, 191–192, 239, 256–258, 268, 283–284, 289, 298, 360, 370 Cell division control protein 42 homolog (Cdc42), 144, 155, 161, 258 Cellocytose, 227–228 Cellular protooncogene homologous to Rous sarcoma virus (Src), 238, 259–260 c-Fos, 237, 242 Chemotaxis, 354, 357–359, 366, 368, 382–384 Choriocarcinomas, 208, 213, 406, 410–411, 413 Chorion, 119 Chorionic epithelium, 119
Index Chorionic villi, 95, 119, 205–206 Chromosomal instability, 104 Class I and II fusion proteins, 67 Class III fusion proteins, 67 Clinical trials, 336–339 Clonal expansion, 222 cMet, 368 c-Myc, 305 Coarse melanin, 356–359, 370–371, 373–379, 381 Cofilin, 54, 154 Coiled coil helices, 19 Coiled coil structures, 18, 19–20, 67, 77 Colitis, 294, 398 Collagen folding, 367 Colorectal cancer, 131, 133, 336, 413 Colorectal tumors, 293 Confocal fusion analysis, 291, 295, 327 Connexin43, 239 Co-receptor, 15–16, 17, 25, 77, 92 Co-receptor blockers, 16–17, 25, 77, 92 Costimulatory ligands, 331 CpG methylation, 94, 99, 130 Creatine kinase, 251 CREB binding protein (CBP), 129, 130, 132 Cre recombinase, 161, 290, 292–293, 303 CRISP1, 195 Crk, 144, 152, 156–157, 160 Crkl, 144, 160 CX6 CC, 70, 72, 90, 123, 125 CXXC motif, 20–21, 24, 32, 70, 72–73 Cyclic AMP, 98, 368 Cytokine, 102, 133, 206–207, 236–237, 255, 259–260, 298, 330, 339, 368 Cytokinesis, 7, 152, 235, 293, 397–398, 401, 406 Cytoskeleton, 2, 4, 6, 52, 54–55, 141, 153, 162, 165, 189–190, 209–210, 257–258, 268, 272 Cytotoxic T lymphocytes (CTL), 319–320, 329, 330, 332, 339–341 Cytotrophoblast (CT), 94–97, 99–102, 119–120, 129, 205–206, 209–210, 212–213, 409 D Dedicator of cytokinesis (DOCK), 140 Delta, 142 Deltaretroviruses, 23–24, 69 Dendritic cell maturation, 330–331 Dendritic cells (DC), 7, 24, 177–178, 180, 221–222, 227, 237, 315–342 Dendritic cell-specific ICAM-3-grabbing nonintegrin (DC-SIGN), 24–25
Index Dendritic cell-specific transmembrane protein (DC-STAMP), 5, 222–223, 226–227, 241–243 Disintegrins, 173, 187, 192–194, 207, 268, 283 Disulfide bond isomerization, 32 Disulfide isomerase motif, 125 Disulfide isomerization, 20, 23–24, 32 DNAX-activating protein 12 (DAP12), 222, 238–240, 242 Dock180, 144, 152, 155–156 DRAP27, 175 Drosophila, 4, 6, 67, 98, 139–165, 244, 257–258, 261, 287–288, 361 Duf/Kirre, 142–144, 150–153, 157, 160, 163 Dumbfounded/Kin of irre, 150 E E-cadherin, 239, 360–361, 367 Ecotropic, 15, 22, 43, 45–46, 48–51 Effector T cells, 338 Egg, 5–6, 17, 66, 154, 177–181, 187, 189–196, 205, 239, 244, 281–285, 399 Egg exosomes, 179–181 Electrofusion, 325–328, 332, 334–336, 338 Electron-dense plaques, 146–148, 157, 160 Embryogenesis, 142, 151, 154, 205, 250, 287, 359–360 Embryonic stem cells, 66, 176, 229, 279–280, 282, 286–287, 290, 296–297, 302–305, 307, 401–402 Endogenous class I processing pathway, 328 Endogenous JSRV (enJSRV), 30–31, 68, 83–84, 86–88, 90–92, 97, 122 Endogenous retroviruses (ERVs), 13–14, 23, 26–32, 68, 81–82, 86, 88, 97, 103, 120, 395–416 Endometrial carcinoma (EnCa), 103, 120, 133, 281, 334–335, 406, 410, 412–413 Endomitosis, 7, 397, 404, 406–407 Endoplasmic reticulum (ER), 15, 17, 44, 47, 69–72, 102, 134–135, 237, 367, 381 Endopolyploid, 397 Endoreduplication, 397–399 Endosomes, 18–19, 25, 50–51, 73–74, 78–79, 127, 175, 319 Endothelial cells, 3, 103–104, 133, 270, 355, 369–370, 405, 413 Enfuvirtide, 19 Enhanced green fluorescent protein (EGFP), 178–179 Enveloped viruses, 15, 43, 69, 75, 400, 402 Envelope (env), 3, 13–14, 16, 27–32, 68–83, 94–97, 120, 126, 211, 267–273, 408–414
429 Envelope protein, 3, 13, 15–17, 19, 21–26, 30–32, 67, 71, 81, 87, 89–90, 92, 97, 99, 101, 120–121, 124–127, 134, 211, 405 Epididymal glycoprotein DE, 195 Epigenetic regulation, 130–131 Epigenetic state, 308 Epithelial fusion failure (EFF-1), 165, 225, 268, 400 Epithelial-mesenchymal transition (EMT), 302, 354, 360–361, 367–368 Epsilonretroviruses, 22–23, 69 Equatorin, 194–195 ER stress, 134–135, 381 ERV-3, 89, 211, 410 ERVWE1 transcripts, 80, 82–83, 85–87, 97, 104 E-selectin, 369–370 Estradiol, 103, 409 Estrogen, 133, 412 Exosomes, 5, 75, 177–181, 191, 243 Experimental hybrids, 355 Extracellular degradation, 224 Extracellular lysosome, 224 Extracellular signal-regulated kinase1/2 (ERK), 207–208 Extravillous trophoblast, 205, 212, 400 Ezrin, 54 Ezrin-radixin-moesin (ERM), 4, 54, 190 F F-actin plugs/foci, 141, 151, 153, 156, 163 FF family of fusogens, 225 FBW2, 130, 132 Fertilin β, 193–194, 283 Fertilization, 5, 17, 44, 172–175, 178–179, 181, 187, 189–192, 194–196, 205, 239, 244, 281–285, 296 Filamentous (polymerized) actin (F-actin), 4, 141, 151–159, 161–164, 272 Filopodia, 3, 145–146, 150, 162, 192, 254–256, 271–272, 401 FK506, 242 Flippase, 100, 208 Fluorescence activated cell sorting (FACS), 327 Foamy viruses, 25–26, 69 α-Fodrin, 209–211 Follistatin, 259–261 Foreign body giant cell (FBGC), 239, 241, 302 Forskolin, 121, 127, 208, 213, 409 Founder cell (FC), 142–143, 145–146, 148–151, 157, 162, 164, 222, 321, 324, 328, 330, 332–336, 338–342
430 Fumarylacetoacetate hydrolase (FAH), 243, 297 Fungus, 176–177 Furin, 47, 70, 72, 90, 123–124 Fusion competent myoblast (FCM), 146–153, 155, 157–159, 162–164, 251, 255, 261, 287 Fusion efficiency, 280–281, 324–329, 332–335, 338, 342 Fusion index, 89, 161, 252, 270 Fusion inhibitors, 19, 25, 31, 91, 208 Fusion peptide (FP), 18–19, 24–25, 31, 46, 50, 67, 69–70, 72, 76–77, 79, 84, 89, 123, 125, 127, 225, 283, 402 Fusion pores, 18, 46, 67, 76–77, 79–80, 157–158, 161–163, 209, 258 Fusion proteins, 20–21, 25, 27, 30, 66–67, 77, 79, 90, 172, 209, 225, 268, 283, 285, 402 Fusion restricted myogenic adhesive structure (FuRMAS), 4, 139–165 Fusion theory, 353–354, 385 Fusion from within, 22 Fusion from without, 5, 22 Fusogen, 3, 162, 164, 211, 225, 229, 318, 321, 324–327, 400 Fusogenic membrane glycoproteins (FMG), 324, 326–327, 335 G β-Galactosidase, 252, 290–292, 295, 404 Galectin-3, 369 Gammaretroviruses, 13, 20–23, 26, 29, 32, 69, 71, 408 Ganglioside, 178 Gap junctions, 207, 239 GCM1, 3–4, 128–132, 134, 207–208 GCM1-knockout, 128–129 GCMa, 98, 272 Genomic instability, 6, 300–301, 397 Germ line infection, 26 Giant cells, 5, 43, 66, 88, 95–97, 212, 221–225, 235, 237, 242–243, 302, 398–400, 402, 405 Giant nuclei, 399, 406 Gleefull, 143 Glial cell missing (GCM), 98, 128, 132, 207–208 Glioblastoma, 370, 399, 405–406, 414 Glioma, 333–334, 336–337, 340, 355, 410 Glucocorticoid receptor, 98 Glucose transporter 1 (GLUT1), 24, 45 Glycosylation, 45–46, 49, 70–72, 84–90, 92, 124, 191, 195, 256, 357, 369
Index Glycosyltransferase, 357–359 Gp160, 127, 134 GPI-anchor, 93 Granulomas, 224, 398 Green fluorescent protein (GFP), 153, 158, 161, 288, 291–293, 295, 297–298, 302–303, 306, 326–327 Group-specific antigen (gag), 13–14, 26, 28, 49, 69, 73–75, 86, 97, 120, 408–413 Growth factor, 119, 206–207, 222, 250, 259–260, 268, 338, 359–360 Growth hormone, 259–260 GSK-3β, 130–132, 134 H Hairpins, 19, 46, 67, 77, 79–80, 161, 268 β-hCG, 213–214 Heart, 119, 279, 298–299, 303, 404, 410 HELLP syndrome, 102, 131 Hemagglutinating virus of Japan (HVJ), 326 Hemagglutinin (HA), 17–19, 50, 79, 127, 134, 402 Hemifusion, 18–19, 66–67, 76–77, 79–80, 147, 149, 282 Hemolysis, elevated liver enzymes and low platelets (HELLP), 102, 131, 400 Hepatocyte growth factor (HGF), 360, 368 Hepatocytes, 176, 243, 280, 291, 294, 297–298, 398, 403–404 Heptad repeats (HR), 18–20, 24–25, 31, 69–70, 72, 77, 124–125, 127 Herpes simplex virus, 102 HERV-FRD, 29, 82, 85, 87, 90, 121, 127, 211, 408–409, 413 HERV-H, 81, 411–412 HERV-K, 27, 30, 81–82, 408–413 HERV-P(b), 29, 409, 413 HERV-R, 89, 410, 412 HERV-W, 27, 29, 80, 82, 85–87, 91–94, 121, 211, 400, 405, 408–410, 412–413 Heterokaryons, 7, 66, 280, 301–302, 304–305, 307, 318–321, 332, 382, 399 Heterophilic cell-cell fusion, 243 Heterotypic fusions, 6–7, 66, 280 Hibris (Hbs), 143–144, 150, 152, 323 Histone acetylation, 130 HIV-1 co-receptors, 77 Hodgkin lymphoma, 405, 407 Homeostasis, 2, 221, 224, 295, 307, 404 Homotypic fusions, 7, 66, 225, 280 Homotypic interaction, 225 HSP70, 178 HSP90, 178–179 HSPC300, 155
Index Human endogenous retrovirus (HERV), 7, 14, 23, 26, 27–32, 44–45, 99, 101, 104, 121, 134, 269, 395–416 Human immunodeficiency virus (HIV), 5, 15–17, 19, 25, 43–45, 47–55, 69, 134, 225 Human immunodeficiency virus type 1 (HIV-1), 13, 15, 17, 19, 23–25, 53, 67, 71–74, 77–80, 94, 172, 177, 180, 402, 409 Human placenta, 87, 90, 95, 99, 119, 121, 178, 205–206, 208, 211–212 Human T-cell lymphotropic virus type 1 (HTLV-1), 23–25, 32, 73 Human Teratocarcinoma-Derived Virus, 81 Hyaluronidase 2 (HYAL2), 30–31, 68, 92–93, 97 Hybridomas, 7, 318, 324, 365–366 Hypoxia, 102, 131, 354 I IFN-γ, 101–102, 331, 340 IgSF8, 190, 192 IL-1β, 98, 133 IL-4, 101, 222, 236–237, 239, 241–242, 259–260, 268, 330 IL-10, 294, 306, 338, 360 IL-12, 329–330, 332–337, 339–340 IL-13, 259–260 Immunoglobulin superfamily (IgSF), 150, 190–192, 257 Immunological synapse, 164 Immunosuppressive, 24, 29–31, 70, 89, 102, 121–122, 134–135, 339 Imprinting, 98 Induced pluripotent stem, 302, 305 Inflammation, 229, 279, 291, 294, 306, 399, 404, 407 Inflammatory bowel disease, 294 Initiator caspase 8, 209 Inner acrosomal membrane (IAM), 188, 195 Insulin growth factor 1 (IGF-1), 259–260 β1-Integrin, 268, 283, 370 Integrin alpha3beta1, 176 Integrins, 53–54, 173, 175–176, 187, 192–194, 238, 254, 257–258, 268, 283–284, 340, 369–370, 383 Intercellular syncytial fusion, 212 Interference, 16–17, 21, 29, 67–69, 78, 84, 93, 99, 101, 126, 129, 133, 227, 271–272, 289 Intestinal epithelial stem cells, 291, 293 Intracytoplasmic sperm injection (ICSI), 181, 194
431 Invadopodia, 164 IZUMO, 5, 66, 174–175, 191–192, 196, 285 J Jaagsiekte sheep retrovirus (JSRV), 15, 23, 25, 30, 50, 72, 78, 83, 86, 92–93, 97 JAR cells, 130 JEG-3 cells, 129 JSRV receptor, 93 K Karyogamy, 397, 404, 414–416 Kette, 145, 153, 155–157, 159, 161 Ki67, 210, 294 Kirrel, 144, 288 Klf4, 305 Koala retrovirus (KoRV), 82 L Labyrinth, 68, 88, 96, 98, 119–120, 122, 128 Labyrinth zone, 96 LacZ, 252, 290 Lameduck (Lmd), 143 Lamellopodia, 254 LAMP-1, 357, 369–370, 383 Langhans’ giant cells, 398 Large extracellular loop (LEL), 175–176, 189–190 Lens, 401 Lentiviruses, 22–23, 25–26, 69, 72–73, 75, 82 Lentivirus lytic peptides (LLP), 73 Leukocyte-specific protein tyrosine kinase (Lck), 55 Leukocyte-tumor hybrids, 330, 360–361, 365, 369, 371, 399 LIN28, 305 Lineage commitment, 279, 290 Lipid rafts, 22, 52–53, 55, 74, 80, 187, 256 Lmd/Minc/Gfl, 143 Long Terminal Repeats (LTR), 4, 26, 69, 85–86, 94, 97–99, 120, 129–132, 408, 412 LoVo cells, 124 Lung cancer, 92, 406, 410 Lung carcinoma, 336, 406–407 Lymphocyte, 24–25, 44, 74, 81, 101, 178, 239, 319, 322, 332, 340, 366–367 M Macrophage activating factor (MAF), 236 Macrophage colony stimulating factor (M-CSF), 207, 220, 222–223, 236, 238, 242, 360 Macrophage fusion factor (MFF), 236
432 Macrophage fusion receptor (MFR), 5, 223, 226–228, 239 Macrophage giant cell (MGC), 235–237, 239, 240–244 Macrophage-melanoma hybrids, 357, 365–366, 368–371, 383 Macrophages, 5–7, 66–67, 74, 102, 180, 190, 206–207, 219–230, 233–244, 280, 297, 301–302, 306, 340, 354–357, 359–363, 365–383, 399, 404–405, 415 Major facilitator superfamily domain containing 2 (MFSD2), 30, 68, 93–94, 96, 123, 126–127, 129, 134, 269, 409, 413 Mammalian target of rapamycin (mTOR), 259, 261 Mannose receptor, 237, 239, 255, 259–260, 337 MARVEL domain, 162–163 Mason-Pfizer monkey virus (MPMV), 23–24, 48, 124 MCF-7 cells, 131, 329 Megakaryocytes, 398 Melanocortin-1 receptor (MC1R), 368, 383 Melanomas, 357, 359–361, 369, 371, 381, 405–406 Melanophages, 371, 376, 379–382 Membrane vesicles, 178–179, 187–189, 196, 213 Membrane vesiculation, 145–149, 151, 162–163, 187 Mesenchymal stem cells, 291 Metastasis, 103–104, 175, 229, 243–244, 308, 331, 333, 336, 338, 340–342, 353–355, 357–363, 365–366, 368–370, 383–385, 404–405, 412, 414–416 Metastatic phenotype, 353–354, 360, 365–383 Methylation, 3–4, 26, 94, 98–99, 104, 130, 173, 304, 408 Methyl-β-cyclodextrin, 52–53 MIC3, 175 Microdomains, 2–4, 52–53, 55, 75, 80, 190, 285 Microphthalmia-associated transcription factor (MITF), 222, 368, 383 Micropores, 325 Microvilli, 178–181, 195 Migration, 3, 6–7, 90, 120, 148, 155, 163, 191, 252–253, 255–256, 260, 289, 331, 354, 361, 363, 369–370, 384, 400–401, 406 Mind bomb 2 (Mib2), 143–144 Mitogen-activated protein kinases, 208 Mitotic catastrophy, 414 MMP9, 240–242
Index MN9, 5, 185, 194–195 Moesin, 4, 54, 190 Monoblasts, 221 Monocytes, 5, 194, 221, 243, 321, 331, 361, 367–368, 383 Mononuclear phagocyte system (MPS), 221 Monosialo ganglioside 3 (GM3), 178–179 Motility, 175, 190, 192, 195, 255, 259–260, 354, 357, 360–361, 367–370, 383 Motility-related protein 1 (MRP-1), 175 Mouse, 15, 23, 25, 27, 43, 45–46, 50, 67–69, 78, 81, 83, 86–89, 92, 94, 96–97, 119–120, 122, 128, 133–134, 144–145, 160, 176, 178–180, 188, 190–193, 195, 251–253, 255–257, 259, 283–284, 292–295, 297–298, 303–306, 327, 333, 355–358, 360, 366–367, 370, 401, 405, 408, 413, 416 placenta, 88, 96, 119, 122 Mucin 1 tumor-associated antigen (MUC1), 329–330, 332–333 Multinucleation, 221–229, 235–237, 239, 241–242, 244, 288, 397, 403, 404, 406, 414–415 Multiple sclerosis (MS), 80–81, 102–103, 133–134 Multiple Sclerosis associated RetroVirus (MSRV), 80–82, 101, 121, 409 Multivesicular bodies (MVB), 74–75, 177, 243 Murine leukemia viruse (MLV), 13, 15–17, 22–24, 31–32, 43, 45–52, 67, 69, 71, 73–75, 77, 80, 90–94, 124–125, 128 Muscle fibers, 6–7, 44, 49, 51, 55, 66, 205, 268, 286–287, 317, 403 Myelin, 102, 133 Myeloma, 6, 227, 318, 330, 333–334, 336, 340, 363, 365, 384, 407 Myoblast adhesion, 256 Myoblast alignment, 148, 254, 256 Myoblast city (Mbc), 144, 152, 155–157, 160–161 Myoblast elongation, 6, 254 Myoblast incompetent(Minc), 143 Myoblast migration, 155, 256, 260 Myoblast-myotube fusion, 259–261 Myoblast recognition, 157, 256 Myoblasts, 3, 6–7, 49, 66–67, 139–165, 209, 239, 249–273, 281, 286–289, 303, 400–401, 403 Myocardial-infarction, 298 Myoferilin, 261 Myofibers, 250, 253, 256, 259–261, 268
Index Myogenesis, 22, 49, 52, 161, 165, 254–256, 258, 261, 268 Myogenin, 250–251, 270 Myonuclear number, 252, 259–260 Myosin heavy chain, 251 Myostatin, 259–260 Myotubes, 6–7, 142–153, 155–160, 162–164, 208, 244, 251–252, 254–256, 258–261, 268, 270–272, 281, 289, 400–401 N N-acetylglucosaminyltransferase V (GnT-V), 357, 369 Nanog, 305 Nap1, 153, 155, 161, 258 Nascent myotube, 6–7, 159–160, 251, 255, 258–261, 289, 400 Neph1, 144, 160 Nephrin, 144, 160, 259, 261 N-ethylmaleimide-sensitive factor (NSF), 188–189 Neural stem cells, 279, 289–291, 304–305, 309 Neurotropin, 361 NF-κB, 98, 102, 222–223, 226, 329 Nitric oxide (NO), 102–103, 195 NO synthetase, 102 Notch, 142 Nuclear factor of activated T-cells (NFAT), 242, 259–260 Nuclear factor of activated T-cells c1 (NFATc1), 222, 242–243, 259 Nuclear factor of activated T-cells c2 (NFATc2), 259–260, 268 Nuclear reprogramming, 6, 290, 302–305 O Oct4, 302–303, 305 Old astrocytes specifically induced substance (OASIS), 102, 118, 134 Oligodendrocytes, 102, 133–134 Oligodeoxynucleotides containing CpG motif (CpG ODN), 326, 331, 335, 339–340 Oncoprotein, 92 Oocyte, 66, 175, 180, 187, 190–191, 193–196, 281, 283–286 OK-432, 331, 338, 340 Osteoblast, 236–237, 241, 244 Osteoclast differentiation factor (ODF), 237 Osteoclastogenesis, 224, 226–227, 237–238, 240, 242, 401 Osteoclastogenesis inhibitory factor (OCIF), 237 Osteoclasts, 3, 5, 221–223, 227, 236–239, 241–242, 244, 363, 367, 384, 404–405
433 Osteonectin, 354, 367 Osteopetrosis, 222, 224, 227, 238, 400 Osteoporosis, 5 Outer acrosomal membrane, 187–188, 196 P P21, 251 Parabiosis, 293 Parathyroid hormone, 237 Pathogen-associated molecular patterns (Pamps), 228 Perivitelline space (PVS), 178–179, 181, 191 Phagocytosis, 5, 78, 228–229, 237, 282, 327, 370, 379, 383–384 PH-dependent fusion, 78 PH-independent fusion, 78 Phosphatidylserine, 5–6, 100–101, 208–209 PI-3K/Akt signaling, 132 Placenta, 3, 5, 27, 29–31, 44–45, 66–68, 80, 83, 87–91, 94–104, 119–122, 127–132, 134–135, 165, 178, 190, 205–206, 208–209, 211–214, 269, 286, 296, 298, 307, 396, 400, 408–410 Placental protein, 213 Plasma membrane, 4–5, 15–16, 22, 25, 43, 51–52, 66, 69–70, 74–75, 89, 99–100, 143, 149, 155, 163–164, 177, 180, 187–196, 205, 208–211, 213, 224, 226–227, 229, 243, 256, 261, 281, 382 Podosomes, 164 Pol, 13–14, 27–28, 69, 86–87, 97, 120, 408–412 Polarisation of budding, 74 Polyethylene glycol, 318, 324, 332, 356, 365 Polyploidy, 291, 396–398, 414 Polytropic, 15–17, 45, 48 Pore model, 79 Preeclampsia (PE), 102, 131–132, 134– 135, 206 Progesterone, 68, 98, 133, 409 Progesterone receptor, 98, 133 Prognostic impact, 103, 133 Programmed cell death, 5 Proliferation, 49, 68, 89, 94, 103–104, 119, 129, 190, 210, 280, 289, 293–294, 306, 328, 368, 371–373, 407, 412, 415 Promoter, 69, 87, 94, 97–99, 129–132, 153, 237, 242, 292–293, 302–303, 327, 408 Proprotein convertase, 123 Prostaglandin F2α , 259–260 Proteasome, 129 Protein kinase A (PKA), 3, 5, 98, 121, 129, 132, 134, 207–208, 254, 272, 409
434 Protein scaffold, 4–5 Protein tyrosine kinase, 55 Provirus, 13, 22, 26, 69, 81, 85–86, 408–412 Pseudotyping, 91, 93–94, 409 PSG17, 190 PTGFRN, 192 PTP1B, 189 R Rab, 188 Rabbit, 3, 31, 83–87, 94, 96–97, 122, 134, 235–236, 269, 322 Rac, 55, 152, 155, 157, 160–161, 258 Radixin, 4, 54, 190 Rafts, 2, 4, 18, 22, 52–53, 55, 74, 80, 96, 100, 187, 256 RANK ligand (RANKL), 5, 221–222, 227, 237–238, 242 RD114 virus, 78, 82 Receptor activator of NFκ B (RANK), 5, 221–223, 229, 237–238 Receptor-binding domain (RBD), 24, 93, 126 Receptor blocking, 92 Receptor interference, 16, 21, 29, 68–69, 126 Recurrence-free survival, 103, 133 Redox mechanims, 102 Reduction division, 293, 301 Regenerating myofibers, 259 Regeneration, 66, 250, 253, 257, 259, 268, 279, 281, 289–291, 293–299, 303, 305–309, 384, 407 Regulatory T cells (Treg), 338–339 Renal cancer, 338 Reproduction, 3, 174, 409 Reprogramming, 6, 290, 292, 301–305, 402, 415–416 Retrotransposons, 82, 98, 103, 408 Retroviridae, 22–23, 69 Retrovirus, 2–3, 11–33, 43–51, 55, 67–69, 71–75, 77–83, 85–93, 97, 99, 101, 103–104, 120–122, 124, 126, 128, 131, 395–416 Retrovirus D-type receptor (RDR), 92–93, 99, 126 Reverse transcriptase (RT), 69, 80, 408 Rho, 52, 55, 398, 406 Rho-associated, coiled-coil containing protein kinase (ROCK), 55, 398 Rolling pebbles (Rols), 145, 151–152, 157, 162 Roughest/Irregular optic Chiasma (Rst/IrreC), 140, 143–144, 150–153 R-peptide, 21–22, 24, 29, 43, 48–50, 70, 75–76, 90–91, 124
Index S Satellite cells, 7, 165, 268 Scar/Wave, 154–157 Schizo/Loner, 145, 152, 155, 157 Schneider cells, 150 Secreted protein acidic and rich in cysteine, osteonectin, BM40 (SPARC), 354, 367–368, 383 SEL, 175, 189 Self recognition, 5, 228 Self-tolerance, 229 Serine proteases, 123, 255 SH2 domain containing phosphatase-2 (SHP-2), 259–260, 268 Sheep3, 15, 23, 25, 30–31, 50, 67–68, 83, 85–86, 88–89, 92, 94, 97, 103, 122 Signal regulatory protein-alpha (SIRP α), 5, 226–229 Simian retrovirus (SRV), 69, 92, 78, 101 Singles bar, 144, 162 Skeletal muscle, 3, 6–7, 44, 66, 93, 100, 159–160, 165, 205, 208–209, 250, 256, 259–260, 268–269, 281, 286–287, 297–298, 307, 403–404, 415 Snail, 354, 360–361, 367–368 SNAP receptors (SNARE), 4–5, 66–67, 187–189, 196 Soluble NSF attachment protein (SNAP), 188–189 Sox2, 305 Sp-1, 98 Sperm, 5–6, 17, 171–181, 187–196, 205, 239, 244, 281–286, 399 Spermatid precursors, 401 Spermatogonia, 401 Sperm-egg fusion, 5–6, 171–181, 190, 192–193, 195, 239, 282, 285 Sperm factor, 194 Sperm-oocyte adhesion, 283 Spontaneous hybrid formation, 7, 355 Spumavirus, 22–23, 25–26, 69 Sra-1/PIR121, 155 Stem cell, 6, 66, 94, 96, 119, 129, 165, 176, 212, 221, 223, 228–229, 243, 250, 277–309, 317, 355–356, 361, 363–365, 398, 401–403, 407, 415–416 Stem cell fusion, 279–282, 285–295, 297, 299–303, 305–307, 309 Sticks and stones (Sns), 143–144, 150–153, 156–157, 160, 163–164, 261 Sumoylation, 130 Superinfection, 16 Superinfection resistance, 16
Index Surface subunit, 15 Surface unit (SU), 15, 17–18, 20–21, 23–25, 32, 43–44, 46–47, 49–50, 69–73, 75–76, 79, 90, 93, 123–127, 132, 272, 409 Syk, 238, 240, 242 Synapse, 4, 74, 164 Synaptobrevin, 187, 189 Synaptotagmin I, 189 Syncytial plaques, 97 Syncytin, 3–5, 7, 29–30, 44–45, 48, 63–104, 121, 123–127, 129, 131, 133–135, 165, 178, 211–212, 214, 225, 267–273 Syncytin-1, 3–4, 6, 31–32, 68, 80, 82–104, 120–134, 211–212, 225, 268–272, 400, 405, 408–410, 412–414 Syncytin-1 transactivation, 129 Syncytin-2 receptor, 4, 27, 29–30, 83, 85–87, 89–93, 95–97, 99, 101–102, 121–122, 124–132, 134, 211, 408–409, 413 Syncytin-A, 3, 68, 86–88, 90, 93, 96–98, 122, 125, 129, 134, 269 Syncytin-B, 68, 87, 93, 96–97, 122, 129, 134 Syncytin-A and -B, 86, 90, 93, 122, 269 Syncytin-Ory1, 68, 83–84, 86–87, 90, 92, 96–97, 122, 269 Syncytiotrophoblats, 96 Syncytium, 21, 43–44, 47–49, 51–55, 89, 119, 159, 205–206, 208–210, 213, 289, 307, 401 Synkaryons, 7, 280, 399 T TATA box, 98 T-cell activation, 318, 331 T-cell tolerance, 330 Teratomas, 406 Testis, 87, 93, 97, 187, 191–193, 285, 410–411 Tetraploid, 296, 303, 397–398, 402, 414 Tetraspanins (Tspan), 4, 66, 173, 175–178, 180, 187, 189–194, 196, 239, 284–286 Three-cell cluster, 341 Thrombspondin-1(TSP-1), 227 Time-lapse microscopy, 150, 252, 254, 287, 291 Toll-like receptor agonists, 331 Toll-like receptors (TLR), 228, 331 Transdifferentiation, 279, 296, 298, 302, 402–403 Transforming growth factor-β (TGF- β), 103–104, 206–207, 222, 338–339, 352, 360, 412 Trans-Golgi network(TGN), 73
435 Translocase, 100, 208 Transmembrane domain (TM), 24, 31, 45, 66–67, 69, 71, 75, 79–80, 84, 91, 93, 123, 127, 175, 189, 284 Transmembrane subunit (TM), 15, 18, 20–21, 23–25, 29, 31–32, 43–44, 46–50, 69–76, 79–80, 84, 90, 92, 102, 123–125, 127, 134, 143, 272, 409 Transplantation, 279, 290–293, 296–297, 300, 306, 407 Trogocytosis, 178, 191 Trophoblast, 3, 5, 30–31, 68, 86, 88–90, 94, 96–98, 100, 120, 122, 128–129, 131, 203–214, 269–270, 286, 307, 398, 400, 404–405 Trophoblast binucleated cells, 97 Tryptophan, 27, 71, 79, 101 T-SNAREs, 4–5, 189 Tspan-29, 175 Tumor associated macrophages, 302, 359–363, 404–405 Tumor cell fusion, 300, 329, 353–355, 359, 363, 368, 371, 383 Tumor-host hybrids, 355 Tumor-initiating cells, 300 Tumor microenvironment, 338, 360 Tumor necrosis factor (TNF-α), 102, 206–207, 222, 237, 329, 340 Tumor stem cells, 229, 302 Tumor suppressor genes, 6, 133, 293, 365 Tumor suppressors, 414 Tumor x tumor hybrids, 355 Tunicamycin, 49, 124 Type D mammalian retrovirus receptor, 126 Type D receptor, 67, 78, 84, 92, 99, 124, 126, 212 Tyrosinase, 327, 357–358 Tyrosine kinase, 55, 221–222, 272 U Ubiquitin, 129–130, 132, 143 V Vacuolar adenosine triphosphatase (v-ATPase), 241 VCA, 54, 154–155 VCAM-1, 54, 298 Vesicle associated membrane protein (VAMP), 189 Vesicles, 5, 66, 69, 73–75, 146–148, 156–157, 159–165, 177–181, 187–189, 191, 196, 205, 210, 213, 243, 256–258, 284, 357, 371, 377–383
436 Vesicle-SNARE (v-SNAREs), 189 Vesicular stomatitis virus, 326 Villous trophoblast, 205–206, 208–213, 400 Viral FMG, 326 Virologic synapse, 74 Virus-like particles, 27, 80–81, 402 Virus receptor types, 92 Vitamin D, 237 Vrp1/sltr, 145, 155–158 W WASP family Verprolin-homologous protein (WAVE), 145, 154–157, 161, 258 WASP-interacting partner (Wip), 145, 154–160 Wheat germ agglutinin (WGA), 259
Index Wiscott-Aldrich syndrome protein, (WASP), 54, 145, 154–160, 258 Wnt, 294, 407 X XC cells, 22, 49–50, 85 Xenotropic, 15–16, 45, 48 Xenotropic/polytropic receptor, 45 Xist, 304 Y Y-chromosome, 291, 298 Z Z-discs, 152 Zebrafish, 4, 141, 144–145, 160, 288 Zona pellucida, 173, 178–179, 181, 187–188, 193, 196, 281