Advances in Immunology Volume 75

ADVANCES IN

Immunology VOLUME 75

This Page Intentionally Left Blank

ADVANCES IN

Immunology EDITED BY FRANK J. DIXON The Scripps Research Institute La Jolla, California ASSOCIATE EDITORS

Frederick Alt K. Frank Austen Tadamitsu Kishimoto Fritz Melchers Jonathan W. Uhr Emil R. Unanue

VOLUME 75

San Diego San Francisco New York Boston London Sydney Tokyo

CONTENTS

CONTRIBUTORS

ix

Exploiting the Immune System: Toward New Vaccines against Intracellular Bacteria

JU¨RGEN HESS, ULRICH SCHAIBLE, BA¨RBEL RAUPACH, AND STEFAN H. E. KAUFMANN I. II. III. IV.

Introduction General Comments Infectious Diseases Caused by Intracellular Bacterial Pathogens Intracellular Bacteria, Their Niches in the Host Cell, and the Immune Responses Elicited V. Vaccines VI. From Genomes to Antigens VII. Concluding Remarks and Outlook References

1 2 7 12 26 59 61 63

The Cytoskeleton in Lymphocyte Signaling

A. BAUCH, F. W. ALT, G. R. CRABTREE, AND S. B. SNAPPER I. Introduction II. Overview III. Current View of the Regulation of Actin Polymerization in Cytoskeletal Rearrangements IV. The Vav Family of Guanine Nucleotide Exchange Factors V. The Regulation of Actin by Rac VI. WASP and the Wiskott–Aldrich Syndrome VII. WASP and the Control of Actin VIII. Shared Aspects in the Phenotype of WASP- and Vav-Deficient Mice IX. Actin Polymerization and the Propagation of Signals to the Nucleus References

v

89 90 94 95 98 100 104 107 108 110

vi

CONTENTS

TGF-웁 Signaling by Smad Proteins

KOHEI MIYAZONO, PETER TEN DIJKE, AND CARL-HENRIK HELDIN I. II. III. IV. V. VI. VII. VIII. IX. X. XI.

Abstract The TGF-웁 Superfamily Serine/Threonine Kinase Receptors Structure and Function of Smads Cytoplasmic Actions of Smads Actions of Smads in the Nucleus I-Smads Signaling Cross-Talk Roles of Smads in Human Cancer In Vivo Functions of Smads: Analyses by Gene Targeting Perspectives/Conclusion References

115 115 116 119 126 128 135 137 140 142 143 145

MHC Class II-Restricted Antigen Processing and Presentation

JEAN PIETERS I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII.

Introduction The MHC Class I and Class II Pathways Structure of MHC Class II Complexes Biosynthesis and Assembly of MHC Class II/Invariant Chain Complexes Entry of MHC Class II/Invariant Chain Complexes into the Endocytic Pathway Subcellular Organelles Involved in Antigenic Peptide Loading onto Class II Molecules: The MHC Class II Compartments Transport of MHC Class II Complexes to the Cell Surface Antigen Internalization Antigen Processing The Biology of a Prime Antigen-Presenting Cell: The Dendritic Cell System Subversion of MHC Class II-Restricted Antigen Presentation by Pathogens Conclusions References

159 160 162 171 173 176 180 181 185 187 189 191 192

T-Cell Receptor Crossreactivity and Autoimmune Disease

HARVEY CANTOR I. Introduction II. Crossreactivity and T-Cell Activation III. Limits of TCR Crossreactivity: Peripheral Purging of Useless T-Cells

209 209 218

CONTENTS

IV. TCR Crossreactivity and Autoimmune Disease: Viral-Derived Peptide Agonists V. Cytokine Checkpoints in the Generation of Autoimmune Disease References

vii 220 223 226

Strategies for Immunotherapy of Cancer

CORNELIS J. M. MELIEF, RENE´ E. M. TOES, JAN PAUL MEDEMA, SJOERD H. VAN DER BURG, FERRY OSSENDORP, AND RIENK OFFRINGA I. Introduction II. Natural Protective Immunity against Cancer III. Antigens Eliciting T Cell Responses Expressed by Virus-Associated Tumors IV. Antigens Eliciting T Cell Responses by Non-Virus-Induced Tumors V. Processing of Tumor Antigens VI. Pivotal Role of Dendritic Cells and Tumor-Specific CD4⫹ Helper Cells in Tumor Immunity VII. Fine Tuning of T Cell Responses by TNF(-R) Family Members VIII. Escape Mechanisms of Tumors IX. Cancer Therapy by Adoptive Transfer of T Cells X. Design of Rational Cancer Vaccines Including Molecularly Defined Adjuvants XI. Tumor Immunotherapy Based on Improved Costimulation via the CD28 Pathway XII. Enhancement of Tumor-Specific T Cell Responses by Cytokines and by Cytokine-Transduced Tumor Cells XIII. Monitoring of Tumor-Specific T Cell Responses XIV. Immunotherapy with Monoclonal Antibodies XV. Epilogue References

235 238 240 241 246 247 249 253 256 258 261 262 263 263 264 264

Tyrosine Kinase Activation in the Decision between Growth, Differentiation, and Death Responses Initiated from the B Cell Antigen Receptor

ROBERT C. HSUEH AND RICHARD H. SCHEUERMANN I. Abstract II. Introduction III. B Cell Antigen Receptor Signaling Controls Several Stages of B Lymphocyte Development IV. Mechanisms Governing Different Cellular Responses from the Same Receptor V. Ig움 and Ig웁 Coreceptors in BCR Signaling VI. The BCR-Associated Protein Tyrosine Kinases VII. The Connections between Tyrosine Kinase Activation and Cellular Responses VIII. Concluding Remarks References

283 283 284 288 292 295 304 307 308

viii

CONTENTS

The 3⬘ IgH Regulatory Region: A Complex Structure in a Search for a Function

AHMED AMINE KHAMLICHI, ERIC PINAUD, CATHERINE DECOURT, CHRISTINE CHAUVEAU, AND MICHEL COGNE´ I. II. III. IV. V. VI. VII. VIII.

Introduction Structure of the 3⬘ IgH Control Region Activity during B Cell Development Synergies between 3⬘ Enhancers DNA-Binding Proteins In a Search for a Function The Case of Humanized Mice Conclusion References

INDEX CONTENTS OF RECENT VOLUMES

317 317 319 320 323 330 337 337 338 347 359

CONTRIBUTORS

Numbers in parentheses indicate the pages on which the authors’ contributions begin.

Frederick W. Alt (89), The Children’s Hospital, Center for Blood Research, Department of Genetics, Harvard Medical School, Boston, Massachusetts 02115 Angela Bauch (89), Department of Pathology and Developmental Biology, Beckman Center for Molecular and Genetic Medicine, Stanford University School of Medicine, Stanford, California 94305 Harvey Cantor (209), Department of Pathology, Harvard Medical School, and Department of Cancer Immunology and AIDS, Dana-Farber Cancer Institute, Boston, Massachusetts 02115 Christine Chaveau (317), Laboratoire d’Immunologie et Institut Universitaire de France, Faculte´ de Me´decine, Limoges Cedex 87025, France Michel Cogne´ (317), Laboratoire d’Immunologie et Institut Universitaire de France, Faculte´ de Me´decine, Limoges Cedex 87025, France G. R. Crabtree (89), Department of Pathology and Developmental Biology, Beckman Center for Molecular and Genetic Medicine, Stanford University School of Medicine, Stanford, California 94305 Catherine Decourt (317), Laboratoire d’Immunologie et Institut Universitaire de France, Faculte´ de Me´decine, Limoges Cedex 87025, France Peter ten Dijke (115), The Netherlands Cancer Institute, 1066 CX Amsterdam, The Netherlands Carl-Henrik Heldin (115), Ludwig Institute for Cancer Research, S-751 24 Uppsala, Sweden Ju¨rgen Hess (1), Department of Immunology, Max-Planck-Institute for Infection Biology, D-10117 Berlin, Germany Robert C. Hsueh (283), Department of Pathology, University of Texas Southwestern Medical Center, Dallas, Texas 75235 Stefan H. E. Kaufman (1), Department of Immunology, Max-PlanckInstitute for Infection Biology, D-10117 Berlin, Germany ix

x

CONTRIBUTORS

Ahmed Amine Khamlichi (317), Laboratoire d’Immunologie et Institut Universitaire de France, Faculte´ de Me´decine, Limoges Cedex 87025, France Rienk Offringa (235), Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, RC Leiden 2300, The Netherlands Ferry Ossendorp (235), Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, RC Leiden 2300, The Netherlands Jan Paul Medema (235), Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, RC Leiden 2300, The Netherlands Cornelis J. M. Melief (235), Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, RC Leiden 2300, The Netherlands Kohei Miyazono (115), Department of Biochemistry, The Cancer Institute, Japanese Foundation for Cancer Research, Tokyo 170-8455, Japan Jean Pieters (159), Basel Institute for Immunology, Basel CH-4005, Switzerland Eric Pinaud (317), Laboratoire d’Immunologie et Institut Universitaire de France, Faculte´ de Me´decine, Limoges Cedex 87025, France Ba¨rbek Raupach (1), Department of Immunology, Max-Planck-Institute for Infection Biology, D-10117 Berlin, Germany Ulrich Schaible (1), Department of Immunology, Max-Planck-Institute for Infection Biology, D-10117 Berlin, Germany Richard H. Scheuermann (283), Department of Pathology, University of Texas Southwestern Medical Center, Dallas, Texas 75235 Scott B. Snapper (89), Gastrointestinal Unit and the Center for the Study of Inflammatory Bowel Disease, Massachusetts General Hospital, and Department of Medicine, Harvard Medical School, Boston, Massachusetts 02115 Rene´ E. M. Toes (235), Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, RC Leiden 2300, The Netherlands Sjoerd H. van der Burg (235), Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, RC Leiden 2300, The Netherlands

ADVANCES IN IMMUNOLOGY, VOL. 75

Exploiting the Immune System: Toward New Vaccines against Intracellular Bacteria ¨ RBEL RAUPACH, AND STEFAN H. E. KAUFMANN ¨ RGEN HESS, ULRICH SCHAIBLE, BA JU Department of Immunology, Max-Planck-Institute for Infection Biology, D-10117 Berlin, Germany

I. Introduction

Vaccination can be described as the induction or modulation of an antigen-specific immune response to prevent or cure the targeted disease. We consider the qualification ‘‘antigen-specific’’ essential for distinguishing vaccination from nonspecific immune modulation. Although we will restrict our general comments to infectious agents, with an emphasis on intracellular bacteria, the scope of future vaccination strategies is much broader, including cancer, autoimmune diseases, and allergic diseases. In other words, the discipline that deals with these issues, often referred to as vaccinology, is based on and intimately linked with immunology. Of equal importance, however, is an understanding of the microbiology of the disease-causing agents. Vaccination was born more than 200 years ago at a time when immunology was unknown and the concept of disease transmission by microbes was not generally accepted. On June 21, 1798, Edward Jenner reported the successful vaccination against smallpox using cowpox material ( Jenner, 1798). As we now know, the cowpox virus is of low virulence for humans but shares sufficient cross-reactivity with the highly virulent smallpox virus to induce protective immunity. This strategy of applying attenuated viable microbes as vaccines was further pursued by Louis Pasteur, who set about attenuating microbes in the laboratory. He successfully attenuated anthrax bacilli by means of in vitro passages and rabies virus by drying off the infected nerve tissue. It was also Pasteur who, in honor of Jenner, generalized the term ‘‘vaccination’’ from the use of cowpox (vaccinia) to include similar strategies for controlling infectious diseases. Pasteur’s emphasis was on microbial agents, with less attention paid to the host immune response. Emil von Behring and Paul Ehrlich were among the first to integrate microbiology and immunology equally into their strategies for vaccination. They successfully applied passive vaccination with standardized antisera to the control of toxin-producing bacteria, such as Clostridium tetani and Corynebacterium diphtheriae. The finetuned standardization of the toxin-neutralizing activity of these antisera illustrated how much vaccinology could learn from basic immunology. 1

Copyright 䉷 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2776/00 $35.00

2

HESS et al.

However, soon thereafter the two disciplines diverged, with immunology becoming mostly concerned with academic issues and vaccinology becoming increasingly empirical. Not surprisingly, vaccinology focused on the beneficial outcome of an immune response, whereas immunology primarily attempted to dissect and understand the mechanisms underlying immunity. Accordingly, the terms vaccination and immunization should not be thought of as interchangeable. Vaccination is restricted to those immunization protocols that provide a definite degree of protection against the targeted pathogen or, at least, against disease (Table I). The success of empirically developed vaccines is without precedent in medicine. They are not only the most successful application of immunology but also one of the most cost-efficient measures in medicine. It has been estimated that at least 8 million lives a year are saved by the application of currently available vaccines. This success story is perhaps best illustrated by the WHO’s declaration on May 8, 1980, that smallpox had been eradicated. However, the empirical approach has reached its limits and it is now time to include the recent advances in immunology and microbiology in the rational design of future vaccines. We will focus on intracellular bacterial pathogens of medical importance, notably Mycobacterium tuberculosis, Salmonella spp., and Chlamydia spp. In addition, we will consider how intracellular bacteria can be used as vaccine carriers for heterologous antigens, particularly attenuated Salmonella and Listeria strains, as well as Mycobacterium bovis bacille Calmette Gue´rin (BCG). These strains not only represent suitable recombinant (r)carriers for protein antigens but also are potential delivery systems for naked DNA constructs. II. General Comments

Before going into the details of the immunologic basis of vaccination against infectious diseases caused by intracellular bacteria, we will provide the reader with some basic vaccinology information relevant to the subsequent discussion. TABLE I REQUIREMENTS FOR AN EFFECTIVE VACCINE AGAINST INTRACELLULAR BACTERIA Appropriate milieu for Th1 cells Antigen processing through adequate MHC pathway and perhaps MHC-like pathways Mucosal and systemic immunity Memory

EXPLOITING THE IMMUNE SYSTEM

3

A. SOME CONNOTATIONS: INFECTION AND DISEASE, PATHOGENICITY AND VIRULENCE Although lay discussions do not often distinguish between ‘‘infection’’ and ‘‘infectious disease’’, it is appropriate to differentiate between these two terms. Intracellular bacterial pathogens frequently establish infection without inevitably causing disease. ‘‘Disease’’ is understood to mean signs of detriment to the host, often reflecting damage to host cells and tissues (Kolle and von Wassermann, 1912; Zinsser and Bayne-Jones, 1934; Casadevall and Pirofski, 1999). Correspondingly, infection is defined as the outcome of an interaction between pathogen and host that results in a measurable host response, that eventually, but not necessarily, causes overt pathologic sequelae. In this context, infection describes a limited tissue reaction, such as a measurable immune response, which is harmless and may or may not prevent the microbial agent from causing disease. In contrast, commensals colonize host surfaces without causing a measurable tissue response. The transition from infection to disease should be viewed as a continuum: as long as tissue responses do not manifest any harm to the host, they are considered to constitute infection, while we categorize them under the term disease once harmful sequelae are noted. It does not matter whether the damage is caused directly by the microbe or by the host immune system. In fact, many of the diseases dealt with here, notably trachoma and tuberculosis (TB), have a marked immunopathological component. The microbial species that are capable of infecting a given host with the possible outcome of disease are described as pathogenic. If it becomes necessary to quantify the degree of damage, we prefer to use the term ‘‘virulent,’’ which is considered a quantitative measure of pathogenicity within different members of a species. Although the distinction between virulence and pathogenicity may be important in certain settings, we will not adhere strictly to this distinction in the context of vaccine development and will use the two terms interchangeably. The generation of attenuated viable vaccine strains provides the most intriguing example that illustrates these issues. Attenuation results in a vaccine strain that still causes an immune response but which has lost the capacity to cause disease in an immunocompetent host. Attenuated vaccine strains retain the inherent risk of causing disease in severely immunodeficient individuals, underlining the mutual impact of pathogen and host immune response on virulence. To provide a specific example, the anti-TB vaccine BCG is an attenuated strain of M. bovis, a pathogenic species for cattle and human. It can safely be given to newborns, but despite its highly reduced virulence, BCG may cause disease in certain immunodeficient individuals.

4

HESS et al.

B. STERILE PATHOGEN ERADICATION PRIOR TO DISEASE DEVELOPMENT AS THE IDEAL GOAL OF VACCINATION As has been stated, active vaccination should prevent the targeted disease. For infectious diseases, this is most reliably achieved by sterile eradication of the pathogen. This is particularly important for two reasons, the first of which is to avoid any later reactivation of the disease by dormant microorganisms. In the case of M. tuberculosis, the reactivation of dormant bacteria, often decades after primary infection, is the major cause of pulmonary TB in adults. The currently available TB vaccine, BCG, can prevent rapid development of TB in children but fails to obviate the reactivation of persistent bacteria. The second reason is to prevent spreading of the pathogen to the population from an apparently healthy individual. Carriers of Salmonella typhi harbor the pathogen in secluded sites such as the gall bladder and, despite being healthy themselves, can spread the pathogen to the population through fecal contamination. The most famous carrier was Mary Mallon, or ‘‘Typhoid Mary,’’ who when working as a cook around the turn of the century was responsible for at least 47 typhoid cases, including 3 deaths. However, several vaccines in use today fail to achieve sterile pathogen eradication despite having successfully proven their efficacy. These include widely used viral and bacterial vaccines, such as those against measles, mumps, rubella, as well as those against diphtheria, pertussis, and tetanus. C. PREVENTIVE AND PREINFECTION VS THERAPEUTIC AND POSTINFECTION VACCINATION Louis Pasteur and Emil von Behring clearly considered the therapy of infectious diseases an important aspect of vaccination. The rabies vaccine of Louis Pasteur, as well as the tetanus and diphtheria vaccines of Behring, was aimed at preventing the disease in individuals who were already infected. The rationale underlying Pasteur’s vaccination strategy was to speed up a slow-developing immune response by active vaccination, whereas Behring’s toxoid vaccines were based on passive vaccination with specific antibodies raised in animals. However, in later times the concept of therapeutic vaccination became less attractive, and vaccination was considered to be primarily preventive, i.e., vaccination was used to induce a primary immune response in naive individuals. It is only recently, mostly through the advent of AIDS (acquired immunodeficiency syndrome), that active therapeutic vaccination has been reconsidered as an important strategy for controlling infectious diseases. Very closely related to this discussion are the terms ‘‘preinfection’’ and ‘‘postinfection’’ vaccination. We define infection as a measurable outcome

EXPLOITING THE IMMUNE SYSTEM

5

of the interplay between pathogenic microbe and host that may, but need not, cause disease, and we consider therapy in a broader sense, as a measure not only of curing disease but also of interfering with infection. Accordingly, ‘‘preventive’’ and ‘‘preinfection’’ vaccines, and ‘‘therapeutic’’ and ‘‘postinfection’’ vaccines, will be considered as more or less synonymous. Importantly, vaccines may induce different immune reactions against a single disease when given pre- or postinfection. Ideally, preinfection vaccines would prevent infection, i.e., the stable establishment of the pathogen in the host. Similarly, it would be best if a postinfection vaccine were able to eradicate the infectious agent. Since the ideal goal may often not be attained, one should be satisfied with the second-best objective, namely, prevention of disease. This aim may be achieved by reducing the microbial burden to such a low level that the immune system can contain the microbes, thus preventing them from causing clinical symptoms. However, pathogen restraint is a dynamic process between the microbe and the immune response that will be kept in balance only as long as the immune system is fully competent. Any deficiency in the immune system can tip the balance in favor of the pathogen and allow outbreak of disease. This is exactly what happens during reactivation of TB. While the chance of developing TB from dormant bacilli during one’s life is less than 10%, it reaches a level of about 10% per year in human immunodeficiency virus (HIV)-infected individuals. Most experimental approaches are aimed at determining the efficacy of potential vaccines given preinfection, and only few studies have considered the postinfection approach. Although adult TB, for example, is primarily a reactivation disease requiring vaccination postinfection, this has only rarely been considered in the design of experimental protocols. Such protocols were already designed in the 1950s (McCune et al., 1957) and are now being revitalized (Lowrie et al., 1999). They use vaccination of M. tuberculosisinfected animals harboring low doses of microorganisms as a model for postinfection vaccination against reactivation disease. D. EXPERIMENTAL ANIMAL MODELS FOR DETERMINING VACCINE EFFICACY Experimental animal models exist that can be used to study the specific questions discussed above. However, the necessary critical judgment is sometimes lacking in the application of these models. The two most widely used models for determining vaccine efficacy are prevention of fatality and reduction of bacterial load in organs. The former experimental system is based on challenge infections with inocula well above the dose that is lethal for 50% of naive animals (in short, LD50), with the purpose of revealing whether the vaccine can prevent fatal disease. Although this type

6

HESS et al.

of experiment provides impressive data and involves relatively little work for the experimenter, it reflects a mode of infection that rarely occurs in nature. In TB, for example, a minute number of microorganisms establish an infection, and high numbers are seen only when overt disease has already manifested itself. The alternative approach uses sublethal inocula and determination of colony-forming units in target organs. This is a more laborious approach, but it is quantitative and allows precise assessment of the reduction of the bacterial load and determination of the time point of sterile pathogen eradication. The major drawback of this system lies in the interpretation made by the experimenter. Small reductions in the bacterial load, although statistically significant, may reveal an immunologically interesting phenomenon that does not necessarily reflect a satisfactory outcome of vaccination. This lowered, but still profound, bacterial burden may increase again and cause disease at a later time. It is therefore strongly recommended that terms such as ‘‘partial protection’’ be used in such situations. As a third experimental approach to measuring vaccine efficacy, the severity of disease can be determined in the experimental animal. However, this kind of information is often ignored and, if it is considered, is frequently restricted to histology of autopsy material. Only rarely are disease symptoms determined in the living animal, although several parameters, including weight loss and stool frequency and solidity, as well as liver enzyme activities such as serum transaminase activity, can be easily ascertained without harm to the animal. As difficult as it may already be to develop a vaccination protocol that achieves satisfactory protection, a vaccine applicable for human use needs to fulfill further requirements. They include all aspects of safety and side effects, heat stability and long shelf life, inexpensive production, and compatibility with existing vaccination and diagnosis schedules. Although these are critical prerequisites for approval for use in humans, we will not dwell on them. E. THE CENTRAL ROLE OF T CELLS IN PROTECTION AGAINST INTRACELLULAR BACTERIA: ANY ROLE FOR ANTIBODIES? It is generally accepted that T cells are central to acquired resistance against intracellular bacteria. Although antibodies are formed against these pathogens, they are of negligible importance. This clearly holds true for TB and listeriosis, and may be questioned only for typhoid fever. It is therefore reasonable to assume that vaccines against TB and related infectious diseases should emphasize activation of T lymphocytes as effectors of vaccine-induced protection. Unfortunately, there is no precedent for demonstrating that such a vaccine is feasible. All successful vaccines cur-

EXPLOITING THE IMMUNE SYSTEM

7

rently in use rely on high levels of circulating antibodies preexisting at the time of infection. These antibodies prevent infection by blocking adhesion or invasion, or they preclude detrimental damage by neutralizing toxic molecules. At the time of pathogen entry, the circulating antibodies are responsible for protection, and reactivation of memory B cells to produce additional antibodies is of little importance. Because antibodies directly interfere with the biological activities of molecules involved in infection or tissue damage, their target antigens are defined by their functions in these processes, e.g., adhesion and invasion. In contrast to antibodies, T cells do not act on infectious agents directly and recognize antigens only after processing. This means that the pathogens have already infected host cells, typically macrophages, which serve as their major habitat. Consequently, whether certain functions predestine antigens as preferred targets for T lymphocytes remains to be established. In the following, we will consider T cells as major targets for vaccines against intracellular bacteria and follow the concept that full protection is best achieved by a combination of different T cell populations that interact in a tightly controlled network. Because distinct activation requirements exist for different T cell populations, a vaccine needs to introduce the antigen to these different T cell sets in appropriate ways. Moreover, it is important to consider the different T cell combinations to be activated by preinfection versus postinfection vaccines. Finally, we will speculate on the possibility that antibodies also play a role in prompt prevention of infection with intracellular bacteria. III. Infectious Diseases Caused by Intracellular Bacterial Pathogens

A. TUBERCULOSIS Mycobacterium tuberculosis, the causative agent of TB, was discovered by Robert Koch in 1882. Tuberculosis remains one of the major killers of humankind. While approximately 95% of TB cases occur in developing countries, several states of Eastern Europe, including Russia, have been witnessing increasing incidences of TB. Global mortality ranges between 1.6 and 2.2 million lives per year, depending on whether the half million individuals suffering co-infection with HIV and M. tuberculosis are included in the toll of AIDS or TB (World Health Organization, 1999). The situation is further worsened by increasing incidences of multidrug resistant strains, which is due in part to incomplete compliance with chemotherapy. In several states, e.g., Estonia and the Dominican Republic, as many as 10–20% of all mycobacterial isolates are of the multidrug-resistant type. Each year, 55 million people become newly infected with the pathogen, resulting in as many as 2 billion infected people worldwide. Of these

8

HESS et al.

infected individuals, 8 million annually will develop disease. Thus, fewer than 10% of infected people will develop TB during their lifetime, although they are all at risk of doing so. This low ratio of diseased to infected individuals indicates that the immune system can control M. tuberculosis efficiently, as long as it remains competent. In contrast, immunodeficiency will markedly increase the risk of active disease to about 10% per year. The lung represents the port of entry and primary organ of disease manifestation. At the site of mycobacterial implantation, granulomas are formed and the microbes are contained efficaciously in immunocompetent persons. However, weakening of the immune response results in reactivation of pathogens, transforming infection into active disease. Infection of immunocompromised individuals, including newborns, results in microbial dissemination, leading directly to miliary TB. M. tuberculosis is shielded by a wax-rich cell wall composed of various glycolipids, notably, lipoarabinomannan and mycolic acids (Brennan and Nikaido, 1995). This cell wall is responsible for numerous unique features of the pathogen, including acid-fastness, adjuvanticity, and resistance to acidic or alkaline solutions, as well as to disinfectants. It also contributes to resistance against complement and to intracellular persistence within resting macrophages. A vaccine against TB was developed in 1908 by ca. 230 serial in vitro passages of a M. bovis strain (Calmette et al., 1927). During in vitro passage, the strain became attenuated due to loss of numerous gene complexes, which have very recently been identified (Behr et al., 1999). In the 1920s, this strain was found to protect newborns against miliary TB. The protection of young children against miliary TB by BCG has been confirmed in several field trials (Huebner, 1996). However, these trials also revealed that the BCG vaccine inadequately protects adults from pulmonary TB, which results from reactivation. Therefore, general agreement exists that in the long run a novel vaccine will be required for satisfactory TB control, mostly in developing countries. Even in the United States, a TB vaccine for high-risk populations has been considered favorable because it incurs small costs compared to the savings in terms of life and health (Ashley and Murray, 1996; Committee to Study Priorities for Vaccine Development, 1999). A similar vaccine need can be envisaged for the European Union, which had more than 75,000 estimated cases of TB in 1996. B. SALMONELLOSIS Members of the genus Salmonella are ubiquitous pathogens found in humans and their livestock, wild mammals, reptiles, birds, and even insects. Antigenic analysis based on envelope, capsular, and flagellar antigens led to the identification of over 1500 distinct antigenic variants, or serotypes,

EXPLOITING THE IMMUNE SYSTEM

9

of Salmonella. However, only about 10 serotypes are responsible for the vast majority of human diseases. Historically, each antigenic variant of Salmonella has been given a separate species designation. While extremely useful as an epidemiological tool, species definition based on serologic typing has complicated the nomenclature. Therefore, the genus Salmonella has been restructured, resulting in a similarly confusing classification system (Le Minor, 1987), with all salmonellae being combined in the same species, Salmonella enterica, and each organism of a unique bio- or serotype representing a variant. For instance, Salmonella typhimurium is now referred to as S. enterica subsp. enterica serovar Typhimurium. However, for the reader’s convenience, we will continue to call this bacterium S. typhimurium. It should also be noted that it has recently been decided to maintain the name Salmonella typhi, primarily for safety reasons (Euzeby, 1999). The clinical pattern of salmonellosis can be divided into enteric fever (typhoidlike disease), gastroenteritis, bacteremia (with and without focal extraintestinal infection), and the asymptomatic carrier state. Salmonella infection in AIDS patients is common and often very severe. Bacteremia occurs in 70% of these patients and can cause septic shock and death. Despite adequate antimicrobial coverage, relapses are common. In addition, patients with lymphoproliferative disease are also highly susceptible to disseminated salmonellosis. Salmonella gastroenteritis usually follows the ingestion of contaminated food or drinking water and accounts for about 15% of foodborne infections in the United States. Approximately 10,000 bacilli are sufficient to cause illness. S. typhimurium and S. enteritidis are the major causes of Salmonella gastroenteritis, which is predominantly a disease of industrialized societies and is caused by improper food handling. Poultry products, including eggs, are most often implicated as the vehicle of infection. The precise pathogenesis of Salmonella gastroenteritis is not known and no equivalent animal model of disease exists. The microorganisms appear to rapidly breach the epithelial barrier of the intestine and proliferate in the lamina propria, inducing a superficial inflammatory response by infiltrating neutrophils. The bacteria only occasionally enter the bloodstream and cause sepsis. However, in most cases, the disease resolves within a week postinfection. Typhoid fever remains an important cause of morbidity and mortality, with more than 16 million cases and 600,000 deaths annually worldwide (Pang et al., 1998). In industrialized nations, the disease has declined as a result of clean water supplies and proper disposal of fecal waste, largely restricting the infection to travelers in endemic areas. Typhoid fever is a systemic infection affecting mainly the mesenteric lymph nodes, liver, and spleen; symptoms include prolonged fever and sustained bacteremia. The etiologic agent of typhoid is S. typhi, which infects only humans and other

10

HESS et al.

primates. The microoganisms enter the host via specialized epithelial cells, the M-cells overlying the Peyer’s patches (Kohbata et al. 1986). Here, the bacteria are ingested by macrophages and rather than being killed, they multiply within these cells and use them as carriers to lymphatic and blood circulation. By the time the patient presents with symptoms, the organisms have significantly proliferated in Peyer’s patches, mesenteric nodes, liver, and spleen. The immune response to enteric fever is both humoral and cell-mediated, subduing the untreated infection over a period of about three weeks. Some patients become asymptomatic carriers, because of chronic infection of the gallbladder and the biliary tract. These individuals are important reservoirs of infection in endemic areas. The first vaccinations against typhoid fever were attempted in the late 19th century. In 1896, Robert Koch’s co-workers Pfeiffer and Kolle used a phenolized suspension of salmonellae for immunization. Only one year later, Wright and Semple prepared heat-killed organisms for intramuscular (i.m.) administration to British troops fighting in the Boer War in South Africa (Winkle, 1997). By 1911, three doses of heat-killed typhoid vaccine were routinely given to members of the U.S. army. This vaccine was used with varying success for a long time, before being replaced by new formulations. Currently, two typhoid vaccines are commercially marketed. A new subunit typhoid vaccine consists of purified Vi antigen, the capsular polysaccharide of S. typhi. Administered by i.m. injection, this vaccine elicits antibodies that confer significant but incomplete protection (Keitel et al., 1994). In contrast, the live attenuated Ty21a typhoid vaccine has been in use for many years. Although the precise reason for the attenuation of Ty21a remains unclear, this vaccine strain is very safe. It is given orally in three doses and elicits both humoral and cellular protective immunity, albeit incomplete. Due to their ability to induce cell-mediated immunity as well as antibodies, live vaccines are generally considered to confer better protection against salmonellosis than killed vaccines. Therefore, a thorough search for improved live Salmonella vaccines that are protective following a single oral dose has been performed, and several new vaccine candidates have been described (see Section V.E.1). In addition to protection against salmonellosis, these vaccine strains carry the potential to serve as vehicles for delivering heterologous recombinant antigens (see Section V.F.3). C. LISTERIOSIS The genus Listeria comprises several species, including L. monocytogenes, L. ivanovii, and L. innocua. Of these, only L. monocytogenes is pathogenic for humans. L. monocytogenes is a gram-positive rod that causes an uncommon, but potentially serious, type of food-borne infection. In healthy adults, L. monocytogenes infection is usually asymptomatic or at

EXPLOITING THE IMMUNE SYSTEM

11

most produces mild flulike symptoms. More serious infections are seen in adults with certain underlying conditions that compromise their immune systems (cancer, diabetes, old age, pregnancy, etc.). In such people, listeriosis can cause central nervous system infections (bacterial meningitis, encephalitis) and fatal bacteremia. Furthermore, L. monocytogenes is one of the few bacteria that can cross the placenta. In pregnant women who contract listeriosis, the pathogen can infect the fetus, resulting in systemic L. monocytogenes infection of the newborn, stillbirths, or preterm labour (Kaufmann, 1988a). Therefore, most of the deaths caused by L. monocytogenes involve fetuses and newborns, as well as immunocompromised people. While human listeriosis is not a major health threat, murine listeriosis has long been used as model system for studying the cell-mediated immune response to intracellular bacteria. L. monocytogenes causes acute systemic infection in mice, and this animal model has been used to assess virulence of bacterial mutants. Infection with L. monocytogenes is one of the rare examples in which the intracellular pathogen is completely eradicated by the immune response. Cells involved in the early host response to L. monocytogenes comprise macrophages, neutrophils, and natural killer (NK) cells. These nonspecific cellular mechanisms are required for limiting bacterial replication in the mammalian host. Ultimately, 움웁 T cells (CD4 T helper 1 (Th1) and CD8 T cells) are responsible for clearance of infection and for long-term immunity (see Section IV.E). At least two secreted antigens, p60 and listeriolysin (Hly) of L. monocytogenes, are recogized by T cells. The p60 protein is involved in listerial septation and probably in cell adhesion of L. monocytogenes (Hess et al., 1995; Hess et al., 1996a; Kuhn and Goebel, 1989; Wuenscher et al., 1993), whereas the pore-forming Hly is essentially required for the phagosomal escape of L. monocytogenes within host cells (Gaillard et al., 1987). D. CHLAMYDIAL INFECTIONS The genus Chlamydia comprises three species (for an overview, see Stephens, 1999). Chlamydia psittaci (ornithosis) and C. pneumoniae (TWAR) are airborne pathogens that cause atypical pneumonia. In contrast, Chlamydia trachomatis is acquired congenitally or transmitted via close personal contact and, depending on the serotype, causes conjunctivitis, trachoma, lymphogranuloma venerum, and urogenital infections. Chlamydial infections have also been implicated in the development of reactive arthritis and atherosclerosis (Kuo and Campbell, 1998). Ocular trachoma, one of the major diseases caused by C. trachomatis, has been recognized since antiquity, reported in China in the 27th century BC and in Egypt in the 19th century BC. However, the organism’s role in genital tract infections did not become apparent until early this century.

12

HESS et al.

Today, C. trachomatis infections impose a significant burden on humans, representing the primary cause of blindness in the sub-Saharan countries. It is estimated that worldwide as many as 500 million people are affected by ocular trachoma, 7–9 million of them with significant visual impairment. Genital tract infections are even more prevalent, and the Centers for Disease Control has estimated that there are approximately 4 million new C. trachomatis infections per year in the United States alone, with a risk for sterility in infected women. The highest rates of chlamydial infections are in 15- to 19-year-old adolescents, regardless of demographics or location. Therefore, it has been suggested that even in the United States a Chlamydia vaccine administered to 12-year-olds would incur small cost for the value gained and hence falls in the group of favorable candidate vaccines (Committee to Study Priorities for Vaccine Development, 1999). Chlamydiae are obligate intracellular bacteria with a complex biphasic developmental cycle. The extracellular infectious form, called elementary body (EB), attaches to susceptible epithelial cells via heparan sulfate bridges and is subsequently internalized (Zhang and Stephens, 1992). Inside the host cell, the EB are reorganized into reticulate bodies (RB), the replicative form. As the RB divide, they fill up the vacuole and form a cytoplasmic inclusion. During maturation of the inclusion, anywhere between 100 and 1000 infectious EB are produced and then released to infect new host cells. While natural infection with C. trachomatis can be easily controlled by the immune system, it appears to confer very little and only short-lived protection against reinfection. Multiple or chronic infections are an essential factor in the pathogenesis of both ocular trachoma and genital infections, as it is the fibrosis and scarring that occur as the infection resolves which ultimately result in blindness or sterility. Thus, it is the host immune response to the organism that leads to pathology in this disease. IV. Intracellular Bacteria, Their Niches in the Host Cell, and the Immune Responses Elicited

Intracellular bacteria are endowed with strategies to survive in their host cells, which protect them from humoral defense mechanisms such as specific antibodies and complement. The macrophage is the central cell in infections with intracellular bacteria, with most of the bacteria discussed in this review using macrophages as host cells. The very same cells have to process and present antigens to T cells and, if activated by interferon웂 (IFN-웂), are the main effector cells for eradicating infections with intracellular bacteria (Kaufmann, 1998). Therefore, T cells expressing a Th1

EXPLOITING THE IMMUNE SYSTEM

13

cytokine pattern, in particular IFN-웂, play a pivotal role in protection against these pathogens. The bacteria discussed in this review have developed several strategies for exploiting distinct intracellular niches that enable survival and replication as well as protection from antibacterial activities of the host (O’Brien et al., 1996; Russell et al., 1997; Sinai and Joiner, 1997). Their seclusion in an intracellular hiding place seems to be the main reason effective vaccines against these bacteria have not yet been developed. Classical vaccines depend on specific antibodies as the protective principle, but these are not critical for defense against intracellular bacteria. Rather, effective vaccines against these pathogens need to induce strong Th1 responses. Although most intracellular bacteria can replicate outside their host cells, their intracellular niche affects the type of immune response elicited. Chlamydiae reside and replicate in a nonacidic compartment, the inclusion body, which does not fuse with endosomes (Sinai and Joiner, 1997). Mycobacteria reside in nonacidified early endosome-like phagosomes, which are blocked in maturation (Russell et al., 1997). In contrast, salmonellae appear to be adapted to survive in spacious, late endosomal/ lysosomal compartments (Alpuche-Aranda et al., 1995). Finally, L. monocytogenes escape from the phagosome into the cytoplasm by expressing a pore-forming hemolysin, Hly, and two phospholipases (Cossart, 1997). A. THE INNATE IMMUNE RESPONSE Preceding the acquired immune response, components of the innate immune system rapidly respond to bacterial structures. Mononuclear phagocytes and NK cells are the principal cells involved in the innate response. They produce cytokines—notably IFN-웂—and effector molecules to contain the infectious agent. The early IFN-웂 burst activates macrophages and triggers a major effector mechanism, inducible nitric oxide synthase (iNOS) (Fang, 1997). This enzyme synthesizes the highly reactive antimicrobial nitric oxide radical (NO ⭈ ), which acts as an oxidizing agent and can interact with O2⫺ to form toxic peroxynitrite (ONOO⫺) (MacMicking et al., 1997). Macrophages activated by IFN-웂 can inhibit the replication of a variety of intracellular bacteria, including BCG and M. tuberculosis, in an NO ⭈ -dependent manner (Chan et al., 1992; Flesch and Kaufmann, 1991; Miyagi et al., 1997; Rhoades and Orme, 1997; Yamamoto et al., 1992). Although the early restriction of microbial growth is important in successfully combating invading pathogens, this part of the innate response does not represent a major target of vaccination because it is short-lived and antigen-nonspecific. However, the innate immune system also promotes the development of an appropriate antigen-specific response and in this respect, constitutes an important target of vaccination.

14

HESS et al.

Early in infection, interleukins (IL) 12 and 18 are produced and in turn stimulate NK cells to express IFN-웂. IL-12 and IL-18 are secreted by macrophages and/or dendritic cells (DC) when triggered by certain cell wall components of intracellular bacteria, nonmethylated CpG motifs contained in bacterial DNA or lipoproteins, or upon activation by IFN-웂 (Brightbill et al., 1999; Chace et al., 1997). Mycobacterial lipoproteins stimulate macrophage IL-12 production by means of Toll-like receptors (Brightbill et al., 1999). Furthermore, macrophages can independently produce IFN-웂 after costimulation with IL-12 and IL-18 (Munder et al., 1998). Taken together, this results in a T-cell-independent proinflammatory feedback loop, with IFN-웂 acting on macrophages to sustain IL-12 production (Flesch et al., 1995). B. ANTIGEN-PROCESSING PATHWAYS Bacterial antigens can be presented by classical and nonclassical antigenpresenting molecules. The classic antigen-presenting molecules encompass the polymorphic molecules of major histocompatibility complex (MHC) class I and II, which present antigenic peptides to conventional CD8 or CD4 T cells, respectively. These cells usually express the 움웁 T cell receptor (TCR). Antigens presented by MHC class I molecules are of endogenous origin, derived either from the host cell or from intracellular pathogens such as viruses. These proteins are processed in the cytosol by proteasomes and other proteases, and the resulting peptides are transported via ATPdriven transporters (TAP1/2) into the endoplasmic reticulum (ER), where they hook up with the newly synthesized MHC class I molecules (York and Rock, 1996). This is done by practically every nucleated cell expressing MHC class I molecules. In contrast, MHC class II molecules are expressed only by professional antigen-presenting cells (APC), such as macrophages, DC, B cells, or epithelial cells. The antigens presented via MHC class II are of exogenous origin and must be engulfed by macropinocytosis, receptormediated endocytosis, or phagocytosis. The receptors involved in antigen uptake encompass membrane immunoglobulins (Ig) of B cells, as well as complement receptors, scavenger receptors, and members of the mannose receptor family (MMR, DEC205) on macrophages and/or DC (Watts, 1997). In late endosomes/lysosomes, these antigens are proteolytically processed by certain cathepsins and/or an asparaginyl endopeptidase and bound to MHC class II molecules upon HLA–DM-catalyzed exchange for an invariant chain derived peptide (CLIP) (Manoury et al., 1998; Pieters, 1999). The invariant chain hooks up to MHC class II molecules in the ER to target class II into endosomes, where Ii is degraded by cathepsins. Accordingly, antigens derived from pathogens engulfed by and residing in macrophages are prone to processing via the MHC class II

EXPLOITING THE IMMUNE SYSTEM

15

pathway. Nonclassical antigen-presenting molecules are nonpolymorphic and probably represent correlates of the coevolution of bacteria and the mammalian immune system. They include murine MHC class Ib (H2-M3, Qa), human CD1a, b, and c, as well as human and murine CD1d molecules. It has been shown that murine H2-M3 presents formylated peptides, and human CD1a, CD1b, and CD1c present glycolipids of bacterial origin to T cells (Fischer Lindahl et al., 1997; Porcelli and Modlin, 1999). Although it is not yet clear if these antigens need processing and where in the cell this is happening, presentation of mycobacterial glycolipids via CD1b requires functional lysosomes (Sieling et al., 1995). C. THE ACQUIRED IMMUNE RESPONSE The acquired immune response, which is mediated by antigen-specific lymphocytes, represents the prime target of any vaccination strategy. First, it is antigen-specific and focuses vaccine-induced responses on a single pathogen, thereby avoiding broad specificities, which may encompass autoimmunity. Second, the acquired immune response possesses memory, allowing long-lived or—as an ideal goal of vaccination—lifelong protection. MHC class II-restricted CD4 T cells expressing the TCR 움웁 represent the central T cell population in the protective immune response against intracellular bacteria residing in phagosomes. These T cells can be divided into Th1 and Th2 cell subsets according to their cytokine secretion patterns (Lucey et al., 1996). Th1 cells produce IFN-웂, and tumor necrosis factor (TNF) 웁, whereas Th2 cells secrete IL-4, IL-5, IL-6, and IL-13. Because of differences in their cytokine patterns, these T cell subsets have different roles in the host immune response. Th1 cells produce IFN-웂 thereby promoting cell-mediated immunity characterized by macrophage activation, stimulation of cytotoxic T lymphocytes (CTL), and production of the opsonizing Ig isotypes IgG2a and IgG2b in the mouse. In contrast, Th2 cells control the differentiation of B cells to Ig-producing plasma cells via the production of IL-4, and promote Ig class switching to IgG1, IgE, and IgA by IL-4 or IL-5. These cells play a pivotal role in infections with extracellular bacteria and parasites. Infections with intracellular pathogens generally induce a Th1 response with potent protective potential. Because vaccines against intracellular bacteria must trigger a Th1-type response, viable bacteria such as salmonellae and BCG, which already induce a Th1 response, have several advantages as vaccines/vaccine carriers over subunit vaccines. In the case of subunit vaccines, strong adjuvants, probably together with IL-12 treatment would be required to induce a Th1 response. The development of a Th1 response is a well-regulated event propelled by the IFN-웂/IL-12/IL-18-driven feedback loop outlined previously, enhancing the expression of IL-12 receptors on CD4 T cells (Flesch et al.,

16

HESS et al.

1995; Guler et al., 1996; Hsieh et al., 1993; Macatonia et al., 1995). More recently, IL-18 has been shown to potentiate IL-12-driven Th1 cell development (Robinson et al., 1997). In the initial phase of a specific immune response, the early T cell cytokine Eta-1 (osteopontin) can push T cells toward a Th1-type by stimulating IL-12 in macrophages and, at the same time, blocking production of the Th1 inhibitory cytokine IL-10 (Ashkar et al., 2000). The importance of the Th1/Th2 T cell dichotomy in human disease states is best exemplified by the host response to M. leprae, where the tuberculoid form of the disease is characterized by a Th1-type response, resulting in activated macrophages, granuloma formation, and few bacteria. In contrast, toward the lepromatous pole of the disease, a Th2-like response occurs, with increased levels of IgE and IgG1, little, if any, granuloma formation, and abundant acid-fast bacilli (Lucey et al., 1996). MHC class I-restricted CD8 T cells exhibit antigen-specific cytolytic activity against infected host cells by secreting perforin and granzymes. Furthermore, they express granulysin, which is toxic to many intracellular bacteria and protozoa, including mycobacteria (Stenger et al., 1998), and secrete IFN-웂 (DeLibero et al., 1988; Lalvani et al., 1998). There are three possible functions of CD8 T cells during the course of infection with intracellular bacteria (Kaufmann, 1999): (i) they kill target cells to curtail the growth of the bacteria (DeLibero et al., 1988; Stenger et al., 1997); (ii) they lyse infected cells that can no longer control the infection and release the bacteria so that other more proficient cells can phagocytose and ultimately kill them; and (iii) they secrete IFN-웂 to activate macrophages, further enhancing their microbicidal activity. Because of their residence in the cytoplasm, L. monocytogenes organisms preferentially induce MHC class I-restricted CD8 T cell responses, which represent the major effector phase in protection (Kaufmann, 1993; Kaufmann and Ladel, 1994; Ladel et al., 1994; White et al., 1996). However, CD8 T cells are also activated upon infection with intraphagosomal pathogens such as M. tuberculosis, M. bovis BCG (DeLibero et al., 1988; Turner and Dockrell, 1996), S. typhimurium (Pope and Kotlarski, 1994; Pope et al., 1994; Sztein et al., 1995), and C. psittaci (Starnbach et al., 1994). The important role of MHC class I-restricted CD8 TCR 움웁 cells in murine TB has been revealed by using knockout mice lacking functional MHC class I molecules (웁2m⫺/⫺ mice; H-2KbDb⫺/⫺ mice) or TAP transporters. These mice succumb to infection more rapidly than wild-type animals (Behar et al., 1999; Flynn et al., 1992; Rolph and Raupach, unpublished). Furthermore, protection against experimental TB in mice could be transferred by CD8 T cell clones recognizing a mycobacterial heat shock protein of 65 kDa (Hsp65) (Silva et al., 1994). The fact that phagosomally residing bacteria can elicit CD8 T cells raises the question of whether and how

EXPLOITING THE IMMUNE SYSTEM

17

the antigens of these bacteria reach the cytoplasm-associated classical MHC class I-processing/presentation pathway. MHC class I presentation of these antigens is not exclusively dependent on this pathway and a number of publications have unraveled the existence of an ‘‘alternative’’ MHC class I pathway. This pathway involves at least some intravesicular steps and is utilized mainly for particulate antigens such as bacteria (Harding et al., 1995; Reimann and Kaufmann, 1997). It has only a low efficiency for soluble proteins. This processing pathway could be facilitated by a subset of MHC class I molecules that associate with the invariant chain and are targeted to endosomal compartments, where they can be loaded with exogenous peptides (Sugita and Brenner, 1995). In addition, empty MHC class I molecules (H-2Ld ), which recycle from the cell surface to endosomes, can sample peptides from exogenous antigens (Schirmbeck and Reimann, 1996). Alternatively, endosomally processed peptides could be regurgitated by infected macrophages, allowing sensitization of bystander APC ( Jondal et al., 1996). Another explanation takes into account that most intracellular bacteria or bacterial lipoproteins induce apoptosis and formation of apoptotic blebs in host cells (Aliprantis et al., 1999; Hayashi et al., 1997). Furthermore, APC such as B cells and DC release extracellular vesicles containing peptides and MHC molecules (exosomes) (Raposo et al., 1996; Zitvogel et al., 1998). Macrophages or DC can engulf antigenloaded exosomes, apoptotic cells, or apoptotic blebs and present these antigens in a TAP-dependent fashion to specific CD8 T cells (Albert et al., 1998; Bellone et al., 1997). Because many intracellular bacteria induce apoptosis in their host cells, this translocation pathway can be easily envisaged. In DC, an endosome-to-cytosol transport of internalized antigens allowing access of antigens to the conventional cytosolic MHC class I antigen processing machinery has recently been described (Rodriguez et al., 1999). Taken together, these data suggest that efficient vaccination against intracellular bacteria requires that CD8 CTL be induced in addition to CD4 Th1 cells (Kaufmann, 1988b). For this purpose, vaccine design must employ either carrier organisms or molecules facilitating entry of the vaccine antigens into the classical or alternative MHC class I pathway. This can be best achieved by recombinant bacteria expressing bacterial hemolysins or by certain adjuvants such as immunostimulatory complexes. A number of human DN or CD8 T cell lines that express TCR 움웁 and recognize mycobacterial glycolipids in the context of group I CD1 molecules have been described. Because these T cells are cytolytic, secrete granulysin, and produce IFN-웂, they could play an important role in the protective immune response against mycobacteria in humans (Porcelli and Modlin, 1999). CD1 molecules share sequence homologies and structural features with MHC class I molecules, but they are nonpolymorphic (Por-

18

HESS et al.

celli, 1995). Like MHC class I molecules, they are noncovalently linked to 웁2m (Bauer et al., 1997; Brutkiewicz et al., 1995; Teitell et al., 1997). The group I CD1 molecules CD1a, CD1b, and CD1c are found in humans, guinea pigs, and other mammals but not in mice. They are primarily expressed on DC (Porcelli et al., 1998). The glycolipids presented by group I CD1 molecules are constituents of the acid-fast mycobacterial cell wall and include lipoarabinomannan (LAM), phosphoinositol mannosides (PIM), mycolic acids, mycolyl glycolipids such as glucose monomycolate (GMM), and phospholipids (Beckman et al., 1994, 1996; Moody et al., 1997; Porcelli et al., 1992; Rosat et al., 1999; Sieling et al., 1995). Taking the nonpolymorphic character of CD1 molecules into account, chemically defined glycolipids represent interesting vaccine candidates. Included in a subunit vaccine mixture, glycolipids will also drive the immune response into a Th1 direction due to their adjuvanticity. Early vaccination studies in the 1960s and 1970s using protein-free preparations from mycobacteria in mice, guinea pigs, and humans favored glycolipids as vaccine candidates (reviewed in Crowle, 1988). Furthermore, other cell wall compounds, namely, peptidoglycan complexes, when entrapped into liposomes, can partially protect mice against a lethal dose of M. tuberculosis (Chugh and Khuller, 1993). The group II CD1 molecule CD1d is present in both humans and mice on epithelial cells, B cells, and other APC. The T cells that respond to CD1d are NK T cells. They are either DN or CD4 and coexpress NK cell markers such as NK1.1, in addition to a limited TCR repertoire (V움14J움281 in mice; V움24V웁11 in humans) (Balk et al., 1991; Bendelac et al., 1995; Exley et al., 1997; Porcelli et al., 1998). CD1d presents glycolipids of nonmicrobial origin, such as 움-galactosylceramide, to NK T cells (Kawano et al., 1997) and binds glycosylphosphatidylinositol (GPI) ( Joyce et al., 1998). NK T cells produce IL-4 and IFN-웂 in response to GPI-linked glycolipids and proteins from parasites (Schofield et al., 1999). NK T cells are thought to play a regulatory role rather than participating directly in the effector phase against intracellular bacteria. Anti-CD1d treatment with specific antibodies in vivo rendered mice more susceptible to TB but slightly ameliorated listeriosis (Szalay et al., 1999a,b). When injected subcutaneously (sc), crude polar glycolipids from mycobacteria induced granuloma-like infiltrations in the skin that contain significant numbers of NK T cells (Apostolou et al., 1999). In contrast, mice genetically defective in CD1d expression are equally as susceptible to TB as their wild-type littermates (Behar et al., 1999). TGF웁 is downregulated in anti-CD1dtreated mice infected with listeriae, pointing to an immunoregulatory role of CD1d-controlled NK T cells (Szalay et al., 1999a). Although the glycolipids so far described as presented by CD1d are of nonbacterial origin, the structures of these carbohydrates resemble many microbial cell wall

EXPLOITING THE IMMUNE SYSTEM

19

constituents. Therefore, future studies must consider bacterial glycolipids as antigens for presentation by CD1d. T cells expressing TCR 웂␦ in humans represent a small population of all peripheral T lymphocytes, but one which rapidly expands upon in vitro stimulation by bacterial components (Garcia et al., 1997; Hara et al., 1992; Modlin et al., 1989). In human adults, ⬎50% of the 웂␦ T cell population express V웂2␦2 (DeLibero et al., 1991; Kabelitz et al., 1991) as this population is expanded from less than 5% during ontogeny. Although the function of 웂␦ T cells is not yet conclusive, they seem to participate in the host response, especially toward mycobacteria. They are abundant in lesions of leprosy patients, and their number is increased upon infection with salmonellae (Hara et al., 1992; Modlin et al., 1989). In the absence of any nominal antigen-presenting molecule, they respond to phospholigands such as isopentenyl pyrophosphate and their alkyl derivatives, nucleotides, phosphosugars, and phosphoesters (Chien et al., 1996; Constant et al., 1994; Pfeffer et al., 1990; Schoel et al., 1994; Tanaka et al., 1994, 1995). In mice, 웂␦ T cells respond to peptides presented by nonpolymorphic MHC class I molecules, such as Qa. These cells accumulate and secrete IFN-웂 during listeriosis and can compensate for the absence of 움웁 T cells (Hiromatsu et al., 1992; Mombaerts et al., 1993; Skeen and Ziegler, 1993). In murine tuberculosis, 웂␦ T cells appear to participate in the containment of the bacilli (Ladel et al., 1995). Upon low-dose aerosol infection, the role of 웂␦ T cells, however, is negligible (D’Souza et al., 1997). It was also observed that TCR 웂␦ KO mice showed an increased influx of neutrophils into the mycobacteria-containing granuloma, in contrast with the primarily lymphocytic infiltrate in wild-type mice (D’Souza et al., 1997). Together with the observation that L. monocytogenes-infected TCR 웂␦ KO mice develop liver abscesses rather than granulomas (Mombaerts et al., 1993), this result suggests that 웂␦ T cells play a regulatory role. They limit the inflammatory response that leads to tissue damage, probably by producing IL-10 to counterregulate IFN-웂-induced inflammation and tissue damage (Hsieh et al., 1996). The nonpolymorphic MHC class I molecule H2-M3 presents N-methylformylated peptides derived from bacteria such as L. monocytogenes to CD8 T cells (Fischer-Lindahl et al., 1997). Because these molecules have only been described yet for mice, the potential of formylated proteins as vaccine candidates for use in humans is low. Moreover, although H2–M3restricted T cells are induced quickly after infection with listeriae, during the memory response they are markedly reduced (Kerksiek et al., 1999). Although the role of antibodies in infections with intracellular bacteria is still elusive, they may have some influence. In fact, studies with mice lacking B cells revealed a significant protective role for specific B cells/

20

HESS et al.

antibodies in both the primary and the secondary response to oral infection with S. typhimurium: Upon vaccination with an attenuated salmonella strain, protection against wild-type Salmonellae became B-cell-dependent not only after oral but also after systemic infection (Mittru¨cker et al., 2000). Furthermore, Fc-receptor-mediated engulfment of antibody-coated listeriae or mycobacteria by macrophages alters the intracellular fate of these bacteria; i.e., their phagosomes fuse with lysosomal compartments (Armstrong and Hart, 1975; Collins et al., 1997; Malik et al., 2000). Fusion with lysosomes alone does not solely seem to influence the survival of mycobacteria, which in addition depends on macrophage activation (Gomes et al., 1999; Schaible et al., 1998; Malik et al., 2000). Interestingly, a monoclonal antibody specific for a surface glycolipid has been described as being protective against TB when the bacilli were coated prior to infection (Teitelbaum et al., 1998). Similarly, passive transfer of antibodies against Hly ameliorates listeriosis in mice (Edelson et al., in press). In experimental C. trachomatis infection in mice, it was found that specific antibodies of the IgG and IgA isotype are relevant for protection against reinfection via the genital mucosa ( Johansson et al., 1997; Peterson et al., 1997; Su et al., 1997). Despite the central role of T cells in defense against bacteria once they have established themselves in their intracellular niche, in principle, there may be a role for antibodies in vaccine-induced protection against these pathogens. Given that high titers of neutralizing and/or opsonizing antibodies can be sustained at ports of entry, it can be envisaged that they prevent microbial invasion of the host, especially when infections are established by only small numbers of bacteria. Because the bacteria under discussion here enter the host at mucosal sites, vaccination strategies that achieve high levels of mucosal antibodies should be elucidated. D. MUCOSAL IMMUNITY For all the bacteria discussed here, the mucosal surface serves as an entry port. The intestinal tract is the largest immune organ in the body and one that contains all the major immune cell populations. Although mucosal immunity is, in itself a topic demanding comprehensive review, vaccination strategies against intracellular bacteria must necessarily consider the importance of the immune response at the site of first encounter between host and infectious agent, i.e., the mucosa. Listeria monocytogenes, S. typhimurium, and some mycobacteria enter the host through the intestinal epithelium, M. tuberculosis and C. pneumoniae primarily through the lung epithelium, and C. trachomatis through epithelia of the genital tract and the eye. Listeriae pass through normal epithelial cells whereas salmonellae and mycobacteria prefer crossing the epithelial border via specialized cells, the membraneous or microfold (M) cells which are distrib-

EXPLOITING THE IMMUNE SYSTEM

21

uted in the mucosa (Kumagai, 1922; Gru¨tzkau et al., 1990; Jones et al., 1994; Sansonetti and Phalipon, A., 1999; Teitelbaum et al., 1999). It has been shown that lymphocytes of the mucosa-associated lymphoid tissue (MALT)—most probably B220⫹B cell precursors—can induce differentiated epithelial cells to become M cells (Kemeis et al., 1997). Generally, there is intimate contact between the epithelium and the MALT, for example, in Peyer’s patches in the small intestine, the tonsils, and other lymphoid tissues associated with the upper and lower airways, the conjunctiva, and the urogenital tract. Antigens such as pathogens, microparticles, macromolecules, or soluble compounds are sampled by M cells and probably epithelial cells, via pinocytosis, macropinocytosis, or receptor-mediated engulfment. Epithelial cells are able to process and present antigens in the context of MHC class I, MHC class II, and CD1d. The expression of MHC molecules and molecules associated with the MHC class II antigenprocessing pathway (Ii, HLA-DM) on epithelial cells is upregulated by IFN-웂 (Hershberg et al., 1997). Although MALT is considered a primary site of the mucosal immune response, intestinal epithelial cells can independently present antigens to T cells (Bland and Warren, 1986; Kaiserlian et al., 1989). They also express a novel 180-kDa membrane protein which serves as an activating ligand for CD8 T cells (Li et al., 1995). Resting intestinal epithelial cells primarily stimulate CD8 T cells and are assumed to tolerize CD4 T cells, whereas IFN-웂-activated intestinal epithelial cells present antigens to stimulate CD4 T cells. Epithelial cells can also secrete proinflammatory cytokines such as IL-1, IL-6, TNF움, M-CSF, GM-CSF, and TGF웁, as well as an array of chemokines (reviewed in Berin et al., 1999). Because HLA–DR expression is faint on intestinal epithelial cells, and costimulatory molecules such as CD80 and CD86 to date have been found only on gastric epithelial cells, other APC in close vicinity to the epithelium are probably needed (Berin et al., 1999). The basolateral surface of the M cell forms a pocket that harbors DC, macrophages, B cells, and T cells of primarily the CD4 subtype. Pocket B cells are predominantly (앑66%) memory cells (sIgD⫺) and express HLA–DR as well as costimulatory molecules (CD80, CD86). These cells are able to specifically capture antigens via their membrane Ig and may therefore represent the most important APC in the M cell pocket to provide cognate stimulation of adjacent T cells (Brandtzaeg et al., 1999). In contrast, naive B cells or macrophages lacking costimulatory molecules will provide signals to tolerize/anergize T cells against mucosal antigens, leading to mucosal tolerance. The T cells in the M cell pocket often express activation markers such as CD69 and the memory marker CD45RO⫹, but no L-selectin (Farstad et al., 1994). Conventional T cells of the MALT are derived from the

22

HESS et al.

thymus and, after priming in the MALT, return either to the lamina propria as CD4 T cells or to the epithelium as CD4 or CD8움웁 intraepithelial lymphocytes (IEL). The mucosal epithelium is also home to a set of probably extrathymically developing IEL, which mostly express CD8움움 (80–90%) and either TCR 움웁 or TCR 웂␦ (Brandtzaeg et al., 1999). In mice, these cells develop from bone marrow precursors in small, primary lymphoid aggregates in the gut, which are called cryptopatches (Saito et al., 1998). As with conventional T cells, B cells primed in the MALT acquire homing receptors directing them back to mucosal sites and exocrine glands, where they differentiate into antibody-producing plasma cells. The primary antibodies secreted at mucosal sites are of the IgA, IgM, and IgG isotypes. Whereas IgG passively diffuses through the epithelium, transfer of sIgA and sIgM to the apical surface of the epithelium is facilitated by the 100-kDa polymeric Ig receptor (pIgR) of epithelial cells (Brandtzaeg et al., 1999). Together with innate mucosal defense mechanisms, luminal antibodies provide noninflammatory surface protection against pathogens by preventing entry into the host. On the one hand, vaccination via mucosal tissues is easy, cheap, and very effective in inducing mucosal, as well as generalized, immunity. Mucosal vaccine delivery systems such as cholera toxoid, attenuated salmonellae, or listeriae or liposome-encapsulated DNA are discussed in Section V. On the other hand, chronic exposure to luminal antigens in the intestine induces hyporesponsiveness, so-called oral tolerance (Chen et al., 1995). This phenomenon, which has also been described in the respiratory tract, minimizes immune and inflammatory reactions against the enormous amount of luminal antigens derived from bacteria, food, and inhaled particles. Furthermore, oral feeding of antigenic peptides can be used to tolerize autoimmune T cells in experimental allergic encephalitis (Miller et al., 1992). Oral tolerance is believed to be maintained by continuous exposure to a stable dose of antigen, whereas specific mucosal immune responses can be induced by the sudden appearance or increase in dosage of an antigen. For example, a single administration of sheep red blood cells is immunogenic, but a continuous rechallenge with the same amount of red cells is tolerizing (Mattingly and Waksman, 1978). It has been suggested that continuous antigen exposure in a noninflamed epithelium will lead to the elimination/tolerization of T cells. In this situation, activation of CD8 T cells producing inhibitory cytokines will lead to hyporesponsiveness, a phenomenon that has been demonstrated by adoptive transfer experiments (McMenamin et al., 1994; Miller et al., 1992). Furthermore, suppressive CD4 T cells have also been described (Chen et al., 1995). In contrast, antigen exposure in the inflamed epithelium will induce a specific Th1dominated immune response. Therefore, mucosal vaccines against infec-

EXPLOITING THE IMMUNE SYSTEM

23

tious agents need to activate CD4 Th1 cells, whereas interfering with autoimmune disease depends on tolerization of CD4 Th1 cells. Vaccines delivered via mucosae induce mucosal immunity at the sites of pathogen entry in addition to systemic immunity. By vaccination via specific mucosal tissues, the immune response is directed specifically to the organs targeted by certain bacteria: oral/gut/salmonellae, intranasal/ lung/mycobacteria and chlamydiae, intravaginal/urogenital tract/chlamydiae. However, tissues such as the gut, that harbor an abundant microbial flora have the disadvantage that the vaccines must compete with other microbial antigens. In contrast, the nasal and vaginal mucosae are less densely colonized and therefore, represent preferred application sites. Furthermore, due to migration of primed effector cells from the mucosa to other exocrine tissues, immunization via one mucosal site will eventually lead to an immune response at other sites (Brandtzaeg et al., 1999). E. THE MEMORY OF THE IMMUNE SYSTEM Vaccination is based on immunological memory. Both the B cell and the T cell compartment are able to generate antigen-specific memory cells. Upon rechallenge with the homologous antigen, memory cells respond much faster and more vigorously to lower doses of antigen than primary cells do. In the case of memory B cells, the response is characterized by antibodies with higher affinity. Memory T cells can produce a wider variety of cytokines than naive T cells and can provide potent B cell help (Duncan and Swain, 1994; Swain, 1994). The mechanisms that lead to the generation and maintenance of memory cells are incompletely understood. Nevertheless, the duration of immunological memory in humans is impressive: the classic example is a study on measles epidemics on the Faeroe Islands, showing that survivors of the first epidemic in 1781 were still protected during the second epidemic 65 years later in 1846 (Panum, 1847). Other studies revealed long-term antibody responses, such as 15 years for vaccinia, 40 years for polio, 75 years for yellow fever, and 10 years for the nonreplicating toxin vaccines like tetanus and diphtheria (Cooney et al., 1991; Paul et al., 1951; Sawyer, 1931; Gottlieb et al., 1964; Kjeldsen et al., 1985). Whereas all these infections/vaccinations are controlled by protective antibody responses, protective immunity against intracellular bacteria is dominated by a specific T cell response. Memory T cells have been described as small resting cells that are resistant to activation-induced cell death and show the following phenotypic marker characteristics: L-selectin⫺, CD44medium, CD45RBlow in mice, and CD45RO⫹, CD45RA⫺ in humans. The mechanisms for the generation and maintenance of memory T cells are still obscure. Antigen does not seem to be required for the transition of T cells from the activated to the memory stage. For the maintenance

24

HESS et al.

of T cell memory, the presence of antigen is needed in some systems but not in others (reviewed in Dutton et al., 1998). In some adoptive transfer models, CD8 memory can persist in the absence of both the specific antigen and the restriction element; in others, maintenance of memory depends on the presence of MHC class I molecules, and antigen is needed for the expansion of memory T cells (Tanchot et al., 1997; Markiewicz et al., 1998). The current view is that memory T cells persist as a slowly but continuously replicating cell pool that can be expanded in the absence of antigen by certain cytokines, including IL-2, IL-12, IL-15, and IFN-움 (reviewed in Freitas and Rocha, 1999). Whereas the generation of memory T cells has not yet been linked clearly to a specialized anatomical site, memory B cells are generated in the germinal centers of lymphoid tissues. Here, follicular dendritic cells, which can store antigen–antibody complexes, play a role in positively selecting long-lived B cells that secrete high-affinity antibodies. Post-GC memory B cells appear in two populations: antibody-secreting plasma cells and nonsecreting memory B cell precursors. The first population may be responsible for long-lived maintenance of circulating antibodies, whereas the second population may act as a reservoir to be expanded in response to rechallenge. Several models have been proposed for the maintenance of B cell memory and long-term antibody production: (i) repeated or chronic exposure to an infectious agent, (ii) antigen-containing immune complexes in follicular dendritic cells, and (iii) long-lived plasma cells, which do not depend on the presence of antigen (reviewed in Slifka and Ahmed, 1998; Manz et al., 1998). Whereas immune complexes decline with a half-life of 8 weeks (Tew and Mandel, 1979), plasma cells can survive for several months. In the bone marrow, these cells have a halflife of at least 4 weeks, and the majority are selected on the basis of antibody affinity (Manz et al., 1997). In considering the mucosae as sites for vaccine application to induce first-line defense antibody responses, it should be noted that in contrast to serum antibody production, mucosal antibody responses appear to be rather short-lived (McHeyzer-Williams and Ahmed, 1999). This poorly understood phenomenon requires further investigation by vaccinologists. In summary, vaccines against intracellular bacteria must achieve long-lived if not lifelong, immunity. Our understanding of the requirements for induction and maintenance of T cell memory will provide an important basis for the development of efficacious vaccination strategies, including the following questions: (i) Are viable vaccines that persist in the host and release antigens continuously preferable over inactivated vaccines? (ii) Which type of adjuvant releases vaccine antigens over a long period of time to induce memory against inactivated vaccines? (iii) How many boosts in which time ranges are needed? (iv) Can certain

EXPLOITING THE IMMUNE SYSTEM

25

prime/boost schedules enhance immunogenicity and memory? Prime/boost schedules can influence the efficacy of vaccines as has been shown for a vaccine against experimental malaria in mice that used priming by recombinant DNA followed by a boost with recombinant vaccinia virus (Schneider et al., 1998). F. QUANTITATIVE CORRELATES OF PROTECTION To measure the effectiveness of vaccination, several parameters of the protective immune response can be determined. Antibody-based vaccinations can easily be controlled by testing patient sera for the presence of specific or even neutralizing antibodies. Vaccination-induced T cell responses require more sophisticated technologies. Evaluation of vaccine efficacy in humans strongly depends on in vitro correlates of protection. Vaccine-induced T cells responding to recall antigens in vitro can be tested for the antigen-induced Th1 cytokines such as IFN-웂 or CTL activity. Recent advances have allowed quantitative assessment of protective correlates. First, it has become possible to determine the numbers of T cells that produce a given cytokine, e.g., by cytofluorimetry. Second, it is now feasible to enumerate peptide-specific T cells using tetramer complexes of major MHC molecules and specific peptides (Altman et al., 1996). A combination of both techniques will allow determination of the frequency of specific T cells producing a given cytokine. The tetramer technology has been mostly applied to the determination of virus-specific T cells (Dunbar et al., 1998; Murali-Krishna et al., 1998; Tan et al., 1999). For several viral pathogens, dominant epitopes are well defined, and it appears that antiviral CTL responses are frequently skewed to a few dominant epitopes. Such studies have come up with the provocative finding that during the height of the antiviral immune response, 10% or more of the CD8 T cell population is directed against a single viral epitope. In the case of intracellular bacteria, dominant epitopes have been defined only for L. monocytogenes. These studies have revealed slightly lower, although still compatible, frequencies of epitope-specific CD8 T cells (Busch et al. 1998). It will be essential to determine whether a relatively linear relationship exists between T cell numbers and vaccine efficacy, i.e., whether such high percentages of antigen-specific T cells are required for optimum protection. It could well be that the majority of T cells activated during infection are superfluous and that the high proportion reflects a marked redundancy in the system. If a linear relationship between vaccine efficacy and T cell numbers does indeed exist, this raises major questions about multivalent vaccination protocols. It could easily be assumed that multivalent vaccines comprising multiple antigens would cause antigen competition, inducing

26

HESS et al.

suboptimal numbers of T cells and, hence suboptimal protection. It is more likely that the system is quite redundant, in which case the threshold numbers of T cells required for protection need to be carefully determined. Once such numbers have been defined, the tetramer technology combined with the cytokine detection by cytofluorometry will provide a valuable tool for ascertaining quantitative correlates of protection in vaccinees. V. Vaccines

Principally, technologies for the development of active vaccines fall into three major categories: live, killed or subunit, and genetic (Table II). The protective potential of killed intracellular bacteria and subunit vaccines as purified protein components of these killed pathogens is limited in the absence of sophisticated adjuvants (Fig. 1). Attenuated viable bacterial strains are generally more potent in evoking protective immunity against the homologous pathogen due to their intrinsic adjuvanticity. Because of the complex composition within live homologous vaccines of various immunogenic and immunostimulatory components, such as lipoproteins, glycoproteins, polysaccharides, and glycolipids in addition to protein antigens, subunit vaccines will hardly achieve protective efficacies comparable to those of viable bacterial vaccine strains. Nevertheless, the strategic decision is generally based on the pathogenesis, immunobiology, and epidemiology of the infection or disease in question, as well as on the technical feasibility of each approach (Ellis, 1999).

TABLE II MAJOR VACCINE CANDIDATES Live vaccines Attenuated strains Defined mutants Undefined mutants Recombinant bacterial carriers Immunostimulatory molecules Heterologous antigens ‘‘Naked’’ DNA Subunit vaccines (Recombinant) antigens in adjuvant Peptides in adjuvant Genetic vaccines ‘‘Naked’’ DNA

EXPLOITING THE IMMUNE SYSTEM

27

FIG. 1. Principles of DNA and subunit vaccination. In vivo transfection of APC with DNA constructs results in the stimulation of MHC class I-restricted CD8 T cells and of MHC class II-restricted CD4 T cells via cross-priming of neighboring APC or by ligand competition for MHC class II molecules on the surface of the same APC. Subunit vaccines (protein/adjuvant) primarily introduce antigen into the MHC class II pathway, although some adjuvant provide access to the MHC class I machinery. Generally, proteins and hence DNA and subunit vaccines do not stimulate 웂␦ T cells or CD1-restricted T cells.

A. CHARACTERISTIC FEATURES OF PROTECTIVE ANTIGENS 1. Physicochemical Properties of Vaccine Antigens During bacterial infections, specific antibodies develop against a broad range of antigens, including carbohydrates, lipids, glycolipids, nucleic acids, and proteins. Antibodies can differentially recognize linearized polypeptides, partially or fully folded proteins, and secondary modifications of the proteins such as acylation, glycosylation, and haptenization. In contrast, T cells usually see their antigens in the context of antigen-presenting molecules. Although it has been found that MHC class II molecules can bind longer amino acid chains and even partially unfolded whole proteins such as hen egg lysozyme, usually MHC class I or II molecules bind linear peptides of 앑9 or between 15 and 22 amino acids, respectively, for presentation to T cells (Lindner and Unanue, 1996; York and Rock, 1996). Hence,

28

HESS et al.

MHC class II molecules can capture antigens and protect from proteolytic degradation those epitopes that bind in the MHC groove. Furthermore, glycosylation of proteins/peptides can influence the immune response in a positive or negative way. Glycosylation of the tetanus toxin antigen blocks its proteolytic processing by the lysosomal asparaginyl endopeptidase and hence presentation by MHC class II molecules (Manoury et al., 1998). In contrast, some T cells recognize specifically glycosylated peptides in the context of MHC class I or MHC class II molecules (reviewed in Carbone and Gleeson, 1997). Furthermore, mannose-capped proteins are preferentially taken up by APC via MMR or other pattern recognition receptors, thus enhancing the immunogenicity of these proteins ( Jiang et al., 1995; Sallusto et al., 1995; Engering et al., 1997; Tan et al., 1997). Similarly, maleylation of proteins facilitates their uptake via scavenger receptors and presentation via MHC class I (Bansal et al., 1999). Processing of maleylated proteins for MHC class I presentation is dependent on degradation in acidified endosomes, suggesting an alternative MHC class I processing pathway for such antigens (Bansal et al., 1999). Frequently, lipoproteins have been shown to be more immunogenic than the deacylated forms of the same proteins (Erdile et al., 1993; Ferru et al., 1996). Lipoproteins bias immune responses toward a Th1 type, probably due to their ability to induce IL-12 (Brightbill et al., 1999). The dogma that T cell antigens encompass only proteins was revised when it was found that certain T cell populations respond to glycolipids or to phospholigands specifically (see Section IV.C). Glycolipids are evolutionarily stable compounds which are less influenced by mutations. Furthermore, the nonpolymorphic nature of the CD1 molecules that present these antigens to T cells would facilitate their use in most vaccines. Finally, a strong adjuvant activity toward Th1 polarization has been attributed to certain microbial glycolipids, thus further improving the immunogenicity of such antigens. Hence, glycolipids represent interesting subunit vaccine candidates. Other antigens encompass those for 웂␦ T cells, such as phosphosugars and other phospholigands which are seen by these T cells independent of nominal antigen-presenting molecules (Kaufmann, 1996). Addition of these compounds to subunit vaccines would be appropriate in case a 웂␦ T cell response is desirable. 2. Conserved vs Specific Proteins It is still a matter of debate whether protective proteins to be selected for vaccine development should be exclusively expressed by pathogenic microbes or whether these antigens could be also produced by related nonpathogenic species. These two different classes of proteins are described as specific and conserved proteins. One of the most ubiquitous

EXPLOITING THE IMMUNE SYSTEM

29

conserved ‘‘protein families’’ developed during evolution is the ‘‘heat shock proteins’’ (Hsp) (see also Section V.C.1). The high cross-species homology of Hsp is best documented by their important role in supporting and maintaining cellular metabolism. The protein homology among different species is not limited to the bacterial world. Rather, mammalian Hsp share conserved regions with bacterial Hsp (Schoel and Kaufmann, 1996). Therefore, Hsp have been considered major antigens in infection and autoimmunity (Schoel and Kaufmann, 1996). To date, several reports have focused on Hsp members as vaccine antigen candidates (Noll et al., 1997). For example, vaccination with a DNA construct coding for Hsp60 of Yersinia enterocolitica succeeded in inducing protective immunity against challenge infection with the homologous pathogen (Noll et al., 1999) (see also Section V.D.1). Prophylactic as well as therapeutic immunization with mycobacterial hsp65 DNA vaccine of M. tuberculosis elicited protective immune responses, that reduced bacterial burden in M. tuberculosisinfected mice (Tascon et al., 1996; Lowrie et al., 1999) (see also Section V.D.5). Whether the protective activity of hsp-based vaccines is due solely to their antigenic activity or is enhanced by their intrinsic adjuvant activity remains to be elucidated (see Section V.D.1). Finally, and as a word of caution, use of Hsp as vaccine antigens must include consideration of their autoreactive potential due to high sequence homology between pro- and eukaryotic Hsp cognates. Autoimmune diseases caused by Hsp-specific CD4 and CD8 T cells have been described in various animal models (Van Eden et al., 1987, 1988; Steinhoff et al., 1999). Several antigens of BCG and M. tuberculosis, such as members of the antigen 85 complex, are immunologically cross-reactive (Wiker and Harboe, 1992; Hess et al., 1999). In contrast, genome analysis of M. tuberculosis revealed that one member of the proline–glutamic acid (PE) ‘‘protein family’’ showed reciprocal insertions and deletions of two peptide regions between BCG and M. tuberculosis (Cole et al., 1998). More importantly, these two putative antigens share 100% sequence homology outside the two regions (Cole et al., 1998). These peptide deletion or insertion events among highly conserved mycobacterial proteins may be a general phenomenon for members of the PE and proline–proline–glutamic acid (PPE) families and could be due to error-prone DNA replication caused by repetitive DNA sequences (Cole et al., 1998). From an immunological point of view, distinct regions among otherwise conserved proteins should also be taken into consideration in the context of new target antigens against TB because, in the case of M. tuberculosis-specific proteins, they contain additional T cell epitopes for immune recognition. In general, cross-reactivity of antigens between virulent and avirulent bacterial species could contribute to the induction and maintenance of T

30

HESS et al.

cells independent of infection with the actual pathogen (Kaufmann, 1991). In experimental murine listeriosis, this notion was recently analyzed for p60-specific CD4 and CD8 T cells by using virulent L. monocytogenes and avirulent L. innocua strains (Geginat et al., 1999). Both listerial strains express the p60 protein, which is involved in bacterial septation and probably in cell adhesion (Hess et al., 1995, 1996a; Kuhn and Goebel, 1989; Wuenscher et al., 1993). In contrast, Hly is a major virulence factor of L. monocytogenes, which is absent in L. innocua. Due to its Hly deficiency, L. innocua remains inside phagosomes of macrophages. A p60-specific CD4 T cell clone raised against purified p60 of L. monocytogenes was able to recognize a p60-specific epitope of L. innocua that differed only in the first amino acid residue. Although L. innocua booster infection expanded p60-specific memory T cells that had been induced by previous sublethal L. monocytogenes infection, primary immunization with L. innocua failed to induce p60-specific T cells (Geginat et al., 1999). These findings provide further evidence that infection with a frequently occurring environmental microbe can contribute to the maintenance of memory T cells specific for a related intracellular pathogen. 3. Antigen Expression by Recombinant Bacterial Carriers Experiments in which mice were immunized with different numbers of BCG provided evidence that the mycobacterial dose defines the Th1/Th2 nature of the immune response independent of the administration route (Bretscher, 1992; Power et al., 1998). For anti-TB vaccination, such a dosedependent polarization of the immune response should be considered important because Th1-biased defense mechanisms are central to control of intracellular bacteria. Accordingly, the type of antigen delivery which influences the amount of r-antigens expressed by heterologous microbial carriers needs to be taken into account in rational vaccine design. Generally, the abundance of an antigen expressed by an r-carrier is influenced by several parameters, including (i) bacterial vector persistence, (ii) in vivo plasmid stability, (iii) promoter strength for r-antigen expression, (iv) compartmentalization of r-antigens by the carrier strain and (v) localization of the r-carrier within APC. Each single parameter must be considered in constructing genetically stable antigen-delivery systems that induce the optimum immune response to r-proteins (Hess and Kaufmann, 1997). For the balanced induction of CD4 and CD8 T cell responses, foreign genes to be expressed by r-L. monocytogenes were inserted into the chromosome and their expression was controlled by the moderate hly promoter (Weiskirch and Paterson, 1999). Because isolates of L. monocytogenes species, which harbor natural plasmids, have not yet been found, the only way to achieve stable r-constructs for in vivo application has been chromosomal

EXPLOITING THE IMMUNE SYSTEM

31

integration of transgenes. In contrast, for r-Salmonella strains, both modes of gene expression, plasmid and chromosomal-directed antigen expression, have been used (Hormaeche and Khan, 1996). Several studies showed that chromosomal integration is not essential for r-Salmonella-based antigendelivery systems to induce MHC-class I-restricted T cell responses, provided that a pBR322-derived plasmid construct with high in vivo stability is applied (Turner et al., 1993; Gentschev et al., 1996; Hess et al., 1996b). In vivo plasmid stability gains further importance for oral vaccination strategies, because r-antigens should still be available for priming immune responses once r-Salmonella-infected APC have reached central lymphoid organs. Comparative analysis revealed that a stronger humoral immune response was induced when the r-antigen was encoded by a high-copynumber plasmid in an unstable r-Salmonella delivery system than when it was encoded by a chromosomally integrated expression cassette in a stable r-Salmonella construct (Cardenas and Clements, 1993). This notion also holds true for DNA vaccination using oral S. typhimurium strains as vehicle for delivery of eukaryotic expression plasmids (Darji et al., 1997). By enhancing the in vivo plasmid stability by means of the pBR322-based replicon, this DNA delivery device could be further improved. B. ADJUVANTS AND SUBUNIT VACCINES Pathogenic microorganisms consist of various immunogenic components, including proteins, glycoproteins, polysaccharides, and glycolipids. Only a finite number of these antigens are likely to induce immune responses that will contribute to protection. Moreover, some antigens could be responsible for detrimental effects and should be excluded from vaccine development. Therefore, the selection of the correct immunogen and its delivery in appropriate adjuvant form are essential prerequisites for an effective subunit vaccine. In general, adjuvants are compounds that, when used in combination with specific vaccine antigens, enhance the resultant immune response. With regard to vaccination against intracellular bacteria, the Th1 rather than the Th2 quality of the immune response elicited by an adjuvant–antigen formulation is decisive for success. Although only alum is allowed for use in humans at present, application of adjuvants represents a well-established practice for improving immune responses in experimental immunology. Adjuvants can be classified by their mechanism of action. Alum or biodegradable polymer microspheres provide sustained release of the immunogen. Emulsions, liposomes, saponins, or monophosphoryl lipid A can be classified as surfactant-like. These types of adjuvants have been used in subunit vaccine formulations against TB (Horwitz et al., 1995). For example, Syntex Adjuvant Formulation, an oil-in-water emulsion, was used to

32

HESS et al.

immunize guinea pigs intradermally (i.d.) with purified extracellular proteins of M. tuberculosis (Horwitz et al., 1995). Vaccination with the 30-kDa major secretory protein (also termed antigen 85B), either alone or in combination with other extracellular proteins of M. tuberculosis, induced substantial protective immunity against aerosol challenge with M. tuberculosis (Horwitz et al., 1995). Syntex Adjuvant Formulation may target antigens to APC in draining lymph nodes and probably even to more distant lymphoid tissues (Vogel and Powell, 1995). Alternatively, MPL, a lipid A derivative with a highly amphiphilic structure (Vogel and Powell, 1995), isolated from the lipopolysaccharide of Salmonella minnesota R595 strain, has been used in combination with culture filtrate proteins (CFP) of M. tuberculosis in the pulmonary TB model of guinea pigs (Baldwin et al., 1998). However, it was only in combination with r-IL-12 and r-IL-2 that the MPL–CFP preparation reduced mycobacterial counts in the lungs of guinea pigs (Baldwin et al., 1998). The enhancement of immune responses by adjuvants is commonly determined by an increase in antibody levels in vaccinated animals. Adjuvants may also lead to significant alterations of the immune response, such as augmented numbers of epitopes that are recognized, changes in the antibody isotype profiles, and induction of T cell responses. Adjuvants that change the numbers of epitopes recognized and cause antibody switches include liposomes, saponins, and block polymers. Antigen-specific cellular immune responses, including CD8 T cell responses, can be strengthened by adjuvant formulations such as saponins or carriers such as liposomes (Vogel and Powell, 1995). For instance, a formulation of the immunodominant MHC class I epitope (amino acid sequence positions 91–99) of Hly of L. monocytogenes within QuilA liposomes induced CD8 T-cell-mediated protection against otherwise lethal listeriosis (Lipford et al., 1994a). Proteins encapsulated into microspheres are released more slowly than soluble formulations, thus reducing the number of booster immunizations required to achieve the desired response. When combined with the 38-kDa lipoprotein of M. tuberculosis, microspheres prepared from synthetic, biodegradable poly(L-lactide) (PLA) and copolymers of lactide and glycolipids such as poly(DL-lactide co-glycolide) (PLG) induced strong antigen-specific Th1type immune responses (Venkataprasad, 1999). With regard to oligonucleotide adjuvants, recent reported advances are based on findings published in the 1980s by Tokunaga and co-workers (Carson and Raz, 1997; Lipford et al., 1998; Tokunaga et al., 1984). In an attempt to purify the smallest entity within complete Freund’s adjuvant, composed of killed mycobacteria dispersed in mineral oil, which induces cellular immune responses (which can now be categorized as Th1 type), these researchers purified DNA from mycobacteria, which induced

EXPLOITING THE IMMUNE SYSTEM

33

IFN-웂 release by murine NK cells (Shimada et al., 1986). Fractionation of this mycobacterial DNA led to the identification of several palindromic immunostimulatory sequences (ISS), centered around an unmethylated CpG dinucleotide core, which were responsible for NK cell stimulation (Yamamoto et al., 1992). Today, these natural CpG sequences are often mimicked by synthetic phosphothionate oligodeoxynucleotides (ODN), which are coinjected with a DNA vaccine or with protein antigens to promote antigen-specific Th1 responses due primarily to potent IL-12 stimulation (Klinman et al., 1996, 1997; Roman et al., 1997; Wagner, 1999). In addition, ODN by themselves, without any specific antigenic component, were shown to induce nonspecific resistance to experimental listeriosis of mice (Elkins et al., 1999; Krieg et al., 1998). This elevated resistance persisted for up to 2 weeks, emphasizing the strong immunopotentiating activity of ISS (Elkins et al., 1999). This feature of ODN is also considered important for the i.d. route of DNA delivery via gene gun (Fensterle et al., 1999). Nevertheless, evidence indicates that ODN may exert inhibitory effects on the immune response because exposure of macrophages to ODN resulted in a decrease in antigen processing and presentation in vitro due to reduced MHC class II expression (Chu et al., 1999). Future in vivo experiments are required to clarify inhibitory and immunstimulatory features of ODN in vaccine-induced immunity. In summary, IL-12/IL-18inducing compounds favoring Th1-biased immunity and adjuvants that induce CD4 and CD8 T cell responses are valuable ingredients for vaccines targeted at intracellular bacterial pathogens. C. USE OF HOST ADJUVANTS 1. Heat Shock Proteins Heat shock proteins act as chaperones for peptides and other proteins. Members of the Hsp60 and Hsp70 families, for instance, participate in the folding and unfolding of other endogenous proteins, in the formation of multimeric protein complexes, and in protein translocation from one cellular compartment to another (Schoel and Kaufmann, 1996; Zu¨gel and Kaufmann, 1999). They have been implicated in the loading of immunogenic peptides onto MHC class I molecules for presentation to T cells. Mammalian Hsp such as Hsp70, Hsp90, and Gp96 can complex with a wide array of immunogenic peptides (Blachere et al., 1997; Tamura et al., 1997; Udono and Srivastava, 1993; Udono et al., 1994). These Hsp serve as carriers for peptides and, as corollary, Hsp–peptide complexes isolated from tumor cells have been functionally identified as potent ‘‘tumor rejection antigens.’’ Vaccination of mice with these Hsp–peptide complexes isolated from tumor cells induced specific immunity to those tumor cells

34

HESS et al.

from which the Hsp had been isolated but not to unrelated tumor cells (Blachere et al., 1997; Tamura et al., 1997; Udono and Srivastava, 1993; Udono et al., 1994). In a murine tumor model, single immunization with Gp96 isolated from 웁-galactosidase-expressing P815 cells induced CTL specific for this antigen. CTL could be induced across the MHC haplotype; both mice that were MHC identical to the Gp96 donor cells and mice of different MHC haplotype could be immunized (Arnold et al., 1995). Gp96 and Hsp90 were capable of binding a broader range of peptides, not only the natural MHC epitope but also larger precursors of the peptides comprising the MHC class I motif (Ishii et al., 1999). Moreover, the murine Hsp73 chaperone, which is located in the cytosol of cells, promotes loading of truncated antigens onto MHC class I molecules. Loading occurred in host cells with deficiency in TAP, thus bypassing the ER (Schirmbeck et al., 1997). TAP-competent and TAP-deficient cell lines were transfected with expression plasmids encoding either the wild-type large tumor antigen of SV40 virus or a truncated cytoplasmic variant of this viral protein. The truncated variant, but not the wild-type antigen, was stably associated with the constitutively expressed, cytosolic Hsp73 chaperone (Schirmbeck et al., 1997). Thus, Hsp represent valuable carriers for efficient loading of endogenous and exogenous peptides onto MHC class I molecules, including peptides from bacterial pathogens. Hence, vaccination strategies with Hsp–peptide complexes should be considered for infectious diseases which require CD8 T cell responses for pathogen clearance. Additionally, this approach could facilitate identification of protective antigens comprising a broad spectrum of MHC-binding specificity. Data from this laboratory revealed that Gp96 isolated from L. monocytogenes- or M. tuberculosisinfected macrophages induced partial protection against the homologous pathogen, indicating the feasibility of this approach (Zu¨gel et al., submitted). 2. Dendritic Cells Dendritic cells have prominent sentinel functions in vivo. This is reflected by their in situ distribution and by their capacity to capture antigen and subsequently migrate into lymphoid organs, where they stimulate CD4 and CD8 T cells (Banchereau and Steinman, 1998; Schaible et al., 1999). DC are able to cross-prime T cells to foreign proteins that require processing, including those from infectious agents and tumor cells (Inaba et al., 1993; Mayordomo et al., 1995; Moll et al., 1993; Paglia et al., 1996). Immunization with DC can break CTL tolerance, as shown in the model of hepatitis B virus (HBV) transgenic mice. Vaccination with DC was superior to that with a DNA construct coding for the HBV surface antigen (Shimizu et al., 1998). Recent data provide evidence that distinct DC

EXPLOITING THE IMMUNE SYSTEM

35

subsets differentially regulate the type of immune response in vivo (Pulendran et al., 1999). The lymphoid-related DC subpopulation induced high levels of Th1 cytokines like IFN-웂, whereas the myeloid-related DC subset produced the Th2 cytokines IL-4 and IL-10 (Pulendran et al., 1999). Several reports addressed the question whether DC can phagocytose intracellular microbes such as S. typhimurium, BCG, and L. monocytogenes (Guzman et al., 1995; Inaba et al., 1993; Marriott et al., 1999; Svensson et al., 1997). DC can take up viable bacteria, although less efficiently than macrophages (Svensson and Wick, 1999). One report described intratracheal immunization of mice with BCG-infected DC, which induced immune responses against mycobacterial antigens and led to partial protection against aerosol challenge with M. tuberculosis (Demangel et al., 1999). In addition, the vaccine efficacy of DC that had been pulsed with killed C. trachomatis was determined in mice (Su et al., 1998). Animals immunized intravenously (i.v.) with Chlamydia-pulsed DC induced Th1-mediated protection against chlamydial infection of the female genital tract similar to that obtained after infection with live microorganisms (Su et al., 1998). In conclusion, in vitro antigen- or peptide-pulsed DC represent a potential vaccination regime against bacterial pathogens. It should be noted, however, that DC-based vaccination requires use of MHC-compatible DC. This virtually excludes the use of DC as vaccines against infectious agents. However, as a presentation device to screen candidate antigens for their protective value, DC will be useful. D. DNA VACCINES Genetic vaccination has evolved as a novel technology during the last decade. Use of ‘‘naked’’ uncoated DNA that codes for an antigen under the expression control of a eukaryotic promoter was a major breakthrough toward relatively simple and efficient immunization protocols. Such DNA vaccines have proven successful as preventive and therapeutic control means in several small animal models of infectious diseases (Alarcon et al., 1999). There is also evidence to suggest immunogenicity of DNA vaccines against HBV in chimpanzees (Davis et al., 1996). Moreover, the first demonstration in healthy humans of the induction of CD8 T cells by DNA vaccination has been reported (Wang et al., 1998). Malaria-naive volunteers who were vaccinated with plasmid DNA encoding the circumsporozoite protein of Plasmodium falciparum developed antigen-specific CTL (Wang et al., 1998). This report provides a basis for further human testing of this promising vaccine technology in humans. 1. Routes of DNA Administration It is now well established that administration of naked DNA induces both humoral and cellular immune responses against encoded antigens

36

HESS et al.

from viral, bacterial, or parasitic agents (Donnelly et al., 1994; Fynan et al., 1993; Sedegah et al., 1994; Shiver et al., 1996; Ulmer et al., 1993). Nevertheless, the choice of routes for delivering plasmid DNA to induce potent immune responses to the expressed antigens is apparently restricted. DNA constructs coding for the small HBV surface antigen efficiently and reliably primed MHC class I-restricted CD8 T cells and serum antibodies in mice only when they were injected via the i.m. or s.c. route, but not when given via the i.v. and intraperitoneal (i.p.) routes (Bo¨hm et al., 1998). Several groups have reported on the failure of im and mucosal DNA vaccination to induce immune defense mechanisms which prevent pathogen entry at mucosal sites (Zhang et al., 1997; Kuklin et al., 1997; Noll et al., 1999). The i.m. injection of DNA constructs induced only systemic, and not mucosal, protective immunity. Immunization of BALB/c and C57BL /6 mice with a naked DNA vaccine encoding the Y. enterocolitica Hsp60 conferred protection against systemic Y. enterocolitica in spleen, but not in Peyer’s patches, which are the sites of mucosal entry (Noll et al., 1999). Intramuscular vaccination with a DNA construct coding for the major outer-membrane protein of C. trachomatis reduced the pathogen burden in lungs more than 100-fold following intranasal (i.n.) challenge infection with the murine pneumonitis isolate of C. trachomatis (Zhang et al., 1997). Protective immunity against C. trachomatis was accompanied by elevated serum antibodies to mouse pneumonitis elementary bodies and by significant delayed-type hypersensitivity (DTH) responses (Zhang et al., 1997). Mucosal DNA immunization via the i.n. route and i.m. administration of plasmid DNA encoding glycoprotein B of herpes simplex virus (HSV) type 1 failed to generate an immune barrier to viral invasion at the mucosa (Kuklin et al., 1997). Although i.n. DNA immunization was an effective means of eliciting mucosal antibodies, it was inferior to i.m. DNA delivery in providing protection against lethal HSV challenge via the vaginal route (Kuklin et al., 1997). In addition, i.n. DNA delivery induced MHC class I-restricted CTL responses against measles virus hemagglutinin (HA). It is noteworthy that oral or intrajejunal immunization with the HA–DNA construct induced a CTL response of smaller magnitude (Etchart et al., 1997). Plasmid DNA complexed with specific lipid compounds showed enhanced expression of the encoded model antigen, firefly luciferase, in the respiratory tract and increased levels of circulating specific IgA and IgG after i.n. DNA immunization (Klavinskis et al., 1999). The most efficient method of i.d. DNA delivery is gene-gun-based acceleration of DNAcoated gold particles into the epidermis (Fynan et al., 1993; Zarozinski et al., 1995). Gene gun inoculation of DNA vaccines induced both neutralizing antibodies and CTL responses capable of controlling influenza or lympho-

EXPLOITING THE IMMUNE SYSTEM

37

cytic choriomeningitis virus (LCMV) infection in mice (Fynan et al., 1993; Zarozinski et al., 1995). For vaccine development against intracellular bacterial pathogens, the protocols for mucosal DNA vaccination (reviewed by McCluskie and Davis, 1999) must be improved to elicit protective immune defense mechanisms that control microorganisms not only at sites of entry but also after dissemination within the body. 2. Preferential APC for DNA Delivery Several groups have attempted to elucidate the biological basis underlying the immune responses induced by gene gun delivery or i.m. injection of DNA vaccines (Akbari et al., 1999; Condon et al., 1996; Corr et al., 1996; Porgador et al., 1998). DNA immunization of mice by scarification of the ear skin or i.m. injection into the quadriceps led to in vivo transfection of keratinocytes or myocytes, respectively (Akbari et al., 1999; Corr et al., 1996). The role of somatic tissues that express protein encoded by the injected DNA constructs may be to serve as a reservoir for antigens that are then transferred to APC. In contrast, gene-gun-based or cutaneous delivery of DNA constructs into the skin allowed direct in vivo transfection of DC in draining lymph nodes, which induced cell-mediated immunity (Fig. 1) (Condon et al., 1996; Porgador et al., 1998). 3. Improved Immunogenicity of DNA Vaccines In addition to these classical approaches to the evaluation and improvement of DNA vaccination strategies (Shiver et al., 1996) and to the attempts to enhance expression of transgenes by adapting codon usage (Uchijima et al., 1998), naked DNA research has entered an extraordinarily innovative era of vaccine design. Six major fields for improving the DNA vaccine technology have emerged thus far: (i) Coinoculation of different DNA cassettes expressing known antigens as well as cytokines such as GM-CSF and IL-12. This approach enhanced the B and T cell responses to rabies virus and induced protective immunity against Leishmania major and HSV type-2 (Gurunathan et al., 1998; Sin et al., 1999; Xiang and Ertl, 1995). (ii) Immune responses induced by DNA vaccines were enhanced by coadministration of DNA expression cassettes for the 움-chemokine IL-8 and 웁-chemokines MIP-1움, RANTES and MCP-1. IL-8 enhanced CD4 T cell and antibody responses, and MIP-1움 had profound effects on antibody responses and modulated the shift to a Th2 immune response. RANTES enhanced the levels of antigen-specific Th1 and CTL responses, and MCP-1 was the most potent activator of CD8 CTL activity, in addition to

38

HESS et al.

increasing IFN-웂 and TNF움 secretion and reducing the IgG1/IgG2a ratio (Kim et al., 1998). (iii) Immune responses induced by DNA vaccines were enhanced by coadministration of DNA expression cassettes for the costimulatory proteins CD80 and CD86. This resulted in a marked increase in MHC class Irestricted CD8 CTL responses in mice and chimpanzees (Kim et al., 1998). (iv) Coadministration of DNA vaccines together with ISS–DNA sequences containing CpG motifs enhanced immune responses of Th1 type (Carson and Raz, 1997; Lipford et al., 1998) (for details, see Section V.B). (v) Prime/boost immunization schedules comprising priming with DNA vaccines and boosting with r-vaccinia constructs, based on the modified vaccinia virus Ankara, showed increased immunogenicity and protective efficacy in models of rodent malaria caused by Plasmodium yoelii or Plasmodium berghei as compared to DNA vaccination alone (Schneider et al., 1998; Sedegah et al., 1998). (vi) Mucosal DNA delivery was improved by using gram-positive or gram-negative bacteria as vehicles for DNA vaccines (Darji et al., 1997; Dietrich et al., 1998; Fennelly et al., 1999) (for details, see Section V.I). 4. DNA Vaccination against Listeriosis Intramuscular vaccination of BALB/c mice with a DNA construct coding for a fusion protein between the murine tissue plasminogen activator protein signal peptide and the nonhemolytic version of Hly induced prominent antigen-specific CTL responses and protected mice against a lethal systemic challenge with L. monocytogenes (Cornell et al., 1999). In contrast, a DNA vaccine encoding the naturally hemolytic Hly fusion protein failed to induce protective immunity via the i.m. route. Fensterle et al., (1999) reported on a similar DNA vaccination approach against experimental listeriosis in BALB/c mice, involving gene gun inoculation of a DNA construct coding for the hemolytic Hly. In this case, the hemolytic version encoded by the hly-DNA vaccine induced CD8 CTL- and IFN-웂dependent protection against L. monocytogenes in mice. Reduction of listerial load was similar to that achieved by gene gun vaccination with a DNA construct expressing p60, a second prominent antigen of L. monocytogenes carrying immunodominant CTL epitopes for H-2Kd (Fensterle et al., 1999). We assume that the hemolytic function of Hly had no impact on vaccine efficacy in cases involving the dermal route of DNA delivery. It is assumed that myocytes are the major targets of i.m. DNA vaccination, whereas DC are primarily transfected by gene-gun-based vaccination. Apparently, myocytes were more susceptible to the hemolytic activity of Hly than DC. Consistent with this, when DNA constructs encoding hemolytic Hly were targeted directly to APC via orally administered Salmonella spp.

EXPLOITING THE IMMUNE SYSTEM

39

carriers, a protective immune response against murine listeriosis ensued (Darji et al., 1997). 5. DNA Vaccination against Tuberculosis Several reports have provided evidence for the feasibility of DNA vaccines as preventive and therapeutic means of TB control (Table III). For TB prevention, a DNA construct encoding the nonsecreted mycobacterial antigen, the Hsp65 (Tascon et al., 1996), as well as DNA vaccines encoding the secreted Ag85A, Ag85B antigens, or the surface-associated phosphatebinding lipoprotein PstS3 of M. tuberculosis, have been found to induce partial protection. However, no evidence for their superior efficacy over BCG vaccination has yet been observed (Huygen et al., 1996; Kamath et al., 1999; Tanghe et al., 1999). Notably, 4 doses of the hsp65-DNA construct given im 8 weeks postinfection significantly enhanced the antimycobacterial response, reducing the load of M. tuberculosis H37Rv in organs of mice (Lowrie et al., 1999). Similarly, immunization with a DNA vaccine coding for the 22-kDa secreted protein MPT70, which may be upregulated during infection, showed therapeutic efficacy in M. tuberculosis-infected animals (Lowrie et al., 1999). Lowrie et al. (1999) additionally reported on DNA vaccination of M. tuberculosis-infected mice treated with the antibiotics isoniazid and pyrazinamide prior to i.m. inoculation of the hsp65 DNA construct. The relatively short treatment with antibiotics meant that mycobacteria were not completely cleared prior to DNA immunization (Lowrie et al., 1999). DNA vaccination caused complete eradication of the remaining M. tuberculosis microorganisms in sevTABLE III DIFFERENT VACCINE CANDIDATES AGAINST TB Vaccine Subunit vaccines Proteins ⫹ adjuvants Proteins adsorbed to microspheres DNA vaccines Live vaccines Homologous vaccines Attenuated M. tuberculosis Heterologous vaccines Auxotrophic BCG BCG expressing cytokines BCG expressing Hly

Reference Horwitz et al., 1995; Baldwin et al., 1998 Venkataprasad et al., 1999 Tascon et al., 1996; Huygen et al., 1996 Tanghe et al., 1999; Kamath et al., 1999 Pelicic et al., 1997; Berthet et al., 1998 Yuan et al., 1998 Guleria et al., 1996 Murray et al., 1996 Hess et al., 1998

40

HESS et al.

eral mice, emphasizing the high value of combined DNA vaccination and chemotherapy in the control of experimental TB. A recent report revealed that protective immune responses induced by the DNA vaccines ag85A and pstS3 follow different kinetics (Tanghe et al., 1999). Consistent with the concept that secreted proteins are preferential antigens for early immune recognition, antigen Ag85A seems to be recognized early in infection, whereas PstS3 apparently represents a late antigenic target in M. tuberculosis infection (Huygen et al., 1996; Tanghe et al., 1999). Accordingly, an improved vaccine against TB could require a combination of antigens that are expressed at different time points during infection. The genome of M. tuberculosis reveals protein families with high sequence homologies between each member (Cole et al., 1998). For instance, the nearly homologous domains of the 167 PE and PPE proteins, which should share several common T and B cell epitopes, are encoded in the M. tuberculosis genome (Cole et al., 1998). Both the PE and PPE proteins are exceptionally glycine-rich, and the PPE proteins also contain considerable amounts of asparagine, an amino acid that is generally rare in the proteome of M. tuberculosis. Asparagine is the preferred nitrogen source for M. tuberculosis, and therefore, PPE proteins may serve as storage polypeptides (Cole, 1999). Due to the remarkable homology between multiple cognates, a single protein of the PPE family could potentially induce an immune response directed at virtually all PPE family members. Provided that different PPE proteins are expressed at different times during infection and disease, a PPE subunit vaccine composed of a single antigen could target PPE cognates expressed at different stages of TB. On this note, one member of the PPE protein family, Mtb39a, has been found to elicit strong proliferative T cell and IFN-웂 responses in peripheral blood mononuclear cells from 9 of 12 purified protein derivative (PPD)-positive individuals (Dillon et al., 1999). Additionally, an mtb39a DNA vaccine induced partial protection in mice against an aerosol challenge with M. tuberculosis (Dillon et al., 1999). Yet, as screening strategy for protective antigens, DNA vaccination also has some disadvantages. Most significantly, critical immunological properties due to posttranslational modifications (e.g., induction of cell-mediated immunity and B cell responses) by lipoproteins such as PstS3 of M. tuberculosis could be missed. With respect to epitope diversity generated during antigen processing, DNA vaccination with the ag85A construct can be superior to natural M. tuberculosis infection (Denis et al., 1998). DNA vaccination directs antigen processing along the MHC class II-processing pathway to normal acidic vacuoles, whereas natural M. tuberculosis infection impairs acidification and therefore reduces proteolytic activities within

EXPLOITING THE IMMUNE SYSTEM

41

these essential processing compartments of APC. Hence, the epitope diversity of mycobacterial antigens generated within this modified microenvironment is reduced (Denis et al., 1998). In conclusion, DNA vaccination represents a powerful technique both for screening protective antigens and, if these are available, for the prevention and therapy of TB. E. LIVE VACCINES Compared to subunit and DNA vaccines, live vaccines are endowed with higher immunogenicity. For attenuated live microorganisms, usually a single inoculation at a modest dose is sufficient for a protective immune response, since the microorganisms will multiply in vivo to a sufficiently large immunogenic dose and at the same time will produce the majority of antigens seen during the natural course of infection (Weiskirch and Paterson, 1997; Rolph and Ramshaw, 1997). In addition, antigen processing and presentation mimic a natural infection, resulting in immune responses similar to those elicited in infected individuals. Live vaccines usually stimulate both humoral and cellular immunity. Moreover, because most live vaccines can be delivered via the oral route, they have the potential to induce mucosal immune responses, which are not produced by systemically administered subunit vaccines. On the other hand, many pathogens have developed elaborate strategies to evade host defense mechanisms and to perturb the immune response to their advantage. Therefore, viable vaccine candidates still possessing such evasion mechanisms will induce a suboptimal immune response. Hence, rational attenuation should result not only in the deletion of factors that directly harm the host but also in the exclusion of evasion factors that perturb the protective immune response. The two most intensively studied live attenuated bacterial vaccines belong to the genera Mycobacterium and Salmonella. BCG, the current TB vaccine, and live attenuated salmonellae, the current vaccines for typhoid fever, have also been tested for their capacity to deliver foreign antigens as multivalent bacterial carriers (see Sections V.F.2 and V.F.3). Two general classes of attenuated bacteria can be distinguished by the kinds of mutation—defined or undefined—that the microorganisms contain. S. typhi Ty21a and BCG, which are the two most widely used attenuated bacterial vaccines worldwide, are examples of vaccines generated by undefined mutations. In contrast, multiple examples exist of experimental attenuated Salmonella and Mycobacterium vaccine strains carrying defined genetic lesions. Attenuating mutations can be further differentiated based on whether the disruption affects metabolic/regulatory genes or virulence genes. Relevant to vaccine design, genetic attenuation of the bacteria may differentially affect the host response. For example, two attenuated

42

HESS et al.

Salmonella mutant strains ( phoP vs aroA) have been found to elicit different types of immune responses in mice (Vancott et al., 1998). 1. Salmonella Mutants Because typhoid fever is still prevalent in various countries, numerous attempts have been made to develop a safe and effective S. typhi livevaccine strain that can be administered via the oral route. The generation of different S. typhi vaccine candidates with defined genetic lesions has been facilitated by the fact that the genus Salmonella, in contrast to Mycobacterium spp., is easily amenable to genetic engineering. However, testing of live-attenuated S. typhi strains as vaccine candidates is hampered by the failure of this bacterium to multiply in nonprimate hosts unless given in highly artificial ways (i.p. injection into mice of either very large inocula or small inocula plus hog gastric mucin). Due to this marked host restriction and hence the lack of a small animal model, most of the studies analyzing Salmonella host–pathogen interactions are conducted with S. typhimurium (literally, mouse typhoid). This organism is a naturally occurring pathogen of mice that causes an enteric fever similar to typhoid fever in humans. A somewhat confusing aspect of this model is that S. typhimurium is also a pathogen in humans, where it leads to self-limiting gastroenteritis and not to typhoid fever. Although the murine model of S. typhimurium has been extensively used for the analysis of pathogenesis, immunity, and candidate vaccines in typhoid fever, it has its limitations. S. typhi has a polysaccharide capsule, called virulence (Vi) antigen, and S. typhimurium does not; reciprocally, S. typhimurium possesses a virulence plasmid, and S. typhi does not. However, the relevance of these molecules in pathogenesis is not clear. While the high degree of host restriction complicates the evaluation of live oral typhoid vaccines, it represents an additional safety feature of S. typhi-based vaccines since the risk of uncontrolled spread from the initial vaccinee to a nonhuman reservoir is reduced. This safety feature of S. typhi emphasizes an important aspect that is frequently neglected in designing live vaccines. Successful eradication of a pathogen can be achieved only when the host spectrum is restricted to humans or when intermediate hosts can be controlled as well. This was the case for the only human pathogen eridicated thus far, smallpox. However, while we are generally concerned with safety in terms of adverse reactions in humans, we tend to neglect that live, attenuated vaccines also require safety assessment and risk analysis before release into the environment (Killeen et al., 1999). a. Undefined Mutations. The first live vaccine against typhoid fever, S. typhi Ty21a, is the only example of a chemically induced attenuation successfully employed in a licensed bacterial vaccine product. The chemical

EXPLOITING THE IMMUNE SYSTEM

43

mutagen nitrosoguanidine was used to introduce random mutations (Germanier and Fuer, 1975). Ty21a is administered in three or four oral doses and is well tolerated in humans. The vaccine elicits both local and systemic humoral immunity as well as cellular responses and confers significant, although incomplete, protection from typhoid fever. This strain was found to be highly effective in a controlled field trial in children in Alexandria, Egypt (Wahdan et al., 1982), whereas a subsequent trial in Santiago, Chile, showed wide variations in efficacy (Ferreccio et al., 1989) and a lack of protection was reported in European travelers (Steffen et al., 1981). S. typhi Ty21a was reported to be a galE mutant (UDP-glucose-4epimerase). Due to their deficient galactose synthesis, the bacteria cannot produce LPS that contains O antigen. However, gal⫹ revertants of Ty21a are still avirulent and protective in mice and when a defined galE mutant of S. typhi was constructed, it proved to be fully virulent (Hone et al., 1988; Silva et al., 1987). Similarly, restoration of Vi antigen expression does not increase immunogenicity or virulence in volunteers (Tacket et al., 1991). These data suggest that the attenuation of S. typhi Ty21a correlates with undefined mutations other than galE. Because of these difficulties in determining the genetic basis underlying the attenuation of S. typhi Ty21a, all subsequent approaches to developing a live attenuated S. typhi vaccine are based on defined genetic lesions. b. Defined Mutations. It is obvious that for pathogenic bacteria that cause disease by extensive replication in the tissues of an infected host, mutations that will reduce multiplication in host tissues can result in attenuation. All currently available vaccine candidates of attenuated S. typhi were constructed by the introduction of complete and nonreverting mutational blocks in either biosynthesis pathways or virulence genes. However, the following paragraphs will show that the combination of genetic lesions in a vaccine candidate strain must be chosen carefully to generate a successful typhoid vaccine that is both well tolerated and highly immunogenic. A variety of S. typhi strains with reduced virulence have been created and analyzed in animal models and human trials. By and large, the results suggest that the most important determinant of the immunogenicity of these Salmonella live-vaccine candidates is their ability to colonize the small intestine and Peyer’s patches, whereas persistence in host tissues and spread to the systemic compartment appears to be less crucial (Redman et al., 1996; Sigwart et al., 1989). 1. Mutations in metabolic genes. During multiplication in the infected host, Salmonella must acquire all the nutrients necessary for its growth. Consequently, mutations impairing metabolic pathways can potentially interfere with replication in the host and thus reduce virulence, provided

44

HESS et al.

the required metabolites are absent from the host compartment that the bacteria reside in or are at levels insufficient for bacterial growth (Stocker, 1993). As early as 1950, Bacon and his colleagues reported that three classes of induced auxotrophic mutants of S. typhi were attenuated in mice: mutants requiring purine, those responding to aspartic acid, and those dependent on p-aminobenzoic acid (Bacon et al., 1950). However, little attention was paid to this work until Hoiseth and Stocker (1981) used transposon mutagenesis to generate an aroA deletion mutant of S. typhimurium which dependend on aromatic amino acids for growth ( p-aminobenzoic acid and 2,3-dihydroxybenzoate). The LD50 of this strain in mice is increased by a factor of 106 compared with the parental strain. Based on these results, Edwards and Stocker (1988) generated nonreverting aromatic-dependent mutants (aroA) of S. typhi strains Ty2 and CDC 1080. As a further guarantee of safety, they dramatically reduced the chance of in vivo reversion by introducing a second attenuating deletion in the purA gene. The resulting ⌬aroA his ⌬purA155 candidate oral-route livevaccine strain 541Ty and its Vi-negative mutant 543Ty were tested in a volunteer trial. Both strains were well tolerated and induced cell-mediated immunity, but elicited minimal humoral responses in a small percentage of volunteers (Levine et al., 1987). Because the strains were less immunogenic than Ty21a, they were not developed further. Subsequent analyses in S. typhimurium revealed that the poor humoral immunogenicity of 541Ty was linked to the presence of the purA mutation (O’Callaghan et al., 1988, 1990; Sigwart et al., 1989). Therefore, further vaccine development was directed toward the evaluation of various S. typhi strains, containing only aroA mutants and derivatives thereof. Two S. typhi ⌬aroC ⌬aroD double mutants were derived from S. typhi ISP1820 and S. typhi Ty21a by the Centre for Vaccine Development (CVD) and designated S. typhi CVD906 and CVD908, respectively (Hone et al., 1991). In volunteers, only a single oral dose was required for immunogenicity (Tacket et al., 1992). Strain CVD906 caused transient bacteremias, whereas CVD908 was tolerated much better with only certain formulations evoking adverse reactions (Sztein et al., 1994; Tacket et al., 1992). Given that both strains carry an identical ⌬aroC ⌬aroD mutation, this result may, at first glance, appear contradictory. However, differential reactogenicity may reflect that additional mutations outside aro may contribute to the attenuation of CVD908. Whereas strain CVD906 was derived from the S. typhi wild-type ISP1820, the strain CVD908 has S. typhi Ty2 as its parent. Ty2, although derived from a typhoid outbreak in Russia in 1913, represents a laboratory-adapted S. typhi strain which is likely to have accumulated additional attenuating mutations during the many years of in vitro passage.

EXPLOITING THE IMMUNE SYSTEM

45

For example, Ty2 was shown to carry a mutation in RpoS, an alternative sigma factor required for virulence. Recent analyses showed that this mutation in S. typhi Ty2 must have been acquired before the generation of Ty21a in 1975, since both strains carry the same rpoS mutant allele (Robbe-Saule and Norel, 1999). Further derivatives of S. typhi CVD906 and CVD908 were constructed by incorporating a deletion in htrA, a gene encoding a bifunctional stress protein/serine protease (Tacket et al., 1997b). In a volunteer study, the resulting strains, CVD906-htrA and CVD908-htrA, were well tolerated across all tested dose ranges, but immunogenicity was reduced compared with that of their progenitors (Tacket et al., 1997b). A recent study by Lowe and co-workers (1999) has shown that htrA influences S. typhi virulence in vivo. The comparative analysis of two candidate S. typhi-based live oral vaccine strains, BRD691 (S. typhi harboring mutations in aroA and aroC ) and BRD1116 (BRD691 ⌬htrA), both in vitro (Caco-2 and HT-29 cells) and in vivo (human intestinal xenografts in SCID mice), revealed that loss of the HtrA protease reduces intracellular survival in epithelial cells. BRD1116 and other S. typhi htrA strains are currently undergoing phase 1 clinical trials. An additional candidate live oral S. typhi vaccine strain, Chi3927 (⌬cya ⌬crp), which carries mutations in the genes encoding adenylate cyclase and cAMP receptor protein, was constructed by Curtiss and co-workers. However, when evaluated in volunteers, Chi3927 caused unacceptable adverse reactions including fever and bacteremia (Tacket et al., 1992a). A different attenuating mutation in the dam gene (DNA adenine methylase) has also been described (Heithoff et al., 1999). DNA adenine methylases are highly conserved in many pathogenic bacteria and therefore represent potential targets for both vaccines and antimicrobial agents. Thus far, dam mutants have been evaluated as live vaccine candidates only in the murine model of typhoid fever. In S. typhimurium, Dam has been shown to regulate the expression of at least 20 genes known to be induced during infection. S. typhimurium lacking DNA adenine methylase are fully proficient in colonizing mucosal sites, i.e., Peyer’s patches, but show severe defects in colonization of deeper tissues. These dam mutants were totally avirulent and were effective as live oral vaccines against murine typhoid. 2. Mutations in virulence genes. In contrast to metabolic attenuation, genetic lesions in virulence genes generally do not affect bacterial multiplication in vivo. The S. typhi strain Ty800 is the first and so far the only example of a live vaccine candidate for typhoid fever that is exclusively attenuated for virulence but metabolically intact. This strain harbors a mutation in the two-component regulatory system PhoP/Q. When evaluated in volunteers as a live attenuated vaccine against typhoid fever, Ty800

46

HESS et al.

proved to be safe, and single oral doses are highly immunogenic in humans (Hohmann et al., 1996). In parallel, another S. typhi strain, TyLH445, which combined the metabolic aroA deletion with the phoP/Q virulence attenuation was generated (Hohmann et al., 1996). This strain was found to be safe, but ineffective at provoking good immune responses even after two oral doses due to overattenuation. Another attenuated S. typhi strain, Chi4073, is a descendant of the ⌬cya ⌬crp mutant S. typhi Chi3927 (Tacket et al., 1997a). Chi4073 contains an additional deletion of cdt, a virulence gene required for the dissemination of S. typhi from the gut-associated lymphoid tissue to deeper organs. In volunteer studies, the strain was found to be immunogenic and better tolerated than its progenitor. In summary, the data presented previously suggest that mutations in virulence genes rather than metabolic attenuation will form the basis for the development of future live bacterial vaccines. In addition, it is crucial for the immunogenicity of the vaccine strain that the bacteria can successfully colonize the small intestine and Peyer’s patches. Genetic lesions in metabolic genes generally compromise bacterial replication and consequently also reduce immunogenicity. The degree of attenuation is hard to predict, because the bioavailability of the respective metabolite within the human body is generally unknown. In contrast, certain virulence genes do not affect in vivo multiplication and immunogenicity. Therefore, safer and more effective live bacterial vaccine candidates may result from the elucidation of novel genes encoding virulence functions. 2. BCG and Deletion Mutants of M. tuberculosis a. BCG. With nearly 2 billion immunizations, the anti-TB vaccine BCG has a long record of safe use in humans. It can be administered at birth in only a single dose and achieves long-lived immunity. Successful BCG vaccination causes minor lesions, local self-limiting bacterial multiplication, and DTH to mycobacterial protein preparations such as PPD which may persist for several years. In the absence of BCG vaccination, the DTH response to PPD is considered a tentative indicator of M. tuberculosis infection, and hence BCG is not recommended for general TB control in countries with low TB incidences, such as the European Union member states or the United States. Although BCG represents the most widely used viable vaccine, its protective value is still questionable. General agreement exists that BCG can protect against, or at least ameliorate, severe forms of systemic TB in children, particularly meningitis (Huebner, 1996). However, it seems to be of low or no protective value in adults (Colditz et al., 1994). The most prevalent form of TB, namely, reactivation of latent

EXPLOITING THE IMMUNE SYSTEM

47

pulmonary TB in adults, cannot be prevented by BCG in a satisfactory way (Parrish et al., 1998). Comparison by a meta-analysis of various controlled trials revealed that the average protective efficacy of BCG in adults reaches 50% with an efficacy range from ineffective to 80% protection (Colditz et al., 1994). Several reasons could contribute to these differences: (i) the genetic variability among, and different ages of, the vaccinees; (ii) the immunological cross-reactivity between BCG and environmental mycobacterial strains prevalent in different parts of the world; (iii) latent M. tuberculosis infection in vaccinees; (iv) lack of comparability between different vaccination studies due to the use of different BCG strains, variable doses, and different immunization schedules. Genetic differences between BCG and M. tuberculosis provide several possible explanations for the failure of BCG as a vaccine (see in detail Section VI.A). From an immunological point of view, one of BCG’s major drawbacks is its failure to stimulate anti-mycobacterial CD8 T cell responses effectively, which are necessary for M. tuberculosis control. Notably, the course of BCG infection in 웁2-microglobulin-deficient mice lacking CD8 T cells revealed that CD4 T cells are virtually sufficient for the control of BCG (Flynn et al., 1992; Ladel et al., 1995). Taken together, it is evident that the balanced combination of CD4 and CD8 T lymphocytes required for protection against TB cannot be induced by the current BCG vaccine. b. Modified BCG or Attenuated M. tuberculosis Strains. The capability of viable recombinant antigen-delivery systems or homologous vaccine strains to induce efficient cell-mediated immune responses is beyond question. Two principal issues should be taken into consideration in pursuing a strategy toward an attenuated bacterial r-carrier strain: (i) an antigen could concomitantly perform virulence function, and therefore genetically modified strains may cause disease; and (ii) the deletion of genes coding for important antigens could result in attenuated homologous vaccine cognates with insufficient immunogenicity. The recent success in the generation of gene-deletion mutants of M. tuberculosis by allelic exchange and transposon mutagenesis is promising (Table III) (Berthet et al., 1998; Pelicic et al., 1997). These mycobacterial mutant strains do not survive in the host, either because they are auxotrophic, e.g., due to purC deficiency (Pelicic et al., 1997), or because they have lost their virulence for unknown reasons (as is the case in erp or acr deletion mutants (Berthet et al., 1998; Yuan et al., 1998). Erp represents a secreted protein of M. tuberculosis, while the 14-kDa Acr protein of M.

48

HESS et al.

tuberculosis is related to the 움-crystallin family of low-molecular-weight Hsp (Verbon et al., 1992). Notably, Acr is primarily produced during the stationary growth phase in vitro but undetectable during logarithmic growth of M. tuberculosis. By growing bacilli at defined oxygen concentrations, acr transcription was strongly induced under mildly hypoxic conditions and during in vitro infection of macrophages (Yuan et al., 1998). The precise functional basis of the roles of both Erp and Acr in virulence and persistence remains to be established. In addition, no data are yet available regarding the protective capacity of such attenuated M. tuberculosis strains. Transposon mutagenesis was also used to generate enhanced attenuated strains of BCG for their potential use in HIV-1-infected individuals during the asymptomatic phase of disease (Berthet et al., 1998; Guleria et al., 1996). To this end, the clearance of these BCG strains in immunodeficient mice was analyzed. SCID (severe combined immunodeficiency disease) mice lacking virtually all T and B cells were infected with met- and leuauxotrophic BCG strains and were able to control these infections for at least 230 days. In contrast, all SCID mice succumbed to a conventional BCG vaccine within 8 weeks (Guleria et al., 1996). An obvious alternative approach is to improve the immunogenicity of BCG by genetic engineering (see also Section V.F.2). Recombinant BCG strains that express cytokines such as IFN-웂 and IL-2 were constructed in an attempt to evoke more potent immune responses against M. tuberculosis (Murray et al., 1996). In an effort to improve access to the MHC class I pathway of antigen processing (see Section V.H), r-BCG strains that secrete a hemolytic fusion protein containing Hly of L. monocytogenes were generated (Hess et al., 1998). Hly secretion did not allow bacterial escape from the phagosomal vacuole, yet it enhanced presentation of co-phagocytosed soluble ovalbumin to CD8 T cells, suggesting that the release of Hly into the phagosome improved the translocation of antigen into the MHC class I pathway (Hess et al., 1998). Future experiments should be directed at clarifying whether these r-BCG strains possess increased vaccine efficacy against TB in experimental guinea pig and mouse models. Preliminary data from an in vitro system consisting of human macrophages or DC infected with Hly-secreting BCG revealed improved CTL responses (Conradt et al., 1999). An additional strategy for improving the current BCG vaccine is based on the recent knowledge of which genes are absent from BCG compared to M. tuberculosis (Behr et al., 1999; Cole, 1999). It is an obvious goal to endow BCG with M. tuberculosis-specific genes to enhance its immunogenicity and protective efficacy against TB. Obviously, genes encoding putative virulence factors of M. tuberculosis should not be introduced in active form into the genome of BCG to maintain its attenuated state.

EXPLOITING THE IMMUNE SYSTEM

49

F. BACTERIA AS ANTIGEN CARRIERS BCG as well as attenuated strains of L. monocytogenes and S. typhimurium represent effective vehicles for delivery of heterologous antigens due to their preferred intracellular replication in professional APC such as macrophages or DC (Fig. 2) (Kaufmann, 1998; Guzman et al., 1995; Inaba et al., 1993; Schaible et al., 1999; Svensson et al., 1997). The subcellular localization within APC differs among these bacteria. L. monocytogenes resides within the cytoplasmic compartment of APC, whereas S. typhimurium and BCG persist in the phagosomal vacuole (Hess and Kaufmann, 1997). The preferred intracellular location of these vector strains determines the trafficking of bacterial antigens through different MHCprocessing pathways. S. typhimurium and BCG target their antigens mainly

FIG. 2. Major delivery pathways for antigens expressed by viable r-bacterial carriers. Shigella spp., L. monocytogenes (Listeria), and S. typhimurium secreting hemolytic Hly (Salmonella-Hly) are endowed with phagosomal escape functions for antigen delivery directly into the cytoplasm of host cells and therefore preferentially stimulate MHC class I-restricted CD8 T cells. In contrast, Salmonella spp. and BCG remain in a modified phagosome, thus delivering antigens to MHC class II-loading compartments for preferentially inducing CD4 T cells. Probably BCG and Escherichia coli equipped with Hly (BCG-Hly, E. coli-Hly) can introduce antigens into both MHC class I and class II processing. Abbreviations: Hly, listeriolysin; ER, endoplasmic reticulum.

50

HESS et al.

to the MHC class II presentation pathway and therefore stimulate CD4 T cells predominantly. In contrast, L. monocytogenes-derived proteins are primarily introduced into the MHC class I-processing machinery which leads to CD8 T cell stimulation (Hess and Kaufmann, 1997). However, with respect to the induction of CD4 versus CD8 T cell responses, this compartmentalization should not be taken as absolute. Various antigendelivery systems based on S. typhimurium (Hess et al., 1996b; Hormaeche and Khan, 1996) and BCG (Aldovini and Young, 1999; Hess et al., 1998) that allow antigen-specific stimulation of CD8 T lymphocytes in addition to the prominent CD4 T cell induction have been described. Vaccine efficacy of r-strains of attenuated L. monocytogenes, S. typhimurium, or BCG is corroborated by their ability to induce potent Th1-like cytokine responses. By using these bacterial vectors, the r-antigens are not only targeted to appropriate pathways of MHC class I and class II antigen processing, but also stimulate the innate immune system to provide the adequate cytokine milieu and appropriate expression of costimulatory molecules for promoting the protective immune response. Nonproteinaceous immunogenic components of these carrier strains, for instance, BCG with its phospholigands for recognition by 웂␦ T cells or glycolipids for stimulating T cells via CD1 binding, not only provide potential target antigens but also represent moieties with additional immunostimulating properties (Hoffmann et al., 1999). Therefore, attenuated r-bacterial strains expressing heterologous target antigens are effective devices for protecting against challenge with bacterial, viral, or parasitic pathogens that are controlled by Th1 cell responses. 1. L. monocytogenes as Antigen Carrier L. monocytogenes resides not only within macrophages but also in certain nonphagocytic cells such as hepatocytes or epithelial cells. Approximately 2 h after host-cell invasion, L. monocytogenes escapes from the vacuolar compartment into the cytoplasm, primarily through the action of the secreted virulence factor, Hly (Gaillard et al., 1987). Due to its cytosolic localization, L. monocytogenes targets its antigens directly to the classical MHC class I-processing and presentation pathway via the ER and Golgi network (Berche et al., 1987; Pamer et al., 1991; Pamer, 1994). For this reason, L. monocytogenes predominantly stimulates CD8 T cells, while CD4 T cells are activated to a lesser degree (Ladel et al., 1995). Therefore, it is obvious that L. monocytogenes should be used as r-carrier for the stimulation of antigen-specific immunity against viral diseases or tumors that are controlled mainly by cell-mediated immunity with emphasis on CD8 T cells (Hess and Kaufmann, 1997; Paterson, 1999). Recombinant L. monocytogenes strains have been constructed that secrete the nucleopro-

EXPLOITING THE IMMUNE SYSTEM

51

tein (NP) of LCMV or influenza virus under control of the hly promoter inserted into the listerial virulence gene cluster. Vaccinated mice were able to clear LCMV infection, suggesting that sterilizing immunity was achieved (Shen et al., 1995; Slifka et al., 1996). Mice immunized with rL. monocytogenes expressing the NP antigen of influenza were able to clear influenza infection more rapidly, as indicated by reduced titers in the lungs 5 days after challenge (Ikonomidis et al., 1997). Similarly to the LCMV and influenza infection models, antigen-specific CTL responses were induced by r-L. monocytogenes secreting the p55 HIV-gag gene product of HIV-1 (Frankel et al., 1995). For exploiting r-L. monocytogenes as carriers for anticancer vaccines (reviewed in Weiskirch and Paterson, 1999), mice were immunized with r-L. monocytogenes secreting NP protein of influenza virus and challenged with NP-transfected tumor cells. The observed tumor regression was mediated by NP-specific CD8 CTL (Pan et al., 1995a,b). In another cancer model in mice, oral immunization with an attenuated L. monocytogenes mutant expressing 웁-galactosidase as tumor-associated antigen triggered long-lasting immunity against a murine fibrosarcoma expressing this antigen (Paglia et al., 1997). Indeed, oral vaccination with r-L. monocytogenes was more effective in inducing antitumor immunity than either immunization with peptides in complete Freund’s adjuvant or loaded onto DC (Paglia et al., 1997). On basis of these data, the potential use of attenuated L. monocytogenes as an antigen-delivery system should be further exploited for diseases that are predominantly controlled by CD8 T cells. 2. BCG as Carrier for Heterologous Antigens At the beginning of the 1990s, two groups succeeded in constructing Escherichia coli–mycobacteria shuttle vectors capable of expressing foreign antigens (Aldovini and Young, 1991; Stover et al., 1991). Since then, several studies have demonstrated that strong cellular and humoral immune responses can be induced against heterologous proteins expressed by BCG. With BCG being the vaccine currently used against TB, the question arises whether immunization with r-BCG strains would successfully induce immune responses to heterologous antigens in populations that were already BCG-vaccinated. With 웁-galactosidase as marker antigen for r-BCG, it was found that proliferative T cell responses were suppressed to approximately 50% of those in naive animals, whereas antibody responses were enhanced (Gheorghiu et al., 1994). These results could mean that preexposure to BCG can reduce T cell responses but is not a limiting factor, especially when antibody responses are important for control of the target pathogen.

52

HESS et al.

Recombinant BCG vaccine candidates have already been developed against pneumonia and Lyme disease caused by Streptococcus pneumoniae or Borrelia burgdorferi, respectively (Langermann et al., 1994a,b; Stover et al., 1993). These diseases are mainly controlled by antibodies. The pneumococcal surface protein A (PspA) of S. pneumoniae and the outer-surface protein (OspA) of B. burgdorferi were expressed by these r-BCG strains. More recently, r-BCG constructs expressing glutathione S-transferase (GST) of Schistosoma haematobium or Schistosoma mansoni induced mixed neutralizing anti-GST serum antibodies of different isotypes such as IgG1, IgG2a, IgG2b, and IgA (Kremer et al., 1996, 1998). After i.n. administration, high levels of anti-GST IgA were found in the bronchoalveolar lavage fluid, demonstrating that r-BCG was capable of inducing long-lasting secretory and systemic immune responses to antigens expressed intracellularly (Kremer et al., 1998). Expression of foreign antigens with the signal sequence of the 19-kDa lipoprotein of M. tuberculosis improved induction of humoral immunity in mice consistent with a B-cellactivating capacity of these lipid residues (Langermann et al., 1994a; Stover et al., 1993). The potential immunostimulatory effects of the 19-kDa lipoprotein have also been shown for the generation of T cell responses as indicated by markedly increased IL-12 production by human macrophages (Brightbill et al., 1999). The IL-12 induction in macrophages is mediated by Toll-like receptors (Brightbill et al., 1999). Note, however, that a rBCG strain expressing a fusion protein between OspA of B. burgdorferi and the 19-kDa lipoprotein failed to elicit primary antigen-specific antibody responses in humans (Edelman et al., 1999). In this first phase 1 study with r-BCG microorganisms, the low immunization doses that had to be used for safety reasons induced a PPD-positive skin test in only half of the vaccinees. This relatively low seroconversion rate could also explain the ineffective induction of OspA-specific antibodies in r-BCG-vaccinated individuals (Edelman et al., 1999). The capacity of the r-BCG carrier strain to efficiently stimulate cellmediated immunity against antigens from infectious agents has been reported: (i) Somatic expression of the leishmanial Gp63 protein by r-BCG evoked potent protective immunity to L. major challenge in resistant and susceptible mice, suggesting that CD4 T cells of Th1-type were induced (Abdelhak et al., 1995; Connell et al., 1993). (ii) Immunization with r-BCG expressing the simian immunodeficiency virus (SIVmac) Gag protein elicited a Gag-specific CTL response in rhesus monkeys (Yasutomi et al., 1993, 1995).

EXPLOITING THE IMMUNE SYSTEM

53

(iii) Vaccination with r-BCG expressing SIVmac Nef protein induced CTL responses against Nef synthetic peptides in mice (Winter et al., 1995). (iv) Recombinant BCG producing the HIV-1 Env-V3-loop epitope evoked Env-specific CD8 CTL responses in mice (Honda et al., 1995; Kameoka et al., 1994). (v) r-BCG strains expressing the amino-terminal half of Env antigen of SIVmac induced neutralizing antibodies and CTL responses in mice (Lim et al., 1997). Moreover, guinea pigs immunized by the oral route produced significant levels of Env-specific IgA, demonstrating that r-BCG is able to induce local humoral immunity in the intestinal mucosa (Lim et al., 1997). In summary, if further attenuated by simultaneously improving its immunogenicity, BCG may serve as a valuable vaccine carrier for heterologous antigens originating from pathogenic microorganisms in populations of unknown HIV-status (see in detail Section V.E.2.b). 3. Attenuated Salmonella as Carrier Strains for Antigen Delivery In preclinical studies, attenuated Salmonella are the most intensively studied bacterial vectors with more than 35 bacterial (e.g., Yersinia pestis, Bordetella pertussis, Helicobacter pylori), 15 viral (e.g., HIV-1, HSV, influenza, HBV), and 15 parasitic antigens (e.g., P. falciparum, L. major, S. mansoni) expressed (for a review, see Hone et al., 1999; Hormaeche and Khan, 1996). All currently used live Salmonella vaccine vectors tested as carriers for heterologous antigens are derived from the new generation of live anti-salmonellosis vaccines with defined, nonreverting genetic lesions in known attenuating genes (described in Section V.E.b). However, novel attenuated Salmonella vectors continue to be developed. For instance, two new live attenuated S. typhimurium strains carrying mutations in the genes sseC or sseD, which encode potential effector proteins of the type III secretion system encoded by Salmonella pathogenicity island 2, have recently been evaluated (Medina et al., 1999). In addition to inducing mucosal and systemic antibody responses against heterologous antigens, Salmonella vectors elicit antigen-specific CD4 and CD8 T cell responses. The ability of live Salmonella vectors to stimulate MHC class I-restricted CD8 T cell responses has received major attention, because this pathogen remains endosome-bound within the host cell (Alpuche-Aranda et al., 1995; Buchmeier and Heffron, 1991) and was not believed to have access to MHC class I antigen presentation pathways originating in the cytoplasm (Turner et al., 1993; Harding and Pfeifer, 1994). However, several groups have independently demonstrated the generation of MHC class I-restricted CD8 T cells after vaccination with

54

HESS et al.

live Salmonella vectors carrying viral (Valentine et al., 1996), bacterial (Catic et al., 1999; Verma et al., 1995), and parasitic antigens (Aggarwal et al., 1990; Flynn et al., 1990; Gonzalez et al., 1994). The majority of efforts to improve the ability of Salmonella to elicit MHC class I-restricted immune responses have concentrated on the efficient delivery of antigens to the host-cell cytosol. One way to achieve this aim is by directing whole bacteria out of the phagosome into the cytosol by creating r-Salmonella strains that express Hly of L. monocytogenes (for details, see Section III.C). Alternatively, the antigen of interest can be secreted. Antigen localization plays a decisive role in optimizing the immune response against heterologous antigens. Secreted antigens are preferable to somatically expressed antigens for two reasons. First, they are more readily processed by APC during in vivo expression. Second, secreted heterologous proteins should not interfere with the housekeeping functions of the vaccine carrier, while intracellular accumulation of somatically expressed antigens could potentially compromise the overall fitness of the live vector. Two specialized, sec-independent systems have been established for the secretion of heterologous antigens in Salmonella. The E. coli 움-hemolysin (HlyA) HlyB/HlyD/TolC secretion apparatus represents a type I secretion system that allows the export of heterologous antigens (ranging between 10 and 1000 amino acids in size) fused to the C-terminal secretion signal of HlyA (Gentschev et al., 1990, 1996; Hess et al., 1990). Alternatively, the contact-dependent type III secretion system encoded by Salmonella pathogenicity island 1 has been successfully used to deliver MHC class Irestricted, viral epitopes to the host-cell cytosol (Russmann et al., 1998). The HlyA secretion system has been adapted for export of heterologous antigens in various gram-negative bacteria, including Salmonella (Gentschev et al., 1997), Shigella (Tzschaschel et al., 1996), and Vibrio (Ryan et al., 1997). A defined set of attenuated S. typhimurium strains exists that allows secretion, surface expression, or cytoplasmic arrest of the antigen of interest, depending on the constructs and carrier strains used. The immunological consequences resulting from differences in localization of the displayed antigens are discussed in Section V.G. The following heterologous antigens delivered by attenuated Salmonella strains via the HlyA secretion system were shown to elicit protective immune responses against the respective intracellular pathogens in animal models of infection: Hly, p60, and superoxide dismutase (SOD) against L. monocytogenes (Hess et al., 1996b, 1997); antigen 85B against M. tuberculosis (Hess et al., in press); and p67 sporozoite antigen against Theileria parva (Gentschev et al., 1998).

EXPLOITING THE IMMUNE SYSTEM

55

The delivery of epitopes via the S. typhimurium SPI1 type III secretion apparatus has, to date, been tested only in the viral system. Two viral epitopes (influenza virus nucleoprotein residues 377–374 and LCMV NPresidues 118–126) were introduced at a permissive site of SptP. This S. typhimurium effector protein is injected into the host cell via the contactdependent type III secretion apparatus but it is not required for efficient bacterial entry into epithelial cells (Fu and Galan, 1998; Kaniga et al., 1996). Oral immunization of the attenuated aroA sptP S. typhimurium strains expressing SptP–LCMV NP118–126 efficiently stimulated MHC class I-restricted immune responses, protecting vaccinated animals against lethal intracerebral challenge with virulent LCMV (Russmann et al., 1998). Future application of this system will expand the efficient use of S. typhimurium as a carrier of heterologous antigens to vaccinate against diseases in which CD8 T cell responses are crucial for protection, including tumors, and infections with intracellular bacteria. Although the size of the heterologous antigens delivered as chimeric type III effectors is currently restricted to epitope length, this delivery system increases the potential of attenuated Salmonella strains to serve as multivalent vaccine carriers. Combining independent antigen expression and delivery systems within the same rlive bacterial vector not only will allow stimulation of all desired effector cell populations of the immune system, but also may have important implications for the design of live multivalent bacterial vaccines against viral, bacterial, and eukaryotic pathogens combined in the same carrier. G. COMPARTMENTALIZED EXPRESSION OF ANTIGENS BY R-CARRIERS It is still an open, though relevant, question whether and how antigen display by bacterial carriers influences the type of T cell response induced. In general, antigen compartmentalization (somatic vs surface-localized) in r–S. typhimurium carriers was without apparent effect on presentation efficacy of heterologous proteins through the MHC class I pathway in vitro (Kovacsovics-Bankowski and Rock, 1995; Pfeifer et al., 1993). A different picture, however, arises under in vivo conditions. For these studies, a S. typhimurium aroA delivery system combined with the HlyB/HlyD/ TolC transport machinery was employed for comparing the influence of different expression modes (somatic vs secreted) of the listerial antigens p60 and SOD proteins on protective immunity against L. monocytogenes (Hess et al., 1996b, 1997) (see also Section V.H). The p60 antigen represents a naturally secreted protein of L. monocytogenes, whereas SOD is retained within the listerial cytoplasm (Hess and Kaufmann, 1997). Vaccination of C57BL/6 mice with r-S. typhimurium strains induced protection against lethal listeriosis, provided that these proteins were secreted by the r-Salmonella carriers. In contrast, nonsecreted forms of these anti-

56

HESS et al.

gens delivered by r-Salmonella strains failed to induce protective immunity against listeriosis. In vitro, both secreted and nonsecreted antigens delivered by r–Salmonella carriers were presented by MHC class I molecules but followed different kinetics (Catic et al., 1999). Secreted antigens were recognized by CD8 T cells more rapidly than their nonsecreted counterparts (Catic et al., 1999). Thus, the faster availability of antigens secreted by r-S. typhimurium constructs apparently accelerated recognition of infected APC by antigen-specific CD8 T cells. Even SOD-secreting r-S. typhimurium bacteria induced protective immunity against listeriosis, although this enzyme represents a naturally somatic antigen of L. monocytogenes (Hess et al., 1997). Shen and co-workers (1998) reported that both secreted and somatic NP antigens of LCMV expressed by r-L. monocytogenes are capable of inducing protective CD8 T-cell responses against LCMV. In a reciprocal immunization/challenge model, however, only the NP antigen secreted by r-L. monocytogenes, and not the somatic antigen, was recognized by CD8 T cells primed by LCMV infection (Shen et al., 1998). There are two major differences between the two types of experimental approaches (Pamer, 1998). First, secreted and somatic antigens of r-L. monocytogenes are preferentially introduced from the cytoplasm into the conventional MHC class I-processing pathway. In contrast, antigens of r-S. typhimurium localized within phagosomal vacuoles of APC probably follow alternative MHC class I-processing pathways (Pamer, 1998; Pfeifer et al., 1993; Wick and Pfeifer, 1996). Second, the antigen fusions p60 and SOD delivered by rSalmonella strains carry CD4 as well as CD8 T-cell-specific epitopes, whereas the r-L. monocytogenes constructs exclusively express the immunodominant CD8 T cell epitope of the LCMV-NP. In general, the secretion of complete antigens carrying both CD4 and CD8 T cell epitopes by rSalmonella spp. and L. monocytogenes apparently represent a potent way for inducing antigen-specific cell-mediated protection in different models of infectious diseases (see also Section V.F). Therefore, antigen secretion should be considered for expression of heterologous antigens by these bacterial r-carrier strains. H. TRANSFER OF PHAGOSOMAL ESCAPE FUNCTIONS TO DIFFERENT BACTERIAL CARRIERS Intracellular bacterial pathogens with phagosomal escape function directly introduce their antigens into the classical MHC class I presentation pathway, thus inducing CD8 T cells more efficiently than pathogens with exclusively phagosomal localization (Hess and Kaufmann, 1997). Due to the importance of the bacterial localization for stimulation of CD8 T cells, pore-forming cytolysins such as Hly of L. monocytogenes or perfringolysin

EXPLOITING THE IMMUNE SYSTEM

57

O of Clostridium perfringens, which cause bacterial escape from the phagosome into the cytoplasm of host cells, are conceivable for the construction of heterologous attenuated microorganisms (Goebel and Kreft, 1997; Jones and Portnoy, 1994; Jones et al., 1996). The poreforming toxin Hly represents the only biologically active cytolysin successfully transferred to several bacterial species. Because of the critical role of Hly for efficient antigen introduction into the classical MHC class I presentation pathway during L. monocytogenes infection (Berche et al., 1997), the transfer of this cytolysin to other bacterial species is an obvious research aim. Bielecki et al. (1990) succeeded in equipping gram-positive Bacillus subtilis microorganisms with the hemolytically active Hly. Since then, gram-positive attenuated Bacillus anthracis, the anti-TB vaccine strain BCG, gram-negative E. coli, attenuated Salmonella dublin, and S. typhimurium mutants carrying the pore-forming Hly have been constructed (Fig. 2). Attenuated r-Salmonella spp., r-B. subtilis, and attenuated r-B. anthracis engineered to express Hly mimic the intracytosolic lifestyle of L. monocytogenes in macrophages (Bielecki et al., 1990; Gentschev et al., 1995; Hess and Kaufmann, 1997; Sirard et al., 1997). In contrast, r-BCG-secreting Hly and r-E. coli microorganisms expressing a nonsecreted hemolytic Hly were not released from the phagosomal vacuole into the cytosol of macrophages (Hess et al., 1998; Higgins et al., 1999). In the case of r-BCG, the failure to escape from the phagosome could be due to the low hemolytic activity expressed by this strain, even under acidified conditions. In vivo this low hemolytic activity may be even more critical due to the maturation failure along the endosomal/lysosomal continuum caused by BCG, which in turn hinders acidification of r-BCG-containing vacuoles, additionally reducing the hemolytic activity. Nevertheless, the pore-forming activity of Hly delivered by r-BCG or r-E. coli facilitated trafficking of a bystander antigen to the MHC class I pathway in the cytosol of the host cells, which efficiently stimulated antigen-specific CD8 T cells (Hess et al., 1998; Higgins et al., 1999). For Hly expressing r-B. subtilis, it was shown that infected macrophages were recognized by L. monocytogenes-immune CD8 T cells (Bouwer et al., 1992). Moreover, the facultative intracellular carriers r-S. typhimurium and r-BCG, with different in vivo clearance rate, and the normally extracellular r-B. anthracis strain expressing the active Hly were shown to induce protection in mice against lethal listeriosis (Hess et al., 1996b; Hess and Kaufmann, unpublished results; Sirard et al., 1997). The phagosomal escape of r-S. typhimurium by Hly further improved the protective efficacy against listeriosis which was already achieved by the secretion of a nonhemolytic version of Hly retaining r-Salmonella within the phagosome of host cells (Hess and Kaufmann, unpublished results) (see also Section V.G). In summary, expression of hemolytic Hly by different carrier strains improves

58

HESS et al.

antigen introduction into classical MHC class I presentation pathway to efficiently stimulate CD8 T cells even without phagosomal escape of the heterologous microorganisms. I. BACTERIAL VECTORS FOR DNA DELIVERY Various strategies have been pursued to improve the immunogenicity of DNA vaccines. From a practical point of view, the mucosa is an important site for priming protective immune responses against many infectious agents (see also Section IV.D). So far, mucosal immunization with naked DNA via the oral or i.n. route has failed to generate a local immune barrier for pathogen invasion and failed to induce systemic protection in various animal models of infectious diseases (see in detail Section V.D.1). To overcome this limitation of ‘‘naked’’ DNA vaccines, oral or i.n. administration of DNA takes advantage of bacterial carriers for delivery. As an additional advantage, bacterial carriers possess adjuvanticity, which promotes the development of Th1 cells. Attenuated bacterial carriers with intracytosolic localization inside host cells such as the asd-deficient Shigella flexneri 2a and attenuated L. monocytogenes, as well as phagosomally persisting S. typhimurium aroA strains, were employed for targeting DNA to the mucosal immune system (Dietrich et al., 1999). These constructs contained eukaryotic expression plasmids coding for model antigens or antigens from pathogens or tumor cells. The seminal experiment analyzed Shigella-based DNA delivery to cell lines and ocular tissue which expressed the encoded 웁-galactosidase under control of the cytomegalovirus promoter (Sizemore et al., 1995). Since then, several studies have attempted to target DNA vaccines directly to APC by bacterial carriers (Darji et al., 1997; Pagilia et al., 1998; Dietrich et al., 1998; GrillotCourvalin et al., 1998; Fennelly et al., 1999). Attenuated L. monocytogenes strains have been used to direct cellular protein expression by delivering eukaryotic expression plasmids into infected host cells in vitro (Dietrich et al., 1998). The listerial mutant strain was further attenuated by addition of a plasmid which codes for the lysis protein PLY118 of a bacteriophage specific for L. monocytogenes. By combining the expression of this ‘‘suicide factor’’ with the intracellularly active promoter of the actA gene locus, availability of the DNA constructs was promoted by autolysis of the r-L. monocytogenes strains (Dietrich et al., 1998). MHC class I presentation of an immunodominant reporter epitope demonstrated the capacity of DNA delivered by suicide L. monocytogenes vectors to induce CD8 T cell responses (Dietrich et al., 1998). Attenuated S. flexneri strain carrying DNA vaccines for structural proteins of measles virus were recently shown to induce antigen-specific immune responses in mice after i.n. administration (Fennelly et al., 1999). Darji and co-workers reported oral application

EXPLOITING THE IMMUNE SYSTEM

59

of r-S. typhimurium aroA harboring an hly DNA vaccine construct that protected mice from a normally lethal listeriosis by inducing strong CD8 CTL responses (Darji et al., 1997). Oral immunization with r-S. typhimurium aroA microorganisms as carrier for 웁-galactosidase-coding DNA resulted in protective immunity against an experimental challenge with a murine fibrosarcoma expressing 웁-galactosidase (Paglia et al., 1998). Grillot-Courvalin and associates described direct transfer of DNA constructs from E. coli K12 bacteria to mammalian cells, provided that E. coli could enter these cells by means of invasin of Yersinia pseudotuberculosis (Grillot-Courvalin et al., 1998). The DNA transfer efficiency was further enhanced by the co-expression of Hly of L. monocytogenes, mainly at lower multiplicities of infection. Recombinant S. typhimurium which secreted Hly fusion proteins with pore-forming capacity generated a route for more efficient transfer of DNA constructs from the phagosome into the cytoplasm of macrophages and therefore augmented MHC class I presentation of a marker epitope (Gentschev et al., 1995; Hess et al., 1996b; Catic et al., 1999). Taken together, bacterial carriers represent versatile DNA delivery vehicles with broad application for mucosal DNA vaccination strategies. VI. From Genomes to Antigens

A. GENOMES AND DNA MICROARRAYS Elucidation of the complete genomes of pathogenic microorganisms, as well as analysis of their global in situ gene-expression profiles by DNA microarrays and their global microbial protein expression by proteomics, will enable researchers to further unravel the complexity of microbial pathogenesis and antimicrobial immunity. New virulence determinants and protective antigens will become known through these technologies. In 1995, Haemophilus influenzae became the first microbial organism to have its entire genome sequence published (Fleischmann et al., 1995). By now, numerous genomes have been sequenced and complete bacterial DNA sequences of three intracellular bacterial pathogens have been made public: the genomes of C. trachomatis and C. pneumoniae, approximately 1.04 Mb in size (http://chlamydia-www.berkeley.edu:4231) (Kalman et al., 1999; Stephens et al., 1998) and the genomic 4.4-Mb DNA sequence of M. tuberculosis H37Rv strain (Cole et al., 1998). The latter DNA sequence is available on two World Wide Web addresses: http:// www.sanger.ac.uk and the more sophisticated http://www. pasteur.fr/Bio/TubercuList. Techniques that allow the large-scale study of the entire genetic complement of microorganisms have been advanced in parallel. One of the major

60

HESS et al.

breakthroughs is DNA microarrays, a platform for functional genomics that has been reviewed in detail by Schena and co-workers (1998). This approach combines the knowledge of entire bacterial genomes with the advantages of miniaturization. For comparative genome analysis of BCG, M. bovis, and M. tuberculosis, the following DNA microarray platform was established: 4896 spots on a microscope slide representing 3902 of the 3924 open reading frames (ORFs) of M. tuberculosis H37Rv were arranged. These spots consisted of ORF internal DNA sequences amplified by polymerase chain reaction at sizes ranging from 250 to 1000 bp (Behr et al., 1999). DNA–DNA hybridization on the DNA microarray slide revealed virtually all genetic differences between M. tuberculosis H37Rv, M. bovis, and some BCG substrains (Behr et al., 1999). In total, 129 M. tuberculosisspecific ORFs were absent in 16 regions of the genome of nearly all BCG substrains (Behr et al., 1999; Cole et al., 1998). In addition to this structural genomic deficiency of BCG, it can be envisaged that BCG and M. tuberculosis have different control mechanisms for gene expression during infection and intracellular persistence (Behr et al., 1999; Cole, 1999). Principally, each of these differentially expressed ORFs with their corresponding antigens could be suitable targets for the development of diagnostic reagents, antimicrobial agents, and vaccine candidates. B. PROTEOMICS Although DNA sequence analysis of entire genomes provides the basic information on predicted gene products, the majority of these have no known function as yet. Combination of two-dimensional gel electrophoresis and mass spectrometry and the use of N-terminal amino acid sequence analysis and protein database searches have facilitated protein identification (for details see Blackstock and Weir, 1999). The proteome is the expressed protein complement of a genome and proteomics represents functional genomics at the protein level. With respect to the proteome analysis of intracellular bacterial pathogens, the following areas may be distinguished: (i) expression proteomics for the identification of yet unknown proteins (Sonnenberg and Belisle, 1997); (ii) the study of global changes in protein expression in response to altered environmental conditions for mycobacterial growth (Wong et al., 1999); and (iii) differential proteomics, the comparative proteome analysis among related species such as BCG and M. tuberculosis ( Jungblut et al., 1999). Proteins from M. tuberculosis Erdman strain grown under low- or high-iron conditions were separated by twodimensional gel electrophoresis and identified by mass spectrometry (Wong et al., 1999). The expression of at least 15 proteins was found to be selectively induced, and that of at least 12 proteins was decreased under low-iron conditions, reflecting changes in the global regulation of gene

EXPLOITING THE IMMUNE SYSTEM

61

expression in response to a single, but essential, growth parameter (Wong et al., 1999). In the first global and comparative proteome analysis between mycobacterial strains, M. tuberculosis H37Rv and Erdman strains versus BCG, 1800 spots from culture filtrate and 800 spots from bacterial lysate proteins were separated, of which 263 were identified ( Jungblut et al., 1999). Sixteen proteins differed in intensity or position between M. tuberculosis H37Rv and Erdman and 25 proteins differed between M. tuberculosis H37Rv and BCG. C. VACCINOMICS Based on the concept of ‘‘naked’’ DNA immunization, a promising approach to vaccine development was initiated by the construction of an expression library comprising Mycoplasma pulmonis-specific DNA fusions with the last exon of the gene coding for the human growth hormone (hGH). The corresponding hGH–M. pulmonis fusion antigens are secreted by in vivo transfected cell (Barry et al., 1995). M. pulmonis is a normal lung pathogen of rodents and i.m. immunization of mice with certain M. pulmonis DNA vaccine constructs rendered them resistant against M. pulmonis challenge (Barry et al., 1995). Recently, a similar i.m. expression library approach succeeded in inducing significant protection against L. major (Piedrafita, 1999). Expression library immunization makes accessible almost the complete protein pattern of the pathogen. This immunization strategy is now applied to the development of malaria vaccines (Hoffman et al., 1998). Because an appropriate animal model for P. falciparum, a major cause of human malaria, is lacking the targeted expression library immunization is pursued using ORFs of the rodent malaria species P. yoelii. The P. falciparum orthologues of protective P. yoelii genes could be identified by sequence comparison (Hoffman et al., 1998). In general, this strategy provides a means to identify protective antigens for vaccine development regardless of the pathogen, independent from the knowledge of genes and complete genomes, as well as from function of candidate antigens. This technology, therefore, accelerates screening times and thus will greatly improve identification of candidate vaccine antigens. Similar expression library immunization approaches are also being performed now in the field of TB vaccine research ( Johnston, 1998). VII. Concluding Remarks and Outlook

In this chapter we have argued that the design of vaccines against intracellular pathogens must extend its scope from the question of which antigens to select for a vaccine to the question of how to introduce the antigens. To date, successful vaccines have been developed either totally

62

HESS et al.

empirically or focused on the identification of protective antigens. This was the right approach, because protective immune responses and consequently, protective antigens could be defined by means of neutralizing antibodies, which are induced relatively straightforwardly by vaccination regimes. Whenever specific T lymphocytes take responsibility for protection, as is the case with intracellular bacterial pathogens, the search for the appropriate presentation of antigens to the immune system becomes essential. In fact, it probably becomes more important than the identification of vaccine antigens. The question of whether distinct proteins qualify as protective antigens remains unsettled. For example, the major virulence factor of L. monocytogenes, Hly, is indeed both a dominant and a protective antigen (Harty and Bevan, 1992; Pamer et al., 1991). Yet, the ubiquitous protein Hsp65 has also been shown to be a protective antigen in the case of M. tuberculosis infection despite being a typical housekeeping molecule present equally in nonpathogenic and pathogenic microbes (Lowrie et al., 1999). Although the rapidly increasing list of microbial pathogens whose genomes have been elucidated will be of great help, no predictions can be made at present about the critical features that qualify a microbial antigen as a vaccine candidate. The concept of vaccinomics has taken this notion seriously by proposing to identify protective antigens directly within the genome without prior knowledge of their existence or even their function (Barry et al., 1995; Hoffman et al., 1998). The question of how to introduce an antigen to the immune system for optimum protection includes as a major issue presentation by the adequate MHC pathways (and perhaps MHC-like pathways in adjunct) as well as induction of costimulatory molecules and the appropriate cytokine milieu. To achieve optimum conditions for protective immunity, the choice of the appropriate vaccine carrier system becomes critical. We have emphasized viable bacterial carriers and naked DNA vaccines, which we consider the most promising vaccine candidates for intracellular bacteria. Because these pathogens naturally enter the host through mucosal surfaces, vaccines aimed at controlling them as early as possible during infection should be operative at these sites. Given that mucosal immune responses are best evoked by vaccination at the very same sites, the optimal conditions for mucosal immunization need to be explored in greater detail. Similarly, we need to know more about the requirements for inducing long-lasting immunity, i.e., immunologic memory. Finally, the conditions for both pre and postinfection vaccines need to be analyzed, and the possibility that they may differ from each other must be entertained. All these issues have been too much neglected in the past. Although there is still a long way to go, the basic knowledge about the cross-talk between bacterial pathogens and the host immune system will

EXPLOITING THE IMMUNE SYSTEM

63

provide the guidelines for rational vaccine development. Without doubt, in pursuing these goals vaccinology will profit tremendously from basic immunology. Given that the immune system evolved as a countermeasure against microbial invaders, these investigations will yield insights into the immune system that will reciprocally benefit the more academic branches of immunology. ACKNOWLEDGMENTS Financial support from the WHO, BMBF, SFB, DAHW, and the Fonds der Chemischen Industrie is gratefully acknowledged. We thank C. McCoull for excellent secretarial help, F. Arndt for help with the figures, and Dr. H. Collins for critically reading the manuscript. We also thank the numerous colleagues who willingly shared their unpublished, submitted, or in press results. We apologize to any colleagues whose work we failed to include because of an oversight on our part.

REFERENCES Abdelhak, S., Louzir, H., Timm, J., Blel, L., Benlasfar, Z., Lagranderie, M., Gheorghiu, M., Dellagi, K., and Gicquel, B. (1995). Recombinant BCG expressing the leishmania surface antigen Gp63 induces protective immunity against Leishmania major infection in BALB/c mice. Microbiology 141, 1585–1592. Aggarwal, A., Kumar, S., Jaffe, R., Hone, D., Gross, M., and Sadoff, J. (1990). Oral Salmonella: Malaria circumsporozoite recombinants induce specific CD8⫹ cytotoxic T cells. J. Exp. Med. 172, 1083–1090. Akbari, O., Panjwani, N., Garcia, S., Tascon, R., Lowrie, D., and Stockinger, B. (1999). DNA vaccination: Transfection and activation of dendritic cells as key events for immunity. J. Exp. Med. 189, 169–178. Alarcon, J. B., Waine, G. W., and McManus, D. P. (1999). DNA vaccines: Technology and application as anti-parasite and anti-microbial agents. Adv. Parasitol. 42, 343–410. Albert, M. L., Sauter, B., and Bhardwaj, N. (1998). Dendritic cells acquire antigen from apoptotic cells and induce class I restricted CTLs. Nature 392, 86–89. Aldovini, A., and Young, R. A. (1999). Recombinant BCG vaccines. In ‘‘Intracellular Bacterial Vaccine Vectors’’ (Y. Paterson, ed.), pp. 151–170, John Wiley & Sons, New York, Chicester, Weinheim, Brisbane, Singapore, Toronto. Aldovini, A., and Young, R. A. (1991). Humoral and cell-mediated immune responses to live recombinant BCG-HIV vaccines. Nature 351, 479–482. Aliprantis A. O., Yang, R.-B, Mark, M. R., Suggett, S., Devaux, B., Radolf, J. D., Klimpel, G. R., Godowski, P., and Zychlinski, A. (1999). Cell activation and apoptosis by bacterial lipoproteins through toll-like receptor-2. Science 285, 736–739. Alpuche-Aranda, C. M., Berthiaume, E. P., Mock, B., Swanson, J. A., and Miller, S. I. (1995). Spacious phagosome formation within mouse macrophages correlates with Salmonella serotype pathogenicity and host susceptibility. Inf. Immun. 63, 4456–4462. Altare, F., Lammas, D., Revy, P., Jouanguy, E., Doffinger, R., Lamhamedi, S., Drysdale, P., Scheel-Toellner, D., Girdlestone, J., Darbyshire, P., Wadhwa, M., Dockrell, H., Salmon, M., Fischer, A., Durandy, A., Casanova, J. L., and Kumararane, D. S. (1998). Inherited interleukin 12 deficiency in a child with bacille Calmette-Guerin and Salmonella enteritidis disseminated infection. J. Clin. Invest. 102, 2035–2040. Altman, J. D., Moss, P. A. H., Goulder, P. J. R., Barouch, D. H., McHeyzer, Williams, M. G., Bell, J. I., McMichael, A. J., and Davis, M. M. (1996). Phenotypic analysis of antigen-specific T lymphocytes. Science 274, 94–96.

64

HESS et al.

Apostolou, I., Takahama, Y., Belmant, C., Kawano, T., Huerre, M., Marchal, G., Cui, J., Taniguchi, M., Nakauchi, H., Fournie, J. J., Kourilsky, P., and Gachelin, G. (1999). Murine natural killer cells contribute to the granulomatous reaction caused by mycobacterial cell walls. Proc. Natl. Acad. Sci. USA 96, 5141–5146. Armstrong, J. A., and Hart, P. D. (1975). Phagosome–lysosome interactions in cultured macrophages infected with virulent tubercle bacilli. Reversal of the usual nonfusion pattern and observations on bacterial survival. J. Exp. Med. 142, 1–16. Arnold, D., Faath, S., Rammensee, H., and Schild, H. (1995). Cross-priming of minor histocompatibility antigen-specific cytotoxic T cells upon immunization with the heat shock protein gp96. J. Exp. Med. 182, 885–889. Ashkar, S., Weber, G. F., Panoutsakopoulou, V., Sanchinco, M. E., Jansson, M., Zawaiden, S., Rittling, S. R., Denhardt, D. T., Gliwicker, M. J., and Cantor, H. (2000). Eta-1 (osteopontin): An early component of type-1 (cell-mediated) immunity. Science 287, 860–864. Ashley, R. V., and Murray, C. J. L. (1996). Economic perspective on vaccine needs. In ‘‘Concepts in Vaccine Development’’ (S. H. E. Kaumann, ed.), pp. 27–70, Walter de Gruyter, Berlin, New York. Bacon, J. A., Burrows, T. W., and Yates, M. (1950). The effects of biochemical mutation on the virulence of Bacterium typhosum: the virulence of mutants. Br. J. 31, 714–724. Baldwin, S. L., D’Souza, C., Roberts, A. D., Kelly, B. P., Frank, A. A., Lui, M. A., Ulmer, JB, Huygen, K., McMurray, D. M., and Orme, I. M. (1998). Evaluation of new vaccines in the mouse and guinea pig model of tuberculosis. Inf. Immun. 66, 2951–2959. Balk, S. P., Ebert, E. C., Blumenthal, R. L., McDermott, F. V., Wucherpfennig, K. W., Landau, S. B., and Blumberg, R. S. (1991). Oligoclonal expansion and CD1 recognition by human intestinal intraepithelial lymphocytes. Science 253, 1411–1415. Banchereau, J., and Steinman, R. M. (1998). Dendritic cells and the control of immunity. Nature 392, 245–252. Bansal, P., Mukherjee, P., Basu, S.K., George, A., Bal, V., and Rath, S. (1999). MHC class I-restricted presentation of maleylated protein binding to scavenger receptors. J. Immunol. 162, 4430–4437. Barry, M. A., Lai, W. C., and Johnston, S. A. (1995). Protection against mycoplasma infection using expression-library immunization. Nature 377, 632–635. Bauer, A., Huttinger, R., Staffler, G., Hansmann, C., Schmidt, W., Majdic, O., Knapp, W., and Stockinger, H. (1997). Analysis of the requirement for beta 2-microglobulin for expression and formation of human CD1 antigens. Eur. J. Immunol. 27, 1366–1373. Beckman, E. M., Melian, A. M., Behar, S. M., Sieling, P. A., Chatterjee, D., Matsumoto, R., Rosat, J.-P., Modlin, R. L., and Porcelli, S. A. (1996). CD1c restricts responses of mycobacteria-specific T cells: Evidence for antigen presentation by a second member of the human CD1 family. J. Immunol. 157, 2803. Beckman, E. M., Porcelli, S. A., Morita, C. T., Behar, S. M., Furlong, S. T., and Brenner, M. B. (1994). Recognition of a lipid antigen by CD1-restricted 움웁⫹ T cells. Nature 372, 691–694. Behar, S. M., Dascher, C. C., Grusby, M. J., Wang, C.-R., and Brenner, M. B. (1999). Susceptibility of mice deficient in CD1D or TAP1 to infection with Mycobacterium tuberculosis. J. Exp. Med. 189, 1973–1980. Behr, M. A., Wilson, M. A., Gill, W. P., Salamon, H., Schoolnik, G. K., Rane, S., and Small, P. M. (1999). Comparative genomics of BCG vaccines by whole-genome DNA microarray. Science 284, 1520–1523. Bellone, M., Iezzi, G., Rovere, P., Galati, G., Rochetti, A., Pia Protti, M., Davoust, J., Rugarli, C., and Manfredi, A. A. (1997). Processing of engulfed apoptotic bodies yields T cell epitopes. J. Immunol. 159, 5391–5399.

EXPLOITING THE IMMUNE SYSTEM

65

Bendelac, A., Lantz, O., Quimby, M. E., Yewdell, J. W., Bennink, J. R., and Brutkiewicz, R. R. (1995). CD1 recognition by mouse NK1⫹T lymphocytes. Science 268, 863–865. Berche, P., Gaillard, J. L., and Sansonetti, P. J. (1987). Intracellular growth of Listeria monocytogenes as a prerequisite for in vivo induction of T cell-mediated immunity. J. Immunol. 138, 2266–2271. Berin, M. C., McKay, D. M., and Perdue, M. H. (1999). Immune-epithelial interactions in host defense. Am. J. Trop. Med. Hygiene 60, 16–25. Bermudez, L. E., and Kaplan, G. (1995). Recombinant cytokines for controlling mycobacterial infections. Trends Microbiol. 3, 22–27. Berthet, F. X., Lagranderie, M., Gounon, P., Laurent-Winter, C., Ensergueix, D., Chavarot, P., Thouron, F., Maranghi, E., Pelicic, V., Portnoi, D., Marchal, G., and Gicquel, B. (1998). Attenuation of virulence by disruption of the Mycobacterium tuberculosis erp gene. Science 282, 759–762. Beyers, A. D., van Rie, A., Adams, J., Fenhalls, G., Gie, R., and Beyers, N. (1998). Signals that regulate the host response to Mycobacterium tuberculosis. Novartis Foundation Symposium 217, 145–157. Bielecki, J., Youngman, P., Connelly, P., and Portnoy, D. A. (1990). Bacillus subtilis expressing a haemolysin gene from Listeria monocytogenes can grow in mammalian cells. Nature 345, 175–176. Blachere, N. E., Li, Z., Chandawarkar, R. Y., Suto, R., Jaikaria, N. S., Basu, S., Udono, H., and Srivastava, P. K. (1997). Heat shock protein–peptide complexes, reconstituted in vitro, elicit peptide-specific cytotoxic T lymphocyte response and tumor immunity. J. Exp. Med. 186, 1315–1322. Blackstock, W. P., and Weir, M. P. (1999). Proteomics: Quantitative and physical mapping of cellular proteins. Trends Biotechnol. 17, 121–127. Bland, P. W., and Warren, L. G. (1986). Antigen presentation by epithelial cells of the rat small intestine.1. Kintetics, antigen-specificity, and blocking by anti-Ia antisera. Immunology 58, 1–7. Bo¨hm, W., Mertens, T., Schirmbeck, R., and Reimann, J. (1998). Routes of plasmid DNA vaccination that prime murine humoral and cellular immune responses. Vaccine 16, 949–954. Bouwer, H. G., Nelson, C. S., Gibbins, B. L., Portnoy, D. A., and Hinrichs, D. J. (1992). Listeriolysin O is a target of the immune response to Listeria monocytogenes. J. Exp. Med. 175, 1467–1471. Brandtzaeg, P., Baekkevold, E. S., Farstad, I. N., Jahnsen, F. L., Johansen, F.-E., Nilsen, E. M., and Yamanaka, T. (1999). Regional specialization in the mucosal immune system: What happens in the microcompartments. Immunol. Today 20, 141–151. Brennan, P. J., and Nikaido, H. (1995). The envelope of mycobacteria. Ann. Rev. Biochem. 64, 29–63. Bretscher, P. A. (1992). A strategy to improve the efficacy of vaccination against tuberculosis and leprosy. Immunol. Today 13, 342–345. Brightbill, H. D., Libraty, D. H., Krutzki, S. R., Yang, R.-B., Belisle, J. T., Bleharski, J. R., Maitland, M., Norgard, M. V., Plevy, S. E., Smale, S. T., Brennan, P. J., Bloom, B. R., Godowski, P. J., and Modlin, R. L. (1999). Host defense mechanisms triggered by microbial lipoproteins through toll-like receptors. Science 285, 732–736. Brutkiewicz, R. R., Bennink, J. R., Yewdell, J. W., and Bendelac, A. (1995). TAP-independent, beta 2-microglobulin-dependent surface expression of functional mouse CD1.1. J. Exp. Med. 182, 1913–1919. Buchmeier, N. A., and Heffron, F. (1991). Inhibition of macrophage phagosome–lysosome fusion by Salmonella typhimurium. Inf. Immun. 59, 2232–2238.

66

HESS et al.

Busch, D. H., Pilip, I. M., Vijh, S., and Pamer, E. G. (1998). Coordinate regulation of complex T cell populations responding to bacterial infection. Immunity 8, 353–362. Calmette, A., Guerin, C., Negre, L., and Bocquet, A. (1927). Sur la vaccination pre´ventive des enfants nouveau-ne´s contre la tuberculose par le BCG. Ann. Inst. Pasteur 3, 201. Carbone, F. R., and Gleeson, P. A. (1997). Carbohydrates and antigen recognition by T cells. Glycobiology 7, 725–730. Cardenas, L., and Clements, J. D. (1993). Stability, immunogenicity, and expression of foreign antigens in bacterial vaccine vectors. Vaccine 11, 126–135. Carson, D. A., and Raz, E. (1997). Oligonucleotide adjuvants for T helper 1 (Th1)-specific vaccination. J. Exp. Med. 186, 1621–1622. Casadevall, A., and Pirofski, L.-A. (1999). Host-pathogen interactions: Redefining the basic concepts of virulence and pathogenicity. Inf. Immun. 67, 3703–3713. Catic, A., Dietrich, G., Gentschev, I., Goebel, W., Kaufmann, S. H. E., and Hess, J. (1999). Introduction of protein or DNA delivered via recombinant Salmonella typhimurium into major histocompatibility complex class I presentation pathway of macrophages. Microbes Inf. 1, 113–121. Chace, J. H., Hooker, N. A., Mildenstein, K. L., Krieg, A. M., and Cowdery, J. S. (1997). Bacterial DNA-induced NK cell IFN-웂 production is dependent on macrophage secretion of IL-12. Clin. Immunol. Immunopathol. 84, 185–193. Chan, J., Xing, Y., Magliozzo, R. S., and Bloom, B. R. (1992). Killing of virulent Mycobacterium tuberculosis by reactive nitrogen intermediates produced by activated murine macrophages. J. Exp. Med. 175, 1111–1122. Chen, Y. H., Inobe, J., Marks, R., Gonnella, P., Kuchroo, V. K., and Weiner, H. L. (1995). Peripheral deletion of antigen-reactive T-cells in oral tolerance. Nature 376, 177–180. Chien, Y.-H., Jores, R., and Crowley, M. P. (1996). Recognition by 웂/␦ T cells. Annu. Rev. Immunol. 14, 511–532. Chu, R. S., Askew, D., Noss, E. H., Tobian, A., Krieg, A. M., and Harding, C. V. (1999). CpG oligonucleotides down-regulate macrophage class II MHC antigen processing. J. Immunol. 163, 1188–1194. Chugh, I. B., and Khuller, G. K. (1993). Immunoprotective behaviour of liposome entrapped cell wall subunit of Mycobacterium tuberculosis against experimental tuberclous infection in mice. Eur. Resp. J. 6, 811–815. Colditz, G. A., Brewer, T. F., Berkey, C. S., Wilson, M. E., Burdick, E., Fineberg, H. V., and Mosteller, F. (1994). Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the published literature. J. Am. Med. Assoc. 271, 698–702. Cole, S. T. (1999). Learning from the genome sequence of Mycobacterium tuberculosis H37Rv. FEBS Lett. 452, 7–10. Cole, S. T., Brosch, R., Parkhill, J., Garnier, T., Churcher, C., Harris, D., Gordon, S. V., Eiglmeier, K., Gas, S., Barry, C. E., Tekaia, F., Badcock, K., Basham, D., Brown, D., Chillingworth, T., Connor, R., Davies, R., Devlin, K., Feltwell, T., Gentles, S., Hamlin, N., Holroyd, S., Hornsby, T., Jagels, K., and Barrell, B. G. (1998). Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393, 537–544. Collins, H. L., Schaible, U. E., Ernst, J. D., and Russell, D. G. (1997). Transfer of phagocytosed particles to the parasitophorous vacuole of Leishmania mexicana is a transient phenomenon preceding the acquisition of annexin I by the phagosome. J. Cell Sci. 110, 191–200. Committee to Study Priorities for Vaccine Development, I. o. M. (1999). Vaccines for the 21st century: A tool for decision-making (prepublication copy, uncorrected proof ), http:// bob.nap.edu/readingroom/books/vacc21/.

EXPLOITING THE IMMUNE SYSTEM

67

Condon, C., Watkins, S. C., Celluzzi, C. M., Thompson, K., and Falo, L. D., Jr. (1996). DNA-based immunization by in vivo transfection of dendritic cells. Nature Medicine 2, 1122–1128. Connell, N. D., Medina-Acosta, E., McMaster, W. R., Bloom, B. R., and Russell, D. G. (1993). Effective immunization against cutaneous leishmaniasis with recombinant bacille Calmette–Guerin expressing the Leishmania surface proteinase gp63. Proc Natl. Acad. Sci. USA 90, 11473–11477. Conradt, P., Hess, J., and Kaufmann, S. H. E. (1999). Cytolytic T-cell responses to human dendritic cells and macrophages infected with Mycobacterium bovis BCG and recombinant BCG secreting listeriolysin. Microbes Inf. 1, 753–764. Constant, P., Davodeau, F., Peyrat, M.-A., Poquet, Y., Puzo, G., Bonneville, M., and Fournie´, J.-J. (1994). Stimulation of human 웂␦ T cells by nonpeptidic mycobacterial ligands. Science 264, 267–270. Cooney, E. L., Collier, A. C., Greenberg, P. D., Coombs, R. W., Zarling, J., Arditti, D. E., Hoffman, M. C., Hu, S. L., and Corey, L. (1991). Safety of and immunological response to a recombinant vaccinia virus vaccine expressing HIV envelope glycoprotein. Lancet 337, 567–572. Cooper, A. M., Magram, J., Ferrante, J., and Orme, I. M. (1997). Interleukin 12 (IL-12) is crucial to the development of protective immunity in mice intravenously infected with mycobacterium tuberculosis. J. Exp. Med. 186, 39–45. Cooper, A. M., Roberts, A. D., Rhoades, E. R., Callahan, J. E., Getzy, D. M., and Orme, I. M. (1995). The role of interleukin-12 in acquired immunity to Mycobacterium tuberculosis infection. Immunology 84, 423–432. Cornell, K. A., Bouwer, H. G. A., Hinrichs, D. J., and Barry, R. A. (1999). Genetic immunization of mice against Listeria monocytogenes using plasmid DNA encoding listeriolysin O. J. Immunol 163, 322–329. Corr, M., Lee, D. J., Carson, D. A., and Tighe, H. (1996). Gene vaccination with naked plasmid DNA: Mechanism of CTL priming. J. Exp. Med. 184, 1555–1560. Cossart, P. (1997). Host/pathogen interactions. Subversion of the mammalian cell cytoskeleton by invasive bacteria. J. Clin. Invest. 99, 2307–2311. Crowle, A. J. (1988). Immunization against tuberculosis–what kind of vaccine. Inf. Immun. 56, 2769–2773. D’Souza, C., Cooper, A. M., Frank, A. A., Mazzaccaro, R. J., Bloom, B. R., and Orme, I. M. (1997). An anti-inflammatory role for 웂␦ T lymphocytes in acquired immunity to Mycobacterium tuberculosis. J. Immunol. 158, 1217–1221. Darji, A., Guzman, C. A., Gerstel, B., Wachholz, P., Timmis, K. N., Wehland, J., Chakraborty, T., and Weiss, S. (1997). Oral somatic transgene vaccination using attenuated S. typhimurium. Cell 91, 765–775. Davis, B. D., Dulbecco. R., Eisen, H. N., and Ginsberg, H. S. (1969). ‘‘Microbiology,’’ pp. 603–652. Harper International, Cambridge, MA. Davis, H. L., McCluskie, M. J., Gerin, J. L., and Purcell, R. H. (1996). DNA vaccine for hepatitis B: Evidence for immunogenicity in chimpanzees and comparison with other vaccines. Proc. Natl. Acad. Sci. USA 93, 7213–7218. de Jong, R., Altare, F., Haagen, I. A., Elferink, D. G., Boer, T., van, B., Vriesman, P. J., Kabel, P. J., Draaisma, J. M., van Dissel, J. T., Kroon, F. P., Casanova, JL, and Ottenhoff, T. H. (1998). Severe mycobacterial and Salmonella infections in interleukin-12 receptordeficient patients. Science 280, 1435–1438. DeLibero, G., Casorati, G., Giachino, C., Carbonara, C., Migone, N., Matzinger, P., and Lanzave˙cchia, A. (1991). Selection by two powerful antigens may account for the presence of the major population of human peripheral 웂/␦ T cells. J. Exp. Med. 173, 1311–1322.

68

HESS et al.

DeLibero, G., Flesch, I., and Kaufmann, S. H. E. (1988). Mycobacteria reactive Lyt2⫹ T cell lines. Eur. J. Immunol. 18, 59–66. Demangel, C., Bean, A. G. D., Feng, C. G., Kamath, A. T., and Britton, W. J. (1999). Protection against aerosol Mycobacterium tuberculosis infection using Mycobacterium bovis Bacillus Calmette Gue´rin-infected dendritic cells. Eur. J. Immunol. 29, 1972–1979. Denis, O., Tanghe, A., Palfiet, K., Jurion, F., van den Berg, T. P., Vanonckelen, Ooms, J., Saman, E., Ulmer, J. B., Content, J., and Huygen, K. (1998). Vaccination with plasmid DNA encoding myocobacterial antigen 85A stimulates a CD4⫹ and CD8⫹ T-cell epitopic repertoire broader than that stimulated by Mycobacterium tuberculosis H37Rv infection. Inf. Immun. 66, 1527–1533. Dietrich, G., Bubert, A., Gentschev, I., Sokolovic, Z., Simm, A., Catic, A., Kaufmann, S. H., Hess, J., Szalay, A. A., and Goebel, W. (1998). Delivery of antigen-encoding plasmid DNA into the cytosol of macrophages by attenuated suicide Listeria monocytogene. Nature Biotechnol. 16, 181–185. Dietrich, G., Gentschev, I., Hess, J., Ulmer, J. B., Kaufmann, S. H. E., and Goebel, W. (1999). Delivery of DNA vaccines by attenuated intracellular bacteria. Immunol. Today 20, 251–253. Dillon, D. C. (1999). Molecular characterization and human T-cell responses to a member of a novel Mycobacterium tuberculosis mtb39 gene family. Inf. Immun. 67, 2941–2950. Donnelly, J. J., Ulmer, J. B., and Liu, M. A. (1994). Immunization with DNA. J. Immunol. Meth. 176, 145–152. Dunbar, P. R., Ogg, G. S., Chen, J., Rust, N., van der Bruggen, P., and Cerundolo, V. (1998). Direct isolation, phenotyping and cloning of low-frequency antigen-specific cytotoxic T lymphocytes from peripheral blood. Curr. Biol. 8, 413–416. Duncan, D. D., and Swain, S. L. (1994). Role of antigen-presenting cells in the polarized development of helper T-cells subsets-evidence for differential cytocine production by tho cells in response to antigen presentation by B-cells and macrophages. Eur. J. Immunol. 24, 2506–2514. Dutton, R. W., Bradley, L. M., and Swain, S. L. (1998). T cell memory. Annu. Rev. Immunol. 16, 201–223. Edelman, R., Palmer, K., Russ, K. G., Secrest, H. P., Becker, J. A., Bodison, S. A., Perry, J. G., Sills, A. R., Barbour, A. G., Luke, C. J., Hanson, M. S., Stover, C. K., Burlein, J. E., Bansal, G. P., Connor, E. M., and Koenig, S. (1999). Safety and immunogenicity of recombinant Bacille Calmette–Guerin (rBCG) expressing Borrelia burgdorferi outer surface protein A (OspA) lipoprotein in adult volunteers: A candidate Lyme disease vaccine. Vaccine 17, 904–914. Edelson, B. T., Cossart, P., and Unanue, E. R. (1999). Paradigm revisited: Antibody provides resistance to Listeria infection. J. Immunol. 163, 4087–4090. Edwards, M. F., and Stocker, B. A. (1988). Construction of delta aroA his delta pur strains of Salmonella typhi. J. Bacteriol. 170, 3991–3995. Elkins, K. L., Rhinehart-Jones, T. R., Stibitz, S., Conover, J. S., and Klinman, D. M. (1999). Bacterial DNA containing CpG motifs stimulates lymphocyte-dependent protection of mice against lethal infection with intracellular bacteria. J. Immunol. 162, 2291–2298. Ellis, R. W. (1999). New technologies for making vaccines. Vaccine 17, 1596–1604. Engering, A. J., Cella, M., Fluitsma, D., Brockhaus, M., Hoefsmit, E. C., Lanzavecchia, A., and Pieters, J. (1997). The mannose receptor functions as a high capacity and broad specificity antigen receptor in human dendritic cells. Eur. J. Immunol. 27, 2417–2425. Erdile, L. F., Brandt, M. A., Warakomski, D. J., Westrack, G. J., Sadziene, A., Barbour, A. G., and Mays, J. P. (1993). Role of attached lipid in immunogenicity of Borrelia burgdorferi OspA. Inf. Immun. 61, 81–90.

EXPLOITING THE IMMUNE SYSTEM

69

Etchart, N., Buckland, R., Liu, M. A., Wild, T. F., and Kalserlian, D. (1997). Class Irestricted CTL induction by mucosal immunization with naked DNA encoding measles virus haemagglutinin. J. Gen. Virol. 78, 1577–1580. Euzeby, J. P. (1999). Revised Salmonella nomenclature: Designation of Salmonella enterica (ex Kauffmann and Edwards, 1952), Le Minor and Popoff, 1987 sp. nov., nom. rev. as the neotype species of the genus Salmonella Lignieres 1900 (Approved Lists 1980), rejection of the name Salmonella choleraesuis (Smith, 1894) Weldin, 1927 (Approved Lists 1980), and conservation of the name Salmonella typhi (Schroeter, 1886) Warren and Scott, 1930 (Approved Lists 1980). Request for an Opinion. Intl. J. System. Bacteriol. 49, 927–930. Exley, M., Garcia, J., Balk, S. P., and Porcelli, S. (1997). Requirements for CD1d recognition by human invariant V alpha 24(⫹) CD4(⫺)CD8(⫺) T cells. J. Exp. Med. 186, 109–120. Fang, F. C. (1997). Mechanisms of nitric oxide-related antimicrobial activity. J. Clin. Invest. 100, S43–S50. Farstad, I. N., Halstensen, T. S., Fausa, O., and Brandtzaeg, P. (1994). Heterogeneity of M-cell-associated B and T cells in human Peyer’s patches. Immunology 83, 457–464. Fennelly, G. J., Khan, S. A., Abadi, M. A., Wild, T. F., and Bloom, B. R. (1999). Mucosal DNA vaccine immunization against measles with a highly attenuated Shigella flexneri vector. J. Immunol. 162, 1603–1610. Fensterle, J., Grode, L., Hess, J., and Kaufmann, S. H. E. (1999). Effective DNA vaccination against listeriosis by prime/boost inoculation with the gene gun. J. Immunol. 163, 4510– 4518. Ferreccio, C., Levine, M. M., Rodriguez, H., and Contreras, R. (1989). Comparative efficacy of two, three, or four doses of TY21a live oral typhoid vaccine in enteric-coated capsules: A field trial in an endemic area. J. Infect. Dis. 159, 766–769. Ferru, I., Rollin, C., Estaquier, J., Sutton, P., Delacre, M., Tartar, A., Gras-Masse, H., and Auriault, C. (1996). Comparison of the immune response elicited by a free peptide and a lipopeptide construct. Peptide Res. 9, 136–143. Fischer Lindahl, K., Byers, D. E., Dahi, V. M., Hovik, R., Jones, E. P., Smith, G. P., Wang, C.-R., Xiao, H., and Yoshino, M. (1997). H2-M3, a full-service class lb histocompatibility antigen. Annu. Rev. Immunol. 15, 851–879. Fleischmann, R. D., Adams, M. D., White, O., Clayton, R. A., Kirkness, E. F., Kerlavage, A. R., Bult, C. J., Tomb, J. F., Dougherty, B. A., and Merrick, J. M. (1995). Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269, 496–512. Flesch, I. E., Hess, J. H., Huang, S., Aguet, M., Rothe, J., Bluethmann, H., and Kaufmann, S. H. (1995). Early interleukin 12 production by macrophages in response to mycobacterial infection depends on interferon gamma and tumor necrosis factor alpha. J. Exp. Med. 181, 1615–1621. Flesch, I. E. A., and Kaufmann, S. H. E. (1991). Mechanisms involved in mycobacterial growth inhibition by gamma interferon-activated bone marrow macrophages: Role of reactive nitrogen intermediates. Infect. Immun. 59, 3213–3218. Flynn, J. L., Goldstein, M. M., Triebold, K. J., Koller, B., and Bloom, B. R. (1992). Major histocompatibility complex class I-restricted T cells are required for resistance to Mycobacterium tuberculosis infection. Proc. Natl. Acad. Sci. USA 89, 12013–12017. Flynn, J. L., Goldstein, M. M., Triebold, K. J., Sypek, J., Wolf, S., and Bloom, B. R. (1995). IL-12 increases resistance of BALB/c mice to Mycobacterium tuberculosis infection. J. Immunol. 155, 2515–2524. Flynn, J. L., Weiss, W. R., Norris, K. A., Seifert, H. S., Kumar, S., and So, M. (1990). Generation of a cytotoxic T-lymphocyte response using a Salmonella antigen-delivery system. Mol. Microbiol. 4, 2111–2118.

70

HESS et al.

Frankel, F. R., Hegde, S., Lieberman, J., and Paterson, Y. (1995). Induction of cell-mediated immune responses to human immunodeficiency virus type 1 Gag protein by using Listeria monocytogenes as a live vaccine vector. J. Immunol. 155, 4775–4782. Freitas, A. A., and Rocha, B. (1999). Peripheral T cell survival. Curr. Opin. Immunol. 11, 152–156. Fu, Y., and Galan, J. E. (1998). The Salmonella typhimurium tyrosine phosphatase SptP is translocated into host cells and disrupts the actin cytoskeleton. Mol. Microbiol. 27, 359–368. Fynan, E. F., Webster, R. G., Fuller, D. H., Haynes, J. R., Santoro, J. C., and Robinson, H. L. (1993). DNA vaccines: Protective immunizations by parenteral, mucosal, and genegun inoculations. Proc. Natl. Acad. Sci. USA 90, 11478–11482. Gaillard, J. L., Berche, P., Mounier, J., Richard, S., and Sansonetti, P. (1987). In vitro model of penetration and intracellular growth of Listeria monocytogenes in the human enterocyte-like cell line Caco-2. Infect. Immun. 55, 2822–2829. Garcia, V. E., Sieling, P. A., Gong, J., Barnes, P. F., Uyemura, K., Tanaka, Y., Bloom, B. R. Morita, C. T., and Modlin, R. L. (1997). Single-cell cytokine analysis of 웂␦ T cell responses to nonpeptide mycobacterial antigens. J. Immunol. 159, 1328–1335. Geginat, G. (1999). Enhancement of the Listeria monocytogenes p60-specific CD4 and CD8 T cell memory by nonpathogenic Listeria innocua. J. Immunol. 162, 4781–4789. Gentschev, I., Dietrich, G., Mollenkopf, H. J., Sokolovic, Z., Hess, J., Kaufmann, S. H., and Goebel, W. (1997). The Escherichia coli hemolysin secretion apparatus—A versatile antigen delivery system in attenuated Salmonella. Behring Institute Mitteilungen 98, 103–113. Gentschev, I., Glaser, I., Goebel, W., McKeever, D. J., Musoke, A., and Heussler, V. T. (1998). Delivery of the p67 sporozoite antigen of Theileria parva by using recombinant Salmonella dublin: Secretion of the product enhances specific antibody responses in cattle. Infect. Immun. 66, 2060–2064. Gentschev, I., Hess, J., and Goebel, W. (1990). Change in the cellular localization of alkaline phosphatase by alteration of its carboxy-terminal sequence. Mol. Gen. Genet. 222, 211–216. Gentschev, I., Mollenkopf, H., Sokolovic, Z., Hess, J., Kaufmann, S. H., and Goebel. (1996). Development of antigen-delivery systems, based on the Escherichia coli hemolysin secretion pathway. Gene 179, 133–140. Gentschev, I., Sokolovic, Z., Mollenkopf, H. J., Hess, J., Kaufmann, S. H., Kuhn, M., Krohne, G. F., and Goebel, W. (1995). Salmonella strain secreting active listeriolysin changes its intracellular localization. Infect. Immun. 63, 4202–4205. Germanier, R., and Fuer, E. (1975). Isolation and characterization of Gal E mutant Ty 21a of Salmonella typhi: A candidate strain for a live, oral typhoid vaccine. J. Infect. Dis. 131, 553–558. Gheorghiu, M., Lagranderie, M. R., Gicquel, B. M., and Leclerc, C. D. (1994). Mycobacterium bovis BCG priming induces a strong potentiation of the antibody response induced by recombinant BCG expressing a foreign antigen. Infect. Immun. 62, 4287–4295. Goebel, W., and Kreft, J., (1997). Cytolysins and the intracellular life of bacteria. Trends Microbiol. 5, 86–88. Gomes, M. S., Paul, S., Moreira, A. L., Appelberg, R., Rabinovitch, M., and Kaplan, G. (1999). Survival of Mycobacterium avium and Mycobacterium tuberculosis in acidified vacuoles of murine macrophages. Infect. Immun. 67, 3199–3206. Gonzalez, C., Hone, D., Noriega, F. R., Tacket, C. O., Davis, J. R., Losonsky, G., Nataro, J. P., Hoffman, S., Malik, A., and Nardin, E. (1994). Salmonella typhi vaccine strain CVD

EXPLOITING THE IMMUNE SYSTEM

71

908 expressing the circumsporozoite protein of Plasmodium falciparum: strain construction and safety and immunogenicity in humans. J. Infect. Dis. 169, 927–931. Gottlieb, S., McLaughlin, F. X., Levine, L., Latham, W. C., and Edsall, G. (1964). Longterm immunity to tetanus—A statistical evaluation and its clinical implication. Am. J. Public Health 54, 961–971. Gru¨tzkau, A., Hanski, C., Hahn, H., and Riecken, E. O. (1990). Involvement of M cells in the bacterial invasion of Peyer’s patches: A common mechanism shared by Yersinia enterocolitica and other enteroinvasive bacteria. Gut 31, 1011–1015. Grillot-Courvalin, C., Goussard, S., Huetz, F., Ojcius, D. M., and Courvalin, P. (1998). Functional gene transfer from intracellular bacteria to mammalian cells. Nature Biotechnol. 16, 862–866. Guler, M. L., Gorham, J. D., Hsieh, C. S., Mackey, A. J., Steen, R. G., Dietrich, W. F., and Murphy, K. M. (1996). Genetic susceptibility to Leishmania: IL-12 responsiveness in TH1 cell development. Science 271, 984–987. Guleria, I., Teitelbaum, R., McAdam, R. A., Kalpana, G., Jacobs, W. R. J., and Bloom, B. R. (1996). Auxotrophic vaccines for tuberculosis. Nature Med. 2, 334–337. Gurunathan, S., Prussin, C., Sacks, D. L., and Seder, R. A. (1998). Vaccine requirements for sustained cellular immunity to an intracellular parasitic infection. Nature Med. 4, 1409– 1415. Guzman, C. A., Rohde, M., Chakraborty, T., Domann, E., Hudel, M., Wehland, J., and Timmis, K. N. (1995). Interaction of Listeria monocytogenes with mouse dendritic cells. Infect. Immun. 63, 3665–3673. Hara, T., Mizuno, Y., Takaki, K., Takada, H., Akeda, H., Aoki, T., Nagata, M., Ueda, K., Matsuzaki, G., Yoshikai, Y. et al. (1992). Predominant activation and expansion of V웂9bearing 웂␦ T cells in vivo as well as in vitro in Salmonella infection. J. Clin. Invest. 90, 204–210. Harding, C. V., and Pfeifer, J. D. (1994). Antigen expressed by Salmonella typhimurium is processed for class I major histocompatibility complex presentation by macrophages but not infected epithelial cells. Immunology 83, 670–674. Harding, C. V., Song, R., Griffin, J., France, J., Wick, M. J., Pfeifer, J. D., and Geuze, H. J. (1995). Processing of bacterial antigens for presentation to class I and II MHCrestricted T lymphocytes. Infect. Agents Disease 4, 1–12. Harty, J. T., and Bevan, M. J. (1992). CD8⫹ T cells specific for a single nonamer epitope of Listeria monocytogenes are protective in vivo. J. Exp. Med. 175, 1531–1538. Hayashi, T., Catanzaro, A., and Rao, S. P. (1997). Apoptosis of human monocytes and macrophages by Mycobacterium avium sonicate. Infect. Immun. 65, 5262–5271. Heithoff, D. M., Sinsheimer, R. L., Low, D. A., and Mahan, M. J. (1999). An essential role for DNA adenine methylation in bacterial virulence. Science 284, 967–970. Hershberg, R. M., Framson, P. E., Cho, D. H., Lee, L. Y., Kovats, S., Beitz, J., Blum, J. S., and Nepom, G. T. (1997). Intestinal epithelial cells use two distinct pathways for HLA class II antigen processing. J. Clin. Invest. 100, 204–215. Hess, J., Dietrich, G., Gentschev, I., Miko, D., Goebel, W., and Kaufmann, S. H. (1997). Protection against murine listeriosis by an attenuated recombinant Salmonella typhimurium vaccine strain that secretes the naturally somatic antigen superoxide dismutase. Infect. Immun. 65, 1286–1292. Hess, J., Dreher, A., Gentschev, I., Goebel, W., Ladel, C., Miko, D., and Kaufmann, S. H. (1996a). Protein p60 participates in intestinal host invasion by Listeria monocytogenes. Zentralblattfuer Bakteriologie 284, 263–272. Hess, J., Gentschev, I., Goebel, W., and Jarchau, T. (1990). Analysis of the haemolysin secretion system by PhoA-HlyA fusion proteins. Mol. Gen. Genetics 224, 201–208.

72

HESS et al.

Hess, J., Gentschev, I., Miko, D., Welzel, M., Ladel, C., Goebel, W., and Kaufmann, S. H. (1996b). Superior efficacy of secreted over somatic antigen display in recombinant Salmonella vaccine induced protection against listeriosis. Proc. Natl. Acad. Sci. USA 93, 1458–1463. Hess, J., Gentschev, I., Szalay, G., Ladel, C., Bubert, A., Goebel, W., and Kaufmann, S. H. (1995). Listeria monocytogenes p60 supports host cell invasion by and in vivo survival of attenuated Salmonella typhimurium. Infect. Immun. 63, 2047–2053. Hess, J., Grode, L., Hellwig, J., Gentschev, I., Goebel, W., Ladel, C. H., and Kaufmann, S. H. E. (1999). Protection against murine tuberculosis by an attenuated recombinant Salmonella typhimurium vaccine strain that secretes the 30 kDa-antigen of Mycobacterium bovis BCG. FEMS Immunol. Med. Microbiol., in press. Hess, J., and Kaufmann, S. H. E. (1997). Principles of cell-mediated immunity underlying vaccination strategies against intracellular pathogens. In ‘‘Host Response to Intracellular Pathogens’’ (S. H. E. Kaufmann, ed.), pp. 75–94. R.G. Landes Company, Austin. Hess, J., Miko, D., Catic, A., Lehmensiek, V., Russell, D. G., and Kaufmann, S. H. (1998). Mycobacterium bovis Bacille Calmette–Guerin strains secreting listeriolysin of Listeria monocytogenes. Proc. Natl. Acad. Sci. USA 95, 5299–5304. Higgins, D. E., Shastri, N., Portnoy, D. A., and Portnoy. (1999). Delivery of protein to the cytosol of macrophages using Escherichia coli K-12. Mol. Microbiol. 31, 1631–1641. Hiromatsu, K., Yoshikai, Y., Matsuzaki, G., Ohga, S., Muramori, K., Matsumoto, K., Bluestone, J. A., and Nomoto, K. (1992). A protective role of 웂␦ T cells in primary infection with Listeria monocytogenes in mice. J. Exp. Med. 175, 49–56. Hoffmann, J. A., Kafatos, F. C., Janeway, C. A., and Ezekowitz, R. A. (1999). Phylogenetic perspectives in innate immunity. Science 284, 1313–1318. Hoffman, S. L., Rogers, W. O., Carucci, D. J., and Venter, J. C. (1998). From genomics to vaccines: Malaria as a model system. Nature Med. 4, 1351–1353. Hohmann, E. L., Oletta, C. A., Killeen, K. P., and Miller, S. I. (1996). phoP/phoQ-deleted Salmonella typhi (Ty800) is a safe and immunogenic single-dose typhoid fever vaccine in volunteers. J. Infect. Diseases 173, 1408–1414. Hohmann, E. L., Oletta, C. A., and Miller, S. I. (1996). Evaluation of a phoP/phoQ-deleted, aroA-deleted live oral Salmonella typhi vaccine strain in human volunteers. Vaccine 14, 19–24. Hoiseth, S. K., and Stocker, B. A. (1981). Aromatic-dependent Salmonella typhimurium are nonvirulent and effective as live vaccines. Nature 291, 238–239. Honda, M., Matsuo, K., Nakasone, T., Okamoto, Y., Yoshizaki, H., Kitamura, K., Sugiura, W., Watanabe, K., Fukushima, Y., and Haga, S. (1995). Protective immune responses induced by secretion of a chimeric soluble protein from a recombinant Mycobacterium bovis bacillus Calmette–Guerin vector candidate vaccine for human immunodeficiency virus type 1 in small animals. Proc. Natl. Acad. Sci. USA 92, 10693–10697. Hone, D. M., Attridge, S. R., Forrest, B., Morona, R., Daniels, D., LaBrooy, J. T., Bartholomeusz, R. C., Shearman, D. J., and Hackett, J. (1988). A galE via (Vi antigen-negative) mutant of Salmonella typhi Ty2 retains virulence in humans. Infect. Immun. 56, 1326– 1333. Hone, D. M., Harris, A. M., Chatfield, S., Dougan, G., and Levine, M. M. (1991). Construction of genetically defined double aro mutants of Salmonella typhi. Vaccine 9, 810–816. Hone, D. M., Shata, M. T., and Pascual, D. W. L. G. K. (1999). Mucosal vaccination with Salmonella vaccine vectors. In ‘‘Intracellular Bacterial Vaccine Vectors’’ (Y. Patterson, ed.), pp. 1–23. John Wiley and Sons. Hormaeche, C. E., and Khan, C. M. A. (1996). Recombinant bacteria as vaccine carriers of heterologous antigens. In ‘‘Concepts in Vaccine Development’’ (S. H. E. Kaufman, ed.), pp. 327–360. Walter de Gruyter, Berlin, New York.

EXPLOITING THE IMMUNE SYSTEM

73

Horwitz, M. A., Lee, B. W., Dillon, B. J., and Harth, G. (1995). Protective immunity against tuberculosis induced by vaccination with major extracellular proteins of Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 92, 1530–1534. Hsieh, B., Schrenzel, M. D., Mulvania, T., Lepper, H. D., DiMolfetto-Landon, L., and Ferrick, D. A. (1996). In vivo cytokine production in murine listeriosis. Evidence for immunoregulation by 웂␦ ⫹ T cells. J. Immunol. 156, 232–237. Hsieh, C. S., Macatonia, S. E., Tripp, C. S., Wolf, S. F., O’Garra, A., and Murphy, K. M. (1993). Development of TH1 CD4⫹ T cells through IL-12 produced by Listeria-induced macrophages. Science 260, 547–549. Huebner, R. E. (1996). BCG vaccination in the control of tuberculosis. Curr. Top. Microbiol. Immunol. 215, 263–282. Huygen, K., Content, J., Denis, O., Montgomery, D. L., Yawman, A. M., Deck, R. R., DeWitt, C. M., Orme, I. M., Baldwin, S., D’Souza, C., Drowart, A., Lozes, E., Vandenbussche, P., Van Vooren, J. P., Liu, M. A., and Ulmer, J. B. (1996). Immunogenicity and protective efficacy of a tuberculosis DNA vaccine. Nature Med. 2, 893–898. Ikonomidis, G., Portnoy, D. A., Gerhard, W., and Paterson, Y. (1997). Influenza-specific immunity induced by recombinant Listeria monocytogenes vaccines. Vaccine 15, 433–440. Inaba, K., Inaba, M., Naito, M., and Steinman, R. M. (1993). Dendritic cell progenitors phagocytose particulates, including bacillus Calmette–Guerin organisms, and sensitize mice to mycobacterial antigens in vivo. J. Exp. Med. 178, 479–488. Ishii, T., Udono, H., Yamano, T., Ohta, H., Uenaka, A., Ono, T., Hizuta, A., Tanaka, N., Srivastava, P. K., and Nakayama, E. (1999). Isolation of MHC class I-restricted tumor antigen peptide and its precursors associated with heat shock proteins hsp70, hsp90, and gp96. J. Immunol. 162, 1303–1309. Jenner, E. (1798). An inquiry into the causes and effects of the Variolae vaccinae, a disease discovered in some of the western counties of England, particularly Gloucestershire, and known by the name of the Cow-Pox, p. 75 with 4 plates, London, UK. Jiang, W., Swiggard, W. J., Heufler, C., Peng, M., Mirza, A., Steinman, R. M., and Nussenzweig, M. C. (1995). The receptor DEC-205 expressed by dendritic cells and thymic epithelial cells is involved in antigen processing. Nature 375, 151–155. Johansson, M., Ward, M., and Lycke, N. (1997). B-cell-deficient mice develop complete immune protection against genital tract infection with Chlamydia trachomatis. Immunology 92, 422–428. Johnston, S. A. (1998). DNA vaccines and related technology. Talk held at the International Symposium on Vaccinology (18–20. Nov., 1998) in Paris. Jondal, M., Schirmbeck, R., and Reimann, J. (1996). MHC class I-restricted CTL responses to exogenous antigens. Immunity 5, 295–302. Jones, B. D., Ghori, N., and Falkow, S. (1994). Salmonella typhimurium initiates murine infection by penetrating and destroying the specialized epithelial M cells of the Peyer’s patches. J. Exp. Med. 180, 15–23. Jones, S., and Portnoy, D. A. (1994). Characterization of Listeria monocytogenes pathogenesis in a strain expressing perfringolysin O in place of listeriolysin O. Infect. Immun. 62, 5608– 5613. Jones, S., Preiter, K., and Portnoy, D. A. (1996). Conversion of an extracellular cytolysin into a phagosome-specific lysin which supports the growth of an intracellular pathogen. Mol. Microbiol. 21, 1219–1225. Joyce, S., Woods, A. S., Yewdell, J. W., Bennink, J. R., Desilva, A. D., Boesteanu, A., Balk, S. P., Cotter, R. J., and Brutkiewicz, R. R. (1998). Natural ligand of mouse CD1d1–cellular glycosylphosphatidylinositol. Science 279, 1541–1544.

74

HESS et al.

Jungblut, P., Schaible, U., Mollenkopf, H.-J., Zimny-Arndt, U., Raupach, B., Mattow, J., Halada, P., Lamer, S., Hagens, K., and Kaufmann, S. H. E. (1999). Comparative proteome analysis of Mycobacterium tuberculosis and Mycobacterium bovis BCG strains: Towards functional genomics of microbial pathogens. Mol. Microbiol. 33, 1103–1117. Kabelitz, D., Bender, A., Prospero, T., Wesselborg, S., Janssen, O., and Pechhold, K. (1991). The primary response of human 웂/␦⫹ T cells to Mycobacterium tuberculosis is restricted to V웂 9-bearing cells. J. Exp. Med. 173, 1331–1338. Kaiserlian, D., Vidal, K., and Revillard, J. P. (1989). Murine enterocytes can present solubleantigen to specific class-II-restricted CD4⫹ T-cells. Eur. J. Immunol. 19, 1513–1516. Kalman, S., Mitchell, W., Marathe, R., Lammel, C., Fan, J., Hyman, R. W., Olinger, Grimwood, J., Davis, R. W., and Stephens, R. S. (1999). Comparative genomes of Chlamydia pneumoniae and C. trachomatis. Nat. Genet. 21, 385–389. Kamath, A. T., Feng, C. G., Macdonald, M., Briscoe, H., and Britton, W. J. (1999). Differential protective efficacy of DNA vaccines expressing secreted proteins of Mycobacterium tuberculosis. Infect. Immun. 67, 1702–1707. Kameoka, M., Nishino, Y., Matsuo, K., Ohara, N., Kimura, T., Yamazaki, A., Yamada, T., and Ikuta, K. (1994). Cytotoxic T lymphocyte response in mice induced by a recombinant BCG vaccination which produces an extracellular alpha antigen that fused with the human immunodeficiency virus type 1 envelope immunodominant domain in the V3 loop. Vaccine 12, 153–158. Kaniga, K., Uralil, J., Bliska, J. B., and Galan, J. E. (1996). A secreted protein tyrosine phosphatase with modular effector domains in the bacterial pathogen Salmonella typhimurium. Mol. Microbiol. 21, 633–641. Kaufmann, S. H. E. (1988a). Listeriosis: New findings—current concern. Microb. Pathogen. 5, 225–231. Kaufmann, S. H. E. (1988b). CD8⫹ T lymphocytes in intracellular microbial infections. Immunol. Today 9, 168–174. Kaufmann, S. H. E. (1991). Heat-shock proteins and pathogenesis of bacterial infections. Springer Semin. Immunpathol. 13, 25–36. Kaufmann, S. H. E. (1993). Immunity to intracellular bacteria. Annu. Rev. Immunol. 11, 129–163. Kaufmann, S. H. E. (1996). Gamma/delta and other unconventional T lymphocytes: What do they see and what do they do? Proc. Natl. Acad. Sci. USA 93, 2272–2279. Kaufmann, S. H. E. (1999). Killing vs suicide in antibacterial defence. Trends Microbiol. 7, 59–61. Kaufmann, S. H. E., and Andersen, P. (1998). Immunity to mycobacteria with emphasis on tuberculosis: Implications for rational design of an effective tuberculosis vaccine. Chem. Immunol. 70, 21–59. Kaufmann, S. H. E. and Ladel, C. H. (1994). Role of T cell subsets in immunity against intracellular bacteria: Experimental infections of knock-out mice with Listeria monocytogenes and Mycobacterium bovis BCG. Immunobiology 191, 509–519. Kaufmann, S. H. E. (1998). Immunity to intracellular bacteria. In ‘‘Fundamental Immunology’’ (W. Paul, ed.), 4th edition, pp. 1335–1371. Raven Press Ltd., New York. Kawano, T., Cui, J., Koezuka, Y., Toura, I., Kaneko, Y., Motoki, K., Ueno, H., Nakagawa, R., Sato, H., Kondo, E., Koseki, H., and Taniguchi, M. (1997). CD1d-restricted and TCRmediated activation of V움 14 NKT cells by glycosylceramides. Science 278, 1626–1629. Keitel, W. A., Bond, N. L., Zahradnik, J. M., Cramton, T. A., and Robbins, J. B. (1994). Clinical and serological responses following primary and booster immunization with Salmonella typhi Vi capsular polysaccharide vaccines. Vaccine 12, 195–199.

EXPLOITING THE IMMUNE SYSTEM

75

Kerksiek, K. M., Busch, D. H., Plip, I. M., Allen, S. E., and Pamer, E. G. (1999). H2-M3 restricted T cells in bacterial infection: Rapid primary but diminished memory responses. J. Exp. Med. 190, 195–204. Kerneis, S., Bogdanova, A., Kraehenbuhl, J. P., and Pringault, E. (1997). Conversion by Peyer’s patch lymphocytes of human enterocytes into M cells that transport bacteria. Science 277, 949–952. Killeen, K., Spriggs, D., and Mekalanos, J. (1999). Bacterial mucosal vaccines: Vibrio cholerae as a live attenuated vaccine/vector paradigm. Curr. Top. Microbiol. Immunol. 236, 237–254. Kim, J. J., Nottingham, L. K., Wilson, D. M., Bagarazzi, M. L., Tsai, A., Morrison, L. D., Javadian, A., Chalian, A. A., Agadjanyan, M. G., and Weiner, D. B. (1998). Engineering DNA vaccines via co-delivery of co-stimulatory molecule genes. Vaccine 16, 1828–1835. Kjeldsen, K., and Heron, I. S. O. (1985). Immunity against diphtheria 25–30 years after primary vaccination in childhood. Lancet 1, 900–902. Klavinskis, L. S., Barnfield, C., Gao, L., and Parker, S. (1999). Intranasal immunization with plasmid DNA-lipid complexes elicits mucosal immunity in the female genital and rectal tracts. J. Immunol. 162, 254–262. Klinman, D. M., Yamshchikov, G., and Ishigatsubo, Y. (1997). Contribution of CpG motifs to the immunogenicity of DNA vaccines. J. Immunol. 158, 3635–3639. Kohbata, S., Yokoyama, H., and Yabuuchi, E. (1986). Cytopathogenic effect of Salmonella typhi-GIFU 100007 on M cells of murine ileal Peyer’s patches in ligated ileal loops: An ultrastructural study. Microbiol. Immunol. 30, 1225–1237. Kolle, W., and von Wassermann, A. (1912). Handbuch der pathogenen Mikroorganismen, pp. 555–631. Gustav Fuscher Verlag Jena. Kouacsovics-Bankowski, M., and Rock, K. L. (1995). A phagosome-to-cytosol pathway for exogenous antigens presented on MHC class I molecules. Science 267, 243–246. Kremer, L., Dupre, L., Riveau, G., Capron, A., and Locht, C. (1998). Systemic and mucosal immune responses after intranasal administration of recombinant Mycobacterium bovis bacillus Calmette–Guerin expressing glutathione S-transferase from Schistosoma haematobium. Infect. Immun. 66, 5669–5676. Kremer, L., Riveau, G., Baulard, A., Capron, A., and Locht, C. (1996). Neutralizing antibody responses elicited in mice immunized with recombinant bacillus Calmette–Guerin producing the Schistosoma mansoni glutathione S-transferase. J. Immunol. 156, 4309–4317. Krieg, A. M., Love-Homan, L., Yi, A. K., and Harty, J. T. (1998). CpG DNA induces sustained IL-12 expression in vivo and resistance to Listeria monocytogenes challenge. J. Immunol. 161, 2428–2434. Kuhn, M., and Goebel, W. (1989). Identification of an extracellular protein of Listeria monocytogenes possibly involved in intracellular uptake by mammalian cells. Infect. Immun. 57, 55–61. Kuklin, N., Daheshia, M., Karem, K., Manickan, E., and Rouse, B. T. (1997). Induction of mucosal immunity against herpes simplex virus by plasmid DNA immunization. J. Virol. 71, 3138–3145. ¨ ber den Resorptionsvergang der corpuscula¨ren Bestandteile im Darm. Kumagai, K. (1922). U Kekkaku-Zassi 4, 429–431. Kuo, C., and Campbell, L. A. (1998). Is infection with Chlamydia pneumoniae a causative agent in atherosclerosis? Mol. Med. Today, 4, 426–430. Ladel, C. H., Blum, C., Dreher, A., Reifenberg, K., and Kaufmann, S. H. (1995). Protective role of gamma/delta T cells and alpha/beta T cells in tuberculosis [published erratum appars in Eur. J. Immunol. 1995 Dec: 25 (12): 3525]. Eur. J. Immunol. 25, 2877–2881.

76

HESS et al.

Ladel, C. H., Flesch, I. E., Arnoldi, J., and Kaufmann, S. H. (1994). Studies with MHCdeficient knock-out mice reveal impact of both MHC I- and MHC II-dependent T cell responses on Listeria monocytogenes infection [published erratum appears in J. Immunol. 1995 Apr 15; 154 (8): 4223]. J. Immunol. 153, 3116–3122. Lalvani, A., Brookes, R., Wilkinson, R. J., Malin, A. S., Pathan, A. A., Andersen, P., Dockrell, H., Pasvol, G., and Hill, A. V. (1998). Human cytolytic and interferon gamma-secreting CD8⫹ T lymphocytes specific for Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 95, 270–275. Langermann, S., Palaszynski, S., Sadziene, A., Stover, C. K., and Koenig, S. (1994a). Systemic and mucosal immunity induced by BCG vector expressing outer-surface protein A of Borrelia burgdorferi. Nature 372, 552–555. Langermann, S., Palaszynski, S. R., Burlein, J. E., Koenig, S., Hanson, M. S., Briles, D. E., and Stover, C. K. (1994b). Protective humoral response against pneumococcal infection in mice elicited by recombinant bacille Calmette–Guerin vaccines expressing pneumococcal surface protein A. J. Exp. Med. 180, 2277–2286. Le Minor, L. (1987). Request for an opinion, designation of Salmonella enterica sp Nov., nom. Rev., as the type and only species of the genus Salmonella. Int. J. System. Bacteriol. 37, 465–468. Levine, M. M., Herrington, D., Murphy, J. R., Morris, J. G., Losonsky, G., Tall, B., Lindberg, A. A., Svenson, S., Baqar, S., and Edwards, M. F. (1987). Safety, infectivity, immunogenicity, and in vivo stability of two attenuated auxotrophic mutant strains of Salmonella typhi, 541Ty and 543Ty, as live oral vaccines in humans. J. Clin. Invest. 79, 888–902. Li, Y., Yio, X. Y., and Mayer, L. (1995). Human intestinal epithelial cell-induced CD8(⫹) T-cell activation is mediated through CD8 and the activation of CD8-associated P 56 (LCK). J. Exp. Med. 182, 1079–1088. Lim, E. M., Lagranderie, M., Le Grand, R., Rauzier, J., Gheorghiu, M., Gicquel, B., and Winter, N. (1997). Recombinant Mycobacterium bovis BCG producing the N-terminal half of SIVmac251 Env antigen induces neutralizing antibodies and cytotoxic T lymphocyte responses in mice and guinea pigs. AIDS Res. Human Retroviruses 13, 1573–1581. Lindner, R., and Unanue, E. R. (1996). Distinct antigen MHC class II complexes generated by separate processing pathways. EMBO J. 15, 6910–6920. Lipford, G. B., Heeg, K., and Wagner, H. (1998). Bacterial DNA as immune cell activator. Trends Microbiol. 6, 496–500. Lipford, G. B., Lehn, N., Bauer, S., Heeg, K., and Wagner, H. (1994a). The immunodominant peptide from listeriolysin in Quil A liposomes vaccinates CD8⫹ cytolytic T cells and confers protection to infection. Immunol. Lett. 40, 101–104. Lipford, G. B., Wagner, H., and Heeg, K. (1994b). Vaccination with immunodominant peptides encapsulated in Quil A-containing liposomes induces peptide-specific primary CD8⫹ cytotoxic T cells. Vaccine 12, 73–80. Lowe, D. C., Savidge, T. C., Pickard, D., Eckmann, L., Kagnoff, M. F., Dougan, G., and Chatfield, S. N. (1999). Characterization of candidate live oral Salmonella typhi vaccine strains harboring defined mutations in aroA, aroC, and htrA. Infect. Immun. 67, 700–707. Lowrie, D. B., Tascon, R. E., Bonato, V. L. D., Lima, V. M. F., Faccioli, L. H., Stavropoulos, E., Colston, M. J., Hewinsohn, R. G., Moelling, K., and Silva, C. L. (1999). Therapy of tuberculosis in mice by DNA vaccination. Nature 400, 269–271. Lucey, D. R., Clerici, M., and Shearer, G. M. (1996). Type 1 and type 2 cytokine dysregulation in human infectious, neoplastic, and inflammatory diseases. Clin. Microbiol. Rev. 9, 532–562. Ma¨kela¨, P. H., and Hormaeche, C. E. (1997). Immunity to Salmonella. In ‘‘Host Response to Intracellular Pathogens’’ (S. H. E. Kaufmann, ed.), pp. 143–166, R. G. Landes Company, Austin.

EXPLOITING THE IMMUNE SYSTEM

77

Macatonia, S. E., Hosken, N. A., Litton, M., Vieira, P., Hsieh, C. S., Culpepper, J. A., Wysocka, M., Trinchieri, G., Murphy, K. M., and O’Gana, A. (1995). Dendritic cells produce IL-12 and direct the development of Th1 cells from naive CD4⫹ T cells. J. Immunol. 154, 5071–5079. MacMicking, J, Xie, Q., and Nathan, C. (1997). Nitric oxide and macrophage function. Annu. Rev. Immunol. 15, 323–350. Malik, Z. A., Denning, G. M., and Kusner, D. J. (2000). Inhibition of Ca2⫹ signaling by Mycobacterium tuberculosis is associated with reduced phagosome-lysosome fusion and increased survival within human macrophages. J. Exp. Med. 191, 287–302. Manoury, B., Hewitt, E. W., Morrice, N., Dando, P. M., Barrett, A. J., and Watts, C. (1998). An asparaginyl endopeptidase processes a microbial antigen for class II MHC presentation. Nature 396, 695–699. Manz, R. A., Lohning, M., Cassese, G., Thiel, A., and Radbruch, A. (1998). Survival of long-lived plasma cells is independent of antigen. Intl. Immunol. 10, 1703–1711. Manz R. A., Thiel, A., and Radbruch, A. (1997). Lifetime of plasma cells in the bone marrow. Nature 388, 133–134. Markiewicz, M. A., Girao, C., Opferman, J. T., Sun, J. L., Hu, Q. H., Agulnik, A. A., Bishop, C. E., Thompson, C. B., and Ashton-Rickardt, P. G. (1998). Long-term T cell memory requires the surface expression of self-peptide major histocompatibility complex molecules. Proc. Natl. Acad. Sci. USA 95, 3065–3070. Marriott, I., Hammond, T. G., Thomas, E. K., and Bost, K. L. (1999). Salmonella efficiently enter and survive within cultured CD11c⫹ dendritic cells initiating cytokine expression. Eur. J. Immunol. 29, 1107–1115. Mattingly, J. A., and Waksman, B. H. (1978). Immunological suppression after oral-administration of antigen. 1. Specific suppressor cells formed in rat Peyers patches after oraladministration of sheep erythrocytes and their systemic migration. J. Immunol. 121, 1878– 1883. Mayordomo, J. I., Zorina, T., Storkus, W. J., Zitvogel, L., Celluzzi, C., Falo, L. D., Melief, C. J., Ildstad, S. T., Kast, W. M., and Deleo, A. B. (1995). Bone marrow-derived dendritic cells pulsed with synthetic tumour peptides elicit protective and therapeutic antitumour immunity. Nat. Med. 1, 1297–1302. McCluskie, M. J., and Davis, H. L. (1999). Mucosal immunization with DNA vaccines. Microbes Infect. 1, 685–698. McCune, R. M., and Tompsett, R. M. W. (1957). Fate of Mycobacterium tuberculosis in mouse tissues as determined by the microbial enumeration technique. II. The conversion of tuberculous infection to the latent state by the administration of pyrazinamide and a companion drug. J. Exp. Med. 763–802. McHeyzer-Williams, M. G., and Ahmed, R. (1999). B cell memory and the long-lived plasma cell. Curr. Opin. Immunol. 11, 172–179. McMenamin, C., Pimm, C., McKersey, M., and Holt, P. G. (1994). Regulation of IgE responses to inhaled antigen in mice by antigen-specific gamma-delta T-cells. Science 265, 1869–1871. Medina, E., Paglia, P., Nikolaus, T., Muller, A., Hensel, M., and Guzman, C. A. (1999). Pathogenicity island 2 mutants of Salmonella typhimurium are efficient carriers for heterologous antigens and enable modulation of immune responses. Infect. Immun. 67, 1093– 1099. Miller, A., Lider, O., Roberts, A. B., Sporn, M. B., and Weiner, H. L. (1992). Suppressor T-cells generated by oral tolerization to myelin basic-protein suppress both in vitro and in vivo immune-responses by the release of transforming growth-factor-beta after antigenspecific triggering. Proc. Natl. Acad. Sci. USA 89, 421–425.

78

HESS et al.

Miller, M. A., Skeen, M. J., and Ziegler, H. K. (1995). Nonviable bacterial antigens administered with IL-12 generate antigen-specific T cell responses and protective immunity against Listeria monocytogenes. J. Immunol. 155, 4817–4828. Miller, M. A., Skeen, M. J., and Ziegler, H. K. (1997). A synthetic peptide administered with IL-12 elicits immunity to Listeria monocytogenes. J. Immunol. 159, 3675–3679. Mittru¨cker, H. W., Raupach, B., Ko¨hler, A., and Kaufmann, S. H. E. (2000). Role of antibodies in protective immunity against Salmonella typhimurium infection. J. Immunol. 164, 1648–1652. Miyagi, K., Kawakami, K., and Saito, A. (1997). Role of reactive nitrogen and oxygen intermediates in gamma interferon-stimulated murine macrophage bactericidal activity against Burkholderia pseudomallei. Infect. Immun. 65, 4108–4113. Modlin, R. L., Pirmez, C., Hofman, F. M., Torigian, V., Uyemura, K., Rea, T. H., Bloom, B. R., and Brenner, M. B. (1989). Lymphocytes bearing antigen-specific gamma delta T-cell receptors accumulate in human infectious disease lesions. Nature 339, 544–548. Moll, H., Fuchs, H., Blank, C., and Rollinghoff, M. (1993). Langerhans cells transport Leishmania major from the infected skin to the draining lymph node for presentation to antigen-specific T cells. Eur. J. Immunol. 23, 1595–1601. Mombaerts, P., Arnoldi, J., Russ, F., Tonegawa, S., and Kaufmann, S. H. (1993). Different roles of alpha beta and gamma delta T cells in immunity against an intracellular bacterial pathogen. Nature 365, 53–56. Moody, D. B., Reinhold, B. B., Guy, M. R., Beckman, E. M., Frederique, D. E., Furlong, S. T., Ye, S., Reinhold, V. N., Sieling, P. A., Modlin, R. L., Besra, G. S., and Porcelli, S. A. (1997). Structural requirements for glycolipid antigen recognition by CD1b-restricted T cells. Science 278, 283–286. Munder, M., Mallo, M., Eichmann, K., and Modolell, M. (1998). Murine macrophages secrete inferferon gamma upon combined stimulation with interleukin (IL)-12 and IL18: A novel pathway of autocrine macrophage activation. J. Exp. Med. 187, 2103–2108. Murali-Krishna, K., Altman, J. D., Suresh, M., Sourdive, D. J. D., Zajac, A. J., Miller, J. D., Slansky, J., and Ahmed, R. (1998). Counting antigen-specific CD8 T cells: A reevaluation of bystander activation during viral infection. Immunity 8, 177–187. Murray, P. J., Aldovini, A., and Young, R. A. (1996). Manipulation and potentiation of antimycobacterial immunity using recombinant bacille Calmette–Guerin strains that secrete cytokines. Proc. Natl. Acad. Sci. USA 93, 934–939. Noll, A., Bucheler, N., Bohn, E., and Autenrieth, I. B. (1997). Microbial heat shock proteins as vaccine. Behring Institute Mitteilungen 98, 87–98. Noll, A., Bucheler, N., Bohn, E., Schirmbeck, R., Reimann, J., and Autenrieth, I. B. (1999). DNA immunization confers systemic, but not mucosal, protection against enteroinvasive bacteria. Eur. J. Immunol. 29, 986–996. O’Brien, L., Roberts, B., and Andrew, P. W. (1996). In vitro interaction of Mycobacterium tuberculosis and macrophages; Activation of anti-mycobacterial activity of macrophages and mechanisms of anti-mycobacterial immunity. Curr. Top. Microbiol. Immunol. 215, 97–101. O’Callaghan, D., Maskell, D., Liew, F. Y., Easmon, C. S., and Dougan, G. (1988). Characterization of aromatic- and purine-dependent Salmonella typhimurium: attention, persistence, and ability to induce protective immunity in BALB/c mice. Infect. Immun. 56, 419–423. O’Callaghan, D., Maskell, D., Tite, J., and Dougan, G. (1990). Immune responses in BALB/ c mice following immunization with aromatic compound or purine-dependent Salmonella typhimurium strains. Immunology 69, 184–189.

EXPLOITING THE IMMUNE SYSTEM

79

Paglia, P., Arioli, I., Frahm, N., Chakraborty, T., Colombo, M. P., and Guzman, C. A. (1997). The defined attenuated Listeria monocytogenes delta mp12 mutat is an effective oral vaccine carrier to trigger a long-lasting immune response against a mouse fibrosarcoma. Eur. J. Immunol. 27, 1570–1575. Paglia, P., Chiodoni, C., Rodolfo, M., and Colombo, M. P. (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. Paglia, P., Medina, E., Arioli, I., Guzman, C. A., and Colombo, M. P. (1998). Gene transfer in dendritic cells, induced by oral DNA vaccination with Salmonella typhimurium, results in protective immunity against a murine fibrosarcoma. Blood 92, 3172–3176. Pamer, E. G. (1994). Direct sequence identification and kinetic analysis of an MHC class I-restricted Listeria monocytogenes CTL epitope. J. Immunol. 152, 686–694. Pamer, E. G. (1998). Cell-mediated immunity: The role of bacterial protein secretion. Curr. Biol. 8, 457–460. Pamer, E. G., Harty, J. T., and Bevan, M. J. (1991). Precise prediction of a dominant class I MHC-restricted epitope of Listeria monocytogenes. Nature 353, 852–855. Pan, Z. K., Ikonomidis, G., Lazenby, A., Pardoll, D., and Paterson, Y. (1995a). A recombinant Listeria monocytogenes vaccine expressing a model tumour antigen protects mice against lethal tumour cell challenge and causes regression of established tumors. Nat. Med. 1, 471–477. Pan, Z. K., Ikonomidis, G., Pardoll, D., and Paterson, Y. (1995b). Regression of established tumors in mice mediated by the oral administration of a recombinant Listeria monocytogenes vaccine. Cancer Res. 55, 4776–4779. Pang, T., Levine, M. M., Ivanoff, B., Wain, J., and Finlay, B. B. (1998). Typhoid feverimportant issues still remain. Trends Microbiol. 6, 131–133. Panum, P. L. (1847). Beobachtungen u¨ber das Masern Contagium. Virchows Archiv. 1, 492–502. Parrish, N. M., Dick, J. D., and Bishai, W. R. (1998). Mechanisms of latency in Mycobacterium tuberculosis. Trends Microbiol. 6, 107–112. Paterson, Y. (1999). The relationship between intracellular bacterial life-styles and immune responsiveness to bacterial delivered antigens. In ‘‘Intracellular Bacterial Vaccine Vectors’’ (Y. Paterson, ed.), pp. 1–23. John Wiley & Sons, New York. Paul, J. R., Riordan, J. T., and Melnick, J. L. (1951). Antibodies to three different antigenic types of poliomyelitis virus in sera from North Alaskan Eskimos. Am. J. Hygiene 54, 275–285. Pelicic, V., Jackson, M., Reyrat, J. M., Jacobs, W. R. J., Gicquel, B., and Guilhot, C. (1997). Efficient allelic exchange and transposon mutagenesis in Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 94, 10955–10960. Peterson, E. M., Cheng, X., Motin, V. L., and delaMaza, L. M. (1997). Effect of immunoglobulin G isotype on the infectivity of Chlamydia trachomatis in a mouse model of intravaginal infection. Infect. Immun. 65, 2693–2699. Pfeffer, K., Schoel, B., Gulle, H., Kaufmann, S. H. E., and Wagner, H. (1990). Primary responses of human T cells to mycobacteria: A frequent set of 웂/␦ T cells are stimulated by protease-resistant ligands. Eur. J. Immunol. 20, 1175–1179. Pfeifer, J. D., Wick, M. J., Roberts, R. L., Findlay, K., Normark, S. J., and Harding, C. V. (1993). Phagocytic processing of bacterial antigens for class I MHC presentation to T cells. Nature 361, 359–362. Piedrafita, D. (1999). Protective immune responses induced by vaccination with an expression genomic library of leishmania major. J. Immunol. 163, 1467–1472.

80

HESS et al.

Pope, M., and Kotlarski, I. (1994). Detection of Salmonella-specific L3T4⫹ and Lyt-2⫹ T cells which can proliferate in vitro and mediate delayed-type hypersensitivity reactivity. Immunology 81, 183–191. Pope, M., Kotlarski, I., and Doherty, K. (1994). Induction of Lyt-2⫹ cytotoxic T lymphocytes following primary and secondary Salmonella infection. Immunology 81, 177–182. Porcelli, S., Morita, C. T., and Brenner, M. B. (1992). CD1b restricts the response of human Cd4-8-T lymphocytes to a microbial antigen. Nature 360, 593–597. Porcelli, S. A. (1995). The CD1 family: A third lineage of antigen-presenting molecules. Adv. Immunol. 59, 1–98. Porcelli, S. A., and Modlin, R. L. (1999). The CD1 system: Antigen-presenting molecules for T cell recognition of lipids and glycolipids. Annu. Rev. Immunol. 17, 297–329. Porcelli, S. A., Segelke, B. W., Sugita, M., Wilson, I. A., and Brenner, M. B. (1998). The CD1 family of lipid antigen-presenting molecules. Immunol. Today 19, 362–368. Porgador, A., Irvine, K. R., Iwasaki, A., Barber, B. H., Restifo, N. P., and Germain, R. N. (1998). Predominant role for directly transfected dendritic cells in antigen presentation to CD8⫹ T cells after gene gun immunization. J. Exp. Med. 188, 1075–1082. Power, C. A., Wei, G., and Bretscher, P. A. (1998). Mycobacterial dose defines the Th1/Th2 nature of the immune response independently of whether immunization is administered by the intravenous, subcutaneous, or intradermal route. Infect. Immun. 66, 5743–5750. Pulendran, B., Smith, J. L., Caspary, G., Brasel, K., Pettit, D., Maraskovsky, E., and Maliszewski, C. R. (1999). Distinct dendritic cell subsets differentially regulate the class of immune response in vivo. Proc. Natl. Acad. Sci. USA 96, 1036–1041. Raposo, G., Nijman, H. W., Stoorvogel, W., Leijendecker, R., Harding, C. V., Melief, C. J., and Geuze, H. J. (1996). B lymphocytes secrete antigen-presenting vesicles. J. Exp. Med. 183, 1161–1172. Redman, T. K., Harmon, C. C., Michalek, S. M., Hoiseth, S. K., and Stocker, B. A. (1996). Oral immunization with recombinant Salmonella typhimurium expressing surface protein antigen A (SpaA) of Streptococcus sobrinus: Effects of the Salmonella virulence plasmid on the induction of protective and sustained humoral responses in rats. Aromatic-dependent Salmonella typhimurium are non-virulent and effect as live vaccines. Vaccine 14, 868–878. Reimann, J., and kaufmann, S. H. (1997). Alternative antigen processing pathways in antiinfective immunity. Cur. Opin. Immunol. 9, 462–469. Rhoades, E. R., and Orme, I. M. (1997). Susceptibility of a panel of virulent strains of Mycobacterium tuberculosis to reactive nitrogen intermediates. Infect. Immun. 65, 1189– 1195. Robbe-Saule, V., and Norel, F. (1999). The rpoS mutant allele of Salmonella typhi Ty2 is identical to that of the live typhoid vaccine vaccine Ty21a. FEMS Microbiol. Lett. 170, 141–143. Robinson, D., Shibuya, K., Mui, A., Zonin, F., Murphy, E., Sana, T., Hartley, S. B., Menon, S., Kastelein, R., Bazan, F., and O. Garra, A. (1997). IGIF does not drive Th1 development but synergizes with IL-12 for interferon-gamma production and activates IRAK and NFkappaB. Immunity 7, 571–581. Rodriguez, A., Regnault, A., Kleijmeer, M., Ricciardi-Castagnoli, P., and Amigorena, S. (1999). Selective transport of internalized antigens to the cytosol for MHC class I presentation in dendritic cells. Nature Cell Biol. 1, 362–368. Rolph, M. S., and Ramshaw, I. A. (1997). Recombinant viruses as vaccines and immunological tools. Curr. Opin. Immunol. 9, 517–524. Roman, M., Martin-Orozco, E., Goodman, J. S., Nguyen, M. D., Sato, Y., Ronaghy, A., Kornbluth, R. S., Richman, D. D., Carson, D. A., and Raz, E. (1997). Immunostimulatory DNA sequences function as T helper-1-promoting adjuvants. Nat. Med. 3, 849–854.

EXPLOITING THE IMMUNE SYSTEM

81

Rosat, J. P., Grant, E. P., Beekman, E. M., Dascher, C. C., Sieling, P. A., Frederique, D., Modlin, R. L., Porcelli, S. A., Furlong, S. T., and Brenner, M. B. (1999). CD1-restricted microbial lipid antigen-specific recognition found in the CD8(⫹) alpha beta T cell pool. J. Immunol. 162, 366–371. Russell, D. G., Sturgill-Koszycki, S., Vanheyningen, T., Collins, H., and Schaible, U. E. (1997). Why intraceullar parasitism need not be a degrading experience for Mycobacterium. Philosoph. Trans. Roy. Soc. London—Series B: Biol. Sci. 352, 1303–1310. Russmann, H., Shams, H., Poblete, F., Fu, Y., Galan, J. E., and Donis, R. O. (1998). Delivery of epitopes by the Salmonella type III secretion system for vaccine development. Science 281, 565–568. Ryan, E. T., Butterton, J. R., Smith, R. N., Carroll, P. A., Crean, T. I., and Calderwood, S. B. (1997). Protective immunity against Clostridium difficile toxin A induced by oral immunization with a live, attenuated Vibrio cholerae vector strain. Infect. Immun. 65, 2941–2949. Saito, H., Kanamori, Y., Takemori, T., Nariuchi, H., Kubota, E., Takahashi-Iwanaga, H., Iwanaga, T., and Ishikawa, H. (1998). Generation of intestinal T cells from progenitors residing in gut cryptopatches. Science 280, 275–278. Sallusto, F., Cella, M., Danieli, C., and Lanzavecchia, A. (1995). Dendritic cells use macropinocytosis and the mannose receptor to concentrate macromolecules in the major histocompatibility complex class II compartment: Downregulation by cytokines and bacterial products. J. Exp. Med. 182, 389–400. Sansonetti, P. J., and Phalipon, A. (1999). M cells as ports of entry for enteroinvasive pathogens: Mechanisms of interaction, consequences for the disease process. Sem. Immunol. 11, 193–203. Sawyer, W. A. (1931). Persistance of yellow fever immunity. J. Prev. Med. 5, 413–428. Schaible, U. E., Collins, H. L., and Kaufmann, S. H. (1999). Confrontation between intracellular bacteria and the immune system. Adv. Immunol. 71, 267–377. Schaible, U. E., Sturgill-Koszycki, S., Schlesinger, P. H., and Russell, D. G. (1998). Cytokine activation leads to acidification and increases maturation of Mycobacterium aviumcontaining phagosomes in murine macrophages. J. Immunol. 160, 1290–1296. Schena, M., Heller, R. A., Theriault, T. P., Konrad, K., Lachenmeier, E., and Davis, R. W. (1998). Microarrays: Biotechnology’s discovery platform for functional genomics. Trends Biotechnol. 16, 301–306. Schirmbeck, R., Bohm, W., and Reimann, J. (1997). Stress protein (hsp73)-mediated TAPindependent processing of endogenous, truncated SV40 large T antigen for D-b-restricted peptide presentation. Eur. J. Immun. 27, 2016–2023. Schirmbeck, R., and Reimann, J. (1996). ‘‘Empty’’ Ld molecules capture peptides from endocytosed hepatitis B surface antigen particles for major histocompatibility complex class I-restricted presentation. Eur. J. Immunol. 26, 2812–2822. Schneider, J., Gilbert, S. C., Blanchard, T. J., Hanke, T., Robson, K. J., Hannan, C. M., Becker, M., Sinden, R., Smith, G. L., and Hill, A. V. (1998). Enhanced immunogenicity for CD8⫹ T cell induction and complete protective efficacy of malaria DNA vaccination by boosting with modified vaccinia virus Ankara. Nat. Med. 4, 397–402. Schoel, B., and Kaufmann, S. H. E. (1996). The unique role of heat shock proteins in infections. In ‘‘Stress Proteins in Medicine,’’ (Van Eden and Young, eds.), pp. 27–51. Marcel Dekker Inc., New York, Basel, Hong Kong. Schoel, B., Sprenger, S., and Kaufmann, S. H. (1994). Phosphate is essential for stimulation of V gamma 9V delta 2 T lymphocytes by mycobacterial low molecular weight ligand. Eur. J. Immunol. 24, 1886–1892.

82

HESS et al.

Schofield, L., McConville, M. J., Hansen, D., Campbell, A. S., Fraser-Reid, B., Grusby, M. J., and Tachado, S. D. (1999). CD1d-restricted immunoglobulin G formation to GPIanchored antigens mediated by NKT cells. Science 283, 225–229. Sedegah, M., Hedstrom, R., Hobart, P., and Hoffman, S. L. (1994). Protection against malaria by immunization with plasmid DNA encoding circumsporozoite protein. Proc. Natl. Acad. Sci. USA 91, 9866–9870. Sedegah, M., Jones, T. R., Kaur, M., Hedstrom, R., Hobart, P., Tine, J. A., and Hoffman, S. L. (1998). Boosting with recombinant vaccinia increases immunogenicity and protective efficacy of malaria DNA vaccine. Proc. Natl. Acad. Sci. USA 95, 7648–7653. Shen, H., Miller, J. F., Fan, X., Kolwyck, D., Ahmed, R., and Harty, J. T. (1998). Compartmentalization of bacterial antigens: Differential effects on priming of CD8 T cells and protective immunity. Cell 92, 535–545. Shen, H., Slifka, M. K., Matloubian, M., Jensen, E. R., Ahmed, R., and Miller, J. F. (1995). Recombinant Listeria monocytogenes as a live vaccine vehicle for the induction of protective anti-viral cell-mediated immunity. Proc. Natl. Acad. Sci. USA 92, 3987–3991. Shimada, S., Yano, O., and Tokunaga, T. (1986). In vivo augmentation of natural killer cell activity, with a deoxyribonucleic acid fraction of BCG. Jap. Natl. Cancer Inst. 74, 808–816. Shimizu, Y., Guidotti, L. G., Fowler, P., and Chisari, F. V. (1998). Dendritic cell immunization breaks cytotoxic T lymphocyte tolerance in hepatitis B virus transgenic mice. J. Immunol. 161, 4520–4529. Shiver, J. W., Ulmer, J. B., Donnelly, J. J., and Liu, M. A. (1996). Naked DNA vaccination. In ‘‘Concepts in Vaccine Development’’ (S. H. E. Kaumann, ed.), pp. 423–436. Walter de Gruyter, Berlin, New York. Sieling, P. A., Chatterjee, D., Porcelli, S. A., Prigozy, T. I., Mazzaccaro, R. J., Soriano, T., Bloom, B. R., Brenner, M. B., Kronenberg, M., Brennan, P. J., and Modlin, R. L. (1995). CD1-restricted T cell recognition of microbial lipoglycan antigens. Science 269, 227–230. Sigwart, D. F., Stocker, B. A., and Clements, J. D. (1989). Effect of a purA mutation on efficacy of Salmonella live-vaccine vectors. Infect. Immun. 57, 1858–1861. Silva, B. A., Gonzalez, C., Mora, G. C., and Cabello F. (1987). Genetic characteristics of the Salmonella typhi strain Ty21a vaccine [published erratum appears in J. Infect. Dis. 1987 Aug; 156(2): 421]. J. Infect. Dis. 155, 1077–1078. Silva, C. L., Silva, M. F., Pietro, R. C., and Lowrie, D. B. (1994). Protection against tuberculosis by passive transfer with T-cell clones recognizing mycobacterial heat-shock protein 65. Immunology 83, 341–346. Silva, R. A., Pais, T. F., and Appelberg, R. (1998). Evaluation of IL-12 in immunotherapy and vaccine design in experimental Mycobacterium avium infections. J. Immunol. 161, 5578–5585. Sin, J. I., Kim, J. J., Arnold, R. L., Shroff, K. E., McCallus, D., Pachuk, C., McElhiney, S. P., Wolf, M. W., Pompa-de Bruin, S.J., Higgins, T. J., Ciccarelli, R. B., and Weiner, D. B. (1999). IL-12 gene as a DNA vaccine adjuvant in a herpes mouse model: IL-12 enhances Th1-type CD4⫹ T cell-mediated protective immunity against herpes simplex virus-2 challenge. J. Immunol. 162, 2912–2921. Sinai, A. P., and Joiner, K. A. (1997). Safe haven: The cell biology of nonfusogenic pathogen vacuoles. Annu. Rev. Microbiol. 51, 415–462. Sirard, J. C., Fayolle, C., Dechastellier, C., Mock, M., Leclerc, C., and Berche, P. (1997). Intracytoplasmic delivery of listeriolysin O by a vaccinal strain of Bacillus anthracis induces CD8-mediated protection against Listeria monocytogenes. J. Immunol. 159, 4435–4443. Sizemore, D. R., Branstrom, A. A., and Sadoff, J. C. (1995). Attenuated Shigella as a DNA delivery vehicle for DNA-mediated immunization. Science 270, 299–302.

EXPLOITING THE IMMUNE SYSTEM

83

Skeen, M. J., and Ziegler, H. K. (1993). Induction of murine peritoneal gamma/delta T cells and their role in resistance to bacterial infection. J. Exp. Med. 178, 971–984. Slifka, M. K., and Ahmed, R. (1998). Long lived plasma cells: A mechanism for maintaining persistent antibody production. Curr. Opin. Immunol. 10, 252–258. Slifka, M. K., Shen, H., Matloubian, M., Jensen, E. R., Miller, J. F., and Ahmed, R. (1996). Antiviral cytotoxic T-cell memory by vaccination with recombinant Listeria monocytogenes. J. Virol. 70, 2902–2910. Sonnenberg, M. G., and Belisle, J. T. (1997). Definition of Mycobacterium tuberculosis culture filtrate proteins by two-dimensional polyacrylamide gel electrophoresis, N-terminal amino acid sequencing, and electrospray mass spectrometry. Infect. Immun. 65, 4515– 4524. Starnbach, M. N., Bevan, M. J., and Lampe, M. F. (1994). Protective cytotoxic T lymphocytes are induced during murine infection with Chlamydia trachomatis. J. Immunol. 153, 5183– 5189. Steffen, R., Schar, G., and Mosimann, J. (1981). Salmonella and Shigella infections in Switzerland, with special reference to typhoid vaccination for travellers. Scandinavian J. Infect. Dis. 13, 121–127. Steinhoff, U., Brinkmann, V., Klemm, U., Aichele, P., Seiler, P., Brandt, U., Bland, P. W., Prinz, I., Zu¨gel, U., Kaufmann, S. H. E. (1999). Autoimmune intestinal pathology induced by hsp60 specific CD8 T cells. Immunity 11, 349–358. Stenger, S., Hanson, D. A., Teitelbaum, R., Dewan, P., Niazi, K. R., Froelich, C. J., Ganz, T., Thoma-Uszynski, S., Melian, A., Bogdan, C., Porcelli, S. A., Bloom, B. R., Krensky, A. M., and Modlin, R. L. (1998). An antimicrobial activity of cytolytic T cells mediated by granulysin. Science 282, 121–125. Stenger, S., Mazzaccaro, R. J., Uyemura, K., Cho, S., Barnes, P. F., Rosat, J. P., Sette, A., Brenner, M. B., Porcelli, S. A., Bloom, B. R., and Modlin, R. L. (1997). Differential effects of cytolytic T cell subsets on intracellular infection. Science 276, 1684–1687. Stenger, S., and Modlin, R. L. (1999). T cell mediated immunity to Mycobacterium tuberculosis Curr. Opin. Microbiol. 2, 89–93. Stephens, R. S. (1999). ‘‘Chlamydia: Intracellular Biology, Pathogenesis, and Immunity.’’ ASM Press. Stephens, R. S., Kalman, S., Lammel, C., Fan, J., Marathe, R., Aravind, L., Mitchell, W., Olinger, L., Tatusov, R. L., Zhao, Q., Koonin, E. V., and Davis, R. W. (1998). Genome sequence of an obligate intracellular pathogen of humans: Chlamydia trachomatis. Science 282, 754–759. Stocker, B. A. (1993). Attenuation in Salmonella by Auxotrophy. In ‘‘Biology of Salmonella’’ (F. Caballo, ed.), pp. 309–322. Plenum Press, New York. Stover, C. K., Bansal, G. P., Hanson, M. S., Burlein, J. E., Palaszynski, S. R., Young, J. F., Koenig, S., Young, D. B., Sadziene, A., and Barbour, A. G. (1993). Protective immunity elicited by recombinant bacille Calmette–Guerin (BCG) expressing outer surface protein A (OspA) lipoprotein: A candidate Lyme disease vaccine. J. Exp. Med. 178, 197–209. Stover, C. K., de, I. C., V, Fuerst, T. R., Burlein, J. E., Benson, L. A., Bennett, LT, Bansal, G. P., Young, J. F., Lee, M. H., and Hatfull, G. F. (1991). New use of BCG for recombinant vaccines. Nature 351, 456–460. Su, H., Feilzer, K., Caldwell, H. D., and Morrison, R. P. (1997). Chlamydia trachomatis genital tract infection of antibody-deficient gene knockout mice. Infect. Immun. 65, 1993– 1999. Su, H., Messer, R., Whitmire, W., Fischer, E., Portis, J. C., and Caldwell, H. D. (1998). Vaccination against chlamydial genital tract infection after immunization with dendritic cells pulsed ex vivo with nonviable Chlamydiae. J. Exp. Med. 188, 809–818.

84

HESS et al.

Sugita, M., and Brenner, M. B. (1995). Association of the invariant chain with major histocompatibility complex class I molecules directs trafficking to endocytic compartments. J. Biol. Chem. 270, 1443–1448. Svensson, M., Pfeifer, J., Stockinger, B., and Wick, M. J., (1997). Bacterial antigen delivery systems: Phagocytic processing of bacterial antigens for MHC-I and MHC-II presentation to T cells. Behring Institute Mitteilungen 98, 197–211. Svensson, M., Stockinger, B., and Wick, M. J. (1997). Bone marrow-derived dendritic cells can process bacteria for MHC-I and MHC-II presentation to T cells. J. Immunol. 158, 4229–4236. Svensson, M., and Wick, M. J. (1999). Classical MHC class I peptide presentation of a bacterial fusion protein by bone marrow-derived dendritic cells. Eur. J. Immunol. 29, 180–188. Swain, S. L. (1994). Generation and in-vivo persistence of polarized Th1 and Th2 memory cells. Immunity 1, 543–552. Szalay, G., Ladel, C. H., Blum, C., Brossay, L., Kronenberg, M., and Kaufmann, S. H. E. (1999a). Cutting edge: Anti-CD1 monoclonal antibody treatment reverses the production patterns of TGF-beta 2 and Th1 cytokines and ameliorates listeriosis in mice. J. Immunol. 162, 6955–6958. Szalay, G., Zu¨gel, U., Ladel, C. H., and Kaufmann, S. H. E. (1999b). Participation of group II CD1 molecules in the control of murine tuberculosis. Microbes Infect. 1, 1153–1158. Sztein, M. B., Tanner, M. K., Polotsky, Y., Orenstein, J. M., and Levine, M. M. (1995). Cytotoxic T lymphocytes after oral immunization with attenuated vaccine strains of Salmonella typhi in humans. J. Immunol. 155, 3987–3993. Sztein, M. B., Wasserman, S. S., Tacket, C. O., Edelman, R., Hone, D., Lindberg, A. A., and Levine, M. M. (1994). Cytokine production patterns and lymphoproliferative responses in volunteers orally immunized with attenuated vaccine strains of Salmonella typhi. J. Infect. Dis. 170, 1508–1517. Tacket, C. O., Hone, D. M., Curtiss, R., Kelly, S. M., Losonsky, G., Guers, L., Harris, A. M., Edelman, R., and Levine, M. M. (1992a). Comparison of the safety and immunogenicity of delta aroC delta aroD and delta cya delta crp Salmonella typhi strains in adult volunteers. Infect. Immun. 60, 536–541. Tacket, C. O., Hone, D. M., Losonsky, G. A., Guers, L., Edelman, R., and Levine, M. M. (1992b). Clinical acceptability and immunogenicity of CVD 908 Salmonella typhi vaccine strain. Vaccine 10, 443–446. Tacket, C. O., Kelly, S. M., Schodel, F., Losonsky, G., Nataro, J. P., Edelman, R., Levine, M. M., and Curtiss, R. (1997a). Safety and immunogenicity in humans of an attenuated Salmonella typhi vaccine vector strain expressing plasmid-encoded hepatitis B antigens stabilized by the Asd-balanced lethal vector system [published erratum appears in Infect. Immun. 1997 Sep; 65 (9): 3981]. Infect. Immun. 65, 3381–3385. Tacket, C. O., Losonsky, G., Taylor, D. N., Baron, L. S., Kopecko, D., Cryz, S., and Levine, M. M. (1991). Lack of immune response to the Vi component of a Vi-positive variant of the Salmonella typhi live oral vaccine strain Ty21a in human studies. J. Infect. Dis. 163, 901–904. Tacket, C. O., Sztein, M. B., Losonsky, G. A., Wasserman, S. S., Nataro, J. P., Edelman, R., Pickard, D., Dougan, G., Chatfield, S. N., and Levine, M. M. (1997b) Safety of live oral Salmonella typhi vaccine strains with deletions in htrA and aroC aroD and immune response in humans. Infect. Immun. 65, 452–456. Tamura, Y., Peng, P., Liu, K., Daou, M., and Srivastava, P. K. (1997). Immunotherapy of tumors with autologous tumor-derived heat shock protein preparations. Science 278, 117–120.

EXPLOITING THE IMMUNE SYSTEM

85

Tan, L. C., Gudgeon, N., Annels, N. E., Hansasuta, P., O’Callaghan, C. A., Rowland-Jones, S., McMichael, A. J., Rickinson, A. B., and Callan, M. F. C. (1999). A re-evaluation of the frequency of CD8(⫹) T cells specific for EBV in healthy virus carriers. J. Immunol. 162, 1827–1835. Tan, M. C., Mommaas, A. C., Drijfhout, J. W., Jordens, R., Onderwater, J. J., Verwoerd, D., Mulder, A. A., van der Heiden, A. N., Scheidegger, D., Oomen, L. C., Ottenhoff, T. H., Tulp, A., Neefjes, J. J., and Koning, F. (1997). Mannose receptor-mediated uptake of antigens strongly enhances HLA class II-restricted antigen presentation by cultured dendritic cells. Eur. J. Immunol. 27, 2426–2435. Tanaka, Y., Morita, C. T., Nieves, E., Brenner, M. B., and Bloom, B. R. (1995). Natural and synthetic non-peptide antigens recognized by human 웂␦ T cells. Nature 375, 155–158. Tanaka, Y., Sano, S., Nieves, E., De Libero, G., Rosa, D., Modlin, R. L., Brenner, M. B., Bloom, B. R., and Morita, C. T. (1994). Nonpeptide ligands for human 웂␦ T cells. Proc. Natl. Acad. Sci. USA 91, 8175–8179. Tanchot, C., Lemonnier, F. A., Perarnau, B., Freitas, A. A., and Rocha, B. (1997). Differential requirements for survival and proliferation of CD8 naive or memory T cells. Science 276, 2057–2062. Tanghe, A., Lefevre, P., Denis, O., D’Souza, S., Braibant, M., Lozes, E., Singh, Montgomery, D., Content, J., and Huygen, K. (1999). Immunogenicity and protective efficacy of tuberculosis DNA vaccines encoding putative phosphate transport receptors. J. Immunol. 162, 1113–1119. Tascon, R. E., Colston, M. J., Ragno, S., Stavropoulos, E., Gregory, D., and Lowrie, D. B. (1996). Vaccination against tuberculosis by DNA injection. Nat. Med. 2, 888–892. Teitelbaum, R., Glatman-Freedman, A., Chen, B., Robbins, J. B., Unanue, E., Casadevall, A., and Bloom, B. R. (1998). A mAb recognizing a surface antigen of Mycobacterium tuberculosis enhances host survival. Proc. Natl. Acad. Sci. USA 95, 15688–15693. Teitelbaum, R., Schubert, W., Gunther, L., Kress, Y., Macaluso, F., Pollard, J. W., McMurray, D. N., and Bloom, B. R. (1999). The M cell as a portal of entry to the lung for the bacterial pathogen Mycobacterium tuberculosis. Immunity 10, 641–650. Teitell, M., Holcombe, H. R., Brossay, L., Hagenbaugh, A., Jackson, M. J., Pond, L., Balk, S. P., Terhorst, C., Peterson, P. A., and Kronenberg, M. (1997). Nonclassical behavior of the mouse CD1 class I-like molecule. J. Immunol. 158, 2143–2149. Tew, J. G., and Mandel, T. E. (1979). Prolonged antigen half-life in the lymphoid follicles of specifically immunized mice. Immunology 37, 69–76. Tokunaga, T., Yamamoto, H., Shimada, S., Abe, H., Fukuda, T., Fujiisawa, Y., Furutani, Y., Yano, O., Kataoka, T., and Sudo, T. (1984). Antitumor activity of deoxyribonucleic acid fraction from Mycobacterium bovis BCG. Isolation, physicochemical characterization, and tumor activity. Jap. Natl. Cancer Inst. 72, 955–962. Turner, J., and Dockrell, H. M. (1996). Stimulation of human peripheral blood mononuclear cells with live Mycobacterium bovis BCG activates cytolytic CD8⫹ T cells in vitro. Immunology 87, 339–342. Turner, S. J., Carbone, F. R., and Strugnell, R. A. (1993). Salmonella typhimurium delta aroA delta aroD mutants expressing a foreign recombinant protein induce specific major histocompatibility complex class I-restricted cytotoxic T lymphocytes in mice. Infect. Immun. 61, 5374–5380. Tzschaschel, B. D., Klee, S. R., Delorenzo, L. V., Timmis, K. N., and Guzman, C. A. (1996). Towards a vaccine candidate against Shigella dysenteriae 1: Expression of the Shiga toxin B-subunit in an attenuated Shigella flexneri aroD carrier strain. Microb. Pathogen. 21 277–288.

86

HESS et al.

Uchijima, M., Yoshida, A., Nagata, T., and Koide, Y. (1998). Optimization of codon usage of plasmid DNA vaccine is required for the effective MHC class I-restricted T cell responses against an intracellular bacterium. J. Immunol. 161, 5594–5599. Udono, H., Levey, D. L., and Srivastava, P. K. (1994). Cellular requirements for tumorspeicific immunity elicited by heat shock proteins: Tumor rejection antigen gp96 primes CD8⫹ T cells in vivo. Proc. Natl. Acad. Sci. USA 91, 3077–3081. Udono, H., and Srivastava, P. K. (1993). Heat shock protein 70-associated peptides elicit specific cancer immunity. J. Exp. Med. 178, 1391–1396. Ulmer, J. B., Donnelly, J. J., Parker, S. E., Rhodes, G. H., Felgner, P. L., Dwarki, V. J., Gromkowski, S. H., Deck, R. R., DeWitt, C. M., and Friedman, A. (1993). Heterologous protection against influenza by injection of DNA encoding a viral protein. Science 259, 1745–1749. Valentine, P. J., Meyer, K., Rivera, M. M., Lipps, C., Pauza, D., Maziarz, R. T., So, and Heffron, F. (1996). Induction of SIV capsid-specific CTL and mucosal sIgA in mice immunized with a recombinant S. typhimurium aroA mutant. Vaccine 14, 138–146. Vancott, J. L., Chatfield, S. N., Robets, M., Hone, D. M., Hohmann, E. L., Pascual, D. W., Yamamoto, M., Kiyono, H., and McGhee, J. R. (1998). Regulation of host immune responses by modification of Salmonella virulence genes. Nat. Med. 4, 1247–1252. Van Eden, W., Holoshitz, J, and Cohen, I. (1987). Antigenic mimicry between mycobacteria and cartilage proteoglycans: The model of adjuvant arthritis. Concepts Immunopathol. 4, 144–170. Van Eden, W., Thole, J. E., van der Zee, R., Noordji, A., van Embden, J. D., Hensen, E. J., and Cohen, I. R. (1988). Cloning of the mycobacterial epitope recognized by T lymphocytes in adjuvant arthritis. Nature 331, 171–173. Venkataprasad, N. (1999). Induction of cellular immunity to a mycobacterial antigen adsorbed on lamellar particles of lactide polymers. Vaccine 17, 1814–1819. Verbon, A., Hartskeerl, R. A., Schuitema, A., Kolk, A. H., Young, D. B., and Lathigra, R. (1992). The 14,000-molecular-weight antigen of Mycobacterium tuberculosis is related to the alpha-crystallin family of low-molecular-weight heat shock proteins. J. Bacteriol. 174, 1352–1359. Verma, N. K., Ziegler, H. K., Stocker, B. A. D., Schoolnik, G. K. (1995). Induction of a cellular immune response to a defined T-cell epitope as an insert in the flagellin of a live vaccine strain of Salmonella. Vaccine 13, 235–244. Vogel, F. R., and Powell, M. F. (1995). A compendium of vaccine adjuvants and excipients. In ‘‘Vaccine Design: The Subunit and Adjuvant Approach’’ (M. F. Powell and M. J. Newman, eds.), pp. 141–228. Plenum Press, New York. Wagner, H. (1999). Bacterial CpG DNA activates immune cells to signal infectious danger. Adv. Immunol. 73, 329–368. Wagner, R. D., Steinberg, H., Brown, J. F., and Czuprynski, C. J. (1994). Recombinant interleukin-12 enhances resistance of mice to Listeria monocytogenes infection. Microb. Pathogen. 17, 175–186. Wahdan, M. H., Serie, C., Cerisier, Y., Sallam, S., and Germanier, R. (1982). A controlled field trial of live Salmonella typhi strain Ty 21a oral vaccine against typhoid: Three-year results. J. Infect. Dis. 145, 292–295. Wang, R., Doolan, D. L., Le, T. P., Hedstrom, R. C., Coonan, K. M., Charoenvit, Y., Jones T. R., Hobart, P., Margalith, M., Ng, J., Weiss, W. R., Sedegah, M., DE, Taisne, C., Norman, J. A., and Hoffman, S. L. (1998). Induction of antigen-specific cytotoxic T lymphocytes in humans by a malaria DNA vaccine. Science 282, 476–480. Watts, C. (1997). Capture and processing of exogenous antigens for presentation on MHC molecules. Annu. Rev. Immunol. 15, 821–850.

EXPLOITING THE IMMUNE SYSTEM

87

Weiskirch, L., and Paterson, Y. (1999). The use of Listeria monocytogenes recombinants as vaccine vectors in infectious and neoplastic disease. In ‘‘Intracellular Bacterial Vaccine Vectors’’ (Y. Paterson, ed.), pp. 223–260, John Wiley & Sons, New York. Weiskirch, L. M., and Paterson, Y. (1997). Listeria monocytogenes: A potent vaccine vector for neoplastic and infectious disease. Immunol. Rev. 158, 159–169. White, D. W., Wilson, R. L., and Harty, J. T. (1996). CD8⫹ T cells in intracellular bacterial infections of mice. Res. Immunol. 147, 519–524. Wick, M. J., and Pfeifer, J. D. (1996). Major histocompatibility complex class I presentation of ovalbumin peptide 257-264 from exogenous sources: Protein context influences the degree of TAP-independent presentation. Eur. J. Immunol. 26, 2790–2799. Wiker, H. G., and Harboe, M. (1992). The antigen 85 complex: A major secretion product of Mycobacterium tuberculosis. Microbiol. Rev. 56, 648–661. Winkle, S. (1997). ‘‘Geisseln der Menschheit: Kulturgeschichte der Seuchen.’’ Winter, N., Lagranderie, M., Gangloff, S., Leclerc, C., Gheorghiu, M., and Gicquel, B. M. (1995). Recombinant BCG strains expressing the SIVmac251nef gene induce proliferative and CTL responses against net synthetic peptides in mice. Vaccine 13, 471–478. Wong, D. K., Lee, B. Y., Horwitz, M. A., and Gibson, B. W. (1999). Identification of fur, aconitase, and other proteins expressed by Mycobacterium tuberculosis under conditions of low and high concentrations of iron by combined two-dimensioanl gel electrophoresis and mass spectrometry. Infect. Immun. 67, 327–336. World Health Organization. (1999). ‘‘The World Health Report 1999: Making a Difference.’’ Wuenscher, M. D., Kohler, S., Bubert, A., Gerike, U., and Goebel, W. (1993). The iap gene of Listeria monocytogenes is essential for cell viability, and its gene product, p60, has bacteriolytic activity. J. Bacteriol. 175, 3491–3501. Xiang, Z., and Ertl, H. C. (1995). Manipulation of the immune response to a plasmidencoded viral antigen by coinoculation with plasmids expressing cytokines. Immunity 2, 129–135. Yamamoto, S., Yamamoto, T., Kataoka, T., Kuramoto, E., Yano, O., and Tokunaga, T. (1992). Unique palindromic sequences in synthetic oligonucleotides are required to induce IFN [correction of INF] and augment IFN-mediated [correction of INF] natural killer activity. J. Immunol. 148, 4072–4076. Yasutomi, Y., Koenig, S., Haun, S. S., Stover, C. K., Jackson, R. K., Conard, P., Conley, A. J., Emini, E. A., Fuerst, T. R., and Letvin, N. L. (1993). Immunization with recombinant BCG-SIV elicits SIV-specific cytotoxic T lymphocytes in rhesus monkeys. J. Immunol. 150, 3101–3107. Yasutomi, Y., Koenig, S., Woods, R. M., Madsen, J., Wassef, N. M., Alving, C. R., Klein, H. J., Nolan, T. E., Boots, L. J., and Kessler, J. A. (1995). A vaccine-elicited, single viral epitope-specific cytotoxic T lymphocyte response does not protect against intravenous, cell-free simian immunodeficiency virus challenge. J. Virol. 69, 2279–2284. York, I. A., and Rock, K. L. (1996). Antigen processing and presentation by the class I major histocompatibility complex. Annu. Rev. Immunol. 14, 369–396. Yuan, Y., Crane, D. D., Simpson, R. M., Zhu, Y. Q., Hickey, M. J., Sherman, D. R., and Barry, C. E. (1998). The 16-kDa alpha-crystallin (Acr) protein of Mycobacterium tuberculosis is required for growth in macrophages. Proc. Natl. Acad. Sci. USA 95, 9578– 9583. Zarozinski, C. C., Fynan, E. F., Selin, L. K., Robinson, H. L., and Welsh, R. M. (1995). Protective CTL-dependent immunity and enhanced immunopathology in mice immunized by particle bombardment with DNA encoding an internal virion protein. J. Immunol. 154, 4010–4017.

88

HESS et al.

Zhang, D., Yang, X., Berry, J., Shen, C., McClarty, G., and Brunham, R. C. (1997). DNA vaccination with the major outer-membrane protein gene induces acquired immuniy to Chlamydia trachomatis (mouse pneumonitis) infection. J. Infect. Dis. 176, 1035–1040. Zhang, J. P., and Stephens, R. S. (1992). Mechanism of C. trachomatis attachment to eukaryotic host cells. Cell 69, 861–869. Zitvogel, L., Regnault, A., Lozier A, Wolfers, J., Flament, C. T., Tenza, D., RicciardiCastagnoli, P., Raposo, G., and Amigorena, S. (1998). Eradication of established murine tumors using a novel cell-free vaccine: Dendritic cell drived exosomes. Nat. Med. 4, 594–600. Zinsser, H., and Bayne-Jones, S. (1934). ‘‘A Textbook of Bacteriology.’’ D. Appleton Century Co., New York, London. Zu¨gel, U., and Kaufmann, S. H. E. (1999). Role of heat shock proteins in protection from and pathogenesis of infectious diseases. Clin. Microbiol. Rev. 12(1), 19–39. Zu¨gel, U., and Kaufmann, S. H. E. (1999). Gp96-peptide vaccination of mice against intracellular bacteria. Submitted.

ADVANCES IN IMMUNOLOGY, VOL. 75

The Cytoskeleton in Lymphocyte Signaling A. BAUCH,* F. W. ALT,† G. R. CRABTREE,* AND S. B. SNAPPER†‡ *Department of Pathology and Developmental Biology, Howard Hughes Medical Institute, Beckman Center for Molecular and Genetic Medicine, Stanford University School of Medicine, Stanford, California 94305; †Howard Hughes Medical Institute, The Children’s Hospital; Center for Blood Research; Department of Genetics, Harvard Medical School; Boston, Massachusetts 02115; ‡Gastrointestinal Unit (Medical Services) and the Center for the Study of Inflammatory Bowel Disease, Massachusetts General Hospital; Department of Medicine, Harvard Medical School, Boston, Massachusetts 02115

I. Introduction

Although lymphocytes are one of the most symmetrical cells, they receive their activating signals directionally, in the sense that these signals come from cell–cell contact, almost always from a single cell bearing antigen bound to MHC. Although there are specialized features of antigen presentation, this cell–cell contact is probably very similar to the cell–cell interactions that lead to activation or developmental decisions in many cell types that receive directional signals limited by a basement membrane or other cells. However, one feature that is almost certainly unique to lymphocytes is the very small number of antigen–MHC complexes of the correct specificity for a neighboring lymphocyte, possibly only a few dozen or even fewer. Thus, signaling must be initiated with a very small number of engaged receptors. With these special and general considerations in mind, we review progress in understanding how actin polymerization and the cytoskeleton become important in lymphocyte signaling. We try to address several unresolved questions: (1) Is the requirement for actin polymerization simply a reflection of the need for an intact cytoskeleton or is actin polymerization more directly involved in signaling? (2) What are the roles of the actin Cap and the supramolecular activation complexes (SMACs) seen after antigen binding? (3) Is actin polymerization necessary only for orienting the lymphocyte to the antigen-presenting cell or is it actually involved in activating signaling molecules? (4) Do the involvement of the cytoskeleton and actin polymerization help to solve the special problems associated with signaling in lymphocytes, such as the low number of engaged receptors and directionality of the activation process? (5) What are the roles and mechanisms of action of upstream regulators of the actin cytoskeleton such as Vav1, Rac, Cdc42, and WASP? (6) Finally, how do Vav1, Rac, and the actin cytoskeleton become involved in the activation of transcription and are they likely to be involved in the function of the actin-dependent, PIP2-dependent BAF chromatin-remodeling complex? 89

Copyright 䉷 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2776/00 $35.00

90

BAUCH et al.

II. Overview

Many cell types appear to have a critical requirement for actin-dependent polarization for signaling. For example, in yeast at least a dozen signaling molecules rapidly cluster after contact with pheromone (Ayscough and Drubin, 1998). In Drosophila the asymmetric clustering of signaling molecules such as numb, prospero, and inscuteable is actin-dependent and is critical for asymmetric cell division (Knoblich et al., 1997). In lymphocytes a variety of evidence indicates that the actin-dependent clustering of signaling molecules is required for activation (Dustin et al., 1998; Penninger and Crabtree, 1999). The evolutionarily conserved use of clustering of signaling molecules in response to both geometrically fixed and soluble activators suggests that clustering of signaling molecules is likely to serve some fundamental function, but as yet this function is undefined. In lymphocytes, these actin-dependent structures have specific morphologies and are referred to as Caps or SMACs and are likely to be related to or require the formation of lipid rafts—the lateral assemblies of glycosphingolipids and cholesterol in the plasma membrane (Harder and Simons, 1999; Montixi et al., 1998; Moran and Carrie Miceli, 1998; Viola et al., 1999; Xavier et al., 1998). A. THE STRUCTURAL FEATURES OF THE CAP AND OTHER ACTIN-DEPENDENT SIGNALING COMPLEXES De Petris and colleagues first noted that cell surface molecules of lymphocytes rapidly cluster into a Cap shortly after contact with activating molecules such as anti-immunoglobulin antibodies or concanavalin A (de Petris, 1975; de Petris and Raff, 1973). Actin polymerization, but not microtubular polymerization, is required for formation of the Cap. Formation of the Cap requires about 3 to 5 min and can be initiated by antibodies against surface-bound immunoglobulins, the T cell receptor (TCR), Thy-1, CD3, CD2, CD5, and CD28. With each stimulus, co-capping of other receptors or associated signaling molecules can be observed. The nonspecific nature of the stimulus led to a lack of interest in this phenomenon and indeed it was viewed by many as an epiphenomon, accompanying activation but not necessarily critical to activation. A structure similar to the Cap forms with similar kinetics at the contact site of a T cell and an antigen-presenting cell (APC). Kupfer and colleagues observed that T cells polarize toward the APC, thereby forming organized spatial domains referred to as SMACs (Dustin et al., 1998; Monks et al., 1998). This process involves rearrangements of the T cell cytoskeleton as well as redistribution of cell surface molecules. Some molecules, such as TCR, CD4, CD2, CD28, and PKC⍜, are enriched in a central zone,

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

91

whereas others, such as LFA-1, CD45, and talin, are localized to peripheral zones. The major difference between classical Caps and SMACs is their induction by different stimuli. Caps are induced by an antibody-mediated (or other soluble stimulant-mediated) activation of either B or T cells. SMACs, on the other hand, are induced by the complex integration of signals from different receptors, as is the case when a T cell encounters an antigenspecific APC. However, both Caps and SMACs require actin polymerization and appear to involve similar signaling molecules (de Petris, 1974; Dustin et al., 1998; Monks et al., 1998). B. POSSIBLE FUNCTIONS OF LIGAND-INDUCED, ACTIN-DEPENDENT ASSEMBLIES OF SIGNALING MOLECULES At present, the evidence that cytoskeletal reorganization and the formation of actin-dependent Caps or SMACs are essential for T cell signaling is correlative, and a mechanistic sequence explaining the role of these actin-dependent structures has not yet been defined. This correlative evidence has emerged from several sources. Perhaps the best evidence comes from the early experiments indicating that addition of cytochalasin D, a drug that interferes with actin polymerization by binding the ends of actin filaments, blocks Cap formation and Ca2⫹ mobilization and inhibits cytokine production when T lymphocytes interact with peptide-pulsed APCs (Valitutti et al., 1995). Similar results are seen with the actin monomer sequestration molecule latrunchulin, strongly indicating that actin-dependent events play an essential role in signaling. Some groups have disputed these results, but the inhibitors were used for extremely long periods of time, making the discrimination of primary and secondary effects of blocking actin polymerization difficult (Valentine and Vaughan, 1986). A second type of evidence for a functional role of the Caps comes from the finding by Janeway and colleagues that monoclonal antibodies that could induce formation of the Cap were also able to induce activation (Rojo et al., 1989; Saizawa et al., 1987). A third line of evidence that actin polymerization is essential to signaling comes from observation of cells undergoing activation by antigen presentation. T lymphocytes interacting with target cells undergo sequential changes in shape while fluxing Ca2⫹ (Donnadieu et al., 1994; Negulescu et al., 1996). In addition, TCR triggering induces actin polymerization (DeBell et al., 1992), an increase in the affinity of LFA-1 for its ligand (Springer, 1990), and a more stable association of LFA-1 with the cytoskeleton (Pardi et al., 1992). Moreover, Wu¨lfing and Davis recently showed an actin-driven transport of receptors and other cytoskeletonlinked molecules to the T cell–APC interface. This actin-driven transport requires the engagement of the costimulatory receptors CD28 and

92

BAUCH et al.

LFA-1 (Wulfing and Davis, 1998; Viola et al., 1999). A role for the organized structure of the SMAC is suggested from the observation that altered peptides that cluster CD3 and talin at the T cell–APC interface, but do not segregate into SMACs, fail to induce cytokine production and cell proliferation (Dustin et al., 1998; Monks et al., 1998). Finally, some progress has been made toward identifying the molecules and mechanism used by actin-dependent structures. Three independent groups showed that T lymphocytes from mice lacking Vav1, a guanine nucleotide exchange factor (GEF) for the Rho/Rac/Cdc42 family of GTPases or Wiskott–Aldrich syndrome protein (WASP), fail to form Caps and induce proliferation following antigen receptor stimulation (Fischer et al., 1998; Holsinger et al., 1998; Snapper et al., 1998). Furthermore, dominant negative alleles of Rac block Cap formation and also block transmission of signals into the nucleus as determined by the activation of NF–AT-dependent transcription in Jurkat cells (L. Holsinger and G. Crabtree, unpublished results). The possibility that actin polymerization was critical to the Vav1-deficient phenotype was suggested from the observation that the signaling defects in the Vav1-deficient cells were similar to those found in cells treated with cytochalasin D or latrunchulin (Holsinger et al., 1998). The requirement for the GTPases Rac/Cdc42 in the formation of lamellipodia and filopodia in fibroblasts in response to a variety of growth factors is well known (Nobes and Hall, 1995). Hence, this observation suggests that actin polymerization is critical for signaling. This speculation is further supported by the observation that T cells containing a mutation in WASP, which binds Cdc42 and clusters actin, are also deficient in proliferation and cytokine production (Molina et al., 1992, 1993; Snapper et al., 1998). These observations support a three-step mechanism of T cell signaling (see Fig. 1) in which initial tyrosine phosphorylation resulting from T cell–APC contacts leads to activation of Vav1. Second, Vav1 activation leads to Rac and/or Cdc42 and concomitant WASP activation, which in

FIG. 1. Three-step model of T cell signaling. (1) After antigen-stimulation early TCRassociated tyrosine-phosphorylated substrates recruit Vav1 to the plasma membrane. There Vav1 becomes activated by tyrosine phosphorylation and by binding to products of activated PI3-kinase. (2) Vav1 links TCR-mediated phosphotyrosine signaling to Rho GTPasecontrolled actin polymerization, which is required for TCR clustering and Cap formation. This may be accomplished by two different mechanisms or by the integration of both. (A) Activated Cdc42 and PIP2 recruit WASP to the membrane and stimulate the actin nucleation activity of the WASP-associated Arp2/3 complex. (B) Activated Rac promotes actin polymerization by an as-yet-unknown mechanism of PIP5-K activation or LIM-K activation (for details see text). (3) Cap formation and clustering of signaling molecules such as PKC⍜ lead to efficient signaling and immune response.

1. Initial TCR signaling leads to Vav activation Antigen Presenting Cell

MHC

CD28

TCR

PI3-K PI3,4-P2 PI3,4,5-P3

ZAP70 -Y- P SLP76 -Y- P

T Lymphocyte

+

+

LCK Pi

Vav 2. Rac and WASP promote actin polymerization and cap formation

PIP

P Vav

+ ?

TP 42-G Cdc WASP

PIP5K

+

2/3

Arp

?

PIP2

? +

n Tali

Rac-GTP

Rac-GDP

F-Actin

PAK

A

+

B

cofilin

P

cofilin

(inactive)

LIM-K

Pi

cofilin

3. Actin-dependent clustering of signaling molecules leads to signal amplification and immune response CD2

CD28 PKC␪

LFA-1

Tali n

F-Acti

n

PKC␪ Ca

2+

PKC␪ substrates? Unknowns?

NF-AT Transcription Complex

?

JNK? other Targets?

Immune Response Genes

94

BAUCH et al.

turn initiates actin polymerization and the clustering of surface molecules. Third, clustering of signaling molecules somehow leads to effective signaling. While the above observations suggest that Cap formation and other actin-dependent events are essential components in lymphocyte signal transduction, complete delineation of a mechanistic sequence leading from clustering to activation of the genes essential for proliferation will be essential to resolve the controversy surrounding the roles of Vav1, WASP, and actin. Furthermore, other molecules such as SLP-76, Nck, Fyn, and Pak, each with connections to the actin cytoskeleton and lymphocyte activation, are likely to play a role in these processes (van Leeuwen and Saumelson, 1999; Wardenburg et al., 1998). As has been noted, the formation of rafts may also contribute to the assembly of activation-critical molecules in the Cap. Interestingly, a link between the actin cytoskeleton and rafts has been suggested (Harder and Simons, 1999; Moran and Carrie Miceli, 1998), further corroborating the fact that actin polymerization and lipid organization may work together as regulators of proximity. III. Current View of the Regulation of Actin Polymerization in Cytoskeletal Rearrangements

If actin polymerization is essential for signaling in lymphocytes and other cells, how might actin polymerization be regulated by ligand binding or, in the case of lymphocytes, by antigen presentation? There has been much progress in understanding in vivo regulation of actin polymerization in recent years, particularly the events necessary for nucleation of actin filaments. In brief, nucleation can be controlled by several factors, including local levels of phosphatidylinositol-4, 5-bisphosphate (PIP2), Cdc42, Ca2⫹, and also by the Arp2/3 complex ( Janmey, 1994; Mullins et al., 1998; Welch et al., 1997b). Although actin filaments are also controlled in other ways, including stabilization and depolymerization, the rate-limiting step for filament assembly in vitro is nucleation—the formation of actin dimers and trimers. Each filament has two distinct types of ends, a pointed (or slowgrowing) end and a barbed (or fast-growing) end. Polymerization usually occurs at the filament’s barbed end. But in quiescent cells the availability of barbed ends is limited because of the high abundance of filamentcapping proteins. Many of these actin-capping proteins are controlled by PIP2 in that PIP2 frees them from the growing end of the actin filament, allowing continued polymerization ( Janmey, 1994). In addition, a number of these actin-binding proteins are controlled by Ca2⫹ levels, for example, gelsolin and CAP32 ( Janmey, 1994), allowing localized actin polymerization in response to increases in intracellular Ca2⫹. Because visible Cap formation follows Ca2⫹ influx during lymphocyte activation, these considerations indi-

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

95

cate that actin polymerization may in part be regulated by the early influx in Ca2⫹. An additional means of regulating actin that might be critical to the control of Cap formation comes from studies on the Arp2/3 complex. The Arp2/3 complex is able to create new barbed ends by nucleating new filaments from actin monomers. The Arp2/3 complex consists of two actinrelated proteins and five novel proteins whose function is presently unclear (Machesky et al., 1994; Welch et al., 1997a; Winter et al., 1997). In addition to its nucleating role it has also been shown to cross-link actin filaments and to organize actin filaments into branching networks (Mullins et al., 1998). Moreover, it functions as a capping protein residing at the pointed end of a filament. The nucleating activity of the Arp2/3 complex is quite low, but transient interaction with an adaptor protein can enhance the nucleating activity of the Arp2/3 complex significantly. Such an adaptor protein has been characterized from studies on the motility of the pathogenic bacterium Listeria monocytogenes in host cells. ActA, a surface protein of this pathogen, acts synergistically with the Arp2/3 complex to nucleate actin assembly (Welch et al., 1998). Since the Arp2/3 complex is present at the leading edge of lymphocytes, a similar type of adaptor protein has been proposed to exist there and to control nucleation at the leading edge of cells. Recently, the Arp2/3 complex was shown to bind to the C terminus of WASP and N-WASP (Machesky et al., 1999; Machesky and Insall, 1998; Rohatgi et al., 1999; Winter et al., 1999; Yarar et al., 1999). Moreover, the induction of actin polymerization upon interaction of WASP and the Arp2/3 complex can be greatly enhanced by Cdc42 and PIP2, suggesting a mechanism of signal-controlled actin polymerization (Rohatgi et al., 1999). These results suggest that N-WASP-like proteins might be the cellular counterparts of ActA. A worthwhile study would be to determine whether antigen receptor signaling in lymphocytes leads to membrane localization and clustering of the Arp2/3 complex. IV. The Vav Family of Guanine Nucleotide Exchange Factors

Formation of the Cap has been shown to be dependent upon the vav gene, which is also required for transmitting signals to the nucleus (Fischer et al., 1998; Holsinger et al., 1998), suggesting a mechanistic connection between the Cap, signaling, and Vav1. The proto-oncogene vav was originally identified as a transforming gene produced by deletion of the Nterminal 67 aa from the proto-oncogene product (Coppola et al., 1991; Katzav et al., 1989, 1991). Three closely related Vav proteins have been identified—Vav1 (Katzav et al., 1989), expressed in hematopoietic cells,

96

BAUCH et al.

and Vav2 and Vav3 (Henske et al., 1995; Schuebel et al., 1996; Trenkle et al., 1998), expressed more generally. A. THE STRUCTURE OF VAV1 Vav1 is a 95-kDa protein that contains a number of distinct structural motifs (see Fig. 2). A calponin-homology (CH) domain (Castresana and Saraste, 1995), found also in other cytoskeletal proteins, is present in the N terminus, which can bind to filamentous actin (F-actin). This CH domain is followed by an acidic domain and the catalytic Db1-homology (DH) domain (Cerione and Zheng, 1996), a guanine nucleotide exchange factor (GEF) domain for the Rho/Rac/Cdc42 family of GTPases. C-terminally, the DH domain is flanked by a Pleckstrin-homology (PH) domain that can bind polyphosphoinositides (Han et al., 1998), which in turn are thought to regulate the GEF activity of Vav1. The C terminus of Vav1 contains one Src homology 2 (SH2) domain flanked by two Src homology 3 (SH3) domains, which are involved in protein–protein interactions. These motifs suggest that Vav1 has several mechanisms of activation or may function at the crossroads of more than one signaling pathway.

P Y0- P P7 ZA 76 -Y P SL

Actin ?

Vav

CH

A

Rho Rac Cdc42 GDP

DH

PH

SH3

SH2

SH3

Rho Rac Cdc42 GTP

FIG. 2. Vav1 functions as an adaptor molecule linking signaling molecules and cytoskeletal proteins to actin. Vav1 contains different structural motifs such as the calponin-homology (CH), acidic (A), Dbl-homology (DH), pleckstrin-homology (PH), and one Src-homology 2 (SH2) flanked by two SH3 domains. The DH domain mediates Vav1 guanine–nucleotide exchange activity for the Rho GTPases. Vav1 activation by tyrosine kinases is stimulated by binding of PI3-kinase products (PI-3,4,5-P3 and PI-3,4-P2) to the PH domain and is further facilitated by its membrane recruitment through the interaction of its SH2 domain with tyrosine-phosphorylated SLP-76 or ZAP70.

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

97

B. PROPOSED MECHANISM OF VAV1 ACTIVATION The GEF activity of Vav1, which has been shown to be specific for the Rho family of GTPases, with Rac being the best substrate, is regulated by tyrosine phosphorylation (Crespo et al., 1997; Han et al., 1997). Coexpression of Src family kinases enhances the GEF activity of Vav1 (Crespo et al., 1997; Han et al., 1997); in particular, Fyn, Zap70, and Syk have been implicated to phosphorylate Vav1 (Fernandez et al., 1999; Michel et al., 1998; Salojin et al., 1999). Moreover, Han et al. have recently shown that the GEF activity of Vav1 is influenced by phosphoinositides, particularly by the products of the phosphoinositide 3-kinase (PI3-kinase) (Han et al., 1997). Thus, two different signaling pathways can regulate the GEF activity of Vav1. The SH2, SH3, and PH domains of Vav1 most likely determine the intracellular localization of Vav1. The SH2 domain of Vav1 has been shown to be involved in the membrane recruitment of Vav1. In T cells, a protein complex between the SH2 domain of Vav1 and the tyrosine phosphorylated adaptor protein SLP-76 is formed upon antigen-receptor stimulation (Wu et al., 1996). This interaction is important for the recruitment of Vav1 to the proximity of a protein tyrosine kinase, thereby allowing Vav1 to become phosphorylated and activated. In B cells, an analogous protein, SLP-65, has been characterized and implicated in the coupling of Vav1 to antigenreceptor molecules (Wienands et al., 1998). In addition, CD19 has been reported to be the docking site for the SH2 domain of Vav1 in B cells (O’Rourke et al., 1998). Alternatively, the interaction of the Vav1 PH domain with membrane lipids could induce the translocation of Vav1 to the plasma membrane. Upon antigen-receptor stimulation, PI3-kinase becomes activated, and thus the membrane levels of D-3 phosphoinositides are elevated and might be transiently restricted to the regions near the antigen receptor. Consequently, the PH domain of Vav1 could bind to PI3,4,5-trisphosphate (Han et al., 1998). At the plasma membrane Vav1 can then be phosphorylated by tyrosine kinases, leading to an increase of the catalytic activity of Vav1 (Han et al., 1998). In addition to the SH2 and PH domains, the SH3 domain of Vav1 can also be used to recruit Vav1 to the membrane. For instance, the myristylated NEF protein of primate lentiviruses has been shown to recruit Vav1 to the membrane by binding to its SH3 domain, thereby increasing the activity of Vav1 (Fackler et al., 1999). C. FUNCTION OF VAV1 IN SIGNALING The importance of Vav1 in lymphocytes was confirmed by the analysis of Vav1-deficient mice. Initial studies using Rag-2-deficient blastocyst com-

98

BAUCH et al.

plementation assays showed that B and T lymphocytes from Vav1-deficient mice have a severe defect in proliferation upon antigen-receptor signaling (Fischer et al., 1995; Tarakhovsky et al., 1995; Zhang et al., 1995). The proliferation defect can be bypassed by PMA and ionomycin stimulation, indicating a receptor-proximal role for Vav1. Furthermore, transient transfection studies have demonstrated that Vav1 may be important in TCRmediated signal transduction pathways that result in NF–AT-dependent transcription and IL-2 promoter activation (Holsinger et al., 1995; Wu et al., 1995). The lymphoid-specific transcription complex, NF–AT, is involved in early gene activation in T cells and is assembled from a preexisting T cell restricted cytoplasmic factor (NF–ATc) and an inducible ubiquitous nuclear component (NF–ATn) after activation through the antigen receptor (Flanagan et al., 1991). Surprisingly, studies of Vav1 knockout mice demonstrated that Vav1 is not essential for major early antigen receptorstimulated signaling pathways including tyrosine phosphorylation, MAPK activation, or SAPK/JNK activation (Fischer et al., 1998; Holsinger et al., 1998). Following activation, Vav1-deficient lymphocytes fail to form TCR Caps and fail to form radiating actin polymers extending from the Cap. Moreover, Ca2⫹ mobilization is impaired in T cells as well as in B cells (Holsinger et al., 1998; O’Rourke et al., 1998; Turner et al., 1997). Studies in B cells (O’Rourke et al., 1998) indicate that CD19 recruits Vav1 to the plasma membrane, which then activates phosphatidylinositol-4-phosphate5-kinase (PIP5-kinase) via Rac, leading to PIP2 production and elevated Ca2⫹ levels. This sequence of events provides a link between Vav1 and Ca2⫹ signaling. In summary, Vav1 is required for capping, IL-2 production, cell proliferation, and Ca2⫹ signaling—all processes that are actin-dependent. Vav1, which contains SH2, SH3, and PH domains, may therefore serve as an adaptor linking signaling molecules and other cytoskeletal proteins to actin. Future challenges will be to identify the mechanisms by which the Vav1mediated actin rearrangements and the Cap formation lead to the activation of genes involved in proliferation and in the immune response. V. The Regulation of Actin by Rac

The essential role of Vav1 in lymphocyte activation appears to be mediated by its ability to regulate the actin cytoskeleton, as illustrated in Fig. 1. Rac is the most likely intermediate in this pathway. Rac belongs to the Rho family of GTPases. GTPases can function as intracellular switches by cycling between the active GTP-bound state and the inactive GDP-bound state. Activated GTPases bind to a variety of downstream effector molecules, thereby initiating distinct effector pathways and regulating different

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

99

biological processes. Three classes of regulatory proteins control the guanine nucleotide-binding cycle of Rho family GTPases. Guanine nucleotide exchange factors (GEFs) promote the transition from the inactive GDPbound state to the active GTP-bound conformation. GTPase-activating proteins (GAPs) stimulate the inactivation of the GTPases. Guanine nucleotide dissociation inhibitors (GDIs) act to lock the GTPase in either the active or the inactive state. The best characterized Rho family GEF in lymphocytes is Vav1, which acts as an exchange factor for the Rho family of GTPases and is regulated by tyrosine phosphorylation (Crespo et al., 1997; Han et al., 1997) and phosphoinositides (Han et al., 1998). Activated Rho GTPases interact with cellular target proteins or effectors to trigger a wide variety of cellular responses, including the reorganization of the actin cytoskeleton and the control of gene transcription. A. ROLE OF THE RHO GTPASES IN CYTOSKELETAL DYNAMICS The first biological function found for Rho GTPases was the dynamic organization of the actin cytoskeleton and the assembly of associated integrin structures in fibroblasts. Cdc42, Rac, and Rho modulate actin reorganization in various ways. Rho regulates the formation of actin stress fibers and focal adhesion, and Rac controls lamellipodia formation, whereas Cdc42 induces the formation of microspikes and filopodia (Nobes and Hall, 1995; Ridley and Hall, 1992; Ridley et al., 1992). Further studies showed the involvement of Rac in the modulation of Jun N-terminal kinase or stressactivated kinase ( JNK or SAPK, respectively) activity (Coso et al., 1995; Minden et al., 1995). These findings suggest that Rac is involved in different distinct signaling pathways. The role of Rac in activating different signaling pathways is also supported by the fact that specific alleles of Rac activate changes in the cytoskeleton and proliferation but do not activate JNK, whereas other alleles of Rac activate JNK but have no effect on cell proliferation or the cytoskeleton ( Joneson et al., 1996; Lamarche et al., 1996). These results raise the vital question of how actin polymerization comes to be essential for cell proliferation. B. REGULATION OF ACTIN POLYMERIZATION Rho family GTPases appear to control actin polymerization by binding and activating PIP5-kinase which in turn leads to the production of PIP2 (Chong et al., 1994; Hartwig et al., 1995; Tolias et al., 1995). PIP2 promotes actin polymerization by binding to a wide variety of actin-binding proteins, preventing them from capping the rapidly polymerizing ends of actin filaments and leading to filament growth. These results provide a direct connection between TCR signaling and an important regulatory mechanism for actin polymerization.

100

BAUCH et al.

In addition to its function of inducing actin polymerization by increasing levels of PIP2, Rac has also been implicated in the inhibition of actin filament disassembly. Arber et al. and Yang et al. have shown that Rac activates the serine threonine kinase LIM-kinase 1 (LIMK), which in turn phosphorylates and thereby inactivates cofilin (Arber et al., 1998; Yang et al., 1998), the key protein required for actin depolymerization (Theriot, 1997) (see Fig. 1). In a 1999 study, a Rho-dependent activation of LIMK has also been shown through the serine threonine kinase ROCK (Maekawa et al., 1999). Therefore, both Rho and Rac are able to stabilize actin filaments by inactivating cofilin but possibly lead to slightly different phenotypes. However, Rac is able to regulate actin polymerization by at least two mechanisms, which may explain the inability of Vav1-deficient cells to form the Cap or SMAC structures. But why do Vav1-deficient cells have defects in the cytoskeleton but no defects in the activation of JNK? One explanation could be the functional redundancy by the other Vav family members. Alternatively, the exchange factor for the Rho family of GTPases could impart on Rac a conformation or composite surface such that its actions are directed away from the JNK pathway and to the actin regulatory pathway. In this way, Vav1 could mimic the mutant alleles of Rac that have a tyrosine-to-lysine mutation at amino acid position 40 which are incapable of activating JNK but do induce actin polymerization and proliferation (Lamarche et al., 1996). Moreover, the Vav1-regulated actions of Rac could not only lead to the actin regulatory pathway but also be responsible for the pathway leading to Ca2⫹ signaling because PIP2, a product of the above-mentioned Rac-activated PIP5-kinase activity, is also the precursor of inositol-1,4,5-trisphosphate (IP3), an essential regulator of intracellular Ca2⫹ levels. Therefore, Rac very likely is the direct downstream effector molecule of Vav1 in the actin-dependent signaling pathway leading to proliferation and immune responses. VI. WASP and the Wiskott–Aldrich Syndrome

The Wiskott–Aldrich syndrome is a rare X-linked immunodeficiency that is characterized by recurrent infections, eczema, thrombocytopenia, and lymphoreticular malignancies. Early studies investigating WAS lymphocytes and platelets suggested that the gene mutated in this disorder might regulate the actin cytoskeleton (Kenney et al., 1986). Lymphocytes isolated from WAS patients had abnormal shapes with fewer cell surface microvilli and an abnormal pattern of actin filaments (Kenney et al., 1986; Molina et al., 1992; Remold-O’Donnell et al., 1996). These defects are not secondary to alterations occurring in the bloodstream of affected individuals since T cell lines from WAS patients cultured in vitro had even greater

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

101

cytoskeletal abnormalities (Molina et al., 1992). WAS T cells were also found to have a defect in anti-CD3-induced proliferation and actin polymerization (Gallego et al., 1997; Molina et al., 1993), suggesting that the responsible gene might encode a signaling molecule linking antigen receptor signaling to the cytoskeleton. With the identification of WASP, the gene mutated in patients with this syndrome (Derry et al., 1994), numerous studies have confirmed a role for this molecule in a complicated signaling cascade in hematopoietic cells that regulates both the actin cytoskeleton and signals to the nucleus (Snapper and Rosen, 1999). WASP encodes for a 502-amino-acid cytoplasmic protein that is expressed solely in hematopoietic cells (Derry et al., 1994). Other WASP-like molecules have also been recently identified, including the more generally expressed mammalian neural-WASP (N-WASP, first identified from bovine brain) (Miki et al., 1996), Bee-1 from yeast (Li, 1997), and Scar1 from Dictyostyleum (Bear et al., 1998). More recently, a human Scar1 homolog, WAVE (for WASP family Verprolin homologous protein), has been characterized (Machesky and Insall, 1998; Miki et al., 1998b). Each of these WASP-like molecules has been demonstrated to regulate the actin cytoskeleton. WASP and N-WASP are both proline-rich molecules and have greater than 50% homology at the protein level with a similar domain structure (see Fig. 3). WASP contains several putative functional domains, including (1) a polyproline-rich region that contains numerous SH3 domain-binding sites (Derry et al., 1994) and a profilin-binding region (Suetsugu et al.,

FIG. 3. WASP interacts with actin cytoskeleton directly as well as with other signaling and cytoskeletal-associated proteins. These interactions are mediated via several functional domains, including (1) two unique domains, WASP homology 1 and 2 (WH1 and WH2), that mediate interactions with WIP and actin, respectively; (2) a pleckstrin homology domain (PH) that overlaps with WH1 and can interact with PIP2; (3) a basic region (BR) and an acidic region (AR) upstream of the GBD and at the extreme C terminus, respectively; (4) a GTPase-binding domain (GBD); (5) a proline-rich region that contains numerous SH3 domain binding sites and a profilin binding region; and (6) a cofilin homology domain (CH). See text for details.

102

BAUCH et al.

1998); (2) a GTP-binding domain (GBD), also known as Cdc42/Rac interactive binding (CRIB) motif (Burbelo et al., 1995); (3) two unique domains, WASP homology 1 and 2 (WH1 and WH2), that contain sequences found in other proline-rich proteins known to interact with the cytoskeleton (Symons et al., 1996); (4) a pleckstrin homology domain that overlaps with WH1 (Miki et al., 1996); (5) a cofilin homology domain (Miki et al., 1996); and (6) basic and acidic regions upstream of the GBD and at the extreme C terminus, respectively (Miki et al., 1996). A. THE STRUCTURE OF WASP WASP has been demonstrated in vitro and in vivo to bind a variety of SH3 domain-containing proteins. SH3-binding domain sequences found within the polyproline-rich sequence mediate these interactions. In fact, each of the in vivo interactors (Fyn, Nck, PSTPIP, and GRB-2) have known links to the actin cytoskeleton. For example, the proline serine threonine phosphatase-interacting protein (PSTPIP) colocalizes with the cortical actin cytoskeleton and can induce lamellipodia when expressed in NIH 3T3 cells (Wu et al., 1998). Mutations within the Drosophila Nck homolog (Dock) result in defects in axonal guidance (Garrity et al., 1996). Interestingly, Nck interacts with other signaling molecules in T cells that may function in a pathway similar to that of WASP including SLP-76 (a known Vav1 interactor) (Wardenburg et al., 1998) the p21-associated kinase (PAK1) (Lu et al., 1997), and the WASP-interacting protein (WIP; see discussion following) (Anton et al., 1998). The interaction between Nck and SLP-76 is required for actin polymerization in lymphocytes (Wardenburg et al., 1998). The polyproline-rich region of N-WASP (and presumably WASP) and WAVE bind profilin, a G-actin-binding protein that promotes actin assembly (Miki and Takenawa, 1998; Suetsugu et al., 1998). The profilin-binding region appears to be distinct from the SH3 domain-binding region (Suetsugu et al., 1998). Three groups have demonstrated that the GBD/CRIB motif within WASP can mediate interaction with Cdc42 (and to a lesser extent Rac), a member of the Rho family of small GTPases (Aspenstrom et al., 1996; Kolluri et al., 1996; Symons et al., 1996). As previously discussed, the Rho family of GTPases can function as molecular switches upon stimulation by GEFs (e.g., Vav) with the ability to induce both cytoskeletal changes (e.g., lamellipodia formation) and cell proliferation (Hall, 1998). A discussion of the potential coordinated role for GEFs, Rho family GTPases, and WASP in the regulation of the cytoskeleton in lymphocytes will be discussed following. Interestingly, although WAVE has been shown to be an effector

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

103

molecule for Rac, it does not contain the consensus GBD/CRIB motif (Miki et al., 1998b). The roles of the putative novel functional domains WH1 and WH2 remain unclear. WH1 domains are found in other cytoskeletal-associated proteins, including VASP, dena, and homer. WH1 is required for the WASP/WIP interaction (Ramesh et al., 1997). Identified as a WASP interactor by two-hybrid screening, WIP is a proline-rich protein with homology to verprolin that binds profilin and can induce actin polymerization (Ramesh et al., 1997; Vaduva et al., 1999). WH1 also has some homology to pleckstrin homology (PH) domains and has been demonstrated to interact with PIP2 (Miki et al., 1996). However, some investigators have argued that the homology between the WH1 domain of WASP and PH domains is too weak to designate WH1 as a true PH domain (Insall and Machesky, 1999). Therefore WH1 may have more than one function—perhaps mediating both localization of WASP to the membrane upon activation (through its homology with PH domains) and interaction with other proteins that regulate the actin cytoskeleton. WH2 domains are found in verprolin (Symons et al., 1996), Scar1/WAVE (Miki et al., 1998b), and the WASPinteracting protein (WIP) (Anton et al., 1998). Both N-WASP and Scar/ WAVE WH2 domains have been shown to bind directly to monomeric actin (Machesky and Insall, 1998; Miki et al., 1996, 1998b). Near the C terminus of WASP is a cofilin homology region (Miki et al., 1996). As has been noted, cofilin is an actin-binding protein that when activated can depolymerize actin either by releasing actin monomers from pointed ends or by severing actin filaments (Maciver, 1998). Full length N-WASP or cofilin homology domain-containing fragments of N-WASP can either decrease actin viscosity or reduce the sedimentibility of actin filaments in vitro (Miki et al., 1996, 1998a). These data are consistent with a role for the cofilin homology region in actin depolymerization. Miki et al. have postulated that actin depolymerization is prevented in the resting state by an interaction between the basic and acidic regions that mask the cofilin homology region (Miki et al., 1998a). Upon Cdc42 binding to WASP, this interaction is prevented, thereby enabling the cofilin homology domain to interact with actin. Although Scar also binds actin, no detectable depolymerization activity has been appreciated. Sequences within the C terminus of WASP and Scar1/WAVE, distinct from the actin-binding region of WH2 but overlapping with the cofilin region and acidic residues, have been shown to interact with components of the Arp2/3 complex (Machesky and Insall, 1998). As has been noted, this seven-protein complex is known to stimulate actin assembly, perhaps by promoting de novo actin nucleation (Zigmond, 1998). In fact, N-WASP and Scar have each been shown in vitro to stimulate actin assembly in an

104

BAUCH et al.

Arp2/3-dependent fashion (Machesky et al., 1999; Rohatgi et al., 1999). At least for N-WASP, maximal stimulation requires Cdc42 and phosphoinositides (Rohatgi et al., 1999). VII. WASP and the Control of Actin

Considerable effort has been devoted to understanding the relationship between WASP and the Rho family GTPase Cdc42 in regulating the actin cytoskeleton. However, because most of the experimental systems employed have relied upon protein overexpression, experiments investigating the relationship between these molecules have been somewhat difficult to interpret. When expressed alone, WASP leads to ectopic actin clusters with a largely perinuclear distribution, whereas N-WASP overexpression leads to actin accumulation in cortical areas (Miki et al., 1996; Symons et al., 1996). Although constitutively active Cdc42 can induce filopodia and can bind either WASP or N-WASP in vivo, the role of these interactions in the generation of filopodia is unclear. WASP overexpression in COS cells blocks the ability of constitutively active Cdc42 to induce cytoskeletal changes (Miki et al., 1998a; Symons et al., 1996). However, in the same assay conditions, N-WASP expression does not block constitutively active Cdc42-induced filopodia but rather appears to be essential for the induction of filopodia by Cdc42 (Miki et al., 1998a). More direct support for an essential role for N-WASP in Cdc42-mediated cytoskeletal effects was shown in a cell culture system demonstrating that the induction of filopodia by bradykinin is blocked by the introduction of anti-N-WASP antibodies (a process known to be Cdc42-dependent). These apparent differences in the roles of WASP and N-WASP in regulating filopodia formation may result either from the specific cell types or experimental conditions in which these studies were performed or from specific differences in the function of these proteins. Supporting the former possibility is a study using an inducible membrane-clustering strategy demonstrating that membrane localization and clustering of either WASP or constitutively active Cdc42 results in the formation of membrane protrusions resembling filopodia (Castellano et al., 1999). In this system, membrane localization of WASP or Cdc42, without clustering, was not sufficient to activate the formation of filopodia (Castellano et al., 1999). The generation of specific cytoskeletal structures by WASP/N-WASP in specific cell types may therefore require both subcellular localization and clustering of protein in order to activate downstream signaling events. Interestingly, WAVE, which has homology to WASP and N-WASP in the WH2/Verprolin homology domain, appears to act specifically as a Rac effector (Miki et al., 1998b). A dominant negative form of WAVE lacking

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

105

the WH2/Verprolin homology domain blocks the ability of the Rac1G12V dominant active mutant to form lamellipodia. However, this dominant negative WAVE mutant does not affect the ability of the Cdc42G12V dominant active mutant to form filopodia. This further supports different roles for WASP-family members in controlling cytoskeletal structures. Studies investigating the motility of intracellular parasites have suggested more general functions for WASP-family members in regulating the actin cytoskeleton. N-WASP has been implicated in directly regulating the actinbased motility of certain intracellular parasites (Suzuki et al., 1998). The propulsion of Shigella and Listeria within host cells requires the formation of filopodia-like structures (F-actin tail). Associated with these structures are both bacterial surface proteins (e.g., VirG) and host-encoded proteins (e.g., VASP, vinculin, 움-actinin). N-WASP interacts with the asymmetrically localized Shigella outermembrane protein VirG and colocalizes with the F-actin tail or comet that is felt to be the driving force behind the spread of Shigella within an epithelial cell (Suzuki et al., 1998). Initial support for a direct role of N-WASP in actin-based motility of Shigella comes from experiments demonstrating that overexpression of dominant negative N-WASP (lacking the cofilin homology domain) abolished the formation of the actin tail in Shigella-infected cells (Suzuki et al., 1998). More direct evidence for a role for N-WASP in these processes are experiments that have shown that the depletion of N-WASP from Xenopus extracts can prevent the formation of actin tails in conditions whereby Shigella surface proteins normally drive actin assembly (Suzuki et al., 1998). A. MECHANISM OF ACTIN CYTOSKELETON REGULATION The mechanism by which WASP-family members regulate the actin cytoskeleton in lymphocytes has not been established. WASP may play an essential role in the reorganization of the cytoskeleton that is required for receptor patterning at APC/T cell contact surfaces and the formation of SMACs (Dustin et al., 1998; Monks et al., 1998). The upstream signals that regulate WASP in lymphocytes may be triggered by antigen receptor stimulation (⫹/⫺ co-receptor stimulation). In this regard, T cells from WAS patients and WASP knockout mice have defects in both antigen receptor-induced proliferation and regulation of the actin cytoskeleton (Gallego et al., 1997; Molina et al., 1993; Snapper et al., 1998). Several possible factors may regulate WASP activity: (1) change in subcellular localization (e.g., movement to the membrane); (2) tyrosine phosphorylation resulting in interactions with other molecules (e.g., SH2 domaincontaining proteins); (3) structural changes upon protein–protein interactions (e.g., interaction with activated Rho family GTPases).

106

BAUCH et al.

WASP may be recruited to the membrane upon TCR activation through an interaction between the WASP PH domain and PIP2 (Miki et al., 1996). Alternatively, changes in the subcellular localization of WASP may occur via an interaction between SH3 domain-containing molecules (e.g., Nck, Fyn) and the proline-rich region of WASP or via an interaction with SH2 domain-containing molecules and specific phosphorylated SH2-binding motifs. As has been noted, numerous in vivo interactions between WASP and SH3-containing molecules have been demonstrated. Although published data in lymphocytes is lacking, in human platelets and mast cells WASP has been shown to be phosphorylated on tyrosine residues upon activation (Guinamard et al., 1998; Oda et al., 1998). WASP phosphorylation in lymphocytes upon antigen-receptor activation (or other surface receptor activation, e.g., CD28) may lead to the association of WASP with SH2 domain-containing molecules. WASP may also interact with the membrane through interactions with recruited Rho family GTPases. Lymphocyte activation may result in the release of certain WASP associators leading to the activation of downstream signaling cascades. For example, the association of WASP- and GTP-loaded Cdc42 may result in both the release of WIP and the ‘‘unmasking’’ of the C-terminal region stimulating direct effects on actin modeling (e.g., actin depolymerization via the cofilin homology, Arp2/3-dependent actin polymerization). In this regard, maximal Arp2/3-dependent actin polymerization by N-WASP requires Cdc42 and phosphoinositides (PIP2) (Rohatgi et al., 1999). Following the ‘‘activation’’ of WASP—i.e., change in localization, association with Rho family GTPases, association with SH2 or SH3 domaincontaining molecules, release of specific interacting molecules, or the resulting change in shape of WASP resulting from such interactions—a number of downstream signaling events are likely induced that result in a specific set of changes in the actin cytoskeleton (e.g., filopodia, lamellipodia, Cap formation, microtubule polarization). Some of these structures may facilitate cell movement, others may facilitate antigen receptor stimulation, and still others may facilitate directed cytokine secretion. There are numerous ways in which activated WASP may lead to cytoskeletal alterations: (1) depolymerization of actin filaments via the cofilin homology region (Miki et al., 1996, 1998a); (2) interaction with profilin-promoting actin assembly (Suetsugu et al., 1998); (3) interaction with WIP, a WASP interactor that also binds profilin, resulting in changes in actin polymerization (Ramesh et al., 1997); (4) direct actin binding (Machesky and Insall, 1998; Miki et al., 1996); and (5) interaction with the Arp2/3 complex promoting actin assembly (Machesky et al., 1999; Machesky and Insall, 1998; Rohatgi et al., 1999). It is likely that some or all of these processes will be used together for specific structural changes. In addition, other

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

107

molecules in lymphocytes with links to WASP may have effects on the actin cytoskeleton upon WASP activation, including SLP-76, Vav, Nck, PSTPIP, Fyn, and Pak (Fischer et al., 1998; Holsinger et al., 1998; RiveroLezcano et al., 1995; Wardenburg et al., 1998; Wu et al., 1998). VIII. Shared Aspects in the Phenotype of WASP- and Vav-Deficient Mice

Several lines of genetic evidence point to a common signaling pathway in lymphocytes that include both Vav1 and WASP in the regulation the actin cytoskeleton. The most striking similarity between Vav1- and WASPdeficient mice is the profound T cell proliferative and capping defects upon antigen receptor stimulation (Fischer et al., 1995, 1998; Holsinger et al., 1998; Snapper et al., 1998; Tarakhovsky et al., 1995; Zhang et al., 1995). In addition, T cells from both Vav1- and WASP-deficient mice secrete reduced amounts of IL-2 upon antigen receptor-induced stimulation (Snapper et al., 1998; Zhang et al., 1995). Furthermore, the signaling defect in WASP- and Vav1-deficient lymphocytes appears to involve membrane-proximal events just as the combination of phorbol esters and calcium ionophores leads to normal proliferation. Despite these similarities, there are some notable differences between Vav1- and WASP-deficient mice. Vav1 deficiency results in more severe phenotypic abnormalities in lymphocyte development and function. Whereas T cell development proceeds normally in WASP-deficient mice (Snapper et al., 1998), Vav1 deficiency results in a block in T cell expansion—with a dramatic decrease in the number of immature CD4⫹/CD8⫹ double positive and CD4⫹ and CD8⫹ single positive thymocytes (Fischer et al., 1995; Tarakhovsky et al., 1995; Zhang et al., 1995). Furthermore, both positive and negative selection of thymocytes are Vav1-dependent processes (Fischer et al., 1995; Turner et al., 1997). Although most evidence suggests that WASP and Vav1 deficiency do not significantly affect peripheral B cell development in mice (except for the lack of B1 cells in the latter mutant mice), Vav1-deficient, but not WASP-deficient, B cells have defects in proliferation upon antigen receptor stimulation (Snapper et al., 1998; Zhang et al., 1995). While the effect of Vav1 deficiency on capping and regulation of the actin cytoskeleton in B cells has not been described, normal capping parallels the normal responses to antigen receptor-induced proliferation in WASP-deficient B cells (Snapper et al., 1998). Unique features of WASP-deficient mice include mild thrombocytopenia and chronic colitis (Snapper et al., 1998). Vav1 deficiency may result in a more severe phenotype than WASP deficiency as a result of its more proximal location in the signaling cascade. Alternatively, a milder phenotype in WASP-deficient mice may result from functional redundancy by related

108

BAUCH et al.

proteins (e.g., N-WASP, Scar) that compensate for WASP deficiency in specific lymphocyte populations (i.e., immature thymocytes and B cells). As has been described, in order to characterize more fully the role of Vav1 in lymphocyte signaling, several studies have analyzed the effect of Vav1 deficiency on various known signaling pathways in lymphocytes. The activation of MAPK, JNK, and p38 and the translocation of NFATc apparently proceed normally upon antigen receptor stimulation in Vav1-deficient lymphocytes (Fischer et al., 1998; Holsinger et al., 1998). Although receptor-proximal events such as antigen receptor-induced CD3␧, Zap70, and SLP-76 phosphorylation occur normally in Vav1-deficient cells, calcium mobilization is markedly affected (Costello et al., 1999). SLP-76 and Vav1 have recently been shown to form a complex with Nck, and together appear to regulate the actin cytoskeleton in lymphocytes (Wardenburg et al., 1998). It seems likely that both Vav1 and WASP are part of a related signaling cascade that regulates certain actin-dependent structures (e.g., contact Cap formation) whose generation is essential for downstream signals that drive transcription and cell cycle progression. A connection between Vav1 and WASP may be expected since Vav family members are known to be GEFs for Rho family GTPases, which in turn are known to interact with several WASP family members. Yet, as has been noted, most evidence seems to suggest that Vav1 has greatest GEF activity for Rac (Crespo et al., 1996, 1997; Olson et al., 1996), a GTPase with only limited binding affinity for WASP (Aspenstrom et al., 1996; Kolluri et al., 1996; Symons et al., 1996). However, Vav1 and WASP may still be part of a common signaling cascade in lymphocytes in vivo either if Vav1 has some GEF activity for Cdc42 (Olson et al., 1996) or if Rac binds WASP (Aspenstrom et al., 1996; Kolluri et al., 1996; Symons et al., 1996). Furthermore, current functional assays used to monitor associations and activities of these molecules (e.g., cell cycle progression, cytoskeletal changes) might not reveal all modes of interaction in lymphocytes (e.g., signaling cascade regulating the contact Cap). IX. Actin Polymerization and the Propagation of Signals to the Nucleus

Tracking the biochemical sequence of events backward from the genes critical for a response to the receptors that initiated these responses has been effective in the elucidation of components of several different signaling pathways. This approach revealed the sequential events underlying signaling to the nucleus through the MAPK (Norman et al., 1988), the JNK (Hibi et al., 1993), the JAK–STAT (Pine et al., 1990), and the Ca2⫹ – Calcineurin pathways (Clipstone and Crabtree, 1992; Flanagan et al., 1991; Shaw et al., 1988). The observations described indicate that the actin cytoskeleton is involved in propagating signals to the nucleus. This require-

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

109

ment likely reflects the need for actin polymerization rather than simply the need for an intact cytoskeleton because the molecules required for signaling (Vav1, Rac, and WASP) appear to be critical for regulating polymerization. One critical step to understanding the role of actin polymerization in signaling is the identification of targets in the nucleus. Weiss and colleagues showed that overexpression of Vav1 in Jurkat T cells leads to the activation of NF–AT, which is involved in IL-2 expression. Furthermore, Vav1 synergized with TCR stimulation in inducing NF–AT- and IL-2-dependent transcription (Wu et al., 1995). Since the NF–AT complex consists of a cytoplasmic component that is dependent on the Ca2⫹ – Calcineurin pathway (NF–ATc) and an inducible nuclear component that depends on the ras–PKC-dependent signaling pathway (NF–ATn), the question of which subunit of NF–AT is affected by the actin cytoskeleton arises. At this point, the data obtained from Vav1-deficient cells do not resolve this question. NF–ATc translocation to the nucleus was reported to be normal in Vav1-deficient lymphocytes, but defects in Ca2⫹ mobilization were noted (Holsinger et al., 1998). The apparent controversy of NF–ATc still being able to translocate to the nucleus despite a Ca2⫹ mobilization defect in these cells remains to be solved. Either the Ca2⫹ concentration reached in Vav1-deficient cells is sufficient to localize most of the NF–ATc to the nucleus or there is indeed a translocation defect that could be reflected by a delay in translocation or by a decreased amount of NF–ATc being translocated. However, the defect in Ca2⫹ mobilization observed in Vav1-deficient cells is further supported by the observation that ionomycin can complement most of the defect in proliferation in these cells (A. Bauch and G. R. Crabtree, unpublished data; Costello et al., 1999). These data would suggest that the Ca2⫹-dependent NF–ATc is a downstream target of Vav1. Alternatively, because constitutively active Vav1 and Rac have been shown to partially complement constitutively active calcineurin to activate NF–AT-dependent transcription, Vav1 may act on the nuclear partner of NF–ATc (Holsinger et al., 1998). In summary, there is evidence that both translocation of the cytoplasmic component and activation of the nuclear component of NF–AT are defective in Vav1deficient mice, suggesting that a common signaling intermediate high in the pathway is at fault in the Vav1-deficient mice. Similar conclusions were reached by Tybulewicz and colleagues (Costello et al., 1999) that support the three-step model shown in Fig, 1. Clearly, it would be of interest also to define the role of WASP in transcription. A. ACTIN AND THE REGULATION OF CHROMATIN REMODELING BY THE BAF COMPLEX A surprising role of actin is the observation that nuclear actin is an integral component of the BAF chromatin-remodeling complex (Zaho et

110

BAUCH et al.

al., 1998). This complex is similar to the yeast SWI/SNF complex and is made up of 11 subunits encoded by multigene families that are assembled combinatorially (Wang et al., 1996). The BAF complex in lymphocytes undergoes rapid association with chromatin after TCR signaling and hence could be involved in the genome-wide chromatin decondensation that occurs after lymphocyte activation. Furthermore, the association of the BAF complex with the nucleus can be induced in vitro by addition of PIP2, suggesting that PIP2 levels normally control the activity of this chromatin-remodeling complex. Although one function of actin in the complex is apparently to enhance rates of ATP hydrolysis by the Brg-1 DNA-dependent ATPase, it is also likely that actin in the complex plays a critical role in linking the complex to the matrix and chromatin. These results raise the possibility that TCR-induced activation of Vav1 and Rac might control nuclear PIP2 levels and hence also regulate the chromatinremodeling activity of BAF in the nucleus, leading to decondensation of chromatin associated with activation. B. THE ROLE OF THE CAP AND OTHER ACTIN-DEPENDENT STRUCTURES AS REGULATORS OF PROXIMITY OR EFFECTIVE MOLARITY The studies reviewed here suggest that actin-dependent assemblies at the cell membrane and possibly in the nucleus are critical for mediating signals from the antigen receptor. Relatively little is known of how these assemblies might be required for activation. However, one central concept to emerge from studies of signaling is the role of proximity or, more precisely, effective molarity in controlling the activities of signaling molecules. Studies with synthetic ligands or chemical inducers of dimerization or proximity (CIDs) indicate that virtually any intracellular-signaling molecule from receptors to exchange factors, to kinases, to import proteins, or to transcription factors can be activated by bringing the relevant molecules into effective contact (Holsinger et al., 1995; Klemm et al., 1997; Spencer et al., 1995, 1996). Here the term ‘‘effective molarity,’’ borrowed from chemistry, is useful since it encompasses aspects of proximity, orientation, and kinetics. Thus, these actin-dependent structures might assemble molecules in the correct configuration for activation or allow them to maintain a configuration that is critical to their function. In either case, the role of these actin-dependent structures would be to control effective molarity between signaling molecules. A formal test of this will require the controlled assembly of the relevant Cap-associated molecules. REFERENCES Anton, I. M., Lu, W., Mayer, B. J., Ramesh, N., and Geha, R. S. (1998). J. Biol. Chem. 273, 20992–20995.

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

111

Arber, S., Barbayannis, F. A., Hanser, H., Schneider, C., Stanyon, C. A., Bernard, O., and Caroni, P. (1998). Nature 393, 805–809. Aspenstrom, P., Lindberg, U., and Hall, A. (1996). Curr. Biol. 6, 70–75. Ayscough, K. R., and Drubin, D. G. (1998). Curr. Biol. 8, 927–930. Bear, J. E., Rawls, J. F., and Saxe, C. L. (1998). J. Cell. Biol. 142, 1325–1335. Burbelo, P. D., Drechsel, D., and Hall, A. (1995). J. Biol. Chem. 270, 29071–29074. Castellano, F., Montcourrier, P., Guillemot, J. C., Gouin, E., Machesky, L., Cossart, P., and Chavrier, P. (1999). Curr. Biol. 9, 351–360. Castresana, J., and Saraste, M. (1995). FEBS Lett. 374, 149–151. Cerione, R. A., and Zheng, Y. (1996). Curr. Opin. Cell. Biol. 8, 216–222. Chong, L. D., Traynor-Kaplan, A., Bokoch, G. M., and Schwartz, M. A. (1994). Cell 79, 507–513. Clipstone, N. A., and Crabtree, G. R. (1992). Nature 357, 695–697. Coppola, J., Bryant, S., Koda, T., Conway, D., and Barbacid, M. (1991). Cell Growth Differ. 2, 95–105. Coso, O. A., Chiariello, M., Yu, J. C., Teramoto, H., Crespo, P., Xu, N., Miki, T., and Gutkind, J. S. (1995). Cell 81, 1137–1146. Costello, P. S., Walters, A. E., Mee, P. J., Turner, M., Reynolds, L. F., Prisco, A., Sarner, N., Zamoyska, R., and Tybulewicz, V. L. J. (1999). Proc. Natl. Acad. Sci. USA 96, 3035–3040. Crespo, P., Bustelo, X. R., Aaronson, D. S., Coso, O. A., Lopez-Barahona, M., Barbacid, M., and Gutkind, J. S. (1996). Oncogene 13, 455–460. Crespo, P., Schuebel, K. E., Ostrom, A. A., Gutkind, J. S., and Bustelo, X. R. (1997). Nature 385, 169–172. de Petris, S. (1974). Nature 250, 54–56. de Petris, S. (1975). J. Cell Biol. 65, 123–146. de Petris, S., and Raff, M. (1973). Nat. New Biol. 241, 257–259. DeBell, K. E., Conti, A., Alava, M. A., Hoffman, T., and Bonvini, E. (1992). J. Immunol. 149, 2271–2280. Derry, J. M., Ochs, H. D., and Francke, U. (1994). Cell 78, 635–644. Donnadieu, E., Bismuth, G., and Trautmann, A. (1994). Curr. Biol. 4, 584–595. Dustin, M. L., Olszowy, M. W., Holdorf, A. D., Li, J., Bromley, S., Desai, N., Widder, P., Rosenberger, F., van der Merwe, P. A., Allen, P. M., and Shaw, A. S. (1998). Cell 94, 667–677. Fackler, O. T., Luo, W., Geyer, M., Alberts, A. S., and Peterlin, B. M. (1999). Mol. Cell 3, 729–739. Fernandez, J. A., Keshvara, L. M., Peters, J. D., Furlong, M. T., Harrison, M. L., and Geahlen, R. L. (1999). J. Biol. Chem. 274, 1401–1406. Fischer, K. D., Kong, Y. Y., Nishina, H., Tedford, K., Marengere, L. E., Kozieradzki, I., Sasaki, T., Starr, M., Chan, G., Gardener, S., Nghiem, M. P., Bouchard, D., Barbacid, M., Bernstein, A., and Penninger, J. M. (1998). Curr. Biol. 8, 554–562. Fischer, K. D., Zmuldzinas, A., Gardner, S., Barbacid, M., Bernstein, A., and Guidos, C. (1995). Nature 374, 474–477. Flanagan, W. M., Corthesy, B., Bram, R. J., and Crabtree, G. R. (1991). Nature 352, 803–807. Gallego, M., Santamaria, M., Pena, J., and Molina, I. (1997). Blood 90, 3089–3097. Garrity, P. A., Rao, Y., Salecker, I., McGlade, J., Pawson, T., and Zipursky, S. L. (1996). Cell 85, 639–650. Guinamard, R., Aspenstrom, P., Fougereau, M., Chavrier, P., and Guillemot, J. C. (1998). FEBS Lett. 434, 431–436. Hall, A. (1998). Science 279, 509–514.

112

BAUCH et al.

Han, J., Das, B., Wei, W., Van Aelst, L., Mosteller, R. D., Khosravi-Far, R., Westwick, J. K., Der, C. J., and Broek, D. (1997). Mol. Cell. Biol. 17, 1346–1353. Han, J., Luby-Phelps, K., Das, B., Shu, X., Xia, Y., Mosteller, R. D., Krishna, U. M., Falck, J. R., White, M. A., and Broek, D. (1998). Science 279, 558–560. Harder, T., and Simons, K. (1999). Eur. J. Immunol 29, 556–562. Hartwig, J. H., Bokoch, G. M., Carpenter, C. L., Janmey, P. A., Taylor, L. A., Toker, A., and Stossel, T. P. (1995). Cell 82, 643–653. Henske, E. P., Short, M. P., Jozwiak, S., Bovey, C. M., Ramlakhan, S., Haines, J. L., and Kwiatkowski, D. J. (1995). Ann. Hum. Genet. 59, 25–37. Hibi, M., Lin, A., Smeal, T., Minden, A., and Karin, M. (1993). Genes Dev. 7, 2135–2148. Holsinger, L. J., Graef, I. A., Swat, W., Chi, T., Bautista, D. M., Davidson, L., Lewis, R. S., Alt, F. W., and Crabtree, G. R. (1998). Curr. Biol. 8, 563–572. Holsinger, L. J., Spencer, D. M., Austin, D. J., Schreiber, S. L., and Crabtree, G. R. (1995). Proc. Natl. Acad. Sci. USA 92, 9810–9814. Insall, R., and Machesky, L. (1999). Trends Cell Biol. 9, 211. Janmey, P. A. (1994). Annu. Rev. Physiol. 56, 169–191. Joneson, T., McDonough, M., Bar-Sagi, D., and Van Aelst, L. (1996). Science 274, 1374– 1376. Katzav, S., Cleveland, J. L., Heslop, H. E., and Pulido, D. (1991). Mol. Cell Biol. 11, 1912– 1920. Katzav, S., Martin-Zanca, D., and Barbacid, M. (1989). EMBO J. 8, 2283–2290. Kenney, D., Cairns, L., Remold-O’Donnell, E., Peterson, J., Rosen, F. S., and Parkman, R. (1986). Blood 68, 1329–1332. Klemm, J. D., Beals, C. R., and Crabtree, G. R. (1997). Curr. Biol. 7, 638–644. Knoblich, J. A., Jan, L. Y., and Jan, Y. N. (1997). Proc. Natl. Acad. Sci. USA 94, 13005–13010. Kolluri, R., Tolias, K. F., Carpenter, C. L., Rosen, F. S., and Kirchhausen, T. (1996). Proc. Natl. Acad. Sci. USA 93, 5615–5618. Lamarche, N., Tapon, N., Stowers, L., Burbelo, P. D., Aspenstrom, P., Bridges, T., Chant, J., and Hall, A. (1996). Cell 87, 519–529. Li, R. (1997). J. Cell. Biol. 136, 649–658. Lu, W., Katz, S., Gupta, R., and Mayer, B. J. (1997). Curr. Biol. 7, 85–94. Machesky, L. M., Atkinson, S. J., Ampe, C., Vandekerckhove, J., and Pollard, T. D. (1994). J. Cell Biol. 127, 107–115. Machesky, L. M., Dyche Mullins, R., Higgs, H. N., Kaiser, D. A., Blanchoin, L., May, R. C., Hall, M. E., and Pollard, T. D. (1999). Proc. Natl. Acad. Sci. USA 96, 3739–3744. Machesky, L. M., and Insall, R. H. (1998). Curr. Biol. 8, 1347–1356. Maciver, S. K. (1998). Curr. Opin. Cell Biol. 10, 140–144. Maekawa, M., Ishizaki, T., Boku, S., Watanabe, N., Fujita, A., Iwamatsu, A., Obinata, T., Ohashi, K., Mizuno, K., and Narumiya, S. (1999). Science 285. Michel, F., Grimaud, L., Tuosto, L., and Acuto, O. (1998). J. Biol. Chem. 273, 31932–31938. Miki, H., Miura, K., and Takenawa, T. (1996). EMBO J. 15, 5326–5335. Miki, H., Sasaki, T., Takai, Y., and Takenawa, T. (1998a). Nature 391, 93–96. Miki, H., Suetsugu, S., and Takenawa, T. (1998b). EMBO J. 17, 6932–6941. Miki, H., and Takenawa, T. (1998). Biochem. Biophys. Res. Commun, 243, 73–78. Minden, A., Lin, A., Claret, F. X., Abo, A., and Karin, M. (1995). Cell 81, 1147–1157. Molina, I. J., Kenney, D. M., Rosen, F. S., and Remold-O’Donnell, E. (1992). J. Exp. Med. 176, 867–874. Molina, I. J., Sancho, J., Terhorst, C., Rosen, F. S., and Remold-O’Donnell, E. (1993). J. Immunol. 151, 4383–4390.

THE CYTOSKELETON IN LYMPHOCYTE SIGNALING

113

Monks, C. R., Freiberg, B. A., Kupfer, H., Sciaky, N., and Kupfer, A. (1998). Nature 395, 82–86. Montixi, C., Langlet, C., Bernard, A. M., Thimonier, J., Dubois, C., Wurbel, M. A., Chauvin, J. P., Pierres, M., and He, H. T. (1998). EMBO J. 17, 5334–5338. Moran, M., and Carrie Miceli, M. (1998). Immunity 9, 787–796. Mullins, R. D., Heuser, J. A., and Pollard, T. D. (1998). Proc. Natl. Acad. Sci. USA 95, 6181–6186. Negulescu, P. A., Krasieva, T. B., Khan, A., Kerschbaum, H. H., and Cahalan, M. D. (1996). Immunity 4, 421–430. Nobes, C. D., and Hall, A. (1995). Biochem. Soc. Trans. 23, 456–459. Norman, C., Runswick, M., Pollock, R., and Treisman, R. (1988). Cell 55, 989–1003. Oda, A., Ochs, H. D., Druker, B. J., Ozaki, K., Watanabe, C., Handa, M., Miyakawa, Y., and Ikeda, Y. (1998). Blood 92, 1852–1858. Olson, M. F., Pasteris, N. G., Gorski, J. L., and Hall, A. (1996). Curr. Biol. 6, 1628–1633. O’Rourke, L. M., Tooze, R., Turner, M., Sandoval, D. M., Carter, R. H., Tybulewicz, V. L., and Fearon, D. T. (1998). Immunity 8, 635–645. Pardi, R., Inverardi, L., Rugarli, C., and Bender, J. R. (1992). J. Cell Biol. 116, 1211–1220. Penninger, J. M., and Crabtree, G. R. (1999). Cell 96, 9–12. Pine, R., Decker, T., Kessler, D. S., Levy, D. E., and Darnell, J. E., Jr. (1990). Mol. Cell Biol. 10, 2448–2457. Ramesh, N., Anton, I. M., Hartwig, J. H., and Geha, R. S. (1997). Proc. Natl. Acad. Sci. USA 94, 14671–14676. Remold-O’Donnell, E., Rosen, F. S., and Kenney, D. M. (1996). Blood 87, 2621–2631. Ridley, A. J., and Hall, A. (1992). Cell 70, 389–399. Ridley, A. J., Paterson, H. F., Johnston, C. L., Diekmann, D., and Hall, A. (1992). Cell 70, 401–410. Rivero-Lezcano, O. M., Marcilla, A., Sameshima, J. H., and Robbins, K. C. (1995). Mol. Cell. Biol. 15, 5725–5731. Rohatgi, R., Ma, L., Miki, H., Lopez, M., Kirchhausen, T., Takenawa, T., and Kirschner, M. W. (1999). Cell 97, 221–231. Rojo, J. M., Saizawa, K., and Janeway, C. A., Jr. (1989). Proc. Natl. Acad. Sci. USA 86, 3311– 3315. Saizawa, K., Rojo, J., and Janeway, C. A., Jr. (1987). Nature 328, 260–263. Salojin, K. V., Zhang, J., and Delovitch, T. L. (1999). J. Immunol. 163, 844–853. Schuebel, K. E., Bustelo, X. R., Nielsen, D. A., Song, B. J., Barbacid, M., Goldman, D., and Lee, I. J. (1996). Oncogene 13, 363–371. Shaw, J. P., Utz, P. J., Durand, D. B., Toole, J. J., Emmel, E. A., and Crabtree, G. R. (1988). Science 241, 202–205. Snapper, S. B., and Rosen, F. S. (1999). Annu. Rev. Immunol. 17, 905–929. Snapper, S. B., Rosen, F. S., Mizoguchi, E., Cohen, P., Khan, W., Liu, C. H., Hagemann, T. L., Kwan, S. P., Ferrini, R., Davidson, L., Bhan, A. K., and Alt, F. W. (1998). Immunity 9, 81–91. Spencer, D. M., Belshaw, P. J., Chen, L., Ho, S. N., Randazzo, F., Crabtree, G. R., and Schreiber, S. L. (1996). Curr. Biol. 6, 839–847. Spencer, D. M., Graef, I., Austin, D. J., Schreiber, S. L., and Crabtree, G. R. (1995). Proc. Natl. Acad. Sci. USA 92, 9805–9809. Springer, T. A. (1990). Nature 346, 425–434. Suetsugu, S., Miki, H., and Takenawa, T. (1998). EMBO J. 17, 6516–6526. Suzuki, T., Miki, H., Takenawa, T., and Sasakawa, C. (1998). EMBO J. 17, 2767–2776.

114

BAUCH et al.

Symons, M., Derry, J. M., Karlak, B., Jiang, S., Lemahieu, V., McCormick, F., Francke, U., and Abo, A. (1996). Cell 84, 723–734. Tarakhovsky, A., Turner, M., Schaal, S., Mee, P. J., Duddy, L. P., Rajewsky, K., and Tybulewicz, V. L. (1995). Nature 374, 467–470. Theriot, J. A. (1997). J. Cell Biol. 136, 1165–1168. Tolias, K. F., Cantley, L. C., and Carpenter, C. L. (1995). J. Biol. Chem. 270, 17656–17659. Trenkle, T., Welsh, J., Jung B., Mathieu-Daude, F., and McClelland, M. (1998). Nucleic Acids Res. 26, 3883–3891. Turner, M., Mee, P. J., Walters, A. E., Quinn, M. E., Mellor, A. L., Zamoyska, R., and Tybulewicz, V. L. (1997). Immunity 7, 451–460. Vaduva, G., Martinez-Quiles, N., Anton, I. M., Martin, N. C., Geha, R. S., Hopper, A. K., and Ramesh, N. (1999). J. Biol. Chem. 11, 17103–17108. Valentine, M. A., and Vaughan, J. A. (1986). Cell. Immunol, 101, 651–658. Valitutti, S., Dessing, M., Aktories, K., Gallati, H., and Lanzavecchia, A. (1995). J. Exp. Med. 181, 577–584. van Leeuwen, J. E., and Saumelson, L. E. (1999). Curr. Opin. Immunol. 3, 242–248. Viola, A., Schroeder, S., Sakakibara, Y., and Lanzavecchia, A. (1999). Science 283, 680–682. Wang, W., Xue, Y., Zhou, S., Kuo, A., Cairns, B. R., and Crabtree, G. R. (1996). Genes Dev. 10, 2117–2130. Wardenburg, J. B., Pappu, R., Bu, J.-Y., Bayer, B., Chernoff, J., Straus, D., and Chan, A. C. (1998). Immunity 9, 607–616. Welch, M. D., DePace, A. H., Verma, S., Iwamatsu, A., and Mitchison, T. J. (1997a). J. Cell. Biol. 138, 375–384. Welch, M. D., Iwamatsu, A., and Mitchison, T. J. (1997b). Nature 385, 265–269. Welch, M. D., Rosenblatt, J., Skoble, J., Portnoy, D. A., and Mitchison, T. J. (1998). Science 281, 105–108. Wienands, J., Schweikert, J., Wollscheid, B., Jumaa, H., Nielsen, P. J., and Reth, M. (1998). J. Exp. Med. 188, 791–795. Winter, D., Lechler, T., and Li, R. (1999). Curr. Biol. 9, 501–504. Winter, D., Podtelejnikov, A. V., Mann, M., and Li, R. (1997). Curr. Biol. 7, 519–529. Wu, J., Katzav, S., and Weiss, A. (1995). Mol. Cell. Biol. 15, 4337–4346. Wu, J., Motto, D. G., Koretzky, G. A., and Weiss, A. (1996). Immunity 4, 593–602. Wu, Y., Spencer, S. D., and Lasky, L. A. (1998). J. Biol. Chem. 273, 5765–5770. Wulfing, C., and Davis, M. M. (1998). Science 282, 2266–2269. Xavier, R., Brennan, T., Li, Q., McCormack, C., and Seed, B. (1998). Immunity 8, 723–732. Yang, N., Higuchi, O., Ohashi, K., Nagata, K., Wada, A., Kangawa, K., Nishida, E., and Mizuno, K. (1998). Nature 393, 809–812. Yarar, D., To, W., Abo, A., and Welch, M. D. (1999). Curr. Biol. 9, 555–558. Zhang, R., Alt, F. W., Davidson, L., Orkin, S. H., and Swat, W. (1995). Nature 374, 470–473. Zhao, K., Wang, W., Rando, O. J., Xue, Y., Swiderek, K., Kuo, A., and Crabtree, G. R. (1998). Cell 95, 625–636. Zigmond, S. H. (1998). Curr. Biol. 8, R654–R657.

ADVANCES IN IMMUNOLOGY, VOL. 75

TGF-␤ Signaling by Smad Proteins KOHEI MIYAZONO,* PETER TEN DIJKE,† AND CARL-HENRIK HELDIN‡ *Department of Biochemistry, The Cancer Institute of Japanese Foundation for Cancer Research ( JFCR), Tokyo 170-8455, Japan; †The Netherlands Cancer Institute, 1066CX Amsterdam, The Netherlands; ‡Ludwig Institute for Cancer Research, S-751 24 Uppsala, Sweden

I. Abstract

Members of the transforming growth factor-웁 (TGF-웁) family bind to type II and type I serine/threonine kinase receptors, which initiate intracellular signals through activation of Smad proteins. Receptorregulated Smads (R-Smads) are anchored to the cell membrane by interaction with membrane-bound proteins, including Smad anchor for receptor activation (SARA). Upon ligand stimulation, R-Smads are phosphorylated by the receptors and form oligomeric complexes with common-partner Smads (Co-Smads). The oligomeric Smad complexes then translocate into the nucleus, where they regulate the transcription of target genes by direct binding to DNA, interaction with various DNA-binding proteins, and recruitment of transcriptional coactivators or corepressors. A third class of Smads, inhibitory Smads (I-Smads), inhibits the signals from the serine/threonine kinase receptors. Since the expression of I-Smads is induced by the TGF-웁 superfamily proteins, Smads constitute an autoinhibitory signaling pathway. The functions of Smads are regulated by other signaling pathways, such as the MAP kinase pathway. Moreover, Smads interact with and modulate the functions of various transcription factors which are downstream targets of other signaling pathways. Loss of function of certain Smads is involved in tumorigenesis, e.g., pancreatic and colorectal cancers. Analyses by gene targeting revealed pivotal roles of Smads in early embryogenesis, angiogenesis, and immune functions in vivo. II. The TGF-␤ Superfamily

The TGF-웁 superfamily is composed of approximately 30 members in mammals and, in addition, several in other species, including Drosophila melanogaster and Caenorhabditis elegans (Kingsley, 1994). The TGF-웁 superfamily includes TGF-웁s, activins/inhibins, BMPs, mu¨llerian inhibiting substance (MIS), and myostatin. Although each ligand has a broad range of biological activities, TGF-웁s and activins activate similar signaling pathways, whereas BMPs activate a distinct set of receptor substrates. Signal 115

Copyright 䉷 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2776/00 $35.00

116

MIYAZONO et al.

transduction pathways for other TGF-웁 superfamily proteins, e.g., inhibins, MIS, and myostatin, have not yet been elucidated. TGF-웁s are multifunctional proteins that regulate the growth, differentiation, adhesion, and apoptosis of various cell types, including epithelial cells, endothelial cells, hematopoietic cells, and lymphocytes (Roberts and Sporn, 1990; Miyazono et al., 1994). TGF-웁 inhibits the growth of most cell types. In addition, TGF-웁 induces tissue fibrosis by stimulating the production of extracellular matrix proteins and inhibiting the activity of enzymes that degrade matrix proteins. TGF-웁 also induces an IgA class switch of B lymphocytes (Coffman et al., 1989; Sonoda et al., 1989). Activin was originally identified as a factor that induces secretion of folliclestimulating hormone from the pituitary gland, but it is now known also to stimulate differentiation of erythroid cells and to induce dorsal mesoderm in Xenopus laevis embryos (Mathews, 1994). BMPs include multiple members, e.g., BMP-2, BMP-4, OP-1 (osteogenic protein-1, also termed BMP-7), and growth-differentiation factor-5 (GDF-5, also termed cartilage-derived morphogenetic protein-1, or CDMP1) (Kawabata et al., 1998a). BMPs induce bone and cartilage formation in vivo but also play pivotal roles in morphogenesis of various other tissues. BMPs regulate the development of hematopoietic stem cells (Bhatia et al., 1999). In Xenopus embryos, BMPs induce ventral mesoderm; blocking the BMP signals by truncated BMP type I receptors or BMP-binding proteins, such as noggin and chordin, leads to the induction of neural tissues. The aim of this review is to discuss the intracellular signaling mechanisms that relay cellular effects of members of the TGF-웁 superfamily. III. Serine/Threonine Kinase Receptors

A. TYPE II AND TYPE I RECEPTORS Members of the TGF-웁 superfamily bind to two different types of serine/ threonine kinase receptors, termed type II and type I receptors (Massague´, 1998). Ligands induce hetero-oligomeric receptor complexes, presumably composed of two molecules each of type II and type I receptors. Type II receptors are constitutively active kinases, which transphosphorylate type I receptors at the GS domain located upstream of the serine/threonine kinase region, as well as in certain other regions (Heldin et al., 1997) (Fig. 1). Type I receptors then activate intracellular substrates such as Smad (Sma- and Mad-related) proteins; thus the type I receptors determine the specificity of the intracellular signals. Mutation of Thr204 in the GS domain of human TGF-웁 type I receptor (T웁R-I) to an aspartic acid residue [T웁RI(TD)] results in the constitutive activation of the serine/threonine kinase. Mutation of the corresponding residues (threonine or glutamine) in other type I receptors to acidic amino acids also leads to the constitutive activation

TGF-웁 SIGNALING BY SMAD PROTEINS

117

FIG. 1. R-Smad activation by serine/threonine kinase receptors, complex formation with Co-Smad, and nuclear translocation. R-Smads are anchored to membrane through SARA before receptor activation. Ligand binding leads to phosphorylation of the GS domain of type I receptor (R-I) by type II receptor (R-II). The R-I kinase phosphorylates R-Smads at the C-terminal SSXS motif. Then, R-Smads form complexes with Co-Smads (presumably, hetero-trimeric complexes) and translocate into the nucleus, where the Smad complexes associate with other DNA binding proteins and transcriptional coactivator or corepressor and regulate transcription of target genes.

of their serine/threonine kinases. T웁R-I(TD) transduces signals in the absence of ligands and type II receptors, which is consistent with the notion that type I receptors act downstream of the type II receptors. B. THREE DISTINCT SUBGROUPS OF TYPE I RECEPTORS Seven different type I receptors have been isolated in mammals; these were originally termed activin receptor-like kinase (ALK) 1 through 7 (ten

118

MIYAZONO et al.

Dijke et al., 1994a, b). These receptors can be divided into three subgroups based on their sequence similarities (Fig. 2). ALK5 and ALK4 are type I receptors for TGF-웁 (T웁R-I) and activin (ActR-IB), respectively. ALK7 is structurally similar to T웁R-I and ActR-IB, but its ligand has not yet been identified. The T웁R-I group includes these three mammalian receptors and Drosophila Atr-I, which transduce similar, but not identical, signals. Another subgroup (BMPR-I group) is constituted by BMP type IA receptor (BMPR-IA or ALK3), BMPR-IB (ALK6), and Drosophila Thick veins (Tkv), which are similar to each other and specifically bind BMPs in the presence of an appropriate type II receptor. ALK1 and ALK2 constitute another subgroup (ALK1 group), which also includes Drosophila Saxophone (Sax). ALK2 has been shown to bind activin, but it is most likely mainly a type I receptor for certain BMPs, e.g., BMP-6 and OP-1/BMP7 (ten Dijke et al., 1994b; Macı´as-Silva et al., 1998; Ebisawa et al., unpublished data). The physiological ligand for ALK1 is still unknown, but recent data suggest that TGF-웁 binds to ALK1 in certain cell types, including endothelial cells (Lux et al., 1999; Imamura et al., unpublished data). C. THREE-DIMENSIONAL STRUCTURE OF T웁R-I The FK506-binding immunophilin, FKBP12, interacts with T웁R-I and other type I receptors (Wang et al., 1994). FKBP12 binds to a Leu–Pro sequence in the GS domain and protects against ligand-independent, spontaneous activation of type I receptors by type II receptors (Wang et al., 1996; Y. G. Chen et al., 1997). The crystal structure of the cytoplasmic domain of T웁R-I was determined in a complex with FKBP12 (Huse et

FIG. 2. Ligands, receptors, and corresponding R-Smads and Co-Smad for the TGF웁 superfamily.

TGF-웁 SIGNALING BY SMAD PROTEINS

119

al., 1999). The inactive conformation of the T웁R-I cytoplasmic domain is maintained by interactions between the GS domain, the N-terminal lobe involved in the ATP-binding, and the activation loop of the kinase. The receptor is converted to its active conformation by phosphorylation in the GS domain by T웁R-II. The L45 loop of the N lobe of type I receptors protrudes out from the kinase domain, so that it can interact with intracellular substrates, such as Smads (see Section V.B). IV. Structure and Function of Smads

In the JAK/STAT pathways activated by cytokine receptors, receptor oligomerization and association with the JAK family of tyrosine kinases induce the phosphorylation of STATs, which translocate into the nucleus and act as transcription factors. In a similar scenario, activation of the serine/threonine kinase receptors leads to phosphorylation of receptorregulated Smads (or pathway-restricted Smads, or R-Smads), which, after forming a complex with common-partner Smads (Co-Smads), translocate into the nucleus. The oligomeric Smad complexes then bind to DNA together with other DNA-binding proteins, including winged-helix transcription factor FAST1, and regulate the transcription of target genes. In contrast to STATs, which require JAKs for activation, R-Smads are direct substrates of the serine/threonine kinase receptors. A. THREE DIFFERENT CLASSES OF SMADS Smads have molecular masses of about 42–65 kDa. Eight different Smads have thus far been identified in mammals and can be classified into three subclasses: R-Smads, Co-Smads, and inhibitory Smads (I-Smads, or Anti-Smads) (Heldin et al., 1997). R-Smads can be further subdivided into two subtypes: those activated by TGF-웁 and activin receptors and those activated by BMP receptors. The former group includes Smad2 and Smad3, and the latter involves Smad1, Smad5, and Smad8. Co-Smad is a shared component by TGF-웁/activin and BMP pathways. Smad4 is thus far the only Co-Smad identified in mammals, whereas two Smad4-like molecules have been isolated in Xenopus and C. elegans. Smad6 and Smad7, which are structurally more distantly related to other Smads, act as I-Smads by antagonizing the activity of R-Smads. Smads have highly conserved N- and C-terminal regions, termed Mad homology domain (MH) 1 and 2 (also called N- and C-domain), respectively (Fig. 3). The MH1 and MH2 domains are bridged by a proline-rich linker region with variable length and amino acid sequence. MH2 domains are observed in all three classes of Smad proteins. MH1 domains are found in R-Smads and Co-Smads, but the N-terminal regions of I-Smads are

120

MIYAZONO et al.

FIG. 3. Structures of Smad2 (R-Smad), Smad4 (Co-Smad), and Smad7 (I-Smad). MH1 and MH2 domains that are conserved between Smads are shown as shaded boxes. In Smad2, two regions in the MH1 domain that are conserved between Smad2 and Smad3 are shown as hatched boxes. In Smad7, N-terminal regions similar between Smad6 and Smad7 are shown as hatched boxes. The 웁-hairpin region in the MH1 domain binds to DNA, whereas the MH1 L3 loop encoded by exon 3 interferes with the DNA binding. The L3 loop in the MH2 domain is essential for the interaction with the type I receptors, and the 움-helix H1 is also required for the association with the type I receptors in the ALK1 group. The 움-helix H2 in the MH2 domain is most important for binding to FAST1, and the interaction is supported by the C-terminal tail.

highly divergent from those in other Smads. In addition, R-Smads have a characteristic Ser-Ser-Val/Met-Ser sequence (SSXS motif ) in their most C-terminal parts, which are phosphorylated by the serine/threonine kinase receptors (Macı´as-Silva et al., 1996; Kretzschmar et al., 1997a; Abdollah et al., 1997; Souchelnytskyi et al., 1997). Smad2 and Smad3 are activated by TGF-웁 and activin in most cell types. In epidermal keratinocytes, activin phosphorylates Smad3 more efficiently than Smad2, whereas TGF-웁 activates both Smad2 and Smad3 (Shimizu et al., 1998). At early stages of development, however, Smad2 may be predominantly activated by activin and may play an important role in the dorsal mesoderm induction, together with FAST1 or related proteins. It is currently not known whether Co-Smads are absolutely required for all the signals in the Smad pathways. Loss of Smad4 expression was shown to lead to loss of responses to TGF-웁/activin (Zhou et al., 1998a) as well as to BMP in human cancer cells ( Jonk et al., 1998). In Drosophila, however, Medea was shown to be dispensable for the activation of the

TGF-웁 SIGNALING BY SMAD PROTEINS

121

Mad-responsive gene, optomoter blind (omb), during wing development (Wisotzkey et al., 1998). Since R-Smads can bind to DNA in the absence of Co-Smad (Kawabata et al., 1998b; Vindevoghel et al., 1998), it is possible that certain genes can be activated in the absence of Co-Smad. Alternatively, there may be yet unidentified Co-Smads in mammalian cells (see Section IV.C). Smad1, Smad5, and Smad8 are preferentially activated by BMPs (Y. Chen et al., 1997; Tamaki et al., 1998). However, it was reported that Smad1 and Smad5 are activated by TGF-웁 in certain cell types. Thus, TGF-웁 induced the phosphorylation and functional activation of Smad1 in human breast cancer cells and rat intestinal epithelial cells (X. Liu et al., 1998; Yue et al., 1999a). Using an antisense oligonucleotide approach, Smad5 was shown to mediate the growth inhibitory effect of TGF-웁 in hematopoietic cells (Bruno et al., 1998). Because TGF-웁 binds to ALK1 with weak affinity in endothelial cells (Attisano et al., 1993; Imamura et al., unpublished data), and because ALK1 induces phosphorylation of Smad1 and Smad5, it is possible that TGF-웁-induced Smad1/5 activation occurs through ALK1. It is also possible that TGF-웁 can indirectly activate Smad1/5 by T웁R-I through other signaling pathways (Yue et al., 1999b). B. STRUCTURES OF SMADS 1. MH2 Domain The MH2 domain is an effector domain with approximately 200 amino acid residues and is important for interaction with the receptors, oligomer formation, interaction with other DNA binding proteins, and transcriptional activation. When fused to the DNA-binding domain of GAL4, the MH2 domains of Smad1 and Smad2 show transcriptional activity in the absence of ligand stimulation (Liu et al., 1996, 1997). Before activation of R-Smads by the receptors, the MH2 and MH1 domains repress the activity of each other by physical interaction (Fig. 1), which is removed by the phosphorylation of the C-terminal SSXS motif in the MH2 domain. The MH2 structure consists of a 웁-sheet sandwich capped at one end by a three-움-helix bundle (움-helices H3, H4, and H5) and at the other end by a loop/helix structure composed of three loops and an 움-helix (L1, L2, and L3 loops and 움-helix H1) (Shi et al., 1997) (Fig. 3). The C-terminal SSXS motif in the R-Smads is located following the 움-helix H5. In between 움-helices H3 and H4, a 35-amino acid region is present in Co-Smads, the function of which remains to be elucidated. The corresponding regions of I-Smads are shorter and dissimilar. The MH2 domain of Smad4 forms a trimer in solution, mainly through the interaction between the 움-helix bundle and the loop/helix structure, although the L3 loop is not involved in this interaction (Shi et al., 1997).

122

MIYAZONO et al.

The L3 loop, 움-helices H1 and H2, and the most C-terminal tail of the MH2 domain are divergent between Smads (Fig. 3). The L3 loop, a region composed of 17 amino acid residues (Lo et al., 1998), protrudes from the surface and determines the specificity of the interaction with type I receptors. The amino acid sequences of the L3 loop are conserved within the subgroups of R-Smads activated by TGF-웁/activin (Smad2/3) and within the subgroups of BMP-activated R-Smads (Smad1/5/8), but are different at two amino acid residues between the two groups. The 움-helix H1 with eight amino acid residues is different in two or three amino acid residues between Smad2/3 and Smad1/5/8. In addition to the L3 loop, the 움-helix H1 is required for the interaction of Smad1/5/8 with the ALK1 group of type I receptors (Chen and Massague´, 1999). The 움-helix H2 and the C-terminal tail of the MH2 domain of R-Smads determine the specificity of the downstream pathways (Y. G. Chen et al., 1998; Lagna and Hemmati-Brivanlou, 1999). The 움-helix H2 is most important for the specific interaction with nuclear DNA binding proteins, including FAST1 and possibly other proteins involved in transduction of BMP signals. In addition, the C-terminal tail upstream of the SSXS motif of R-Smads supports this interaction and acts as a secondary determinant (Lagna and Hemmati-Brivanlou, 1999). 2. MH1 Domain MH1 domains have approximately 130 amino acid residues, composed of a core structure with three 움-helices on one side and two small 웁-sheets and a 웁-hairpin on the other side (Shi et al., 1998). MH1 is responsible for direct DNA binding, which occurs mainly through a protruded, highly conserved 11 amino acid 웁-hairpin (Fig. 3). Smad2 does not bind to DNA because of interference by an insert of 30 amino acid residues just upstream of the DNA binding 웁-hairpin. An alternatively spliced variant of Smad2, lacking the 30 amino acid residues, binds to DNA and has functions similar to Smad3 (Yagi et al., 1999; Dennler et al., 1999). Drosophila dSmad2 does not bind to DNA because of the difference in amino acid sequence in the 웁-hairpin (Das et al., 1999). Another unique region of 10 amino acid residues (L1; Fig. 3) is found in Smad2 but not in Smad3, but the deletion of this region does not induce the DNA binding ability of Smad2 (Dennler et al., 1999). The C-terminal half of 움-helix 2 is rich in basic residues (basic helix); the possibility that this region serves as a nuclear localization signal is unlikely, however, judging by the observation that isolated MH1 domains have been shown to localize in the cytoplasm (Zhang et al., 1997). L2 and L4 loops from a double-loop region, which is located on the opposite site of the DNA-binding 웁-hairpin. The double-loop region is exposed on the

TGF-웁 SIGNALING BY SMAD PROTEINS

123

surface and may be involved in the interaction with other molecules. The MH1 domain binds to several DNA-binding proteins, including ATF-2 and vitamin D receptor (see Sections VI.B and VIII.D). 3. Proline-Rich Linker Region The functions of the linker region have not been fully determined. PXS/TP (or S/TP) motifs are found in R-Smads, which may be phosphorylated by MAP kinases (see Section VIII.B) (Fig. 3). A PY motif, which is known to interact with WW domains, is found in all Smad proteins except for Smad4 (Attisano and Wrana, 1998). A WW domain-containing protein that specifically interacts with Smads has been identified (Zhu et al., 1999). A C-terminal part of the Smad4 linker region (amino acid 274–321) (de Caestecker et al., 1997; Zhang et al., 1997) is not conserved in other Smads and is essential for the signaling activity of Smad4, presumably through stabilization of oligomer structures with R-Smads or through binding to other proteins with transcriptional activity, such as melanocyte-specific gene 1 (MSG1) (Shioda et al., 1998; see Section VI.C). An alternatively spliced variant of Smad8, lacking 37 amino acid residues in the linker region, is observed in rat (Y. Chen et al., 1997) and human (also termed MADH6) (Watanabe et al., 1997). Sequences similar to the 37 amino acid region are also found in Smad1 and Smad5; whether there is a functional difference between these two forms of Smad8, however, remains to be determined. C. SMADS IN XENOPUS AND INVERTEBRATES 1. Smads in Xenopus Smad1 and Smad2 are highly conserved between mammals and Xenopus. In contrast, Xenopus Smad6 (Nakayama et al., 1998a) and Smad7 (formerly termed XSmad8; Nakayama et al., 1998b) have only 52 and 74% amino acid sequence identity, respectively, to their amino acid sequences in their mammalian orthologs. The MH2 domain (71%) and linker region (60%) of Smad6 are relatively well conserved between mouse and Xenopus, but the N-terminal region is poorly conserved (35%), suggesting important functional roles of the MH2 and linker regions of I-Smads. In addition to the Smad4 ortholog (XSmad4움), XSmad4웁 (also termed Smad10 or Smad4b) with 70% sequence similarity has been identified in Xenopus (LeSueur and Graff, 1998; Masuyama et al., 1999; Howell et al., 1999). XSmad4움 and XSmad4웁 are structurally divergent at the linker region. XSmad4웁 is expressed in early developmental stages and begins to decrease by midgastrulation, when the expression of XSmad4움 is induced. In contrast to XSmad4움, XSmad4웁 is constitutively phosphorylated and located in the nucleus (Masuyama et al., 1999). XSmad4웁 has a C-

124

MIYAZONO et al.

terminal SSVN sequence, but its phosphorylation by type I receptors has not been shown. Similar to XSmad4움, XSmad4웁 forms a complex with RSmads of the TGF-웁/activin pathways and of the BMP pathways (Masuyama et al., 1999; Howell et al., 1999). Using a Xenopus assay, however, XSmad4웁 was suggested to preferentially act in the BMP pathways (Masuyama et al., 1999). In addition, XSmad4웁 was shown to mimic the action of Spemann organizer and to induce anterior and posterior neural tissues in Xenopus embryos (LeSueur and Graff, 1998). An XSmad4웁-like molecule has not yet been identified in mammals. 2. Smads in Drosophila Four different Smads have been identified in Drosophila. Mad (Mothers against Dpp) is a founding member of the Smad family and is activated by a BMP-like ligand, Decapentaplegic (Dpp) (Raftery et al., 1995; Sekelsky et al., 1995; Newfeld et al., 1996). dSmad2 may be activated by an as-yetunidentified activin-like ligand (Brummel et al., 1999; Das et al., 1999). The type I receptor Tkv activates Mad, whereas Atr-I activates dSmad2. Medea is a Co-Smad (Das et al., 1998; Hudson et al., 1998; Wisotzkey et al., 1998; Inoue et al., 1998). Dad (daughters against Dpp) acts as an ISmad, and, similar to Smad6 and Smad7 in mammals, it is induced by Dpp signals (Tsuneizumi et al., 1997; Inoue et al., 1998). 3. Smads in C. elegans In C. elegans, TGF-웁-like pathways play important roles in the dauer pathway and in the body size determination and male tail patterning. Sma2, Sma-3, and Sma-4 are functionally nonredundant and act downstream of the type I receptor SMA-6 in the pathway involved in body size determination and male tail patterning (Savage et al., 1996; Krishna et al., 1999). Sma-2 and Sma-4 may belong to R-Smads and Co-Smad, respectively. Sma-3 is similar to R-Smads, but it has a C-terminal NSMT sequence instead of the SSXS motif. Whether Sma-3 is phosphorylated by type I receptors remains to be determined. In the dauer pathway, the type I receptor DAF-1 transduces signals possibly via three Smad-like molecules, DAF-8, DAF-14, and DAF-3 (Krishna et al., 1999). DAF-3 is structurally most similar to Co-Smads. Thus, there may be two Smad4-like molecules in C. elegans. However, instead of acting positively in signal transduction, DAF-3 is partially localized in the nucleus in the absence of ligand stimulation and antagonizes the receptor action. DAF-3 binds to DNA via its MH1 domain to ‘‘GTCTG’’ sequence and negatively regulates the enhancer activity of the target genes (Patterson et al., 1997; Thatcher et al., 1999).

TGF-웁 SIGNALING BY SMAD PROTEINS

125

D. BIOLOGICAL FUNCTION OF SMADS Whether Smads are major signal mediators of the TGF-웁 superfamily and whether Smads can mimic the biological effects induced by the ligands or receptors are important questions. In mammalian cells, overexpression of Smad4 induces growth inhibition, G1 cell cycle arrest, and apoptosis (Atfi et al., 1997a; Hunt et al., 1998; Le Dai et al., 1999). Smad2 and Smad3 can also induce apoptosis in epithelial cells in vitro (Yanagisawa et al., 1998). TGF-웁 induces the transcriptional activation of various genes, which have been shown to be regulated by Smad3/Smad4 and Smad2/ Smad4. p21/WAF1/Cip1, an inhibitor of cyclin/cyclin-dependent kinase (CDK) complexes, is induced by TGF-웁 in certain cell types and plays a critical role in the growth inhibition induced by TGF-웁 (Datto et al., 1995). Expression of p21/WAF1/Cip1 is induced by Smad3/Smad4 and, to a much lower extent, by Smad2/Smad4 complexes in hepatic cells (Moustakas and Kardassis, 1998). p15/INK4B, a CDK inhibitor, is induced by TGF-웁 (Li et al., 1995), whereas Cdc25A, a tyrosine phosphatase acting on G1 CDKs, is repressed (Iavarone and Massague´, 1997, 1999); however, involvement of Smads in the transcriptional regulation of these genes has not been reported. Expression of extracellular matrix proteins is also tightly regulated by the action of Smad3/Smad4. Smad-binding elements (SBE) are observed in the promoter regions of plasminogen activator inhibitor (PAI)-1, type I collagen (Dennler et al., 1998), and type VII collagen genes (Vindevoghel et al., 1998). p3TP-Lux, widely used in TGF-웁 response assays, contains a part of the promoter region of the PAI-I gene and three tandem repeats of AP-1 binding sites of the collagenase I gene. In contrast to induction of most extracellular matrix proteins, induction of fibronectin by TGF-웁 can occur independent of Smad4 action but requires the c-Jun N-terminal kinase ( JNK)/stress-activated protein kinase (SAPK) pathway and activation of the cAMP responsive elements (CRE) by c-Jun/ATF-2 heterodimer (Hocevar et al., 1999). Smad1 and Smad5 act downstream of BMP type I receptors, and expression of these R-Smads induces osteoblast differentiation of osteoprogenitor cells, similar to the effects of BMPs or constitutively active forms of BMP type I receptors (Nishimura et al., 1998; Yamamoto et al., 1997; Fujii et al., 1999). However, the effects of Smad1 and Smad5 are less potent than those of the ligands or receptors; nuclear translocation of Smads appears to be required for the efficient signaling activity. In contrast to the promoters that respond to TGF-웁/activin, not much is known about the promoter sequences that respond to BMPs. Tlx-2 is a

126

MIYAZONO et al.

homeobox gene related to human Hox11 gene and is expressed in the primitive streak of mouse embryos (Tang et al., 1998). BMPs, BMP type I receptors, and Smad1 induce the transcriptional activation of the Tlx-2 promoter in murine embryonal carcinoma P19 cells (Tang et al., 1998). Xvent.2 is a homeobox gene that mediates the early effect of BMP-4 in Xenopus embryos. The Xvent.2 promoter (Candia et al., 1997) is also activated by the action of Smad1 in the P19 cells and in the C2C12 cells (Y. G. Chen et al., 1998; Masuyama et al., 1999). E. REGULATION OF SMAD GENE EXPRESSION Expression of I-Smads is regulated by various stimuli, including TGF웁 and BMP signals (see Section VII.B). However, less is known about the regulation of R-Smad and Co-Smad expression. Smad2, but not Smad1, expression is induced by TGF-웁 stimulation in granulosa cells (Li et al., 1997). Smad3 expression is downregulated in response to TGF-웁 in normal human lung epithelial cells (Yanagisawa et al., 1998). Overexpression of Smad2 induces upregulation of the Smad4 gene in osteoblastic cells (Li et al., 1998).

V. Cytoplasmic Actions of Smads

A. MEMBRANE ANCHORING OF R-SMADS R-Smads transiently interact with and become phosphorylated by activated type I receptors. A plasma membrane-associated protein, termed Smad anchor for receptor activation (SARA), was identified, which facilitates recruitment of Smad2/3 to TGF-웁 receptors by directly interacting with nonphosphorylated Smad2/3 and TGF-웁 receptors (Tsukazaki et al., 1998). SARA contains a putative membrane phospholipid interaction module, termed FYVE domain, through which targeting to the plasma membrane is achieved. Ectopic expression of SARA mutants, which lack the FYVE domain, is diffusely distributed in the cytoplasm and potently interferes with TGF-웁/Smad-mediated responses. Upon SARA-mediated Smad presentation and phosphorylation by type I receptor, the components within the SARA/Smad/receptor complex dissociate, allowing Smad2/3 to associate with Smad4 and freeing SARA for recruitment of other Smads (Fig. 1). SARA may not only enhance the efficiency of receptor signaling but may also increase selectivity by favoring presentation of specific substrates. SARA does not interact with Smad1. A SARA-like molecule in the BMP receptor/Smad pathway likely exists, considering the high mechanistic similarity between TGF-웁 and BMP signaling.

TGF-웁 SIGNALING BY SMAD PROTEINS

127

B. ACTIVATION OF R-SMADS BY THE RECEPTORS R-Smads are phosphorylated by serine/threonine kinase receptors at the C-terminal SSXS motif, in which the last two serine residues are phosphorylated by type I receptors (Abdollah et al., 1997; Souchelnytskyi et al., 1997). Kinase activity of T웁R-II is required for the interaction of T웁R-I with Smad2 or Smad3. Smad2/3 interact with T웁R-I only transiently and they are released from the receptor complexes after phosphorylation. The association can, thus, be seen in vitro only with the kinase inactive form of T웁R-I or with Smad2/3 mutants with the C-terminal SSXS motif replaced with an AAXA sequence (Macı´as-Silva et al., 1996; Nakao et al., 1997b). The interaction between the type I receptors and R-Smads can be detected within 5 min and starts to decrease at 15 min after ligand stimulation (Lebrun et al., 1999). Interaction of Smad2/3 with Smad4 is then observed 30 min after ligand stimulation and increases up to 90 min after stimulation. The Smad complex translocates into the nucleus and can be detected there for at least 3 h after ligand simulation (Shimizu et al., 1998; Kretzschmar et al., 1999). It is unknown how Smads are degraded or whether they are exported again to the cytoplasm. The L45 loop between kinase subdomain IV and V is composed of nine amino acid residues and determines the specificity of the intracellular signals (Feng and Derynck, 1997; Y. G. Chen et al., 1998; Persson et al., 1998; Armes et al., 1999). The amino acid sequences of the L45 loops are conserved within each group of type I receptors. However, four amino acids are different between the T웁R-I group and BMPR-I group. Analysis by chimeric receptors revealed that Smad2 and Smad3 bind to the L45 loop of the T웁R-I group, whereas Smad1, and presumably Smad5 and Smad8, interact with the L45 loop of the BMPR-I group (Y. G. Chen et al., 1998). The sequences of L45 loops of the ALK1 group are most distantly related to the other type I receptors, and only two amino acid residues are conserved between the ALK1 group and BMPR-I group. However, ALK1 and ALK2 can interact with Smad1, and presumably Smad5 and Smad8, and transduce BMP-like signals in the cells similar to those in the BMPR-I group (Macı´as-Silva et al., 1998; Chen and Massague´, 1999). R-Smads interact with the type I receptors mainly through the L3 loop in the MH2 domain (Lo et al., 1998; Y. G. Chen et al., 1998) (Fig. 3). In addition to the L3 loop, the 움-helix H1 of Smad1, located in the vicinity of the L3 loop in the three-dimensional structure is essential for the interaction with ALK1 (Chen and Massague´, 1999) (see Section IV.B). C. OLIGOMER FORMATION AND NUCLEAR TRANSPORT OF SMADS In the absence of TGF-웁 stimulation, Smads exist as monomers in the cytoplasm in vivo. Smad2 and Smad3 form homo-oligomers and hetero-

128

MIYAZONO et al.

oligomers upon phosphorylation by TGF-웁, and this oligomerization does not necessarily require Smad4. (Nakao et al., 1997b; Kawabata et al., 1998b; Topper et al., 1998). In the presence of Smad4, Smad2/3 may form a heterotrimer, which is probably composed of two molecules of Smad2/3 and one molecule of Smad4 (Kawabata et al., 1998b; Tada et al., unpublished data). Data supporting the notion that a complex containing Smad2/ Smad3/Smad4 is most potent in the transcriptional activation of p3TP-Lux has been presented (Nakao et al., 1997b). Most, but not all, Smads exist in the cytoplasm in unstimulated cells. Activated R-Smads/Co-Smad complexes translocate into the nucleus. Nuclear translocation of R-Smads occurs in the absence of Co-Smad, whereas Smad4 requires the presence of R-Smads to accumulate into the nucleus (F. Liu et al., 1997; Souchelnytskyi et al., 1997). It is not known how the activated Smads are transported to the nucleus. Smad2 or Smad3 lacking the MH1 domain is constitutively localized in the nucleus (Baker and Harland, 1996; Meersseman et al., 1997; Zhang et al., 1997). The linker region of Smad3, but not of Smad4, also has a tendency to accumulate into the nucleus (Zhang et al., 1997). In contrast, isolated MH1 domains are seen in the cytoplasm, although a region rich in basic amino acid residues (basic helix) is observed in the MH1 domain (Shi et al., 1998). This suggests that Smad3 may have two portions important for nuclear localization, one in the MH2 domain and another in the linker region. Smad4 does not have the latter region. In addition, the basic helix in the MH1 domain may serve as a nuclear localization signal after oligomerization of Smads. D. INTERACTION OF SMAD2 WITH CALMODULIN Calmodulin was shown to interact with Smad2 in a calcium-dependent manner (Zimmerman et al., 1998). The interaction occurs through the MH1 and linker regions of Smad2. Overexpression of calmodulin repressed the transcriptional activation of p3TP-Lux. The functional role of calmodulin in signal transduction by Smads remains to be elucidated. VI. Actions of Smads in the Nucleus

A. DIRECT DNA BINDING OF SMADS Certain Smads have been shown to possess an intrinsic sequence-specific DNA-binding activity. The MH1 domain of Drosophila Mad was shown to specifically bind a G ⫹ C-rich sequence that is essential for Dppdependent transcriptional activation of the vestigial gene. The interaction of Mad with DNA is weak and occurs only when the MH2 is removed (Kim et al., 1997). Following these observations, Smad3 and Smad4 were

TGF-웁 SIGNALING BY SMAD PROTEINS

129

also shown to bind to DNA. Smad binding-site selection from random pools of oligonucleotides revealed that the Smad3 and Smad4 MH1 domains specifically recognize a GTCT (or inverted AGAC)-containing sequence (Zawel et al., 1998). Whereas full-length Smad4 interacts with DNA, Smad3 must be released from an intramolecular inhibitory activity by MH2 domain deletion (Dennler et al., 1998; Jonk et al., 1998) or receptor-induced phosphorylation (Kawabata et al., 1998b). Smad3/4-binding elements (SBEs, containing AGAC sequences) were identified in several promoters of TGF-웁-induced genes, including PAI-1 (Dennler et al., 1998; Song et al., 1998; Stroschein et al., 1999), JunB ( Jonk et al., 1998), type VII collagen (Vindevoghel et al., 1998), and germline immunoglobulin (Ig) C움 region (Pardali et al., 2000; Hanai et al., 1999). Mutational analysis of the SBE (CAGACA) in the JunB promoter revealed that the GAC core sequence is most critical for Smad binding ( Jonk et al., 1998). Multimers of SBEs confer TGF-웁 responsiveness to a minimal promoter (Zawel et al., 1998; Dennler et al., 1998; Jonk et al., 1998), and mutation of these sequences in the PAI-1 and Ig C움 promoters impairs TGF-웁 responsiveness (Dennler et al., 1998; Stroschein et al., 1999; Pardali et al., 2000; Hanai et al., 1999). The sequences adjacent to the SBE can effect the efficiency of Smad DNA binding (Stroschein et al., 1999). The DNA-interacting 웁-hairpin loop that protrudes from the MH1 core is highly conserved among R- and Co-Smads (see Section IV.B). Smad1, however, binds less efficiently to CAGA boxes than do Smad3 and Smad4 (Pardali et al., 2000; Imamura et al., unpublished data). Whether Smad1 and Smad5 are able to bind other specific DNA sequences with higher affinity remains to be elucidated. In the goosecoid promoter, Smad3 and Smad4 have been shown to bind a G ⫹ C-rich sequence that is unrelated to the AGAC-containing sequence (Labbe´ et al., 1998). B. INTERACTING PARTNERS IN THE NUCLEUS Smads have DNA-binding ability, but the affinity is relatively low. In order to efficiently bind to DNA and activate transcription, sequencespecific DNA-binding proteins are recruited to Smad complexes. Thus, two distinct sequences are required for the promoters, Smad-binding sequence and another adjacent (or overlapping) site for DNA-binding proteins that cooperatively act with Smads (Fig. 4). When the interaction between a DNA-binding protein and the Smad complex occurs with a relatively high affinity, the DNA-binding protein may serve as a primary DNA-binding component in the complex. If DNA-binding proteins and Smads interact only weakly or not at all with each other, the DNA-binding proteins and Smads may bind to DNA independently; after binding to

130

MIYAZONO et al.

FIG. 4. DNA binding of Smads with other DNA binding proteins and transcriptional regulation in the nucleus. (A) When the interaction between DNA binding proteins (e.g., FAST1) and the Smad complex occurs with a relatively high affinity, the DNA binding protein may serve as a primary DNA binding component in the complex. (B) When DNA binding proteins (e.g., TFE3) and Smads interact only weakly or do not interact directly, the DNA binding proteins and Smads may bind to DNA independently; after binding to adjacent regions in the promoters, the DNA binding proteins and Smads may act synergistically in transcriptional regulation. (C) Certain transcriptional repressors may bind to DNA in the absence of TGF-웁/activin or BMP signals, and the repression may be relieved after interaction with Smads.

adjacent regions in the promoters, the DNA-binding proteins and Smads may synergistically induce transcriptional regulation. 1. Xenopus FAST1 and Mammalian FASTs Xenopus FAST1 (forkhead activin signal transducer 1) is a winged-helix transcription factor, which binds to the activin responsive element (ARE) in the Mix.2 promoter and plays an essential role in early development (Chen et al., 1996). FAST1 interacts with Smad2/3 and Smad4 upon activin stimulation and forms an activin responsive factor (ARF) complex (X. Chen et al., 1997; Liu et al., 1997) (Fig. 4A). FAST1 is the principal DNAbinding component in ARF. In addition, the Mix.2 gene promoter contains two SBEs around the FAST1-binding site, and Smad4 directly binds to DNA. Since Smad2 is as potent as Smad3 in the transcriptional activation of the Mix.2 gene promoter (Yagi et al., 1999), direct DNA binding of Smad3 may not be critical in the ARF complex. The interaction between FAST1 and Smad2 occurs through the C-terminal Smad-interacting domain (SID) of FAST1 and the MH2 domain of Smad2. Smad4 binds to FAST1 in the presence of Smad2, and binding of Smad4 stabilizes the Smad2/FAST1 complex. FAST1 does not interact with Smad1.

TGF-웁 SIGNALING BY SMAD PROTEINS

131

Two mammalian homologs of FAST1, human FAST1 (referred to as hFAST1 in this review) and murine FAST2 (mFAST2), have been identified (Zhou et al., 1998b; Labbe´ et al., 1998; Liu et al., 1999). The sequence similarities between these three molecules are very low except for the winged-helix region and C-terminal SID. hFAST1 is widely expressed in adult tissues, whereas mFAST2 is expressed only at early developmental stages. Both hFAST1 and mFAST2 form complexes with Smad2 and activate transcription upon TGF-웁 or activin stimulation. Smad2 stimulates the transcription of the goosecoid promoter in a complex with Smad4 and mFAST2, but Smad3 was shown to strongly repress the transcription (Labbe´ et al., 1998). This may be because of competition of Smad3 with Smad4 for the Smad-binding G ⫹ C-rich sequence located in the goosecoid promoter, which may lead to a conformational change of the DNA binding complex. Thus, Smad2 and Smad3 act oppositely in the transcriptional activation of the goosecoid promoter. 2. AP-1 ( Jun/Fos) Complex and ATF-2 AP-1 transcription factors composed of c-Jun and c-Fos bind to the TPA-responsive elements (TREs) known as AP-1-binding sites. Smad3/ Smad4 acts cooperatively with c-Jun and c-Fos to mediate transcriptional activation of, e.g., the collagenase I and the c-Jun promoters in response to TGF-웁 (Zhang et al., 1998; Wong et al., 1999). Smad3 physically interacts with c-Jun in a ligand-inducible manner, the interaction occurs through the MH1 and linker regions of Smad3 and the C-terminal portion of cJun. JunB and JunD also interact with Smad3 (Liberati et al., 1999). cFos interacts weakly with Smad3 through the MH2 domain of Smad3. Smad2, but not Smad1, weakly cooperates with c-Jun (Zhang et al., 1998). Since TGF-웁 activates SAPK/JNK in certain cell types (see Section VIII.B) and phosphorylates c-Jun, the SAPK/JNK pathway may contribute to the Smad-mediated transcriptional activation (Liberati et al., 1999). ATF-2 (also called CRE-BP-1) is a member of the ATF/CREB family, which binds to CRE. The sequences of the CRE and AP-1 sites are similar to each other, and the proteins of the ATF/CREB family act as homodimers as well as heteordimers with c-Jun. TGF-웁 induces the activation of the p38 MAP kinase in certain cell types, which then phosphorylates ATF-2 and enhances its activity. TGF-웁 induces the interaction between ATF2 and Smad3/Smad4 through the MH1 domain of Smad3 and induces synergistic transcriptional responses (Sano et al., 1999). Smad pathways and MAPK/SAPK/p38 signaling pathways may thus converge at the AP1- or CRE-binding promoters. 3. PEBP2움/CBFA/AML Polyoma virus enhancer binding protein 2 (PEBP2; also called core binding factor, CBF) is a transcription factor complex composed of 움 and

132

MIYAZONO et al.

웁 subunits (Ito and Bae, 1997). Three mammalian members of PEBP2움/ CBFA have a conserved Runt domain critical for DNA binding, whereas the 웁 subunit enhances the DNA-binding activity of the 움 subunits. PEBP2움A/ CBFA1-null mice exhibit complete loss of bone formation, whereas PEBP2움B/CBFA2 (also called AML1)-deficient mice show defects in definitive hematopoiesis. PEBP2움C/CBFA3 is required for stimulation of germline Ig C움 promoter prior to immunoglobulin class switch to generate IgA (Shi and Stavnezer, 1998; Xie et al., 1999). PEBP2움B/AML1 gene is disrupted by chromosomal translocations in several types of human leukemias. All three mammalian members of PEBP2움/CBFA interact with R-Smads, and, for example, Smad3 and PEBP2움B/CBFA2/AML1 or PEBP2움C/CBFA3 synergistically stimulate the transcription of the germline Ig C움 promoter (Hanai et al., 1999; Pardali et al., 2000). Thus, PEBP2/ CBFA may function as a nuclear target of Smads in specific tissues. 4. Other Transcription Factors Certain transcription factors synergize with Smads, although direct physical interactions with Smads have not been demonstrated. These include TFE3, Sp1, and Drosophila Tinman. Transcription factor 애E3 (TFE3) synergistically functions with Smad3/4 in the transcriptional activation of the PAI-1 promoter upon TGF-웁 stimulation, through binding to the adjacent sites in the PAI-1 promoter (Hua et al., 1998) (Fig. 4B). TFE3 was originally isolated as a basic helix–loop–helix transcription factor that binds to the E-box sequence in the enhancer of an immunoglobulin gene (Beckmann et al., 1990). In the p21/WAF1/Cip1 promoter, multiple Sp1 sites are observed, which are essential for the constitutive activation of the promoter (Datto et al., 1995). Although physical interaction has not been demonstrated, Smad3/ Smad4 and, to a much lesser extent, Smad2/Smad4 were shown to functionally synergize with Sp1 in the transcriptional activation of the p21/WAF1/ Cip1 promoter (Moustakas and Kardassis, 1998). In Drosophila tinman gene expression, Mad and Medea were shown to bind to the tinman promoter at G ⫹ C-rich sequences and to function synergistically with Tinman (Xu et al., 1998). Tinman is a homeobox protein in the NK family, including Nkx2.5, which is induced by BMPs and is required for normal heart development in vertebrates (Lyons et al., 1995; Schultheiss et al., 1997). Drosophila Schnurri (Shn) gene encodes a zincfinger protein with similarity to mammalian transcription factors of the MBP/HIV-EP/PRDII-BF family (Arora et al., 1995; Grieder et al., 1995; Staehling-Hampton et al., 1995). Because Shn is a downstream component of the Dpp signaling pathway, it may also interact with Mad/Medea to regulate transcriptional responses.

TGF-웁 SIGNALING BY SMAD PROTEINS

133

Zinc-finger transcription factors of the Gli family are involved in hedgehog signaling. Although full-length Gli proteins do not interact with Smads, C-terminally truncated Gli3, which occurs in Pallister–Hall syndrome and polydactyly type A, interacts with Smads 1 through 4 (F. Liu et al., 1998). Complex formation between R-Smads and the truncated Gli3 is repressed by ligand stimulation. The functional significance of formations of the Smad–Gli3 complex remains to be determined. 5. Transcriptional Repressors Certain nuclear proteins act as transcriptional repressors, and the repression may be relieved by the action of Smads (Fig. 4C). A homeodomain transcription factor, Hoxc-8, binds to the osteopontin promoter at a specific binding sequence and represses the transcription. Smad1 interacts with Hoxc-8 and dissociates it from the DNA-binding element, leading to transcriptional activation in response to BMP signaling (Shi et al., 1999). SIP1, a member of the ␦EF1/Zfh-1 family of the two-handed zinc finger/ homeodomain proteins, binds to R-Smads through the MH2 domain in a ligand-dependent manner. SIP1 can bind to 5⬘-CACCT sequences in the brachyury promoter, and ectopic expression of SIP1 inhibits brachyury expression. Interaction of R-Smads with SIP1 may prevent binding of SIP1 to the brachyury promoter and thereby relieve SIP1 from its transcriptional repressor function (Verschueren et al., 1999). Drosophila Brinker (Brk) is a nuclear protein which may also act as a transcriptional repressor. In the nucleus, Brk represses the transcription of Dpp target genes. In Xenopus, ectopically expressed Brk antagonizes the action of BMP-4 as well as that of Mad. Dpp signals relieve the repression by Brk by downregulation of the expression of brk mRNA or by antagonizing the repressor activity of Brk at the promotors of Dpp target genes ( Jazwinska et al., 1999; Campbell and Tomlinson, 1999; Minami et al., 1999). 6. Evi-1 Evi-1 is a transcriptional regulator with two zinc-finger domains. The t(3;21)(q26;q22) translocation seen in blastic crisis of chronic myelogenous leukemia generates a fusion product of AML1 (PEBP2움B/CBFA2) and Evi-1. Evi-1 interacts with Smad3, but not with other Smads, through the first zinc-finger domain of Evi-1 and the MH2 domain of Smad3 (Kurokawa et al., 1998a). Evi-1 and the fusion product AML1/Evi-1 repress TGF-웁 signaling by preventing the binding of Smad3 to DNA and may contribute to leukemogenesis by blocking the TGF-웁-mediated growth inhibition (Kurokawa et al., 1998a; 1998b).

134

MIYAZONO et al.

C. TRANSCRIPTIONAL COACTIVATORS AND COREPRESSORS The acetylation state of core histones is critical for the regulation of transcription (Struhl, 1998; Travers, 1999). p300 and CREB-binding protein (CBP) have histone acetyl transferase (HAT) activity. These transcriptional coactivators facilitate transcription by loosening the nucleosomal structure and by increasing the accessibility to the general transcription machinery (Giles et al., 1998). In contrast, histone deacetylases (HDACs) induce nucleosomal condensation and thereby repress transcription. 1. Transcriptional Coactivators p300 and CBP p300 and CBP are structurally similar proteins, which were originally identified as proteins interacting with the adenoviral E1A protein and the CRE-binding protein (CREB), respectively (Chrivia et al., 1993; Eckner et al., 1994). Many different DNA-binding proteins, including p53, CREB, AP-1, STATs, MyoD, NF-␬B, and nuclear steroid receptors, have been shown to interact with p300/CBP. Through binding to multiple transcription factors, p300/CBP may integrate various signals, e.g., by bridging various transcription factors (see Section VIII.C). R-Smads, including Smad1, Smad2, and Smad3, interact with p300/CBP after ligand stimulation (Feng et al., 1998; Janknecht et al., 1998; Nishihara et al., 1998; Pouponnot et al., 1998; Topper et al., 1998; Shen et al., 1998). The MH2 domains of R-Smads are responsible for the interaction with p300/CBP. The C-terminal part of p300/CBP adjacent to the E1A-binding domain is most important for the interaction with Smads, although the N-terminal portion of p300/CBP can also weakly bind to Smads (Feng et al., 1998; Nakashima et al., 1999). p300/CBP induce TGF-웁- and Smad-induced transcriptional activation in a Smad4-dependent manner. E1A blocks the Smad3/Smad4-induced transcriptional activation by competing for binding to p300/CBP, as well as by direct binding to Smads (Nishihara et al., 1999). Mad recruits Drosophila CBP (also termed Nejire or Nej) and activates the transcription of Dpp-responsive genes during fly development. Lossof-function mutants of Drosophila CBP/nej cause embryonic lethality and various patterning defects (Akimaru et al., 1997), which are similar to those seen in mutants that lack Dpp or Mad. Moreover, nej interacts with a weak allele of dpp (Waltzer and Bienz, 1999). Drosophila CBP binds to the MH2 domain of Mad through its C-terminal region, in a manner similar to that by which mammalian p300/CBP binds to Smads. 2. Transcriptional Corepressor, TGIF In contrast to p300/CBP, TGIF, which belongs to the TALE class of homeodomain proteins, acts as a transcriptional corepressor by recruiting HDACs. TGIF interacts with Smad2 and Smad3, and weakly with Smad1,

TGF-웁 SIGNALING BY SMAD PROTEINS

135

in a ligand-dependent manner (Wotton et al., 1999). Interaction of Smad2/ 3 with TGIF occurs through the MH2 domain. TGIF binds to the Smad2/ Smad4/mFAST2 complex and stabilizes the complex. Activated Smads thus recruit TGIF and HDACs to the promoters of target genes and repress transcription. Because the interaction of Smads with p300 and TGIF are mutually exclusive, TGIF may play a role in modulating the magnitude of TGF-웁 responses through competing with p300/CBP for the interaction with Smads. 3. MSG1 MSG1 is a 27-kDa nuclear protein with a strong transcriptional activity, although it does not bind to DNA by itself. In yeast two-hybrid system, MSG1 was shown to interact with the C-terminal domain of Smad4, including the C-terminal part of the linker region (de Caestecker et al., 1997), and thus it may confer transcriptional activity to Smad4 upon ligand stimulation (Shioda et al., 1998). VII. I-Smads

A. FUNCTION OF I-SMADS A distinct subgroup of Smads, including mammalian Smad6 (Imamura et al., 1997) and Smad7 (Hayashi et al., 1997; Nakao et al., 1997a), Xenopus XSmad6 (Nakayama et al., 1998a) and XSmad7 (Nakayama et al., 1998b), and Drosophila Dad (Tsuneizumi et al., 1997), has been identified as intracellular antagonists of TGF-웁 family signaling. These inhibitory Smads, or I-Smads, stably interact with activated type I receptors and prevent R-Smads from being phosphorylated by these receptors (Imamura et al., 1997; Hayashi et al., 1997; Nakao et al., 1997a; Inoue et al., 1998). Smad7 can inhibit the binding of Smad2 to activated T웁R-I (Hayashi et al., 1997); I-Smads may thus compete with R-Smads for type I receptor interaction (Fig. 5). Another mechanism of action has been proposed for Smad6: Smad6 prevents Smad1/Smad4 complex formation by competing with Smad4 for binding to Smad1 (Hata et al., 1998). Smad7 interacts with activated TGF-웁, activin, and BMP type I receptors and is a general inhibitor of TGF-웁 superfamily-induced responses (Hayashi et al., 1997; Nakao et al., 1997a; Souchelnytskyi et al., 1998; Ishisaki et al., 1998, 1999; Lebrun et al., 1999). Smad6 binds many, but not all, type I receptors and was found to preferentially inhibit BMP signaling in mammalian cells (Imamura et al., 1997; Itoh et al., 1998; Ishisaki et al., 1999). In Xenopus embryos, XSmad6 and in particular XSmad7 can antagonize activin-like signaling, but both are much more potent inhibitors of BMPs (Nakayama et al., 1998a, b; Hata et al., 1997; Casellas and Hemmati-Brivanlou, 1998;

136

MIYAZONO et al.

FIG. 5. Mechanism of action of I-Smads. Smad7 exists in the nucleus before ligand stimulation and is exported to the cytoplasm after activation by TGF-웁. Smad6 is observed both in the nucleus and in the cytoplasm. Both Smad6 and Smad7 stably interact with type I receptors and inhibit the signaling by R-Smads (1). Smad6 was also shown to compete with Smad4 for the complex formation with R-Smads (2).

Bhushan et al., 1998). Structure–function analysis has revealed that an intact C-terminal MH2 domain of I-Smads is sufficient for type I receptor interaction and inhibitory activity (Souchelnytskyi et al., 1998). A short Cterminal deletion abrogates the receptor interaction, suggesting that an intact MH2 domain is required for its inhibitory function (Hayashi et al., 1997). The N-terminal domains may function to target specific pathways for inhibition (Souchelnytskyi et al., 1998). In Xenopus, XSmad6 was found to be partially or completely localized in the nuclei of most cells (Nakayama et al., 1998a). In mammalian cells, Smad7 is localized in the nucleus but is predominantly located in the cytoplasm upon TGF-웁 stimulation (Itoh et al., 1998) (Fig. 5). The mechanism for ligand-induced nuclear export of Smad7 remains to be elucidated. It is not known whether nuclear ISmads have a role in transcriptional regulation and effects distinct from their antagonistic role in receptor-mediated R-Smad activation.

TGF-웁 SIGNALING BY SMAD PROTEINS

137

B. REGULATION OF I-SMAD EXPRESSION I-Smads are ubiquitously expressed in various human tissues, most abundantly in the lung, a tissue rich in endothelial cells (Imamura et al., 1997; Nakao et al., 1997a). Smad6 and Smad7 were found to be highly expressed in vascular endothelial cells, and their expression was induced after laminar shear stress (Topper et al., 1997). In addition, Smad7 is highly expressed in the developing vascular system during mouse embryogenesis (Zwijsen et al., unpublished data). The role of I-Smad expression in endothelial cell function remains to be elucidated. The expression of I-Smads is quickly and directly induced upon stimulation by TGF-웁 superfamily members, and I-Smads may thus participate in a negative autocrine feedback loop (Tsuneizumi et al., 1997; Nakao et al., 1997a; Nakayama et al., 1998b; Takase et al., 1998; Afrakhte et al., 1998; Ishisaki et al., 1998, 1999). In addition to regulating the intensity or duration of the ligand responses that triggered their expression, I-Smads may regulate each other’s responses. For example, TGF-웁-induced expression of Smad6 or Smad7 may inhibit BMP signaling. Furthermore, regulation of I-Smad levels may provide the possibility for cross-talk of other signaling pathways with the TGF-웁 superfamily/Smad pathway. Of note in this respect, epidermal growth factor (EGF) induced the expression of I-Smads, and phorbol ester TPA potentiated the TGF-웁-induced expression of Smad7 (Afrakhte et al., 1998). Interestingly, interferon-웂 (IFN-웂) acting through Jak1 and STAT1 was found to induce Smad7 expression and thereby inhibit TGF-웁-induced responses (Ulloa et al., 1999; see Section VIII.C). Overexpression of ISmads could be a mechanism by which cancer cells escape TGF-웁-induced growth inhibition. In one study, Smad6 was found to be overexpressed in pancreatic cells (Kleeff et al., 1999), whereas in another report no change was observed in I-Smad expression ( Jonson et al., 1999). VIII. Signaling Cross-Talk

A. CROSS-TALK BETWEEN SMAD PATHWAYS Smad4 is a shared component in the signaling pathways for TGF-웁/ activin and BMPs. Using Xenopus embryo assays, it was shown that if the pool of Smad4 in the cells is limited, these two signaling pathways may antagonize each other by competition for Smad4 (Candia et al., 1997) (Fig. 6). When the BMP pathways are activated, Smad4 is sequestered by Smad1/ 5/8 and the cells become resistant to the action of activin/TGF-웁. When the activin/TGF-웁 pathways are activated, a similar limitation of the BMP signals occurs. If Smad4 expression is increased, this antagonistic effect can be overridden. I-Smads may also play a role for cross-talk between TGF-웁/activin and BMP pathways (see Section VII.B).

138

MIYAZONO et al.

FIG. 6. Regulation of signaling by serine/threonine kinase receptors and Smads. FKBP12 represses the activity of type I receptors through binding to the GS domains. I-Smads inhibit the signals by R-Smads (see Fig. 5). Co-Smad is shared by different signaling pathways; thus, signals will be repressed when Co-Smad is sequestered by other TGF-웁 superfamily pathways. Erk MAP kinase phosphorylates the linker regions of R-Smads, which prevent the nuclear accumulation of R-Smads. Transcriptional corepressor (TGIF) competes with transcriptional coactivators p300/CBP, leading to the inhibition of transcriptional activation.

B. MAP KINASE PATHWAYS Members of the TGF-웁 superfamily have synergistic and/or opposing effects with other signaling molecules, and Smads play pivotal roles in the cross-talk with these pathways. Three distinct MAP kinase pathways in mammals, Erk, SAPK/JNK, and p38, have been shown to be activated by TGF-웁/BMPs in certain cell types (Hartsough and Mulder, 1995; Frey and Mulder, 1997; Atfi et al., 1997b; Liberati et al., 1999; Zhou et al., 1999) and also to positively or negatively modulate the Smad pathways. Growth factors, such as hepatocyte growth factor (HGF) and EGF, bind to tyrosine kinase receptors and activate the Erk MAP kinase pathway,

TGF-웁 SIGNALING BY SMAD PROTEINS

139

which leads to phosphorylation of Smad1, as well as Smad2 and Smad3, and inhibits their nuclear accumulation (Fig. 6). Signals induced by tyrosine kinase receptors thus antagonize the BMP and TGF-웁 pathways (Kretzschmar et al., 1997b; 1999). Phosphorylation by the Erk MAP kinase occurs on serine (or threonine) residues in PXS/TP or S/TP motifs, four copies of which are observed in the linker regions of Smad1, 2, and 3 (Fig. 3). Ras-transformed cells become resistant to TGF-웁, which may be induced by the inhibition of nuclear translocation of Smad2/3 by oncogenically mutated, hyperactive Ras (Kretzschmar et al., 1999). In contrast, MAP kinase pathway activated by HGF was shown to induce phosphorylation of Smad2 and to facilitate Smad2 accumulation into the nucleus, although to a lesser extent than TGF-웁 (de Caestecker et al., 1998). A Smad2 mutant at the C-terminal SSXS motif blocked the HGF-mediated Smad2 signals. Thus, Smad2 may transduce common signals of HGF and TGF-웁 under certain conditions. However, it is not known under which conditions the Erk MAP kinase inhibits or mediates Smad signaling pathways. The SAPK/JNK pathway has been shown to be activated by TGF-웁 in certain cell types (Atfi et al., 1997b; Frey and Mulder, 1997; Liberati et al., 1999) and to activate the Smad signaling pathway. MEKK1 is a MAP kinase kinase kinase acting upstream of SEK1/MKK4–SAPK/JNK. MEKK1 phosphorylates Smad2 at sites other than the C-terminal SSXS motif in endothelial cells, which leads to oligomerization and nuclear accumulation of Smad2 and Smad4 and transcriptional activation (Brown et al., 1999). Smad7 inhibits the Smad2-dependent transcriptional activation induced by MEKK1. In addition, SAPK/JNK phosphorylates c-Jun, which may act in concert with Smad3 in transcriptional activation (Liberati et al., 1999; see Section VI.B). The p38 MAP kinase pathway is also activated by TGF-웁/BMP stimulation in mammalian cells as well as in Drosophila (Adachi-Yamada et al., 1999). p38 induces the phosphorylation of the transcription factor ATF-2 and synergistically acts with Smads in transcriptional activation (Sano et al., 1999; see Section VI.B). TGF-웁 activated kinase (TAK) 1 has been suggested to act as a downstream component of TGF-웁/BMP receptors and to activate the SAPK/JNK and p38 MAP kinase pathways (Yamaguchi et al., 1995). However, activation of TAK1 by TGF웁/BMPs may occur only under certain conditions (Sakurai et al., 1999); TAK1 was recently reported to be efficiently activated by interleukin-1 and tumor necrosis factor (TNF)-움 (Ninomiya-Tsuji et al., 1999; Sakurai et al., 1999). C. STAT PATHWAYS Positive and negative cross-talks between STAT and Smad pathways have also been demonstrated. Leukemia inhibitory factor (LIF) and BMP-

140

MIYAZONO et al.

2 synergistically induce the differentiation of neuronal progenitors into astrocytes (Nakashima et al., 1999). LIF binds to a cell surface LIF receptor–gp130 complex, which activates JAKs and STAT3. STAT3 interacts with the N-terminal part of the transcriptional coactivator p300, whereas Smad1 associates with its C-terminal position. Thus, STAT3 and Smad1 are bridged by p300, which contributes to the cooperative signaling of LIF and BMP2. Transmodulation of Smad pathways by STAT pathways is also observed via induction of Smad7 (Ulloa et al., 1999). Treatment of cells with IFN웂 leads to resistance to TGF-웁 action. IFN-웂 activates STAT1 and the MAP kinase pathway. Although the Erk MAP kinase induces phosphorylation of Smad3 in a fashion similar to EGF or HGF, the MAP kinase pathway is not involved in the TGF-웁 resistance induced by IFN-웂. Instead, STAT1 induces the expression of Smad7, which consequently prevents the activation of Smad3 by TGF-웁. D. NUCLEAR STEROID RECEPTOR Smad pathways may also modulate the action of nuclear receptor for vitamin D. Cooperative actions of TGF-웁 and vitamin D have been reported in various cell types. Smad3, but not the other Smads, interacts with ligand-activated vitamin D receptor (VDR) and forms a complex with a member of the steroid receptor coactivator (SRC)-1/TIF2 family (Yanagisawa et al., 1999). Interaction between Smad3 and VDR occurs via the MH1 domain of Smad3. Other nuclear receptors, including estrogen receptor 움, retinoic acid receptor, or retinoid X receptor, do not interact with Smads. The Smad3–VDR–SRC-1 complex enhances transactivation of target genes containing vitamin D-responsive elements in their promoters, and Smad7, but not Smad6, efficiently inhibits this transcriptional activity (Yanagi et al., 1999). IX. Roles of Smads in Human Cancer

A. CHROMOSOMAL LOCALIZATION OF SMADS Smad2, Smad4, and Smad7 are colocalized on human chromosome 18q21.1 (Hahn et al., 1996; Riggins et al., 1996; Jonson et al., 1999). Smad3 and Smad6 are also colocalized on chromosome 15q21-22. Smad1 is located on 4q28, Smad5 on 5q31, and Smad8 on 13q12–14 (Watanabe et al., 1997). B. MUTATIONS OF SMADS IN HUMAN CANCER Losses of 15q and 18q are frequently observed in various tumors, including pancreatic cancers. Smad4 was originally isolated as a product of the tumor suppressor gene, DPC4 (deleted in pancreatic carcinoma, locus 4),

TGF-웁 SIGNALING BY SMAD PROTEINS

141

located on human chromosome 18q21.1 (Hahn et al., 1996). Mutations of Smad4 are observed in significant numbers of pancreatic cancers (27–53%), biliary tract carcinomas (16%), and colorectal cancers (16–22%) (Hahn et al., 1996, 1998; Thiagalingam et al., 1996), but mutations in other tumors, including breast carcinomas, ovarian carcinomas, head and neck cancers, and lung cancers, are less frequent (less than 12%) (reviewed in Duff and Clarke, 1998). Smad2 is altered in only small fractions of colorectal cancer and lung cancers (Riggins et al., 1996; Eppert et al., 1996; Uchida et al., 1996; Takagi et al., 1998). Familial juvenile polyposis is an autosomal dominant disease, which is characterized by hamartomatous polyps and gastrointestinal cancer. Germline mutations in the Smad4 gene have been found in some of the juvenile polyposis families (Howe et al., 1998). Abnormalities in the long arm of human chromosome 5 are frequently observed in myeloid dysplasia and acute myelogenous leukemia. However, mutations of Smad5 in the retained allele are not found in these hematological malignancies or human leukemia cell line HL60 (Hejlik et al., 1997; Zavadil et al., 1997). C. BIOLOGICAL ACTIONS OF SMAD MUTANTS 1. Mutations in the MH2 Domain Mutations in the MH2 domains can be classified into (1) those affecting the core structure, (2) those in the loop/helix region or the three-움-helix bundle region which may prevent the Smad oligomerization, and (3) those at the L3 loop. Mutations of Asp450 (located in 움-helix H5) in Smad2 (Fig. 3) and the corresponding Asp407 in Smad3 to glutamic acid result in strong interaction with T웁R-I but defective receptor-induced phosphorylation (Eppert et al., 1996; Lo et al., 1998; Goto et al., 1998). The Smad2 mutant is unable to interact with Smad4, although binding to FAST1 is not impaired (Hoodless et al., 1999). In contrast, mutations in the L3 loop, e.g., Gly421 to serine in Smad2, cause decrease in receptor binding and loss of TGF-웁-induced phosphorylation (Lo et al., 1998). A mutation of the corresponding Gly508 to serine in Smad4 abolishes the ability to form a complex with R-Smads, indicating that the L3 loop of Smad4 mediates the interaction with R-Smads (Shi et al., 1997). 2. Mutations in the MH1 Domain Four tumor-derived mutations in the MH1 domain have been reported, all of which are located in the L2 or L4 loop in the double loop region. Mutations of Arg100 to Thr in Smad4 (Schutte et al., 1996) and the corresponding Arg133 to cysteine in Smad2 (Eppert et al., 1996) (Fig. 3) have been shown to lead to increase in the affinity between the MH1 and MH2 domains and may prevent their dissociation after phosphorylation

142

MIYAZONO et al.

of the SSXS motif (Hata et al., 1997). Loss of DNA binding in the Arg100 mutant of Smad4 was also reported (Stroschein et al., 1999). X. In Vivo Functions of Smads: Analyses by Gene Targeting

A. ROLES OF SMADS IN EMBRYOGENESIS Analyses by gene targeting revealed that most Smads play pivotal roles during embryonic development. Mice lacking Smad4, Smad2, or Smad5 die during embryonic development. Homozygous Smad4-mutant mice die before embryonic day 7.5 (E7.5) with gastrulation defect and abnormal visceral endoderm (Sirard et al., 1998; Yang et al., 1998). In chimeras of the Smad4-mutant ES cells with wild-type cells, the gastrulation defect could be rescued, indicating that the primary requirement of Smad4 in gastrulation resides in extraembryonic tissues rather than in the embryos (Sirard et al., 1998). Smad4 was also shown to be involved in anterior patterning during gastrulation. The Smad4-mutant phenotype is reminiscent of those of the BMP-4- and BMPR-IA/ALK3-deficient mice (Mishina et al., 1995; Winnier et al., 1995). Smad2-mutant embryos die before E8.5 with lack of egg cylinder elongation and of mesoderm induction (Nomura and Li, 1998; Weinstein et al., 1998) and a failure in the establishment of anterior–posterior axis (Waldrip et al., 1998). The phenotype mimics those of nodal- and ActR-IB/ALK-4-deficient mice (Conlon et al., 1994; Gu et al., 1998), which thus may be physiological ligand and receptor, respectively, acting upstream of Smad2 during early embryogenesis. Smad5mutant mice show multiple defects and die between E9.5 and E11.5 (Yang et al., 1999a; Chang et al., 1999). The Smad5-deficient embryos have defects in angiogenesis, exhibiting enlarged vessels with decreased numbers of smooth muscle cells. The phenotype is similar to those of TGF-웁1- and T웁R-II-deficient mice (Dickson et al., 1995; Oshima et al., 1996). Because ALK1 binds TGF-웁 (Attisano et al., 1993; Lux et al., 1999; Imamura et al., unpublished data) and activates Smad1 and Smad5 (Macı´as-Silva et al., 1998; Fujii et al., 1999), Smad5 may function as a downstream component in the TGF-웁 pathway in vivo. B. SMADS IN IMMUNE FUNCTION AND TUMOR DEVELOPMENT Although Smad3 is expressed in embryos, Smad3-deficient mice are viable and die 1 to 8-months after birth (Zhu et al., 1998; Yang et al., 1999b; Datto et al., 1999). The Smad3⫺/⫺ mice are smaller than the wildtype littermates. In adult tissues, the highest expression of Smad3 is observed in spleen and thymus (Datto et al., 1999). The Smad3-null mice exhibit leukocytosis and impaired mucosal immunity with massive inflammation and bacterial abscesses. The B cells of the Smad3-null mice are

TGF-웁 SIGNALING BY SMAD PROTEINS

143

sensitive to TGF-웁, whereas mutant T cells and neutrophils do not respond to TGF-웁. Embryonic fibroblasts from the Smad3-null mice do not induce p3TP-Lux transcriptional activation upon TGF-웁 stimulation, and the growth inhibitory action of TGF-웁 is largely lost. The Smad3⫺/⫺ mice suffer from and die of chronic infection. Some of the TGF-웁1-knockout mice are also alive until birth, but they die of massive infiltration of leukocytes in multiple organs (Shull et al., 1992; Kulkarni et al., 1993; Diebold et al., 1995). The phenotype of TGF-웁1-null mice and that of Smad3-null mice are thus dissimilar. The results of gene targeting revealed that Smad3 cannot compensate for the defect in Smad2-knockout mice during early development and that Smad3 may have exclusive roles in certain adult tissues, e.g., T cells and leukocytes. Smad3-mutant mice have also been shown to develop metastatic colorectal cancer in mice with certain genetic backgrounds (Zhu et al., 1998), although mutations in the Smad3 gene have not been found in human tumors. In heterozygous mice carrying mutations of Smad4 and Apc (responsible for human familial adenomatous polyposis) on the same chromosome, loss of heterozygosity and reduplication of the gene carrying the mutations result in the intestinal polyposis with more malignant phenotypes than the simple Apc heterozygotes (Takaku et al., 1998). Thus, Smad3 and Smad4 may play critical roles in the malignant progression of colorectal tumors. XI. Perspectives/Conclusion

Our knowledge regarding how serine/threonine kinase receptors are activated and signal through Smad molecules has been increasing dramatically. Activation of R-Smads by receptor-induced phosphorylation is followed by complex formation with Co-Smads and translocation to the nucleus, where the transcription of specific genes is affected. A negative feedback mechanism involving induction of I-Smads has also been elucidated. Although Smads are clearly very important in the signal transduction downstream of serine/threonine kinase receptors, it is also likely that other parallel pathways are activated, which are important for the cellular responses to stimulation by TGF-웁 superfamily members. Several interesting examples of cross-talk between Smads and other signaling pathways, including MAP kinase, STATs, and vitamin D, have been elucidated. Cross-talk between different signaling pathways is a common theme in signal transduction and appears to be a mechanism whereby a cell can integrate different signal inputs and respond in an appropriate manner. Considering the highly context-dependent character of the action

144

MIYAZONO et al.

of TGF-웁 family members, many more examples of cross-talk involving Smad molecules are likely to be discovered in the near future. It is striking that each serine/threonine kinase receptor generally binds more than one ligand, and at the same time any given ligand often binds to more than one signaling receptor. Moreover, each of the type I receptors often activates more than one R-Smad molecule. One possible reason why different members of the TGF-웁 superfamily still give different cellular responses is cell type-specific expression of the different components in the signaling pathways. Smad2 and Smad3, for example, have distinct, sometimes even antagonistic properties, and thus their relative expression levels within the cell are important determinants for TGF-웁-induced cellular response. It is likely that additional mechanisms for generation of signal specificity in individual cells will be unraveled. A picture is emerging whereby Smad molecules in a cell type-specific manner act in concert with other transcription factors, and co-activators and co-repressors, to control the expression of a certain set of genes. An important future task will be to identify nuclear Smad partners and their roles in transcriptional complexes. Moreover, the availability of gene array techniques will make it possible to identify in detail the target genes in individual cell types. Together with results from proteome techniques, which will give additional information, e.g., regarding proteins which are specifically degraded in response to ligand, our understanding of the functional role of individual Smad pathways will be enhanced. The mechanism whereby activated R-Smad/Co-Smad complexes are rapidly and efficiently translocated into the nucleus is not understood. Upon activation, Smad molecules may be released from a putative cytoplasmic anchor, or, alternatively, a nuclear localization signal may be exposed. Another interesting possibility, which remains to be investigated, is that translocation involves the interaction with specific transport proteins. Several other questions remain to be addressed, e.g., regarding the stoichiometry of Smad complexes, whether Smads undergo posttranslational modifications other than phosphorylation, regarding the mechanism for deactivation of Smads, whether Smads have functions not related to transcriptional regulation, and whether Smads may function in pathways not regulated by TGF-웁 superfamily members. Given the high activity in the field, answers to these questions are likely to come soon. Overactivity of TGF-웁 has been implicated in the progression of certain disorders, e.g., in different types of fibrotic conditions in which the potent effect of TGF-웁 on matrix accumulation is important. Thus, clinically useful antagonists of TGF-웁 action would be warranted. On the other hand, lack of TGF-웁 responsiveness is implicated in certain forms of cancer, whereby loss of growth inhibitory signals contributes to loss of

TGF-웁 SIGNALING BY SMAD PROTEINS

145

growth control. Thus, TGF-웁 agonists may also be useful. Because TGF웁 has such a wide variety of effects on cells, complete blocking or stimulation of all TGF-웁 signals might be associated with severe side effects. The recent insight into the mechanism of signal transduction of TGF-웁 has now given numerous targets for drug discovery. Thus, it might be possible to design inhibitors or activators of specific pathways of TGF-웁, which, through more precise actions, might be clinically useful. REFERENCES Abdollah, S., Macı´as-Silva, M., Tsukazaki, T., Hayashi, H., Attisano, L., and Wrana, J. L. (1997). T웁RI phosphorylation of Smad2 on Ser465 and Ser467 is required for Smad2–Smad4 complex formation and signaling. J. Biol. Chem. 272, 27678–27685. Adachi-Yamada, T., Nakamura, M., Irie, K., Tomoyasu, Y., Sano, Y., Mori, E., Goto, S., Ueno, N., Nishida, Y., and Matsumoto, K. (1999). p38 mitogen-activated protein kinase can be involved in transforming growth factor 웁 superfamily signal transduction in Drosophila wing morphogenesis. Mol. Cell. Biol. 19, 2322–2329. Afrakhte, M., More´n, A., Jossan, S., Itoh, S., Sampath, K., Westermark, B., Heldin, C.-H., Heldin, N.-E., and ten Dijke, P. (1998). Induction of inhibitory Smad6 and Smad7 mRNA by TGF-웁 family members. Biochem. Biophys. Res. Commun. 249, 505–511. Akimaru, H., Chen, Y., Dai, P., Hou, D. X., Nonaka, M., Smolik, S. M., Armstrong, S., Goodman, R. H., and Ishii, S. (1997). Drosophila CBP is a co-activator of cubitus interruptus in hedgehog signalling. Nature 386, 735–738. Armes, N. A., Neal, K. A., and Smith, J. C. (1999). A short loop on the ALK-2 and ALK-4 activin receptors regulates signaling specificity but cannot account for all their effects on early Xenopus development. J. Biol. Chem. 274, 7929–7935. Arora, K., Dai, H., Kazuko, S. G., Jamal, J., O’Connor, M. B., Letsou, A., and Warrior, R. (1995). The Drosophila schnurri gene acts in the Dpp/TGF웁 signaling pathway and encodes a transcription factor homologous to the human MBP family. Cell 81, 781–790. Atfi, A., Buisine, M., Mazars, A., and Gespach, C. (1997a). Induction of apoptosis by DPC4, a transcriptional factor regulated by transforming growth factor-웁 through stress-activated protein kinase/c-Jun N-terminal kinase (SAPK/JNK) signaling pathway. J. Biol. Chem. 272, 24731–24734. Atfi, A., Djelloul, S., Chastre, E., Davis, R., and Gespach, C. (1997b). Evidence for a role of Rho-like GTPases and stress-activated protein kinase/c-Jun N-terminal kinase (SAPK/ JNK) in transforming growth factor 웁-mediated signaling. J. Biol. Chem. 272, 1429–1432. Attisano, L., Ca´rcamo, J., Ventura, F., Weis, F. M. B., Massague´, J., and Wrana, J. L. (1993). Identification of human activin and TGF웁 type I receptors that form heteromeric kinase complexes with type II receptors. Cell 75, 671–680. Attisano, L., and Wrana, J. L. (1998). Mads and Smads in TGF웁 signalling. Curr. Opin. Cell Biol. 10, 188–194. Baker, J. C., and Harland, R. M. (1996). A novel mesoderm inducer, Madr2, functions in the activin signal transduction pathway. Genes Dev. 10, 1880–1889. Beckmann, H., Su, L. K., and Kadesch, T. (1990). TFE3: A helix-loop-helix protein that activates transcription through the immunoglobulin enhancer 애E3 motif. Genes Dev. 4, 167–179. Bhatia, M., Bonnet, D., Wu, D., Murdoch, B., Wrana, J., Gallacher, L., and Dick, J. E. (1999). Bone morphogenetic proteins regulate the developmental program of human hematopoietic stem cells. J. Exp. Med. 189, 1139–1148.

146

MIYAZONO et al.

Bhushan, A., Chen, Y., and Vale, W. (1998). Smad7 inhibits mesoderm formation and promotes neural cell fate in Xenopus embryos. Dev. Biol. 200, 260–268. Brown, J. D., DiChiara, M. R., Anderson, K. R., Gimbrone, M. A., Jr., and Topper, J. N. (1999). MEKK-1, a component of the stress (stress-activated protein kinase/c-Jun Nterminal kinase) pathway, can selectively activate Smad2-mediated transcriptional activation in endothelial cells. J. Biol. Chem. 274, 8797–8805. Brummel, T., Abdollah, S., Haerry, T. E., Shimell, J. M., Merriam, J., Raftery, L., Wrana, J. L., and O’Connor, M. B. (1999). The Drosophila activin receptor baboon signals through dSmad2 and controls cell proliferation but not patterning during larval development. Genes Dev. 13, 98–111. Bruno, E., Horrigan, S. K., Van Den Berg, D., Rozler, E., Fitting, P. R., Moss, S. T., Westbrook, C., and Hoffman, R. (1998). The Smad5 gene is involved in the intracellular signaling pathways that mediate the inhibitory effects of transforming growth factor-웁 on human hematopoiesis. Blood 91, 1917–1923. Campbell, G., and Tomlinson, A. (1999). Transducing the Dpp morphogen gradient in the wing of Drosophila: Regulation of Dpp targets by brinker. Cell 96, 553–562. Candia, A. F., Watabe, T., Hawley, S. H., Onichtchouk, D., Zhang, Y., Derynck, R., Niehrs, C., and Cho, K. W. (1997). Cellular interpretation of multiple TGF-웁 signals: Intracellular antagonism between activin/BVg1 and BMP-2/4 signaling mediated by Smads. Development 124, 4467–4480. Casellas, R., and Hemmati-Brivanlou, A. H. (1998). Xenopus Smad7 inhibits both the activin and BMP pathways and acts as a neural inducer. Dev. Biol. 198, 1–12. Chang, H., Huylebroeck, D., Verschueren, K., Guo, Q., Matzuk, M. M., and Zwijsen, A. (1999). Smad5 knockout mice die at mid-gestation due to multiple embryonic and extraembryonic defects. Development 126, 1631–1642. Chen, X., Rubock, M. J., and Whitman, M. (1996). A transcriptional partner for MAD proteins in TGF-웁 signalling. Nature 383, 691–696. Chen, X., Weisberg, E., Fridmacher, V., Watanabe, M., Naco, G., and Whitman, M. (1997). Smad4 and FAST-1 in the assembly of activin-responsive factor. Nature 389, 85–89. Chen, Y., Bhushan, A., and Vale, W. (1997). Smad8 mediates the signaling of the ALK-2 receptor serine kinase. Proc. Natl. Acad. Sci. USA 94, 12938–12943. Chen, Y. G., Liu, F., and Massague´, J. (1997). Mechanism of TGF웁 receptor inhibition by FKBP12. EMBO J. 16, 3866–3876. Chen, Y. G., Hata, A., Lo, R. S., Wotton, D., Shi, Y., Pavletich, N., and Massague´, J. (1998). Determinants of specificity in TGF-웁 signal transduction. Genes Dev. 12, 2144–2152. Chen, Y. G., and Massague´, J. (1999). Smad1 recognition and activation by the ALK1 group of transforming growth factor-웁 family receptors. J. Biol. Chem. 274, 3672–3677. Chrivia, J. C., Kwok, R. P., Lamb, N., Hagiwara, M., Montminy, M. R., and Goodman, R. H. (1993). Phosphorylated CREB binds specifically to the nuclear protein CBP. Nature 365, 855–859. Coffman, R. L., Lebman, D. A., and Shrader, B. (1989). Transforming growth factor 웁 specifically enhances IgA production by lipopolysaccharide-stimulated murine B lymphocytes. J. Exp. Med. 170, 1039–1044. Conlon, F. L., Lyons, K. M., Takaesu, N., Barth, K. S., Kispert, A., Herrmann, B., and Robertson, E. J. (1994). A primary requirement for nodal in the formation and maintenance of the primitive streak in the mouse. Development 120, 1919–1928. Das, P., Maduzia, L. L., Wang, H., Finelli, A. L., Cho, S. H., Smith, M. M., and Padgett, R. W. (1998). The Drosophila gene Medea demonstrates the requirement for different classes of Smads in dpp signaling. Development 125, 1519–1528.

TGF-웁 SIGNALING BY SMAD PROTEINS

147

Das, P., Inoue, H., Baker, J. C., Beppu, H., Kawabata, M., Harland, R. M., Miyazono, K., and Padgett, R. W. (1999). Drosophila dSmad2 and Atr-I transmit activin/TGF웁 signals. Genes Cells 4, 123–134. Datto, M. B., Yu, Y., and Wang, X.-F. (1995). Functional analysis of the transforming growth factor 웁 responsive elements in the WAF1/Cip1/p21 promoter. J. Biol. Chem. 270, 28623–28628. Datto, M. B., Frederick, J. P., Pan, L., Borton, A. J., Zhuang, Y., and Wang, X.-F. (1999). Targeted disruption of smad3 reveals an essential role in transforming growth factor 웁mediated signal transduction. Mol. Cell. Biol. 19, 2495–2504. de Caestecker, M. P., Hemmati, P., Larisch-Bloch, S., Ajmera, R., Roberts, A. B., and Lechleider, R. J. (1997). Characterization of functional domains within Smad4/DPC4. J. Biol. Chem. 272, 13690–13696. de Caestecker, M. P., Parks, W. T., Frank, C. J., Castagnino, P., Bottaro, D. P., Roberts, A. B., and Lechleider, R. J. (1998). Smad2 transduces common signals from receptor serine–threonine and tyrosine kinases. Genes Dev. 12, 1587–1592. Dennler, S., Itoh, S., Vivien, D., ten Dijke, P., Huet, S., and Gauthier, J. M. (1998). Direct binding of Smad3 and Smad4 to critical TGF웁-inducible elements in the promoter of human plasminogen activator inhibitor-type 1 gene. EMBO J. 17, 3091–3100. Dennler, S., Huet, S., and Gauthier, J. M. (1999). A short amino-acid sequence in MH1 domain is responsible for functional differences between Smad2 and Smad3. Oncogene 18, 1643–1648. Dickson, M. C., Martin, J. S., Cousins, F. M., Kulkarni, A. B., Karlsson, S., and Akhurst, R. J. (1995). Defective haematopoiesis and vasculogenesis in transforming growth factor-웁1 knockout mice. Development 121, 1845–1854. Diebold, R. J., Eis, M. J., Yin, M., Ormsby, I., Boivin, G. P., Darrow, B. J., Saffitz, J. E., and Doetschman, T. (1995). Early-onset multifocal inflammation in the transforming growth factor 웁1-null mouse is lymphocyte mediated. Proc. Natl. Acad. Sci. USA 92, 12215–12219. Duff, E. K., and Clarke, A. R. (1998). Smad4 (DPC4)—A potent tumour suppressor? Br. J. Cancer 78, 1615–1619. Eckner, R., Ewen, M. E., Newsome, D., Gerdes, M., DeCaprio, J. A., Lawrence, J. B., and Livingston, D. M. (1994). Molecular cloning and functional analysis of the adenovirus E1A-associated 300-kD protein (p300) reveals a protein with properties of a transcriptional adaptor. Genes Dev. 8, 869–884. Eppert, K., Scherer, S. W., Ozcelik, H., Pirone, R., Hoodless, P., Kim, H., Tsui, L. C., Bapat, B., Gallinger, S., Andrulis, I. L., Thomsen, G. H., Wrana, J. L., and Attisano, L. (1996). MADR2 maps to 18q21 and encodes a TGF웁-regulated MAD-related protein that is functionally mutated in colorectal carcinoma. Cell 86, 543–552. Feng, X.-H., and Derynck, R. (1997). A kinase subdomain of transforming growth factor-웁 (TGF-웁) type I receptor determines the TGF-웁 intracellular signaling specificity. EMBO J. 16, 3912–3923. Feng, X.-H., Zhang, Y., Wu, R. Y., and Derynck, R. (1998). The tumor suppressor Smad4/ DPC4 and transcriptional adaptor CBP/p300 are coactivators for Smad3 in TGF-웁induced transcriptional activation. Genes Dev. 12, 2153–2163. Frey, R. S., and Mulder, K. M. (1997). Involvement of extracellular signal-regulated kinase 2 and stress-activated protein kinase/Jun N-terminal kinase activation by transforming growth factor 웁 in the negative growth control of breast cancer cells. Cancer Res. 57, 628–633. Fujii, M., Takeda, K., Imamura, T., Aoki, H., Sampath, T. K., Enomoto, S., Kawabata, M., Kato, M., Ichijo, H., and Miyazono, K. (1999). Roles of bone morphogenetic protein type

148

MIYAZONO et al.

I receptors and Smad proteins in osteoblast and chondroblast differentiation. Mol. Biol. Cell 10, 3801–3813. Giles, R. H., Peters, D. J., and Breuning, M. H. (1998). Conjunction dysfunction: CBP/ p300 in human disease. Trends Genet. 14, 178–183. Goto, D., Yagi, K., Inoue, H., Iwamoto, I., Kawabata, M., Miyazono, K., and Kato, M. (1998). A single missense mutant of Smad3 inhibits activation of both Smad2 and Smad3, and has a dominant negative effect on TGF-웁 signals. FEBS Lett. 430, 201–204. Grieder, N. C., Nellen, D., Burke, R., Basler, K., and Affolter, M. (1995). Schnurri is required for Drosophila Dpp signaling and encodes a zinc finger protein similar to the mammalian transcription factor PRDII-BF1. Cell 81, 791–800. Gu, Z., Nomura, M., Simpson, B. B., Lei, H., Feijen, A., van den Eijnden-van Raaij, J., Donahoe, P. K., and Li, E. (1998). The type I activin receptor ActRIB is required for egg cylinder organization and gastrulation in the mouse. Genes Dev. 12, 844–857. Hahn, S. A., Schutte, M., Hoque, A. T. M. S., Moskaluk, C. A., da Costa, L. T., Rozenblum, E., Weinstein, C. L., Fischer, A., Yeo, C. J., Hruban, R. H., and Kern, S. E. (1996). DPC4, a candidate tumor suppressor gene at human chromosome 18q21.1. Science 271, 350–353. Hahn, S. A., Bartsch, D., Schroers, A., Galehdari, H., Becker, M., Ramaswamy, A., SchwarteWaldhoff, I., Maschek, H., and Schmiegel, W. (1998). Mutations of the DPC4/Smad4 gene in biliary tract carcinoma. Cancer Res. 58, 1124–1126. Hanai, J.-i., Chen, L. F., Kanno, T., Ohtari-Fujita, N., Kim, W. Y., Imamura, T., Ishidou, Y., Fukuchi, M., Shi, M.-J., Stavnezer, J., Kawabata, M., Miyazono, K., and Ito, Y. (1999). Interaction and functional cooperation of PEBP2/CBF with Smads. J. Biol. Chem. 274, 31577–31582. Hartsough, M. T., and Mulder, K. M. (1995). Transforming growth factor 웁 activation of p44mapk in proliferating cultures of epithelial cells. J. Biol. Chem. 270, 7117–7124. Hata, A., Lo, R. S., Wotton, D., Lagna, G., and Massague´, J. (1997). Mutations increasing autoinhibition inactivate tumour suppressors Smad2 and Smad4. Nature 388, 82–87. Hata, A., Lagna, G., Massague´, J., and Hemmati-Brivanlou, A. (1998). Smad6 inhibits BMP/ Smad1 signaling by specifically competing with the Smad4 tumor suppressor. Genes Dev. 12, 186–197. Hayashi, H., Abdollah, S., Qiu, Y., Cai, J., Xu, Y. Y., Grinnell, B. W., Richardson, M. A., Topper, J. N., Gimbrone, M. A., Jr., Wrana, J. L., and Falb, D. (1997). The MAD-related protein Smad7 associates with the TGF웁 receptor and functions as an antagonist of TGF웁 signaling. Cell 89, 1165–1173. Hejlik, D. P., Kottickal, L. V., Liang, H., Fairman, J., Davis, T., Janecki, T., Sexton, D., Perry, W., Tavtigian, S. V., Teng, D. H., and Nagarajan, L. (1997). Localization of SMAD5 and its evaluation as a candidate myeloid tumor suppressor. Cancer Res. 57, 3779–3783. Heldin, C.-H., Miyazono, K., and ten Dijke, P. (1997). TGF-웁 signalling from cell membrane to nucleus through SMAD proteins. Nature 390, 465–471. Hocevar, B. A., Brown, T. L., and Howe, P. H. (1999). TGF-웁 induces fibronectin synthesis through a c-Jun N-terminal kinase-dependent, Smad4-independent pathway. EMBO J. 18, 1345–1356. Hoodless, P. A., Tsukazaki, T., Nishimatsu, S., Attisano, L., Wrana, J. L., and Thomsen, G. H. (1999). Dominant-negative Smad2 mutants inhibit activin/Vg1 signaling and disrupt axis formation in Xenopus. Dev. Biol. 207, 364–379. Howe, J. R., Roth, S., Ringold, J. C., Summers, R. W., Jarvinen, H. J., Sistonen, P., Tomlinson, I. P., Houlston, R. S., Bevan, S., Mitros, F. A., Stone, E. M., and Aaltonen, L. A. (1998). Mutations in the SMAD4/DPC4 gene in juvenile polyposis. Science 280, 1086–1088. Howell, M., Itoh, F., Pierreux, C. E., Valgeirsdottir, S., Itoh, S., ten Dijke, P., and Hill, C. S. (1999). Xenopus Smad4웁 is the Co-Smad component of developmentally regulated

TGF-웁 SIGNALING BY SMAD PROTEINS

149

transcription factor complexes responsible for induction of early mesodermal genes. Dev. Biol. 214, 354–369. Hua, X., Liu, X., Ansari, D. O., and Lodish, H. F. (1998). Synergistic cooperation of TFE3 and Smad proteins in TGF-웁-induced transcription of the plasminogen activator inhibitor1 gene. Genes Dev. 12, 3084–3095. Hudson, J. B., Podos, S. D., Keith, K., Simpson, S. L., and Ferguson, E. L. (1998). The Drosophila Medea gene is required downstream of dpp and encodes a functional homolog of human Smad4. Development 125, 1407–1420. Hunt, K. K., Fleming, J. B., Abramian, A., Zhang, L., Evans, D. B., and Chiao, P. J. (1998). Overexpression of the tumor suppressor gene Smad4/DPC4 induces p21waf1 expression and growth inhibition in human carcinoma cells. Cancer Res. 58, 5656–5661. Huse, M., Chen, Y. G., Massague´, J., and Kuriyan, J. (1999). Crystal structure of the cytoplasmic domain of the type I TGF웁 receptor in complex with FKBP12. Cell 96, 425–436. Iavarone, A., and Massague´, J. (1997). Repression of the CDK activator Cdc25A and cell-cycle arrest by cytokine TGF-웁 in cells lacking the CDK inhibitor p15. Nature 387, 417–422. Iavarone, A., and Massague´, J. (1999). E2F and histone deacetylase mediate transforming growth factor 웁 repression of cdc25A during keratinocyte cell cycle arrest. Mol. Cell. Biol. 19, 916–922. Imamura, T., Takase, M., Nishihara, A., Oeda, E., Hanai, J.-i., Kawabata, M., and Miyazono, K. (1997). Smad6 inhibits signalling by the TGF-웁 superfamily. Nature 389, 622–626. Inoue, H., Imamura, T., Ishidou, Y., Takase, M., Udagawa, Y., Oka, Y., Tsuneizumi, K., Tabata, T., Miyazono, K., and Kawabata, M. (1998). Interplay of signal mediators of decapentaplegic (Dpp): Molecular characterization of mothers against dpp, Medea, and daughters against dpp. Mol. Biol. Cell 9, 2145–2156. Ishisaki, A., Yamato, K., Nakao, A., Nonaka, K., Ohguchi, M., ten Dijke, P., and Nishihara, T. (1998). Smad7 is an activin-inducible inhibitor of activin-induced growth arrest and apoptosis in mouse B cells. J. Biol. Chem. 273, 24293–24296. Ishisaki, A., Yamato, K., Hashimoto, S., Nakao, A., Tamaki, K., Nonaka, K., ten Dijke, P., Sugino, H., and Nishihara, T. (1999). Differential inhibition of Smad6 and Smad7 on bone morphogenetic protein- and activin-mediated growth arrest and apoptosis in B cells. J. Biol. Chem. 274, 13637–13642. Ito, Y., and Bae, S.-C. (1997). The Runt domain transcription factor, PEBP2/CBF, and its involvement in human leukemia. In ‘‘Oncogenes as Transcriptional Regulators, Vol. 2’’ (M. Yaniv and J. Ghysdael, eds.), pp. 107–132. Birka¨user Verlag, Basel, Switzerland. Itoh, S., Landstro¨m, M., Hermansson, A., Itoh, F., Heldin, C.-H., Heldin, N.-E., and ten Dijke, P. (1998). Transforming growth factor 웁1 induces nuclear export of inhibitory Smad7. J. Biol. Chem. 273, 29195–29201. Janknecht, R., Wells, N. J., and Hunter, T. (1998). TGF-웁-stimulated cooperation of Smad proteins with the coactivators CBP/p300. Genes Dev. 12, 2114–2119. Jazwinska, A., Kirov, N., Wieschaus, E., Roth, S., and Rushlow, C. (1999). The Drosophila gene brinker reveals a novel mechanism of Dpp target gene regulation. Cell 96, 563–573. Jonk, L. J., Itoh, S., Heldin, C.-H., ten Dijke, P., and Kruijer, W. (1998). Identification and functional characterization of a Smad binding element (SBE) in the JunB promoter that acts as a transforming growth factor-웁, activin, and bone morphogenetic protein-inducible enhancer. J. Biol. Chem. 273, 21145–21152. Jonson, T., Gorunova, L., Dawiskiba, S., Andren-Sandberg, A., Stenman, G., ten Dijke, P., Johansson, B., and Hoglund, M. (1999). Molecular analyses of the 15q and 18q SMAD genes in pancreatic cancer. Genes Chromosomes Cancer 24, 62–71.

150

MIYAZONO et al.

Kawabata, M., Imamura, T., and Miyazono, K. (1998a). Signal transduction by bone morphogenetic proteins. Cytokine Growth Factor Rev. 9, 49–61. Kawabata, M., Inoue, H., Hanyu, A., Imamura, T., and Miyazono, K. (1998b). Smad proteins exist as monomers in vivo and undergo homo- and hetero-oligomerization upon activation by serine/threonine kinase receptors. EMBO J. 17, 4056–4065. Kim, J., Johnson, K., Chen, H. J., Carroll, S., and Laughon, A. (1997). Drosophila Mad binds to DNA and directly mediates activation of vestigial by Decapentaplegic. Nature 388, 304–308. Kingsley, D. M. (1994). The TGF-웁 superfamily: New members, new receptors, and new genetic tests of function in different organisms. Genes Dev. 8, 133–146. Kleeff, J., Maruyama, H., Friess, H., Buchler, M. W., Falb, D., and Korc, M. (1999). Smad6 suppresses TGF-웁-induced growth inhibition in COLO-357 pancreatic cancer cells and is overexpressed in pancreatic cancer. Biochem. Biophys. Res. Commun. 255, 268–273. Kretzschmar, M., Liu, F., Hata, A., Doody, J., and Massague´, J. (1997a). The TGF-웁 family mediator Smad1 is phosphorylated directly and activated functionally by the BMP receptor kinase. Genes Dev. 11, 984–995. Kretzschmar, M., Doody, J., and Massague´, J. (1997b). Opposing BMP and EGF signalling pathways converge on the TGF-웁 family mediator Smad1. Nature 389, 618–622. Kretzschmar, M., Doody, J., Timokhina, I., and Massague´, J. (1999). A mechanism of repression of TGF웁/Smad signaling by oncogenic Ras. Genes Dev. 13, 804–816. Krishna, S., Maduzia, L. L., and Padgett, R. W. (1999). Specificity of TGF웁 signaling is conferred by distinct type I receptors and their associated SMAD proteins in Caenorhabditis elegans. Development 126, 251–260. Kulkarni, A. B., Huh, C. G., Becker, D., Geiser, A., Lyght, M., Flanders, K. C., Roberts, A. B., Sporn, M. B., Ward, J. M., and Karlsson, S. (1993). Transforming growth factor 웁1 null mutation in mice causes excessive inflammatory response and early death. Proc. Natl. Acad. Sci. USA 90, 770–774. Kurokawa, M., Mitani, K., Irie, K., Matsuyama, T., Takahashi, T., Chiba, S., Yazaki, Y., Matsumoto, K., and Hirai, H. (1998a). The oncoprotein Evi-1 represses TGF-웁 signalling by inhibiting Smad3. Nature 394, 92–96. Kurokawa, M., Mitani, K., Imai, Y., Ogawa, S., Yazaki, Y., and Hirai, H. (1998b). The t(3;21) fusion product, AML1/Evi-1, interacts with Smad3 and blocks transforming growth factor웁-mediated growth inhibition of myeloid cells. Blood 92, 4003–4012. Labbe´, E., Silvestri, C., Hoodless, P. A., Wrana, J. L., and Attisano, L. (1998). Smad2 and Smad3 positively and negatively regulate TGF웁-dependent transcription through the forkhead DNA-binding protein FAST2. Mol. Cell 2, 109–120. Lagna, G., and Hemmati-Brivanlou, A. (1999). A molecular basis for Smad specificity. Dev. Dyn. 214, 269–277. Le Dai, J., Bansal, R. K., and Kern, S. E. (1999). G1 cell cycle arrest and apoptosis induction by nuclear Smad4/Dpc4: Phenotypes reversed by a tumorigenic mutation. Proc. Natl. Acad. Sci. USA 96, 1427–1432. Lebrun, J. J., Takabe, K., Chen, Y., and Vale, W. (1999). Roles of pathway-specific and inhibitory Smads in activin receptor signaling. Mol. Endocrinol. 13, 15–23. LeSueur, J. A., and Graff, J. M. (1999). Spemann organizer activity of Smad10. Development 126, 137–146. Li, J. M., Nichols, M. A., Chandrasekharan, S., Xiong, Y., and Wang, X.-F. (1995). Transforming growth factor 웁 activates the promoter of cyclin-dependent kinase inhibitor p15INK4B through an Sp1 consensus site. J. Biol. Chem. 270, 26750–26753. Li, M., Li, J., Hoodless, P. A., Tzukazaki, T., Wrana, J. L., Attisano, L., and Tsang, B. K. (1997). Mothers against decapentaplegic-related protein 2 expression in avian granulosa

TGF-웁 SIGNALING BY SMAD PROTEINS

151

cells is up-regulated by transforming growth factor 웁 during ovarian follicular development. Endocrinology 138, 3659–3665. Li, J., Tsuji, K., Komori, T., Miyazono, K., Wrana, J. L., Ito, Y., Nifuji, A., and Noda, M. (1998). Smad2 overexpression enhances Smad4 gene expression and suppresses CBFA1 gene expression in osteoblastic osteosarcoma ROS17/2.8 cells and primary rat calvaria cells. J. Biol. Chem. 273, 31009–31015. Liberati, N. T., Datto, M. B., Frederick, J. P., Shen, X., Wong, C., Rougier-Chapman, E. M., and Wang, X.-F. (1999). Smads bind directly to the Jun family of AP-1 transcription factors. Proc. Natl. Acad. Sci. USA 96, 4844–4849. Liu, F., Hata, A., Baker, J. C., Doody, J., Ca´rcamo, J., Harland, R. M., and Massague´, J. (1996). A human Mad protein acting as a BMP-regulated transcriptional activator. Nature 381, 620–623. Liu, F., Pouponnot, C., and Massague´, J. (1997). Dual role of the Smad4/DPC4 tumor suppressor in TGF웁-inducible transcriptional complexes. Genes Dev. 11, 3157–3167. Liu, F., Massague´, J., and Ruiz i Altaba, A. (1998). Carboxy-terminally truncated Gli3 proteins associate with Smads. Nat. Genet. 20, 325–326. Liu, X., Yue, J., Frey, R. S., Zhu, Q., and Mulder, K. M. (1998). Transforming growth factor 웁 signaling through Smad1 in human breast cancer cells. Cancer Res. 58, 4752–4757. Liu, B., Dou, C. L., Prabhu, L., and Lai, E. (1999). FAST-2 is a mammalian winged-helix protein which mediates transforming growth factor 웁 signals. Mol. Cell. Biol. 19, 424–430. Lo, R. S., Chen, Y. G., Shi, Y., Pavletich, N. P., and Massague´, J. (1998). The L3 loop: A structural motif determining specific interactions between SMAD proteins and TGF-웁 receptors. EMBO J. 17, 996–1005. Lux, A., Attisano, L., and Marchuk, D. A. (1999). Assignment of transforming growth factor 웁1 and 웁3 and a third new ligand to the type I receptor ALK-1. J. Biol. Chem. 274, 9984–9992. Lyons, I., Parsons, L. M., Hartley, L., Li, R., Andrews, J. E., Robb, L., and Harvey, R. P. (1995). Myogenic and morphogenetic defects in the heart tubes of murine embryos lacking the homeo box gene Nkx2-5. Genes Dev. 9, 1654–1666. Macı´as-Silva, M., Abdollah, S., Hoodless, P. A., Pirone, R., Attisano, L., and Wrana, J. L. (1996). MADR2 is a substrate of the TGF웁 receptor and its phosphorylation is required for nuclear accumulation and signaling. Cell 87, 1215–1224. Macı´as-Silva, M., Hoodless, P. A., Tang, S. J., Buchwald, M., and Wrana, J. L. (1998). Specific activation of Smad1 signaling pathways by the BMP7 type I receptor, ALK2. J. Biol. Chem. 273, 25628–25636. Massague´, J. (1998). TGF-웁 signal transduction. Annu. Rev. Biochem. 67, 753–791. Masuyama, N., Hanafusa, H., Kusakabe, M., Shibuya, H., and Nishida, E. (1999). Identification of two Smad4 proteins in Xenopus: Their common and distinct properties. J. Biol. Chem. 274, 12163–12170. Mathews, L. S. (1994). Activin receptors and cellular signaling by the receptor serine kinase family. Endocr. Rev. 15, 310–325. Meersseman, G., Verschueren, K., Nelles, L., Blumenstock, C., Kraft, H., Wuytens, G., Remacle, J., Kozak, C. A., Tylzanowski, P., Niehrs, C., and Huylebroeck, D. (1997). The C-terminal domain of Mad-like signal transducers is sufficient for biological activity in the Xenopus embryo and transcriptional activation. Mech. Dev. 61, 127–140. Minami, M., Kinoshita, N., Kamoshida, Y., Tanimoto, H., and Tabata, T. (1999). brinker is a target of Dpp in Drosophila that negatively regulates Dpp-dependent genes. Nature 398, 242–246. Mishina, Y., Suzuki, A., Ueno, N., and Behringer, R. R. (1995). Bmpr encodes a type I bone morphogenetic protein receptor that is essential for gastrulation during mouse embryogenesis. Genes Dev. 9, 3027–3037.

152

MIYAZONO et al.

Miyazono, K., ten Dijke, P., Ichijo, H., and Heldin, C.-H. (1994). Receptors for transforming growth factor-웁. Adv. Immunol. 55, 181–220. Moustakas, A., and Kardassis, D. (1998). Regulation of the human p21/WAF1/Cip1 promoter in hepatic cells by functional interactions between Sp1 and Smad family members. Proc. Natl. Acad. Sci. USA 95, 6733–6738. Nakao, A., Afrakhte, M., More´n, A., Nakayama, T., Christian, J. L., Heuchel, R., Itoh, S., Kawabata, M., Heldin, N.-E., Heldin, C.-H., and ten Dijke, P. (1997a). Identification of Smad7, a TGF웁-inducible antagonist of TGF-웁 signalling. Nature 389, 631–635. Nakao, A., Imamura, T., Souchelnytskyi, S., Kawabata, M., Ishisaki, A., Oeda, E., Tamaki, K., Hanai, J., Heldin, C.-H., Miyazono, K., and ten Dijke, P. (1997b). TGF-웁 receptormediated signalling through Smad2, Smad3 and Smad4. EMBO J. 16, 5353–5362. Nakashima, K., Yanagisawa, M., Arakawa, H., Kimura, N., Hisatsune, T., Kawabata, M., Miyazono, K., and Taga, T. (1999). Synergistic signaling in fetal brain by STAT3-Smad1 complex bridged by p300. Science 284, 479–482. Nakayama, T., Gardner, H., Berg, L. K., and Christian, J. L. (1998a). Smad6 functions as an intracellular antagonist of some TGF-웁 family members during Xenopus embryogenesis. Genes Cells 3, 387–394. Nakayama, T., Snyder, M. A., Grewal, S. S., Tsuneizumi, K., Tabata, T., and Christian, J. L. (1998b). Xenopus Smad8 acts downstream of BMP-4 to modulate its activity during vertebrate embryonic patterning. Development 125, 857–867. Newfeld, S. J., Chartoff, E. H., Graff, J. M., Melton, D. A., and Gelbart, W. M. (1996). Mothers against dpp encodes a conserved cytoplasmic protein required in DPP/TGF-웁 responsive cells. Development 122, 2099–2108. Ninomiya-Tsuji, J., Kishimoto, K., Hiyama, A., Inoue, J., Cao, Z., and Matsumoto, K. (1999). The kinase TAK1 can activate the NIK-I␬B as well as the MAP kinase cascade in the IL-1 signalling pathway. Nature 398, 252–256. Nishihara, A., Hanai, J.-i., Okamoto, N., Yanagisawa, J., Kato, S., Miyazono, K., and Kawabata, M. (1998). Role of p300, a transcriptional coactivator, in signalling of TGF-웁. Genes Cells 3, 613–623. Nishihara, A., Hanai, J.-i., Imamura, T., Miyazono, K., and Kawabata, M. (1999). E1A inhibits transforming growth factor-웁 signaling through binding to Smad proteins. J. Biol. Chem. 274, 28716–28723. Nishimura, R., Kato, Y., Chen, D., Harris, S. E., Mundy, G. R., and Yoneda, T. (1998). Smad5 and DPC4 are key molecules in mediating BMP-2-induced osteoblastic differentiation of the pluripotent mesenchymal precursor cell line C2C12. J. Biol. Chem. 273, 1872–1879. Nomura, M., and Li, E. (1998). Smad2 role in mesoderm formation, left-right patterning and craniofacial development. Nature 393, 786–790. Oshima, M., Oshima, H., and Taketo, M. M. (1996). TGF-웁 receptor type II deficiency results in defects of yolk sac hematopoiesis and vasculogenesis. Dev. Biol. 179, 297–302. Pardali, E., Xie, X.-Q., Tsapogas, P., Itoh, S., Arvanitidis, K., Heldin, C.-H., ten Dijke, P., Grundstro¨m, T., and Sideras, P. (2000). Smad and AML proteins synergistically confer transforming growth factor 웁1 responsiveness to human germ-line IgA genes. J. Biol. Chem. 275, 3552–3560. Patterson, G. I., Koweek, A., Wong, A., Liu, Y., and Ruvkun, G. (1997). The DAF-3 Smad protein antagonizes TGF-웁-related receptor signaling in the Caenorhabditis elegans dauer pathway. Genes Dev. 11, 2679–2690. Persson, U., Izumi, H., Souchelnytskyi, S., Itoh, S., Grimsby, S., Engstro¨m, U., Heldin, C.-H., Funa, K., and ten Dijke, P. (1998). The L45 loop in type I receptors for TGF-웁 family members is a critical determinant in specifying Smad isoform activation. FEBS Lett. 434, 83–87.

TGF-웁 SIGNALING BY SMAD PROTEINS

153

Pouponnot, C., Jayaraman, L., and Massague´, J. (1998). Physical and functional interaction of SMADs and p300/CBP. J. Biol. Chem. 273, 22865–22868. Raftery, L. A., Twombly, V., Wharton, K., and Gelbart, W. M. (1995). Genetic screens to identify elements of the decapentaplegic signaling pathway in Drosophila. Genetics 139, 241–254. Riggins, G. J., Thiagalingam, S., Rozenblum, E., Weinstein, C. L., Kern, S. E., Hamilton, S. R., Willson, J. K., Markowitz, S. D., Kinzler, K. W., and Vogelstein, B. (1996). Madrelated genes in the human. Nat. Genet. 13, 347–349. Roberts, A. B. and Sporn, M. B. (1990). The transforming growth factor-웁s. In ‘‘Peptide Growth Factors and Their Receptors, Part I’’ (M. B. Sporn and A. B. Roberts, eds.), pp. 419–472. Springer-Verlag, Berlin, Germany. Sakurai, H., Miyoshi, H., Toriumi, W., and Sugita, T. (1999). Functional interactions of transforming growth factor 웁-activated kinase 1 with I␬B kinases to stimulate NF-␬B activation. J. Biol. Chem. 274, 10641–10648. Sano, Y., Harada, J., Tashiro, S., Gotoh-Mandeville, R., Maekawa, T., and Ishii, S. (1999). ATF-2 is a common nuclear target of Smad and TAK1 pathways in transforming growth factor-웁 signaling. J. Biol. Chem. 274, 8949–8957. Savage, C., Das, P., Finelli, A. L., Townsend, S. R., Sun, C. Y., Baird, S. E., and Padgett, R. W. (1996). Caenorhabditis elegans genes sma-2, sma-3, and sma-4 define a conserved family of transforming growth factor 웁 pathway components. Proc. Natl. Acad. Sci. USA 93, 790–794. Schultheiss, T. M., Burch, J. B., and Lassar, A. B. (1997). A role for bone morphogenetic proteins in the induction of cardiac myogenesis. Genes Dev. 11, 451–462. Schutte, M., Hruban, R. H., Hedrick, L., Cho, K. R., Nadasdy, G. M., Weinstein, C. L., Bova, G. S., Isaacs, W. B., Cairns, P., Nawroz, H., Sidransky, D., Casero, R. A., Jr., Meltzer, P. S., Hahn, S. A., and Kern, S. E. (1996). DPC4 gene in various tumor types. Cancer Res. 56, 2527–2530. Sekelsky, J. J., Newfeld, S. J., Raftery, L. A., Chartoff, E. H., and Gelbart, W. M. (1995). Genetic characterization and cloning of Mothers against dpp, a gene required for decapentaplegic function in Drosophila melanogaster. Genetics 139, 1347–1358. Shen, X., Hu, P. P., Liberati, N. T., Datto, M. B., Frederick, J. P., and Wang, X.-F. (1998). TGF-웁-induced phosphorylation of Smad3 regulates its interaction with coactivator p300/ CREB-binding protein. Mol. Biol. Cell 9, 3309–3319. Shi, Y., Hata, A., Lo, R. S., Massague´, J., and Pavletich, N. P. (1997). A structural basis for mutational inactivation of the tumour suppressor Smad4. Nature 388, 87–93. Shi, M. J., and Stavnezer, J. (1998). CBF움3 (AML2) is induced by TGF-웁1 to bind and activate the mouse germline Ig 움 promoter. J. Immunol. 161, 6751–6760. Shi, Y., Wang, Y. F., Jayaraman, L., Yang, H., Massague´, J., and Pavletich, N. P. (1998). Crystal structure of a Smad MH1 domain bound to DNA: insights on DNA binding in TGF-웁 signaling. Cell 94, 585–594. Shi, X., Yang, X., Chen, D., Chang, D., and Cao, X. (1999). Smad1 interacts with homeobox DNA-binding proteins in bone morphogenetic protein signaling. J. Biol. Chem. 274, 13711–13717. Shimizu, A., Kato, M., Nakao, A., Imamura, T., ten Dijke, P., Heldin, C.-H., Kawabata, M., Shimada, S., and Miyazono, K. (1998). Identification of receptors and Smad proteins involved in activin signalling in a human epidermal keratinocyte cell line. Genes Cells 3, 125–134. Shioda, T., Lechleider, R. J., Dunwoodie, S. L., Li, H., Yahata, T., de Caestecker, M. P., Fenner, M. H., Roberts, A. B., and Isselbacher, K. J. (1998). Transcriptional activating

154

MIYAZONO et al.

activity of Smad4: Roles of SMAD hetero-oligomerization and enhancement by an associating transactivator. Proc. Natl. Acad. Sci. USA 95, 9785–9790. Shull, M. M., Ormsby, I., Kier, A. B., Pawlowski, S., Diebold, R. J., Yin, M., Allen, R., Sidman, C., Proetzel, G., Calvin, D., Annunziata, N., and Doetschman, T. (1992). Targeted disruption of the mouse transforming growth factor-웁1 gene results in multifocal inflammatory disease. Nature 359, 693–699. Sirard, C., de la Pompa, J. L., Elia, A., Itie, A., Mirtsos, C., Cheung, A., Hahn, S., Wakeham, A., Schwartz, L., Kern, S. E., Rossant, J., and Mak, T. W. (1998). The tumor suppressor gene Smad4/Dpc4 is required for gastrulation and later for anterior development of the mouse embryo. Genes Dev. 12, 107–119. Song, C. Z., Siok, T. E., and Gelehrter, T. D. (1998). Smad4/DPC4 and Smad3 mediate transforming growth factor-웁 (TGF-웁) signaling through direct binding to a novel TGF웁-responsive element in the human plasminogen activator inhibitor-1 promoter. J. Biol. Chem. 273, 29287–29290. Sonoda, E., Matsumoto, R., Hitoshi, Y., Ishii, T., Sugimoto, M., Araki, S., Tominaga, A., Yamaguchi, N., and Takatsu, K. (1989). Transforming growth factor 웁 induces IgA production and acts additively with interleukin 5 for IgA production. J. Exp. Med. 170, 1415–1420. Souchelnytskyi, S., Tamaki, K., Engstro¨m, U., Wernstedt, C., ten Dijke, P., and Heldin, C.-H. (1997). Phosphorylation of Ser465 and Ser467 in the C terminus of Smad2 mediates interaction with Smad4 and is required for transforming growth factor-웁 signaling. J. Biol. Chem. 272, 28107–28115. Souchelnytskyi, S., Nakayama, T., Nakao, A., More´n, A., Heldin, C.-H., Christian, J. L., and ten Dijke, P. (1998). Physical and functional interaction of murine and Xenopus Smad7 with bone morphogenetic protein receptors and transforming growth factor-웁 receptors. J. Biol. Chem. 273, 25364–25370. Staehling-Hampton, K., Laughon, A. S., and Hoffmann, F. M. (1995). A Drosophila protein related to the human zinc finger transcription factor PRDII/MBPI/HIV-EP1 is required for dpp signaling. Development 121, 3393–3403. Stroschein, S. L., Wang, W., and Luo, K. (1999). Cooperative binding of Smad proteins to two adjacent DNA elements in the plasminogen activator inhibitor-1 promoter mediates transforming growth factor 웁-induced Smad-dependent transcriptional activation. J. Biol. Chem. 274, 9431–9441. Struhl, K. (1998). Histone acetylation and transcriptional regulatory mechanisms. Genes Dev. 12, 599–606. Takagi, Y., Koumura, H., Futamura, M., Aoki, S., Yamaguchi, K., Kida, H., Tanemura, H., Shimokawa, K., and Saji, S. (1998). Somatic alterations of the SMAD-2 gene in human colorectal cancers. Br. J. Cancer 78, 1152–1155. Takaku, K., Oshima, M., Miyoshi, H., Matsui, M., Seldin, M. F., and Taketo, M. M. (1998). Intestinal tumorigenesis in compound mutant mice of both Dpc4 (Smad4) and Apc genes. Cell 92, 645–656. Takase, M., Imamura, T., Sampath, T. K., Takeda, K., Ichijo, H., Miyazono, K., and Kawabata, M. (1998). Induction of Smad6 mRNA by bone morphogenetic proteins. Biochem. Biophys. Res. Commun. 244, 26–29. Tamaki, K., Souchelnytskyi, S., Itoh, S., Nakao, A., Sampath, K., Heldin, C.-H., and ten Dijke, P. (1998). Intracellular signaling of osteogenic protein-1 through Smad5 activation. J. Cell. Physiol. 177, 355–363. Tang S. J., Hoodless, P. A., Lu, Z., Breitman, M. L., McInnes, R. R., Wrana, J. L., and Buchwald, M. (1998). The Tlx-2 homeobox gene is a downstream target of BMP signalling and is required for mouse mesoderm development. Development 125, 1877–1887.

TGF-웁 SIGNALING BY SMAD PROTEINS

155

ten Dijke, P., Yamashita, H., Sampath, T. K., Reddi, A. H., Estevez, M., Riddle, D. L., Ichijo, H., Heldin, C.-H., and Miyazono, K. (1994a). Identification of type I receptors for osteogenic protein-1 and bone morphogenetic protein-4. J. Biol. Chem. 269, 16985– 16988. ten Dijke, P., Yamashita, H., Ichijo, H., Franze´n, P., Laiho, M., Miyazono, K., and Heldin, C.-H. (1994b). Characterization of type I receptors for transforming growth factor-웁 and activin. Science 264, 101–104. Thatcher, J. D., Haun, C., and Okkema, P. G. (1999). The DAF-3 Smad binds DNA and represses gene expression in the Caenorhabditis elegans pharynx. Development 126, 97–107. Thiagalingam, S., Lengauer, C., Leach, F. S., Schutte, M., Hahn, S. A., Overhauser, J., Willson, J. K., Markowitz, S., Hamilton, S. R., Kern, S. E., Kinzler, K. W., and Vogelstein, B. (1996). Evaluation of candidate tumour suppressor genes on chromosome 18 in colorectal cancers. Nat. Genet. 13, 343–346. Topper, J. N., Cai, J., Qiu, Y., Anderson, K. R., Xu, Y. Y., Deeds, J. D., Feeley, R., Gimeno, C. J., Woolf, E. A., Tayber, O., Mays, G. G., Sampson, B. A., Schoen, F. J., Gimbrone, M. A., Jr., and Falb, D. (1997). Vascular MADs: two novel MAD-related genes selectively inducible by flow in human vascular endothelium. Proc. Natl. Acad. Sci. USA 94, 9314– 9319. Topper, J. N., DiChiara, M. R., Brown, J. D., Williams, A. J., Falb, D., Collins, T., and Gimbrone, M. A., Jr. (1998). CREB binding protein is a required coactivator for Smaddependent, transforming growth factor 웁 transcriptional responses in endothelial cells. Proc. Natl. Acad. Sci. USA 95, 9506–9511. Travers, A. (1999). How to put a HAT on the histones. Curr. Biol. 9, R23–R25. Tsukazaki, T., Chiang, T. A., Davison, A. F., Attisano, L., and Wrana, J. L. (1998). SARA, a FYVE domain protein that recruits Smad2 to the TGF웁 receptor. Cell 11, 779–791. Tsuneizumi, K., Nakayama, T., Kamoshida, Y., Kornberg, T. B., Christian, J. L., and Tabata, T. (1997). Daughters against dpp modulates dpp organizing activity in Drosophila wing development. Nature 389, 627–631. Uchida, K., Nagatake, M., Osada, H., Yatabe, Y., Kondo, M., Mitsudomi, T., Masuda, A., and Takahashi, T. (1996). Somatic in vivo alterations of the JV18-1 gene at 18q21 in human lung cancers. Cancer Res. 56, 5583–5585. Ulloa, L., Doody, J., and Massague´, J. (1999). Inhibition of transforming growth factor-웁/ SMAD signalling by the interferon-웂/STAT pathway. Nature 397, 710–713. Verschueren, K., Remacle, J. E., Collart, C., Kraft, H., Baker, B. S., Tylzanowski, P., Nelles, L., Wuytens, G., Su, M.-T., Bodmer, R., Smith, J. C., and Huylebroeck, D. (1999). SIP1, a novel zinc finger/homeodomain repressor, interacts with Smad proteins and binds to 5⬘-CACCT sequences in candidate target genes. J. Biol. Chem., 274, 20489–20498. Vindevoghel, L., Lechleider, R. J., Kon, A., de Caestecker, M. P., Uitto, J., Roberts, A. B., and Mauviel, A. (1998). SMAD3/4-dependent transcriptional activation of the human type VII collagen gene (COL7A1) promoter by transforming growth factor 웁. Proc. Natl. Acad. Sci. USA 95, 14769–14774. Waldrip, W. R., Bikoff, E. K., Hoodless, P. A., Wrana, J. L., and Robertson, E. J. (1998). Smad2 signaling in extraembryonic tissues determines anterior–posterior polarity of the early mouse embryo. Cell 92, 797–808. Waltzer, L., and Bienz, M. (1999). A function of CBP as a transcriptional co-activator during Dpp signalling. EMBO J. 18, 1630–1641. Wang, T., Donahoe, P. K., and Zervos, A. S. (1994). Specific interaction of type I receptors of the TGF-웁 family with the immunophilin FKBP-12. Science 265, 674–676.

156

MIYAZONO et al.

Wang, T., Li, B. Y., Danielson, P. D., Shah, P. C., Rockwell, S., Lechleider, R. J., Martin, J., Manganaro, T., and Donahoe, P. K. (1996). The immunophilin FKBP12 functions as a common inhibitor of the TGF웁 family type I receptors. Cell 86, 435–444. Watanabe, T. K., Suzuki, M., Omori, Y., Hishigaki, H., Horie, M., Kanemoto, N., Fujiwara, T., Nakamura, Y., and Takahashi, E. (1997). Cloning and characterization of a novel member of the human Mad gene family (MADH6). Genomics 42, 446–451. Weinstein, M., Yang, X., Li, C., Xu, X., Gotay, J., and Deng, C. X. (1998). Failure of egg cylinder elongation and mesoderm induction in mouse embryos lacking the tumor suppressor smad2. Proc. Natl. Acad. Sci. USA 95, 9378–9383. Winnier, G., Blessing, M., Labosky, P. A., and Hogan, B. L. (1995). Bone morphogenetic protein-4 is required for mesoderm formation and patterning in the mouse. Genes Dev. 9, 2105–2116. Wisotzkey, R. G., Mehra, A., Sutherland, D. J., Dobens, L. L., Liu, X., Dohrmann, C., Attisano, L., and Raftery, L. A. (1998). Medea is a Drosophila Smad4 homolog that is differentially required to potentiate DPP responses. Development 125, 1433–1445. Wong, C., Rougier-Chapman, E. M., Frederick, J. P., Datto, M. B., Liberati, N. T., Li, J. M., and Wang, X.-F. (1999). Smad3-Smad4 and AP-1 complexes synergize in transcriptional activation of the c-Jun promoter by transforming growth factor 웁. Mol. Cell. Biol. 19, 1821– 1830. Wotton, D., Lo, R. S., Lee, S., and Massague´, J. (1999). A Smad transcriptional corepressor. Cell 97, 29–39. Xie, X. Q., Pardali, E., Holm, M., Sideras, P., and Grundstrom, T. (1999). AML and Ets proteins regulate the I움 1 germ-line promoter. Eur. J. Immunol. 29, 488–498. Xu, X., Yin, Z., Hudson, J. B., Ferguson, E. L., and Frasch, M. (1998). Smad proteins act in combination with synergistic and antagonistic regulators to target Dpp responses to the Drosophila mesoderm. Genes Dev. 12, 2354–2370. Yagi, K., Goto, D., Hamamoto, T., Takenoshita, S., Kato, M., and Miyazono, K. (1999). Alternatively spliced variant of Smad2 lacking exon 3: Comparison with wild-type Smad2 and Smad3. J. Biol. Chem. 274, 703–709. Yamaguchi, K., Shirakabe, K., Shibuya, H., Irie, K., Oishi, I., Ueno, N., Taniguchi, T., Nishida, E., and Matsumoto, K. (1995). Identification of a member of the MAPKKK family as a potential mediator of TGF-웁 signal transduction. Science 270, 2008–2011. Yamamoto, N., Akiyama, S., Katagiri, T., Namiki, M., Kurokawa, T., and Suda, T. (1997). Smad1 and Smad5 act downstream of intracellular signalings of BMP-2 that inhibits myogenic differentiation and induces osteoblast differentiation in C2C12 myoblasts. Biochem. Biophys. Res. Commun. 238, 574–580. Yanagi, Y., Suzawa, M., Kawabata, M., Miyazono, K., Yanagisawa, J., and Kato, S. (1999). Positive and negative modulation of vitamin D receptor function by TGF-웁 signaling through Smad proteins. J. Biol. Chem. 274, 12971–12974. Yanagisawa, K., Osada, H., Masuda, A., Kondo, M., Saito, T., Yatabe, Y., Takagi, K., and Takahashi, T. (1998). Induction of apoptosis by Smad3 and down-regulation of Smad3 expression in response to TGF-웁 in human normal lung epithelial cells. Oncogene 17, 1743–1747. Yanagisawa, J., Yanagi, Y., Masuhiro, Y., Suzawa, M., Watanabe, M., Kashiwagi, K., Toriyabe, T., Kawabata, M., Miyazono, K., and Kato, S. (1999). Convergence of transforming growth factor-웁 and vitamin D signaling pathways on SMAD transcriptional coactivators. Science 283, 1317–1321. Yang, X., Li, C., Xu, X., and Deng, C. (1998). The tumor suppressor SMAD4/DPC4 is essential for epiblast proliferation and mesoderm induction in mice. Proc. Natl. Acad. Sci. USA 95, 3667–3672.

TGF-웁 SIGNALING BY SMAD PROTEINS

157

Yang, X., Castilla, L. H., Xu, X., Li, C., Gotay, J., Weinstein, M., Liu, P. P., and Deng, C. X. (1999a). Angiogenesis defects and mesenchymal apoptosis in mice lacking SMAD5. Development 126, 1571–1580. Yang, X., Letterio, J. J., Lechleider, R. J., Chen, L., Hayman, R., Gu, H., Roberts, A. B., and Deng, C. (1999b). Targeted disruption of SMAD3 results in impaired mucosal immunity and diminished T cell responsiveness to TGF-웁. EMBO J. 18, 1280–1291. Yue, J., Hartsough, M. T., Frey, R. S., Frielle, T., and Mulder, K. M. (1999a). Cloning and expression of a rat Smad1: Regulation by TGF웁 and modulation by the Ras/MEK pathway. J. Cell. Physiol. 178, 387–396. Yue, J., Frey, R. S., and Mulder, K. M. (1999b). Cross-talk between the Smad1 and Ras/ MEK signaling pathways for TGF웁. Oncogene 18, 2033–2037. Zavadil, J., Brezinova, J., Svoboda, P., Zemanova, Z., and Michalova, K. (1997). Smad5, a tumor suppressor candidate at 5q31.1, is hemizygously lost and not mutated in the retained allele in human leukemia cell line HL60. Leukemia 11, 1187–1192. Zawel, L., Le Dai, J., Buckhaults, P., Zhou, S., Kinzler, K. W., Vogelstein, B., and Kern, S. E. (1998). Human Smad3 and Smad4 are sequence-specific transcription activators. Mol. Cell 1, 611–617. Zhang, Y., Musci, T., and Derynck, R. (1997). The tumor suppressor Smad4/DPC4 as a central mediator of Smad function. Curr. Biol. 7, 270–276. Zhang, Y., Feng, X.-H., and Derynck, R. (1998). Smad3 and Smad4 cooperate with c-Jun/ c-Fos to mediate TGF-웁-induced transcription. Nature 394, 909–913. Zhou, S., Buckhaults, P., Zawel, L., Bunz, F., Riggins, G., Le Dai, J., Kern, S. E., Kinzler, K. W., and Vogelstein, B. (1998a). Targeted deletion of Smad4 shows it is required for transforming growth factor 웁 and activin signaling in colorectal cancer cells. Proc. Natl. Acad. Sci. USA 95, 2412–2416. Zhou, S., Zawel, L., Lengauer, C., Kinzler, K. W., and Vogelstein, B. (1998b). Characterization of human FAST-1, a TGF웁 and activin signal transducer. Mol. Cell 2, 121–127. Zhou, G., Lee, S. C., Yao, Z., and Tan, T.-H. (1999). Hematopoietic progenitor kinase 1 is a component of transforming growth factor 웁-induced c-Jun N-terminal kinase signaling cascade. J. Biol. Chem. 274, 13133–13138. Zhu, H., Kavsak, P., Abdollah, S., Wrana, J. L., and Thomsen, G. H. (1999). A SMAD ubiquitin ligase targets the BMP pathway and affects embryonic pattern formation. Nature 400, 687–693. Zhu, Y., Richardson, J. A., Parada, L. F., and Graff, J. M. (1998). Smad3 mutant mice develop metastatic colorectal cancer. Cell 94, 703–714. Zimmerman, C. M., Kariapper, M. S., and Mathews, L. S. (1998). Smad proteins physically interact with calmodulin. J. Biol. Chem. 273, 677–680.

This Page Intentionally Left Blank

ADVANCES IN IMMUNOLOGY, VOL. 75

MHC Class II-Restricted Antigen Processing and Presentation JEAN PIETERS Basel Institute for Immunology, CH-4005 Basel, Switzerland

I. Introduction

Despite the continuous exposure to a wide variety of pathogenic microbes, vertebrate organisms usually do not succumb to infections. This ability to withstand these challenges is due to a two-level defense mechanism: A first level of defense is provided by the innate (inborn) immune system, and a second by the capacity of vertebrates to mount an adaptive immune response. The innate immune system relies on the action of neutrophiles, macrophages, dendritic cells, and natural killer cells to nonspecifically clear the organism from infectious agents. In contrast, the adaptive immune system has evolved as an efficient mechanism to neutralize incoming pathogens through the generation of an antibody-mediated (also termed humoral) as well as a cell-mediated response, highly specific for the microbes that have invaded the organism. Both the antibody- and the cell-mediated response rely on the activation of different sets of T lymphocytes, which, in turn, either directly eliminate the source of infection (in case of the T killer cells [also referred to as CD8⫹ cells due to their expression of the CD8 coreceptor]) or aid in the generation of an antibody response as well as induce the innate immune system to become more active (in case of the T helper cells (also called CD4⫹ cells, as these express the CD4 coreceptor)). Activation of both T cell subsets occurs through the action of a set of molecules encoded in the major histocompatibility complex (MHC), a gene cluster containing various genes encoding proteins involved in immune defense. Whereas T killer cells are activated by MHC class I molecules, activation of T helper cells is induced by MHC class II molecules. MHC class I and class II molecules can activate T cells after binding small peptide fragments derived from foreign material and presenting such peptides at the cell surface for scrutiny by the T cell receptors of T lymphocytes. Within the past two decades, research from a variety of disciplines has contributed to a close to complete understanding of the biology of MHC class I- and class II-mediated antigen presentation. This is largely due to the ability to reconstitute the various steps involved in the generation of MHC class I- or class II-restricted responses in vitro. We have by now also been able to assess the contributions of each of the different molecules that function in antigen presentation though the generation of mice devoid 159

Copyright 䉷 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2776/00 $35.00

160

PIETERS

of these molecules. This chapter reviews the research that has contributed to the elucidation of the MHC class II-presenting pathway, focusing on structure/function of the class II molecules, the various chaperones that play a role in the proper functioning of the generation of a T helper response, and the subcellular organelles involved. Prior to presentation, antigens must be degraded into peptide fragments, and the known mechanisms that are responsible for antigen processing will also be discussed. At the same time, it should be realized that the MHC-restricted activation of the immune system is not the only, and may not even be the most important, component of immunity to infections. In addition to the alreadymentioned innate immunity, the mechanisms of which remain poorly defined, recent research highlights the involvement of other players in host defense. These components include CD1-mediated antigen presentation as well as the 웂/␦ T lymphocytes, whose biology is only partially understood and may be characterized as belonging to both the innate and the adaptive immune system. Furthermore, the more we learn about the mechanisms that activate host immune defense mechanisms, the clearer it becomes that various microbes have evolved mechanisms to circumvent immune recognition, thereby becoming pathogenic. In the final paragraph of this review, a brief description is provided of some of the known mechanisms of subversion of the immune system. II. The MHC Class I and Class II Pathways

Vertebrate organisms are usually challenged by two types of infectious agents. On the one hand, cells can become infected with viruses, which acquire access to the cytosolic compartment of the cell, or alternatively, the host organisms may become infected with bacteria that are present extracellularly and enter the endosomal pathway. As a result of these two distinct sites of pathogen residence, the immune system has evolved two different pathways to bring antigens to the cell surface for recognition by T lymphocytes: whereas the class I molecules sample the cytosolic compartment for infectious agents, the class II molecules bind endosomally derived (and thus those agents having entered from the extracellular milieu) antigens prior to their transport to the cell surface (Fig. 1). Whereas virtually all cell types possess class I molecules, allowing presentation of cytosolically located antigens, only a few specialized cell types (B lymphocytes, dendritic cells, and macrophages) are capable of presenting extracellularly residing antigens by class II molecules to T lymphocytes (Germain, 1994; Wolf and Ploegh, 1995; Pieters, 1997b). MHC molecules do not present intact proteins, but rather peptide fragments to T lymphocytes. Therefore, the cytosollically located as well as

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

161

FIG. 1. The MHC class I and class II pathways. MHC class I molecules acquire antigenic peptides from the cytosol that are degraded by the proteasome and translocated into the endoplasmic reticulum by the TAP complex. The MHC class I–peptide complex is transported through the Golgi complex directly to the cell surface for presentation to CD8⫹ T cells. In contrast, MHC class II molecules acquire antigenic peptides derived from antigens that are internalized in the endocytic pathway. (Modified from Pieters, 1999, by permission of JAI Press/Elsevier Science, Stamford, CT.)

the endosomal antigens must be degraded (Babbitt et al., 1985; Townsend et al., 1985, 1986; Gould et al., 1989). Cytosollically located antigens can be recognized as foreign and readily destroyed by the ubiquitin/proteosome system (Tanaka et al., 1997; Varshavsky, 1997; Baumeister et al., 1998). Peptides generated by the proteosome are translocated into the endoplasmic reticulum via transporter molecules, the TAP dimer (for transporter associated with antigen presentation) (for a recent review, see Pamer and Cresswell, 1998). In the endoplasmic reticulum, binding of peptides to class I molecules leads to the export of the class I–peptide complex from the endoplasmic reticulum followed by stable expression at the cell surface (Townsend et al., 1989; Ljunggren et al., 1990) (Fig. 1). MHC class II-restricted peptides are generated in endosomal/lysosomal organelles, which are well equipped to carry out such degradation, due to an abundant presence of proteolytic enzymes (Cresswell et al., 1987; Korn-

162

PIETERS

feld and Mellman, 1989; Fineschi and Miller, 1997; Chapman, 1998). Thus, MHC class II molecules must be transported from their site of biosynthesis, the endoplasmic reticulum, to these sites, from where they are further directed to the cell surface (see Fig. 1). How this is achieved, which molecules are involved, and the intracellular pathways followed will be the major focus of this chapter. III. Structure of MHC Class II Complexes

A. STRUCTURE OF MHC CLASS II MOLECULES The cell surface form of the MHC class II complex consists of a heterodimer of two transmembrane glycoproteins, the alpha chain and beta chain, that are encoded within the major histocompatibility complex located on chromosome 6 in humans (where it is called HLA) and chromosome 17 in mice (termed H-2 complex) (McDevitt and Bodmer, 1974; Wiman et al., 1982; Hansen et al., 1993). The 움 chain has a molecular weight of 앑33 kDa, bearing two consensus sites for N-linked glycosylation, whereas the 웁 chain is an 앑27-kDa glycoprotein containing one N-linked glycosylation site (Fig. 2) (Kvist et al., 1982). Both the 움 and the 웁 chain consist of a short cytoplasmic domain, a transmembrane portion, followed by an 움2 and a 웁2 domain, respectively, with amino-acid sequences and structure similarities to immunoglobulin constant domains (see Fig. 2) (Cresswell et al., 1987). Elucidation of the MHC class II structure by X-ray crystallography revealed that the class II complex is structurally almost identical to the MHC class I molecule (Bjorkman et al., 1987; Brown et al., 1993). The most remarkable feature of the MHC lies in the antigenic peptide-binding domain, consisting for the class II molecules of the 움1 and 웁1 domains, which fold together to form a peptide-binding cleft (Brown et al., 1993) (Figs. 2b and 2c). This peptide-binding site is composed of eight antiparallel 웁 sheets that are bordered by two 움 helices. Both ends of the cleft are open, and, as a consequence, they allow relatively long (poly)peptides that can protrude from the peptide-binding groove (Demotz et al., 1989; Chicz et al., 1993; Brown et al., 1993a; Rammensee et al., 1993a; Falk et al., 1994). Interestingly, one notable difference in the structures of the class I and class II complexes is that the peptide-binding cleft in the class I complex is closed, restricting the peptide length that can be accommodated (Bjorkman et al., 1987). The peptide-binding clefts of the MHC class I and class II molecules represent a unique structural domain that has features optimal for selectively activating T lymphocytes in a peptide-specific manner. Furthermore, the peptide nature of the MHC ligand ensures that not only those proteins

FIG. 2. Structure of the MHC class I and class II complex. (a) The MHC class I molecule folds into three domains, 움1, 움2, and 움3 that pairs with the 웁2-microglobulin molecule. The 움3 domain, as well as the 웁2M molecule, has a structure similar to those of immunoglobulin constant domains. The 움1 and 움2 domains fold into each other to form the peptide binding cleft. The overall organization of the class II molecules is very similar, consisting of a compact four-domain structure. Whereas the 움2 and 웁2 have a structure similar to those of immunoglobulin constant domains, the 움1 and 웁1 domains fold into each other, forming a peptide binding cleft highly similar to the class I peptide-binding domain. One notable difference is that while the MHC class I peptide-binding cleft is closed at both ends, limiting the size of peptides that can be accommodated, the class II peptide-binding cleft is open at both sides. (Modified from Immunobiology, Janeway and Travers, 1996, Fig. 4, p. 4.5 with kind permission from Elsevier Science Ltd./Garland Publishing, London, UK.) (b) The left side shows the MHC class I molecule, which in association with 웁2microglobulin can bind antigenic peptides that are imported into the endoplasmic reticulum. In the middle the MHC class II complex, consisting of an 움 chain, a 웁 chain, and the invariant chain, is depicted. The MHC class II and I assemble into a nonameric complex in the endoplasmic reticulum. It should be noted that whereas MHC class I molecules bind peptides of 앑9 amino acid residues, MHC class II molecules are able to bind relatively large polypeptide fragments after removal of the invariant chain. A model of the MHC class II–peptide complex as it appears at the plasma membrane is represented in the right panel. (Adapted from Pieters, 1997b, Curr. Opin. Immunol. 9, 89–96, with kind permission from Elsevier Science Ltd., London, UK.)

164

PIETERS

that are readily accessible, but virtually every (proteinous) part of the microbe can serve as a substrate for the MHC molecule, allowing the generation of a broad repertoire of T cell reactivity. B. POLYGENICITY AND POLYMORPHISM OF THE MHC MOLECULES Activation of T lymphocytes by peptide–MHC class II complexes results in the destruction of the microorganisms from which the peptides are derived either through a humoral (antibody) or cell-mediated immune response (Zinkernagel and Doherty, 1974a,b; Benacerraf and Germain, 1978; Benacerraf, 1981). The question therefore arises why microbes cannot easily circumvent T cell activation by merely mutating their structural genes in a way that resulting peptides will no longer be presented by MHC molecules. Part of the answer lies in the fact that the MHC class I and II molecules are both polygenic and polymorphic (Benacerraf and McDevitt, 1972; Kaufman et al., 1984; Mengle-Gaw and McDevitt, 1985; Snell, 1986; Trowsdale and Campbell, 1992). Polygenicity means that for each of the class I and class II molecules, several genes encode different proteins with different peptide-binding specificities. An overview of the different genes encoding the classical MHC class I and class II molecules in the human and the murine systems is given in Table I. In the murine system, two class II genes exist, named I-A and I-E. In humans, there are three pairs of class II genes, named HLA-DR, HLA-DP, and HLA-DQ (Trowsdale et al., 1991; Trowsdale and Campbell, 1992), as well as the nonclassical class II genes HLA-DM and HLA-DO (see later). All genes are co-dominantly expressed, and all of the resulting proteins have different peptide-binding specificities and are expressed at the cell surface, thereby broadening the range of peptides that can be presented by a given individual (Falk et al., 1991, 1994; Rammensee et al., 1993b; Vogt et al., 1994). In addition to the polygenicity, the MHC is polymorphic; i.e., there are multiple alleles at each locus. In fact, the MHC genes are among the most polymorphic genes known today (over 800 alleles are known in the human MHC to encode class I and class II molecules (Bodmer et al., 1997)). Interestingly, most of the polymorphic residues are localized within the TABLE I NOMENCLATURE OF THE CLASSICAL MHC CLASS I AND CLASS II GENES

Human Murine

MHC Class I

MHC Class II

HLA-A, HLA-B, HLA-C H2-K, H2-D, H2-L

HLA-DR, HLA-DP, HLA-DQ I-A, I-E

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

165

peptide-binding cleft, thereby greatly expanding the binding specificity of each MHC molecule. One exception to the polymorphism of MHC class I and class II molecules is the HLA–DR (in human) and I–E (in mice) 움 chain, which is identical in different individuals (‘‘monomorphic’’) for reasons that are unknown. The polymorphism of the MHC class I and class II genes greatly expands the diversity of the different MHC molecules, thereby providing each individual with the potency to bind a wide variety of peptides. C. STRUCTURAL FEATURES OF MHC CLASS II-ASSOCIATED PEPTIDES The different structure of the class I and class II peptide-binding cleft is reflected in the array of peptides that associate with the two types of complexes. Insight into the properties of peptides bound to MHC molecules came both from structure determinations of MHC molecules and from direct analysis of peptide sequences that could be eluted from purified MHC molecules (Rotzschke et al., 1990; Van Bleek and Nathenson, 1990; Rammensee et al., 1993b). As predicted from the class I crystal structure, the length of the peptides eluted was restricted to 앑9 amino acid residues, consistent with the space available within the peptide-binding groove (Bjorkman et al., 1987; Schumacher et al., 1991; Rammensee et al., 1993b). The same method was subsequently used to determine the specificity of the peptides bound to class II molecules, revealing a much more variable length of the peptides eluted, ranging from 15 to 25 residues, consistent with the observed crystal structure (Demotz et al., 1989; Rudensky et al., 1991b; Chicz et al., 1992, 1993; Riberdy et al., 1992; Newcomb and Cresswell, 1993; Falk et al., 1994). Given the high polymorphism of the MHC class I and class II molecules, one might wonder whether there are any rules governing the sequence specificity of class I and class II binding peptides at all. In fact, one of the important results of characterizing peptides eluted from purified MHC molecules was the realization of the presence of so-called anchor residues (Falk et al., 1991), whose existence also became clear through the cocrystallization of MHC molecules with peptide (Garrett et al., 1989; Fremont et al., 1992; Matsumura et al., 1992). Anchor residues are amino acids within MHC binding peptides that are identical or very similar in peptides eluted from a given allelic variant, and ‘‘anchor’’ the peptide within the MHC peptide-binding cleft at fixed positions (Rammensee et al., 1993a). The anchor residues in class I eluted peptides have been much better characterized than the ones in class II eluted peptides. Class II binding peptides seem to be more degenerate, probably related to the open structure of the peptide-binding pocket, as opposed to the closed structure in class I molecules. Class II molecules bind peptides through

166

PIETERS

the central region in the peptide-binding cleft, and both the amino-terminal and the C-terminal residues protrude from the binding cleft (Stern et al., 1994; Rammensee, 1995). Why the two classes of MHC molecules apply different rules with respect to their peptide-binding specificity is not clear. The difference may lie in the different sources of peptides (viral vs bacterial proteins), the different subcellular site for peptide loading (endoplasmic reticulum vs endosomal/lysosomal pathway), or the type of T cell to be activated (killer T cells vs T helper cell). Since the original identification of the specificities of MHC binding peptides, a wealth of information has become available on the binding characteristics of a wide variety of MHC alleles. In principle, this information can be used to predict epitopes for T lymphocytes (Rotzschke et al., 1991; Falk et al., 1995). More importantly, a number of autoimmune diseases are associated with the expression of certain HLA molecules, in particular, class II (Nepom and Erlich, 1991; Martin et al., 1992; Vyse and Kotzin, 1998; McDevitt and Wakeland, 1998). The knowledge obtained from analyzing HLA-associated peptides might enable the characterization of T lymphocytes involved in autoimmunity by screening candidate peptides derived from self proteins that may activate patients’ T lymphocytes (Wordsworth and Bell, 1992; Falk et al., 1995). D. THE MHC CLASS II-ASSOCIATED INVARIANT CHAIN At the end of the seventies, it was realized that in addition to the 움 and 웁 subunits of the MHC class II molecules, a third polypeptide with a Mr of 앑31-kDa was present in MHC class II immunoprecipitates obtained from radiolabeled cells ( Jones et al., 1979; Charron and McDevitt, 1979; Shackelford and Strominger, 1980; Owen et al., 1981; Charron et al., 1983; Rudd et al., 1985). In contrast to the MHC class II molecules, this molecule was found to be nonpolymorphic and therefore was called the invariant chain (Ii) (Machamer and Cresswell, 1982; Koch et al., 1989). Cloning and sequencing of the cDNA encoding human and murine invariant chain revealed that the invariant chain is encoded outside the MHC (on chromosome 5 in human and chromosome 18 in mice) (Claesson-Welsh et al., 1984; Koch et al., 1987; Genuardi and Saunders, 1988) and encodes a type II transmembrane protein (N terminus cytoplasmically, C terminus lumenally) (Wiman et al., 1982). Despite its name, multiple forms of Ii exist (Claesson et al., 1983; Quaranta et al., 1984; Singer et al., 1984; Strubin et al., 1984, 1986a,b). These different forms arise both through alternative initiation sites and through alternative splicing and can be found expressed within the same cell (Fig. 3) (Strubin et al., 1986a; Koch et al., 1987).

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

167

FIG. 3. Structural and functional domains in the MHC class II-associated invariant chain (Ii). Four basic forms of Ii can be expressed in human cells, IiP33, IiP41 (upper part), IiP35, and IiP43 (lower part). In the mouse, only the shorter forms of Ii are expressed, IiP31 and IiP41. IiP41 and IiP43 contain an additional exon encoding a 64-amino-acid, cysteine-rich domain that can function as an inhibitor of endosomal/lysosomal cysteine proteases. IiP35 and IiP43 arise by alternative initiation and contain an additional segment of 16 amino acid residues that encode a signal for retention in the endoplasmic reticulum (ret, residues ⫺16–0, which is not found in mice). Trimerization of Ii molecules occurs via the C-terminal trimerization domain (residues 163–183). The transmembrane segment (TM, residues 30–60) may also contribute to trimerization. Association of Ii with MHC class II molecules is mediated predominantly via the CLIP region (class II-associated invariant chain peptide, residues 81–104). The N-terminal cytoplasmic tail of these invariant chains contains signals for targeting to MHC class II compartments in the endocytic pathway. Branched structures indicate carbohydrate side chains. (From Pieters, 1997b, Curr. Opin. Immunol. 9, 89–96, with kind permission from Elsevier Science Ltd., London, UK.)

In human cells, alternative initiation gives rise to an additional segment of 16 amino acid residues, producing IiP35, which functions in retaining the Ii molecules in the endoplasmic reticulum (Marks et al., 1990; Lamb and Cresswell, 1992); see Fig. 3. In addition, both in human and in mouse cells, alternative splicing gives rise to the inclusion of an additional exon (exon 6b), producing a 43- and/or 45-kDa molecule in human cells and a 41-kDa molecule in mouse cells (Strubin et al., 1986a; Koch et al., 1987). This exon encodes a cysteine-rich segment consisting of 64 amino acid residues, exhibiting a high degree of homology to a motif found in thyroglobulin that has been found to display protease inhibitory activity (Koch et al., 1989; Bevec et al., 1996); see below. Although in some cell types the capacity of Ii to enhance antigen presentation is restricted to this extended form (Peterson and Miller, 1992), when reconstituted in Ii⫺/⫺

168

PIETERS

mice, both the normal and the additional spliced Ii form were shown to function to similar degrees (Naujokas et al., 1995; Shachar et al., 1995; Takaesu et al., 1995). However, the precise functions of each Ii form may vary in different antigen-presenting cell types (Engering et al., 1998). Finally, the invariant chain contains both N-linked and O-linked glycosylation sites and can be modified by chondroitin sulfate addition as well as by phosphorylation (Fig. 3) (Claesson et al., 1983; Miller et al., 1988; Spiro and Quaranta, 1989; Cresswell, 1992). The chondroitin sulfate form of Ii has been reported to function in co-stimulation of T lymphocytes (Naujokas et al., 1993), whereas phosphorylation of Ii may regulate its intracellular transport (Anderson and Roche, 1998; Pieters, unpublished). The invariant chain is usually synthesized in excess over the class II molecules, ensuring that in the endoplasmic reticulum all 움웁 dimers associate with Ii chains (Kvist et al., 1982; Sekaly et al., 1986; Pieters et al., 1991). Research over the past decade has revealed several important functions for the invariant chain that allow the generation of a proper MHC class II–peptide complex at the cell surface of antigen-presenting cells. First, the invariant chain functions as a chaperone in the endoplasmic reticulum, assisting the MHC class II complex to fold properly (Anderson and Miller, 1992). Second, at early stages after biosynthesis, the invariant chain prevents binding of peptides to MHC class II complexes, as peptides imported into the endoplasmic reticulum must bind to MHC class I molecules (Teyton et al., 1990; Roche and Cresswell, 1990). Third, the invariant chain determines the targeting and transport of MHC class II complexes to the appropriate organelles within antigen-presenting cells in which peptides are loaded onto class II molecules (Bakke and Dobberstein, 1990; Lotteau et al., 1990; Pieters et al., 1993). Interaction of Ii with class II molecules is predominantly mediated through a region in the lumenal domain of Ii, spanning amino acid residues 81–104 (Freisewinkel et al., 1993). This region has been termed CLIP (for class II-associated invariant chain peptide; see Figs. 2 and 3), because this peptide was found to be frequently associated with class II molecules as analyzed by peptide elution experiments (Rudensky et al., 1991a; Riberdy et al., 1992; Chicz et al., 1992; Sette et al., 1992). This suggested that the CLIP region binds to class II molecules in the peptide-binding groove. Indeed, when the crystal structure of the MHC class II-CLIP complex was solved by X-ray crystallography, it turned out that CLIP binds to class II molecules in a manner similar to the binding of an antigenic peptide (Stern et al., 1994; Ghosh et al., 1995). However, as not all class II alleles can bind CLIP, other segments in Ii may also contribute to the binding of class II molecules to Ii (Sette et al., 1995; Cresswell, 1996).

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

169

As mentioned, an important function of the invariant chain is to target the class II complex to the organelles, where class II molecules can associate with antigenic peptides. Once the class II/Ii complex arrives at the transGolgi network (see Fig. 1), the complex is deviated from the secretory pathway into the endosomal/lysosomal pathway (Peters et al., 1991; Pieters et al., 1991). The signals responsible for this targeting phenomenon were found to be located within the N-terminal, cytoplasmic portion of the invariant chain (Lotteau et al., 1990; Bakke and Dobberstein, 1990; Pieters et al., 1993) and will be separately discussed (see Section IV). Once the Ii has targeted the class II complex to the appropriate intracellular organelle, where antigenic peptides can be acquired, its function is finished and it must be removed. In addition, because the invariant chain CLIP region occupies the peptide-binding site, this Ii region must be removed to allow antigenic peptides to bind to class II molecules. Intracellular Ii removal is achieved by limited proteolysis within endosomal organelles, which results in the stepwise degradation of the Ii lumenal domain (Blum and Cresswell, 1988; Pieters et al., 1991), through the activity of cathepsins (Riese et al., 1996; Nakagawa et al., 1998, 1999). Interestingly, it was recently found in mice that the protease that degrades the invariant chain is cell type specific; whereas cathepsin S is responsible for Ii degradation in B lymphocytes and dendritic cells (Shi et al., 1999; Nakagawa et al., 1999), in thymic cortical epithelial cells, the cells that carry out positive selection (see also following), cathepsin L is the protease that degrades Ii (Cresswell, 1998; Nakagawa et al., 1998). The reason for this dichotomy is not known. As a result of this limited proteolysis, a small Ii fragment remains associated with the class II molecules, which includes the CLIP fragment. Exchange of CLIP for antigenic peptides is to a large degree dependent on the activity of the HLA-DM/HLA-DO molecules (Cresswell, 1994a; Busch and Mellins, 1996). E. THE NONCLASSICAL CLASS II-LIKE MOLECULES HLA-DM AND HLA-DO The HLA-DM and HLA-DO molecules, in contrast to the invariant chain, are encoded in the MHC (Karlsson et al., 1991; Cho et al., 1991; Kelly et al., 1991) and show a low degree of polymorphism (Inoko et al., 1985; Servenius et al., 1987; Karlsson et al., 1991; Karlsson and Peterson, 1992; Sanderson et al., 1994; Peleraux et al., 1996). The HLA-DM molecules, called H2-M in mice, are membrane-bound heterodimers that were discovered in cells defective in the processing and presentation of intact antigen, but which could readily present peptides (Mellins et al., 1990; Fling et al., 1994; Denzin et al., 1994; Morris et al., 1994). In fact, most

170

PIETERS

of the class II molecules in these cells were loaded with the CLIP fragment (Sette et al., 1992; Riberdy et al., 1992). This suggested that either antigen processing was deficient or the mechanism for loading peptides onto class II molecules in exchange for CLIP was compromised. A series of biochemical experiments from various laboratories established that HLA-DM indeed functions as a catalyst to remove CLIP in favor of antigenic peptides (Sherman et al., 1995; Sloan et al., 1995; Denzin and Cresswell, 1995). In accordance with a function in peptide exchange on class II molecules, HLA-DM is targeted to the endocytic pathway via a tyrosine motif present in its cytoplasmic tail (Lindstedt et al., 1995; Marks et al., 1995b). The precise mechanisms by which HLA-DM catalyzes release of CLIP and binding of antigenic peptides to class II molecules are still unclear. It has been proposed that HLA-DM functions as a true chaperone in that it binds class II–peptide complexes, such as class II–CLIP, that are relatively unstable. These associations may endure until the class II molecules encounter peptides that induce a more stable conformation, possibly allowing these complexes to be exported to the cell surface (Roche, 1995). It should also be noted that the requirement for HLA-DM in peptide loading is dependent on the class II allele (Stebbins et al., 1995), suggesting that CLIP release from class II molecules can occur in the absence of HLA-DM (Kropshofer et al., 1997). To assess the role of HLA-DM in antigen presentation in vivo, mice lacking the homologous H2-M molecules were generated by targeted deletion (Fung-Leung et al., 1996; Miyazaki et al., 1996; Martin et al., 1996). As expected, antigen-presenting cells from these mice were unable to present peptides derived from intact antigens, while most class II molecules remained stably associated with the CLIP fragment, both intracellularly and at the cell surface. Interestingly, these mice proved to be extremely valuable in assessing the role of MHC class II–peptide complexes in the thymus. Normally, within the thymus, T lymphocytes that are especially effective in the recognition of foreign peptides are being selected, a process called ‘‘positive selection’’ (von Boehmer, 1994). An unresolved question, however, was the precise contribution of specific peptides in the positive selection of T lymphocytes. In mice lacking H2-M molecules, which almost exclusively contain the CLIP fragment associated with self-MHC molecules, positive selection is nevertheless quite efficient (Fung-Leung et al., 1996; Miyazaki et al., 1996; Martin et al., 1996). This suggests that there is no need for a wide range of different peptides to be present on thymic antigen-presenting cells for efficient selection of T lymphocytes to occur. More likely, stabilization of class II molecules is required, and such stabilization can be induced by the CLIP peptide. Indeed, in a separate study, a single class II–peptide

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

171

combination (distinct from CLIP) was shown to be sufficient for the generation of a large T cell repertoire (Ignatowicz et al., 1996). Recent studies have revealed the contribution of another chaperone molecule, the heterodimer HLA-DO, in HLA-DM-dependent peptide loading onto class II molecules. HLA-DO (or H-2O in mice) was originally identified as a molecule encoded in the class II region of the MHC that was expressed on a subset of antigen-presenting cells (Karlsson et al., 1991). HLA-DO associates with HLA-DM in the endoplasmic reticulum, after which the DM/DO complex is targeted to the endocytic pathway, where it remains associated (Liljedahl et al., 1996). HLA-DO functions as an inhibitor of HLA-DM activity (Denzin et al., 1997; van Ham et al., 1997), possibly regulating the specificity of antigen presentation in different cell types ( Jensen, 1998). In accordance with this, targeted deletion of the H-2O gene in mice showed that B cells from these mice had an altered capacity to present protein antigens compared to wild-type mice (Liljedahl et al., 1998). HLA-DO may function as a pH-dependent inhibitor of DM, modifying its peptide-editing function, and may favor presentation of antigens that are internalized by B cell receptor-mediated uptake, as opposed to fluid-phase endocytosis (Liljedahl et al., 1998). IV. Biosynthesis and Assembly of MHC Class II/Invariant Chain Complexes

A. TRANSCRIPTIONAL CONTROL OF MHC CLASS II/Ii EXPRESSION The expression of MHC class II molecules, in contrast to that of MHC class I molecules, is restricted to certain cell types only. B lymphocytes, macrophages, and dendritic cells express MHC class II molecules constitutively, albeit to varying degrees, whereas in many cell types MHC class II expression can be induced by interferon-웂 (Pober et al., 1983). Regulation of class II expression occurs at the level of transcription, and research on the molecular defects underlying the bare lymphocyte syndrome (BLS), a rare form of immunodeficiency characterized by the absence of class II molecules from all cell types, led to the discovery of several factors important for class II expression (Mach et al., 1994; Reith et al., 1995; Boss, 1997). Many of the identified factors are DNA-binding proteins and are ubiquitously expressed (Steimle et al., 1993, 1995; Boss, 1997). An additional factor, named CIITA, for MHC class II trans-activator molecule, is not a direct DNA-binding molecule (Steimle et al., 1993; Mach et al., 1994; Chang et al., 1996; Boss, 1997). CIITA is, however, an important mediator of class II expression, in cells that express class II both constitutively and after induction with interferon-웂 (Steimle et al., 1994; Reith et al., 1995). Interestingly, CIITA also regulates the expression of the invariant chain and the HLA-DM molecules (Chang and Flavell, 1995; Kern et al.,

172

PIETERS

1995). Deletion of the CIITA gene in mice leads to a severe reduction of class II molecules on B lymphocytes and macrophages, although a subset of dendritic cells still expresses MHC class II, raising the possibility of the existence of alternative transactivation mechanisms (Chang et al., 1996; Williams et al., 1998). B. FOLDING AND ASSEMBLY OF NEWLY SYNTHESIZED CLASS II/Ii COMPLEXES IN THE ENDOPLASMIC RETICULUM Due to the transcriptional regulation by CIITA, MHC class II molecules and Ii are usually coexpressed within the same cell. The class II 움 and 웁 chains as well as Ii are cotranslationally inserted into the membrane of the endoplasmic reticulum (ER) through signal-peptide-mediated processes (Blobel and Dobberstein, 1975a,b; Kvist et al., 1982; Dobberstein, 1987). Once in the ER, the invariant chain associates to form a trimer, upon which dimers of 움웁 molecules are assembled to form a nonameric structure (Roche et al., 1991; Cresswell, 1994b, 1996). Trimers of Ii can accommodate any of the different Ii forms mentioned (Fig. 2), and the final composition will depend on the relative expression levels in different cell types (Marks et al., 1990; Arunachalam et al., 1994; Engering et al., 1998). In addition, different MHC class II molecules can assemble within one oligomeric complex, and HLA-DR, DQ, and DP mixtures may exist within one complex (De Kretser et al., 1982), such that in heterozygous individuals different class II alleles can occupy one nonamer. As for all ER resident proteins, both Ii and the MHC class II subunits are retained by ER resident chaperones until they are properly folded (Hurtley and Helenius, 1989; Schaiff et al., 1992; Romagnoli and Germain, 1995; Marks et al., 1995a; Arunachalam and Cresswell, 1995). However, in many cell types both class II molecules and Ii can be exported independently from one another from the endoplasmic reticulum to later stages of the biosynthetic pathway (Teyton et al., 1990; Lotteau et al., 1990; Bakke and Dobberstein, 1990; Pieters et al., 1993; Chervonsky et al., 1994). Despite the ability of class II molecules to be transported to the cell surface in the absence of Ii, Ii-negative cells are severely compromised in the ability to process and present antigens to T lymphocytes (Stockinger et al., 1989; Viville et al., 1993), illustrating the importance of Ii in proper functioning of MHC class II-restricted antigen presentation. C. TRANSPORT OF MHC CLASS II/Ii COMPLEXES THROUGH THE ENDOPLASMIC RETICULUM– GOLGI COMPARTMENTS After the MHC class II/Ii complex has been properly folded, it is exported from the endoplasmic reticulum. The precise mechanisms that are involved in export from the ER are still unclear. In general, the endoplasmic reticu-

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

173

lum functions as a quality control station in which newly synthesized proteins are properly folded, assisted by various chaperones (Hammond and Helenius, 1995). Correctly folded proteins and oligomers are then competent to be transported to later stages of the exocytic pathway, whereas misfolded proteins are being degraded, most probably by the proteosome, after their exit from the endoplasmic reticulum (Klausner and Sitia, 1990; Bonifacino, 1996; Wiertz et al., 1996a,b). In case of the MHC class II/Ii complex, proper formation of the nonameric structure might be sufficient to release this complex from the ER into the Golgi compartments (Hurtley and Helenius, 1989; Marks et al., 1995a; Arunachalam and Cresswell, 1995). Whether a distinct set of chaperones accompanies the class II/Ii complex to later stages of the pathway is unclear, although no candidate accessory molecules at those stages have been observed to be associated with class II molecules. The Ii trimer in the absence of class II molecules is normally retained in most cells. In addition, the N-terminally extended forms of Ii in human cells, IiP35 and IiP43 (Fig. 3), contain an endoplasmic reticulum retention signal, which may have a specific function within a particular cell type (Bakke and Dobberstein, 1990; Lotteau et al., 1990; Pieters et al., 1993). For example, in human dendritic cells, the export of class II/Ii complexes from the endoplasmic reticulum during their development is highly regulated, which may be related to differences in the Ii composition of the class II/Ii complexes at these different stages (Engering et al., 1998). After export of the class II/Ii complex to the Golgi complex, extensive glycosylation on both 움웁 molecules as well as Ii occurs (Kvist et al., 1982). The role of glycosylation of the class II molecules is not clear. In the absence of glycosylation, the transport and stability of MHC class II molecules are unaltered (A. Engering and J. Pieters, manuscript in preparation) and peptide presentation does also not seem to require N-linked glycans (Nag et al., 1992, 1994). One possibility is that glycosylation events contribute to proper folding (Hurtley and Helenius, 1989). Alternatively, glycosylation of class II complexes may be important for the regulation of peptide acquisition in the endocytic pathway (see following). V. Entry of MHC Class II/Invariant Chain Complexes into the Endocytic Pathway

Peptides presented by class II molecules are derived from endocytosed antigens, and the observation that class II molecules, Ii, and endocytosed material could be colocalized within the same organelles (Pieters et al., 1991) opened the search for the targeting events as well as the precise subcellular organelles involved (Cresswell, 1994b; Wolf and Ploegh, 1995; Pieters, 1997a). The work defining the specific sorting signals involved in

174

PIETERS

post-Golgi targeting of class II/Ii, as well as the characterization of these compartments, not only was instrumental in elucidating the MHC class II pathway, but also contributed to insight into basic cell biological mechanisms of protein and membrane trafficking. A. SORTING IN THE TRANS-GOLGI NETWORK MEDIATED BY Ii The original model of vectorial transport through the secretory pathway as outlined by Palade (1975) still serves as a framework for intracellular transport processes. Within this framework, the trans-Golgi network has been viewed as a sorting station for three groups of proteins translocated into the endoplasmic reticulum: first, plasma membrane and constitutively secreted proteins; second, secretory proteins to be packaged into secretory vesicles; and third, proteins destined for endosomes and lysosomes (Griffiths and Simons, 1986; Burgess and Kelly, 1987). The localization of class II/Ii comlexes in endocytic organelles and the absence of the Ii from the cell surface class II complex made the invariant chain an attractive candidate for mediating sorting at the trans-Golgi network. A series of straightforward experiments revealed the presence of targeting information in the Ii cytoplasmic tail (Bakke and Dobberstein, 1990; Lotteau et al., 1990). Whereas full-length Ii, expressed alone or in combination with class II molecules, is readily targeted to endosomal/ lysosomal organelles, deletion of the cytoplasmic tail results in the expression of Ii at the cell surface only. More specific mutagenesis revealed the presence of two distinct targeting signals within the cytoplasmic tail of Ii, the critical residues being a Leu–Ile and Met–Leu pair (Pieters et al., 1993; Odorizzi et al., 1994), each of which separately was sufficient to mediate endosomal targeting. These experiments also revealed the contribution of the Ii transmembrane region to the correct sorting and retention of Ii; chimeric molecules containing also the transmembrane region of Ii were more efficiently sorted to the endocytic pathway, as is the case for molecules containing exclusively the cytoplasmic tail (Pieters et al., 1993; Odorizzi et al., 1994). Taken together, the complete Ii molecule contains both active sorting signals and a retention signal for the proper endocytic organelles to which it has to be targeted after its egress from the transGolgi network. The molecular machinery responsible for endosomal targeting from the trans-Golgi network has not yet been fully elucidated. The current model proposes that the cytoplasmic portions of transmembrane molecules to be transported interact with components of a proteinaceous coat surrounding the Golgi membrane that is also required for vesicle formation (Kirchhausen et al., 1997). The major constituent of this protein coat is clathrin, a scaffold protein (Smith et al., 1998; ter Haar et al., 1998) to which several

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

175

adaptor molecules can bind (Le Borgne and Hoflack, 1998; Pishvaee and Payne, 1998; Jackson, 1998). Different adaptor complexes have been identified, two of which, AP-1 and AP-3, appear to be specifically located at the trans-Golgi network (Robinson, 1994; Simpson et al., 1997; Dell’Angelica et al., 1997). To date, two distinct targeting signals are known to be responsible for endosomal targeting of transmembrane proteins: the tyrosine-based motif (Davis et al., 1987; Chen et al., 1990) and the dileucine motif, the latter of which is present in the Ii cytosolic tail (Letourneur and Klausner, 1992; Pieters et al., 1993). Protein targeting by either tyrosine-based or dileucine motifs occurs via a distinct mechanism (Marks et al., 1996), and these two signals appear to bind to different subunits of the AP-1 adaptor complexes (Ohno et al., 1995; Rapoport et al., 1998). In addition to endosomal targeting from the trans-Golgi network, most tyrosine- and dileucinecontaining proteins, including the invariant chain (Letourneur and Klausner, 1992; Pieters et al., 1993), can also reach endosomes after internalization from the cell surface by processes likely to be analogous to transGolgi-mediated targeting (Ohno et al., 1995). An interesting question is why two distinct (tyrosine- and dileucinebased) signals are used within the same cell to reach the endosomal pathway. Most probably, these two signals allow transport to endosomes via different routes and, in addition, may lead to targeting to distinct subcompartments of the endosomal/lysosomal pathway (see following). Why does the invariant chain contain multiple sorting determinants to ensure a proper localization of the class II complex to endosomal organelles? One possibility is that, depending on the type of antigen presenting cell, the endosomal targeting mechanisms may be different. The other possibility is that within one and the same cell multiple routes exist to target Ii and the associated class II molecules to endosomal organelles. For example, dileucine signals are also involved in targeting proteins from the plasma membrane to endosomal/lysosomal organelles (Letourneur and Klausner, 1992), and this route may be followed by Ii/class II complexes that are first transported to the cell surface and subsequently internalized (Warmerdam et al., 1996). The efficiency of targeting may be dictated by the composition of the Ii trimer within the class II/Ii nonameric complex: complexes containing at least two Ii molecules with an intact cytoplasmic domain are efficiently targeted to the endocytic pathway, whereas when only one intact Ii is present, the complex is expressed at the cell surface, followed by rapid internalization (Arneson and Miller, 1995). This suggests that multimerization of the Ii sorting determinants dictates efficient transGolgi network–endosome transport. The different routes taken by class II/Ii molecules may be relevant for the set of epitopes loaded onto class II molecules, as both in different cell types and in different subcellular

176

PIETERS

organelles, distinct antigenic peptides may reside. The various targeting phenomena mediated by Ii to escort class II molecules to different organelles may thus contribute to a maximal acquisition of antigenic peptides. VI. Subcellular Organelles Involved in Antigenic Peptide Loading onto Class II Molecules: The MHC Class II Compartments

Antigenic peptides to be presented to T lymphocytes by MHC class II molecules are derived from material that has entered the endocytic pathway of the antigen-presenting cells. Therefore, it was revealing to find that MHC class II molecules colocalized with endocytosed proteins within the endosomal/lysosomal pathway (Pieters et al., 1991; Peters et al., 1991). The organelles highly enriched in class II molecules had a typical multivesicular morphology and were therefore called ‘‘multivesicular bodies’’ (Pieters et al., 1991) or ‘‘MHC class II compartments’’ (Peters et al., 1991). Interestingly, the class II compartments appear to be set aside from the normal endosomal pathway, because various endosomal markers such as the mannose 6-phosphate receptor, as well as lysosomal markers, were absent from these organelles (Peters et al., 1991; Pieters et al., 1991; Pieters, 1997a). This suggested that antigen-presenting cells might possess specialized organelles dedicated to the process of peptide loading onto class II molecules. Confirmation of this idea came with the biochemical isolation and characterization of the intracellular MHC class II compartment (West et al., 1994; Tulp et al., 1994; Amigorena et al., 1994). These compartments receive newly synthesized MHC class II complexes by virtue of the targeting domain in the Ii cytoplasmic tail. Once the complex has been correctly targeted, the Ii is degraded from its lumenal, C-terminal domain, while the N-terminal cytoplasmic region remains associated for prolonged times (Blum and Cresswell, 1988; Pieters et al., 1991). This ensures retention of the class II complex in these organelles until appropriate peptide loading has occurred (Pieters et al., 1991; Amigorena et al., 1994; Tulp et al., 1994; Ferrari et al., 1997). Partially loaded class II complexes may recycle between the MHC class II compartments and the plasma membrane, and, when stably loaded with antigenic peptides, become a resident of the plasma membrane for the activation of T lymphocytes. All antigenprocessing and -presenting cells examined were found to contain MHC class II compartments; these are particularly abundant in dendritic cells, which are important antigen-processing and -presenting cells. This is illustrated in Fig. 4, which shows an overview of a dendritic cell in an immature state, containing large amounts of MHC class II compartments (Figs. 4a and 4b).

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

177

FIG. 4. MHC class II expression in immature and mature dendritic cells. Immature (a, b) and mature (c) dendritic cells were prepared for immunocytochemistry and ultrathin sections were labeled with anti-class II antibodies followed by 15-nm protein A–gold. (a) An overview of the intracellular and plasma membrane expression in an immature dendritic cell. Magnification 48,000⫻. (From Engering et al., 1998). (b) Several MHC class II compartments from immature dendritic cells. Small gold indicates localization of the mannose receptor. Magnification 48,000⫻. (c) Mature dendritic cell with predominantly plasma membrane expression. Magnification 64,000⫻. (Micrographs courtesy of A. Engering, D. M. Fluitsma, and E. C. M. Hoefsmit.)

178

PIETERS

Of course, the process of peptide loading onto class II molecules depends to a large degree on the presence of HLA-DM molecules. HLA-DM molecules are located within MHC class II compartments, but a puzzling feature was that they arrive there due to the presence of a tyrosine-based sorting motif, whereas the class II molecules are directed to these organelles via the dileucine motif present in the Ii cytoplasmic tail (Pieters et al., 1993; Marks et al., 1995b; Lindstedt et al., 1995). This discrepancy was solved with the demonstration that the MHC class II compartments actually consist of two physically and functionally distinct organelles (Ferrari et al., 1997). The first of these compartments receives newly synthesized class II/Ii complexes from the trans-Golgi network, and in these organelles the invariant chain is partially degraded, resulting in the formation of class II–CLIP complexes. These class II–CLIP complexes are then transported, possibly via a targeting signal present in the class II 웁 chain (Chervonsky et al., 1994), to another compartment. HLA-DM resides in these latter organelles, which assist in the exchange of the CLIP fragment for peptides and which have therefore been termed the compartment of peptide loading, or CPL (Fig. 5). After being fully loaded with antigenic peptides, the MHC class II molecules exit from these intracellular sites to be stably expressed at the cell surface (Tulp et al., 1994; Ferrari et al., 1997). Importantly, the HLA-DM-positive compartment of peptide loading is not accessible to endocytosed material, whereas the class II compartment where Ii is degraded is readily connected to the endocytic network. This is important for peptide loading, because MHC class II molecules, as noted previously, not only can bind small peptide fragments, but also can accommodate relatively large protein fragments. In fact, even intact polypeptides have been shown to associate with class II molecules (Lindner and Unanue, 1996). Thus, if antigen internalization and degradation would occur in the same organelles where Ii is cleaved from the class II molecules, endocytosed material would be able to bind directly to the class II molecules. Such binding, however, cannot occur, because it is prevented by the presence of the CLIP fragment to the class II complex in those organelles. Instead, peptides are loaded in a distinct compartment that is not accessible to the bulk endocytic pathway (Ferrari et al., 1997). By separating the (HLA-DM-dependent) peptideloading reaction from endocytosed material, peptides can be loaded in a much more specific manner, as well as at much higher concentrations, because competition from, for example, serum proteins that enter the endocytic pathway is absent. The question of how peptides are selectively transported to the compartment of peptide loading remains unresolved, but this process may be assisted by heat shock proteins that are present

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

179

FIG. 5. The MHC class II loading pathway. Antigenic peptide loading of MHC class II molecules requires the sequential transit of newly synthesized complexes through two distinct MHC class II compartments. After assembly in the endoplasmic reticulum, the class II/Ii complex is targeted at the trans-Golgi network to MHC class II compartments through the action of cathepsins (compartment of Ii degradation), leading to the generation of MHC class II molecules complexed with CLIP. Subsequently, the class II-CLIP complex is transported to a distinct organelle that contains HLA–DM. Here, CLIP is exchanged for antigenic peptides in a reaction catalyzed by HLA–DM. From this compartment of peptide loading (CPL) peptide-loaded class II molecules are transported to the cell surface. The CPL, in contrast to the compartment of invariant chain degradation, is not directly accessible to antigens internalized via fluid phase and therefore may receive peptides generated at a distinct site; antigens taken up through receptor-mediated endocytosis can, however, be rapidly transported to the CPL, either in an intact form or after degradation into peptides.

in both the endocytic and the peptide loading organelles (VanBuskirk et al., 1991; Pieters, unpublished); see also Fig. 5. In most antigen-presenting cells, the MHC class II compartments are the main sites of antigenic peptide loading. However, class II molecules

180

PIETERS

are more widely distributed through the different endosomal organelles (Pieters et al., 1991; Pieters, 1997a), and class II molecules may recycle between the class II compartments and the plasma membrane via the early endosomes. Recycling is thought to occur until the class II molecules have become fully loaded with antigenic peptides (Ferrari et al., 1997; Pieters, 1999), although class II molecules may acquire peptides during the recycling process. Indeed, for a subset of epitopes, it has been shown that loading occurs exclusively on recycling class II molecules in an invariant chain and HLA-DM-independent manner, most probably in early endosomes (Pinet et al., 1995; Zhong et al., 1997). Most likely, the class II system has evolved in such a way that it maximizes the encounter and binding of peptides throughout the endosomal/lysosomal pathway. The efficiency of the peptide-loading process is partially dictated by the entry mechanisms of the antigen; for example, receptor-mediated uptake processes generate class II–peptide complexes by different mechanisms as material that enters via fluid phase uptake (see following). Furthermore, it should be realized that the different types of antigen-presenting cells, such as macrophages, dendritic cells and B lymphocytes, each via cell-type-specific mechanisms, act in concert to generate a proper T cell response. VII. Transport of MHC Class II Complexes to the Cell Surface

How MHC class II molecules are transported to the cell surface is still unclear. It is likely that different routes are involved depending on the type of the antigen-presenting cell, the state of the MHC class II complex (i.e., fully loaded with antigenic peptides or not), and the presence of other molecules in the class II complex (such as Ii fragments). As mentioned, the MHC class II complex can recycle through different organelles of the endosomal pathway, as well as through the compartment of peptide loading via the plasma membrane. During recycling, peptides may be loaded onto class II molecules until a stable peptide–class II complex has formed (Ferrari et al., 1997). Whether this complex is then actively transported to the cell surface or, alternatively, cannot be internalized from the plasma membrane remains to be established. In human B lymphocytes, transfer of peptide-loaded class II molecules seems to involve organelles that are distinct from early endocytic structures, suggesting that active processes are involved (Pond and Watts, 1997). Another possibility that has been proposed is that the MHC class II molecules are externalized in so-called ‘‘exosomes,’’ which would represent the internal membranes of the intracellular MHC class II compartments (Raposo et al., 1996), a process proposed to be similar

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

181

to the release of secretory granules by cytotoxic T lymphocytes (Griffiths, 1995, 1996). The physiological relevance of such a pathway for peptideloaded class II molecules, however, remains unclear. Dendritic cells, as will be discussed in more detail later, are extremely efficient in the generation of cell surface peptide–class II complexes, and in these cells class II transport is highly regulated. Analysis of the composition of the intracellularly residing class II complexes revealed the presence of several molecules that may be involved in regulating the subcellular distribution of class II molecules, including their presence at the plasma membrane (A. Engering and J. Pieters, manuscript in preparation) (see also Section X). VIII. Antigen Internalization

The pathway by which an antigen is internalized is dictated both by the antigen itself and by the cell type that takes up the antigen. For example, intact bacteria and large protein aggregates can be internalized by professional phagocytes such as macrophages and dendritic cells through phagocytosis, whereas single molecules enter the cell through either fluid-phase endocytosis (which is nonselective) or receptormediated (more specific) uptake. An overview of the different modes of uptake is given in Fig. 6. A. FLUID-PHASE UPTAKE All cells are capable of internalizing material via fluid-phase uptake, although the efficiency varies greatly among different cell types. Macrophages and dendritic cells have a large endocytic capacity, sometimes referred to as macropinocytosis (Steinman and Swanson, 1995; Watts, 1997a), whereas B and T lymphocytes are extremely poor in fluid phase endocytosis. Material that enters the cell via fluid-phase uptake first enters early endosomes and is then transferred to late endosomes, finally reaching lysosomes (Griffiths et al., 1988; Mellman, 1996). These different endocytic organelles may harbor distinct proteases, and therefore both the endocytic traffic of the antigen-processing cell and the proteolytic susceptibility of the antigen itself determine the site of antigen degradation. B. RECEPTOR-MEDIATED UPTAKE A more specific way to internalize antigens occurs through receptormediated uptake. Antigens can bind either directly (in case of the mannose and B cell receptors) or after binding to circulating immunoglobulins (to Fc receptors). In most cases, receptor-mediated uptake ensures

182

PIETERS

FIG. 6. Possible pathways for antigen internalization. Particulate material such as viable and killed bacteria is taken up by phagocytosis. In addition, most cells can take up fluids as well as soluble molecules via fluid-phase endocytosis, whereas a more specific way to internalize molecules occurs through receptor-mediated endocytosis. After internalization, material is transported along the endosomal/lysosomal pathway, where it can be degraded. Note that some pathogens, such as mycobacteria, have evolved mechanisms to avoid lysosomal delivery and can survive within the phagosome. (From Pieters, 1999, by permission of JAI Press/Elsevier Science, Stamford, CT.)

a high intracellular concentration of antigens that circulate extracellularly at relatively low levels (Lanzavecchia, 1990). Alternatively, receptormediated uptake leads to specific targeting of the antigen to the class II peptide-loading compartment, thereby enhancing the presentation of antigenic epitopes by class II molecules (Bonnerot et al., 1995; Ferrari et al., 1997). Ligand binding is followed by receptor–ligand concentration in coated pits and delivery of the ligand to early endosomes, late endosomes, and lysosomes, although the precise organelle to which ligands are delivered varies according to the specific receptors. Simultaneously, in many cases, receptor clustering leads to the activation of intracellular signaling cascades, often via accessory molecules that are associated with the receptors. These signaling cascades may function as

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

183

activating or inhibiting signals, depending on the ligand bound as well as on the type of receptor (Ravetch and Kinet, 1991; Caron and Hall, 1998). B cell receptors and Fc receptors are targeted together with their ligands to the various endosomal/lysosomal organelles, whereas the mannose receptor releases its ligand in early endosomes and recycles to the cell surface for another round of internalization. 1. B Cell Receptor The B cell receptor basically consists of membrane-bound, surfaceexpressed immunoglobulin (mIgG) that is noncovalently associated to the Ig움/Ig웁 heterodimer (Neuberger et al., 1993; Reth and Wienands, 1997). Whereas mIgG provides the binding site for a particular antigen, the Ig움/웁 heterodimer functions in signaling events that accompany B cell activation (Pleiman et al., 1994; Reth and Wienands, 1997). The power of B-cell mediated antigen internalization is illustrated by the fact that antigen-specific B cells (bearing mIgG molecules specific for one antigen) can stimulate T lymphocytes at concentrations up to 10,000fold lower than those of nonspecific B lymphocytes (Rock et al., 1984; Lanzavecchia, 1985). The B cell receptor not only increases the ligand concentration intracellularly, but also may influence the types of epitopes that will be presented on class II molecules (Watts and Lanzavecchia, 1993; Aluvihare et al., 1997). This may be due to either the protection of certain domains of the antigen through the mIgG molecule itself or the differential intracellular trafficking of the mIgG–antigen complex (Aluvihare et al., 1997; Ferrari et al., 1997). Recent evidence suggests that both the cytoplasmic tail of the mIgG molecule and those from the Ig움/웁 heterodimer may play a role in intracellular targeting to a specific compartment (Bonnerot et al., 1995; Knight et al., 1997; Aluvihare et al., 1997). Thus, depending on the precise intracellular signaling events induced by ligand binding, the B cell receptor may target antigens to distinct subcellular organelles where they may be differentially proteolyzed and the resulting peptides bound to class II molecules (Tarlinton, 1997; Amigorena and Bonnerot, 1998). 2. Fc Receptors Fc receptors are expressed on most cells of the immune system and constitute a large family of membrane-integrated molecules that mediate the internalization of immunoglobulin-bound ligands, via recognition of the Fc portion of immunoglobulins (Daeron, 1997). These may be either soluble antigens that have encountered specific immunoglobulins or large particles such as bacteria, which will then be internalized through

184

PIETERS

phagocytosis by professional phagocytes. Fc receptor-mediated internalization is quite similar to the processes mentioned above for B cell receptor-mediated uptake. For each antibody class, there is a specific Fc receptor, Fc웂R (R for receptor) binding IgG, Fc움R binding IgA, Fc␧R binding IgE, and Fc애R binding IgM (Ravetch and Kinet, 1991; Hulett and Hogarth, 1994). Apart from these subtypes, two classes of Fc receptors exist: those bearing signal transduction motifs much like those found in B cell receptors leading to cell activation and those that do not contain any signaling motif and hence cannot activate cells, but whose ligand binding sometimes results in the generation of inhibiting signals (Reth, 1989; Daeron, 1997). Interestingly, the types of peptides presented from a given antigen appear to depend on the nature of the Fc receptor that mediates internalization (Amigorena et al., 1998). This may be because of different intracellular targeting events, as mentioned previously for the B cell receptor. 3. Mannose Receptor The mannose receptor is expressed on most professional phagocytes, contains multiple carbohydrate-binding domains, and is involved in the internalization of a variety of sugar-containing proteins (Ezekowitz et al., 1990; Stahl, 1992; Drickamer and Taylor, 1993; Engering et al., 1997a). Two features distinguish the mannose receptor from the previously mentioned B cell and Fc receptors: First, the mannose receptor recycles constitutively between the plasma membrane and early endosomes, where its ligands are released (Stahl et al., 1980; Engering et al., 1997a,b). Therefore, the mannose receptor allows multiple rounds of ligand internalization, in contrast to Fc and B cell receptors, which are degraded together with their ligand (see also Fig. 5). A second feature that distinguishes the mannose receptor from these other receptors is that it functions almost exclusively in the internalization of microbial ligands in which mannosylated proteins are relatively abundant (Ezekowitz et al., 1990, 1991; Pontow et al., 1992). Proteins from higher eukaryotes normally lack mannose residues in an accessible form. Thus, the mannose receptor functions as a broad specificity receptor for the internalization of microbial proteins and, in addition, can function in the internalization of entire microorganisms by phagocytosis (Gaynor et al., 1995; Schlesinger et al., 1996); see following. In contrast to the receptors mentioned, uptake via the mannose receptor does not appear to lead to cellular activation (Fraser et al., 1998; AstarieDequeker et al., 1999). Dendritic cells also express DEC-205, a member of the mannosereceptor family with a structure quite similar to that of the mannose receptor ( Jiang et al., 1995). Although anti-DEC-205 antibodies are rapidly

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

185

internalized by dendritic cells and presented to T cells, it is unclear which other ligands are internalized by DEC-205 for processing and presentation on class II molecules. C. PHAGOCYTOSIS Phagocytosis is normally defined as the engulfment of large particles such as microbes and tissue debris that are present during an infection or inflammation (Silverstein, 1995; Brown, 1995). After phagocytosis, this material arrives in the phagosome, from which it is normally transferred to late endosomes and lysosomes (Rabinowitz et al., 1992; Desjardins et al., 1994). In these organelles, antigens may be processed and from there be targeted to MHC class II compartments for loading of the resulting peptides onto class II molecules (Pieters, 1999). Indeed, most bacteria that have accessed phagocytes induce MHC class II-restricted responses (Pancholi et al., 1993; Unanue, 1997; Schaible et al., 1999). In some phagosomes, MHC class II molecules have been found to colocalize with bacteria (Antoine et al., 1991; Russell et al., 1992; Lang et al., 1994; Clemens and Horwitz, 1995). It is unclear, however, whether these arrive in the phagosome as a consequence of the internalization of large pieces of plasma membrane (Hasan et al., 1997), because bacterial degradation is not thought to occur prior to phagosome–lysosome fusion (Rockey and Rosquist, 1994; Mordue and Sibley, 1997; Schaible et al., 1999; Kaufmann and Hess, 1999). Phagocytosis can either occur nonspecifically, through the invagination of large pieces of plasma membrane, or be mediated by specific receptors. Molecules that have been implicated in phagocytosis are the Fc receptors (Greenberg et al., 1990), the mannose receptors (Ezekowitz et al., 1991; Kruskal et al., 1992), the integrins (Brown, 1991), and the scavenger receptor (Kodama et al., 1996). To which subcellular organelle phagocytosed material is delivered depends partially on the mode of entry; each of the above-mentioned receptors can trigger signal transduction events that can differentially modulate traffic of phagocytosed material intracelluarly (Greenberg et al., 1993; Greenberg, 1995; Caron and Hall, 1998). In addition, microorganisms engulfed by phagocytes are able to alter their intracellular fate by actively interfering with the normal endosomal trafficking events (Armstrong and D’Arcy Hart, 1971; Joiner et al., 1990; Lang et al., 1994; Sturgill-Koszycki et al., 1994; Hasan et al., 1997; Ferrari et al., 1999). IX. Antigen Processing

Despite a reasonably good understanding of the biology of the endosomal/lysosomal organelles through which MHC class II molecules and

186

PIETERS

antigens traffic, the proteases that mediate processing of these antigens into peptide fragments are still poorly characterized. The endosomal/ lysosomal pathway is filled with proteases that contribute to an efficient degradation of internalized material. Most of these are cysteinyl and aspartyl proteases, such as the cathepsins (Fineschi and Miller, 1997; Chapman, 1998). In vitro, several cathepsins have been shown to generate antigenic peptide fragments from intact antigens (Bennett et al., 1992; van Noort and Jacobs, 1994; Rodriguez and Diment, 1995). However, none of the known cathepsins have been shown in vivo to act in antigen processing, and animals lacking cathepsin B, D, or L show no obvious defect in antigen processing and presentation (Villadangos et al., 1997; Deussing et al., 1998) other than as a result of impaired Ii cleavage in a subset of cells, as has been discussed, (Cresswell, 1998; Nakagawa et al., 1998). A protease quite distinct from the cathepsins, namely, asparaginyl endopeptidase (AEP), was recently shown to mediate processing of a model antigen in vitro, and inhibitors of this protease slowed down presentation of peptides from this antigen in B lymphocytes (Manoury et al., 1998). Interestingly, the presence of N-glycans on antigens blocked degradation by asparaginyl endopeptidase, leading to the speculation that this protease might show a preference for prokaryotic, non-N-glycosylated antigens (Bogyo and Ploegh, 1998; Manoury et al., 1998). The question of whether in vivo proteases present in the endosomal/ lysosomal pathway are functionally redundant may be answered through targeted deletion of each of these proteases in mice. It is, however, unlikely that the class II system has evolved in such a way that the generation of presentable peptides would depend on one or a few well-defined proteases. More likely, antigenic peptides are generated throughout the whole endocytic pathway rather randomly, and the multiple proteases present within endosomes and lysosomes will ensure the generation of a multitude of peptides, some of which will serve as class II-restricted epitopes. As an alternative to processing of antigen prior to loading of the peptides onto class II molecules, the mechanism of so-called ‘‘determinant capture’’ has been proposed (Sette et al., 1989; Sercarz et al., 1993; Deng et al., 1993). In this model, proteins or large protein fragments are ‘‘captured’’ by receptive class II complexes, after which proteases trim down these relatively large fragments to a size appropriate for T cell stimulation. Proteolysis of these relatively large protein (fragments) could occur after degradation of Ii, for example, in the endocytic pathway (Busch et al., 1996; Fineschi and Miller, 1997). Also, many cells express proteases at their cell surface, and these proteases could act in concert

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

187

with class II complexes to generate epitopes from protein bound to the class II complex (Larsen et al., 1996). The relative contribution of the mechanism of determinant capture to the generation of epitope repertoire in antigen-presenting cells remains, however, to be determined. The multiple functions of the invariant chain in the generation of class II–peptide complexes have already been discussed. Recently, an additional function of Ii in the direct modulation of antigen processing was described. This function was attributed to the segment containing the 65-amino-acidresidue insertion in IiP43 (or IiP41 and in mouse cells; see Fig. 3). Whereas in mice both the normal and the extended forms have an identical function in class II targeting and maturation (Shachar et al., 1995; Naujokas et al., 1995; Takaesu et al., 1995), IiP41 can enhance the presentation of a subset of antigens (Peterson and Miller, 1992). Interestingly, the additional exon present in IiP41 was found to be complexed with one of the endosomal cathepsins (cathepsin L), for which it was a potent and specific inhibitor (Deng et al., 1993; Bevec et al., 1996). Moreover, overexpression of IiP41 alters the proteolytic environment where Ii degradation occurs and might also influence processing of antigens (Peterson and Miller, 1992; Fineschi et al., 1996). IiP41 levels vary greatly among different antigen-presenting cells (from 앑5% in B lymphocytes to greater than 40% in dendritic cells) (Kampgen et al., 1991; Engering et al., 1998) and therefore may contribute to the generation of different peptide repertoires in different antigenpresenting cell types. X. The Biology of a Prime Antigen-Presenting Cell: The Dendritic Cell System

Dendritic cells were first described by Zanvil Cohn and colleagues as a cell type present in lymphoid organs that was morphologically distinct from all other known cells present in that tissue (Steinman and Cohn, 1973). These cells do not express markers of B and T lymphocytes, lack certain properties known from phagocytes, and were found to be potent stimulators of T cell proliferation (Steinman and Witmer, 1978). Perhaps most importantly, dendritic cells have a unique capacity to initiate immune responses through the stimulation of naive T cells (Nussenzweig et al., 1980), whereas other antigen-presenting cells such as B cells and macrophages can only stimulate previously activated T lymphocytes. Dendritic cells are bone marrow derived and circulate through the blood in a precursor state, from where they seed peripheral organs. During this circulation, these precursor cells acquire the capacity to internalize antigens, and the expression of MHC class II complexes as well as that of Ii is induced (Engering et al., 1998). However, the class II/Ii complexes

188

PIETERS

synthesized in dendritic cells are only partially exported from the endoplasmic reticulum, providing a pool of class II molecules to be transported to class II compartments whenever they have arrived at the peripheral tissues (Engering et al., 1998). Upon arrival of precursor dendritic cells in these peripheral tissues, they differentiate into ‘‘immature’’ dendritic cells. Although immature dendritic cells are present in low numbers, they have a large capacity to internalize ligands through both fluid-phase and receptormediated uptake, as well as through phagocytosis (Inaba et al., 1993; Maurer and Stingl, 1995; Sallusto et al., 1995; Steinman and Swanson, 1995; Svensson et al., 1997; Engering et al., 1997a,b). In these immature dendritic cells, class II molecules remain to be synthesized but do not accumulate at the cell surface due to extensive and continuous recycling of these class II complexes between the plasma membrane and the intracellular MHC class II compartments (Cella et al., 1997a,b; Engering et al., 1998); see Fig. 4. In such a state, dendritic cells can rapidly acquire and process any incoming antigen, as well as load the resulting peptides onto class II molecules. The presence of dendritic cells in peripheral tissues becomes especially important in the early phases of an infection. In a bacterial infection, for example, both bacterial products (such as lipopolysaccharide) and inflammatory stimuli (tumor necrosis factor or TNF) induce the maturation of dendritic cells (Cella et al., 1997a, 1999; Watts, 1997b; Pierre et al., 1997). During maturation, several morphological and biochemical changes drastically alter the functional state of the dendritic cell. Within the first 10–20 h, the biosynthesis of class II molecules is transiently upregulated while the endocytic capacity remains high. Thus, dendritic cells retain a high capacity for peptide loading, and these peptides can assemble with both newly synthesized and recycling class II complexes (Cella et al., 1997a; Pierre et al., 1997). At the same time, inflammatory stimuli and bacterial products induce migration of dendritic cells to lymphoid tissue such as the spleen and lymph nodes (Nakamura et al., 1995; McWilliam et al., 1996; Adema et al., 1997). To ensure that during this migration the antigenic peptides derived from the infectious organisms are not replaced with peptides derived from material encountered during migration, both the endocytic capacity and the recycling of class II molecules are downregulated. In addition, the class II complexes, now loaded with antigenic peptides, become very stable, their half-life expanding from 앑30 to over 100 h (Cella et al., 1997a). These coordinated changes lead to an extremely potent antigenpresenting cell, precisely at the site where this is required, namely, in lymphoid tissue where naive T lymphocytes can be stimulated (Banchereau and Steinman, 1998). The molecular mechanisms that regulate these phe-

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

189

notypic changes are only beginning to be identified. In general, the precise subcellular distribution of class II complexes may be dictated by the degree and rate of Ii proteolysis (Pieters et al., 1991). In murine dendritic cells the degree of Ii proteolysis is altered during maturation: in immature dendritic cells Ii is inefficiently cleaved, directing the class II molecules to lysosomes. During maturation, Ii becomes more extensively proteolyzed, most probably due to enhanced cathepsin S activity, and this leads to expression of class II complexes at the plasma membrane (Pierre and Mellman, 1998). In human dendritic cells, however, no differences can be detected with respect to Ii proteolysis (Cella et al., 1997a; Engering and Pieters, unpublished). Instead, as already mentioned, in human immature dendritic cells class II molecules associate with the tetraspan CD63, a molecule known to be present in class II compartments (Metzelaar et al., 1991; Nijman et al., 1995; Hammond et al., 1998). Maturation induces the addition of glycans to CD63 (in the form of poly-Nlactosamine residues), a process that may be involved in the redistribution of class II molecules intracellularly (Engering et al., manuscript in preparation). Several other molecules are being characterized to analyze their involvement in MHC class II function at distinct stages during human dendritic cell development. XI. Subversion of MHC Class II-Restricted Antigen Presentation by Pathogens

The MHC class II antigen-presenting system in higher vertebrates has evolved as a means to combat extracellular-residing infectious agents, in most cases bacteria. Given the long-standing coevolution of microorganisms with higher vertebrates, it is not surprising that several bacteria have developed strategies to circumvent the processing and presentation of their antigens. Similar strategies have been developed by viruses, which can manipulate cell biological processes that generate MHC class I–peptide complexes for the activation of T killer cells (Ploegh, 1998). Whenever a bacterial species manages to survive and replicate within a susceptible host, it has the potential to become pathogenic. While the bacterium will replicate within the host and possibly be transmitted to other susceptible individuals, the host will develop immunity against subseqent challenges of these bacteria. Thus, host and bacteria are usually in a careful balance with one another (Falkow, 1991). But even the bacterial pathogen that is not of immediate threat to the healthy host needs to employ certain survival strategies, because otherwise it would be rapidly recognized as foreign and cleared by the innate and adaptive immune systems.

190

PIETERS

The strategies that pathogenic bacteria use to subvert immune mechanisms are not well known, but the evidence thus far suggests that bacteria may have developed the ability to circumvent several of the steps leading to their processing and the generation of class II-restricted epitopes. To avoid endosomal/lysosomal destruction, certain pathogens target cells that are unable to destroy them, which is the case for Salmonella and Yersinia spp. (Falkow, 1991; Bliska et al., 1993), while others (Listeria monocytogenesis, Shigella flexneri (Nhieu and Sansonetti, 1999)) escape into the cytoplasm; some (Listeria) can even spread from cell to cell without ever entering the circulation again (Kocks et al., 1992; Theriot, 1995; Sanders and Theriot, 1996; Ireton and Cossart, 1997). Yet another class of pathogens readily enters the endosomal pathway, but manages to arrest normal endosomal/lysosomal traffic once inside the phagocyte. An intriguing example of such a pathogen is provided by several Mycobacterium spp. (Russell et al., 1996). Mycobacteria can be phagocytosed through several cell surface receptor molecules, including the previously mentioned mannose receptors, as well as several complement receptors. The complement pathway (which received its name because it was found to ‘‘complement’’ the antibacterial activity of antibodies) is activated by antibodies bound to antigen or directly by pathogenic surfaces after these have bound certain serum components (Carroll, 1998). Rather then relying on the activation of serum complement, mycobacteria express a protein at their cell surface that can directly bind and activate complement locally, leading to the binding to complement receptors on the phagocyte, thus ensuring their internalization into macrophages (Schorey et al., 1997; Lachmann, 1998). Once inside the macrophage phagosome, mycobacteria manage to avoid the normal delivery to lysosomes by avoiding phagosome–lysosome fusion (Armstrong and D’Arcy Hart, 1971; Barker et al., 1997; Hasan et al., 1997). The mycobacterial phagosome lacks the proton-ATPase that is normally responsible for endosomal acidification (Mellman et al., 1985; SturgillKoszycki et al., 1994), and the resulting lack of acidification may contribute to mycobacterial survival inside these phagosomes. But what is the molecular mechanism involved in the inhibition of phagosome–lysosome fusion? Very recently, a molecule that is responsible for preventing the fusion of the mycobacterial phagosome with lysosomes was characterized. This protein, termed TACO (for tryptophane aspartatecontaining coat protein), is recruited to and actively retained at the mycobacterial phagosome, thereby preventing delivery of the mycobacteria to lysosomes (Ferrari et al., 1999). TACO is a normal component of the host phagocyte cytoskeleton, intimately associated with the microtubule network, and its precise function in uninfected macrophages is currently

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

191

unclear. During phagocytosis, TACO forms a scaffold, or coat, around newly formed phagosomes. Most probably, this TACO coat then must be removed from the phagosomal membrane in order to allow the machinery involved in phagosome–lysosome fusion to be recruited. It is likely that mycobacteria, in the course of their coevolution with their mammalian hosts, have gained the capacity to actively retain TACO on the phagosomal membrane, thereby preventing their delivery to lysosomes. By residing within the mycobacterial phagosome and thereby surviving within macrophages, mycobacteria can evade the normal bactericidal defense mechanisms of the host at several levels: First, the bacilli prevent their destruction in lysosomes by altering normal intracellular trafficking pathways (Ferrari et al., 1999). Second, by avoiding lysosomes, degradation of mycobacterial proteins, and thus the generation of MHC class II epitopes, is prevented, resulting in aberrant class II-restricted presentation, severely compromising the immune response (Pancholi et al., 1993). Third, once inside the macrophage, the mycobacteria also become invisible to circulating immune defense compounds, such as complement and antibodies. Thus, the strategies mentioned here that are employed by mycobacteria nicely illustrate the limitation of several levels of immune defense. This not only emphasizes the importance of alternative activation pathways (CD1, 웂/␦ T lymphocytes) but should also lead us to realize that whatever clever system has been developed by higher organisms, there might always be some pathogen that will manage to circumvent these defense mechanisms. XII. Conclusions

Activation of T lymphocytes through the action of MHC class II molecules complexed with antigenic peptides is now a well-understood process. The general mechanisms involved in the intracellular transport, peptide loading, and surface expression of these complexes have been thoroughly analyzed. In addition, through the generation of mice lacking one or more molecules that function in the MHC class II antigen-processing and presentation pathway, we can now also appreciate the relative contributions of the individual components in vivo. In contrast, the proteolytic systems that are involved in the generation of antigenic peptide from internalized material remain poorly characterized. Finally, understanding the biology of non-MHC-restricted activation of the immune system, as well as a better knowledge of the mechanisms of various pathogens to subvert the immune system, should be a major goal as we enter the new millennium. Not only will this research contribute to designing better strategies for antimicrobial treatments but, as has resulted

192

PIETERS

from work on MHC-restricted antigen presentation, it might also unravel basic cell biological mechanisms currently unknown. ACKNOWLEDGMENTS I thank Siamak Bahram, Heinz Jacobs, and Raul Torres for comments on the manuscript, and Anneke Engering, Donna Fluitsma, and Elizabeth Hoefsmit for providing the electron micrographs. The Basel Institute for Immunology was founded and is supported by Fa. Hoffmann–La Roche Ltd., Basel, Switzerland.

REFERENCES Adema, G. J., Hartgers, F., Verstraten, R., de Vries, E., Marland, G., Menon, S., Foster, J., Xu, Y., Nooyen, P., McClanahan, T., Bacon, K. B., and Figdor, C. G. (1997). A dendritic cell-derived C–C chemokine that preferentially attracts naive T cells. Nature 387, 713–717. Aluvihare, V. R., Khamlichi, A. A., Williams, G. T., Adorini, L., and Neuberger, M. S. (1997). Acceleration of intracellular targeting of antigen by the B-cell antigen receptor: importance depends on the nature of the antigen–antibody interaction. EMBO J. 16, 3553–3562. Amigorena, S., and Bonnerot, C. (1998). Role of B-cell and Fc receptors in the selection of T-cell epitopes. Curr. Opin. Immunol. 10, 88–92. Amigorena, S., Drake, J. R., Webster, P., and Mellman, I. (1994). Transient accumulation of new MHC molecules in a novel endocytic compartment in B lymphocytes. Nature 369, 113–120. Amigorena, S., Lankar, D., Briken, V., Gapin, L., Viguier, M., and Bonnerot, C. (1998). Type II and III receptors for immunoglobulin G (IgG) control the presentation of different T cell epitopes from single IgG-complexed antigens. J. Exp. Med. 187, 505–515. Anderson, H. A., and Roche, P. A. (1998). Phosphorylation regulates the delivery of MHC class II invariant chain complexes to antigen processing compartments. J. Immunol. 160, 4850–4858. Anderson, M. S., and Miller, J. (1992). Invariant chain can function as a chaperone protein for class II major histocompatibility complex molecules. Proc. Natl. Acad. Sci. USA 89, 2282–2286. Antoine, J. C., Jouanne, C., Lang, T., Prina, E., de Chastellier, C., and Frehel, C. (1991). Localization of major histocompatibility complex class II molecules in phagolysosomes of murine macrophages infected with Leishmania amazonensis. Infect. Immun. 59, 764–775. Armstrong, J., and D’Arcy Hart, P. (1971). Response of cultured macrophages to Mycobacterium tuberculosis, with observations on fusion of lysosomes with phagosomes. J. Exp. Med. 134, 713–740. Arneson, L. S., and Miller, J. (1995). Efficient endosomal localization of major histocompatibility complex class II-invariant chain complexes requires multimerization of the invariant chain targeting sequence. J. Cell. Biol. 129, 1217–1228. Arunachalam, B., and Cresswell, P. (1995). Molecular requirements for the interaction of class II major histocompatibility complex molecules and invariant chain with calnexin. J. Biol. Chem. 270, 2784–2790. Arunachalam, B., Lamb, C. A., and Cresswell, P. (1994). Transport properties of free and MHC class II-associated oligomers containing different isoforms of human invariant chain. Int. Immunol. 6, 439–451. Astarie-Dequeker, C., N’Diaye, E. N., Le Cabec, V., Rittig, M. G., Prandi, J., and Maridonneau-Parini, I. (1999). The mannose receptor mediates uptake of pathogenic and non-

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

193

pathogenic mycobacteria and bypasses bactericidal responses in human macrophages. Infect. Immun. 67, 469–477. Babbitt, B. P., Allen, P. M., Matsueda, G., Haber, E., and Unanue, E. R. (1985). Binding of immunogenic peptides to Ia histocompatibility molecules. Nature 317, 359–361. Bakke, O., and Dobberstein, B. (1990). MHC class II-associated invariant chain contains a sorting signal for endosomal compartments. Cell 63, 707–716. Banchereau, J., and Steinman, R. M. (1998). Dendritic cells and the control of immunity. Nature 392, 245–252. Barker, L. P., George, K. M., Falkow, S., and Small, P. L. (1997). Differential trafficking of live and dead Mycobacterium marinum organisms in macrophages. Infect. Immun. 65, 1497–1504. Baumeister, W., Walz, J., Zuhl, F., and Seemuller, E. (1998). The proteasome: Paradigm of a self-compartmentalizing protease. Cell 92, 367–380. Benacerraf, B. (1981). Role of MHC gene products in immune regulation. Science 212, 1229–1238. Benaceraff, B., and Germain, R. N. (1978). The immune response genes of the major histocompatibility complex. Immunol. Rev. 38, 70–119. Benacerraf, B., and McDevitt, H. O. (1972). Histocompatibility-linked immune response genes. Science 175, 273–279. Bennett, K., Levine, T., Ellis, J. S., Peanasky, R. J., Samloff, I. M., Kay, J., and Chain, B. M. (1992). Antigen processing for presentation by class II major histocompatibility complex requires cleavage by cathepsin E. Eur. J. Immunol. 22, 1519–1524. Bevec, T., Stoka, V., Pungercic, G., Dolenc, I., and Turk, V. (1996). Major histocompatibility complex class II-associated p41 invariant chain fragment is a strong inhibitor of lysosomal cathepsin L. J. Exp. Med. 183, 1331–1338. Bjorkman, P. J., Saper, M. A., Samraoui, B., Bennett, W. S., Strominger, J. L., and Wiley, D. C. (1987). Structure of the human class I histocompatibility antigen, HLA-A2. Nature 329, 506–512. Bliska, J. B., Galan, J. E., and Falkow, S. (1993). Signal transduction in the mammalian cell during bacterial attachment and entry. Cell 73, 903–920. Blobel, G., and Dobberstein, B. (1975a). Transfer of proteins across membranes I. Presence of proteolytically processed and unprocessed nascent immunoglobulin light chains on membrane-bound ribosomes of murine myeloma. J. Cell Biol. 67, 835–851. Blobel, G., and Dobberstein, B. (1975b). Transfer of proteins across membranes II. Reconstitution of functional rough microsomes from heterologous components. J. Cell Biol. 67, 852–862. Blum, J. S., and Cresswell, P. (1988). Role for intracellular proteases in the processing and transport of class II HLA antigens. Proc. Natl. Acad. Sci. USA 85, 3975–3979. Bodmer, J. G., Marsh, S. G., Albert, E. D., Bodmer, W. F., Bontrop, R. E., Charron, D., Dupont, B., Erlich, H. A., Fauchet, R., Mach, B., Mayr, W. R., Parham, P., Sasazuki, T., Schreuder, G. M., Strominger, J. L., Svejgaard, A., and Terasaki, P. I. (1997). Nomenclature for factors of the HLA system, 1996. Eur. J. Immunogenet 24, 105–151. Bogyo, M., and Ploegh, H. L. (1998). Antigen presentation. A protease draws first blood. Nature 396, 625, 627. Bonifacino, J. S. (1996). Reversal of fortune for nascent proteins. Nature 384, 405–406. Bonnerot, C., Lankar, D., Hanau, D., Spehner, D., Davoust, J., Salamero, J., and Fridman, W. H. (1995). Role of B cell receptor Ig alpha and Ig beta subunits in MHC class IIrestricted antigen presentation. Immunity 3, 335–347. Boss, J. M. (1997). Regulation of transcription of MHC class II genes. Curr. Opin. Immunol. 9, 107–113.

194

PIETERS

Brown, E. J. (1991). Complement receptors and phagocytosis. Curr. Opin. Immunol. 3, 76–82. Brown, E. J. (1995). Phagocytosis. Bioessays 17, 109–117. Brown, J. H., Jardetzky, T. S., Gorga, J. C., Stern, L. J., Urban, R. G., Strominger, J. L., and Wiley, D. C. (1993). Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1. Nature 364, 33–39. Burgess, T. L., and Kelly, R. B. (1987). Constitutive and regulated secretion of proteins. Annu. Rev. Cell Biol. 3, 243–293. Busch, R., Cloutier, I., Sekaly, R.-P., and Hammerling, G. J. (1996). Invariant chain protects class II histocompatibility antigens from binding intact polypeptides in the endoplasmic reticulum. EMBO J. 15, 418–428. Busch, R., and Mellins, E. D. (1996). Developing and shedding inhibitions: How MHC class II molecules reach maturity. Curr. Opin. Immunol. 8, 51–58. Caron, E., and Hall, A. (1998). Identification of two distinct mechanisms of phagocytosis controlled by different Rho GTPases. Science 282, 1717–1721. Carroll, M. C. (1998). The role of complement and complement receptors in induction and regulation of immunity. Annu. Rev. Immunol. 16, 545–568. Cella, M., Engering, A., Pinet, V., Pieters, J., and Lanzavecchia, A. (1997a). Inflammatory stimuli induce accumulation of MHC class II complexes on dendritic cells. Nature 388, 782–787. Cella, M., Sallusto, F., and Lanzavecchia, A. (1997b). Origin, maturation and antigen presenting function of dendritic cells. Curr. Opin. Immunol. 9, 10–16. Cella, M., Salio, M., Scheidegger, D., Engering, A., Pieters, J., Sallusto, A., and Lanzavecchia, A. (1999). Regulation of antigen capture, MHC biosynthesis and degradation optimizes presentation of infectious antigens in maturing dendritic cells. In ‘‘Dendritic cells— Biology and Clinical Applications’’ (M. Lotze and A. Thomson, eds.), Academic Press. Chang, C. H., and Flavell, R. A. (1995). Class II transactivator regulates the expression of multiple genes involved in antigen presentation. J. Exp. Med. 181, 765–767. Chang, C. H., Guerder, S., Hong, S. C., van Ewijk, W., and Flavell, R. A. (1996). Mice lacking the MHC class II transactivator (CIITA) show tissue-specific impairment of MHC class II expression. Immunity 4, 167–178. Chapman, H. A. (1998). Endosomal proteolysis and MHC class II function. Curr. Opin. Immunol. 10, 93–102. Charron, D. J., Aellen-Schulz, M. F., St. Geme, J. 3d, Erlich, H. A., and McDevitt, H. O. (1983). Biochemical characterization of an invariant polypeptide associated with Ia antigens in human and mouse. Mol. Immunol. 20, 21–32. Charron, D. J., and McDevitt, H. O. (1979). Analysis of HLA-D region-associated molecules with monoclonal antibody. Proc. Natl. Acad. Sci. USA 76, 6567–6571. Chen, W. J., Goldstein, J. L., and Brown, M. S. (1990). NPXY, a sequence often found in cytoplasmic tails, is required for coated pit-mediated internalization of the low density lipoprotein receptor. J. Biol. Chem. 265, 3116–3123. Chervonsky, A. V., Gordon, L., and Sant, A. (1994). A segment in the MHC class II 웁 chain plays a critical role in targeting class II molecules to the endocytic pathway. Int. Immunol. 6, 973–982. Chicz, R. M., Urban, R. G., Gorga, J. C., Vignali, D. A. A., Lane, W. S., and Strominger, J. L. (1993). Specificity and promiscuity among naturally processed peptides bound to HLA-DR alleles. J. Exp. Med. 178, 27–47. Chicz, R. M., Urban, R. G., Lane, W. S., Gorga, J. C., Stern, L. J., Vignali, D. A. A., and Strominger, J. L. (1992). Predominant naturally processed peptides bound to HLA-

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

195

DR1 are derived from MHC-related molecules and are heterogeneous in size. Nature 358, 764–768. Cho, S., Attaya, M., and Monaco, J. J. (1991). New class II-like genes in the murine MHC. Nature 353, 573–576. Claesson, L., Larhammar, D., Rask, L., and Peterson, P. A. (1983). cDNA clone for the human invariant chain of class II histocompatibility antigens and its implications for the protein structure. Proc. Natl. Acad. Sci. USA 80, 7395–7399. Claesson-Welsh, L., Barker, P. E., Larhammar, D., Rask, L., Ruddle, F. H., and Peterson, P. A. (1984). The gene encoding the human class II antigen-associated gamma chain is located on chromosome 5. Immunogenetics 20, 89–93. Clemens, D. L., and Horwitz, M. A. (1995). Characterization of the Mycobacterium tuberculosis phagosome and evidence that phagosomal maturation is inhibited. J. Exp. Med. 181, 257–270. Cresswell, P. (1992). Chemistry and functional role of the invariant chain. Curr. Opin. Immunol. 4, 87–92. Cresswell, P. (1994a). Antigen presentation. Getting peptides into MHC class II molecules. Curr. Biol. 4, 541–543. Cresswell, P. (1994b). Assembly, transport, and function of MHC class II molecules. Annu. Rev. Immunol. 12, 259–293. Cresswell, P. (1996). Invariant chain structure and MHC class II function. Cell 84, 505–507. Cresswell, P. (1998). Proteases, processing, and thymic selection. Science 280, 394–395. Cresswell, P., Blum, J. S., Kelner, D. N., and Marks, M. S. (1987). Biosynthesis and processing of class II histocompatibility antigens. Crit. Rev. Immunol. 7, 31–53. Daeron, M. (1997). Fc receptor biology. Annu. Rev. Immunol. 15, 203–234. Davis, C. G., van Driel, I. R., Russell, D. W., Brown, M. S., and Goldstein, J. L. (1987). The low density lipoprotein receptor. Identification of amino acids in cytoplasmic domain required for rapid endocytosis. J. Biol. Chem. 262, 4075–4082. De Kretser, T. A., Crumpton, M. J., Bodmer, J. G., and Bodmer, W. F. (1982). Twodimensional gel analysis of the polypeptides precipitated by a polymorphic HLA-DR1,2,w6 monoclonal antibody: Evidence for a third locus. Eur. J. Immunol. 12, 600–606. Dell’Angelica, E. C., Ohno, H., Ooi, C. E., Rabinovich, E., Roche, K. W., and Bonifacino, J. S. (1997). AP-3: An adaptor-like protein complex with ubiquitous expression. EMBO J. 16, 917–928. Demotz, S., Grey, H. M., Appella, E. and Sette, A. (1989). Characterization of a naturally processed MHC class II-restricted T- cell determinant of hen egg lysozyme. Nature 342, 682–684. Deng, H., Apple, R., Clare-Salzler, M., Trembleau, S., Mathis, D., Adorini, L., and Sercarz, E. (1993). Determinant capture as a possible mechanism of protection afforded by major histocompatibility complex class II molecules in autoimmune disease. J. Exp. Med. 178, 1675–1680. Denzin, L. K., and Cresswell, P. (1995). HLA-DM induces CLIP dissociation from MHC class II alpha beta dimers and facilitates peptide loading. Cell 82, 155–165. Denzin, L. K., Robbins, N. F., Carboy Newcomb, C., and Cresswell, P. (1994). Assembly and intracellular transport of HLA-DM and correction of the class II antigen-processing defect in T2 cells. Immunity 1, 595–606. Denzin, L. K., Sant’Angelo, D. B., Hammond, C., Surman, M. J., and Cresswell, P. (1997). Negative regulation by HLA-DO of MHC class II-restricted antigen processing. Science 278, 106–109. Desjardins, M., Huber, L. A., Parton, R. G., and Griffiths, G. (1994). Biogenesis of phagolysosomes proceeds through a sequential series of interactions with the endocytic apparatus. J. Cell. Biol. 124, 677–688.

196

PIETERS

Deussing, J., Roth, W., Saftig, P., Peters, C., Ploegh, H. L., and Villadangos, J. A. (1998). Cathepsins B and D are dispensable for major histocompatibility complex class II-mediated antigen presentation. Proc. Natl. Acad. Sci. USA 95, 4516–4521. Dobberstein, B. (1987). Structure and function of the signal recognition particle (SRP). Mol. Biol. Rep. 12, 213–217. Drickamer, K., and Taylor, M. E. (1993). Biology of animal lectins. Annu. Rev. Cell Biol. 9, 237–264. Engering, A. J. (1998). Regulation of major histocompatibility complex class II-restricted antigen presentation by human dendritic cells. In ‘‘Department of Cell Biology and Immunology,’’ pp. 143. Free University, Amsterdam. Engering, A. J., Cella, M., Fluitsma, D. M., Hoefsmit, E. C. M., Lanzavecchia, A., and Pieters, J. (1997a). The mannose receptor functions as a high capacity and broad specificity antigen receptor in human dendritic cells. Eur. J. Immunol. 27, 2417–2425. Engering, A. J., Cella, M., Fluitsma, D. M., Hoefsmit, E. C. M., Lanzavecchia, A., and Pieters, J. (1997b). Mannose receptor mediated antigen uptake and presentation in human dendritic cells. Adv. Exp. Med. Biol. 417, 183–187. Engering, A. J., Richters, C. D., Fluitsma, D. M., van Pelt, A. M., Kamperdijk, E. W., Hoefsmit, E. C., and Pieters, J. (1998). MHC class II and invariant chain biosynthesis and transport during maturation of human precursor dendritic cells. Int. Immunol. 10, 1713–1723. Ezekowitz, R. A., Sastry, K., Bailly, P., and Warner, A. (1990). Molecular characterization of the human macrophage mannose receptor: Demonstration of multiple carbohydrate recognition-like domains and phagocytosis of yeast in Cos-1 cells. J. Exp. Med. 172, 1785– 1794. Ezekowitz, R. A., Williams, D. J., Koziel, H., Armstrong, M. Y., Warner, A., Richards, F. F., and Rose, R. M. (1991). Uptake of Pneumocystis carinii mediated by the macrophage mannose receptor. Nature 351, 155–158. Falk, K., Rotzschke, O., Stevanovic, S., Jung, G., and Rammensee, H. G. (1991). Allelespecific motifs revealed by sequencing of self-peptides eluted from MHC molecules. Nature 351, 290–296. Falk, K., Rotzschke, O., Stevanovic, S., Jung, G., and Rammensee, H. G. (1994). Pool sequencing of natural HLA-DR, DQ, and DP ligands reveals detailed peptide motifs, constraints of processing, and general rules. Immunogenetics 39, 230–242. Falk, K., Rotzschke, O., Takiguchi, M., Gnau, V., Stevanovic, S., Jung, G., and Rammensee, H. G. (1995). Peptide motifs of HLA-B51, -B52 and -B78 molecules, and implications for Behcet’s disease. Int. Immunol. 7, 223–228. Falkow, S. (1991). Bacterial entry into eukaryotic cells. Cell 65, 1099–1102. Ferrari, G., Knight, A. M., Watts, C., and Pieters, J. (1997). Distinct intracellular compartments involved in invariant chain degradation and antigenic peptide loading of major histocompatibility complex (MHC) class II molecules. J. Cell Biol. 139, 1433–1446. Ferrari, G., Naito, M., Langen, H., and Pieters, J. (1999). A coat protein on phagosomes involved in the intracellular survival of Mycobacteria. Cell 97, 435–447. Fineschi, B., and Miller, J. (1997). Endosomal proteases and antigen processing. Trends Biochem. Sci. 22, 377–382. Fineschi, B., Sakaguchi, K., Appella, E., and Miller, J. (1996). The proteolytic environment involved in MHC class II-restricted antigen presentation can be modulated by the p41 form of invariant chain. J. Immunol. 157, 3211–3215. Fling, S. P., Arp, B., and Pious, D. (1994). HLA-DMA and -DMB genes are both required for MHC class II/peptide complex formation in antigen-presenting cells. Nature 368, 554–560.

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

197

Fraser, I. P., Koziel, H., and Ezekowitz, R. A. (1998). The serum mannose-binding protein and the macrophage mannose receptor are pattern recognition molecules that link innate and adaptive immunity. Sem. Immunol. 10, 363–372. Freisewinkel, I. M., Schenck, K., and Koch, N. (1993). The segment of invariant chain that is critical for association with major histocompatibility complex class II molecules contains the sequence of a peptide eluted from class II polypeptides. Proc. Natl. Acad. Sci. USA 90, 9703–9706. Fremont, D. H., Matsumura, M., Stura, E. A., Peterson, P. A., and Wilson, I. A. (1992). Crystal structures of two viral peptides in complex with murine MHC class I H-2Kb Science 257, 919–927. Fung-Leung, W.-P., Surh, C. D., Liljedahl, M., Pang, J., Leturcq, D., Peterson, P. A., Webb, S. R., and Karlsson, L. (1996). Antigen presentation and T cell development in H2-M-deficient mice. Science 271, 1278–1281. Garrett, T. P., Saper, M. A., Bjorkman, P. J., Strominger, J. L., and Wiley, D. C. (1989). Specificity, pockets for the side chains of peptide antigens in HLA-Aw68. Nature 342, 692–696. Gaynor, C. D., McCormack, F. X., Voelker, D. R., McGowan, S. E., and Schlesinger, L. S. (1995). Pulmonary surfactant protein. A mediates enhanced phagocytosis of Mycobacterium tuberculosis by a direct interaction with human macrophages. J. Immunol. 155, 5343–5351 Genuardi, M., and Saunders, G. F. (1988). Localization of the HLA class II-associated invariant chain gene to human chromosome band 5q32. Immunogenetics 28, 53–56. Germain, R. N. (1994). MHC-dependent antigen processing and peptide presentation: Providing ligands for T lymphocyte activation. Cell 76, 287. Ghosh, P., Amaya, M., Mellins, E., and Wiley, D. C. (1995). The structure of an intermediate in class II MHC maturation: CLIP bound to HLA-DR3. Nature 378, 457–462. Gould, K., Cossins, J., Bastin, J., Brownlee, G. G., and Townsend, A. (1989). A 15 amino acid fragment of influenza nucleoprotein synthesized in the cytoplasm is presented to class I-restricted cytotoxic T lymphocytes. J. Exp. Med. 170, 1051–1056. Greenberg, S. (1995). Signal transduction of phagocytosis. Trends Cell Biol. 5, 93–99. Greenberg, S., Burridge, K., and Silverstein, S. C. (1990). Colocalization of F-actin and talin during Fc receptor-mediated phagocytosis in mouse macrophages. J. Exp. Med. 172, 1853–1856. Greenberg, S., Chang, P., and Silverstein, S. C. (1993). Tyrosine phosphorylation is required for Fc receptor-mediated phagocytosis in mouse macrophages. J. Exp. Med. 177, 529–534. Griffiths, G., Hoflack, B., Simons, K., Mellman, I., and Kornfeld, S. (1988). The mannose 6-phosphate receptor and the biogenesis of lysosomes. Cell 52, 329–341. Griffiths, G., and Simons, K. (1986). The trans Golgi network: Sorting at the exit site of the golgi complex. Science 234, 438. Griffiths, G. M. (1995). The cell biology of CTL killing. Curr. Opin. Immunol. 7, 343–348. Griffiths, G. M. (1996). Secretory lysosomes—A special mechanism of regulated secretion in haemopoietic cells. Trend Cell Biol. 6, 329–332. Hammond, C., Denzin, L. K., Pan, M., Griffith, J. M., Geuze, H. J., and Cresswell, P. (1998). The tetraspan protein CD82 is a resident of MHC class II compartments where it associates with HLA-DR, -DM, and -DO molecules. J. Immunol. 161, 3282–3291. Hammond, C., and Helenius, A. (1995). Quality control in the secretory pathway. Curr. Opin. Cell Biol. 7, 523–529. Hansen, T. H., Carreno, B. M., and Sachs, D. H. (1993). The major histocompartibility complex. In ‘‘Fundamental Immunology’’ (W. E. Paul, ed.), pp. 577–628. Raven Press, Ltd., New York.

198

PIETERS

Hasan, Z., Schlax, C., Kuhn, L., Lefkovits, I., Young, D., Thole, J., and Pieters, J. (1997). Isolation and characterization of the mycobacterial phagosome: Segregation from the endosomal/lysosomal pathway. Mol. Microbiol. 24, 545–553. Hulett, M. D., and Hogarth, P. M. (1994). Molecular basis of Fc receptor function. Adv. Immunol. 57, 1–127. Hurtley, S. M., and Helenius, A. (1989). Protein oligomerization in the endoplasmic reticulum. Annu. Rev. Cell Biol. 5, 277–307. Ignatowicz, L., Kappler, J., and Marrack, P. (1996). The repertoire of T cells shaped by a single MHC/peptide ligand. Cell 84, 521–529. Inaba, K., Inaba, M., Naito, M., and Steinman, R. (1993). Dendritic cell progenitors phagocytose particulates, including Bacillus Calmette–Guerin organisms, and sensitize mice to mycobacterial antigens in vivo. J. Exp. Med. 178, 479–488. Inoko, H., Ando, A., Kimura, M., and Tsuji, K. (1985). Isolation and characterization of the cDNA clone and genomic clones of a new HLA class II antigen heavy chain, DO alpha. J. Immunol. 135, 2156–2159. Ireton, K., and Cossart, P. (1997). Host–pathogen interactions during entry and actin-based movement of Listeria monocytogenes. Annu. Rev. Genet. 31, 113–138. Jackson, T. (1998). Transport vesicles: Coats of many colours. Curr. Biol. 8, R609–612. Janeway, C. J., and Travers, P. (1996). ‘‘Immunobiology.’’ Current Biology Ltd./Garland Publishing Inc., London, U.K. Jensen, P. E. (1998). Antigen processing: HLA-DO—A hitchhiking inhibitor of HLA-DM. Curr. Biol. 8, R128–131. Jiang, W., Swiggard, W. J., Heufler, C., Peng, M., Mirza, A., Steinman, R. M., and Nussenzweig, M. C. (1995). The receptor DEC-205 expressed by dendritic cells and thymic epithelial cells is involved in antigen processing. Nature 375, 151–155. Joiner, K. A., Fuhrman, S. A., Miettinen, H. M., Kasper, L. H., and Mellman, I. (1990). Toxoplasma gondii: Fusion competence of parasitophorous vacuoles in Fc receptor-transfected fibroblasts. Science 249, 641–646. Jones, P. P., Murphy, D. B., Hewgill, D., and McDevitt, H. O. (1979). Detection of a common polypeptide chain in I–A and I–E sub-region immunoprecipitates. Mol. Immunol. 16, 51–60. Kampgen, E., Koch, N., Koch, F., Stoger, P., Heufler, C., Schuler, G., and Romani, N. (1991). Class II major histocompatibility complex molecules of murine dendritic cells: Synthesis, sialylation of invariant chain, and antigen processing capacity are downregulated upon culture. Proc. Natl. Acad. Sci. USA 88, 3014–3018. Karlsson, L., and Peterson, P. A. (1992). The alpha chain gene of H-2O has an unexpected location in the major histocompatibility complex. J. Exp. Med. 176, 477–483. Karlsson, L., Surh, C. D., Sprent, J., and Peterson, P. A. (1991). A novel class II MHC molecule with unusual tissue distribution. Nature 351, 485–488. Kaufman, J. F., Auffray, C., Korman, A. J., Shackelford, D. A., and Strominger, J. (1984). The class II molecules of the human and murine major histocompatibility complex. Cell 36, 1–13. Kaufmann, S. H., and Hess, J. (1999). Impact of intracellular location of and antigen display by intracellular bacteria: Implications for vaccine development. Immunol. Lett. 65, 81–84. Kelly, A. P., Monaco, J. J., Cho, S. G., and Trowsdale, J. (1991). A new human HLA class II-related locus, DM. Nature 353, 571–573. Kern, I., Steimle, V., Siegrist, C. A., and Mach, B. (1995). The two novel MHC class II transactivators RFX5 and CIITA both control expression of HLA-DM genes. Int. Immunol. 7, 1295–1299.

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

199

Kirchhausen, T., Bonifacino, J. S., and Riezman, H. (1997). Linking cargo to vesicle formation: Receptor tail interactions with coat proteins. Curr. Opin. Cell Biol. 9, 488–495. Klausner, R. D., and Sitia, R. (1990). Protein degradation in the endoplasmic reticulum. Cell 62, 611–614. Knight, A. M., Luccocq, J. M., Prescott, A. R., Ponnambalam, S., and Watts, C. (1997). Antigen endocytosis and presentation mediated by human membrane IgG1 in the absence of the Ig움/Ig웁 dimer. EMBO J. 16, 3842–3850. Koch, N., Lauer, W., Habicht, J., and Dobberstein, B. (1987). Primary structure of the gene for the murine Ia antigen-associated invariant chains (Ii). An alternatively spliced exon encodes a cysteine-rich domain highly homologous to a repetitive sequence of thyroglobulin. EMBO J. 6, 1677–1683. Koch, N., Lipp, J., Pessara, U., Schenck, K., Wraight, C., and Dobberstein, B. (1989). MHC class II invariant chains in antigen processing and presentation. Trends Biochem. Sci. 14, 383–386. Kocks, C., Gouin, E., Tabouret, M., Berche, P., Ohayon, H., and Cossart, P. (1992). L. monocytogenes-induced actin assembly requires the actA gene product, a surface protein. Cell 68, 521–531. Kodama, T., Doi, T., Suzuki, H., Takahashi, K., Wada, Y., and Gordon, S. (1996). Collagenous macrophage scavenger receptors. Curr. Opin. Lipidol. 7, 287–291. Kornfeld, S., and Mellman, I. (1989). The biogenesis of lysosomes. Annu. Rev. Cell Biol. 5, 483–525. Kropshofer, H., Hammerling, G. J., and Vogt, A. B. (1997). How HLA-DM edits the MHC class II peptide repertoire: Survival of the fittest? Immunol. Today 18, 77–82. Kruskal, B. A., Sastry, K., Warner, A. B., Mathieu, C. E., and Ezekowitz, R. A. (1992). Phagocytic chimeric receptors require both transmembrane and cytoplasmic domains from the mannose receptor. J. Exp. Med. 176, 1673–1680. Kvist, S., Wiman, K., Claesson, L., Peterson, P. A., and Dobberstein, B. (1982). Membrane insertion and oligomeric assembly of HLA-DR histocompatibility antigens. Cell 29, 61–69. Lachmann, P. J. (1998). Microbial immunology: a new mechanism for immune subversion. Curr. Biol. 8, R99–R101. Lamb, C. A., and Cresswell, P. (1992). Assembly and transport properties of invariant chain trimers and HLA-DR-invariant chain complexes. J. Immunol. 148, 3478–3482. Lang, T., Hellio, R., Kaye, P. M., and Antoine, J. C. (1994). Leishmania donovani-infected macrophages: Characterization of the parasitophorous vacuole and potential role of this organelle in antigen presentation. J. Cell Sci. 107, 2137–2150. Lanzavecchia, A. (1985). Antigen-specific interaction between T and B cells. Nature 314, 537–539. Lanzavecchia, A. (1990). Receptor-mediated antigen uptake and its effect on antigen presentation to class II-restricted T lymphocytes. Annu. Rev. Immunol. 8, 773–793. Larsen, S. L., Pedersen, L. O., Buus, S., and Stryhn, A. (1996). T cell responses affected by aminopeptidase N (CD13)-mediated trimming of major histocompatibility complex class II-bound peptides [published erratum appears in J. Exp. Med. 1996, Nov. 1; 184(5): 2073]. J. Exp. Med. 184, 183–189. Le Borgne, R., and Hoflack, B. (1998). Mechanisms of protein sorting and coat assembly: Insights from the clathrin-coated vesicle pathway. Curr. Opin. Cell Biol. 10, 499–503. Letourneur, F., and Klausner, R. D. (1992). A novel di-leucine motif and a tyrosine-based motif independently mediate lysosomal targeting and endocytosis of CD3 chains. Cell 69, 1143–1157. Liljedahl, M., Kuwana, T., Fungleung, W. P., Jackson, M., Peterson, P. A., and Karlsson, L. (1996). HLA-DO is a lysosomal resident which requires association with HLA-DM for efficient intracellular transport. EMBO J 15, 4817–4824.

200

PIETERS

Liljedahl, M., Winqvist, O., Surh, C. D., Wong, P., Ngo, K., Teyton, L., Peterson, P. A., Brunmark, A., Rudensky, A. Y., Fung-Leung, W. P., and Karlsson, L. (1998). Altered antigen presentation in mice lacking H2-O. Immunity 8, 233–243. Lindner, R., and Unanue, E. R. (1996). Distinct antigen MHC class II complexes generated by separate processing pathways. EMBO J. 15, 6910–6920. Lindstedt, R., Liljedahl, M., Peleraux, A., Peterson, P. A., and Karlsson, L. (1995). The MHC class II molecule H2-M is targeted to an endosomal compartment by a tyrosinebased targeting motif. Immunity 3, 561–572. Ljunggren, H.-G., Stam, N. J., Ohlen, C., Neefjes, J. J., Hoglund, P., Heemels, M. T., Bastin, J., Schumacher, T. N. M., Townsend, A., Karre, K., and Ploegh, H. L. (1990). Empty MHC Class I molecules come out in the cold. Nature 346, 476–480. Lotteau, V., Teyton, L., Peleraux, A., Nilsson, T., Karlsson, L., Schmid, S. L., Quaranta, V., and Peterson, P. A. (1990). Intracellular transport of class II MHC molecules directed by invariant chain. Nature 348, 600–605. Mach, B., Steimle, V., and Reith, W. (1994). MHC class II-deficient combined immunodeficiency: A disease of gene regulation. Immunol. Rev. 138, 207–221. Machamer, C. E., and Cresswell, P. (1982). Biosynthesis and glycosylation of the invariant chain associated with HLA-DR antigens. J. Immunol. 129, 2564–2569. Manoury, B., Hewitt, E. W., Morrice, N., Dando, P. M., Barrett, A. J., and Watts, C. (1998). An asparaginyl-endopeptidase processes a microbial antigen for class II MHC presentation. Nature 396, 695–699. Marks, M. S., Blum, J. S., and Cresswell, P. (1990). Invariant chain trimers are sequestered in the rough endoplasmic reticulum in the absence of association with HLA class II antigens. J. Cell Biol. 111, 839–855. Marks, M. S., Germain, R. N., and Bonifacino, J. S. (1995a). Transient aggregation of major histocompatibility complex class II chains during assembly in normal spleen cells. J. Biol. Chem. 270, 10475–10481. Marks, M. S., Roche, P. A., van Donselaar, E., Woodruff, L., Peters, P. J., and Bonifacino, J. S. (1995b). A lysosomal targeting signal in the cytoplasmic tail of the beta chain directs HLA-DM to MHC class II compartments. J. Cell Biol. 131, 351–369. Marks, M. S., Woodruff, L., Ohno, H., and Bonifacino, J. S. (1996). Protein targeting by tyrosine- and di-leucine-based signals: Evidence for distinct saturable components. J. Cell Biol. 135, 341–354. Martin, R., McFarland, H. F., and McFarlin, D. E. (1992). Immunological aspects of demyelinating diseases. Annu. Rev. Immunol. 10, 153–187. Martin, W. D., Hicks, G. G., Mendiratta, S. K., Leva, H. I., Ruley, H. E., and Vankaer, L. (1996). H2-M mutant mice are defective in the peptide loading of class II molecules, antigen presentation, and T cell repertoire selection. Cell 84, 543–550. Matsumura, M., Fremont, D. H., Peterson, P. A., and Wilson, I. A. (1992). Emerging principles for the recognition of peptide antigens by MHC class I molecules. Science 257, 927–934. Maurer, D., and Stingl, G. (1995). Immunoglobulin E-binding structures on antigenpresenting cells present in skin and blood. J. Invest. Dermatol. 104, 707–710. McDevitt, H. O., and Bodmer, W. F. (1974). HL-A, immune-response genes, and disease. Lancet 1, 1269–1275. McDevitt, H. O., and Wakeland, E. K. (1998). Autoimmunity. Curr. Opin. Immunol. 10, 647–648. McWilliam, A. S., Napoli, S., Marsh, A. M., Pemper, F. L., Nelson, D. J., Pimm, C. L., Stumbles, P. A., Wells, T. N., and Holt, P. G. (1996). Dendritic cells are recruited into

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

201

the airway epithelium during the inflammatory response to a broad spectrum of stimuli. J. Exp. Med. 184, 2429–2432. Mellins, E., Smith, L., Arp, B., Cotner, T., Celis, E., and Pious, D. (1990). Defective processing and presentation of exogenous antigens in mutants with normal HLA class II genes. Nature 343, 71–74. Mellman, I. (1996). Endocytosis and molecular sorting. Annu. Rev. Cell Dev. Biol. 12, 575–625. Mellman, I., Fuchs, R., and Helenius, A. (1985). Acidification of endocytic and exocytic pathways. Annu. Rev. Biochem. 55, 663–700. Mengle-Gaw, L., and McDevitt, H. O. (1985). Genetics and expression of mouse Ia antigens. Annu. Rev. Immunol. 3, 367–396. Metzelaar, M. J., Wijngaard, P. L., Peters, P. J., Sixma, J. J., Nieuwenhuis, H. K., and Clevers, H. C. (1991). CD63 antigen. A novel lysosomal membrane glycoprotein, cloned by a screening procedure for intracellular antigens in eukaryotic cells. J. Biol. Chem. 266, 3239–3245. Miller, W. J., Hatch, J. A., Simonis, S., and Cullen, S. E. (1988). Identification of the glycosaminoglycan-attachment site of mouse invariant-chain proteoglycan core protein by site-directed mutagenesis. Proc. Natl. Acad. Sci. USA 85, 1359–1363. Miyazaki, T., Wolf, P., Tourne, S., Waltzinger, C., Dierich, A., Barois, N., Ploegh, H., Benoist, C., and Mathis, D. (1996). Mice lacking H2-M complexes, enigmatic elements of the MHC class II peptide-loading pathway. Cell 84, 531–541. Mordue, D. G., and Sibley, L. D. (1997). Intracellular fate of vacuoles containing Toxoplasma gondii is determined at the time of formation and depends on the mechanism of entry. J. Immunol. 159, 4452–4459. Morris, P., Shaman, J., Attaya, M., Amaya, M., Goodman, S., Bergman, C., Monaco, J. J., and Mellins, E. (1994). An essential role for HLA-DM in antigen presentation by class II major histocompatibility molecules. Nature 368, 551–554. Nag, B., Passmore, D., Kendrick, T., Bhayani, H., and Sharma, S. D. (1992). N-linked oligosaccharides of murine major histocompatibility complex class II molecule. Role in antigenic peptide binding, T cell recognition, and clonal nonresponsiveness. J. Biol. Chem. 267, 22624–22629. Nag, B., Wada, H. G., Arimilli, S., Fok, K., Passmore, D., Sharma, S. D., and McConnell, H. M. (1994). The role of N-linked oligosaccharides of MHC class II antigens in T cell stimulation. J. Immunol. Methods 172, 95–104. Nakagawa, T., Roth, W., Wong, P., Nelson, A., Farr, A., Deussing, J., Villadangos, J. A., Ploegh, H., Peters, C., and Rudensky, A. Y. (1998). Cathepsin L: Critical role in Ii degradation and CD4 T cell selection in the thymus. Science 280, 450–453. Nakagawa, T. Y., Brissette, W. H., Lira, P. D., Griffiths, R. J., Petrushova, N., Stock, J., McNeish, J. D., Eastman, S. E., Howard, E. D., Clarke, S. R., Rosloniec, E. F., Elliott, E. A., and Rudensky, A. Y. (1999). Impaired invariant chain degradation and antigen presentation and diminished collagen-induced arthritis in cathepsin S null mice. Immunity 10, 207–217. Nakamura, K., Williams, I. R., and Kupper, T. S. (1995). Keratinocyte-derived monocyte chemoattractant protein 1 (MCP-1): Analysis in a transgenic model demonstrates MCP1 can recruit dendritic and Langerhans cells to skin. J. Invest. Dermatol. 105, 635–643. Naujokas, M. F., Arneson, L. S., Fineschi, B., Peterson, M. E., Sitterding, S., Hammond, A. T., Reilly, C., Lo, D., and Miller, J. (1995). Potent effects of low levels of MHC class II-associated invariant chain on CD4⫹ T cell development. Immunity 3, 359–372. Naujokas, M. F., Morin, M., Anderson, M. S., Peterson, M., and Miller, J. (1993). The chondroitin sulfate form of invariant chain can enhance stimulation of T cell responses through interaction with CD44. Cell 74, 257–268.

202

PIETERS

Nepom, G. T., and Erlich, H. (1991). MHC class-II molecules and autoimmunity. Annu. Rev. Immunol. 9, 493–525. Neuberger, M. S., Patel, K. J., Dariavach, P., Nelms, K., Peaker, C. J., and Williams, G. T. (1993). The mouse B-cell antigen receptor: definition and assembly of the core receptor of the five immunoglobulin isotypes. Immunol. Rev. 132, 147–161. Newcomb, J. R., and Cresswell, P. (1993). Characterization of endogenous peptides bound to purified HLA-DR molecules and their absence from invariant chain-associated alpha beta dimers. J. Immunol. 150, 499–507. Nhieu, G. T., and Sansonetti, P. J. (1999). Mechanism of Shigella entry into epithelial cells. Curr. Opin. Microbiol. 2, 51–55. Nijman, H. W., Kleijmeer, M. J., Ossevoort, M. A., Oorschot, V. M., Vierboom, M. P., van de Keur, M., Kenemans, P., Kast, W. M., Geuze, H. J., and Melief, C. J. (1995). Antigen capture and major histocompatibility class II compartments of freshly isolated and cultured human blood dendritic cells. J. Exp. Med. 182, 163–174. Nussenzweig, M. C., Steinman, R. M., Gutchinov, B., and Cohn, Z. A. (1980). Dendritic cells are accessory cells for the development of anti-trinitrophenyl cytotoxic T lymphocytes. J. Exp. Med. 152, 1070–1084. Odorizzi, C. G., Trowbridge, I. S., Xue, L., Hopkins, C. R., Davis, C. D., and Collawn, J. F. (1994). Sorting signals in the MHC class II invariant chain cytoplasmic tail and transmembrane region determine trafficking to an endocytic processing compartment. J. Cell Biol. 126, 317–330. Ohno, H., Stewart, J., Fournier, M. C., Bosshart, H., Rhee, I., Miyatake, S., Saito, T., Gallusser, A., Kirchhausen, T., and Bonifacino, J. S. (1995). Interaction of tyrosine-based sorting signals with clathrin-associated proteins. Science 269, 1872–1875. Owen, M. J., Kissonerghis, A. M., Lodish, H. F., and Crumpton, M. J. (1981). Biosynthesis and maturation of HLA-DR antigens in vivo. J. Biol. Chem. 256, 8987–8993. Palade, G. E. (1975). Intracellular aspects of the process of protein synthesis. Science 189, 347–358. Pamer, E., and Cresswell, P. (1998). Mechanisms of MHC class I-restricted antigen processing. Annu. Rev. Immunol. 16, 323–358. Pancholi, P., Mirza, A., Bhardwaj, N., and Steinman, R. M. (1993). Sequestration from immune CD4⫹ T cells of mycobacteria growing in human macrophages. Science 260, 984–986. Peleraux, A., Karlsson, L., Chambers, J., and Peterson, P. A. (1996). Genomic organization of a mouse MHC class II region including the H2-M and Lmp2 loci. Immunogenetics 43, 204–214. Peters, P. J., Neefjes, J. J., Oorschot, V., Ploegh, H. L., and Geuze, H. J. (1991). Segregation of MHC Class II molecules from MHC Class I molecules in the Golgi complex for transport to lysosomal compartments. Nature 349, 669–676. Peterson, M., and Miller, J. (1992). Antigen presentation enhanced by the alternatively spliced invariant chain gene product p41. Nature 357, 596–598. Pierre, P., and Mellman, I. (1998). Developmental regulation of invariant chain proteolysis controls MHC class II trafficking in mouse dendritic cells. Cell 93, 1135–1145. Pierre, P., Turley, S. J., Gatti, E., Hull, M., Meltzer, J., Mirza, A., Inaba, K., Steinman, R. M., and Mellman, I. (1997). Developmental regulation of MHC class II transport in mouse dendritic cells. Nature 388, 787–792. Pieters, J. (1997a). MHC class II compartments: Specialized organelles of the endocytic pathway in antigen presenting cells. Biol. Chem. 378, 751–758. Pieters, J. (1997b). MHC class II restricted antigen presentation. Curr. Opin. Immunol. 9, 89–96.

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

203

Pieters, J. (1999). Processing and presentation of phagocytosed antigens to the immune system. In ‘‘Phagocytosis: The Host’’ (S. Gordon, ed.), JAI Press, Stamford, CT. Pieters, J., Bakke, O., and Dobberstein, B. (1993). The MHC class II-associated Invariant chain contains two endosomal sorting signals within its cytoplasmic tail. J. Cell Sci. 106, 831–846. Pieters, J., Horstmann, H., Bakke, O., Griffiths, G., and Lipp, J. (1991). Intracellular transport and localization of major histocompatibility complex class II molecules and associated invariant chain. J. Cell Biol. 115, 1213–1223. Pinet, V., Vergelli, M., Martin, R., Bakke, O., and Long, E. O. (1995). Antigen presentation mediated by recycling of surface HLA-DR molecules. Nature 375, 603–606. Pishvaee, B., and Payne, G. S. (1998). Clathrin coats—Threads laid bare. Cell 95, 443–446. Pleiman, C. M., D’Ambrosio, D., and Cambier, J. C. (1994). The B-cell antigen receptor complex: Structure and signal transduction. Immunol. Today 15, 393–399. Ploegh, H. L. (1998). Viral strategies of immune evasion. Science 280, 248–253. Pober, J. S., Collins, T., Gimbrone, M. A., Jr., Cotran, R. S., Gitlin, J. D., Fiers, W., Clayberger, C., Krensky, A. M., Burakoff, S. J., and Reiss, C. S. (1983). Lymphocytes recognize human vascular endothelial and dermal fibroblast Ia antigens induced by recombinant immune interferon. Nature 305, 726–729. Pond, L., and Watts, C. (1997). Characterization of transport of newly assembled, T cellstimulatory MHC class II-peptide complexes from MHC class II compartments to the cell surface. J. Immunol. 159, 543–553. Pontow, S. E., Kery, V., and Stahl, P. D. (1992). Mannose receptor. Int. Rev. Cytol. 137B, 221–244. Quaranta, V., Majdic, O., Stingl, G., Liszka, K., Honigsmann, H., and Knapp, W. (1984). A human Ia cytoplasmic determinant located on multiple forms of invariant chain (gamma, gamma 2, gamma 3). J. Immunol. 132, 1900–1905. Rabinowitz, S., Horstmann, H., Gordon, S., and Griffiths, G. (1992). Immunocytochemical characterization of the endocytic and phagolysosomal compartments in peritoneal macrophages. J. Cell Biol. 116, 95–112. Rammensee, H. G. (1995). Chemistry of peptides associated with MHC class I and class II molecules. Curr. Opin. Immunol. 7, 85–96. Rammensee, H. G., Falk, K., and Rotzschke, O. (1993a). MHC molecules as peptide receptors. Curr. Opin. Immunol. 5, 35–44. Rammensee, H. G., Falk, K., and Rotzschke, O. (1993b). Peptides naturally presented by MHC class I molecules. Annu. Rev. Immunol. 11, 213–244. Rapoport, I., Chen, Y. C., Cupers, P., Shoelson, S. E., and Kirchhausen, T. (1998). Dileucinebased sorting signals bind to the beta chain of AP-1 at a site distinct and regulated differently from the tyrosine-based motif-binding site. EMBO J. 17, 2148–2155. Raposo, G., Nijman, H. W., Stoorvogel, W., Leijendekker, R., Harding, C. V., Melief, C. J. M., and Geuze, H. J. (1996). B lymphocytes secrete antigen-presenting vesicles. J. Exp. Med. 183, 1161–1172. Ravetch, J. V., and Kinet, J. P. (1991). Fc receptors. Annu. Rev. Immunol. 9, 457–492. Reith, W., Steimle, V., Durand, B., Kobr, M., and Mach, B. (1995). Regulation of MHC class II gene expression. Immunobiology 193, 248–253. Reth, M. (1989). Antigen receptor tail clue. Nature 338, 383–384. Reth, M., and Wienands, J. (1997). Initiation and processing of signals from the B cell antigen receptor. Annu. Rev. Immunol. 15, 453–479. Riberdy, J. M., Newcomb, J. R., Surman, M. J., Barbosa, J. A., and Cresswell, P. (1992). HLA-DR molecules from an antigen-processing mutant cell line are associated with invariant chain peptides. Nature 360, 474–477.

204

PIETERS

Riese, R. J., Wolf, P. R., Bromme, D., Natkin, L. R., Villadangos, J. A., Ploegh, H. L., and Chapman, H. A. (1996). Essential role for cathepsin S in MHC class II-associated invariant chain processing and peptide loading. Immunity 4, 357–366. Robinson, M. S. (1994). The role of clathrin, adaptors and dynamin in endocytosis. Curr. Opin. Cell. Biol. 6, 538–544. Roche, P. A. (1995). HLA-DM: An in vivo facilitator of MHC class II peptide loading. Immunity 3, 259–262. Roche, P. A., and Cresswell, P. (1990). Invariant chain association with HLA-DR molecules inhibits immunogenic peptide binding. Nature 345, 615–618. Roche, P. A., Marks, M. S., and Cresswell, P. (1991). Formation of a nine-subunit complex by HLA class II glycoproteins and the invariant chain. Nature 354, 392–394. Rock, K. L., Benacerraf, B., and Abbas, A. K. (1984). Antigen presentation by hapten-specific B lymphocytes. I. Role of surface immunoglobulin receptors. J. Exp. Med. 160, 1102–1113. Rockey, D. D., and Rosquist, J. L. (1994). Protein antigens of Chlamydia psittaci present in infected cells but not detected in the infectious elementary body. Infect. Immun. 62, 106–112. Rodriguez, G. M., and Diment, S. (1995). Destructive proteolysis by cysteine proteases in antigen presentation of ovalbumin. Eur. J. Immunol. 25, 1823–1827. Romagnoli, P., and Germain, R. N. (1995). Inhibition of invariant chain (Ii)–calnexin interaction results in enhanced degradation of Ii but does not prevent the assembly of alpha beta Ii complexes. J. Exp. Med. 182, 2027–2036. Rotzschke, O., Falk, K., Deres, K., Schild, H., Norda, M., Metzger, J., Jung, G., and Rammensee, H. G. (1990). Isolation and analysis of naturally processed viral peptides as recognized by cytotoxic T cells. Nature 348, 252–254. Rotzschke, O., Falk, K., Stevanovic, S., Jung, G., Walden, P., and Rammensee, H. G. (1991). Exact prediction of a natural T cell epitope. Eur. J. Immunol. 21, 2891–2894. Rudd, C. E., Bodmer, J. G., Bodmer, W. F., and Crumpton, M. J. (1985). HLA-D region antigen-associated invariant polypeptides as revealed by two-dimensional gel analysis. Glycosylation and structural interrelationships. J. Biol. Chem. 260, 1927–1936. Rudensky, A., Preston Hurlburt, P., Hong, S. C., Barlow, A., and Janeway, C. A., Jr. (1991a). Sequence analysis of peptides bound to MHC class II molecules. Nature 353, 622–627. Rudensky, A., Preston-Hurlburt, P., Hong, S. C., Barlow, A., and Janeway, C. A., Jr. (1991b). Sequence analysis of peptides bound to MHC class II molecules. Nature 353, 622–627. Russell, D. G., Dant, J., and Sturgillkoszycki, S. (1996). Mycobacterium avium- and Mycobacterium tuberculosis-containing vacuoles are dynamic, fusion-competent vesicles that are accessible to glycosphingolipids from the host cell plasmalemma. J. Immunol. 156, 4764– 4773. Russell, D. G., Xu, S., and Chakraborty, P. (1992). Intracellular trafficking and the parasitophorous vacuole of Leishmania mexicana-infected macrophages. J. Cell Sci. 103, 1193– 1210. Sallusto, F., Cella, M., Danieli, C., and Lanzavecchia, A. (1995). Dendritic cells use macropinocytosis and the mannose receptor to concentrate macromolecules in the major histocompatibility complex class II compartment: downregulation by cytokines and bacterial products. J. Exp. Med. 182, 389–400. Sanders, M. C., and Theriot, J. A. (1996). Tails from the hall of infection: Actin-based motility of pathogens. Trends Microbiol. 4, 211–213. Sanderson, F., Powis, S. H., Kelly, A. P., and Trowsdale, J. (1994). Limited polymorphism in HLA-DM does not involve the peptide binding groove. Immunogenetics 39, 56–58. Schaible, U. E., Collins, H. L., and Kaufmann, S. H. (1999). Confrontation between intracellular bacteria and the immune system. Adv. Immunol. 71, 267–377.

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

205

Schaiff, W. T., Hruska, K. A., Jr., McCourt, D. W., Green, M., and Schwartz, B. D. (1992). HLA-DR associates with specific stress proteins and is retained in the endoplasmic reticulum in invariant chain negative cells. J. Exp. Med. 176, 657–666. Schlesinger, L. S., Kaufman, T. M., Iyer, S., Hull, S. R., and Marchiando, L. K. (1996). Differences in mannose receptor-mediated uptake of lipoarabinomannan from virulent and attenuated strains of Mycobacterium tuberculosis by human macrophages. J. Immunol. 157, 4568–4575. Schorey, J. S., Carroll, M. C., and Brown, E. J. (1997). A macrophage invasion mechanism of pathogenic mycobacteria. Science 277, 1091–1093. Schumacher, T. N. M., De Bruijn, M. L. H., Vernie, L. N., Kast, W. M., Melief, C. J. M., Neefjes, J. J., and Ploegh, H. L. (1991). Peptide selection by MHC class I molecules. Nature 350, 703–706. Sekaly, R. P., Tonnelle, C., Strubin, M., Mach, B., and Long, E. O. (1986). Cell surface expression of class II histocompatibility antigens occurs in the absence of the invariant chain. J. Exp. Med. 164, 1490–1504. Sercarz, E. E., Lehmann, P. V., Ametani, A., Benichou, G., Miller, A., and Moudgil, K. (1993). Dominance and crypticity of T cell antigenic determinants. Annu. Rev. Immunol. 11, 729–766. Servenius, B., Rask, L., and Peterson, P. A. (1987). Class II genes of the human major histocompatibility complex. The DO beta gene is a divergent member of the class II beta gene family. J. Biol. Chem. 262, 8759–8766. Sette, A., Adorini, L., Colon, S. M., Buus, S., and Grey, H. M. (1989). Capacity of intact proteins to bind to MHC class II molecules. J. Immunol. 143, 1265–1267. Sette, A., Ceman, S., Kubo, R. T., Sakaguchi, K., Appella, E., Hunt, D. F., Davis, T. A., Michel, H., Shabanowitz, J., Rudersdorf, R., et al. (1992). Invariant chain peptides in most HLA-DR molecules of an antigen- processing mutant. Science 258, 1801–1804. Sette, A., Southwood, S., Miller, J., and Appella, E. (1995). Binding of major histocompatibility complex class II to the invariant chain-derived peptide, CLIP, is regulated by allelic polymorphism in class II. J. Exp. Med. 181, 677–683. Shachar, I., Elliott, E. A., Chasnoff, B., Grewal, I. S., and Flavell, R. A. (1995). Reconstitution of invariant chain function in transgenic mice in vivo by individual p31 and p41 isoforms. Immunity 3, 373–383. Shackelford, D. A., and Strominger, J. L. (1980). Demonstration of structural polymorphism among HLA-DR light chains by two-dimensional gel electrophoresis. J. Exp. Med. 151, 144–165. Sherman, M. A., Weber, D. A., and Jensen, P. E. (1995). DM enhances peptide binding to class II MHC by release of invariant chain-derived peptide. Immunity 3 197–205. Shi, G. P., Villadangos, J. A., Dranoff, G., Small, C., Gu, L., Haley, K. J., Riese, R., Ploegh, H. L., and Chapman, H. A. (1999). Cathepsin S required for normal MHC class II peptide loading and germinal center development. Immunity 10, 197–206. Silverstein, S. C. (1995). Phagocytosis of microbes: Insights and prospects. Trends Cell Biol. 5, 141–142. Simpson, F., Peden, A. A., Christopoulou, L., and Robinson, M. S. (1997). Characterization of the adaptor-related protein complex, AP-3. J. Cell Biol. 137, 835–845. Singer, P. A., Lauer, W., Dembic, Z., Mayer, W. E., Lipp, J., Koch, N., Hammerling, G., Klein, J., and Dobberstein, B. (1984). Structure of the murine Ia-associated invariant (Ii) chain as deduced from a cDNA clone. EMBO J. 3, 873–877. Sloan, V. S., Cameron, P., Porter, G., Gammon, M., Amaya, M., Mellins, E., and Zaller, D. M. (1995). Mediation by HLA-DM of dissociation of peptides from HLA-DR. Nature 375, 802–806.

206

PIETERS

Smith, C. J., Grigorieff, N., and Pearse, B. M. (1998). Clathrin coats at 21 A resolution: A cellular assembly designed to recycle multiple membrane receptors. EMBO J. 17, 4943– 4953. Snell, G. D. (1986). Some recollections of Peter Gorer and his work on this fiftieth anniversary of his discovery of H-2. Immunogenetics 24, 339–340. Spiro, R. C., and Quaranta, V. (1989). The invariant chain is a phosphorylated subunit of class II molecules. J. Immunol. 143, 2589–2594. Stahl, P., Schlesinger, P. H., Sigardson, E., Rodman, J. S., and Lee, Y. C. (1980). Receptormediated pinocytosis of mannose glycoconjugates by macrophages: Characterization and evidence for receptor recycling. Cell 19, 207–215. Stahl, P. D. (1992). The mannose receptor and other macrophage lectins. Curr. Opin. Immunol. 4, 49–52. Stebbins, C. C., Loss, G. E., Jr., Elias, C. G., Chervonsky, A., and Sant, A. J. (1995). The requirement for DM in class II-restricted antigen presentation and SDS-stable dimer formation is allele and species dependent. J. Exp. Med. 181, 223–234. Steimle, V., Durand, B., Barras, E., Zufferey, M., Hadam, M. R., Mach, B., and Reith, W. (1995). A novel DNA-binding regulatory factor is mutated in primary MHC class II deficiency (bare lymphocyte syndrome). Genes Dev. 9, 1021–1032. Steimle, V., Otten, L. A., Zufferey, M., and Mach, B. (1993). Complementation cloning of an MHC class II transactivator mutated in hereditary MHC class II deficiency (or bare lymphocyte syndrome). Cell 75, 135–146. Steimle, V., Siegrist, C. A., Mottet, A., Lisowska-Grospierre, B., and Mach, B. (1994). Regulation of MHC class II expression by interferon-gamma mediated by the transactivator gene CIITA. Science 265, 106–109. Steinman, R. M., and Cohn, Z. A. (1973). Identification of a novel cell type in peripheral lymphoid organs of mice. I. Morphology, quantitation, tissue distribution. J. Exp. Med. 137, 1142–1162. Steinman, R. M., and Swanson, J. (1995). The endocytic activity of dendritic cells. J. Exp. Med. 182, 283–288. Steinman, R. M., and Witmer, M. D. (1978). Lymphoid dendritic cells are potent stimulators of the primary mixed leukocyte reaction in mice. Proc. Natl. Acad. Sci. USA 75, 5132–5136. Stern, L. J., Brown, J. H., Jardetzky, T. S., Gorga, J. C., Urban, R. G., Strominger, J. L., and Wiley, D. C. (1994). Crystal structure of the human class II MHC protein HLADR1 complexed with an influenza virus peptide. Nature 368, 215–221. Stockinger, B., Pessara, U., Lin, R. H., Habicht, J., Grez, M., and Koch, N. (1989). A role of Ia-associated invariant chains in antigen processing and presentation. Cell 56, 683–689. Strubin, M., Berte, C., and Mach, B. (1986a). Alternative splicing and alternative initiation of translation explain the four forms of the Ia-antigen associated invariant chain. EMBO J. 5, 3483–3488. Strubin, M., Mach, B., and Long, E. O. (1984). The complete sequence of the mRNA for the HLA-DR associated invariant chain reveals a polypeptide with an unusual transmembrane polarity. EMBO J. 3, 869–872. Strubin, M., Long, E. O., and Mach, B. (1986b). Two forms of the Ia antigen-associated invariant chain result from alternative initiations at two in-phase AUGs. Cell 47, 619–625. Sturgill-Koszycki, S., Schlesinger, P. H., Chakraborty, P., Haddix, P. L., Collins, H. L., Fok, A. K., Allen, R. D., Gluck, S. L., Heuser, J., and Russell, D. G. (1994). Lack of acidification in Mycobacterium phagosomes produced by exclusion of the vesicular proton-ATPase. Science 263, 678–681. Svensson, M., Stockinger, B., and Wick, M. J. (1997). Bone marrow-derived dendritic cells can process bacteria for MHC-I and MHC-II presentation to T cells. J. Immunol. 158, 4229–4236.

MHC CLASS II-RESTRICTED ANTIGEN PROCESSING AND PRESENTATION

207

Takaesu, N. T., Lower, J. A., Robertson, E. J., and Bikoff, E. K. (1995). Major histocompatibility class II peptide occupany, antigen presentation, and CD4⫹ T cell function in mice lacking the p41 isoform of invariant chain. Immunity 3, 385–396. Tanaka, K., Tanahashi, N., Tsurumi, C., Yokota, K. Y., and Shimbara, N. (1997). Proteasomes and antigen processing. Adv. Immunol. 64, 1–38. Tarlinton, D. (1997). Antigen presentation by memory B cells: The sting is in the tail. Science 276, 374–375. ter Haar, E., Musacchio, A., Harrison, S. C., and Kirchhausen, T. (1998). Atomic structure of clathrin: A beta propeller terminal domain joins an alpha zigzag linker. Cell 95, 563–573. Teyton, L., O’Sullivan, D., Dickson, P. W., Lotteau, V., Sette, A., Fink, P., and Peterson, P. A. (1990). Invariant chain distinguishes between the exogenous and endogenous antigen presentation pathways. Nature 348, 39–44. Theriot, J. A. (1995). The cell biology of infection by intracellular bacterial pathogens. Annu. Rev. Cell Dev. Biol. 11, 213–239. Townsend, A., Ohlen, C., Bastin, J., Ljunggren, H.-G., Foster, L., and Karre, K. (1989). Association of class I major histocompatibility heavy and light chains induced by viral peptides. Nature 340, 443–448. Townsend, A. R. M., Gotch, F. M., and Davey, J. (1985). Cytotoxic T cells recognize fragments of the influenza nucleoprotein. Cell 42, 457–467. Townsend, A. R. M., Rothbard, J., Gotch, F. M., Bahadur, G., Wraith, D., and McMichael, A. J. (1986). The epitopes of influenza nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides. Cell 44, 959–968. Trowsdale, J., and Campbell, R. D. (1992). Complexity in the major histocompatibility complex. Eur. J. Immunogenet. 19, 45–55. Trowsdale J., Ragoussis, J., and Campbell, R. D. (1991). Map of the human MHC. Immunol. Today 12, 443–446. Tulp, A., Verwoerd, D., Dobberstein, B., Ploegh, H. L., and Pieters, J. (1994). Isolation and characterization of the intracellular MHC class II compartment. Nature 369, 120–126. Unanue, E. R. (1997). Inter-relationship among macrophages, natural killer cells and neutrophils in early stages of Listeria resistance. Curr. Opin. Immunol. 9, 35–43. Van Bleek, G. M., and Nathenson, S. G. (1990). Isolation of an endogenously processed immunodominant viral peptide from the class I H-2Kb molecule. Nature 348, 213–216. van Ham, S. M., Tjin, E. P. M., Lillemeier, B. F., Gruneberg, U., van Meijgaarden, K. E., Pastoors, L., Verwoerd, D., Tulp, A., Canas, B., Rahman, D., Ottenhoff, T. H., Pappin, D. J., Trowsdale, J., and Neefjes, J. (1997). HLA-DO is a negative modulator of HLADM-mediated MHC class II peptide loading. Curr. Biol. 7, 950–957. van Noort, J. M., and Jacobs, M. J. (1994). Cathepsin D, but not cathepsin B, releases T cell stimulatory fragments from lysozyme that are functional in the context of multiple murine class II MHC molecules. Eur. J. Immunol. 24, 2175–2180. VanBuskirk, A. M., DeNagel, D. C., Guagliardi, L. E., Brodsky, F. M., and Pierce, S. K. (1991). Cellular and subcellular distribution of PBP72/74, a peptide-binding protein that plays a role in antigen processing. J. Immunol. 146, 500–506. Varshavsky, A. (1997). The ubiquitin system. Trends Biochem. Sci. 22, 383–387. Villadangos, J. A., Riese, R. J., Peters, C., Chapman, H. A., and Ploegh, H. L. (1997). Degradation of mouse invariant chain: Roles of cathepsins S and D and the influence of major histocompatibility complex polymorphism [published erratum appears in J. Exp. Med. 1997 Dec, 1; 186(11): 5945]. J. Exp. Med. 186, 549–560. Viville, S., Neefjes, J., Lotteau, V., Dierich, A., Lemeur, M., Ploegh, H., Benoist, C., and Mathis, D. (1993). Mice lacking the MHC class II-associated invariant chain. Cell 72635– 72648.

208

PIETERS

Vogt, A. B., Kropshofer, H., Kalbacher, H., Kalbus, M., Rammensee, H. G., Coligan, J. E., and Martin, R. (1994). Ligand motifs of HLA-DRB5*0101 and DRB1*1501 molecules delineated from self-peptides. J. Immunol. 153, 1665–1673. von Boehmer, H. (1994). Positive selection of lymphocytes. Cell 76, 219–228. Vyse, T. J., and Kotzin, B. L. (1998). Genetic susceptibility to systemic lupus erythematosus. Annu. Rev. Immunol. 16, 261–292. Warmerdam, P. A., Long, E. O., and Roche, P. A. (1996). Isoforms of the invariant chain regulate transport of MHC class II molecules to antigen processing compartments. J. Cell. Biol. 133, 281–291. Watts, C. (1997a). Capture and processing of exogenous antigens for presentation on MHC molecules. Annu. Rev. Immunol. 15, 821–850. Watts, C. (1997b). Immunology. Inside the gearbox of the dendritic cell. Nature 388, 724–725. Watts, C., and Lanzavecchia, A. (1993). Suppressive effect of antibody on processing of T cell epitopes. J. Exp. Med. 178, 1459–1463. West, M. A., Lucocq, J. M., and Watts, C. (1994). Antigen processing and class II MHC peptide-loading compartments in human B-lymphoblastoid cells. Nature 369, 147–151. Wiertz, E. J., Tortorella, D., Bogyo, M., Yu, J., Mothes, W., Jones, T. R., Rapoport, T. A., and Ploegh, H. L. (1996a). Sec61-mediated transfer of a membrane protein from the endoplasmic reticulum to the proteasome for destruction. Nature 384, 432–438. Wiertz, E. J. H. J., Jones, T. R., Sun, L., Bogyo, M., Geuze, H. J., and Ploegh, H. L. (1996b). The human cytomegalovirus US11 gene product dislocates MHC class I heavy chains from the endoplasmic reticulum to the cytosol. Cell 84, 769–779. Williams, G. S., Malin, M., Vremec, D., Chang, C. H., Boyd, R., Benoit, C., and Mathis, D. (1998). Mice lacking the transcription factor CIITA—a second look. Int. Immunol. 10, 1957–1967. Wiman, K., Larhammar, D., Claesson, L., Gustafsson, K., Schenning, L., Bill, P., Bo¨hme, J., Denaro, M., Dobberstein, B., Hammerling, U., Kvist, S., Servenius, B., Sundelin, J., Peterson, P. A., and Rask, L. (1982). Isolation and identification of a cDNA clone corresponding to an HLA-DR antigen beta-chain. Proc. Natl. Acad. Sci. USA 79, 1703– 1707. Wolf, P. R., and Ploegh, H. L. (1995). How MHC class II molecules acquire peptide cargo: Biosynthesis and trafficking through the endocytic pathway. Annu. Rev. Cell. Dev. Biol. 11, 267–306. Wordsworth, B. P., and Bell, J. I. (1992). The immunogenetics of rheumatoid arthritis. Springer Semin. Immunopathol. 14, 59–78. Zhong, G., Romagnoli, P., and Germain, R. N. (1997). Related leucine-based cytoplasmic targeting signals in invariant chain and major histocompatibility complex class II molecules control endocytic presentation of distinct determinants in a single protein. J. Exp. Med. 185, 429–438. Zinkernagel, R. M., and Doherty, P. C. (1974a). Immunological surveillance against altered self components by sensitised T lymphocytes in lymphocytic choriomeningitis. Nature 251, 547–548. Zinkernagel, R. M., and Doherty, P. C. (1974b). Restriction of in vitro T cell-mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system. Nature 248, 701–702.

ADVANCES IN IMMUNOLOGY, VOL. 75

T-Cell Receptor Crossreactivity and Autoimmune Disease HARVEY CANTOR Department of Pathology, Harvard Medical School, Department of Cancer Immunology and AIDS, Dana-Farber Cancer Institute, Boston, Massachusetts 02115

I. Introduction

There is increasing appreciation that T-cell receptors (TCR) display a high degree of crossreactivity and degeneracy in their recognition of peptide antigens. These findings, coupled with the division of peptide ligands into antagonists or agonists depending on their affinity for the TCR, have shed new light on mechanisms of self-tolerance and autoimmunity and are the subject of this review. We summarize current evidence for two opposing effects of TCR crossreactivity. These data suggest that TCR crossreactivity of autoreactive T-cells with antagonistic self-antigens may inhibit autoimmune disease. Additional evidence suggests that crossreactivity between autoreactive T-cells and microbial antigens may facilitate the development of Type 1 immunity and trigger autoimmune disease. II. Crossreactivity and T-Cell Activation

One of the earliest indications that TCR crossreactivity might play an important role in the immune response came from studies of the phenomenon of alloreactivity. Finberg et al. (1978) showed that CD8⫹CTL that recognized and lysed syngeneic (self ) cells infected with Sendai virus also recognized and specifically lysed uninfected allogeneic cells expressing particular MHC class I alleles. Crossreactive recognition of these foreign class I MHC products depended on T-cells specific for Sendai virus/selfMHC because lysis of uninfected foreign cells was inhibited completely by unlabeled syngeneic cells infected with Sendai virus. Bevan and colleagues arrived at a similiar conclusion (Bevan, 1977) using T-cell lines raised initially against foreign minor H antigens that had subsequently been expanded by stimulation with allogeneic cells. Lysis of allogeneic targets and H-2 identical targets by these T-cell lines occurred with similar efficiency and indicated crossreactivity, with the proviso that the two reactivities were mediated by the same T-cell clones. Subsequent studies of CD4 cells reactive to cytochrome c (Sredni and Schwartz, 1980; Ashwell et al., 1986) and insulin (Abromson-Leeman and Cantor, 1983) extended this paradigm to include crossrecognition of allogeneic class II MHC products by T-cell clones that recognized self class II and foreign peptides. The 209

Copyright 䉷 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2776/00 $35.00

210

CANTOR

demonstration that T-cells specific for self-MHC/foreign peptide crossrecognized foreign MHC products provided a long-sought cellular explanation for the phenomenon of alloreactivity and the associated clinical barrier to transplantation across MHC differences. A second measure of crossreactive recognition is the range of peptides that can successfully elicit activation of a single T-cell clone. Two complementary experimental approaches have been used to address this question, both pointing to a high degree of TCR crossreactivity. In one approach, a single T-cell clone with a well-characterized receptor is used to report on the number of peptides within a peptide library that bind with sufficient strength to activate the clone. In the second approach, a single peptide MHC complex is used to positively select T-cells in vivo; the number of clones with differing TCR that develop from this early antigenic experience is used to measure TCR crossreactivity. In an example of the first approach, random combinatorial peptide libraries were used to analyze a T-cell clone specific for a peptide derived from the myelin basic protein (MBP) (Hemmer et al., 1997). The library consisted of 2 ⫻ 1014 different peptides, with concentrations of about 5 ⫻ 10⫺19 gm/ml for each. The observation that the clone responded to this mixture of peptides almost as strongly as it did to the parent MBP peptide strongly suggests a high degree of degeneracy in antigen recognition. Additional screening of peptide sublibraries yielded a number of substituted peptide analogs that stimulated maximal activation at concentrations that were several orders of magnitude lower than that of the starting peptide. Analyses of MBP-specific T-cell clones demonstrated that many amino acid substitutions of the ‘‘antigenic’’ peptide were tolerated and that some substituted peptides displayed superagonist activity compared to the starting peptide. These findings followed the earlier observation of Nanda et al. (1995) that a single T-cell clone might recognize at least five different overlapping peptides, each with a distinct core structure in the context of the same MHC molecule. That is, the TCR may express multiple sets of contact residues for different peptide MHC ligands and, in that sense, represent a multiunit recognition structure. Interpretation of these data is somewhat constrained by the possibility that the T-cell clones used for these analyses may be exceptions to the rule, and definition of a larger number of T-cell clones might reveal more restricted individual repertoires. A second experimental approach overcomes this objection by placing constraints on peptide heterogeneity rather than on the T-cell receptor (Ignatowicz et al., 1996, 1997; Tourne et al., 1997; Surh et al., 1997). In this case, mice engineered to express a single (or highly restricted) class II-associated peptide generated a large number of CD4⫹ T-cell clones that represented as much as 20–50% of the T-cell repertoire of their normal littermates. These studies emphasize a second

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

211

level of flexibility of the TCR molecule that allows many receptor variations to interact with a single peptide/MHC complex and undergo positive selection. Although these observations suggest that the T-cell repertoire can be extraordinarily broad, several factors are likely to reduce the effective repertoire of mature peripheral T-cells in vivo. Processing of intrinsic cellular proteins generates only a small fraction of potential peptide sequences due to structural restrictions in enzymatic cleavage, and only a small fraction of cleaved peptides are displayed on the surface of cells at high enough concentrations (estimated at 103 to 104 peptides/cell e.g., Marrack et al., 1993) to trigger a T-cell clone. An additional element that may reduce the potential for autoreactive triggering of T-cell clones is interactions with antagonistic peptides in peripheral tissues. A. LOW AFFINITY LIGANDS AND T-CELL ANTAGONISM T-cell crossreactivity is usually discussed in connection with the triggering of autoimmune disease. Here, we outline the role of T-cell crossreactivity in the maintenance of self-tolerance through T-cell antagonism. The phenomenon of T-cell antagonism was first described by Rao et al. (1984a,b). These studies used a panel of CD4 clones that responded to arsonate (Ar-) conjugated to various proteins. In these experiments, the Ar- structure represented an essential and dominant portion of the TCR ligand while contiguous amino acids contributed the remaining (minor) portion of the ligand. This system allowed a detailed analysis of the effects of systematic structural changes to either the Ar- portion or the contiguous amino acid portion of the ligand to both TCR binding and T-cell activation. One example used a murine CD4 clone termed Ar5 that is activated by arsonate-coupled proteins in association with I-Ad (Rao et al., 1984b). This clone also responded to the p-azobenzoate (C-), p-azobenzenesulfonate (S-), and p-azobenzenesulfonamide (SNH2-), haptens conjugated to the same protein (ovalbumin) but not top-azobenzamide (CONH2), p-azonitrobenzene (NO2-)1, o-azobenzenearsonate (OAA-), and thiourea-linked benzenearsonate (AR[NCS]-) (Rao et al., 1984 a,b) (Table I). A comparison of the proliferative T-cell response to these ligands indicated that C-, S-, and SNH2- were 30-, 100-, and 1000-fold less effective stimulatory ligands, respectively, than were Ar- conjugates (e.g., Fig. 1). Reactivity was then compared to specific binding of these and other radio-labeled ligands to the TCR. All four activity ligands (C-, S-, SNH2and Ar-) bound to the Ar5 TCR specifically, while nonactivating Ar analogs, including CONH2-, NO2-, OAA-, and AR[NCS]-, did not bind (Table I). The key observation was that addition of low concentrations of weakly activating ligands such as C-OVA, S-OVA, or SNH2-OVA antagonized the

212

CANTOR

TABLE I STRUCTURAL ANALOGS OF THE P-AZOBENZENEARSONATE HAPTEN Structure OH tyr/his

N

N

Abbreviation

Activation

Binding

AR⫺

⫹⫹⫹

⫹⫹⫹

S⫺

⫹

C⫺

⫹⫹

⫹⫹⫹

SNH2⫺

⫹

⫹

CONH2⫺

⫺

⫺

NO2⫺

⫺

⫺

OOA⫺

⫺

⫺

AR(NCS)⫺

⫺

⫺

O

As O

O S

O

⫹⫹⫹

O O C O O S

NH2

O O C NH2 O N O OH O As O N

N OH

lys

N

C S

N

tyr/his O

As O

‘‘Activation’’ refers to the ability, when conjugated to carrier proteins, to activate DNA synthesis by clone Ar-5 in the presence of antigen-presenting cells (e.g., Fig. 1); ‘‘binding’’ refers to the ability to interact with the arsonate binding site of clone Ar-5. From Rao, A., Ko, W. W.-P., Fass, S. J., and Cantor, H. (1984). Binding of antigen in the absence of histocompatibility proteins by arsonate-reactive T-cell clones. Cell 36, 879–888. Reprinted with the permission of Cell Press.

T-cell clone’s physiological response to Ar-OVA (Fig. 2). The finding that low concentrations of the crossreactive C-, S-, and SNH2- analogs blocked activation by the Ar agonist but that they could activate the T-cell clone themselves at high concentrations indicated that these crossreactive ligands possessed mixed agonist/antagonist properties for this TCR (Rao et al.,

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

213

FIG. 1. DNA synthesis by clone Ar-5 in response to structural analogs of arsonate conjugated to ovalbumin. All ovalbumin conjugates contained between 6 and 9 moles hapten/ mole protein. 3H-thymidine incorporation by 3 ⫻ 104 Ar-5 cells was measured in response to irradiated CAF, spleen cells, and antigen. From Rao, A., Ko, W. W.-P., Fass, S. J., and Cantor, H. (1984). Binding of antigen in the absence of histocompatibility proteins by arsonate-reactive T-cell clones. Cell 36, 879–888. Reprinted with the permission of Cell Press.

1984a). Although this phenomenon had not yet been described for T-cell ligands, they were well-known properties of ligands that interacted with enzymes or with other cell receptors, e.g., hexamethonium and decamethonium antagonism of the aceylcoline receptor, aconitine antagonism of the sodium channel, and cyanogen bromide-cleaved EGF antagonism of the EGF receptor. The binding kinetics of the mixed agonist/antagonist ligands for the TCR were not reported, but it is likely that TCR antagonism by the S-, C-, and SNH2-, ligands reflected increased off-rates compared to the full agonist, Ar-. These findings were extended to conventional peptide ligands by Kersh and Allen (1996), who showed that crossreactive recognition of peptide antagonists led to partial activation and inhibition of responses to coadministered peptide agonists. Earlier studies had also shown that engagement of CD8⫹ thymocytes by suboptimal peptide ligands (antagonists) sustained positive selection while agonists induced negative selection (Hogquist et al., 1994, 1995). The mechanism that underlies TCR antagonism is still not fully understood. Two observations are particularly relevant. First, recognition of antagonistic peptide ligands is accompanied by diminished levels of phosphorylation of the CD3 complex compared to that obtained with agonist ligands (Kersh and Allen, 1996; Madrenas and Germain, 1996). Second, rapid dissociation of antagonist complexes and rebinding to other TCRs is favored by the increased off-rate of antagonists (Lyons et al.,

214

CANTOR

FIG. 2. Effect of S-OVA on activation of clone Ar-5. Clone Ar-5 cells (3 ⫻ 104) were incubated (2 h at 37⬚C) with the indicated concentrations of S-OVA and irradiated CAF1 spleen cells before additional of AR-OVA (3 애g/ml). (A) 3H-thymidine incorporation between 20 and 24 h. (B) Interleukin-2 (IL-2) produced in the supernatant at 16 h. Background IL-2 production in response to 100 애g/ml S-OVA and CAF1 APC was ⬍0.01 U/ml. From Rao, A., Fass, S. J., and Cantor, H. (1984). Analogs that compete for antigen binding to an arsonate-reactive T-cell clone inhibit the functional response to arsonate. Cell 36, 889–895. Reprinted with the permission of Cell Press.

1996). These two observations suggest that prolonged sterile engagement of an excess number of CD3 complexes by antagonistic peptides reduces receptor availability for engagement of full agonists. The finding that agonist-driven formation of TCR dimers and clusters is important for effective activation (Boniface et al., 1998; Alam et al., 1999; Penninger and Crabtree, 1999) adds weight to the view that TCR antagonism has an allosteric as well as a kinetic (off-rate) explanation. The possibility that antagonistic activity of self-peptide ligands may offer a mechanism that helps maintain self tolerance is supported by the finding that transgenic expression of antagonist ligands inhibits the primary T-cell response to a crossreactive agonistic ligand in appropriate TCR transgenic mice (Basu and Williams, 1998). As might be expected from studies on the Ar system, the inhibiting effects of these antagonistic ligands in vivo are overcome by administration of excess agonist. B. THE ROLE OF T-CELL ANTAGONISM IN PERIPHERAL T-CELL TOLERANCE Although thymocytes that recognize self-peptide agonists are instructed to die (Moller, 1989; Kappler et al., 1987), self peptides derived from

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

215

proteins restricted to sequestered tissues may be hidden from this process, and T-cell recognition of these sequestered peptides represents a potential risk for organ-specific autoimmune disease. How might this potential inciting event be regulated? One mechanism is based on TCR crossreactivity. Engagement of autoreactive T-cell clones by antagonistic self-peptide ligands on professional APC may inhibit reactivity to peptide agonists expressed in sequestered target organs and tissues. 1. Immunoglobulin-Derived Peptide Antagonists and Peripheral Tolerance Immunoglobulin-derived self-peptides are readily available to the thymus and peripheral T-cells and, when displayed by B-cells, efficiently induce T-cell tolerance through anergy (Eynon and Parker, 1992; Fuchs and Matzinger, 1992). Polymorphisms of immunoglobulin genes can affect susceptibility to a number of autoimmune disorders, although these effects have not been attributed to Ig-derived peptides. In mice, susceptibility to anti-erythrocyte autoantibody formation (van Snick, 1981), spontaneous production of rheumatoid factor (van Snick, 1981; Pereira and Coutinho, 1989), and murine myasthenia gravis is associated with Igh-linked polymorphisms (Berman and Patrick, 1980). An increased relative risk for multiple sclerosis has also been linked to IgG polymorphisms detected serologically (Salier et al., 1981, 1983) or by RFLP (Gaiser et al., 1987). Susceptibility to certain forms of myasthenia gravis has been associated with particular IgH allotypes (Smith et al., 1983, 1984), and patients with combinations of particular HLA haplotypes and Gm allotypes are at increased risk for this disease (Field et al., 1984; Schernthaner and Mayr, 1984). Linkage to particular Gm allotypes has been correlated with an increased incidence of Hashimoto’s diseases (So et al., 1987), systemic lupus erythematosus (Hoffman et al., 1991), and polymyalgia rheumatica (Demaine et al., 1983). Although these associations usually have been attributed to the effects of Ig polymorphisms on the antibody activity of intact immunoglobulins (Garchon and Bach, 1991), an alternative explanation is based on observation that T-cell recognition of self-Ig-derived peptides on B-cells (Weiss and Bogen, 1991; Woodland and Cantor, 1978; Bikoff et al., 1988; Bikoff, 1991; Rudensky and Yurin, 1989) induces T-cell tolerance (Eynon and Parker, 1992; Fuchs and Matzinger, 1992). Peptides derived from Ig bind with high affinity to multiple MHC haplotypes of human (Chicz et al., 1993) and mouse (Rudensky et al., 1992) class II molecules and play a role in positive thymocyte selection (Chen et al., 1992). For example, Tcell clones that recognize influenza viral peptides are likely to be positively selected by self-Ig V region-derived peptides (Cao et al., 1994). These considerations have suggested the hypothesis that Ig-allotype-linked resis-

216

CANTOR

tance to autoimmune disease might reflect regulation of T-cell clones specific for self-peptides derived from polymorphic regions of Ig. 2. Immunoglobulin Allotype-Linked Resistance to Autoimmune Disease Herpes stromal keratitis (HSK) is a virally induced T-cell-dependent immune reaction that destroys corneal tissue and represents a leading cause of blindness (Dawson and Togni, 1976). Studies of murine HSK have shown that susceptibility is controlled in a unigenic dominant fashion by genes linked to the Igh locus (Opremcak et al., 1988) and MHC polymorphisms have not been implicated (Foster et al., 1987). Resistance/susceptibility has been located within the IgCH region using a panel of Ig-congenic recombinant mice (Dutt et al., 1993). Inbred mouse strains that carry the Igha/d/e alleles are susceptible to necrotizing stromal keratitis, whereas strains that carry the Ighb allele are highly resistant (Foster et al., 1987; Opremcak et al., 1988, 1990). HSK is reproducibly elicited 8–14 days following corneal inoculation of HSV-1 in mouse strains that possess an intact T-cell system and appropriate Igh allotype. Corneal HSV-1 infection of nu/nu or SCID mice does not result in HSK, and adoptive transfer of T-cells restores HSK susceptibility (Metcalf et al., 1979; Russell et al., 1984; Mercadal et al., 1993). This murine disease model is attractive because of the clarity of the resistance/susceptiblity effect on a T-cell-mediated disease that is conferred by alleles of the Igh locus in the absence of other genetic influences, and the ability to obtain noninvasive measurements of disease development. The idea that IgCH polymorphisms regulate HSK resistance through an effect on the immunologic activities of immunoglobulins (e.g., complement fixation and Fc-receptor binding) is unlikely, because HSK is mediated by T-cells and is independent of the host antibody response (Opremcak et al., 1990; Foster, 1989). Ig-linked resistance of HSK might reflect the tolerogenic action of Ig-derived peptides on T-cell clones that recognize corneal autoantigens. Indeed, susceptible (C.AL-20) recipients of serum Ighb (the allotype carried by the resistant strain) but not Ighe-containing material do not develop HSK (Avery et al., 1995). Additional studies showed that the Ig fraction of the sera is responsible for these effects, and that reduced disease levels following intravenous Ighb Ig reflected inhibition of the host T-cell response: T-cells from Ighb-treated donors lost the ability to transfer HSK to syngeneic nu/nu recipients (Avery et al., 1995). The Igh C-region cluster includes genes encoding different Ig heavy chain isotypes, including IgG1, IgG2a, IgG2b, and IgG3. Intravenous injection of the IgG2ab isotype (but not IgG2bb, IgG1b, or IgG2ac/a) hybridoma proteins into C.AL-20 mice inhibits HSK. T-cells that mediate HSK were shown to recognize an IgG2ab-derived peptide, and CD4⫹ clones specific

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

217

for this peptide (Bikoff and Eckhardt, 1989; Bikoff, 1991) transfer HSK into nu/nu hosts (Avery et al., 1995). These IgG2ab-specific T-cell clones cross-recognize a self-antigen expressed by murine corneal cells (but not murine kidney, skin, or other cells). Interestingly, reactivity of IgG2abspecific clones to corneal protein required CD28 ligation (which had no effect on the response of the control clones), suggesting that increased costimulatory activity associated with HSV infection might be required to enhance this response in vivo (Avery et al., 1995). Analysis of peptides that represented allotype-specific amino acid differences (clustered in five distinct regions of the IgG2a molecule: 103–118, 215–230, 249–264, 270–285, and 292–308) revealed a peptide (pos 292– 308) that efficiently activated the IgG2ab-specific clones. In vivo administration of this soluble peptide, but not control peptides, resulted in resistance to HSK. Moreover, recipients of peptide 292–308-immune T-cells, but not T-cells from mice immunized with other peptides, developed HSK (Avery et al., 1995). Sequestration of tissue such as cornea from the immune system has been attributed to the relative absence of blood and lymphatic drainage of this peripheral tissue (Niederkorn, 1990) and consequent failure of corneal antigens to traffic to regional lymphoid tissues and sensitize the immune system (Ross et al., 1991; Callanan et al., 1988). Despite the poor representation of corneal antigens in the thymus, clinical tolerance to this tissue is generally maintained throughout life, and ⬎90% of corneal grafts are successful in the absence of HLA typing (Smith et al., 1980). In the uninfected eye, this may reflect the failure of autoreactive T-cells to gain entry into this tissue and inefficient presentation of self-antigens to the few T-cells that stray into an environment that is virtually devoid of professional APC. However, after microbial infection, T-cells reactive to corneal peptides have access to this anatomic site and may be activated. Sequestered tissues may escape the penalties of isolation from thymic negative selection by including peptide subsequences that crossreact with peptide antagonists such as the IgG2ab peptide that are presented by professional APC. Polymorphisms of genes that influence resistance to autoimmune disease have generally been thought to act through their protein products rather than peptides derived from them (Garchon and Bach, 1991). However, genetic polymorphisms that regulate susceptibility to autoimmune disease may also act by changing the structure of self-peptides. In the case of HSK, studies of mice carrying a keratogenic TCR transgene may further delineate in vivo mechanisms that underlie Ig-linked susceptibility to autoimmune disease.

218

CANTOR

III. Limits of TCR Crossreactivity: Peripheral Purging of Useless T-Cells

The selection process in the thymus that regulates the repertoire of Tcell subsets is effective but not foolproof. T-cells bearing antigen receptors that do not functionally engage self-MHC products can also emerge from the thymus (Schulz et al., 1996). One pitfall comes from the fact that some peptides are expressed well in the thymic epithelium but not in the periphery. Positive selection by this peptide set is no guarantee that the peptide will be processed and presented in the same way in the peripheral tissues, and some clones that are selected by these peptides in the thymus may fail to see the same self-peptide in the periphery. In the case of CD8 cells, additional problems arise. Class II-reactive CD8 cells can be generated through interactions either with strong peptide agonists or with self-peptides that antagonize CD4, but not CD8, cell activation ( Jameson and Bevan, 1998), perhaps reflecting reduced p56lck signaling (Matechak et al., 1996) or upregulation of Notch-1 (Robey et al., 1996). Although CD8 cells that bear mismatched TCRs may be functionally active in vitro ( Jameson and Bevan, 1998; Kirberg et al., 1994), they do not contribute significantly to normal immune responses (Rao et al., 1983; Suzuki et al., 1994). Thus, CD8 cells that express TCRs with an affinity that is too low to interact with class I-associated peptides may normally be inactivated by a post-thymic mechanism. We have examined the fate of CD8 cells that express a TCR that does not interact with class I-associated peptides. A portion of T-cells bearing the class II-restricted DO11.10 TCR transgene that survive thymic selection express CD8 and are exported in significant numbers to peripheral tissues (Matechak et al., 1996; Murphy et al., 1990). However, within 2 to 3 days after adoptive transfer into syngeneic BALB/c hosts, these CD8 cells downregulate CD8 coreceptor expression, despite the availability of class I MHC for CD8 engagement and class II MHC for TCR engagement. Because CD8 cells that express a TCR transgene specific for a class Irestricted antigen (HY) (Kisielow et al., 1988) continued to express the CD8 coreceptor after a similar period in adoptive syngeneic (female) hosts, CD8 co-engagement appears necessary for stable CD8 expression (Pestano et al., 1999). The hypothesis that continued expression of the CD8 receptor may require co-engagement of CD8 and TCR by MHC class I–peptide complexes in peripheral tissues was also tested by an examination of the effects of class I deficiency on expression of CD8 by peripheral T cells. Within 5 to 10 days, virtually all CD8 cells transfused into 웁2M⫺/⫺ mice expressed a CD4⫺ CD8⫺ (‘‘double negative’’–DN) phenotype, similar to the fate of DO11.10 class II-specific CD8 cells in a normal MHC environment (Pes-

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

219

tano et al., 1999). Thus, even correctly selected CD8 cells rapidly downregulate their CD8 coreceptor in a lymphoid environment that is deficient in class I MHC products. Coreceptor downregulation is specific for the CD8 lineage, because purified CD4 cells transferred to MHC class II-deficient (IA웁b⫺/⫺) hosts maintained normal levels of CD4 coreceptor expression (Pestano and Cantor, unpublished data). Mice that carry the 웁2M⫺/⫺ mutation, although severely deficient in expression of class I MHC–self-peptide complexes, harbor significant numbers of CD8 cells in their peripheral lymphoid tissues (Koller et al., 1990). However, CD8 expression on these cells does not reflect stable and active synthesis of the CD8 protein. CD8 cells from 웁2M⫺/⫺ but not 웁2M⫹/⫹ mice lose expression of surface CD8 protein within 4 hours of in vitro culture and do not resynthesize CD8 after enzymatic removal (Pestano et al., 1999). Analysis of genes expressed by purified CD8 cells from 웁2M⫺/⫺ mice also showed that these cells express little or no CD8움 or 웁 mRNA consistent with their ‘‘lame duck’’ status with respect to CD8 expression. These cells also downregulated expression of the granzyme B gene, which mediates cytotoxic effector activity of CD8 cells, whereas expression of housekeeping genes, such as actin and GAPDH, was unchanged. Additional analysis of gene expression in CD8 cells unable to engage class I MHC in peripheral tissues indicated reduced expression of the LKLF transcription factor, which normally acts to inhibit FasL expression, and markedly elevated levels of FasL (Pestano et al., 1999). The molecular mechanism that underlies downregulation of CD8 gene expression after failed CD8 ligation includes remethylation of the CD8 gene. Demethylation of the CD8움 gene accompanies initial CD8 expression in the thymus, as judged by increased susceptibility of the CD8 gene in DP thymocytes to methylation-sensitive enzymes (Carbone et al., 1988). Thus, the CD8움 Hha I site is fully sensitive to enzymatic cleavage in double positive (CD4⫹8⫹ ) thymocytes and in lineage-committed CD8 SP thymocytes from 웁2M⫹/⫹ mice. However, the CD8 gene in CD8 cells from 웁2M⫺/⫺ mice is resistant to Hha I cleavage, indicative of complete remethylation, and single positive CD8 thymocytes show an intermediate Hha I sensitivity, indicative of partial remethylation. The decline in absolute numbers of CD8 cells after transfer into 웁2M⫺/⫺ recipients (Tanchot et al., 1997; Nesic and Vukmanovic, 1998; von Boehmer and Fehling, 1997) is due, in part, to Fas-dependent cell death. Up to 50% of donor CD8 cells express elevated levels of Fas and about 75 to 90% of the DN progeny express both Fas and FasL. Moreover, the small population of lame duck CD8 cells in 웁2M⫺/⫺ mice that had CD44hi also upregulated Fas expression and underwent apoptosis at two to three times the rate of CD8 cells from MHC ⫹/⫹ mice. Most significantly, genetic

220

CANTOR

defects in Fas or FasL expression enhanced by 15- to 20-fold the accumulation of DN cells from CD8 precursors in 웁2M⫺/⫺ hosts. Additional analysis of B6lpr CD8 cells and DN progeny in adoptive 웁2M⫺/⫺ hosts showed that, although CD8 cells did not replicate during this process, about two-thirds of the newly generated DN cells underwent at least one round of replication, as judged by CFSE dye-dilution analysis. These findings identify a pathway taken by CD8 cells that cannot efficiently engage class I MHC-self-peptide molecules as a result of incorrect thymic selection or defects in peripheral MHC class I expression. In either case, failed CD8–TCR co-engagement results in downregulation of genes that account for specialized CTL function, resistance to cell death (CD8움웁, granzyme B, LKLF), and upregulation of the Fas and FasL death genes. The finding that MHC engagement is required to inhibit expression and delivery of a Fas-dependent death program rather than to supply a putative trophic stimulus for T-cell survival bears directly on the fate of T-cells in the context of autoimmune disease. Expansion of DN cells in animals that carry mutations in Fas or FasL has been attributed to a proliferative response of autoreactive T-cells (Perkins et al., 1996). Instead, these cells are likely to represent the detritus of the CD8 repertoire that is unable to engage host MHC products but is spared elimination through mutations in Fas-related death molecules. These mistakenly selected T-cells include clones that are positively selected by class I–self-peptide complexes (Maldonado et al., 1995; Jevnikar et al., 1994; Koh et al., 1995) expressed in the thymus but not peripheral lymphoid tissues (Laufer et al., 1996; Chung et al., 1996) and CD8 cells that are selected by thymic class II–self-peptide complexes ( Jameson and Bevan, 1998; Marrack and Kappler, 1997; von Boehmer and Fehling, 1997; Albert Basson et al., 1998). Defects in delivery of the death signal to these cells may underlie the accumulation of DN T-cells in patients that carry inherited mutations of Fas and FasL and in approximately 25% of SLE patients that display the cellular signature of defects in this mechanism of CD8 cell quality control (Cohen and Eisenberg, 1991; Lim et al., 1998; Devi et al., 1998; LeDeist et al., 1996). IV. TCR Crossreactivity and Autoimmune Disease: Viral-Derived Peptide Agonists

Several mechanisms have been suggested to explain the clinical connection between viral infections and induction or exacerbation of autoimmune disease. Release of self-antigens, enhancement of costimulatory activity, and migration of dendritic cells into sites of infection have all been suggested as contributing factors leading to activation of autoreactive T-cells. A critical question concerns the potential requirement for expression of

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

221

self-antigen by the microbial invader. This mechanism postulates that microbial determinants that mimic host antigens stimulate self-reactive Tcell clones to react against sequestered host tissues. For example, diabetes does not develop in mice containing pancreatic islet cells that express a transgenic LCMV gene product (nucleoprotein [NP] or glycoprotein [GP]) (Oldstone et al., 1991; Ohashi et al., 1991), because T-cells specific for these viral proteins are functionally unresponsive to the (and presumably other) self-antigens expressed on islet cells. However, LCMV infection and associated macrophage/dendritic cell presentation of viral antigens induce a strong and destructive T-cell response to islet cell viral determinants and associated diabetes. Specific triggering of autoreactive T-cells by microbial epitopes that crossreact with autoantigen has been hypothesized to account for the clinical link between infection and autoimmunity (Oldstone et al., 1991). Although viral peptides can cross-stimulate autoreactive T-cells (Wucherpfennig and Strominger, 1995; Hemmer et al., 1997) and, as discussed previously, mice that express a viral transgene in islet cells can develop diabetes after infection with the relevant virus (Oldstone et al., 1991), these observations do not provide direct evidence that viral infection may precipitate autoimmune disease by molecular mimicry. In the case of HSV1-induced keratitis, viral infection might contribute to the induction of disease by providing nonspecific inflammatory stimulus that activates efficient presentation of corneal self-peptides by professional APC. Although the intensity of the T-cell response to these corneal self-antigen(s) is inhibited by expression of the crossreactive IgG2ab-derived peptide antagonists in the thymus (Avery et al., 1995), these studies did not address the role of crossreactive viral peptides in this disease. Initial support for viral mimicry came from the finding that HSK-inducing T-cell clones also recognize HSV-1-infected cells according to proliferative responses in vitro (Zhao et al., 1998). A search of the Genbank database of HSV-1 proteins uncovered a peptide subsequence embedded in the UL6 virion-associated protein (McGeoch and Guidotti, 1997; Hay and Ruyechan, 1992; Patel and MacLean, 1995), which contained identical or homologous amino acids at 6/8 sequential residues that contribute to T-cell recognition (Zhao et al., 1998). The HSK-producing CD4 clones C1–6 and C1–15 were activated by a synthetic 15-mer peptide that contained this UL6 sequence. Intravenous administration of a soluble UL6 peptide prevented HSK, while transfer of T cells from UL6 peptide-immune C.AL-20 mice into BALB/c nu/nu mice induced severe HSK. The role of the UL6-derived peptide in HSK was further defined using an HSV-1 (strain KOS) mutant virus that contained a single base pair mutation that generated a stop codon in the 5⬘ end of the UL6 open

222

CANTOR

reading frame (Zhao et al., 1998). Pathogenic CD4 clones that are activated by protein extracts of wild-type HSV-1 did not respond to KOS/UL6m viral extracts and CD4 cells immune to wild-type, but not mutant, KOS/UL6m transferred disease. A comparison of HSK induction by KOS/UL6m with two other replication-defective HSV-1 (KOS) mutant viruses was revealing. The HSV-1–KOSICP27m mutant is deficient in expression of late viral proteins including UL6 (McCarthy et al., 1989). The HSV-1 mutant K082 expresses all viral proteins except the late viral protein glycoprotein B (gB) (Cai et al., 1987). Corneal infection by the K082 mutant caused severe HSK in about 85% of mice whereas corneal infection with either KOS/ UL6m or KOS/ICP27m did not result in detectable HSK (Zhao et al., 1998). The finding that deletion of UL6, but not gB, prevents HSK after infection by these mutant viruses provides evidence that peptide mimics play an important role in the development of a virally induced autoimmune disease. These results, taken together with the studies discussed earlier, suggest that susceptibility to the development of autoimmunity may be regulated by two opposing peptide-sharing mechanisms. Genetic polymorphisms that affect a protein’s sequence (Avery et al., 1995) and tissue expression (von Herrath et al., 1994) can generate endogenous molecular mimics that antagonize the T-cell response to sequestered self-antigens, thus decreasing susceptibility to autoimmune disease. On the other side of this coin are exogenous microbial mimics that enhance the development of autoimmunity by attraction and activation of autoreactive T-cell clones. Peptide crossreactivity by microbial determinants may allow translation of low-level viral infections into autoimmune responses, whereas more virulent infections may induce inflammatory responses that provoke autoimmune disease with less help from crossreactive viral peptides. A comparison of HSK in transgenic mice containing T-cells which all carry a receptor specific for the HSV-1 UL-6 peptide or for an unrelated peptide should clarify the relative roles of crossreactivity and inflammation in the pathogenesis of virally induced autoimmune disease. Although these data pinned down a role for crossreactive activation in a murine model of disease, the contribution of this mechanism to human disease has not been established. Recently, crossreactive activation has been implicated in the pathogenesis of Lyme disease (Gross et al., 1998) and a murine model of heart disease (Bachmaier et al., 1999). The clue in Lyme disease came from the clinical observation that a cohort of patients develops chronic arthritis despite curtailment of Borrelia burgdorferi infection by antibiotics. Treatment-resistant Lyme arthritis is associated with T-cell reactivity to the outer surface protein A (OspA) of the pathogen and expression of a particular class II MHC allele. CD4 cell reactivity in these patients is directed mainly at a dominant OspA peptide that is closely related to a peptide derived

224

CANTOR

tion of IL-10. According to this view, early and prolonged IL-12 production and IL-10 inhibition is an essential prelude to Type 1 immunity and disease progression (Segal et al., 1998). However, the events that lead to IL-12 rather than IL-10 expression early in the course of a normal immune response have been clouded with uncertainty. Although an interaction between CD40 ligand on activated T-cells and CD40 on macrophages can induce IL-12 expression (Koch et al., 1996), recent analyses of the significance of this interaction in the absence of other cellular factors suggest that it is not sufficient for physiologic induction of IL-12 (Ria et al., 1998), and the role of other pathways to macrophage IL-12 expression has not been clear. A gene product that may play an early and critical role in these responses is a T-cell cytokine termed Eta-1 (for early T-lymphocyte activation-1) (Patarca et al., 1989). This gene encodes the most abundant newly transcribed mRNA species expressed after TCR ligation of CD4 cells and specifies a phosphoprotein that is secreted by activated CD4 T-cells very early (within 12 h) after viral or bacterial infection in vivo, in contrast to standard cytokines, which are not detected at the protein level until 48–72 h after in vivo infections (Patarca et al., 1989). This gene product is also expressed by osteoblasts, may play a role in bone remodeling, and has been termed osteopontin (Opn) to reflect this function. An interaction between Eta-1/Opn and its receptors on macrophages that leads to macrophage activation (Lampe et al., 1991) and chemotactic migration in vitro (Weber et al., 1996) and in vivo (Singh et al., 1990) may account for genetic resistance to several intracellular pathogens (Patarca et al., 1989, 1993). Eta-1/Opn is transcribed at higher levels in Eta-1/Opna mouse strains and is associated with resistance to rickettsial pathogens and viral pathogens, including those responsible for yellow fever and Japanese/St. Louis B encephalitis (Patarca et al., 1993). Eta-1b strains, which transcribe low levels of Eta-1/Opn, are highly susceptible to lethal intracellular infection by these organisms (Patarca et al., 1989). For example, inbred strains carrying the resistant Eta-1a allele completely suppress rickettsial growth after inoculation of 106 organisms, whereas susceptible Eta-1b strains die from overwhelming bacterial infection after inoculation of as few as 103 organisms (Patarca et al., 1989). Genetic resistance to these pathogens reflects an Eta-1/Opn interaction with both CD44 and 움v웁3 integrins that promotes macrophage migration and activation (Singh et al., 1990; Weber et al., 1996; Mullbacher and Lobigs, 1995). Overexpression of Eta-1/Opn may also play a central role in the pathogenesis of autoimmune disease in MRL/lpr mice (Patarca et al., 1990; Lampe et al., 1991). Our studies indicate that mice deficient in Eta-1/Opn gene expression secondary to targeted gene mutation fail to develop DTH responses and

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

223

from the human LFA-1 protein. The demonstration of autoreactivity against a crossreactive epitope of LFA-1 suggests that treatment-resistant Lyme disease is an autoimmune process against an LFA peptide that is triggered by crossreactive activation by B. burgdorferi. An attractive feature of this hypothesis is that upregulation of LFA-1 expression on activated T-cells and macrophages in inflamed joints may allow more efficient presentation of this self-antigen to OspA-activated T-cells. A second example comes from the finding that peptides derived from chlamydia that are structurally related to peptides derived from murine myosin can trigger T-cell-dependent inflammatory responses in heart muscle (Bachmaier et al., 1999). T-cells from mice immune to these chlamydia peptides also transfer myocardial inflammatory responses to adoptive hosts, and chlamydial infection provokes heart inflammation that is associated with antibodies to peptides derived from both the bacterium and the heart. Although these findings indicate that the association between the chlamydial infection and inflammatory heart disease reflects molecular mimicry, the role of this process in clinical atherosclerosis is less clear. The inflammatory lesions in this murine model are confined to the heart muscle itself, with little apparent extension into coronary arteries. Nonetheless, these studies lay the groundwork for studies of T-cell reactivity to chlamydial peptides in patients with documented chlamydial infection in atherosclerotic plaques. V. Cytokine Checkpoints in the Generation of Autoimmune Disease

Recognition of autoantigen is necessary but not sufficient for autoimmune disease. A prolonged Type 1 immune response may also be required for induction and progression of organ-specific disease. These disorders have been associated with Th1 effector cytokines including 애-interferon and TNF-움 and are inhibited by Th2 cytokines including IL-4. However, other studies suggest that the focus on effector Th1 cytokines per se may be too narrow, because EAE (experimental autoimmune encephalomyelitis) can be induced in mice that have targeted mutations in either the 애IFN gene or the 애-IFN-receptor (Ferber et al., 1996; Krakowski and Owens, 1996; Willenborg et al., 1996; Segal et al., 1998), while injection of neutralizing anti 애-IFN antibodies exacerbates EAE (Billiau et al., 1988; Duong et al., 1992; Heremans et al., 1996; Lublin et al., 1993). Moreover, mice doubly deficient in expression of TNF-움 and lymphotoxin-움 develop EAE (Frei et al., 1977), indicating an extensive level of redundancy at this late stage of a Type 1 autoimmune response. A more critical and nonredundant component of organ-specific autoimmunity may be macrophage/dendritic cell production of IL-12 and inhibi-

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

225

associated Th1-driven autoimmunity after viral infection (Ashkar et al., 2000). The defective immune responses of Eta-1/Opn⫺/⫺ mice are associated with diminished in vivo production of IL-12 and excessive production of IL-10. The interaction of Eta-1/Opn with macrophages is mediated through two functional receptors. An interaction between the N-terminal portion of Eta-1/Opn and its integrin receptor, 움v웁3, induces IL-12 secretion whereas a sequential interaction of Eta-1/Opn with the CD44 receptor inhibits of IL-10 expression. These findings identify Eta-1/Opn as a critical and nonredundant cytokine that initiates Type 1 immunity and provides a mechanism by which T-cell recognition of foreign antigens may regulate the development of antimicrobial inflammatory responses. Eta-1/Opn is secreted in nonphosphorylated and phosphorylated forms (Stordeur et al., 1995). Phosphorylation may allow Eta-1/Opn to associate with the cell surface rather than the extracellular matrix (Stordeur and Goldman, 1998) through a contribution to integrin binding. In contrast, serine phosphorylation of recombinant Eta-1/Opn is not required for CD44-dependent interactions leading to chemotactic migration (Weber et al., 1996). Dephosphorylation of naturally produced Eta-1/Opn abolishes IL-12 stimulatory activity whereas phosphorylation of bacterial recombinant Eta-1/Opn at specific sites restores activity. Moreover, although bacterial recombinant Eta-1/Opn lacking phosphate groups cannot induce IL12, this molecule retains inhibitory activity for the macrophage IL-10 response. There is abundant evidence that phosphorylation can regulate the biological activity of intracellular enzymes and their substrates. These results indicate that serine phosphorylation can also provide molecular information that regulates the biological activity of a secreted protein. The potential definition of Eta-1/Opn as an essential molecular link between T-cell recognition and cell-mediated immunity fills a logical gap in our understanding of the cellular and molecular mechanisms of Type 1 immunity. Although downregulation of CD40 ligand expression by 애IFN and soluble CD40 occurs within 24 h after viral infection, IL-12 is detected in serum over the next 7–10 days (Ria et al., 1998). Replacement of the CD40L signal by Eta-1/Opn may be essential to potentiate the IL12 response and allow Type 1 immunity. An intriguing feature of the regulatory effects of Eta-1/Opn is its ability to inhibit IL-10 and, thus, upregulate production of Type 1 cytokines other than IL-12. Although mice deficient in IL-12 gene expression develop fewer 애-IFN-producing T-cells following immunization, they retain the ability to mount a reduced but significant DTH response, the cellular signature of Type 1 immunity (Singh et al., 1990). Concomitant inhibition of IL-10 expression by Eta1/Opn may allow increased production of Type 1 cytokines that together

226

CANTOR

give rise to efficient development of cell-mediated immunity and may increase the likelihood of autoimmune sequelae. These studies may form the basis for new therapeutic approaches to several diseases. For example, Eta-1/Opn analogs that mediate CD44dependent inhibition of IL-10 may reduce septic complications in burn patients (Schmits et al., 1997), while antagonists of this pathway may ameliorate bacterial arthritis (Salih et al., 1995). Engineered forms of Eta1/Opn that imprint Type 1 responses following immunization may be valuable components of viral and cancer vaccines. In the context of this review, Eta-1/Opn analogs that inhibit the interaction between Eta1/Opn and the 움v웁3 receptor leading to IL-12 and/or CD44-dependent enhancement of the IL-10 response represent potentially useful therapeutic approaches to organ-specific autoimmune disease. REFERENCES Abromson-Leeman, S. R., and Cantor, H. (1983). Specificity of T-cell clones for antigen and autologous major histocompatibility complex products determined specificity for foreign major histocompatibility complex products. J. Exp. Med. 158, 428–437. Alam, S. M., Davies, G. M., Nasholds, W., Jameson, S. C., Hogquist, K. A., Gascoigne, N. R. J., and Travers, P. J. (1999). Qualitative and quantitative differences in T cell receptor binding of agonist and antagonist ligands. Immunity 10, 227–237. Albert Basson, M., Bommhardt, U., Cole, M. S., Tso, J. Y., and Zamoyska, R. (1998). CD3 litigation on immature thymocytes generates antagonist-like signals apporpriate for CD8 lineage commitment, independently of T cell receptor specificity. J. Exp. Med. 187(8), 1249–1260. Ashkar, S., Weber, G. F., Panoutsakopoulou, V., Sanchirico, M. E., Jannson, M., Zawaideh, S., Rittling, S. R., Denhardt, D. T., Glimcher, M. G., and Cantor, H. (2000). Eta-1 (osteopontin): An early component of type 1 (cell-mediated) immunity. Science 287, 860–864. Ashwell, J. D., Chen, C., and Schwartz, R. H. (1986). High frequency and nonrandom distribution of alloreactivity in T cell clones selected for recognition of foreigh antigen in association with self class II molecules. J. Immunol. 136, 389–395. Avery, A. C., Zhao, Z.-S., Rodriquez, A., Bikoff, E. K., Soheilian, M., Foster, C. S., and Cantor, H. (1995). Resistance to herpes stromal keratitis conferred by an IgG2a-derived peptide. Nature 376, 431–434. Bachmaier, K., Neu, N., delaMaza, L. M., Pal, S., Hessel, A., and Penninger, J. M. (1999). Chlamydia infections and heart disease linked through antigenic mimicry. Science 283(5406), 1238–1239. Basu, D., and Williams, C. B., Allen, P. M. (1998). In vivo antagonism of a T cell response by an endogenously expressed ligand. Proc. Natl. Acad. Sci. USA 95, 14332–14336. Berman, P. W., and Patrick, J. (1980). Linkage between the frequency of muscular weakness and loci that regulate immune responsiveness in murine experimental myasthenia gravis. J. Exp. Med. 152, 507. Bevan, M. J. (1977). Killer cells reactive to altered-self antigens can also be alloreactive. Proc. Natl. Acad. Sci. USA 74(5), 2094–2098. Bikoff, E., and Eckhardt, L. A. (1989). Presentation of IgG2a antigens to class II-restricted T cells by stably transfected B lymphoma cells. Eur. J. Immunol. 19, 1903–1909.

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

227

Bikoff, E. K., Yu, H., and Eckhardt, L. A. (1988). T cell recognition of endogenous IgG2a expressed in B lymphoma cells. Eur. J. Immunol. 18, 341–348. Bikoff, E. K. (1991). Class II-restricted IgG2ab-specific T cells recognize a signal-minus form of the V-CH3b antigen. Eur. J. Immunol. 21, 1411–1417. Billiau, A., Heremans, H., Vandekerckhove, F., Dijkmans, R., Sobis, H., Meulepas, E., and Carton, H. (1988). Enhancement of experimental allergic encephalomyelitis in mice by antibodies against IFN-gamma. J. Immunol. 140, 1506–1510. Boniface, J. J., Rabinowitz, J. D., Wulfing, C., Hampl, J., Reich, Z., Altman, J. D., Kantor, R. M., Beeson, C., McConnell, H. M., and Davis, M. M. (1998). Initiation of signal transduction through the T cell receptor requires the multivalent engagement of peptide/ MHC ligands. Immunity 9, 459–466. Cai, W. Z., Person, S., Warner, S. C., Zhou, J. H., and DeLuca, N. A. (1987). Linkerinsertion nonsense and restriction-site deletion mutations of the gB glycoprotein gene of herpes simplex virus type 1. J. Virol. 61, 714–721. Callanan, D., Peeler, J., and Niederkorn, J. Y. (1988). Characterization of rejection of orthotopic corneal allogrants in the rat. Transplantation 45, 437–443. Cao, W., Myers-Powell, B. A., and Braciale, T. J. (1994). Recognition of an immunoglobulin VH epitope by influenza virus specific class I major histocompatibility complex-restricted cytolytic T lymphocytes. J. Exp. Med. 179, 195–202. Carbone, A. M., Marrack, P., and Kappler, J. W. (1988). Demethylated CD8 gene in CD4⫹ cells suggests that CD4⫹ cells develop from CD8⫹ precursors. Science 242(4882), 1174– 1176. Chen, J.-J., Kaveri, S.-V., and Kohler, H. (1992). Cryptic T cell epitopes in polymorphic immunoglobulin regions: Evidence for positive repertoire selection during fetal development. Eur. J. Immunol. 22, 3077. Chicz, R. M., Urban, R. G., Gorga, J. C., Vignali, D. A. A., Lane, W. S., and Strominger, J. L. (1993). Specificity and promiscuity among naturally processed peptides bound to HLA-DR alleles. J. Exp. Med. 178, 27–47. Chung, D. H., Choi, E. Y., Lee, G. K., Bae, Y. M., Hahn, J. H., Kim, S. S., Song, H. G., Park, W. S., Ahn, H. S., and Park, S. H. (1996). DN4: A novel cell surface molecule on cortical epithelial cells of the human thymus. Hum. Immunol. 42(2), 124–130. Cohen, P. L., and Eisenberg, R. A. (1991). Lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu. Rev. Immunol. 9, 243–269. Dawson, C., and Togni, G. (1976). Herpes simplex eye infections: Clinical manifestations, pathogenesis and management. Surv. Ophthalmol. 21, 11–29. Demaine, A. G., Vaughan, R. W., Behn, A. R., Myles, A. B., Panayi, G. S., and Welsh, K. I. (1983). Immunoglobulin (Gm) allotype frequencies in patients with giant cell arteritis and polymyalgia rheumatica. J. Immunogenet. 10, 343. Devi, B. S., Van Noordin, S., Krausz, T., and Davies, K. A. (1998). Peripheral blood lymphocytes in SLE-hyperexpression of CD154 on T and B lymphocytes and increased number of double negative T cells. J. Autoimmun. 11(5), 471–475. Duong, T., St. Louis, J., Gilbert, J. J., Finkelman, F. D., Singh, B., and Strejan, G. H. (1992). Effect of anti-interferon-gamma and anti-interleukin-2 monoclonal antibody treatment on the development of actively and passively induced experimental allergic encephalomyelitis in the SJL/J mouse. J. Neuroimmunology 36, 105–115. Dutt, J. E., Riblet, R., and Foster, C. S. (1993). The influence of Igh-C linked loci on murine herpes keratitis. Invest. Ophthalmol. Vis. Sci. 34, 1347. Eynon, E. E., and Parker, D. C. (1992). Small B cells as antigen presenting cells in the induction of tolerance to soluble protein antigens. J. Exp. Med. 175, 131–138.

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

229

Jameson, S. C., and Bevan, M. J. (1998). T-cell selection. Curr. Opin. Immunol. 10, 214–219. Jevnikar, A. M., Grusby, M. J., and Glimcher, L. H. (1994). Prevention of nephritis in major histocompatibility complex class II-deficient MRL-lpr mice. J. Exp. Med. 179, 1137–1141. Kappler, J. W., Roehm, N., and Marrack, P. (1987). T-cell tolerance by clonal elimination in the thymus. Cell 4, 273. Kersh, G. J., and Allen, P. M. (1996). Essential flexibility in the T-cell recognition of antigen. Nature 380, 495–498. Kirberg, J., Baron, A., Jakob, S., Rolink, A., Karjalainen, K., and vonBoehmer, H. (1994). Thymic selection of CD8⫹ single positive cells with a class II major histocompatibility complex-restricted receptor. J. Exp. Med. 180(1), 25–34. Kisielow, P., Bluthmann, H., Staerz, U. D., Steinmetz, M., and vonBoehmer, H. (1988). Tolerance in T-cell-receptor transgenic mice involves deletion of nonmature CD4⫹8⫹ thymocytes. Nature 333(6175), 742–746. Koch, F., Stanzl, U., Jennewein, P., Janke, K., Heufler, C., Kampgen, E., Romani, N., and Schuler, G. (1996). High level IL-12 production by murine dendritic cells: Upregulation via MHC class II and CD40 molecular and downregulation by IL-4 and IL-10. J. Exp. Med. 184, 741–746. Koh, D. R., Ho, A., Rahemtulla, A., Fung-Leung, W. P., Griesser, H., and Mak, T. W. (1995). Murine lupus in MRL/lpr mice lacking CD4 or CD8 T cells. Eur. J. Immunol. 25(9), 2558–2562. Koller, B. H., Marrack, P., Kappler, J. W., and Smithies, O. (1990). Normal development of mice deficient in beta 2M, MHC class I proteins, and CD8⫹ T cells. Science 248, 1227– 1230. Krakowski, N., and Owens, T. (1996). Interferon-gamma confers resistance to experimental allergic encephalomyelitis. Eur. J. Immunol. 26, 1641–1646. Lampe, M. A., Patarca, R., Iregui, M. V., and Cantor, H. (1991). Polyclonal B-cell activation by the Eta-1 cytokine and the development of autoimmune disease. J. Immunol. 147, 2902–2906. Laufer, T. M., DeKoning, J., Markowitz, J. W., Lo, D., and Glimcher, L. H. (1996). Unopposed positive selection and autoreactivity in mice expressing class II MHC only on thymic cortex. Nature 383(6595), 81–85. LeDeist, F., Emile, J. F., Rieux-Laucat, F., Benkerrou, M., Roberts, I., Brousse, N., and Fischer, A. (1996). Clinical, immunological and pathological consequences of Fas-deficient conditions. Lancet 348(9029), 719–723. Lim, M. S., Straus, S. E., Dale, J. K., Fleisher, T. A., Stetler-Stevenson, M., Strober, W., Sneller, M. C., Puck, J. M., Lenardo, M. J., Elenitoba-Johnson, K. S., Lin, A. V., Raffeld, M., and Jaffe, E. S. (1998). Pathological findings in human autoimmune lymphoproliferative syndrome. Am. J. Pathol. 153(5), 1541–1550. Lublin, F. D., Knobler, R. L., Kalman, B., Goldhaber, M., Marini, J., Perrault, M., D’Imperio, C., Joseph, J., Alkan, S. S., and Korngold, R. (1993). Monoclonal anti-gamma interferon antibodies enhance experimental allergic encephalomyelitis. Autoimmunity 16, 267–274. Lyons, D. S., Lieberman, S. A., Hampl, J., Boniface, J. J., Chien, Y., Berg, L. J., and Davis, M. M. (1996). A TCR binds to antagonist ligands with lower affinities and faster dissociation rates than to agonists. Immunity 5, 53–61. Madrenas, J., and Germain, R. N. (1996). Variant TCR ligands: New insights into the molecular basis of antigen-dependent signal transduction and T-cell activation. Immunol. 8, 83–101. Maldonado, M. A., Eisenberg, R. A., Roper, E., Cohen, P. L., and Kotzin, B. L. (1995). Greatly reduced lymphoproliferation in lpr mice lacking major histocompatibility complex class I. J. Exp. Med. 181, 641–648.

228

CANTOR

Ferber, I. A., Brocke, S., Taylor-Edwards, C., Ridgway, W., Dinisco, C., Steinman, L., Dalton, D., and Fathman, C. G. (1996). Mice with a disrupted IFN-gamma gene are susceptible to the induction of experimental autoimmune encephalomyelitis (EAE). J. Immunol. 156, 5–7. Field, L. L., Anderson, C. E., Neiswanger, K., Hodge, S. E., Spence, M. A., and Rotter, J. I. (1984). Interaction of HLA and immunoglobulin antigens in type I (insulin-dependent) diabetes. Diabetol. 27, 504. Finberg, R., Burakoff, S. J., Cantor, H., and Benacerraf, B. (1978). Biological significance of alloreactivity: T cells stimulated by Sendai virus-coated syngeneic cells specifically lyse allogeneic target cells. Proc. Natl. Acad. Sci. USA 75, 5145–5149. Foster, C. S., Opremcak, E. M., Rice, B., Wells, P., Chung, H., Thompson, P., Fong, L. P., and Raizman, M. (1987). Clinical, pathologic and immunopathologic characteristics of experimental murine herpes simplex virus stromal keratitis and uveitis is controlled by gene products from the Igh-1 locus on chromosome 12. Tr. Am. Ophth. Soc. 85, 293. Foster, C. S. (1989). Herpes simplex virus-induced destructive corneal disease. Eye 3, 194–203. Frei, K., Eugster, H.-P., Constantinescu, C. S., Lavi, E., and Fontana, A. (1977). Tumor necrosis factor alpha and lymphotoxin alpha are not required for induction of acute experimental autoimmune encephalomyelitis. J. Exp. Med. 185, 2177–2182. Fuchs, E. J., and Matzinger, P. (1992). B cells turn off virgin but not memory T cells. Science 258, 1156. Gaiser, C. N., Johnson, M. J., de Lange, G., Rassenti, L., Cavalli-Sforza, L. L., and Steinman, L. (1987). Susceptibility to multiple sclerosis associated with an immunoglobulin gamma 3 restriction fragment length polymorphism. J. Clin. Invest. 79, 309. Garchon, H.-J., and Bach, J.-F. (1991). The contribution of non-MHC genes to susceptibility to autoimmune diseases. Hum. Immunol. 32, 1–30. Gross, D. M., Forsthuber, T., Tary-Lehmann, M., Etling, C., Ito, K., Nagy, Z. A., Field, J. A., Steere, A. C., and Huber, B. T. (1998). Identification of LFA-1 as a candidate autoantigen in treatment-resistant Lyme arthritis. Science 281(5377), 703–706. Hay, J., and Ruyechan, W. T. (1992). Regulation of herpes simplex virus type 1 gene expression. Curr. Topics Microbiol. Immunol. 179, 1–14. Hemmer, B., Fleckenstein, B. T., Vergelli, M., Jung, G., McFarland, H., Martin, R., and Wiesmuller, K. H. (1997). Identification of high potency microbial and self ligands for a human autoreactive class II-restricted T cell clone. J. Exp. Med. 185, 1651–1659. Heremans, H., Dillen, C., Groenen, M., Martens, E., and Billiau, A. (1996). Chronic relapsing experimental autoimmune encephalomyelitis (CREAE) in mice: Enhancement by monoclonal antibodies against interferon-gamma. Eur. J. Immunol. 26, 2393–2398. Hoffman, R. W., Sharp, G. C., Irvin, W. S., Anderson, S. K., Hewett, J. E., and Pandey, J. P. (1991). Association of immunoglobulin Km and Gm allotypes with specific antinuclear antibodies and disease susceptibility among connective tissue disease patients. Arth. Rheum. 34, 453–458. Hogquist, K. A., Jameson, S. C., Heath, W. R., Howard, J. L., Bevan, M. J., and Carbone, F. R. (1994). T cell receptor antagonist peptides induce positive selection. Cell 76, 17–27. Hogquist, K. A., Jameson, S. C., and Bevan, M. J. (1995). Strong agonist ligands for the T cell receptor do not mediate positive selection of functional CD8⫹ T cells. Immunity 3, 79–86. Ignatowicz, L., Kappler, J., and Marrack, P. (1996). The repertoire of T cells shaped by a single MHC/peptide ligand. Cell 84, 521–529. Ignatowicz, L., Rees, W., Pacholczyk, R., Ignatowicz, H., Kushnir, E., Kappler, J., and Marrack, P. (1997). T cells can be activated by peptides that are unrelated in sequence to their selecting peptide. Immunity 7, 179–186.

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

233

Willenborg, D. O., Fordham, S., Bernard, C. C. A., Cowden, W., and Ramshaw, I. A. (1996). IFN-gamma plays a critical downregulatory role in the induction and effector phase of myelin oligodendrocyte glycoprotein-induced autoimmune encephalomyelitis. J. Immunol. 157, 3223–3227. Woodland, R., and Cantor, H. (1978). Idiotype-specific T-helper cells are required to induce idiotype-positive B memory cells to secrete antibody. Eur. J. Immunol. 8, 600–606. Wucherpfennig, K. W., and Strominger, J. L. (1995). Molecular mimicry in T cell-mediated autoimmunity: Viral peptides activate human T cell clones specific for myelin basic protein (MBP). Cell 80, 695–705. Zhao, Z-S., Granucci, F., Yeh, L., Schaffer, P. A., and Cantor, H. (1998). Molecular mimicry by Herpes Simplex Virus-1: Autoimmune disease after viral infection. Science 279, 1344– 1347.

230

CANTOR

Marrack, P., Ignatowicz, L., Kappler, J. W., Boymelm, J., and Freed, J. H. (1993). Comparison of peptides bound to spleen and thymus class II. J. Exp. Med. 178, 2173–2183. Marrack, P., and Kappler, J. W. (1997). Positive selection of thymocytes bearing alpha beta T cell receptors. Curr. Opin. Immunol. 9(2), 250–255. Matechak, E. O., Killeen, N., Hedrick, S. M., and Fowlkes, B. J. (1996). MHC class II-specific T cells can develop in the CD8 lineage when CD4 is absent. Immunity 4(4), 337–347. McCarthy, A. M., McMahan, L., and Schaffer, P. A. (1989). Herpes simplex virus type 1 ICP27 deletion mutants exhibit altered patterns of transcription and are DNA deficient. J. Virol. 63, 18–27. McGeoch, J. E., and Guidotti, G. (1997). AO.1-700 Hz current through a voltage-clamped pore: Candidate protein for initiator of neural oscillations. Brain Res. 766(1-2), 188–194. Mercadal, C. M., Bouley, D. M., DeStephano, D., and Rouse, B. T. (1993). Herpetic stromal keratitis in the reconstituted scid mouse model. J. Virol. 67, 3404–3408. Metcalf, J. F., Hamilton, D. S., and Reichert, R. W. (1979). Herpetic keratitis in athymic (nude) mice. Infect. Immun. 26, 1164–1171. Moller, G. (1989). T-cell tolerance, Mls and MHC antigen. Immunol. Rev. 107, 1–200. Mullbacher, A., and Lobigs, M. (1995). Up-regulation of MHC class I by flavivirus-induced peptide translocation into the endoplasmic reticulum. Immunity 3, 207–214. Murphy, K. M., Heimberger, A. B., and Loh, D. Y. (1990). Induction by antigen of intrathymic apoptosis of CD4⫹CD8⫹TCRlo thymocytes in vivo. Science 250, 1720–1723. Nanda, N. K., Arzoo, K. K., Geysen, H. M., Sette, A., and Sercarz, E. E. (1995). Recognition of multiple peptide cores by a single T cell receptor. J. Exp. Med. 182, 531–539. Nesic, D., and Vukmanovic, S. (1998). MHC class I is required for peripheral accumulation of CD8⫹ thymic emigrants. J. Immunol. 160(8), 3705–3712. Niederkorn, J. Y. (1990). Immune privilege and immune regulation in the eye. Adv. Immunol. 48, 191–226. Ohashi, P. S., Oehen, S., Buerki, K., Pircher, H., Ohashi, C. T., Odermatt, B., Malissen, B., Zinkernagel, R. M., and Hengartner, H. (1991). Ablation of ‘‘tolerance’’ and induction of diabetes by virus infection in viral antigen transgenic mice. Cell 65, 305–317. Oldstone, M. B. A., Nerenberg, M., Southern, P., Price, J., and Lewicki, H. (1991). Virus infection triggers insulin-dependent diabetes mellitus in a transgenic model: Role of antiself (virus) immune response. Cell 64, 319–331. Opremcak, E. M., Wells, P. A., Thompson, P., Daigle, J. A., Rice, B. A., Millin, J., and Foster, C. S. (1988). Immunogenetic influence of Igh-1 phenotype on experimental herpes simplex virus type-1 corneal infection. Invest. Ophthalmol, Vis. Sci. 29, 749–754. Opremcak, E. M., Rice, B. A., Wells, P. A., and Foster, C. S. (1990). Histology and immunohistology of Igh-1-restricted herpes simplex keratitis in BALB/c congenic mice. Invest. Ophthalmol. Vis. Sci. 31, 305–312. Patarca, R., Freeman, G. J., Singh, R. P., Wie, F.-Y., Durfee, T., Blattner, F., Regnier, D. C., Kozack, C. A., Mock, B. A., Morse, H. C., III, Jerrells, T. R., and Cantor, H. (1989). Structural and functional studies of the early T lymphocyte activation 1 (Eta-1) gene. Definition of a novel T cell-dependent response associated with genetic resistance to bacterial infection. J. Exp. Med. 170, 145–161. Patarca, R., Wei, F.-Y., Singh, P., Morasso, M. I., and Cantor, H. (1990). Dysregulated expression of the T-cell cytokine Eta-1 in CD4-8-lymphocytes during the development of murine autoimmune disease. J. Exp. Med. 172, 1177–1183. Patarca, R., Saavedra, R. A., and Cantor, H. (1993). Molecular and cellular basis of genetic resistance to bacterial infection: The role of the Early T-lymphocyte activation-1/ Osteopontin gene. Crit. Rev. Immunol. 13, 225–246.

T-CELL RECEPTOR CROSSREACTIVITY AND AUTOIMMUNE DISEASE

231

Patel, A. H., and MacLean, J. B. (1995). The product of the UL-6 gene of herpes simplex virus type 1 is associated with virus capsids. Virology 206, 465–478. Penninger, J. M., and Crabtree, G. R. (1999). The actin cytoskeleton and lymphocyte activation. Cell 96, 9–12. Pereira, P., and Coutinho, A. (1989). I-E-linked control of spontaneous rheumatiod factor production in normal mice. J. Exp. Med. 170, 1825. Perkins, D. L., Listman, J. A., Marshak-Rothstein, A., Kozlow, W., Kelley, V. R., Finn, P. W., and Rimm, I. J. (1996). Restriction of the TCR repertoire inhibits the development of memory T cells and prevents autoimmunity in lpr mice. J. Immunol. 156, 4961–4968. Pestano, G. A., Zhou, Y., Daley, J., Trimble, L. A., Weber, G. F., and Cantor, H. (1999). Inactivation of misselected CD8 T cells by CD8 gene methylation and cell death. Science 284, 1187–1191. Rao, A., Allard, W. J., Hogan, P. G., Rosenson, R. S., and Cantor, H. (1983). Alloreactive Tcell clones. Ly phenotypes predict both function and specificity for major histocompatibility complex products. Immunogenetics 17(2), 147–165. Rao, A., Faas, S., and Cantor, H. (1984a). Analogues which compete for antigen binding to an arsonate-reactive T-cell clone inhibit the functional response to arsonate. Cell 36, 889–895. Rao, A., Ko, W. W., Faas, S. J., and Cantor, H. (1984b). Binding of antigen in the absence of histocompatibility proteins by arsonate-reactive T-cell clones. Cell 36, 879–888. Ria, F., Penna, G., and Adorini, L. (1998). Th1 cells induce and Th2 inhibit antigendependent IL-12 secretion by dendritic cells. Eur. J. Immunol. 28(6), 2003–2016. Robey, E., Chang, D., Itano, A., Cado, D., Alexander, H., Lans, D., Weinmaster, G., and Salmon, P. (1996). An activated form of Notch influences the choice between CD4 and CD8 T cell lineages. Cell 87(3), 483–492. Ross, J., Callanan, D., Kunz, H., and Niederkorn, J. Y. (1991). Evidence that fate of class II-disparate corneal grafts is determined by the timing of class II expression. Transplantation 51, 532–536. Rudensky, A. Y., and Yurin, V. L. (1989). Immunoglobulin-specific T-B interaction. I. Presentation of self immunoglobulin determinants by B lymphocytes. Eur. J. Immunol. 19, 1677–1683. Rudensky, A. Y., Preston-Hurlburt, P., Al-Ramadi, B. K., Rothbard, J., and Janeway, C. A., Jr. (1992). Truncation variants of peptides isolated from MHC class II molecules suggest sequence motifs. Nature 359, 429. Russell, R. G., Nasisse, M. P., Larson, H. S., and Rouse, B. T. (1984). Role of T-lymphocytes in the pathogenicity of herpetic stromal keratitis. Invest. Ophthalmol. Vis. Sci. 25, 938–944. Salier, J. P., Goust, J. M., Pandey, J. P., and Fudenberg, H. H. (1981). Preferential synthesis of the G1m (1) allotype of IgG1 in the central nervous system of multiple sclerosis patients. Science 213, 1400. Salier, J. P., Glynn, P., Goust, J. M., and Cuzner, M. L. (1983). Distribution of nominal and latent IgG (Gm) allotypes in plaques of multiple sclerosis brain. Clin. Exp. Immunol. 54, 634. Salih, E., Ashkar, S., Gerstenfeld, L. C., and Glimcher, M. J. (1995). Identification of the in vivo phosphorylated sites of secreted osteopontin from chicken osteoblasts. Ann. NY Acad. Sci. 760, 357–360. Schernthaner, G., and Mayr, W. R. (1984). Immunoglobulin allotype markers and HLADR genes in type I diabetes mellitus. Metabolism 33, 833. Schmits, R., Filmus, J., Gerwin, N., Senaldi, G., Kiefer, F., Kundig, T., Wakeham, A., Shahinian, A., Catzavelos, C., Rak, J., Furlonger, C., Zakarian, A., Simard, J. J., Ohashi, P. S., Paige, C. J., Gutierrez-Ramos, J. C., and Mak, T. W. (1997). CD44 regulates

232

CANTOR

hematopoietic progenitor distribution, granuloma formation and tumorigenicity. Blood 90, 2217–2233. Schulz, R. J., Parkes, A., Mizoguchi, E., Bhan, A. K., and Koyasu, S. (1996). Development of CD4-CD8- alpha beta TCR ⫹ NK1.1 ⫹ T lymphocytes: Thymic selection by self antigen. J. Immunol. 157(10), 4379–4389. Segal, B. M., Dwyer, B. K., and Shevach, E. M. (1998). An interleukin (IL)-10/IL-12 immunoregulatory circuit controls susceptibility to autoimmune disease. J. Exp. Med. 187, 537–546. Singh, R. P., Patarca, R., Schwartz, J., Singh, P., and Cantor, H. (1990). Definition of a specific interaction between Eta-1 protein and murine macrophages in vitro and in vivo. J. Exp. Med. 171, 1931–1942. Smith, C. I. E., Grubb, R., Hammarstrom, L., and Matell, G. (1983). Gm allotypes in Swedish myasthenia gravis patients. J. Immunogenet. 10, 1. Smith, C. I. E., Grubb, R., Hammarstrom, L., and Pirskanen, R. (1984). Gm allotypes in Finnish myasthenia gravis patients. Neurology 34, 1604. Smith, R. B., McDonald, H. R., Nesburn, A. B., and Minckler, D. S. (1980). Arch. Ophthalmol. 98, 1226–1229. So, A., John, S., Bailey, C., and Owen, M. J. (1987). A new polymorphic marker of the Tcell antigen receptor 움 chain genes in man. Immunogenetics 25, 141. Sredni, B., and Schwartz, R. H. (1980). Alloreactivity of an antigen-specific T-cell clone. Nature 287, 855. Stordeur, P., Schandane, L., Durez, P., Gerard, C., Goldman, M., and Velu, T. (1995). Spontaneous and cycloheximide-induced interleukin-10 mRNA expression in human mononuclear cells. Mol. Immunol. 32, 233–239. Stordeur, P., and Goldman, M. (1998). Interleukin-10 as a regulatory cytokine induced by cellular stress: Molecular aspects. Int l. Rev. Immunol. 16, 501–522. Surh, C. D., Lee, D. S., Fung-Leung, W., Karlsson, L., and Sprent, J. (1997). Thymic selection by a single MHC/peptide ligand produces a semidiverse repertoire of CD4⫹ T cells. Immunity 7, 209–219. Suzuki, H., Eshima, K., Takagaki, Y., Hanaoka, S., Katsuki, M., Yokoyama, M., Hasegawa, T., Yamazaki, S., and Shinohara, N. (1994). Origin of a T cell clone with a mismatched combination of MHC restriction and coreceptor expression. J. Immunol. 153(10), 4496– 4507. Tanchot, C., Lemonnier, F. A., Perarnau, B., Freitas, A. A., and Rocha, B. (1997). Differential requirements for survival and proliferation of CD8 naive or memory T cells. Science 276(5321), 2057–2062. Tourne, S., Miyazaki, T., Oxenius, A., Klein, L., Fehr, T., Kyewski, B., Benoist, C., and Mathis, D. (1997). Selection of a broad repertoire of CD4⫹ T cells in H-2MaO/O mice. Immunity 7, 187–195. van Snick, J. L. (1981). A gene linked to the Igh-C locus controls the production of rheumatoid factor in the mouse. J. Exp. Med. 152, 738. von Boehmer, H., and Fehling, H. J. (1997). Early alpha beta T cell development in the thymus of normal and genetically altered mice. Curr. Opin. Immunol. 9(2), 263–275. von Herrath, M. G., Dockter, J., and Oldstone, M. B. (1994). How virus induces a rapid or slow onset insulin-dependent diabetes mellitus in a transgenic model. Immunity 1, 231–242. Weber, G. F., Ashkar, S., Glimcher, M. J., and Cantor, H. (1996). Receptor-ligand interaction between CD44 and Osteopontin/Eta-1. Science 271, 509–512. Weiss, S., and Bogen, B. (1991). MHC class II-restricted presentation of intracellular antigen. Cell 64, 767.

This Page Intentionally Left Blank

236

MELIEF et al.

feasible at present. However, as subgroups of cancer induced on the basis of known genetic predisposition are being identified, this situation may change, and preventive vaccines may be directed against prospective tumor constituents that are likely to be overexpressed or mutated. The only current form of therapy that may be effective against tumor burdens larger than minimal residual disease in a high proportion of patients is therapy with specific T cells. This is illustrated by the successful therapy of relapsed chronic myelogenous leukemia (CML). Of the CML patients treated by allogeneic bone marrow transplantation, approximately 20% show leukemia recurrence. Full remission in most of these patients can subsequently be achieved by adoptive immunotherapy involving the infusion of lymphocytes from the marrow donor. The beneficial effect of lymphocyte transfusion is intimately connected to the anti-leukemic activity of transplanted T cells (Kolb and Holler, 1997). Similarly, EBV-induced lymphomas in recipients of allogeneic bone marrow were treated successfully by adoptive transfer of donor-derived T cells enriched for EBVspecific cells (O’Reilly et al., 1998; Rooney et al., 1998). EBV-specific cytotoxic T lymphocytes (CTL) were shown to be responsible for the therapeutic efficacy of the infused T cell populations (Heslop et al., 1996; O’Reilly et al., 1998; Rooney et al., 1998), bearing out the remarkable capacity of adoptively transferred CTL to eradicate large tumor masses in mouse models (Kast et al., 1989; Greenberg, 1991; Melief, 1992). Adoptively transferred CD4⫹ T cells also possess substantial antitumor activity, even against MHC class II-negative tumors (Greenberg, 1991; Toes et al., 1999). Adoptive transfer of T cells thus can serve as rescue therapy in individual patients when other forms of therapy have failed or are futile by current technology because the disease is too far advanced. On the other hand, this type of therapy is tailored to the individual patient, is laborious and costly, and may not work as efficiently with autologous expanded T cells directed against tissue-specific tumor-associated antigens, because the most potentially reactive therapeutic T cells may have been deleted by tolerance-inducing mechanisms. Traditionally, tumor escape mechanisms such as production by tumor cells of immunosuppressive factors (e.g., TGF웁, IL-10) and loss of MHC expression or tumor-associated antigens have been considered the greatest enemies of successful immunotherapy. Although tumor escapes are well documented, it seems likely that an even more formidable barrier to immunotherapy is successful maintenance of (self ) tolerance by cancers. In this respect tumors do not differ from the normal tissues in which they arose. In particular, high-dose antigens in normal tissues are probably processed and presented in a continuous fashion by a phenomenon known as ‘‘cross-presentation’’ (see following), leading to both MHC class I- and

ADVANCES IN IMMUNOLOGY, VOL. 75

Strategies for Immunotherapy of Cancer CORNELIS J. M. MELIEF, RENE´ E. M. TOES, JAN PAUL MEDEMA, SJOERD H. VAN DER BURG, FERRY OSSENDORP, AND RIENK OFFRINGA Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, 2300 RC Leiden, The Netherlands

I. Introduction

The prevailing mood among immunologists and oncologists concerning immunotherapy of cancer is still one of gloom. Indeed, as of this writing, the impact of immunotherapy on oncological practice is modest at best. Nevertheless, as this review will illustrate, many of the reasons for the failures can now be understood on the basis of which strategies can be shifted. These insights and the undeniable striking successes in defined clinical conditions and animal models can lead to more rationally designed and sophisticated forms of immunotherapy. Let us first examine some of the striking clinical successes. Despite the fact that many human cancerassociated viruses establish persistent infection in a high proportion of infected individuals, preventive vaccination against such viruses can be highly effective. This was first demonstrated by preventive vaccination of young children against hepatitis B virus (HBV). This vaccine reduced the rate of chronic infection from 10% to less than 1% and was associated with a striking reduction in the incidence of hepatocellular carcinoma many years later (Chang et al., 1997). The important message from this work is that preventive vaccination against tumor viruses, which are associated with 15–20% of all cancers, can be successful, whereas so-called therapeutic. vaccination in patients with advanced virus-associated cancer is likely to fail miserably. An exception to this rule is the cancer patient whose tumor burden has been reduced significantly by conventional therapy. In this situation of minimal residual disease, vaccination could be used as adjuvant therapy, provided that neither the disease process nor the prior therapy has compromised the patient’s capacity to respond to the vaccine. Preventive vaccines against human tumor viruses other than HBV are being developed. A case in point is human papilloma virus (HPV), oncogenic variants of which constitute a leading cause of cancer-related death among women in less developed countries. Large field trials of preventive HPV vaccination are likely to start soon (WHO report, 1999). These HPV vaccines are based on the immunizing potency of viruslike particles (VLP), constituting naturally folded viral envelopes without the viral genome. Preventive vaccination for the major human cancer types not associated with viruses is not 235

Copyright 䉷 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2776/00 $35.00

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

237

II-restricted peptide display on professional antigen-presenting cells (APC) (Heath et al., 1998). Because normal tissues and many early-stage tumors do not provide the inflammatory stimuli required to turn the professional APC into T-cell-activating agents, and therefore keep them in a resting tolerizing state, T cells recognizing the cross-presented antigens will be tolerized or anergized (Matzinger, 1994; Heath et al., 1998; Toes et al., 1999). A large part of this review will therefore be spent on a discussion of strategies to counteract immunological tolerance induced by tumors. If tumor cells express lower doses of antigens, chances are that the immune system simply ignores antigen expression by the tumor or preneoplastic lesions, even in the case of virus-induced tumors. This state of ‘‘immunological ignorance’’ differs from tolerance in that T cells from such individuals behave as virgin lymphocytes that have never encountered antigen (Starzl and Zinkernagel, 1998). To what extent tumor antigens induce tolerance or are associated with ignorance (indifference) is unclear at present. Our own bias is that there is much more active tolerance induction going on than previously realized, particularly because the sensitivity of the exogenous pathway for processing of MHC class I-presented antigens has been grossly underestimated. For example, infection by oncogenic human papillomaviruses (HPV) is considered to be associated with immunological ignorance (Starzl and Zinkernagel, 1998), but this does not explain why many young women apparently manage to eradicate early HPV-induced lesions and become completely virus negative (Ho et al., 1998). Perhaps the balance between ignorance and immunity is very delicate, because many other women do not manage to rid themselves of the virus and some go on to develop preneoplastic lesions which may progress to cervical cancer. Another example of the importance of cross-presentation for class I-restricted responses is the exquisite efficiency by which professional APC acquire exogenous antigen for priming of CTL responses following intramuscular gene vaccination (Tighe et al., 1998). Transgene-encoded antigens expressed by the muscle cells apparently have no problem in arriving in professional APC of draining lymph nodes. This might go unnoticed if strong immunostimulatory sequences in DNA vaccines, causing APC activation, would not make sure that CTL immunity rather than tolerance is the outcome (Tighe et al., 1998). The inescapable concept thus stands out that, depending on their state of activation, professional APC, also named dendritic cells (DC), either tolerize or activate T cells. CD4⫹ helper T cells appear somewhat easier to activate than CD8⫹ cytotoxic T lymphocyte precursors. Indeed, activated CD4⫹ cells appear to be required to activate DC into a state permitting CTL induction (‘‘license to kill’’ concept,’’ see following) in a remarkable three-cell-type interaction (Schoenberger et al., 1998b; Bennett et al., 1998; Ridge et al., 1998). Novel therapeutic ap-

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

239

immune vitiligo (Ogg et al., 1998), suggesting that these CTL are at least in part responsible for the destruction of skin melanocytes in this disease. CTL against melanocyte differentiation antigens can also be retrieved from healthy donors (Visseren et al., 1995; Bakker et al., 1995; Van Elsas et al., 1996; Chen et al., 1998). This suggests that these CTL, like those in nonregressing melanoma patients, are checked and balanced in vivo, too low in numbers, or of insufficient affinity to mediate significant destruction. Assays are therefore needed to better monitor T cell responses in cancer patients, particularly assays that identify truly in vivo tumoricidal CTL and distinguish these from tumor-reactive CTL, sometimes referred to as tumor-observing lymphocytes, that lack biting force or at least need additional prodding. The second major line of evidence for natural protective immunity against cancer comes from the observation that immunodeficient or immunosuppressed patients suffer from a markedly increased cancer incidence, particularly cancer induced by tumor viruses. Patients with AIDS exhibit a strongly increased incidence of EBV-induced lymphomas and of Kaposi sarcomas associated with the sexually transmitted human herpesvirus type 8 (HHV8; Kedes et al., 1996; Gao et al., 1996). EBV-induced lymphomas also emerge frequently in the immunodeficient period following bone marrow transplantation. These tumors, in contrast to the rare EBV lymphomas arising in immunocompetent individuals, express a wide array of EBVassociated antigens and can therefore efficiently be treated with adoptive transfer of donor-derived EBV-specific T cells (Heslop et al., 1996; O’ Reilly et al., 1998; Rooney et al., 1998). Patients receiving long-term immunosuppressive treatment following renal allograft transplantation also experience a markedly increased incidence of malignancies, which has risen further since the introduction of cyclosporin A (Hiesse et al., 1997; Newstead, 1998). Again, many of these appear to have a viral etiology. Particularly frequent are skin cancers, which are likely the result of failure of immunosurveillance against ultraviolet light- and HPV-induced neoplasia, EBV-induced lymphomas, HHV8-associated Kaposi sarcoma, and HPVpositive cervical cancer (Newstead, 1998). The fields of tumor immunity and autoimmunity converge in the case of the so-called paraneoplastic neurologic disorders. Patients affected typically carry a malignancy of neuroendocrine origin, such as small-cell lung carcinoma or certain types of gynecologic cancer, in the course of which they develop severe paraneoplastic cerebellar degeneration (PCD). In these patients both antibody and CTL responses are found against the antigen cdr2, which is shared between neuronal tissues and tumor (Darnell, 1996; Albert et al., 1998). Cancer patients with PCD appear to have a more favorable clinical course than patients without it, supporting the notion

238

MELIEF et al.

proaches must deal with this model. While the exact DC activation state required for optimal protective T cell response remains to be identified, the realm of molecularly defined adjuvants that directly activate DC to the desired state in vivo comes within view. In two recent studies addressing this issue, in vivo triggering of CD40 converted tolerance of CD8⫹ or CD4⫹ tumor-specific T cells into protective immunity (Diehl et al., 1999; Sotomayor et al., 1999). Rather than focusing on improved performance of DC, one might also consider taking the brakes off costimulation at the T cell level by blockade of the inhibitory receptor CTLA-4 on T cells (Hurwitz et al., 1998). This approach has been shown to act synergistically with vaccination by GM–CSF-transduced tumor cells (Hurwitz et al., 1998; Van Elsas et al., 1999). In the ensuing pages we discuss the several aspects of antitumor immunity, ranging from natural protective immunity against cancer to the multiple immunoregulatory mechanisms that must be considered in developing effective immunotherapeutic strategies for patients in whom such natural immunity is lacking or failing. II. Natural Protective Immunity against Cancer

Evidence that natural immune responses protect against cancer comes from two main sources. First, for tumors of different histologic types, including melanoma, medullary breast carcinoma, gastric carcinoma, bladder carcinoma, seminoma, choriocarcinoma, neuroblastoma, and glioblastoma, the presence of T lymphocytes constitutes an important prognostic factor (reviewed in Clemente et al., 1998). Recently, infiltration of human colorectal cancer tissue by CD8⫹ T cells was also found to be a favorable prognostic sign. The impact of this was similar to—but independent of— Duke’s staging (Naito et al., 1998). Although with some exceptions, particularly melanoma, the identity of nonviral tumor antigens recognized by tumor-specific T cells has not been elucidated, these data provide a strong impetus to the search of tumor antigens recognized by T cells of the autochthonous host in human cancers of diverse histologic types. In melanoma patients, tumor-specific CD8⫹ T cells usually coexist with the tumors. Apart from the evidence already cited, it is therefore hard to prove that these CD8⫹ T cells are beneficial. However, if CD8⫹ T cell responses are stimulated, e.g., by vaccination with HLA class I-binding peptides with or without additional interleukin-2 (IL-2) (Rosenberg et al., 1998; Nestle´ et al., 1998; Marchand et al., 1995), significant therapeutic responses are seen. Favorable responses are often associated with depigmentation reminiscent of autoimmune vitiligo (Rosenberg et al., 1998). Indeed, high frequencies of skin-homing melanocyte-specific CTL are found in auto

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

241

IV. Antigens Eliciting T Cell Responses Expressed by Non-Virus-Induced Tumors

A list of the most important currently known human cancer antigens on non-virus-induced tumors is provided in Table II. Most of these antigens were indentified in melanomas. The antigens of the MAGE, BAGE, and GAGE families as well as the genes encoding them have been reviewed elsewhere (Boon et al., 1994; Van den Eynde and Van der Bruggen, 1997; Rosenberg, 1999). The LAGE-1 gene was identified by representational difference analysis (Lethe et al., 1998). It is closely related to NY-ESO-1 (Chen et al., 1997), which was originally detected by antibodies in the serum of cancer patients (Chen et al., 1997). Importantly, naturally processed HLA–A2-restricted epitopes derived from NY–ESO-1/LAGE-1 have now been reported ( Ja¨ger et al., 1998; Aarnoudse et al., 1999). The PRAME antigen was originally identified as a melanoma-associated antigen recognized by a CTL clone expressing the NK inhibitory receptor p58.2. This clone could therefore recognize only the HLA–A24-presented PRAME peptide on a melanoma metastasis lacking HLA–Cw7 (Ikeda et al., 1997). Nevertheless, PRAME expression is found on a variety of cancers, including TABLE II COMMON ANTIGENS RECOGNIZED BY T CELLS ON NONVIRAL HUMAN CANCERS Tumor type Differentiation antigens Tyrosinase Gp 100 MART/1 (Melan-A) TRP-1 (gp 75) TRP-2 CEA (carcinoembryonic antigen) Testis tumor antigens MAGE-1 MAGE-2 MAGE-3 BAGE GAGE RAGE LAGE-1 NY-ESO-1 PRAME Overexpressed antigens (as a result of mutation) P53 HER-2/neu

Melanoma Melanoma Melanoma Melanoma Melanoma Colorectal carcinomas Melanoma and variety of carcinomas Melanoma and variety of carcinomas Melanoma and variety of carcinomas Melanoma and variety of carcinomas Melanoma and variety of carcinomas Renal cancer and variety of carcinomas Melanoma and variety of carcinomas Melanoma and variety of carcinomas Melanoma and variety of carcinomas and leukemias Variety of cancers Breast, ovary, and colorectal carcinomas

240

MELIEF et al.

that the same effector mechanisms are involved in both autoimmunity and antitumor responses. In some patients PCD-associated tumors even regress coincident with the onset of autoimmune neurologic disease (Darnell and DeAngelis, 1993). III. Antigens Eliciting T Cell Responses Expressed by Virus-Associated Tumors

A list of currently known human cancer-inducing viruses and the viral antigens expressed by the cancer cells is provided in Table I. An extensive description of these antigens, including literature references, is provided elsewhere (Van der Burg et al., 1999). TABLE I ANTIGENS EXPRESSED BY VIRUS-INDUCED HUMAN CANCER Virus EBV

Disease

Antigen Expression

Burkitt’s lymphoma Hodgkin’s lymphoma Nasopharyngeal carcinoma T cell lymphoma Immunoblastic lymphoma In immunocompromised individuals

冧

HTLV-1

Adult T cell leukemia

HHV-8

Kaposi’s sarcoma Primary effusion lymphoma

HBV

Hepatocellular carcinoma in HBV surface antigen (HbsAg) positive patients Hepatocellular carcinoma in HbsAg negative patients

HCV

Hepatocellular carcinoma

HPV

Cervical intraepithelial neoplasia Cervical carcinoma

EBNA-1 EBNA-1 LMP-1*, LMP-2a*, LMP-2b*

冎

EBNA-1, 2*, 3* EBNA leader protein* LMP1*, LMP2a*, LMP2b* Gag*, envelope*, reversetranscriptase, integrase, protease, tax*, rex HHV-8 shows sequence Similarity to two other oncogenic 웂herpes viruses: Herpes virus saimiri (HVS) and EBV, similarly early and late antigens will probably be expressed. Pre-S, HbsAg* (surface) HbeAg, HbcAg* (core) HbxAg, polymerase* HbxAg (HbeAg and HbcAg are detected in numerous tumors) Core*, Envelope 1* and Envelope 2*, Nonstructural proteins: NS2*, NS3*, NS4* and NS5* E1, E2, E4, E5, E6, E7*, L1 and L2 E6 and E7*

* Antigens against which natural T cell responses in patients have been observed.

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

243

cells proliferating to wild-type p53 have indeed been found (Tilkin et al., 1995). CTL responses of healthy donor T cells against both mutant and wild-type p53 sequences have been reported (Houbiers et al., 1993). One of the CTL lines generated in this study, directed against the HLA–A2binding sequence LLGRNSFEV (p53 wild-type AA 264–272), also recognized endogenously processed wt p53 on HLA-A*0201-positive human cancer lines (Ro¨pke et al., 1996). This suggests that tolerance to wild-type p53, like tolerance to the melanoma-associated self antigens discussed earlier, is not complete. On the other hand, it is much more difficult to generate high-affinity CTL against H-2- or HLA–A2-binding sequences of wild-type mouse p53 in, respectively, normal C57BL/6 or HLA-A2 transgenic mice than in their p53-deficient counterparts (Vierboom et al., 1997; Theobald et al., 1998). Irrespective of this tolerance issue, the difference in expression of p53 between tumors carrying mutant p53 and normal tissues is such that large tumors can be eradicated by adoptive transfer of cloned wild-type p53-specific CTL and IL-2 in the absence of any demonstrable toxicity to normal tissues (Vierboom et al., 1997). This makes a strong case for wild-type p53-directed CTL therapy to be explored further for human cancer, provided the tolerance issue can be adequately addressed. One possibility for circumventing this problem involves exploitation of the allogeneic T cell repertoire. This approach is based on the observation that T cell tolerance is self MHC-restricted. Stauss and coworkers were the first to demonstrate that allogeneic murine T cell cultures can be enriched for T cells that recognize the ‘‘self ’’ MHC/peptide complex of interest and that the resulting CTL exhibit antitumor reactivity in vitro and in vivo (Sadovnikova and Stauss, 1996). Of note, the frequent observation of p53-specific IgG responses in cancer patients (see above) indicates that self tolerance seems less pronounced with respect to the p53-specific Th response. Because the relevance of the tumor-specific Th response is becoming more and more apparent (e.g., Ossendorp et al., 1998), future investigation of p53-directed immunotherapy should also include the Th arm of the immune system. Various other aspects of this immunotherapy are discussed elsewhere (Vierboom et al., 1999; Chen and Carbone, 1997; DeLeo, 1998; McCarty et al., 1998). Because of its pivotal role as a tumor suppressor, gene p53 is also used for nonimmunologically oriented (gene) therapies (Gallagher and Brown, 1999). B. CEA AS A TUMOR ANTIGEN Like p53, CEA was originally detected as an overexpressed antigen, this time in human tumors, against which serum antibodies accumulate spontaneously (Gold, 1967). For a long time CEA-directed immunotherapy did not receive the attention it deserves, because considerable quantities

242

MELIEF et al.

leukemias, in particular, more than 50% of the M2 and M3 types of acute myelogenous leukemias and approximately 20% of common ALL (Van Baren et al., 1998). In our laboratory we have raised CTL clones against two HLA–A2-presented PRAME peptides that lysed a wide variety of PRAME expressing HLA–A2-positive cancer cell lines, including melanoma lines, renal cancer lines, non-small-cell lung cancer lines, and leukemia lines (our unpublished data). Autoreactive CTL against an HLA–A2-presented CEA-encoded peptide have been generated and shown to lyse HLA–A2-compatible, CEA-overexpressing tumor cells (Tsang et al., 1997). A CEA peptide with enhanced A2-binding ability as a result of single amino acid substitution also showed improved immunogenicity (Zaremba et al., 1997). A similar observation was made for the GP 100 melanoma antigen, and this peptide was used for clinical vaccination (Rosenberg et al., 1998). Accordingly, we have shown that stability of MHC class I/peptide complexes is directly correlated with immunogenicity of the peptide epitopes (Van der Burg et al., 1996). It is conceivable that the T cell immune system exhibits a certain degree of tolerance against ‘‘naturally optimal’’ epitopes derived from tumorassociated auto-antigens, leaving particularly the subdominant and cryptic determinants from these proteins as targets for immunotherapy (Schoenberger and Sercarz, 1996; Sherman et al., 1998). By using optimized variants of such epitopes, it may be possible to elicit a T cell response with maximal antitumor reactivity and minimal reactivity to normal tissue. Because melanoma-associated antigens have received ample attention in recent reviews, we have opted to discuss several other categories of tumor-associated antigens in greater detail. A. p53 AS A TUMOR ANTIGEN p53 was originally identified by antibodies found in the serum of mice bearing chemically induced sarcomas (De Leo et al., 1979). Mutation of the p53 gene is one of the most frequent events in human oncogenesis. Aberrant expression of p53 is found in approximately 50% of all human malignancies (Hollstein et al., 1991; Lane, 1994). Formation of antibodies against p53 occurs only in cancer patients, not in healthy people (Lubin et al., 1995a; Hammel et al., 1997). The highest prevalence of these antibodies is found in lung cancer patients (30%), and such antibodies are early markers of disease (Lubin et al., 1995b). Antibodies against p53 are also frequently (26%) detected in sera of patients with colorectal cancer, where p53 mutation is common, but not all patients with p53 overexpression have demonstrable serum antibodies (Houbiers et al., 1995; Hammel et al., 1997). Many of the p53 antibodies are of the IgG class (Lubin et al., 1995a), indicating an underlying CD4⫹ T helper cell response. Such helper

244

MELIEF et al.

of the antigen are shed in the circulation. As a result, antibody-mediated therapy was considered futile, because therapeutic antibodies would be trapped by circulating antigen long before they might reach the tumor. However, these considerations do not apply to processed CEA peptides recognized by T cells, in particular CTL. Tolerance of the T cell immune system to CEA, as shown for other self antigens, again appears incomplete. This was illustrated by the fact that CTL specific for an HLA–A2-presented CEA peptide, and capable of tumor cells lysis, could be generated in patients immunized with recombinant vaccinia–CEA tumor vaccine (Tsang et al., 1997). Furthermore, CEA-specific vaccination of (human) CEAtransgenic mice was shown to induce protective T cell immunity against a challenge with CEA-expressing tumor cells (Kass et al., 1999). C. HER-2/NEU AS A TUMOR ANTIGEN The HER-2/neu oncogene-encoded protein is a member of the tyrosine kinase family of growth factor receptors. Overexpression, which is correlated with poor prognosis, is found frequently in adenocarcinomas of breast, ovary, and colorectum (Slamon et al., 1987). CTL isolated from cancer patients or generated in vitro against several HLA–A2-binding peptides of HER-2/neu were found to specifically kill antigen-positive HLA–A2matched cancer cell lines (Yoshino et al., 1994; Brossart et al., 1998; Rongcun et al., 1999). However, in the hands of another group of investigators, CTL against one of these peptides, raised in secondary stimulation in vitro of PBL from patients vaccinated with HER-2/neu peptide AA369– 377, despite excellent peptide specificity, failed to lyse HLA–A2-matched HER-2/neu-overexpressing or even vaccinia-HER-2/neu recombinant virus-infected target cells (Zaks and Rosenberg, 1998). A possible explanation for this latter failure is bias for low-affinity peptide-specific CTL selection by the peptide vaccination approach. However, in other examples of clinical peptide vaccination, CTL recognizing endogenously processed MHC class I peptides were readily found (Rosenberg et al., 1998; Nestle´ et al., 1998). D. MUCIN AS A TUMOR ANTIGEN Tumor-associated mucins, in particular, MUC1, which is highly expressed at the surface of human cancer cells, can elicit both antibody (IgM) and CD4⫹ T cell responses (Kotera et al., 1994; Hiltbold et al., 1998). Until recently, the search for anti-MUC1 immune responses, as well as the development of MUC1-specific antitumor vaccines, has been focused on the large tandem repeat (TR) region of MUC1, mainly because most of the above-mentioned responses were found to be directed against epitopes in this region. CD8⫹ T cell responses have also been observed

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

245

against this TR region, although it should be noted that the peptides identified showed only low-affinity binding to the restricting MHC molecules (HLA–A11 and A2) and did not contain the preferred binding motifs (Domenech et al., 1995; Apostolopoulos et al., 1997). Yet other MUC1specific CD8⫹ T cells that recognized their targets in an MHC-unrestricted fashion were described (Magarian-Blander et al., 1998). Since then, a number of classical, HLA–A2-restricted CTL epitopes have been identified. Of note, these epitopes map outside the MUC1 TR region (Brossart et al., 1999; our unpublished data). Although human MUC1-transgenic mice were shown to exhibit tolerance for MUC1, this tolerance could be broken by a dendritic cell-based vaccine (Tempero et al., 1998; Gong et al., 1998). On the other hand, MUC1 was found to be expressed at considerable levels on activated human T cells (Agrawal et al., 1998). The significance of MUC1 as a target antigen for immunotherapy of cancer is therefore unclear at present. E. IDIOTYPE IG AS A TUMOR ANTIGEN A significant number of B cell non-Hodgkin’s lymphoma (NHL) and multiple myeloma (MM) patients do not benefit from chemo- and radiotherapy. Because the immunoglobulin (Ig) produced by the malignant B cell clone comprises a tumor-specific marker, it appears a highly attractive target for immunotherapy of NHL. Experiments in mouse models of myeloma and NHL have shown that idiotypic vaccination can induce protective antitumor immunity (Lynch et al., 1972; Campbell et al., 1987; King et al., 1993). Although both humoral and cellular anti-idiotype responses were observed, more recent evidence suggests that the tumor-protective effect of this vaccination can be largely attributed to the humoral response (Campbell et al., 1990; Syrengelas and Levy, 1999). In clinical studies, patients with B cell lymphoma have been vaccinated with tumor lg protein purified from custom-made tumor-derived hybridomas. The antigen was either coupled to keyhole limpet hemocyanin and emulsified in adjuvant or pulsed onto autologous dendritic cells. In the study with the KLH-based vaccine, 50% of the patients generated anti-idiotype humoral responses, whereas in the study with the DC-based vaccine all patients (4/4) were shown in addition to exhibit cellular immunity (Hsu et al., 1996, 1997). Importantly, several of the vaccinated patients showed complete or partial tumor regression. In a similar study, 2 of 12 MM patients that received idiotype-specific vaccination were shown to develop idiotype-specific humoral and cellular immunity, which correlated with the fact that these patients remained in complete remission (Reichardt et al., 1999). Taken together, these data show that idiotype-specific vaccination constitues a highly promising immunotherapeutic approach against Ig-expressing tu-

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

247

macrophages, whereas both cell populations are capable of processing exogenous antigen into MHC class II (Mitchell et al., 1998). Some class I-restricted viral epitopes may be exclusively generated by exogenous processing, whereas others arise commonly from exogenous as well as endogenous processing (Gil-Torregrosa et al., 1998; Stolze et al., 1998; Geier et al., 1999). Recently, priming of CTL immunity to virus-infected nonhematopoietic cells was found to require processing of exogenous antigen by professional APC (Sigal et al., 1999). It is hoped that continued analysis of the specificity of the proteasome complex and other processing enzyme systems will allow more accurate prediction of processing (Nussbaum et al., 1998). The identity of MHC class I-restricted tumor peptides can also be determined by direct elution of peptides from class I molecules and mass spectrometry sequencing (Henderson et al., 1993; Flad et al., 1998; Schirle et al., 1999; Skipper et al., 1999). Processing of exogenously derived proteins and long peptides has long been recognized as a major pathway for presentation by MHC class II molecules (reviewed in Lindner and Unanue, 1996; Nelson et al., 1997). Ingested antigens are broken down within endosomes or lysosomes where MHC class II loading with peptides takes place. Proteolytic cleavage is mediated by lysosomal cathepsins. Recently, an asparaginyl endopeptidase was found to be another important processing activity for class II peptide generation (Manoury et al., 1998). Interestingly, incubation of dendritic cells with interleukin-6 (IL-6) was shown to alter the hierarchy by which class II-restricted epitopes derived from a model antigen were processed and presented. Dendritic cells modified in this manner were capable of activating T cells against determinants that were otherwise cryptic because of poor presentation (Drakesmith et al., 1999). Defective antigen processing and MHC expression in tumor cells are discussed in Section VIII. VI. Pivotal Role of Dendritic Cells and Tumor-Specific CD4ⴙ Helper Cells in Tumor Immunity

As discussed in Section I (Introduction), bone-marrow-derived APC, conceivably DC, can either tolerize or prime T cell responses (Heath et al., 1998; Sallusto and Lanzavecchia, 1999). DC that tolerize CD8⫹ CTL precursors could be either a specialized lineage of so-called lymphoid DC (Kronin et al.,1996) or resting and activated forms (immature and mature) of the same DC lineage. Appreciation of the fundamental role of the DC activation state to tune the outcome of T cell responsiveness helps to explain why CTL responses against tumors, including those induced by noninflammatory persistent tumor viruses such as HBV, HCV, HPV, HHV8, and EBV and murine leukemia virus (MuLV), are dependent on T

246

MELIEF et al.

mors. So far, vaccination studies have been performed with patients from whom tumor cells could be harvested for generation of idiotype-producing hybridomas. It is conceivable that the PCR methodology used for detection and identification of tumor-specific idiotypes (e.g., van Belzen et al., 1997) can also be used to clone idiotype-encoding genes and to incorporate those into expression vectors for production of recombinant idiotype protein (McCormick et al., 1999). Alternatively, such a vaccine could consist of a set of synthetic peptides covering the idiotype-specific sequence (approx, 100 residues). This would permit idiotype-specific vaccination of patients for which hybridomas cannot be generated. F. DISCOVERY OF NEW TUMOR ANTIGENS ON NON-VIRUS-INDUCED HUMAN TUMORS BY SEROLOGY Undoubtedly, molecular cloning of tumor antigens recognized by CTL has strongly revitalized tumor immunology (Boon et al., 1994; Van den Eynde and van der Bruggen, 1997). Similar methodology allows the cloning of genes encoding proteins recognized by serum antibodies of tumorbearing patients (Sahin et al., 1995). In fact, many of the gene products eliciting antibody responses are also likely targets of tumor-specific autoreactive T cells, as exemplified by the aforementioned results with the NYESO-1 antigen (Chen et al., 1997). A recent survey of sera from 234 cancer patients showed that antibodies to proteins of the testis cancer family of antigens can be found regularly, whereas antibodies to the MART-1/MelanA and tyrosinase, antigens expressed by both melanoma cells and their normal counterparts, were not demonstrable in sera of 127 melanoma patients (Stockert et al., 1998; York and Rock, 1996). V. Processing of Tumor Antigens

Processing of exogenous antigens for presentation by MHC class I molecules is now a widely recognized second pathway of class I-restricted antigen presentation, next to the well-known endogenous route (for reviews see Bevan, 1995; Rock, 1996; York and Rock, 1996; Jondal et al., 1996; Reimann and Kaufmann, 1997; Heath et al., 1998). By a variety of routes, including delivery of apoptotic bodies, antigen enters into the cytoplasm and hence into the proteasome–TAP-dependent pathway. In particular, immature DC possess this capacity and ingest apoptotic cells via 움v웁5 integrin and CD36 for cross-presentation to CTL (Albert et al., 1998b). Conceivably, the normal outcome of cross-presentation is CTL tolerance, unless APC activation by CD4⫹ T cells or inflammatory stimuli takes place (Kurts et al., 1997; Matzinger, 1998). Processing of exogenous antigen into MHC class I appears to be a capacity of dendritic cells but not of

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

249

and the costimulatory molecules CD80 and CD86 (Cella et al., 1996; Shinde et al., 1996). Expression of these molecules is important for CTL priming by DC, but for full induction of protective CTL responses associated with longterm memory, CD40 ligation alone, either naturally through CD40L on CD4⫹ helper cells or by CD40 activating antibody, is unlikely to be sufficient. Although we have succeeded in overcoming peptide-induced peripheral CTL tolerance and augmenting antitumor peptide vaccine efficacy by monoclonal antibody-mediated CD40 activation in vivo (Diehl et al., 1999), this burst of tumor-specific CTL activity appears not to be long-lasting (our unpublished observations), in contrast to the long-term CTL memory associated with vaccination with the same peptide delivered by peptide-loaded DC or by adenovirus (Toes et al., 1997a, 1998a). Nonetheless, CD40 ligation led to markedly improved results of therapeutic MHC class I- or II-presented peptide vaccines, allowing complete eradication of established tumors in the absence of toxicity, whereas the same peptide vaccines in the absence of CD40 ligation had no impact on the tumors or even caused peptide-induced tolerance (Diehl et al., 1999; Sotomayor et al., 1999). Thus CD40 ligation acts as a molecularly defined adjuvant. This can reverse the induction of T cell tolerance, which appear to be an early event in tumor progression (Staveley-O’Carroll et al., 1998). Triggering or blocking of other defined molecules on DC or T cells may help to tip the balance further toward tumor eradication. CD40 is a member of the tumor necrosis factor receptor (TNF-R) super family, whereas CD40L is a member of the TNF family. Other members of both families can have profound effects on the regulation of T cell responses and T cell survival (see next paragraph). Another important target molecule for immunotherapy of cancer is the regulator of costimulation cytotoxic Tlymphocyte-associated antigen 4 (CTLA-4; see Section XI). VII. Fine Tuning of T Cell Responses by TNF(-R) Family Members

TNF-R family molecules and their ligands of the TNF family profoundly affect T cell responses and in defined cases also T cell effector function. In this review we restrict ourselves to a brief discussion of the TNF-R molecules CD95, CD40, CD30, 4-1BB, and OX-40. Detailed reviews of these molecules and their ligands can be found elsewhere (Krammer, 1999; Gravestein and Borst, 1998; Vogel and Noelle, 1998; Toes et al., 1998; Horie and Watanabe, 1998; Weinberg et al., 1998; Vinay and Kwon, 1998). Initiation of T cell responses requires proper duration and intensity of TCR-mediated signals (Lezzi et al., 1998; Hamad et al., 1998; Boniface et al., 1998; Motta et al., 1998; Delon et al., 1998). Full activation of naive T cells also requires costimulation of T cells through engagement of CD28

248

MELIEF et al.

cell help, whereas CTL responses against acute disease-causing cytopathic viruses, such as influenza virus, are without a clear need for CD4⫹ Th activity (Allan et al., 1990). In a well-defined MuLV tumor model, the need for tumor-specific help for protection against an MHC class IInegative tumor was well characterized (Ossendorp et al., 1998). Crucial CD4⫹ T cell help for CTL-mediated protection was also documented in other tumor models in mice (reviewed in Hung et al., 1998, and Toes et al., 1999). How does this CD4⫹ help operate? Assembled evidence shows that for induction of MHC class I-restricted tumor-specific immunity crosspresentation of antigens captured by DC plays a dominant role (Seung et al., 1993; Huang et al., 1994; Toes et al., 1996a). Analysis of the cellular interactions involved in CTL priming revealed that Th cells must recognize antigen on the same APC that cross-presents the CTL epitope (Bennett et al., 1997). This explains the requirement for epitope linkage between Th epitopes and CTL epitopes for induction of CTL responses (Cassel and Forman, 1988). The traditional explanation for the need of this proximity is that the CD4⫹ T cell, upon recognition of antigen on MHC class II, releases cytokines such as IL-2, which allow more pronounced activation and proliferation of CD8⫹ CTL precursors recognizing antigen presented by MHC class I. Some evidence indicates that this is not the major mechanism of CD4⫹ help for CTL, although factors released by CD4⫹ cells such as IFN-웂 do have an additional role (see following). The decisive events, however, are of a cell–cell cognate nature, in which CD40L on CD4⫹ T helper cells, which becomes upregulated on CD4⫹ cells during activation, triggers CD40 on DC (Schoenberger et al., 1998b; Bennett et al., 1998). These results confirm a central role for CD40–CD40L interactions in the generation of protective T-cell-mediated tumor immunity (Mackey et al., 1997, 1998). CD4⫹ T cell–DC interaction via CD40L–CD40 most likely empowers the DC to prime CTL, because help for CTL priming can be bypassed by activation of DC through CD40 in vitro (Ridge et al., 1998) or in vivo (Schoenberger et al., 1998; Bennett et al., 1998). Other lines of evidence indicate that CD40 signaling is part of an important pathway in T-cell-dependent APC activation. Recombinant soluble CD40L stimulates human monocytes to release proinflammatory cytokines (Kiener et al., 1995), whereas ligation of CD40 on DC or interaction between CD4⫹ T cells, notably T helper 1 cells, and DC triggers the production of IL-12 (Koch et al., 1996; Ria et al., 1998). This IL-12 production can be inhibited by blockade of CD40L on the CD4⫹ T cell (Koch et al., 1996). Although CD40 engagement is essential, IFN-웂 is required as a second signal for induction of IL-12 secretion (Snijders et al., 1998). CD40 signaling is also a robust stimulus to upregulate expression of the adhesion molecule ICAM-1

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

251

In tissues such as testis and the eye, CD95L expression is essential to maintain immune privilege (Griffith et al., 1995; Nagata 1997), which seems to be due to elimination of the infiltrating T cells. However, CD95L expression is in itself not sufficient to obtain privilege but requires additional factors. For instance, TGF-웁 was shown to inhibit the infiltration/ activation of neutrophils, which are attracted by CD95L (Chen et al., 1998). Moreover, this immune privilege is not absolute since we have recently shown that adoptive CTL therapy of intraocular tumors in mice was very efficient in the absence of any toxicity. These data suggest that the immune privilege of ocular tumors is related to their low immunogenicity rather than to the failure of effector cells to track down and destroy such tumors (Schurmans et al., 1999). The CD95/CD95L pair is essential not only for (auto-) regulation of the T cell system but also for its effector function. CD95-expressing targets can be efficiently eradicated by CTL through this mechanism (Lowin et al., 1994; Hanabuchi et al., 1994; Ka¨gi et al., 1994). In certain cases CD95L-induced cytotoxicity even appears to dominate over the perforin/granzyme B pathway (see following). B. CD40 The pivotal role of CD40 on DC in cognate interactions between CD4⫹ Th cells, immature DC, and CD8⫹ CTL precursors has been discussed in the previous paragraph, but CD40 has much broader biological effects as could be anticipated from its wide tissue distribution, including not only DC and monocytes but also B cells and a variety of other cell types outside the immune system. CD40 on B lymphocytes is required for B cell activation, proliferation, isotype switching, germinal center formation, and memory cell generation (Vogel and Noelle, 1998). In contrast to CD95, which is also expressed on B cells, CD40 prevents apoptosis of activated B cells. B cells that have received signals through the B cell receptor become resistant to CD95-mediated apoptosis, allowing a full display of the stimulatory effects of CD40 triggering (Vogel and Noelle, 1998) and of the helper activity of CD95L positive CD4⫹ helper cells. Inappropriately triggered B cells die at the hands of the latter cells. CD40 is also expressed in a variety of human lymphomas and epithelial cancers, including breast cancer. Although CD40 signaling promotes normal B cell responses, it can inhibit neoplastic B cell growth both in vitro and in vivo, the latter in SCID mice (Funakoshi et al., 1994; Murphy et al., 1995). In the same SCID model, treatment with soluble CD40L caused significantly increased survival of mice bearing CD40-expressing human breast cancer tumors (Hirano et al., 1999). Although CD40 triggering can induce apoptosis in its own right, prior exposure of a fibroblast line to anti-CD40 antibody

250

MELIEF et al.

by CD80 or CD86 on DC (reviewed in Allison, 1994). However, even in successful CD4⫹ or CD8⫹ T cell activation, for instance, following infection by LCMV or influenza virus, many of the immune T cells eventually die or become inactivated (Belz et al., 1998; Oxenius et al., 1998). It is increasingly clear that the ultimate fate of T cells is determined to a large extent by subtle signaling events involving TNF-R family members and their ligands. All TNF-R family molecules just mentioned have a proven role in this regulation, and an important role of other family members is likely to become apparent in the near future. A. CD95 CD95-mediated T cell death is necessary to prevent splenomegaly, lymphoadenopathy, and autoimmunity, as can be deduced form lpr/lpr mice, which carry a mutation in the CD95 gene, as well as from the CD95L-mutated gld/gld mice (Krammer, 1999). Upon activation, T cells express both CD95 and CD95L, yet it is only later when the T cells have exerted their function that they actually become sensitive to CD95-induced apoptosis (Klas et al., 1993; Peter et al., 1997). Clearly, resistance and sensitivity to CD95 need to be carefully regulated to allow the T cells to function, yet to prevent them from inducing autoimmunity. Several mechanisms that govern CD95-induced apoptosis have been proposed over the years. For instance, the expression of Bcl-2 is higher in resistant T cells than in sensitive ones. Costimulation through CD28 increases this expression even further (Boise et al., 1995). However, T cells from Bcl-2 transgenic mice are not resistant to CD95-induced apoptosis, indicating that expression of Bcl-2 may be required but is not sufficient for complete resistance (Strasser et al., 1995). An alternative mechanism involves the anti-apoptotic molecule cFLIP, which efficiently inhibits CD95-induced apoptosis. cFLIP was shown to be induced in resistant T cells and almost absent in the senstive cells (Irmler et al., 1997). Although this would provide an elegant way to warrant resistance of T cells, it was subsequently shown that resistant T cells fail to form a death-inducing signaling complex upon CD95 triggering (Peter et al., 1997), Since cFLIP requires this complex to exert its function (Scaffidi et al., 1999), it is conceivable that other factors are needed to guarantee a CD95-resistant status of T cells. Whatever the mechanism may be, it should endow peripheral T cells with resistance to CD95-mediated apoptosis early in the immune response, while allowing deletion of most activated T cells, except for the memory cells, when their target has been eliminated. CD95 expression and CD95induced apoptosis in lymphoid cells thus limit clonal expansion and seem to preserve tolerance, as is exemplified by the phenotype of CD95/CD95L mutant mice.

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

253

ligand (4-1BBL) is expressed on mature B cells and macrophages, as well as on activated B cells and DC (Vinay and Kwon, 1998). 4-1BBL is upregulated on B-cells by CD40 ligation and on DC by culture-induced maturation (De Benedette et al., 1997) and possibly also by CD40 ligation. E. OX-40 The Ox-40 TNF-R family member is expressed on the surface of activated CD4⫹ T cells. This appears to be yet another receptor on T cells that inhibits activation-induced T cell death, thereby enhancing both effector and memory CD4⫹ responses (Weinberg et al., 1998). Ox-40L is expressed on activated professional APC. Its expression is upregulated on B cells by activation through the antigen receptor and through CD40, and on DC through CD40. As described for 4-1BB in the previous section, ligation of Ox-40 (through a soluble Ox-40L : Ig fusion protein) enhances tumor immunity in murine tumor models (Weinberg et al., 1998; Muy-Rivera et al., 1998). F. OTHER TNF-R FAMILY MEMBERS A role for other TNF-R family members in the regulation of T cell responses is less well defined. An intriguing and potentially important player in the field is CD27, next to its ligand CD70 (Loenen, 1998; Lens et al., 1998). CD27 clearly is a costimulatory receptor on both T and B cells (Lens et al., 1998). CD70 is largely confined to lymphoid cells. Since both molecules are absent from dendritic cells, this receptor–ligand pair is not thought to be involved in T cell priming, but rather appears to control size and function of lymphocyte populations following antigen triggering (Loenen, 1998; Lens et al., 1998) VIII. Escape Mechanisms of Tumors

As has been argued, two of the most powerful escape mechanisms of tumors may be either to be ignored by the immune system (Starzl and Zinkernagel, 1998; Sogn, 1998) or to induce peripheral T cell tolerance. Tolerance occurs in particular because the majority of the tumors, like most normal tissue, lack costimulatory properties themselves and, following cross-presentation of tumor antigens by DC, fail to activate the DC to an immunizing state and instead leave them in their normal tolerizing state. In doing so, tumors do not need to develop any special feature. Just by masquerading as normal tissues, cancer lesions can cause tolerance following cross-presentation of tumor antigens (Heath et al., 1998). Other escape mechanisms of tumors usually depend on their genetic instability, thus allowing loss or gain of gene expression permitting specific

252

MELIEF et al.

rendered the cells resistant to TNF-mediated apoptosis, again pointing to opposing effects of TNF-R family members (Hess et al., 1998). C. CD30 CD30 is expressed by activated, but not by resting, B or T cells, as well as on a subset of thymocytes (Horie and Watanabe, 1998). Mice deficient in CD30 experience an impairment of thymic negative selection (Amakawa et al., 1996). The T-cell-dampening role of CD30 is also borne out by the fact that signaling through CD30 protects against autoimmune diabetes mediated by CD8⫹ CTL (Kurts et al., 1999). CD30-deficient pancreatic islet-specific CTL were roughly 6000-fold more auto-aggressive. Complete destruction of islets and diabetes were induced by the transfer of a few as 160 CD30-deficient T cells (Kurts et al., 1999). Whereas the same group of investigators had previously shown that autoreactive CD8⫹ CTL induced by cross-presentation of self antigens are deleted by a mechanism involving signaling through CD95 (Kurts et al., 1998), CD30 signaling is apparently a second distinct mechanism for deletion of autoreactive T cells. In tumor immunology, however, control of tumor-specific T cell proliferation by CD30 interacting with its ligand is highly undesirable, and ways to block this interaction, perhaps at the expense of (temporary) autoimmunity, are attractive. Interplay between different TNF-R members is illustrated by the finding that CD30 is a CD40-inducible molecule that negatively regulates CD40-mediated immunoglobulin class switching in non-antigen-selected human B cells (Cerutti et al., 1998). This suggests that CD30 dampens the effects not only of CD40-triggered T cell responses, but also of B cell responses. D. 4-1BB 4-1BB is specifically expressed on T cells. Whereas CD40 ligation provides activation and survival stimuli for DC and B cells, 4-1BB triggering promotes activation and survival of T cells, in particular, Th1 cells and CD8⫹ CTL. It thus provides a CD28-independent costimulation pathway for T cells (De Benedette et al., 1997; Kim et al., 1998; Saoulli et al., 1998; Vinay & Kwon, 1998; Takahashi et al., 1999). Entirely in line with these findings, agonistic monoclonal antibody to 4-1BB caused eradication of established poorly immunogenic tumors, accompanied by a marked augmentation of tumor-selective CTL activity (Melero et al., 1997). Conceivably, because CD40 signaling activates DC to promote CTL activation whereas 4-1BB signaling directly triggers CD8⫹ CTL, simultaneous activation of these two TNF-R members might act synergistically in promoting tumor-specific CD8⫹ CTL responses. Preliminary studies in our laboratory indicate that this is indeed the case (our unpublished observations). 4-1BB

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

255

the exception of EBNA-1 are switched off. EBNA-1 shrewdly subverts proteasome-mediated class I processing through a glycine–alanine repeat sequence that prevents the interaction of ubiquitinated polypetides with the proteasome, effectively inhibiting proteolysis and thereby MHC class I epitope generation (Shapiro et al., 1998). We have observe two instances of epitope-specific immune evasion events affecting proteasome-mediated antigen processing of a tumorvirus (MuLV)-encoded CTL epitope. One virus variant contained a single amino acid substitution within the viral epitope which caused a novel major cleavage site within the epitope. This led to premature destruction of the viral sequence (Ossendorp et al., 1996). In the second example a single amino acid substitution immediately flanking the C terminus of the viral epitope prevented precise C-terminal cleavage of the epitope which is required for proteasome-mediated generation of CTL epitopes (Beekman et al., 1999). Similarly, a mutation in p53 flanking a CTL epitope precluded proper proteasome-mediated generation of this epitope, protecting cells from lysis by specific CTL (Theobald et al., 1998). Additional strategies utilized by (human) cancer viruses to evade specific immune responses have been reviewed by us (Van der Burg et al., 1999). Another level of immune escape is production by tumor cells of immunosuppressive factors such as TGF-웁 and IL-10. IL-10 production leads to strongly reduced levels of TAP and MHC by the tumor cells (Petersson et al., 1998). Finally, tumors sometimes develop mechanisms that directly interfere with their demise by the action of cytolytic T lymphocytes. One strategy is the production of a soluble decoy receptor, termed decoy receptor 3 (DcR3), that binds to CD95L and inhibits CD95L-induced apoptosis (Pitti et al., 1998). In this study, the DcR3 gene was amplified in about half of 35 primary lung and colon cancers studied. Another strategy involves overexpression of an apoptosis-inhibiting protein called FLICE inhibitory protein (c-FLIP). This protein is the most receptor-proximal inhibitor of CD95-induced apoptosis and interferes with CD95- but not perforindependent killing by CTL in vitro (Kataoka et al., 1998). We have demonstrated that overexpression of c-FLIP in two murine tumors, whose eradication critically depends on CTL induction, results in their escape from this response. Moreover, these tumors were found to be selected for elevated c-FLIP expression in immunocompetent but not immunodeficient hosts (Medema et al., 1999). The c-FLIP-overexpressing tumor cells were normally lysed in vitro in the usual assays, for which the perforin pathway apparently sufficed. These results indicate that the CD95-dependent pathway is more important for tumor cell killing in vivo than is to be anticipated from in vitro cytolytic assays. Furthermore, our data demonstrate that blockade of the CD95 pathway through overexpression of c-FLIP, as was

254

MELIEF et al.

immune evasion strategies. Such mechanisms may be especially important for tumors that fail to tolerize—or fail to become ignored by—the immune system, perhaps because they express highly immunogenic antigens and/ or because their expansion causes inflammation. Once tumor-specific T cell immunity has been elicited, loss of MHC class I expression (H-2 or HLA) is a frequent route of escape from CTL-mediated immune attack (reviewed in Garrido et al., 1995, 1997; Ruiz-Cabello and Garrido, 1998). MHC class I expression can be of various kinds, ranging from loss of a single HLA locus to total HLA loss (Garrido et al., 1997). Loss of 웁2microglobulin synthesis is a frequent cause of total HLA loss. Even in tumor types not considered to be strongly immunogenic, loss of HLA class I expression of one type or another is very frequent (Cabrera et al., 1998). In melanoma HLA-B locus downregulation was found in association with HLA–haplotype loss (Real et al., 1998). Recently, 웁2-microglobulin gene mutations were shown to result in lack of HLA class I molecules on melanoma cells of two patients immunized with MAGE peptide (Benitez et al., 1998). Further studies are required to establish a causal relationship between vaccination and loss of HLA class I expression. Loss of HLA class I expression and melanocyte differentiation antigen expression were shown to be independent events in metastatic melanoma ( Ja¨ger et al., 1997). In addition to loss of MHC class I or 웁2-microglobulin genes or defects in MHC class I regulation, human cancer cell lines were frequently shown to exhibit deficiencies in genes involved in antigen processing ( Johnson et al., 1998). Affected gene products included the TAP1/ TAP2 peptide transporter and the LMP2, LMP7, and LMP10 (MECL-1) proteasome components. Importantly, most of these deficiencies could be restored by treatment with interferon-웂. However, cancer cells can develop unresponsiveness to IFN-웂 or even all interferons by deficiencies in, respectively, expression of the IFN-웂 receptor or the STAT protein that is involved in transducing the intracellular signal (Garrido et al., 1997). Indeed, human cancer lines that are resistant to MHC induction by IFN-움 and/or 웂 have been described (Garrido et al., 1997; Abril et al., 1998). The importance of IFN-induced antigen processing and MHC expression in cancer immunosurveillance was recently demonstrated in mice lacking either the IFN웂 receptor or the STAT1 protein (Kaplan et al., 1998). These mice showed both increased spontaneous tumor incidence and more rapid tumor development following application of the chemical carcinogen methycholanthrene. In particular, the lack of IFN-웂 responsiveness appears to decrease the tumor immunogenicity and might conceivably also act at the effector cell level by decreased sensitivity to IFN-웂-deficient T cells and NK cells. Tumor viruses can evade proteasome-mediated processing in various ways. In the case of EBV-induced Burkitt lymphomas, all viral antigens with

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

257

host disease mediated by irrelevant alloreactive T cells among the desired tumor-specific T cells. A category of tumor antigens of particular usefulness in the allogeneic setting are minor histocompatibility antigens with restricted expression to the hemopoietic system (Goulmy 1997; Simpson et al., 1998). The feasibility of expanding human CTL against such antigens from the blood of healthy allogeneic donors has been reported (Mutis et al., 1999). If the effectiveness of T cell culture and expansion from allogeneic donors can be made sufficiently high, HLA-mismatched or minor H-mismatched allogeneic stem cell transplantation with highly purified stem cells, in combination with high avidity specific T cell therapy, has much to offer and could be applied to both leukemias/lymphomas and solid tumors. Interestingly, in a murine model graft versus host disease could be suppressed and graft versus tumor effects amplified by infusion of activated NK cells following allogeneic bone marrow transplantation (Asai et al., 1998). One of the challenging questions with respect to adoptive T cell therapy is whether therapeutic T cells can be obtained in sufficient quantity and quality from tumor-bearing patients. Scanty experience from melanoma studies indicates that this is possible (Rosenberg et al., 1994). Another case in point is virus-induced cancers such as HPV-induced cervical carcinomas which express non-self antigens. Although in patients bearing HPV-positive malignancies a certain level of tumor-induced tolerance cannot be excluded, deletion of high-avidity T cells by professional APC that crosspresent antigens from normal tissues is not an issue here. In accordance with this notion, specific CD8⫹ CTL and CD4⫹ T helper cells have been observed in patients with HPV-16-induced lesions (Ressing et al., 1996; Evans et al., 1997; De Gruijl et al., 1996, 1998; Luxton et al., 1996). In the years to come proof of concept of tumor eradication with T cells retrieved from the autochthonous host will have to be delivered. These efforts should include cloning and expansion of CD4⫹ tumor-specific T cells for therapy trials by adoptive transfer. As outlined in the previous section, effective T cell immunity, even against MHC class II-negative tumors, depends heavily on tumor-specific CD4⫹ T cells, in both the induction and effector phases (Ossendorp et al., 1998). Adoptive transfer of tumor-specific CD4⫹ cells, in particular IFN-웂-producing Th1 cells, can be expected to activate DC both locally and in the draining lymph nodes. This can strongly promote the activity and expansion of preexistent or simultaneously transferred tumor-specific CD8⫹ T cells. Quite apart from this, tumor-specific CD4⫹ T cells can be expected to themselves exert effector function, because cytokines produced by Th1- or Th2-type CD4⫹ cells can recruit and activate macrophages and eosinophils, respectively (Hung et al., 1998). In this study, protection against tumor challenge was

256

MELIEF et al.

found in human melanomas (Irmler et al., 1997; Griffith et al., 1998), can indeed serve as an efficient mechanism of immune-escape by tumors. IX. Cancer Therapy by Adoptive Transfer of T Cells

The marked success of adoptive T cell therapy in tumor eradication in a setting of allogeneic bone marrow or stem cell transplantation has already been mentioned. How can we capitalize on these findings and achieve success in a wider variety of cancers? First, we should realize that the eradication of the highly antigenic EBV lymphomas that arose in immundeficient hosts is light years away from dealing with EBV lymphomas that arose in immunocompetent hosts, which express much less or even no antigens detectable by tumor-specific CTL and likely have evolved a variety of other immune evasion strategies. Nevertheless, some EBV-encoded antigens against which CTL responses have been recorded are expressed on Reed–Sternberg cells in Hodgkin’s diseases (Table 1), and CTL against these antigens could be used for immunotherapy of Hodgkin’s disease (Sing et al., 1997). Poor antigen expression and immune evasion do not apply, however, to relapsed CML in allogeneic stem cell recipients. CML constitutes a disease involving all hemopoietic lineages, including DC. It is tempting to speculate that inclusion of DC among the leukemia population contributes to the success of allogeneic T cell therapy in relapsed CML. These DC may fulfill all of the criteria for successful expansion and survival of infused therapeutic T cells, including CD40, CD28, 4-1BB, and Ox-40 signaling, preventing premature death, or lack of expansion of therapeutic T cells. In addition, allogeneic T cells have the advantage of not having undergone self MHC-restricted negative selection in the thymus. Moreover, positive selection in the context of a certain MHC molecule does not seem to be required to generate high-avidity TCR that are restricted by the same molecule. Based on this principle, high-avidity peptide-specific CTL against viral and tumor antigens can be generated from allogeneic donors in both mouse (Sadovnikova and Stauss, 1996; Obst et al., 1998) and human (Sadovnikova et al., 1998; Mu¨nz et al., 1999). From the allogeneic repertoire of T cells, highly specific T cells can therefore be retrieved not only against leukemias but also against solid tumors. The likelihood that this will be successful against solid tumors is already indicated by results of polyclonal allogeneic cell therapy in a mouse mammary carcinoma model (Morecki et al., 1998). Provided that MHC-restricted epitopes are known, advanced cell-sorting technology, involving, for instance, MHC tetramers (see Section XIII), can be employed to select and expand T cells from the allogeneic donors (Dunbar et al., 1998), minimizing the risk of graft versus

258

MELIEF et al.

strongly associated with the presence of eosinophil granulocytes and the production of oxygen radicals by tumoricidal macrophages. Of note, these CD4⫹-dependent effector mechanisms should also operate in the case of tumors that have lost MHC class I expression. Although adoptive T-cellmediated immunotherapy seems a highly promising approach, major hurdles remain in establishing technologies for the efficient and reliable expansion of specific T cells. The major problems are limitation in the number of cell divisions (life span), induction of apoptosis rather than proliferation as a result of an inappropriate ensemble of growth factors and stimuli, and loss of antigen specificity due to overgrowth of the culture by nonspecific rapidly growing lymphocytes. All tricks discussed in the preceding pages should be tried to obtain sufficient cell numbers for adoptive T cell therapy. Undoubtedly, this will lead to major biotechnological advances in adoptive T cell therapy and it is hoped, will make this therapy available to more than the handful of patients who profit by it today. X. Design of Rational Cancer Vaccines Including Molecularly Defined Adjuvants

The area of cancer vaccines has been reviewed by several investigators (Toes et al., 1997b; Offringa et al., 1999; Pardoll, 1998; Restifo and Rosenberg 1999). This discussion will therefore be restricted to what future tumor vaccines might look like and to the feasibility of therapeutic rather than preventive vaccines. Processing of exogenous protein not only for MHC class II but also for class I presentation is now a widely recognized fact. Previous failure to view exogenous class I processing as an important and efficient pathway is probably due to the fact that much exogenous class I processing and presentation were never visualized, because APC presenting the epitopes were not appropriately stimulated for CD8⫹ CTL induction. The signals for appropriate DC stimulation are now being uncovered (Sections VI and VII) and several DC and/or CD8⫹ stimuli (adjuvants) have been found capable of supporting CD8⫹ CTL responses following exogenous antigen delivery. Indeed, appropriate delivery of exogenous antigens in different formulations, including synthetic peptides in incomplete Freund’s adjuvant (IFA), proteins in IFA or other adjuvants, denatured protein, and liposome or particle-associated proteins, leads to efficient CTL induction and protection (reviewed in Jondal et al., 1996, and Offringa et al., 1999). In our hands excellent protection against murine HPV16 tumors associated with CTL memory was achieved by different types of vaccination. Effective vaccines include HPV16 synthetic peptides of the exact MHC class I fitting length (9-mer) or a 19-mer peptide with the same epitope in the middle, delivered in IFA (Feltkamp et al., 1993) or loaded onto ex vivo-activated DC (Mayordomo et al., 1995; Ossevoort

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

259

et al., 1995). Recently we showed that an E7 32-mer peptide with the epitope in the middle induced a level of CTL-mediated protection similar to that of the minimal MHC class I-binding 9-mer if delivered in IFA (our unpublished obervations). HPV16 E7 recombinant protein delivered in the MF 59 adjuvant or in IFA also induced protective T cell immunity (Zhu et al., 1995; De Bruijn et al., 1998). The immune system is apparently capable of excising the exact MHC class I-binding peptides from exogenously offered proteins and long peptides. On the basis of these results, we favor long peptides or proteins for future anticancer vaccination trials. The advantage of such an approach is that, if delivered in the appropriate adjuvant (with DC stimulatory capacity), all potential MHC class II and class I epitopes within the delivered peptides or protein will be processed and presented to host T cells. These vaccines will thus become independent of MHC binding motif prediction or processing algorithms and can be administered to subjects independent of their HLA type. Moreover, we have observed that downregulation of CTL responses, observed by us in the case of two exact MHC class I-binding adenovirus E1 peptides (Toes et al., 1996b,c), probably because the short peptides rapidly leak out of the adjuvant depot and fail to become loaded on activated DC (Diehl et al., 1999), does not occur if these epitopes are delivered as middle portions of long 32-mer peptides (our unpublished observations). The main message from this work is that, as long as proper adjuvanticity is ensured, MHC class I-restricted protective responses will be induced, regardless of the length of the peptide/protein sequences offered. The same holds in fact for DNA vaccines and viral vector-based vaccines. DNA vaccines need to have immunostimulatory sequences that activate DC. In most cases of DNA vaccination, the DC probably acquire the antigen via exogenous processing (Tighe et al., 1998). In certain cases, however, when DNA was delivered in mice via gene gun or by scarification of the ear, a major involvement in T cell priming of direct presentation by gene-transduced DC was demonstrated (Porgador et al., 1998; Akbari et al., 1999). An alternative approach to obtaining direct presentation of the antigen of interest involves DC that are transfected or infected in vitro prior to their application in vivo (e.g., Kim et al., 1997; Tuting et al., 1998; Nair et al., 1998). Is it possible to further improve the potency of peptide/protein-based cancer vaccines by the use of improved molecularly defined adjuvants? The answer is probably yes. Mouse studies have indicated that prophylactic or tolerizing peptide vaccines acquire therapeutic tumor-eradicating potential by antibody-mediated CD40 signaling (Diehl et al., 1999; Sotomayor et al., 1999). CD40–ligand trimer constitutes an alternative to CD40 signaling monoclonal antibody (Gurunathan et al., 1998). Further improvements

260

MELIEF et al.

can be expected from agonistic signaling through 4-1BB (Melero et al., 1997) and Ox-40 (Weinberg et al., 1998) and from blockade of CD30. If applied clinically, these measures should be monitored carefully for induction of autoimmune disease (see, for instance, Roskrow et al., 1999; Overwijk et al., 1999). The idea to temporarily shift the T cell activation and survival balance within the immune system by signaling and blocking of TNFR family members and their ligands (see Section VII) appears useful not only for tumor vaccine strategies but also for adoptive T-cell-based cancer therapies. Eventually it should be possible to design small molecular compounds that mimic agonistic or blocking action of individual TNFR family members. A second mode of improvement of peptide/protein-based cancer vaccines is to offer epitopes with improved immunogenicity to the immune system. This has been achieved by Rosenberg et al. (1998), who used a gp100 peptide vaccine with a methionine instead of a threonine at position 2 of the peptide. This ensured increased binding to the HLA-A *0201 molecule and resulted in CTL responses in 91% of vaccinated patients and objective clinical responses in combination with IL-2. A similar strategy to break tolerance against gp100 was used in a mouse melanoma model. The CTL elicited with a high-affinity altered peptide ligand crossreacted with self gp100, lysed tumor cells, and were therapeutically active (Overwijk et al., 1998). Improved immunogenicity was also obtained with an HLA-A*0201-binding CEA peptide, this time by substituting a TCR contact residue in the peptide, thus more easily breaking tolerance and eliciting CTL that cross-reacted with the wild-type CEA peptide and lysed CEA-overexpressing cancer cells (Zaremba et al., 1997). Recombinant viruses used in cancer vaccines include attenuated influenza or vaccinia viruses, avian poxviruses, which do not replicate in mammalian cells, or gene-deleted adenoviruses. Such viruses, encoding either entire tumor antigens (reviewed in Offringa et al., 1999; Restifo and Rosenberg, 1999) or string-of-beads arrangements of several tumor-associated CTL epitopes (Toes et al., 1997a), have been shown to elicit tumor-specific CTL responses associated with tumor protection. In one phase I vaccination study with recombinant avipox CEA virus in advanced cancer patients, CEA-specific CTL responses were induced and the vaccine was well tolerated (Marshall et al., 1999). As argued previously for the peptide/protein vaccines, disruption of tolerizing or downregulatory T cell circuits is probably required for optimal in vivo effects of these antitumor vaccines. It is unlikely that this can be achieved alone by incorporation of additional immunostimulatory sequences within the recombinant vaccines. The effects of the vaccine will be local and very temporary, particularly upon boosting with the same vaccine because of accompanying immune responses against the virus

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

261

vector, whereas a prolonged systemic T cell expansion and survival strategy is needed. Therefore, vaccination strategies are being developed that involve subsequent vaccination with different vectors encoding the same antigen (e.g., Irvine et al., 1997; Hodge et al., 1997). Furthermore, stringof-beads constructs (polyepitope constructs) can be substantially improved by the incorporation of CTL epitopes with increased binding ability as well as of tumor-specific T helper epitopes (Toes et al., 1999; Topalian et al., 1996; Manici et al., 1999; Pieper et al., 1999; Chaux et al., 1999) and perhaps by incorporation of cytokine genes such as IL-12. Like recombinant viruses, DNA vaccination also induced CTL-mediated tumor protection (e.g., Schreurs et al., 1998), although it should be noted that DNA vaccination can also result in strong T cell responses against immunodominant epitopes encoded by the vector backbone (Van Hall et al., 1998). An interesting variant of the genetic vaccination concept is the use of a selfreplicating RNA vaccine. In this vaccine design, the function of the naked nucleic acid immunogen was amplified by the incorporation of a geneencoding and RNA replicase from Semliki Forest virus. A single injection of this vaccine elicited antibody and CD8⫹ tumor-specific CTL responses, protected mice from tumor challenge, and prolonged survival of mice bearing established tumors (Ying et al., 1999). XI. Tumor Immunotherapy Based on Improved Costimulation via the CD28 Pathway

Expression of the costimulatory molecules CD80 and CD86 on professional APC is required for initiation of T cell responses. These molecules costimulate TCR-triggered lymphocytes via the CD28 costimulatory molecule. The same ligands downregulate T cell responses by binding to the CTLA-4 molecule on T cells, serving as a brake on CD28-mediated T cell costimulation. Like TNFR-mediated T cell homeostasis, this costimulatory event and its downregulation are finely regulated (Chambers and Allison, 1997). Tumor cells show enhanced immunogenicity when transfected with CD80 or CD86 (Hellstro¨m et al., 1995). Mouse embryo cells transfected with CD80 and a construct expressing an endoplasmatic reticulumtargeting signal sequence followed by a tumor-associated CTL epitope sequence, associated with very high specific peptide/MHC expression, even completely bypass professional APC for in vivo CTL induction and are capable of direct CTL-priming (Schoenberger et al., 1998a). Such a strategy overcomes poor antigen uptake by professional APC and can be useful for therapeutic purposes. Indeed, established spontaneous mammary carcinoma metastases could be eliminated or reduced following immunotherapy with a cellular vaccine consisting of tumor cells transfected with a combina-

262

MELIEF et al.

tion of CD80 and MHC class II genes. However, the primary tumor was not affected by this treatment. Still such a strategy would be useful if the primary tumor is resectable or sensitive to irradiation treatment (Pulaski and Ostrand-Rosenberg, 1998). Another sophisticated strategy is to use a single-chain monoclonal antibody recognizing a tumor antigen and couple this to the signal transduction domain of CD28. Human primary T cells retrovirally transduced with this construct proliferated and produced IL2 upon recognition of the antigen (Krause et al., 1998). Although similar ‘‘T-bodies’’ expressing tumor-specific antibody domains fused to the CD3␨ signaling domain have been shown to effectively eliminate tumor cells in vitro (review: Eshhar, 1997), true in vivo efficacy of such engineered effector cells remains to be demonstrated. The successful treatment of established poorly immunogenic tumors by combined vaccination with GM–CSF-producing tumor cells and CTLA-4 blockade (Hurwitz et al., 1998; Van Elsas et al., 1999) has already been referred to in the introduction of this review. XII. Enhancement of Tumor-Specific T Cell Responses by Cytokines and by Cytokine-Transduced Tumor Cells

The antitumor effects of IL-2 alone or in combination with adoptive CTL therapy are well known and have been reviewed elsewhere (Rosenberg et al., 1997). High-dose IL-2, however, causes severe toxicity. Other cytokines that have received considerable attention for immunotherapy of cancer are IL-12 and GM–CSF. In a murine tumor model, tumor cells transfected by IL-12 and IL-18 acted synergistically in protection against concurrently injected tumor cells (Coughlin et al., 1998b). The mechanism of action involves inhibition of angiogenesis. Tumor cells defective in IFN-웂 receptor 1 are less responsive to IL-12 therapy in vivo (Coughlin et al., 1998a). The most potent cytokine gene to enhance tumor-specific responses, if used to transfect tumor cells to create a cellular vaccine, is GM–CSF (Dranoff et al., 1993). Although GM–CSF transduction does not enhance the immunogenicity of all murine tumors, a recent clinical phase I study using GM–CSF-transduced irradiated autologous melanoma cells showed much more densely infiltrated tumor lesion after vaccination than prior to vaccination in 11 of 16 patients examined. The infiltrates included T cells and plasma cells as well as granulocytes (eosinophils and neutrophils) and were associated with evidence of substantial tumor destruction (Soiffer et al., 1998). In addition, local pharmacological administration of GM–CSF in patients with colorectal cancer enhanced antibody and proliferative T cell responses to recombinant carcinoembryonic antigen (Samanci et al., 1998). Systemic

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

263

administration of GM–CSF enhanced CTL and DTH responses to MHC class I-binding tumor-associated peptides in melanoma patients ( Ja¨ger et al., 1996). XIII. Monitoring of Tumor-Specific T Cell Responses

It has been notoriously difficult to sensitively monitor tumor antigenspecific CTL responses in cancer patients and thus document the effect of vaccination and other forms of immunotherapy. An important step forward is TCR staining with fluorescent tetrameric high-avidity peptide/ MHC complexes (Altman et al., 1996). This technology has already allowed sensitive and reliable detection of melanoma peptide-specific CTL in the blood of melanoma patients and in metastatic lymph nodes (Romero et al., 1998a, b; Dunbar et al., 1998). Alternatively, the frequency of specific T cells (either CTL or Th) can be monitored by measuring cytokine production at the single cell level, through Elispot assays (Herr et al., 1997; Romero et al., 1998a), intracellular cytokine staining (Murali-Krishna et al., 1998), or staining of secreted cytokines that have been captured at the surface of the secreting cells through bispecific antibodies (Manz et al., 1995). An advantage of tetramer staining is the possibility of discriminating between low- and high-affinity T cells (Davis, 1999), the latter of which are likely to have more impact on the eradication of tumors in vivo. On the other hand, the cytokine staining techniques can discriminate between responsive and hyporesponsive (anergic) T cells. Although tetramer and cytokine staining, at least in certain settings, yielded the same outcome in numbers of antigen-specific T cells (Murali-Krishna et al., 1998), it is very likely that this will not always be the case. For instance, in patients with advanced cancer tumor, reactive T cells may have become anergic. Such T cells would still be visible with tetramers, but not through cytokine staining. It seems therefore advisable to use both detection approaches in parallel. Of note, both tetramers and staining of extracellularly captured cytokines allow sorting of live cells on the basis of their antigen specificity. The resulting T cell populations can subsequently be expanded in vitro and are of potential interest for adoptive immunotherapy (as discussed in Section IX). XIV. Immunotherapy with Monoclonal Antibodies

While not the chief subject of this review, the impact of monoclonal antibody therapy is important and calls for intensified efforts to improve it. Anti-CD20 antibodies have shown significant clinical responses in many B cell lymphoma patients who have failed chemotherapy (Maloney et al.,

264

MELIEF et al.

1997). Anti-CD20 is now a registered anti cancer therapeutic compound. Conceivably, the therapeutic effect can be further improved by activation of Fc-receptor-bearing effector cells with cytokines such as IL-2 (Hooijberg et al., 1995). Postoperative adjuvant treatment of patients with colorectal cancer with the monoclonal antibody 17-1A, specific for epithelial cell adhesion molecule (Ep-CAM), led to an approximately 30% reduction in mortality from Duke’s stage C colorectal cancer 7 years later (Riethmu¨ller et al., 1998). Apparently this monoclonal antibody dealt effectively with micrometastases, despite the fact that it was an intact mouse monoclonal antibody with less than optimal interaction with human Fc receptors for maximal effector function. XV. Epilogue

This review shows that a plethora of different immunotherapeutic strategies mediates considerable biological effects in a variety of experimental systems and clinical studies. It will be an art to extract from these approaches those therapies that will be effective in the various clinical situations with a minimum number of toxic side effects. Probably we have to live with at least some autoimmunity if we are to cure patients with a significant burden of tumors expressing tumor-associated autoantigens. An alternative is allogeneic bone marrow transplantation in combination with immunotherapy. After the current phase of extensive phase I trials in patients with advanced cancer, the odds for success are by far the best in patients with early cancer and with minimal residual disease. Substantial reduction of the high incidence of virally induced cancer in many countries by preventive vaccination is an achievable goal. Rescue therapies for patients suffering from tumors that have escaped T cell immunity, for example, by loss of MHC expression, must be considered. CD4⫹ T cells may be active in this case, next to monoclonal antibodies. What about NK cells, whose therapeutic potential in human disease remains largely unexplored? Monoclonal antibody therapy, adoptive transfer of T cells, and various vaccination strategies must all be rigorously explored without prejudice for—or against—any particular strategy. Progress is likely to be slow and based on a shrewd combination of animal experiments and clinical trials. It can no longer be denied that this is a most exciting area of immunology. Malignant tumors are the ultimate challenge to immunologists, posing profound and basic questions concerning tolerance, escape, stimulation, death, and survival. REFERENCES Aarnoudse, C. A., van den Doel, P. B., Heemskerk, B., and Schrier, P. I. (1999). Interleukin2-induced, melanoma-specific T cells recognize CAMEL, an unexpected translation product of LAGE-1. Int. J. Cancer 82, 442–448.

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

265

Abril, E., Real, L. M., Serrano, A., Jimenez, P., Garcia, A., Cauton, J., Trigo, I., Garrido, F., and Ruiz-Cabello, F. (1998). Unresponsiveness to interferon associated with STAT1 protein deficiency in a gastric adenocarcinoma cell line. Cancer Immunol. Immunother. 47, 113–120. Agrawal, B., Krantz, M. J., Parker, J., and Longenecker, B. M. (1998). Expression of MUC1 mucin on activated human T cells: Implications for a role of MUC1 in normal immune regulation. Cancer Res, 58, 4079–4081 Akbari, O., Panjwani, N., Garcia, S., Taseon, R., Lowrie, D., and Stockinger, B. (1999). DNA vaccination: Transfection and activation of dendritic cells as key events for immunity. J. Exp. Med. 189, 169–178. Albert, M. L., Darnell, J. C., Bender, A., Francisco, L. M., Bhardwaj, N., and Darnell, R. B. (1998a). Tumor-specific killer cells in paraneoplastic cerebellar degeneration. Nature Med. 4, 1321–1324. Albert, M. L., Pierce, S. F. A., Francisco, L. M., Santer, B., Roy, P., Silberstein, R. L., and Bhardwai, N. (1998b). Immature dendritic cells phagocytose apoptotic cells via 움v웁5 and CD36, and cross-present antigens to CTL. J. Exp. Med. 188, 1359–1368. Albert, M. L., Santer, B., and Bhardwaj, N. (1998). Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs. Nature 392, 86–89. Allan, W., Tabi, Z., Cleary, A., and Doherty, P. C. (1990). Cellular events in the lymph node and lung with influenza. Consequences of depleting CD4⫹ T cells. J. Immunol. 144, 3980–3986. Allison, J. P. (1994). CD28-B7 interactions in T cell activation. Curr. Opin. Immunol. 6, 414–419. Altman, J. D., Moss, P., Goulder, P., Barouch, D. H., McHeyzer, W. M., Bell, J. I., McMichael, A. J., and Davis, M. M. (1996). Phenotypic analysis of antigen specific T lymphocytes. Science 274, 94–96. Amakawa, R., Hakem, A., Kundig, T. M., Matsuyama, T., Simard, J. J. L., Timms, E., Wakeham, A., Mittruecker, H. W., Griesser, H., Takimoto, H., H., Schmits, R., Shahinian, A., Ohashi, P. S., Penninger, J. M., and Mak, T. W. (1996). Impaired negative selection of T cells in Hodgkin’s disease antigen CD30-deficient mice. Cell 84, 551–562. Apostolopoulos, V., Karanikas, V., Haurum, J. S., and McKenzie, I. F. (1997). Induction of HLA–A2-restricted CTLs to the mucin 1 human breast cancer antigen. J. Immunol. 159, 5211–5218. Asai, O., Longo, D. L., Tian, Z. G., Hornung, R. L., Taub, D. D., Ruscetti, F. W., and Murphy, W. J. (1998). Suppression of GVHD and amplification of graft versus tumor effects by activated NK cells after allogeneic bone marrow transplantation. J. Clin. Invest. 101, 1835–1842. Bakker, A. B., Marland, G., de Boer, A. J., Huijbens, R. J., Danen, E. H., Adema, G. J., and Figdor, C. G. (1995). Generation of antimelanoma cytotoxic T lymphocytes from healthy donors after presentation of melanoma-associated antigen-derived epitopes by dendritic cells in vitro. Cancer Res. 55, 5330–5334. Beekman, N., van Veelen, P. A., van Hall, T., Camps, M., Drijfhout, J. W., Kloetzel, P. M., Melief, C. J. M., and Ossendorp, F. (2000). Flanking amino acids influence C-terminal cleavage preference for CTL epitope generation by the 20S proteasome. J. Immunol. 164, 1898–1905. Bellgrau, D., Gold, D., Selawry, H., Moore, J., Franzusoff, A., and Duke, R. C. (1995). A role for CD95 ligand in preventing graft rejection. Nature 377, 630–632 Belz, G. T., Altman, J. D., and Doherty, D. C. (1998). Characteristics of virus-specific CD8⫹ T cells in the liver during the control and resolution phases of influenza pneumonia. Proc. Natl. Acad. Sci. USA 95, 13812–13817.

266

MELIEF et al.

Benitez, R., Godelaine, D., Lopez-Nevot, M. A., Brasseur, F., Jimenez, P., Marchand, M., Oliva, M. R., Van Baren, N., Cabrera, T., Andry, G., Landry, C., Cabello, F. R., Boon, T., and Garrido, F. (1998). Mutations of the 웁2-microglobulin gene result in a lack of HLA class I molecules on melanoma cells of two patients immunized with MAGE peptides. Tissue Antigens 52, 520–529. Bennett, S. R. M., Carbone, F. R., Karamalis, F., Miller, J. F. A. P., and Heath, W. R. (1997). Induction of a CD8⫹ CTL response by crosspriming requires cognate CD4⫹ T cell help. J. Exp. Med. 186, 65–70. Bennett, S. R. M., Carbone, F. R., Karamalis, F., Flavell, R. A., Miller, J. F. A. P., and Heath, W. R. (1998). Help for cytotoxic T lymphocyte responses is mediated by CD40 signalling. Nature 393, 478–480. Bevan, M. J. (1995). Antigen presentation to cytotoxic T lymphocytes in vivo. J. Exp. Med. 182, 639–641. Boise, L. H., Minn, A. J., Noel, P. J., June, C. H., Accavitti, M. A., Lindsten, T., and Thompson, C. B. (1995). CD28 costimulation can promote T cell survival by enhancing the expressing of BcL-xl. Immunity 3, 87–98. Boniface, J. J., Rabinowitz, J. D., Wu¨lfing, C., Hampl, J., Reich, Z., Altman, J. D., Kantor, R. M., Beeson, C., McConnell, H. M., and Davis, M. M. (1998). Initiation of signal transduction through the TCR requires the peptide multivalent engagement of MHC ligands. Immunity 9, 459–466. Boon, T., Cerottini, J. C., Van den Eynde, B., Van der Bruggen, P., and Van Pel, A. (1994). Tumor antigens recognized by T lymphocytes. Annu. Rev. Immunol. 12, 337–365. Brossart, P., Stuhler, G., Flad, T., Stevanovic, S., Rammensee, H. G., Kanz, L., and Brugger, W. (1998). Her-2/neu-derived peptides are tumor-associated antigens expressed by human renal cell and colon carcinoma lines and are recognized by in vitro induced specific cytotoxic T lymphocytes. Cancer Res. 58, 732–736 Brossart, P., Heinrich, K. S., Stuhler, G., Behnke, L., Reichardt, V. L., Stevanovic, S., Muhm, A., Rammensee, H. G., Kanz, L., and Brugger, W. (1999). Identification of HLA–A2-restricted T cell epitopes derived from the MUC1 tumor antigen for broadly applicable vaccine therapies. Blood 93, 4309–4317. Cabrera, T., Collado, A., Fernandez, M. A., Ferron, A., Sancho, J., RuizCabello, F., and Garrido, F. (1998). High frequency of altered HLA class I phenotypes in invasive colorectal carcinomas. Tissue Antigens 52, 114–123. Campbell, M. J., Carroll, W., Kon, S., Thielemans, K., Rothbard, J. B., Levy, S., and Levy, R. (1987). Idiotype vaccination against murine B cell lymphoma. Humoral and cellular responses elicited by tumor-derived immunoglobulin M and its molecular subunits. J. Immunol. 139, 2825–2833. Campbell, M. J., Esserman, L., Byars, N. E., Allison, A. C., and Levy, R. (1990). Idiotype vaccination against murine B cell lymphoma. Humoral and cellular requirements for the full expression of antitumor immunity. J. Immunol. 145, 1029–1036. Cassel, D., and Forman, J. (1988). Linked recognition of helper and cytotoxic determinants for the generation of CTL. Ann. NY Acad. Sci. 532, 51–60. Cella, M., Scheidegger, D., Palmer-Lehmann, K., Lane, P., Lanzavecchia, A., and Alber, G. (1996). Ligation of CD40 on dendritic cells triggers production of high levels of IL12 and enhances T cell stimulatory capacity: T-T help via APC activation. J. Exp. Med. 184, 747–752. Cerutti, A., Schaffer, A., Shah, S., Zan, H., Liou, H. S., Goodwin, R. G., and Casali, P. (1998). CD30 is a CD40-inducible molecule that negatively regulates CD40-mediated immunoglobulin class switching in non-antigen-selected human B cells. Immunity 9, 247–256.

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

267

Chambers, C. A., and Allison, J. P. (1997). Costimulation in T cell responses. Curr. Opin. Immunol. 9, 396–404. Chang, M. H., Chen, C. J., Lai, M. S., et al. (1997). Universal hepatitis B vaccination in Taiwan and the incidence of hepatocellular carcinoma in children. N. Engl. J. Med. 336, 1855–1859. Chaux, P., Vantomme, V., Coulie, P., Boon, T., and Van der Bruggen, P. (1998). Estimation of the frequencies of anti-MAGE-3 cytolytic T lymphocyte precursors in blood from individuals without cancer. Int. J. Cancer 77, 538–542. Chaux, P., Vantomme, V., Stroobant, V., Thielemans, K., Corthals, J., Luiten, R., Eggermont, A. M. M., Boon, T., and Van der Bruggen, P. (1999). Identification of MAGE-3 epitopes presented by HLA-DR molecules to CD4⫹ T lymphocytes. J. Exp. Med. 189, 767–777. Chen, H. L., and Carbone, D. P. (1997). p53 as a target for anti-cancer immunotherapy. Mol. Med. Today 3, 160–167. Chen, J. J., Sun, Y., and Nabel, G. J. (1998). Regulation of the proinflammatory effects of Fas ligand (CD95L). Science 282, 1714–1717. ¨ ., Gure, A. O., Tsang, S., Williamson, B., Chen, Y. T., Scanlan, M. J., Sahin, U., Tu¨reci, O Stockert, E., Pfreundschuh, M., and Old, L. J. (1997). A testicular antigen aberrantly expressed in human cancers detected by autologous antibody screening. Proc. Natl. Acad. Sci. USA 94, 1914–1918. Clemente, C., Rao, S., Lupetti, R., Tragni, G., Pisarra, P., Bersani, I., Parminai, G., Mihm, M. C., and Sensi, M. (1998). Immunohistochemical analysis of the T cell receptor-웁 chain variable regions expressed by T-lymphocytes infiltrating primary human melanoma. Lab. Invest. 78, 619–627. Coughlin, C. M., Salhany, K. E., Gee, M. S., La Temple, D. C., Kotenko, S., Ma, S. J., Gri, G., Wysocka, M., Kim, J. E., Liu, L., Liao, F., Farber, J. M., Pestka, S., Trinchieri, G., and Lee, W. M. F. (1998a). Tumor cell responses to IFN웂 affect tumorigenicity and response to IL-12 therapy and antiangiogenesis. Immunity 9, 25–34. Coughlin, C. M., Salhany, K. E., Wysocka, M., Aruga, E., Kurzawa, H., Chang, A. E., Hunter, C. A., Fox, J. C., Trinchieri, G., and Lee, W. M. F. (1998b). IL-12 and IL-18 synergistically induce murine tumor regression which involves inhibition of angiogenesis. J. Clin. Invest. 101, 1441–1452. Darnell, R. B., and DeAngelis, L. M. (1993). Regression of small-cell lung carcinoma in patients with paraneoplastic neuronal antibodies. Lancet 341, 21–22. Darnell, R. B. (1996). Onconeural antigens and the paraneoplastic neurologic disorders: At the intersection of cancer immunity and the brain. Proc. Natl. Acad. Sci. USA 93, 4529– 4536. Davis, M. (1999). Presentation on Keystone Symposium on Tolerance and Autoimmunity, March 26–April 1, Keystone, CO. De Benedette, M. A., Shahinian, A., Mak, T., and Watts, T. H. (1997). Costimulation of CD28-T lymphocytes by 4-1 BB ligand. J. Immunol. 158, 551–559. De Bruijn, M. L. H., Schuurhuis, D. H., Vierboom, M. P. M., Vermeulen, H., De Cock, K. A. J., Ooms, M. E., Ressing, M. E., Toes, R. E. M., Franken, K. L. M. C., Drijfhout, J. W., Ottenhoff, T. H. M., Offringa, R., and Melief, C. J. M. (1998). Immunization with HPV16 oncoprotein-loaded DC as well as protein in adjuvant induces MHC class I restricted protection to HPV16-induced tumor cells. Cancer Res. 58, 724–731. De Gruijl, T. D., Bontkes, H. J., Stukart, M. J., Walboomers, J. M. M., Remmink, A. J., Verheijen, R. H. M., Helmerhorst, T. J. M., Meijer, C. J. L. M., and Scheper, R. J. (1996). T cell proliferative responses against HPV16 E7 oncoprotein are most prominent in cervical intraepithelial neoplasia patients with a persistent viral infection. J. Gen. Virol. 77, 2183–2191.

268

MELIEF et al.

De Gruijl, T. D., Bontkes, H. J., Walboomers, J. M. M., Stukart, M. J., Doekhie, F. S., Remmink, A. J., Helmerhorst, T. J. M., Verheijen, R. H. M., Duggan-Keen, M. F., Stern, P. L., Meijer, C. J. L. M., and Scheper, R. J. (1998). Differential T helper responses to HPV16 E7 related to viral clearance or persistence in patients with cervical neoplasia. Cancer Res. 58, 1700–1706. DeLeo, A. B., Jay, G., Appella, E., DuBois, G. C., Law, L. W., and Old, L. J. (1979). Detection of a transformation-related antigen in chemically induced sarcomas and other transformed cells of the mouse. Proc. Natl. Acad. Sci. USA 76, 2420–2423. DeLeo, A. B. (1998). P53-based immunotherapy of cancer. Crit. Rev. Immunol. 18, 29–35. Delon, J., Gre´goire, C., Malissen, B., Darche, S., Lemaitre, F., Kourilsky, P., Abastado, J. P., and Trautmann, A. (1998). CD8 expression allows T cell signaling by monomeric peptide-MHC complexes. Immunity 9, 467–473. Diehl, L., Den Boer, A. Th., Schoenberger, S. P., Van der Voort, E. I. H., Schumacher, T. N. M., Melief, C. J. M., Offringa, R., and Toes, R. E. M. (1999). CD40 activation in vivo overcomes peptide-induced peripheral cytotoxic T-lymphocyte tolerance and augments anti-tumor vaccine efficacy. Nature Med. 5, 774-779. Domenech, N., Henderson, R. A., and Finn, O. J. (1995). Identification of an HLA–A11restricted epitope from the tandem repeat domain of the epithelial tumor antigen mucin. J. Immunol. 155, 4766–4774. Drakesmith, H., O’Neil D., Schneider, S. C., Binks, M., Medd, P., Sercarz, E., Beverley, P., and Chain, B. (1999). In vivo priming of T cells against cryptic determinants by dendritic cells exposed to interleukin 6 and native antigen. Proc. Natl. Acad. Sci. USA 95, 14903–14908 Dranoff, G., Jaffee, E., Lazenby, A., Golumbek, P., Levitsky, H., Brose, K., Jackson, V., Hamada, H., Pardoll, D., and Mulligan, R. C. (1993). Vaccination with irradiated tumor cells engineered to secrete murine GM-CSF stimulates potent, specific and long-lasting anti-tumor immunity. Proc. Natl. Acad. Sci. USA 90, 3539–3543. Dunbar, P. R., Ogg, G. S., Chen, J., Rust, N., Van der Bruggen, P., and Cerundolo, V. (1998). Direct isolation, phenotyping and cloning of low frequency antigen-specific CTL from peripheral blood. Current Biol. 8, 413–416. Eshhar, Z. (1997). Tumor-specific T-bodies: Towards clinical application. Cancer Immunol. Immunother. 45, 131–136. Evans, E. M., Man, S., Evans, A. S., and Borysiewicz, L. K. (1997). Infiltration of cervical cancer tissue with HPV-specific CTL. Cancer Res. 57, 2943–2950. Feltkamp, M. C. W., Smits, H. L., Vierboom, M. P. M., Minnaar, R. P., de Jongh, B. M., Drijfhout, J. W., Ter Schegget, J., Melief, C. J. M., and Kast, W. M. (1993). Vaccination with CTL epitope containing peptide protects against a tumor induced by HPV16-transformed cells. Eur. J. Immunol. 23, 2242–2249. Flad, T., Spengler, B., Kalbacher, H., Brossart, P., Baier, D., Kaufmann, R., Bold, P., Metzger, S., Blu¨ggel, Meyer, H. E., Kurz, B., and Mu¨ller, C. A. (1998). Direct identification of MHC class I-bound tumor-associated peptide antigens of a renal carcinoma cell line by a novel mass spectrometric method. Cancer Res. 58, 5803–5811. Funakoshi, S., Longo, D. L., Beckwith, M., Conley, D. K., Tsarfaty, G., Tsarfaty, I. et al. (1994). Inhibition of human B-cell lymphoma growth by CD40 stimulation. Blood 83, 2787–2794. Gallahger, W. M., and Brown, R. (1999). P53-oriented cancer therapies: Current progress. Ann. Oncol. 10, 139–150. Gallimore, A., Fumrese, T., Hengartner, H., Zinkernagel, R. M., and Rammensee, H. G. (1998). Protective immunity does not correlate with the hierarchy of virus-specific CTL responses to naturally processed peptides. J. Exp. Med. 187, 1647–1657.

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

269

Gao, S. J., Kingsley, L., Li, M., Zheng, W., Parravicini, C., Ziegler, J., Newton, R., Rinaldo, C. R., Saah, A., Phair, J., Detels, R., Chang, Y., and Moore, P. S. (1996). KSHV antibodies among Americans, Italians and Ugenands with and without Kaposi’s sarcoma. Nature Med. 2, 925–929. Garrido, F., Cabrera, T., Lopez-Nevot, M. A., and Ruiz-Cabello, F. (1995). HLA class I antigens in human tumors. Adv. Cancer Res. 67, 155–195. Garrido, F., Ruiz-Cabello, F., Cabrera, T., Perez-Villar, J. J., Lopez-Botet, M., DugganKeen, M., and Stern, P. L. (1997). Implication for immunosurveillance of altered HLA class I phenotypes in human tumours. Immunol. Today 18, 89–95. Geier, E., Pfeifer, G., Wilm, M., Lucchiari-Hartz, M., Baumeister, W., Eichmann, K., and Niedermann, G. (1999). A giant protease with potential to substitute for some functions of the proteasome. Science 283, 978–981. Gil-Torregrosa, B. C., Castano, A. R., and Del Val, M. (1998). Major histocompatibility complex class I viral antigen processing in the secretory pathway defined by the transGolgi network protease furin. J. Exp. Med. 188, 1105–1106. Gold, P. (1967). Circulating antibodies against carcinoembryonic antigens of the human digestive system. Cancer 20, 1663–1667. Gong, J., Chen, D., Kashiwaba, M., Li, Y., Chen, L., Takeuchi, H., Qu, H., Rowse, G. J., Gendler, S. J., and Kufe, D. (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. Goulmy, E. (1997). Human minor histocompatibility antigens: New concepts for marrow transplantation and adoptive immuno-therapy. Immunol. Rev. 157, 125–140. Gravestein, L. A., and Borst, J. (1998). Tumor necrosis factor receptor family members in the immune system. Sem. Immunol. 10, 423–434. Greenberg, P. D. (1991). Adoptive T cell therapy of tumors: Mechanisms operative in the recognition and elemenation of tumor cells. Adv. Immunol. 49, 281–355. Griffith, T. S., Brunner, T., Fletcher, S. M., Green, D. R., and Ferguson, T. A. (1995). Fas ligand-induced apoptosis as a mechanism of immune privilege. Science 270, 1189–1192 Griffith, T. S., Chin, W. A., Jackson, G. C., Lynch, D. H., and Kubin, M. Z. (1998). Intracellular regulation of TRAIL-induced apoptosis in human melanoma cells. J. Immunol. 161, 2833–2840. Gurunathan, S., Irvine, K. R., Wu, C. Y., Cohen, J., Thomas, E., Prussin, C., Restifo, N. P., and Seder, R. A. (1998). CD40 ligand/trimer DNA enhances both humoral and cellular immune responses and induces protective immunity to infections and tumor challenge. J. Immunol. 161, 4563–4571. Hamad, A. R. A., O’Herrin, S. M., Lebowitz, M. S., Srikrisknau, A., Bieler, J., Schneck, J., and Pardoll, D. (1998). Potent T cell activation with dimeric peptide–MHC class II ligand: The role of the CD4 coreceptor. J. Exp. Med. 188, 1633–1640. Hammel, P., Boissier, B., Chaumette, M. T., Piedbois, P., Rotman, N., Koujoumdjian, Lubin, R., Delchier, J. C., and Soussi, T. (1997). Detection and monitoring of serum p53 antibodies in patients with colorectal cancer. Gut 40, 356–361. Hanabuchi, S., Koyaniagi, M., Kawasaki, A., Shinohara, N., Matsuzawa, A., Nishimura, Y., Kobayaski, Y., Yonehara, S., Yagita, H., and Okumura, K. (1994). Fas and its ligand in a general mechanism of T cell mediafted cytotoxicity. Proc. Natl. Acad. Sci. USA 91, 4930– 4934. Heath, W. R., Kurts, C., Miller, J. F. A. P., and Carbone, F. (1998). Cross-tolerance: A pathway for inducing tolerance to peripheral tissue antigens. J. Exp. Med. 187, 1549–1553. Hellstro¨m, I., Hellstro¨m, K., and Chen, L. (1995). Can co-stimulated tumor immunity be therapeutically efficacious? Immunol. Rev. 145, 123–145.

270

MELIEF et al.

Henderson, R. A., Cox, A. L., Sakaguchi, K., Appella, E., Shabanowitz, J., Hunt, D. F., and Engelhard, V. H. (1993). Direct identification of an endogenous peptide recognized by multiple HLA–A2. 1-specific CTL. Proc. Natl. Acad. Sci. USA 90, 10275–10279. Herr, W., Linn, B., Leister, N., Wandel, E., Meyer zum Bu¨schenfelde, K. H., and Wo¨lfel, T. (1997). The use of computer-assisted video images for the quantitation of CD8⫹ T lymphocytes producing TNF움 spots in response to peptide antigens. J. Immunol. Meth. 203, 141–152. Heslop, H. E., Ng C. Y., Li, C., Smith, C. A., Loftin, S. K., Krance, R. A., Brenner, M. K., and Rooney, C. M. (1996). Long-term restoration of immunity against Epstein–Barr virus infection by adoptive transfer of gene-modified virus-specific T lymphocytes. Nature Med. 2, 551–555. Hess, S., Gottfried, E., Smola, H., Grunwald, U., Schuchmann, M., and Engelmann, H. (1998). CD40 induces resistance to TNF-mediated apoptosis in a fibro-blast cell line. Eur. J. Immunol. 28, 3594–3604. Hiesse, C., Rieu, P., Kriaa, F., Larue, J. R., Goupy, C., Neyrat, N., and Charpentier, B. (1997). Malignancy after renal transplantation analysis of incidence and risk factors in 1700 patients followed during a 25-year period. Transplant Proc. 71, 517–520. Hiltbold, E. M., Ciborowski, P., and Finn, O. J. (1998). Naturally processed class II epitope from the tumor antigen MUC1 primes human CD4⫹ T cells. Cancer Res. 58, 5066–5070. Hirano, A., Longo, D. L., Ferris, D. K., Taub, D. D., Young, L. S., Eliopoulos, A. G., Cullen, N., Macartney, J., Fanslow, W. C., and Murphy, W. J. (1999). Inhibition of human breast carcinoma growth by a soluble recombinant human CD40 ligand. Blood 93, 2999–3007. Ho, G. Y. F., Bierman, R., Beardsley, L., Chang, C. J., and Burk, R. D. (1998). Natural history of cervicovaginal papillomavirus infection in young women. N. Eng. J. Med. 338, 423–428. Hodge, J. W., McLaughlin, J. P., Kantor, J. A., and Schlom, J. (1997). Diversified prime and boost protocols using recombinant vaccinia virus and recombinant nonreplicating avian pox virus to enhance T cell immunity and antitumor responses. Vaccine 15, 759–768 Hollstein, M., Sidransky, D., Vogelstein, B., and Harris, C. (1991). P53 mutations in human cancer. Science 253, 49–53. Hooijberg, E., Sein, J. J., van den Berk, P. C., Hart, A. A., Van der Valk, M. A., Kast, W. M., Melief, C. J. M., and Hekman, A. (1995). Eradication of large human B cell tumors in nude mice with unconjugated CD20 monoclonal antibodies and interleukin2. Cancer Res. 55, 2627–2634. Horie, R., and Watanabe, T. (1998). CD30: Expression and function in health and disease. Semin. Immunol. 10, 457–470. Houbiers, J. G. A., Nijman, H. W., Van der Burg, S. H., Drijfhout, J. W., Kenemans, P., van de Velde, C. J. H., Brand, A., Momburg, P., Kast, W. M., and Melief, C. J. M. (1993). In vitro induction of human cytotoxic T Lymphocyte responses against peptides of mutant and wild-type p53. Eur. J. Immunol. 23, 2072–2077. Houbiers, J. G., van der Burg, S. H., van de Watering, L. M., Tollenaar, R. A., Brand, A., van de Velde, C. J., and Melief, C. J. (1995). Antibodies against p53 are associated with poor prognosis of colorectal cancer. Br. J. Cancer 72, 637–641. Hsu, F. J., Caspar, C. B., Czerwinski, D., Kwak, L. W., Liles, T. M., Syrengelas, A., TaidiLaskowski, B., and Levy, R. (1997). Tumor-specific idiotype vaccines in the treatment of patients with B-cell lymphoma—long-term results of a clinical trial. Blood 89, 3129–3135. Hsu, F. J., Benike, C., Fagnoni, F., Liles, T. M., Czerwinski, D., Taidi, B., Engleman, E. G., and Levy, R. (1996). Vaccination of patients with B-cell lymphoma using autologous antigen-pulsed dendritic cells. Nat. Med. 2, 52–58.

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

271

Huang, A. Y. C., Golumbek, P., Ahmad-Zadeh, M., Jaffee, E., Pardoll, D., and Levitsky, H. (1994). Role of bone marrow-derived cells in presenting MHC class I-restricted tumor antigens. Science 264, 961–965. Hung, K., Hayashi, R., Lafond-Walker, A., Lowenstein, C., Pardoll, D., and Levitsky, H. (1998). The central role of CD4⫹ T cells in anti-tumor immune response. J. Exp. Med. 188, 2357–2368. Hurtado, J. C., Kim, Y. J., and Kwon, B. S. (1997). Signals through 4-1 BB are costimulatory to previously activated splenic T cells and inhibit activation induced cell death. J. Immunol. 158, 2600–2609. Hurwitz, A. A., Yu, T. F. Y., Leach, D. R., and Allison, J. P. (1998). CTLA-4 blockade synergizes with tumor-derived granulocyte–macrophage colony-stimulating factor for treatment of an experimental mammary carcinoma. Proc. Natl. Acad. Sci. USA 95, 10067– 10071. Ikeda, H., Lethe´, B., Lehmann, F., Van Baren, N., Baurain, J.-F., De Smet, C., Chambost, H., Vitale, M., Moretta, A., Boon, T., and Coulie, P. G. (1997). Cytolytic T cells expressing NK inhibitory receptor P58.2 recognize on a melanoma metastasis lacking HLA-Cw7, an antigen encoded by a gene which is preferentially expressed in tumors. Immunity 6, 199–208. Irmler, M., Thome, M., Habne, M., Schneider, P., Hofmann, K., Steiner, V., Bodmer, J. L., Schro¨ter, M., Burns, K., Mattmann, C., Rimoldi, D., French, L. E., and Tschopp, J. (1997). Inhibition of death receptor signals by cellular FLIP. Nature 388, 190–195. Irvine, K. R., Chamberlain, R. S., Shulman, E. P., Surman, D. R., Rosenberg, S. A., and Restifo, N. P. (1997). Enhancing efficacy of recombinant anticancer vaccines with prime/ boost regimens that use two different vectors. J. Natl. Cancer Inst. 89, 1595–1601. Ja¨ger, E., Ringhoffer, M., Dienes, H. P., Brand, M., Karbach, J., Ja¨ger, D., Ilsemann, C., Hagedorn, M., Oesch, F., and Knuth, A. (1996). GM–CSF enhances immune responses to melanoma-associated peptides in vivo. Int. J. Cancer 67, 54–62. Ja¨ger, E., Ringhoffer, M., Altmannsberger, M., Brand, M., Karback, J., Ja¨ger, D., Oesch, F., and Knuth, A. (1997). Immunoselection in vivo: Independent loss of MHC class I and melanocyte differentiation antigen expression in metastative melanoma. Int. J. Cancer 71, 142–147. Ja¨ger, E., Chen, Y. T., Drijfhout, J. W., Karbach, J., Ringhoffer, M., Ja¨ger, D., Brand, M., Wada, H., Noguchi, Y., Stockert, E., Old, L. J., and Knuth, A. (1998). Simultaneous humoral and cellular immune response against cancer-testis antigen NY-ESO-1: Definition of human histocompatibility leukocyte antigen (HLA)–A2-binding peptide epitopes. J. Exp. Med. 187, 265–270. Johnson, A., France, J., Sy, M. S., and Harding, C. V. (1998). Down-regulation of the transporter for antigen presentation, proteasome subunits and MHC class I in tumor cell lines. Cancer Res 58, 3660–3667. Jondal, M., Schirmbeck, R., and Reimann, J. (1996). MHC class I-restricted CTL responses to exogenous antigens. Immunity 5, 295–302. Ka¨gi, D., Vignaux, F., Ledermann, B., Ba¨rki, K., Depraetere, V., Nagata, S., Hengartner, H., and Golstein, P. (1994). FAS and perforin pathways as major mechanisms of T cell mediated cytotoxicity. Science 265, 528–530. Kaplan, D. H., Shankaran, V., Dighe, A. S., Stockert, E., Aguet, M., Old, L. J., and Schreiber, R. D. (1998). Demonstration of an interferon-웂 dependent tumor surveillance systems in immunocompetent mice. Proc. Natl. Acad. Sci. USA 95, 7556–7561. Kass, E., Schlom, J., Thompson, J., Guadagni, F., Graziano, P., and Greiner J. W. (1999). Induction of protective host immunity to carcinoembryonic antigen (CEA), a self-antigen

272

MELIEF et al.

in CEA transgenic mice, by immunizing with a recombinant vaccinia–CEA virus. Cancer Res. 59, 676–683. Kast, W. M., Offringa, R., Peters, P. J., Voordouw, A. C., Meloen, R. H., Van der Eb, A. J., and Melief, C. J. M. (1989). Eradication of adenovirus E1-induced tumors by E1Aspecific cytotoxic T lymphocytes. Cell 59, 603–614. Kataoka, T., Schr&ouacte;ter, M., Hahne, M., Schneider, P., Irmler, M., Thome, M., Froelich, C. J., and Tschopp, J. (1998). FLIP prevents apoptosis induced by death receptors but not by perforin/granzyme B, chemotherapeutic drugs and gamma irradication. J. Immunol. 161, 3936–3942. Kedes, D. H., Operskalski, E., Busch, M., Kohn, R., Flood, J., and Ganem, D. (1996). The seroepidemiology of human herpesvirus 8 (Kaposi’s sarcoma-associated herpesvirus: Distribution of infection in KS risk groups and evidence for sexual transmission. Nature Med. 2, 918–921. Kiener, P. A., Moran-Davis, P., Rankin, B. M., Wahl, A. F., Ajuffo, A., and Hollenbaugh, P. (1995). Stimulation of CD40 with purified soluble gp39 induces proinflammatory responses in human monocytes. J. Immunol. 155, 4917–4925. Kim, C. J., Prevette, T., Cormier, J., Overwijk, W., Roden M., Restifo, N. P., Rosenberg, S. A., and Marincola, F. M. (1997). Dendritic cells infected with poxviruses encoding MART-1/Melan A sensitize T lymphocytes in vitro J. Immunother. 20, 276–286 Kim, Y. J., Kim, S. H., Mantel, P., and Kwon, B. S. (1998). Human 4-1 BB regulates CD28 co-stimulation to promote TH1 responses. Eur. J. Immunol. 28, 881–890. King, C. A., Wills, M. R., Hamblin, T. J., and Stevenson, F. K. (1993). Idiotypic IgM on a B-cell surface requires processing for recognition by anti-idiotypic T cells. Cell. Immunol 147, 411–424 Klas, C., Debatin, K. M., Jonker, R. R., and Krammer, P. H. (1993). Activation interferes with the APO-1 pathway in mature human T cells. Int. Immunol. 5, 625–630 Koch, F., Stanzl, U., Jennewein, P., Janke, K., Heufler, C., Kampgen, E., Romani, N., and Schuler, G. (1996). High level IL-12 production by murine dendritic cells: Upregulation via MHC class II and CD40 molecules and downregulation by IL-4 and IL-10. J. Exp. Med. 184, 741–746. Kolb, H. J., and Holler, E. (1997). Adoptive immunotherapy with donor lymphocyte transfusions. Curr. Opin. Oncol. 9, 139–145. Kotera, Y., Fontenot, D. G., Pecher, G., Metzgar, R. S., and Finn, O. (1994). Humoral immunity against a tandem repeat epitope of human mucin MUC-1 in sera from breast, pancreatic and colon cancer patients. Cancer Res. 54, 2856–2860. Krammer, P. (1999). CD95 (APO-1/Fas)-mediated apoptosis: Live and let die. Adv. Immunol. 71, 163–210. Krause, A., Guo, H. F., Latouche, J. B., Tan, C., Cheung, N. K., and Sadelain, M. (1998). Antigen-dependent CD28 signalling selectively enhances survival and proliferation in genetically modified activated human primary T lymphocytes. J. Exp. Med. 188, 619–626. Kronin, V., Winkel, K., Suss, G., Kelso, A., Heath, W., Kirberg, J., von Boehmer, H., and Shortman, K. (1996). A subclass of dendritic cells regulates the response of naive CD8 T cells by limiting their IL-2 production. J. Immunol. 157, 3819–3827. Kurts, C., Carbone, F. R., Barnden, M., Blanas, E., Allison, J., Heath, W. R., and Miller, J. F. A. P. (1997). CD4⫹ T cell help impairs CD8⫹ presentation of self-antigens and favors autoimmunity. J. Exp. Med. 186, 2057–2062. Kurts, C., Heath, W. R., Kosaka, H., Miller, J. F. A. P., and Carbone, F. R. (1998). The peripheral deletion of autoreactive CD8⫹ T cells induced by cross-presentation of selfantigens involves signaling through CD95 (fas, Apo-1). J. Exp. Med. 188, 415–420.

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

273

Kurts, C., Carbone, F. R., Krummel, M. F., Koch, K. M., Miller, J. F. A. P., and Heath, W. R. (1999). Signaling through CD30 protects against autoimmune diabetes mediated by CD8 T cells. Nature 398, 341–344. Lane, D. P. (1994). p53 and human cancer. British. Med. Bull. 50, 582–599. Lens, S. M. A., Tesselaar. K., van Oers, M. H. J., and Van Lier, R. A. W. (1998). Control of lymphocyte function through CD27–CD70 interactions. Semin. Immunol. 10, 491–499. Lethe´, B., Lucas, S., Michaux, L., De Smet, C., Godelaine, D., Serrano, A., De Plaen, E., and Boon, T. (1998). LAGE-1, a new gene with tumor specificity. Int. J. Cancer 76, 903–908. Lezzi, G., Karjalainen, K., and Lanzavecchia, A. (1998). The duration of antigenic stimulation determines the fate of naive and effector T cells. Immunity 8, 89–95. Lindner, R., and Unanue, E. R. (1996). Distinct MHC class II complexes generated by separate processing pathways. EMBO J. 15, 6910–6920. Loenen, W. A. M. (1998). CD27 and TNFR relatives in the immune system: Their role in health and disease. Semin. Immunol. 10, 407–422. Lowin, B., Hahne, M., Mattman, C., and Tschopp, J. (1994). Cytotoxic T cell cytotoxicity is mediated through perforin and FAS lytic pathways. Nature 370, 650–653. Lubin, R., Schlichtholz, B., Teilland, J. L., Garay, E., Bussel, A., Wild, C. P., and Soussi, T. (1995a). P53 antibodies in patients with various types of cancer: Assay, identification and characterization. Clin. Cancer Res. 1, 1463–1469. Lubin, R., Zalcman, G., Bouchet, L., Tre´damiel, J., Legros, Y., Cazals, D., Hirsch, A., and Soussi, T. (1995b). Serum p53 antibodies as early markers of lung cancer. Nature Med. 1, 701–702. Luxton, J. C., Rowe, A. J., Cridland, J. C., Coletart, T., and Sheperd, P. S. (1996). Proliferative T cell response to the HPV16 E7 protein in women with cervical dysplasia and cervical carcinoma and in healthy individuals. J. Gen. Virol. 77, 1585–1593. Lynch, R. G., Graff, R. J., Sirisinha, S., Simms, E. S., and Eisen, H. N. (1972). Myeloma proteins as tumor-specific transplantation antigens. Proc. Natl. Acad. Sci. USA 69, 1540– 1544. Mackey, M. F., Gunu, J. R., Ting, P. P., Kukutani, H., Dranoff, G., Noelle, R. J., and Barth, R. J. (1997). Protective immunity induced by tumor vaccines requires interactions between CD40 and its ligand, CD154. Cancer Res. 57, 2569–2574. Mackey, M. F., Gunn, J. R., Maliszewski, C., Kikutani, H., Noelle, R. J., and Barth, R. J. (1998). Dendritic cells require maturation via CD40 to generate protective anti-tumor immunity. J. Immunol. 161, 2094–2098. Magarian-Blander, J., Ciborowski, P., Hsia, S., Watkins, S. C., and Finn, O. (1998). Intercellular and intracellular events following the MHC-unrestricted TCR recognition of a tumorspecific peptide epitope on the epithelial antigen MUC1. J. Immunol. 160, 3111–3120. Maloney, D. G., Grillo-Lopez, A. J., White, C. A., Bodkin, D., Schuller, R. J., Neidhart, J. A., Janakiraman, N., Foon, K. A., Liles, T. M., Dallaire, B. K., et al. (1997). IDECC2B8 (Rituximab) anti-CD20 monoclonal antibody therapy in patients with relapsed lowgrade non Hodgkin’s lymphoma. Blood 90, 2188–2195. Manici, S., Sturmiolo, T., Imro, M. A., Hammer, J., Sinigaglia, F., Noppen, C., Spagnoli, G., Mazzi, B., Bellone, M., Dellabona, P., and Protti, M. P. (1999). Melanoma cells present a MAGE-3 epitope to CD4⫹ CTL in association with histocompatibility leukocyte antigen DR II. J. Exp. Med. 189, 871–876. Manoury, B., Hewitt, E. W., Morrice, N., Dando, P. M., Barrett, A. J., and Watts, C. (1998). An asparaginyl endopeptidase processes a microbial antigen for class II MHC presentation. Nature 396, 695–699.

274

MELIEF et al.

Manz, R., Assenmacher, M., Pfluger, E., Miltenyi, S., and Radbruch, A. (1995). Analysis and sorting of live cells according to secreted molecules, relocated to a cell-surface affinity matrix. Proc. Natl. Acad. Sci. USA 92, 1921–1925. Marchand, M., Weynants, P., Rankin, E., Arienti, F., Belli, F., Parmiani, G., Cascinelli, N., Bourlond, A., Vanwijck, R., Humblet, Y., et al. (1995). Tumor regression responses in melanoma patients treated with a peptide encoded by gene MAGE-3. Int. J. Cancer 63, 883–885. Marshall, J. L., Hawkins, M. J., Tsang, K. Y., Richmond, E., Pedicano, J. E., Zhu, M. Z., and Schlom, J. (1999). Phase I study in cancer patients of a replication-defective Avipox recombinant vaccine that expresses human carcinoembryonic antigen. J. Clin. Oncol. 17, 332–337. Matzinger, P. (1994). Tolerance, danger, and the extended family. Annu. Rev. Immunol. 12, 991–1045. Matzinger, P. (1998). An innate sense of danger. Semin. Immunol. 10, 399–415. Mayordomo, J. I., Zorina, T., Storkus, W. J., Zitvogel, L., Celluzi, C. Falo, L. D., Melief, C. J. M., Ildstad, S. T., Kast, W. M., DeLeo, A. B., and Lotze, M. T. (1995). Bone-marrowderived DC pulsed with synthetic tumour peptides elicits protective and therapeutic antitumour immunity. Nature Med. 1, 1297–1301. McArthur, J. G., and Mulligan, R. C. (1998). Induction of protective anti-tumor immunity by gene-modified dendritic cells. J. Immunother. 21, 41–47. McCormick, A. A., Kumagai, M. H., Hanley, K, Turpen T. H., Hakim I, Grill, L. K., Tuse, D., Levy, S., and Levy, R. (1999). Rapid production of specific vaccines for lymphoma by expression of the tumor-derived single-chain Fv epitopes in tobacco plants. Proc. Natl. Acad. Sci. USA 96, 703–708. Medema, J. P., de Jong, J., Van Hall, T., Melief, C. J. M., and Offringa, R. (1999). Immune escape of tumors in vivo by expression of C-FLIP. J. Exp. Med., 190, 1033–1038. Melero, I., Shuford, W. W., Ashe-Newby, S., Aruffo, A., Ledbetter, J. A., Hellstro¨m, K. E., Mittler, R. S., and Chen, L. (1997). Monoclonal antibodies against the 4-1 BB T cell activation molecule eradicate established tumors. Nature Med. 3, 682–685. Melief, C. J. M. (1992). Tumor eradication by adoptive transfer of cytotoxic T lymphocytes. Adv. Cancer Res. 58, 143–175. Mitchell, D. A., Noir, S. K., and Gilboa, E. (1998). Dendritic cell/macrophage precursors capture exogenous antigen for MHC class I presentation by dendritic cells. Eur. J. Immunol. 28, 1923–1933. Morecki, S., Yacovlev, E., Diab, A., and Slavin, S. (1998). Allogeneic cell therapy for a murine mammary carcinoma. Cancer Res. 58, 3891–3895. Motta, I., Lone, Y. C., and Kourilsky, P. (1998). In vitro induction of CTL with complexes of peptide and recombinant MHC class I molecules coated onto beads: Role of TCR/ ligand density. Eur. J. Immunol. 28, 3685–3695. Mu¨nz, C., Obst, R., Osen, W., Stevanovic, S., and Rammensee, H. G. (1999). Alloreactivity as a source of high avidity peptide-specific human CTL. J. Immunol. 162, 25–34. Murali-Krishna K, Altman JD, Suresh M, Sourdive DJ, Zajac AJ, Miller JD, Slansky J, and Ahmed R. (1998). Counting antigen-specific CD8 T cells: A reevaluation of bystander activation during viral infection. Immunity 8, 177–187. Murphy, W. J., Funakoshi, S., Beckwith, M., Rushing, S. E., Conley S. K., Armitage, R. J., et al. (1995). Antibodies to CD40 prevent Epstein-Barr virus-mediated human B cell lymphomagenesis in SCID mice given human PBL. Blood 86, 1946–1953. Mutis, T., Verdijk, R., Schrama, E., Esendam, B., Brand, A., and Goulmy, E. (1999). Feasibility of immunotherapy of relapsed leukemia with ex vivogenerated CTL specific for hematopoietic system-restricted minor histocompatibility antigens. Blood 93, 2336–2341.

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

275

Muy-Rivera, M., Lemon, M., Funatake, C., and Weinberg, A. D. (1998). Enhancement of antitumor immunity by engaging the OX40 receptor (CD134) in vivo. Am. Assoc. Cancer Res. Meeting Abstracts. 3610, 193. Nagata, S. (1997). Apoptosis by death factor. Cell 88, 355–365. Nair, S. K., Boczkowski, D., Morse, M., Cumming, R. I., Lyerly, H. K., and Gilboa, E. (1998). Induction of primary carcinoembryonic antigen (CEA)-specific cytotoxic T lymphocytes in vitro using human dendritic cells transfected with RNA. Nat. Biotechnol, 16, 364–369. Naito, Y., Saito, K., Shiiba, K., Ohuchi, A., Saigenji, K., Nagura, H., and Ohtani, H. (1998). CD8⫹ T cells infiltrated within cancer cells nests as a prognostic factor in human colorectal cancer. Cancer Res. 58, 3491–3494. Nelson, C. A., Vidafsky, I., Viner, N. J., Gross, M. L., and Unanue, E. R. (1997). Aminoterminal trimming of peptides for presentation on major histocompatibility complex class II molecules. Proc. Natl. Acad. Sci. USA 94, 628–633. Nestle´, F. O., Alijagic, S., Gilliet, M., Sun, Y., Grabbe, S., Dummer, R., Burg, G., and Schadendorf, D. (1998). Vaccination of melanoma patients with peptide- or tumor lysatepulsed dendritic cells. Nature Med. 4, 328–332. Newstead, C. G. (1998). Assessment of risk of cancer after renal transplantation. Lancet 351, 610–612. Nussbaum, A. K., Dick, T. P., Keilholz, W., Schirle, M., Stevanovic, S., Dietz, K., Heinemeyer, W., Groll, M., Wolf, D. H., Huber, R., Rammensee, H. G., and Schild, H. (1998). Cleavage motifs of the yeast 20S proteasome 웁 subunits deduced from digest of enolase 1. Proc. Natl. Acad. Sci. USA 95, 12504–12509. Obst, R., Mu¨nz, C., Stevanovic, S., and Rammensee, R. G. (1998). Allo- and self-restricted CTL against a peptide library: Evidence for a functionally diverse allo-restricted T cell repertoire. Eur. J. Immunol. 28, 2432–2443. Offringa, R., Toes, R. E. M., Ressing, M. E., De Bruijn, M. L. H., and Melief, C. J. M. (1999). T cell-mediated immunotherapy of cancer. In ‘‘Textbook of Malignant Haematolgy,’’ pp. 447–461. Martin Dunitz, Cambridge Univ. Press, UK. Ogg, G. S., Dunbar, P. R., Romero, P., Chen, J. L., and Cerundolo, V. (1998). High frequency of skin-homing melanocyte-specific cytotoxic T lymphocytes in Autoimmune Vitiligo. J. Exp. Med. 188, 1203–1208. O’Reilly, R. J., Small, T. N., Papadopoulos, E., Lucas, K., Lacerda, J., and Koulova, L. (1998). Adoptive immunotherapy for Epstein–Barr virus-associated lympho-proliferative disorders complicating marrow allografts. Springer Semin. Immuno-Pathol. 20, 445–491. Ossevoort, M. A., Feltkamp, M. C. W., Van Veen, K. J. H., Melief, C. J. M., and Kast, W. M. (1995). Dendritic cells as carriers for a CTL epitope-based peptide vaccine in protection against an HPV16 induced tumor. J. Immunother. 18, 86–94. Ossendorp, F., Eggers, M., Neisig, A., Ruppert, T., Groettrup, M., Sijts, A., Mengede, E., Kloetzel, P. M., Neefjes, J. J., Koszinowski, U., and Melief, C. J. M. (1996). A single residue exchange within a viral CTL epitope alters proteasome-mediated degradation resulting in lack of antigen presentation. Immunity 5, 115–124. Ossendorp, F., Mengede´, E., Camps, M., Filius, R., and Melief, C. J. M. (1998). Specific T helper cell requirement for optimal induction of CTL against MHC class II-negative tumors. J. Exp. Med. 187, 693–702. Overwijk, W. W., Tsung, A., Irvine, K. R., Parkkurst, M., Goletz, T. J., Tsung, K., Carroll, M. W., Liu, C., Moss, B., Rosenberg, S. A., and Restifo, N. P. (1998). Gp100/pmel 17 is a murine tumor rejection antigen: Induction of ‘‘self ’’ reactive, tumoricidal T cells using high-affinity, altered peptide ligand. J. Exp. Med. 188, 277–286.

276

MELIEF et al.

Overwijk, W. W., Lee, D. S., Surman D. R., Irvine, K. R., Touloukian C. E., Chan C. C., Carroll M. W., Moss B, Rosenberg S. A., and Restifo N. P. (1999). Vaccination with a recombinant vaccinia virus encoding a ‘‘self ’’ antigen induces autoimmune vitiligo and tumor cell destruction in mice: Requirement for CD4⫹ T lymphocytes. Proc. Natl. Acad. Sci. USA 96, 2982–2987. Owen-Schaub, L., van Golen, K. L., Hill, L. L., and Price, J. E. (1998). Fas and Fas ligand interactions suppress melanoma. J. Exp. Med. 188, 1717–1723. Oxenius, A., Zinkernagel, R. M., and Hengartner, H. (1998). Comparison of activation versus induction of unresponsiveness of virus-specific CD4⫹ and CD8⫹ T cells upon acute versus peristent viral infection. Immunity 9, 449–457. Pardoll, D. M. (1998). Cancer vaccines. Nature Med. 4, 525–531. Peter M. E., Kischkel, F. C., Scheuerpflug C. G., Medema, J. P., Debatin K. M., and Krammer P. H. (1997). Resistance of cultured peripheral T cells towards activationinduced cell death involves a lack of recruitment of FLICE (MACH/caspase 8) to the CD95 death-inducing signaling complex. Eur. J. Immunol. 27, 1207–1212. Petersson, M., Charo, J., Salazar-Onfray, F., Noffz, G., Mohaupt, M. Qin, Z., Klein, G., Blankenstein, T., and Kiessling, R. (1998). Constitutive IL-10 production accounts for the high NK sensitivity, low MHC class I expression and poor TAP 1/2 function in the prototype NK target YAC-1. J. Immunol. 161, 2099–2105. Pieper, R., Chrisian, R. E., Gonzales, M. I., Nishimura, M. I., Gupta, G., Settlage, R. E., ShabanowitzJ., Rosenberg, S. A., Hunt, D. F., and Topalian, S. L. (1999). Biochemical indentification of a mutated human melanoma antigen recognized by CD4⫹ T cells. J. Exp. Med. 189, 757–765. Pitty, R. M., Marsters, S. A., Lawrence, D. A., Roy, M., Kischkel, F. C., Dowd, P., Huang, A., Donahue, C. J., Sherwood, S. W., Baldwin, D. T., Godowski, P. J., Wood, W. I., Gurney, A. L., Hillan, K. J., Cohen, R. L., Goddard, A. D., Botstein, D., and Ashkenazi, A. (1998). Genomic amplification of a decoy receptor for FAS ligand in lung and colon cancer. Nature 396, 699–702. Porgador, A., Irvine, K. R., Iwasaki, A., Barber, B. H., Restifo, N. P., and Germain, R. N. (1998). Predominant role for directly transfected dendritic cells in antigen presentation of CD8⫹ T cells after gene gun immunization. J. Exp. Med. 188, 1075–1082. Pulaski, B. A., and Ostrand-Rosenberg, S. (1998). Reduction of established spontaneous mammary carcinoma metastases following immunotherapy with MHC class II and B7.1 cell based tumor vaccines. Cancer Res. 58, 1486–1493. Real, L. M., Junenez, P., Canton, J., Kirkin, A., Garcia, A., Abril, E., Zeuthen, J., RuizCabello, F., and Garrilo, F. (1998). In vivo and in vitro generation of a new altered HLA phenotype in melanoma-tumour-cell variants expressing a single HLA-class-I allele. Int. J. Cancer 75, 317–323. Reichardt, V. L., Okada, C. Y., Liso, A., Benike, C. J., Stockerl-Goldstein, K. E., Engleman, E. G., Blume K. G., and Levy, R. (1999). Idiotype vaccination using dendritic cells after autologous peripheral blood stem cell transplantation for multiple myeloma—A feasibility study. Blood 93, 2411–2419. Reimann, J., and Kaufmann, S. H. E. (1997). Alternative antigen processing pathways in anti-infective immunity. Curr. Opin. Immunol. 9, 462–469. Ressing, M. E., Van Driel, W. J., Celis, E., Sette, A., Brandt, R. M. P., Hartman, M., Anholts, J. D., Schreuder, G. M. T., Ter Harmsel, W. B., Fleuren, G. J., Trimbos, J. B., Kast, W. M., and Melief, C. J. M. (1996). Occasional memory CTL responses of patients with human HPV-16 positive cervical lesions against a human HlA-A*0201 restricted E7 encoded epitope. Cancer Res. 56, 582–588.

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

277

Restifo, N. P., and Rosenberg, S. A. (1999). Recent progress toward the development of recombinant and synthetic vaccines for the treatment of melanoma. Curr. Opin. Oncol. 11, 50–57. Ria, F., Penna, G., and Adorini, L. (1998). Th1 cells induce and Th2 inhibit antigendependent IL-12 secretion by dendritic cells. Eur. J. Immunol. 28, 2003–2016. Ridge, J. P., Di Rosa, F., and 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. Riethmu¨ller, G., Holz, E., Schmiok, G., Schiegel, W., Raab, R., Ho¨ffken, K., Gruber, R., Funke, I., Pichlmaier, H., Hirche, H., et al. (1998). Monoclonal antibody therapy for resected Duke’s C colorectal cancer: Seven year outcome of a multicenter randomized trial. J. Clin. Oncol. 16, 1788–1794. Rock, K. L. (1996). A new foreign policy: MHC class I molecules monitor the outside world. Immunol. Today 17, 131–137. Romero, P., Cerottini, J. C., and Waanders, G. A. (1998a). Novel methods to monitor antigen-specific CTL responses in cancer immunotherapy. Mol. Med. Today 4, 305–312. Romero, P., Dunbar, P. R., Valenori, D. Pittet, M., Ogg, G. S., Rimoldi, D., Chen, J. L., Lienard, D., Cerottini, J. C., and Cerundolo, V. (1998b). Ex vivo staining of metastatic lymph nodes by MHC class I tetramers reveals high numbers of antigen-experienced tumor-specific CTL. J. Exp. Med. 188, 1641–1650. Rongcun, Y., Salazar-Onfray, F., Charo, J., Malmberg, K. J., Evrin, K., Maes, H., Kono, K., Hising C., Petersson M., Larsson, O., Lan, L., Appella E., Sette A., Celis E., and Kiessling R. (1999). Identification of new HER2/neu-derived peptide epitopes that can elicit specific CTL against autologous and allogeneic carcinomas and melanomas. J. Immunol. 163, 1037–1044. Rooney, C. M., Smith, C. A., Ng, C. Y. C., Loftin, S. K., Sixberg J. W., Gan, Y., Srivastava, D. K., Bowman, L. C., Krance, R. A., Brenner, M. K., and Heslop, H. E. (1998). Infusion of CTL for the prevention and treatment of EBV-induced lymphoma in allogeneic transplant recipients. Blood 92, 1549–1555. Ro¨pke, M., Hald, J., Guldenberg, P., Zeuthen, J., Norgaard, L., Fugger, L., Svejgaard, A., Van der Burg. S. H., Nijman, H. W., Melief, C. J. M., and Claesson, M. H. (1996). Spontaneous human squamous cell carcinomas are killed by a human cytotoxic T lymphocyte clone recognizing a wild-type p53-derived peptide. Proc. Natl. Acad. Sci. USA 93, 14704–14707. Rosenberg, S. A., Yanelli, J. R., Yang, J. C., Topalian, S. L., Schwarzentruber, D. J., Weber, J. S., et al. (1994). Treatment of patients with metastatic melanoma with autologous tumor-infiltrating lymphocytes and IL-2. J. Natl. Cancer Inst. 86, 1159–1166. Rosenberg, S. A., Yang, J. C., Schwartzentruber, D. J., Hwu, P., Marincola, F., Topalian, S. L., Restifo, N. P., Dudley, M. E., Schwartz, S. L., Spiess, P. J. et al. (1998). Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma. Nature Med. 4, 321–327. Rosenberg, S. A. (1999). A new era for cancer immunotherapy based on the genes that encode cancer antigens. Immunity 10, 281–287. Roskrow, M. A., Dilloo, D., Suzuki, N., Zhong, W., Rooney, C. M., and Brenner, M. K. (1999). Autoimmune disease induced by dendritic cell immunization against leukemia. Leuk. Res. 23, 549–557. Ruiz-Cabello, F., and Garrido, F. (1998). HLA and cancer: From research to clinical impact. Immunol. Today 19, 539–542. Sadovnikova, E., and Stauss, H. J. (1996). Peptide-specific cytotoxic T lymphocytes restricted by nonself major histocompatibility complex class I molecules: Reagents for tumor immunotherapy. Proc. Natl. Acad. Sci. USA 93, 13114–13118.

278

MELIEF et al.

Sadovnikova, E., Jopling, L. A., Soo, K. S., and Stauss, H. J. (1998). Generation of human tumor-reactive cytotoxic T cells against peptides presented by non-self HLA class I molecules. Eur. J. Immunol. 28, 193–200. ¨ ., Schmitt, H., Cochlovius, B., Johannes, T., Schmitts, R., Stenner, F., Sahin, U., Tu¨reci, O Luo, G., Schobert, I., and Pfreundschuh, M. (1995). Human neoplasms elicit multiple specific immune responses in the autologous host. Proc. Natl. Acad. Sci. USA 92, 11810– 11813. Sallusto, F., and Lanzavecchia, A. (1999). Mobilizing dendritic cells for tolerance, priming and chronic inflammation. J. Exp. Med. 189, 611–614. Samanci, A., Yi, Q., Fagerberg, J., Strigard, K., Smith, G., Rude´n, U., Wahren, B., and Mellstedt, H. (1998). Pharmacological administration of granulocyte/macrophage-colonystimulating factor is of significant importance for the induction of a strong humoral and cellular response in patients immunized with recombinant careinoembryonic antigen. Cancer Immunol. Immunother. 47, 131–142. Saoulli, K., Lee, S. L., Cannons, J. L., Yeh, W. C., Santana, A., Goldstein, M. D., Bangia, N., De Benedette, M. A., Mak, T. W., Choi, Y., and Watts, T. H. (1998). CD28-independent, TRAF2-dependent costimulation of resting T cells by 4-1 BB ligand. J. Exp. Med. 187, 1849–1862. Scaffidi, C., Schmitz I., Krammer P. H., and Peter, M. E. (1999). The role of c-FLIP in modulation of CD95-induced apoptosis. J. Biol. Chem. 274, 1541–1548. Scaffidi, C., Fulda, S., Srinivasan, A., Friesen, C., Li, F., Tomaselli, K. J., Debatin, K. M., Krammer, P. H., and Peter, M. E. (1998). Two CD95 (APO-1/FAS) signalling pathways. EMBO J. 17, 1675–1687. Schirle, M., Keilholz, W., Gouttefangeas, C., Weber, B., Dumrese, T., Becker, H. D., Stevanovic, S., and Rammensee, H. G. (2000). Identification of tumor-associated MHC class I ligands by a novel T cell independent approach. Submitted for publication. Schirmbeck, R., Wild, J., and Reimann, J. (1998). Similar as well as distinct MHC class Ibinding peptides are generated by exogenous and endogenous processing of hepatitis B virus surface antigen. Eur. J. Immunol. 28, 4149–4161. Schoenberger, S. P., and Sercarz, E. E. (1996). Harnessing self-reactivity in cancer immunotherapy. Semin. Immunol. 8, 303–309. Schoenberger, S. P., Jonges, L. E., Mooijaart, R., Hartgers, F., Toes, R. E. M., Kast, W. M., Melief, C. J. M., and Offringa, R. (1998a). Efficient direct priming of tumor-specific CTL in vivo by an engineered APC. Cancer Res. 58, 3094–3100. Schoenberger, S. P., Toes, R. E., van der Voort, E. I., Offringa, R., and Melief, C. J. M. (1998b). T cell help for cytotoxic T lymphocyte responses is mediated by CD40–CD40 L interactions. Nature 393, 480–483. Schreurs, M. W. J., de Boer, A., Figdor, C. G., and Adema, G. J. (1998). Genetic vaccination against the melanocyte lineage-specific antigen gp100 induces CTL-mediated tumor protection. Cancer Res. 58, 2509–2514. Schurmans, L. R. H. M., Den Boer, A., Diehl, L., V.d. Voort, E. I. H., Kast, W. M., Melief, C. J. M., Toes, R. E. M., and Jager M. J. (1999). Successful immunotherapy of an intraocular tumor in mice. Cancer Res., 59, 5250–5254. Seung, S., Urban, J. L., and Schreiber, H. (1993). A tumor escape variant that has lost one MHC class I restriction element induces specific CD8⫹ T cells to an antigen that no longer serves as a target. J. Exp. Med. 178, 933–940. Shapiro, A., Imreh, M., Leonchiks, A., Imreh, S., and Masucci, M. G. (1998). A minimal glycine–alanine repeat prevents the interaction of ubiquitinated I␬웁움 with the proteasome: A new mechanism for selective inhibition of proteolysis. Nature Med. 4, 939–944.

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

279

Sherman, L. A., Theobald, M., Morgan, D., Hernandez, J., Bacik, I., Yewdell, J., Bennink, J., and Biggs, J. (1998). Strategies for tumor elimination by cytotoxic T lymphocytes. Crit. Rev. Immunol. 18, 47–54. Shinde, S., Wu, Y., Guo, Y., Niu, Q, Xu, J., Grewal, I. S., Flavell, R. A., and Liu, Y. (1996). CD40L is important for induction of, but not response to, costimulatory activity. ICAM1 as the second costimulatory molecule rapidly upregulated by CD40L. J. Immunol. 157, 2764–2768. Sigal, L. J., Crotty, S., Andino, R., and Rock, K. L. (1999). CTL immunity to virus-infected non-haematopoietic cells requires presentation of exogenous antigen. Nature 398, 77–80. Simpson, E., Roopenian, D., and Goulmy, E. (1998). Much ado about minor histocompatibility antigens. Immunol. Today 19, 108–112. Sing, A. P., Ambinder, R. F., Hong, D. J., Jensen, M., Batten, W., Petersdorf, E., and Greenberg, P. D. (1997). Isolation of EBV-specific cytotoxic T lymphocytes that lyse Reed–Sternberg cells: Implications for immune-mediated therapy of EBV⫹ Hodgkin’s disease. Blood 89, 1978–1986. Skipper, J. C. A., Gulden, P. H., Hendrickson, R. C., Harthun, N., Caldwell, J. A., Shabanowitz, J., Engelhard, V. H., Hunt, D. F., and Slingluff, C. L. (1999). Mass-spectrometric evaluation of HLA-A*0201-associated peptides identifies dominant naturally processed forms of CTL epitopes MART-1 and gp100. Int. J. Cancer 82, 669–677 Slamon, D. J., Clark, G. M., Wong, S. G., Levin, W. J., Ulrich, A., and McGuire, W. L. (1987). Human breast cancer: Correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235, 177–182. Snijders, A., Kalinski, P., Hilkens, C. M. U., and Kapsenberg, M. L. (1998). High-level IL12 production by human dendritic cells requires two signals. Int. Immunol. 10, 1593–1598. Sogn, J. A. (1998). Tumor immunology: The glass is half full. Immunity 9, 757–763. Soiffer, R., Lynch, T. Mihm, M., and 24 others (1998). Vaccination with irradiated autologous melanoma cells engineered to secrete human GM–CSF generates potent anti-tumor immunity in patients with metastatic melanoma. Proc. Natl. Acad. Sci. USA 95, 13141– 13146. Sotomayor, E. M., Borrello, I., Tubb, E., Rattis, F. M., Bien, H., Lu, Z., Fein, S., Schoenberger, S., and Levitsky, H. I. (1999). Conversion of tumor-specific CD4⫹ T cell tolerance to T cell priming through in vivo ligation of CD40. Nature Med. 5, 780–787. Starzl, T. E., and Zinkernagel, R. M. (1998). Antigen localization and migration in immunity and tolerance. N. Engl. J. Med. 339, 1905–1913. Staveley-O’Carroll, K., Sotomayor, E., Montgomery, J., Borrello, I., Hwang, L., Fein, S., Pardoll, D., and Levitsky, H. (1998). Induction of antigen-specific T cell anergy: An early event in the course of tumor progression. Proc. Natl. Acad. Sci. USA 95, 1178–1183. Stockert, E., Ja¨ger, E., Chen, Y. T., Scanlan, M. J., Gout, I., Karbach, J. Arand, M., Knuth, A., and Old, L. J. (1998). A survey of the humoral immune response of cancer patients to a panel of human tumor antigens. J. Exp. Med. 187, 1349–1354. Stoltze, L., Dick, T. P., Deeg, M., Po¨mmerl, B., Rammensee, H. G., and Schild, H. (1998). Generation of the VSV nucleoprotein CTL epitope requires proteasome-dependent and -independent proteolytic activities. Eur. J. Immunol. 28, 4029–4036. Strasser, A., Harris, A. W., Huang, D. C., Krammer, P. H., and Cory, S. (1995). Bcl-2 and Fas/APO-1 regulate distinct pathways to lymphocyte apoptosis. EMBO J. 14, 6136–6147. Syrengelas, A. D., and Levy. R. (1999). DNA vaccination against the idiotype of a murine B cell lymphoma: Mechanism of tumor protection. J. Immunol. 162, 4790–4795. Takahashi, C., Mittler, R. S., and Vella, A. T. (1999). Cutting edge: 4-1BB is a bona fide CD8 T cell survival signal. J. Immunol. 162, 5037–5040.

280

MELIEF et al.

Tempero, R. M., VanLith, M. L., Morikane, K., Rowse, G. J., Gendler, S. J., and Hollingsworth, M. A. (1998). CD4⫹ lymphocytes provide MUC1-specific tumor immunity in vivo that is undetectable in vitro and is absent in MUC1 transgenic mice. J. Immunol. 161, 5500–5506. Theobald, M., Ruppert, T., Kuckelkorn, U., Hernandez, J., Ha¨nssler, A., Ferreira, E. A., Liewer, U., Biggs, J., Levine, A. J., Huber, C., Koszinowski, U. H., Kloetzel, P. M., and Sherman, L. A. (1998). J. Exp. Med. 188, 1017–1028. Tighe, H., Corr, M., Roman M., and Raz, E. (1998). Gene vaccination: Plasmid DNA is more than just a blueprint. Immunol. Today 19, 89–97. Tilkin, A. F., Lubin, R., Soussi, T., et al. (1995). Primary proliferative T cell response to wild-type p53 protein in patients with breast cancer. Eur. J. Immunol. 25, 1765–1769. Toes, R. E. M., Blom, R. J. J., Van der Voort, E., Offringa, R., Melief, C. J. M., and Kast, W. M. (1996a). Protective anti-tumor immunity induced by immunization with completely allogeneic tumor cells. Cancer Res. 56, 3782–3789. Toes, R. E. M., Offringa, R., Blom, R. J. J., Melief, C. J. M., and Kast, W. M. (1996b). Peptide vaccination can lead to enhanced tumor growth through specific tolerance induction. Proc. Natl. Acad. Sci. USA 93, 7855–7860. Toes, R. E. M., Blom, R. J. J., Offringa, R., Kast, W. M., and Melief, C. J. M. (1996c). Functional deletion of tumor-specific CTL induced by peptide immunization can lead to the inability to reject tumors. J. Immunol. 156, 3911–3918. Toes, R. E. M., Hoeben, R. C., Van der Voort, E. I., Melief, C. J. M., and Offringa, R. (1997a). Protective anti-tumor immunity induced by vaccination with recombinant adenoviruses encoding multiple tumor-associated cytotoxic T lymphocyte epitopes in a string-of-beads fashion. Proc. Natl. Acad. Sci. USA 94, 14660–14665. Toes, R. E. M., Schoenberger, S. P., Van der Voort, E. I. H., Kast, W. M., Hoeben, R. C., Melief, C. J. M., and Offringa, R. (1997b). Activation or frustration of anti-tumor responses by T cell based immune modulation. Semin. Immunol. 9, 323–327. Toes, R. E. M., van der Voort, E. I. H., Schoenberger, S. P., Drijhout, J. W., van Blooijs, L., Storm, G., Kast, W. M., Offringa, R., and Melief, C. J. M. (1998a). Enhancement of tumor outgrowth through CTL tolerization after peptide vaccination is avoided by peptide presentation on dendritic cells. J. Immunol. 160, 4449–4456. Toes, R. E. M., Schoenberger, S. P., Van der Voort, E. I. H., Offringa, R., and Melief, C. J. M. (1998b). CD40–CD40 L interactions and their role in CTL priming and antitumor immunity. Semin. Immunol. 10, 443–448. Toes, R. E. M., Ossendorp, F. O., Offringa, R., and Melief, C. J. M. (1999). CD4⫹ T cells and their role in antitumor immune responses. J. Exp. Med. 189, 753–756. Tollefson, A. E., Herimston, T. W., Lichtenstein, D. L., Colle, C. F., Tripp, R. A., Dimitrov, T., Toth, K., Wells, C. E., Doherty, P. C., and Wold, W. S. M. (1998). Forced degradation of fas inhibits apoptosis in adenovirus-infected cells. Nature 392, 726–730. Topalian, S. L., Gonzales, M. I., Parkhurst, M., Li, Y. F., Southwood, S., Sette, A., Rosenberg, S. A., and Robbins, P. F. (1996). Melanoma-specific CD4⫹ T cells recognize non-mutated HLA–DR-restricted tyrosinase epitopes J. Exp. Med. 183, 1965–1971. Tsang, K. Y., Zhu, M., Nieroda, C. A., Correale, P., Zaremba, S., Hamilton, J. M., Cole, D., Lam, C., and Schlom, J. (1997). Phenotypic stability of a cytotoxic T cell line directed against an immunodominant epitope of human carcinoembryonic antigen. Clin. Cancer Res. 3, 2439–2449. Tuting, T., Wilson, C. C., Martin, D. M., Kasamon, Y. L., Rowles, J., Ma, D. I., Slingluff, C. L., Jr, Wagner, S. N., van der Bruggen, P., Baar, J., Lotze, M. T., and Storkus, W. J. (1998). Autologous human monocyte-derived dendritic cells genetically modified to express melanoma antigens elicit primary cytotoxic T cell responses in vitro: Enhancement

STRATEGIES FOR IMMUNOTHERAPY OF CANCER

281

by cotransfection of genes encoding the Th1-biasing cytokines IL-12 and IFN-alpha. J. Immunol. 160, 1139–1147. Van Baren, N., Chambost, H., Ferrant, A., Michaux, L., Ikeda, H., Millard, I., Olive, D., Boon, T., and Coulie, P. G. (1998). Prame, a gene encoding an antigen recognized by cytolytic T cells, is expressed in acute leukaemia cells. Br. J. Haematol. 102, 1376–1379. Van Belzen, N., Hupkes, P. E., Doekharan, D., Hoogeveen-Westerveld, M., Dorssers, L. C., and van’t Veer, M. B. (1997). Detection of minimal disease using rearranged immunoglobulin heavy chain genes from intermediate- and high-grade malignant B cell non-Hodgkins lymphoma. Leukemia 11, 1742–1752. Van den Eynde, B., and Van der Bruggen, P. (1997). T cell-defined tumor antigens. Curr. Opin. Immunol. 9, 684–693. Van der Burg, S. H., Visseren, M. J., Brandt, R. M., Kast, W. M., and Melief, C. J. M. (1996). Immunogenicity of peptides bound to MHC class I molecules depends on the MHC-peptide complex stability. J. Immunol. 156, 3308–3314. Van der Burg, S. H., Offringa, R., and Melief, C. J. M. (1999). Viral cancer antigens. In ‘‘Cancer’’ (V. T. De Vita, S. Hellman, and S. A. Rosenberg, eds.), pp. 514–526. LippincottRaven, Philadelphia. Van Elsas, A., van der Burg, S. H., van der Minne, C. E., Borghi, M., Mourer, J. S., Melief, C. J. M., and Schrier, P. I. (1996). Peptide-pulsed dendritic cells induce tumoricidal cytotoxic T lymphocytes from healthy donors against stably HLA-A*0201 binding peptides from the melan-A/MART-1 self antigen. Eur. J. Immunol. 26, 1683–1689. Van Elsas, A., Hurwitz, A. A., and Allison, J. P. (1999). Combination immunotherapy of B16 melanoma using anti-CTLA4 and GM-CSF producing vaccines induces rejection of subcutaneous and metastatic tumors accompanied by autoimmune depigmentation. J. Exp. Med. 190, 355–366. Van Hall, T., van de Rhee, N. E., Schoenberger, S. P., Vierboom, M. P., Verreck, F. A., Melief, C. J., and Offringa, R. (1998). Cryptic open reading frames in plasmid vector backbone sequences can provide highly immunogenic cytotoxic T-lymphocyte epitopes. Cancer Res. 58, 3087–3093. Vierboom, M. P. M., Nijman, H. W., Offringa, R., Van der Voort, E., Van Hall, T., Van den Broek, L., Fleuren, G. J., Kenemans, P., Kast, W. M., and Melief, C. J. M. (1997). Tumor eradication by wild-type p53-specific cytotoxic T lymphocytes. J. Exp. Med. 186, 695–704. Vierboom, M. P. M., Gabrilovich, D. I., Offringa, R., Kast, W. M., and Melief, C. J. M. (1999). p53: A target for T cell-mediated immunotherapy. In ‘‘Peptide-based vaccines’’ (W. M. Kast, ed.). Landes Biosciences. Vinay, D. S., and Kwon, B. S. (1998). Role of 4-1 BB in immune responses. Semin. Immunol. 10, 481–489. Visseren, M. J. W., Van Elsas, A., van der Voort, E., Ressing, M. E., Kast, W. M., Schrier, P. I., and Melief, C. J. M. (1995). CTL specific for the tyrosinase autoantigen can be induced from healthy donor blood to lyse melanoma cells. J. Immunol. 154, 3991–3998. Vogel, L. A., and Noelle, R. J. (1998). CD40 and its crucial role as a member of the TNFR family. Semin. Immunol. 10, 435–442. Weinberg, A. D., Vella, A. T., and Croft, M. (1998). OX-40: Life beyond the effector T cell stage. Semin. Immunol. 10, 471–480. WHO Report of a technical meeting. (1999). The current status of development of prophylactic vaccines against human papillomavirus infection. Genoa, Italy. Wood, P., and Elliott, T. (1998). Glycan-regulated antigen processing of a protein in the endoplasmic reticulum can uncover cryptic CTL epitopes. J. Exp. Med. 188, 773–778.

282

MELIEF et al.

Ying, H., Zaks, T. Z., Wang, R.-F., Irvine, K. R., and Restifo, N. P. (1999). Cancer therapy using a self-replicating RNA vaccine. Nature Med. 5, 823–827. York, I. A., and Rock, K. L. (1996). Antigen processing and presentation by class I major histocompatibility complex. Annu. Rev. Immunol. 14, 369–396. Yoshino, I., Peoples, G. E., Goedegebuure, P. S., Maziarz, R., and Eberlein, T. J. (1994). Association of HER2/neu expression with sensitivity to tumor-specific CTL in human ovarian cancer. J. Immunol. 152, 2393–2400. Zaks, T. Z., and Rosenberg, S. A. (1998). Immunization with a peptide epitope (p 369–377) from HER-2/neu leads to peptide-specific cytotoxic T lymphocytes that fail to recognize HER-2/neu tumors. Cancer Res. 58, 4902–4908. Zaremba, S., Barraga, E., Zhu, M. Z., Soares, N., Tsang, K. Y., and Schlom, J. (1997). Identification of an enhancer agonist CTL peptide from human carcinoembryonic antigen. Cancer Res. 57, 4570–4577. Zhu, X., Tommasino, M., Vousden, K., Sadovnikova, E., Rappuoli, R., Crawford, L., Kast, W. M., Melief, C. J. M., Beverley, P. C. L., and Stauss, H. J. (1995). Both immunization with protein and recombinant vaccinia virus can stimulate CTL specific for the E7 protein of HPV16. Scand. J. Immunol. 42, 557–563.

ADVANCES IN IMMUNOLOGY, VOL. 75

Tyrosine Kinase Activation in the Decision between Growth, Differentiation, and Death Responses Initiated from the B Cell Antigen Receptor ROBERT C. HSUEH AND RICHARD H. SCHEUERMANN Laboratory of Molecular Pathology and Department of Pathology, University of Texas Southwestern Medical Center, Dallas, Texas 75235

I. Abstract

Immunoglobulin-containing receptors expressed on B lineage lymphocytes play critical roles in the development and function of the humoral arm of the immune system. The preB cell antigen receptor (preBCR) contains the immunoglobulin 애 heavy chain (Ig애) and signals to the preB cell that heavy chain rearrangement has been successful, a process termed heavy chain selection. The B cell antigen receptor (BCR) contains both Ig heavy and light chains and is expressed on immature and mature B cells before and after antigen encounter. Both receptor types form a complex with the Ig움 and Ig웁 proteins that link the predominantly extracellular Ig with intracellular signal transduction pathways. Signaling through the BCR induces different cellular responses depending on the nature of the signaling agent and the development stage of the target cell. These responses include clonal anergy and apoptotic deletion in immature B cells and survival, proliferation, and differentiation in mature B and preB cells. Several protein tyrosine kinases are activated rapidly following engagement of the BCR/preBCR complexes, including members of the Src family (Lyn and Blk), the Syk/ ZAP70 family (Syk), and the Tec family (Btk). In this review, we discuss possible mechanisms by which engagement of these similar receptor complexes can give rise to different cellular responses and the role that these kinases play in this process. II. Introduction

The B cell compartment contributes to the pliability and specificity of an immune system that must effectively respond to a multitude of antigens. A critical function of B cells is to provide the host with an arsenal of specific antibodies capable of neutralizing foreign molecules and mobilizing various defense mechanisms to eliminate invading pathogens. For example, recognition and opsinization of pathogens by soluble antibodies lead to ingestion and degradation by professional phagocytes. Thus, in this circum283

Copyright 䉷 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2776/00 $35.00

284

HSUEH AND SCHEUERMANN

stance, effective elimination of the pathogen requires that effector B cells synthesize and secrete pathogen-specific antibodies. Production of secreted antibodies requires effective clonal activation, expansion, and differentiation of the antigen-specific mature B cells, as originally predicted by Burnet (1959). However, the production of antigen-specific B cells must be balanced by the elimination of cells that express B cell antigen receptors (BCRs) that recognize normal cellular components in order to prevent autoimmune disease. The BCR plays critical roles in orchestrating the efficient development and subsequent activation of B lymphocytes at several discrete stages during B lineage differentiation. It is clear that other factors, including cytokines and co-stimulatory receptors, play important roles in modifying the cellular response to BCR engagement. Several reviews have been published that deal with these complex topics (e.g., Callard et al., 1996; Grewal and Flavell, 1998). However, it is also clear that the B cell response to BCR engagement in the absence of other influences can be quite variable. For example, in some cases, the cellular responses to BCR engagement include apoptosis and unresponsiveness and in other cases proliferation and activation, depending on the exact cell type and circumstance. In this review, we discuss the question of how signaling from the same receptor can have different consequences, and we focus on the function of three families of protein tyrosine kinase that physically associate with the BCR and that appear to be involved in these qualitative response decisions. III. B Cell Antigen Receptor Signaling Controls Several Stages of B Lymphocyte Development

Responses to signaling through membrane complexes that include immunoglobulin heavy chain can be classified into four different categories: heavy chain selection, tolerance induction, population maintenance, and antigen activation (Fig. 1). Each category appears to involve distinct cell types that represent different stages of B lineage differentiation. The cellular responses in each category can be quite different and include changes in cell surface and cytoplasmic protein expression, proliferative status, and cell function. A. HEAVY CHAIN SELECTION OF PREB CELLS B cell development can be broadly categorized into two stages. The first stage involves the production of immature B cells through a series of differentiation steps originating from the hematopoietic stem cell and is considered antigen-independent. The process begins with an orchestrated series of events that results in somatic recombination of gene segments of

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

285

FIG. 1. Cellular responses to BCR/preBCR signaling during B lineage differentiation. B cell development can be divided into several discrete stages, based on a variety of molecular and functional characteristics, beginning with the lineage-committed proB cell and ending with the immunoglobulin-secreting plasma cell. The BCR/preBCR plays a role in regulating cell kinetics at several stages. The earliest detectable receptor complex is the preBCR expressed on preB cells. The BCR receptor complexes include IgM expressed on immature cells and IgM and IgD expressed at high levels on mature B cells. The BCR/preBCR complexes play important roles in four characterized cellular processes: heavy chain selection, tolerance induction, population maintenance, and antigen activation. The cellular outcome of signaling is highly dependent on the strength of the signal initiated through the BCR/ preBCR and on the developmental stage of the target cell and can include differentiation, cell cycle entry, anergy, or death.

the immunoglobulin heavy chain variable (V) region (VHDHJH recombination) (Neuberger, 1997). Productive, in-frame rearrangements at the heavy chain locus lead to the expression of a preB cell receptor complex (preBCR) composed of 애m, the two surrogate light chain proteins ␭5 and VpreB, and two Ig움 and 웁 (CD79a and b) heterodimers at the preB cell stage. The expression of this preBCR is essential for the survival of the developing lymphocyte. Both ␭5 surrogate-light chain and 애 heavy chain knockout mice exhibit a blockade in early preB cell development (Kitamura et al., 1991, 1992). For example, targeted ablation of exon one of C애 by homologous recombination prevents the expression of membrane IgM, the first of the various classes of immunoglobulin expressed on B cells during their lifetime. Mice homozygous for this heavy chain disruption have few peripheral B cells, and the precursors found in the bone marrow are developmentally arrested at the early preB cell stage. Thus, signaling through the preBCR is required for cell survival, clonal expansion, and/ or further differentiation of these preB cells. This process is thought to function as a checkpoint during the development of B lineage cells, to ensure the generation of a functional heavy chain and signal transduction

286

HSUEH AND SCHEUERMANN

cascade before further differentiation. Thus, this process has been termed heavy chain selection. Signaling through the preBCR also appears to regulate V(D)J recombination. ProB cells express the RAG-1 and RAG-2 proteins involved in the recombination process and are competent for Ig heavy chain, but not Ig light chain, recombination. Preferential recombination of the heavy chain at this stage is thought to occur by regulation of recombination factor accessibility, a process that remains poorly defined at the biochemical level (Sleckman et al., 1998). Once a completed Ig heavy chain is generated and the preBCR is expressed on the surface, RAG gene expression is repressed and recombination activity drops (Bauer and Scheuermann, 1993; Grawunder et al., 1995; Ma et al., 1992). As a result of this preBCR signal, accessibility to the other IgH allele is closed, while accessibility to the light chain loci is opened. Ig light chain rearrangement then proceeds after re-expression of the RAG genes. As a result of this process, preBCR signaling prevents the expression of more than one Ig heavy chain in the same cell, a process known as allelic exclusion (Alt et al., 1980, 1982; Coleclough et al., 1981). B. TOLERANCE INDUCTION OF IMMATURE B CELLS Once the developing B cell expresses a functional BCR on the surface, consisting of Ig heavy and light chain proteins, it enters an antigenregulated phase. During V(D)J recombination in preB cells, the initial repertoire is generated in an antigen-independent process, and B cells expressing Ig specific for both foreign and self-antigens are generated. In order to fulfill its capacity to ignore self-antigens and respond to foreign antigens, the immune system has developed selection processes for inducing self-tolerance. In this process, self-reactive B cells are physically and/ or functionally eliminated from the pool of developing B cells, leaving only those B cells that do not recognize self to respond to foreign antigens. Tolerance induction is an active process because inhibitors of energy metabolism and protein or RNA synthesis can block tolerance induction in vitro (Klinman, 1996; Teale and Klinman, 1980, 1984). A large body of evidence suggests that engagement of the BCR on immature B cells gives rise to tolerance. Immature B cells can be distinguished from mature, naive B cells by virtue of the fact that they express low levels of IgD on the cell surface, even though both express comparable levels of IgM. In vitro engagement of the BCR on these immature IgM⫹ IgDlo cells derived from the adult bone marrow or neonatal spleen results in cell death, most likely through the process of apoptosis (Cambier et al., 1976; Metcalf and Klinman, 1976; Monroe, 1996; Norvell et al., 1995; Raff et al., 1975; Teale and Klinman, 1984; Yellen et al., 1991). In contrast, mature IgM⫹ IgDhi

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

287

cells enter the cell cycle in response to BCR engagement, as determined by 3H-thymidine incorporation and DNA content measurements with propidium iodide staining and FACS analysis (Monroe, 1996; Norvell et al., 1995; Yellen et al., 1991). Evidence for two types of tolerance-inducing mechanisms in vivo has come from a series of studies using mice transgenic for Ig heavy and light chains specific for either a self-antigen (Erikson et al., 1991; Nemazee and Burki, 1989; Okamoto et al., 1992) or a ‘‘pseudo-self-antigen’’ encoded by a second transgene (Goodnow et al., 1988, 1991; Klinman, 1996). In the presence of self-antigen expressed on the surface of cells in the animal, tolerance induction appears to occur at, or shortly after, the preB to immature B cell transition through deletion of autoreactive cells. However, if the self-antigen is soluble, autoreactive B cells survive but are unresponsive to subsequent antigenic stimulation. This unresponsive state has been termed anergy. These studies indicate that the stage of development dictates not only the cellular response to BCR engagement, but also the nature of the antigen. Interestingly, some B lymphoma cells lines, such as WEHI-231, BCL1. 3B3, CH33, Daudi, and DT40 with immature surface receptor phenotypes, undergo both BCR-induced apoptosis and cell-cycle arrest in the G1 phase of the cell cycle in vitro (for example, see Pennell and Scott, 1986; Scheuermann et al., 1994; Scott et al., 1986). This has led to the proposition that BCR stimulation of these cell lines provides a model of B cell tolerance in vitro. Taken together, these data indicate that engagement of the BCR in immature B cells leads to negative growth responses, including cell cycle arrest and apoptosis. C. ACTIVATION OF MATURE B CELLS Once precursors develop into mature, naive B cells, they emigrate to secondary lymphoid organs, where they circulate and recirculate awaiting activation by their specific antigen. The recognition of foreign antigen by cognate BCR complexes on select B cells drives their expansion and differentiation into either antibody-secreting, plasma cell effectors or memory B cells (Neuberger, 1997). Again, this process can be categorized as selective and antigen-dependent. Thus, in mature B cells, BCR interaction with foreign antigen is essentially a type of receptor–ligand activation response (Hollowood and Goodlad, 1998). As has been discussed, mature B cells express high levels of both IgM and IgD. Stimulation of mature splenic B cells takes on a much different profile than stimulation of immature B cells. Mature B cells that have not encountered antigen are small in size and are predominantly in the G0 phase of the cell cycle (Hollowood and Goodlad, 1998). Recognition of antigen by the BCR complex on mature B cells results in cell-cycle entry

288

HSUEH AND SCHEUERMANN

by progression from the G0 to the G1 phase. One to two days after activation by antigen through the specific BCR, cells enlarge and have a blast-like morphology. In vitro, splenic B cells incorporate 3H-thymidine in response to incubation with anti-IgM antibodies (Yellen et al., 1991), a response in stark contrast to the immature B cell response described previously. In addition, changes in the expression of nuclear protein, such as c-fos and egr1 (Yellen et al., 1991), and cell surface proteins, such as CD25 (Lowenthal et al., 1985) and CD40 (Klaus et al., 1994), occur in response to BCR stimulation. Thus, engagement of the BCR on mature B cells leads to positive growth responses, including cell activation and proliferation. However, stimulation of mature B cells does not always result in activation and differentiation into effector cells. Several studies have indicated that activation of mature cells can result in a change in BCR specificity through a process of receptor editing (Gay et al., 1993; Hertz and Nemazee, 1997; Radic et al., 1993; Retter and Nemazee, 1998; Tiegs et al., 1993). Whether this process serves as a fail-safe mechanism of tolerance induction or some other purpose remains to be determined and is likely controlled by the interplay between BCR signaling and co-stimulation. D. MAINTENANCE OF MATURE B CELLS The BCR also appears to play a role in mature B cells before they encounter antigen by providing a general survival signal (Lam et al., 1997; Neuberger, 1997). Deletion of the BCR on mature B cells in vivo, using a Cre-loxP-mediated inducible gene-targeting approach, resulted in a decrease in the number of peripheral, mature B lymphocytes (Lam et al., 1997). The loss of B220hi B lymphocytes was paralleled by an increase in apoptosis within the population of cells that had lost the heavy chain. This B cell phenotype could be rescued by the reintroduction of a transgenic immunoglobulin gene (Lam et al., 1997). This series of experiments suggests that the continued presence of the BCR is required for the maintenance of long-lived mature cells, even in the absence of cognate antigen. IV. Mechanisms Governing Different Cellular Responses from the Same Receptor

Due to the importance of the BCR in development, maintenance, and effector functions of B lymphocytes, studies have examined the differences in BCR signaling in B cells at different stages of development. Much of this work has focused on the response of immature versus mature cells to BCR signaling. In the simplest context, BCR engagement on immature cells results in growth inhibition, whereas BCR engagement on mature cells results in proliferation (Monroe, 1996). Characterization of the responses of mature and immature cells to BCR crosslinking has raised two broad

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

289

questions that warrant attention. First, how does engagement of the same antigen receptor elicit different responses at the mature and immature B cell stages? Second, within the immature B cell population, what are the biochemical mechanisms that dictate whether a cell will respond by cell cycle arrest and anergy or apoptosis and deletion from engagement of the same receptor? A. STAGE-SPECIFIC DIFFERENCES IN SIGNALING RESPONSE The signaling mechanisms modulating the differing responses of immature and mature B cells to BCR crosslinking are complex, but three models have garnered the most attention. The first model proposes that the difference in response to BCR signaling lies in the difference in signaling capabilities of IgM versus IgD. Immature cells initially express IgM only (Goodman et al., 1975; Vitetta et al., 1975), whereas mature cells express 5- to 10fold higher levels of IgD than IgM (Havran et al., 1984). This differential ratio of IgD to IgM expression has led to the hypothesis that the ratio of IgD to IgM is important in determining the response of B cells. Evolution of this model suggests that signaling through IgM would result in a growth inhibitory or negative signal. In contrast, signaling through IgD would give a proliferative or positive signal and could counter the negative signal sent through IgM. Evidence in support of this model comes from in vitro studies (Cambier et al., 1977; Scott et al., 1977; Vitetta et al., 1977). Culturing of splenic B cells with antigen results in plasma cell differentiation, as measured by antigen-specific antibody secretion using a plaque assay. However, when IgD is modulated on splenic B cells by anti-IgD pretreatment (to induce the internalization of the receptor), they become unresponsive to antigen. The WEHI-231 mouse B cell lymphoma line has also been used to test this hypothesis. WEHI-231 is IgM⫹ IgDlo (Lanier and Warner, 1981). However, if levels of IgD were increasd by ectopic expression following transfection, IgM crosslinking resulted in cell cycle arrest (Lanier and Warner, 1981; Scott et al., 1986) and apoptosis (Benhamou et al., 1990; Hasbold and Klaus, 1990), but IgD crosslinking did not (Ales-Martinez et al., 1988; Tisch et al., 1988). These results support the idea that the signaling capabilities of IgM and IgD are different. However, other data argue against this hypothesis. In experiments using CH33, another immature murine B cell lymphoma line capable of responding by apoptosis and cell cycle arrest (Pennell and Scott, 1986), no differences in the effects of IgM and IgD crosslinking were observed (Webb et al., 1989). In addition, the fact that IgD-positive B cells are deleted and anergized in mice expressing both the anti-hen egg lysozme (HEL) immunoglobulin and HEL transgenes appears to be inconsistent with a

290

HSUEH AND SCHEUERMANN

model in which IgD signals can overcome the negative signal delivered by IgM (Goodnow et al., 1988; Hartley et al., 1991). Finally, the most convincing argument against this model is that there are no obvious B cell defects in IgD knockout mice, which would be predicted by this model to lack activatable cells (Nitschke et al., 1993). For these reasons, the hypothesis that IgM mediates a negative signal and IgD mediates a positive signal does not adequately explain the difference in responses of immature and mature B cells (Monroe, 1996). A second model suggests that the responses of immature and mature B cells to BCR crosslinking are dictated not by the IgM or IgD receptors but by the presence or absence of T cell help (Monroe, 1996). Indeed, activated T cells are capable of blocking tolerance induction in B lymphoctyes isolated from neonatal spleen or bone marrow, sources of immature B cells (Chang et al., 1991; Metcalf and Klinman, 1976, 1977). This hypothesis is consistent with the known effects of other receptor/ligand interactions, such as IL-4 and its receptor or CD40 and CD40 ligand, on B cell activation. For example, BCR-induced apoptosis can be blocked by soluble CD40L (Rothstein et al., 1995; Wu et al., 1998). Perhaps T cell help is limiting within the bone marrow and neonatal spleen, where there is an abundance of immature B cells. So the physical separation of immature and mature B cells into different tissue compartments with different levels of T cell help could be resposible for tolerance of immature and activation of mature B cells—an effect of the environment that is not intrinsic to the cells. However, in vitro studies in the absence of T cell help indicate that there are intrinsic differences between the two cell types, because neonatal splenic and bone marrow B cells are susceptible to apoptosis following BCR engagement, whereas adult splenic B cells are not (Monroe, 1996; Norvell et al., 1995). This intrinsic difference in susceptibility to apoptosis supports a third model, in which the maturational stage of the B cell determines the response to signaling through the B cell receptor. The stage-specific differences in response could be dictated by differences in the signal transduction proteins present in each cell type (Monroe, 1996). Differences in the responses of immature and mature cells at the biochemical level have been well documented. For example, the BCR-dependent induction of the immediate/early genes egr-1 and c-fos seen in mature B cells is absent in immature cells (Yellen et al., 1991). Differences in BCR-induced inositol phospholipid hydrolysis were also reported in a comparison between neonatal and adult splenic B cells incubated in vitro with anti-IgM antibodies, suggesting that the biochemical differences must be membrane proximal (Yellen et al., 1991). The nature of this difference between immature and mature cells could lie in the expression and activation of protein tyrosine

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

291

kinases (PTKs) that lie between the receptor and phospholipid hydrolysis in the signaling cascade. In support of this idea, immature cells from day3 mouse spleen are deficient in the src-family kinases, p59fyn and p55fgr-1 (Wechsler and Monroe, 1995). Although the significance of this observation is as yet unclear, these results support the hypothesis that immature and mature B cells differ in the expression of membrane-proximal PTKs and that different PTKs or combinations of PTKs regulate different cellular responses. Other hypotheses involving PTK modulation of BCR responses have been offered to explain the differences at the biochemical level (Healy and Goodnow, 1998; Reth and Wienands, 1997) and will be discussed in detail. B. DIFFERENT RESPONSES WITHIN THE IMMATURE B CELL POPULATION As has been mentioned, tolerance induction to self-antigens within the immature B cell population occurs by two different mechanisms. In some cases, self-antigen causes elimination of autoreactive cells by inducing apoptosis. In other cases, self-antigen induces a state of unresponsiveness (anergy) such that cells remain viable but will not proliferate when subsequently stimulated with the same antigen. The decision between anergy and apoptosis appears to be dependent on the nature of the antigen. Using the antigen-specific IgM transgenic systems already described, it was found that membrane-bound antigen caused deletion of the antigen-specific B cells (Hartley et al., 1991). On the other hand, the same antigen in a soluble form caused anergy (Goodnow et al., 1988). Similar choices in cellular responses have been observed in vitro. B cell lines, including the Daudi human Burkitt’s lymphoma and the mouse WEHI-231 and BCL1.3B3 cell lines, provide a homogenous population of genetically identical cells for this analysis. These cells can be stimulated through the BCR using antibodies against membrane Ig (anti-Ig) in the absence of antigen. When these cells are stimulated with high concentrations of polyclonal anti-Ig preparations, a mixed cellular response is observed. Some of the cells become arrested in the G1 phase of the cell cycle, some of the cells die by apoptosis, and the remainder continue to proliferate (Benhamou et al., 1990; Hasbold and Klaus, 1990; Scheuermann et al., 1994; Scott et al., 1986; Yefenof et al., 1993). However, if lower concentrations of polyclonal anti-Ig are used, cell cycle arrest occurs without apoptosis (Hsueh et al., 1999; Marches et al., 1995). A similar response is observed using monoclonal anti-Ig antibodies in which cell cycle arrest occurs without apoptosis (Hsueh et al., 1999; Marches et al., 1995). However, if the monoclonal anti-Ig is hyper-crosslinked with a secondary antibody, apoptosis is induced (Hsueh et al., 1999; Marches et al., 1995). Taken

292

HSUEH AND SCHEUERMANN

together, these data suggest that the type of receptor clusters that form on the surface of stimulated, immature cells determines whether the cells become cell cycle arrested and unresponsive or deleted by apoptosis. Relatively small receptor clusters induced with soluble antigen, monoclonal antibodies, or low concentrations of polyclonal antibodies appear to induce unresponsiveness. Large receptor clusters induced by multivalent, membrane-bound antigen, hyper-crosslinked monoclonal antibodies or high concentrations of polyclonal antibodies appear to induce apoptosis. The mechanism by which the different receptor clusters are translated into different intracellular signals will be considered later. Although the different responses appear to be associated with different signaling pathways, they also appear to be connected in interesting ways. In the cell lines stimulated with high concentrations of polyclonal antibody, cells arrested in the cell cycle appear to be relatively resistant to apoptosis, because the apoptotic population comes mainly from the cycling pool (A. Hammill, R. C. H., and R. H. S., unpublished results). The cell cycle arrest response involves the induction of the p21 cyclin-dependent kinase inhibitor in Daudi cells (Marches et al., 1999). Cells transfected with a p21 antisense expression construct lose the cell cycle arrest response and the level of apoptosis increases (Marches et al., 1999). This suggests that either cell cycle arrest and/or p21 provide a protective effect against the apoptotic response. Apoptosis and proliferation are also tightly linked. Even though these two processes continue to occur for weeks after stimulation in culture, the number of viable cells remains remarkably constant (Hammill et al., 1999). This population stability indicates that apoptosis and cell cycle progression must be precisely balanced. This could occur either if cell cycle entry is involved in rescuing cells from apoptotic death or if apoptosis requires cell cycle entry in this system. Indeed, a linkage between apoptosis and cell cycle entry has been reported in cells that overexpress the c-myc protooncogene (Askew et al., 1991). In this system, overexpression of c-myc stimulates cell cycle entry and apoptosis in the absence of specific survival factors (e.g., interleukins). Apoptosis can be prevented and proliferation stimulated if bc12 expression is added (Bissonnette et al., 1992; Fanidi et al., 1992). In B cells, the linkage between cell cycle entry and apoptosis and the survival effects of cell cycle arrest may provide an explanation for some of the complex phenotypes observed in the BCR and tyrosine kinase knockout mice to be discussed. V. Ig␣ and Ig␤ Coreceptors in BCR Signaling

The BCR complex is a hetero-oligomeric receptor, expressed on B cells, whose extracellular domain recognizes antigen. One component of the BCR

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

293

complex is immunoglobulin, consisting of two extracellular light chains and two heavy chains with extracellular and transmembrane domains and a short cytoplasmic tail. In addition, the BCR complex contains two disulfidebonded heterodimers of Ig움 and Ig웁 that physically associate with Ig heavy chain (Campbell et al., 1991; Hombach et al., 1988, 1990). In vitro experiments using chimeric-molecule approaches were used to study the roles of Ig움 and Ig움 on the biochemical events induced by BCR crosslinking in the mature B cell line, A20 (Sanchez et al., 1993). Ig heavy chain molecules, carrying a mutation of two conserved polar residues within the transmembrane region, are not functional. They can neither associate with Ig움/Ig웁 heterodimers nor stimulate increases in intracellular Ca2⫹ and tyrosine phosphorylation. However, chimeric immunoglobulin molecules fused to the cytoplasmic tails of Ig움 or Ig웁 (Ig애/Ig움 and Ig애/Ig웁) reconstitute Ca2⫹ flux. Interestingly, differences between the functions of the Ig움 or Ig웁 tails were noted. Only the Ig애/Ig움 chimera could reconstitute increases in both Ca2⫹ and tyrosine phosphorylation levels induced by crosslinking (Sanchez et al., 1993). The Ig애/Ig웁 chimeric molecule reconstituted the Ca2⫹ flux only (Sanchez et al., 1993). In vivo and in vitro experiments using similar chimeric molecule approaches illustrate the importance of the cytoplasmic tails of Ig움 and Ig웁 in the cellular responses to signal transduction from the BCR complex. The B cell lymphoma, WEHI-231, has been used to examine the role of Ig움 and Ig웁 in BCR signaling of an immature B cell line. As has been described, engagement of the BCR gives both apoptotic and cell cycle arrest signals to WEHI-231 cells. Chimeric molecules, composed of the extracellular and transmembrane portions of the platelet-derived growth factor receptor (PDGFR) and the cytoplasmic tail of either Ig움 or Ig웁, are capable of inducing apoptosis in WEHI-231 following PDGF treatment, but only when both are present (Tseng et al., 1997); there was no effect of PDGF on cells transfected with PDGFR/Ig움 or PDGFR/Ig웁 alone. Although BCR crosslinking induces apoptosis and cell cycle arrest, the effects of these chimeric molecules on cell cycle arrest were not investigated. It would be of great interest to determine if the ability of WEHI-231 to induce both cell cycle arrest and apoptosis is somehow regulated by differential contributions of Ig움 and Ig웁. Introduction of similar chimeric molecules between the cytoplasmic domains of Ig움 and Ig웁 and the extracellular and transmembrane domains of Ig애 into Ig애⫺/⫺ mice has been used to investigate the importance of Ig움 and Ig웁 signaling in B cell development in vivo (Teh and Neuberger, 1997). Loss of either RAG or Ig애 proteins results in a block in B cell development at the proB cell stage. Introduction of chimeric transgenes into these knockout backgrounds allows the rescue of B cells from the

294

HSUEH AND SCHEUERMANN

developmental block imparted by the original mutation. One interpretation of this set of experiments is that signals provided by these chimeric molecules are sufficient to stimulate early B cell development. However, an alternative explanation is that because these molecules contain the entire extracellular and transmembrane domains of Ig애, they provide an appropriate scaffold for the assembly of a functional preBCR complex. In theory, stimulation of early B cell development could occur from this unusal preBCR complex through components other than Ig움 and Ig웁. Analysis of knockout animals further supports the hypothesis that Ig움 and Ig웁 are critical BCR signaling components. Complete ablation of Ig웁 in mice results in a complete impairment of B cell development. Oddly, the block appears to occur before VH to DHJH recombination within the B220⫹ CD43⫹ precursor B cell population, even before Ig애 is expressed. This suggests that there is a checkpoint that requires Ig웁 signaling during the VHDHJH recombination process that occurs before Ig애 is expressed (Gong et al., 1996; Kurosaki, 1999). Thus, in addition to the BCR and preBCR, a pro-BCR that lacks completed heavy and light Ig chains must exist. This putative pro-BCR complex may include partial D애 protein or may lack Ig components altogether. Because of this early block, the role of Ig웁 in BCR signaling of more mature lineage cell types cannot be examined in these knockout mice. The role of signal transduction through Ig움 was examined through a deletion of its cytoplasmic domain. The cytoplasmic domain of Ig움 is not required for proper assembly and surface expression of the BCR because Ig is found coupled with Ig웁 and truncated Ig움 (Torres et al., 1996). PreB cell development in Ig움 mutant mice is comparable to its littermate controls, suggesting that signal transduction through Ig움 is not required for early development. It is possible that Ig웁 provides the signal transduction requirements of early B cell development. However, a 10-fold decrease in the number of B cells can be found in the periphery (Torres et al., 1996). The cause for the decrease in peripheral mature B cells in Ig움⫺/⫺ mice might be similar to that observed in the transgenic mice subjected to an inducible ablation of the BCR from mature B cells (Lam et al., 1997). Both in vivo systems demonstrate a need for continued signal transduction provided through the BCR for population maintenance of mature, naive B cells (Kurosaki, 1999). From the results obtained from the Ig웁⫺/⫺ and the conditional Ig움⫺/⫺ knockout mice, it appears that the contributions of Ig움 and Ig웁 are not identical and potentially impact different stages of B cell development. Evidence that accentuates the potential differences of Ig움 and Ig웁 is shown by the diversity of glycosylated forms of Ig움 that vary at different stages of B cell development. In contrast, only one glycosylated form of Ig웁 is

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

295

detected in the human pro-, pre-, and mature B cell lines (Benlagha et al., 1999). The significance of these results is yet to be determined, but it is interesting to speculate that the different glycosylated forms of Ig움 may have different functions. VI. The BCR-Associated Protein Tyrosine Kinases

It is clear from these studies that the Ig움 and Ig웁 coreceptors play a critical role in BCR/pre-BCR signaling. However, these two proteins lack any defined catalytic activity that could function in signal transduction. The first clue as to how they might function came from the observation that immediately after BCR engagement, Ig움 and Ig웁 become phosphorylated on tyrosine residues (Flaswinkel and Reth, 1994). A comparison of these two proteins with other lymphocyte coreceptors identified a conserved tyrosine-containing region that might serve as the phosphorylation site (Fig. 2). The discovery of the immunoreceptor tyrosine-based activation motif (ITAM) found within the cytoplasmic tails of the Ig움 and Ig웁 is perhaps the most significant contribution to understanding how signal transduction occurs through the BCR complex (Reth, 1989; Samelson and Klausner, 1992; Weiss, 1993). The ITAMs of the lymphocyte coreceptors are characterized by the sequence motif (D/E-x7-D/Ex2-Y-x2-L/I-x7-Y-x2I/ L). Phosphorylated tyrosines in the ITAMs serve as potential docking sites for Src homology 2 (SH2) domain-containing proteins that are, in turn,

FIG. 2. The BCR complex found on the surface of immature and mature B cells. The phosphorylation state of the ITAM sequences within the Ig움 and Ig웁 (CD79a and b) coreceptors are depicted for the resting and the activated BCR complex.

296

HSUEH AND SCHEUERMANN

responsible for perpetuating signal transduction. At this stage, the identities of the kinase(s) responsible for phosphorylation of Ig움 and Ig웁 and the protein(s) that bind to the phosphorylated ITAMs are still debated. What is clear is that through the actions of Ig움 and Ig웁, three different families of tyrosine kinases are rapidly activated—the Src, Syk/ZAP70, and Tec families. Specific kinases from each of these families appear to play a unique role in connecting the BCR with downstream responses, as will be discussed in detail and summarized in Table I. A. SRC FAMILY TYROSINE KINASES The Src family of kinases is defined based on structural similarity with the Src kinase originally identified in the Rous Sarcoma Virus (Mustelin, 1994). Of the nine Src-family members, Lyn, Blk, and Fyn are expressed in B lineage cells and are activated following BCR engagement. A consideration of the Lyn kinase illustrates the essential structural components of this important class of signaling enzymes (Fig. 3). Like other Src family kinases, Lyn has a site of myristoylation in its amino terminus, followed by a Src homology domain 3 (SH3), an SH2, and a kinase domain (Src homology domain 1-SH1) (Courtneidge et al., 1980; Cross et al., 1984). The SH2 and SH3 domains are primarily docking domains for hetero- or homodimerization (Cohen et al., 1995; Pawson, 1995). SH2 domains bind inter- or intramolecularly to phosphorylated tyrosines, with the specificity regulated by flanking sequences (Felder et al., 1993; Panayotou et al., 1993). Proline-rich sequences intrinsic to some signaling adaptor proteins interact with SH3 domains (Feng et al., 1994; Lim et al., 1994; Pawson, 1995). Two downstream targets of activated Lyn have been described—the HS1 protein (Yamanashi et al., 1993) and the regulatory subunit of PI3 kinase (Pleiman et al., 1994). Like all other members of the Src family, Lyn contains a carboxy-terminal regulatory tyrosine, Tyr508 (Fig. 3) (Brown and Cooper, 1996). Mutation of this tyrosine increases the constitutive activity of the kinase in the absence of receptor engagement. This has led to the proposition that the C termini of these kinases might bind to their own SH2 domains when phosphorylated, thereby masking the catalytic domain located in between (Sicheri et al., 1997; Xu et al., 1997). The DT40 chicken B cell lymphoma has been particularly useful to test this hypothesis. DT40 cells have the peculiar property of having an extraordinarily high frequency of homologous recombination when DNA constructs are introduced into the cells by transfection (Buerstedde and Takeda, 1991). Thus, DT40 is an excellent cell line for the generation of targeted mutations in genes and has been an invaluable tool for the investigation of signal transduction from the BCR (Scheuermann and Uhr, 1995). Using this system, it was found that

TABLE I FUNCTION OF NONRECEPTOR TYROSINE KINASES IN BCR/PREBCR M Lyn Cell lines In vitro Heavy chain selection Tolerance induction Apoptosis CCA/Anergy Population maintenance Antigen activation Proliferation Effector cell differentiation a

Syk

Recombinant mice In vitro In vivo

Recombinant Cell lines In vitro

In vitro

No (Blk yes)a (Yes)b Yes

No Yes

Yes No

(Yes)c

Although there is no evidence that Lyn plays a role in heavy chain selection, another Src-family member, In this case, Lyn appears to normally block this response, because Lyn-deficient mice demonstrate -independent settings. c Again, Lyn appears to function normally to dampen this response, because splenocytes from Lyn-deficient b

298

HSUEH AND SCHEUERMANN

FIG. 3. Protein structure of three nonreceptor cytoplasmic tyrosine kinases activated after BCR crosslinking. Shown are Lyn, of the Src family; Syk, of the Syk/Zap 70 family; and Btk, of the Tec family of tyrosine kinases. Critical residues and domains regulating enzymatic activity of Lyn, Syk, and Btk are indicated and discussed in the text.

the phosphorylation of Tyr508 appears to be catalyzed by the Csk kinase, since it is hypophosphorylated in Csk-deficient DT40 cells. In addition, Lyn activity in unstimulated, Csk-deficient cells is equivalent to the level observed in wild-type cells after BCR stimulation (Hata et al., 1994). The CD45 phosphatase appears to be responsible for removal of the phosphate from Tyr508. Thus, in CD45-deficient DT40 cells, Lyn activity is diminished and Tyr508 is hyperphosphorylated (Yanagi et al., 1996). Thus, phosphorylation of Tyr508 by Csk is associated with repression of Lyn activity, whereas dephosphorylation by CD45 occurs upon BCR stimulation and is associated with activation. Although Tyr508 in Lyn is dephosphorylated, Tyr394 within the catalytic domain is rapidly phosphorylated upon BCR engagement (Fig. 3). In Src, phosphorylation of this residue appears to be required for maximal activity because its mutation severely impairs activation following receptor engagement (Kmiecik and Shalloway, 1987; Piwnica-Worms et al., 1987). It is likely that this residue serves a similar function in Lyn. The significance of Lyn kinase activation in BCR signaling comes from a variety of in vitro and in vivo studies. Again, the DT40 chicken B cell system has been particularly useful in this regard. Loss of Lyn protein through targeted gene disruption changes the biochemical responses of

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

299

the cell to BCR engagement. The normal rise in intracellular free Ca2⫹ is delayed and the plateau concentration is lower than observed in wildtype DT40 (Takata et al., 1994). As with other B lymphoma lines, DT40 undergoes apoptosis after BCR crosslinking. However, apoptosis is induced normally in Lyn-deficient DT40 (Takata et al., 1995). Thus, although Lyn can be linked to biochemical changes associated with signal transduction, its relevance to the downstream cellular responses in this system remains unclear. Other studies have implicated Lyn in changing cellular proliferation in response to BCR crosslinking (Scheuermann et al., 1994). The Daudi human Burkitt B cell lymphoma and the BCL1.3B3 mouse B cell lymphoma undergo G1 cell cycle arrest and apoptosis after BCR crosslinking and are thought to represent the immature B cell stage of development. Depletion of Lyn protein using Lyn-specific antisense oligonucleotides resulted in a loss of BCR-induced cell cycle arrest without affecting the apoptotic response in both cell lines, implicating Lyn in cell cycle arrest induction. This result suggests that the processes of apoptosis and cell cycle arrest induction in immature B cells are separable and potentially result from the differential activation of specific kinases. Further support for the importance of Lyn in controlling B cell proliferation comes from the analysis of Lyn-deficient mice (Chan et al., 1997; Wang et al., 1996). In contrast to what is observed for Ig웁⫺/⫺ mice, early B-lineage cells develop normally within the bone marrow of Lyn⫺/⫺ mice. However, peripheral mature B cell development and function are altered. The numbers of B cells considered recent emigrants to the periphery (B220lo HSAhi) are unchanged when compared to controls. However, the mature B cells in young mice are lower in number and hyper-responsive to crosslinking (Chan et al., 1997; Wang et al., 1996). Thus, the presence of Lyn, has a negative influence on proliferation but a positive influence on mature cell survival. As mice age, the splenocytes become less responsive to BCR stimulation, as indicated by lower levels of tyrosine phosphorylation (Hibbs et al., 1995; Nishizumi et al., 1995). This is most likely due to the accumulation of large numbers of myeloid cells and Mac1⫹CD5⫹ B cells, which are unresponsive to B cell stimulation (Chan et al., 1997; Wang et al., 1996). As in the DT40 system, mature cells that are generated in Lyn⫺/⫺ mice demonstrate a delayed rise in intracellular Ca2⫹. Lyn⫺/⫺ mice carrying a transgenic immunoglobulin receptor against hen egg lysozyme (Cornall et al., 1998) have exaggerated levels of cell deletion to self-antigen when crossed to mice expressing soluble hen egg lysozyme. Thus, not only is tolerance induction still active in the absence of Lyn, but also the anergic signal normally seen with soluble antigen appears to be converted into a deletional signal. Perhaps the loss of the cell cycle

300

HSUEH AND SCHEUERMANN

arrest response due to the loss of Lyn expression prevents the rescue of stimulated cells from apoptosis. However, even though negative selection appears to be exaggerated in these mice, there is an increase in immunization-independent formation of plasma cells in Lyn⫺/⫺ mice leading to the production of auto-antibodies. Investigating the function of the Fyn and Blk tyrosine kinases in B cell development has been less fruitful. Fyn-deficient mice have normal numbers of B lineage cells. Ca2⫹ flux, proliferation, and tyrosine phosphorylation in response to BCR signaling also appear to be normal. These results suggest either that BCR signal transduction through Fyn is not required for B cell development and function or that the loss of Fyn can be compensated for by another Src family member (Appleby et al., 1995; Sillman and Monroe, 1994). Indeed, mice deficient in both Lyn and Fyn demonstrate a decrease in peripheral B cells as compared with either knockout alone (Yasue et al., 1997). Like Fyn-deficient mice, Blk-deficient mice exhibit normal B cell development and function. In addition, B cell development, antibody production, and BCR-dependent proliferation are normal in cells overexpressing an inactive or SH2-defective form of Blk (Malek et al., 1998). However, transgenic mice expressing a constitutively active Blk mutant have a propensity to develop lymphoma. At 4–6 months of age, these mice develop clonal lymphomas. Although the ␬ locus was in the germline configuration, the tumors analyzed showed DH to JH rearrangements, but not VH to DHJH, as assessed by PCR analysis. Cell surface marker expression analysis confirmed that these tumors represented early B lineage cells, because they were CD19⫹, CD22⫹, B220lo, and mIg⫺. Within the bone marrow, a B220⫹ CD43⫹ population is overrepresented before tumor formation, suggesting that the tumors result from an expansion of cells with an early B cell progenitor phenotype (Malek et al., 1998). Thus, although loss of Blk function can be tolerated, overly active Blk influences early B lineage growth. B. SYK OF THE SYK/ZAP70 FAMILY OF KINASES Another nonreceptor, cytoplasmic, tyrosine kinase whose activity increases rapidly after BCR crosslinking is Syk, of the Syk/ Zap70 family. Unlike the related kinase Zap70, whose expression is more restricted, Syk is expressed in B cells, T cells, mast cells, natural killer (NK) cells, and other cells of the hematopoietic lineage (Law et al., 1994). The structure of this family of kinases is slightly different from the Src family. At the amino terminus, Syk has two tandem SH2 domains and lacks any SH3 domains (Fig. 3). In addition, Syk contains three tandem tyrosine residues at its carboxyl terminus. At this time, the role of these tyrosine residues

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

301

in Syk regulation remains to be determined. Phosphorylation of the tyrosine at position 323 is likely to lead to Syk activation by analogy with effects of a similar residue in ZAP70 (Chu et al., 1998). Based on the recognition sequence of the Syk SH2 domains, it has been proposed that Syk binds to diphosphorylated ITAMs of Ig웁, resulting in its activation (Chu et al., 1998). Two downstream targets of activated Syk have been identified—the Vav guanine nucleotide exchange factor (Deckert et al., 1996) and phospholipase C웂2 (PLC웂2) (Law et al., 1996). In the case of PLC웂2, Syk phosphorylation leads to enzymatic activation, resulting in the generation of the inositol trisphosphate (IP3) and diacylglycerol second messengers. The role that Syk plays in signal transduction from the BCR has been analyzed using targeted gene disruption in DT40 cells. Depletion of Syk in DT40 results in complete abrogation of BCR-induced calcium flux, tyrosine phosphorylation of PLC웂2, and apoptosis (Takata et al., 1994, 1995). The lack of an apoptotic response in Syk-depleted cells has also been observed in Daudi cell antisense and dominant-negative transfectants (R. C. H. and R. H. S., unpublished results). Because PLC웂2 phosphorylation and activation is required for Ca2⫹ release and because PLC웂2 phosphorylation is thought to be mediated by Syk in B cells, it is no surprise that the Ca2⫹ flux is ablated. Further demonstration of the link between Syk and PLC웂2 is the absence of BCR-induced apoptosis in DT40 deleted of PLC웂2 (Takata et al., 1995). Thus, as opposed to Lyn, Syk appears to be essential for the apoptotic response in immature B cells through the action of PLC웂2. Syk-deficient mice die perinatally, indicating that Syk activity must also be important for organ systems other than the immune system (Cheng et al., 1995; Chu et al., 1998; Turner et al., 1995; Tybulewicz, 1998). The early death of Syk-deficient mice has made the investigation of the immune system in these mice problematic. However, analysis of donor B lineage cells in radiation chimeras reconstituted with fetal liver from Syk-deficient mice demonstrates that Syk is required for B cell development. Few B220⫹ IgM⫹ IgD⫺ B cells accumulate within the bone marrow of reconstituted animals. These radiation chimeric mice have normal numbers of B220⫹ CD25⫺ IgM⫺ pro B cells but substantially reduced numbers of B220⫹ CD25⫹ IgM⫺ pre B cells (Turner et al., 1995). VHDHJH recombination appears normal. Taken together, these data suggest that the differentiation block is due to abnormal signal transduction through the preBCR. Because no mature Syk⫺/⫺ B cells are generated, it remains to be determined if Syk plays a role in signaling in mouse B cells. This type of analysis requires conditional gene disruption of Syk in B cells with a system similar to the inducible Cre-loxP system already mentioned. Because both Lyn and Syk are activated rapidly following BCR engagement, the question arises as to whether their activation is linked in any

302

HSUEH AND SCHEUERMANN

way. Lyn-deficient DT40 cells have reduced levels of tyrosine phosphorylation, as determined by whole cell lysate anti-phosphotyrosine immunoblot analysis (Kurosaki et al., 1994; Takata et al., 1994). Within these cells, tyrosine phosphorylation levels of the Syk protein are diminished, suggesting that Syk activation may lie downstream of Src family kinases (Kurosaki et al., 1994). However, the fact that some Syk activation occurs indicates that Lyn activation is not absolutely required. The intracellular Ca2⫹ flux in Lyn⫺/⫺ DT40 cells is altered, showing a slower rise after receptor stimulation, and plateuing at lower levels as compared with wild-type cells. This increase in intracellular Ca2⫹ is thought to be driven by the cleavage of inositol phospholipids by PLC웂2. Therefore, it is surprising that Ca2⫹ concentrations increase after BCR crosslinking within Lyn⫺/⫺ cells if Syk is really a downstream target of Lyn. One hypothesis is that an IP3independent mechanism of Ca2⫹ modulation exists (Tybulewicz, 1998). A second hypothesis is that Lyn and Syk are part of independent pathways and that the slow rise in Ca2⫹ is in fact due to Syk activation, leading to subsequent PLC웂2 phosphorylation and activation. A third hypothesis is that functional redudancy within the Src family allows Syk activation in the absence of Lyn. Another series of experiments argues against the requirement for Lyn in Syk activation. As discussed, the CD45 phosphatase participates in activating Src family kinases. In the CD45-deficient lymphoma cell line, J558L애m3, Src family kinase recruitment to and activation by the BCR is inhibited (Pao and Cambier, 1997; Tamir and Cambier, 1998). However, Syk phosphorylation still occurs to a limited extent, again suggesting that Syk phosphorylation can occur in a Src family-independent manner. In CD45-deficient DT40 cells, where Lyn activation is dramatically reduced, the kinetics and levels of Syk activation are the same as in wild-type DT40. Furthermore, if Syk activation is dependent on Lyn, then Lyn⫺/⫺ and Syk⫺/⫺ mice might be predicted to exhibit a similar lymphoid phenotype; however, this is not the case. Thus, it appears that Syk activation does not always require Lyn function. However, at this stage it is impossible to rule out the possibility that other Src family members can compensate for the loss of Lyn in these systems. This issue is unlikely to be resolved until Lyn, Fyn, and Blk triple-deficient mice or cell lines are generated to investigate Syk activation. C. BTK OF THE TEC FAMILY OF KINASES The Tec family of nonreceptor, cytoplasmic tyrosine kinases is composed of four family members—Itk, Bmx, Tec, and Btk (Desiderio, 1997). Btk, Bruton’s tyrosine kinase, is expressed in B lymphocytes and myeloid and erythroid cells but not in T cells. It is expressed at virtually all stages of

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

303

B cell development except the plasma cell stage (de Weers et al., 1994; Smith et al., 1994; Tsukada et al., 1993; Vetrie et al., 1993). Btk has been suggested to function not only in signaling pathways from the BCR but also in signaling pathways from CD40, the IL-5 receptor, and CD38 (Go et al., 1990; Santos-Argumedo et al., 1995; Sato et al., 1994). Btk was identified as the protein mutated in X-linked agamma-globulinemia (XLA) in humans (Tsukada et al., 1993; Vetrie et al., 1993). Analysis of the Btk gene in XLA patients has shown that the XLA phenotype correlates with deletions, insertions, and point mutations that lead to amino acid substitutions or premature stop codons. Virtually all domains within the Btk protein are susceptible to alteration, making it difficult to correlate mutations in different domains with function (Desiderio, 1997; Sideras and Smith, 1995). Members of this family of tyrosine kinases are distinguished by a pleckstrin homology (PH) domain at the amino terminus (Fig. 3), followed by a cysteine- and proline-rich Tec homology (TH) domain. SH3, SH2, and SH1 (kinase) domains follow the TH domain. Despite their resemblance to Src family tyrosine kinases, they are not myristoylated, nor do they contain a negative regulatory phosphorylation site corresponding to Tyr527 in the carboxy terminus (Brown and Cooper, 1996). However, Tyr551 appears to serve a function analogous to the tyrosines located adjacent to the catalytic domains in the other families, since mutations lead to lower activity (Rawlings et al., 1996). The roles of the PH and TH domains are unknown but have been suggested to be involved in protein–protein interactions (Desiderio, 1997). The high susceptibility of XLA patients to viral disease and pyrogenic bacterial infections correlates with an absence of plasma cells and the low levels of all classes of serum immunoglobulin (Tsukada et al., 1993; Vetrie et al., 1993). (The Btk deficiency is reviewed in depth elsewhere [Desiderio, 1997].) This phenotype is due to a block in B cell development. Peripheral B cells are rare in XLA patients, and the few peripheral B cells present have an immature phenotype, IgMhi and IgDlo. Analysis of the bone marrow compartment within these patients demonstrates that the number of preB cells expressing IgM is comparable between XLA patients and normal individuals, suggesting either an inability to expand or an increased susceptibility to cell death in the immature or mature B cell compartment. A single amino acid substitution of a conserved amino acid in the PH domain of Btk in CBA/N mice is also responsible for murine X-linked immunodeficiency (Xid) (Rawlings et al., 1993; Thomas et al., 1993). However, the Xid phenotype is less serve than that seen in XLA. Although T cell-independent responses are absent, Xid mice are capable of generating T cell-dependent responses. Furthermore, whereas all classes of serum Ig are reduced in XLA, only serum IgM and IgG3 are low in Xid. The number

304

HSUEH AND SCHEUERMANN

of peripheral B cells is slightly reduced, with an overrepresentation of the IgMhi and IgDlo immature population. Most interestingly, the same amino acid substitution in humans results in a severe phenotype, not like that seen in murine Xid, suggesting that other factors, perhaps genetic modifiers, may come into play in human XLA. Supporting this hypothesis is the fact that Btk⫺/⫺ mice exhibit a phenotype similar to Xid rather than XLA (Kerner et al., 1995; Khan et al., 1995). In vitro crosslinking of IgM on Xid and Btk⫺/⫺ splenic B cells fails to promote proliferation. This block in proliferation can be bypassed by treatment with ionomycin and phorbol ester, agents that increase Ca2⫹ concentrations and activate protein kinase C (PKC) (Khan et al., 1995). The effects of these reagents suggest that the action of Btk must be upstream of these events (Desiderio, 1997). In normal mature B cells, Bcl-xL is induced. B cells of Xid mice are incapable of inducing Bcl-xL. However, ectopic expression of Bcl-xL restores the proliferative capacity of Xid B cells (Solvason et al., 1998). This result suggests that Btk provides a survival signal required for proliferation. 웂-Irradiation induces apoptosis in DT40 cells but not in those that are Btk-deficient, suggesting that Btk also modulates survival in DT40 (Uckun et al., 1996). However, in this case Btk appears to promote apoptosis of these immature cells, in contrast to its effects in promoting survival of mature B cells in the animal. Finally, there is some indication that Btk phosphorylation and activation lies downstream of both Lyn and Syk. In Lyn-deficient DT40, Btk phosphorylation and activation show delayed kinetics. In Syk-deficient DT40, Btk activation kinetics are roughly normal, but the overall levels of activation are reduced. Thus, although Lyn and Syk may not be absolutely required for Btk activation, they do have an impact. VII. The Connections between Tyrosine Kinase Activation and Cellular Responses

Three points should be emphasized from a consideration of these genetic and biochemical data. First, it is clear that cells in different developmental stages respond differently to signaling through the BCR/preBCR. For example, immature B cells receive a negative signal, whereas mature B cells receive a positive signal. Second, cells within the same stage of development can respond differently depending on the nature of the stimulus. Immature B cells appear to respond to weak signals by arresting in the cell cycle and becoming anergic and to strong signals by undergoing apoptosis. Third, the membrane-proximal tyrosine kinases have different effects on these cellular responses. The assimilation of the data in hand into an encompassing model for cellular responses to antigen receptor signaling is perhaps premature, and certainly risky. Many questions remain unresolved,

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

305

and some of the data in the literature appears contradictory. Nevertheless, we will present a very simplistic and specific model to attempt to put together the information from this complex field. Obviously, the model may not be correct in some (or many) of its features, but the ideas are testable and will stimulate experimentation. The effects of interleukins and co-stimulation have not been incorporated for the sake of simplicity. However, it is clear that these factors can have a profound influence on the outcome of the cellular response by intersecting with the signaling pathways presented. With this in mind, the model is presented in Fig. 4. The model begins with a separation of the three major cell types responding to BCR/preBCR signaling—the preB cell, the immature B cell, and the mature B cell (Fig. 4). The responses of each cell type are subdivided based on the nature of the stimulus. Agents that cause relatively small receptor clusters are considered to give weak signals, whereas agents that cause large receptor clusters are considered to give strong signals. As they enter the cell, these signals that might be considered quantitative are converted into qualitatively different signals through the differential activation of the membrane-proximal tyrosine kinases. Although other kinases may be activated, the kinases listed in the model are proposed to predominate in determining the response. In the end, the combined actions of the activated kinases dictate the cellular responses—proliferation, differentiation, anergy, or death. The Syk tyrosine kinase is proposed to play a central and consistent role in each cell type. Much of the information regarding the effects of Syk activation comes from the analysis of immature B cells. Strong signals through the BCR in immature cells lead to the activation of Syk (Hsueh et al., 1999). As described, depletion of Syk in immature cells has been found to prevent BCR-induced apoptosis in several in vitro systems. In these systems, apoptosis and cell cycle progression are tightly linked. Therefor, Syk is proposed to initially induce cell cycle progression that leads to apoptosis and deletional tolerance (Fig. 4A). Weak signals through the BCR lead to Lyn activation, without Syk activation, and cell cycle arrest (Hsueh et al., 1999). This cell cycle-arrested state is proposed to be equivalent to clonal anergy in vivo. Depletion of Lyn abrogates cell cycle arrest, suggesting that Lyn activation is required for this response. Lyn is also activated by strong signals, and a population of arrested cells is observed in vitro. Therefore, one might expect to see the presence of some level of anergy in addition to deletion in vivo in systems that provide multivalent, hyper-crosslinking antigen since arrested cells appear to be resistant to apoptosis. In Lyn knockout mice, the peripheral B cells appear hyperresponsive because the cell cycle arrest pathway has been disrupted. However, fewer mature cells are found since signaling through the BCR gives an apoptotic response without the rescue effect of cell cycle arrest normally

FIG. 4. Model integrating BCR/preBCR-mediated tyrosine kinase activation with growth, differentiation, and death responses. The details of this model are discussed in the text.

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

307

provided by Lyn. In the absence of any signal through the BCR, the default response of immature B cells would be differentiation to the mature stage. In mature cells, strong signals through the BCR are again proposed to activate Syk, leading to cell cycle entry (Fig. 4B). However, in this case, the effects of Btk activation and Bcl-xL expression block the further progression into apoptosis. Thus, the combined action of Syk and Btk results in proliferation and clonal expansion that leads to subsequent differentiation within the population. A weak signal through the BCR is proposed to give Btk activation without Syk activation, leading to survival and population maintenance without proliferation and clonal expansion. In the absence of a signal, the default response in mature B cells would be apoptosis. Thus, in Btk knockout animals, few mature B cells would survive due to the lack of population maintenance in the absence of signaling, and little differentiation to the plasma cell stage would occur even in the presence of antigen due to Syk-induced apoptosis. The signaling pathways for preB cells are proposed to be similar to those present in mature B cells, except that the Blk kinase replaces the function of Btk. Thus, strong signals through the BCR are again proposed to activate Syk leading to cell cycle entry in preB cells (Fig. 4C). In this case, the additional activation of Blk blocks progression into apoptosis and allows proliferation, population expansion, and subsequent differentiation. A weak signal is proposed to activate Blk without Syk and would promote the survival of preB cells without proliferation. Thus, the propensity for transformation of early B lineage cells after an extended latency period in Blk transgenic mice would be related to its effect on survival of the target cells, in a manner similar to that observed in bc12 transgenics (Strasser, 1991). The reduced numbers of preB cells in radiation chimeras generated with Syk-deficient bone marrow would arise due to a lack of proliferation and population expansion. Again, for preB cells, the default response in the absence of a signal would be apoptosis due to the failure of heavy chain selection. Thus, differential tyrosine kinase activation is proposed to control differential cell fates in developing B lineage cells and the different responses to various antigen structures. Inherent in this model is that, as B lineage cells differentiate, the identities of the tyrosine kinases that dominate the signaling response also change. Whether this is dictated by changes in expression, activation, or influence of these kinases remains to be determined. VIII. Concluding Remarks

Signaling through the B cell antigen receptor has provided an excellent paradigm for understanding the complexities of signal transduction sys-

308

HSUEH AND SCHEUERMANN

tems. The challenge for the future will be to integrate the information regarding antigen structure, receptor structure, kinase activation, and costimulation into a scheme that can explain the various cellular responses. In addition, the linkage between cell cycle progression, apoptosis, and differentiation remains poorly understood, even though it is likely that the balance between these processes controls the development of organ systems and how they respond to their environment. Once again, the immune system promises to provide fertile ground for investigating these fundamental issues. ACKNOWLEDGMENTS We thank Jonathan Uhr and Angus Sinclair for critical review of this manuscript. This work was supported by grants CA78793 and CA09082 from the National Institutes of Health and RPG-95-079-03 from the American Cancer Society.

REFERENCES Ales-Martinez, J. E., Warner, G. L., and Scott, D. W. (1988). Immunoglobulins D and M mediate signals that are qualitatively different in B cells with an immature phenotype. Proc. Natl. Acad. Sci. USA 85, 6919–6923. Alt, F. W., Enea, V., Bothwell, A. L., and Baltimore, D. (1980). Activity of multiple light chain genes in murine myeloma cells producing a single, functional light chain. Cell 21, 1–12. Alt, F. W., Rosenberg, N., Enea, V., Siden, E., and Baltimore, D. (1982). Multiple immunoglobulin heavy-chain gene transcripts in Abelson murine leukemia virus-transformed lymphoid cell lines. Mol. Cell Biol. 2, 386–400. Appleby, M. W., Kerner, J. D., Chien, S., Maliszewski, C. R., Bondada, S., Perlmutter, R. M., and Bondadaa, S. (1995). Involvement of p59fynT in interleukin-5 receptor signaling [published erratum appears in J. Exp. Med. 1995 Oct 1;182(4):1179]. J. Exp. Med. 182, 811–820. Askew, D. S., Ashmun, R. A., Simmons, B. C., and Cleveland, J. L. (1991). Constitutive cmyc expression in an IL-3-dependent myeloid cell line suppresses cell cycle arrest and accelerates apoptosis. Oncogene 6, 1915–1922. Bauer, S. R., and Scheuermann, R. H. (1993). Expression of the VpreB/␭5/애 Pseudo-Ig complex correlates with downregulated RAG-1 expression and V(D)J type recombination: A mechanism for allelic exclusion at the IgH locus. Transgene 1, 33–45. Benhamou, L. E., Cazenave, P. A., and Sarthou, P. (1990). Anti-immunoglobulins induce death by apoptosis in WEHI-231 B lymphoma cells. Eur. J. Immunol. 20, 1405–1407. Benlagha, K., Guglielmi, P., Cooper, M. D., and Lassoued, K. (1999). Modifications of Ig alpha and Ig beta expression as a function of B lineage differentiation. J. Biol. Chem. 274, 19389–19396. Bissonnette, R. P., Echeverri, F., Mahboubi, A., and Green, D. R. (1992). Apoptotic cell death induced by c-myc is inhibited by bcl-2. Nature 359, 552–554. Brown, M. T., and Cooper, J. A. (1996). Regulation, substrates and functions of src. Biochim. Biophys. Acta 1287, 121–149. Buerstedde, J. M., and Takeda, S. (1991). Increased ratio of targeted to random integration after transfection of chicken B cell lines. Cell 67, 179–188.

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

309

Burnet, F. M. (1959). ‘‘The Clonal Selection Theory of Acquired Immunity.’’ Vanderbuilt Univ. Press, Nashville, TN. Callard, R. E., Matthews, D. J., and Hibbert, L. (1996). IL-4 and IL-13 receptors: Are they one and the same? Immunol. Today 17, 108–110. Cambier, J. C., Kettman, J. R., Vitetta, E. S., and Uhr, J. W. (1976). Differential susceptibility of neonatal and adult murine spleen cells to in vitro induction of B-cell tolerance. J. Exp. Med. 144, 293–297. Cambier, J. C., Vitetta, E. S., Kettman, J. R., Wetzel, G. M., and Uhr, J. W. (1977). Bcell tolerance. III. Effect of papain-mediated cleavage of cell surface IgD on tolerance susceptibility of murine B cells. J. Exp. Med. 146, 107–117. Campbell, K. S., Hager, E. J., Friedrich, R. J., and Cambier, J. C. (1991). IgM antigen receptor complex contains phosphoprotein products of B29 and mb-1 genes. Proc. Natl. Acad. Sci. USA 88, 3982–3986. Chan, V. W., Meng, F., Soriano, P., DeFranco, A. L., and Lowell, C. A. (1997). Characterization of the B lymphocyte populations in Lyn-deficient mice and the role of Lyn in signal initiation and downregulation. Immunity 7, 69–81. Chang, T. L., Capraro, G., Kleinman, R. E., and Abbas, A. K. (1991). Anergy in immature B lymphocytes. Differential responses to receptor-mediated stimulation and T helper cells. J. Immunol. 147, 750–756. Cheng, A. M., Rowley, B., Pao, W., Hayday, A., Bolen, J. B., and Pawson, T. (1995). Syk tyrosine kinase required for mouse viability and B-cell development. Nature 378, 303–306. Chu, D. H., Morita, C. T., and Weiss, A. (1998). The Syk family of protein tyrosine kinases in T-cell activation and development. Immunol. Rev. 165, 167–180. Cohen, G. B., Ren, R., and Baltimore, D. (1995). Modular binding domains in signal transduction proteins. Cell 80, 237–248. Coleclough, C., Perry, R. P., Karjalainen, K., and Weigert, M. (1981). Aberrant rearrangements contribute significantly to the allelic exclusion of immunoglobulin gene expression. Nature 290, 372–378. Cornall, R. J., Cyster, J. G., Hibbs, M. L., Dunn, A. R., Otipoby, K. L., Clark, E. A., and Goodnow, C. C. (1998). Polygenic autoimmune traits: Lyn, CD22, and SHP-1 are limiting elements of a biochemical pathway regulating BCR signaling and selection. Immunity 8, 497–508. Courtneidge, S. A., Levinson, A. D., and Bishop, J. M. (1980). The protein encoded by the transforming gene of avian sarcoma virus (pp60src) and a homologous protein in normal cells (pp60proto-src) are associated with the plasma membrane. Proc. Natl. Acad. Sci. USA 77, 3783–3787. Cross, F. R., Garber, E. A., Pellman, D., and Hanafusa, H. (1984). A short sequence in the p60src N terminus is required for p60src myristylation and membrane association and for cell transformation. Mol. Cell. Biol. 4, 1834–1842. de Weers, M., Mensink, R. G., Kraakman, M. E., Schuurman, R. K., and Hendriks, R. W. (1994). Mutation analysis of the Bruton’s tyrosine kinase gene in X-linked agammaglobulinemia: Identification of a mutation which affects the same codon as is altered in immunodeficient xid mice. Hum. Mol. Genet. 3, 161–166. Deckert, M., Tartare-Deckert, S., Couture, C., Mustelin, T., and Altman, A. (1996). Functional and physical interactions of Syk family kinases with the Vav proto-oncogene product. Immunity 5, 591–604. Desiderio, S. (1997). Role of Btk in B cell development and signaling. Curr. Opin. Immunol. 9, 534–540. Erikson, J., Radic, M. Z., Camper, S. A., Hardy, R. R., Carmack, C., and Weigert, M. (1991). Expression of anti-DNA immunoglobulin transgenes in non-autoimmune mice. Nature 349, 331–334.

310

HSUEH AND SCHEUERMANN

Fanidi, A., Harrington, E. A., and Evan, G. I. (1992). Cooperative interaction between c-myc and bcl-2 proto-oncogenes. Nature 359, 554–556. Felder, S., Zhou, M., Hu, P., Urena, J., Ullrich, A., Chaudhuri, M., White, M., Shoelson, S. E., and Schlessinger, J. (1993). SH2 domains exhibit high-affinity binding to tyrosinephosphorylated peptides yet also exhibit rapid dissociation and exchange. Mol. Cell. Biol. 13, 1449–1455. Feng, S., Chen, J. K., Yu, H., Simon, J. A., and Schreiber, S. L. (1994). Two binding orientations for peptides to the Src SH3 domain: Development of a general model for SH3-ligand interactions. Science 266, 1241–1247. Flaswinkel, H., and Reth, M. (1994). Dual role of the tyrosine activation motif of the Ig-alpha protein during signal transduction via the B cell antigen receptor. EMBO J. 13, 83–89. Gay, D., Saunders, T., Camper, S., and Weigert, M. (1993). Receptor editing: An approach by autoreactive B cells to escape tolerance. J. Exp. Med. 177, 999–1008. Go, N. F., Castle, B. E., Barrett, R., Kastelein, R., Dang, W., Mosmann, T. R., Moore, K. W., and Howard, M. (1990). Interleukin 10, a novel B cell stimulatory factor: Unresponsiveness of X chromosome-linked immunodeficiency B cells. J. Exp. Med. 172, 1625–1631. Gong, S., Sanchez, M., and Nussenzweig, M. C. (1996). Counterselection against D mu is mediated through immunoglobulin (Ig)alpha-Igbeta. J. Exp. Med. 184, 2079–2084. Goodman, S. A., Vitetta, E. S., Melcher, U., and Uhr, J. W. (1975). Cell surface immunoglobulin. XIII. Distribution of IgM and IgD-like molecules on small and large cells of mouse spleen. J. Immunol. 114, 1646–1648. Goodnow, C. C., Brink, R., and Adams, E. (1991). Breakdown of self-tolerance in anergic B lymphocytes. Nature 352, 532–536. Goodnow, C. C., Crosbie, J., Adelstein, S., Lavoie, T. B., Smith-Gill, S. J., Brink, R. A., Pritchard-Briscoe, H., Wotherspoon, J. S., Loblay, R. H., Raphael, K. et al. (1988). Altered immunoglobulin expression and functional silencing of self-reactive B lymphocytes in transgenic mice. Nature 334, 676–682. Grawunder, U., Leu, T. M., Schatz, D. G., Werner, A., Rolink, A. G., Melchers, F., and Winkler, T. H. (1995). Down-regulation of RAG1 and RAG2 gene expression in preB cells after functional immunoglobulin heavy chain rearrangement. Immunity 3, 601–608. Grewal, I. S., and Flavell, R. A. (1998). CD40 and CD154 in cell-mediated immunity. Annu. Rev. Immunol. 16, 111–135. Hammill, A. K., Uhr, J. W., and Scheuermann, R. H. (1999). Annexin V staining due to loss of membrane asymmetry can be reversible and precede commitment to apoptotic death. Exp. Cell Res. 251, 16–21. Hartley, S. B., Crosbie, J., Brink, R., Kantor, A. B., Basten, A., and Goodnow, C. C. (1991). Elimination from peripheral lymphoid tissues of self-reactive B lymphocytes recognizing membrane-bound antigens. Nature 353, 765–769. Hasbold, J., and Klaus, G. G. (1990). Anti-immunoglobulin antibodies induce apoptosis in immature B cell lymphomas. Eur. J. Immunol. 20, 1685–1690. Hata, A., Sabe, H., Kurosaki, T., Takata, M., and Hanafusa, H. (1994). Functional analysis of Csk in signal transduction through the B-cell antigen receptor. Mol. Cell. Biol. 14, 7306– 7313. Havran, W. L., DiGiusto, D. L., and Cambier, J. C. (1984). mIgM:mIgD ratios on B cells: Mean mIgD expression exceeds mIgM by 10-fold on most splenic B cells. J. Immunol. 132, 1712–1716. Healy, J. I., and Goodnow, C. C. (1998). Positive versus negative signaling by lymphocyte antigen receptors. Annu. Rev. Immunol. 16, 645–670.

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

311

Hertz, M., and Nemazee, D. (1997). BCR ligation induces receptor editing in IgM⫹IgDbone marrow B cells in vitro. Immunity 6, 429–436. Hibbs, M. L., Tarlinton, D. M., Armes, J., Grail, D., Hodgson, G., Maglitto, R., Stacker, S. A., and Dunn, A. R. (1995). Multiple defects in the immune system of Lyn-deficient mice, culminating in autoimmune disease. Cell 83, 301–311. Hollowood, K., and Goodlad, J. R. (1998). Germinal centre cell kinetics. J. Pathol. 185, 229–233. Hombach, J., Leclercq, L., Radbruch, A., Rajewsky, K., and Reth, M. (1988). A novel 34kd protein co-isolated with the IgM molecule in surface IgM-expressing cells. EMBO J. 7, 3451–3456. Hombach, J., Tsubata, T., Leclercq, L., Stappert, H., and Reth, M. (1990). Molecular components of the B-cell antigen receptor complex of the IgM class. Nature 343, 760–762. Hsueh, R. C., Hammill, A. K., Marches, R., Uhr, J. W., and Scheuermann, R. H. (1999). Antigen receptor signaling induces differential tyrosine kinase activation and population stability in B-cell lymphoma. Curr. Top. Micro. Immunol. 246, 299–305. Kerner, J. D., Appleby, M. W., Mohr, R. N., Chien, S., Rawlings, D. J., Maliszewski, C. R., Witte, O. N., and Perlmutter, R. M. (1995). Impaired expansion of mouse B cell progenitors lacking Btk. Immunity 3, 301–312. Khan, W. N., Alt, F. W., Gerstein, R. M., Malynn, B. A., Larsson, I., Rathbun, G., Davidson, L., Muller, S., Kantor, A. B., Herzenberg, L. A. et al. (1995). Defective B cell development and function in Btk-deficient mice. Immunity 3, 283–299. Kitamura, D., Kudo, A., Schaal, S., Muller, W., Melchers, F., and Rajewsky, K. (1992). A critical role of lambda 5 protein in B cell development. Cell 69, 823–831. Kitamura, D., Roes, J., Kuhn, R., and Rajewsky, K. (1991). A B cell-deficient mouse by targeted disruption of the membrane exon of the immunoglobulin mu chain gene. Nature 350, 423–426. Klaus, G. G., Holman, M., and Hasbold, J. (1994). Properties of mouse CD40: The role of homotypic adhesion in the activation of B cells via CD40. Eur. J. Immunol. 24, 2714–2719. Klinman, N. R. (1996). The ‘‘clonal selection hypothesis’’ and current concepts of B cell tolerance. Immunity 5, 189–195. Kmiecik, T. E., and Shalloway, D. (1987). Activation and suppression of pp60c-src transforming ability by mutation of its primary sites of tyrosine phosphorylation. Cell 49, 65–73. Kurosaki, T. (1999). Genetic analysis of B cell antigen receptor signaling. Ann. Rev. Immunol. 17, 555–592. Kurosaki, T., Takata, M., Yamanashi, Y., Inazu, T., Taniguchi, T., Yamamoto, T., and Yamamura, H. (1994). Syk activation by the Src-family tyrosine kinase in the B cell receptor signaling. J. Exp. Med. 179, 1725–1729. Lam, K. P., Kuhn, R., and Rajewsky, K. (1997). In vivo ablation of surface immunoglobulin on mature B cells by inducible gene targeting results in rapid cell death. Cell 90, 1073–1083. Lanier, L. L., and Warner, N. L. (1981). Cell cycle related heterogeneity of Ia antigen expression on a murine B lymphoma cell line:analysis by flow cytometry. J. Immunol. 126, 626–631. Law, C. L., Chandran, K. A., Sidorenko, S. P., and Clark, E. A. (1996). Phospholipase Cgammal interacts with conserved phosphotyrosyl residues in the linker region of Syk and is a substrate for Syk. Mol. Cell. Biol. 16, 1305–1315. Law, C. L., Sidorenko, S. P., Chandran, K. A., Draves, K. E., Chan, A. C., Weiss, A., Edelhoff, S., Disteche, C. M., and Clark, E. A. (1994). Molecular cloning of human Syk. A B cell protein-tyrosine kinase associated with the surface immunoglobulin M-B cell receptor complex. J. Biol. Chem. 269, 12310–12319.

312

HSUEH AND SCHEUERMANN

Lim, W. A., Richards, F. M., and Fox, R. O. (1994). Structural determinants of peptidebinding orientation and of sequence specificity in SH3 domains [published erratum appears in Nature 1995 Mar 2;374(6517):94]. Nature 372, 375–379. Lowenthal, J. W., Zubler, R. H., Nabholz, M., and MacDonald, H. R. (1985). Similarities between interleukin-2 receptor number and affinity on activated B and T lymphocytes. Nature 315, 669–672. Ma, A., Fisher, P., Dildrop, R., Oltz, E., Rathbun, G., Achacoso, P., Stall, A., and Alt, F. W. (1992). Surface IgM mediated regulation of RAG gene expression in E mu-N-myc B cell lines. EMBO J. 11, 2727–2734. Malek, S. N., Dordai, D. I., Reim, J., Dintzis, H., and Desiderio, S. (1998). Malignant transformation of early lymphoid progenitors in mice expressing an activated Blk tyrosine kinase. Proc. Natl. Acad. Sci. USA 95, 7351–7356. Marches, R., Hsueh, R., and Uhr, J. W. (1999). Cancer dormancy and cell signaling: Induction of p21(waf1) initiated by membrane IgM engagement increases survival of B lymphoma cells. Proc. Natl. Acad. Sci. USA 96, 8711–8715. Marches, R., Racila, E., Tucker, T. F., Picker, L., Mongini, P., Hsueh, R., Vitetta, E. S., Scheuermann, R. H., and Uhr, J. W. (1995). Tumour dormancy and cell signalling—III: Role of hypercrosslinking of IgM and CD40 on the induction of cell cycle arrest and apoptosis in B lymphoma cells. Ther. Immunol. 2, 125–136. Metcalf, E. S., and Klinman, N. R. (1977). In vitro tolerance induction of bone marrow cells: A marker for B cell maturation. J. Immunol. 118, 2111–2116. Metcalf, E. S., and Klinman, N. R. (1976). In vitro tolerance induction of neonatal murine B cells. J. Exp. Med. 143, 1327–1340. Monroe, J. G. (1996). Tolerance sensitivity of immature-stage B cells: Can developmentally regulated B cell antigen receptor (BCR) signal transduction play a role? J. Immunol. 156, 2657–2660. Mustelin, T. (1994). ‘‘Src Family Tyrosine Kinases in Leukocytes’’ (C. Kerkaporta, ed.). R. G. Landes Company, Austin, TX. Nemazee, D. A., and Burki, K. (1989). Clonal deletion of B lymphocytes in a transgenic mouse bearing anti-MHC class I antibody genes. Nature 337, 562–566. Neuberger, M. S. (1997). Antigen receptor signaling gives lymphocytes a long life. Cell 90, 971–973. Nishizumi, H., Taniuchi, I., Yamanashi, Y., Kitamura, D., Ilic, D., Mori, S., Watanabe, T., and Yamamoto, T. (1995). Impaired proliferation of peripheral B cells and indication of autoimmune disease in lyn-deficient mice. Immunity 3, 549–560. Nitschke, L., Kosco, M. H., Kohler, G., and Lamers, M. C. (1993). Immunoglobulin Ddeficient mice can mount normal immune responses to thymus-independent and -dependent antigens. Proc. Natl. Acad. Sci. USA 90, 1887–1891. Norvell, A., Mandik, L., and Monroe, J. G. (1995). Engagement of the antigen-receptor on immature murine B lymphocytes results in death by apoptosis. J. Immunol. 154, 4404– 4413. Okamoto, M., Murakami, M., Shimizu, A., Ozaki, S., Tsubata, T., Kumagai, S., and Honjo, T. (1992). A transgenic model of autoimmune hemolytic anemia. J. Exp. Med. 175, 71–79. Panayotou, G., Gish, G., End, P., Truong, O., Gout, I., Dhand, R., Fry, M. J., Hiles, I., Pawson, T., and Waterfield, M. D. (1993). Interactions between SH2 domains and tyrosinephosphorylated platelet-derived growth factor beta-receptor sequences: Analysis of kinetic parameters by a novel biosensor-based approach. Mol. Cell. Biol. 13, 3567–3576. Pao, L. I., and Cambier, J. C. (1997). Syk, but not Lyn, recruitment to B cell antigen receptor and activation following stimulation of CD45- B cells. J. Immunol. 158, 2663–2669. Pawson, T. (1995). Protein modules and signalling networks. Nature 373, 573–580.

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

313

Pennell, C. A., and Scott, D. W. (1986). Lymphoma models for B cell activation and tolerance. IV. Growth inhibition by anti-Ig of CH31 and CH33 B lymphoma cells. Eur. J. Immunol. 16, 1577–1581. Piwnica-Worms, H., Saunders, K. B., Roberts, T. M., Smith, A. E., and Cheng, S. H. (1987). Tyrosine phosphorylation regulates the biochemical and biological properties of pp60csrc. Cell 49, 75–82. Pleiman, C. M., Hertz, W. M., and Cambier, J. C. (1994). Activation of phosphatidylinositol3⬘ kinase by Src-family kinase SH3 binding to the p85 subunit. Science 263, 1609–1612. Radic, M. Z., Erikson, J., Litwin, S., and Weigert, M. (1993). B lymphocytes may escape tolerance by revising their antigen receptors. J. Exp. Med. 177, 1165–1173. Raff, M. C., Owen, J. J., Cooper, M. D., Lawton, A. R. d., Megson, M., and Gathings, W. E. (1975). Differences in susceptibility of mature and immature mouse B lymphocytes to anti-immunoglobulin-induced immunoglobulin suppression in vitro. Possible implications for B-cell tolerance to self. J. Exp. Med. 142, 1052–1064. Rawlings, D. J., Saffran, D. C., Tsukada, S., Largaespada, D. A., Grimaldi, J. C., Cohen, L., Mohr, R. N., Bazan, J. F., Howard, M., Copeland, N. G., et al. (1993). Mutation of unique region of Bruton’s tyrosine kinase in immunodeficient XID mice. Science 261, 358–361. Rawlings, D. J., Scharenberg, A. M., Park, H., Wahl, M. I., Lin, S., Kato, R. M., Fluckiger, A. C., Witte, O. N., and Kinet, J. P. (1996). Activation of BTK by a phosphorylation mechanism initiated by SRC family kinases. Science 271, 822–825. Reth, M. (1989). Antigen receptor tail clue [letter]. Nature 338, 383–384. Reth, M., and Wienands, J. (1997). Initiation and processing of signals from the B cell antigen receptor. Annu. Rev. Immunol, 15, 453–479. Retter, M. W., and Nemazee, D. (1998). Receptor editing occurs frequently during normal B cell development. J. Exp. Med. 188, 1231–1238. Rothstein, T. L., Wang, J. K., Panka, D. J., Foote, L. C., Wang, Z., Stanger, B., Cui, H., Ju, S. T., and Marshak-Rothstein, A. (1995). Protection against Fas-dependent Th1mediated apoptosis by antigen receptor engagement in B cells. Nature 374, 163–165. Samelson, L. E., and Klausner, R. D. (1992). Tyrosine kinases and tyrosine-based activation motifs. Current research on activation via the T cell antigen receptor. J. Biol. Chem. 267, 24913–24916. Sanchez, M., Misulovin, Z., Burkhardt, A. L., Mahajan, S., Costa, T., Franke, R., Bolen, J. B., and Nussenzweig, M. (1993). Signal transduction by immunoglobulin is mediated through Ig alpha and Ig beta. J. Exp. Med. 178, 1049–1055. Santos-Argumedo, L., Lund, F. E., Heath, A. W., Solvason, N., Wu, W. W., Grimaldi, J. C., Parkhouse, R. M., and Howard, M. (1995). CD38 unresponsiveness of xid B cells implicates Bruton’s tyrosine kinase (btk) as a regular of CD38 induced signal transduction. Int. Immunol. 7, 163–170. Sato, S., Katagiri, T., Takaki, S., Kikuchi, Y., Hitoshi, Y., Yonehara, S., Tsukada, S., Kitamura, D., Watanabe, T., Witte, O., et al. (1994). IL-5 receptor-mediated tyrosine phosphorylation of SH2/SH3-containing proteins and activation of Bruton’s tyrosine and Janus 2 Kinases. J. Exp. Med. 180, 2101–2111. Scheuermann, R. H., Racila, E., Tucker, T., Yefenof, E., Street, N. E., Vitetta, E. S., Picker, L. J., and Uhr, J. W. (1994). Lyn tyrosine kinase signals cell cycle arrest but not apoptosis in B-lineage lymphoma cells. Proc. Natl. Acad. Sci. USA 91, 4048–4052. Scheuermann, R. H., and Uhr, J. W. (1995). Connections between signal transduction components and cellular responses initiated by antigen receptor on B lymphocytes. J. Exp. Med. 182, 903–906.

314

HSUEH AND SCHEUERMANN

Scott, D. W., Layton, J. E., and Nossal, G. J. (1977). Role of IgD in the immune response and tolerance. I. Anti-delta pretreatment facilitates tolerance induction in adult B cells in vitro. J. Exp. Med. 146, 1473–1483. Scott, D. W., Livnat, D., Pennell, C. A., and Keng, P. (1986). Lymphoma models for B cell activation and tolerance. III. Cell cycle dependence for negative signalling of WEHI231 B lymphoma cells by anti-mu. J. Exp. Med. 164, 156–164. Sicheri, F., Moarefi, I., and Kuriyan, J. (1997). Crystal structure of the Src family tyrosine kinase Hck. Nature 385, 602–609. Sideras, P., and Smith, C. I. (1995). Molecular and cellular aspects of X-linked agammaglobulinemia. Adv. Immunol. 59, 135–223. Sillman, A. L., and Monroe, J. G. (1994). Surface IgM-stimulated proliferation, inositol phospholipid hydrolysis, Ca2⫹ flux, and tyrosine phosphorylation are not altered in B cells from p59fyn-1- mice. J. Leukoc. Biol. 56, 812–816. Sleckman, B. P., Bassing, C. H., Bardon, C. G., Okada, A., Khor, B., Bories, J. C., Monroe, R., and Alt, F. W. (1998). Accessibility control of variable region gene assembly during T-cell development. Immunol. Rev. 165, 121–130. Smith, C. I., Baskin, B., Humire-Greiff, P., Zhou, J. N., Olsson, P. G., Maniar, H. S., Kjellen, P., Lambris, J. D., Christensson, B., Hammarstrom, L., et al. (1994). Expression of Bruton’s agammaglobulinemia tryosine kinase gene, BTK, is selectively down-regulated in T lymphocytes and plasma cells. J. Immunol. 152, 557–565. Solvason, N., Wu, W. W., Kabra, N., Lund-Johansen, F., Roncarolo, M. G., Behrens, T. W., Grillot, D. A., Nunez, G., Lees, E., and Howard, M. (1998). Transgene expression of bcl-xL permits anti-immunoglobulin (Ig)-induced proliferation in xid B cells. J. Exp. Med. 187, 1081–1091. Strasser, A. (1991). B cell tumors. In ‘‘Mechanisms of B Cell Neoplasia 1991’’ (F. Melchers and M. Potter, eds.), pp. 181–186. Hoffmann-La Roche, Basel, Switzerland. Takata, M., Homma, Y., and Kurosaki, T. (1995). Requirement of phospholipase C-gamma 2 activation in surface immunoglobulin M-induced B cell apoptosis. J. Exp. Med. 182, 907–914. Takata, M., Sabe, H., Hata, A., Inazu, T., Homma, Y., Nukada, T., Yamamura, H., and Kurosaki, T. (1994). Tyrosine kinases Lyn and Syk regulate B cell receptor-coupled Ca2⫹ mobilization through distinct pathways. EMBO J. 13, 1341–1349. Tamir, I., and Cambier, J. C. (1998). Antigen receptor signaling: Integration of protein tyrosine kinase functions. Oncogene 17, 1353–1364. Teale, J. M., and Klinman, N. R. (1984). Membrane and metabolic requirements for tolerance induction of neonatal B cells. J. Immunol. 133, 1811–1817. Teale, J. M., and Klinman, N. R. (1980). Tolerance as an active process. Nature 288, 385–387. Teh, Y. M., and Neuberger, M. S. (1997). The immunoglobulin (Ig) alpha and Ig-beta cytoplasmic domains are independently sufficient to signal B cell maturation and activation in transgenic mice. J. Exp. Med. 185, 1753–1758. Thomas, J. D., Sideras, P., Smith, C. I., Vorechovsky, I., Chapman, V., and Paul, W. E. (1993). Colocalization of X-linked agammaglobulinemia and X-linked immunodeficiency genes. Science 261, 355–358. Tiegs, S. L., Russell, D. M., and Nemazee, D. (1993). Receptor editing in self-reactive bone marrow B cells. J. Exp. Med. 177, 1009–1020. Tisch, R., Roifman, C. M., and Hozumi, N. (1988). Functional differences between immunoglobulins M and D expressed on the surface of an immature B-cell line. Proc. Natl. Acad. Sci. USA 85, 6914–6918. Torres, R. M., Flaswinkel, H., Reth, M., and Rajewsky, K. (1996). Aberrant B cell development and immune response in mice with a compromised BCR complex. Science 272, 1804–1808.

TYROSINE KINASE ACTIVATION IN RESPONSES INITIATED FROM THE BCR

315

Tseng, J., Eisfelder, B. J., and Clark, M. R. (1997). B-cell antigen receptor-induced apoptosis requires both Ig alpha and Ig beta. Blood 89, 1513–1520. Tsukada, S., Saffran, D. C., Rawlings, D. J., Parolini, O., Allen, R. C., Klisak, I., Sparkes, R. S., Kubagawa, H., Mohandas, T., Quan, S., et al. (1993). Deficient expression of a B cell cytoplasmic tyrosine kinase in human X-linked agammaglobulinemia. Cell 72, 279–290. Turner, M., Mee, P. J., Costello, P. S., Williams, O., Price, A. A., Duddy, L. P., Furlong, M. T., Geahlen, R. L., and Tybulewicz, V. L. (1995). Perinatal lethality and blocked Bcell development in mice lacking the tyrosine kinase Syk. Nature 378, 298–302. Tybulewicz, V. L. (1998). Analysis of antigen receptor signalling using mouse gene targeting. Curr. Opin. Cell. Biol. 10, 195–204. Uckun, F. M., Waddick, K. G., Mahajan, S., Jun, X., Takata, M., Bolen, J., and Kurosaki, T. (1996). BTK as a mediator of radiation-induced apoptosis in DT-40 lymphoma B cells. Science 273, 1096–1100. Vetrie, D., Vorechovsky, I., Sideras, P., Holland, J., Davies, A., Flinter, F., Hammarstrom, L., Kinnon, C., Levinsky, R., Bobrow, M., et al. (1993). The gene involved in X-linked agammaglobulinaemia is a member of the src family of protein-tyrosine kinases [published erratum appears in Nature 1993 Jul 22;364(6435):362]. Nature 361, 226–233. Vitetta, E. S., Cambier, J. C., Ligler, F. S., Kettman, J. R., and Uhr, J. W. (1977). Bcell tolerance. IV. Differential role of surface IgM and IgD in determining tolerance susceptibility of murine B cells. J. Exp. Med. 146, 1804–1808. Vitetta, E. S., Melcher, U., McWilliams, M., Lamm, M. E., Phillips-Quagliata, J. M., and Uhr, J. W. (1975). Cell surface immunoglobulin. XI. The appearance of an IgD-like molecule on murine lymphoid cells during ontogeny. J. Exp. Med. 141, 206–215. Wang, J., Koizumi, T., and Watanabe, T. (1996). Altered antigen receptor signaling and impaired Fas-mediated apoptosis of B cells in Lyn-deficient mice. J. Exp. Med. 184, 831–838. Webb, C. F., Nakai, C., and Tucker, P. W. (1989). Immunoglobulin receptor signaling depends on the carboxyl terminus but not the heavy-chain class. Proc. Natl. Acad. Sci. USA 86, 1977–1981. Wechsler, R. J., and Monroe, J. G. (1995). Src-family tyrosine kinase p55fgr is expressed in murine splenic B cells and is activated in response to antigen receptor cross-linking. J. Immunol. 154, 3234–3244. Weiss, A. (1993). T cell antigen receptor signal transduction: A tale of tails and cytoplasmic protein–tyroisine kinases. Cell 73, 209–212. Wu, M., Bellas, R. E., Shen, J., and Sonenshein, G. E. (1998). Roles of the tumor suppressor p53 and the cyclin-dependent kinase inhibitor p21WAF1/CIP1 in receptor-mediated apoptosis of WEHI 231 B lymphoma cells. J. Exp. Med. 187, 1671–1679. Xu, W., Harrison, S. C., and Eck, M. J. (1997). Three-dimensional structure of the tyrosine kinase c-Src. Nature 385, 595–602. Yamanashi, Y., Okada, M., Semba, T., Yamori, T., Umemori, H., Tsunasawa, S., Toyoshima, K., Kitamura, D., Watanabe, T., and Yamamoto, T. (1993). Identification of HS1 protein as a major substrate of protein–tyrosine kinase(s) upon B-cell antigen receptor-mediated signaling. Proc. Natl. Acad. Sci. USA 90, 3631–3635. Yanagi, S., Sugawara, H., Kurosaki, M., Sabe, H., Yamamura, H., and Kurosaki, T. (1996). CD45 modulates phosphorylation of both autophosphorylation and negative regulatory tyrosines of Lyn in B cells. J. Biol Chem. 271, 30487–30492. Yasue, T., Nishizumi, H., Aizawa, S., Yamamoto, T., Miyake, K., Mizoguchi, C., Uehara, S., Kikuchi, Y., and Takatsu, K. (1997). A critical role of Lyn and Fyn for B cell responses to CD38 ligation and interleukin 5. Proc. Natl. Acad. Sci. USA 94, 10307–10312.

316

HSUEH AND SCHEUERMANN

Yefenof, E., Picker, L. J., Scheuermann, R. H., Tucker, T. F., Vitetta, E. S., and Uhr, J. W. (1993). Cancer dormancy: Isolation and characterization of dormant lymphoma cells. Proc. Natl. Acad. Sci. USA 90, 1829–1833. Yellen, A. J., Glenn, W., Sukhatme, V. P., Cao, X. M., and Monroe, J. G. (1991). Signaling through surface IgM in tolerance-susceptible immature murine B lymphocytes. Developmentally regulated differences in transmembrane signaling in splenic B cells from adult and neonatal mice. J. Immunol. 146, 1446–1454.

ADVANCES IN IMMUNOLOGY, VOL. 75

The 3ⴕ IgH Regulatory Region: A Complex Structure in a Search for a Function AHMED AMINE KHAMLICHI, ERIC PINAUD, CATHERINE DECOURT, CHRISTINE CHAUVEAU, AND MICHEL COGNE´ CNRS EP 118, Laboratoire d’Immunologie et Institut Universitaire de France, 87025 Limoges Cedex, France

I. Introduction

During precursor B cell differentiation, the genes encoding heavy and light chains of an antibody molecule are somatically assembled from germline DNA (reviewed in Tonegawa, 1983). This process occurs prior to antigenic challenge. Newly generated B cells will express on their surface homogeneous immunoglobulin (Ig) molecules with unique antigen-recognition sites that function as antigen receptors. Recognition of a given antigen by the specific B cell receptor (BCR) of a particular clone leads to activation and proliferation of that clone. Throughout these antigendependent stages, cells derived from the activated clone may be the target of hypermutation process and/or class switching and may differentiate into plasma cells or memory cells (reviewed in Rajewsky, 1996). A series of well-characterized and coordinated events takes place in the Ig locus during B cell differentiation converting unrearranged Ig genes into assembled, functional Ig genes (reviewed in Okada and Alt, 1995). Transcriptional regulation plays a central role in this process with time-, tissue-, and inducer-dependent features (reviewed in Libermann and Baltimore, 1991). Intensive functional and biochemical studies allowed the identification of several transcriptional control elements (promoters, enhancers, silencers) as targets for DNA-binding factors and unraveled the role of these elements during B cell development. These topics have been reviewed extensively, with special attention to the intronic enhancer E애 (reviewed in Max, 1993; Ernst and Smale, 1995; Henderson and Calame, 1998). Work with different systems (cell lines, transgenic, and knockout mice) has shed some light on the putative function of a highly complex control region lying downstream of the Ig heavy chain (IgH) locus, which will be the focus of this review. II. Structure of the 3ⴕ IgH Control Region

Soon after the identification of the IgH intron-enhancer E애 (Banerji et al., 1983; Gillies et al., 1983; Neuberger, 1983), it became clear that 317

Copyright 䉷 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2776/00 $35.00

318

KHAMLICHI et al.

other enhancers may exist outside the JH –C애 intron. Cell lines have been reported that efficiently transcribe their IgH genes despite deletion of E애 (Klein et al., 1984; Wabl and Burrows, 1984; Aguilera et al., 1985; Eckhardt and Birshtein, 1985). The c-myc proto-oncogene was found to be activated upon translocation into the IgH locus without necessarily being juxtaposed to E애 (Neuberger and Calabi, 1983). Furthermore, a large deletion of sequences downstream of the mouse C움 was found to be associated with decreased IgH transcription in a cell line that retained its endogenous E애 (Gregor and Morrison, 1986). Indeed, a second B cell-specific enhancer was discovered some 25 kilobases (kb) downstream of the rat C움 (Pettersson et al., 1990). The mouse IgH 3⬘E was subsequently identified at approximately 16 kb downstream of C움 (Dariavach et al., 1991; Lieberson et al., 1991). The rat and the mouse enhancers show high sequence identity although they have opposite orientations. Strikingly, the rat and the mouse IgH 3⬘E core enhancers are flanked by matched inverted repeats (see following) (Dariavach et al., 1991). Another weak B cell-specific enhancer (C움 3⬘E) was identified about 2 kb downstream of the mouse C움 membrane exon (Matthias and Baltimore, 1993). Search for additional regulatory elements based on DNase I hypersensitivity assay led to the discovery of two HSs (DNase I hypersensitive sites) named HS1 and 2 that actually mapped the IgH 3⬘E and of two other B cell-specific enhancers (Giannini et al., 1993; Madisen and Groudine, 1994; Michaelson et al., 1995). One termed HS3 is located 앑29 kb downstream of the mouse C움 while HS4 lies some 4 kb farther (Fig. 1). Striking features were revealed by extensive sequencing of this region: the inverted repeats flanking the IgH 3⬘E were actually part of a larger palindromic structure (probably the largest reported in mammals), the center of which is IgH 3⬘E. The boundaries of this palindrome spanning more than 20 kb are HS3 and C움 3⬘E that proved to be 97% identical although in opposite orientations (Chauveau and Cogne´, 1996; Saleque et al., 1997). For the sake of clarity, the following terminology has been adopted: IgH 3⬘E : HS1,2; C움3⬘E : HS3a; and HS3 : HS3b (Saleque et al., 1997) (Fig. 1). The human 3⬘ IgH regulatory region proved more difficult to clone, partly because of the duplication of arrays of CH genes and the presence of 20 bp tandem repeats downstream of both C움 genes (Chen and Birshtein, 1995; Gualandi et al., 1995; Kang and Cox, 1996). Three groups independently cloned and characterized the human 3⬘ IgH regulatory region downstream of both C움 genes, including the three enhancers: HS3a (Mills et al., 1997; Pinaud et al., 1997), HS1,2 (Chen and Birshtein, 1997;

THE 3⬘ IgH REGULATORY REGION

319

FIG. 1. Structure of the mouse and human IgH loci (not to scale). Shown are the schematics of a rearranged 애 gene. (A) The mouse 3⬘ IgH regulatory region is located downstream of the C움 gene. IR and TR denote inverted and tandem repeats, respectively. Sx and Ix denote switch sequences and germline I promoters, respectively, upstream of constant genes. Arrows indicate the orientation of the germline transcripts. Also indicated are the matrix attachment regions (MARs) surrounding E애 and within some IgH introns. The dotted arrows indicate the extent of the deletions in LP1-2 and 70Z/3 cell lines. (B) The 3⬘ IgH regulatory region is duplicated in human IgH locus and is located downstream of both C움 genes. The 20 bp tandem repeats just downstream of the C움 genes are indicated by small arrows, and CpG islands are shown downstream of HS4. See text for details and references.

Mills et al., 1997; Pinaud et al., 1997), and HS4 (Mills et al., 1997). Again, large inverted repeats were found to symmetrically flank HS1,2. Interestingly, HS3b does not appear to exist in humans but the significance of this finding is not clear. Regarding expression during B cell development, the human enhancers generally displayed the same expression pattern as their mouse homologs (Chen and Birshtein, 1997; Mills et al., 1997; Pinaud et al., 1997). In addition, potential CpG islands have been identified downstream of both HS4s and are candidates for an as-yet-unidentified regulatory elements (Sadhu et al., 1997) (Fig. 1). III. Activity during B Cell Development

Activity of the 3⬘ IgH region enhancers was mainly assayed by transient transfection in cell lines thought to represent different stages of B cell differentiation and by transgenic models.

320

KHAMLICHI et al.

The rat HS1,2 enhancer was initially reported to have a transient enhancer activity comparable to E애 in a mouse plasmacytoma cell line and was suggested to function mainly at late B cell development (Pettersson et al., 1990). In contrast, the mouse HS1,2 enhancer proved to be rather weak in mouse plasmacytoma cell lines in transient transfection assays (Dariavach et al., 1991; Lieberson et al., 1991) and contrasting results were found in the mature B cell lines (Dariavach et al., 1991; Fulton and Van Ness, 1994). Transgenic mice were generated in which the mouse HS1,2 was inserted upstream of reporter genes. High-level expression was achieved in activated B cells. However, aberrant expression was noted in thymus, heart, and kidney and a strict B cell-specific expression could not be achieved with HS1,2 alone (Arulampalam et al., 1994; Andersson et al., 1999). As to HS3a, transient transfection assay showed that it was a weak enhancer, with only 5–15% as active as E애 in B cell and plasmacytoma cell lines (Matthias and Baltimore, 1993). Although variation in transcriptional activity was noted among the cell lines used, HS3b and HS4 were also generally considered to be weak activators (Madisen and Groudine, 1994; Michaelson et al., 1995). Overall, whereas HS1,2, HS3a, and HS3b were active at late B cell differentiation stage, HS4 seemed to be active throughout B cell development (Madisen and Groudine, 1994; Michaelson et al., 1995). This expression pattern correlates quite well with the methylation pattern of the 3⬘ region. In general, the 3⬘ region is hypermethylated at the pre-B cell stage and becomes demethylated at the plasma cell stage, but it is not clear whether this applies to HS4 (Giannini et al., 1993). IV. Synergies between 3ⴕ Enhancers

An important question concerning the multiple enhancers lying within or 3⬘ of the IgH locus is their potential synergy in transcription activation. To fully appreciate the complexity of the problem, one has to bear in mind that the 3⬘ IgH regulatory region is some 앑200 kb from E애 (before class switching) and that the four HSs are kilobases distant from each other within a palindromic structure (except for HS4). Many groups have tackled the question of synergy between the 3⬘ enhancers and E애 in activating transcription. This was generally achieved by using reporter genes under the control of a combination of the enhancers under study. Transcriptional activation was then monitored either by transient or stable transfection or in transgenics. These studies differ in the promoters (and their potential compatibility with enhancers), the arrangement of the enhancers, and the cell lines and reporter genes used, which

THE 3⬘ IgH REGULATORY REGION

321

may explain the discrepancies found between the different groups in some instances. Consequently, we tried to extract the most consistent findings from these studies. The first experimental data to suggest some kind of synergy between E애 and HS1,2 came from analysis of stable plasmacytoma transfectants carrying a PVH-driven-rearranged 애 gene. When the mouse HS1,2 was included in this construct carrying E애, a twofold increase in the expression rate was noted (Dariavach et al., 1991). By using the same approach, another group found an increase in expression rate, but only when HS1,2 was located downstream of the constant exons. Increasing the distance between HS1,2 and E애 further enhanced transcription. Strikingly, inclusion of HS1,2 upstream of E애 in this construct abrogated transcription, this inhibition being strictly dependent on the presence of E애. Thus, HS1,2 appears to interact with E애 in a position- and distance-dependent manner (Mocikat et al., 1993, 1995). In another study, a search for synergy between E애 and HS1,2 was done by using Luciferase reporter gene under the control of thymidine kinase promoter. Transient transfection in pre-B, B, plasma cell lines concluded to a lack of cooperation between the two enhancers at any stage of B cell development (Fulton and Van Ness, 1994). Subsequent identification of HS3b and HS4 allowed the study of their contributions to transcriptional enhancement. HS1,2, HS3b, and HS4 were cloned individually or in combinations upstream of the human growth hormone gene under the control of either the human c-myc promoter or the mouse V␭1 light chain promoter (Madisen and Groudine, 1994). Both promoters were only moderately activated in a plasmacytoma cell line when HS1,2, HS3b, and HS4 were used individually. In contrast, a strong synergistic activation of transcription resulted from combination of the different HSs (HS123, HS124, or HS34), the maximal activation being exhibited by HS1234. When the same constructs were assayed in a preB cell line, it was found that neither HS1,2 nor HS3 stimulated transcription, even when present in multiple copies. The presence of HS4 was mandatory for activating transcription from both promoters (Madisen and Groudine, 1994). It was also shown that HS1,2 was strongly boosted when it was flanked by two copies of HS3 in inverted orientation (mimicking the endogenous arrangement) and this 3⬘ regulatory palindrome proved able to stimulate transcription by itself in mature B cell lines whereas it was only a coactivator of HS4 in a pre-B cell line (Chauveau et al., 1998). Two groups studied the cooperation between the 3⬘ enhancers and E애 by using a VH promoter driving either Luciferase (Ong et al., 1998) or Chloramphenicol Acetyl Transferase (Chauveau et al., 1998) as reporter genes. In a pre-B cell line, both groups found that, in the absence of E애,

322

KHAMLICHI et al.

HS4 was generally required to achieve some transcriptional activity either alone or, more efficiently, in combination with HS1,2 or HS3 (Ong et al., 1998; Chauveau et al., 1998). Addition of E애 to any individual 3⬘ element did not significantly increase transcription above the level exhibited by E애 alone, but a strong effect was obtained when E애 was combined with HS3a123b4 (a combination of HS3a, HS1,2, HS3b, and HS4) (Ong et al., 1998; Chauveau et al., 1998). The cooperation between E애 and HS3a123b4 for maximal activation appears to be a constant finding irrespective of the developmental stage of the B cell line used. Thus, it seems that while the 3⬘ enhancers are by themselves rather weak activators, they act as powerful coenhancers when optimally (i.e., in a manner that mimics their position within the endogenous locus) combined to E애 (Chauveau et al., 1998). The presence of these tissue-specific and cell-stage-specific HSs led to the suggestion that this region could act as a locus control region, that is, a region able to direct high-level, tissue-specific expression of a linked gene in a position-independent, copy number-dependent manner (Madisen and Groudine, 1994; reviewed in Martin et al., 1996). This suggestion was based on the finding that HS123b4 conferred the above properties to a linked c-myc gene in stably transfected plasmacytoma cell lines (Madisen and Groudine, 1994). In a recent work, transgenic mice bearing a VH promoter-driven human 웁-globin gene linked to HS3a123b4 were generated. High-level, integration site-independent and B cell-specific expression was found, but no strict correlation could be established regarding copy number dependence (Chauveau et al., 1999). These results were interpreted as an indication that additional sequences are necessary to allow HS3a123b4 to act as a classical LCR (Chauveau et al., 1999). The contrast between the strict B cell-specific expression found with the LCRbearing transgene (Chauveau et al., 1999) and the aberrant expression of HS1,2 only-driven transgenes (Arulampalam et al., 1994; Andersson et al., 1999) suggests that another function of the LCR, which requires the full LCR but not an incomplete one, is to silence IgH expression in non-B cells (see Ortiz et al., 1997, for the case of the TCR움 LCR). From these studies, the picture that emerges is that the fine tuning of transcription within the IgH locus may result from the balanced and synergistic effects of E애 and the 3⬘ IgH regulatory region throughout B cell development. E애 and HS4 would be leaders at the pre-B stage. As B cells mature, the HS1,2 and HS3a,b activators would become predominant but still synergize with E애 and HS4 for full transcriptional activity at the plasma cell stage. Whether these differential activities coincide with the localization of HS4 outside the palindrome is a matter of speculation (other aspects of synergy will be discussed following with knockout mice and cell lines).

THE 3⬘ IgH REGULATORY REGION

323

V. DNA-Binding Proteins

Given the tight regulation of the 3⬘ enhancers activity during B cell development, it is of interest to know whether they are inducible by extracellular signals and if so, the nature of the nuclear factors that bind to their cognate sites following such stimuli. In this section, we take into account only those studies that functionally or biochemically characterized interactions between trans-acting factors and their binding sites within the HSs. Suggestions based only on sequence homologies, though indicated in Fig. 2, will not be considered here. Two imperfect but functional octamer (OCT) sites were identified within the mouse HS3a (Fig. 2). Electrophoretic mobility shift assay (EMSA) and competition experiments showed that they specifically bound, though with low affinity, octamer factors Oct-1 and Oct-2 from cell line-derived nuclear

FIG. 2. Schematic of the binding sites identified within the mouse E애 and the 3⬘ IgH regulatory region (not to scale). Filled symbols are the binding sites that have been characterized biochemically or functionally, and open symbols are the sites thought to bind trans-acting factors based on sequence homology. IR denotes the inverted repeats surrounding the core enhancer in the rat and the mouse HS1,2. Note that the rat HS1,2 lacks the higher-affinity BSAP b site and is in an inverted orientation to its mouse homolog. A, B, and C denote the domains used in some functional studies. MAR denotes matrix attachment regions surrounding E애. In HS3a, the MARE (Maf recognition elements) encompasses an AP-1 site and, in HS3b, it encompasses both an AP-1 and an E2 site.

324

KHAMLICHI et al.

extracts or from in vitro translations. But no cooperative binding of Oct factors was detected (Matthias and Baltimore, 1993). Mouse or human Band plasma-cell lines transfected with HS3a constructs were treated with LPS, PMA, IL-4, or TGF-웁 either singly or in combination but none of these treatments could significantly increase HS3a activity (Matthias and Baltimore, 1993). This would imply that HS3a activity might not be modulated by extracellular signals when it is considered alone and leaves open the question of how its activity is triggered at late B cell development stages. The rat HS1,2 was dissected into three domains (named A, B, and C, see Fig. 2) which were inserted upstream of the human 웁-globin gene in transient transfection assays. Only domains A and B were lymphoidspecific. Domain A alone yielded 앑60% of the enhancer activity, domain B 앑30%, and domain C 앑25%. As for the whole enhancer, deletion of OCT site in domain A did not abolish transcription activity, suggesting the existence of other motifs within this domain. Domain B contained three motifs, 애1, 애A, and an inverted 애B-like motif. Removal of 애1 completely abolished transcriptional activity of domain B, whereas deletion of 애A and 애B-like motifs has less dramatic effects. Domain C contains a 애E3 motif and was not lymphoid-specific. However, when combined with domains A and B, lymphoid specificity was recovered. Both lymphoid and nonlymphoid nuclear extracts could bind to the 애B-like motif. It was suggested that 애A and 애B-like motifs could bind proteins from Ets family whereas 애E3 could bind a basic helix–loop–helix protein (bHLH). Thus, the full enhancer activity would be achieved by both ubiquitous and lymphoidspecific trans-acting factors, the lymphoid-restricted activity being mediated by factors binding to OCT, 애A, and 애B-like motifs (Grant et al., 1992). Another motif, 애E5, was identified downstream of 애E1 within the mouse HS1,2 (Fig. 2). Mutation of the E5 site decreased enhancer activity. When multimerized upstream of a reporter gene, E5 drove transcription in a plasmacytoma cell line. E5 site was shown to bind E12 and E47, two members of the bHLH family. Activity of bHLH family members is regulated by dominant negative regulatory proteins of the Id family (reviewed in Kadesh, 1992). One member of Id family, Id3, is expressed in pre-B and B cell lines but not in plasmacytoma cell lines (Meyer et al., 1995). By using a GST–Id3 fusion protein and in vitro translation products, it was shown that Id3 could physically interact with E12 and E47. Cotransfection, in a plasmacytoma cell line, of a 웁-globin reporter gene driven by HS1,2 with an Id3 expression vector resulted in a fivefold decrease in the enhancer activity. It was proposed that the inhibitory effect of Id3 on HS1,2 activity would be mediated by abrogation of the binding of the E12/ E47 complex to the E5 site (Meyer et al., 1995).

THE 3⬘ IgH REGULATORY REGION

325

NF– ␬B (Michaelson et al., 1996; Linderson et al., 1997) and NFE (for nuclear factor Ets-like, the mouse homolog to the rat 애B site) (Linderson et al., 1997) sites were identified within domain B of HS1,2. Mutation of either of these sites led to a significant decrease in reporter gene transient transcription in plasmacytoma cell lines (Michaelson et al., 1996; Linderson et al., 1997), whereas in B cell lines, mutation of ␬B site led to a substantial increase in HS1,2 activity, which suggests that the ␬B site acts as an activator at the plasma cell stage but as a repressor at the B cell stage (Michaelson et al., 1996). Multimerized NF– ␬B site could drive some reporter gene activity in a plasmacytoma cell line, and addition of an NFE site to a single-copy NF– ␬B site strongly enhanced transcription (Linderson et al., 1997). In the latter configuration, mutation of either site abrogated promoter activity, which suggests some kind of synergy between the cognate DNA-binding proteins (Linderson et al., 1997). EMSA and supershift experiments against Rel-family proteins (reviewed in Ghosh et al., 1998) allowed the identification of p50, p65, and c-Rel factors as components of the complex that binds the NF– ␬B site (Michaelson et al., 1996; Linderson et al., 1997). Complementation experiments in COS cells and a plasmacytoma cell line suggested that whereas the NF– ␬B site is predominantly bound by p50 and c-Rel, NFE-binding factors would rather have a facilitating role in p50 and c-Rel-dependent activation (Linderson et al., 1997). The most extensively studied trans-acting factor binding to sites within the 3⬘ IgH regulatory region is the B cell-specific activator protein (BSAP). BSAP is encoded by the Pax-5 gene and belongs to a family of transcription factors containing a paired DNA-binding domain. Pax-5 is expressed in B lymphocytes, in the developing central nervous system, and in adult testis (Adams et al., 1992). Mice lacking Pax-5 usually die within 3 weeks, have an altered posterior midbrain morphogenesis, and display a complete block at an early stage of B cell development (Urba´nek et al., 1994). Further analysis of Pax-5 mutant mice indicated different functions in fetal liver and adult bone marrow and a dramatic reduction in V-to-DJ recombination at the IgH locus (Nutt et al., 1996). BSAP may function as a transcriptional activator or repressor, depending on the context: it positively regulates mb-1, CD19, N-myc, and LEF-1 genes but efficiently represses immunoglobulin J chain and cell-surface protein PD-I genes (Fitzsimmons et al., 1996; Rinkenberger et al., 1996; Nutt et al., 1998). Within the IgH locus, BSAP was shown to bind multiple sites in between CH genes (upstream and/or within several switch regions) (reviewed in Busslinger and Urba´nek, 1995; Michaelson et al., 1996; Stavnezer, 1996). The first segment within the 3⬘ IgH regulatory region shown to bind BSAP was located upstream of the mouse HS1,2 within a cluster of repeated

326

KHAMLICHI et al.

sequences, however, its precise function is unknown (Liao et al., 1992). Insight into the function of BSAP came from the identification of two sites within the mouse HS1,2 with degenerate sequences and with different affinities for BSAP: the higher-affinity b site and the lower-affinity a site (Fig. 2). Interestingly, this binding activity was B cell-specific and was detected in pro-B, pre-B, and B cell lines but not in plasmacytoma cell lines. More importantly, mutational analysis of BSAP-binding sites, transient transcription of reporter constructs, downregulation of BSAP expression by antisense oligonucleotides, and overexpression of a BSAP cDNA conclusively showed that BSAP acted as a repressor in B- but not in plasmacytoma cell lines (Singh and Birshtein, 1993; Neurath et al., 1994). Identification of additional transcription factor-binding sites within HS1,2 revealed a more complex role of BSAP. In vivo analysis revealed an 움P footprint between the BSAP sites (Fig. 2). This site was occupied in the plasma-cell lines but not in the B cell line tested, although the nuclear factor NF-움P, an Ets-like factor, was detected in both B and plasma-cell lines. Transient transcription of reporter constructs using either a wild-type or a mutated 움P site showed that inhibition of NF–움P binding substantially decreased transcription activity in a plasmacytoma cell line. Moreover, in vivo blocking of BSAP-binding site b with triple-helix forming oligonucleotides in a B cell line led to a significant binding of NF–움P to the 움P site and an increase in transcriptional activity of reporter constructs. By using cell lines expressing different isotypes, it was shown that binding of NF–움P to its site resulted in an increase in the synthesis of 웂2b, 웂3, and 움 mRNAs but had less effect on that of 애 (Neurath et al., 1995). The most likely explanation is that a competition exists between BSAP and NF–움P: up to the B cell stage, binding of BSAP prevents association of NF–움P to its site, thus repressing HS1,2 activity. At later stages of B cell development, BSAP is downregulated and NF–움P binds to its site to exert its positive regulatory action on HS1,2 (Neurath et al., 1995). Surprisingly, while full repression was achieved when both sites were occupied, binding of BSAP to the isolated higher-affinity b site activated transient transcription of a reporter gene in a B cell line. In a plasma-cell line, the wild-type and a mutated version of the b site displayed decreased activity (Singh and Birshtein, 1996), which could be overcome by cotransfection of the reporter gene with a Pax-5 expression vector (Andersson et al., 1996). These findings suggest that the repressive function of BSAP could be ascribed to other transcription factor-binding sites. Three such motifs were identified: OCT, ␬B, and a G-rich motif. Mutation of either OCT, ␬B, or the G-rich motif resulted in an increase of transcriptional activity in B cell lines. Combinations of mutations within the different sites concluded to the requirement of BSAP for the repressive function of

THE 3⬘ IgH REGULATORY REGION

327

OCT, ␬B, and the G-rich motifs in the B cell lines. EMSA, competition experiments, and the use of specific anti-Oct-1 and anti-Oct-2 antisera showed that OCT motif was bound by both Oct-1 and Oct-2 factors. BSAP was shown to specifically interact with glutathione S-transferase–POU domain fusion proteins of Oct-1 and Oct-2 (Singh and Birshtein, 1996; Michaelson et al., 1996). Thus, it appears that BSAP plays a pivotal role in the repression of HS1,2. At the B cell stage, BSAP would hinder the activating effect of the other binding proteins, ensuring a concerted repression of HS1,2. After B cell activation and differentiation to plasma-cell stage, BSAP is downmodulated and the other binding factors could then exert their positive regulatory effect, probably by interacting with their coactivators (Singh and Birshtein, 1996). In contrast to HS1,2 and HS3, HS4 is thought to be active throughout B cell differentiation stages. Multiple BSAP-binding sites, two OCT sites, and one ␬B site have been identified within HS4 (Fig. 2). The ␬B site was bound by an NF– ␬B-like complex in the pre-B, B, and plasma-cell lines used. OCT-binding activity on HS4 was detected in pre-B and B but not in plasma-cell nuclear extracts. In contrast to HS1,2, competition experiments using anti-Oct-1 and anti-Oct-2 antibodies attributed the binding activity to Oct-1 alone. BSAP-binding complexes were detected in preB and B cell-derived nuclear extracts but not in plasma cell-derived nuclear extracts. Mutation of the ␬B site resulted in a dramatic decrease in transient transcription of a reporter gene in a plasmacytoma cell line, which indicates an important role for ␬B motif in the activation of HS4 (incidentally, this could explain why HS4 is inactive in J558L plasmacytoma cell line, which lacks constitutive NF– ␬B binding activity). In transient transfection of B cell lines, mutation of either BSAP, OCT, or ␬B sites resulted in a dramatic reduction in enhancer activity. Whereas the Oct and ␬B site mutations led to a substantial decrease in HS4 activity, in a pre-B cell line, BSAP site mutation resulted in a twofold increase. Therefore, it appears that Oct-1- and ␬B-binding activities positively regulate HS4 activity, whereas BSAP is a repressor (Michaelson et al., 1996). Thus, in this model based on in vitro studies, the same set of factors seems to have different functions, depending on the structure of the enhancer and the B cell differentiation stage. At the pre-B stage, the same combination of trans-acting factors presumably activates HS4 but represses HS1,2. At the plasma-cell stage, activation of both enhancers is the outcome. BSAPbinding activity would be crucial in all these events. Whether this BSAPmediated regulation functions in a concentration-dependent manner as has been shown for other genes (Rinkenberger et al., 1996; Nutt et al., 1998; Wallin et al., 1998) is unknown.

328

KHAMLICHI et al.

What do in vivo studies of BSAP-mediated repression of HS1,2 have to say about the above models? Cross-linking of OX40L, a member of the TNF/NGF receptor family (reviewed by Armitage, 1994), on CD40Lstimulated splenic B cells led to a 60–80% decrease in BSAP levels, the reduction being detected at both the protein and the messenger levels. However, it was difficult to ascertain whether such an effect was direct (the maximum effect was obtained after 6–7 days). In vivo footprinting experiments showed a loss of the footprint at BSAP b site and the appearance of a footprint at the 움P site, with an occupancy pattern similar to that observed within HS1,2 in plasma cells (Stu¨ber et al., 1995). However, this variation in BSAP levels does not seem to be a constant finding, which could relate to the different signaling pathways triggered by the extracellular stimuli used. When resting splenic cells from HS1,2dependent 웁-globin transgenic mice were stimulated with LPS or CD4OL, a strong induction of activity of the transgene was observed whereas Pax5 expression levels remained virtually the same. A similar result was obtained with a B cell line transfected with the same reporter gene after cross-linking of surface IgM. The most likely explanation is that triggering of HS1,2 by these external signals is not blocked by BSAP (Andersson et al., 1996). On the other hand, the rat HS1,2 lacks the higher-affinity b site (Fig. 2) although the remaining site was occupied in both splenic B cells and LPS-stimulated B cells (Andersson et al., 1996). These data suggest that BSAP-mediated repression of HS1,2 activity is probably more complicated than what was anticipated from in vitro studies. Note that the Pax-5 messenger is alternatively spliced into four different isoforms. One isoform (Pax-5b) remains expressed through terminal B cell differentiation, whereas the other isoforms are downregulated during this stage. However, the physiological consequences of the differential expression of Pax-5 isoforms during B cell development have not been elucidated yet (Zwollo et al., 1997). It would have been of great interest to evaluate the physiological significance of the above models in vivo in Pax-5 mutant mice. However, the complete block of B cell differentiation at an early stage prevented the study of the late processes such as antigen-induced B cell activation, class switching, or plasma cell differentiation. Footprinting experiments allowed identification of a novel motif: ETS–AP-1 (Fig. 2), which was protected following stimulation of splenic B cells with LPS and/or TPA. Trimerization of this motif upstream a reporter gene stably transfected in a surface IgM-expressing cell line conferred a strong expression following cross-linking of surface IgM or treatment with TPA. EMSA, in vitro translation, and antibodies against specific factors allowed identification of a surface Ig-inducible complex NFAB (nuclear factors of activated B cells) that could bind to the ETS–AP-1

THE 3⬘ IgH REGULATORY REGION

329

motif. Various factors appear to form this complex, including Elf-1 protein (an Ets-related factor) and two AP-1 factors, c-Fos, and Jun-B (Grant et al., 1995). Further analysis of the signaling pathways leading to transactivation of HS1,2 identified a CD40 pathway that could account for recruitment of the NFAB complex to the ETS–AP-1 motif. In contrast to IgM, CD40 signaling appears to recruit Elf-1 and JunB but not c-Fos (Grant et al., 1996). Thus, distinct signals which induce B cell activation and proliferation through different intracellular signaling pathways may regulate HS1,2 activity via recruitment of different complexes of trans-acting factors. Putative Maf recognition elements (MAREs) have been identified within HS3a and HS3b (recall that they are 97% identical but in inverted orientations). Maf family proteins are important regulators of cell differentiation, they all possess a conserved basic-region leucine zipper (bZip) domain that mediates protein–protein interactions and DNA binding. Some members such as MafF, MafG, and MafK (referred to as the small Maf family proteins) lack the N-terminal canonical transactivation domain (found in large Maf family proteins) and are essentially composed of a conserved bZip domain. Dimeric combinations of Maf family proteins bind in vitro to the MARE motif (reviewed in Blank and Andrews, 1997). Several proteins have been shown to form heterodimers with the small Maf proteins and bind to the MARE in vitro. Among these transcription factors, Bach2 has a restricted expression to the brain and B cells. Within B cells, Bach2 is expressed during the earlier stages of B cell differentiation. Its expression progressively decreases during B cell maturation and seems to be switched off at the plasma cell stage (Muto et al., 1998). Accordingly, specific binding of Bach2-binding complex to the MARE was found with pro-B and mature B cell-derived nuclear extracts but not with plasmacytoma cell line-derived nuclear extracts. Cotransfection of MafK and Bach2 expression vectors drastically repressed HS123b4-driven reporter gene activity in a plasmacytoma cell line. This synergistic repression suggests that MafK, or another small Maf protein, and Bach2 form a heterodimer that represses gene activity through HS123b4. An interpretation of these findings was that the 3⬘ IgH regulatory region would be repressed by Bach2 in undifferentiated B cells, thus preventing strong transcription of the IgH locus. Downregulation of Bach2 at the plasma cell stage would abrogate Bach2mediated repression by a mechanism that is reminiscent of BSAP-mediated repression of the 3⬘ IgH regulatory region (Muto et al., 1998). Thus, with hindsight (see Max, 1993; Ernst and Smale, 1995, for the case of E애), it comes as no surprise that regulation of the 3⬘ IgH regulatory region activity involves both ubiquitous and lymphoid-specific factors that may act as positive or negative regulators. Identification of new trans-

330

KHAMLICHI et al.

acting factors is likely to reveal a more complex regulation than inferred from the cited studies. Still, the regulatory program would probably be specified by an interplay between positive and negative regulators. How such an interplay is tuned during B cell development and how it is triggered by extracellular signals remain major challenges for the near future. VI. In a Search for a Function

What could be the role of the 3⬘ IgH regulatory region? What does it really control? How could it communicate with regulatory elements located far upstream in the IgH locus? And if such communication does occur, how does it affect physiological processes such as V(D)J recombination, transcription, hypermutation, or class switching? In this section, we review work that has addressed these questions and the models that account for such processes. A. REPLICATION The suggestion that the intronic enhancer E애 was associated with a putative origin of replication (Ariizumi et al., 1993) led to a search for a similar role for the 3⬘ IgH regulatory region. It was suggested that the IgH locus was part of a single replicon and that replication was likely to initiate downstream of C움 gene towards the upstream CH genes. In non-B cells, C움 seems to replicate earlier whereas upstream CH genes replicate at progressively later intervals. In contrast, all CH genes seem to replicate early in S phase in B cells. Some data on replication timing in a murine B cell line suggested the involvement of sequences downstream of the C움 gene in regulating early replication of the IgH locus (Brown et al., 1987; Ermakova et al., 1999). However, the finding that early replication of the IgH locus was maintained in the mouse cell line LP1–2 lacking the 3⬘ IgH regulatory region excludes a role for this region in the control of replication (Michaelson et al., 1997). Thus, there is currently no demonstration that the 3⬘ IgH regulatory region plays any role in IgH replication. The cloning of large segments further downstream of the 3⬘ IgH regulatory region (Michaelson et al., 1997; Ermakova et al., 1999) may help clarify this issue. B. HYPERMUTATION By comparison with transgenic models in which the ␬ light chain intronic and 3⬘ enhancers were suggested to play a critical role in somatic hypermutation (Betz et al., 1994; but see Van Der Stoep et al., 1998), it was tempting to speculate that a similar association between E애 and 3⬘ enhancers would have the same effect on a heavy chain (HC) transgene. However, contradictory results were obtained. Association of E애 and HS1,2 in an HC transgene

THE 3⬘ IgH REGULATORY REGION

331

could not sustain hypermutation in several lines of transgenic mice despite an efficient transcription of the transgene (Tumas-Brundage et al., 1997). The same apparently holds true in transfected cultured B cells (Lin et al., 1998). In contrast, we produced transgenic mice in which a human ␬ light chain gene was under the control of a combination of E애–PVH with the 3⬘ regulatory palindrome (HS3a123b) inserted downstream of the transgene. In preliminary results, sequence analysis in one line revealed a mutation pattern comparable to the endogenous locus (Le Morvan and M. Cogne´, unpublished data). This suggests that cooperation between the 3⬘ IgH regulatory region and E애 may be sufficient to ensure full hypermutation. But additional analysis is needed to rigorously demonstrate the involvement of the LCR in hypermutation. C. GERMLINE TRANSCRIPTION AND Ig CLASS SWITCHING Excellent reviews have been devoted to the many parameters involved in Ig class switch recombination (CSR) (cytokines, I promoters, switch sequences, protein-binding sites, etc.) (Coffman et al., 1993; Zhang et al., 1995; Stavnezer, 1996). Therefore, these components of the Ig class switching process will not be reviewed here. In this section, we review knockout experiments on mice as well as the models proposed to account for the role of the 3⬘ IgH regulatory region in CSR and transcription. Chimeric mice were generated by replacing HS1,2 with a neomycin resistance gene driven by the phosphoglycerate kinase (pgk-neoR). With regard to VDJ assembly, rearrangements were detected at similar rates on both JH loci in heterozygous mutant mice, and B cells were found to express sIgM, which indicates that the mutation does not affect VDJ rearrangement. In contrast, the homozygous mutation suppressed the ability of in vitro cultured splenic B cells to induce germline transcription of CH 웂3, 웂2b, 웂2a, and ␧, and in agreement with the accessibility model (reviewed in Sleckman et al., 1996), CSR to IgG3, IgG2b, IgG2a, and IgE (and to some extent IgA) following LPS or LPS⫹IL-4 treatment. However, whereas a deficiency in serum levels was found for IgG3 and IgG2a, normal levels of IgG2b and IgA were detected in homozygous mice. In vitro as well as in vivo, normal levels of IgM and IgG1 were found (Cogne´ et al., 1994). At this stage, it may be of interest to compare the isotype deficiency in HS1,2 knockout mice with that of mice deficient in some members of the NF– ␬B family (reviewed in Ghosh et al., 1998). In p50⫺/⫺ mice, a substantial decrease was observed for IgE, IgG1, and IgA serum levels (Sha et al., 1995). Upon appropriate stimulation in vitro, p50⫺/⫺ resting B cells underwent substantial switching to IgG1 but markedly less to IgG3, IgE, or IgA. Interestingly, the defects in germline transcription mirrored those in CSR, except for IgA where normal levels of CH 움 germline transcript

332

KHAMLICHI et al.

were expressed by p50⫺/⫺ B cells (Snapper et al., 1996). Whereas B cells from RelB knockout mice underwent normal class switching (Snapper et al., 1996), mice deficient in c-Rel had a severe deficiency in IgG1 and IgG2a (Ko¨ntgen et al., 1995). In vitro, B cells from mice deficient in the C-terminal, transactivation domain of the c-Rel protein failed to switch to IgG3, to IgG1, or to IgE in response to appropriate stimuli. Here again, the failure to switch to IgE is associated with normal levels of CH ␧ germline transcripts (Zelazowski et al., 1997; see also Snapper et al., 1997, for a critical assessement of the accessibility model). Targeted disruption of RelA (p65) leads to a decrease in serum levels of IgG1 and IgA (Doi et al., 1997). In vitro, RelA (p65)-deficient B cells displayed a defect in switching to IgG3, which was associated with a decrease in CH 웂3 germline transcripts (Horwitz et al., 1999). Thus, deficiency in each member of the NF– ␬B family appears to be associated with a specific pattern of Ig deficiency and selective defects in germline transcription. As reported in Section IV, NF– ␬B sites have been identified within HS1,2 and HS4. One might then speculate about a potential link between NF– ␬B family members’ deficiency and an impaired function of the 3⬘ IgH regulatory region leading to a defect in germline transcription and/or CSR. However, a direct link has not been demonstrated. Rather, the NF– ␬B deficiency could down-regulate cytokine or other accessory molecule genes whose products control CSR to specific isotypes. Another question concerning HS1,2 mutation is how could it affect CSR? It was suggested that the primary effect of the mutation would be a decreased accessibility of the affected CH genes, perhaps through inhibition of germline transcription, changes in chromatin structure of affected CH genes or both (Cogne´ et al., 1994). It was also proposed that the replacement mutation could block the activity of a long-range regulatory region that regulates transcriptional activation/CSR of certain CH genes. Such a regulatory region may be associated with the putative LCR described in the 3⬘ region (see Section III). Therefore, insertion of pgk-neoR cassette may compete for such a long-range regulatory effect (Cogne´ et al., 1994). In line with this interpretation, an enhancer shift model has been proposed (Arulampalam et al., 1997). With regard to the germline transcription, which, in the accessibility model, is a prerequiste to isotype switching, the model postulates that individual enhancers within the 3⬘ IgH regulatory region would selectively interact with specific I promoters. Alternatively, the four enhancers may act as a unit on I promoters but would shift from an accessible I promoter to another following appropriate stimuli (Arulampalam et al., 1997). Bringing such remote regions into proximity could be faciliated by looped DNA domains that may form due

THE 3⬘ IgH REGULATORY REGION

333

to the matrix attachment regions (MARs) lying in some introns within the IgH locus (Fig. 1) (Cockerill, 1990), although no MAR has so far been identified within the 3⬘ IgH regulatory region. There are many aspects of gene expression control that look similar between the 웁-globin LCR (reviewed in Martin et al., 1996) and the 3⬘ IgH regulatory region. In fact, the 웁-globin LCR paradigm was underlying the assumptions about the functioning of the 3⬘ IgH regulatory region. In many respects, this proved to be true: whereas targeted deletion of 5⬘ HS2 or HS3 of the murine 웁-globin LCR by insertion of pgk-neoR had severe effects, virtually no abnormality was noted when the selectable marker was eliminated (Fiering et al., 1995; Hug et al., 1996). Therefore, it was interesting to see whether the effect of a clean targeted delection of components of the 3⬘ IgH regulatory region would mirror that of components of the 웁-globin LCR. Mice were generated with targeted replacements or delections of either HS1,2 or HS3a: replacement mutations (with a pgk-neoR) of either HS1,2 or HS3a led to essentially the same phenotype and CSR alterations, although HS1,2 and HS3a are separated by 12 kb (Cogne´ et al., 1994; Manis et al., 1998). Deletion of pgk-neoR restored normal CSR. Therefore, neither HS1,2 nor HS3a is required for germline transcription (and subsequent CSR) or IgH expression, which might simply result from redundancy within the 3⬘ IgH regulatory region (Manis et al., 1998). Alternatively, in addition to this elusive role of the 3⬘ IgH regulatory region, cis-acting elements at individual switch regions may be the master control elements of CSR (Gu et al., 1993; Sakai et al., 1999). Insights into the molecular basis for the polarized effect of the 3⬘ IgH regulatory region may come from studies of knockout mice with the pgkneoR cassette inserted into their I웂2b promoter or their C␧ exons (Seidl et al., 1999). In the former case, germline transcription of, and CSR to, C웂3 was inhibited, whereas in the latter, germline transcription of and CSR to C웂3, C웂2b, and C웂2a was blocked. In both cases, the notable exceptions were C웂1 and C움, which were slightly affected. These findings may be interpreted as the blockade of a long-range and polarized 3⬘ IgH regulatory region effect: competition with pgk-neoR would thus suppress germline transcription and CSR to upstream but not downstream isotypes (Seidl et al., 1999). Because insertion of the cassette in HS1,2 or HS3a does not affect CSR to C웂1 and C움, such a model would imply that CSR to these isotypes may be regulated independently of that 3⬘ elements (Manis et al., 1998; Seidl et al., 1999; see also Elenich et al., 1996). Strikingly, pgk-neoR was shown to be LPS-inducible but only when inserted in the IgH locus (including the 3⬘ IgH regulatory region) (Manis et al., 1998; Seidl et al., 1998; reviewed in Zhang et al., 1995). Therefore, other control elements lying downstream of HS1,2 may be required for

334

KHAMLICHI et al.

LPS induction of germline transcription. Candidates for such a polarized effect could be HS3b and/or HS4. However, HS3b is unlikely to perform such a function because it is 97% identical to HS3a although in opposite orientation (Chauveau and Cogne´, 1996). Moreover, humans lack HS3b but perform their IgH expression, hypermutation, and CSR well. Therefore, HS4 is the prime candidate, assuming that no other regulatory elements lie downstream of HS4. Alternatively, the whole 3⬘ IgH regulatory region with its redundant elements may carry an LPS inducibility so that each elements would be by itself dispensable. Note, however, that the inhibiting effect on upstream isotypes is not obvious when neoR gene is driven by a combination of polyoma enhancer– thymidine kinase promoter ( Jung et al., 1993; Zhang et al., 1993) or a combination of E애–PVH (Bottaro et al., 1994). On the other hand, the effect on upstream isotypes is not a constant finding even when using the pgk promoter. In knockout mice in which the I움 promoter was replaced with a pgk–hypoxanthine phosphoribosyltransferase cassette, CSR to all C웂 genes was substantially impaired (Qiu et al., 1999). When insertion of a pgk-neoR cassette deleted I움, S움, and the 5⬘ half of C움, there was an increase in serum levels of IgG1 and IgG2b and a decrease in IgG3, whereas serum IgG2a level did not vary (Harriman et al., 1999), but the issue of germline transcription and CSR on splenic B cells was not addressed in the latter study. D. IgH EXPRESSION An insertional targeting was performed on an IgG2a,␬-producing cell line lacking E애. Replacement of HS1,2 with pgk-neoR downstream of the expressed allele completely abolished 웂2a gene transcription. This finding is consistent with the notion that HS1,2 (although not alone) plays an important role in transcription at late B cell developmental stage (Lieberson et al., 1995). However, it is difficult to ascertain whether suppression of 웂2a transcription results from the lack of HS1,2 per se, likely to be more critical in the absence of E애, or from disruption of the putative polarized effect of the 3⬘ IgH regulatory region. In HS1,2 knockout mice that retain E애, transcription levels of 애, 웂l, and 움 genes were found to be normal; only germline transcription of selective CH genes was affected (Cogne´ et al., 1994). On the other hand, the potential competition problems mentioned in the previous section also apply to this knockout. It will be interesting to perform a clean deletion of HS1,2 in this cell line, which would provide a good model to study the cooperation between E애 and HS1,2 on 웂2a gene transcription. Recently, a large deletion (앑30 Kb) including both HS3a and HS1,2 has been characterized in the 70Z/3 pre-B cell line (Saleque et al., 1999)

THE 3⬘ IgH REGULATORY REGION

335

(Fig. 1). Despite the lack of HS3a and HS1,2, no decrease in 애 expression could be detected, which fits with the activity pattern of HS3a and HS1,2 (see Section II). However, upon fusion of 70Z/3 with the NSO myeloma cell line, an up-regulation of 애 expression was noted on the 70Z/3 allele of the (single) fusion hybrid. Therefore, HS3a and HS1,2 seem not to be required for 애 expression at the plasma cell stage. The presence of E애, HS3b, and HS4 seems sufficient to drive high levels of IgH expression at that stage. Interestingly, the 70Z/3–NSO hybridoma spontaneously underwent class switching to C웂1, suggesting that the two enhancers are not essential for CSR to C웂1 (Saleque et al., 1999), as previously noticed in HS3a or HS1,2 knockout mice (Cogne´ et al., 1994; Manis et al., 1998). A suggestive comparison can be drawn between the 애-expressing 70Z/ 3 pre-B cell line and an IgA-producing cell line (LP1-2). The latter naturally performed a deletion of the whole 3⬘ IgH regulatory region while retaining E애 (Fig. 1). In this cell line, transcription level of the 움 gene was sevenfold lower than that of its parental cell line with intact 3⬘ IgH regulatory region. The defect resulted from a lower transcription rate, not from an alteration in processing or polyadenylation of 움 pre-messenger RNA (Gregor and Morrison, 1986). At the minimum, this would suggest that the whole 3⬘ IgH regulatory region plays an important role in transcription regulation of class-switched Ig genes through the cooperation of multiple and somehow redundant enhancers so that a clean deletion of only one or part of the 3⬘ IgH regulatory region will have little effect by itself. E. CHROMATIN STRUCTURE It is now accepted that the chromatin structure is an essential component of the transcriptional regulation machinery. Although the exact molecular mechanisms governing the opening of the chromatin compact structure and the accessibility to promoters and enhancers of trans-acting factors are still debatable, it is clear that changes in chromatin structure near transcriptionally active genes require cooperation between transcription factors, histones, and other cofactors in order to remodel and displace nucleosomes (reviewed in Felsenfeld, 1992; Kadonaga, 1998). One way for the enhancers to counteract the repressive chromatin structure may be by recruiting and/or directing histone acetyltransferase (HAT)-containing molecules (such as transcriptional cofactors) to critical regulatory regions. Indeed, transcriptionally active genes are associated with acetylated core histones (reviewed in Majumder and DePamphilis, 1995). Therefore, it was interesting to look at the effect of the 3⬘ IgH regulatory region on chromatin structure of linked genes. A combination of HS1,2, HS3b, and HS4 (HS123b4) was suggested to act as an LCR (Madisen and Groudine, 1994) (see Section III). This

336

KHAMLICHI et al.

cassette was sufficient to deregulate transcription of linked c-myc genes associated with a shift from P2 to P1 promoter usage (Madisen and Groudine, 1994). By using episomally maintained c-myc genes with both promoters or devoid of P1 promoter in Raji Burkitt lymphoma cell line, it was shown that the P2 to P1 shift was not required for HS123b4 to achieve highlevel expression of linked c-myc genes (Madisen et al., 1998). Surprisingly, analysis of hypersensitivity, nucleosome positioning, and restriction enzyme accessibility around the promoter region did not reveal gross changes in chromatin structure despite the differential expression of the templates. However, chromatin immunoprecipitation assays revealed an increase in histone acetylation over the entire episomal vector bearing HS123b4, not only to the c-myc gene (Madisen et al., 1998). Treatment of the transfectants with an inhibitor of histone deacetylases, leading to general histone acetylation, inhibited the HS123b4-mediated high-level expression of P1 but not that of the P2 promoter. Thus, although increased acetylation does not completely explain the HS123b4-mediated activation of the P1 promoter, it may be one mechanism by which the LCR establishes and maintains a transcriptionally active state along linked genes over entire chromatin domains. To achieve this, one possibility would be the recruitment of HAT-containing cofactors that might induce local histone acetylation, further propagating throughout the chromatin domain under the influence of the LCR (Madisen et al., 1998). Thus, one important aspect of the 3⬘ IgH LCR function would be to recruit HAT activity in order to counteract the repressive effect of chromatin structure that may be maintained by deacetylases. If we assume that the 3⬘ IgH LCR and the 웁-globin LCR have evolved similar strategies to regulate differential gene expression, then the described model would be difficult to reconcile with recent results in which the whole mouse 웁globin LCR was deleted (Epner et al., 1998). In this study, the LCR was deleted on both alleles and the mutated chromosomes were then transferred into K562 erythroleukemia cell line. DNase I sensitivity of the mouse 웁-globin locus, formation of an active 웁-globin locus chromatin structure, and developmental regulation of 웁-globin gene transcription were not significantly affected by deletion of the LCR. However, 웁-like globin RNA levels were reduced to 5–25% of normal. Therefore, the primary function of the 웁-globin LCR may be to maintain normal levels of transcription whereas the other critical aspects, such as remodeling of chromatin structure or developmental regulation of transcription, may eventually rely on other elements. Thus, the 웁-globin LCR would have a contributory rather than a dominant role (Epner et al., 1998). With regard to the 3⬘ IgH regulatory region, a major challenge would be to perform a similar complete knock out in mice. Although the LP1-

THE 3⬘ IgH REGULATORY REGION

337

2 cell line has performed such a large deletion and shows a decreased transcription of the 움 gene, the existence of developmental regulatory programs that are switched off in LP1-2 cannot be excluded. VII. The Case of Humanized Mice

With regard to hypermutation and CSR, the data from humanized mice are rather puzzling. In mice transgenic for human miniloci containing only a small number of V gene segments and devoid of any 3⬘ enhancer (Wagner et al., 1994, 1996) or bearing the rat HS1,2 downstream of the translocus (Lonberg et al., 1994; Taylor et al., 1994), hypermutation and CSR occurred (though at lower frequency) within the expressed genes. Interestingly, the use of a larger number of human V gene segments (made possible by the use of yeast artificial chromosomes) in a translocus bearing HS1,2 drastically improved CSR and hypermutation rate, allowing generation and efficient secretion of high affinity human antibodies in transgenic mice (Fishwild et al., 1996; Mendez et al., 1997; Green and Jakobovits, 1998). These observations suggest that other elements upstream of the 3⬘ IgH regulatory region are sufficient for hypermutation and CSR to occur. Note however that the studies with human loci were performed in mice rendered deficient in endogenous Ig production, which may impose different contraints on positively selected B cells than in the situation where the transgene has to compete with the endogenous locus; a delay in the processes of V(D)J recombination, CSR, or somatic hypermutation would be difficult to appreciate without a comparison and/or competition with a normal locus. Analysis of B cells accumulating in the periphery after completion of their maturation does not provide such a dynamic picture. VIII. Conclusion

Over the past years, intensive effort has been devoted to delineating the function of the 3⬘ IgH regulatory region. Much has been learned about the structure of this region, the sites of DNA-binding factors, and the synergy between its components. But in vivo, a definitive and unambiguous function has not yet been elucidated. There is still much to learn: throughout this review, we have assumed that the 3⬘ IgH regulatory region encompasses sequences from HS3a to HS4 with particular emphasis on the HSs. This obviously does not exclude the possibility that other regulatory elements may lie outside or even within this region. Identification and characterization of such elements will add to the complexity of the model of this region but may help clarify some aspects of its function. Whatever the process under study, which may somehow be controlled by the 3⬘

338

KHAMLICHI et al.

IgH regulatory region, a persistent problem will be the redundancy of its enhancers. It will be interesting to delete the whole 3⬘ IgH regulatory region in mice in order to evaluate its contribution to the different processes thought to be under its control. Whereas the comparison between the 웁globin LCR and the 3⬘ IgH regulatory region seems operational, it may be conceptually misleading. After all, the IgH locus is not the 웁-globin locus in regard to neither the tissue specificity nor the processes involved (hypermutation, CSR, etc). Another challenge will be to identify the signals that trigger the 3⬘ enhancers’ activity during B cell development and the mechanisms underlying the interplay between positive and negative regulation. NOTE ADDED IN PROOF A recent study by H. Tang and P. A. Sharp demonstrated a specific role for Oct-2/OCA-B but not Oct-1 in the regulation of HS1,2, and showed that activation of HS1,2 was dependent upon phosphorylation of Oct-2 by protein kinase C in activated B cells (Immunity 11, 517–526; 1999).

ACKNOWLEDGMENTS We are indebted to Jean-Claude Weill and Fred Alt for critical reading of the manuscript and advice and to Yves Denizot for helpful comments. Work in our laboratory is funded by Association pour la Recherche sur le Cancer, Ligue Nationale contre le Cancer and Conseil Re´gional du Limousin.

REFERENCES Adams, B., Do¨rfler, P., Aguzzi, A., Kozmik, Z., Urba´nek, P., Maurer-Fogy, I., and Busslinger, M. (1992). Pax-5 encodes the transcription factor BSAP and is expressed in B lymphocytes, the developing CNS, and adult testis, Genes Dev. 6, 1589–1607. Aguilera, R. J., Hope, T. J., and Sakano, H. (1985). Characterization of immunoglobulin enhancer deletions in murine plasmacytomas. EMBO J. 4, 3689–3693. Andersson, T., Furebring, C., Borrebaeck, C. A. K., and Pettersson, S. (1999). Temporal expression of a VH promoter–C애 transgene linked to the IgH HS1,2 enhancer. Mol. Immunol. 36, 19–29. Andersson, T., Neurath, M. F., Grant, P. A., and Pettersson, S. (1996). Physiological activation of the IgH 3⬘ enhancer in B lineage cells is not blocked by Pax-5. Eur. J. Immunol. 26, 2499–2507. Ariizumi, K., Wang, Z., and Tucker, P. W. (1993). Immunoglobulin heavy chain enhancer is located near or in an initiation zone of chromosomal DNA replication. Proc. Natl. Acad. Sci. USA 90, 3695–3699. Armitage, R. J. (1994). Tumor necrosis factor receptor superfamily members and their ligands. Curr. Opin. Immunol. 6, 407–413. Arulampalam, M., Grant, P. A., Samuelsson, A., Lendahl, U., and Pettersson, S. (1994). Lipopolysaccharide-dependent transactivation of the temporally regulated immunoglobulin heavy chain 3⬘ enhancer. Eur. J. Immunol. 24, 1671–1677. Arulampalam, V., Eckhardt, L., and Pettersson, S. (1997). The enhancer shift: A model to explain the development control of IgH gene expression in B-lineage cells. Immunol. Today 18, 549–554.

THE 3⬘ IgH REGULATORY REGION

339

Banerji, J., Olson, L., and Schaffner, W. (1983). A lymphocyte-specific cellular enhancer is located downstream of the joining region in immunoglobulin heavy-chain genes. Cell 33, 729–740. Betz, A. G., Milstein, C., Gonza´lez-Ferna´ndez, A., Pannell, R., Larson, T., and Neuberger, M. S. (1994). Elements regulating somatic hypermutation of an immunoglobulin ␬ gene: Critical role for the intron enhancer/matrix attachment region. Cell 77, 239–248. Blank, V., and Andrews, N. C. (1997). The Maf transcription factors: Regulators of differentiation. Trends Biochem. Sci. 22, 437–441. Bottaro, A., Lansford, R., Xu, L., Zhang, J., Rothman, P., and Alt, F. W. (1994). S region transcription per se promotes basal IgE class switch recombination but additional factors regulate the efficiency of the process. EMBO J. 13, 665–674. Brown, E. H., Iqbal, M. A., Stuart, S., Hatton, K. S., Valinsky, J., and Schildkraut, C. L. (1987). Rate of replication of the murine immunoglobulin heavy-chain locus: Evidence that the region is part of a single replicon. Mol. Cell. Biol. 7, 450–457. Busslinger, M., and Urba´nek, P. (1995). The role of BSAP (Pax-5) in B-cell development. Curr. Opin. Genet. Dev. 5, 595–601. Chauveau, C., and Cogne´, M. (1996). Palindromic structure of the IgH 3⬘ locus control region. Nat. Genet. 14, 15–16. Chauveau, C., Jansson, E. A., Mu¨ller, S., Cogne´, M., and Pettersson, S. (1999). Immunoglobulin heavy chain 3⬘HS1-4 directs correct spatial position independent expression of a linked transgene to B lineage cells. J. Immunol. 163, 4637–4641. Chauveau, C., Pinaud, E., and Cogne´, M. (1998). Synergies between regulatory elements of the immunoglobulin heavy chain locus and its palindromic 3⬘ locus control region. Eur. J. Immunol. 28, 3048–3056. Chen, C., and Birshtein, B. K. (1995). A region of 20 bp repeats lies 3⬘ of human Ig C움1 and C움2 genes. Int. Immunol. 8, 115–122. Chen, C., and Birshtein, B. K. (1997). Virtually identical enhancers containing a segment of homology to murine 3⬘ IgH-E(hs1,2) lie downstream of human Ig C움1 C움2 genes. J. Immunol. 159, 1310–1318. Cockerill, P. N. (1990). Nuclear matrix attachment occurs in several regions of the IgH locus. Nucleic Acids Res. 18, 2643–2648. Coffman, R. L., Lebman, D. A., and Rothman, P. B. (1993). Mechanism and regulation of immunoglobulin isotype switching. Adv. Immunol. 54, 229–270. Cogne´, M., Lansford, R., Bottaro, A., Zhang, J., Gorman, J., Young, F., Cheng, H.-L., and Alt, F. W. (1994). A class switch control region at the 3⬘ end of the immunoglobulin heavy chain locus. Cell 77, 737–747. Dariavach, P., Williams, G. T., Campbell, K., Pettersson, S., and Neuberger, M. S. (1991). The mouse IgH 3⬘-enhancer. Eur. J. Immunol. 21, 1499–1504. Doi, T. S., Takahashi, T., Taguchi, O., Azuma, T., and Obata, Y. (1997). NF-␬B RelAdeficient lymphocytes: Normal development of T cells and B cells, impaired production of IgA and IgGI, and reduced proliferative responses. J. Exp. Med. 185, 953–961. Eckhardt, L. A., and Birshtein, B. K. (1985). Immunoglobulin class-switch events occuring in a single myeloma cell line. Mol. Cell. Biol. 5, 856–868. Elenich, L. A., Ford, C. S., and Dunnick, W. A. (1996). The 웂l heavy chain gene includes all of the cis-acting elements necessary for expression of properly regulated germ-line transcripts. J. Immunol. 157, 176–182. Epner, E., Reik, A., Cimbora, D., Telling, A., Bender, M. A., Fiering, S., Enver, T., Martin, D. L. K., Kennedy, M., Keller, G., and Groudine, M. (1998). The 웁-globin LCR is not necessary for an open chromatin structure or developmentally regulated transcription of the native mouse 웁-globin locus. Mol. Cell 2, 447–455.

340

KHAMLICHI et al.

Ermakova, O. V., Nguyen, L. H., Little, R. D., Chevillard, C., Riblet, R., Ashouian, N., Birshtein, B. K., and Schildkraut, C. L. (1999). Evidence that a single replication fork proceeds from early to late replicating domains in the IgH locus in a non-B cell line. Mol. Cell 3, 321–330. Ernst P., and Smale S. T. (1995). Combinatorial regulation of transcription II: The immunoglobulin mu heavy chain gene. Immunity 2, 427–438. Felsenfeld, G. (1992). Chromatin as an essential part of the transcriptional mechanism. Nature 355, 219–224. Fiering, S., Epner, E., Robinson, K., Zhuang, Y., Telling, A., Hu, M., Martin, D. I. K., Enver, T., Ley, T. J., and Groudine, M. (1995). Targeted deletion of 5⬘ HS2 of the murine 웁-globin LCR reveals that it is not essential for proper regulation of the 웁-globin locus. Genes Dev. 9, 2203–2213. Fishwild, D. M., O’Donnell, S. L., Bengoechea, T., et al. (1996). High-avidity human IgG␬ monoclonal antibodies from a novel strain of minilocus transgenic mice. Nature Biotech. 14, 845–851. Fitzsimmons, D., Hodsdon, W., Wheat, W., Maira, S-M., Wasylyk, B., and Hagman, J. (1996). Pax-5 (BSAP) recruits Ets proto-oncogene family proteins to form functional ternary complexes on a B-cell-specific promoter. Genes Dev. 10, 2198–2211. Fulton, R., and Van Ness, B. (1994). Selective synergy of immunoglobulin enhancer elements in B-cell development: A characteristic of kappa light chain enhancers, but not heavy chain enhancers. Nucleic Acids Res. 22, 4216–4223. Ghosh, S., May, M. J., and Kopp, E. B. (1998). NF-kappa B and Rel proteins: Evolutionarily conserved mediators of immune responses. Annu. Rev. Immunol. 16, 225–260. Giannini, S. L., Singh, M., Calvo, C.-F., Ding, G. F., and Birshtein, B. K. (1993). DNA regions flanking the mouse Ig 3⬘ 움 enhancer are differentially methylated and DNase I hypersensitive during B cell differentiation. J. Immunol. 150, 1772–1780. Gillies, S. D., Morrison, S. L., Oi, V. T., and Tonegawa, S. (1983). A tissue-specific transcriptional enhancer element is located in the major intron of a rearranged immunoglobulin heavy-chain gene. Cell 33, 717–728. Grant, P. A., Andersson, T., Neurath, M. F., Arulampalam, V., Bauch, A., Mu¨ller, R., Reth, M., and Pettersson, S. (1996). A T cell controlled molecular pathway regulating the IgH locus: CD40-mediated activation of the IgH 3⬘ enhancer. EMBO J. 15, 6691–6700. Grant, P. A., Arulampalam, V., Ahrlund-Richter, L., and Pettersson, S. (1992). Identification of Ets-like lymphoid specific elements within the immunoglobulin heavy chain 3⬘ enhancer. Nucleic Acids Res. 20, 4401–4408. Grant, P. A., Thompson, C. B., and Pettersson, S. (1995). IgM receptor-mediated transactivation of the IgH 3⬘ enhancer couples a novel Elf-1–AP-1 protein complex to the developmental control of enhancer function. EMBO J. 14, 4501–4513. Green, L. L., and Jakobovits, A. (1998). Regulation of B cell development by variable gene complexity in mice reconstituted with human immunoglobulin yeast artificial chromosomes. J. Exp. Med. 188, 483–495. Gregor, P. D., and Morrison, S. L. (1986). Myeloma mutant with a novel 3⬘ flanking region: Loss of normal sequence and insertion of repetitive elements leads to decreased transcription but normal processing of the alpha heavy-chain gene products. Mol. Cell. Biol. 6, 1903–1916. Gu, H., Zou, Y.-R., and Rejewsky, K. (1993). Independent control of immunoglobulin switch recombination at individual switch regions evidenced through Cre-loxP-mediated gene targeting. Cell 73, 1155–1164. Gualandi, G., Frezza, D., Scotto d’Abusco, A., Bianchi, E., Gargano, S., Giorgi, S., Fruscalzo, A., and Calef, E. (1995). Integration of an Epstein–Barr virus episome 3⬘ into the gene encoding 움1 in a lymphoblastoid cell line. Gene 166, 221–226.

THE 3⬘ IgH REGULATORY REGION

341

Harriman, G. R., Bogue, M., Rogers, P., Finegold, M., Pacheco, S., Bradley, A., Zhang, Y., and Mbawuike, I. N. (1999). Targeted deletion of the IgA constant region in mice leads to IgA deficiency with alterations in expression of other Ig isotypes. J. Immunol. 162, 2521–2529. Henderson, A., and Calame, K. (1998). Transcriptional regulation during B cell development. Annu. Rev. Immunol. 16, 163–200. Horwitz, B. H., Zelazowski, P., Shen, Y., Wolcott, K. M., Scott, M. L., Baltimore, D., and Snapper, C. M. (1999). The p65 subunit of NF-␬B is redundant with p50 during B cell proliferative responses, and is required for germline CH transcription and class switching to IgG3. J. Immunol. 162, 1941–1946. Hug, B., Wesselschmidt, R. L., Fiering, S., Bender, M. A., Groudine, M., and Ley, T. J. (1996). Analysis of mice containing a targeted deletion of 웁-globin locus control region 5⬘ hypersensitive site 3. Mol. Cell. Biol. 16, 2906–2912. Jung, S., Rajewsky, K., and Radbruch, A. (1993). Shutdown of class switch recombination by deletion of a switch region control element. Science 259, 984–987. Kadesh, T. (1992). Helix–loop–helix proteins in the regulation of immunoglobulin gene transcription. Immunol. Today 13, 31–36. Kadonaga, J. T. (1998). Eukaryotic transcription: An interlaced network of transcription factors and chromatin-modifying machines. Cell 92, 307–313. Kang, H. K., and Cox, D. W. (1996). Tandem repeats 3⬘ of the IgHA genes in the human immunoglobulin heavy chain cluster. Genomics 35, 189–195. Klein, S., Sablitzky, F., and Radbruch, A. (1984). Deletion of the IgH enhancer does not reduce immunoglobulin heavy chain production of a hybridoma IgD class switch variant. EMBO J. 3, 2473–2476. Ko¨ntgen, F., Grumont, R. J., Strasser, A., Metclaf, D., Li, R., Tarlington, D., and Gerondakis, S. (1995). Mice lacking the c-rel proto-oncogene exhibits defects in lymphocyte proliferation, humoral immunity, and interleukin-2 expression. Genes Dev. 9, 1965–1977. Liao, F., Giannini, S. L., and Birshtein, B. K. (1992). A nuclear DNA-binding protein expressed during early stages of B cell differentiation interacts with diverse segments within and 3⬘ of the IgH chain gene cluster. J. Immunol. 148, 2909–2917. Libermann, T. A., and Baltimore, D. (1991). Transcriptional regulation of immunoglobulin gene expression. In ‘‘The Hormonal Control Regulation of Gene Trancription’’ (P. Cohen and J. G. Foulkes, eds.), pp. 399–421. Elsevier Science Publishers. Lieberson, R., Giannini, S. L., Birstein, B. K., and Eckhardt, L. (1991). An enhancer at the 3⬘ end of the mouse immunoglobulin heavy chain locus. Nucleic Acids Res. 19, 933–937. Lieberson, R., Ong, J., Shi, X., and Eckhardt, L. (1995). Immunoglobulin gene transcription ceases upon deletion of a distant enhancer. EMBO J. 14, 6229–6238. Lin, M. M., Green, N. S., Zhang, W., and Scharff, M. D. (1998). The effects of Emu, 3⬘ alpha (hs 1,2) and 3⬘ kappa enhancers on mutation of an Ig–VDJ–Cgamma2a Ig heavy gene in cultured B cells. Int. Immunol. 10, 1121–1129. Linderson, Y., Cross, D., Neurath, M. F., and Pettersson, S. (1997). NFE, a new transcriptional activator that facilitates p50 and c-Rel-dependent IgH 3⬘ enhancer activity. Eur. J. Immunol. 27, 468–475. Lonberg, N., Taylor, L. D., Harding, F. A., et al. (1994). Antigen-specific human antibodies from mice comprising four distinct genetic modifications. Nature 368, 856–859. Madisen, L., and Groudine, M. (1994). Identification of a locus control region in the immunoglobulin heavy-chain locus that deregulates c-myc expression in plasmacytoma and Burkitt’s lymphoma cells. Genes Dev. 8, 2212–2226. Madisen, L., Krumm, A., Hebbes, T. R., and Groudine, M. (1998). The immunoglobulin heavy chain locus control region increases histone acetylation along linked c-myc genes. Mol. Cell. Biol. 18, 6281–6292.

342

KHAMLICHI et al.

Majumder, S., and DePamphilis, M. L. (1995). A unique role for enhancers is revealed during early mouse development. BioEssays 17, 879–889. Manis, J. P., Van der Step, N., Tian, M., Ferrini, R., Davidson, L., Bottaro, A., and Alt, F. W. (1998). Class switching in B cells lacking 3⬘ immunoglobulin heavy chain enhancers. J. Exp. Med. 188, 1421–1431. Martin, D. I., Fiering, S., and Groudine, M. (1996). Regulation of 웁-globin gene expression: Straightening out the locus. Curr. Opin. Genet. Dev. 6, 488–495. Matthias, P., and Baltimore, D. (1993). The immunoglobulin heavy chain locus contains another B-cell-specific 3⬘ enhancer close to the 움 constant region. Mol. Cell. Biol. 13, 1547–1553. Max, E. E. (1993). Immunoglobulins: Molecular genetics. In ‘‘Fundamental Immunology’’ (3rd ed.) (W. E. Paul, ed.), pp. 315–382. Raven Press, New York, NY. Mendez, M. J., Gren, L. L., Corvalan, J. R. F., et al. (1997). Functional transplant of megabase human immunoglobulin loci recapitulates human antibody response in mice. Nat. Genet. 15, 146–156. Meyer, K. B., Skogberg, M., Margenfeld, C., Ireland, J., and Pettersson, S. (1995). Repression of the immunoglobulin heavy chain 3⬘ enhancer by helix–loop–helix protein Id3 via a functionally important E47/E12 binding site: Implications for developmental control of enhancer function. Eur. J. Immunol. 25, 1770–1777. Michaelson, J. S., Ermakova, O., Birshtein, B. K., Ashouian, N., Chevillard, C., Riblet, R., and Schildkraut, C. L. (1997). Regulation of the replication of the murine immunoglobulin heavy chain gene locus: Evaluation of the role of the 3⬘ regulatory region. Mol. Cell Biol. 17, 6167–6174. Michaelson, J. S., Giannini, S. L., and Birshtein, B. K. (1995). Identification of 3⬘움-hs4, a novel Ig heavy chain enhancer element regulated at multiple stages of B cell differentiation. Nucleic Acids Res. 23, 975–981. Michaelson, J. S., Singh, M., and Birshtein, B. K. (1996). B cell lineage-specific activator protein (BSAP): A player at multiple stages of B cell development. J. Immunol. 156, 2349– 2351. Michaelson, J. S., Singh, M., Snapper, C. M., Sha, W. C., Baltimore, D., and Birshtein, B. K. (1996). Regulation of 3⬘ IgH enhancers by a common set of factors, including ␬Bbinding proteins. J. Immunol. 156, 2828–2839. Mills, F. C., Harindranath, N., Mitchell, M., and Max, E. E. (1997). Enhancer complexes located downstream of both human immunoglobulin C움 genes. J. Exp. Med. 186, 845–858. Mocikat, R., Harloff, C., and Ku¨temeier, G. (1993). The effect of the rat immunoglobulin heavy-chain 3⬘ enhancer is position dependent. Gene 136, 349–353. Mocikat, R., Kardinal, C., and Klobeck, H.-G. (1995). Differential interactions between the immunoglobulin heavy chain 애 intron and 3⬘ enhancer. Eur. J. Immunol. 25, 3195–3198. Muto, A., Hoshino, H., Madisen, L., Yanai, N., Obinata, M., Karasuyama, H., Hayashi, N., Nakauchi, H., Yamamoto, M., Groudine, M., and Igarashi, K. (1998). Identification of Bach2 as a B-cell-specific partner for small Maf proteins that negatively regulate the immunoglobulin heavy chain gene 3⬘ enhancer. EMBO J. 17, 5734–5743. Neuberger, M. S. (1983). Expression and regulation of immunoglobulin heavy chain gene transfected into lymphoid cells. EMBO J. 2, 1373–1378. Neuberger, M. S., and Calabi, F. (1983). Reciprocal chromosome translocation between cmyc and immunoglobulin 웂2b genes. Nature 305, 240–243. Neurath, M. F., Max, E. E., and Strober, W. (1995). Pax5 (BSAP) regulates the murine immunoglobulin 3⬘움 enhancer by suppressing binding of NF-움P, a protein that controls heavy chain transcription. Proc. Natl. Acad. Sci. USA 92, 5336–5340.

THE 3⬘ IgH REGULATORY REGION

343

Neurath, M. F., Strober, W., and Wakatsuki, Y. (1994). The murine Ig 3⬘움 enhancer is a target site with repressor function for the B cell lineage-specific transcription factor BSAP (NF-HB, S움-BP). J. Immunol. 153, 730–742. Nutt, S. L., Morrison, A. M., Do¨rfler, P., Rolink, A., and Busslinger, M. (1998). Identification of BSAP (Pax-5) target genes in early B-cell development by loss- and gain-of-function experiments. EMBO J. 17, 2319–2333. Nutt, S. L., Urba´nek, P., Rolink, A., and Busslinger, M. (1996). Essential functions of Pax5 (BSAP) in pro-B cell development: Difference between fetal and adult B lymphopoiesis and reduced V-to-DJ recombination at the IgH locus. Genes Dev. 11, 476–491. Okada, A., and Alt, F. W. (1995). The variable region gene assembly mechanism. In ‘‘The Immunoglobulin Genes’’ (2nd ed.) (F. W. Alt and T. Honjo, eds), pp. 205–234. Academic Press San Diego, CA. Ong, J., Stevens, S., Roeder, R. G., and Eckhardt, L. A. (1998). 3⬘ IgH enhancer elements shift synergistic interactions during B cell development. J. Immunol. 160, 4896–4903. Ortiz, B. D., Cado, D., Chen, V., Diaz, P. W., and Winoto, A. (1997). Adjacent DNA elements dominantly restrict the ubiquitous activity of a novel chromatinpening region to specific tissues. EMBO J. 16, 5037–5045. Pettersson, S., Cook, G. P., Bruggemann, M., Williams, G. T., and Neuberger, M. S. (1990). A second B cell-specific enhancer 3⬘ of the immunoglobulin heavy-chain locus. Nature 344, 165–168. Pinaud, E., Aupetit, C., Chauveau, C., and Cogne´, M. (1997). Identification of a homolog of the C움3⬘/hs3 enhancer and of an allelic variant of the 3⬘ IgH/hs1,2 enhancer downstream the human immunoglobulin 움1 gene. Eur. J. Immunol. 27, 2981–2985. Qiu, G., Harriman, G. R., and Stavnezer, J. (1999). I움 exon-replacement mice synthetize a spliced HPRT-C움 transcript which may explain their ability to switch to IgA. Inhibition of switching to IgG in these mice. Int. Immunol. 11, 37–46. Rajewsky, K. (1996). Clonal selection and learning in the antibody system. Nature 381, 751–758. Rinkenberger, J. L., Wallin, J. J., Johnson, K. W., and Koshland, M. E. (1996). An interleukin2 signal relieves BSAP (Pax5)-mediated repression of the immunoglobulin J chain gene. Immunity 5, 377–386. Sadhu, A., Shen, M.-L., Hackbarth, M., Hume, E., and McKeithan, T. W. (1997). CpGrich sequences close to the site of duplication within the human IgH constant region. Immunogenetics 45, 365–370. Sakai, E., Bottaro, A., and Alt, F. W. (1999). The Ig heavy chain intronic enhancer core region is necessary and sufficient to promote efficient class switch recombination. Int. Immunol. 11, 1709–1713. Saleque, S., Singh, M., and Birshtein, B. K. (1999). Ig heavy chain expression and class switching in vitro from an allele lacking the 3⬘ enhancers DNase I-hypersensitive hs3a and hs1,2. J. Immunol. 162, 2791–2803. Saleque, S., Singh, M., Little, R. D., Giannini, S., Michaelson, J. S., and Birshtein, B. K. (1997). Dyad symmetry within the mouse 3⬘ IgH regulatory region includes two virtually identical enhancers (C움3⬘E and hs3). J. Immunol. 158, 4780–4787. Seidl, K. J., Bottaro, A., Vo, A., Zhang, J., Davidson, L., and Alt, F. W. (1998). An expressed neor cassette provides required functions of the I웂2b exon for class switching. Int. Immunol. 10, 1683–1692. Seidl, K. J., Manis, J. P., Bottaro, A., Zhang, J., Davidson, L., Kisselgof, A., Oettgen, H., and Alt, F. W. (1999). Position-dependent inhibition of class-switch recombination by PGK-neor cassettes inserted into the immunoglobulin heavy chain constant region locus. Proc. Natl. Acad. Sci. USA 96, 3000–3005.

344

KHAMLICHI et al.

Sha, W. C., Liou, H.-C., Tuomanen, E. I., and Baltimore, D. (1995). Targeted disruption of the p50 subunit of NF– ␬B leads to multifocal defects in immune responses. Cell 80, 321–330. Singh, M., and Birshtein, B. K. (1993). NF–HB (BSAP) is a repressor of the murine immunoglobulin heavy-chain 3⬘움 enhancer at early stages of B-cell differentiation. Mol. Cell. Biol. 13, 3611–3622. Singh, M., and Birshtein, B. K. (1996). Concerted repression of an immunoglobulin heavychain enhancer, 3⬘움E(hs1,2). Proc. Natl. Acad. Sci. USA 93, 4392–4397. Sleckman, B. P., Gorman, J. R., and Alt, F. W. (1996). Accessibility control of antigenreceptor variable-region gene assembly: Role of cis-acting elements. Annu. Rev. Immunol. 14, 459–481. Snapper, C. M., Marcu, K. B., and Zelazowsky, P. (1997). The immunoglobulin class switch: Beyond ‘‘accessibility’’. Immunity 6, 217–223. Snapper, C. M., Rosas, F. R., Zelazowsky, P., Moorman, M. A., Kehry, M. R., Bravo, R., and Weih, F. (1996). B cells lacking RelB are defective in proliferative responses, but undergo normal B cell maturation to Ig secretion and Ig class switching. J. Exp. Med. 184, 1537–1541. Snapper, C. M., Zelazowsky, P., Rosas, F. R., Kehry, M. R., Tian, M., Baltimore, D., and Sha, W. C. (1996). B cells from p50/NF-␬B knock-out mice have selective defects in proliferation, differentiation, germ-line CH transcription, and Ig class switching. J. Immunol. 156, 183–191. Stavnezer, J. (1996). Antibody class switching. Adv. Immunol. 61, 79–146. Stu¨ber, E., Neurath, M., Calderhead, D., Fell, H. P., and Strober, W. (1995). Cross-linking of OX40 ligand, a member of the TNF/NGF cytokine family, induces proliferation and differentiation in murine splenic B cells. Immunity 2, 507–521. Taylor, L. D., Carmack, C. E., Huszar, D., et al. (1994). Human immunoglobulin transgenes undergo rearrangement, somatic mutation and class switching in mice that lack endogenous IgM. Int. Immunol. 6, 579–591. Tonegawa, S. (1983). Somatic generation of antibody diversity. Nature 302, 575–581. Tumas-Brundage, K. M., Vora, K. A., and Manser, T. (1997). Evaluation of the role of the 3⬘움 heavy chain enhancer [3⬘움E(hs1,2)] in Vh gene somatic hypermutation. Mol. Immunol. 34, 367–378. Urba´nek, P., Wang, Z-Q., Fetka, I., Wagner, E. F., and Busslinger, M. (1994). Complete block of early B cell differentiation and altered patterning of the posterior midbrain in mice lacking Pax5/BSAP. Cell 79, 901–912. Van Der Stoep, N., Gorman, J. R., and Alt, F. W. (1998). Reevaluation of 3⬘E␬ function in stage- and lineage-specific rearrangement and somatic hypermutation. Immunity 8, 743–750. Wabl, M., and Burrows, P. D. (1984). Expression of immunoglobulin heavy chains at a high level in the absence of a proposed immunoglobulin enhancer element in Cis. Proc. Natl. Acad. Sci. USA 81, 2452–2455. Wagner, S. D., Gross, G., Cook, G. P., Davies, S. L., and Neuberger, M. S. (1996). Antibody expression from the core region of the human IgH locus reconstructed in transgenic mice using bacteriophage P1 clones. Genomics 35, 405–414. Wagner, S. D., Popov, A. V., Davies, S. L., Xian, J., Neuberger, M. S., and Bru¨ggemann, M. (1994). The diversity of antigen-specific monoclonal antibodies from transgenic mice bearing human immunoglobulin gene miniloci. Eur. J. Immunol. 24, 2672–2681. Wallin, J. J., Gackstetter, E. R., and Koshland, M. E. (1998). Dependence of BSAP repressor and activator functions on BSAP concentration. Science 279, 1961–1964.

THE 3⬘ IgH REGULATORY REGION

345

Zelazowski, P., Carrasco, D., Rosas, F. R., Moorman, M. A., Bravo, R., and Snapper, C. M. (1997). B cells genetically deficient in the c-rel transactivation domain have selective defects in germline CH transcription and Ig class switching. J. Immunol. 159, 3133–3139. Zhang, J., Alt, F. W., and Honjo, T. (1995). Regulation of class switch recombination of the immunoglobulin heavy chain genes. In ‘‘The Immunoglobulin Genes’’ (2nd ed.) (F. W. Alt and T. Honjo, eds.), pp. 235–265. Academic Press, San Diego. Zhang, J., Bottaro, A., Li, S., Stewart, V., and Alt, F. W. (1993). A selective defect in IgG2b switching as a result of targeted mutation of the I애2b promoter and exon. EMBO J. 12, 3529–3537. Zwollo, P., Arrieta, H., Ede, K., Molinder, K., Desiderio, S., and Pollock, R. (1997). The Pax-5 gene is alternatively spliced during B-cell development. J. Biol. Chem. 272, 10160– 10168.

This Page Intentionally Left Blank

INDEX

A Acquired immune response, intracellular bacteria infection, 15–20 Actin, lymphocyte signaling actin cap effective molarity regulation, 110 function, 91–94 proximity regulator role, 110 structure, 90–91 cytoskeleton regulation, 105–107 polymerization regulation current view, 94–95 nucleus signal propagation, 108–110 RAC role, 94–100 Rho family GTPases role, 99–100, 104 WASP role, 92–94, 104–107 Activin signal transducer, transforming growth factor-웁 signaling, 130–131 Adjuvants cancer immunotherapy, 258–261 intracellular bacteria vaccination, 31–35 dendritic cells, 34–35 heat shock proteins, 29, 33–34 subunit vaccines, 31–33 AML, transforming growth factor-웁 signaling, 131–132 Animal models B cell differentiation via IgH 3⬘ regulatory region, humanized mice, 337 transforming growth factor-웁 signaling C. elegans, 124 Drosophila, 124 human cancer, 140–143 Xenopus, 123–124 vaccine efficacy studies, 5–6

Antibodies B cell differentiation via IgH 3⬘ regulatory region, 317–338 B cell development activity, 319–320 control region function possibilities, 330–337 chromatin structure, 335–337 germline transcription, 331–334 hypermutation, 330–331, 337 immunoglobulin class switching, 331–334, 337 immunoglobulin-H expression, 334–335 replication, 330–337 control region structure, 317–319 DNA-binding proteins, 323–330 3⬘ enhancer synergies, 320–322 humanized mice, 337 overview, 317, 337–338 B cell receptor signaling, 292–295 cancer immunotherapy, 263–264 intracellular bacteria protection role, 6–7 tumor T-cell response induction, 245–246 Antigen-presenting cells cancer immunotherapy role, 237 dendritic cell system model, 187–189 DNA vaccine delivery, 37 major histocompatibility complex class II subversion by pathogens, 189–191 Antigens B cell response differentiation, see B cells expression in cancer hepatitis B virus, 235, 240 human herpesvirus type 8, 239–240 human papilloma virus, 237, 240 non-virus-induced tumors, 241–246 CEA, 243–244 HER-2/neu, 244 347

348

INDEX

immunoglobulin, 245–246 MUC1, 244–245 p53, 242–243 virus-associated tumors, 240 intracellular bacteria infection antigen transition mechanisms, 59–61 DNA microarrays, 59–60 genome microarrays, 59–60 proteomics, 60–61 vaccinomics, 61 bacteria as antigen carriers, 49–56 compartmentalized expression, 55–56 heterologous antigens, 51–53 Listeria monocytogenes, 50–51 R-carriers, 55–56 Salmonella, 53–55 conserved versus specific proteins, 28–30 processing pathways, 14–15 recombinant carrier expression, 30–31 vaccine physicochemical properties, 27–28 processing by major histocompatibility complexes antigen internalization, 181–186 B cell receptor-mediated uptake, 183 Fc receptor-mediated uptake, 183–184 fluid-phase uptake, 181–182 mannose receptor-mediated uptake, 184–185 phagocytosis, 185 receptor-mediated uptake, 182–185 tumor antigens, 246–247 biosynthesis and assembly, 171–173 endoplasmic reticulum–Golgi transport, 172–173 folding mechanisms, 172 transcriptional control, 171–172 cell surface transport, 180–181 complex structures class II-associated peptides, 165–166 HLA-DM molecules, 169–171, 177–180 HLA-DO molecules, 169–171 MHC class II complex, 162–164 polygenicity, 164–165 polymorphism, 164–165 dendritic cell system model, 187–189 endocytic pathway entry, 173–176 overview, 159–160, 191–192 pathogen subversion, 189–191

pathway mechanisms, 14–15, 160–162 peptide loading, 176–180 processing mechanisms, 186–187 subcellular organelles role, 176–180 trans-Golgi-mediated sorting, 174–176 tumor antigens, 246–247 vaccinology, 14–15, 62 T cell response, see T cells AP-1 transcription factor complex, transforming growth factor-웁 signaling, 131 ATF-2, transforming growth factor-웁 signaling, 131 Autoimmune disease, T-cell receptor crossreactivity role, 209–226 cytokine checkpoints, 223–226 low affinity ligands, 211–214 overview, 209 peripheral purging of useless T cells, 218–220 peripheral T-cell tolerance, 214–217 T-cell antagonism, 211–217 viral-derived peptide agonists, 220–223

B Bacille Calmette Gue´rin antigen carrier role, 49–53 tuberculosis vaccination, 29, 46–47 Bacteria major histocompatibility complex class II antigen presentation subversion, 189–191 vaccinology, see Vaccines Bacteriophage, DNA delivery, 58–59 BAF chromatin-remodeling complex, lymphocyte signaling role, 109–110 B cell receptor antigen uptake role, 183 tyrosine kinase activation mechanisms, 288–292 response differences, 291–292 stage-specific differences, 289–291 B cells differentiation via IgH 3⬘ regulatory region, 317–338 B cell development activity, 319–320 control region function possibilities, 330–337

INDEX

chromatin structure, 335–337 germline transcription, 331–334 hypermutation, 330–331, 337 immunoglobulin class switching, 331–334, 337 immunoglobulin-H expression, 334–335 replication, 330–337 control region structure, 317–319 DNA-binding proteins, 323–330 3⬘ enhancer synergies, 320–322 humanized mice, 337 overview, 317, 337–338 differentiation via tyrosine kinase activation, 283–308 activation–cellular response connection, 304–307 development regulation, 284–288 heavy chain selection of preB cells, 284–286 mature B cell activation, 287–288 mature B cell maintenance, 288 tolerance induction of immature B cells, 286–287 immunoglobulin-움 and -웁 coreceptors, 292–295 overview, 283–284, 307–308 receptor response mechanisms, 288–292 response differences, 291–292 stage-specific differences, 289–291 tyrosine kinase families, 295–304 Btk of the Tec family, 302–304 Src family, 296–300 Syk of the Syk/Zap70 family, 300–302 memory, 23–25 Bone morphogenetic proteins, transforming growth factor-웁 signaling role, 115–116, 125–126 Btk tyrosine kinase, cellular response differentiation in B lymphocytes, 302–304

C Caenorhabditis elegans, transforming growth factor-웁 signaling by Smad proteins, 124 Calmodulin, transforming growth factor-웁 signaling, 128

349

Cancer immunotherapy strategies, 235–264 adjuvants, 258–261 antigens eliciting T-cell responses CEA, 243–244 future research directions, 246–247 HER-2/neu, 244 immunoglobulin, 245–246 MUC1, 244–245 non-virus-associated tumors, 241–246 p53, 242–243 virus-associated tumors, 240 CD4⫹ cells role, 237, 247–249 CD28 costimulation pathway, 261–262 cytokines, T cell response enhancement, 262–263 dendritic cells role, 247–249 monitoring methods, 263 monoclonal antibodies, 263–264 natural protection, 238–240 overview, 235–238, 264 T cell adoptive transfers, 256–258 tumor antigen processing, 246–247 tumor escape mechanisms, 253–256 tumor necrosis factor family, 249–253 4-1BB, 252–253, 256 CD30, 252 CD40, 238, 248–252, 256 CD95, 250–251 minor family members, 253 Ox-40, 253 transforming growth factor-웁 signaling by Smad proteins, 140–143 CBFA, transforming growth factor-웁 signaling, 131–132 CBP transcriptional coactivator, transforming growth factor-웁 signaling, 134 CD1, acquired immune response role, 18–19 CD4⫹ acquired immune response role, 15–20 antigen-processing pathways, 14–15 tumor immunity role, 237, 247–249 CD8⫹, see also Cytotoxic T lymphocytes acquired immune response role, 15–20 antigen-processing pathways, 14–15 tumor immunity role, 237–240 CD27, T-cell response in cancer, 253 CD28, tumor immunotherapy via costimulation, 261–262 CD30, T-cell response in cancer, 252

350

INDEX

CD40, T-cell response in cancer, 238, 248–252, 256 CD70, T-cell response in cancer, 253 CD80, tumor immunotherapy via costimulation, 249, 261–262 CD86, tumor immunotherapy via costimulation, 249, 261–262 CD95, T-cell response in cancer, 250–251 CEA, tumor T-cell response induction, 243–244 Chlamydial infection, characteristics, 11–12 Chronic myelogenous leukemia, immunotherapy, 236 CRE-BP-1, transforming growth factor-웁 signaling, 131 Crossreactivity T-cell receptor role in autoimmune disease, 209–226 cytokine checkpoints, 223–226 low affinity ligands, 211–214 overview, 209 peripheral purging of useless T cells, 218–220 peripheral T-cell tolerance, 214–217 T-cell antagonism, 211–217 viral-derived peptide agonists, 220–223 transforming growth factor-웁 signaling, Smad proteins, 137–140 MAP kinase pathways, 138–139 nuclear steroid receptor, 140 Smad pathways, 137 STAT pathways, 139–140 Cytokines, see also specific types autoimmune disease generation role, 223–226 innate immune response role, 13–14 tumor-specific T cell response enhancement, 262–263 Cytoskeleton, lymphocyte signaling, 89–110 actin cap effective molarity regulation, 110 function, 91–94 proximity regulator role, 110 structure, 90–91 actin cytoskeleton regulation, 105–107 actin polymerization regulation current view, 94–95 nucleus signal propagation, 108–110 RAC role, 94–100

Rho family GTPases role, 99–100, 104 WASP role, 92–94, 104–107 BAF chromatin-remodeling complex, 109–110 guanine nucleotide exchange factors, Vav family, 95–98 ligand-induced actin-dependent assemblies, functions, 91–94 overview, 89–94 Rho family GTPases, 98–100 actin polymerization, 99–100, 104 description, 98–99 supramolecular activation complexes functions, 91–94 structure, 90–91 WASP actin regulation, 92–94, 104–107 cytoskeleton regulation, 105–107 structure, 102–104 Vav-deficient mouse phenotype compared, 107–108 Wiskott–Aldrich syndrome, 100–104 Cytotoxic T lymphocytes, see also CD8⫹ cancer immunity, 236–240, 247–248, 256 host adjuvant use, 33–34 protection correlates, 25–26

D Dendritic cells antigen-presenting properties, 187–189 cancer immunotherapy role, 247–249 intracellular bacteria vaccination, 34–35 Disease, see Autoimmune disease; Infectious disease; specific diseases DNA transforming growth factor-웁 signaling, direct binding, 128–129 vaccines bacterial vector delivery mechanisms, 35–41 intracellular bacteria vaccination, 35–41 administration routes, 35–37 APC delivery methods, 37 DNA microarrays, 59–60 immunogenicity advances, 37–38 listeriosis vaccination, 38–39 tuberculosis vaccination, 39–41

351

INDEX

DNA-binding proteins, B cell differentiation via IgH 3⬘ regulatory region, 323–330 Drosophila, transforming growth factor-웁 signaling by Smad proteins, 124

E Embryogenesis, transforming growth factor-웁 signaling by Smad proteins, 142 Em intronic enhancer, B cell differentiation via IgH 3⬘ regulatory region, 317–318, 321, 330 Endocytic pathway, major histocompatibility complex class II role antigen processing mechanisms, 173–176 peptide loading, 176–180 trans-Golgi-mediated sorting, 174–176 Endoplasmic reticulum, major histocompatibility complex class II transport mechanisms, 172–173 Epstein–Barr virus, antigen expression in human cancer, 235–236, 240, 256 Evi-1, transforming growth factor-웁 signaling, 133 Experimental models, see Animal models

Golgi complex, major histocompatibility complex class II transport mechanisms biosynthesis and assembly, 172–173 sorting, 174–176 Guanine nucleotide exchange factors, lymphocyte signaling, Vav family, 95–98 Guanosine triphosphatases, Rho family role in lymphocyte signaling actin polymerization, 99–100, 104 description, 98–99

H Heat shock proteins, intracellular bacteria vaccination, 29, 33–34 Hepatitis B virus, antigen expression in human cancer, 235, 240 HER-2/neu, tumor T-cell response induction, 244 Human herpesvirus type 8, antigen expression in human cancer, 239–240 Human leukocyte antigen complex, see also Major histocompatibility complex antigen processing role, 169–171, 177–180 Human papilloma virus, antigen expression in human cancer, 237, 240

F I Fc receptor, antigen uptake role, 183–184 Forkhead activin signal transducer, transforming growth factor-웁 signaling, 130–131 4-1BB, T-cell response in cancer, 252–253, 256

G Genes, see also specific genes antigen transition mechanisms in intracellular bacteria, 59–61 DNA microarrays, 59–60 genome microarrays, 59–60 proteomics, 60–61 vaccinomics, 61 GM–CSF, tumor-specific T-cell response enhancement, 262–263

Immune response, see also B cells; T cells B cell response differentiation via tyrosine kinase activation, 283–308 activation–cellular response connection, 304–307 development regulation, 284–288 heavy chain selection of preB cells, 284–286 mature B cell activation, 287–288 mature B cell maintenance, 288 tolerance induction of immature B cells, 286–287 immunoglobulin-움 and -웁 coreceptors, 292–295 overview, 283–284, 307–308 receptor response mechanisms, 288–292 response differences, 291–292 stage-specific differences, 289–291

352

INDEX

tyrosine kinase families, 295–304 Btk of the Tec family, 302–304 Src family, 296–300 Syk of the Syk/Zap70 family, 300–302 cancer immunotherapy strategies antigens eliciting T-cell responses CEA, 243–244 future research directions, 246–247 HER-2/neu, 244 immunoglobulin, 245–246 MUC1, 244–245 non-virus-associated tumors, 241–246 p53, 242–243 virus-associated tumors, 240 cytokines, T-cell response enhancement, 262–263 intracellular bacteria infection acquired response, 15–20 innate response, 13–14 tumor T-cell response induction, 245–246 Immunoglobulins B cell differentiation via IgH 3⬘ regulatory region, 317–338 B cell development activity, 319–320 control region function possibilities, 330–337 chromatin structure, 335–337 germline transcription, 331–334 hypermutation, 330–331, 337 immunoglobulin class switching, 331–334, 337 immunoglobulin-H expression, 334–335 replication, 330–337 control region structure, 317–319 DNA-binding proteins, 323–330 3⬘ enhancer synergies, 320–322 humanized mice, 337 overview, 317, 337–338 B cell receptor signaling, 292–295 T cell tolerance to peptide antagonists, 215–216 tumor T-cell response induction, 245–246 Immunological memory, intracellular bacteria infection, 23–25 Immunotherapy, see also Vaccines cancer treatment, 235–264 adjuvants, 258–261 antigens eliciting T-cell responses CEA, 243–244

future research directions, 246–247 HER-2/neu, 244 immunoglobulin, 245–246 MUC1, 244–245 non-virus-associated tumors, 241–246 p53, 242–243 virus-associated tumors, 240 CD4⫹ cells role, 237, 247–249 CD28 costimulation pathway, 261–262 cytokines, T-cell response enhancement, 262–263 dendritic cells role, 247–249 monitoring methods, 263 monoclonal antibodies, 263–264 natural protection, 238–240 overview, 235–238, 264 T-cell adoptive transfers, 256–258 tumor antigen processing, 246–247 tumor escape mechanisms, 253–256 tumor necrosis factor family, 249–253 4-1BB, 252–253, 256 CD30, 252 CD40, 238, 248–252, 256 CD95, 250–251 minor family members, 253 Ox-40, 253 Infectious disease, see also Vaccines; specific diseases infection compared, 3 intracellular bacteria vaccination, 7–12 chlamydial infections, 11–12 listeriosis, 10–11, 38–39 salmonellosis, 8–10 tuberculosis, 3, 7–8, 39–41, 46–47 vaccination goals, 4 Innate immune response, intracellular bacteria infection, 13–14 Interferon-웂, innate immune response role, 13–14 Interleukin-2, tumor-specific T-cell response enhancement, 262–263 Interleukin-12 autoimmune disease generation role, 223–226 tumor-specific T-cell response enhancement, 262–263 Intronic enhancers, B cell differentiation via IgH 3⬘ regulatory region, 317–318, 321, 330

353

INDEX

L Leukemia, immunotherapy, 236 Ligands lymphocyte signaling, actin-dependent assembly induction, 91–94 T-cell activation in autoimmune disease, 211–214 Listeriosis antigen carriers, 50–51 characteristics, 10–11 DNA vaccines, 38–39 Lymphocytes, see B cells; Cytotoxic T lymphocytes; T cells

M Major histocompatibility complex class I acquired immune response role, 15–20 antigen-processing pathways overview, 160–162 tumor antigens, 237, 246–247, 256–259 vaccinology, 14–15, 62 Major histocompatibility complex class II acquired immune response role, 15–20 antigen processing antigen internalization, 181–186 B cell receptor-mediated uptake, 183 Fc receptor-mediated uptake, 183–184 fluid-phase uptake, 181–182 mannose receptor-mediated uptake, 184–185 phagocytosis, 185 receptor-mediated uptake, 182–185 biosynthesis and assembly, 171–173 endoplasmic reticulum–Golgi transport, 172–173 folding mechanisms, 172 transcriptional control, 171–172 cell surface transport, 180–181 complex structures class II-associated peptides, 165–166 HLA-DM molecules, 169–171, 177–180 HLA-DO molecules, 169–171 MHC class II complex, 162–164 polygenicity, 164–165 polymorphism, 164–165

dendritic cell system model, 187–189 endocytic pathway entry, 173–176 overview, 159–160, 191–192 pathogen subversion, 189–191 pathway mechanisms, 14–15, 160–162 peptide loading, 176–180 processing mechanisms, 186–187 subcellular organelles role, 176–180 trans-Golgi-mediated sorting, 174–176 vaccinology, 14–15, 62 Mannose receptor, antigen uptake role, 184–185 MAP kinase, signaling pathway cross-talk, 138–139 Memory, intracellular bacteria infection, 23–25 Monoclonal antibodies, cancer immunotherapy, 263–264 Mononuclear phagocytes, innate immune response role, 13–14 MSG1, transforming growth factor-웁 signaling, 135 MUC1, tumor T-cell response induction, 244–245 Mucosal immunity, intracellular bacteria infection, 20–23 Myelogenous leukemia, immunotherapy, 236

N Natural killer cells, innate immune response role, 13–14 Nuclear steroid receptor, signaling pathway cross-talk, 140 Nuclear transport, transforming growth factor-웁 signaling, 127–128

O Ox-40, T-cell response in cancer, 253, 256

P p53, tumor T-cell response induction, 242–243 p300 transcriptional coactivator, transforming growth factor-웁 signaling, 134

354

INDEX

Pathogens, see also Antigens; Vaccines major histocompatibility complex class II antigen presentation subversion, 189–191 pathogenicity–virulence compared, 3 PEBP2움, transforming growth factor-웁 signaling, 131–132 Phagocytosis, antigen uptake role, 185 Phagosomal escape, bacterial carriers, 56–58 Plasmids, DNA delivery, 58–59 Protection cancer immunotherapy strategies, natural protection, 238–240 intracellular bacteria infections antibodies role, 6–7 antigen characteristics conserved versus specific proteins, 28–30 processing pathways, 14–15 recombinant carrier expression, 30–31 vaccine physicochemical properties, 27–28 quantitative correlates, 25–26 Protein folding, major histocompatibility complex class II complex, 172 Proteomics, intracellular bacteria immunization analysis, genome to antigen transition, 60–61

R RAC protein, lymphocyte signaling role, actin polymerization regulation, 94–100 Resistance, see Tolerance Rho family GTPases, lymphocyte signaling, 98–100 actin polymerization, 99–100, 104 description, 98–99

S Salmonellosis antigen carriers, 53–55 characteristics, 8–10 live mutant vaccines, 42–46 defined mutations, 43–46 undefined mutations, 42–43

Serine kinase receptor, transforming growth factor-웁 signaling, 116–119 T웁R-I structure, 118–119 type I and II receptor characteristics, 116–117 type I subgroups, 117–118 Smad proteins, transforming growth factor-웁 signaling, 115–145 biological function, 125–126 cross-talk, 137–140 MAP kinase pathways, 138–139 nuclear steroid receptor, 140 Smad pathways, 137 STAT pathways, 139–140 cytoplasmic actions, 126–128 calmodulin interactions, 128 membrane anchoring, 126 nuclear transport, 127–128 oligomer formation, 127–128 expression regulation, 126 in vivo functions, 142–143 embryogenesis, 142 immune function, 142–143 tumor development, 142–143 I-Smads, 135–137 expression regulation, 137 function, 135–136 models C. elegans, 124 Drosophila, 124 human cancer, 140–143 Xenopus, 123–124 mutation biology, 141–142 nuclear actions, 128–135 AML interactions, 131–132 AP-1 complex interactions, 131 ATF-2 interactions, 131 CBFA interactions, 131–132 CBP transcriptional coactivator, 134 direct DNA binding, 128–129 Evi-1, 133 FASTs interactions, 130–131 interacting partners, 129–134 minor transcription factors, 132–133 MSG1, 135 p300 transcriptional coactivator, 134 PEBP2움 interactions, 131–132 TGIF transcriptional corepressor, 134–135 transcriptional repressors, 133

INDEX

overview, 115, 143–145 serine/threonine kinase receptors, 116–119 T웁R-I structure, 118–119 type I and II receptor characteristics, 116–117 type I subgroups, 117–118 structure, 121–123 MH1 domain, 122–123 MH2 domain, 121–122 proline-rich linker region, 123 subclasses, 119–121 superfamily characteristics, 115–116 Src tyrosine kinase family, cellular response differentiation in B lymphocytes, 296–300 STAT pathways, signaling cross-talk, 139–140 Supramolecular activation complexes, lymphocyte signaling functions, 91–94 structure, 90–91 Syk/Zap70 tyrosine kinase family, cellular response differentiation in B lymphocytes, 300–302

T T-cell receptor, crossreactivity, autoimmune disease role, 209–226 cytokine checkpoints, 223–226 low affinity ligands, 211–214 overview, 209 peripheral purging of useless T cells, 218–220 peripheral T-cell tolerance, 214–217 T-cell antagonism, 211–217 viral-derived peptide agonists, 220–223 T cells, see also specific types activation, T-cell receptor crossreactivity role in autoimmune disease, 209–217 low affinity ligands, 211–214 peripheral T-cell tolerance, 214–217 T-cell antagonism, 211–217 antigen processing, see Antigens cancer immunotherapy, see Cancer cytotoxic T lymphocytes cancer immunity, 236–240, 247–248, 256 host adjuvant use, 33–34 protection correlates, 25–26

355

purging useless cells, 218–220 signaling, cytoskeleton role, 89–110 actin cap effective molarity regulation, 110 function, 91–94 proximity regulator role, 110 structure, 90–91 actin cytoskeleton regulation, 105–107 actin polymerization regulation current view, 94–95 nucleus signal propagation, 108–110 RAC role, 94–100 Rho family GTPases role, 99–100, 104 WASP role, 92–94, 104–107 BAF chromatin-remodeling complex, 109–110 guanine nucleotide exchange factors, Vav family, 95–98 ligand-induced actin-dependent assemblies, functions, 91–94 overview, 89–94 Rho family GTPases, 98–100 actin polymerization, 99–100, 104 description, 98–99 supramolecular activation complexes functions, 91–94 structure, 90–91 WASP actin regulation, 92–94, 104–107 cytoskeleton regulation, 105–107 structure, 102–104 Vav-deficient mouse phenotype compared, 107–108 Wiskott–Aldrich syndrome, 100–104 tolerance, 214–217 autoimmune disease resistance, 216–217 immunoglobulin-derived peptide antagonists role, 215–216 vaccinology host adjuvant use, 33–34 intracellular bacteria protection role, 6–7 memory, 23–25 protection correlates, 25–26 Tec tyrosine kinase family, cellular response differentiation in B lymphocytes, 302–304 TGIF transcriptional corepressor, transforming growth factor-웁 signaling, 134–135

356

INDEX

Threonine kinase receptor, transforming growth factor-웁 signaling, 116–119 Tolerance B cell activation, 286–287 T-cell activation, 214–217 autoimmune disease resistance, 216–217 immunoglobulin-derived peptide antagonists role, 215–216 Transcription B cell differentiation regulation via IgH 3⬘ region, 331–334 major histocompatibility complex class II expression regulation, 171–172 Transcriptional repressors, transforming growth factor-웁 signaling, 133 Transforming growth factor-웁, Smad protein signaling, 115–145 biological function, 125–126 cross-talk, 137–140 MAP kinase pathways, 138–139 nuclear steroid receptor, 140 Smad pathways, 137 STAT pathways, 139–140 cytoplasmic actions, 126–128 calmodulin interactions, 128 membrane anchoring, 126 nuclear transport, 127–128 oligomer formation, 127–128 expression regulation, 126 in vivo functions, 142–143 embryogenesis, 142 immune function, 142–143 tumor development, 142–143 I-Smads, 135–137 expression regulation, 137 function, 135–136 models C. elegans, 124 Drosophila, 124 human cancer, 140–143 Xenopus, 123–124 mutation biology, 141–142 nuclear actions, 128–135 AML interactions, 131–132 AP-1 complex interactions, 131 ATF-2 interactions, 131 CBFA interactions, 131–132 CBP transcriptional coactivator, 134 direct DNA binding, 128–129

Evi-1, 133 FASTs interactions, 130–131 interacting partners, 129–134 minor transcription factors, 132–133 MSG1, 135 p300 transcriptional coactivator, 134 PEBP2움 interactions, 131–132 TGIF transcriptional corepressor, 134–135 transcriptional repressors, 133 overview, 115, 143–145 serine/threonine kinase receptors, 116–119 T웁R-I structure, 118–119 type I and II receptor characteristics, 116–117 type I subgroups, 117–118 structure, 121–123 MH1 domain, 122–123 MH2 domain, 121–122 proline-rich linker region, 123 subclasses, 119–121 superfamily characteristics, 115–116 Tuberculosis BCG vaccines, 29, 46–47 characteristics, 3, 7–8 DNA vaccines, 39–41 Tumor necrosis factor family autoimmune disease generation role, 223–226 T-cell responses in cancer, 249–253 Tumors, see Cancer Tyrosine kinase, cellular response differentiation in B lymphocytes, 283–308 activation–cellular response connection, 304–307 development regulation, 284–288 heavy chain selection of preB cells, 284–286 mature B cell activation, 287–288 mature B cell maintenance, 288 tolerance induction of immature B cells, 286–287 immunoglobulin-움 and -웁 coreceptors, 292–295 overview, 283–284, 307–308 receptor response mechanisms, 288–292 B cell response differences, 291–292 stage-specific differences, 289–291

357

INDEX

tyrosine kinase families, 295–304 Btk of the Tec family, 302–304 Src family, 296–300 Syk of the Syk/Zap70 family, 300–302

V Vaccines cancer treatment, 235–236 intracellular bacteria immunization, 1–63 animal models, 5–6 bacteria characteristics, 12–26 acquired immune response, 15–20 antigen-processing pathways, 14–15 innate immune response, 13–14 memory, 23–25 mucosal immunity, 20–23 quantitative correlates of protection, 25–26 genome to antigen transition, 59–61 DNA microarrays, 59–60 genome microarrays, 59–60 proteomics, 60–61 vaccinomics, 61 immunotherapy, see Immunotherapy infectious disease characteristics, 7–12 chlamydial infections, 11–12 listeriosis, 10–11, 38–39 salmonellosis, 8–10 tuberculosis, 3, 7–8 nomenclature, 3–5 overview, 1–7, 61–63 T cells role, 6–7 vaccine characteristics, 26–59 adjuvant vaccines, 31–33 antigen carrier role, 49–55 antigen physicochemistry, 27–28 attenuated mutations, 47–48, 53 BCG mutations, 46–47 conserved versus specific proteins, 28–30 defined mutations, 43–46 dendritic cells, 34–35

DNA delivery vectors, 58–59 DNA vaccines, 35–41 heat shock proteins, 29, 33–34 host adjuvants, 33–35 live vaccines, 41–49 M. tuberculosis mutations, 46–47 phagosomal escape function transfer, 56–58 protective antigen features, 27–31 R-carrier compartmentalized expression of antigens, 55–56 recombinant bacterial carrier expression, 30–31 Salmonella mutants, 42–46 undefined mutations, 42–43 Vav protein family, lymphocyte signaling characteristics, 95–98 activation mechanisms, 97 signaling mechanisms, 97–98 Vav1 structure, 96 WASP-deficient mouse phenotype compared, 107–108 Vectors, bacterial DNA delivery, 58–59 Virulence, pathogenicity compared, 3 Viruses, antigen expression in human cancer, 235–240, 256

W Wiskott–Aldrich signaling protein, lymphocyte signaling actin regulation, 92–94, 104–107 cytoskeleton regulation, 105–107 structure, 102–104 Vav-deficient mouse phenotype compared, 107–108 Wiskott–Aldrich syndrome, 100–104

X Xenopus, transforming growth factor-웁 signaling by Smad proteins, 123–124

This Page Intentionally Left Blank

CONTENTS OF RECENT VOLUMES

Volume 71

Volume 72

움웁/웂␦ Lineage Commitment in the Thymus of Normal and Genetically Manipulated Mice HANS JO¨ RG FEHLING, SUSAN GILFILLAN, AND RHODRI CEREDIG

The Function of Small GTPases in Signaling by Immune Recognition and Other Leukocyte Receptors AMNON ALTMAN AND MARCEL DECKERT

Immunoregulatory Functions of 웂␦ T Cells WILLI BORN, CAROL CADY, JESSICA JONESCARSON, AKIKO MUKASA, MICHAEL LAHN, AND REBECCA O’BRIEN

Function of the CD3 Subunits of the Pre-TCR and TCR Complexes during T Cell Development BERNARD MALISSEN, LAURENCE ARDOUIN, SHIH-YAO LIN, ANNE GILLET, AND MARIE MALISSEN

STATs as Mediators of Cytokine-Induced Responses TIMOTHY HOEY AND MICHAEL J. GRUSBY

Inhibitory Pathways Triggered by ITIMContaining Receptors SILVIA BOLLAND AND JEFFREY V. RAVETCH

CD95(APO-1/Fas)-Mediated Apoptosis: Live and Let Die PETER H. KRAMMER

ATM in Lymphoid Development and Tumorigenesis YANG XU

A CXC Chemokine SDF-1/PBSF: A Ligand for a HIV Coreceptor, CXCR4 TAKASHI NAGASAWA, KAZUNOBU TACHIBANA, AND KENJI KAWABATA

Comparison of Intact Antibody Structures and the Implications for Effector Function LISA J. HARRIS, STEVEN B. LARSON, AND ALEXANDER MCPHERSON

T Lymphocyte Tolerance: From Thymic Deletion to Peripheral Control Mechanisms BRIGITTA STOCKINGER

Lymphocyte Trafficking and Regional Immunity EUGENE C. BUTCHER, MARNA WILLIAMS, KENNETH YOUNGMAN, LUSIJAH ROTT, AND MICHAEL BRISKIN

Confrontation between Intracellular Bacteria and the Immune System ULRICH E. SCHAIBLE, HELEN L. COLLINS, AND STEFAN H. E. KAUFMANN

Dendritic Cells DIANA BELL, JAMES W. YOUNG, AND JACQUES BANCHEREAU

INDEX 359

360

INDEX

Integrins in the Immune System YOJI SHIMIZU, DAVID M. ROSE, AND MARK H. GINSBERG

Murine Models of Thymic Lymphomas: Premalignant Scenarios Amenable to Prophylactic Therapy EITAN YEFENOF

INDEX INDEX

Volume 73 Mechanisms of Exogenous Antigen Presentation by MHC Class I Molecules in Vitro and in Vivo: Implications for Generating CD8⫹ T Cell Responses to Infectious Agents, Tumors, Transplants, and Vaccines JONATHAN W. YEWDELL, CHRISTOPHER C. NORBURY, AND JACK R. BENNINK Signal Transduction Pathways That Regulate the Fate of B Lymphocytes ANDREW CRAXTON, KEVIN OTIPOBY, AIMIN JIANG, AND EDWARD A. CLARK Oral Tolerance: Mechanisms and Therapeutic Applications ANA FARIA AND HOWARD L. WEINER Caspases and Cytokines: Roles in Inflammation and Autoimmunity JOHN C. REED T Cell Dynamics in HIV-1 Infection DAWN R. CLARK, BOB J. DE BOER, KATJA C. WOLTHERS, AND FRANK MIEDEMA Bacterial CpG DNA Activates Immune Cells to Signal Infectious Danger HERMANN WAGNER Neutrophil-Derived Proteins: Selling Cytokines by the Pound MARCO ANTONIO CASSATELLA

Volume 74 Biochemical Basis of Antigen-Specific Suppressor T Cell Factors: Controversies and Possible Answers KIMISHIGE ISHIZAKA, YASUYUKI ISHII, TATSUMI NAKANO, AND KATSUJI SUGIE The Role of Complement in B Cell Activation and Tolerance MICHAEL C. CARROLL Receptor Editing in B Cells DAVID NEMAZEE Chemokines and Their Receptors in Lymphocyte Traffic and HIV Infection PIUS LOETSCHER, BERNHARD MOSER, AND MARCO BAGGIOLINI Escape of Human Solid Tumors from T-Cell Recognition: Molecular Mechanisms and Functional Signifcance FRANCESCO M. MARINCOLA, ELIZABETH M. JAFFEE, DANIEL J. HICKLIN, AND SOLDANO FERRONE The Host Response to Leishmania Infection WERNER SOLBACH AND TAMA´ S LASKAY INDEX